EPA/540/2-89/014
SUPERFUND TREATABILITY
CLEARINGHOUSE
Document Reference:
Atlantic Research Corp. "Engineering and Development Support of General Decon
Technology for the U.S. Army's Installation/Restoration Program." Prepared for
USATHMA under contract DAAK11-80-C0027. Four volumes with a total of
approximately 500 pp. April-June 1982.
EPA LIBRARY NUMBER:
Superfund Treatabillty Clearinghouse -EUWW-1
-------
SUPERFUND TREATABILITT CLEARINGHOUSE ABSTRACT
Treatment Process: Thermal Treatment - Incineration
Media: Soil/Lagoon Sediment
Document Reference: Atlantic Research Corp. "Engineering and Develop-
ment Support of General Decon Technology for the
U.S. Army's Installation/Restoration Program."
Prepared for USATHMA under contract DAAK11-80-
C0027. Four volumes with a total of approximately
500 pp. April-June 1982.
Document Type: Contractor/Vendor Treatability Study
Contact: Wayne Sisk
U.S. DOD/USATHAMA
Aberdeen Proving Ground, MD 21010-5401
301-671-2054
Site Name: Louisiana Army Ammunition Plant (NPL - Federal
facility)
Location of Site: Atlantic Research Corp., Alexandria, VA
BACKGROUND; This document reports on the results of bench-scale tests of
treatment technologies for explosive-containing sediment located in lagoons
at Army ammunition plants. A companion literature search identified the
appropriate explosives remediation technologies to be evaluated. Cost
estimates for various treatment technologies were made based on the
laboratory data.
OPERATIONAL INFORMATION: Sediment samples contaminated with the explosives
TNT, RDX, tetryl and nitro cellulose from the Louisiana Army Ammunition
Plant were used in the laboratory tests. Explosive levels in lagoon #4
sediments were at or below 1000 ug/g. Samples from lagoons 9 and 11 had
much higher RDX and TNT levels (1000 to 109,000 ug/gm of soil). The report
contains a detailed QA/QC plan and analytical protocol.
PERFORMANCE; Incineration tests were conducted by placing approximately 4g
of sediment in a crucible and placing the crucibles in a muffle furnace for
varying amounts of time. Residues were analyzed for contaminants of inter-
est. Table I shows the results of the incineration tests. Incineration at
temperatures as low as 300-500°C for 30 minutes time can remove all the
contaminants from the sediments. While all of the explosives can be
reduced to their detection limits at the lower temperatures, it is possible
that some toxic decomposition products may remain. It is, therefore,
important to use temperatures which reduce the total organic contents as
measured by chemical oxygen demand (COD) to acceptable levels. This can be
accomplished at temperatures of 500 -700°C and reaction times of 30 min-
utes. Since explosive volatilization may occur, it will be important in a
pilot scale study to determine whether any vaporized explosives can be
detected in the exhaust gases. Costs for treatment can vary from $100,0007
3/89-28 Document Number: EUVW-1
NOTE: Quality assurance of data may not be appropriate for all uses.
-------
year to $2,000,000/year depending on the water content of the slurry that
is incinerated.
In addition to incineration, acetone extraction, gamma irradiation, wet
air oxidation, and water extraction tests were conducted and results
reported in this document. Of the five procedures tested only incineration
and acetone extraction proved effective in removing contaminants from sedi-
ments. Incineration equipment is available and pilot tests were
recommended.
CONTAMINANTS;
Analytical data is provided in the treatability study report. The
breakdown of the contaminants by treatability group is:
Treatability Group
W06-Nitrated Aromatics and
Aliphatics
VIO-Non-Volatile Metals
Wll-Volatile Metals
W12-0ther Inorganics
CAS Number
121-82-4
118-96-7
479-45-8
7440-47-3
7439-92-1
7440-43-9
COD
Contaminants
Hexahydro-1,3,5-trinitro-
1,3,5-triazine (RDX)
Trinitrotoluene (TNT)
Trinitrophenylmethyl-
nitramine (tetryl)
Chromium
Lead
Cadmium
Chemical Oxygen Demand
3/89-28 Document Number: EUW-1
NOTE: Quality assurance of data may not be appropriate for all uses.
-------
TABLE I
INCINERATION OP LAGOON 9 SEDIMENT - EXPLOSIVES LEVELS
Concentration in Dry Sediment
Temperature
No heat
200
300
500
700
900
Time
(min)
5
30
60
5
30
60
5
30
60
5
30
60
5
30
60
TNT RDX
(ug/g) (ug/g)
424,000 159,000
10,000 <1
1,500 <1
1,350 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
<2 <1
Tetryl
(ug/g)
15,800
114
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
COD
(ug/g)
206,000
124,500
116,500
149,200
55,200
52,300
30,000
5,900
2,190
1,280
8,720
1,310
2,320
12,200
2,410
1,670
Note: This is a partial listing of data. Refer to the document for more
information.
3/89-28 Document Number: EUWV-1
NOTE: Quality assurance of data aay not be appropriate for all uses.
-------
DRXTH-TE- C_^ /•+< AD
) ENGINEERING AND DEVELOPMENT SUPPORT OF GENERAL DECON
H TECHNOLOGY FOR THE U.S. ARMY'S INSTALLATION
RESTORATION PROGRAM
Task 2. Treatment of Explosives Contaminated Lagoon Sediment
Phase IL Laboratory Evaluation
Randall S. Wentsel
Suzette Sommerer
Judith F. Kitchens
ATLANTIC RESEARCH CORPORATION
-Alexandria, Virginia—823M
June 1982 c
Unclassified/Limited Distribution
Prepared for:
U.S. ARMY TOXIC AND HAZARDOUS MATERIALS AGENCY
Aberdeen Proving Grounds-Maryland
ATLANTIC RESEARCH CORPORATION
ALEXANDRIA,VIRGINIA
-------
Disclaimer
The views opinions and/or findings contained in this report are those of
the authors and should not be construed as an official Department of the Army
position, policy or decision unless so designated by other documentation.
The use of trade names in this report does not constitute an official
endorsement or approval of the use of such commercial products. This report
may not be cited for purposes of advertisement.
Disposition
Destroy this report when it is no longer needed. Do not return it to the
originator.
-------
UNCLASSIFIED
SECURITY CLASSIFICATION 0? THIS PAGE r»7i»n D»t» Enttrtd)
REPORT DOCUMENTATION PAGE
READ INSTRUCTIONS
BEFORE COMPLETING FORM
NUMBER
2. GOVT ACCESSION NO,
3 RECIPIENT'S CATALOG NUMBER
4. TITLE <•** subnn») Engineering and Development Support
of General Decon Technology for the U.S. Army's
Installation Restoration Program. Task 2. Laboratory
Evaluation - Phase II
5. TYPE Of REPORT * PERIOC COVERED
Final Report
November - July 1981
e. PERFORMING ORG. REPORT NUMBER
49-5002-02-0002
7. AUTHORC*;
Randall S. Wentsel, Suzette Sommerer, and
Judith F. Kitchens
8. CONTRACT OR GRANT NUMBESri;
DAAK11-80-C-0027
9- PERFORMING ORGANIZATION NAME AND ADDRESS
Atlantic Research Corporation
5390 Cherokee Avenue
Alexandria, Virginia 22314
to. PROGRAM ELEMENT. PROJECT. TASK
AREA ft WORK UNIT NUMBERS
II. CONTROLLING OFFICE NAME AND ADDRESS
DCASR Philadelphia
P.O. Box 7730
Philadelphia, Pennsylvania 19101
12. REPORT DATE
June 1982
13. NUMBER OF PAGES
142
14. MONITORING AGENCY NAME » ADDRESS^// dlllunal Irem Controlling Otllct)
U.S. Army Toxic and Hazardous Materials Agency
Aberdeen Proving Ground,
Maryland 21010
IS. SECURITY CLASS, 'at thu rtpon)
UNCLASSIFIED
IS«. OECLASSIFICATION' DOWHIBRAOING
SCHEDULE
16. DISTRIBUTION STATEMENT (ol (hit Rtport)
Unclassified/Limited Distribution
'?. DISTRIBUTION STATEMENT (ol th* taftruct *ntrrfn D*<* £*>•'•<*)
-------
UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS f AGCrmiwi £•<• £n(.r»d)
toxic byproduct. Gamma irradiation was not effective at the high explosives levels
tested, and water extraction would.be prohibitively expensive.
2 UNCLASSIFIED
SECJRI'Y CLASSIFICATION Of THIS P»GE--»hen Car« Ente--a
-------
SUMMARY
The objective of this study was to continue the evaluation of promising
technologies for the treatment of explosives containing sediment. After a literature
evaluation to identify appropriate methods, several techniques were chosen for
laboratory study. This report summarizes the findings of the laboratory •experiments
and recommends further evaluation for those processes which proved most effective on
a small scale.
Five treatment processes were evaluated to decontaminate lagoon sediment
containing TNT, RDX, and tetryl or containing nitrocellulose, or the five, only
incineration and acetone extraction are recommended fojj^vfur^herconsio^ration.
Incineration represents established technology for disposal of waste explosives and has
been proven to be effective in completely destroying such compounds. Acetone
extraction is a rapid and effective method of removing the explosives {rom the
sediment. While the process has not yet been demonstrated on a commercial scale for
lagoon sediment decontamination, it is expected that conventional solvent extraction
equipment can be used. The major disadvantage of acetone extraction is 4hat the
waste explosives are not destroyed and will require some other method of ultimate
disposal or re-processing.
The remaining processes, wet-air oxidation, gamma irradiation and water
extraction, are not recommended for further study for various reasons. Wet-air
oxidation effectively destroyed RDX, tetryl and nitrocellulose, but it converted TNT to
a more toxic by-product, 1,3,5-trinitrobenzene. Gamma irradiation was not effective
enough to be useful for the highly contaminated sediments tested in this study,
however, it may find application in treating low levels of explosives contamination.
Water extraction at high temperature and pressure was effective at removing TNT,
RDX and tetryl from lagoon sediment, but the process would be expensive to implement
and may have some serious safety problems.
Revised cost estimates based on the laboratory data were made for the various
treatment methods. For comparison pirpoMSj )^&P*MyLjgd^npual operating costs were
based on a scenario in ^rtHfcn eleven standard IsfoonsvVHMiaiCir the report, are
treated per year.
-3
-------
ACKNOWLEDGEMENTS
The laboratory study experimental effort for decontamination of explosives
in lagoon sediment was conducted over an eight month period and required a coopera-
tive effort. The assistance of the following personnel in accomplishing this task is
greatly appreciated:
Carl Gulp - water extraction equipment design and construction
Alice DeSouza - quality control for analysis
Antoine Ennis - preparation of samples for explosives analysis
William E. Harward, HI - incineration, acetone extraction, water
extraction experimental work
William E. Jones, III - analysis of explosives and by-products
Debra Price - wet chemical analysis
Janet Mahannah - editing
Richard Miller, IT Enviroscience - wet-air oxidation
James Pierce, Sandia Laboratories - gamma irradiation
-------
TABLE OF CONTENTS
I. Introduction 11
A. Objective 11
B. Background 11
C. Standard Lagoon Scenario 11
D. Experimental Approach 11
II. Lagoon Sediment Analysis 13
III. Incineration 20
A. Process Description 20
B. Objective 25
C. Experimental Procedure 25
D. Results 27
E. Conclusions 32
F. Future Work 36
G. Economic Analysis 36
IV. Gamma Irradiation 40
A. Process Description 40
B. Experimental Procedure 40
1. Sample Preparation and Irradiation 40
2. Analysis of Irradiated and Control
Slurries 43
C. Results and Discussion 43
D. Conslusions 46
E. Future Work 49
F. Economic Analysis 49
V. Wet-Air Oxidation 51
A. Process Description 51
-------
B. Experimental Procedure 51
1. Slurry Preparation and Wet-Air Oxidation Treatment
Methodology 51
2. Sample Analysis 55
C. Results and Discussion 55
D. Conclusions 64
E. Future Work 64
F. Economic Analysis 64
VI. Acetone Extraction 66
A. Process Description 66
B. Experimental Procedure 68
C. Results 72
D. Conclusions 75
E. Future Work 75
F. Economic Analysis 79
VII. Water Extraction 82
A. Process Description 82
B. Experimental Procedure 82
C. Results 84
D. Conclusions 84
E. Cost Analysis 84
F. Future Work 84
vni. Conclusions and Recommendations 87
IX. References 89
Appendix A. Analytical Methods A-l
Appendix B. Wet Chemistry Methods B-l
Appendix C. Raw Data from Incineration and Acetone Extraction
Experiments C-l
-------
LIST OF TABLES
Number Page
1. Characteristics of the Standard Lagoon 12
2. Core Sample Analysis for LAAP Lagoon 4 15
3. Core Sample Analysis for LAAP Lagoon 9 16
4. Core Sample Analysis for LAAP Lagoon 11 17
5. Core Sample Analysis for LAAP Lagoons - Metals 18
6. Detection Limits for Analytical Methods 19
7. Elemental Analysis and Heating Values of Sediments 28
8. Incineration of Lagoon 9 Sediment - Explosives Levels 29
9. Incineration of Lagoon 11 Sediment - Explosives Levels 30
10. \blatilization of Explosives at 200°C 31
11. Incineration of Lagoon 9 Sediment - Metals Levels 33
12. Incineration of Lagoon 11 Sediment - Metals Levels 34
13. Incineration of Nitrocellulose Contaminated Sediment 35
14. Incineration Costs 37
15. Effect of Sediment Water Content on Fuel Costs 38
16. Incinerator Operating Temperatures and Excess Air Requirements. 39
17. Characteristics of Unslurried LAAP Lagcon 9 and Spiked Nitro-
cellulose Sediments 42
18. Experimental Design for Gamma Irradiation Study 44
19. Effects of Gamma Irradiation on TNT/RDX Sediment Slurries . . 45
20. Effects of Gamma Irradiation on Nitrocellulose Slurries 47
21. Duncan's Multiple Range Test for Nitrocellulose Sediment ... 48
22. Capital Costs for Gamma Irradiation 50
23. Annual Operating Costs for Gamma Irradiation 50
24. Treatment Scheme for Wet-Air Oxidation of Sediment Slurries . 54
25. Wet-Air Oxidation Treatment of LAAP Sediment 56
26. Evaluation of Wet-Air Oxidation Data of Duncan's Multiple
Range Test 59
27. Off-Gas Analysis for Wet-Air Oxidation of TNT/RDX Sediment
Slurries 61
-------
Number
28. Wet-Air Oxidation of Nitrocellulose Sediment 62
29. Off-Gas Analysis for Wet-Air Oxidation of Nitrocellulose Sediment
Slurries 63
30. Capital Costs for Wet-Air Oxidation 65
31. Annual Operating Costs for Wet-Air Oxidation 65
32. Solubility of Explosives in Acetone 67
33. Acetone Extraction of TNT/RDX/Tetryl Contaminated Sediment . 73
34. Serial Acetone Extraction 74
35. Acetone Extraction of Nitrocellulose Contaminated Sediment . . 76
36. Experimental Conditions to Establish TNT/RDX/Tetryl
Equilibrium Curves 78
37. Experimental Conditions to Establish Nitrocellulose Equilbrium
Curves 78
38. Capital Costs for Acetone Extraction 80
39. Acetone Extraction Annual Operating Costs 81
40. Water Extraction 85
C-l. Sediment Weights for Incineration Experiments C-2
C-2. Acetone Extraction of TNT/RDX/Tetryl Sediments - Raw Data . C-3
C-3. Acetone Extraction of Nitrocellulose - Raw Data C-4
-------
LIST OF FIGURES
Number page
1. Sample Points at Louisiana Army Ammunition Plant 14
2. Rotary Hearth Incinerator 21
3. Multiple Hearth Incinerator 22
4. Electric Furnace 23
5. Fluidized Bed Incinerator 24
6. Rotary Kiln 26
7. Gamma Irradiation Treatment System for Dried Sewage Sludge. . 41
8. Schematic of Catalyzed Wet-Air Oxidation Batch Reactor .... 52
9. Picture of Wet-Air Oxidation Reactor 53
10. Picture of Wet-Air Oxidation Control Equipment 53
11. GC-MS of Slurry Water Extract Before Wet-Air Oxidation
Treatment 58
12. GC-MS of Slurry Water Extract After Wet-Air Oxidation at
250°C/980 psig 59
13. Two Steps in the Operating Cycle of an Extraction Battery ... 69
14. Rotocel Solvent Extractor 70
15. Dorr Thickener 7i
16. Water Extraction Apparatus 83
10
-------
I. INTRODUCTION
A. Objective
The objective of this study was to determine on a laboratory scale the
effectiveness of several decontamination processes for explosives containing lagoon
sediments. Five processes were recommended for laboratory study after an initial
literature search to identify possible treatments (Wentsel et_al., 1982 ). These
processes were incineration, wet-air oxidation, gamma irradiation, acetone extraction,
and water extraction.
B. Background
For many years, solid and liquid wastes from Army Ammunition Plants have
been stored in waste lagoons. Since most of these lagoons were not lined, the
sediments on the bottoms of these lagoons have become highly contaminated with
explosives and explosives by-products. The toxicity of these compounds and the
potential for their leaching into the local ground water present some serious
environmental problems. Treatment of these sediments to remove explosives is
required if the environmental hazards associated with them are to be avoided. Two
types of sediments which must be treated have been identified at Army Ammunition
Plants. One type of sediment contains 2,4,6-trinitrotoluene (TNT), 2,4-dinitrotoluene
(DNT), hexahydro-l,3,5-trinitro-l,3,5-triazine (RDX), and tetryl. The other contains
nitrocellulose.
C. Standard Lagoon Scenario
In order to compare capital and operating costs for the various treatment
systems, calculations were made on the basis of treating 11 standard lagoons as
defined in Table 1. This standard lagoon is intended to be a typical explosives waste
lagoon, and its characteristics were defined on the basis of the best data that were
available at the time. Capital costs include the cost of equipment for removing the
sediment from the lagoon and processing it. Operating costs are calculated on an
annual basis.
D. Experimental Approach
The laboratory experiments conducted in this study were not intended to
simulate the operation of commercial-scale equipment. Instead, the contaminated
sediments were subjected to simple, small-scale experiments which indicated whether
or not the process was an effective treatment method. In addition, process
parameters were optimized as much as possible on this scale. All of the processes
recommended for additional study will require further development and demonstra-
tion before they are ready for field use.
11
-------
Table L Characteristics of the Standard Lagoon
Size
30.5 m x 45.7 m x 2.7 m deep (100 ft x 150 ft x 9 ft.)
Depth of Sediment to be Treated
0.3 m (1 ft)
Volume of Sediment to be Treated (wet)
447.5 m3 (118,200 gal)
Weight of Sediment to be Treated (wet)
499,400 - 544,800 kg (1,100,000 - 1,200,000 Ib)
Sediment Characteristics
moisture content: 50-80%
composition: (dry basis)
10% TNT
5% RDX
10% other organics
75% ash
heat content (dry basis): 1028 kcal/kg (1850 Btu/lb)
Treatment rate (11 lagoons/yr, 300 days/yr, 24 hr/day)
763 kg/hr (1680 Ib/hr)
684 1/hr (180 gal/hr)
12
-------
n. LAGOON SEDIMENT ANALYSIS
Sediment core samples were collected from waste lagoons at Louisiana Army
Ammunition Plant (LAAP) to select sediment for the laboratory phase of Task 2. The
sediment sampling sites are presented in Figure L Sediment samples were taken with
a core sampler at two different sites in Lagoons 4, 9 and 1L Each core sample was
divided into three 6-inch (15.2 cm) segments.
The sediment samples were analyzed for TNT, RDX, nitrate, COD, percent
volatiles and metals. The methods used in these analyses are detailed in Appendix A.
The data from lagoon 4 are presented in Table 2. Explosive levels in lagoon
4 sediments were at or below 1000 yg/g. COD levels ranged from 8100-37,900yg/g.
The highest explosive levels were found in lagoon 9 (Table 3). In the top
six inches of site 1, TNT and RDX levels were 109,000 and 91,900 yg/g,
respectively. However, explosive levels at site 2 were less than 1000 yg/g. Lagoon 11
was moderately contaminated with explosives (Table 4). In the top six inches (15.2 cm)
of sediment, TNT and RDX levels ranged from 1000 to 8,000 yg/g.
