EPA-600/2 74-006
July 1974
Environmental Protection Technology Series
•••BBBBl
Study of Feasibility of
Herbicide Orange Chlorinolysis
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to fche ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and -non-point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards.
-------
EPA-600/2-7U-006
July
STUDY OF FEASIBILITY OF HERBICIDE ORANGE
CHLORINOLYSIS
by
Dr. Edgar A. Lavergne
Contract No. 68-01-0457
Program Element 1BB036
Interagency Agreement No. EPA-IAG-008(R)
Project Officers
Mr. Paul E. des Rosiers
Office of Research and Development
Washingtoni D. C. 20460
and
Dr. Robert R. Swank
Southeast Environmental Research Laboratory
Athens, Georgia 30601
Prepared For
Office of Research and Development
United States Environmental Protection Agency
Washington, D. C. 20460
-------
EPA Review Notice
This report has been reviewed by the EPA, and approved
for publication. Approval does not stgnify that the
contents necessarily reflect the views and policies of
the Environmental Protection Agency, nor does mention
of trade names or commercial products constitute
endorsement or recommendation for use.
ii
-------
ABSTRACT
A process termed chlorinolysis (exhaustive chlorination) was
applied to samples of USAF herbicide ORANGE, a discontinued military
defoliant. The herbicide (a 50/50 mixture of the n-butyl esters of
2,4-D and 2,4,5-T) contained a production impurity, 2,3,7,8-tetrachloro-
dibenzo-p-dioxin (TCDD or dioxin) - a powerful teratogen.
The research objective was to demonstrate the feasibility of
chlorinolysis as a resource saving method to convert such herbicides
into useful and marketable products, namely, carbon tetrachlorfde CCCl^,
carbonyl chloride (COC^K and hydrogen chloHde (HC1), while destroying/
converting any dioxin present therein.
Bench scale (100 g/hr) chlorinolysis of herbicide ORANGE was
evaluated over a range of conditions. The critical reaction parameters
were found to be: chlorine to carbon ratio (4.4-7.2), temperature
(600-800°C), pressure (225-300 psig), and retention time (0.5-1.0 minute).
Thermodynamic analysis had indicated that CC14, hexachlorobenzene
(HCB), and chlorine (Cl2) would exist in equilibrium at the reaction con-
ditions utilized. Because of the balance required between reaction rate
(reactor size) and HCB content of the effluent, recycle of unconverted
HCB from the product recovery system was found to be necessary. Recycle
tests demonstrated that single pass HCB conversion rates of greater than
80% could be realized.
Destruction of dioxin was complete to below the detectable limit of
10 parts per trillion (ppt) in a single pass operation of the reactor
system. Preliminary toxicological tests of the recovered CC14 on rabbits
and rats showed no evidence of dioxin contamination.
Recovery of the 2,4-D ester component via fractional distillation
was shown to be feasible in an Oldershaw column operated under vacuum.
Data indicated that about 80% of the 2,4-D ester could be recovered as
a distillate free of detectable dioxin. Economics of a combined
fractionation/conversion process for herbicide ORANGE disposition appears
to be superior to chlorinolysis alone.
This successful study opens the path to an effective method for
conversion of environmentally unacceptable materials into useful products
of commerce. Investigation of the application of the process for the
disposal of other materials, such as DDT, lindane, dieldrin, PCBs, waste
chlorocarbons, and certain military chemicals, is recommended.
iii
-------
ACKNOWLEDGEMENTS
The author and the Diamond Shamrock Corporation wish to
acknowledge the efforts of Mr. Paul E. des Roslers of the EPA Office
of Research and Development, Project Director for this contract,
Dr. Robert R. Swank of the EPA Southeast Environmental Research
Laboratory, Project Officer of this contract, and Dr. Billy Welch,
the Assistant Deputy Secretary of the A1r Force for Environmental
Matters, 1n providing guidance, materials, and financial support 1n
the execution of this program,
Dr. Edgar A. La verge 1s the Associate Director of Research for
Process Development for the Diamond Shamrock Corporation, T. R. Evans
Research Center, Pa1nesv1lle, Ohio 44077.
1v
-------
CONTENTS
SECTION PAGE
I CONCLUSIONS I
II RECOMMENDATIONS 2
III INTRODUCTION 3
IV THEORETICAL CONSIDERATIONS 6
V DISCUSSION OP RESULTS 9
VI PROCESS ECONOMICS 19
VII EXPERIMENTAL EQUIPMENT AND OPERATING PROCEDURE 24
VIII ANALYTICAL METHODS 33
IX DISTILLATION RECOVERY OP 2,4-D ESTER 42
APPENDIX
A ANALYSIS OP 001$ POR TCDD 44
B TOXICOLOGICAL TESTS ON CCl^ RESIDUES 49
C USAP/EPA INTERAGENCY AGREEMENT 61
-------
I - CONCLUSIONS
A. High temperature chlorinolysis under the proper conditions effec-
tively converts Herbicide ORANGE and the contained
tetrachlorodibenzo-p-dioxin to carbon tetrachloride, carbonyl
chloride, and hydrogen chloride (gas).
B. Destruction of tetrachlorodibenzo-p-dioxin is complete to the
detectable limit of <10 parts per trillion (ppt) in a single pass
through the reaction-refining system.
C. Single pass operation of the reactor results in the production of
some hexachlorobenzene and chlorinated polybenzene residue.
D. A residue-free process is feasible because residues from single pass
operation when mixed with fresh feed and passed through the reactor
were converted to carbon tetrachloride.
E. Preliminary toxicology tests of the recovered carbon tetrachloride
on rabbits and rats showed no evidence of tetrachlorodibenzo-p-dioxin
contamination.
F. The chlorinolysis process as outlined offers an environmentally safe
method of disposing of Herbfcide ORANGE.
G. The chlorinolysis process offers an opportunity to convert many
environmentally unacceptable materials to useful industrial chemicals
and to conserve resources.
1
-------
II - RECOMMENDATIONS
Chlorinolysis is recommended as a resource saving method of convert-
ing Herbicide ORANGE and the contained tetrachlorodibenzo-p-dioxin to
useful industrial chemicals, carbon tetrachloride, hydrogen chloride,
and carbonyl chloride.
Investigation of the applicability of the chlorinolysis process for
the disposal of other materials, such as DDT, Lindane, Dieldrin, PCBs,
and certain military chemical agents is also recommended.
-------
Ill - INTRODUCTION
Herbicide ORANGE, a military defoliant, was withdrawn from use in
Vietnam in 1970 because of potential teratogenic effects. About 2.3
million gallons of this materiel was on hand when further use was
banned. Since this time, the U. S. Air Force (USAF) has been attempt^
ing to dispose of this materiel inventory.
Herbicide ORANGE is a 50:50 mixture of the n-butyl 2,4-dichloro-
and 2,4,5-trichlorophenoxyacetates. A small concentration of
tetrachlorodibenzo-p-dioxin (dioxin), an artifact in the preparation of
2,4,5-trfchlorophenoxyacetic acid, is present in ORANGE. Because
unrestricted use of 2,4,5-T herbicides was banned by EPA and dioxin is
a highly toxic teratogen, the method for disposing of ORANGE must be
environmentally acceptable.
Environmental acceptability precludes the normal use for control
of vegetation on ran gel and, pastures, and rights-of-way because the
dioxin content of ORANGE exceeds the 0.5 ppm limit tentatively
recommended by EPA, Thus, the methods of disposal available are reduced
to incineration and chemical conversion.
Presently known incineration processes are not completely acceptable
when assessed against the zero pollutant discharge concept of environ-
mental statutes, particularly with respect to potentially hazardous
materials.
Prior to 1972, the only known chemical conversion process, base
hydrolysis, e.g., treatment with caustic or an organic base, was equally
-------
unacceptable because of the large volumes of aqueous wastes that are
produced. In 1972, a new process was developed in the laboratories of
Diamond Shamrock Corporation for the conversion of refractory
chlorocarbon wastes to useful industrial chemicals. This process is
completely closed, that is, no vapor, solid or liquid discharges are
generated other than conversion products. Through the auspices of EPA
and with support by the Air Force, a research program was developed to
evaluate the applicability of this process to the disposal of ORANGE.
Inherent in this program were requirements to define suitable con-
ditions, to delineate a process to recover the products, and to demon-
strate that the recovered products were free of dtoxin. Evaluation of
quality and safety of the recovered carbon tetrachloride would be accom-
plished by instrumental analysis and toxicology effects found in test
rabbits and rats exposed to the product. Because of the diverse and
complex nature of the various evaluations, the program was conducted in
cooperation with the Bioeffects Branch of EPA in Chamblee, Georgia,_and
the Pesticide Degradation Laboratory of the USDA in Beltsville, Maryland.
The project was coordinated by the Office of Research arid Development of
EPA through its Industrial Pollution Control Division, Washington, D. C.
and the Southeast Environmental Research Laboratory, Athens, Georgia.
The overall objective was to demonstrate that the chlorinolysis
process would convert ORANGE and the contained dioxin to carbon
tetrachloride, carbonyl chloride, and hydrogen chloride. The allowable
residual dioxin content in the refined carbon tetrachloride was set at
-------
10 ppt maximum (the proven limit of detectability). Instrumental
analysis of the refined carbon tetrachloride was performed in the USDA
laboratory by Dr. E. A. Woolson under the direction of Dr. P. C. Kearney.
lexicological evaluation of the refined carbon tetrachloride was per-
formed by Dr. R. D. Kimbrough, Acting Chief of the Bioeffects Branch of
EPA.
