EPA-600/2-76-109
July 1976
Environmental Protection Technology Series
MOBILE TREATMENT
SPILLED HAZARDOUS
-------
RESEARCH REPORTING SERIES
Research reports of the Oflice of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. [Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The five series are'
1. Environmental Health Effects Research
2 Environmental Protection Technology
3 Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-76-109
July 1976
DEVELOPMENT OF A MOBILE TREATMENT SYSTEM
FOR HANDLING SPILLED HAZARDOUS MATERIALS
by
Mahendra K. Gupta
Envirex Inc.
A Rexnord Company
Environmental Sciences Division
Milwaukee, Wisconsin 53201
Contract No. 68-01-0099
Project Officer
Joseph P. Lafornara
Oil and Hazardous Materials Spills Branch
Industrial Environmental Research Laboratory-Cincinnati
Edison, New Jersey 08817
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental
Research Laboratory, Cincinnati, U.S. Environmental Protection Agency,
and approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the U.S. Environmental
Protection Agency, nor does mention of trade names or commercial pro-
ducts constitute endorsement or recommendation for use.
-------
FOREWORD
When energy and material resources are extracted, processed, and
used, these operations usually pollute our environment. The resultant
air, land, solid waste and other pollutants may adversely impact our
aesthetic and physical well-being. Protection of our environment
requires that we recognize and understand the complex environmental
impacts of these operations and that corrective approaches be applied.
The Industrial Environmental Research Laboratory - Cincinnati
assesses the environmental, social and economic impacts of industrial
and energy-related activities and identifies, evaluates, develops and
demonstrates alternatives for the protection of the environment.
This report is a product of the above efforts. It documents the
laboratory and engineering studies conducted to determine the design
of a Mobile Hazardous Material Spills Treatment Trailer. The report
also contains a description of the "Trailer" and its main components
(pumps, portable reaction and storage tanks, mixed media filters,
activated carbon columns, etc.) and summarizes its spill response
capabilities. The trailer, the fabrication of which is complete, is
essentially a mobile self-contained filtration/carbon adsorption water
treatment system which, for the first time, provides the capability to
rapidly treat and remove spilled water soluble organics.
This report should be of interest to federal, state and local
government personnel as well as to individuals from the chemical
process and transportation industries who are faced with the problem
of how to treat spills of hazardous chemicals in water.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
i i i
-------
CONTENTS
REVIEW NOTICE ii
ABSTRACT iii
TABLE OF CONTENTS v
LIST OF TABLES vii
LIST OF FIGURES viii
ACKNOWLEDGMENTS ix
Sections
I CONCLUSIONS 1
II RECOMMENDATIONS 2
III INTRODUCTION 3
IV TEST METHODS AND PROCEDURES 6
CHEMICAL TREATMENT AND CLARIFICATION 6
ACTIVATED CARBON ADSORPTION 7
REVERSE OSMOSIS TREATMENT 7
V TEST RESULTS AND EVALUATIONS 8
ACETONE CYANOHYDRIN 8
Chlorination Tests 8
Activated Carbon Treatment 11
ACRYLONITRILE 15
Volatility Tests 15
Chlorination Tests 15
Carbon Adsorption Tests 16
AMMONIA 18
CHLORINE 21
CHLORINATED HYDROCARBONS 24
METHANOL 29
PHENOL 31
Chlorination 31
Activated Carbon Adsorption 34
-------
CONTENTS (continued)
Sections Page
ORGANO LEAD COMPOUNDS 37
Solubility Tests 37
Treatment Tests - Tetraethy 1 lead (TEL) 40
Activated Carbon Adsorption 41
Treatment Tests - Tetramethyllead (TML) 41
Activated Carbon Adsorption 43
REVERSE OSMOSIS TESTS 45
VI DESIGN CRITERIA FOR A MOBILE TREATMENT SYSTEM 48
FIocculat ion/Sedimentation System 50
Dual Media Filters 51
Carbon Columns 56
Treatment System Layout 58
VII REFERENCES 61
VIII APPENDIX A - ANALYTICAL APPARATUS AND PROCEDURES 63
ANALYTICAL INSTRUMENTS AND APPARATUS 64
ANALYTICAL PROCEDURES 65
Extraction Procedure for DDT 67
APPENDIX B - BENCH SCALE TEST PROCEDURES 69
CARBON COLUMN TEST PROCEDURE 70
PROCEDURE FOR CARBON ISOTHERM TEST 72
REVERSE OSMOSIS BENCH TESTS 73
VI
-------
LIST OF TABLES
No. Page
1 LIST OF HAZARDOUS MATERIALS 5
2 OXIDATION OF ACETONE CYANOHYDRIN WITH NaOCl -
30 MINUTE CONTACT TIME 9
3 OXIDATION OF ACETONE CYANOHYDRIN WITH CHLORINE DIOXIDE 10
A CHLORINATION OF ACRYLONITRILE 16
5 CHLORINATION OF AMMONIA WITH NaOCl - 30 MINUTE
CONTACT TIME 22
6 ADSORPTION OF PESTICIDES AND HERBICIDES ON SOILS 25
7 RESULTS OF CARBON ISOTHERMS TESTS PERFORMED ON
METHANOL SOLUTION 29
8 RESULTS OF CHLORINATION OF PHENOL SOLUTION 31
9 PHENOL ADSORPTION EFFICIENCIES ON FILTRASORB 400 34
10 STABILITY OF TETRAMETHYLLEAD IN WATER 39
11 RESULTS OF CHEMICAL TREATMENT TESTS ON TETRAETHYLLEAD (TEL) 40
12 CHEMICAL FLOCCULATION, SETTLING AND FILTRATION OF
TETRAETHYLLEAD WATER 42
13 TREATMENT OF TETRAETHYLLEAD WITH POTASSIUM PERMANGANATE 43
14 CARBON ISOTHERM TEST RESULTS FOR TETRAETHYLLEAD 44
15 TEST OF HAZARDOUS MATERIALS ELIMINATED FOR BENCH SCALE
RO TESTS BECAUSE OF THEIR PROPERTIES AND EXPECTED
ADVERSE EFFECT ON MEMBRANE 45
16 RESULTS OF RO FEASIBILITY TESTS 46
v i
-------
LIST OF FIGURES
NO.
1 Carbon Isotherm Result for Acetone
Cyanohydrin 12
2 Carbon Column Test for Acetone Cyanohydrin
(High Concentration) 13
3 Carbon Column Test For Acetone Cyanohydrin
(Low Concentration) I**
4 Carbon Isotherm Results for Acrylonitrile 17
5 Carbon Column Tests For Acrylonitrile
(High Concentration) 19
6 Carbon Column Tests For Acrylonitrile
(Low Concentration) 20
7 Carbon Isotherm Results For Chlorine 23
8 Carbon adsorption Isotherms For Various
Pesticides & Herbicides 26
9 Carbon Isotherm Plot for pp'-DDT 28
10 Carbon Isotherm Results For Methanol 30
11 Carbon Column Tests For Methanol
(High Concentration) 32
12 Carbon Column Test For Methanol
(Low Concentration) 33
13 Carbon Isotherm Results For Phenol 35
14 Carbon Column Test For Phenol
(High Concentration) 36
15 Waste Treatment Flow Diagram For A Mobile Spill
Response Vehicle 49
16 Completely Assembled Flocculation/Sedimentation
Tank In Field 52
17 Photograph Of The Mounted Dual Media Filters 53
18 Closeup Of The Filter Plumbing Arrangements 54
19 Carbon Column As Mounted On The Mobile
Treatment System 57
20 Recommended Plan View Layout Of System 59
21 Laboratory Carbon Column Test Apparatus 71
22 Schematic Of Reverse Osmosis Test Apparatus 74
v i i
-------
ACKNOWLEDGMENTS
This investigation was carried out by the Environmental Sciences
Division of Envirex Inc. The bench scale studies, analytical
determinations and statistical investigations were conducted by
Mr. Richard Wullschleger and Mr. Gupta's staff. Their efforts and
diligence were essential to the collection and evaluation of the
data presented. Mr. R. Scholz ts responsible for the mechanical
design of the hardware and selection of the materials of construction.
The support of the project by the U.S. Environmental Protection Agency,
and the willing assistance and helpful advice of Project Officers
Mr. Paul Minor and Dr. Joseph LaFornara in the conduct of the project
is acknowledged with sincere thanks.
IX
-------
SECTION I
CONCLUSIONS
Based upon the laboratory data developed during this study for treatment
of selected hazardous materials, it was concluded that:
I. A mobile waste treatment system consisting of chemical reaction,
ftocculation, sedimentation, granular media filtration and
activated carbon adsorption would provide the most suitable
and versatile system for on-site removal and treatment of
hazardous materials.
2. Reverse Osmosis can be utilized on a selective basis for the
treatment of water soluble inorganics. RO was found unsuitable
for many hazardous materials because of their low water solu-
bility and effect on membranes.
