.S. DEPARTMENT OF COMMERCE
i Service
PB-265 385
5tate-of-the-Art: Military Explosives and
Propellants Production Industry. Volume I
The Military Explosives and PropefSants Industry
American Defense Preparedness Association, Washington, D.C.
Prepared for
Industrial Environmental Kssscirch Lab.-Cincinnati, Edison, N.J.
-------
EPA-600/2-76-213a
October 1976 Environmental Protection Technology Series
STATE-OF-THE-ART:
MILITARY EXPLOSIVES AND
PROPELLANTS PRODUCTION INDUSTRY
Vol. I - The Military Explosives
and Propellants Industry
\
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
TECHNICAL REPORT DATA
(Please read lutiniciions on the reverie be fort completing)
Mo.
l-l'A-600/2-76-213a
3. RECIPIENT'S ACCESSIOWNO.
S TITLE AND SUBTITLE
State-of-the-Art: Military Explosives
and Propellants ProdCiction Industry (3 vols)
Vol. I - The Military Explosives & Propellants
5. REPORT DATE
October 1976 (Issue Date)
6. PERFORMING ORGANIZATION CODE
Industry
7. AUTHOR(S)
Patterson, James; J. Brown; W. Duckert;
J. Poison; and N. I. Shapira
8. PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORG \NIZATION NAME AND ADDRESS
American Defense Preparedness Association
Union Trust Building
15th and H Street, N. W.
Washington, D. C. 20005
10. PROGRAM ELEMEN f NO.
1BB610
11. CONTRACT/GRANT NO.
R 802872
I2. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Laboratory - Cin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/12
IS. SUPPLEMENTARY NOTES
Vol. II - Wastewater Characterization
Vol. Ill - Wastewater Treatment
16. ABSTRACT
This study has surveyed the military explosives and propellant
manufacturing industry, covering both "GOGO" and "GOCO" facilities. Sources of
wastewater, volumes, and pollutant constituents have been reported where such
data existed.
Treatment technology currently in use at the various installations has been
described, including effectiveness of pollutant removal and secondary (air and
solid) waste generation. Systems under development at these military
installations have also been examined and evaluated in light of available
information.
The report consists of three volumes. Volume I presents general conclusions
and recommendations and describes the industry's manufacturing operations. '
Volume -II presents the bulk of the data concerning the wastewaters and the
treatment systems now in place. Volume III reviews and summarizes data from the
first two volumes and describes and evaluates the new treatment processes under
development at this time.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Wastewater
Industrial wastes
Explosives
Waste treatment
Propellants
Water pollution control
Chemical wastes
Military
Manufacturing
13B
15B
13. DISTRIBUTION STATEMENT
Public Distribution
19. SECURITY CLASS (Tliis Report)
Unclassified
20. SECURITY CLASS (Thispage)
Unclassified
122. PRICE
EPA\Farnt 2220-1 (9-73)
ftUSGPOi 1977 - 7S7-056/5484 Region 5-11
-------
EPA-600/2-76-213a
October 1976
STATE -OF -THE -ART
MILITARY EXPLOSIVES AND
PROPELLANTS PRODUCTION INDUSTRY
VOLUME I
The Military Explosives and Propellents Industry
by
James Patterson
Norman I. Shapira
John Brown
William Duckert
Jack Poison
American Defense Preparedness Association
Washington, DC 20005
Project No. 802872
Project Officer
Richard Tabakin
Industrial Environmental Research Laboratory - Cincinnati
Edison, New Jersey 08817
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S: ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
ib
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental
Research Laboratory, Cincinnati, of the U. S. Environmental
Protection Agency, and approved for publication. Approval does
not signify that the contents necessarily reflect the views and
policies of the U. S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
ii
-------
FOREWORD
When energy and material resources are extracted, processed,
converted, and used, the related pollutional impacts on our
environment and even on our health often require that new and
increasingly more efficient pollution control methods be used.
The Industrial Environmental Research Laboratory - Cincinnati
(lERL-Ci) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and
economically.
This project, "State-of-the-Art: Military Explosives and
Propellants Production Industry", was undertaken as part of
Environmental Protection Agency's Miscellaneous Chemical Industries
program to establish a baseline of information concerning the
military explosives industry, the magnitude of its waste problems,
and the adequacy of the industry's treatment technology. The
results of the study have indicated that many of the wastes do
present significant problems of toxicity and/or resistance to
treatment, in addition to problems unique to explosives. Although
some treatment systems in use do protect the nation's waterways
from contamination, others are inadequate, generate secondary air
or solid waste problems, or are not widely used due to budgetary
limitations. Further research effort is needed by EPA and/or
Department of Defense to control pollutants generated by certain
sectors of the industry. The data and results of the investigation
have been used extensively by EPA's Office of Water Programs in
developing standards for the explosives industry. It will also
allow engineering staffs at several commercial military manufac-
turing facilities to examine their wastes and compare control
technology with that being used or developed at other installations.
Finally, it will enable EPA to determine our own research efforts
in this industry and how they would relate to other programs.
Questions or requests for additional information should be directed
to the Industrial Environmental Research Laboratory - Cincinnati,
Field Station - Edison, New Jersey.
David G. Stephan
Director
Industrial Environmental Research Laboratory
iii
-------
ABSTRACT
This study, contained in three volumes, addresses the
wastewater effluents of the military explosives and propel-
lants production industry. Both manufacturing and LAP (Load,
Assemble, and Pack) activities are covered. Volume I des-
cribes the industry, as well as the production processes and
technology. Volume II details the wastewater effluents of
manufacture and LAP operations by product, process, and mil-
itary .installation, to the extent that data are available.
Volume.Ill describes and evaluates the effectiveness of
various treatment technologies for water pollution abatement
now in use or under investigation by product, process, and
military installation.
A comprehensive long-term effort has been underway by
the Department of Defense for a number of years for the pur-
pose of modernizing munitions production plants. Pollution
abatement is an integral part of the modernization program.
Although extensive study, research and development investi-
gations have been undertaken, and although significant water
pollution abatement and water management plans have been
developed, implementation is generally in only the initial
stages at selected military facilities. Major Government
emphasis and very substantial funding are essential to: the
continuation of necessary pollution abatement research and
development; the demonstration of promising new treatment
technologies; and the implementation of effective and econ-
omical treatment system construction programs. Recommenda-
tions are set forth in detail in Volume I.
iv
-------
TABLE OF CONTENTS
Page
ABSTRACT ly
ACKNOWLEDGMENTS vi
LIST OF FIGURES, VOLUME I viii
LIST OF TABLES, VOLUME I ix
VOLUME I
CHAPTER I - Summary, Conclusions, and
Recommendations 2
SECTION I - Executive Summary 2
SECTION II - Conclusions and Recommendations... 6
SECTION III - New Technology Investigations 15
CHAPTER II - Introduction 19
CHAPTER III - The Military Explosives and
Propellants Industry 36
CHAPTER IV - Explosives and Propellants
Production Technology 54
SECTION I - Manufacture of Explosives
and Propellants 54
SECTION II - Manufacture of Nitrating Acids.... 80
SECTION III - Loading Operations 84
SECTION IV - Approaches to Water Pollution
Abatement 90
-------
ACKNOWLEDGMENTS
This study was conducted for EPA by The American De-
fense Preparedness Association (ADPA) under Grant No.
R802872, which covered a portion of the Study expenses.
The balance was contributed by ADPA and by The AdHoc Com-
mittee members. The purpose of this paragraph is to iden-
tify the principal participants and their relationships.
a. Sponsor. The study Sponsor was The Office of
Research and Development of The U. S. Environmental Protec-
tion Agency. The project officer within EPA was Mr. Richard
Tabakin of the Industrial Environmental Research Laboratory - Cincinnati,
Field Station Edison, NJ.
b. ADPA. The American Defense Preparedness Association
assembled an AdHoc Committee from the industrial and educa-
tional communities for the specific purpose of conducting
this study. The Committee membership is as follows:
(1) Dr. James W. Patterson, Chairman of the
Department of Environmental Engineering, Illinois Institute
of Technology, Chicago, Illinois. Responsible for Chapter
VI, "Wasterwater Treatment."
(2) Dr. John A. Brown, President of John Brown
Associates, Berkeley Heights, New Jersey; and member of the
U. S. Army Senior Scientist Steering Group for Munitions
Plant Modernization and Pollution Abatement. Responsible
for Chapters III and IV, "The Military Explosives and
Propellant Industry" and "Explosives and Propellants Pro-
duction Technology."
(3) Mr. William Duckert, WAPORA, Inc., Washing-
ton, D. C. Responsible for Chapter V, "Wastewater Charac-
terization. "
(4) Mr. Jack R. Poison, Mason and Hanger Co.,
Iowa Army Ammunition Plant. Special Consultant.
(5) Mr. Norman I. Shapira (Col. U. S. Army -
Ret.), Consultant, Dunkirk, Maryland; ADPA AdHoc Committee
vi
-------
Chairman. Responsible for Chapters I-II and overall manage-
ment of the study program.
c. Department of Defense. The study would not have
been possible without the full cooperation of the Military
Services, and their support was generously and enthusias-
tically provided, specifically by the following commands:
(1) Headquarters, U. S. Army Materiel Command
(2) Headquarters, U. S. Naval Material Command
(3) Office of Development and Acquisition, Head-
quarters, U. S. Air Force
(4) Headquarters, U. S. Army Armament Command
(5) Headquarters, U. S. Navy Sea Systems Command
(6) Manufacturing Technology Directorate, U. S.
Army Picatinny Arsenal
(7) U. S. Army Environmental Hygiene Agency
This study is based substantially upon data and factual
information provided by the Military Services.
d. In particular, the following persons are cited for
their valuable and significant contributions:
(1) Lt. General Woodrow Vaughan, Deputy Commander,
U. S. Army Materiel Command
(2) Mr. Irving Forsten, U. S. Army Picatinny
Arsenal
(3) Mr. Gerald Eskelund, U. S. Army Picatinny
Arsenal
(4) Mr. Thomas Wash, Headquarters, U. S. Army
Armament Command
(5) Mr. Daniel Quagliarello, Headquarters, U. S.
Naval Material Command /;
(6) Col. William Gilley, U. S. Army Environmental
Hygiene Agency
vii
-------
LIST OP FIGURES
Page
Figure 1 - The Military Explosives and
Propellants Industry..: 21
Figure 2 - Major Explosives and Propellant
Facilities in the U. S 38
Figure 3 - Production of Basic Explosive
and Propellant Materials 43
Figure 4 - Production of Formulations
and Blends 45
Figure 5 - Production of Nitration Acids 46
Figure 6 - Major Wastewater Problems 49
Figure 7 - Batch Process for TNT Manufacture 57
Figure 8 - GIL Continuous Process for TNT
Manufacture 60
Figure 9 - TNT Purification GIL Continuous
Process 6l
Figure 10 - Tetryl Manufacturing Process at
Joliet AAP 64
Figure 11 - RDX-HMX Process 68
Figure 12 - Batch Nitrocellulose Manufacture 69
Figure 13 - Biazzi Nitroglycerin Process 73
Figure 14 - Water Balance for Illustrative,
52,800 LB/DAY, Biazzi NG Line at
Radford AAP 75
Figure 15 - Ball Powder Process at Badger AAP 78
Figure l6 - Red Water Reclamation Concept 91
Figure 17 - Current Water Balance, Batch NC
Line 93
Figure 18 - Proposed Water Balance, NC
Batch Line 94
viid
-------
LIST OP TABLES
Page
Table 1 - Picatinny Arsenal Pollution Abatement
Project - Manufacturing Methods and
Technology Engineering ................ 32
Table 2 - Major Projects ..........................
Table 3 - Principal Ingredients of Explosive and
Propellant Formulations ............... 35
ix
-------
VOLUME I
THE MILITARY EXPLOSIVES AND PROPELLANTS INDUSTRY
Introduction, Summary and Conclusions
The reader of this report is advised that it consists
of six chapters, contained in three volumes, each addressing
separate aspects of the explosives and propellants waste-
water effluents and treatment situations, and that duplica-
tion and repetition among these chapters has been kept to a
minimum. Thus, the reader is cautioned that the use or
interpretation of statements or evaluations taken out of
context from the study in its entirety could lead to serious
misunderstandings and incorrect assessments.
-------
CHAPTER I
SUMMARY, CONCLUSIONS. AND RECOMMENDATIONS
SECTION I - EXECUTIVE SUMMARY
1. Background.
This study was conducted by the American Defense Pre-
paredness Association (ADPA)* at the request of the Office
of Research and Development of the U. S. Environmental
Protection Agency.
2. Purpose.
The purposes of this study are as follows:
a. First, to define and characterize the typical
wastewater effluents from military explosives and propellants
production.
b. Second, to identify and evaluate explosives and
propellant wastewater effluent treatment processes, including
those now being used as well as those under investigation,
and to reach conclusions regarding their effectiveness.
c. And finally, to suggest RD&D programs which EPA
might support in order to make available to the explosives
and propellants production industry the most effective and
economical treatment technology for achievement of the ef-
fluent quality prescribed by the U. S. Congress.
3. Organization.
a. Volume I contains four chapters including an Intro-
duction to The Military Explosives and Propellants Industry,
as well as Conclusions and Recommendations.
b. Volume II contains only Chapter V, "The Wastewater
Effluents." This chapter includes all of the monitoring and
sampling data and is quite voluminous.
*
Formerly American Ordnance Association (AOA).
-------
c. Volume III contains only Chapter VI, "Wastewater
Treatment," and is devoted to a discussion and evaluation of
the treatment technologies, by product and process, and
constitutes the primary basis for the conclusions and
recommendations.
4. Scope of the Industry.
a. The industry is characterized by two major
"activities":
(1) "Manufacture" involves production of an ex-
plosive or propellant or intermediate product from raw
materials. Examples of manufacturing include TNT, nitro-
cellulose, nitroglycerin, nitric acid, sulphuric acid, etc.
(2) "Load, Assemble, and Pack (LAP)" involves
the loading of an explosive or propellant product into a
munition. It also usually involves "blending" of various
ingredients in the loading process.
b. The industry produces three major categories of
products: acids, explosives, and propellants. The explo-
sives and propellants are final products, whereas the acids
are intermediate products used in the manufacture of such
products as TNT, NC, RDX, etc.
5. Overview of the Military Explosives/Propellants Industry.
a. The U. S. Army Profile. The Army owns 17 "Army Ammuni-
tion Plants" (AAP) engaged in explosive and/or propellant
activities. One additional AAP is planned for construction
in Mississippi. Seven AAPs engage only in manufacture;
eight are involved only in "LAP" (Load, Assemble, and Pack)
activities, which involve the mixing, blending, and loading
of explosives or propellants into munitions; and two engage
both in manufacture and LAP.
(1) The Army manufactures all explosives used by
the U. S. Air Force, and all explosives used by the U. S.
Navy with the exception of nitroglycerin, which, although
an explosive, is used primarily as an ingredient of
propellants.
-------
(2) . The Army also manufactures propellants such
as nitrocellulose which are used by the Navy and Air Force.
*
(3) The Army's LAP plants not only load Army
munitions, but some Air Force munitions as well.
b. The U. S. Navy Profile. The Navy operates six in-
stallations, two of which are engaged in the manufacture of
explosives (nitroglycerin only) and propellants. These two
facilities also engage in LAP activities. The other four
Navy installations are involved only in LAP activities. The
Navy manufactures propellants for their own use as well as
use by the U. S. Air Force, and is a major contractor to the
U. S. Air Force for loading of explosives into bombs.
c. The U. S. Air Force Profile. The Air Force owns
only one plant within the scope of this study, which is
engaged in loading of propellants.
6. Limitations.
This is probably the most comprehensive study which has
been conducted both to characterize the wastewater effluents
of the military explosives and propellants production indus-
try and to evaluate associated wastewater treatment tech-
nology. Nevertheless, this study is limited to the degree
that it is based upon available sampling and monitoring data,
supplemented by on-site visits to the extent permitted by
time and financial resources. No actual monitoring or sam-
pling was done by the ADPA Committee.
7. General Conclusions.
a. The Department of Defense has displayed exemplary
initiative by engaging in an intensive effort to modernize
munitions production plants, including plans for extensive
programs designed for the abatement of pollution. A wide
range of improved treatment technologies is under investiga-
tion in research and development programs. Implementation
of improved treatment systems has been undertaken, but com-
pletion of the program is, in part, dependent upon future
availability of financial resources as well as the success
of current and future research and development efforts, and
is scheduled over a period of years. Substantial progress
has been made in pollution abatement. However, because of
-------
the magnitude of the effort and resources required, contin-
ued application of substantial funds will be necessary in
the areas of:
(1) Research and development of new treatment
technologies.
(2) Implementation of established treatment tech-
nologies as part of the long-term modernization program.
(3) Establishment of realistic criteria as a
valid basis for measurement of pollution parameters.
(4) Establishment of standardized analytical
procedures for monitoring programs.
(5) Implementation of more extensive, long-term
monitoring programs in a major waste characterization effort
to fill data gaps and to establish a more valid basis for
evaluation of effectiveness of treatment technologies.
(6) Standardization of treatment technologies
among the various plants for identical or similar products
or processes.
(7) Investigation of the energy consumption im-
plications of alternate treatment technologies.
(8) Continued implementation of water management
improvement programs.
b. The U. S. Environmental Protection Agency can con-
tribute substantially to improved pollution abatement by
initiation of an Inter-Agency program to fund the demonstra-
tion of promising treatment technologies now in research and
development by Department of Defense.
-------
SECTION II - CONCLUSIONS AND RECOMMENDATIONS
8. Introduction.
a. Because conclusions and recommendations are so
closely inter-related and interdependent in this study, it
has been found preferable to group them together. The rec-
ommendations are the obvious consequences of the Conclusions,
and are clearly stated herein as actions propqsed 'or
suggested.
b. The conclusions and recommendations are based upon
detailed professional evaluations of available data. Basic
data are presented in Chapter V (Volume II), wherein the
wastewater effluents are characterized. The evaluations of
current treatment technologies, as well as technologies
under investigation, are set forth in Chapter VI (Volume
III). No attempt is made here in Chapter I to repeat the
justifications given in Chapter VI for the .conclusions and
recommendations.
c. The evaluations of treatment technology presented
in Chapter VI (Volume III)a as well as the conclusions pre-
sented in this Chapter are based upon a wide range of data,
collection conditions and analytical procedures, as well as
a wide variety of data sources, many of which are incomplete,
and some of which are conflicting. As a result, the con-
clusions and recommendations assessing the effectiveness of
treatment technologies represent the best professional
judgements of the authors, within the constraints and limi-
tations of available factual data. In some instances, lack
of data in published form on current investigations have
precluded incorporation of adequate assessments of these
investigations. Data sources are cited in detail throughout
Chapters V-VI, and are listed in Appendix I, "References."*
i
d. It has not been possible, within the limited re-
sources available for this study, to investigate the appli-
cation of every possible treatment technology to every
product or process wastewater stream in the military explo-
sives and propellants production industry. However, all
technologies currently in use or under investigation by the
*Volume III
6
-------
Military, on which reports have been published, have been
included and are evaluated in this .study.