-------
IV - THEORETICAL CONSIDERATIONS
The overall chemical reactions projected to occur in the
chlorinolysis of ORANGE are shown in Figure 1. However, several inter-
mediate reactions may be expected to occur with their products also
being destroyed under the reaction conditions. Mechanistic analysis of
the reactions indicates that hexachlorobenzene will be formed as an
intermediate and because of its inherent stability can be expected to
\
\
be present in the product mix from a reactor of a finite size.
Analysis of the thermodynamfc data available indicates that carbon
tetrachloride (CCl^, hexachlorobenzene (HCB) , and chlorine (C^) exist
in equilibrium at the reaction conditions proposed for chlorinolysts:
HCB + 9 C12 ^ 6 CC14 (1)
. in.tr* (2)
where K = exp (-78.6 + ) (3)
and K = equilibrium constant
P = Partial pressure, atm
T = Absolute temperature, in °K
This relation is of concern because, in any practical operation,
-------
FIGURE 1
THEORETICAL REACTIONS FOR
HERBICIDE ORANGE CHLORIHOLYSIS
H
H
o
CL
H
27 Ct2 » 9 CCuj + 14 HO. + 3 COCL£
CL
N BUTYL - 2,4-DlCHLOROPHENOXYACETATE
«
26 CL2 > 9 GCU| + 13 HCL-+ 3 COO.2
H
N BUTYL - 2/4-5-TRICHLOROPHEMI1KYACETATE
+ 22 fig—> 10 CCL/j + 'I HCL + 2 COO.2
CL
.(2/3,7*8 TETRACHUK?ODIBENZO-P-DIOXIN)
7
-------
HCB will be found 1n the product from the reactor. Examination of the
equilibrium expression as shown 1n Equation 4 gives clues to reaction
conditions necessary to minimize the HCB content 1n the reactor effluent.
XHCB - - (4)
K(xcl2)9
where:
x • mole fraction
Tf- total pressure, atm.
Favorable effects will obviously be realized by operating at high
chlorine concentration and elevated total pressure. High carbon tetra-
chloHde concentrations and elevated temperature will tend to produce
unfavorable equilibrium effects. However, elevated temperatures were
found to be desirable from a process standpoint because they promote high
rates of reaction, I.e., highly favorable kinetics at the expense of
equilibrium conversion.
Because of the necessary balance between rates of reaction (reactor
size) and HCB content of the effluent, 1t 1s obvious that some recycle
of unconverted HCB from the product recovery system back through the
reactor may be necessary to ensure that no buildup of this environmentally
unacceptable chemical will occur. The trade-off will essentially be
between reactor size and recycle stream size or possibly allowable HCB
concentration 1n the feed mixture.
8
-------
V - DISCUSSION OF RESULTS
A. Feed
ORANGE containing 14 ppm of 2,3,7,8 tetrach1orod1benzo-p-d1ox1n
supplied by the USAF was used throughout this study. Analyses of three
replicate samples of the material supplied (single container) performed
by the US DA, showed an average dloxln content of 14.8 ppm (actual
values were 15.8, 13.6, 14.9 (see Appendix A)). Commercial grade carbon
tetrachloride was used to dilute the herbicide as needed to moderate the
reaction temperature.
B. Experimental Results
Several attempts to feed ORANGE on an "as received" basis were
unsuccessful because prereaction and carbonization occurred at the Inlet
end of the reactor. Dilution of ORANGE with carbon tetrachlorlde to
reduce the reactivity of the feed resulted 1n satisfactory performance
of the system. All subsequent test work was done with ORANGE diluted
on a 1:1 by volume basis.
Reactor performance was studied over a range of conditions. Critical
factors 1n this feasibility test were chlorine-to-carbon feed ratio,
temperature, pressure, and retention time. The ranges of these variables
studied are summarized 1n Table 1.
TABLE 1 - REACTOR CONDITIONS DURING FEASIBILITY STUDIES
Chlorine to Carbon Ratio, C12/C 4.4 -7.2
Temperature, °C 600 - 800
Pressure, pslg 225 • 300
Retention Time, minute 0.5 • 1.0
-------
Single pass tests were made to establish comparisons of reactor
performance prior to attempting recycle operation. A summary of perti-
nent experimental results is shown in Table 2. Several trends can be
observed in these results. As predicted by the CCl^HCB equilibrium
relationship, an increase in system pressure, while holding all other
conditions constant, produces an increase in 0014 yield and a reduction
in HCB yield (see liests 46 and 52). The reduction in HCB yield is
desirable as it reduces the volume of recycle required to convert the
hexachlorobenzene by-product make.
The effect of increased C12/C ratio upon selectivity to carbon
tetrachloride is also favorable as evidenced by the results of tests 46
and 54.
Adjusting reaction temperature may be expected to have two effects
in this reaction. The first effect postulated is that increasing tem-
perature will increase the concentration of HCB at equilibrium. Esti-
mates of these concentrations from equations 3 and 4, shown in Table 3,
indicate that this effect will be inconsequential in a reactor of finite
size.
10
-------
Table 2
Summary of Selected
Run
No.
46A
52A
54A
58A
68A
72A
74A
100B
102B
Dura-
tion
Hrs
13.5
9.0
16.0
15.5
12.0
5.0
5.5
10.0
5.0
A. 50:50
Organic
Feed
Rate
qms/hr
69
54
101
202
46
63
78
78
58
.3
.9
.0
.3
.2
.5
.0
.0
.6
by volume
B. Mixture of
Temp
°C
700
700
700
700
700
800
600
•800
800
mixture
Press
psig
300
225
300
300
300
300
Ret.
Time
min
Le Pass
1.0
1.0
1.0
0.5
1.4
1.0
Chlorinolysis
Runs
Orange
Carbon
Account-
Conv. selectivity to ability
C17/C % CCl^ HCB COCls %
Oper
6.8
7.0
4.6
4.4
5.7
8.0
100
100
100
100
100
100
300 1.0 6.7 100
•Simulated Recycle Operation-
300 1.0 7.2 100
80
300 1.0
of Orange and
Orange-residues and.
CC14
9.0
CC14
(See
100
81
analysis
63
52
53
46
SB
69
41
75
100
75
100
in Table
8
12
16
26
9
4
17
0
0
0
0
3
100
94
101
100
99
25 104
90
25}* 97
25>* 89
0'
for simulated rec
operation.) Residues are essentially HCB in
The Upper figures represent the conversion and selectivity of Orange tp products.
The lower figures represent the conversion and selectivity of residues (HCB)
to products.
-------
TABLE 3
EFFECT OF TEMPERATURE ON
EQUILIBRIUM HCB CONCENTRATIONS
TEMP
°C
700
700
700
800
800
800
PRESS .
psig
300
300
300
300
300
300
C12/C
5
6
7
5
6
7
WT% HCB IN
PRODUCT*
0.5 x 10-7
0.15 x 10-7
0.05 x ID'7
0.4 x 10-4
0.11 x 10~4
0.04 x 10-4
*Product excludes HC1, excess C12, and COC12
The second effect, which is the most significant, is the increased
rate at which the reaction proceeds as the temperature is increased.
Thus, with all other conditions being fixed, an increase in temperature
should permit a closer approach to equilibrium conditions. This effect
is shown by the results of tests 46 and 74 in which operation at 700°C
results in a lower HCB concentration than operation at 600°C.
The effect of contact time when the feed is completely converted is
similar to that of temperature. Increased contact time permits a closer
approach to equilibrium in a finite size reactor. The selectivity to
12
-------
HCB is reduced as contact time is increased. The results of tests 54
and 58 demonstrate this effect.
These results all indicate that maximum selectivity to CC14 will be
achieved at the higher end of the range of conditions studied. To demon-
strate this effect in single pass operation, test 72 was made. The
results show the highest selectivity to carbon tetrachloride and lowest
selectivity to hexachlorobenzene of all the single pass tests.
Recycle operation to demonstrate that residues from the carbon
tetrachloride recovery distillation could be processed effectively was
also examined. The feed for these tests was made up by blending two
volumes of 0014, with one volume of ORANGE plus residue. This feed had
the composition and density as shown in Table 4. The high percentage of
was necessary to ensure complete solubility of the HCB.
TABLE 4
SIMULATED RECYCLE OPERATION - RESIDUE CONVERSION
Percent By Weight
Specific Gravity
Run No. HCB CC14 ORANGE at 20°C
100
102
7.5
7.2
66.3
59.0
26.2
33.7
1.550
1.524
13
-------
Operating conditions for these two tests are shown in Table 2. Several
one hour off-gas samples were collected by total condensation for each
fif these runs. The HCB balance was then examined to determine what
portion Of the HCB in the feed was converted. The results for tests
100 and 102 show HCB conversions of 80 and 81 percent, respectively.
An overall material balance for recycle operation based upon the
results of test 100 is shown in Figure 2. Ninety-four percent of the
carbon fed to the reaction system as ORANGE was recovered in the reactor
effluent* The 6 percent of carbon unaccounted for in the material
balance is attributed to metering erros, work-up of samples, and weigh-
ing errors.