3. For the various hazardous materials evaluated in this study,
it was found that:
a. Activated carbon can be effectively used for the
treatment of acetone cyanohydrin, acrylonitrile,
chlorine and phenol.
b. Chlorination can be used effectively for the oxidation
of ammonia and for about 50% removal of acetone
cyanohydrin and acrylonitrile.
c. Most pesticides and herbicides (chlorinated hydrocarbons)
can be removed by coagulation, filtration and carbon
adsorption.
d. Physical-chemical methods were not found to be suitable
for the treatment of methano1. However, biological
treatment may be applicable in some spill situations.
e. TEL and TML can be effectively oxidized with potassium
permanganate and also can be adsorbed partially via
activated carbon after chemical flocculation, settling
and fi1tration.
k. A 12.6 I/sec (200 gpm) mobile treatment system consisting of
the treatment processes outlined in item 1 was constructed
based on the design data outlined in this report.
-------
SECTION II
RECOMMENDATIONS
It Is recommended that:
1. The treatment vehicle built under this contract respond to a
test spill to evaluate the effectiveness and capability of
this treatment system.
2. Standard operating procedures and guidelines be developed for
various hazardous materials, so that the hazardous spills
vehicle can be deployed in an optimum manner in minimum time.
3. Trained personnel be employed to operate the vehicle and
develop the standard operating procedures.
4. Multiple units similar to the 12.6 I/sec (200 gpm) mobile
treatment system developed under this contract be built and
placed at strategic locations around the United States for the
on-site handling and treatment of spilled hazardous materials.
-------
SECTION III
INTRODUCTION
"Hazardous substances" as a descriptive term covers a wide variety of
materials, primarily because the word "hazardous" (like most qualitative
adjectives) is essentially relative. Although many substances are
obviously hazardous to aquatic and marine life due to direct inherent
toxicity, other substances are equally detrimental to water ecosystems
by virtue of indirect toxicity resulting from disturbance of the
ecological balance such as the depletion of the dissolved oxygen supply
required to support various life forms. Still others, although originally
present at less than toxic or inhibitory levels, may become concentrated
to these levels by residual accumulation through a succession of food
chain transfers. Furthermore, even a substance as innocuous as molasses
can become "hazardous", if a large quantity is released into a delicately
balanced water ecosystem like a trout stream.
Handling and transport of these hazardous materials is necessary because
of the wide spread usage of such materials in an industrial society.
Although the essential feature of any overall program designed to minimize
the risks involved in producing, handling, transporting, and using
hazardous substances is prevention of spills, it should be recognized
that spills will occur despite the most comphrehensive precautionary
measures. When a spill does occur, the immediate primary concern is
containment of the spilled material -- isolating the contaminating
substance and preventing, in so far as possible, its encroachment on the
surrounding environment. Ideally such containment will prevent the
spilled material from reaching any water resources. However, such an
ideal situation seldom exists in practice. Generally, containment
is not possible, and a water body may be contaminated by the spilled
hazardous materials which must therefore be purified. Thus, it is
reasonable to expect that a development effort needs to be directed to
minimize the effects of spilled hazardous materials that have reached
a watercourse of some type (ditch, pond, stream, lake, river, etc.).
This concern for the accidental entrance of hazardous materials into a
watercourse has initiated several programs by the Environmental
Protection Agency for the control of this problem. The program described
herein was undertaken to develop a transportable treatment system for
on-site removal and treatment of spilled hazardous materials in aqueous
solutions. This program has the following objectives:
-------
1. Develop bench scale design data for the treatment of various
hazardous materials.
2. Design and fabricate a mobile treatment system for spilled
hazardous materials.
A listing of the hazardous materials evaluated during this study is
shown in Table 1. This list was selected based on the priority ranking
system for hazardous materials developed by EPA (I). As can be seen, the
materials have been divided into two categories but overlap in several
cases. Initially, only the nine materials listed in the first section
of Table 1 were included for evaluation in the original contract
68-01-0099- However, later the above contract was amended to include
a laboratory scale evaluation of the applicability of reverse osmosis
for the pre-concentration of materials listed in the second section of
Table 1. The intent of the laboratory evaluations was to screen the
listed hazardous materials based on a brief review of available
literature and to develop bench scale treatment feasibility data for
those materials for which available information was lacking.
This report summarizes the laboratory, design and fabrication studies
performed during the course of this project. Detailed engineering
drawings and equipment specifications suitable for duplicating the
mobile treatment system are in the possession of the Environmental
Protection Agency as are a set of manuals with complete instructions for
operating, repairing and maintaining the component devices. This
comprehensive material can be made available to interested parties
through EPA's Industrial Environmental Research Laboratory, Edison, New
Jersey 08817.
-------
Table 1. LIST OF HAZARDOUS MATERIALS
Materials considered in Selection of the Mobile Treatment System.
Acetone Cyanohydrin
Acryloni tr?le
Ammonia
Chlorinated Hydrocarbon Pesticides
Chlorine
Methanol (Methyl Alcohol)
Phenol
Tetra Ethyl Lead (TEL)
Tetra Methyle Lead (TML)
Materials considered for reverse osmosis treatment feasibility.
Acetone
Acetone Cyanohydrin
Acryloni trile
Alum (Aluminum Sulfate)
Ammoni urn Sa1ts
Benzene
Chlorine
Chlorinated Hydrocarbon Pesticides (DDT, 2-A-D, 2-4-5-T)
Chlorosulfonic Acid
Copper Sulfate
Formaldehyde
Lead as in TEL and TML
Methanol
Mercuric Chloride
Phenol
Phosphorus Penta-Sulfide
Styrene
-------
SECTION IV
TEST METHODS AND PROCEDURES
The unit treatment processes investigated for the hazardous materials
listed earlier were chemical treatment, clarification via settling
and/or filtration, activated carbon adsorption and reverse osmosis.
These processes were selected based on the results of an earlier EPA
study (2) which indicated that many of the hazardous materials could be
precipitated by chemical treatment or oxidized to a more innocuous
state by chemical treatment followed by sedimentation and/or filtration
to remove the precipitated and other parttculate matter such as soil or
debris. In addition, activated carbon was found to be one of the most
versatile and affirmative processes for the treatment of a wide variety
of water soluble materials.
Reverse osmosis treatment feasibility evaluations were based on the promise
of utilizing RO in conjunction with activated carbon. It was thought
that this combination could provide tremendous versatility of treatment
for both inorganic and organic hazardous materials. Reverse osmosis was
promising as a pre-concentration step prior to activated carbon
treatment for various organic materials because of its compact size, high
volume and low weight in a mobile treatment system. Reverse osmosis
treatment could also be utilized for the treatment of various
inorganics, such as toxic heavy metals, that otherwise could not be
treated via activated carbon.
A brief description of the various bench scale treatment test procedures
employed is presented in the following sections. A description of the
laboratory procedures and the analytical apparatus utilized in this
study is included in Appendix A. Bench scale treatment tests were
conducted only in those areas where available information from literature
was lacking.
CHEMICAL TREATMENT AND CLARIFICATION
The bench scale chemical treatment tests were generally conducted in 100
to 1,000 ml beakers and graduated cylinders. Chemicals were added in
measured quantities according to preselected dosages and suitable reaction
times were provided to evaluate optimum chemical dosages. Sedimentation
time generally ranged between 30 to 60 minutes. The pollutant concentra-
tions were analyzed after clarification to evaluate the degree of removal
achieved during various tests. The hazardous materials for which the
-------
chemical treatment tests were conducted were: acetone cyanohydrin,
acrylonitrile, ammonia, chlorinated hydrocarbons, chlorine, phenol,
TEL and TML.
ACTIVATED CARBON ADSORPTION
Both isotherm and dynamic column type carbon adsorption tests were
conducted to supplement existing data for the selected hazardous
materials listed in Table 1. A detailed description of the bench scale
procedures for the two types of carbon tests is included in Appendix B.
The materials for which the carbon adsorption was evaluated were:
acetone cyanohydrin, acrylonitrile, chlorinated hydrocarbons, chlorine,
methanol, phenol, TEL and TML. All carbon tests were conducted with
filtrasorb 400 manufactured by Calgon Co., Pittsburgh, Pennsylvania.
REVERSE OSMOSIS TREATMENT
Reverse osmosis treatment feasibility tests were conducted with a hollow
fine fiber membrane made of a polyamide nylon and designated as B-9
by the manufacturer, Du Pont Co. A half size B-9 permeator (membrane
and the pressure vessel) with a nominal capacity of 0.09 I/sec (2100 gpd)
was utilized for these studies.
The test solutions were made in tap water in desired concentrations.
Membrane rejection capabilities were evaluated by analyzing the raw
product and brine streams for pertinent constituents. Flow rates,
temperatures and pressures were recorded during each test. The tests
were generally of short duration and lasted between 1 to k hours. A
detailed description of the test setup and procedures is shown in
Appendix B. All the materials listed in the second section of Table 1
were screened for RO feasibility tests based on available literature;
however, actual RO tests were conducted for acetone, acetone cyanohydrin,
acrylonitrile, ammonium nitrate, cooper sulfate, formaldehyde,
methanol and mercuric chloride.
-------
SECTION V
TEST RESULTS AND EVALUATIONS
The test results and data have been divided into two separate sections as
listed earlier in Table 1. The nine materials considered in the selection
of the mobile treatment system are discussed first for chemical treatment,
clarification and carbon adsorption treatment. A second section presents
the results of the RO feasibility tests.
ACETONE CYANOHYDRIN
It was indicated in the literature (2) that acetone cyanohydrin may be
detoxified at high pH levels by the precipitation of cyanides via
chlorination or carbon adsorption treatment. Therefore, chlorination
and carbon adsorption tests were conducted for this material.