9. General.
a. A major program has been underway by the U. S. Army
for several years to investigate and implement technologies
for wastewater treatment and water management at Army ammu-
nition plants. A similar program of limited scope has been
undertaken by the U. S. Navy in a number of study and
research and development efforts. However, because of the
magnitude of the work remaining to be done, it is recommended
that the U. S. EPA participate in future research and de-
velopment programs by providing substantial funding support
for the demonstration of promising new treatment technologies.
b. Several pollution abatement procedures used or under
consideration by the military are significant energy consu-
mers. Examples are red water concentration and incineration
at several plants, and the proposed evaporation of pink and
yellow water at Newport AAP. There is a need to develop and
assess alternative, energy-conservative pollution abatement
technologies.
c. Because of the unique consituents and possibly toxic
nature of several process effluents of the military explosives
and propellants industry, there is a need to establish a valid
scientific basis for the levels of control required for in-
dividual constituents. Furthermore, such constituents as may
require control should be regulated and monitored as specific
parameters (such as TNT, NC, or KDX) rather than as general,
non-specific parameters such as BOD or COD. This is recom-
mended as the subject of a major new effort.
d. Although substantial progress has been made by the
U. S. AEHA, U. S. Army Edgewood Arsenal, and U. S. NSWL,
White Oak, a major, coordinated program should be initiated
for the establishment of standardized analytical procedures
and monitoring criteria for the production of statistically
significant data which are needed for the proper character-
ization of wastewater effluents.
e. Many product and- process effluent streams are not
adequately characterized with respect to waste constituents,
-------
constituent concentrations, or effluent volumes. Identifi-
cation of effective treatment technology requires individual
process effluent characterization, rather than monitoring of
combined waste flows. A major new waste characterization
effort should be undertaken to establish a more comprehen-
sive basis for the selection of future treatment technologies
f. Because of major variations in the types and amount
of products manufactured from month to month and year to
year, no brief or superficial monitoring program will pro-
vide adequate waste characterization information. Compre-
hensive, long-term monitoring programs are required both to
establish reliable waste characteristics and to assess the
effectiveness of current and future pollution abatement
programs.
g. The wide variation of proposed treatment processes
among various plants with identical products and similar
waste streams reflects the lack of establishment of industry-
wide pollution abatement standards. This diversity is ex-
emplified by the rigorous treatment proposed for nitric acid
wastes at Radford AAP and Volunteer AAP, as compared to the
plans for simple pH adjustment at Badger and Newport AAPs.
Pollution abatement criteria for specific products and
processes should be unified and standardized for the explo-
sives and propellants industry as a whole.
h. The combining of contaminated process wastewater
with non-contact or other non-contaminated discharges has
been a widespread practice. Combined wastewater adversely
impacts both on effective and on economic pollution abate-
ment. A major program of effluent segregation, such as has
been implemented at Radford AAP for NC, is recommended.
i. The use of evaporation/percolation ponds represents
an economical and effective disposal' method for some explo-
sive and propellant wastes. In the semi-arid regions of
the West and Southwest, the use of pits and lagoons by the
U. S. Navy is quite effective. However, in other areas,
such systems must properly balance seasonal rainfall and
evaporation against process waste volumes.
j. The potential for soil and groundwater contamina-
tion from percolation basins may be significant, and moni-
toring studies are recommended to establish design and
8
-------
operating criteria for ponds and lagoons. In instances
where percolation might result in unacceptable pollution of
ground or near-surface waters, sealed ponds or alternate
disposal methods are indicated.
k. There are numerous proposals for the application
of "biological waste treatment in the explosives and propel-
lants industry. In some instances, the proposed application
is for a wastewater containing constituents, such as phos-
phate, which are not amenable to biological treatment.
Furthermore, serious questions concerning possible bio-
transformation of wastewater constituents, and the environ-
mental impact of treatment by-products, remain unanswered.
Pending further study of these issues, caution should be
used in the adoption of biological waste treatment processes
for specific wastewaters of the explosives and propellant
industry.
1. Demilitarization represents a separate and distinct
activity of the military explosives and propellants indus-
try, as well as a substantial contributor of wastewater
effluents and air pollution. It is not covered in this
report, and should be the subject of a separate study.
10. Data Gaps.
a. In the characterization of the effluents from the
military explosives and propellant production and LAP proc-
esses, there are major data gaps in several areas for the
Army; and essentially no statistically meaningful data are
yet available for the Navy.
b. Waste characterization data for propellant products
are generally deficient.
c. Characteristics are generally not available for
pink water wastes from secondary sources, such as red water
evaporation condensate.
d. Essentially no waste characterization data are
available for several minor LAP operations and products.
-------
11. TNT. .
a. Insufficient attention appears to have been directed
toward optimization of activated carbon treatment of pink
water, despite evidence that both avoidance of color devel-
opment and maintaining acidic pH enhance treatment effic-
iency. In the light of the widespread proposed use of
carbon, studies to better define process design and operat-
ing criteria are recommended.
b. The current practices of use of carbon to exhaus-
tion, and incineration of spent carbon, are uneconomical
and represent a potential for air pollution. Intensive
studies on carbon regeneration are recommended. Both ther-
mal and solvent regeneration technologies appear promising,
although thermal regeneration apparently imposes a restric-
tion on the percent of carbon adsorption capacity available
and also represents a potential air pollution problem.
c. In the absence of economic carbon regeneration
techniques for pink water carbon systems, the potential
for use of polymeric resin adsorption or the combination
of resin plus carbon systems appears to have the advantages
of good regeneration capability for the resin and reduction
in carbon consumption.
d. While attractive from the economic standpoint,
biological treatment of pink water must be considered with
caution. Points of concern include: the need for large
quantities of supplemental nutrient; pollutant biotrans-
formations; and the nature and potential impact of treatment
by-products on the receiving environment, or the need for
application of supplementing polishing steps such as carbon
treatment and resin adsorption.
12. Acids.
a. Significant opportunities exist within the acid
manufacturing facilities for reduction in waste volumes,
water reuse, and similar water management efforts. One
example is the potential use of NAG effluent at Holston AAP
as a water source for nitric acid production and NO abatement.
. x
b. The results of current control of red water by
evaporation/incineration are: a pink water condensate
10
-------
requiring further treatment; a high energy demand; the pro-
duction of air pollutants; and land disposal of large
quantities of ash. Among available alternatives, the
Tampella process for reclamation of the inorganics, and
UV-catalyzed ozonation to destroy dissolved TNT hold suf-
ficient promise to warrant further study. In addition,
investigation of new treatment processes for red water is
recommended to supplement current studies on existing
techniques.
c. Effective pH control of wastes from acid manufac-
ture and use has not yet been fully implemented by the
military. The technology for pH control is well established,
and efforts should be made to implement effective control
procedures.
d. Plans to biologically treat wastewater from acetic
acid and anhydride manufacture will not be acceptable if
preliminary evidence of the presence of toxic constituents
and biorefractory organics is valid. Pending identification
of the extent of toxicity or resistance to degradation,
consideration of separate non-biological treatment is in-
dicated. Projected treatment of nitrate and sulfate waste-
waters by aerobic biological processes such as is proposed
at Holston AAP for acid area wastes is questionable, in the
light of the lack of demonstrated effectiveness of biolog-
ical treatment in abating these waste constituents.
e. Where sulfate control is sought through addition
of lime and precipitation of gypsum, treatment should be
directed to the most concentrated (in sulfate) waste streams
prior to their dilution by flows low in sulfate.
f. A more intensified effort is warranted for nitrate
control, in light of successful pilot studies carried out
by both the military and other industries. These studies
have established design and operation criteria, which can
be implemented for full-scale treatment.
g. Applicable processes for nitrate control include
biological denitrification and ion exchange. The suspended
growth columnar denitrification process appears promising
among biological processes, and the "Chem-Seps" process has
potential application for ion exchange control. These
should be investigated further.
11
-------
13. Nitrocellulose.
a. An abatement effort is recommended for NC produc-
tion at Badger AAP comparable to the successful programs
which have been implemented at Radford AAP.
b. Significant progress has been made in reducing
wastewater volumes and pollutant discharges from NC manu-
facture. Based upon programs implemented at Radford AAP,
it appears that wastewater resulting from NC manufacture
can be totally controlled with available technology.
c. Capability for sulfate control is limited either
to precipitation (as the calcium or barium salt), or to
non-selective processes such as ion exchange deionization.
If sulfate control is required in the explosives industry,
improved economical processes are needed. Further research
and development on sulfate control is recommended.
14. RDX/HMX.
a. Current, proven technology for RDX and HMX control
is limited to carbon adsorption, with polymeric resin ad-
sorption now showing some promise. Carbon treatment without
the capability for regeneration has similar disadvantages
to those encountered in carbon treatment of pink water.
Investigation of alternate techniques for treatment of RDX
and HMX is recommended.
b. Current treatment practice for RDX and HMX efflu-
ents is limited to carbon treatment of LAP wastewaters at
Joliet AAP- Carbon treatment appears effective to a degree,
although the preferential sorption of TNT indicates that
the design capacity of carbon systems intended to remove
TNT in addition to RDX or HMX from wastewaters must be op-
timized for RDX or HMX removal. This may result in over-
design of the system relative to TNT control.
c. Adsorption of RDX or HMX on polymeric resin is a
promising technique, with potential opportunity for resin
regeneration through alkaline hydrolysis of the explosive
upon the resin. Further study of the polymeric resin sys-
tem is recommended.
12
-------
d. The proposal of Holston AAP to biologically treat
wastewater from KDX and HMX manufacture is questionable, in
light of prior failure to show effective biodegradation, and
because of uncertainties concerning biotransformation and
potential environmental impact of treatment by-products.
15. Nitroglycerin.
a. The water management program proposed for Radford
AAP for nitroglycerin wastewaters should be considered for
implementation at Badger AAP and at NOS Indianhead.
b. Biological treatment of high NG wastewaters does
not appear feasible without dilution or pretreatment for
NG removal, and is not expected to control the high nitrate
and sulfate and alkalinity levels of the wastewaters.
c. Other than evaporative ponds, the only effective
treatment for NG wastewaters is the sulfide decomposition
process, and this has several undesirable side effects.
Therefore, a new program of research and development should
be undertaken to explore and assess alternative processes
both for decomposition and for recovery of NG (and DNG).
Such a program would be of value both to the military and
commercial explosives industries.
16. Sellite.
Longer aeration periods and additional mechanical
aeration capacity are needed for adequate sulfite treatment
of effluents from the sellite process at Joliet AAP.
17. Propellants.
a. Wastewaters associated with solvent propellant
manufacture are generally not treated prior to discharge,
except for some pH adjustment. There is significant oppor-
tunity for water management to reduce water use, and
achieve product recovery by recycling.
b. The water management program implemented at Badger
AAP for reduction of process waste volume, and use of
percolation/evaporation ponds Cor solventless propellant
wastewater, should be considered for application at Radford
AAP and NOS Indianhead.
13
-------
c. Propellant conveyance water used in the solvent
process of propellant production is relatively uncontamin-
ated. Recycle of this water should be considered.
d. "Water-dry" wastewater used in the solvent process
of propellant production should be recycled as part of the
water management program, in conjunction with solvent
recovery from the "water-dry" waste by distillation.
e. The wet screening water used in the manufacture
of ball powder should be recycled as part of a water man-
agement program to reduce such wastewater by as much as
50 percent.
f. Ion exchange should be investigated for the treat-
ment of lead-bearing wastewaters from the slurry mix of
solventless propellant production.
g. Still bottoms from solvent propellant manufacture
may be blended with solvent wastes, with the combined flow
treated by activated carbon.
18. LAP.
a. One source of wastewater at many manufacturing and
LAP facilities is air pollution control scrubber water.
Dry collection and control techniques such as used at AF
Plant #78, NAD Hawthorne and NWS Yorktown both avoid gen-
eration of liquid waste and provide the opportunity for
product recovery. The applicability of dry particulate
control should be assessed for other plants.
b. Completely enclosed, forced air conveyance systems
for the transport of powdered materials such as ammonium
perchlorate result in avoidance of the need for building
and equipment washdown and eliminate water pollution. Ex-
amples of such systems are at the Aerojet plant in Sacra-
mento, and at Redstone Arsenal in Alabama. Air can be
recirculated, or processed through dry collectors.
14
-------
SECTION III - NEW TECHNOLOGY INVESTIGATIONS
19. Introduction.
In addition to the established and developmental
wastewater treatment technologies which have been evaluated
in detail in this study, there are a number of experimental
technologies which merit consideration for possible inclu-
sion in future RD&D programs. The following list is not
exhaustive, but it highlights a number of new approaches
which have surfaced in this study.
20. Cross-Flow Filtration.
A new class of filter elements consisting of micro-
porous permeation tubes approximately one-quarter to one-
half inch in diameter is under development. Suspensions
to be filtered are pumped through these tubes at pressures
varying from 15-40 psi and at velocities of approximately
15 feet per second; clear liquid permeates through the
porous walls, and a concentrated slurry is produced. The
high slurry velocity and the turbulence of the flow keeps
the inner walls of the tubes scoured clean, and there is
normally little or no plugging of the pores. Suspended
solids are removed nearly quantitatively from primary sew-
age, fine fibers and suspended solids are removed virtually
quantitatively from a variety of industrial wastewaters,
and at least some oil-water emulsions are efficiently sep-
arated. Different developers offer these tubes in stainless
steel, heavy-wall polymer and membrane form. Filters such
as these are suggested for investigation^ for operations
where conventional filtration, centrifugation or settling
basins are not quite adequate.
21. Evaporative Dewatering.
There is a well-established, energy-conservative,
evaporative dewatering process which may offer better and
lower-cost disposal of streams such as red water. In this
process, an oil phase is added to the slurry or solution
to be dewatered, and the mixture is passed through a heat
exchanger and several effects of falling film evaporators.
The oil serves to reduce the viscosity, to carry the
15
-------
precipitated solids and to prevent fouling and caking on the
heat exchanger surfaces. The water is taken overhead, and
the solids are taken out the bottom of the last effect as a
now-anhydrous slurry in oil. The oil is centrifuged off and
recycled to the feed tank, and the solids can either be
burned for process heat or recovered for reuse or sale. The
condensed water is free of non-volatiles and is suitable for
reuse or discharge with only minimal polishing treatment.
If the initial feed contains as much as 4.2$ of non-volatiles
of 10,000 BTU/lb fuel value or better, an industrial-scale
unit can be self-sustaining and needs no external source of
process heat.
22. Catalyzed Ozonation.
It has recently been shown by a number of investigators
that ozone will readily oxidize normally refractory waste-
water constituents such as TNT and RDX if the ozone is
sparged into the wastewater in the presence of ultra-violet
radiation. Army Surgeon General sponsored studies have
shown that other refractory materials are also destroyed.
The economics have not yet been established, because the
process parameters have not yet been optimized; but UV-
catalyzed ozonation should be investigated for polishing
such dilute effluents as pink water, RDX waters, and the
condensate from red water concentration.
23. Reverse Osmosis.
a. Reverse osmosis has frequently proved disappointing
in attempted industrial applications, perhaps because insuf-
ficient research has been devoted to improved membranes.
There are a number of greatly improved membranes in the
developmental stage in various laboratories, including one
(polybenzimidazole) which withstands chromic acid, free
chlorine, strong nitric acid, strong caustic and tempera-
tures up to at least 75°C. In view of the many attractive
wastewater treatment processes which really good RO membranes
would make possible, a significant, government-supported,
membrane research and development effort is recommended.
b. In addition to fragility of current membranes, a
second limitation of RO processes for industrial applications
is the rapid rise of osmotic pressure with solute concentra-
tions, so that the concentration of aqueous solutions to more
16
-------
than about 3-5$ has generally not been feasible. Given
improved membranes, this limitation could be circumvented
by the use of countercurrent flow; and solute concentrations
could be raised to the saturation level. There are of
course engineering problems involved, and a significant
research and development effort is needed.
24. Powdered Activated Carbon.
The only really serious drawback to the use of activated
carbon for polishing wastewater streams such as pink water is
economic. Up to now at least, regeneration of spent carbon
beds has failed to restore adequate capacity at supportable
cost, and once-through use is too expensive. However, all
past work has been with granular carbon, and the much cheaper
powdered carbon may offer an attractive alternative. The Jet
Propulsion Laboratory has recently demonstrated the economic
use of powdered carbon for treatment of municipal sanitary
sewage, on a 10,000 gallon per day scale, adding the powdered
carbon directly to the incoming raw sewage as a filter aid,
in two stages, producing effluent intermediate in quality
between conventional secondary and conventional tertiary ef-
fluent. The spent, organic-laden carbon is pyrolyzed to make
new powdered carbon for recycle to the feed tank. Alterna-
tively, the sludge-laden carbon could be burned for fuel.
The JPL investigations, supported by EPA, are continuing and
should be followed closely for possible applicability to AAP
wastewater streams. Note that this process would readily
combine with on-site sanitary sludge disposal problems and
with the evaporative dewatering process described above.
2.5. Wastewater Extraction.
a. Once water management improvements have reduced con-
taminated wastewater flows to minimum-volume, maximum-
concentration streams, it may become practical to extract
them in countercurrent fashion with plant fuel oil. Soluble
organics would pass into the oil phase to be destroyed in
the plants' power furnaces where their small NOX, etc.,
contributions would be lost in that already in the stack
gas. The effluent water stream would be saturated with
hydrocarbon oil; but oil - unlike such species as TNT and
RDX - is readily biodegradable by conventional, well-
accepted treatment without any problems of "biotransforma-
tion" to possibly toxic species.