14
-------
FIGURE 2
MATERIAL BALANCE FOR
HERBICIDE ORANGE CHLORINOLYSIS
WITH HCB RECYCLE
1,000 G ORANGE
25.300 G 0.9
7 CL2/C
800 °C
300 PSIG
1 HIM
1.010 G CQCL9 r
19.165 G Ct9
1.680 G HCL
4,314 6 CCL/i
55 G HCB
2,575 G CCU|
GAS ANALYSIS
fvnn
3,1%
82,8*
14,1*
CARBON BALANCE
IE DHL
ORANGE - 490 G COCL2 - 123 6
CCil| - 336 q
159 G
94*
-------
C. Carbon Tetrachloride Product
1. Pioxi n Content
Six 2 liter samples of carbon tetrachloride were recovered from
the various runs made. Five of these samples represent the results
attainable with single pass operation of the reaction system and the
sixth sample represents the results of the recycle operation, in which
distillation residues from the preceding single pass runs were incorpor-
ated in the feed to the reactor. These samples were analyzed for dioxin
content by the USDA laboratory.
Their final report, which is appended [See Appendix A), shows
that the chlorinolysis process effectively converts dfoxln and ORANGE to
carbon tetrachloride. The analytical results and run conditions are
summarized in Table 5. It should be noted that each of the samples of
carbon tetrachloride are composites of several runs. It should also be
noted that during the early phase of the analytical program the method
of analysis was capable of detecting 10 parts per trill ton of dioxin
in pesticide quality carbon tetrachloride. The first three samples,
AD-1, -2, and -3, of carbon tetrachloride recovered from the chlorfnolysis
were contaminated with silicone grease from the distillation apparatus.
The presence of the grease interfered with the dioxin analysis and the
sensitivity of the analytical procedure was limited to 0.1 - 1.0 parts
per billion dioxin. Evaluation of samples of AD-2 and -3 using combined
gas chromatograph/mass spectrometry indicated no detectable dioxin.
The carbon tetrachloride recovery system was refined to eliminate
16
-------
sill cone grease contamination. Samples AD-4, -5, and -6 were then pre-
pared. Gas chromatographic analysis of these samples showed "no detect-
able" dioxin at a sensitivity of OO parts per trillion. The analytical
results are summarized in Table 5.
Analysis for dioxin in the residues from the carbon tetrach-
loride distillation was complicated by the presence of hexachlorobenzene
and chlorinated polybenzenes. Thus, the level of detectabfltty of
dioxin in the residues was in the parts per billion range. Analysis of
the two samples of residue submitted, RD-1 and RD-4, showed dioxin
levels less than 100 parts per billion. These materials would normally
be recycled through the reactor. The analytical results are summarized
in Table 6.
2. Toxicology
Preliminary toxicology tests of the carbon tetrachloride
samples AD-5 and -6 were made on rabbits and rats. The results of these
tests indicated no ill effects attributable to dioxin. A report sum-
marizing the results of these tests is appended (See Appendix B).
Further work in this area using larger animal populations is necessary
to confirm these results.
17
-------
TABLE 5 - DIOXIN CONTENT OF RECOVERED CC14
SAMPLE
NO.
AD-1
AD-2
AD-3
AD-4
AD-5
AD-6
N.D. -
TEST TEMP PRESS.
NO. °C psig C12/C
1- 8 800 150-225 3/1-10/1
9-15 600-700 300 7/1-30/1
16-21 700 225-300 4/1-13/1
22-30 600-800 300 5/1- 8/1
31-37 700 300 5/1- 7/1
38-43 700-800 300 4/1- 9/1
._._ ' ' r'*~JSr ^ i
None Detected
DIOXIN CONTENT, ppt
GC GC-MS
< 100 N.D.
<1000 N.D.
<100 N.D.
N.D.
N.D.
N.D.
TABLE 6 - DIOXIN CONTENT OF RESIDUES
SAMPLE
NO.
h
RD-1
RD-4
TEST TEMP PRESS.
NO. °C psig
1-8 800 150 - 225
22-30 600-800 300
DIOXIN
CONTENT
C12/C ppm
3/1 - 10/1 N.D.
5/1 - 8/1 <0.1
N. D. - None Detected
18
-------
VI - PROCESS ECONOMICS
The 2.3 MM gallons of Herbicide ORANGE can be converted into the
following products:
Carbon Tetrachloride 125.3 MM Ibs
Carbonyl Chloride 26.9 MM Ibs
Hydrogen Chloride 44.6 MM Ibs
The gross economics were examined on the basis of a price range for
each of the products and raw materials involved (See Table 7). The
values used were chosen to reflect optimistic, pessimistic, and most
probable prices necessary to market the products in a competitive situa-
tion, recognizing that these materials would be available only over an
assumed two year period for ORANGE processing.
A plant capacity of 25 tons of herbicide per day was chosen to
prepare a capital estimate. The flow scheme of this plant is shown in
Figure 3. The investment for this plant was estimated to be $6.0 MM.
The sales and cost of sales for each case are shown in Table 8. Also
shown is the total cost for conversion of Herbicide ORANGE assuming that
the plant is written off . These figures show that the total cost will
be about $11 MM in the worst case, about $8 MM in the most probable case,
and about $4 MM in the best case. These costs can be mitigated to some
degree by erecting a plant dedicated as a regional disposal unit to be
operated over a 10-year period. This would reduce the annual deprecia-
tion charges and provide the opportunity to establish a more favorable
market!ng posi tf on.
19
-------
TABLE 7 - HERBICIDE DISPOSAL PRICING ASSUMPTIONS
PRODUCT AND RAW MATERIAL PRICES
CHLORINE 50- 55- 60 $/TON
CARBON TETRACHLORIDE 60- 75- 90 $/TON
CARBONYL CHLORIDE 0- 100- 200 $/TON
HYDROGEN CHLORIDE (30)- 0- 30 $/TON
*( ) Denotes negative value, i.e., cost for disposal.
20
-------
FIGURE 3
HERBICIDE ORANGE DISPOSAL
TOTAL CHLORINOLYSIS
FLOW SHEET .
HCL
CONDENSER
ro
ORANGE
CL2
CHLORINOLYSIS
REACTOR
CL2. RECYCLE
COCL2
CONDENSER
HEAVIES. COCL2
RESIDUE RECYCLE
r—-MX
PRODUCT
QUENCH
COLUMN
HCL-CL2
SEPARATING
COLUMN
COCL2-CCuj
SEPARATING
COLUMN
REFINING
COLUMN
-------
TABLE 8 - HERBICIDE ORANGE DISPOSAL
GROSS ECONOMICS TOTAL
SALES
cci4
COC12
HC1
TOTAL
COST OF SALES (EXCLUDES DEPRECIATION)
cci4
COC12
HC1
TOTAL
REVENUE* J
CAPITAL
NET TOTAL COST ~1
CHLORINOLYSIS
WORST
CASE
3.759
0
(.669)
3.090
7.800
0.094
0.0
7.894
[4.804)
6.0
1 1 • •*•
($MM)
PROBABLE
CASE
4.698
1.245
0
6.043
7.317
0.094
0.139
7.550
(1.507)
6.0
BEST
CASE
5.638
2.690
0.669
8.997
6.834
0.094
0.139
7.067
1.930
6.0
""'
*( ) means expense, i.e., negative revenue.
22
-------
Further refinement of the technology is necessary to establish a
firmer estimate of the plant investment and potential usefulness as a
regional disposal facility. In addition, an extensive marketing and
raw material Csuitable chlorocarbon waste) survey must be made before
any technical decision to construct a full scale facility for ultimate
write-off as a regional disposal unit can be justified.
23
-------
VII - EXPERIMENTAL EQUIPMENT AND OPERATING PROCEDURE
A. Chlorinolysis Reaction System
1. Equi pment
The equipment for bench-scale chlorinolysis of Herbicide
ORANGE, shown schematically in Figure 4, consisted of a quartz tube
1-inch in diameter by 48-inches long. Nickel stuffing boxes were used
on the ends of the reactor. The entire reactor and stuffing box assembly
was mounted on a Unistrut frame using standard Unistrut pipe clamps to
restrain the stuffing boxes. This Unistrut frame was positioned external
to an electrically powered oven used to heat the quartz reactor. Sec-
tionalized, individually controllable, clamshell heaters were used to
*
heat the reactor tube. Heater temperatures were monitored by thermo-
couples inserted through the heater walls. Two additional thermocouples
were provided in a clear quartz thermowell mounted axially in the reactor
to measure the reaction temperature profile.
A metering balance and air operated motor valve were used to control
the chlorine flow to the reactor. Chlorine from a pressurized cylinder
was vaporized, filtered and then passed through the control valve. The
chlorine was then preheated to its feed temperature in an electrically
heated nickel tube prior to mixing with the organic feed stream.
24
-------
FIGURE 4
Bench-Scale Chloronolysis Unit
N)
in
VfcNT
fctAM
K.o. POT
-------
The organic materials were fed to the reactor at a controlled
rate by a Whitey diaphragm metering pump with a steam heated pumping
head. The flow rate of organics to the reactor was monitored by pumping
out of a burette. Readings were taken over specified time intervals.
When the organic feed was not normally liquid at room temperature, as in
the case when hexachlorobenzene was present in the mixture, the feed
vessel was changed to a heated, pressurized, calibrated glass pipe. In
all cases, the feed lines were heated with electrical heaters.
The feed lines for the organics and chlorine were 1/8-inch and
1/4-inch diameter tubing, respectively. (Nickel was used for the chlorine
and stainless steel for the organics.)
Products from the reactor passed through a monel control valve
which maintained the reactor pressure at the desired level. The pressure
sensing tap for the pressure control system was located on the chlorine
feed line. The products after the control valve were either totally
condensed, for samples, or sent to a caustic scrubber for disposal.