Chlorination Tests
Solutions containing approximately 900 mg/1 acetone cyanohydrin were
treated with sodium hypochlorite at ratios of 2.9 and 1.1 parts of
chlorine per part of acetone cyanohydrin at pH values of 9, 10, 11,
and 12 (Table 2) for 30 minutes. Chlorine concentrations were measured by
iodometric tit ration and acetone cyanohydrin was measured by gas
chromatography. Sodium hydroxide was used to adjust the pH at the start
of the tests. From these tests it was found that chlorine dosages of
at least 3 parts chlorine per part of acetone cyanohydrin must be used
at a pH of at least 12 to achieve any significant reduction (about 50%)
in acetone cyanohydrin concentration. This treatment therefore, was
not considered to be a feasible method of treatment for spills of this
material. Additional tests were conducted with anthiurn-dioxide.
Anthium dioxide is a stable solution containing 50,000 mg/1 chlorine
dioxide. Chlorine dioxide is reported to be a more powerful oxidizing
agent than chlorine and results in fewer chlorinated byproducts. The
anthi urn dioxide sample tested had an oxidation capacity equivalent £o
1^5,000 mg/1 chlorine as determined by iodometric titration. The
dosage of anthium-dioxide used in these tests is expressed here in
terms of equivalent chlorine concentration to facilitate comparison
with the previous sodium hypochlorite test work. The results of tests
performed on solutions containing about 8,000 mg/1 acetone cyanohydrin
are listed in Table 3- These tests were performed using ratios of
-------
r>"»oo LA o on o LA cr>
oo vo o *~
co r>- o o CM
IAOO i-» o r^ r-. o co LA LA
O^ CM vO «*> t
o oo oo o o o
oo on on o en o on LA o
on on on fi CM
cor-v oooo
m o P^ LAOO r-». o -a* LA vo
o LAND crv _ . . _
cnoo o o ON oo
o ooo OSO-TOCM oo o
r^» lACMC^ Jf CM CO vD
Q oo -a- -w -a- -a- pg CM
I CM CM CM IA
O
O » OMA OlTVlTVOvO OOO
ra voooi** co ff\^r com
z rO OOCM
o
I
LU
O
O
i
OOOO LAvOOOCTtOA -CM
CM CM
u
o
* §
u
CM
9)
(D (0 4)
X *J L. >
U I/I 4) Q
*- E
4) *J u- 4)
C ID 10 I.
O
4)
CM
U
I/I *
m c
- -oi
o> -o 10
e -o 4)
ID u
4J I/I
O 'c 1- £
§3 4) -O 10 -O 4J
0 « 4) * C (0
U E «*- in 1/1 4) *-
10 <0 3
O <0 10 5
o *».
at ^f "f" * *
nj ^c_ j. +^
z a a 3
c -o
4) 4)
w >
4, g
0. 4)
t_
C
^ C
w *~
o u
>. -o
-C X
O £
i §
X «0
0 X
u
4)
84)
» 1
U V
(0 U
V. «0
"O *»»
41 -0
0 4)
"S 3
-------
Table 3. OXIDATION OF ACETONE CYANOHYDRIN WITH CHLORINE DIOXIDE
Test No. 1 2 3 *
Acetone cyanohydrin cone., mg/1
at start 8660 8070 7600 8190
after 0.5 hours 8120 7440 5240 7460
after 1.5 hours 7000 5400 4980 4160
Acetone cyanohydrin removed, mg/1
after 0.5 hours 540 630 2360 730
after 1.5 hours 1660 2670 2620 4030
Anthium dioxide cone., mg/1 as Cl2
added to sample 5400 10000 14200 10200
after 0.5 hours 4752 8153 4927 8720
used in 0.5 hours 648 1847 9273 1480
pH at start 5.00 5.00 5.00 6.00
pH at end 5.00 5.10 4.40 5-90
Wt/wt ratios:
Cl2 added/acetone cyanohydrin 0.62 1.24 1.87 1.24
C12 used/acetone cyanohydrin (1/2 hr) 1.20 2.90 3.90 2.00
10
-------
chlorine and acetone cyanohydrtn of 0.6:1 to 1.9:1 at pH levels of 5 and
6. The chlorine dioxide is released by reducing the pH of the solution
to less than 7; the lower the pH, the faster the reaction. After 30
minutes, the amount of chlorine dioxide remaining was measured by iodo-
metric titration. Acetone cyanohydrin concentration was measured by
gas chromatography at thirty minutes and 90 minutes. It appears that
relatively long contact times are needed (more than 1/2 hour) for
treatment at the pH levels tested. Lower pH levels are undesirable
because of the hazards of cyanide formation. For these reasons and
since only a maximum of 50% removal was achieved, anthium dioxide was
not given further consideration as an acetone cyanohydrin spill counter
measure.
Activated Carbon Treatment
Isotherm tests were performed on two solutions containing 888 mg/1 and
9120 mg/1 acetone cyanohydrin in tap water. Lime was added to each
solution to raise the pH to 11 to prevent hydrogen cyanide gas formation.
The solutions were filtered prior to running the isotherm tests. The
results of the isotherm tests are plotted in Figure 1. The estimated
adsorption capacity for the two solutions were 0.27 mg/mg carbon for
the 888 mg/1 solution.
Two carbon column tests were also made on acetone cyanohydrin at feed
concentrations of 9,225 mg/l and 95 mg/1. The results of the first
test, at a hydraulic loading rate of 3.27 l/sec/m2 (k.8 gpm/ft2) and a
concentration of 9200 mg/1 is shown in Figure 2. Nearly complete removal
of acetone cyanohydrin was achieved. The adsorption capacity was found
to be 0.203 mg/mg carbon compared to the isotherm value of 0.27 mg/mg
carbon.
The second carbon column run with acetone cyanohydrin was made at a feed
concentration of 95 mg/1 and a hydraulic loading of 2.7 1/sec/m2
(k.\ gpm/ft2). The results of this test are presented in Figure 3.
Again, complete removal of acetone cyanohydrin was achieved. The adsorp-
tion capacity was found to be 0.016 mg/mg carbon or 85% of the carbon
isotherm value (0.019 mg/mg carbon). During the second run the presence
of cyanide in the effluent was also investigaged. Cyanide was found
to be present in measurable quantities as shown in Figure 3. This
presents a problem, as cyanide is toxic, and is evidently released by
the adsorption reaction. However, cyanide can be precipitated by the
ferric ion (2).
As evidenced by the carbon column breakthrough curves, activated carbon
effectively reduces the discharge of acetone cyanohydrin and can be
used for such spills. Minimum contact time was found to be approximately
50 minutes.
-------
1.0
g o.i
en
03
OC
O
0>
0.01
0.001
Co -
100 1 ,000
RESIDUAL CONCENTRATION, mg/1
Co - 9,120
I .
10,000
Figure 1. Carbon isotherm results for acetone cyanohydrin
12
-------
o
o
o
C
0>
o
C
O)
o
-T
O
CM
O
O
s e
o
0
o
o
m
I
u
m
o
o
-------
O
o
10
4J
c
u
8
Vt
U
1
5L
u
fl>
c
o
u
o
I
u
19
o
u
3
O>
OO
o
vO
l/6ui
-------
ACRYLONITRILE
Literature (2) indicated that this compound must also be handled at
elevated pH values of above 9-0 to prevent the formation of hydrogen
cyanide gas. Chlorination and activated carbon adsorption were again
indicated to be viable methods of treatment for this compound. Since
this material was considered to be quite volatile, tests were also
conducted to study the volatalizatton of this solution prior to chemical
treatment tests.
Volatility Tests
Two solutions containing an initial concentration of 7600 mg/1
acrylonitrile were allowed to stand in open beakers at room temperature.
One solution had a pH of 7.8 and the other had a pH of 11.7. The ratio
of solution surface to volume was approximately 0.08 m2/! (3.3 ft2/gal.).
Samples were removed periodically for analysis of acrylonitrile. The
rate loss of acrylonitrile was the same for each solution; the loss
followed a first-order reaction with a K constant of 0.015 min*'.
The "half life" (elapsed time required for loss of half the acrylonitrile)
was 47 minutes at ambient temperature. A check for cyanide ion showed
that the acrylonitrile loss was not due to cyanide formation in either
solution. Although conditions of temperature, wind and ratio of surface
area to volume of contaminated water may vary widely for different spill
situations, it is quite likely that appreciable quantities of acrylonitrile
will be vaporized before carbon adsorption equipment is ready to treat
the contaminated water.
Because of the above observation, all further chemical tests for
acrytonitrite were performed in stoppered erlenmeyer flasks equipped
with a septum such that samples for analysis could be withdrawn with a
syringe to insure reproducibi1Ity.
Chlorination Tests
Three tests were performed to determine the effect of chlorination on
gas chromatograph acrylonitrile solutions. Solutions containing 5900
mg/1, 4300 mg/1 and 353 mg/l acrylonitrile were used. Sodium hypochlorite
was added to these solutions at ratios of 0.3, 1.4 and 16.4 mg per
mg acrylonitrile. The remaining acrylonitrile and chlorine concentrations
were measured after one hour of contact. The results of these tests
are listed in Table 4. The amount of chlorine required varied from 2.5
to 4.9 mg chlorine per mg acrylonitrile destroyed. It should be noted,
however, that as the concentration of acrylonitrile decreased, other
compounds were formed as Indicated by the appearance of two additional
peaks on the chromatogram. Although It was not possible to establish
15
-------
the identity or quantity of the two byproduct compounds, an approximation
of the relative amount of material obtained from the area of the two
peaks (75 mm2) relative to the acrylonitrile peak (188 mm2) demonstrate
that a large portion of the acrylonitrile was converted to other organic
materials which will require further treatment. Because of the need for
large chlorine dosage and the production of byproducts, further work
with sodium hypochlorite was not done.