The core samples from the three lagoons showed a great degree of variability
in explosive concentrations. Explosive concentrations varied at different lagoon sites,
core depths, and between lagoons.
The results of the analysis for metals in the lagoon sediments are presented in
Table 5. Lagoon 9 sediment had elevated zinc levels. Site 1 of Lagoon 11 had higher
than normal lead and chromium levels.
Actual field samples of nitrocellulose contaminated sediment were not available
for these laboratory studies. Therefore sediment samples were obtained from a pond
located near Atlantic Research Corporation and spiked by addition of solid nitro-
cellulose to the sediment followed by mixing of the sediment to evenly disperse the
nitrocellulose. Nitrocellulose concentrations in the sediment ranged from 13,800
yg/g for the incineration studies to approximately 60,000 yg/g for the wet-air oxidation
tests.
All analyses were conducted according to the procedures presented in
Appendices A and B. The methods in Appendix A were verified according to the
USATHAMA QC plan. Detection limits were calculated according to the method of
Hubaux and Vos. These detection b'mits are summarized in Table 6. Several analyses
were performed for which the detection limit determinations were not deemed to be
necessary by the Project Officer. These analytical procedures are presented in
Appendix B.
13
-------
c/5
D
-------
Table 2. Core Sample Analysis for LAAP Lagoon 4
Site
4-1
4-2
Depth
(cm)
0
15.2
30.5
0
15.2
30.5
- 15.2
- 30.5
- 45.7
- 15.2
- 30.5
- 45.7
TNT
(^g/g)
510
330
610
1050
1050
780
RDX
-------
Table 3. Core Sample Analysis for LAAP Lagoon 9
Depth
Site (cm)
9-1 0
15.2
30.5
9-2 0
15.2
30.5
- 15.2
- 30.5
- 45.7
- 15.2
- 30.5
- 45.7
TNT
-------
Table 4. Core Sample Analysis for LAAP Lagoon 11
Site
11-1
11-2
Depth
(cm)
0 - 15.2
15.2 - 30.5
30.5 - 45.7
0 - 15.2
15.2 - 30.5
30.5 - 45.7
TNT
(pg/g)
7110
810
620
1400
930
830
RDX
(Pg/g)
8020
4010
2710
990
680
960
Nitrates
(Pg/g)
57
36
45
22
22
22
COD
(pg/g)
47,700
14,000
14,000
39,600
12,300
10,200
Vo la tiles
(%)
4.4
2.2
2.0
3.1
4.7
3.3
-------
Table 5. Core Sample Analysis for LAAP Lagoons - Metals
00
Site
4-1
4-2
9-1
9-2
ll-l
11-2
Lead
(pg/g)
20.8
6.2-8.8
9.6-35.0
12.1-34.0
54.9-86.7
33.6
Chromium
(pg/g)
23.2
21.7
25.5-30.6
20.4-35.2
64.4-113.0
16.1-21.7
Zinc
(Mg/g)
NA
NA
132-209
845-1408
195
NA
Cadmium
(yg/g)
NA
NA
2.5-4.0
14.9-23.0
NA
NA
NA - not analyzed
-------
Table 6. Detection Limits for Analytical Methods
Methods Detection Limits
Nitrate in Water 0.2 yg/ml
Lead in Water 0.25 yg/ml
Chromium in Water 0.3 yg/ml
Cadmium in Water 0.050 yg/ml
Zinc in Water 0.2 yg/ml
Nitrate in Sediment 2 yg/g
Lead in Sediment 7 yg/g
Chromium in Sediment - 7 yg/g
Cadmium in Sediment 3 Ug/g
Zinc in Sediment 5 yg/g
Nitrocellulose in Sediment 17 yg/g
TNT in Sediment - High Level 178 yg/g
RDX in Sediment - High Level 490 yg/g
TNT in Sediment - Low Level 2 yg/g
RDX in Sediment - Low Level 1 yg/g
Tetryl in Sediment - Low Level 0.3 yg/g
19
-------
III. INCINERATION
A. Process Description
Incineration or high temperature oxidation of materials is an extremely
effective decontamination method. When oxidation of the explosives of interest is
complete, gaseous carbon dioxide, nitrogen oxides, and water are the only products. For
explosives contaminated sediment, most of the inorganic portion of the sediment will
remain as ash.
Several incinerator types have been proposed for the incineration of explosives
contaminated lagoon sediment. Those which were examined in the Phase I report of
this study (Wentsel et aL, 1S82) are the air curtain incinerator, the cyclone or rotary
hearth furnace, the multiple hearth furnace, the electric furnace, the fluidized bed
incinerator and the rotary kiln. A brief description of each type of equipment is given
below.
The cyclone furnace or rotary hearth furnace, shown in Figure 2, is a refractory
lined steel cylinder with a single rotating hearth. Wastes are introduced at the outer
edge of the hearth and are gradually moved inward by a fixed plow toward an ash
discharge chute at the center of the hearth. Combustion air is injected above the
hearth at high tangential velocity in the opposite direction to hearth rotation. The air
sweeps over the burning waste, then spirals upward toward the exhaust. The cyclone is
operated at high temperatures with exit gases leaving the system at around 820°C
(1500°F) (EPA, 1979). Combustion products must pass through this high temperature
central vortex before leaving the furnace, and this passage essentially completes
combustion of all organics without any need for an afterburner (Stribling, 1972).
A schematic drawing of a multiple hearth incinerator is presented in Figure 3.
The furnace is a refractory-lined steel cylinder containing a number of horizontal
refractory hearths stacked vertically. Internally air cooled rabble arms pivot about the
center, raking burning material spirally towards the center or towards the edge on
alternating hearths. Waste material enters the top of the furnace, dropping from
hearth to hearth through the drying zone (310 to 540°C), the combustion zone (760°C
to 980°C), and the ash cooling zone (200°C to 315°C) (EPA, 1979). Combustion air is
introduced at the bottom of the incinerator, and may be preheated by first circulating
as cooling air through the rabble arms.
The electric furnace, shown in Figure 4, has a horizontal metal mesh conveyor
belt which passes under a series of electric heating elements. The steel shell is lined
with ceramic fiber insulation which can be heated and cooled very rapidly without
•damage. Waste material enters at one end of the furnace, and combustion air is
admitted to the other to provide countercurrent exchange of heat and oxygen.
A fluidized bed incinerator, shown in Figure 5, is a vertical refractory-lined
cylinder with a bed of sand or some other particulate suspended by air flowing up
through the bottom of the incinerator. Burners above the fluidized bed at the air inlet
provide heat to raise tne oed temperature ana to provide comoustion heat to low
20
-------
EXHAUST
COMBUSTION AIR
TANGENTIAL
AIR PORTS
BURNER (TYP)
CYCLONIC ACTION
ROTATING HEARTH
FIXED PLOW
SLUDGE
INLET
ASH DISCHARGE IN
CENTER OF FURNACE
Figure 2. Rotary Hearth Incinerator (EPA, 1979)
21
-------
COOLING AIR
DISCHARGE
SLUDGE CAKE,
SCREENINGS,
AND GRIT-
SCUM
RABBLE
ARM
DRIVE
AUXILIARY
AIR PORTS
RABBLE ARM
2 OR 4 PER
HEARTH
GAS FLOW
CLINKER
BREAKER
BURNERS
SUPPLEMENTAL
FUEL
COMBUSTION AIR
SHAFT COOLING
AIR RETURN
SOLIDS FLOW
DROP HOLES
ASH
DISCHARGE
Figure 3. Multiple Hearth Incinerator (EPA, 1979)
22
-------
to
CO
BELT
DRIVE -
SLUDGE FEED
Ox\ j
AIR
GAS
EXHAUST « |
1
LOCK --TfTjr
RADIANT
INFRARED
ROLLER HEATING r- WOVEN WIRE
- LEVELER ELEMENTS (TYP) \ CONTINUOUS BELT
COOLING 1 COOLING I
AIR 1 AIR I
r- RABBLING j 111
} \DEVICE j^^ \ j, I
•\i> oooooooooooooooo \
LJ»
r i i "r j
> COMBUSTION
^ AIR
ASM
DISCHARGE
Figure 4. Electric Furnace (EPA, 1979)
-------
^-EXHAUST AND ASH
SAND/M
FEED^S
THERMOCOUPLE
SLUDGE
INLET
FLUIDIZING
AIR INLET
FLUiDIZED/v/:':'.;/
SAND BED .'•:•:•:•/::
PRESSURE TAP
^SIGHT
Y GLASS
BURNER
REFRACTER
ARCH
TUYERES
FUEL
GUN
PRESSURE TAP
WINDBOX
STARTUP
h-i PREHEAT
DBURNER
_TFOR HOT
WINDBOX
Figure 5. Fluidized Bed Incinerator (EPA, 1979)
24
-------
heating value wastes (EPA, 1979). Sludge is fed either into the fluidized sand or into
the freeboard space above the sand bed (EPA, 1979). Bed temperature is maintained
between 750°C and 850°C (Dawson, 1978). The excellent mixing of the system and the
uniform high bed temperature generally make an afterburner unnecessary. Ash is
carried from the bed by the fluidizing air and is then removed from the gas stream by
a particulate scrubber, electrostatic precipitator, or baghouse.
A rotary kiln is a horizontal refractory-lined cylinder which rotates on a pair
of steel tires. The kiln is sloped downward from inlet to exit, with combustion air inlet
and the burner usually located at the lower end (Conway and Ross, 1980). The rotational
speed and angle of incline determine the residence time of waste material in the kiln
(Conway and Ross, 1980) which varies from several seconds to several hours (Scurlock
etaL, 1975). Combustion temperatures can range from 810 - 1650°C (Dawson, 1978).
A small skid-mounted rotary kiln unit is shown in Figure 6.
Air pollution is a potential difficulty which must be considered in the design of
any incineration system. Incomplete combustion of the material and entrainment of
particulates in the exhaust gas can pose serious environmental problems. Incomplete
combustion and volatilization of the waste materials are corrected by the addition of
an afterburner. Ash particles carried by the gas stream can be removed by scrubbers,
electrostatic precipitators, and baghouses. Vaporized metals can be removed from the
exhaust by a baghouse after the gas has been cooled.
B. Objectives
The objectives of the laboratory incineration experiments were to demonstrate
the feasibility of incineration as a sediment decontamination process and to establish
the temperatures and treatment times required to obtain complete destruction of the
explosives. Because the process conditions in most types of incinerators are non-
homogeneous, they are difficult to duplicate on a small scale. Air/sediment mixing
patterns and appropriate heating distributions are among the most difficult of the
process parameters to modeL The incineration experiments in this study, therefore,
made no attempt to simulate the operation of commercial incinerators.
C. Experimental Procedure
For the incineration experiments, large samples (approximately 100 g) of each
of the sediments from LAAP lagoons 9 and 11 and the spiked nitrocellulose sediment
were taken. These samples were thoroughly mixed to increase the explosives
uniformity and air dried. Duplicate subsamples of each of the three sediment samples
were taken and sent to Galbraith Laboratories in Knoxville, Tennessee for elemental
analysis and bomb colorimetry for heating value determinations.
Incineration tests were conducted by weighing approximately 4 g subsamples of
the dried sediments into preweighed open ceramic crucibles. Each crucible was heated
in a muffle furnace for the prescribed time. The subsamples were treated in random
order at 200°C, 300°C, 700°C and 900°C for intervals of 5, 30 and 60 minutes. After
the treatment interval, the crucibles were removed from the furnace, cooled in a
25
-------
Figure 6. Rotary Kiln (C-E Raymond, 1980)
26
-------
desiccator and reweighed. The heated subsamples were analyzed for residual explosives
and heavy metals by the methods described in Appendix A. The chemical oxygen
demand (COD) of each subsample was determined by the method described in Appendix
B. The residual explosives concentrations were then compared to the concentrations
in non-heated control subsamples to determine the effectiveness of heat treatment.
D. Results
The results of the elemental analyses and heating value determinations
performed by Galbraith Laboratories on TNT/RDX/tetryl containing sediments from
Lagoons 9 and 11 and on sediment spiked with nitrocellulose are listed in Table 7. The
sediment from Lagoon 9 contained high levels of explosives and consequently had high
percentages of carbon, nitrogen and oxygen and a high heating value. The sample from
Lagoon 11 contained only small amounts of organic material and had a very low heating
value. Sediment spiked with approximately 1.5% nitrocellulose had an intermediate
heating value. Very little sulfur was found in any of the samples.
The results of incineration experiments on sediment from Lagoon 9 (which
contained high levels of TNT, RDX and tetryl) are presented in Table 8. Sample
weights before and after treatment are presented in Table C-l of Appendix C. At
200"C, 0.3% of the original TNT remained after one hour. Liquid chromatography
analysis detected no RDX or tetryl in these samples. After heating at temperatures
of 300°C and higher, the explosives could not be detected, even after contact times
as snort as five minutes. Chemical oxygen demand at each temperature was generally
reduced with increasing treatment time, however, some exceptions were observed.
These exceptions were most likely due to variations in the original sediment and to the
fact that the lowest COD levels reported were close to the minimum that can be
determined with the analytical procedure. Reaction temperature had a much larger
effect on COD levels than did reaction time. COD reductions ranged from only 30-40%
at 200°C to more than 99% at 500°C to 900°C for reaction times of 30 minutes or
more.
Results of incineration experiments on Lagoon 11 sediment are presented in
Table 9. Explosives in these experiments did not decompose as rapidly as in the
previous experiments with Lagoon 9 sediment and COD levels did not drop as rapidly
or as completely. After 30 minutes at a temperature of 500°C, 0.04% of the original
TNT, 0.06% of the original RDX and 0.04% of the original tetryl remained in the
sediment. Increasing reaction time had little effect on the remaining explosives, but
COD was reduced. Similar results were observed at 700°C. While explosives reduction
was dramatic, TNT and RDX were both detectable in the treated sediment. One
possible explanation for the more rapid and complete combustion of the highly
contaminated Lagoon 9 sediment is that the large amount of heat released by oxidation
of the explosives drives the temperature of the sample higher than the controlled
furnace temperature. This excess heat results in spontaneous combustion of the residual
explosives in the sample.
One area of concern was the potential for volatilization of the explosives
instead of decomposition. To determine if explosives volatilization occurs and
therefore the necessity of an afterburner, a volatilization experiment was performed on
Lagoon 9 sediment. The results of this experiment are presented in Table 10. Exhaust
27
-------
to
oo
Table 7. Elemental Analysis and Heating Value of Air-Dried Sediments
Heating Valu<
Lngoon 9
Lagoon 11
Nitrocellulose
°C
Ave
17.49 15.82
14.14
0.36 0.37
0.38
4.86 4.87
4.88
°H
Ave
1.22 1.14
1.06
0.11 .11
0.10
1.06 1.04
1.01
% N
Ave
10.28 9.25
8.21
0.18 .17
0.15
1.08 1.10
1.01
% S
Ave
0.01 .01
0.01
0.05 .06
0.06
0.06 .06
0.06
% 0
Ave
17.52 17.88
18.23
2.45 2.55
2.65
14.34 14.35
14.35
BTU/lb.
2326
2255
58
63
763
728
Av
2291
61
751
-------
Table 8. Incineration of Lagoon 9 Sediment - Explosives Levels
Concentration in Dry Sediment
NS
CO
Temperature
<°C)
No heat
200
300
500
700
900
Time
(min)
5
30
60
5
30
60
5
30
60
5
30
60
5
30
60
TNT
(Pg/g)
424,000
10,000
1,500
1,350
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
RDX
(Pg/g)
159,000
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
Tetryl
(Pg/g)
15,800
114
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
COD
(Pg/g)
206,000
124,500
116,500
149,200
55,200
52,300
30,000
5,920
2,190
1,280
8,720
1,310
2,320
12,200
2,510
1,670
-------
Table 9. Incineration of Lagoon 11 Sediment - Explosives Levels
Concentration in Dry Sediment
Temperature Time
(°C) (min)
No heat
200 5
30
60
300 5
30
60
500 5
30
60
700 5
30
60
TNT
(pg/g)
47,300
27,800
1,290
1,170
<178
<178
40
20
20
35
30
3
<1
RDX
(Mg/g)
15,900
4,430
< 490
146
< 490
< 490
20
10
10
10
20
5
30
Tetryl
fcg/g)
2,100
780
< 70
< 70
< 70
< 70
6
< .3
.9
< .3
.8
< .3
< .3
COD
(pg/g)
81,800
32,800
46,900
50,900
57,700
29,800
2,100
16,800
13,400
300
14,000
12,200
8,100
-------
Table 10- Volatilization of Explosives at 200°C
Explosive
TNT
RDX
Tetryl
Explosive Level
Before Heating
(Pg/g)
218,000
32,200
6,600
Explosives in Sediment
After Heating
(%)
34.6
27.3
19.0
Explosives
Volatilized
(%)
14.5
3.3
< 30
-------
gases were passed through a condenser and the condensate was rinsed with known
quantities of acetone and the rinse analyzed for explosives. After one hour at 200°C,
19-35% of each of the original explosives remained in the sediment and 3.3% of the
RDX and 14.5% of the TNT were recovered in the condensate. Tetryl was not detected
in the condensate, but the low initial level of tetryl and the dilution of the rinse
required that more than 30% of the original amount be present for detection. This
experiment shows that some volatilization without combustion can occur. If the
incinerator is one in which the exhaust gases will not pass through a high temperature
region to complete combustion, an afterburner will be required to ensure complete
destruction of the explosives.
Another possibility which must be considered in incineration design is the
potential for vaporization of heavy metals. Lagoon 9 and Lagoon 11 sediment samples
were analyzed for several metals before and after the incineration experiments. The
results of these analyses are presented in Tables 11 and 12. Metal levels before and
after treatment remained approximately the same up to temperatures of 700°C for
sediments from both Lagoons 9 and 11. Increases in metal levels after treatment are
the result of burning away the organic portions of the sediment but leaving the same
total amount of metals. Other variations in the data are due primarily to variations
in the orignal sediment. At 900°C, lead, chromium, and cadmium levels were reduced
significantly, indicating that all three metals were vaporized or carried out in dust with
the air stream during treatment. In a commercial scale incinerator, either the
temperature will have to be controlled to prevent metals vaporization or the metals
will have to be recovered from the gas stream. Such recovery can be accomplished by
a baghouse after the gas has been cooled.
The results of incineration experiments on sediments spiked with nitrocellulose
are shown in Table 13. At all treatment temperatures and times, nitrocellulose was
reduced to below its detection limits. COD levels were reduced by longer treatment
times at any given temperature, but as was observed with the sediments from Lagoons
9 and 11, furnace temperature produced a much larger effect. At 900°C, COD levels
were reduced by 99.8% at treatment times of 30 and 60 minutes. At 200°C, the COD
level was reduced by 83% after one hour.
E. Conclusions
These experiments demonstrated that incineration is an effective method for
destroying TNT, RDX, tetryl and nitrocellulose in lagoon sediment. Essentially
complete decontamination of the sediment for all of the explosives tested can be
achieved at temperatures of 300-500°C with treatment times of 30 minutes or less.
While all of the explosives can be reduced to their detection limits at the lower
temperatures, it is possible that some toxic decomposition products may remain. It is
therefore important to use temperatures which reduce the total organic content as
measured by COD to acceptable levels. This can be accomplished at temperatures of
500-700°C and reaction times of 30 minutes. Since explosives volatilization may occur,
it will be important in a pilot scale study to determine whether any vaporized
explosives can be detected in the exhaust gases.
Metals are vaporized if the sediment is incinerated at 900°C, but are not
vaporized if the temperatures are maintained at 700°C or less. Vaporization can be
minimized by controlling incineration temperature at/or below 700°C. Alternatively,
the incinerator could be operated at higher temperatures if a cooling system and
baghouse were provided to recover the vaporized metals. The incineration laboratory
studies were conducted on dried sediment containing relatively high concentrations of
explosives. The dried sediment represents a worst case from a potential hazards point
of view. However, in the field, the sediment will contain significant amounts of water
e.g. 20-80%. This water wiU render the sediment safer to handle in the digging, mixing
32
-------
Table 11. Incineration of Lagoon 9 Sediment - Metals Levels
CO
GO
Temperature
(°C)
No heat
300
500
700
900
Time
(min)
5
30
60
5
30
60
5
30
60
5
30
60
Lead
(ug/g)
75
74
99
74
93
120
84
90
100
160
81
5.2
18
Chromium
-------
Table 12. Incineration of Lagoon 11 Sediment - Metals Levels
CO
.fc.