When the products were sent to the scrubber, they were first passed
through a nickel knockout vessel where solids, such as hexachlorobenzene,
which would plug the scrubber, were removed. The knockout pot was
charged with hexachlorobutadiene or carbon tetrachloride to dissolve the
solids. This material was replaced periodically with fresh solvent. The
spent solvent was treated with caustic. All the vent lines from the
reactor were heated with electrical heaters to minimize plugging.
26
-------
2. Operating Procedure
Chlorinolysis was carried out by feeding a mixture of chlorine
and the organic to be investigated directly into the quartz reactor tube.
The system was started up by establishing an approximate temperature
profile in the reactor using the external electric heating mantles. The
chlorine flow was then initiated at the desired rate and the reactor
temperature readjusted. The organic flow was then established at a
predetermined rate to meet a specified chlorine to carbon ratio.
An axial temperature profile was then determined using the two
moveable thermocouples in the reactor centerwell. Reaction conditions
and feed rates were gradually fine-tuned to the final specified condi-
tions. Sampling was initiated following lineout at the final settings.
Sampling of the reactor effluent was accomplished by complete
condensation of all components except HC1 using a dry ice-isopropanol
cooled condenser, as shown in Figure 5. Chlorine was refluxed into the
packed stripping column to ensure recovery of all organic components.
To accomplish sample collection, the sampling system was connected to
the reactor outlet piping downstream from the pressure control valve.
The condenser was vented back to a caustic scrubber during sampling
(See Figure 4). A sample was started by simultaneously closing the
valve in the knockout pot line and opening the valve to the sample port,
thereby forcing all vapors into the sample collection unit. Samples were
collected for a specified time, usually 15, 30, or 60 minutes, accord-
ing to the organic feed rate. The longest sample period were used at
the lower flow rates. After completing the collection of a sample, the
27
-------
valving sequence was reversed and the reactor effluent redirected to
the knockout pot. The sampling system (Figure 5) was then disconnected
from the reactor vent line. The collection flask was then removed from
the sampling apparatus and weighed. After weighing, the sample flask
was connected to the distillation apparatus (Figure 6) and a distilla-
tion performed to remove residual hydrogen chloride, chlorine, and
carbonyl chloride. All glassware used in the sample collection and
distillation was taped with electrician's tape to prevent the occurence
of ultraviolet initiated chlorination during sample collection and
preparation.
28
-------
Figure 5
Collection System
COI.L.E.C.TIOM POT
29
-------
Figure 6
Chlorine Distillation system
30
-------
B. Carbon Tetrachloride Recovery
1. Equipment
A standard 2-inch diameter by 30-tray perforated-plate Oldershaw
distillation column was used for recovery of carbon tetrachloride. The
column was equipped with a liquid dividing head, a 5-liter kettle, product
receiver, and thermometers in both the kettle and the head. The kettle
was provided with a Glas-Cal mantle for heating purposes (See Figure 7).
2. Procedure
The kettle was charged with 3 to 4 liters of degassed crude
product, i.e., the pot residue from the chlorine distillation (Figure 6).
Heating was initiated and cooling water applied to the condenser. The
system was allowed to both come up to temperature and load the column
while on total reflux. When stable conditions were established, the
reflux setting was adjusted to give a reflux to make ratio of 10:1.
The initial cut removed was methylene chloride which had been added as
part of the product collection procedure. Following the removal of the
methylene chloride, as indicated when the vapor temperature at the head
had attained the boiling part of CCl^, collection of carbon
tetrachloride was initiated and continued until 2 liters of product had
been accumulated. The system was then shut down and the contents of
the kettle allowed to cool.
31
-------
Figure 7
Carbon Tetrachloride Recovery Still
CONDENSER
30 TRAY
OLDERSHAW
COLUMN
LIQUID
DIVIDING
HEAD
PRODUCT
RECEIVER
OPERATING CONDITIONS
PRESS.- ATMOSPHERIC
REFLUX RATIO - 410/1
DISTILLATION
POT
HEATING
MANTLE
32
-------
VIII - ANALYTICAL METHODS
V
A. General Procedure and Equipment
The liquid portion of the reaction product after degassing to
remove hydrogen chloride, chlorine, and carbonyl chloride, generally
consisted of carbon tetrachloride and dissolved and undissolved hexa-
chlorobenzene in methylene chloride solvent. This necessitated a
multiple step treatment of the sample to obtain an analysis. Vapor
phase chromatography (VPC) was used to determine the CCl^ and dissolved
HCB content of the liquid portion of the sample. The undissolved HCB
was recovered by filtration, washed with methylene chloride, air dried
and weighed.
All VPC analyses were performed on a Model 5720 Hewlett-Packard
gas chromatograph equipped with a 1/4-inch diameter by 10-foot long
column packed with 8-percent OV210 on "Gas Chrom Q" support. The carrier
gas was helium, and a thermal conductivity cell detected the components.
Details of the procedure follow.
The gaseous portion of the reaction product consisted of hydrogen
chloride, chlorine, and carbonyl chloride. This mixture was analyzed
by the wet chemistry procedure outlined.
B. Detailed Sample Analysis
1. Analysis of Chlorinolysis Samples - Non-Gaseous Components
Standards
Three sets of standards were required as follows:
a. Carbon Tetrachloride - CC14
33
-------
(1) A standard containing 25.00 g CC14 diluted to 100 ml
with methylene chloride.
(2) A standard containing 50.00 g CC14 diluted to 100 ml
with methylene chloride.
b. Hexachloroethane - HCA
(1) A standard containing 6.00 g HCA diluted to 100 ml
with perchloroethylene.
(2) A standard containing 12.00 g HCA diluted to 100 ml
with perchloroethylene.
c. Hexachlorobenzene - HCB
(1) A standard containing 0.50 g HCB diluted to 100 ml
with perchloroethylene.
(2) A standard containing 1.00 g HCB diluted to 100 ml
with perchloroethylene.
No standard was used after it was three weeks old. Loss of
solvent by evaporation and the possibility of contamination made stan-
dards suspect when they were used beyond this length of time.
VPC
The analysis was carried out on a 5720 dual column Hewlett-
Packard gas chromatograph equipped with a 10 ft. x 0.25 in. column packed
with 8-percent OV210 on "Gas Chrom Q." The carrier gas was helium, and
a thermal conductivity cell detected components as they eluted.
Sample Preparation
Samples of unusual color or solid content were noted. The sample
was diluted exactly to the mark with methylene chloride and well mixed
34
-------
by inverting the volumetric flask ten times immediately prior to analysis
The total volume was recorded.
Sample Preparation
Samples of unusual color or solid content were noted. The
sample was diluted exactly to the mark with methylene chloride and well
mixed by inverting the volumetric flask ten times immediately prior to
analysis. The total volume was recorded.
Analysis Procedure
a. Carbon Tetrachloride
(1) The VPC column temperature was adjusted to 80°C
Recorder - on 1/4 in/min
Filament Current - 165 ma
Helium Flow - 30 ml/rnin
Attenuator - 128
(2) Duplicate 10 jjl shots of the 25.0 g CC14 standard were
run. If the peak heights of CC14 did not agree within
+2 chart divisions, shots were repeated until the
agreement was satisfactory.
(3) Duplicate 10 yl shots of the 50.0 g CC14 standard were
run. Again, agreement between the shots had to be
+2 chart divisions for CCl^. -
(4) A 10 jil shot of unknown was run. If the peak remained
on scale, a duplicate shot was made. If the peak went
35
-------
off scale, the shot size was reduced sufficiently to
keep the peak on scale. Duplicate shots were required
to agree within +3 chart divisions for CC14.
b. Hexachloroethane
(1) The column temperature was raised to 180°C. The
attenuator was lowered to 64.
(2) The duplicate HCA standards were shot in the same
manner as used in the CCl^ analysis. Peak heights had
to agree within +1 chart division.
(3) The unknowns were run in the same manner as for the
CC14 analysis above. Peak heights for duplicate shots
were required to agree within +2 chart divisions.
c. Insoluble Hexachlorobenzene
(1) If no insoluble HCB was present, we went directly to
Step d.
(2) The insoluble HCB was filtered through a tared filter
paper. The solids were rinsed with 20 ml of methylene
chloride.
(3) The filtrate and washings were retained for Step d.
(4) The solids were air dried to constant weight. The
solids plus paper were weighed. The difference
between total weight and weight of filter paper was
the weight of insoluble HCB.
d. Soluble Hexachlorobenzene
(1) The filtrate from Step c was transferred to an
36
-------
appropriate sized volumetric flask and diluted to the
mark. The total volume was noted.
(2) The VPC column temperature was raised to 240°C, and
the attenuator lowered to 4.
(3) Duplicate shots of the 0.50 and 1.00 g HCB standards
were made. Peak heights were required to agree
within +2 chart divisions.
e. Data Treatment
Peak heights were measured by drawing the most suitable
base line and dropping a perpendicular from the highest point on the peak
to the base line. This perpendicular was taken as the peak height.
Calculations were carried out on a Wang Calculator equipped with an
80-step card programmer. The data were entered manually on the keyboard,
and the programmer carried out the manipulation of the data. Basically,
the programmer automatically carried out the computation of the
constants, a and b, for the equation y = ax + b using the data from the
standard shots. In the linear equation, y was taken to be grams of
component per 100 ml of sample, while x was the observed peak height.