A few tests were performed to investigate the effect of ferric ion on
acrylonitrile. No appreciable reduction in acrylonitrile was observed.
Iron salts may have an application only if the acrylonitrile has
hydrolyzed (in nature) to form cyanide. It is not recommended that
hydrolysis be induced if it has not occurred via natural processes.
Table 4. CHLORINATION OF ACRYLONITRILE
Test 1 Test 2 Test 3
Initial acrylonitrile cone., mg/1 5900 4330 353
pH 10.8 10,7 8.9
Acrylonitrile cone, at I hr, mg/1 5600 3120 165
Acrylonitrile removed, mg/1 300 1210 188
Chlorine added, mg/1 1640 6015 5800
Chlorine at 1 hr, mg/1 560 3170 4885
Chlorine used, mg/1 1080 2845 915
Chlorine added: Acrylonitrile added 0.3:1 1.4:1 16.4:1
Chlorine used: acrylonltrile removed 3.6:1 2.4:1 4.9:1
Carbon Adsorption Tests
Carbon isotherm tests were performed on three solutions of acrylonitrile
diluted with tap water with a minimum pH of 9. Isotherm tests were
performed on solutions containing 73, 805 and 7600 mg/1 acrylonitrile.
The results of these tests are plotted in Figure 4. The estimated
adsorption capacities from this figure are: 0.25 mg/mg carbon for the
7600 mg/1 solution; 0.072 mg/mg carbon for the 805 mg/1 solution and
0.019 mg/mg carbon for the 73 mg/1 solution.
16
-------
o
0
o
_. A
°£v
o\p
\
\
\ 0 -
\ IA 0
\ f> O ^
\ r»» co ^
*\ ill
\ O O 0
~ u u o
I
V < o
\
*\
\
\a
^
\
\
^
\
\
\
\
\
\
V3
\
\
V.
<3
\
i i i
o o
o o
o
o
o
o u
O *J
"" %
*>.
u
^ o
i- -
« £
o
Ul
O 1- **
0 rf
& 3
K- Wl
Z 0
III
UJ b
u
g i
0 0)
_l 4->
^ 8
O ~
O Ul C
ui o
iZ
6ui/o3gyoSOV
17
-------
Carbon column tests with acrylonitri1e at a concentration of approximately
600 mg/1 were performed. Two hydraulic rates utilized were 1.8 and
3.4 1/sec/m2 (2.7 and 5.0 gpm/ft2). The results of these tests are
presented in Figure 5. There was essentially no difference in the adsorp-
tion capacity as a function of hydraulic loading. Nearly complete
removal of acryloni tri le was achieved. Adsorption capacities were 0.143
and 0.134 mg acrylonitrile per mg carbon for hydraulic loadings of
3.4 and 1.8 1/sec/m2 (5.0 and 2.7 gpm/ft2) respectively. The carbon
isotherm for acrylonitrlie predicted 0.22 mg acrylonitrile per mg of
carbon. Hence the column test showed only about 63% adsorption capacity
compared to the isotherm value.
A second test of acrylonitrile was conducted at a lower concentration,
70 mg/1 and at a hydraulic loading of 2.9 1/sec/m2 (4.3 gpm/ft2). The
results of this run are shown in Figure 6. The adsorption capacity,
as determined by the column test was 0.0142 mg acrylonitrile per mg
carbon as compared to the isotherm value of 0.019. It should be noted
that the rate of breakthrough of acrylonitrile through the carbon column
at the lower concentration of 70 mg/1 was significantly slower than at
the higher concentration of 6000 mg/1.
This testing demonstrated that activated carbon could be utilized
effectively for the treatment of acrylonitrile solutions. Minimum contact
time was found to be approximately 40-50 minutes. Because of the volatile
nature of acrylonitrIle, a significant amount will evaporate to the
atmosphere and create an air pollution hazards. It is therefore
especially imperative that rapid action be taken by personnel protected
by adequate safety equipment.
AMMONIA
Oawson et al (2) indicated that the most applicable technique for the
handling of ammonia spills is dilution. It is also indicated that
stripping of ammonia at high pH values may be applicable in certain
situations. Chlorination can also be used for the oxidation of ammonium
ion to nitrogen. Information concerning the oxidation of dilute
concentrations of ammonia with chlorine is available in the literature.
Generally, weight/weight ratios of C^NHj of 4:1 to 10:1 are recommended
for complete destruction of ammonia. Lower concentrations of chlorine
results in chloramine formation. The reaction is reported to be fastest
at pH 8.3* The actual amount of chlorine needed at a spill site would
vary, of course, depending on the amount of other materials in the water
that would also exert a chlorine demand. One test was performed on the
chlorination of solutions containing approximately 850 mg/1 of ammonia.
Sodium hypochlorite was added to solutions of ammonium chloride at
ratios of 1.16, 2.92 and 5*85 parts chlorine per part of ammonia at pH
values of 8.5, 11 and 12. After a thirty minute reaction period, the
amount of chlorine remaining was measured by iodometric titration.
18
-------
c
O
ra
u
4-1
0)
u
c
O
u
41
U
4J
C
O
u
Q
u
O
0)
O
u
O
JQ
0
l_
B.
O
O
O
sO
O
O
O
h
LTV
O
O
O
O
§
04
O
O
O
l/6ui 4
Nl NOIlV>JiN33NOD
19
-------
01
E
UJ
o
o
t/6ui '
Nl
20
-------
A portion of the sample was then dechlorinated and the remaining ammonia
and the chloramines converted to ammonia by the dechlorination and
measured by nesslerization.
The results are listed in Table 5- At pH 8.5, a C12:NH3 ratio of 5.85
was needed for a significant reduction in ammonia. There was no
ammonia destruction at lower ratios at a pH level of 8.5. Some ammonia
was however, destroyed at a ratio of 2.9:1 at higher pH values.
Although some of the ammonia may have been lost by volatilization at the
high pH values, the amount of chlorine used indicated some destruction.
Recent literature (A) indicates that a 7:1 to 10:1 C12NH^ ratio is
needed to effectively remove all the ammonia. Hence, ammonia removal by
breakpoint chlorination is chemically feasible for relatively dilute
concentrations of ammonia spills, but such treatment does not appear
to be logistically possible owing to the large amounts of chlorine
required and the secondary pollution it would cause.
CHLORINE
Activated carbon may be utilized for the dechlorination of chlorine
contaminated solutions. One isotherm test has been performed on a sample
of sodium hypochlorlte containing 9290 mg/1 chlorine. The results of
this test are plotted in Figure 7- The estimated adsorptive capacity
at 9290 mg/1 was 1.7 mg chlorine/mg carbon.
Column tests were performed on a 9000 mg/1 solution of chlorine to
further investigate the feasibility of chlorine removal by activated
carbon at a hydraulic loading rate of $.k l/sec/m^ (5 gpm/ft^). An
adsorption capacity of 2.5 mg C^/rog carbon was achieved, which was
greater than predicted by the isotherm tests. The minimum contact
time was found to be of the order of only a few minutes. It is
indicated that chlorine removal by activated carbon is not an adsorption
phenomenon but an oxidation-reduction reaction. Chlorine is reduced to
chloride utilizing an activated carbon molecule which in turn gets
oxidized to carbon dioxide. This production of C0£ causes an extra
buildup of pressure inside the carbon column and may cause short
circuiting by gas pocket formation. The treated carbon column effluent
was a characteristic dark brown color imparted by the extremely fine
carbon particles produced due to the oxidation of activated carbon. The
color of the treated effluent was much less intense, however, when the
influent waste pH was controlled between 5 and 6. Dilution of chlorine
spills prior to carbon treatment might also minimize this problem.
It is therefore indicated that granular activated carbon treatment is
generally feasible for mitigating chlorine spills.
21
-------
O CO
CM
O
V)
m
c
a>
V)
v
a.
T)
0)
O
o
V)
u
o
u
re
'c
a
c
0)
^J
n>
-D
4)
>
£
0)
L.
»
u
c
o
u
3
o
re
z
T3
a c
(0 « -0
T3 0)
o +> 1/1
re re D
4->
IM
re
4-1
in
4J
re
X
a
-o
c
0)
4J
re
X
a.
(A
O
4-1
re
u
4*^
5
*v
4->
3
4)
TJ
"O
re
CM
e~
0
PA
X
z
^^
CM
^M
O
22
-------
100 J-
2
o>
Ct
UJ
m
ee
o
o>
10
1.0
0.1
Co - 9,290
J_
10 100 1,000
RESIDUAL CONCENTRATION, mg/1
Figure 7- Carbon isotherm results for chlorine
10,000
23
-------
CHLORINATED HYDROCARBONS
White activated carbon adsorption was anticipated to be among the most
effective methods for the removal of most spilled chlorinated hydrocarbons,
pesticides and herbicides, a brief review of the literature showed that
a great deal of work has also been done on the adsorption of pesticides
on various other materials. The data of King, Yeh, Warren and Randall
(Table 6) show substantial removals of lindane from aqueous solutions
when concentration of soils was high (8). In contrast, Lotse, Graetz,
Chesters, Lee and Newland (Table 6) found much lower removal efficiency
at a low concentration (372 mg/1) of soil (9). Efficiency was reported
to be even lower when soils containing less organic materials was used.