Temperature Time
(°C) (min)
No heat
200 5
30
60
300 5
30
60
500 5
30
60
700 5
30
60
Lead
(Pg/g)
31
16
28
18
19
29
31
33
26
31
21
26
21
Chromium
(pg/g)
24
7
10
10
10
7
10
30
10
10
20
40
5
-------
Table 13. Incineration of Nitrocellulose Contaminated Sediment
CO
en
Temperature
No heat
300
500
700
900
Time
(min)
5
30
60
5
30
60
5
30
60
5
30
60
Nitrocellulose COD
(Pg/g) (ug/g)
13,800 106,000
< 17 38,400
<17 22,200
<17 17,900
<17 12,000
<17 3,500
<17 1,200
<17 2,600
<17 690
<17 1,020
<17 2,500
<17 220
<17 200
-------
and incinerator feed stages. The largest problem will be the amount of energy required
to simply vaporize the water.' Incinerator design or optimization of sediment through-
put will thus largely be based on the energy requirements to vaporize the water and
heat the sediment to the desired temperature. Essentially the explosives will go along
for the ride.
Incineration has been demonstrated on a commercial scale for destruction of
waste explosives, and equipment i§ available which would not require signifi-
cant modification to process explosives contaminated sediment. Of all the processes
evaluated, incineration is the one which could be directly implemented at an Army
Ammunition Plant most quickly and with the least developmental costs.
F. Future Work
Pilot scale tests of incineration to treat TNT/RDX/tetryl and nitrocellulose
containing sediments are scheduled for this fall. Tests will probably be conducted in
a rotary kiln incinerator, a piece of equipment which has been used in the past for
munitions incineration. A subcontractor will provide facilities and personnel to
perform pilot scale incineration tests of TNT/RDX/tetryl and nitrocellulose
contaminated sediments. The results of the tests will be used to determine the
technical feasibility, scale-up potential and costs of incinerating approximately
5,500,000 kg/yr (12,100,000 Ib/yr) of sediment.
Each of the two sediment types will be tested at a minimum of four different
incineration conditions for at least two hours after stabilization of the system.
Temperature, air flow, fuel consumption, and feed rate will be monitored during the
tests. The ash, exhaust gas, and any scrubber or other effluents will be sampled
periodically. These samples will be analyzed by ARC to monitor system performance.
After the pilot scale tests are completed, the results will be used to provide a better
estimate of incineration costs and appropriate operating parameters for treating
explosives containing sediments.
G. Economic Analysis
Capital and operating costs (based on the limited labortory tests) to treat the
sediment from 11 standard lagoons per year are presented in Table 14. Capital costs
include a dredge, a holding tank, a flat bed trailer on which to carry the equipment
and the incinerator with afterburner and/or other pollution control devices. Operating
costs were calculated on the basis of the sediment composition and moisture content
described for the standard lagoon. Capital and operating costs are roughly equivalent
for all of the incinerators evaluated so ability of the system to handle the explosives
contaminated sediments will be the major factor in choosing between them. The
effects of sediment moisture content on fuel costs, a major component of the operating
costs, are shown in Table 15. These costs were based on the excess air requirements
and operating temperatures for the various systems, as listed in Table 16. The rotary
kiln has a maximum fuel cost higher than the other incinerators because it can operate
at temperatures up to 1650°C. The high explosives level sediment is capable of
sustainiiig combustion without additional fuel when it is dry, while the low level
sediment requires $100,000 - $200,000 per year in fueL ^,watar..$oatejit,goes up, fuel
-------
Table 14. Incineration Costs (1980 Dollars)
Incinerator Capital Costs Annual Operating Costs
Rotary Hearth Furnace $ 705,800 $310,000 - $440,000
Multiple Hearth Furnace $ 751,000 $460,000 - $500,000
Electric Furnace $ 560,000 $380,000 - $550,000
Fluid Bed Incinerator $ 891,000 $430,000 - $490,000
Rotary Kiln $ 496,000 $370,000 - $420,000
-------
CO
00
Table 15. Effect of Sediment Water Content on Fuel Costs
Annual Costs in Thousands of Dollars/Year (1980 Dollars)
Incinerator
High Heating Value (2000 Btu/lb)
Rotary Hearth
Multiple Hearth
Electric Furnace
Fluidized Bed
Rotary Kiln
Low Heating Value (100 Btu/lb)
Rotary Hearth
Multiple Hearth
Electric Furnace
Fluidized Bed
Rotary Kiln
Dry Sediment
0
0
0
0
0
104-156
115-168
92-152
92-113
104-256
60% Solid Slurry
20-91
25-102
1-86
1-34
'"-265
305-377
304-388
286-371
286-320
305-552
15% Solid Slurry
1525-1736
1483-1747
1460-1730
1460-1586
1524-2485
1810-2021
1768-2032
1745-2016
1745-1871
1810-2771
-------
Table 16. Incinerator Operating Temperatures and Excess
Air Requirements
Incinerator
Temperature Range
Excess Air Requirements
Rotary Hearth
Multiple Hearth
Electric Furnace
Fluidized Bed
Rotary Kiln
820
760-980
760-980
750-850
810-1650
30-80
95-100
20-70
20-30
30-50
39
-------
IV. GAMMA IRRADIATION
A. Process Description
Gamma irradiation can breakdown organic molecules by either direct or
indirect methods. Direct ionization by gamma rays will often cleave chemical bonds
leading to the formation of free radicals. Gamma irradiation also produces H and OH
free radicals which can react to form ozone and hydroperoxide. These oxidizing agents
also contribute to the degradation process through oxidative break down of organic
matter present in the medium.
Two isotopes, Cesium-137 (Cs-137) and Cobalt-60 (Co-60), are currently
employed as sources in the gamma irradiation systems used in sludge disinfection
treatment processes. Cs-137 is available as a by-product from the processing of nuclear
weapon wastes. Co-60 is produced by exposing cobalt metal to accelerated neutrons
(EPA, 1979).
Gamma irradiation, unlike some other ionization processes, is very penetrating.
Sixty-four cm of water are required to stop 90% of the radiation from a Co-60 source.
This penetration range makes gamma irradiation useful for the treatment of relatively
significant amounts of sediment or sludge.
Two types of gamma irradiation systems are used to disinfect sludge. One
system, located in Germany, uses a Co-60 source and pumps liquid sludge around the
source. Another system, shown in Figure 7, is located at Sandia Laboratories in
Albuquerque, New Mexico. This unit has a one-million curie source of Cs-137 and can
process up to 8 tons of dried sludge per day. The sludge is loaded onto a conveyor and
passed by the source. The exposure is determined by the speed of the conveyor. The
Sandia sludge treatment system was used for irradiation of sediment samples for this
study. Sandia also has a 90,000 curie Co-60 source which is used for research purposes.
Several sediment samples were also irradiated with this source.
B. Experimental Procedure
1. Sample Preparation and Irradiation
Actual samples from LAAP lagoon 9 were used for TNT/RDX contamin-
ated sediment. Uncontaminated sediment was spiked with nitrocellulose (10% by weight)
for use in the gamma irradiation study. Analysis of the LAAP lagoon 9 and the spiked
nitrocellulose sediments are presented in Table 17.
For the gamma irradiation studies, aqueous slurries containing 10% solids
(w/w) were formulated from these sediments according to the following procedure. A
weighed amount of dried sediment was placed in a two-gallon Helicone® mixer.
Distilled water was added to the sediment in the proper ratio. The sediment and water
were slowly agitated to wet the particles. After the initial mixing, the mixer speed
was increased to 185 rpm to evenly suspend the particles and provide a uniform slurry.
One hundred ml aliquots of the slurry were dispensed from the valve in the bottom of
the mixer into preweighed polyethylene bottles. Agitation was continued during sample
40
-------
Input
Conveyor
"Source Plaque
Figure 7. Gamma Irradiation Treatment System for Dried
Sewage Sludge (Morris et al., 1979)
41
-------
Table 17. Characteristics of Unslurried LAAP Lagoon 9 and Spikea
Nitrocellulose Sediments
TNT/RDX Nitrocellulose
Sediment Sediment
ND - Not Determined
TNT
(yg/g, dry wt) 287,000
RDX
(ug/g, dry wt) 58,000
Nitrocellulose
(ug/g, dry wt) - 34,000
COD (yg/g) 470,000 166,000
Nitrate (pg/g) 4.5 2
Chromium (ug/g, dry wt) 4 ND
Cadmium (yg/g, dry wt) 5 ND
Lead (ug/g, dry wt) 40 ND
Zinc (yg/g, dry wt) 500 ND
Percent Volatiles 37.7 8.4
42
-------
withdrawal. The polyethylene bottles were reweighed, capped, labelled and refriger-
ated until treatment. Thirteen bottles of each sediment type were prepared in this
manner. Two samples (a and b) were irradiated at each set of conditions excepting the
samples receiving 4.1 megarads. These samples were included to provide additional
information and were not in the laboratory test plan. Samples were randomly assigned
aeration or non-aeration labels and the dosage levels of 0, 0.5, and 1.5 megarads. The
experimental design is presented in Table 18.
The non-aerated samples were irradiated by the Cs-137 sludge treat-
ment unit. The samples were irradiated in two batches. The first batches were exposed
for 45 minutes and received 0.5 megarads. The second batch received 1.5 megarads in
135 minutes of irradiation. Because only sealed bottles could be treated on the Cs-137
sludge plant, the aerated samples were exposed to a Co-60 source. The samples were
aerated during exposure by bubbling air through the solution via airstones and a fish
tank pump. These samples received 0.5, 1.5 or 4.1 megarads with treatment times of
4, 13, and 35 minutes, respectively. After treatment, the samples were refrigerated
and transported to Atlantic Research Corporation for analysis.
2. Analysis of Irradiated and Control Slurries
The analyses of the irradiated and control slurry samples were accom-
plished in the following manner. The water and sediments were separated by
decantation and filtration. The sediments were air dried on the dull side of aluminum
foil. The volume of water and the weight of the sediment were then determined. Each
sediment and water sample was then analyzed in duplicate for the following
parameters:
explosives (TNT, RDX and tetryl, or nitrocellulose)
Chemical Oxygen Demand (COD)
Nitrate
Analyses were according to the procedures presented in Appendices A and B. For the
non-aerated samples, the air above the liquid was sampled and analyzed to determine
if any CO, C02 or NOX was produced upon irradiation.
C. Results and Discussion
The characteristics of the TNT/RDX slurries formulated from lagoon 9
sediment are presented in Table 19. The sediment from lagoon 9 had highly variable
explosives concentrations from sample to sample. This variability was obvious when
the dried sediment was examined since crystals and lumps of TNT and RDX could be
observed in the dried material. The slurries formulated from the sediment were also
high variable in explosive concentrations. A one way analysis of variance was
performed on the TNT, RDX, tetryl and COD data to determine if any of the treated
samples was significantly different from the control or any other treated samples.
None of the treated values was significantly different although it appears that the
higher dose rate (4.1 megarads) had some effect on the degradation of the explosives.
43
-------
Table 18. Experimental Design for Gamma Irradiation Study
RDX/TNT Nitrocellulose
dose (megarads) aeration Non aeration aeration Non aeration
0
0.5
1.5
4.1
a
a
a
a
b
b
b
a
a
a
b
b
b
a
a
a
a
b
b
b
a b
a b
a b
a = Sample bottle a
b = Sample bottle b
44
-------
Table 19. Effects of Gamma Irradiation on TNT/RDX Sediment Slurries
Dose Water Volume Sediment
-------
The nitrocellulose slurries characteristics are summarized in Table 20.
Nitfocellulose concentrations in this table are presented as nitrite and have not been
converted back to nitrocellulose. In the analysis of nitrocellulose, the nitroceDulose is
hydrolyzed and reduced to nitrite. If the numbers are converted back to nitrocellulose,
the data would show an increase in the nitrocellulose concentration. This increase is
due to the breakdown of the nitrocellulose by gamma irradiation. The products of the
irradiation evidently are easier to hydrolyze than the nitrocellulose (25% hydrolysis
efficiency) yielding a higher nitrite leveL This explanation is further supported by the
COD data. * COD levels were greater in the treated sediment than in the control
sediment. Nitrocellulose is poorly oxidized by the COD method. Gamma irradiation
partially degrades the nitrocellulose rendering it more susceptible to oxidation by the
COD procedure. Nitrate levels in the slurry water also increased.
The nitrocellulose data were subjected to one-way analysis of variance. This
analysis showed that the nitrite data was significant at the less than 1% level, the
nitrate data at the 3% level, and the COD data at the 2% level. A table of means for
the Duncan's Multiple Range Test is presented in Table 21. These data show a
significant difference in nitrite and COD levels between the controls and the non-
aerated samples at a dose of 0.5 megarads. At a dose of 1.5 rnegarads, the data were
not significantly different than the controls.
Analysis of the air above the solutions in the non-aerated irradiated samples of
both Lagoon 9 and nitrocellulose sediment samples did not show the presence of any
CO. In addition, CO2 and NOX levels were not significantly different between the
control and irradiated samples.
D. Conclusions
The gamma irradiation treatment of TNT/RDX contaminated sediment was
partially effective at a dose of 4.1 megarads. Approximately 30% of the explosives and
COD levels were degraded. The high amount of explosives (over one-third of the
sediment weight) made it difficult for the method to be effective. Nitrocellulose was
also partially degraded in the sediment by gamma irradiation. However, the
exact quantity of nitrocellulose degraded by the gamma irradiation could not be
determined, due to problems associated with the analytical method for nitrocellulose.
Overall, gamma irradiation did not reduce the explosive levels in the sediment
to acceptable levels in these experiments. Extrapolation of the data indicate that a
dose of 16.2 megarads or greater would probably be required to completely degrade the
explosives. This dose would require a holding time of about 140 minutes in the presence
of the 1,000,000 Curie Co-60 source with aeration or 24.7 hours in the presence of the
90,000 Curie Cs-137 source. This treatment procedure may be effective for removing
low levels of explosives contamination in soils, sediment or liquids. However, reaction
conditions must first be optimized before this method can be useful in toxic materials
destruction.
46
-------
Table 20. Effects of Gamma Irradiation on Nitrocellulose Sediment Slurries
Nitrocellulose
Sample
Control a
b
Control a
b
Non-aerated a
b
Non-aerated a
b
Aerated a
b
Aerated a
b
Dose
(megarads)
0
0
0
0
0.5
0.5
1.5
1.5
0.5
0.5
1.5
1.5
Water Vol.
(mL)
78
91
94
92
85
87
83
88
84
82
54
87
Sediment
dry wt. (g)
7.70
7.92
7.11
7.75
7.97
6.92
5.45
7.11
6.22
7.03
6.38
7.84
% Solids
in Slurry
11.3
8.7
4.8
7.5
9.4
8.0
6.6
8.1
7.4
8.6
11.8
9.0
Expressed as
Nitrite (mg/g)
8.6
8.4
8.7
8.9
26.0
24.7
12.2
23.3
13.0
11.0
18.5
14.0
COD
(mg/g)
167
167
170
160
225
227
204
156
163
163
191
194
Nitrate
(ug/mL)
1.0
0.3
0.5
0.3
0.5
1.0
2.0
1.0
1.0
1.0
1.0
2.0
Aerated a
4.1
105
8.67
8.3
12.3
190
3.0
-------
00
Table 21. Duncan's Multiple Range Test for Nitrocellulose Sediment
Parameter Measured
Dosage (megarads) COD (ug/g) Nitrite (mg/g) Nitrate (ug/mL)
Non-aerated 0
0.5
1.5
Aerated 0.5
1.5
4.1
166,000 b
226,000 a
180,000 b
163,000 b
192,500 a.b
190,000 a.b
8,650 c
25,350 a
17,750 a.b
12,000 b.c
16,250 b
12,300 b,c
0.5 c
0.8 c
1.5 b
1.0 b,c
1.5 b
3.0 a
Values within a column not followed by the same letter are significantly different at the 5% level
of probability according to Duncan's Multiple Range Test.
-------
E. Future Work
Gamma irradiation is not an effective method for decontaminating lagoon
sediments containing high levels of explosives. To obtain complete degradation, large
doses of radiation are required, driving equipment costs up as residence time is
increased. Since the process has not been used in the past to destroy explosives, a
considerable amount of experimentation and process development is required to
optimize the system configuration and operating conditions. For these reasons, further
development of gamma irradiation is not recommended for this application. In other
situations where explosives levels are low and would require shorter residence times for
a particular source, gamma irradiation may prove to be a very inexpensive treatment
method which would warrant further development.
F. Economic Analysis
Capital costs for a gamma irradiation facility using a Cs-137 radiation source
to process 900 kg/hr of explosives contaminated lagoon sediment are presented in Table
22. Costs for the facility include an insulated concrete building, an emergency water
dump tank for source shielding, cesium capsules, a source handling pool, aeration
equipment, control equipment, pumps, piping, flow meters, a radiation alarm and a fire
suppression system (EPA, 1979). Annual operating costs, including annual costs for
replacement Cs-137, are presented in Table 23. Costs in the literature (EPA, 1979) for
gamma irradiation facilities are based on a system which delivers only a 1 megarad
dose. To provide a 16.2 megarad dose, the required size of the system was increased
to give a much longer residence time. The major operating cost for this system is
labor which can probably be reduced if there is other work available not related to the
process to which the operators can devote any extra time. Projected labor
requirements for a system of this size are only 2,400 man-hours per year (EPA, 1979),
much less than the 6,000 operator man-hours and 2,000 supervisor man-hours provided
in this estimate. Further cost reductions can be made if Cs-137 can be obtained at
reduced rates from the processing wastes of nuclear weapons manufacturing. While a
gamma irradiation facility would be expensive to build, it would be relatively
inexpensive to operate once the system was installed.
49
-------
Table 22, Capital Costs for Gamma Irradiation
Equipment
Cost
Reference
Dredge: Porta-Dredge, 1136 L/min $ 58,800
Holding Tank 145,600
Gamma Irradiation Facility 1,200.000
TOTAL COST $1,404,400
Salemink, 1980
Peters and Timmerhaus,
1968
EPA, 1979
Table 23. Annual Operating Costs for Gamma Irradiation
Cesium-137
Electricity (70,000 kwh/yr @ $0.07/kwh)
Maintenance (3% of capital)
Labor
TOTAL COST
$ 42,000
4,900
42,000
210,000
$298,900
50
-------
V. WET-AIR OXIDATION
A. Process Description
Wet-air oxidation is a process which uses a catalyst and oxygen to destroy
organic compounds in aqueous mixtures at elevated temperatures and pressures. The
waste slurry is pumped by means of a high pressure pump through a heat exchanger
where it is heated to reaction temperatures. It then passes to a high pressure reactor
along with compressed air. Specific catalysts are added to the mixture to increase the
reaction rate. Retention times in the reactor are on the order of 40 to 60 minutes. The
treated slurry is passed from the reactor through the heat exchanger where it is cooled.
The cooled slurry then goes to a gas-liquid separator where gaseous products are
removed. Depending on the type of waste, the gaseous or liquid products may require
further treatment before they are released into the environment. Slurries containing
between 5 and 10 percent solids are treated by this process. Higher amounts of solids
create problems with mixing and mass transfer of the oxygen.
B. Experimental Procedure
L Slurry Preparation and Wet-Air Oxidation Treatment Methodology
The wet-air oxidation experiments were conducted at IT-Enviroscience in
Knoxville, Tennessee. The reactions were carried out in a stirred one-liter titanium
clad reactor. The reactor, catalyst addition system and associated equipment are
shown in Figure 8. Photographs of the reactor and control equipment are presented in
Figures 9 and 10.
Ten percent Lagoon 9 sediment slurries in water were formulated as
described for the gamma irradiation study. The same sediments were used in the
formulation. Aliquots of 350 ml (TNT/RDX) or 150 ml (nitrocellulose) were dispensed
from the mixer into pre-weighed polyethylene bottles. The bottles were re-weighed,
capped, labelled and refrigerated until treatment.