The constants were stored and automatically recalled during the calcula-
tion of an unknown. This technique virtually eliminated computational
errors, in addition to dramatically decreasing the time required. Appro-
priate corrections to the calculated answer were carried out manually if
the standard and unkn'owns were in different sized volumetric flasks or if
standard and unknown.shot sizes were not equal.
37
-------
2. Analysis of Chlorlnolysls Samples - Gaseous Components
The following procedure presupposes that no Interferences are
present. Hexachlorobenzene and carbon tetrachlorfde do not Interfere,
but C02 does.
Principle
When a mixture of chlorine, carbonyl chloride, and HC1 1s
reacted with an excess of NaOH, the following reactions will take place:
C12 + 2 NaOH-*- NaCl + NaOCl + HgO (5)
C12CO + 4 NaOH-* 2 NaCl + Na2C03 + 2 H20 (6)
HC1 + NaOH—*• H20 + NaCl (7)
The NaOCl 1s determined by an lodometrfc tltratlon and calcu-
lated back to chlorine according to (5) above.
The Na2C03 1s determined by precipitation with BaCl2, filtered
off, titrated with standard HC1, and calculated as carbonyl chloride
according to (6) above. Finally, total chlorine 1s determined with
AgN03, the chlorine from carbonyl chloride and chlorine substracted, and
the difference calculated as HC1.
Reagents and Materials
a. A/IN NaOH
b. 0.-1N HC1
c. Phenol phthaleln
d. KI
e. (1 + 1) Sulfurfc add, at room temperature
f. 0.1N Sodium thlosulfate
38
-------
g. Starch solution,^0.3%
h. 0.1 N silver nitrate
1. 0.1N thlocyanate
J. FerHc alum., 40fc
k. Nitrobenzene
1. 302 Hydrogen peroxide
m. 250 ml volumetric flasks
n. Plpets (50, 100 ml)
o. Glass funnel (about 50-100 ml capacity)
p. 500 ml Erlenmeyer Flasks
q. 105& Barium chloride solution
r. Centrifuge (250 ml tubes)
Procedure
a. Make sure that the volume, pressure, and temperature of the
gas mixture are known. For convenience 1n the equations (8-10), we have
assumed 30°C and 760 mm of mercury.
b. Absorb all the gases 1n an excess of IN NaOH, This 1s
done by cooling the gas bulb with 1ce for 5-10 minutes, attaching a
funnel with rubber tubing to one of the stopcocks, pouring the excess
of NaOH on the funnel, and carefully opening the stopcock to let the
NaOH Into the bulb. Do not allow air Into the bulb. Close the stopcock.
Shake well to absorb all the mixture.
c. Transfer all the NaOH to a 250 ml flask, wash the gas bulb
and add the washings to the flask. Make up to 250 ml and mix well.
d. Place 100-150 ml of water 1n an Erlenmeyer flask, add 4-5 g
39
-------
of KI, dissolve, add an aliquot of the NaOH solution from Step c above,
(50 ml suggested) mix well, add about 5 ml of 0 + 1) sulfuric actd Can
excess), and titrate the liberated iodine with thiosulfate and starch.
(Vol. of thiosulfate = V-j, normality = N-j, aliquot factor = fj).
e. Transfer another aliquot of the NaOH solution from Step c
(150 ml suggested) to a 250 ml centrifuge tube, add a slight excess of
BaCl£ solution, stopper, mix well, and centrifuge off the precipitate of
barium carbonate. Carefully, pour out the liquid, add some BaCl2 solu-
tion to wash the precipitate and centrifuge again. Discard the liquid.
f. Add a slight, measured excess of 0.1N HC1 to the precipi-
tate in the centrifuge tube and dissolve. Transfer all to a beaker,
wash the tube, boil to expel C02, cool, and backtitrate the excess of
HC1 with 0.1N NaOH and phenolphthalein. (Volume of HC1 = V2,
normality = N£, volume of 0.1N NaOH = V3, normality = N3, aliquot
factor = f£, from Step e above.)
g. Transfer another aliquot of the NaOH solution from Step c
above (25 ml suggested) to a 500 ml Erlenmeyer, carefully add 30% hydro-
gen peroxide dropwise with stirring to decompose all the hypocKlorites
until no more effervescence occurs where the drop falls. Then add 2-3
more drops. Mix well.
h. Add 100 ml of water, mix, and boil for 10 minutes.
i. Cool to room temperature or below, acfdify with a slight
excess of (1 + 1) sulfuric acid, and titrate the chlorides with silver
nitrate and thiocyanate by the Volhard procedure (Volume of silver
40
-------
nitrate = V4, normality = N4, volume of thiocyanate = Vg, normality =
N5, aliquot factor = f3, Step g above.)
j. Calculate the milliliters of chlorine, carbonyl chloride,
and hydrogen chloride in the whole sample by the following equations
(30°C, 760 mm mercury):
ml of chlorine = 12.4 ^ V-, N-, (See Step d) (8)
ml of carbonyl chloride = 12.4 f2 (V2N2 - V3N3) (9)
(See Step f)
ml of hydrogen chloride = 24.8 f3 (V4N4 - V5N5) - 2 (10)
(ml of C12 + ml of C12CO)
(See Step i)
k. Once the above results have been calculated, determine the
percentage of each component in the mixture.
Notes
a. The above equations give the milliliters of each gas in
the whole gas bulb taken for analysis, at 30°C and 760 mm of mercury.
b. Results for chlorine and carbonyl chloride are considered
to be reasonably accurate. Since HC1 is reported by difference, it
might be in error as happens in all indirect procedures.
c. The ideal gas law was assumed to be applicable to each com-
ponent of the mixture.
41
-------
IX - DISTILLATION RECOVERY OF n-BUTYL 2,4-DICHLOROPHENOXYACETATE
During the chlorinolysis study it occurred to the author that
recovery of the n-butyl 2,4-dichlorophenoxyacetate portion of ORANGE
might be achieved by distillation. Although this approach to treatment
of the problem was not part of the contract, some work was done to
determine if this approach was feasible.
Investigation of the literature indicated that there was sufficient
difference between the boiling points of the major components in ORANGE
to permit a separation to be made.
Such a distillation was actually performed using an Oldershaw
perforated plate column operated under vacuum. This distillation
demonstrated that about 80-percent of the n-butyl 2,4-
dichlorophenoxyacetate contained in ORANGE could indeed be recovered as
a distillate. The tetrachlorodibenzo-p-dioxin content of the distillate
was demonstrated by USDA analysis to be less than 1 part per billion.
An overall material balance and operating conditions for this distilla-
tion are shown in Figure 8.
The use of a distillation procedure to pretreat ORANGE could
materially reduce the volume of material to be processed by chlorinolysis
and provide directly a useful and valuable herbicide fn addition to the
carbon tetrachloride, hydrogen chloride, and carbonyl chloride products.
Economics of such a combined process for ORANGE disposal appear to be
superior to those of chlorinolysis alone.
42
-------
FIGURE 8
DISTILLATION OF HERBICIDE ORANGE
1,340 G
3,912 G ORANGE
50% 2,4-D ESTER
50% 2,4,5-T ESTER
M PPM DIOXIN
1,275 G 2,4-D ESTER
65 G CHLOROPHENOLS
<1 PPB DIOXIN
OPERATING CONDITIONS
NUMBER OF PLATES - 20
REFLUX RATIO - 6-7/1
PRESSURE, MM HG - 65
TEMPERATURE, °C - 240
2,452 G^
490 G 2,4-D ESTER
1,962 G 2,4,5-T ESTER
22 PPM DIOXIN
43
-------
APPENDIX A
44
-------
ANALYSIS OF CARBON TETRACHLORIDE FOR TCDD
A cooperative project with the Diamond Shamrock Company, Environmental
Protection Agency and the United States Department of Agriculture.
Prepared by: E. A. Woolson
J. R. Plimmer
P. C. Kearney
Pesticide Degradation Laboratory
Agricultural Environmental Quality Institute
Agricultural Research Center, West
Beltsville, Maryland 20705
May 8, 1973
45
-------
Final Report on Conversion of TCDD to Carbon Tetrachloride
On October 10, 1972, the Pesticide Degradation Laboratory, Agricul-
tural Environmental Quality Institute, Agricultural Research Service,
United States Department of Agriculture entered into a Reimbursable
Agreement for Services between Federal Agencies (under WRU 308-1125-14761)
with the Industrial Pollution Control Division, Environmental Protection
Agency, to analyze a number of samples of carbon tetrachloride for dioxin
content. Specifically, the agreement was to determine the quantity of
2,3,7,8-tetrachloro-dibenzo-p-dioxin (TCDD) in samples of carbon tetra-
chloride produced experimentally by chlorinolysis of Herbicide Orange.
Chlorinolysis was conducted by the Diamond Shamrock Company of Painesville,
Ohio, as part of an overall study being conducted for the U.S. Air Force
for the disposal of approximately 26.5 million pounds of surplus Herbicide
Orange.
This pilot study was conducted in a small chlorinolysis apparatus
which causes the destruction of Herbicide Orange by reaction with chlorine
gas at high temperature under moderate pressure (300 psig). If economi-
cally and chemically feasible, a large scale operation would be constructed
for the conversion of 26.5 million pounds of Herbicide Orange into 125.3
million pounds of carbon tetrachloride, 26.9 million pounds of phosgene
and 45.6 million pounds of hydrogen chloride.