The adsorption of 2, 4-D on montmori1lonite clay was found to be minimal
by Schwartz (10). On the other hand, Huang and Liao have reported
removal of DDT, heptachlor and dieldrin using clay (11). It appears from
this data that the removal of suspended solids in contaminated water at
a spill site will remove the chlorinated organics which have been
adsorbed on the suspended particles of a suitable soil.
Carbon isotherm tests have been reported for several pesticides and
herbicides using a variety of carbons. Morris and Weber (12) reported
results of carbon adsorption on several herbicides including 2,4-D, 2,4,5-T,
Sllvex and parathion. The adsorption isotherms in the Morris and Weber
papers are plotted as Langmulr isotherms. In order to be consistent
with the carbon isotherm plots used in our studies, the data points
from the Langmiur plots were recalculated and plotted as Freundlich
isotherms and are shown in Figure 8. King, et al also performed a
carbon isotherm on parathion. In this test, acetone was used as a
bridge solvent (8). The carbon isotherm plot is also shown in Figure 8.
Aly and Faust performed carbon isotherms on several herbicides including
2,4-D and the iso-octyl and butyl esters of 2,4-D. The isotherm plots
in Figure 8 were constructed from the Freundlich adsorption parameters
reported in their paper (13). Carbon adsorption tests were performed
on 2,k,$-% butoxy-ethanol ester, dieldrin, endrln, lindane, and
parathion by Robeck, et al (14). These tests were performed using distilled
water and water from the Little Miami River in Ohio. The carbon isotherm
lines for these herbicides are also shown in Figure 8. The effect of
other organics in the river water on the adsorption of pesticides
(as on other compounds) is clearly illustrated by this data. It is
quite likely, of course, that the organic content of the contaminated
water at a spill site may be quite high and that these organics will
compete with the hazardous material for the activated carbon, and that
the pesticides and herbicides will not be well adsorbed. All the isotherms
mentioned are plotted in Figure 8 to help estimate the carbon treatment
capacity of these materials.
Very little information was found in the literature concerning the
adsorption of DDT. Although Huang and Liao reported on the adsorption
of DDT on various clays, no reference was found concerning the
adsorption on activated carbon (11). Indeed, information concerning the
-------
ID
t>
U
4)
v.
4>
V4-
4>
OC
OO CO CO CO CO CO
v
C O>
»^ ^1
«4-
"
^ ^^.
*J 0)
a
c
CM
VO vO tA
O
SO
OO
oooooooooo
«nooooooooo
C
o
u
4)
CO
1/1
u
0)
U
3
*J
ID
4)
U.
O
O.
oe
o
CO
o
V
-O
ID
0)
(A
0)
Q.
O
o
I/I 4->
ID
l_ 1.
O 4->
> w
ID u
o o
u
QJ *J «- p
(11 Q. Q. Q. 0) 0> 0)
- a o a vai 4) .-
(Mooeaxxxooa
C7)
OOOOOOOOOO
oooooooooo
OOOOOOOOOOO
T3
4)
U
4)
c
0)
4)
»t-
a>
4)
4)
4)
4)
4)
a.
1- X
o *
O I/I
ID ID
-O T3
W tfl
^£ ^ ._ ._ U U
i- u u u o o
V 0) «) 4>
CO CO O <_> Li- U.
C
o
C
re
CD
L.
.2
~ 0)
^ -W «"
t_ U
J5
4) 4)
C
O
4)CO4)CQ4)CQ
o E*j. E4J~ E*J» E
c^"oc "oc "oc
4>
U
C
4>
l_
4)
li-
ft)
DC
ID
-------
1.0
O.I
c£
O
O)
CO
cc
O
CO
o
i1
0.001
0.0001
0.00001
0.1
1 10 100 1,000
RESIDUAL CONCENTRATION, yg/1
10,000
I.
2.
3.
k.
5.
6.
7.
8.
Parathion
Parathion
Parathion - D.W.
Parathion - river
2,4-D
2,4-D at pH 3
2,k-D at pH 7
2,*l-D at pH 11
9. 2,4-D 16.
10. 2,4-D buthyl ester 17.
1). 2,4-D iso cotyl ester 18.
12. 2,4,5-T 19.
13. 2,4,5,-T butoxyethenol ester 20.
in distil led water 21.
14. 2,4,5-T butoxyethanol ester 22,
in river water
15. Lindane - D.W.
1.0
0.1
0.01
0.001
0.0001
0.00001
Lindane - river
Endrin - D.W.
Endrin - river
Dieldrin - D.W.
Dieldrin - river
DDT
Silves
Figure 8. Carbon adsorption isotherms for various pesticides & herbicides
26
-------
solubility of DDT is confusing, Hartung and Klingler (15) cite
references (16)(17) which report the solubility of p,p'-DDT as being
37 ug/1 and 1.2 yg/1 respectively. Robeck, et at (14) filtered a
suspension of a technical grade of DDT (containing o,p-DDT, p,p'-DDT
and DDE) through a 0.05 u filter and a 5 y filter. The total concentration
of the two DDT isomers and DDE passing the filters was 16 ug/1 and 40 yg/1
respectively. It was concluded that the apparent solubility was due
to the size of pesticide particles remaining in suspension.
Because of the scarcity of data regarding DDT, a carbon isotherm test was
performed in the laboratory. A one gram portion of p,p'-DDT was mixed
with three liters of tap water and was allowed to sit for seven days.
The DDT remaining in suspension was removed by filtering through an 0.45 u
membrane filter. Analysis of the filtrate for p,p'-DDT revealed less
than 0.2 ug/1 concentration. Because of this, a second mixture of DDT
and water was prepared and mixed for five days. A 200 ml portion was
filtered through a 0.45 u membrane filter and another 200 ml portion
was filtered through a Type A Gelman glass fiver filter (47 mm diameter).
Again, no p,p'-DDT was found in the filtrate from the membrane filter
(<0.2 ug/1). The filtrate from the glass fiber filter contained 61 yg/1
p,p'-DDT. The remaining mixture, approximately 2.5 liters in volume,
was filtered through a second glass fiber filter and was found to contain
26 yg/1 p.p'-DDT. This filtrate was used for the carbon isotherm tests.
The carbon isotherm tests were performed in 250 ml erlenmeyer flasks
containing 200 ml filtrate, a selected volume of carbon slurry
(powdered Filtrasorb 400) and distilled water to bring the total volume
of 210 ml. The two carbon slurries used contained 1 mg carbon/ml and
10 mg carbon/ml respectively. The isotherm flasks contained 0, O.I mg,
1 mg, 10 mg and 100 mg carbon. After one hour of shaking on a Burrell
shaker, the contents of each flask were filtered through a clean glass
fiber filter. The filtrates were analyzed for p,p'-DDT. The results
of this isotherm test are plotted in Figure 9- A reduction in DDT
concentration from 26 ug/1 before the isotherm test, to 5-1 ug/1 in the
sample containing no carbon may have been due to removal of more DDT
by the second filtration. However, the adsorption capacity demonstrated
in Figure 9 is 0.00053 mg DDT/mg carbon. Evidently activated carbon
does adsorb some DDT.
Based on the DDT testing and the literature search, it is evident that
activated carbon has applicability in removing water soluble herbicides
and pesticides of various varieties. The lethal nature of these
compounds makes their control at spill sites essential. The relative
removal capacity of activated carbon varies widely. Because of the low
solubilities of pesticides and herbicides, carbon capacity for adsorbing
them should not be a major problem, as long as adequate contact time is
avaliable.
27
-------
0.001
0.0001
ADSORPTION CAPACITY
- 0.00053 mg DDT
mg CARBON
Ul
CD
ee.
s
o
CTI
§
01
0.00001
0.000001
0.1
Co - 5.1
1.0
10
RESIDUAL CONCENTRATION, wg/1
Figure 9. CARBON ISOTHERM PLOT FOR pp'-DDT
28
-------
MET HANOI
Dawson et al indicated that the only method of removing a methanol
spill was to allow dilution and subsequent blodegradation to occur (2).
Although the literature indicated poor removal of methanol by activated
carbon, no concrete data could be found. Therefore, carbon tests were
conducted to establish the extent of removals.
Carbon isotherm tests were performed on two solutions of methanol in
tap water whose concentrations were 73 mg/1 and 7253 mg/1 respectively.
From the results of the two isotherm tests listed in Table 7, it can
easily be seen that the adsorption of methanol by the activated carbon
used is very poor. A log-log plot of the data is shown in Figure 10.
The adsorption capacity was estimated to be 0.028 mg/mg carbon for the
7253 mg/1 methanol solution and 0.0011 mg/mg carbon for the 73 mg/1
solution.