The treatment scheme for the wet-air oxidation studies is presented in
Table 24. The experimental procedure consisted of adding the catalyst mixture to the
catalyst addition tanks. The catalyst mixture typically consisted of 3.81 g of MnSO4
• HoO, 35.47 g of 70.5% HNO3, 6.75 g of 47.2% HBr, and 67.2 g of deionized H2O. The
sediment slurry sample was added to the reactor along with deionized water rinses (50
g) of the sample bottle. The reactor was sealed and pressurized with helium to 800 psig
to check the system for leaks. The reactor was vented and batch-purged with cylinder
oxygen to remove inerts. It was then pressurized to 200 psig with cylinder oxygen and
heated to the desired reaction temperature. The catalyst mixture was added to the
reactor with helium pressure to initiate the catalyzed oxidation of the sample. The
reaction was terminated after the desired length of time by cooling the reactor with
cooling water through the reactor jacket. The reactor was cooled to 20°C before
venting the final reactor gas through a sampling valve into a gas sample bag for
analysis by gas chromatography for C02- The final liquid reaction effluent was
aspirated into a clean bottle. These samples and the final gas samples were shipped
to Atlantic Research Corporation for analysis.
51
-------
0
OXYGEN/CATALYST
ADDITION VALVE
VENT/CATALYST
ADDITION VALVE
VENT TO BUILDING ALLEY
en
ro
CHECK
VALVES
j
J\
-T-V-^ w
i, [) sssasr ssa?
? TANK 9 ?
1 , r [^ 1 n
RELIEF
OXYGEf| PRESSURE GAGE VALVE
CYLINDER
r— in-
COOLING WATER!
DRAIN
^
MAGNETICALLY
r— j COUPLED \
{ \ AGITATOR <
of
fl
_— I !
1 ^^
1,
1 LITER
RUPTURI
VENT/CATAI
-------
Figure 9. Picture of Wet-Air Oxidation Reactor
Figure 10. Picture of Wet-Air Oxidation Control Equipment
53
-------
Table 24. Treatment Scheme for Wet-Air Oxidation of Sediment Slurries
Order
Sediment Type
Temperature (°C)/
Pressure (psig)
Treatment Time
(min)
1 RDX/TNT
2 RDX/TNT
3 RDX/TNT
4 RDX/TNT
5 RDX/TNT
6 RDX/TNT
7 Nitrocellulose
8 Nitrocellulose
9 Nitrocellulose
10 Nitrocellulose
200/600
225/750
225/750
200/600
250/980
250/980
200/600
165/450
200/600
165/450
60
60
90
90
90
60
60
60
90
90
54
-------
2. Sample Analysis
The sample handling and analysis procedures for the slurries were
essentially the same as described under gamma irradiation.
Analysis was also conducted for by-products from the wet-air oxidation of
the TNT/RDX sediment. Ten ml of water from two sediment slurries (before and after
wet-air treatment) were extracted with 1 ml of methylene chloride and the extracts
concentrated to 0.25 ml. This extract was analyzed by gas chromatography-mass
spectrometry using a Hewlett-Packard 5992A Mass Spectrometer and the following
conditions:
Column: 2% Dexsil 300 GC on Anakrom Q packed in a
6 ft. by 0.25 in and 2 mm I.D.
Gas Flow: Helium (9. 15 cc/min
Temperatures:
injector: 210°C
column: 160°C to 240°C <§. 15°C/min
C. Results and Discussion
The results of the wet-air oxidation treatment of TNT/RDX contaminated
sediment are presented in Table 25. The data indicate that RDX and tetryl were
reduced to below detection limits of the GC analysis method at the lowest treatment
scheme of 200°C/600 psig. The low levels of RDX and tetryl in the final product could
not be determined by the more sensitive HPLC method because the presence of large
quantities of TNT breakdown and addition products obscured the peaks.
The TNT was rapidly decarboxylated in the wet-air oxidation reaction. No TNT
was observed in any of the product solutions or sediments. However, 1,3,5-
trinitrobenzene (TNB) was observed in large amounts. The TNB residual decreased as
the treatment temperature and pressure increased. At the highest treatment scheme
of 250°C/980 psig, the trinitrated ring (from TNT or TNB) was reduced by
approximately 99%. At this treatment level, COD was also reduced by 90-95%.
A one-way analysis of variance table showed that the data for TNB
removal by treatment and time interaction were significant at the 0.004% level.
Further evaluation of the data by Duncan's Multiple Range Test is presented in Table
26. Temperature/ pressure combinations had a significant effect on the disappearance
of TNB, however, treatment times above 60 minutes at a specific temperature/pressure
combination had no increased effect on TNB disappearance.
The chromatrograms of the slurry water extracts from before and after wet-
air oxidation are presented in Figures 11 and 12. Before wet-air oxidation, only three
main peaks were observed. These peaks were identified as TNT, RDX and tetryl by
comparison with SARM materials. After wet-air oxidation, no RDX or tetryl was
observed. 1,3,5-Trinitrobenzene, 2,4-dinitrobenzene, bromodinitrobenzene and dibromo-
55
-------
Table 25. Wet-Air Oxidation Treatment of LAAP Sediment
Remaining After Treatment
Pressure (psig)
Control
200/600
200/600
225/750
225/750
250/980
250/980
(min)
0
60
90
60
90
60
90
TNB (Mg/g)
_*
89,000
+5,600
87,000
+18,000
142,000
+27,000
172,000
+21,000
2,600
+1,600
1,800
+1,700
RDX (Mg/g)
42,000
+14,300
< 490
< 490
< 490
< 490
< 490
< 490
Tetryl (Mg/g)
10,400
+3,000
<75
<75
<75
<75
<75
<75
COD (Mg/g)
172,000
NA
NA
NA
NA
7,900
17,000
NA - Not Analyzed
*Initial TNT walues 236,000 + 25,000 Mg/g
-------
Table 26. Evaluation of Wet-Air Oxidation Data by
Duncan's Multiple Range Test
Treatment Temperature (°C)/ Treatment Time Trinitrated Ring Structure After
Pressure (psig) _ (min) _ Treatment
200/600 0 230177 a
60 80171 c
90 76743 c
225/750 0 237428 a
60 123976 b
90 155322 b
250/980 0 239030 a
60 1634 d
90 2586 d
Values with a column not followed by the same letter are significantly different at
the 5% level of probability according to Duncan's Multiple Range Test.
57
-------
TETRYL
Figure 11. GC-MS of Slurry Water Before Wet-Air Oxidation Treatment
58
-------
dlbromonitrobenzene
bromodinitrobenzene
Figure 12. GC-MS of Slurry Water Extract After Wet-Air
Oxidation at 250°C/980 psig
59
-------
nitrobenzene were identified as the main constituents of this extract. Thus, it appears
that the mechanism for TNT destruction by wet-air oxidation proceeds via rapid
oxidation of the methyl group followed by decarboxylation to form TNB. This initial
step is followed by removal of the nitro group and in some cases a substitution of
bromide for the nitro. The compounds formed from wet-air oxidation of TNT are more
hazardous than TNT itself. TNB and dinitrobenzene (DNB) are explosives, and are more
soluble in water than TNT (460 mg/1 for 1,3-DNB and 278 mg/1 for TNB compared to
130 mg/1 for TNT) (Wentsel et aL, 1979; U.S. Army, 1967). TNB and 1,3-DNB are known
to be toxic to humans. Both compounds are potent methemoglobin formers (Wentsel
et aL. 1979).
The reactor air analysis from the treatment of TNT/RDX sediments is
presented in Table 27. The initial atmosphere in the reactor was pure oxygen. The CO 2
levels for the treated sediment ranged from 12.6% of the total gas volume at
200°C/600 psig for 60 minutes to 34.1% at 250°C/980 psig for 90 minutes. Overall the
C02 levels increased as the temperature and treatment time increased. A brown gas
was also observed in the air sampler bags. This gas was identified as bromine from
the catalyst. The bromine gas was not quantified. No significant quantities of NO or
were found.
The amounts of CO2 and N2 which could theoretically be released by complete
decomposition of the explosives present in the treated slurry were calculated to be 5.18
g CO2 and .895 g of N2« At the highest temperature/pressure/time conditions, 34.1%
of the reactor exhaust was C0%. If this number is adjusted for the pressure, 5.36 g of
CO2 were given off. This amount is 0.18 g greater than the theoretical. Thus, it
appears that a high percentage of the explosives was essentially completely oxidized
and that other organic material present in the sediment was also oxidized. The actual
amount of N2 measured (adjusted for pressure) was 1.10 g. Again the actual amount
measured was slightly higher than the theoretical. The additional N£ could be the
result of leakage or could have been produced from the nitric acid. In the higher
temperature/pressure reactions, all of the atmospheric constituents were not accounted
for by expected breakdown products of the explosives. Bromine was present in the
atmospheres of these samples and was probably the major portion of the missing
constituents.
The results of the wet-air oxidation treatment of nitrocellulose sediment are
presented in Table 28. A treatment of 200°C/600 psig for 90 minutes reduced the
nitrocellulose or nitrate levels in the sediment by 96%. COD levels were also lowered
by a similar amount at these conditions.
The data on the percent reduction of nitrocellulose in the sediment were
transformed into log form. An ANOVA table was then constructed. Treatment
pressure/temperature were not significant, but treatment time was significant at the
0.15% level. In other words, the lowest temperature/pressure was as effective in
degrading nitrocellulose as the higher temperature/pressure scheme. Further evaluation
of the data by Duncan's Multiple Range Test showed that the time 0 and 60 or 90
minute data were significantly different at the 5% level of probability but the 90
minute data were not significantly different from the 60 minute data.
The air analysis data for the nitrocellulose sediment treatments are presented
in Table 29. Oxygen was initially present in the reactor. As treatment temperatures
increased, the amount of CO2 increased. CO% levels for the four treatment conditions
ranged from 10.9 - 14.5 percent of the total gas volume.
60
-------
Table 27. Off-Gas Analysis for Wet-Air Oxidation of TNT/RDX
Sediment Slurries
Percent
Treatment
(°C/psig)
200/600
200/600
225/750
225/750
250/980
250/980
Time
(min)
60
90
60
90
60
90
C02
12.60
19.85
29.10
31.97
28.86
34.13
°2
79.71
65.40
46.37
26.39
46.82
43.21
N2
6.54
12.56
13.65
10.86
13.62
11.04
CO
0.1
0.1
0.1
0.4
0.1
0.1
Summation
(percent)
98.95
97.91
89.22
69.62
89.40
88.48
-------
Table 28. Wet-Air Oxidation of Nitrocellulose Sediment
Remaining After Treatment
Temperature (°C)/
Pressure (psig)
Control
165/450
165/450
200/600
200/600
Treatment Time
(min)
0
60
90
60
90
Nitrocellulose
Mg/g
59,800 f 12,800
3,250 + 500
2,280 + 260
2,320 + 540
2,250 + 180
COD Sediment
Wg/g
163,500
NA
NA
32,300
3,400
COD Liquid
ug/ml
19,200
NA
NA
255
260
NA - Not Analyzed
-------
Table 29. Off-Gas Analysis for Wet-Air Oxidation of Nitrocellulose Sediment Slurries
Percent
Treatment
(°C/psig)
165/450
165/450
200/600
200/600
Treatment
Time (min)
60
90
60
90
CO2
10.91
13.18
14.05
14.50
<*
79.54
81.23
78.87
70.56
N2
3.29
5.13
6.29
16.37
CO Summation
.30 94.04
.14 99.68
<.10 99.31
<.10 101.53
o>
CO
-------
D. Conclusions
Wet-air oxidation of the TNT/RDX sediment was effective. The treatment did
reduce RDX, TNT and tetryl levels in the sediment to below GC detection limits.
GC/MS analysis of the treated slurry water indicated that the toxic compounds
trinitrobenzene and dinitrobenzene were present. Relatively high and long treatment
conditions (250°C/980 psig) were required for 99% disappearance of the trinitrated ring
structure.
Nitrocellulose levels in the sediment were also significantly reduced by wet-air
oxidation. However, the increase in treatment temperature and time did not seem to
increase the destruction of the nitrocellulose.
E. Future Work
While wet-air oxidation is an effective treatment method for RDX deconta-
mination, it is not a totally acceptable method for use on TNT containing sediments
because of the toxic by-products formed. Destruction of nitrocellulose by the process
was not complete. In addition, both the capital and operating costs for wet-air
oxidation are extremely high, as can be seen in the next section. For these reasons,
wet-air oxidation is not recommended for further study in the decontamination of
lagoon sediments.
F. Economic Analysis
Capital and operating costs for wet-air oxidation of 3,800 kg/hr of a 10% solids
lagoon sediment slurry are presented in Tables 30 and 31. This treatment rate is the
same as that in the standard lagoon scenario except that water has been added to
facilitate operation of the wet-air oxidation equipment. Costs were taken from a
report prepared by IT-Enviroscience (1981) at the end of the experimental work.
64
-------
Table 30. Capital Costs for Wet-Air Oxidation
Equipment Cost Reference
Dredge $ 58,800 Salemink, 1980
Holding Tank 145,600 Peters and Timmerhaus,
1968
Wet-Air Unit 8,262,000 IT-Enviroscience, 1981
TOTAL $8,466,400
Table 31. Annual Operating Costs for Wet-Air Oxidation
(IT-Enviroscience, 1981)
Maintenance (5% of capital) $ 413,000
Labor (3 operators, 1 supervisor) 210,000
Fuel Oil (490 Btu/lb @. $9/M Btu) 270,000
Cooling Water (L86 gal/lb @ $25/1000 gal) 133,000
Electricity (.05 kwh/lb @ $0.07/kwh) 212,000
HBr (.005 Ib/lb @. $0.09/lb) 236,000
HNOs (.07 Ib/lb @ $0.09/lb) 381,000
NaOH (.05 Ib/lb @ $0.06/lb) 181,000
TOTAL $3,036,000
65
-------
VI. ACETONE EXTRACTION
A. Process Description
Solvent extraction or leaching is the process of separating soluble compounds
from insoluble solids by bring. _: the solids into contact with an appropriate solvent.
The soluble compounds dissolve and the solution is then mechanically separated from
the remaining solids. Additional process steps may be added to remove the solute from
solution or to recover a solvent which is either undesirable in the finished product or
is too expensive to waste. Important factors influencing the effectiveness of solvent
extraction processes are the ease with which the solvent can reach the solute and the
ease of separation of desired products from any remaining solvent.
Acetone extraction of explosives from lagoon sediment may be an economical
and effective decontamination method. As indicated in Table 32, all of the explosives
of interest to this study are either somewhat soluble or easily dispersible in acetone
at room temperature. Extraction of the explosives should be efficient and safe.
Residual acetone could be removed for recovery by gentle heating. Since the normal
boiling point of acetone is well below the detonation temperatures of the explosives,
most of the acetone could be recovered by boiling the mixture and then condensing the
vaporized acetone. Some acetone would be allowed to remain with the explosives to
maintain them in a wet state to reduce the danger of accidental detonation. It is
essential that the impact and heat sensitivities of the explosives in question when
mixed with small amounts of acetone be carefully examined before such a solvent
recovery scheme is attempted. The remaining acetone and extracted explosives could
be incinerated to ensure complete destruction. Heat from the incineration could be
used to power the solvent recovery steps.
As with most mass transfer processes, continuous countercurrent exchange is
the most efficient solvent extraction method in terms of equipment, energy and raw
materials requirements. For solvent extraction, continuous countercurrent ex-
change requires several contacting stages in which solvent and solid are mixed, then
separated. Solids move from stage to stage with a little more solute being extracted
at each stage. Fresh solvent is introduced at the opposite end of the process so that
clean solvent contacts nearly solute-free solids in the last stage. Solvent from the last
stage is fed to the next-to-last stage, and so on from stage to stage until the most
highly solute-laden solvent contacts the untreated solids in the first stage. Counter-
current extraction can produce an essentially solute-free solid stream and a solute-
saturated solvent stream, thus minimizing solvent requirements.
Solid-liquid solvent extraction has been used for a wide variety of separation
processes. A brief review of several types of solvent extraction equipment and their
possible applications to decontamination of lagoon sediment is presented in the
following paragraphs.
66
-------
Table 32. Solubility of Explosives in Acetone (Departments of
the Army and Air Force, 1967)
TNT 0°C 57 g/100 g solvent
20°C 109 g/100 g solvent
25°C 132 g/100 g solvent
30°C 156 g/100 g solvent
50°C 346 g/100 g solvent
RDX 20°C 7.4 g/100 g solvent
50°C 12.8 g/100 g solvent
Tetryl very soluble in acetone
Nitrocellulose not truly soluble, dispersed to
colloidal form in acetone
-------
One of the simplest solvent extraction schemes is the extraction battery or
Shanks system (Treybal, 1955). Each stage is a percolation tank which is filled with
solid through which the solvent can trickle to the bottom of the tank. Solvent draining
from the bottom of each tank is pumped to the top of the next tank. At any given
time, one tank will be out of operation for emptying of solute-free solids and refilling
with fresh solids. In the next step, the tank of fresh solids will receive the most
concentrated solution while the tank which received fresh solvent in the last step is
emptied and refilled. The process continues in this fashion until the first tank emptied
is again ready to be emptied, and the cycle begins again. A typical operation of a five
tank extraction battery is shown in Figure 13.
A more sophisticated version of the same type of percolation scheme is shown
in Figure 14. Designed for the extraction of vegetable seeds such as soybeans or
linseeds, the Rotocel consists of 18 percolation cells which revolve about a stationary
compartmented tank (Treybal, 1955). Solids are continuously loaded, countercurrently
contacted with solvent, then unloaded through hinged screens in the bottom of each
celL A complete cycle in the extraction battery corresponds to one revolution of the
Rotocel. The chief advantage of this scheme is the continuous addition and removal of
solids.
Finely ground solids which are readily suspended can be leached in a series of
alternating agitated vessels and mechanical separators such as centrifuges or
thickeners. Each agitator and separator pair is a single stage, and countercurrent
operation is again the most efficient method of operation. Many types of agitators
have been used, but the turbine type mixer is generally the most suitable (Treybal,
1955). Continuous centrifuges are useful for those solids which are easily separated
from the solution, while thickeners are used for more difficult separations, typically for
very fine solids in dilute suspension. A thickener manufactured by Dorr Oliver is shown
in Figure 15.
Any of these three types of equipment might be useful for solvent extraction
of explosives from lagoon sediment. Effectiveness of the process and economic
considerations will determine which is most suitable.
B. Experimental Procedure
The first step in examining solvent extraction is to determine whether the
process is effective at removing explosives from contaminated lagoon sediment. Several
temperatures and extraction times for sediments contaminated with TNT, RDX and
tetryl from LAAP Lagoon 9 and with nitrocellulose were examined for a single stage
extraction according to the following procedure.
Approximately 10 grams of dry contaminated sediment were placed in a 250 ml
round bottom flask outfitted with a reflux condenser. Acetone (150 mL) was added to
the flask with continuous agitation to ensure good contact between the acetone and
sediment. The flask was placed in a water bath for the required time interval to
maintain a constant temperature. After the proper time period, the 250 ml round
bottom flask was quickly removed from the water bath and the contents filtered to
separate the sediment and solvent. A sample of the solvent was placed in a screw cap
culture tube while the entire sediment filter cake was recovered with the filter paper
and allowed to dry at room temperature.
68
-------
Fresh Solid
cr>
CO
1
L Solvent
•-'•'-•• • • ••
^«^___
,
<^
-^
i
Solut
Sol
,
.'.'• '•''•
i
s
e-Free
id
Fresh
t Solvent
c ^
• t t
• . . •
Soluto-F
Sol id
i
*\V *.."• !•".*•* *.-•!*• "• •
• *.".*."•"•.' * . * i. * • •*
.".*. *- •**•* ~ *."•*••*!••
\^_ _^
1
Saturated
Sol vent
{
f ^*x
rep
;
v^^
Fresh
Solid
^
' ** x " '•'•
.* •' " "*
,
• • J • ; .
• "" * * "•
r
' • *•"«*•* "
:" ' ' * : "•
'" *.•*.*•
Saturated
Sol vent
11
-^"**-1 "^*
V
*• *' •" - ' . '• '• " ,
Two Si ens in the Oncr.-il inp Cvrlr oT ,-ni h'.x I i-.ir i i .-in K/I
-------
Solids
t
Solvent and
Solute
Rotating Cells
Leached
Solids
Solvent
Figure 14 . Rotocel Solvent Extractor
70
-------
TOP VIEW
SIDE VIEW
Figure 15. Dorr Thickener
71
-------
The weights of the sediment before and after treatment were recorded along
with the amount of solvent recovered. Samples of the solvent and the dry
sediment were analyzed for TNT, RDX and tetryl or for nitrocellulose by the methods
described in Appendix A.