This report covers the analysis of experimentally produced carbon
tetrachloride from samples of Herbicide Orange containing approximately
14 ppm TCDD, specifically to determine the concentrations of TCDD in the
carbon tetrachloride at a level of sensitivity of 10 ppt by electron
capture gas chromatography.
46
-------
2
Methods and Materials
Method for TCDD in Carbon Tetrachloride - Condense CC14 (0.1 - 0.5
liters) by using a 3-ball Snyder column on a 1 liter erlenmeyer flask.
Gently remove final traces of 0014 with a ^ stream. Take up residue in
pesticide quality hexane. Transfer onto a 15 g activated alumina column
(19 mm. id., prewash with 100 ml hexane, Fisher's A-540) and wash with
100 ml hexane. Elute with 100 ml 1:1 diethyl ether:hexane and condense
for GLC analysis using a Kuderna-Danish concentrator. Gas chromatographic
columns and conditions are described by Woolson et al_. (1972). Recovery
at 0.4 ppb ranged from 80-110% from CC14.
2,4-D, 2,4,5- T - Transfer an aliquot to a 1 liter separatory funnel.
Add 100 ml hexane, 100 ml 5N NaOH and shake. Draw off and discard the
aqueous phase and repeat the NaOH wash. Rinse with water, followed by
100 ml H2S04. Shake vigorously and allow to settle. Remove I-^SO.
and repeat treatment until F^SO^ is clear and colorless after sitting
overnight. Wash with water, dry with anhydrous Na2§04 and proceed with
the alumina column as before.
Results
Sample Composition TCDD Content
AD-1 CC14 <0.1 ppb
AD-2 CC14 <1 ppb
AD-3 CC14 <0.1 ppb
AD-4 CC14 <10 ppt
AD-5 CC14 < 10 ppt
AD-6 CC14 <10 ppt
AD-1 residue None detected
AD-4 residue < 0.1 ppm
47
-------
Herbicide Orange - 1 2,4-0/2,4,5-T 15.8 ppm
Herbicide Orange - 2 2,4-0/2,4,5-T 13.6 ppm
Herbicide Orange - 3 2,4-0/2,4,5-T 14.9 ppm
2,4-0 Distilled < 1 ppb
2,4-0 Distillation residue 22 ppm
tech. CC14 CC14 < .1 ppb
pest, quality CC14 <10 ppt
Summary
The chlorinolysis of Herbicide Orange to CC14 appears to be a feas-
ible means of chemical disposal. At no time were we able to detect TCDD
in any sample of Herbicide Orange subjected to chlorinolysis. Fractional
distillation of 2,4-D from Herbicide Orange likewise appears to be a
means of recovering part of a useful mixture. No TCDD (cl ppb) was found
in the 2,4-0. Grease, apparently from joints in the distillation appara-
tography failed to reveal the presence of dioxins. Impurities were gener-
ally not chlorinated, but were similar in nature to silicone oils.
In conclusion, TCDD present in Herbicide Orange at a concentration
of about 14 ppm was converted by the pilot chlorinolysis procedure used
by Diamond Shamrock to a concentration of <10 ppt in CCl^. Additional
safety evaluations will have to be conducted by toxicologists to assess
any human health hazards associated with carbon tetrachloride produced
from Herbicide Orange.
Reference
Woolson, E. A., R. F. Thomas, and P. D. J. Ensor, J. Agric. Fd. Chem.
20:351, 1972.
48
-------
APPENDIX B
49
-------
TOXICOLOGICAL TESTS ON CARBON TETRACHLORIDE RESIDUES
by
RENATE D. KIMBROUGH, M.D.
Bioeffects Branch
Environmental Protection Agency
4770 Buford Highway
Chamblee, Georgia 30341
April 5, 1973
50
-------
INTRODUCTION
Two separate toxicologlcal studies were conducted on three
samples received from the Pesticide Degradation Laboratory, Agricultural
Environmental Quality Institute, Northeastern Region Agricultural
Research Center. A study in white rabbits was designed to investigate
the acnegenic activity of the test materials due to possible dioxin con-
tamination. A post implantation exposure study in rats was designed to
provide data on the prepartal, perinatal, and postpartum toxicity of the
test materials.
METHODS
The samples of CCl^ were described as follows:
Sample A - The residue from 1000 ml AD-4, 1000 ml AD-5, and
1000 ml AD-6 CC14 from Diamond Shamrock.
Sample B - The residue from 3000 ml technical CC1^-MCBCX425 (this
sample replaced a sample received originally that was
improperly sealed).
Sample C - The residue from 3000 ml Baker Analyzed Reagent
CC14-1512.
The actual weights of the samples received were 4.06, 2.34 and
3.42 g for Samples A, B, and C, respectively. Each residue was adjusted
to a weight of 6 g by the addition of A.C.S. Spectroanalyzed CCl^ from
Fisher. All subsequent references to dosage and formulations in this
report will indicate quantities of this "adjusted residue." These
adjusted residues were formulated as 30% W/V solutions in acetone for
51
-------
the rabbit study, and as 2% W/V solutions in peanut oil for the rat
study. Sample A contained a grayish sediment that imparted an opaque
appearance to the peanut oil formulation.
The 2,3,7,8 tetrachlorodibenzo-p-dioxin (Dow Chemical) was formu-
lated to final concentrations of 2.51 ug/ml in acetone and 2.51 ug/ml
in 30% CCl^ in acetone solution. Originally the experimental design
called for a 3 ug/ml solution, but we were unable to dissolve this
amount in acetone. The CCl^ used in the control treatments was A.C.S.
Spectroanalyzed CCl^ (Fisher) from our laboratory.
Twelve weanling male white rabbits about 5 weeks old and weighing
633-830 g were divided into 4 groups of 3 rabbits each. Each ear that
was treated with test or control material received a total volume of
1 ml of formulation (0.333 ml/day for 3 consecutive days). Each rabbit
received a total dose of 300 mg adjusted residue of samples A, B or C
on the left ear. Control treatments of CCl^/acetone and acetone only
were applied to the right ear of Groups 1 and 2, respectively. The
right ear of Group 3 received no treatment. Group 4 received dioxin/
acetone on the left ear and dioxin/acetone/CC14 on the right ear for a
total dose of 5.02 ug (See Table 1 for details of dosage). Before the
first dose each ear was swabbed with 70% alcohol. The materials were
applied with a hypodermic needle and syringe as uniformly as possible
to the inner surface of the ear. The rabbits were observed daily for
any response to the treatments. Eighteen days after the last dose the
rabbits were weighed and sacrificed. Organs were examined grossly and
the livers weighed, fixed in formalin and studied microscopically.
52
-------
In the rat study Sherman strain SPF female rats 90-100 days old
were pair-mated with young adult males. When a copulatory plug was
observed insemination was verified by microscopic inspection of a
vaginal smear. When 32 females had been inseminated they were distri-
buted into 4 groups of 8 rats each. Counting the day of insemination
as day 0 the rats were dosed daily by stomach tube on days 7 through 15
of pregnancy. Each rat received 100 mg/kg/day of adjusted residue of
Samples A, B, or C for a total of 900 mg/kg. Control rats received
peanut oil only. Records of offspring were maintained through weaning
(21 days). Offspring from several litters in each group were autopsied
at weaning and examined grossly. Fourteen days after her litter was
weaned each dam was sacrificed, examined grossly, implantation sites
counted and liver weighed and fixed in formalin for further study.
RESULTS
One rabbit in Group 2 (Sample B) developed faint reddish areas on
the left ear about 1 week after the last dose. One rabbit in Group 4
(dioxin) developed similar areas on both ears. The condition persisted
for about 4 days and then was no longer noticeable. One rabbit in
Group 1 (Sample A) and one rabbit in Group 3 (Sample C) developed
diarrhea and died after 2 and 1 applications, respectively. Diarrhea
and post weaning mortality is not uncommon in young rabbits and the
deaths were probably not dose related. All of the rabbits in Group 4
(dioxin) exhibited noticeably depressed growth rates.
Jones and Krizek (1962) reported grossly visible acneform responses
53
-------
in 5/7 adult rabbits treated with 3 ug of 2,3,7,8
tetrachlorodibenzo-p-dioxin on the inner surface of the ear, but found
no systemic toxicity.
No well defined topical response was seen in ears of rabbits in
any group in our study. However, the liver proved to be a sensitive
indicator of dioxin toxicity. Compared to the other groups the livers
from rabbits in Group 4 (dioxin) were markedly enlarged. Two livers
had grossly visible lesions and all 3 livers had histological changes
when examined microscopically. The liver weights and pathology are
presented in Table 2. All livers of animals dosed with Samples A, B,
and C were normal grossly and microscopically.
Microscopic examination of the livers of rabbit dosed with dioxin
showed anisocytosis and hypertrophy of the liver cells throughout the
sections. The cytoplasm of the hepatocytes was vacuolated or had a
foamy appearance. Some liver cells contained a light brown pigment and
inclusions. Many multinucleated cells were also seen. Slight inter-
stitial fibrosis was also present. Two of three livers examined showed
also foci of necrosis surrounded by fibrosis and multinucleated giant
cells. Organisms were not demonstrated within the necrotic lesions by
microscopic examination.
The reproduction and survival of pups in the rat study are
summarized in Table 3. No significant difference in the number, size,
or condition of pups was observed. The overall survival rate (89%) was
54
-------
slightly reduced in rats dosed with Sample A, however, this falls within
the lower range of normal for our rats. Most mortality in this group
occured before the fourth day.