Table 7. RESULTS OF CARBON ISOTHERM TESTS
PERFORMED ON METHANOL SOLUTION
Test Carbon used, Methanol remaining, mg Adsorbed
No. grams mq/1 per mg carbon
1 0.0 7253
J 0.2 7220 0.0165
1 0.5
1 1.0 7120 0.0133
I 2.5 6930 0.0129
1 5.0 6810 0.0089
2 0.0 73.0
2 0.1 Ik.k
2 0.5 70.0 0.0006
2 1.0 69.5 0.00035
2 2.5 66.8 0.00025
2 5.0 59-6 0.00027
29
-------
0.1
2 o.oi
ee
(71
CO
a:
o
0.001
Co - 73 Co - 7,253
0.000l| I , I L_4.
10 100 1,000 10,000
RESIDUAL CONCENTRATION, mg/l
Figure 10. Carbon Isotherm results for methanol
30
-------
A carbon column test was also performed to verify the isotherm results.
The results of this test are presented in Figure 11. The feed concen-
tration was approximately 8000 mg/1. The adsorption capacities were
found to be quite low at 0.028 and 0.926 methano I per mg carbon at
hydraulic loading rates of 3-4 and 1.8 1/sec/m2 (5 and 2.7 gpm/ft2)
respectively. Breakthrough of the carbon column was almost instantaneous.
An additional carbon column test performed at a methanol concentration
of 77 mg/1 provided similar results which are shown in Figure 12. The
adsorption capacity at this influent concentration was found to be
extremely low at 0.00038 mg methanol per mg carbon at a hydraulic loading
of 3.4 1/sec/m (5 gpm/ft2). Therefore, it was concluded that activated
carbon is not suitable for the treatment of methanol spills.
PHENOL
Oawson et al (2) indicated that phenol spills may be successfully
handled both by chemical treatment with chlorine and activated carbon
adsorption. It was also indicated that if the phenol concentration was
excessively high, dilution prior to treatment may be necessary.
Laboratory tests on phenol were conducted both with chlorine and
activated carbon.
Chlorinatlon Tests
A test was performed to determine the approximate amount of sodium
hypochlorite needed to oxidize a dilute phenol solution. The phenol
solution was dosed with the desired amount of sodium hypochlorite, the
pH adjusted to 8.5 and the solution allowed to react for thirty
minutes. The amount of chlorine and phenol remaining was then measured.
The results of these tests are listed in Table 8. It is apparent from
this data that a chlorine dose of two to four times the amount of phenol
present is needed for a significant reduction in phenol concentration.
Chamber 1 in and Griffin (18) report that the amount of chlorine needed
is six to ten times the phenol concentration, when treating a similar
concentration of mixed phenolics. Therefore, it is concluded that
chlorination of phenolic spills is unattractive not only because of
the large amounts of chemical requirements but also becuase of possible
chlorinated reaction products.
Table 8. RESULTS OF CHLORINATION OF PHENOL SOLUTION
Test 1 Test 2 Test 3 Test 4
Initial phenol concentration, mg/1 117 115 114 110
Chlorine added, mg/1 109 268 495 1920
mg chlorine added/mg phenol 0.1/1 2.2/1 4.3/1 17-5/1
Final phenol concentration, mg/1 95 15 2 <0.05
Final chlorine concentration, mg/1 <1 28 0 1307
mg chlorine used/mg phenol destroyed 4.9/1 2.4/1 4.4/1 5.6/1
31
-------
c
o
«
u
o>
g
5
1.
o
c
0
U
3
Ol
O
o
o
oo
o
o
o
o
§
o
o
o
«
ol
Nl NOIJ.V»1N30N03 10NVH13W
32
-------
CM
i.
Ol
in
I
4J
n
o
8
o
o
vO
O
IA
M 1.
o « o
-a- * <*.
E «
«j
*
c
O £
o
-------
Activated Carbon Adsorption
Carbon isotherm tests were performed on four solutions of phenol. The
concentration of phenol was measured using the 4-amino antipyrine
color(metric method with carbon tetrachloride extraction. Three of the
phenol solutions tested were solutions of phenol in Milwaukee tap water
without pH adjustment; the pH of these solutions varied from 7-4 to
7.5. The pH of the fourth solution was adjusted to 10 with sodium
hydroxide before the isotherm test. The results of the isotherm tests
are shown in Figure 13- The results from the test performed on the high
pH phenol solution were not significantly different from the results of
the other tests. The adsorption capacity was estimated to be 0.41
mg/mg carbon for the 9040 mg/1 phenol solution, 0.24 mg/mg carbon for
924 mg/1 and the 1026 mg/1 solutions and 0.14 mg/mg carbon for the 102
mg/1 solution.
Activated carbon column testing was performed on a phenol solution of
9300 mg/1 at a hydraulic loading rate of 3.3 1/sec/m2 (4.9 gpm/ft2).
The results of this run are presented in Figure 14. Complete removals
of phenol were achieved by carbon adsorption. The adsorption capacity
was found to be Q.212 mg per mg carbon which was 4?% of the carbon isotherm
value of 0.45 mg/mg carbon. The above values of carbon adsorption
capacities for phenol match quite closely to the data shown in Table 9
provided by the Calgon Corporation (manufacturer of Filtrasorb 400).
Since sufficient data was available from the manufacturer on the
carbon adsorption efficiencies at lower phenol concentrations, no further
carbon column tests were conducted with phenol. From this data, it is
apparent that activated carbon can effectively treat phenol spills.
Table 9. PHENOL ADSORPTION EFFICIENCIES ON FILTRASORB 400a
Concentration of Contact time, Phenol loading,
phenol, ppm minutes % by weight
16 30 5 - 10
10,000 75 20 - 25
1,140 18 10 - 20
1 15 3-5
2,500 30 12 - 15
225 75 20
2,500 80 23
a. Data provided by Calgon Corporation.
-------
o
o
o
V 1
V
\
\
V
»L
\
\ ^ ^ I1 1
v E £ ^0 O
\
\
\
\
\
^0
\
\
\
\0
\
1 I 1
0
O O
o o
o
o
o
0 0
o e
"~ ex
i.
o
vt-
^Z
? £
0-1.
0 Z
- 2 g
!» ^)
S 5
z 8
UJ "-
o
z c
O Q
O -Q
-1 ft
O < O
^ cA
QJ P^
oe
>
k»
3
u.
o
35
-------
§
o>
"c.
I
-------
ORGANO LEAD COMPOUNDS (TML AND TEL)
Both tetraethyllead (TEL) and tetramethyllead (TML) are heavy liquids
having specific gravities of tOO and 165 respectively. Both these
compounds are considered highly toxic and have limited solubilities in
water. According to Shapiro and Frey (19), the solubility of these
compounds is about 0.2 to 0.3 mg/1 in water. However, higher concentra-
tions of these compounds can occur in contaminated waters at a spill
site due to emuls!fication or adsorption on solids. TML is also an
unstable compound and generally contains 20% to 30% toluene for stability.
TEL is generally difficult to recover, but can be complexed with a calcium
salt or EDTA or dimercaprol (2). Laboratory tests were conducted to
establish the expected concentrations of these materials at a spill site.
Chemical treatment and carbon adsorption tests were then conducted on
solutions containing the expected concentrations of TEL and TML.
Solubility Tests
Some preliminary tests were conducted by mixing known amounts of TML and
TEL in water. in one test, approximately 5 ml of TEL was added to
750 ml tap water in a glass stoppered one liter bottle. The contents of
the bottle were shaken vigorously and allowed to stand for 15 minutes to
allow the TEL droplets to settle, immediate TEL analysis of a supernatant
showed a concentration of 12.8 mg/1 tetraethyllead and 3.3 mg/1 as ionic
lead. A second sample of supernatant, removed three hours after mixing
in the glass bottle, showed 6.8 mg/1 tetraethyllead and 13 mg/1 ionic
lead. Another mixture of TEL and water was mixed several times over
a period of four days and allowed to settle. The supernatant contained
less than 2 mg/1 of TEL but had an ionic lead content of 46.5 mg/1.
Evidently the TEL was degrading during mixing and storage. Also, a highly
volatile organic liquid was observed floating on the surface of the
water when stored in stoppered bottles.
Shapiro and Frey (19) report that organolead compounds may be decomposed
by ultraviolet light and that prolonged contact with air, especially
in the presence of light, can cause gradual decompositions. A test
was performed to study the effect of light and air on a TML water mixture.
Approximately two liters of tap water that had been deaerated by bubbling
nitrogen through it, was vigorously mixed with 2 ml TML (80% TML in
toluene). The air space above the mixture was flooded with nitrogen
prior to mixing and during a 20 minute settling period. After settling,
portions of the supernatant were transferred to the test flasks under
the following conditions:
Flask 1 - covered with foil to exclude light and stoppered to
exclude air
Flask 2 - stoppered to exclude air, but exposed to light
37
-------
Flask 3 - covered with foil to exclude light, but exposed to air
Flask k - exposed to both air and light.
The stoppered flasks were flushed with nitrogen prior to and after the
addition of the samples. The stoppers were fitted with septurns to
allow the removal of sample by syringe for gas chromatography without
exposing to air.
A fifth flask was prepared containing deaerated water and 0.5 ml TML.
The flask was covered with foil and stoppered. Vigorous mixing of the
TML and water was avoided to prevent the formation of suspended
droplets of TML. The purpose of the fifth flask was to determine the
solubility of TML in water which is in contact with excess TML.
The results of these tests are listed in Table 10. The following
observations can be made:
1. The TML concentration of the solutions exposed to air dropped
to less than half the concentration of solutions protected
from air within two hours.
2. The TML concentration of all four test solutions decreased to
about the same level after 2 days.