17
-------
b. A similar result, with perhaps fewer complications,
could be achieved by filtering the raw wastewater through
beds of crushed coal destined for the plants' power furnaces.
Coal is mostly organic in composition, and the Army's Con-
struction Engineering Research Lab has reported preliminary
success in removing TNT from water in this fashion.
26. Ion Exchange.
Ion exchange is being investigated by the U. S. Navy for
treatment of several AAP wastewater streams, but its full
potential is not yet being exploited in such areas as KDX/HMX
removal.. For example, there are a number of anion exchange
resins - some commercial and some experimental - which, when
in the hydroxide form, hydrolize RDX/HMX to products such as
nitrite, ammonia, etc., which are then adsorbed on the
column, so that the effluent no longer contains RDX or its
hydrolysis products. The adsorbed compounds can be purged
during regeneration, with no loss of active resin sites. An
extensive research and development program is recommended for
investigation of resin applications.
18
-------
CHAPTER II
INTRODUCTION
27. Background.
This study was conducted by the American Defense Pre-
paredness Association (ADPA)* at the request of the Office
of Research and Development of the U. S. Environmental
Protection Agency. The American Defense Preparedness As-
sociation is a national association of individuals and
industry whose primary interest is the defense preparedness
of the United States. It is a non-profit, non-political
organization which strives to improve the efficiency of
science-industry-Government relationships. This study is
one of many which ADPA has conducted for the U. S. Govern-
ment. The study was initiated in December 1973, and com-
pleted in October 1975.
28. Purpose.
The purposes of this study are as follows:
a. First, to define and characterize the typical
wastewater effluents from explosives and propellants pro-
duction. This has been done primarily by the assembly and
collation of data already available within the Military
Services.
b. Second, to identify explosives and propellant
wastewater effluent treatment processes, including those
now being used as well as those under investigation.
c. Third, to compare the above mentioned wastewater
effluent treatment processes with other treatment technolo-
gies available or under development, and evaluate their
effectiveness.
d. And fourth, to define RD&D programs which EPA might
support in order to make available to the explosives and
# •
Formerly American Ordnance Association (AOA).
19
-------
propellants production industry the most effective and econ-
omical treatment technology for achievement of the effluent
quality prescribed by the U.. S. Congress.
29. Scope.
The scope of the study can be defined in terms of:
o Military organizations,
o Military products,
o Military processes,
o Effluents, and
o Treatment technology.
The military organizations, products, and processes are
tabulated in Figure 1.
a. The organizations encompassed in this study include:
the U. S. Army; the U. S. Navy; and the U. S. Air Force.
From Figure 1, it can be seen that seventeen U. S. Army Am-
munition Plants (AAP) are included, as well as six U. S. Navy
installations and three U. S. Air Force plants. These in-
clude all of the military facilities engaged in the produc-
tion and loading of military explosives and propellants.
-~ b. The products and processes in the military explosives
and propellants industry are also listed in Figure 1, together
with principal ingredients used in formulations. Figure 1
lists "Product Capabilities." This terminology was selected
deliberately, since the activities of any specific military
installation vary significantly from time to time in terms
of both product and quantity, depending upon the production
demands of the Department of Defense.
(1) The industry is characterized by two major
"activities":
(a) "Manufacture" involves production of an
explosive or propellant or intermediate
product from raw materials. Examples
20
-------
Army
Ammunition Plants
Badger
Cnrnhusker
Holston.
Iowa
Indiana
Jollet
Kansas
Louisiana
Lake City
Longhorn
tone Star
Milan
Newport
Radford
Redstone
Sunflower
Volunteer
Navy
Ammunition Plants
NAD Crane
NAD Hawthorne
NOS Indianhead
NAD Me Alester
NIROP Magna
NWS Yorktown
Air Force
Ammunition Plants
Plant "lit
Plant 77
Plant 78
Activity
MFRE
•O iH Q.
•H O. O
3. $ k
XXX
X X
X X
X X
X
X X
XXX
X X
A
X X
X
LAP
2 *
U Bt
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
X
X
Product Capabilities
Acids
•0
>> oc«»«i*>CH at
JC fHUO^IV^HO3 rH
<5 04 UK » Z U W 0
X X X X
X X X X
XXX
X X X X
\
XXX
X X X. X
XXX
X X X X
Explosives
a
r-4 t.
i? §
EH E-i *> X X 3 »H
gz o « o z o i.
H SB E* (B S Z 0<
X
X -X
X
XXX X
X
X
X X
X
•
X
^
X
Propellahts
v . a
u to a
a a a o>
a « H *j
JQ +* ^1
01 1 C U
•H -H 01 O
M v t> a.
C 09
41 -P 11 *>
M •rt *> 3 1. C
nj oc°)J33oi -n
*) BOh>>>>> -
U EH X 3 j-BOrHrHXHEHXZ
•)0.d KXUUOX 1-H 0 0 D 0 Z 0 B
<
-------
of manufacturing include TNT, nitrocel-
lulose, nitroglycerin, nitric acid,
sulphuric acid, etc.
(b) "Load, Assemble, and Pack (LAP)" involves
the loading of an explosive or propellant
product into a munition. It also usually
involves "blending" of various ingredi-
ents in the loading process.
(2) The industry produces three major categories
of products: acids, explosives, and propellents. The ex-
plosives and propellants are final products, whereas the
acids are intermediate products used in the manufacture of
such products as TNT, NC, RDX, etc. The principal products
in each category are described in more detail in Chapters
III and IV. The category in Figure 1 entitled "Dry Assembly"
refers to the assembly of finished rocket motors and war-
heads, together with cases, electronics, etc., into finished
munitions. Dry assembly involves no wastewater effluents.
c. The effluents considered in this study include both
those from manufacturing operations and from LAP operations.
The effluents are discussed in detail, by product or process,
in Chapter V, and the effectiveness of their treatment is
evaluated in Chapter VI.
30. Limitations and Exclusions.
"--' a. Exclusions. Specifically excluded from the scope
of this study are the following:
(1) Metal parts used in munitions,
(2) Toxic chemical agents,
(3) Herbicides,
(4) Illuminants and incendiaries,
(5) Liquid propellants, which are primarily used
in non-military roles, such as NASA space activities, and
which are primarily manufactured by civilian industry rather
than in military facilities, and
22
-------
(6) Demilitarization of obsolete or defective
military munitions.
b. Limitations . This is probably the most comprehen-
sive study which has been conducted both to characterize the
wastewater effluents of the military explosives and propel-
lants production industry and to evaluate associated
wastewater treatment technology. Nevertheless, this study
is limited to the degree that it is based upon available
sampling and monitoring data, supplemented by on-site visits
to the extent permitted by time and financial resources. No
actual monitoring or sampling was done by the ADPA Committee.
(l) Available data for AAPs come primarily from
the U. S. Army Environmental Hygiene Agency (AEHA), and are
derived from sampling and monitoring programs usually of
weeks to months in duration.
(2) Essentially no statistically meaningful data
are available from the U. S. Navy, which has only recently
initiated sampling and monitoring programs. However, the
Navy manufacturing and LAP processes are sufficiently sim-
ilar to those used at AAPs in most cases, that one can
validly attribute to the Navy comparable effluents to those
generated at AAPs.
(3) The U. S. Air Force owns only one installation
involved with LAP operations, and no explosive/propellant
manufacturing facilities. As will be shown later, there is
no wastewater discharge from this installation.
It is believed that available data are suffi-
cient in scope and detail to be valid and representative of
all major products and processes, although certainly not
representative of every military facility, since there are
sometimes significant variations in product mix, production
volume, effluents and treatment technology from one plant to
another .
(5) Because of the significant variations in type
products and in production volumes from month to month, and
from year to year, it is clear that no brief or superficial
sampling program .can provide valid, representative, and
meaningful data.
23
-------
31. Methodology.
11
The methodology used in this study is relatively simple
and straightforward, and involves the identification, col-
lection, collation, analysis, and evaluation of data on the
effluents and wastewater treatment technology associated
with the production and LAP operations in the military ex-
plosives and propellants industry. The data are derived
from published reports and studies, and from observations
made through on-site plant visits.
a. Tasks. The approach is summarized in the following
tasks which constitute the scope of work of this study:
(1) Task I. Identification of existing data on:
explosives/propellants production wastewater effluents;
explosives/propellants production wastewater treatment
methods.
(a) Identification of existing data.
(b) Collection of existing data.
(c) Evaluation of adequacy of existing data.
(2) Task II. Determination of types and categor-
ies of basic data needed on: explosives/propellant produc-
tion wastewater effluents; explnsives/propellants production
wastewater treatment methods.
""• (3) Task III. Determination and identification
of additional data needed by evaluation and comparison of
results of Task I with the results of Task II.
(4) Task IV. Collection of additional data by
visits to selected production facilities.
(5) Task V. Collation of all data collected.
(6) Task VI. Evaluation of effectiveness of
present wastewater effluent treatment methods, and compari-
son of present treatment methods with other wastewater
treatment technology available or under development.
24
-------
(a) Evaluation of effectiveness of present
treatment processes.
(b) Collection of data on other wastewater
treatment methods.
(c) Comparison of other treatment technology
with present treatment processes, and
determination of potential for applica-
tion to the explosives/propellants pro-
duction industry.
(7) Task VII. Preparation of conclusions3 and
recommendations for EPA RD&D programs for improved treatment
of explosives/propellants production wastewater effluents.
(8) Task VIII. Final report.
b. References. The reports and studies used in this
program are listed in the Appendix to Volume III. Over
230 references were collected and evaluated.
c. Plant Visits. Visits were made by the Committee to
the following military installations for the following pur-
poses: to observe manufacturing and LAP operations at first
hand; to visit facilities used for the treatment or disposal
of wastewater effluents; to be briefed on modernization and
pollution abatement programs; and, to discuss research pro-
grams for the development of new treatment methods.
Installation Operator Location
Air Force Headquarters - Washington, D. C.
Air Force Logistics Command - Dayton, Ohio
Air Force Plant 44 Hughes Tucson, Arizona
Air Force Plant 77 Boeing Odgen, Utah
Air Force Plant 78 Thiokol Brigham City, Utah
Army Environmental
Hygiene Agency - Edgewood, Maryland
Army Materiel Command - Washington, D. C.
Edgewood Arsenal - Edgewood, Maryland
25
-------
Installation
Operator
Location
Hill Air Force Base
Holston AAP
Joliet AAP
Longhorn AAP
Naval Sea Systems Command
NAD Crane
NAD McAlester
NAD Hawthorne
NIROP Magna
NSWC White Oak
NOS Indianhead
NWC China Lake
NWS Yorktown
Pleatinny Arsenal
Radford AAP
Redstone Arsenal
32. Report Organization.
Holston
Defense
Corp.
Ogden, Utah
Kingsport, Tennessee
Uniroyal Joliet, Illinois
Thiokol Marshall, Texas
Washington, D. C.
Navy Crane, Indiana
Navy McAlester, Oklahoma
Navy Hawthorne, Nevada
Hercules Magna, Utah
Navy -White Oak, Maryland
Navy Indianhead, Maryland
Navy China Lake, California
Navy Yorktown, Virginia
Dover, New Jersey
Hercules Radford, Virginia
Thiokol Huntsville, Alabama
Because of the size of this study, it is organized into
three volumes, as shown in the table of contents.
a. Volume I contains the first four chapters: "Execu-
tive Summary, Conclusions, and Recommendations"; "introduc-
tion"; "The Military Explosives and Propellants Industry";
and "Explosives and Propellants Production Technology."
*
b. Volume II contains only Chapter V, "The Wastewater
Effluents." This chapter includes all of the monitoring and
sampling data, and is quite voluminous.
c. Volume III contains only Chapter VI, "Wastewater
Treatment," and is devoted to a discussion and evaluation of
the treatment technologies, by product and process, and
26
-------
constitutes the primary basis for the conclusions and
recommendations.
•^
d. A certain amount of duplication is deliberately
incorporated into the various chapters to provide for con-
tinuity and clarity, as well as for a certain degree of
integrity in each chapter.
33. Overview of the Military Explosives/Propellants Industry
a. The U. S. Army Profile. The Army owns 17 "Army Ammuni-
tion Plants"(AAP) engaged in explosive and/or propellant
activities. One additional AAP is planned for construction
in Mississippi. Seven AAPs engage only in manufacture;
eight are involved only in "LAP" (Load, Assemble, and Pack)
activities, which involve the mixing, blending, and loading
of explosives or propellants into munitions; and two engage
both in manufacture and LAP. Figure 1 lists also the prin-
cipal ingredients used in LAP activities, as well as the
explosives or propellants manufactured at each AAP.
(1) The Army manufactures all explosives used by
the U. S. Air Force, and all explosives used by the U. S.
Navy with the exception of nitroglycerin, which, although
an explosive, is used primarily as an ingredient of
propellants.
(2) The Army also manufactures propellants such
as nitrocellulose which are used by the Navy and Air Force.
(3) The Army's LAP plants not only load Army
munitions, but some Air Force munitions as well.
(4) All of the AAPs listed in Figure 1 are "GOCO"
plants - Government-Owned, Contractor-Operated.
(5) All acid manufacture by the military is made
at Army AAPs, although both the Army and the Navy also use
acids procured from civilian industry, as well.
b. The U. S. Navy Profile. The Navy operates six
installations, two of which are engaged in the manufacture
of explosives (nitr-oglycerin only) and propellants. These
two facilities also engage in LAP activities. The other
four Navy installations are involved only in LAP activities.
27
-------
(1). The Navy manufactures propellants for their
own use as well as use by the U. S. Air Force, and is a
major contractor to the U. S. Air Force for loading of ex-
plosives into bombs.
(2) All Navy facilities listed in Figure 1 are
"GOGO" plants - Government-Owned, Government-Operated.
(3) All explosives used by the Navy are manufac-
tured by the U. S. Army (except for NG).
c. The U. S. Air Force Profile. The Air Force owns
only one plant within the scope of this study. This plant,
AF Plant 78, is operated by Thiokol in Utah, and is engaged
in loading of propellants. The other two Air Force plants
listed in Figure 1, Plants 44 and 77, are typical of "dry
assembly" activities. They assemble finished missiles,
including rocket motors, cases, electronics, fuzes, etc.
There are no explosive or propellant wastewater effluents
associated with their activities.
d. Private Industry. In addition to operating Army
Ammunition Plants, private industrial contractors also
manufacture rocket motors which are used in military muni-
tions, as well as some explosives and numerous ingredients
which are used by the military- For example, Aerojet manu-
factures the first stage Polaris motor at a company-owned
facility. DNT, nitroguanidine and ammonium perchlorate are
manufactured in private industrial plants for use by military
installations.
e. Munitions Plant Modernization and Pollution Abate-
ment Programs. In recent years within the Army and Navy,
two fundamental factors have begun to have an impact on the
Army and Navy installations producing and loading explosives
and propellants. The first factor is the age (and obsoles-
cence) of the plants (thirty-plus years), and the consequent
need for modernization. The second factor is the onset of
action deadlines as a consequence of .pollution-control
legislation. These two considerations have resulted in the
development, and initial implementation, both in the Army
and Navy, of Munition Plant Modernization and Pollution
Abatement Programs, which integrate:
28
-------
Pollution Abatement
New Plant Construction
New Production Processes
Plant Modernization
Safety Measures
Water Management
Product Recovery
Waste Treatment Research and Development
(1) Army. The Army programs are entitled:
"Munitions Production Base Modernization and Expansion" and
"Pollution Abatement Engineering Program for Munition Plant
Modernization." Table 1 lists the principal Army programs
which are oriented toward pollution abatement. These pro-
grams are administered by the Manufacturing Technology Di-
rectorate of Picatinny Arsenal. Additional research and
development programs are supported by the AMC Environmental
Technology Office at Edgewood Arsenal. Table 2 lists
modernization projects each with a value of $1 million or
more, extending through 1981. The latter is a $7 billion
program, of which $192.5 million have been funded in FY 1975
alone. At Radford AAP alone, $25 million were funded for
water pollution abatement through FY 1973. The degree to
which this program will continue to be implemented will de-
pend, partly, on success of research and development pro-
grams, and substantially on Congressional appropriations in
future years. Clearly, this has been, thus far, a signifi-
cant program of major proportions.
(2) Navy. The Navy program is entitled: "Mod-
ernization of Naval Ordnance Field Activities." This program
is essentially still in the planning stage, with no substan-
tial implementation thus far, except for studies, design
engineering, etc. The basic study was initiated in November
1972 and completed in March 1974. This is estimated to be
a $106 million program, and includes five of the six Navy
installations involved in explosive/propellant manufacturing
and LAP.
29
-------
(3) Air Force. No program exists within the Air
Force comparable to those in the Army and Navy,' since the
Air Force essentially has no water pollution abatement prob-
lems resulting from explosives and propellants.
(4) Army versus Navy versus Air Force. Among the
three services3 the Army is the major producer of wastewater
effluents from explosives manufacture and LAP activities.
The Navy is responsible for a much smaller, but still sig-
nificant quantity and variety of effluents, and essentially
none are attributable to the Air Force.
(5) PL 92-500. Significant impetus has been
added to the military pollution abatement programs as a
result of the enactment of public law 92-500, the Federal
Water Pollution Control Act Amendments of 1972.
f. Demilitarization. This is a subject which is not
addressed in this study. Demilitarization frequently in-
volves large quantities of wastewater and air contaminants,
since much demilitarization of obsolete munitions involves
the removal of propellants or explosives by water under
high pressure or as steam as well as incineration. Large
volumes of water are used, and the wastewater is saturated
with a variety of effluents. Because of the magnitude of
this activity, it should be the subject of a separate
study.
g. Variations in Topography and Climate. Methods of
treatment and disposal of wastewater effluents are influenced
substantially by topography and climate. In the semi-arid
regions of the West and Southwest, with annual evaporation
rates of 50 or more inches of water, the use of pits, ponds,
and lagoons is fully effective for disposal of wastewater
effluents, since the water is lost by evaporation and per-
colation and the sludge is readily disposed of by burning
or detonation. There is no evidence of ground water con-
tamination, and drinking water comes from reservoirs or
wells at substantial distances fyom the effluent disposal
sites. In the Central and Eastern parts of the country,
which are characterized by high rainfall, low evaporation,
and close proximity of public waterways, much more severe
requirements exist for treatment and management of .the waste-
water streams so as to remove pollutants prior to release
into main outfalls.