The liver weights of the dams are presented in Table 4. No
difference in weight or gross appearance was observed. The livers of
these rats were also normal on microscopic examination.
CONCLUSION
The mean total dose of 2,3,7,8 tetrachlorodibenzo-p-dioxin which
caused changes in the rabbit liver was 7.05 ug/kg body weight. The
mean total doses of Samples A, B, and C were 434, 444, and 464 mg/kg
body weight, respectively. If the dioxin is present in the material
from which Sample A was derived our studies indicate a concentration
of les's than 0.029 ug/g or less than 0.047 ug/ml of original material.
It is emphasized that the rabbit test is specific for the presence
of 2,3,7,8 tetrachlorodibenzo-p-dioxin or less toxic materials that
elicit the same response. Because of limited quantities of test
material both the rabbit and rat tests utilized small numbers of animals.
Detailed characterization of the toxicity of the test material would
require additional tests and larger numbers of animals.
REFEKENCES
E. Linn Jones, M.D. and Helen Krizek, Ph.D., "A Technique for Testing
Acnegenic Potency in Rabbits, Applied to the Potent Acnegen, 2,3,7,8
55
-------
Tetrachlorodibenzo-p-Dioxin," Journal of Investigative Dermatology.
39, pp. 511-517 (1962).
ACKNOWLEDGMENTS
The following individuals contributed significantly to the test
project: Mr. Ralph Linder of RTF was responsible for the animal dosing
and Dr. Robert Moseman, also of RTF, was instrumental in the preparation
of some of the solutions.
56
-------
Table 1. Dosage rates for rabbits dosed with CC14 samples and
2,3,7,8 tetrachlorodibenzo-p-d1ox1n (3 rabbits per group).
Group
1
2
3
4
Test
material
A
B
C
Dioxin
Treatment***
Left ear
30%
30%
30%
2.51
Sample
Sample
Sample
ug/ml
A
b
C
In
1n
1n
In
acetone
acetone
acetone
acetone
Right. ear
30* CC14 1n acetone**
Acetone
No treatment
2. 51 ug/ml In CC14/
Dose
Dally
100 mg*
100 mg*
100 mg*
1.67ug
Total
300
300
300
mg*
mg*
mg*
5.02ug
acetone**
* Adjusted residue.
** A.C.S. Spectroanalyzed CC14 (Fisher).
*** 0.333 ml/day for 3 consecutive days to Inner surface of ear.
-------
01
CD
Test
•aterial
Saaple A*
SaapleB*
SanpleC*
Dioxin**
Rabbit
No.
1942
1943
1944
1945
1946
1947
1948
1949
1950
19S1
1952
1953
Body
Original
(9»)
686
830
698
679
662
688
662
633
692
711
672
757
Table 2
of
Height
Final
(9-)
1390
Died
1484
1423
1312
1496
1649
1491
Died
1161
1200
1194
. Body Height, liver Height, and autopsy findings
rabbits sacrificed 18 days after last dose.
Liver Height
(1 of
(
-------
Table 3. Reproduction and survival of pups fro* daas
dosed orally on days 7-15 of pregnancy.
Test Feaales
Material dosed
en
ID
Peanut Oil
(control)
Sample A*
Saaple B*
Sample C*
7**
8
8
8
Mean
Implantation Ho. of litters Average litter sizi
Sites
14.3
12.8
14.0
13.1
Born
7
8
8
8
yeaned
7
8
8
8
Birth
12.7
11.4
12.0
12.1
Meaning
12.6
10.1
n.o
11.6
Total pups per group
> Born( found) All ye
Dead
2
1
1
1
Alive
89
91
96
97
4 days
88
82
88
94
at
weaning
88
81
88
93
Survival
98.9
89.0
91.7
95.9
Average
Birth
(9-)
6.0
5.9
5.7
5.8
body wt.
Ifesninq
(g")
32.7
33.2
34.2
33.5
*100 ag/kg/day of adjusted residue.
**0ne rat died from dosing injury, not used in calculations.
-------
Table 4. Liver weights from dams dosed
orally on days 7-15 of pregnancy.**
Test
material No. of rats
Peanut oil
(control)
Sample A*
Sample B*
Sample C*
7
8
8
8
Liver Wt. (gm) Liver wt. (% of body wt.)
Mean +. S.E. Mean + S.E.
10.66
10.51
10.86
10.63
+ 0.217
+ 0.518
+ 0.313
+ 0.465
3.66
3.57
3.65
3.49
+ 0.036
+ 0.152
+ 0.049
+ 0.090
* 100 mg of adjusted residue/kg/day
** Rats sacrificed 14 days after weaning.
60
-------
APPENDIX C
61
-------
EPA-IAC-003(It)
/.,'" UNITED STATL;S ENVIRONJ/ENTAL PROTECTION AGENCY
^ WASHINGTON. D.C. 20.'.00
MEMORANDUM OF AGREEMENT
between
ENVIRONMENTAL PROTECTION AGENCY
and
U.S. AIR FORCE LOGISTICS COMMAND,
WRIGHT PATTERSON AIR FORCE EASE
DAYTON, OHIO
for a project covering the Research Program:
DETOXIFICATION AND DISPOSAL OF AGENT "ORANGE"
1. Purpose
The objective of this agreement is the development of a laboratory
program to evaluate the practicality of the application of chlorinolysis
for the disposal of Agent ORANGE. The investigation will also include
a verification of the destruction of the teratogenic impurity "dioxin"
present in the 2,4,5-T ester component. Information and data obtained
in this- research will be utilized by the EPA to ascertain whether the
proposed concept (1) can be applied and used as a viable means of
disposal of stockpiles of this defoliant and (2) can contribute
significantly towards solving the dilcma of disposal of biorefractory
chlorinated hydrocarbon residues that exists in the Petrochemical Industry
today.
2.. Services to be provided
The Industrial Pollution Control Section (IPCS), Applied Science
and Technology Drench, Technology Division, Office of Research and
Monitoring, Environmental Protection Agency will manage the research
62
-------
program for the U.S. Air Force. The Diamond Shamrock Corporation,
Paincsvillc, Ohio, \.ill conduct on tlieir premises, the laboratory
evaluation of the chlorinolysis concept for destruction of Agent ORANGE
and detoxification of the dioxin impurity present therein. It is
understood that the USAF will supply at least one 55 gallon drum of
Agent ORANGE for this investigation. This program to demonstrate
the feasibility of the chlorinolysis concept will entail 9 man-months
of effort for completion (2 calendar months) and will cost $35,000.
Verification of the fate of dioxin during the chlorination process will
be conducted by the U.S. Department of Agriculture (USDA) team
(Dr. Philip Kearney, Leader) at Beltsville, Md. at a nominal charge of
$100 per sample. It is envisioned that 50 to 100 samples will be
provided to the USDA by the Diamond Shamrock Corporation, thus
requiring an additional $5,000 - $10,000.
The EPA-IPCS will prepare a final report containing all data
collected, together with conclusions and recommendations. It is
understood that proprietary information is to be disclosed, and as such>
is considered FOR OFFICIAL USE ONLY.
3. Period of agreement.
This agreement is effective for a three (3) months from date of
acceptance of both parties.
4. Names, titles, addresses of respective project officers:
Mr. Paul E. DCS Rosiers
Staff Engineer
Industrial Pollution Control (RD-679)
Environmental Protection Agency
Washington, D.C. 20460
(202) A 26-4171
63
-------
Mr. Carlton W. Carter
HqUSAF (LGSKE)
Washington, D.C. 20330
(202) 697-8635
Dr. Philip C. Kearney
Pesticide Degradation Laboratory
Agricultural Environmental Quality Institute
Agricultural Research Service
Agricultural Research Center
Beltsville, 1-ID 20705
(301) 344-3082
5. Funding arrangement
The USAF has -transferred funds in the amount of $45,000
(see Enclosures 1 and 2). to the' EPA-IPCS for the delivery of
information as outlined in the Research Proposal (reference, letter
from ARRX to Col. O.J. Sundstroa, HqUSAF, AF/LGS, Pentagon, dated
July 10, 1972). If the EPA-IPCS determines that work called for in
this agreement will be delayed, it will notify the USAF in writing
immediately. Billing will be accomplished by the EPA to the U.S.
Air Force on Standard Form 1080.
6. Authorization
Economy Act of 1932 as amended (31 U.S.C. 686)
Federal Water Pollution Control Act as amended (33 U.S.C. 466 et seq.)
7, Signature
Agreed:
6 SEP 1972
W. H. Fairbrother, Leland D. Attaway
Brigadier General, USAF Deputy Assistant Administrator for Resca-
Deputy Chief of Staff, Office of Research & Monitoring
Distribution Environmental Protection Agency
Department of the Air Force
U.S. Air Force Logistics Command
Wright- Patterson Air Force Base
Dayton, Ohio 45433
Enclosures
64
-------
DEPAKTI.'.CNT O~ THE AJR. FORCE
»:. Air< roRcc LOCIOTICS
WMIGHT.PAVri:l»:;ON AIM FOrtCC IIA^C. OHIO 4U~33 v . ~' ,
•I1.*-. --
nri'iv 10
f.iiH ori
*uujtcT, Disposition of Herbicide Oroj-.Go
«• SA/UU/DS
1. The attached JiQ USAF/LGSK letter, IS Jul 1972, is fowardc-d for
iisjiediate action as requested in paragraph 3 thereof.