One set of serial extraction experiments was conducted at room temperature
according to the following procedure. Approximately 10 grams of dry TNT/RDX/tetryl
contaminated sediment were weighed into each of two 250 ml Erlenmeyer flasks.
Acetone (150 ml) was added to each flask and the flasks were sealed with plastic film.
One flask was shaken for 15 minutes while the other was shaken initially and then
allowed to stand for the same length of time. Both flasks were unsealed, and the
solvent was decanted off leaving only the sediment and a small amount of solvent in
the bottom of each flask. Fresh acetone (150 ml) was added to each flask and the
entire process was repeated. A third 150 ml aliquot of fresh acetone was added to each
flask and shaken or allowed to stand for 15 minutes, then the entire contents of the
flasks were filtered. The sediment filter cakes were allowed to air dry. Volumes of
solvent recovered from each extraction were recorded, together with the initial and
final weights of the sediment. Extracts and sediment samples were analyzed for TNT,
RDX, and tetryl by the same methods as before.
C. Results
The results of the single stage acetone extraction experiments using
TNT/RDX/tetryl contaminated lagoon sediment are presented in Table 33. The data
shown are the average data for two runs at each set of experimental conditions. Raw
data for these experiments are presented in Table C-2 of Appendix C. With a solvent
to dry sediment weight ratio of approximately 12:1 (15 ml acetone/g sediment),
extraction was essentially complete for all of the explosives at all of the experimental
conditions. Since only a very small amount of explosives was detected after extraction
at 25°C for 15 minutes, there was little room for improvement at the higher
temperatures and longer extraction times. Explosives recoveries in acetone averaged
70% for TNT, 83% for RDX and 97% for tetryl. Possible reasons for incomplete
recovery of TNT and RDX are filtration and handling losses and variation in the
explosives levels in the original sediment.
The results of two room temprature serial extractions using Lagoon 9 sediment
are presented in Table 34. Only 0.1% of the original TNT and none of the RDX or
tetryl was detected in the third extraction of the shaken sample. The sample which was
not shaken showed a somewhat lower recovery in the first extraction and essentially
complete removal of explosives in the second extraction. Overall recoveries of
explosives were higher than in the previous experiments. This higher recovery was
probably because a smaller amount of sediment was mixed and analyzed for the set of
tests, thus minimizing variation in the inital explosives concentration.
72
-------
Table 33. Acetone Extraction of TNT/RDX/Tetryl Contaminated
Sediment
Temperature
<°r)
25
25
25
50
50
50
75
75
75
Time
(min)
15
30
60
15
30
60
15
30
60
TNT
in sediment
(% original TNT)
0.6
0.8
0.7
0.4
0.5
0.7
0.4
0.3
0.3
TNT
in Acetone
(% original TNT)
71
72
71
69
62
67
78
68
69
RDX
in Sediment
(% original RDX)
<0.9
<0.9
<0.9
<0.9
<0.9
<0.9
<0.9
<0.9
<0.9
RDX
in Acetone
(% original RDX)
84.5
85.5
86.6
81.2
73.1
81.7
83.6
85.1
85.1
Tetryl
in Sediment
(% original tetryl)
<0.7
<0.7
<0.7
<0.7
<0.7
<0.7
«;0.7
<0.7
<0.7
Tetryl
in Acetone
(X original tetryl)
1(14
105
104
87
80
94
106
95
99
Original Explosives levels in Sediment
TNT: 205,000 Mg/g
RDX: 52.600 ug/g
Tetryl: 10,500 (jg/g
-------
Table 34. Serial Acetone Extraction
Shaken
Not Shaken
TNT RDX Tetryl
fraction (% explosives in acetone) Extraction
First
Second
Third
TOTAL
87.5
2.2
0.1
89.8
97.4
0.9
0
98.3
95.7 First
0 Second
0 Third
95.7
TNT RDX Tetryl
(% explosives in acetone)
78.5
3.3
0.1
81.9
88.2
4.0
0
92.2
99.4
0
0
99.4
Original Explosive Levels in Sediment:
TNT: 192,700 pg/g
RDX: 40,000 Mg/g
Tetryl: 12,700 pg/g
-------
The results of single stage acetone extraction experiments using nitrocellulose
contaminated sediment are presented in Table 35. Each entry represents the average
value for two runs and the raw data from which this table was generated is in Appendix
C as Table C-3. Extraction efficiencies were not as high as for the other explosives,
but they were high enough to indicate that acetone extraction may be a feasible
method of removing nitrocellulose from soiL The data show that increasing temperature
improves the extraction efficiency. Nitrocellulose recoveries in the acetone ranged
from an average of 27% at 25°C to an average of 73% at 75°C. Increasing contact
time did not produce an improvement in extraction efficiency, and in fact the amounts
of nitrocellulose extracted showed a tendency to be somewhat lower at longer contact
times. Since nitrocellulose does not form a true solution in acetone, this reduced
efficiency is probably the result of nitrocellulose falling out of suspension upon standing
after the initial agitation.
D. Conclusions
Acetone extraction is a very effective method of removing TNT, RDX and
tetryl from lagoon sediments. Extraction efficiencies expressed as percent explosives
recovered in acetone for TNT, RDX and tetryl were 70%, 83% and 97% respectively.
The results indicate that acetone extraction of these explosives is relatively easy. Only
a few contacting stages will be required for essentially complete extraction.
Acetone extraction of nitrocellulose from lagoon sediment is not as effective
as extraction of the other explosives, however, the process is still technically feasible.
More contacting stages would be required for nitrocellulose extraction, and care should
be taken in the equipment design to ensure that nitrocellulose is not allowed to
redeposit on the sediment.
Commerical extraction equipment is currently available which could be used for
decontamination of lagoon sediment. The extraction battery and the Rotocel apparatus
are likely to be useful in this application. While flammable solvents such as acetone
are avoided where possible in solvent extraction processes, such solvents have been
used successfully in the past. Technically, acetone extraction of all of the explosives
considered is a rapid and effective method of explosives decontamination.
E. Future Work
Since acetone extraction proved to be an effective method of removing
explosives from lagoon sediment, some further laboratory tests followed by a pilot
scale demonstration of the process are recommended. Before a demonstration unit can
be designed, it is important to experimentally determine the equilibrium curves relating
the concentration of explosives in the extract, the concentration of explosives and the
amount of solvent associated with the solids after separation, and the amount of
explosives free solid before and after the extraction step. This equilibrium data will
provide the basis for determining the number of countercurrent equilibrium stages
required for the process.
75
-------
Table 35. Acetone Extraction of Nitrocellulose Contaminated Sediment
Nitrocellulose
Nitrocellulose
CD
Temperature
(°C)
25
25
25
50
50
75
75
75
Time
(min)
15
30
60
15
30
15
30
60
in Sediment
(96 original nitrocellulose)
44.2
36.9
75.3
64.7
37.4
15.6
12.9
20.1
in Acetone
(°/-. original nitrocellulose)
27.9
24.0
30.3
38.6
61.5
83.8
71.7
62.5
Original Level in Sediment
Nitrocellulose: 73,400 pg/g
-------
To control explosives levels and to provide concentrations over the entire range
of interest, sediment samples for the equilibrium .experiments should be spiked with
explosives. Since TNT and RDX are found at higher levels in sediment and are less
soluble in acetone than tetryl, only the equilibrium curves for TNT and RDX need be
determined individually. Another set of curves should be found for TNT, RDX and
tetryl together. Nitrocellulose equilibrium curves should be determined separately.
The experimental procedure and proposed experiments to obtain equilibrium data are
described below.
A ten gram sample of spiked sediment is weighed and placed in a 150 ml
Erlenmeyer flask. The required volume of acetone is added, and the mixture is shaken
for ten minutes. The mixture is vacuum filtered on weighed filter paper, the volume
of acetone recovered is measured, and the sediment with filter paper is weighed wet.
The sediment is then allowed to air dry and is reweighed. Extract and sediment are
analyzed for the appropriate explosives.
The experimental conditions which are to be used to generate the equilibrium
curves for TNT/RDX/tetryl containing sediment and for nitrocellulose containing
sediment are listed in Tables 36 and 37, respectively. The complete experimental
program for TNT/RDX/tetryl containing sediment outlined thus far would require a
total of 96 experiments to determine equilibrium curves for each explosive individually
and 864 experiments to find the curves for all of the explosives together. Running
these experiments in duplicate boosts these figures to 200 and 1700 experiments
respectively. Since 1700 experiments are far too many to be practical for the current
program, the earliest experiments should be for one acetone volume and two explosives
levels for each explosive, both individually and in tandem. If statistical analysis shows
that the results for each explosive are not significantly different when run alone or in
tandem, then the individual experiments should be sufficient and the tandem
experiments should be dropped. If the tandem experiments are required, the total
number of experiments should be reduced to give a more manageable experimental
program. In addition, for all experiments, if any acetone volume reduces an explosive
level to its detection limit, all experiments using larger volumes of acetone for that
explosives level should be dropped. While such a program does not allow for sample
randomization, the possibilities for greatly reducing the total number of samples, and
therefore the costs, outweigh any loss in statistical reliability. While the nitrocellulose
should be tested alone, the same procedure of throwing out high volumes of acetone
if the explosive is not detected in the sediment should be used.
Two possible process difficulties which should be addressed in the laboratory
phase of this study are the effects of water on the extraction efficiencies and the
hazards of crystallized explosives under acetone. Before the equilibrium experiments
are conducted, a series of five experiments should be run to establish whether moisture
in the sediment has any significant effect on the extraction. For one acetone volume
and one relatively high level of each explosive (in tandem), experiments should be done
in which the only variable is the sediment moisture level. If no significant variation is
observed, the experimental program should be continued as planned. If water has a
significant effect on the extraction, then the program will have to be redesigned to
take this factor into account. Explosives hazards of the sediment extracts should be
determined by boiling the acetone containing explosives down to approximately 10% of
its original volume, then impact testing the crystallized explosives and acetone
mixture.
77
-------
Table 36. Experimental Conditions to Establish TNT/RDX/Tetryl
Equilibrium Curves
Acetone Volumes
5 ml
10 ml
20 ml
50 ml
75 ml
100 ml
TNT Levels
0.5%
1%
5%
10%
25%
50%
RDX Levels
0.5%
1%
5%
10%
20%
30%
Tetryl Levels
0.5%
1%
5%
10%
Table 37. Experimental Conditions to Establish Nitrocellulose
Equilibrium Curves
Acetone Volumes Nitrocellulose Levels
5 ml 0.5%
10 ml 1%
20 ml 5%
50 ml 10%
75 ml 20%
100 ml
78
-------
At this point, it is necessary to determine whether a bench scale simulation or
pilot scale tests are needed. Bench scale simulation using an extraction battery would
be relatively simple to set-up. A separate process vessel is required for each stage as
determined by the equilibrium data. Each vessel would need a perforated support for
the sediment, a solvent distributor at the top of the vessel, and a collection system and
pump at the bottom. Each vessel must also be easily opened to remove treated
sediment. A continuous still at the end of the process line collects waste explosives
under acetone. The distilled acetone is reused in the process. Incineration of the
waste explosives should not be a part of this simulation. Explosives can be disposed
of by incineration performed by Atlantic Research Corporation's Propellent Division.
Information to be gained from a system on this scale using actual contaminated
sediment includes solvent percolation rates, minimum amounts of acetone required for
desired extraction, and demonstration of a semi-continuous process.
If a pilot scale demonstration is recommended, the options are for Atlantic
Research to build or purchase a solvent extraction unit or to use the pilot equipment
of a manufacturer. In either case, equipment must be added to the system to dispose
of the extracted explosives immediately. This disposal can be accomplished either with
a liquid incinerator to burn the entire extract or with a solvent recovery still followed
by an incinerator to burn the crystallized explosives. If this latter option is chosen,
it may prove more cost effective to skip the pilot scale demonstration and purchase
the equipment directly for a field demonstration. In this case, it might be possible to
use existing incineration facilities at the installation to dispose of test products.
F. Economic Analysis
The capital and yearly operating costs for acetone extraction are presented in
Tables 38 and 39. The capital costs were obtained from the references shown.
Necessary adjustments for changing economic conditions were made with the use of the
appropriate chemical engineering plant cost index. The total capital cost in 1981 dollars
for this system is $383,300 and the yearly operating costs are $394,400. These costs are
on the same order as those for the incineration system. A major factor contributing
to the relatively high operating costs is the cost of the acetone.
79
-------
Table 38. Capital Costs for Acetone Extraction
Quantity
Equipment
Installed Cost
Reference
oo
o
5
5
1
1
Dredge: Porta-Dredge PD-4LS-
1136 1/min
Holding Tank: 454,200 1
(120,000 gal) w/75 hp side
entry turbine
Slurry Pump: 1/2 hp
Flat Bed Trailer
Incineration System (includes
feed system, incinerator, after
burner and/or scrubber) (50 gal/
hr)
Percolation Tank: 470 1 (124 gal)
Pump: 12 1/min (3 gal/min)
Still: 148 1 (40 gal)
Heat Exchanger (Shell and Tube)
4 rn2 (44 ft2)
$ 58,800
145,600
880
11,030
150,000
7,100
3,640
5,300
940
Salemink, 1980
Peters and Timmerhaus, 1968
Peters and Timmerhaus, 1968
Peterbilt, 1980
Met-Pro, 1981
Peters and Timmerhaus, 1979
Peters and Timmerhaus, 1979
Peters and Timmerhaus, 1979
Popper, 1968
TOTAL CAPITAL COST
$ 383,290
-------
Table 39. Acetone Extraction Annual Operating Costs
Labor (3 operators @ $45,000/man year $ 210,000
(1 supervisor @ $75,000/man year)
Maintenance (3% of total capital) 11,500
Electricity
(4000 kwh/yr @ $0.07/kwh) 300
Cooling Water
(45.8 x 106 liter/yr @ $0.0264/1000 1) 1,210
•— Acetone
(210,650 kg/yr @ $0.60/kg) 126,400
TOTAL ANNUAL OPERATING COSTS $ 394,410
-------
VH. WATER EXTRACTION
A. Process Description
Water extraction is a solvent extraction process using a cheap, non-flammable
solvent. Since TNT, RDX and tetryl are not very soluble, fairly large volumes of water
are required to dissolve the explosives. In addition, the water is heated to increase the
solubility. Nitrocellulose is so insoluble in water that this process cannot be applied
to removing it from sediment. If temperatures greater than 100°C are desired, the
process must include a high pressure chamber in which the extraction is carried out.
The process is operated on a cycle in which the water and contaminated sediment are
heated together, then separated. The extracted sediment proceeds to another
extraction stage or is discharged. The explosives containing water is then cooled
causing the explosives precipitate out of solution. The water is then reheated and
returned to the extraction chamber.
The equipment in which the extractions are performed will be similar to that
used in acetone extraction except that the equipment must be heated and must be able
to withstand the appropriate pressures if temperatures higher than the normal boiling
point of water are desired.
B. Experimental Procedure
Atmospheric water extraction experiments were conducted in 250 ml round
bottom flasks equipped with water cooled reflux condensers and heated by heating
mantles. LAAP Lagoon 9 sediment samples weighing approximately 10 g were heated
with 100 ml of water under reflux for one hour. The extract was decanted off, and
the remaining sediment was air dried and analyzed for TNT, RDX and tetryl.
High pressure extractions were conducted in the apparatus shown in
Figure 16. The reactor consists of two pieces of stainless steel pipe joined with high
pressure flanges and sealed at each end with a welding cap. The faces of the flanges
have been machined to hold in place a piece of 50 mesh wire screen which separates
the halves of the reactor. A thermocouple is inserted into one end of the tube. The
opposite end has a pressure gage, a pressure relief valve, and a valve to allow pressure
release or drainage when desired. The thermocouple output, is monitored by a strip
chart recorder and a temperature controller. The temperature controller drives a 600
watt band heater which is placed on the thermocouple half of the reactor.
In the experiments, 10 g samples of LAAP Lagoon 9 or Lagoon 11 sediment and
100 ml of water were placed in the thermocouple half of the reactor. The screen was
put on top of the flange, and the other half of the reactor was bolted on with the gage
and valves upward. The reactor was then purged for several minutes with nitrogen to
prevent oxidation of the explosives to ensure that extraction is the only method by
which explosives are removed. The controller was set, and the reactor was allowed to
come to temperature and remain at temperature for one hour. When the experiment
was over, the heater was shut off and the reactor turned upside down and allowed to
cool for several hours. This maneuver trapped the sediment on the screen while
allowing the water extract to drain through. When the reactor was cool, the water and
crystallized explosives were drained into a container of acetone to dissolve the
explosives. The reactor was taken apart, and the sediment was removed and air dried.
The sediment was then analyzed for TNT, RDX and tetryl. The water extracts could
not be analyzed for explosives because the water in the solution would damage the gas
chromatograph column used in this analysis.
82
-------
00
Figure 16. Water Extraction Apparatus
-------
C. Results
In the atmospheric pressure experiments, the high level explosives sediment
formed a scum over the top of the liquid as the sample was heated. Boiling broke up
the layer, and spattered the scum on the sides of the flask. This scum presumably
contained a significant fraction of the TNT which melted and floated on the surface
of the liquid rather than dissolved in the solution.
Results of the water extraction experiments are presented in Table 40. Each
entry represents the average results of two experiments. The atmospheric extraction
shows excellent explosives removal, however, this is probably the result of pouring off
the finer portion of the sediment in the decanting procedure and leaving sandy soil.
More reliable results were obtained in the high pressure reactor where rapid filtering
at the treatment temperature was possible. Significant reduction in the sediment
explosives levels was observed at both 150°C and at 200°C. The extraction efficiency
was quite high at 200°C and a complete extraction could probably be accomplished in
only two stages at this temperature.
D. Conclusions
Water extraction is an effective method of removing TNT, RDX and tetryl
from sediment at 200°C. However, if TNT is present in excess, the amount that is not
dissolved will melt and float on the liquid. A layer of melted TNT floating on the
liquid could pose serious combustion hazards. If this is the case, then the non-
flammability of water which was supposed to be one of the main advantages of this
method is no longer important. The requirements for a hot extraction process and for
high pressure to achieve an efficient extraction temperature dictate that specialized
equipment be designed. Finally, the technology and equipment for this process have not
yet been developed.
E. Cost Analysis
It is difficult to arrive at a reliable cost estimate for the water extraction
process because the equipment required has not been developed. There are several
factors inherent in this process which have the potential of driving equipment costs up.
Primary among these are the operating conditions of the process. The equipment must
be designed for the generation of high temperatures. In addition, the equipment must
withstand high pressures. Equipment designed to meet these requirements is much
more costly than equipment designed for operation at standard conditions such as those
at which the acetone extraction system is operated. Therefore, it is apparent that
capital costs for water extraction will surpass those for acetone extraction.
F. Future Work
Further work in the area of water extraction of explosives contaminated
sediment is not recommended. There are several factors which make this process
84
-------
Table 40. Water Extraction
oo
01
Temperature
°c
100
149
204
Explosives
Level
High
Low
Low
Percent Explosives Remaining
TNT RDX
1.0 4.8
50.5 6.3
2.4 <4.1
in Sediment
Tetryl
-
31.1
8.2
Original Explosives Levels in Sediment
High Level Low Level
TNT: 385,300 pg/g 39,200 pg/g
RDX: 80,000 pg/g 12,000 pg/g
Tetryl: 25,400 pg/g 3,800 pg/g
-------
unattractive for this application. One of the most important factors is the safety
uncertainty of the process. As mentioned previously, a potential safety hazard exists
as a result of the layer of molten TNT in the heated pressure chamber. Another
problem with this process is the potentially high equipment cost due to the extreme
operating conditions. Finally, a problem exists in that the technology and equipment
have not been developed. For the above reasons, water extraction does not compare
favorably to acetone extraction.
86
-------
Vffi. CONCLUSIONS AND RECOMMENDATIONS
Laboratory scale experiments on various treatment methods for decontamina-
ting lagoon sediment have been conducted to determine the effectiveness of each
process. The techniques evaluated were incineration, gamma irradiation, wet-air
oxidation, acetone extraction and water extraction.
Incineration quickly and effectively destroyed explosives in lagoon sediment.