30
-------
(1) Three plants are located in the West and
Southwest, as follows:
NAD Hawthorne, Nevada
NIROP Magna, Utah
AF Plant 78, Brigham City, Utah
(2) All other plants are located in areas of
moderate to substantial rainfall, and usually with outfalls
emptying into public waterways. Complete containment of
effluent streams within the military installation may be
possible, but usually not. An example where effective con-
tainment has been achieved by use of holding lagoons is NAD
McAlester, Oklahoma.
h. Effluents versus Pollutants. The effluents con-
sidered in this study include both those from manufacturing
operations and from LAP operations. They include such
things as: acid water; "red water"; "pink water"; dissolved
explosives; inorganic chemicals such as nitrates, sulphates,
etc.; organic chemicals such as solvents, polymers; and
suspended and dissolved solids. The word "effluent" is not
identical with "pollutant." Depending upon treatment and
disposal methods, as well as discharge limits, effluents
may or may not constitute pollutants. Certainly, effluents
are potential pollutants. For example, the effluents from
NG manufacture with the Biazzi process in the arid region
of Magna, Utah, are not pollutants; whereas the effluents
from the same process in the humid region of Indianhead,
Maryland are potential pollutants. Pink water from LAP op-
erations at NAD McAlester, Oklahoma, is safely disposed of
in lagoons without treatment. Pink water from LAP operations
at Joliet AAP must undergo carbon-column treatment prior to
release to the outfall.
i. Formulations. For convenience, the major explosive
and propeliant formulations and their principal ingredients
are listed in Table 3.
31
-------
TABLE 1
PICATINNY ARSENAL POLLUTION ABATEMENT
PROJECT - MANUFACTURING METHODS AND TECHNOLOGY ENGINEERING
CO
to
TECHNOLOGY INVESTIGATIONS
Bomb Wash-Out Water Treatment
Sellite Waste Treatment
Treatment of Nitrobody Wastes from TNT Production
Disposal of Wastes from Propellant Manufacture
Elimination of Sulfate Wastes
Elimination of Organic Solvent Wastes
Pollution Abatement in Processing Metal Parts
for Ammunition
Improved Treatment of Waste Materials from Primer
Mix Planks
Water Management
Utilization of Waste Energetic Materials
Regeneration of Activated Carbon Contaminated
with Explosives
Removal Techniques for Soluble RDX and HMX
Hazards Analysis of Pollution Abatement Techniques
SOx Abatement Methods
Solid Waste Soil Disposal Techniques
Removal of Small Quantities of Ammonia in Streams
Pollution Control in Loading of Explosives Con-
taining Ammonium Nitrate, Aluminum, RDX and TNT
Decontamination and Elimination of Bulk Materials
TASK APPROVED PLANNED
FY-75 FY-76 FY-77/80
2 COMPLETED - 4/70
3 COMPLETED - 12/72
9 x
10 x x x
11 x
12 xx
14 TRANSFERRED TO FRANKFORD
ARS
15 TRANSFERRED TO FRANKFORD
ARS
16 x x x
18 Now an R&D program, not
MM T
22 x x x
24
25
26
27
28
29
30
x
x
X
X
X
X
X
X
X
-------
CO
CO
TABLE 1 (Continued)
PICATINNY ARSENAL POLLUTION ABATEMENT
PROJECT - MANUFACTURING METHODS AND TECHNOLOGY ENGINEERING
TASK APPROVED
FY-75 PY-76
INSTRUMENTATION AND MONITORING
Methods and Equipment to Measure, Monitor, and
Control Pollutants
PILOT PLANT DEMONSTRATIONS
Abatement Methods
Explosive Contaminated Inert Waste Disposal
Explosive and Propellant Waste Incineration
Elimination of Nitrate Wastes
Disposal of Red Water from TNT Purification
Elimination of Phosphate Wastes from Cooling
Water and Boiler Blow-Down Effluents
Disposal of Lead Azide
Disposal of Methyl Nitrate from HMX and RDX
Manufacture
PLANNED
FY-77/80
19
x
X
4 X X X
5 COMPLETED - 12/72
6 x x x
7 x x
8 x x
17 SCHEDULED COMPLETION
7/74
20 x
23 x
-------
TABLE 2
MAJOR PROJECTS*
PROGRAM
AREAS
Single Base
Multi Base
TNT
Nitro-
Guanidine
Black Powder
RDX/Comp B
Melt Pour
Scamp
Major Cal
Support FAG
FISCAL YEAR
70
2
12
2
1
1
9
71
5
1
1
5
1
72
3
2
4
5
3
73
1
1
3
10
2
74 75
1
1
1
1
1
1
2
13 14
6 5
76
1
1
1
17
5
77
1
1
3
1
1
12
9
78
1
1
1
8
6
79
1
2
1
4
7
7
80
3
1
1
1
3
9
81
3
1
3
3
7
TOTAL 27 13
*$1.0 Million or over.
17
17
20
26
25
28
17
22
18
17
Note - This project list is only a projection. It is reviewed continually and is
subject to revision in response to developing technology and changing requirements
and funding.
-------
TABLE 3
PRINCIPAL INGREDIENTS OF EXPLOSIVE
AND PROPELLANT FORMULATIONS
Explosives
Composition A (1-6)
Composition B
Cyclotol
Explosive D
HEX
H-6
Minol
Octol
PBX
Tritonal
Principal Ingredients
RDX, Wax
TNT, RDX, Wax
TNT, RDX
Ammonium Picrate
RDX, AJ, Wax, CaC^a
TNT, RDX, AJ, Wax, CaCJtz
TNT, NH4N03, At
HMX, TNT
RDX or HMX and Polymeric
Binder
TNT, A&
Pr ope Hants
Single Base
Double Base
Cross-Linked Double Base
Triple Base
Composite
Base Grain Casting Powder
Nitrocellulose (NC), DNT
NC and Nitroglycerin (NG)
NC, NG, HMX, AP*, At, Poly-
urethane
NC and NG and Nitroguanidine
AP*, A&3 Polyurethane or
Polybutadiene
NC, NG, HMX, AP
*Ammonium Perchlorate
35
-------
CHAPTER III
THE MILITARY EXPLOSIVES AND PROPELLANTS INDUSTRY
Introduction.
This chapter presents an executive overview of the
structure and production operations of the military explo-
sives and propellants production industry, to give perspec-
tive to the more detailed technical discussions in later
chapters.
35. The Composition of the Military Munitions Industry
a. It is national policy that American heavy munitions
production shall be primarily a Government operation, and
not an industrial empire as it is in some other countries.
Most of the actual plants are operated by private contrac-
tors, but they are owned and controlled by the Federal
Government. The contractor is typically a division of a
major U. S. chemical company, which has bid for and received
the operating contract for a limited number of years. The
company is responsible for all industrial and business op-
erations and for quality control, but usually not for the
design of the product or of the chemical processes used.
The Government owns the plants and owns the products and is
the only customer. Such facilities are known as "GOCO" -
Government-Owned, Contractor-Operated.
b. There are some exceptions to the above description.
A few plants, mostly Navy, are operated directly by Govern-
ment employees (GOGO)*. Most small arms, most metal parts
and most rocket-propelled missiles are made by private
industry for the Government on contract, and so are some
specialty explosive and propellant materials; but the great
bulk of military explosive manufacture and loading is done
in Government plants.
c. Virtually all of the high explosives and most of
the propellants for all three Services are manufactured in
a network of Army Ammunition Plants (AAPs) scattered mainly
*
Government-Owned, Government-Operated.
36
-------
over the eastern and southern portions of the country- They
are loaded in other facilities of the same AAPs, in other
AAPs which do not themselves manufacture chemicals, and in
a series of Naval Ammunition facilities. Two Navy facilities
manufacture some supplementary ingredients, notably nitro-
glycerin and casting powder; but mostly they load explosives
obtained from the AAPs. The Navy, in fact, loads most of
the bombs for all three services. They also assemble and
load a variety of solid rocket motors with propellants partly
obtained from the AAPs and partly manufactured in-house. The
Air Force does no manufacturing of explosives and propellants
at all. They do oversee the manufacture of large solid
rocket motors such as Minuteman, and there are a few GOCO Air
Force plants for this purpose; but the total size of this
activity is small compared to the Army and Navy activity, and
these operations do not result in wastewater effluents. The
Army also loads some large rocket motors in GOCO facilities,
and still others are purchased from private contractor
facilities.
d. The map in Figure 2, following, shows the distribu-
tion of the AAPs and NADs which do the bulk of the high
explosive and propellant manufacturing and loading. There
are approximately as many more which deal in small arms or
metal parts or serve as storage depots, but they are outside
the scope of this study. There are also several AAPs which
load warheads but do not manufacture the fills; they are not
discussed in any detail in this chapter because their rela-
tively minor wastewater problems are well typified by those
facilities which are discussed. There are also a dozen or
so private facilities which make and load large rocket motors
for sale to one or another of the services, but they are also
beyond the scope of this study except for brief mention.
They do not present any different operations or problems.
e. Figure 1 itemizes the major operations of those
plants which do explosives and propellants manufacturing -
which accounts for the great bulk of the wastewater problems -
as well as those which fill warheads or rocket motors but
may or may not manufacture the energetic ingredients.
36. Major Steps in the Manufacture of Munitions.
a. The manufacture of end item munitions can be divided
into four major steps:
37
-------
CO
00
-T
coT— -L
Manufacturing AAP.
Loading AAP.
Manufacturing Navy Facility
Loading Navy Facility
FIGURE 2 - MAJOR EXPLOSIVES AND PROPELLANT FACILITIES IN THE U. S.
-------
o Manufacture or purchase of Ingredients
such as nitrocellulose, nitroglycerin,
TNT, HMX, binders, etc.
o Combination of ingredients into blends,
grains or formulations, such as cannon
powder, Composition B, extruded double-
base propellant grains, etc.
o Loading of warheads, bombs, rocket motors,
etc., with the blend-s or 'formulations.
o Final assembly and pack-out of complete
munitions, including metal parts.
b. Manufacture of Ingredients. The manufacture of
chemical explosives and propellants such as TNT or nitro-
cellulose is completely similar to any other chemical
manufacturing operation; and the plants are much like any
other chemical plant except that they tend to be more iso-
lated from their neighbors, more spread out within their
own fences, and rather less modern than average. They are
isolated and spread out for safety, since their products
are hazardous materials; and they are less modern than most
chemical plants because they tend to be rather old. That
last factor is changing rapidly at this writing. A massive
modernization program is underway, spurred by the need to
reduce air and water pollution but also taking full advan-
tage of opportunities to improve production rates and ef-
ficiency by the installation of new and modern equipment.
(1) A typical chemical manufacturing operation
is the production of trinitrotoluene - TNT, the dominant
military explosive. It involves the manufacture of huge
quantities of nitric and sulfuric acids, from sulfur and
ammonia, in chemical plants exactly similar to those which
make nitric and sulfuric acids in the civilian economy.
Mixed nitric and sulfuric acid is then combined with liquid
toluene in a series of reaction kettles to produce mononi-
trotoluene, dinitrotoluene, and finally trinitrotoluene
which is a solid. The trinitrotoluene is separated from
the mother liquor, washed with water, and then treated with
a sulfite solution called "Sellite" which extracts undesir-
able isomers and impurities, leaving purified a-TNT. The
TNT is washed, solidified, flaked, and finally shipped out
in boxes.
39
-------
(2) The operation is a thoroughly typical chemi-
cal manufacturing operation not unlike those found in the
plastics, petroleum, food, paint or solvents industries.
It involves pumping of liquids, stirring of slurries, heat-
ing and cooling of streams, filtrations, evaporations,
dryings, bagging and boxing, and all the other usual chemi-
cal manufacturing operations. It also involves wastewater
streams containing unreacted material and by-products, waste
acids, decomposition products and extracted impurities, and
exhausted reagents. These wastewater streams, if untreated,
constitute major pollution problems, and will be discussed
in detail in Chapter VI of this report.
(3) The manufacture of nitrocellulose is similar
in that it involves the same nitrating acids as used for
TNT, but used to treat cotton linters or wood pulp (raw
cellulose) in a series of vats and reactors similar to the
ones described for TNT. The crude nitrocellulose is sim-
ilarly subjected to a series of water and aqueous solution
washes until it is finally delivered as a purified, fibrous
mat - ordinarily wet with water or alcohol for safety.
Again, there are major wastewater streams laden with spent
reagents and extracted impurities.
(4) Tetryl, HMX, RDX, nitroglycerin, and a whole
host of minor ingredients are all made similarly - in typi-
cal chemical process plants with typical equipment and
controls and typical wastewater streams.
^ c. Production of Grains or Blends. Most explosives
and all propellants are blends of ingredients rather than
single substances. Composition B, for example, is an inti-
mate mixture of 60$ RDX and 40$ TNT, with a little wax
added for stability. It is normally made in the RDX plant
by blending in TNT and wax right on the production line
without ever storing the RDX. There are four different
Cyclotols, which are blends of RDX/TNT without wax: 75/25,
70/30, 65/35, and 60/40; and there are two Octols, blends
of HMX/TNT: 70/30 and 75/25. They are also made on the
RDX/HMX line. There are dozens of other formulations in-
corporating other ingredients such as tetryl, aluminum
powder^ PETN, etc. Most of them are made in the chemical
manufacturing plant and shipped to the loading plant as
flakes or chunks, but some of them can also be made up in
the melt-mix kettles in the loading plant.
40
-------
(1) Nitrocellulose and nitroglycerin are rarely
used singly, "but are usually incorporated into double-base
or multi-base propellants which may have up to a dozen
other ingredients including metal powders, oxidizers and
various stabilizers and burning rate catalysts. The pro-
pellants are processed into shapes which range from pinhead
size pellets to cylinders two or three feet in diameter and
six feet or so long. The pellets may be used as-is as gun
propellants, or they may be incorporated into still more
complex formulations in the casting of large rocket motors.
The larger cylinders are usually complete rocket motor
grains themselves, ready to be inserted into finished rocket
motor hardware.
(2) A small but important class of formulations
comprises the Composite Solid Rocket Propellants. Compos-
ites typically contain a major amount of an oxidizer such
as ammonium perchlorate or HMX, a metal powder such as
aluminum, a binder which is one or another type of rubber
(or a double-base), and up to a dozen trace ingredients
such as catalysts, stabilizers, etc. There are literally
hundreds of formulations, all to a degree similar; and the
choice comes down to specific missions, economics, and
special requirements.
d. Loading of End Items. The blends and formulations
described above may be loaded into their hardware in the
plant where they are made, or they may be shipped to another
plant for Load/Assemble/Pack (LAP). There are a number of
AAPs which are solely LAP plants, and four of the six Navy
plants engage only in LAP.
(1) Most current explosive fills are blends of
TNT with other ingredients, and are melt-cast loaded into
bombs and warheads. Typically, several thousand pounds of
the flaked blend are charged to a stirred melt kettle and
heated to just above the melting point of TNT. It forms a
mobile slurry when molten, and is poured into the empty
bomb or shell cases, where it solidifies. Additional in-
gredients such as aluminum may be added in the melt kettle,
depending on the particular formulation being poured and
the particular munition product.
(2) Small rocket motors are usually, but not
always, loaded with pre-shaped double-base grains which
41
-------
are simply slipped into the motor case like a battery into
a flashlight. Large motors are usually, but not always,
loaded with a cast-in-place, composite grain which is mixed
in special kettles and poured into the motor case to harden.
An alternate loading method is to fill the motor case with
tiny, loose, double-base pellets and then infuse the bed
with a Casting Solvent consisting of nitroglycerin and/or
similar other energetic solvents for double-base. The
solvent semi-dissolves the pellets and blends the whole
into a tough, rubbery mass of adequate mechanical strength.
Both the pellets and the casting solvent can, of course,
have numerous other, minor ingredients. There are enough
exceptions and overlapping cases to disprove any manufac-
turing classification; for example, pre-formed, slip-in
grains are also us.ed for certain rocket motors up to a
couple of feet in diameter, which is quite large by most
standards.
(3) There is also an as-yet small but growing
class of formulations known as "PBX", Plastic Bonded
Explosive. PBXs are similar to rubber-base, composite
rocket propellants in that they consist of 85$ or so of
powdered high-energy explosive incorporated into a "plas-
tic" matrix (which can be a conventional plastic or a
double-base) and cast into place.
e. Final Assembly and Pack-Out. The loaded warheads,
rocket motors, etc., are finally assembled with their cases,
electronics, etc., into finished munitions. This "dry"
assembly involves no wastewater effluents and so is only
mentioned in this report. Moreover, the manufacture of
metal parts, electronic components, and arming and fuzing
devices are beyond the scope of this study and are not
discussed.
37. Major Explosive and Propellant Products.
a. Although there are literally hundreds of different
explosive and propellant compositions, just four materials -
TNT, nitrocellulose, KDX, and nitroglycerin - account for
nearly all the production tonnage; and the mainstay is TNT.
Figure 3 makes this comparison graphically.
b. As noted earlier, the basic explosives - except for
TNT - are rarely used alone, but are mostly used in blends
42
-------
• 1969- 1971 PRODUCTION
CD CAPACITY
J
10 20 30 40 50 60 70 80 90
MILLIONS OF POUNDS PER MONTH
FIGURE 3 - PRODUCTION OF BASIC EXPLOSIVE
AND PROPELLANT MATERIALS
100
43
-------
or formulations; and the basic tonnages reappear in the
production of the composites. Figure 4 makes this comparison.
c. There are literally hundreds of different formula-
tions. Holston AAP alone, to pick an example, makes about
50 variants of TNT/RDX/HMX/etc., and a sampling of them is
listed below for familiarization:
o Composition B: KDX/TNT/Wax
o Cyclotol 75/25: RDX/TNT
o Octol 75/25: HMX/TNT
o Composition A-4: RDX/Wax
o LX-03-0: HMX/DATB/Viton A
There are as many variations on nitrocellulose, with and
without nitroglycerin; and there are hundreds of variations
of solid rocket propellants. The variations are really
minor in the present context, and a detailed consideration
of them would add little to the wastewater picture.
d. The manufacture of nitration acids, however, is
significant. AAP sulfuric acid production for example is
about four times the tonnage of TNT and amounts to about
4$ of the total US production for all purposes. Figure 5
shows the production rates of nitric and sulfuric acids.
Note the change of scale in Figure 5.