3. The concentration of TML in the four test solutions was approx-
imately half the concentration of TML in the water in flask 5
after 2 days.
It appears that the decrease in TML concentration in water with time
is not due entirely to exposure to light and air, or to settling of
the suspended TML droplets. The dynamic nature of these solutions demon-
startes that extreme care must be taken in testing them. Analysis for
TML should be made as quickly as possible. In order to evaluate the
maximum possible concentration levels of TML in water, about 5 ml of
TML (80% TML and 2Q% toluene) was added to about 800 ml tap water in
a glass stoppered bottle. The contents of the bottle were shaken
vigorously, then allowed to stand for various periods of time to allow
the large drops of TML to settle. The supernatant was then decanted for
tests. The properties of the TML-water mixture were generally similar
to that of the TEL-water mixtures previously tested. The major difference
observed is that concentrations of TML in water were higher than the
corresponding TEL concentration and ranged generally between 20-30 mg/1.
38
-------
Table 10. STABILITY OF TETRAMETHYLLEAD IN WATER
CONCENTRATION OF TML, mg/1
Exposed to 1 ight
Exposed to air
Elapsed Time, Hrs.
0.3
0.5
0.7
0.9
.0
.2
.4
.6
.7
.9
2.1
2.2
46.2
46.3
46.5
46.7
47-0
47.1
47.3
47.5
47.7
47.8
48.0
48.2
48.3
48.5
FLASK 1
NO
NO
M.5
12.5
--
--
11.2
--
--
--
3.2
1.6
--
2.4
--
FLASK 2
Yes
No
__
17.4
14.1
13.6
--
3.6
2.7
2.2
--
2.6
FLASK 3
No
Yes
_ _
--
10.4
--
6.2
6.5
2.8
--
--
--
--
--
2.6
FLASK 4
Yes
Yes
__
12.2
--
--
--
6.8
--
--
--
5.1
--
--
--
2.1
--
--
--
--
--
1.6
FLASK 5
No
No
__
--
--
--
--
5.7
--
--
5.6
--
--
--
6.9
48.6
5.3
39
-------
Treatment Tests - Tetraethyllead (TEL)
Chemical treatment tests were conducted on a sample containing 12.8 mg/1
TEL. Southern bentonite clay was added to two portions of the sample to
simulate dirt that might be suspended in water at a spilt site.
Concentrations of 500 mg/1 and 1000 mg/1 clay were used. Each sample
was then treated with 50 mg/1 ferric chloride (as FeCl^), flocculated for
two minutes and allowed to settle five minutes. Samples of the clarified
water were taken for tetraethyllead analyses. The clarified water was
then passed through a one Inch deep sand filter to remove the small floe
particles remaining in suspension. Samples were taken of the filtrate
for tetraethyllead analyses. The results of these tests are listed in
Table 11. From these results, it appeared that flocculation and
filtration would adequately remove tetraethyllead. However, it should
be noted that the chemical treatment test results may have been influenced
by the natural degradation of TEL with time, as indicated in earlier
solubility tests. The increase in water soluble lead between settling
and filtration tests would seem to further endorse this idea. However,
the bench scale test work does indicate that TEL can be controlled by
chemical clarification. The residual soluble lead may then be removed
by conventional means of raising the solution pH and precipitating out
the lead as hydroxide.
Table 11. RESULTS OF CHEMICAL TREATMENT TESTS ON TETRAETHYLLEAD (TEL)
Initial TEL concentration, mg/1
Concentration of clay, mg/1
FeClj dosage, mg/1
Mix time, min.
Flocculation time, min.
Settling time, min.
TEL concentration, mg/1
After settling
After sand filtration
Ionic Lead concentration, mg/1
After settling
After sand filtration
Test 1
12.80
500
50
0.50
2
5
0.10
<0.05
0.80
1.00
Test 2
12.80
1000
50
0.50
2
5
<0.05
<0.05
0.70
1.50
-------
Activated Carbon Adsorption - One attempt was made to run a carbon
Isotherm test on suspended TEL. The carbon isotherm test procedure was
modified by preparing a dilute carbon slurry and measuring the desired
amount of carbon slurry by pi pet. The amounts of carbon used were 0 mg,
0.75 mg, 1.5 mg, 2.5 mg and 5 mg. A mixture of TEL and water was shaken
and allowed to settle for one hour. The supernatant was used for the
isotherm tests. The initial TEL concentration was 5 mg/1. After shaking
one hour with 0 mg carbon, the TEL content was reduced to 0.5 mg/1. The
TEL content of all the samples that were mixed with carbon was less
than 0.1 mg/1.
As a result of the above described tests, it appears that only low
levels of TEL may persist in waters at a spill site and that these
amounts can be reduced to safe levels by chemical flocculation, settling
and filtrations. The longer the length of time the TEL contaminated water
is impounded, the lower the concentration of TEL is likely to be. The
amount of ionic lead, however, is expected to increase. The soluble lead
can be satisfactorily treated by conventional pH adjustment for the
precipitation of lead hydroxide and removal of the precipitate by
sedimentation and filtration.
Treatment Tests - Tetramethyllead (TML)
The results of chemical flocculation, settling and sand filtration tests
on several TML-water mixtures are listed in Table 12. The TML-water
mixtures used varied from 22 to 33 mg/1 TML. A one hour to 2i hour
settling time was used to allow large droplets to settle after the
initial mixing of the TML. Southern bentonlte clay was added to the
TML-water mixture in concentrations of 100,500 and 1,0000 mg/1 to
simulate dirt that might be suspended in the water to be treated. Each
sample was then treated with 50 mg/1 ferric chloride, mixed, flocculated
for two minutes and allowed to settle ten minutes. Samples of the
supernatant were decanted and passed through a one inch deep sand
filter to remove suspended solids. Portions of the supernatant were
analyzed for TML before and after sand filtration. Appreciable amounts
of TML remained after chemical treatment and filtration.
41
-------
Table 12. CHEMICAL FLOCCULATION, SETTLING AND FILTRATION
OF TETRAMETHYLLEAD WATER
Test No.
Elapsed time after mixing, hrs
Clay added, mg/1
Ferric Chloride added, mg/1
Flocculation time, min
Settl ing time, min
TML concentration, mg/1
initial concentration
after settl Ing
after filtration
Toluene concentration, mg/1
initial c-ncentrat ton
after settl ing
after filtration
1
1
100
50
2
10
33
19.2
12.3
142
106
84
2
2.5
500
50
2
101
22
14.1
13.6
114
90
82
3
0.75
1 ,000
50
2
10
32.5
21.3
16.2
101
88
Tests were also conducted with potassium permanganate to evaluate its
effect on the removal of TML. The results of these tests are shown in
Table 13. The potassium permanganate dosages utilized during these
tests were 50, 100 and 200 mg/1. Also, separate samples of TML-water
mixture were kept exposed to light and air without the addition of
KMnOit to compare the removal efficiencies with and without the addition
of potassium permanganate. It was found that for all the three samples
treated with 200 mg/1 KMnOj,, the TML content of the treated water was
reduced to less than 5 mg/1 within 10 minutes while for the untreated
samples, the TML content was between 7 and 14 mg/1 after standing for
one hour. The TML concentrations of the samples treated with 50 and
100 mg/1 potassium permanganate decreased slowly. The effluent TML
concentration for the 200 mg/1 potassium permanganate treated sample
did not change at all after the initial 10 minute reduction. It was
evident that the potassium permanganate treatment was effective in
handling TML solutions and can be utilized for the treatment of such
spills.
-------
Table 13. TREATMENT OF TETRAMETHYLLEAD
WITH POTASSIUM PERMANGANATE
Test No.
KMnOj. added, mg/1
1
50
2
100
3
200
TML concentrations, mg/1
immediately before KMnOj, 15-2 22.0 20.8
within 10 min after KMnO/,
addition 4.6 1.7 3.7
KMnO/t treated sample after 2.1 1.2 3.7
one hour
untreated portion of sample
after one hour 7.0 9-3 14.4
Activated Carbon Adsorption - Two carbon isotherm tests were performed
on a TML-water mixture.Procedure used was similar to the TEL isotherm
test. The results are listed in Table 14. Substantial reductions in
TML concentrations were observed in the flasks containing no carbon.
This may have been due to the volatilization or degradation of the TML.
In the first test, the TML concentratton was 20 mg/1 when the mixture
was added to the isotherm test flasks. The TML concentration dropped to
1.8 mg/1 after one hour of shaking in the flask containing no carbon.
In the second test, the TML concentration after shaking one hour without
carbon was 3*8 mg/1. A sample of the solution added to the flasks in
the second test contained 12.3 mg/1 TML approximately 2 hours after the
isotherm test was started. Although this sample was not shaken, it was
exposed to light and air during storage and it is probable that some
TML was lost during this period. In each test, the concentration of TML
in the flasks containing carbon was reduced to 1.5 to 2.1 mg/1. The
amount of carbon used had little apparent effect on the residual TML
concentration. The carbon did effect some TML reduction, though it was
very smal1.
From the above data, it is evident that conventional coagulation and
suspended solids removal treatment probably would not be effective for
satisfactory removals of residual TML concentrations at a spill site.
However, a combination of potassium permanganate treatment and
activated carbon may provide more satisfactory treatment.