TNT, RDX, tetryl and nitrocellulose are all oxidized to environmentally acceptable
gases when combustion is complete. Metals in the sediment could be a potential
problem. There are two potential alternatives to the metals problem depending on the
metal involved. First operating temperatures and dust emissions can be controlled so
that very little of the metals are removed from the sediment. The metal concentrations
in the residual sediment will, of course, increase due to losses of organics. The
resulting sediment may have to be placed in a secure landfill if final metal levels are
higher than allowed. Another alternative is to deliberately vaporize any volatile metals
and collect them in a baghouse. If the sediment from this type of operation has less
than the maximum allowable concentration of metals, it could be returned to the
lagoon thus saving landfill costs. However, a penalty in fuel costs for operating
conditions could vary from lagoon to lagoon depending on the wastes found in the
sediments. Incineration is recommended for pilot-scale study to optimize system
parameters for the burning of actual explosives contaminated lagoon sediment.
Gamma irradiation at a dose of 4.1 megarads degraded approximately 30% of
the TNT and RDX in sediment. The same treatment was partially effective for
nitrocellulose in sediment; however, this effectiveness could not be quantified because
of interference with the analytical method. Gamma irradiation will require high dosage
levels and long retention times to affect complete decontamination of the explosives
in the sediment. A large number of unknowns is associated with this process for
destruction of chemicals, therefore, further studies of gamma irradiation for treatment
of highly contaminated explosives sediments are not recommended.
Wet-air oxidation of sediments containing TNT, RDX and tetryl reduced these
explosives to below the detection limits at all reaction conditions. Toxic 1,3,5-
trinitrobenzene and 1,3-dinitrobenzene as well as bromoinated nitrobenzenes were
observed in the aqueous effluent. High temperatures and pressures (250°C/980 psig)
were required to degrade 99% of the trinitrated benzene ring structure. Wet-air
oxidation of nitrocellulose produced 95-96% reduction in nitrocellulose levels. Due to
the potential toxic by-products and high costs, wet-air oxidation is not recommended
for further study.
Acetone extraction effectively removed TNT, RDX and tetryl from lagoon
sediment. Approximately 27% of the nitrocellulose was extracted. While a single stage
extraction did not completely remove all explosives, solvent extraction is usually
conducted as a multi-stage operation to obtain complete extraction. The low recovery
of nitrocellulose means it will require more stages than will TNT, RDX and tetryl
sediment. Additional acetone extraction experiments followed by a somewhat larger
scale demonstration are recommended as the next steps in the investigation of this
process.
87
-------
Water extraction at high temperatures and pressures proved to be a moderately
effective means of removing TNT, RDX and tetryl from sediment. The heating, cooling
and high pressure environment required by this process will greatly increase capital and
operating costs over those required for acetone extraction. In addition, the much lower
solubilities of the explosives in water, even at high temperature, will require a larger
system to contain the liquid volume. Finally, the floating of molten TNT on the
surface of the water could pose a serious safety hazard. For these reasons, water
extraction is not recommended for further study.
Of the five process tested, only incineration and acetone extraction produced
effective decontamination of explosives containing sediments with no major process
complications. Incineration equipment is readily available, and pilot testing of this
process should be top priority. Acetone extraction has not been used in this
application, but existing solvent extraction equipment should be usable without major
modifications. Through the use of these technologies, it is expected that relatively
cost effective decontamination of lagoon sediments can be accomplished with only a
fairly small process development effort.
88
-------
IX. REFERENCES
C-E Raymond (1980), "Bartlett-Snow," Bulletin 793.
Conway, R.A. and Ross, R.D. (1980), Handbook of Industrial Waste Disposal, Van
Nostrand Reinhold Company, New York.
Dawson, G.W. (1978), "Appendix A to the EPA Kepone Mitigation Project Report: The
Feasibility of Mitigating Kepone Contamination in the James River Basin,"
BatteUe, Pacific Northwest Laboratory. NTIS, PB 286 085.
Department of the Army and the Air Force (1967}, Military Explosives, Technical
Manual No. 9-1300-214, November 28, 1967.
EPA (1979), Process Design Manual - Sludge Treatment and Disposal, Environmental
Protection Agency, Cincinnati, Ohio, EPA 628/1-79-01L
IT-Enviroscience (1981), "Evaluation of Catalyzed Wet Air Oxidation for Treating Army
Lagoon Sediments," unpublished report submitted to Atlantic Reerch Cor-
poration.
Met-Pro Corporation (1980), Personal Visit, Harleyville, Pennsylvania.
Morris, M.; Sivinski, J.; Brandon, J.; Neuhausser, K. and Wood, R. (1979), "A Summary
of Recent Developments in the Sludge Irradiation Program at Sandia Lab-
oratories," Sandia Laboratories, Albuquerque, New Mexico.
Peterbilt (1980), Personal Communication, Landover, Maryland.
Peters, M.S. and Timmerhaus, K.D. (1968), Plant Design and Economics for Chemical
Engineers, 2nd edition, McGraw-Hill Book Company, New York.
Peters, M.S. and Timmerhaus, K.D. (1979), Plant Design and Economics for Chemical
Engineers, 3rd edition, McGraw-Hill Book Company, New York.
Popper, H. ed. (1970), Modern Cost Engineering Techniques, McGraw-Hill Book
Company, New York.
Salemink, W. (1980), Personal Communication, Letter, Assemblers Inc., West Liberty,
Iowa.
Scurlock, A.D.; Lindsey, A.W.; Fields, T. Jr. and Huber, D.R. (1975), "Incineration in
Hazardous Waste Management," Environmental Protection Agency,
EPA/530/SW-141. NTIS, PB 261 049.
Stribling, J.B. (1972), "Sludge Incineration By Cyclone Furnace," Effluent and Water
Treatment Journal, 12(8): 395-400.
89
-------
Treybal, R.C. (1955), Mass-Transfer Operations, McGraw-Hill Book Company, New York.
Wentsel, R.S.; Sommerer, S. and Kitchens, J.F. (1982), "Engineering and Development
Support of General Decon Technology for the DARCOM Installation Restoration
Program. Task E: Treatment of Explosives Contaminated Lagoon Sediment
- Phase I. Literature Review and Evaluation," Atlantic Research Corporation.
Wentsel, R.S.; Wilkinson, M.S.; Hyde, R.G.; Harward, W.E.; Jones, W.E. and Kitchens,
J.F. (1979), "Problem Definition Study on 1,3-Dinitrobenzene, 1,3,5-Trinitro-
benzene and Di-n-Propyl Adipate," Contract No. DAMD17-77-C-7057, U.S.
Army Medical Research and Development Command, Fort Detrick, Frederick,
Maryland.
90
-------
APPENDIX A. ANALYTICAL METHODS
A-l
-------
Appendix A
Analytical Methods
Analysis of High Levels of TNT and RDX in Sediment - Quantitative
Determination of Nitrocellulose in Sediment - Quantitative
Lead and Cadmium in Water - Quantitative
Chromium and Zinc in Water - Semi-Quantitative
Lead and Cadmium in Sediment - Quantative
Chromium and Zinc in Sediment - Semi-Quantitative
Analysis of Nitrate-N in Sediment - Semi-Quantative
Analysis of Nitrate-N in Water - Semi-Quantative
Analysis of Low Levels of TNT, RDX and Tetryl in Sediment -
SemirQuantative
A-2
-------
ANALYSIS OF HIGH LEVELS OF TNT AND RDX IN SEDIMENT - QUANTITATIVE
ARC METHOD NO. 1 for
DECON Technology DAAK11-80-C-0027
1. APPLICATION
Method used to determine the concentration of TNT and RDX in sediment.
A. Tested Concentration Range; (pg/g dry sediment)
TNT - 126.0 to 2520.0 pg/g
RDX - 249.1 to 5340.0 ^g/g
B. Sensitivity;
TNT - 2.93 area units/ng based on 288 ng injection
RDX - 0.42 area units/ng based on 304 ng injection
C. Detection Limit; (ug/g dry sediment)
TNT - 178.5 Vg/g
RDX - 490.1 Mg/g
D. Interferences; None encountered during analysis
E. Analysis Rate; Each sample requires 15 minutes for extraction and 20
minutes for GC analysis. With a GC autosampler, one analyst can perform 30-40
extractions and load the autosampler vials in an 8-hour day.
II. CHEMISTRY:
Toluene, 2,4,6-trinitro-
CAS RN 118-96-7
Melting Point; 80.75°C Boiling Point; 240°C (explodes)
Hexahydro-1 ,3,5-trinitro-l ,3,5-triazine
CAS RN 121-82-4
Melting Point; 204°C Boiling Point; not available
TNT - Use caution in handling TNT. Potential explosive, skin absorption
and toxic inhalation hazards exist.
RDX - Use caution in handling RDX. Potential explosive and toxic inhalation
hazards exist.
A-3
-------
III. APPARATUS
A. Instrumentation;
Gas Chromatograph - Hewlett-Packard 5880A with flame ionization
detector, auto injector, computer controller and integrator.
B. Parameters;
Column - 2% Dexsil 300 GC on Anakrom Q packed in a 2 mm
I.D., 0.25 in. O.D. by a 2 ft. column.
Gas Flow - nitrogen - 32 ml/min at detector
Temperature - injection port 210°C
oven 140-230°C
Detector 250°C
Temperature programming - 8°C/min
Injection volume - 4 yl
Detector - flame ionization detector
Retention times - TNT - 2.0 minutes
RDX - 3.4 minutes
C. Hardware/Glassware;
1 ml pipets (1 for each standard; 1 per sample)
10 ml pipets (1 for solvent; 1 per sample)
culture tubes - 16 mm x 150 mm, teflon lined
screw cap (2 per sample)
10 ml volumetric flask (4)
GC vials - teflon septum (1 per sample)
centrifuge (1)
refrigerator (1)
25 yl Hamilton syringe (1)
50 yl Hamilton syringe (1)
10 yl Hamilton syringe (1)
aluminum foil
ASTM #10 sieve (1)
D. Chemicals;
TNT "SARM" - #PA 364, Lot #2714
RDX "SARM" - #PA 361, Lot #1101475-1
TNT - recrystallizod (determinedto be equivalent
purity as SARM)
RDX - rccrystallizcd (determined to be equivalent
purity ns SARM)
Acetone, certified (Fisher Scientific)
A-4
-------
IV. STANDARDS:
Concentrated stock solutions of RDX and TNT are prepared by weighing out
the following amounts of SARM material (or equivalent) into volumetric flasks and
bringing to volume with acetone.
RDX
534.1 mg in 10 ml = 53,411 mg/1 (I)
498.3 mg in 50 ml = 9,965 mg/1 (II)
111.9 mg in 25 ml = 4,474 mg/1 (III)
TNT
561.5 mg in 10 ml = 56,150 mg/1 (I)
360.0 mg in 50 ml = 7,200 mg/1 (II)
32.5 mg in 100 ml = 325 mg/1 (III)
These volumetrics are wrapped in aluminum foil and stored in a refrigerator until
needed. Storage time should not exceed one month.
A. Calibration Standards
Working standards are prepared by diluting the concentrated stock
solutions according to the following scheme in acetone.
RDX
5 ml of n to 10 ml = 4,983 mg/1 (A)
2.5 ml of I to 100 ml = 1,335 mg/1 (B)
1 ml of II to 10 ml - 997 mg/1 (C)
new stock = 206 mg/1 (D)
new stock = 153.9 mg/1 (E)
1 ml of C to 10 ml = 99.7 mg/1 (F)
5 ml of E to 10 ml = 76 mg/1 (G)
5 ml of F to 10 ml = 38 mg/1 (H)
TNT '
5 ml of II to 10 ml = 3,600 mg/1 (A)
1 ml of II to 5 ml = 1,440 mg/1 (B)
1 ml of II to 10 ml = 720 mg/1 (C)
325 mg/1 (D)
5 ml of III to 10 ml = 162 mg/1 (E)
1 ml of C to 10 ml = 72 mg/1 (F)
5 ml of F to 10 ml = 3G mg/1 (G)
5 ml of G to 10 ml = 18 mg/1 (H)
-------
Working standards should be freshly prepared at least every 2 days.
•
B. Control Spikes;
The concentrated stock solutions were used for spiking the sediment
samples. The dried sediment samples are spiked with the TNT or RDX by
pipeting known amounts of the control stock solutions U5 yl to 1 ml) onto the
sediment in a culture tube with a microliter syringe or 1 ml pipct. Spike should
be double expected TNT or RDX concentration or 2 to 10 times detection limit.
The sediment is allowed to dry in the dark and extracted and analyzed as in procedure.
Perform extraction, calibration, and analysis.
The following spiking procedure was implemented to determine pre-
cision, accuracy and detection limit for TNT and RDX in sediment: Four sets of
duplicate dried sediment samples weighing 2.00 + 0.02 g each were spiked as
follows:
Sample set A 50 yl of RDX Stock II = 249.1 yg/g RDX
25 yl of TNT Stock II = 126.0 yg/g TNT
Sample set B 20 yl of RDX Stock I = 534.0 yg/g RDX
70 yl of TNT Stock II = 252.0 yg/g TNT
Sample set C 50 yl of RDX Stock I = 1335.3 yg/g RDX
20 yl of TNT Stock I = 561.5 yg/g TNT
Sample set D 0.5 ml of RDX Stock II = 2491 yg/g RDX
50 yl of TNT Stock I = 1404 yg/g TNT
Sample set E 1.0 ml of RDX Stock II = 4982 yg/g RDX
0.7 ml of TNT Stock II = 2520 yg/g TNT
Four sets of duplicate 2.00 g + 0.02 g of unspiked sediment samples served as
blanks.
V. PROCEDURE:
The sediment samples are air dried in the dark by spreading on individual
labeled pieces of aluminum foil. Once the samples are dried, each sample is
sieved through an A.S.T.M. No. 10 sieve.
Each dried sieved sample is thoroughly mixed. Duplicate 2.0 g samples
are taken for analysis. Each subsamplc is weighed and placed into a new culture
tube with a teflon-lined screw cap.
A-6
-------
Ten ml of acetone are pipeted into the tube. The tube contents are shaken
well and allowed to settle after which they are centrifuged. The supernatant
is removed with a clean 10 ml pipct and placed in a second culture tube. The
contents of the tube arc shaken to thoroughly mix the sample. The mixed extract
is pipetted into a teflon septum sealed GC vial.
Samples are ready for GC analysis
Inject 4 pi of sample extract onto GC column in duplicate and
record peak area
Calibration, inject working calibration solutions singly at beginning
and conclusion of each analytical run: Plot peak area in nano-
grams injected of each standard to obtain working curve.
VI. CALCULATIONS;
Peak area of working standards used to prepare a calibration curve from
which apparent concentrations can be determined.
The concentration in the original sediment is calculated by the following
formula:
C(S) = C(e) x v(e)
w(s)
where C(s) = concentration in the original sediment
C(e) = concentration in extract
w(s) = weight of dry sediment
V(e) = volume of extract
The concentration is corrected for extraction efficiency by the following formula:
C(cs) = + b°
where C(cs) = corrected concentration in
dry sediment
bi = slope of the regression line
for found vs. target for
spiked samples
bo = intercept of the regression
line for found vs. target for
spiked samples
A-7
-------
VII. REFERENCE:
Lindner, V. (1980), "Explosives and Propellents," Kirk-Othmer Encyclopedia
Chemical Technology, 3rd edition, John Wiley and Sons, NY, 9, 561-671.
-------
DETERMINATION OF NITROCELLULOSE
IN SEDIMENT - QUANTITATIVE
ARC Method No. 2
L APPLICATION:
Method used to determine the concentration of nitrocellulose in sediment.
A, Tested Concentration Range: (yg/g sediment)
4.4 to 110.2 yg/g
B. Sensitivity:
0.23 absorbance units/yg (nitrite-nitrogen) based on a 0.5 yg sample.
C. Detection Limit: (yg/g sediment)
17.17 yg/g
D. Interferences: Organic nitrates and nitrite and inorganic nitrates and
nitrites in high concentrations.
E. Analysis Rate: Extraction, conversion and evaporation require 8 hours
(about 2 hours of "labor). Analysis requires 30 minutes for 6 samples. Total time
required for analysis is 1.5 days.
2. CHEMISTRY:
(CsH702(ON02)3N Nitrocellulose, Cellulose Trinitrate
CAS.RN: 9004-70-0
Melting Point: Decomposes* at 125°C
Nitrocellulose used in propellants has a nitrogen content ranging from 12.6 to 13.596.
Care should be taken in handling this explosive, especially in the dry state.
This method for analysis of nitrocellulose in sediment involves the following steps:
removal of interfering species by extraction with methanol'
extraction of the nitrocellulose with acetone
treatment with NaOH and evaporation to convert nitrocellulose
to nitrate/nitrite
conversion of nitrate to nitrite by treatment with a Hach
Nitri Ver VI Reagent
analysis for nitrite by EPA Method 354.1
A-9
-------
As no "standard sediment" has been approved and since the method is not adaptable
to the analysis of "standard distilled water," a sample of sediment was obtained
from a pond near Atlantic Research, dried and sieved. This material was
employed as a "provisional standard sediment."
3. APPARATUS:
A. Instrumentation:
Hach DR/2 Spectrophotometer with absorbance scale
B. Parameters:
Wavelength - 540 nm
C. Glassware/Hardware:
16x125 mm culture tubes with screw caps (teflon liners) (112)
25 ml volumetric flasks (6)
125 ml Erlenmeyer flasks (7)
Hach 25 ml sample cells (2 - matched)
pipets- 1 ml
pipets- 5 ml
pipets- 10 ml
D. Chemicals:
Acetone, ACS certified (Fisher Scientific)
Methanol, ACS certified (Fisher Scientific)
NaOH, reagent grade
Nitrocellulose, provisional SARM PA 365, Lot No. 36181
Hach Nitri Ver VI Reagent Powder Pillows
Nitrogen
Water, Nitrite Free Distilled
Sodium Nitrite, ACS certified (Fisher Scientific)
Buffer Color Reagent
HC1, Reagent Grade
Sulfanilamide, Baker Analyzed
N-U-Napthyl) Ethylene Diamine Dihydrochloride
Sodium Acetate, Baker Analyzed
4. STANDARDS
A 100 mg/1 (nitrite-nitrogen) stock solution (NI) is prepared by wieighing
out 0.493 g of sodium nitrite and diluting to 1000 ml with nitrite-nitrate free
distilled water.
A. Calibration:
Calibration standards arc prepared from the stock solution as follows:
A-10
-------
Calibration
Standard
1
2
3
4
Method of
Preparation
1 ml of NI to
1000 ml
50 ml of 1 to
100 ml
20 ml of 1 to
100 ml
. 10 ml of 1 to
Final Concentration
mg/1 nitrite-N
0.100
0.050
0.020
0.010
100 ml
These standards are run with each analysis.
B. Control Spikes:
The following stock solutions of nitrocellulose are prepared in acetone.
Stock No.
I
II
III
Method of
Preparation
44.1 mg NC to
100 ml
1 ml of I to
10 ml
1 ml of I to
50 ml
Final Concentration
mg/1 NC
- 441
44.1
8.82
Soil samples of 2.00 + 0.05 grams are weighed out and placed in 6 different culture
tubes. The sediment "is spiked with the following target solutions:
Blank
Tube A
Tube B
Tube C
Tube D
Tube E
Volume
of Spike
0
0.5 ml
2.0 ml
1.0 ml
0.5 ml
1.0 ml
Stock Solution
Cone., mg/1
0
441
44.1
44.1
44.1
8.82
Target Cone, in
Dry Sed., ug/g
0
110.2
44.1
22.0
11.02
4.41
A-ll
-------
5. PROCEDURE:
Add 6 ml of methanol to the sediment in the culture tube, replace cap,
shake well and centrifuge for 5-6 minutes, decant the methanol and discard.
Repeat the extraction scheme with 6 additional ml of methanol. Extract the
sediment, add 3 ml of acetone to the culture tube. Shake well, centrifuge,
pipet off the acetone and place in a clean culture tube. Repeat the extraction
scheme 3 times and combine the extracts. Add 3 ml of 1.0 N NaOH and evaporate
at 30°C under a stream of nitrogen to a final volume of less than 3 ml.
Quantitatively transfer the sample into a 25 ml volumetric flask and
bring up to volume with nitrite free distilled water. Shake well and transfer to
125 ml Erlenmeyer flask. Add contents of 1 Hach Nitri Ver VI Reagent Pillow
and swirl sample.