38. Wastewater Streams - Sources and Types.
a. The greatest number of AAP water pollutants (or
potential pollutants) are similar to those from any chemi-
cal manufacturing operation, e.g., acid drippings, solvent
spills, rust, stack scrubber drainings, floor washdown, and
the like; but an important few are military-unique, e.g.,
"Red Water" from TNT purification and water solutions of
various explosives themselves. They are discussed briefly
here under four headings, and in more detail in later
chapters.
44
-------
BALL POWDER
CYCLOTOL
SOLVENTLESS
DOUBLE BASE
SINGLE BASE
COMPOSITE
EXPLOSIVES
1969- 1971 PRODUCTION
CAPACITY
J
I
I
10 20 30 40 50 60 70 80 90
MILLIONS OF POUNDS PER MONTH
100
FIGURE 4 - PRODUCTION OF FORMULATIONS AND BLENDS
45
-------
OLEUM
DILUTE HNO
CONC. HNO
CONC.
• 1969- 1971 PRODUCTION
CD CAPACITY
100 200 300
MILLIONS OF POUNDS PER MONTH
FIGURE 5 - PRODUCTION OF NITRATION ACIDS
400
46
-------
(1) Acid Manufacture. Relatively small; mostly
leakage plus drainings from air pollution abatement
scrubbers.
o Acid waters, neutralized with lime or
soda ash.
o Sometimes azeotroping agents such as
n-propyl acetate.
o Sometimes heavy metals from equipment
corrosion.
o Nitrobodies from acid recovery.
(2) Basic Explosives Manufacture. The major
quantities and the toughest problems are here.
o Acid waters, treated with lime or soda
ash, chemical washes, spills, washdowns.
o "Red Water" from TNT purification. A
complex, brick-red solution of sodium
nitrate, sodium sulfate, sodium sulfite,
sodium nitrite, and about 17$ organics
which include sulfonated nitrotoluene
isomers and complex, unidentified dye-
bodies.
o Dissolved explosives, e.g., "Pink Water"
which is approximately 100 ppm TNT in
water.
o Suspended explosive particles - dust and
chips.
o Sometimes solvents such as acetone,
benzene, dimethyl aniline.
(3) Compounding of Explosives and Propellants.
Small. Generally similar to streams from manufacturing,
except that the chemical purification wastes are absent.
o "Dust and chips.
o Dissolved explosives. Generally a few to
100 ppm.
47
-------
o Solvents.
o Organic materials such as collagen.
o Ammonium nitrate or perchlorate.
(4) Load/Assemble/Pack Operations (LAP). Small.
Mostly floor washdowns and generally similar to Pink Water.
o Dissolved explosives.
o Dust and chips.
o Heavy metals from paints and corrosion
and. metal cleaning.
The washout of reject munitions at some LAP plants can con-
tribute substantial additional effluents when washout op-
erations are running.
b. The following Figure 6 summarizes - and oversimpli-
fies for clarity and emphasis - the most characteristic
wastewater problems of each of the products or operations
highlighted in Figure 1. Every operation has some aspects
of every problem, but these are the outstanding ones. The
major problems are the large volume, high concentration,
chemical wastes from the manufacturing operations, the
streams from acid manufacture and loading operations are
much smaller and much simpler. The most noticeable waste
streams are the Red Water and Pink Water from TNT manufac-
ture and loading, respectively. Red and Pink Water will
be discussed in detail in later sections of this report,
but they deserve mention in this introductory chapter too
because of their importance and their military-uniqueness.
(1) Pink Water is simply a solution of TNT in
water. a-TNT is soluble in water to the extent of approxi-
mately 100 ppm at ambient conditions, the exact value
depending strongly upon temperature and the presence or
absence of other solutes. Freshly-made solutions of TNT
in water are virtually colorless; but exposure to ultra-
violet light, including sunlight, causes the formation of
highly-colored, complex, incompletely identified substances
similar to dyes. They impart a characteristic pink color
which persists even after dilution down to a few ppm with
48
-------
Wastes
Acid waters, nitrate &
sulfate salts, etc.
Red Water
Pink Water
Other dissolved
explosives and/or
dust and chips
Organic solvents
and resins
Chromium and other
metals from corrosion
Perchlorate and
other oxidizers
Manufacture of Chemicals
EH
523
p
°8
0)
C
CD
j3
H
O
•P
O
£-1
-P
•H
C
•H
Ss
EH
+
+
+
+
H
£>j
£4
-P
EH
+
+
+
to
H
0$
•H
f*i
-------
clean water. The release of Pink Water to receiving streams
is thus objectionable.
(a) Pink Water is ubiquitous where TNT is
made or handled, because of the need to
wash down equipment and working areas
from time to time. The largest volumes
and the highest concentrations are found
in plants which manufacture TNT or un-
load it from obsolete warheads, but it
is also the major component in the (rel-
atively small) wastewater streams from
loading plants.
(b) There are a number of technically feasible
approaches to removing dissolved TNT
and the derived dyebodies from waste-
waters, and one or another is planned
for each installation with the problem.
They will be discussed in detail in
Chapter VI.
(2) Red Water - which can be almost black at
times - also contains dissolved TNT; however, it is not an
incidental stream like Pink Water; it is a major by-product
stream from TNT manufacture. The mixed-acid nitration of
toluene yields not only 2, 4, 6 -trinitrotoluene, the desired
product, but also a host of other isomers and by-products
amounting to approximately 4.5$ of the total yield; and it
is necessary to remove them from the product. They are re-
moved by extraction with a sodium sulfite - sodium carbonate
solution which sulfonates and dissolves them; the extract
is called Red Water.
(a) Red Water is a very complex and some-
what variable mixture containing 15$ or
so of sulfonated or sellited nitrobodies
and a number of inorganic salts . Typi-
cal components are:
Water
NaN02-NaN03
50
-------
Sulfonated or sellited nitro
compounds
Solids
Ash
(b) Red Water is not released into receiving
waters. Some of it is sold to paper
manufacturers who can use it as a proc-
ess feedstream, and the rest is evap-
orated to dryness and the residue
incinerated. Extensive research and
development is in progress on better
processes, and eventually its components
will probably all be reclaimed for re-
cycle. These processes are discussed in
greater detail in Chapter VI.
(c) The above discussion has purposely been
presented from a chemical viewpoint in
order to give a feel for the origin and
nature of the wastewater streams. It
has been usual to describe them in the
conventional terms of BOD, pH, suspended
solids, etc., and most of the surveys
and permits are couched in these terms,
although specific, military-unique par-.
ameters would be preferable. The waste
streams will be treated in detail in
this manner in Chapter V.
39. Munition Plant Modernization Program.
a. Both the Army and the Navy have extensive plant
modernization and pollution abatement programs planned and
underway, and the wastewater picture is changing rapidly.
So rapidly, in fact, that field data are out of date by the
time they can be put into a report. Perhaps the best way
to present the story is to force-fit all information into
one of three time eras.
(1) Before Modernization. Manufacturing processes
from the early 40's to the late 60's. In general, this meant
little or no pollution awareness or abatement, with things
51
-------
like Red Water being discharged directly into rivers; and
effluent characterization was essentially a description of
the raw process discharges.
(2) Transition Period. The past few years,
today, and the next few years, while new treatment processes
are being developed and installed. Presentday plants ex-
hibit a mixed picture of modern abatement facilities along-
side antique process units, and effluent streams range from
raw discharge to zero discharge. New construction is
everywhere, and even more construction is in the planning
stage. Research and development is in progress on even
better abatement processes. Consequently, even current
wastewater data have only the most limited significance
except as trend indicators.
(3) After Modernization. That day in the future
(1980's) when all the planned installations are complete
and on stream. There will still be a few problem areas
then, and they are highlighted in the Conclusions and Rec-
ommendations section of this report; but by and large, the
military munitions plants are planned to be models of clean
outfalls. Those effluent data will be the first to be truly
indicative of state-of-the-art or best-available. Future
data are not available, of course; but forecasts based on
laboratory and pilot plant data and on such actual instal-
lations as have been put on stream are available; and they
form the basis for conclusions herein as to what can be
achieved.
b. All of the AAPs and Navy facilities included in
this study were constructed during World War II (1941-1943).
They were built in areas which, at that time, were generally
remote from population centers; and they were built in re-
sponse to an urgent political-military need (the Second
World War) with no regard to environmental impact, which
was not a consideration'in those days. Thus, there is a
discrete period of approximately"*25-30 years (1940-1965/70)
during which these plants operated - some intermittently -
in response to political-military requirements such as Korea
and Vietnam, with essentially no change in manufacturing
technology or waste treatment methods.
c. The present time frame represents a dynamic, in-
terim situation during which important changes are taking
52
-------
place. Furthermore, there is a substantial overlap from
the condition of "traditional operations" into the present,
"interim" situation. It would be inappropriate and mislead-
ing to describe as "representative" many plant conditions
and many sampling and monitoring data incorporated into this
study. Generally, those sampling and monitoring data which
are available and which were used in this study (Chapter V)
were taken during the 1968-1973 time frame, and usually
represent the pre-"interim", or "traditional" 1940-70 manu-
facturing and waste treatment situation. The degree to
which implementation of modernization and pollution abate-
ment programs thus far impacts on treatment effectiveness
is discussed in detail in Chapter VI, "Treatment Technology.
d. The modernization program encompasses far more
than the mere installation of effluent treatment units. It
involves the replacement of obsolete process equipment, the
automation of hand-lines, the substitution of continuous
processes for batch processes, the reduction of outfalls
via recycle, re-use and cascading, and the clean up of -the
remaining outfalls by terminal treatment. There are also
a number of new manufacturing processes with reduced pollu-
tion potential under development, for example, new chemistry
for TNT manufacture which results in reduced production of
Red Water through the generation of lesser amounts of unde-
sired isomers. There are also new processes under develop-
ment to convert former wastewater streams into feed streams
within the plant, for example, the re-work of Red Water into
fresh Sellite solution and the reclamation of nitric and
sulfuric acids from spent nitration acid. Those moderniza-
tion programs which impact the wastewater picture are
discussed in detail under their respective technologies
later in this report.
e. The timetable for modernization is hazy, because
it depends upon the future availability of funding. Bil-
lions of dollars are involved, and national priorities
change. Most serious problems are scheduled for abatement
within the next two to five years, but some of the remainder
may stretch out into the I960's.
53
-------
CHAPTER IV
EXPLOSIVES AND PROPELLANTS PRODUCTION TECHNOLOGY
SECTION I - MANUFACTURE OP EXPLOSIVES .AND PROPELLANTS
40. Introduction.
This chapter presents the essential manufacturing
processes involved in the production of the major explosive
and propellant chemicals and their assembly into end-item
munitions, organized by product type or by major process
type as appropriate. Minor plant-by-plant variations are
not dealt with here. The basic explosives discussed below
account for virtually all of the tonnage of U. S. production,
There are approximately as many more minor items, but their
combined production is so small that their inclusion would
have no impact on the wastewater picture.
41. Trinitrotoluene (TNT).
a. TNT plants vary considerably in their physical
facilities, but they all embody the same chemical process,
the treatment of liquid toluene with mixed nitric and sul-
furic acids followed by removal of undesired isomers and
residual dinitrated toluene by conversion to soluble species
and extraction. The undesired isomers plus residual dini-
trated species are removed from the reaction mixture by
95.5% *-5%
HN03
treatment with aqueous sodium sulfite solution - called
"sellite" - which reacts with everything except the desired
2, 4, 6 isomer to form water-soluble sulfonate derivatives.
The remaining 2, 4, 6 material, a-TNT, is washed and cast
molten onto a flaker belt for pack-out.
b. The spent sellite solution, containing the extracted
sulfonate derivatives, is known as "red water"; and it
54
-------
constitutes a major, inherent waste stream. The kinetics
of the nitration reaction inherently yield approximately
i|.5# of the indicated by-products; and at 1969-19?! produc-
tion rates, they alone amount to approximately two million
pounds per month.
c. Red Water is a deep red, almost black, aqueous
stream containing an extremely complex mixture of nitroaro-
matics and various inorganic salts. The organic portion is
mostly accounted for by all the various sulfonated deriva-
tives of the three undesired TNT isomers, typified by the
following generalized structure:
S03Na
Other organic constituents include smaller fragments, dis-
solved a-TNT, and complex, unidentified, dye bodies formed
from the photolysis of a-TNT by sunlight. The inorganic
portion consists mainly of unreacted sulfite plus nitrite
and nitrate formed in the extraction reactions . Red Water
varies somewhat in composition from plant-to-plant and from
run-to-run, but typical components are:
Water
NaN02-NaN03
Sulfonated or sellited nitro compounds
Solids
Ash
d. At this writing, no red water is discharged into
receiving waters. It is all either sold for paper manufac
turing feedstream, or else evaporated to dryness and incin
erated to destroy, the organics.
e. There are two major variations of the TNT manu-
facturing process in current use: the conventional,
55
-------
three-stage, batch process; and the newer, continuous,
Canadian Industries Limited (GIL) process. Both use the
same chemistry - the nitration of toluene with a mixture
of nitric and fuming sulfuric acid (oleum). The sulfuric
acid acts as a catalyst and also as a dehydrating agent to
absorb the water formed by the nitration reactions.
f. Figure 7 shows the schematic flow diagram of the
batch process. The nitration reactions are carried out in
three consecutive batch units referred to as "Mono", "Bi",
and "Tri" Houses. The feed chemicals to the Mono House
are toluene and the waste acid from the Bi House fortified
with 60$ HNC>3. Reaction to mononitrotoluene goes smoothly
and exothermically, and cooling coils are used to keep the
temperature at about 40°C during mixing and then at about
6o°C for an hour or so afterward. All three isomers,
ortho-, meta-, and para-, are formed; but the ortho- is
predominant.
HN03
H2S04
The ortho- and para- isomers nitrate further to a-TNT in
later stages, but the meta- isomer represents an impurity
which ultimately shows up in the Red Water.
g. After reaction, the charge is allowed to settle,
the waste acid is transferred to a storage tank for subse-
quent recovery, and the partially nitrated toluene ("mono
oil") is pumped to the "Bi House" where further nitration
is effected with waste acid from the "Tri House" fortified
with 60$ HN03- This time the temperature is raised in steps
to 90°C, and the result is a mixture of all possible di-
nitrated isomers - "Bi Oil".
h. After settling and separation, the Bi Oil is
pumped to the Tri House where the feed acid is a mixture
of 98$ nitric acid and oleum. Temperatures are staged up
to 120°C for approximately two hours. The nitrated product
from this third nitration stage operation is crude TNT con-
taining a-TNT (2,4,6 trinitrotoluene) which is the desired
product, and unsymmetrical TNT isomers which are the
56
-------
Toluene
Mono-House
(First
N itration)
Nitric Acid
T
Mono-Oil
Mixed Acids
and Oleum
Spent Acid
Flake TNT
Storage or
Shipment
Bi-Waste
Bi-House
(Second
Nitration)
Bi-Oil
M Tri-House
Tri-Waste
(Third
Nitration)
Water-
Sell ite-
Filter Water-
Washhouse
Waste Acid
Red Yellow
Water Water
33
FIGURE 7 - BATCH PROCESS FOR TNT MANUFACTURE
57
-------
impurities. The crude TNT is fed to the Wash House for
purification.
i. The purification of crude TNT involves crystalli-
zation in water, neutralization of free acid with soda ash
and solubilization and removal of undesirable nitrated pro-
ducts by treatment with a solution of sodium sulfite
(sellite). The wastewater from the sellite purification
stage is the Red Water which is sent to the red water treat-
ment plant for disposal by evaporation-concentration and
concentrate incineration.
J. The TNT slurry is transferred to a filter tank
where it is washed and filtered on a screen leaving layers
of TNT crystals. The .crystals are reslurried with water
and pumped to a melt tank where TNT is melted and most of
the water is removed by evaporation. The molten product is
run into hot air dryers for the removal of residual water.
The water-free product is solidified on a water-cooled
flaker drum or stainless steel belt, and the resultant film
is removed in the form of small flakes by scraping the drum
or belt with a beryllium blade. The flaked TNT is boxed
and sent to a packing house for transfer to a storage or
loading area.
k. Four waste streams are shown in Figure 7: spent
acid, waste acid, red water, and yellow water. The spent
acid is not discharged; the nitric acid is distilled off and
re-used, and the residual sulfuric acid is sold for commer-
cial use. The waste acid (mostly spillage, floor drainage
and the like) is neutralized with lime or soda ash and dis-
charged to the chemical sewer. The yellow water, essentially
a dilute solution of crude TNT in water plus acids is re-
cycled to the second nitrator; and the red water is destroyed
as described above.
1. Acid manufacture and recycle is an integral part
of the TNT manufacturing operation; but since it is such a
major operation in its own right and acid has other plant
uses besides TNT, it is discussed separately in Section II.
m. Sellite manufacture is also a discrete operation
in batch TNT manufacture. A typical operation consists of:
(1) burning sulfur to produce SOgJ (2) countercurrent
scrubbing of the product gas with water to remove sulfur
58
-------
trloxlde and other impurities; and (3) absorption of SOg by
a solution of sodium carbonate (soda ash) in a countercur-
rent packed tower. The liquid discharged from the tower
flows into a tank and is recirculated until the desired
strength (16$ ^2803) is obtained. The soda ash solution
used in the process is about 22$ in strength and is prepared
by dissolving commercial soda ash in water. The wastewaters
from the sellite manufacturing operation result from gas
scrubbing, spills of soda ash and sellite solution, and
floor washdowni. They are minor.
n. Figures 8 and 9* following, present schematic flow
diagrams of the GIL continuous TNT process. It is not dif-
ferent in principle or chemistry from the batch process
described above, but is a modernized version. The nitration
of toluene is carried out in six nitrator-separator stages
with the organic phase (toluene-nitrobody mixture) flowing
countercurrent to the acid .phase. The first and third
nitration stages have two nitration vessels per separator
whereas the remaining four stages have only one nitrator
vessel per separator. The process features lower hold-up
of hazardous materials due to its continuous rather than
batch nature. It also has somewhat reduced waste stream
production due to its more efficient control of process
conditions and better utilization of recycle streams.
o. The CIL purification operation is also an incre-
mental improvement over the older process. The crude molten
TNT first passes through a mixer-settler washer where five
separate countercurrent water washes remove the free acids.