2. RSi funds under ESIC 586 in the aniount of'$45,000 will be provided
your installation in the August change to the Operations Operating
Budget (OOB).
3. It is recommended, that obligation authority be provided to Z?A
in two increments of $35,000 and $10,000 each as discussed 27 Jul I$72
between Mr. G. Kiska, SA/ui/i/i)&-I and Kr. S. Keaton of this office.
FOR THE COMKAKDER
1 Atch
W. K. FAIRBROTKER, Brigadier Genorat USAF KQ USAF/LGSK Ltr, .
. Deputy Chief of Sloff,. Distribution 1S Jul W?2
Cy to: KQ USAF/LGSK, v/z.-
SAAKA/SF, w/o stcr
PRIDE IN THE PAST : ^-rxvT7TVi'— . FAITH IN THE FUTUBI
-------
OBLIGATION AUTHORITY
- INSTRUCTIONS TO
Duplicate copy hereof should be returned to the issuinc.
office no: l.iicr than the J.ue specified above, or as soon as
all obligations hereundcr have been completed, whichever
occurs first. Obligation-: itii'urred will be listed on the reveise.
A copy of each obluraiion document will be transmitted
to the issuing office coincident with incurrencc of the obli-
gation. A copy of each paid voucher wi!l be transmitted
direct to The issuing office by the accounting and finance of-
ficer, and documents submitted to the accounting and finance
'"United States Environmental Protection Age
Aitn: Paul £. Desfcosiers
Washington, D C 20460
'"^-•.-""YOU ARE Httttl
'"* WfTHIN THE AMOUNT
••^ ^THE FUNDS crre
f AUTHORIZED TO INCUR OBLIGATIONS I
ADVIQt HUMU«
S73-59
CHANCf HUMUd 1
KKCDVING OFFICE
office should contain instructions to that effect. The name.
address, and accounting and disbursing station number of the
issuing office and the number of the obligation authority will
always be shown on ail obligation and expenditure docu-
ments, the ob!ij:i;ion authority number being shown in
parentheses irnmedateiy after the document number..
TChen items are furnished from stock for which reim-
bursement is required. Standard Form 1080 will be prepared
and processed by your office, charging the funds cited below.
I&SUCD ftV | A«w and ocac: tf unang tffict)
lcy.--SAAMA/DSRA - -
ihiriL S K ovcniDet*
1S72
. FOR THE rURPOSES ANO
B.ARE PROfERLY CHAROEAHE-AND HAVE BEEN RESERVED BY WB OFFKE.
SICNATUCt. TtftSntjMl. CKAOf . A^D/filtt Pf IHIT1MING Off ICBt
^%$gtf&&64&i
Chiel, Surveillance & Support branch
DATE ISSUED
10 A"iia 1£72
AmOfdIATON JYM*CH
5733400
SIOWAIUM, mre NAME. AMU GIADE of ACCOUNTING ANI> HNANCC omaR
(Kvpmt) ^^ r .
1 MAJO* ACCOUHTH
' MlOTMBn'snlAt 1 FtOJECT
3--
ACCOUHTIMC AND
OUIUXSIMO 5TA- •
TIOMMUMK*
5504300 '
AMOUNT1
-$357000
- _ .: . _ _. AUTHORIZED PURPOSES AND DESCRIPTION -.
-To conduct a.study and perform tests for the disposal of Herbicide Orange. See attache.
£A£I^A/DS letter, 7 Aug 1972, Disposition of Herbicide Orange, with attachments.
,
~ ................. .-.- , 1Atcn --_•••
DS Ltr, 7 Aug 72, w/Ateh
'- -
.
_ /| 17' .
iilz.-'-"-
>
s~
rs
. - .. -s - . • t .- -.
i '••
... ._ . ....
-tt
_,_:..;
.
— ;
- r '
.
'•
.'
1
- \ ' '. '
— - '•»
^
'OA/iVir/ioiu WILL NOT fwfd the o*m»tm/ of this evtfumtt.
«e '•'•-'*
•W AUC »» '
Miviouj toniOMs or THIS KHUA MAY IE useo
66
-------
OBUGATION AUTHORITY
ADVICE HUMKI
INSTRUCTIONS TO RHClilVING OFFICE
Duplicate copy hereof should be returned to the issuing
office not later than the date specified above, or as soon as
ail cci'jidons hereunder have betn completed, whichever
occurs n.'-<. Oblii-acions incurred will Iv: Ihttt! on the revrrstf.
A crpy of r?;fi >>b!v;it:.->n H.;>oirr.enc vi\\ L.« ir.iti--tr.ined
to the 3suiag office coincident with incurrcnce of the obli-
gaticn. A copy of each paid voucher will be transmitted
citc.T TO «ie j*si::p4 ot::ce by :'•:<: acn^ntiiu; and fi.i.mtcof-
ficer. uid documents submitted to the accounting anil finance
.: Mail Obligations to 2851ABGp/ACFSB-^l
office should contain instructions to that cITect. The name,
address, and accounting and disbursing station number of the
issuing office and the number of the obligation authority will
alw.iys be shown on .til obligation and expenditure docu-
:ncm>. the ciblicjcirjn nuthoriiy number bcin^r shov.n in
partiiLliests immediately after the document number.
When items are furnished from stock for which reim-
h'lrSf :r.^nt is refiuired, Sr.incU/d Form 10oO o.-:ll b-- prcp.uuJ
and ptix^ssed by your oriicc, charging the funds ciceU below.
ISSUCD TO *>«*v. m4<(tta, mi memtuilt*c a*dJitkaling statitn
***tjai~ecl States EcviroozsentaJL Protection
ATTTf: Paul E. OesHoeiers
' Washington DC 20U60
Agency
ISWfO «Y (/To
*ffa)
SAAMA/DSRA
Kelly AFB TX 782Ul
TOO ME HEREBY AUTHORIZED TO INCUR OBLIGATIONS UNTIL .
irg
. FOR THE PURPOSES AND
VtTHN THE AMOUNT (I) STATED JFLOW.
' ^ THS fUNOS I
AND HAVE BEEN RESEXVED BY THIS OFFICE. •
Chief, Surveillance & Support Branch
*T*ft'f'-*» ftf TH g'f.'
SIOKATUCT, TYPED NAMl, AND OUAOf Or ACCOUKT1NO AND riHANCC OfUdH
f*(?EB&riEVE R. SCLERANDI ^ ' /C$ ,-1
Funds Certifying Officer
APPHOnlATION SYAUOl
MAXX ACCOUMIINO CtASStf ICAT1OM
AUOTMINT UnAl
ACCOUNTING AND
CHS»U«St«G STA-
IIONNUMmi
AMOUNT'
10 Aug 1975
5733l»00
303 6306 28»86g386
550^300
$10.000
AUTHORIZED PURPOSES AND DESCRIPTION
"o conduct sampling tests as cited in US Environmental Protection Agency letter,
10 Jul 1972, addressed to HQ USAF AF/LGS. Copy filed with Obligation Authority
(Acrice Mo. S73-59) issued by SAAMA/DSRA, 10 Aug 1972.
,V/T rwrf th< ammail a/Mil mlhmlt.
67
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
4. Title
STUDY OP FEASIBILITY OP HERBICIDE ORANGE CHLORINOLYSIS
7. Authors)
Lavergne, Edward A.
ReiiortDste
6,
8. Perfor
Rt port Ho.
Project No.
9. Organization
Diamond Shamrock Corporation
T. R. Evans Research Center
Painesville, Ohio 44077 -
Organisation
111. Contract /Grant No.
68-01-0457
13.' Type of Repot; tod
Period Coveted
I:"'. Supplementary Notes
Environmental Protection Agency report number, EPA-600/2-7^-006, July
16. Abstract
A process termed chlorlnolysis (exhaustive chlorination) was applied to samples of
USAF Herbicide ORANGE. The ORANGE (50/50 volume mixture of the n-butyl esters of
2,4-D and 2,4,5-T) contained a production impurity called diozLn - a powerful teratogen.
The research objective was to demonstrate the feasibility of converting such
herbicides into marketable products, namely, carbon tetrachloride (CCl^), carbonyl
chloride (COCLj), and hydrogen chloride (HC1), while destroying any dioxin present.
Bench scale (100 g/hr) chlorlnolysis of ORANGE was evaluated over a range of
reactor conditions. The critical reaction parameters were found to be: chlorine to
carbon ratio (4.4 - 7.2); temperature (600 - 800°C); pressure (225 - 300 psig); and
retention time (0.5 - 1.0 minute).
Thermodynamlc analysis had indicated that CCl^, hexachlorobenzene (HCB), and
chlorine (C^) would exist in equilibrium at the reaction conditions utilized. Because
of the balance required between reaction rate (reactor size) and HCB content of the
effluent, recycle of unconverted HCB from the product recovery system was found to be
necessary. Recycle tests Indicated that single pass HCB conversion rates of 80% could be
realized.
17a. Descriptors
Chlorlnolysis, Herbicide ORANGE, Exhaustive Chlorination, Dioxin, 2,4-D, 2,4,5-T,
Chemical Conversion, Fractionation/Distillation, Chlorinated Still Bottom Residues
17b. Identifiers
17c. COWRR Field <* Group
18. Availability
19. Stourftyf'Ma,
(Report)
Abstractor PT-.
WASHINGTON, D. C. 8OS4O
Institution
SERL. Athena. Gaot-yta
WRSIC 102 BtV JUNE
V.
------- |