Prepare buffer color reagent, adding 105 ml cone. HC1, 5.0 g sulfanilamide,
0.5 g N-(l-Napthyl) ethylene-diamine dihydrochloride and 136 g sodium acetate
to 250 ml nitrite free distilled water in a 500 ml volumetric flask.
Stir until dissolved and bring to volume with nitrite free distilled water.
Reagent is stable for several weeks if stored in the dark.
Add 2 ml of buffer color reagent to solution in Erlenmeyer flask. Swirl
flask contents and allow color to develop for 15 minutes. Place sample in 25 ml
sample cell and blank in a second cell. Zero machine on blank before each
sample. Read absorbance of sample at 540 nm. Dilute as necessary to maintain
absorbance readings between 0.2 and 0.8. Read absorbance of calibration standards,
plot calibration curves of mg/1 Nitrite-N versus absorbance.
6. CALCULATIONS:
The nitrite-nitrogen concentration is read off the calibration curve. The
nitrocellulose concentration is calculated assuming 13.0% nitrogen and taking
into account the dilution factor.
Ug nitrocellulose = nitrite-N (yg/ml) x
g sediment
LOO g nitrocellulose x 25 ml x 10
0.13 g N 2 g
A-12
-------
7. REFERENCES:
"EPA Methods for Chemical Analysis of Water and Wastes," March, 1979,
EPA-600/4-79-020.
".Water Analysis Handbook," (1979), Hach Chemical Company, Loveland,
CO., 12338-08.
Lindner, W. (1980), "Explosives and Propellants," Kirk-Othmer Encyclopedia of
Chemical Technology, 3rd edition, John Wiley and Sons, NY, 9, 561-671.
A-13
-------
LEAD AND CADMIUM IN WATER - QUANTITATIVE
CHROMIUM AND ZINC IN WATER - SEMI QUANTITATIVE
ARC METHOD H
DECON TECHNOLOGY
DAAK11-80-C-0027
I. APPLICATION;
Method used to determine lead and cadmium in water (quantitative) and chromium
and zinc in water (semiquantitative).
A. Concentration Range Tested;
Lead - 0.25 to 5.0
Cadmium - 0.05 to 1.0
Chromium - 0.25 to 5.0 ug/ml
Zinc - .0.05 to 1.0
B. Sensitivity; Not applicable.
C: Detection Limit;
Lead - 0.250 yg/ml
Cadmium - 0.050 ug/ml
Chromium - 0.277 yg/ml
Zinc - 0.177 yg/ml
D. Interferences; None observed during analysis.
E. Analysis Time; Analysis of a set of six samples can be accomplished
in 30 minutes per metal (including preparation of standards and standard curve.)
II. CHEMISTRY;
Lead: CAS Registry No. - 7439-92-1
Melting Point - 327.4°C Atomic Weight - 207.2
Lead exhibits acute human toxic effects for blood levels above
0.05 mg %.
Cadmium: CAS Registry No. - 7440-43-9
Melting Point - 321°C Atomic Weight - 112.40
Cadmium and its salts are highly toxic by inhalation and ingestion.
Chromium: CAS Registry No.. - 7^40-47-3
Melting Point - 1900°C Atomic Weight - 51.996
A-14
-------
Chromium(VI) salts have irritant effects' on skin and respiratory pas-
sages and are highly toxic if. ingested. Chromium (III) compounds show little
or no toxic effects.
Zinc: CAS Registry No. - 7440-66-6
Melting Point - 419.5°C Atomic Weight - 65.38
Zinc salts have irritant effects on skin and respiratory passages and
are highly toxic if ingested.
III. APPARATUS;
A. Instrumentation;
Van'an Atomic Absorption Spectrophotometer AA-775
Varian hollow cathode lamps
Cadmium #56-100008-00
Lead #56-100029-00
Chromium #56-100012-00
Brass (combination Copper - Lead - Zinc) #56-100192-00
B. Parameters;
Lamp Current Cadmium 3.5m Amp
Lead 5.0 m Amp
Chromium 7.0 m Amp
Zinc 5.0 m Amp
Fuel Air-acetylene flame
Wavelength
Spectral Band Pass
C. Hardware/Glassware;
16 x 150mm culture tubes with teflon lined screw caps
1, 5 and 10 ml disposable pipets
D. Chemicals;
Cadmium (certified Atomic Absorption, Reference Standard 1,000
ppm - Fisher Scientific Co.)
Lead (certified Atomic Absorption, Reference Standard 1,000
ppm - Fisher Scientific Co.)
A-15
Cadmium
Lead
Chromium
Zinc
Cadmium
Lead
Chromium
Zinc
228.8 nm
283.3 nm
357.9 nm
213.9 nm
0.5 nm
0.5 nm
0.2 nm
1.0 nm
-------
Chromium (Dilut-It Chromium Standard, 1 g Cr + as K2Cr04
J.T. Baker Chemical Co.)
Zinc (certified Atomic Absorption Reference Standard 1,000
ppm - Fisher Scientific Co.)
IV. STANDARDS;
schemes.
Calibration Standards;
The working standards were prepared according to the following
Lead Standard = 1,000 yg/ml (I)
1 ml of I diluted to 10 ml = 100
2 ml of II diluted to 10 ml = 20
1 ml of II diluted to 10 ml = 10
1 ml of IV diluted to 10 ml 1
5 ml of V diluted to 10 ml = 0.5
2.5 ml of V diluted to 10 ml = 0.25
pg/ml
yg/ml
yg/ml
yg/ml
yg/ml
pg/ml
(II)
(III)
(IV)
(V)
(VI)
(VII)
Cadmium Standard = 1,000 yg/ml (I)
1 ml of I diluted to 10 ml = 100 ug/ml (II)
1 ml of II diluted to 10 ml = 10 wg/ml (III)
3 ml of III diluted to 10 ml = 3 pg/ml (IV)
2 ml of III diluted to 10 ml = 2 yg/ml (V)
1 ml of III diluted to 10 ml = 1 yg/ml (VI)
1 ml of VI diluted to 10 ml = 0.1 yg/ml (VII)
5 ml of VII diluted to 10 ml = 0.05 yg/ml (VIII)
Chromium Standards = 1000 yg/ml (I)
1 ml of I diluted to 10 ml
2 ml of II diluted to 10 ml
1 ml of II diluted to 10 ml
1 ml of IV diluted to 10 ml
5 ml of V diluted to 10 ml
5 ml of VI diluted to 10 ml
100
20
10
1
0.5
0.25
yg/ml
yg/ml
yg/ml
yg/ml
ug/ml
(II)
(III)
(IV)
(V)
(VI)
(VII)
Zinc Standard = 1000 ug/ml (I)
1 ml of I diluted to 10 ml
1 ml of I diluted to 10 ml
1.5 ml of III diluted to 10
1.0 ml of III diluted to 10
0.5 ml of III diluted to 10
1 ml of VI diluted to 10 ml
ml =
ml =
ml =
100 ug/ml (II)
10 yg/ml (III)
1.5 yg/ml (IV)
1.0 ug/ml (V)
0.5 ug/ml (VI)
0.05 ug/ml (VII)
A-16
-------
B. Control Spikes
Lead and cadmium were spiked together into corresponding test
tubes and brought up to a volume of 15 ml with distilled water.
Lead Concentration
Test Tube 1. 3.75 ml of V = 0.25 yg/ml
Test Tube 2. 0.75 ml of IV = 0.5 yg/ml
Test Tube 3. 1.5 ml of IV -1.0 ug/ml
Test Tube 4. 3.75 ml of IV = 2.5 yg/ml
Test Tube 5. 7.5 ml of IV = 5.0
Test Tube 6. Blank
Cadmium Concentration
Test Tube 1. 0.75 ml of VI = 0.05
Test Tube 2. 1.5 ml of VI =0.1 ug/ml
Test Tube 3. 3.0 ml of VI - 0.2 pg/ml
Test Tube 4. 0.75 ml of III = 0.5 gg/ml
Test Tube 5. 1.5 ml of III = 1.0 yg/ml
Test Tube 6. Blank
Zinc and chromium standards were spiked toegther into corres-
ponding test tubes and brought up to a volume of 15 ml with distilled water.
Chromium Concentration
Test Tube 1. 3.75 ml of V = 0.25 ug/ml
Test Tube 2. 0.75 ml of IV = 0.5 yg/ml
Test Tube 3. 1.5 ml of IV =1.0 yg/ml
Test Tube 4. 3.75 ml of IV = 2.5 yg/ml
Test Tube 5. 7.5 ml of IV = 5.0 yg/ml
Test Tube 6. Blank
Zinc Concentration
Test Tube 1. 0.75 ml of V = 0.05 ug/ml
Test Tube 2. 1.5mlofV = 0.1 ug/ml
Test Tube 3. 3.0 ml of V = 0.2 ng/ml
Test Tube 4. 0.75 ml of III = 0.5 yg/ml
Test Tube 5. 1.5 ml of IIT = 1.0 ug/ml
Test Tube 6. Blank
A-17
-------
V. PROCEDURE;
The samples to be analyzed are aspirated into the flame for analysis.
Distilled water is aspirated between samples to ensure no cross contamination.
Each sample is run in duplicate. Standards are run daily to construct a
calibration curve which is input into the AA so that the sample concentra-
tions are read out directly in ppm.
VI. CALCULATIONS;
The apparent concentration of the metal is read directly from the
AA and multiplied by the appropriate dilution factor, if any. The actual
concentration is read from the target versus found line.
VII. REFERENCES;
Windholz, M. (1976), "The Merck Index," 9th edition, Merck Company,
Inc., Rahway, New Jersey.
Varian Techtrcn Pty, Ltd. (1979), "Analytical Methods for flame
spectroscopy," Springvale, Australia, Publication No. 85-100009-00.
A-18
-------
LEAD AND CADMIUM IN SEDIMENT. - QUANTITATIVE
CHROMIUM AND ZINC IN SEDIMENT -SEMIQUANTITATIVE
ARC METHOD #5
DECON TECHNOLOGY
DAAK11-80-C-0027
1. APPLICATION;
The method is applicable to the analysis of lead, cadmium, chromium and
zinc in soil.
A. Tested concentration range;
B.
C.
Lead
Cadmium -
Chromium -
Zinc
Sensitivity; not
Detection Limit
Lead
Cadmium -
Chromium -
Zinc
3.125 yg/g to 31.25 yg/g
0.625 ug/g to 12.5 ug/g
2.5 yg/g to 50.0 yg/g
2.0 ug/g to 40.0 Mg/g
applicable
7.07 yg/g
2.53 yg/g
7.21 yg/g
4.54 yg/g
D. Interference; None observed during operation.
E. Analysis Time; Extraction and filtering of 6 samples can be accomplished
in approximately 2 1/2 hours. Analysis of the samples on the instrument can be
accomplished in about 20 minutes.
II. CHEMISTRY
Lead: CAS Registry No. - 7439-92-1
Melting Point - 327.4°C Atomic Weight - 207.2
Lead exhibits acute human toxic effects for blood levels above 0.05 mg %.
Cadmium: CAS Registry No. - 7440-43-9
Melting Point - 321°C Atomic Weight - 112.40
Cadmium and its salts are highly toxic by inhalation and ingestion.
Chromium: CAS Registry No. - 7440-47-3
Melting Point - 1900°C Atomic Weight - 51.996
Chromium(VI) salts have irritant effects on skin nnd respiratory passages and
are highly toxic if ingested. Chromium (III) compounds show little or no toxic
effects.
A-19
-------
Zinc: CAS Registry No. - 7440-66-6
Melting Point - 419.5°C Atomic Weight - 65.38
Zinc salts have irritant effects on skin and respiratory passages and are
highly toxic if ingested.
III. APPARATUS;
A. Instrumentation;
Varian Atomic Absorption Spectrophotometer AA-775
Varian Hollow Cathode Lamps
Cadmium #56-100008-00
Lead #56-100029-00
Chromium #56-100012-00
Brass (combination copper - lead - zinc) #56-100192-00
B. AA Parameters
I. Lamp Current -
Cadmium
Lead
Chromium
Zinc
2.
3.
Fuel - Air-acetylene flame
Wavelength
4. Spectral Band Pass-
C. Hardware/Glassware
Cadmium
Lead
Chromium
Zinc
Cadmium
Lead
Chromium
Zinc
3.5 millilamps
5.0 millilamps
7.0 millilamps
5.0 millilamps
228.8 nm
283.3 nm
357.9 nm
213.9 nm
0.5 nm
0.5 nm
0.2 rim
1.0 nm
16 x 150 mm culture tubes with Teflon lined screw caps
25 ml volumetrics
50 ml graduated cylinders
glass funnels
15 cm filter paper for quantitative analysis
weight of ash below 0.1 milligram
H2O water bath
Thermolyne heating plate
25 mm x 150 mm test tubes •
25 ml flasks
1, 5 and 10 ml disposable pipets
A-20
-------
D. Chemicals;
Nitric acid - Reagent A.C.S. (Fisher Scientific Company)
Cadmium Certified Atomic Absorption Reference
Standard 1000 ppm (Fischer Scientific Company)
Lead Certified Atomic Absorption Reference Standard
1000 ppm (Fisher Scientific Company)
Dilut-It™ Chromium Standard, 1 g Cr6+ as I^CrCXi
(J.T. Baker Chemical Company)
Zinc Certified Atomic Absorption Reference Standard
1000 ppm (Fisher Scientific Company)
IV. STANDARDS;
A. Calibration Standards;
Lead Standard = 1000 ug/ml (I)
1 ml of I diluted to 10 ml = 100 pg/ml (II)
2 ml of II diluted to 10 ml = 20 ug/ml (III)
1 ml of II diluted to 10 ml = 10 yg/ml (IV)
1 ml of IV diluted to 10 ml = 1 Pg/ml (V)
5 ml of V diluted to 10 ml = 0.5 ug/ml (VI)
2.5 ml of IV diluted to 10 ml = 0.25 yg/ml (VH)
Cadmium Standard = 1000 ug/ml (I)
1 ml of I diluted to 10 ml = 100 ug/ml (II)
1 ml of II diluted to 10 ml = 10 ug/ml (III)
3 ml of III diluted to 10 ml = 3 ug/ml (IV)
2 ml of III diluted to 10 ml = 2 ug/ml (V)
1 ml of III diluted to 10 ml 1 ug/ml (VI)
1 ml of VI diluted to 10 ml = 0.1 ug/ml (VII)
5 ml of VII diluted to 10 ml = 0.05 ug/ml (VIII)
Chromium Standard = 1000 ug/ml (I)
1 ml of I diluted to 10 ml = 100 ug/ml (II)
2 mi of II diluted to 10 ml = 20 ug/ml (III)
1 ml of II diluted to 10 ml = 10 ug/ml (IV)
1 ml of IV diluted to 10 ml = 1 ug/ml (V)
5 ml of V diluted to 10 ml = 0.5 ug/ml (VI)
5 ml of VI diluted to 10 ml = 0.25 ug/ml (VII)
Zinc Standard = 1000 ug/ml (1)
1 ml of I diluted to 10 ml = 100 ug/ml (II)
1 ml of I diluted to 10 ml = 10 ug/ml (III)
1.5 ml of III diluted to 10 ml = 1.5 ug/ml (IV)
1.0 ml of III diluted to 10 ml = 1.0 ug/ml (V)
0.5 ml of III diluted to 10 ml = 0.5 ug/ml (VI)
1 ml of VI diluted to 10 ml = 0.05 ug/ml (VII)
A-21
-------
B. Control Spikes;
Lead and cadmium standards were spiked together into corresponding
test tubes containing 2.00 + .005 gram sediment samples.
Test Tube 1.
Test Tube 2.
Test Tube 3.
Test Tube 4.
Test Tube 5.
Test Tube 6.
Test Tube 1.
Test Tube 2.
Test Tube 3.
Test Tube 4.
Test Tube 5.
Test Tube 6.
Lead Concentrations
0.625 ml of IV = 3.125 yg/g
1.25 ml of IV = 6.25 yg/g
2.5 ml of IV = 12.5 ug/g
0.625 ml of H = 31.25 ug/g
125 ml of II * 62.50 yg/g
Blank
Cadmium Concentration
1.25 ml of VI
2.5 ml of VI
0.5 ml of HI
1.25 ml of III
2.5 ml of III
Blank
0.625 yg/g
1.25 yg/g
2.5 yg/g
6.25 yg/g
12.5 yg/g
Chromium and zinc standards were spiked together into corresponding
test tubes containing 2.00 + .005 gram sediment samples.
Chromium Concentration
Test Tube 1.
Test Tube 2.
Test Tube 3.
Test Tube 4.
Test Tube 5.
Test Tube 6.
.625 ml of IV
1.25 ml of IV
2.5 ml of IV
.525 ml of II
1.25 ml of II
Blank
=
—
—
—
=
2.5 yg/g
5.0 yg/g
10.0 yg/g
25 yg/g
50 yg/g
Zinc Concentration
Test Tube 1.
Test Tube 2.
Test Tube 3.
Test Tube 4.
Test Tube 5.
Test Tube 6.
.4 ml of III
.8 ml of III
2 ml of III
.4 ml of II
.8 ml of II
Blank
—
=
—
=
=
2.0 yg/g
4.0 yg/g
10.0 yg/g
20.0 yg/g
40.0 yg/g
A-22
-------
V. PROCEDURE
Two g + .005 g sediment samples are digested in 20 ml of 1:1 nitric acid/
distilled water Tor 2 hours on a hot water bath. The extract fronn the digestion
is filtered, diluted to 20 ml with distilled water and then poured into a flask for
analysis. Additional dilutions are. made of samples that are outside the optimum
working range for the AA. The samples are aspirated into the AA for analysis.
The aspirator is rinsed with distilled water between analyses to prevent cross
contamination. The standards are run before the samples and the curve input
into the AA so that the sampled concentrations are read out directly in ppm.
\L CALCULATIONS;
The apparent concentration of the metal in the solution is read directly
from the AA. The concentration in the sediment is calculated by:
ug/g sediment = pg/ml from A A x 20 ml x any additional dilution factor
2 g
The actual concentration is read from the target versus found line.
YD.- REFERENCES:
Windholz, M. (1976), 'The Merck Index," 9th edition, Merck Company, Inc.
R'ahway, New Jersey.
Varian Techtron Pty. Ltd. (1979), "Analytical Methods for Flame Spectro-
scopy," Springvale, Australia, Publication No. 85-100009-00.
A-23
-------
ANALYSIS OF NITRATE-N IN SEDIMENT - SEMIQUANTITATIVE
ARC Method #6
DECON Technology DAAK11-80-C-0027
I. APPLICATION
Method used to determine the concentration of nitrate-N in soils and
sediment.
A. Tested Concentration Range (yg/g dry sediment)
Nitrate-N - 1.0 to 20.0 yg/g
B. Sensitivity;
50 mV/yg
C. Detection Limit: (yg/g dry sediment)
Nitrate-N - 2.056 Mg/g
D. Interferences; none encountered during analysis
E. Analysis Rate; Each sample requires 20 minutes for extraction and about
3 minutes to test; A set of six samples can be analyzed in 40 minutes.
II. CHEMISTRY
Nitrate usually occurs in soils and sediment as a salt and is a common
plant nutrient.
III. APPARATUS
A. Instrumentation;
Orion 93-07 nitrate specific ion electrode
Orion 91-05 combination pH gel filled electrode
Orion 399A pH/mV meter
Mettler H5AR Analytical Balance
B. Hardware/Glassware;
volumetric flask - 1000 ml (1)
volumetric flask - 100 ml (1)
volumetric flasks - 10 ml (4)
beakers - 100 ml (6)
pipet - 10 ml (6)
pipet - 1 ml (3)
pipet - 2__ml (2)
quart dark reapent bottle (1)
A-24
-------
C. Chemicals;
Potassium Nitrate, analytical reagent grade, (Mallinckordt, Inc.).
Silver Sulfate, Certified ACS, (Fisher Scientific Co.)
Aluminum Sulfate, Certified ACS, (Fisher Scientific Co.)
Boric Acid, Photographic/Technical, (Eastman Kodak Co.)
Sulfamic Acid, Certified (Fisher Scientific Co.)