The acid wash (Yellow Water) is returned to nitrator No. 2
as acid make-up. The TNT then flows through two sellite
washers in series where it is neutralized with soda ash and
treated with sodium sulfite. Each of the sellite washers
is followed by a separator which separates the aqueous phase
(Red Water) from the purified TNT phase. The dilute Red
Water from the second separator is returned to the first
separator, and the more concentrated Red Water from the
first separator is sent to the Red Water Treatment Plant.
The sellite-treated TNT receives final countercurrent water
washes and is pumped to the Finishing Building for drying,
flaking, and packaging.
p. The purification for this modernized TNT process
differs from that for the older operation in that:
59
-------
Weak HNO
Recycle HNO
Weak HNO
- Yellow Water
Strong HNO.
Separator
Lhir-L I
Spent Acid
FIGURE 8 -CIL CONTINUOUS PROCESS FOR TNT MANUFACTURE
-------
o»
—Process Water
Soda Ash-i r-SO
Add 1
Washer
Sodium Sulfite
Process Water—i
Transfer Water-
Yellow Water
r
Sellite |^—H
Washer U »
, ^
^^
Sellite H|~"
Washer (*— »
p»
Red Water
Post
Sellite
Washer
To Finishing Building^
Purified
TNT Slurry
FIGURE 9 - TNT PURIFICATION, CIL CONTINUOUS PROCESS
-------
(1) water is used in place of sodium carbonate solutions
for the initial removal of free acids; (2) the TNT is al-
ways molten; and (3) sellite solution is prepared directly
from dry sodium sulfite instead of through the SC>2-carbonate
reaction. Both these changes result in reduced waste
streams.
q. Dinitrotoluene (DNT), a closely related explosive,
can be made in a TNT plant; but at present, crude DNT is
simply purchased from commercial sources and purified by
fractional crystallization. The crude DNT is charged to a
"sweat pan" and heated to 75°C. The temperature is then
decreased in 5°C steps until a temperature of 25°C is
reached, at which time most of the pure DNT is in crystal
form and the liquid .phase containing most of the impurities
is drained off to storage. The small amounts of impurities
adhering to the crystals are removed by gradually increasing
the temperature and draining the further "sweat". When the
freeze point reaches 3^°C, sweating is discontinued and the
charge is rapidly heated to 95°C with additional draining
of impurities until tests indicate the freezing point of
the remaining charge is above 63°C. .The molten purified
DNT is then transferred to a water-cooled kettle where
"graining" is effected by agitation as the charge cools.
The resulting DNT crystal powder is discharged to a hopper,
screened, and packaged in drums for storage and shipment.
r. DNT purification is essentially a pollution-free
operation, because the separated impurities are all fed to
the TNT manufacturing operation for conversion to TNT.
Aqueous effluents are limited to uncontaminated cooling
water.
s. The thrust of current TNT modernization plans
viz-a-viz wastewater clean-up is primarily the installation
of collection and treatment systems plus the conversion of
some old lines to the modern GIL process. There are some
new processes in the research and development stage which
offer much lower pollution potential, for example a low
temperature process under study at Stanford Research Insti-
tute; but they are not close to commercialization, and for
the foreseeable future TNT is going to be made by one or
both of the processes described above. There are also
studies in progress aimed at eliminating the Red Water prob-
lem by reclaiming its constituents for feeding back to the
62
-------
TNT manufacturing process, and these studies are discussed
in detail in Chapter VI.
42. Tetryl.
a. Like TNT, tetryl is manufactured by nitrating an
aromatic feed-stock in mixed nitric/sulfuric acid, followed
by washing and recrystallization. Demand for tetryl is
small compared to that for TNT, and the entire current U. S,
capacity (1.3 million pounds per month) is represented by
the twelve production lines at Joliet Army Ammunition
Plant. At this writing (1975), tetryl is not being manu-
factured; and is not expected to be manufactured in the
future. It is discussed anyway because it has been impor-
tant and could conceivably be again.
b. Tetryl production is a batch operation involving
sulfation of dimethylaniline (DMA) with 93# sulfuric acid
to produce dimethylaniline sulfate (DMAS), and subsequent
nitration of DMAS with a mixture of sulfuric and nitric
acids (see Figure 10, following). The nitrated mix is
cooled and sent to a stainless steel neutch (a container
with filter) where the free acid is separated. The crude
tetryl is washed and slurried through a tank, filtered,
and transferred to a dissolving tank where acetone from a
reflux condenser dissolves the tetryl and washes it into
a still charged with 45$ acetone. When all the tetryl
has been transferred to the still, the acetone is removed
by evaporation; and the precipitated tetryl is slurried to
a neutch, partially dried and transferred to the wet stor-
age area (lag house). When enough purified tetryl is
accumulated in the lag house, it is carted by powder buggy
to the drying house, where it is dried by hot air blown
through bins and then transferred to the packing-shipping
house for shipment and/or storage.
c. The major waste stream from tetryl manufacture
consists of spent nitric and sulfuric acids. These are
recovered and sent to the sulfuric and nitric acid concen-
tration facilities, described elsewhere in this report.
Other wastewater sources are the wash water from tetryl
purification, tank cleanings, chemical spills, cooling
water, and water and sellite solutions used for floor wash-
down and equipment cleanup.
63
-------
DMA H2S04
Sulfator
| DMAS
Cooling Wafer
_c
u
0
.5
|
Z
Floor Wash
Acid Tank
Drainage
Nitrator
J 1
t
Acid
Neutch
I
Wash Water Water
£
5
D>
1
Cool ing
Water
Floor
Wash
Wash
Water
Neutch
Jo,
Refining
1
t
Water
Neutch
L
•«
HNO /H SO. Mixture
J t. *r
Spent Acid
P
Acid
_ Recover'
1
Bubble Cap Water
Cooling Water
(to Nitrating Ditch)
jde Tetryl
Acetone
••Mi
Lag
• House •*
Storage
( To Teti
Dry
• House -<
Wash Water \
H0SO . and HNO0
24 3
V (to NAG /SAC Units)
Sellite Tank Cleanout
Floor Wash
Flush of decomposition unit
with 40 - 50# HNO3
ryl Ditch)
Pack
1 Floor Wash
Dry House
Ditch
^Outfall
Tetryl Ditch
FIGURE 10 - TETRYL MANUFACTURING PROCESS AT JOLIET AAP.
64
-------
43. RDX and HMX.
a. RDX, cyclotrimethylenetrinitramine, is the third
most important munitions explosive from a tonnage view-
point, with production running as much as 15 million pounds
per month. HMX, cyclotetramethylenetetranitramine, is a
by-product produced in much smaller amounts and separated
in the RDX purification process.
b. The chemistry of RDX/HMX manufacture is straight-
forward. The starting material, hexamine, is purchased
from commercial suppliers and nitrated with ammonium ni-
trate, nitric acid in acetic acid, and acetic anhydride
solution to yield the two products directly:
CH2 CH
ru rH PU
\f\\n ^"9 \>nn
1 o^ L
2 ^
'2
HEXAMINE RDX HMX
c. Manufacture starts with dissolving hexamine in
glacial acetic acid. This solution is pumped to a batch
nitration vessel where acetic anhydride is added, followed
by pre-prepared ammonium nitrate - nitric acid solution.
Nitration is carried out in a water-cooled reactor until
reaction ceases, and the mixture is then "aged" to effect
chemical equilibrium. The mixture is then diluted with
recycle wash water and pumped to the washing operation.
d. The crude RDX/HMX slurry is filtered, washed with
water, and re-slurried for transfer to the recrystalliza-
tion facility. There, an organic solvent (which can be
icyclohexanone, acetbhe or toluene, depending upon the type
I
65
-------
of crystal wanted) is added and the explosive dissolved.
Water is separated and solvent distilled off to leave a
hot, concentrated solution of explosive in solvent. The
solution is allowed to cool, and the purified explosive
crystallizes out.
e. The crystals are filtered off, re-slurried in
water and usually sent to a compounding operation. There,
the water-wet explosive cake is blended with molten TNT
(for example) in steam-heated kettles; the water is de-
canted; wax (if used) is added; and the molten slurry is
solidified on a cooled flaking belt for boxing and shipment.
f. HMX, produced as a side reaction in RDX manufac-
ture, is separated in the recrystallization step and is
recovered from the mother liquor by a separate recrystal-
lization.
i
g. Neither RDX nor HMX is ever handled pure and dry,
because the danger of accidental explosion is too great.
Most of the production is immediately incorporated into
formulations such as Composition B wherein the RDX crystals
are coated with TNT and/or wax. Smaller amounts are formu-
lated into compositions wherein the RDX is desensitized
with small amounts of wax or polyisobutylene. Where the
pure explosive is needed elsewhere for blending into other
compositions, it is shipped water- or solvent-wet.
h. Wastewater streams are partly recycled and partly
sewered, with additional recycle and/or recovery facilities
under construction at this writing. In general, major exit
streams are reprocessed; and cooling water, pump seal water,
floor washdowns, and the like are sewered.
i. The filtrate from the post-nitration filtration,
along with the first wash water, is sent to a recovery
facility where the nitric acid is neutralized with sodium
hydroxide and pumped to a primary evaporator. About 80$
of the feed is volatilized and condensed as 60$ acetic acid
for re-use. The remaining 20$ is withdrawn as a sludge,
diluted and heated to about 100°C. A seed slurry of RDX
is added; and the mixture is cooled, whereupon a crop of
RDX-HMX crystals separates and is sent back to the washing
step of the manufacturing line. The remaining liquid is
sent to a secondary evaporator that recovers more acetic
66
-------
acid, and the sludge from the secondary evaporator is
steam stripped to recover the remaining acetic acid. The
recovered acetic acid is all purified, concentrated and
re-used.
j. Sodium hydroxide is added to the stripped sludge,
converting the ammonium nitrate in the sludge to sodium ni-
trate and ammonia, the residual acetic acid to sodium
acetate, and the residual RDX and HMX to ammonia, formates,
amines, and sodium nitrate. The ammonia and amines re-
leased in the reactor are absorbed in water and distilled
to recover anhydrous ammonia and amines. The sludge, Con-
taining mostly sodium nitrate, is currently being made into
fertilizer. Small amounts of ammonia and amines are dis-
charged in the effluent wastewater.
k. Other than the above, most of the wastewater
consists of cooling water, pump seal water, condensate,
washdown, and the like; and is currently sewered to the
river. Modernization currently in progress will provide
treatment, and is discussed later in this report.
1. Figure 11 diagrams the RDX-HMX manufacturing
process. The acid manufacturing and recovery processes
are discussed separately in Section II, "Manufacture of
Nitrating Acids."
44. Nitrocellulose.
a. Nitrocellulose, at 25 million pounds per month
recent production, is the second largest (after TNT) volume
AAP product in the U. S. It is made by the mixed-acid
nitration of cellulose, a natural high polymer of sugar ob-
tained from cotton linters or wood pulp. Figure 12 outlines
H ON02 CH2ON02
Cellulose
Nitrocellulose
J n
67
-------
Ammonia
Oxidation
1
Weak HNO
HN03
Concn
I
Cone HNO,
Mfg. & Mix
Acetone or
Cyclohexanone
I
3
i _
soln
^
-\ examine
Store
^ —
r^
Acid
Concentration
1 1
^^^ f
^ Acetic
Anhydride
Mfg.
1
_L
Mix
!
Nitration
\
Wash
i
• m
<
<
(
»«
4
<
1
<
*
e
Distill
Recrystallize
Wax, e.g.
Q>
Mix
I I TN.T'
I
Bottoms
Caustic
e.g.
Melt, Mix
!* *
Comp C-4, e.g. Comp B, e.g.
Disposal
FIGURE 11 - RDX - HMX PROCESS
68
-------
Nitrator
I
Wringer
I
Drowning Tub
I
Boil ing Tub
1
'Spent Acids to Recovery Plant
Boiling Tub Pit
• Acid Water to
Neutralization
Plant
oearer
t
Poacher
i
wringer
MMM
— »i
I
Water To
Recovered
Water Tank
Press
Damp NC
FIGURE 12 - BATCH NITROCELLULOSE MANUFACTURE
69
-------
the manufacturing process. Pre-purified cotton linters or
wood pulp are shredded and dried to remove excess moisture
and then treated with mixed nitric and sulfuric acids in
"dipping pots" fitted with agitators to esterify most of
the hydroxyl groups. In theory, it should be possible to
nitrate all of the hydroxyl groups as shown in the ideal-
ized formula above for a nitrogen content of 14.14$; but
in practice the most desirable compositions fall between
10.5$ and about 13.8$, representing a hydroxyl substitution
of about 1.8 to 2.9 per glucose anhydride unit. These
values are controlled by the proportions of acids used.
The reactor is cooled, and treatment is continued until
reaction stops. The result is a slurry of nitrocellulose
fibers in mixed acid now somewhat diluted with the water
formed in the reaction.
b. The NC/acid slurry is passed through a centrifugal
wringer which removes the bulk of the spent acids for re-
covery, and is then dumped into a drowning tub filled with
water. The NC/water slurry then enters a large diameter
flume line and is conveyed to the boiling tub house. The
water used to convey the NC acquires about 3$ total acidity
and is discharged to the boiling tub pits. Water is then
added to the NC in the boiling tub to attain a total acid-
ity level of 0.25 to 0.50 percent. The mixture is heated
to 96°C and boiled for 15-40 hours (acid, boil).
c. At completion of the acid boil operation, the
water is drained to the boiling tub pits. The NC is washed
for five minutes with recovered or filtered water. The NC
then receives two neutral boils, using filtered water, at
96°C for 5-8 hours. The NC is washed with recovered or
filtered water for 30 minutes after the first boil and five
minutes after the second boil. The water from the two
neutral boils and washes is discharged to the boiling tub
pits.
d. The NC is then slurried to the beater house and
pumped through the primary and secondary Jordan beaters.
Prior to entering the beaters, approximately one pound of
sodium carbonate is added for each 1000 pounds of NC to
neutralize any residual acids. The NC is then slurried
to the poacher house. In the poachers, the NC is boiled
in a soda ash solution (four pounds sodium carbonate per
1000 pounds of NC) at 96°C for four hours, and in a fresh
70
-------
water (neutral boil) step at 96°C for two hours. The NC
is allowed to settle for one hour and the water is drained.
e. The NC is passed over screens to remove any for-
eign material and finally over a vacuum filter to remove
water and excess soda ash. The NC is then slurried to the
blender house where it is circulated as a water slurry for
five hours and sampled for final product quality. The NC
is conveyed by water slurry to the final wring house where
it is wrung to approximately 30$ moisture by weight. The
water from ail of these operations is discharged to the
poacher pits. Water flow is shown in more detail in Figure
17, paragraph 60.
f. At this writing, a continuous NC line is under
construction at Radford AAP. In this process, the NC is
continuously nitrated at a rate of 150 Ib/min and dumped
into a centrifuge. The centrifuge is operated to remove
acid from the NC, and the acid is recovered. The NC is
then washed by a spray system and centrifuged to remove
the spent acid. The NC, containing 58$ total acids by
weight, is then discharged to a slurry tub where water is
added. The NC/water slurry is agitated and then pumped
to the boiling tub house as a 10$ slurry having an acid
strength of 1 to 1.5 percent. Prom the slurry tub on
through the process, the system is identical to the con-
ventional process.
g. In the modernized process, virtually all water
will be recycled. The drain water from the first acid
boil will be used as make-up water for the flume line; the
water from the first neutral boil will be used as make-up
water in the slurry tub which follows the acid wringer;
and the balance of the water used in the process will be
sent to the poacher pits, clarified by centrifugation and
pumped to the recovered water tank for re-use as shown.
Note that this eliminates the use of the acid neutraliza-
tion plant indicated in Figure 12 with its high dissolved
solids effluent, and leaves little wastewater except the
usual washdown, spills and leaks, etc., and the squeezings
from the dehydration press.
45. Nitroglycerin.
a. Nitroglycerin, "NG", is an explosive plasticizer
largely used as an ingredient in double and multi base
71
-------
propellants and in the casting of plastisol double base
rocket motors (cf. paragraph 46 and 56). Chemically, it
is glyceryl trinitrate:
CHo-ON00
I 2 2
CH-ON02
I
CH2-ON02
and is made by the nitration of glycerin with mixed nitric
and sulfuric acids. Nearly all U. S. military production
is made by the Biazzi Process, and it is largely made where
it is used to avoid having to ship it. The various plants
are all nearly identical, and the following description is
a composite one.
b. Like all continuous processes, the Biazzi Process
is characterized by a very small inventory of material in
the reactor and by a careful balancing of flow rates and
cooling. The entire unit, including the nitrator, two acid
separators and three soda water washers, is contained in
one room (units 1 through 11 in Figure 13). To give a feel
for the small scale, the nitrator is a vessel approximately
two feet in diameter and four feet tall, jammed full of
cooling coils; the acid separator is approximately five
feet in diameter and 18 inches high; and the soda water
washers are approximately 18 inches in diamter by three
feet high.
c. Glycerin and Mixed Acid are metered into and
through the nitrator, with strong cooling, at a rate de-
pendent on the strength of the nitrating acid. The efflu-
ent stream goes through the acid separator where the spent
acids are removed, and then into the first washer to be
contacted with a 16$ soda ash solution for neutralization
of residual acid. Then through two fresh water washes and
another separator. After washing, the NG goes to an emul-
sifier where a 3$ soda ash solution equal in volume to that
of the NG is added. The emulsified NG is conveyed to a
temporary storehouse to await use. At the time of use, the
emulsified liquid goes through a separator, and the spent
water is discharged to waste.
72
-------
TO MAGAZINES
BUGGY
x:
flSCALE
EC
-1
*jm i
••••ia^P^HH^RP
CART
i
A
|
«0*!!»LJiwJ
SEPARATOR
TANK
aVifl
r*J5U
w.t
1S1
>••»••• JN!B*B»B«**I
Q
DIVERTER
BUILDING
ITOR
CATCH TANK
DESENSITIZING
BUILDING
USEM
2 WO CONSTANT HEM TANK
1 MUSURNM NT
4 MTRATOR
S UVEL CONTROLLER
0 SEPARATOR
7 SOOA WATER WASHERS
I fRESH WATER WASHERS
0 RE-EMULSIFIER
10 SOOA WATER TANK
II HOT WATER TANK
INTERRUPTER
FUNNEL
SODA DISSOLVER
ft REFRIGERATION
1 BUILDING
SODA
OlSSOLVERl
SPENT ACID
BUILDING
SODA WATER
TANK STORAGE
BIAZZI PROCESS
BUILDING
MIXED ACID
STORAGE
TANKS
i UNDERGROUND
I ACETONE STORAGE
SPENT ACID
TANK ROOM
: M-Tbuon
iAYHtJLDTNli
TANK ROOM
TRIACETIN STORAGE TANKS
CATCH TANK
ROOM
FIGURE 13 - BIAZZI NITROGLYCERIN PROCESS ( Courtesy U .S. Naval
Ordnance Station, Indian Head )
-------
d. The water balance at maximum production (52,800
Ib/day) for one NG line (at Radford AAP) which has been
extensively studied is shown in Figure 14. The total dis-
charge of 86,500 gpd includes 0.51 gallon per pound of NG
of wash and transfer water; the rest is uncontaminated
cooling water. Planned modernization is projected to re-
duce the discharge to 27,300 gpd with 14,400 gpd recycled.
e. There are also a few of the older, batch NG lines
still in operation. They use the same chemistry and proc-
esses as the Biazzi, are being phased out, and will not be
discussed in any additional detail.
46. Nitrocellulose-Base Propellents.
a. Nearly all the nitrocellulose and nitroglycerin
made in the various Ammunition Plants go into the manufac-
ture of Single Base, Double Base, or Multi Base propellants
for shells and rocket motors. These propellants are made
by colloiding and molding processes not unlike those used
in the plastics industry; in fact, nitrocellulose was the
first of the modern plastics. Single Base propellants are
compositions consisting mostly of nitrocellulose with minor
amounts of plasticizers, stabilizers, burning rate cata-
lysts, etc. Double Base implies nitrocellulose plus a
liquid nitrate ester, usually nitroglycerin, with stabili-
zers, catalysts, etc.; and Multi Base implies a combination
of several nitrate materials such as nitrocellulose, nitro-
glycerin, nitroguanidine, triethyleneglycol dinitrate, with
stabilizers and the like. Although the formulations are
legion, the materials are all quite similar in manufacture
and properties; and they will be discussed as classes.
b. A major portion of U. S. military nitrocellulose
goes into Smokeless Powder for shells and into Casting Pow-
der for missiles. Although called "powder", these materials
are physically cylindrical pellets ranging in diameter from
a fraction of a millimeter to a centimeter or more. The
larger ones may contain multiple perforations to increase
their surface area.
c. Solvent Extrusion is a major manufacturing process
for smokeless and casting powders. In this process, water-
or alcohol-wet nitrocellulose fibers from the NC Plant are
masticated with solvents such as alcohol-ether and/or
74
-------
Oi
45,000 gpd
7200 gpd Cooling Water
1300 gpd Cleanllp
Drinking Water
75,700 gpd (Summer)
66,500 gpd (Winter)
I
-TT
7200 gpd
Cooling
Water
L
Air Aspirator
3000 gpd
(D)
10,800 gpd Line Heating Water (Winter Only)
8000 gpd
1
4000 gpd In
16 £ Soda Ash
Solution
Combi- NG/Acid
nation -••Nitrator-•» ««f -UfSeparator]^ Water nf
House I [[Separator" "'-L n T .. . T
Soda
Wash
T
20,000 gpd
(Summer Only)
Refrigeration
House
I
20,000 gpd
10,000 gpd
I
\^
4000 gpd
J_
Fresh
Water
Wash
I
1300 gpd
Clean Up
I 7200 gpd 4000 gpd
1 I
(D) (D) (D)
• 15,000 gpd
4000 gpd In
3# Soda Ash
Solution
J
Separator ' »Emulsifier
4000 gpd 14., 800 gpd
(D)
Spent Acid
Storage
Air Compressor
House
15,000 gpd
Store Houses
(3)
14,800 gpd
I
17,200 gpd
(D)
( D) - Discharge to Sewer ( Total - 86,500 gpd)
FIGURE 14 - WATER BALANCE FOR ILLUSTRATIVE, 52,800 LB/DAY, BIAZZI NG LINE AT RADFORD AAP
-------
acetone in a heavy duty mixer until the mass forms a homo-
geneous plastic dough. The stabilizers, catalysts, etc.,
are blended in at this point, as are nitroglycerin, nitro-
guanidine, etc., if the formulation is to be Double or
Multi Base. The dough is pressed into preliminary blocks
for feeding to a finishing press which extrudes long strands
of approximately the desired diameter through single or
multiple dies. Smaller strands are usually solid, while
larger ones frequently have longitudinal channels in them.
The soft, "green" strands are cut to convenient lengths (or
to the final pellet length) and placed in drying ovens to
drive off the volatile solvents. Considerable shrinkage
takes place during drying, and the strands come to their
final diameters. The dried strands, which may be brittle'
or rubbery depending upon the formulation, are then cut to
final pellet length. In some processes, the pellets are
given a final hot water soak to drive off traces of sol-
vents and then final-dried.
d. In the Solventless Process, a pre-mix is typically
made from chopped or powdered nitrocellulose plus liquid
nitroglycerin and the minor ingredients, dispersed in a
water slurry as a carrier. The pre-mix paste is centri-
fuged to remove most of the water, bagged and air-dried.
The air-dried paste is tumble-blended with any other in-
gredients and then passed through remotely-controlled
differential-speed-roll calenders which colloid it into a
homogeneous sheet. Several sheets are then passed through
a series of evenspeed rolls which compact them into a
single sheet of specified thickness. A slitter machine
cuts the evenspeed sheets into strips. These are rolled
into "carpet rolls" which are fed into the finishing press
just as in the Solvent Process. This time, though, there
is little or no shrinkage of the strand because it contains
little or no volatile material.
e. Rocket motor grains up to two fe'et or more in
diameter and up to six feet long (although most of them
are much smaller) are made by the above extrusion process,
in mammoth, heated extrusion presses; using either solvent
or solventless pre-mixing of the ingredients.
f. Certain double or multi base formulations have
additional ingredients that make them High Energy, for ex-
ample metal powder fuels such as aluminum and ammonium
76
-------
perchlorate or HMX oxidizers. These are usually made in
the form of Casting Powder or Base Grain, rather than as
large extrusions. Casting powder particles are very fine
pellets, a fraction of a millimeter in diameter, and are
intended for further combination with energetic plasticizers
in a subsequent rocket motor casting operation.
47. Ball Powder.
a. Ball powder is a special case of the solvent NC
propellant process in which the propellant is formed into
spheres not more than 0.030 inches in diameter. The oper-
ation is not large by munitions standards - recently around
900,000 pounds per month, for about seven million gallons
of contaminated wastewater - but is of interest because the
manufacturing process bears little resemblance to conven-
tional propellant manufacturing as described above.
b. Figure 15 shows a flow diagram of the ball powder
line at Badger AAP. The feed at Badger is surplus single
base propellant. It is ground in water slurry, extracted
with benzene to remove non-NC ingredients and then dissolved
in ethyl acetate which forms a separate liquid phase on top
of the transfer water. Agitation is adjusted so as to form
an emulsion of the NC solution dispersed in the water, and
a protective colloid such as collagen is added to prevent
the globules from coalescing in the subsequent solvent re-
moval step. Next, sodium sulfate is added to the water
phase to set up an osmotic pressure gradient and draw out
most of the water from the globules. The temperature is
then raised to 69°C (the boiling point of ethyl acetate)
to remove the NC solvent. Agitation is maintained, the
protective colloid prevents globule coalescence, and the
globules slowly harden to firm spheres suspended in the
water phase.
c. The spheres are next coated with dinitrotoluene to
retard their initial burning rate upon ignition, by adding
a solution of DNT in a non-solvent for NC or as an emulsion
and agitating until the DNT is completely absorbed into the
surface layer of the spheres.
d. The spheres, still in water slurry, are next passed
between rolls to flatten them to ellipsoids of controlled
minor dimension, and at last filtered off and dried. They
77
-------
00
!L _j X 1 L
FIGURE 15 - BALL POWDER PROCESS AT BADGER AAP
-------
are coated with graphite to minimize static pickup by tum-
bling in a "sweetie barrel" until glazed, then screened and
packaged.
e. The cooling water used in the ball powder line is
100$ recycled, but the process water is currently mostly
discarded. It contains benzene, ethyl acetate, nitrocellu-
lose, nitroglycerin, sodium sulfate, and collagen. At this
writing, the line is under engineering study looking towards
maximum re-use of water and optimum outfall treatments.
48. Black Powder.
a. Black powder is the oldest explosive in the inven-
tory, the composition and manufacturing process having
remained essentially unchanged for more than 100 years. It
is still a critical component in the ignition charges and
primers used in 90$ of U. S. artillery ammunition, so it is
by no means obsolete.
b. Manufacturing consists of mechanically compacting
potassium nitrate, sulfur, and charcoal in a wheel mill.
The pulverized materials are mulled together water-wet,
and mulling is continued until all the water is evaporated.
The mill cake or "clinker" is broken up and re-compacted at
about 6000 psi in a hydraulic press. The hydraulic press
cake is "corned" by passing it between rollers with corru-
gated surfaces, and the resulting pellets are tumbled with
a small amount of graphite until they are dry, hard, and
glazed.
c. No water is used in the process except for a small
amount in the wheel mill, and that is all evaporated in the
glazing step.
d. At this writing, all black powder is purchased from
a commercial supplier; but a process is under design for
installation in one of the AAPs, probably Indiana.
49. Pyrotechnics and Primer Mixes.
This category of explosives consists of a host of
initiator materials such as lead azide, lead styphnate, tet-
racene, trinitroresorcinol, and others; made in small quan-
tities, often on a mere bench scale. They are beyond the
scope of this study.
79
-------
SECTION II - MANUFACTURE OP NITRATING ACIDS
50. Introduction.
All of the basic explosives manufactured in the U. S.
munitions complex are nitrated products, and this makes
acid manufacture the largest operation in the system. At
recent production rates, the tonnage of sulfuric acid manu-
factured was four times the tonnage of TNT and four percent
of the U. S. total for all purposes. All explosives
manufacturing plants make their own acid, and some of them
have as many as 50 separate units .
51. Nit ric/Sulf uric/Oleum .
a. Nitric Acid is invariably made by the DuPont
Ammonia Oxidation Process (AOP) . In this process, liquid
ammonia is vaporized, mixed with air and passed under pres-
sure over a Pt-Hg-Pd catalyst for conversion to nitric
oxide. The nitric oxide is further oxidized to N02j and
the N02 is absorbed in water in an internally cooled bubble
cap column to form weak (approximately 60-65$)
(1) The weak nitric acid is concentrated by one
or the other of two processes. In the usual Nitric Acid
Concentrator (NAG), the weak HNOo is mixed with concentra-
ted sulfuric acid and injected into the bottom of a distil-
lation tower along with live steam. The sulfuric acid,
being a powerful dehydrating agent, combines with the free
water; and 98-99$ HNO^ distills overhead and is condensed.
The still bottoms, consisting of approximately 68$ I^SO^,
are recovered and sent to one of the Sulfuric Acid Concen-
tration (SAC) units discussed below. In an alternate
nitric acid concentration process, the weak HNO^ from the
AOP unit is mixed with hot 72$ aqueous magnesium nitrate,
another powerful dehydrating agent. The hot mixture is
passed through a stripping column, 99$ HNO^ distills out
the top, and 60$ aqueous magnesium nitrate drains out the
bottom. The still bottoms are reconcent rated by evapora-
tion and re-used.
b. Sulfuric acid is made by one or another version
of the classic Contact Process wherein S0j> is oxidized to
80
-------
S03 and absorbed in 100# H2S(ty to yield 40^ oleum which
can be diluted with water to yield additional 100# H2SOjj.
or used as oleum in the nitration of feeds such as toluene.
In the conventional version of this process, sulfur is
burned in air to produce the 302* an(^ *ne ^°2 ^s oxidized
with more air over a vanadium pentoxide catalyst to pro-
duce the 803. In the closely related SAR (sulfuric acid
recovery) process, the S02 is derived from the decomposi-
tion of spent H2SO^ in a cracking furnace. Either way,
the 303 is fed to the bottom of a stripper column and
either water or 100# H2SO^ is fed into the top. Either
100# H2SO^ or oleum, respectively, comes out the bottom;
and unreacted S02 goes out the top to recycle or to the
sellite plant if there is one.
c. Acid recovery takes several forms. In the older
SAC (sulfuric acid concentration) process, spent HNC^/^SOjj.
from the nitrators is distilled. HNOo passes overhead, is
condensed and. is re-used. The still bottoms, consisting
of concentrated H2SO^, are not re-used because of the slow
buildup of impurities that would result; they are sold for
commercial,uses for which they are satisfactory. In the
newer SAR (sulfuric acid regeneration) process, the spent
HrjSO^ is cracked in a furnace to yield a mixture of sulfur
oxides and water. The sulfur oxides are oxidized to SO^
as above and used to make new H2SOij. or oleum which is sat-
isfactory for re-use.
(1) The eventual complete conversion from the
old SAC process to the new SAR process, which is part of
the modernization program, will have a major impact on the
levels of acid production. HN03 is consumed in the manu-
facture of explosives, because it becomes part of the ex-
plosive molecule; but H2SOjj. is only a reaction medium and
catalyst. Consequently, as H2SOj, recovery and re-use
rises, the necessary new production will fall; and the high
production figures quoted above will become quite out of
date.
d. Wastewater streams from the above operations are
in principle practically nil, but in practice they are con-
siderable due to the large scale of the operations. Nothing
is 10Q# efficient, =a^fe there are always losses which appear
in the wastewater. Each process has a tail gas stream which
contains volatilized acids and oxides of nitrogen and sulfur,
81
-------
These gases are passed through water scrubbers to abate air
pollution, and their contaminants reappear as dissolved
acids in the wastewater .stream. In addition, there are al-
ways minor leaks, spills and washdowns; and these too appear
in the wastewater. There are also sometimes sludges from
still bottoms or from equipment corrosion, and there are
salts from the lime or soda ash neutralization of spills and
residues.
52. Acetic Acid/Acetic-Anhydride/Ammonium-Nitrate.
The nitration of hexamine to make RDX/HMX, unlike most
other nitrations, is carried out in acetic acid - acetic
anhydride solution. New acetic acid is purchased; but
acetic anhydride is manufactured on-site, and spent acetic
acid is recovered for re-use.
a. Acetic acid is recovered from spent nitrating acid
mix. The spent acid, containing some nitric acid (which is
neutralized with sodium hydroxide) and some RDX/HMX, is
sent to a primary evaporator where about 80$ of the acetic
acid is vaporized, condensed, and recovered as 60$ acetic
acid. The remaining 20$ is withdrawn as primary sludge,
diluted, heated, and cooled. During cooling^ RDX seed
crystals are added to promote crystallization of the RDX
content; and the crystals are removed by cyclone separa-
tors. The resulting liquid, called filtrate feed, is sent
to a secondary evaporator and/or fed back to the primary
evaporator. The recovered 60$ acetic acid is concentrated
to glacial (99$) acetic acid by azeotropic distillation
with n-propyl acetate and sent to storage.
b. Acetic anhydride is made from glacial acetic acid
by thermal cracking of acetic acid to ketene gas and reac-
tion of the ketene with more acetic acid to form acetic
anhydride. Glacial acetic acid is passed through a crack-
ing furnace, and the wail&P. of reaction is removed by con-
densation. Any uncracked acid is concentrated by azeotropic
distillation and recycled. The ketene stream is scrubbed
with glacial acetic acid, and crude acetic anhydride is
withdrawn from the bottom of the scrubber. It is refined
by simple distillation and sent to storage.
^c. Ammonium nitrate is made in an adjunct to the AOP
nitric acid plant. Anhydrous ammonia and concentrated
82
-------
nitric acid are injected into a circulating stream of the
product, with strong cooling, and react to form more product,
The resulting solution of ammonium nitrate in nitric acid is
transported by tank car to the RDX manufacturing area where
it is used as the nitrating agent.
d. Wastewater streams from these operations are sev-
eral. In addition to the usual spill and leak washdown
water, there are weak acid streams from tail gas scrubbers,
organics in solution from the decantation of the water-n-
propyl-acetate azeotrope and from the ketene scrubber, and
sludge residues from acetic acid recovery and from equipment
corrosion.
83
-------
SECTION III - LOADING OPERATIONS
53. Introduction.
The final operation in the manufacture of munitions is
the loading of the explosive or propellant into field hard-
ware.
54. Melt-Pour Loading of High Explosive Warheads.
a. Bombs, shells, and similar ordnance items are typ-
ically filled with TNT or a mixture of TNT and other high
explosives, poured in.molten from overhead kettles and al-
lowed to solidify. Water use is minimal compared to that
in explosive manufacturing, consisting mainly of equipment
washout and floor washdown. The following description is
a composite of several different loading operations, fabri-
cated to illustrate all major unit operations. A specific
loading plant somewhere may be a considerably simpler
operation.
b. Flake TNT arrives at the loading plant in box cars
filled with one-cubic-foot cardboard boxes which were filled
at the TNT manufacturing plant. The boxes are unloaded
manually and sent up a conveyor belt to a dump station on
the second floor where they are opened and dumped through
a screen into a hopper. In some cases, the hopper dis-
charges directly down into the melt kettle.
c. The melt kettle is typically a several-thousand-
gallon stainless steel chemical kettle fitted with a slow-
speed, anchor-type agitator and a steam jacket; and is
located on the floor above the pouring floor. The TNT,
Composition B or other formulation is charged through a
hatch in the top of the kettle and melted with gentle agi-
tation. Agitation is particularly''important in the case
of mixtures such as Composition B, where the RDX is in the
form of a suspended, non-molten powder, and in the case of
composites such as Tritonal, which has aluminum powder
suspended in the molten TNT. The charge is heated to just
above the melting point of TNT, typically about 78 to 82°C,
at which temperature the mixture forms a mobile slurry.
84
-------
d. Empty, prepared bomb or shell cases are positioned
underneath the melt kettle; and the molten explosive is run
into them by gravity until the case is filled to the mark.