IV. STANDARDS
A. Extracting Solution (I)
16.66 g Aluminum Sulfate
1.24 g Boric Acid
4.67 g Silver Sulfate
2.43 g Sulfamic Acid
Diluted to 1 liter with distilled water
B. Calibration Standards;
Calibration standards are prepared according to the following scheme:
1.44 g Potassium Nitrate/100 ml = 2000 ppm (II)
1 ml II diluted to 10 ml = 200 ppm (III)
1 ml in diluted to 10 ml = 20 ppm (IV)
1 ml IV diluted to 10 ml = 2 ppm (V)
1 ml V diluted to 10 ml = .2 ppm (VI)
Standards IH through VI prepared from n daily. Calibration standards
were all diluted with extracting solution I, spiking standards were diluted with
distilled water.
C. Control Spikes;
Prepared standards were added to dried, preweighed sediment samples
weighing 2.000 10.005 g according to the following scheme:
(a) 1 ml of VI to give target of 1 y g/g
(b) 2 ml of VI to give target of 2 u g/g
(c) 4 ml of VI to give target of 4 u g/g
(d) 1 ml of V to give target of 10 ug/g
(e) 1 ml of III to give target of 100 ug/g
(f) Blank
The sediment was allowed to dry before extraction and analysis. Or.e
set of spiked controls is run.
A-25
-------
V. PROCEDURE
A calibration curve is plotted daily using standards III through VI which
were mixed with extracting solution.
Each 2 gram sediment sample is dispersed in 10 ml of extracting solution.
After 20 minutes, mV readings are taken using the nitrate specific ion electrode
and the reference portion of the combination pH electrode with the pH/mV meter.
Between samples, the electrodes are rinsed with distilled water and blotted dry.
VI. CALCULATIONS
From the calibration plot, concentrations are determined in ug/ml for the
sample dispersions and are converted to yg/g of sediment by multiplying by:
10 ml dilution
2.00 g sample
Vn. REFERENCES:
Orion Research, Inc. (1978) "Methods Manual, 93 series electrodes", Form
93 MM/8740, Cambridge, Massachusetts.
Sawyer, C.N. and McCarty, P.L. (1978), Chemicals for Environmental
Engineering, 3rd Edition, McGraw-Hill Book Company, New York.
A-26
-------
ANALYSIS OF NITRATE - N IN WATER - SEMIQUANTITATIVE
ARC Method #7 ..
DECON Technology DAAK11-80-C-0027
I. APPLICATION
A. Tested Concentration Range;
Nitrate-N - 0.2 to 4.0.ug/ml
B. Sensitivity;
50 mV/vg
C. Deteciion Limit;
Nitrate-N - 0.243 ug/ml
D. Interferences; none encountered during analysis.
E. Analysis Rate; Samples require about 5 minutes to prepare and test.
D. CHEMISTRY
The nitrate ion, NO3~, often occurs in natural waters as a result of aerobic
oxidation of amrr.onia-N. It was determined that drinking waters with a high nitrate
content caused methemoglooinemia in infants. EPA therefore proposed that nitr&te-N
levels should not exceed 10 mg/1 in public water supplies.
IE. APPARATUS
A. Instrumentation;
Orion 93-07 nitrate specific ion electrode
Orion 91-05 combination pH gel filled electrode
Orion 399A pH/mV meter
Mettler H5AR Analytical Balance
B. Hardware/Glassware;
100 ml volumetric flasks
100 ml beakers
10 ml disposable pipets
1 ml disposable pipets
1 quart dark reagent bottles
C. Chemicals;
Potassium Nitrate, analytical reagent grade (Mallinckrodt, Inc.)
A-27
-------
Ammonium sulfate, t:Baker Analyzed" reagent grade (J.T. Baker
Chemical Co.)
IV. STANDARDS
A. Ionic Strength Adjuster (I);
26.42 g ammonium sulfate diluted to 100 ml with distilled water
B. Calibration Standards;
1.444 g Potassium nitrate/100 ml distilled water = 2000 ppm (II)
1 ml II diluted to 10 ml = 200 ppm (III)
1 ml ID diluted to 10 ml = 20 ppm (IV)
1 ml IV diluted to 10 ml = 2 ppm (V)
1 ml V diluted to 10 ml = .2 ppm (VI)
Standards ni through VI are prepared daily from II.
C. Control Spikes;
Prepared standards were diluted with distilled water according to the
following scheme:
(a) 25 ml of VI undiluted to give target of 0.2
(b) 5 ml of V diluted to 25 ml to give target of 0.4 yg/ml
(c) 10 ml ofV diluted to 25 ml to give target of 0.8 yg/ml
(d) 25 ml of V undiluted to give target of 2.0 yg/ml
(e) 5 ml of IV diluted to 25 ml to give target of 4.0 ug/ml
One set of spiked controls were run.
V. PROCEDURE
A calibration curve is plotted daily using standards III through VI. Prior to
testing, all samples and standards have 0.5 ml Ionic Strength Adjuster added to
25 ml of sample to provide a constant background.
Millivolt readings are taken of each sample and standard using the nitrate
specific ion electrode and the reference portion of the combination pH electrode
with the pH/mV meter. Between samples the electrodes are rinsed with distilled
water and blotted dry.
VI. CALCULATIONS
From the calibration plot, concentrations are determined directly in ug/ml.
VH. REFERENCE
Orion Research, Inc. (1978) "Methods Manual, 93 series electrodes," form
93 MM/8740, Cambridge, Massachusetts.
A-28
-------
Sawyer, C.N. and McCarty, P.L. (1978), Chemistry for Environmental
Engineering, 3rd Edition, McGraw-Hill Book Company, New York.
A-29
-------
ANALYSIS OF LOW LEVELS OF TNT, RDX AND TETRYL IN
SEDIMENT - SEMIQUANTITATIVE
ARC Method #12 '
. I. APPLICATION:
Method used to determine the concentration of TNT, RDX and tetryl in
sediment.
*
A. Tested Concentration Range; (yg/g dry sediment)
TNT 0.502 to 10.05 ug/g
RDX 0.540 to 10.80 pg/g
Tetryl 0.495 to 9.90 pg/g
B. Sensitivity;
TNT 7.8 cm based on a 0.196 yg injection
RDX 5.6 cm based on a 0.194 yg injection
Tetryl 5.7 cm based on a 0.155 yg injection
C. Detection Limit; (pg/g dry sediment)
TNT 1.5551 yg/g
' RDX 1.4039 yg/g
Tetryl 0.2965 yg/g
D. Interferences; No interferences were observed with TNT, RDX and
tetryL
E. Analysis Rate; Six samples can be extracted in 20-30 minutes. With
an LC autosamplcr, one analyst can extract and perform duplicate analyses on
approximately 18 samples in an 8-hour day, including standards.
H. CHEMISTRY;
Toluene, 2,4,6-trinitro-
CAS RN 118-96-7
Melting Point: 80.75°C Boiling Point; 24()°C (explodes)
C3H6N6° 6 Hexahydro-1,3 ,5-trinitro-l,3 ,5-triazine
CAS RN 121-82-4
Melting Point: 204CC Boiling Point; not available
Tetryl; Aniline, M-methyl-N-2,4,6-tetranitro
CAS RN 479-45-8
Melting Point: 131°C Boiling Point; 187°C (explodes)
A-30
-------
Hazards. Use caution in handling TNT, RDX and tetryl, explosive and toxic hazards
exist.
ID. APPARATUS:
~ A, Instrumentation;
HPLC - Perkin-Elmer "601 Liquid Chromatograph with Perkin-
Elmer UV Spectrophotometer
LC-55 variable wavelength detector
Waters Radial Compression Unit
Cole-Parmer Recorder
B. Parameters;
Column - Waters 8"Cig 10 fi reverse phase radial compression
column
Mobile Phase - isocratic; 4596 methanol/5596 water @ 3.0 ml/min
At other methanol concentrations, interfering, peaks present
problems, particularly with RDX.
Detector - 230 nm
Injection Volume - 175 pi for all samples
Retention Times of Compounds:
TNT 10.5 min
RDX 4.0 min
• . Tetryl 8.0 min
C. Hardware/Glassware;
Gelman suction filter apparatus (1)
Aqueous 0.45 //m membrane filters (1 for each solvent run)
Volumetric flask - 100 ml (4)
Volumetric flask - 50 ml (1)
Culture tubes - 16 mm x 150 mm, teflon lined screw cap (6)
Pipets - 1 ml (10)
Pipets - 10 ml (6)
Pasteur Pipets (12)
Aluminum foil
Refrigerator
Centrifuge
1 ml LC sample bottles with teflon lined caps (12)
A-31
-------
D. Chemicals;
TNT "SARM" - PA 364, Lot #2714
RDX "SARM" - PA 361, Lot #1101475-1
Tetryl "SARM" - PA 608, Lot #2714
Acetone, HPLC (Fisher Scientific)
Methanol, HPLC (Fisher Scientific)
Distilled water
IV. STANDARDS:
A. Calibration Standards;
Concentrated individual stock standards of TNT, RDX and tetryl are
prepared by weighing out the following amounts of SARM material into volumetric
flasks and bringing them to volume with HPLC grade methanol:
TNT 33.5 mg in 100 ml = 335 mg/1 (I)
RDX 27.75 mg in 100 ml = 277.5 mg/1 (II)
Tetryl 6.65 mg in 50 ml = 133 mg/1 (in)
The concentrated individual stock solutions are mixed and diluted with HPLC grade
methanol according to the following scheme to produce the calibration standards:
Standard Solution A
TNT 1 ml I to 100 ml = 3.35 mg/1
RDX 1 ml II to 100 ml = 2.775 mg/1
Tetryl 2 ml III to 100 ml = 2.66 mg/1
Standard Solution B \
TNT 0.5 ml I to 10 ml = 16.75 mg/1
RDX 0.5 ml II to 10 ml = 13.88 mg/1
Tetryl 1.0 ml III to 10 ml = 13.30 mg/1
Standard Solution C
TNT 0.5 ml I to 15 ml = 11.16 mg/1
RDX . 0.6 ml II to 15 ml = 11.10 mg/1
Tetryl 1.0 ml III to 15 ml = 8.86 mg/1
A-32
-------
Standard Solution D
TNT 1 ml Standard Solution C to 10' ml
RDX 1 ml Standard Solution C to 10 ml
Tetryl 1 ml Standard Solution C to 10 ml
Standard Solution E
TNT i ml Standard Solution C to 25 ml
RDX 1 ml Standard Solution C to 25 ml
Tetryl 1 ml Standard Solution C to 25 ml
Standard Solution F
1.12 mg/1
1.11 mg/1
0.886 mg/1
0.446 mg/1
0.444 mg/1
0.354 mg/1
TNT 1 ml Standard Solution B to 100 ml
RDX 1 ml Standard Solution B to 100 ml
Tetryl 1 ml Standard Solution B to 100 ml
0.168 mg/1
0.139 mg/1
0.133 mg/1
Working standards should be freshly prepared at least every 2 days.
B. Control Spikes;
.Concentrated stock solutions of TNT, RDX and tetryl are prepared
by weighing out the following amounts of SARM material into 100 ml volumetric
flasks and bringing them to volume with HPLC grade methanol.
TNT 2L6 mg in 100 ml =
RDX 33.5 mg in 100 ml =
Tetryl 330.2 mg in 100 ml =
216 mg/1 (a)
335 mg/1 (b)
3302 mg/1 (c)
The concentrated stock solutions are then diluted with HPLC grade methanol as follows:
1 ml of I to 10 ml
TNT i mi Of n to 10 ml
1 ml of I to 10 ml
RDX 1 ml of II to 10 ml
1 ml of n to 100 ml
Tetryl 1 ml of III to 10 ml
21.6 mg/1 (d)
2.16 mg/1 (e)
33.5 mg/1 (f)
3.35 mg/1 (g)
33.02 mg/1 (h)
3.30 mg/1 (i)
These volumetrics are wrapped in aluminum foil and stored in a refrigerator until
needed. Storage time should not exceed one month.
A-33
-------
To spike the sediments, six 1 + 0.05 gram samples cf dry sediment
are weighed out. Each sample is placed in a culture tube with teflon lined screw
cap. The spiked samples are prepared as follows:
L TNT 0.3 ml of d = 10.05 yg
RDX 0.5 ml of f = 10.80 ug
Tetryl 0.3 ml of h = 9.90 yg
2. TNT 0.15 ml of d = 5.03 yg
RDX 0.25 ml of f = 5.40 yg
Tetryl L5 ml of i = 4.95 yg
3. TNT 0.6 ml of e = 2.01 yg
RDX LO ml of g = 2.16 yg
Tetryl 0.6 ml of i = L98 yg
4. TNT 0.3 ml of e = LOO yg
RDX 0.5 ml of g = 1.08 yg
Tetryl 0.3 ml of i = 0.99 yg
5. TNT 0.15 ml of e = 0.50 yg
RDX 0.25 ml of g = 0.54 yg
Tetryl 0.15 ml of i = 0.50 yg
6. Blank
The spikes are covered with aluminum foil and refrigerated until analysis. These
control spike samples are used to determine precision, accuracy and detection
limits for TNT, RDX and tetryl in sediment.
V. PROCEDURE:
The spiked sediment samples are evaporated to dryness under a stream
of nitrogen. Once the sediment is dry, 2 ml of acetone are added to the culture
tube. The tube is capped and shaken to extract the explosives from the sediment.
The tubes are then centrifuged for 5 minutes. The acetone is carefully drawn
off the sediment with a pasteur pipet and placed in LC vials for analysis.
The samples are ready for LC analysis .
Inject 175 yl of methanol standard singly before each run.
Inject 175 yl of each sample, unknown and spikes, in duplicate.
In order to keep the peaks on the scale of the recorder, the detector and/or
recorder sensitivity must be adjusted to the appropriate range.
A-34
-------
VI. CALCULATIONS:
Plot peak height of standard (mm) versus nanograms of material injected
on the column [fil injected x concentration of standard (ug/L x 10~6)3 to obtain
standard curve. Obtain apparent concentrations of unknown by reading the nano-
grams in the sample off the standard curve. The actual concentration for analytes
in the samples are then determined from the target versus found line.
VII. REFERENCE :
Lindner, V. (1980), "Explosives and Propellents," Kirk-Othmer Encyclopedia
of Chemical Technology, 3rd edition, John Wiley and Sons, NY, £, 561-671.
A-35
-------
APPENDIX B. WET CHEMISTRY METHODS
B-l
-------
Appendix B
A. p_H
Twenty grams of the dried sediment were mixed with 20 ml of distilled water
and placed on a shaker for an hour. The mixture was then allowed to settle for half
an hour and the pH was taken using an Orion 399A pH/mV meter.
B. Chloride
An additional 20 ml of distilled water were added to the pH samples to give
a final weight ratio of 1:2. The mixtures were again placed on the shaker for an hour.
MgSC>4 was added to flocculate the sediment which was then centrifuged. Measured
amounts of the extract were diluted to 50 ml and sodium bicarbonate and K2CrO4
indicator were added. The samples were finally titrated with .0025 N AgNOs to the
appearance of the reddish-brown precipitate indicating the endpoint.
C. Percent Moisture
A 5 gram portion of the sediment was added to a tared crucible and dried in
an oven (100-105°C) to constant weight. The crucible was then cooled in a dessicator
and reweighed. The percent moisture was calculated from the following formula:
% moisture = 100 x weight before drying - weight after drying
weight before drying
D. Percent Volatiles
The weight lost by a dry sediment on heating to 550°C provides a general idea
of the amount of organic matter in the sediment. The previously oven dried sample
from the percent moisture analysis was heated in a muffle furnace at 250°C for 30
minutes. The sample was then cooled in a dessicator and reweighed. The percent
volatiles was calculated by the following formula:
% volatiles = 100 x weight of dried sample - weight after ignition
wieght of dried sample
E. Chemical Oxygen Demand
The Standard Methods (1979) procedure for chemical oxygen demand (COD) in
Section 508 was followed. Weighed samples were diluted to 20 ml with distilled water
and prepared with mercuric sulfate, potassium dichromate and concentrated sulfuric
acid ' fixed with silver sulfate. The samples were refluxed for 2 hours, diluted and
titrated with ferrous ammonium sulfate and ferroin indicator. A blank and a standard
were run with each set of samples.
B-2
-------
F. Biochemical Oxygen Demand (BOD)
The procedure followed for BOD is listed in Section 507 of the Standard
Methods (1979). Dilutionsof 1,2 and3 grams of sediment were made with standard BOD
diluent and initial dissolved oxygen readings were taken after 15 minutes. The samples
were incubated at 20°C for 5 days. At the end of the incubation period,
dissolved oxygen readings were taken with a calibrated YSI Dissolved Oxygen Meter and
the soluble BOD calculated.
G. Gas Analyses
Gas samples were collected in 22 liter plastic air sampling bags (Plastic Film
Eng.). The gases analyzed were N2, O2, CO2 and CO.
The analyses were performed in a Varian 3700 gas chromatograph. The gas
chromatograph parameters were:
Detector: Thermal conductivity
Column: Altech "CTR" coaxial/molecular seive/
porapak column
Gas Flow: 35 cc/min
Temperature:
injection port: 210°C
oven: 60°C isothermal
detector: 200°C
Retention Time:
CO2 0.25 min
CO 2.72 min (and 0.16 min)
N2 1.55 min (and 0.16 min)
O2 L10 min (and 0.16 min)
B-3
-------
APPENDIX C. RAW DATA FROM INCINERATION AND
ACETONE EXTRACTION EXPERIMENTS
C-l
-------
Table C-I . Sediment Weights for Incineration Experiments
Sample
Weight Before (g)
Weight After ( g)
Lagoon 9
300°, 5 minutes
300°, 30 minutes
300°, 60 minutes
500°, 5 minutes
500°, 30 minutes
500°, 60 minutes
700°, 5 minutes
700°, 30 minutes
700°, 60 minutes
900°, 5 minutes
900°, 30 minutes
900°, 60 minutes
Lagoon 11
200°, 5 minutes
200°, 30 minutes
200°, 60 minutes
300°, 5 minutes
300°, 30 minutes
300°, 60 minutes
500°, 5 minutes
500°, 30 minutes
500°, 60 minutes
700°, 5 minutes
700°, 30 minutes
700°, 60 minutes
4.442
4.112
3.989
4.120
4.164
4.299
4.112
3.999
3.987
3.905
4.010
4.022
4.072
4.129
4.031
4.070
4.077
4.102
4.149
4.155
4.682
4.005
4.031
4.039
3.471
3.212
3.127
3.027
2.484
3.079
3.016
2.429
2.426
2.357
2.937
2.397
3.795
3.916
3.804
3.846
3.769
3.795
3.733
3.710
4.260
3.563
3.597
3.585
C-2
-------
Table C-2. Acetone Extraction of TNT/RDX/Tetryl Contaminated Sediment -
Raw Data
O
i
CO
Run No.
1
2
3
4
5
ft
7
8
9
10
II
12
13
14
15
16
17
18
Temperature
(°C)
25
25
25
25
25
25
50
50
50
50
50
50
75
75
75
75
75
75
Time
(•in)
15
15
30
36
60
60
15
15
30
30
60
60
15
15
30
30
60
60
Weight
Pefore
10.04
10.06
10.08
10.00
10.00
10.02
10.10
10.15
10.12
10.00
10.00
10.00
10.15
10.00
10.15
10.01
10.19
10.02
Weight
After
-------
Table C-3. Acetone Extraction of Nitrocellulose - Raw Data
o
lun Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
Ifi
17
18
Temperature
(°C)
25
25
25
25
25
25
50
50
50
50
50
50
75
75
75
75
75
75
Time
(min)
15
15
30
30
60
60
15
15
30
30
60
60
15
15
30
30
60
60
Weight
Before
(g)
10.05
10.07
10.00
10.07
10.01
10.05
10.01
10.06
10.01
10.04
10.02
10.02
10.01
10.03
10.03
10.01
10.01
10.03
Weight
After
(g)
9.20
9.59
9.10
8.80
9.33
9.62
9.15
9.60
8.78
9.45
9.13
9.61
8.84
9.24
9.30
8.66
8.11
9.04
Volume Solvent
Solvent Recovered Nitrocellulose Acetone Nil
(ml) O'g/g) l«vel (ug
95
117
115
117
105
120
110
115
120
117
95
113
105
95
110
105
90
117
38,900
31,800
17,800
44,700
54.700
62,700
47.000
55.200
32,000
29,900
44.800
64,600
11,700
15,400
-
12,000
15,100
20,600
2,340
1.832
1,832
1.393
2,271
1,956
3.161
2,024
3,409
4.681
2.463
1 ,956
7,327
5,627
5.683
4.681
5.052
4,422
------- |