The "mark" may be near the top, or it may be part way up a
temporary funnel attached to the top of the munition. TNT
shrinks about 13$ upon solidification, and the over-filled
funnel feeds additional TNT down into the forming shrinkage
cavity to minimize void formation. The filled munitions are
sent to a holding area to cool and solidify.
e. After solidification, the funnels, if any, are re-
moved; and the fuze cavity is trimmed to specification by
hand. Fuzes and other hardware are attached, and the fin-
ished munition is packaged in boxes or on pallets and
shipped out. Clean riser scrap, TNT trimmings and the like
(but not floor scrap) are sent back to the melt kettle for
re-use.
f. Once a day or once a shift, the melt kettles are
allowed to drain as empty as they will; and are then washed
out with water and dried. The casting areas, the wagons,
the floors, the trimming areas, etc., are also washed down
with water; as is the dumping area, the screens and the
hopper. In some plants, a continuous spray of water is
kept going in the boxcar unloading area to keep down TNT
dust; and in some plants the working floors are kept wet
with running hoses for dust and spark safety. Riser fun-
nels, trays, reject munitions, and the like are steamed
out; and the condensate is combined with the wash waters.
The combined wash waters are collected in gutters and led
to catch basins outside the building.
g. The water in the catch basins is known as "Pink
Water". It is a saturated solution of TNT (which is about
100 ppm at room temperature); and it turns pink in the sun-
light due to photolysis of dissolved TNT to form complex
dye-like molecules. (The Navy calls Pink Water "Red Water",
but this report follows Army usage in reserving the term
"Red Water" for the more complex - and redder - wastewater
from the sellite purification of crude TNT.) A sludge of
solid TNT and other explosive ingredients collects in the
bottoms of the catch basins and is periodically removed and
burned.
85
-------
h. Some melt-pour loading plants have exhaust fans
over the flake dumping stations, the kettle hatches, etc.,
for dust removal. The dust may be caught in wet scrubbers
or simply vented to the atmosphere. If it is caught in
scrubbers, the scrubber drainings are added to the catch
basins.
i. There is no process water as such in a melt-pour
operation, and the only wastewater stream is the Pink Water
from clean-up operations. It is small compared to the
volumes of wastewaters from explosive manufacturing, and
its treatment is a problem only in larger operations in
the Northern and Eastern parts of the country. In the dry
Western and Southern parts of the country, treatment con-
sists simply of evaporation followed by burning of the dry
residue. Treatment processes are discussed in detail in
Chapter VI.
55. Extruded, Nitrocellulose-Base, Rocket Motor Grains.
Double base (NC/NG) propellants comprise an important
class of solid rocket motor grains, made by hot extrusion
as described in paragraph 46 above. The dies have channel
inserts of various shapes to give hollow grains with round,
cruciform or star-shaped perforations as desired. The
extrusions are cut to length, and are dried and cured until
their dimensions have stabilized. They are then trimmed to
exact dimensions on a lathe. Their outer surfaces and ends
are painted with a flame-retardant called a "restrictor" or
"inhibitor" to confine burning to the inside surface. The
finished grains are then boxed and shipped to the motor
assembly plant where they are inserted into motor cases
like batteries into a flashlight. Large ones may be ce-
mented in place, and small ones may be simply slid in.
56. Cast-in-Place Rocket Motor Grains.
There are two general types of cast rocket motor grains:
plastisol, and polymerization-cured. They differ little in
their handling and still less in their wastewater aspects,
and will be discussed together.
a. Plastisol cast grains are typified by the high-
energy compositions used in advanced, solid rocket motor
ICBM's and IRBM's. In this casting operation, a high-energy
86
-------
"base1grain" or "casting powder" such as described in para-
graph 46 is poured into an upside down rocket motor case
and vibrated to maximum bulk density. The air is exhausted
by means of a vacuum bell, and liquid nitroglycerin or
nitroglycerin mixture is run in so as to fill all the voids
with liquid and give a dense slurry with no included bubbles,
An internal star-shaped cavity is provided for by a remov-
able lengthwise mandrel inserted during case preparation.
(1) The loaded motor is then placed in an oven
at about 40°C for days or even weeks, during which time the
liquid semi-dissolves the base grain pellets to yield a
uniform, rubbery, solid gel. After cooling, the mandrel is
pulled, any ragged propellant edges are trimmed up, the
grain end is painted with restrictor, and the nozzle and
ignitor hardware are attached.
(2) No water is used in the process, except minor
amounts for clean-up and sometimes floor washdown, depending
upon the particular plant. There may be steam coils in the
curing ovens.
b. Polymerization-cured grains are cast in similar
vacuum bells in similar-sized motors, and cured in similar
curing ovens; but the mix preparation at the casting site
is more complex.
(1) A pre-mix consisting of all the ingredients
except the oxidizer is usually made in manually-operated
equipment. This will typically be a thin slurry of aluminum
powder in a liquid pre-polymer which will eventually become
the polymerized binder, plus any minor ingredients such as
stabilizers, burn rate catalysts, etc. The pre-mix is
charged to the bowl of a several-thousand-gallon vertical
mixer and shipped to the mixing building that way.
(2) The oxidizer, typically ammonium perchlorate,
is fine^ground just prior to use in one or more hammer mills
and collected in tote bins for shipment to the mixing build-
ing. The tote bins are positioned above the mixer, ready
for remote addition upon command.
• V"
(3) The loaded bowl is attached to the mixer,
the operators leave the area, and stirring is commenced un-
der remote control. Oxidizer is added slowly, with mixing,
87
-------
and with cooling -to remove the heat of mixing, to form the
final complete mix; and mixing is continued until the mass
is uniform and smooth. Addition of the polymer cross-
linking agent is made some time during this step, as early
as possible for good mixing and as late as possible for
longest pot life. The mixing may be done under vacuum for
de-aeration, but usually is not for safety reasons.
After mixing, the mixing bowl is lowered and
transported to the casting site which may be another build-
ing some distance away. There it is positioned over the
prepared motor case contained in a vacuum bell, and the
propellant - a thick paste - allowed to flow down into the
case. When the case is filled, the motor is either cured
in situ by heating the bell or else removed to a curing
oven as in the plastisol procedure. The empty bowl is taken
away for cleaning.
(5) Clean-up procedures vary widely from plant
to plant, and constitute the only use of water except for
heating and cooling coils . Most plants wipe down the
mixers and containers with solvent -soaked rags only and
use no water there, but some hose down the floor and some
hose down the oxidizer grinders and chutes. Still other
plants rigidly exclude any water at all (except for emer-
gency fire deluge systems) and hermetically seal their
oxidizer preparation systems with pressurized, dry air or
nitrogen. However, to keep things in perspective, even
those plants which use water "liberally" use much less
than, say, a melt-load plant, and vastly less than an ex-
plosive or propellant manufacturing plant.
57. Pressed Explosive and Pyrotechnic Charges.
a. Pressing or compression molding is used to load
a few specialty items such as illuminating flares, black
powder delay trains, and a few developmental explosives;
but the quantities used and the numbers of items produced
are miniscule compared to the major items discussed above.
Almost no water is used except for some floor washdown,
and only brief mention of the process will be included
here.
b. Typically, a semi-dry, granular ingredient mix
including the binder if one is used is made in a muller or
88
-------
similar mixer and sent to the loading station in gallon-
sized buckets. There, an empty illuminating shell (for
example) is positioned under a raised plunger behind a
safety barricade. An operator pours in an inch-or-two
layer of mix from a hand scoop and presses it home with
the hydraulically-actuated plunger. Additional lifts are
added in the same way until the shell is filled. Any ex-
cess is trimmed, and the shell is sent on to have other
hardware attached.
c. A few, developmental, shaped, high explosive
charges are also now being made by compression molding;
but the operations are very limited in scope, use almost
no water, and have no appreciable impact on the industry
wastewater picture.
-------
SECTION IV - APPROACHES TO WATER POLLUTION ABATEMENT
58. Introduction.
The major discussions of water pollution abatement in
this report are presented in the Chapter VI, but some of
the approaches are so integrally tied in with manufacturing
technology that they deserve introductory mention here.
59 • Manufacturing Improvements.
a. Some explosives manufacturing activities involve
operations which are inherently generators of wastewater
pollutants, such as the sellite purification of crude TNT;
and major improvements can be made if the process itself
can be improved.
b. The GIL continuous process, described earlier, is
one such improvement. It produces approximately 20$ less
Red Water than the older batch TNT process by producing
smaller quantities of undesired isomers, by using some
preliminary water washes which can be fed back to the ni-
trators, and by using multi-stage, countercurrent sellite
treatment which is more efficient.
c. Another and more extensive process improvement
currently under study at Stanford Research Institute prorn^-
ises to make approximately an 80$ reduction in Red Water
if. all goes well. Its key is operation at low temperature •
around -10°C - under which conditions a much lower percen-
tage of undesired isomers are formed. The process is
currently somewhere between bench and pilot scale, and it
will not have an immediate impact.
d. A program which will have an earlier impact is
research and development towards reclamation and re-use of
the constituents of Red Water. Figure 16 illustrates the
concept schematically. In this concept, the ash from Red
Water incineration is reduced chemically to yield sodium
carbonate and hydrogen sulfide which are recycled to make
new sellite solution. Fumes from the evaporation and in-
cineration steps are caught in activated carbon which is
subsequently incinerated or regenerated, rather than caught
90
-------
Soda Ash (Na2CO3)
Water
1
Mixing
Soda Ash
Solution
Either or
Absorption
Reaction
SO,
Sulfur
Set lite
(Na2S03)
Solution
TNT
Purifi-
cation
Red
Water
Neutral-
ization
Combustion
Reaction
I
H2S
Activated
Carbon
Adsorption
T
Water to Stream
or Recycle
Spent Activated Carbon
\
Incineration
Fume Treatment
Absorption or
Catalytic Methods
F1 uid Bed
Reduction
Red Water Incineration
with Asrt Recycle
FIGURE 16 - RED WATER RECLAMATION CONCEPT
-------
in a water scrubber which would itself be a water pollution
generator. At this writing, the ash reduction step is in
the laboratory stage and the carbon regeneration step is in
the pilot stage. It is not yet clear how successful they
are going to be.
e. Longer range again, exploratory laboratory research
is underway to reclaim the organic components of Red Water
by desulfonation and conversion to materials of commercial
interest. The promise of this program has not yet been
established.
f. Similar, although generally less dramatic, process
improvements are under study for RDX/HMX, nitrocellulose,
ball powder, acetic acid-manufacture and sulfuric acid
regeneration.
60. Water Management.
a. In contrast to the inherent waste streams consid-
ered in the above section, there are numerous wastewater
streams which could be greatly reduced or eliminated by
relatively simple measures such as cascading, recycle, or
re-use; and the Army is currently studying this approach
in great detail. Among the processes being analyzed for
possibilities are: Nitrocellulose, nitroglycerin, TNT, NC
propellants, ball powder, RDX/HMX, tetryl, acetic acid,
nitric acid, black powder, and sulfuric acid recovery. Ni-
trocellulose will serve as an illustrative example.
b. Figure 17 shows the general product flow and the
water balance for a typical batch nitrocellulose line. This
particular line operating at capacity uses 2.24 MGD of new
water and 0.20 MGD of recovered water for three basic pur-
poses: (1) conveying NC as a 3-10 percent slurry from one
operation to another; (2) boiling and neutralizing the NC
for stabilization; and (3) washing the NC.
c. A modified process under review at this writing
is shown in Figure 18. In this system, the flume line
water carrying NC from the nitrator is recovered by filtra-
tion and largely re-used; and the slurry water itself is
used for. the first, acid boil. Water from the first neutral
boil is saved and used as make-up water for the flume line.
The balance of the water used is sent to the poacher pits,
92
-------
CO
Rite red Water
2,744,600
500,000
Discharged
to Sewer
Nitraror
M
r-
30
3,000
1—1,441,400
From Poacher P
Recovered Water
1,441,400
784,300
n
J
I
Boiling Tubs
1,604,300
606,300
109.800
Beater
716,100
315,300^
455,000
Poacher
770,300
427,300
I
226,000
Blender
653,300
229,500
I
49,500
998,000
400.800
343,000
I
423,800
Wringer
279,000
273,800
Boiling
Tub Pits
T
Total Discharged: 2,744,600
Total Reused: 200,000
i
Poacher
Pits
1,441,400
2,239,400 To Neutralization
Facility and Discharge to Sewer
5,200
Dehy
Press
I
4,850 to Sewer
FIGURE 17 - CURRENT WATER BALANCE, BATCH NC LINE
-------
to
Filtered Water
623,609
500,000
to Sewer i
e
o
Nitrator
1,120,000
I
620,000
i-i:
•1,441,400 From Centrifuges
30,989
Recovered Wafer
1,441,400
39,809
fered
,677
8,688
|
1 ,
I r
ling
Acid
ails
,420
,541
»
ng and
SI urry Tub
120,420
80,611
i
120,420
1 ,879 ,
t 1
•M^MH
Boiling Tub
Neutral
Boils
695,599
89,299
*
Storage
109,800
Beater
716,100
o-
8
^
CO
400,800
315,300
424,011
30,989
182,880
1,120
Poacher 427,300 I Blender 1229,500
»t •
770,300
653,300
343,000
1
49,500
Wringer
279,000
423,800
t
273,800
Total Discharged: 623,609
Total Reused: 1,441,400
Poacher
Pits
Storage
738,409
118,409
To Acid Recovery
89,299
entrifuge
Bank
5,200
Dehy
Press
4,850 to Sewer
FIGURE 18 - PROPOSED WATER BALANCE, NC BATCH LINE
-------
clarified by centrifuges, and pumped to the recovered water
tank for re-use. The cooling water from the nitrator is
uncontaminated except for the added heat, and could be
cooled and re-used, or re-used elsewhere in the plant if
desired.
d. The net result of these improvements in this op-
eration would be to reduce the new water requirement from
2.24 MGD. to 0.12 MGD. The remaining discharge would be to
an acid recovery plant where approximately 170,000 pounds
of acid would be recovered daily, eliminating the need for
neutralization plants and further nitrate and sulfate treat-
ment. The total water discharged would be 623,609 gpd
instead of the current. 2,744,600 gpd; and it would be
cleaner water due to the recovery instead of the neutral-
ization of the waste acid.
e. Similarly impressive results are projected for
other major operations, including TNT, RDX/HMX, and acetic
anhydride manufacture.
f. It might be wondered why the plants were originally
finalized with so much opportunity left for improvement,
but it must be realized that they were designed and built
over 30 years ago in a very different era when water was
cheap and plentiful and pollution had not entered the na-
tional thinking. Moreover, these modernization improvements
do not come without cost. Capital funds are needed for the
required new construction and require the approval of the
U. S. Congress. In addition, the improved plants are more
complex and will most more to operate and maintain. A
balancing of national values and priorities is still needed
before making changes.
61. Outfall Treatment.
When process improvements and water management have
done all they can, there still usually remains an irreduc-
ible flow of wastewater carrying by-products and impurities,
and spills and washdowns. These residual flows have been
and are the subject of intensive research and development in
DOD and contractor laboratories, and numerous final treat-
ments are under development for ultimate clean-up. These
outfall treatments are discussed and evaluated in detail in
III, Chapter VI, and conclusions and recommendations
presented in Chapter I.
95
-------
MIOM NATIONAL TECHNICAL INFORMATION SERVICE
Fire Safety Aboard LNG Vessels
ADA-030 619/PAT 295 p. PC$9.25/MF$3.00
Solar Energy Heat Pump Systems for Heating and
Cooling Buildings, Proceedings of a Workshop
Conducted by the Pennsylvania State University
College of Engineering
June 12-14, 1975
COO-2560-1/PAT 240 p. PC$8.00/MF$3.00
Data Encryption Standard. Category: ADP Operations.
Subcategory: Computer Security
FIPSPUB 46/ PAT 20 p. PC$3.50/MF$3.00
Criteria for Evaluation of the Health Aspects of Using
Flavoring Substances as Food Ingredients
PB-257 1847 PAT 109 p. PC$5.50/MF$3.00
Fundamental Aspects of Nuclear Reactor Fuel
Elements
TID-26711-P1/PAT 625 p. PC$16.257 MF$3.00
Patrol Force Allocation for Law Enforcement: An
Introductory Planning Guide
N76-24084/PAT 51 p. PC$4.50/MF$3.00
FAA Statistical Handbook of Aviation. Calendar
Year 1975
ADA-033 2107 PAT 158 p. PC$6.75/MF$3.00
Neutron Cross Sections. Volume II
BNL-325ED3V2/PAT 519 p. PC$12.757MF$3.00
A Study of Hospital Patient Injury Prevention Programs
PB-260 7337 PAT 200 p. PC$7.50/MF$3.00
NIOSH Analytical Methods for Set N
PB-258 4337 PAT 56 p. PC$4.50/MF$3.00
Ocean Carrier Service Image and Marketing Practices
PB-261 1817 PAT 150 p. PC$6.007 MF$3.00
Manual of Methods for Chemical Analysis of
Water and Wastes
PB-259 9737 PAT 317 p. PC$9.75/MF$3.00
HOW TO ORDER
When you indicate the method of pay-
ment, please note if a purchase order is not
accompanied by payment, you will be billed
an additional $5.00 ship and bill charge. And
please include the card expiration date when
tring American Express.
Normal delivery time takes three to five
weeks. It is vital that you order by number
or your order will be manually filled, insur-
ing a delay. You can opt for airmail delivery
for $2.00 North American continent; $3.00
outside North American continent charge per
item. Just check the Airmail Service box. If
you're really pressed for time, call the NT1S
Rush Handling Service (703)557-4700. For a
$10.00 charge per item, your order will be
airmailed within 48 hours. Or, you can pick
up your order in the Washington Informa-
tion Center & Bookstore or at our Springfield
Operations Center within 24 hours for a
$6.00 per item charge.
You may also place your order by tele-
phone or if you have an NTIS Deposit Ac-
count or an American Express card order
through TELEX. The order desk number is
(703) 557-4650 and the TELEX number is
89-9405.
Thank you for your interest in NTIS. We
appreciate your order.
METHOD OF PAYMENT
D Charge my NTIS deposit account no.
Q Purchase order no
Check enclosed for $
Bill me. Add $5.00 per order and sign below. (Not avail-
able outside North American continent.)
Charge to my American Express Card account number
NAME
ADDRESS-
CITY. STATE. ZIP.
Card expiration date.
Signature
Airmail Services requested
Clip and mail to:
National Technical Information Servke
1J.S. DEPARTMENT OF COMMERCE
Springfield. Va. 22161
(703) 557-4650 TELEX 89-9405
Item Number
Quantity
Paper Copy
(PC)
All prices subject to change. The pric
above are accurate as of 6/77
Microfiche
(MF)
Unit Price*
es Sub Total
Additional Charge
Enter Grand Total
Total Price' •
._
Foreign Prices on Request.
------- |