-------
o
<
LU
_1
i
>-
^
|
LU
1
|
LU
1-
CC
O
ll_
I/I
t
|
Z3
co
LU
o:
I-
i/>
1 1 1
H-
X
a:
LU
^
h-
o
I/)
^
O
CD
0£
<
O
-a-
0)
ja
(D
1-
CA
O
CM CA
-a-
LA
CM -a-
CA
o
N CA
-a-
LA
CM -3-
CM CA
J2
CM O
CA
O
CA
-a-
LA
-a-
CA
LA
-a-
m
0
_
\
g
»
c
O
4J
C
4)
o
o
o
O
z. c
o
^< J\
in «-
4>
4-1
ID
u
4-1
o
o
s
0
n
o
i/i
£
CA
CO
*
O
CA
CM
O
CM
CM
CO
PA
LA
vO
~
IA
vO
*
vO
LA
co
*
_i
s:
o
-3-
O
-3-
vD PA
J- -3-
1 LA
-»
CA LA
vO -3-
CM vO
r^ -a-
PA
1
LA 1
CM
vO
|
O 1
PA
LA
1
O 1
-3-
CM
|
PA 1
-3-
CO
1
CM 1
-a-
^^
jQ
O.
V)
4) <0
C '
4)
3 13
ID
O 4)
t- -1
O
_Q
l_
ID
U
4-1
3
Adsorbed/
4-1
S
CM
O
-3-
O
CM
CO
PA
CA
co
i
i
CM
O
PA
*
O
CM
CM
r--
SO
1
1
^-^
\£>
1
O
^
_l
s
co
r>>
co
fA
LA
co
CM
1
1
PA
O
1
vO
-3-
CM
r^
r^
PA
cA
r-
o
o
1
1
so
1
O
^
Toluene
V.
01
PA
n
4)
4)
>x 3
1^0
4J
f^*
r^ PA
-3-
CM
H -*
4) N
4) -1
3 S
1
0
4J
* *J
~ t/1
V^ fl)
o> *-
o I!
CM 4)
II *<
O
1 lA
«J In
^T
*E. *"
M-
O
**
*J *J
IA «-
4) "3
4J> 4-1
|A
wl
^ ^
4) 4)
JZ 4-1
4J U-
O ID
I/I
t/»
u
<4- 3
° °
I- CM
ID
XJ ' *
** *
M 3
xj 5
*< uj
(0
-------
REVERSE OSMOSIS TESTS
The hazardous materials listed earlier in Table 1 were screened for
reverse osmosis application feasibility based on available literature.
Of the 20 materials listed in this table, data indicated that ten could
not be treated suitably with reverse osmosis and no such treatment tests
were conducted for these materials. These materials were: benzene,
chlorine, chlorinated hydrocarbon pesticides, chlorosulfonic acid, organ!
lead compounds, phosphorus penta sulfide and styrene. The reasons for
elimination of these materials from consideration are shown in Table 15.
Generally, the reasons pertained to low solubility of the hazardous materials
or expected adverse effects on the RO membranes. Among the remaining ten
of the listed materials, it was indicated that sufficient application
feasibility data was available from the RO membrane manufacturer for two
materials and therefore, bench scale feasibility tests were conducted on
only the remaining eight compounds.
Table 15. LIST OF HAZARDOUS MATERIALS ELIMINATED FOR
BENCH SCALE RO TESTS BECAUSE OF THEIR PROPERTIES
AND EXPECTED ADVERSE EFFECT ON MEMBRANE
Material Reason for elimination from feasibility tesfs"
1. Benzene A solvent is expected to have adverse effects
on the B-9 membrane. Also limited solu-
bility in water.
2. Chlorine Extremely low tolerance for B-9 membranes
(<0.25 mg/1). Expected tolerance for
cellulose acetate membranes up to <5 mg/1.
3. Chlorinated hydrocarbons Low solubility in water.
k. Chlorosulfonic acid Decomposes violently in contact with water.
5. Lead as in TEL & TML Unstable compounds with low solubility in water.
6. Phosphorus pentasulfide Decomposes with water to H,POj, and ^S.
Hydrogen sulfide is expected to adversely
affect the membranes.
7. Styrene Low solubility in water.
The materials for which feasibility tests were conducted were: acetone,
acetone cyanohydrine, acrylonitr1le, ammonium salts, copper sulfate,
formaldehyde, metHanoi and mercuric chloride. A summary of the results
of the feasibility data is shown In Table 16. A description of the RO
test procedures and terminology is presented in Appendix B. No specific
change in rejection capabilities was noticed for any of the compounds
tested because of a change in feed concentration. Generally, the inorganic
compounds showed good rejection capabilities except mercuric chloride.
The rejection of mercuric ion at 12% was found to be lower than expected
since generally most divalent and trivalent ions can be rejected in
-------
to
1-
co
UJ
1-
>-
h-
_l
CO
CO
LU
U.
o
oc.
L.
O
to
1-
_l
ZJ
to
LU
^_
cu
XI
TO
ation, Molecular % Rejection B
1 pH Weight % Recovery B~9 Membrane
i_ ^^
4-> en
c E
cu
o
c
o
o
en o LA
NO f*^ LA
LA t^ LA
\O LA LA
OO LA
LA OO
on NO on
on NO cy\
r-^ CM
t
r~~
c
^_
T3
x:
o
c
TO
,
TO
L.
cu
4J
fO
2:
(_j
cu cu
c c
O O
4-> 4-1
cu cu
0 U
-a-
en
on
LA
oo
,
cu
._
1_
4-1
._
C
O
>>
L.
U
<
oo on
oo oo
CM
NO
0
CO
LA
NO
o
4->
on
-a- CM
LA en
d)
4-J jZH Z
TO
U I/I (SI
4-J TO TO
...
z -a- on
a: o
E z z
3
.
C
Q
£j
E
oo oo
LA
LA
CM
NO
LA
O O
CM NO
on o
M *
^~ r^
cu
4-1 Z
TO
<4- in
TO
3
CO LA -3"
3: o
E Z co
._
C
E
NO NO
OO O
oo oo
en
LA
O LA
LT\-3-
0 0
-a- o
-a- oo
t **
^ 0^
CU
4-1
TO
* 4*
3 3
CO 0
U
cu
CL
Q.
O
o
0
CM
LA
O
-a-
0
i
»>
oo
cu
o
x:
cu
T5
r-.
TO
E
u
O
u.
on
-a-
CM
on
LA
r>«
o
o
o
V
o
^_
»
o
c
TO
x:
4-1
CU
z:
CM
LA
CM
CM
0
4-J
-a-
0
oo
-a-
\
CM
cu
TJ
1_
o
xl +
0 +
en
o 3:
t_
^
0
i_
CU
^.
OO LA
CT> LA
LA LTV
^"^ r^»
v«O ^f
NO on
o
on
-a- o
o
r-^
0
o
LA O
-~ M
CM
O
4->
O
o
LA
Jf
+
+
r
<
X)
*
TO O
E c
3 CU
x:
< o-
LA
CT>
LA
r^.
o
o
o
LA
*.
CM
XI
CU
4-1
TO
^~
>^
C
0)
x:
0.
E
3
-a
o
to
o
CM
CU
O
C
cu
L.
0)
M-
0)
O£
TO
pont Company.
3
a
>*.
XI
^o
0
Q.
Q.
3
CO
TO
4-*
TO
O
XI
ts conducted at ^00 psi and at ambient temperature.
(/)
cu
4-1
^
L.
O
4-1
TO
O
Xl
TO
^~-
^
<
LU
1-
o
z
CM
LA
oo
O CM
-------
excess of 90%. Also during the brief test duration (2k hours) for
mercuric chloride, it was noticed that the headless across the membrane
increased rapidly, suggesting high membrane fouling potential. Therefore,
it was concluded that RO would not be suitable for the treatment of
mercuric chloride.
Rejections for ammonium ion varied between 88% and 98% depending upon
the type of anion associated with the ammonium ion. The higher rejection
for ammonium ion when associated with sulfate ion was expected because
of the generally better rejection characteristics of RO membranes in
the presence of a divalent ion. Copper and aluminum ions were rejected
in excess of 95% as expected.
The various organic compounds evaluated in this study showed considerably
lower membrane rejection capabilities compared to inorganic materials.
Methanol, formaldehyde and acrylonitrtle showed poor rejection
capabilities (13-201) while acetone, acetone cyanohydrin and phenol
showed medium separation capabilities (55 to 70%). All of the above
mentioned organic compounds have a molecular weight of less than 100.
Such organic compounds are generally expected to exhibit lower membrane
rejection characteristics. However, it was found that the change in pH
for some of these compounds can provide a marked improvement in the
rejection capabilities. The pH values shown in Table 16 are considered
to be the optimum values for membrane rejection. The effect of pH
change was most pronounced for phenol where the conversion of phenol to
sodium phenylate enhanced the rejection capability from 55% to 95% at
comparable operating conditions.
From the above data it is obvious that reverse osmosis has limited
applicability for the treatment of only a selected few hazardous
materials such as soluble inorganics and high molecular weight organics
(mol. wt. >100). Out of the 20 hazardous materials screened only eight
showed medium to good removal effectiveness. Many of these spilled
hazardous materials will require extensive pretreatment for the removal
of particulate solids and/or other treatment such as pH adjustment,
softening etc. prior to processing by RO. Furthermore, the reverse
osmosis process being a high pressure process, requires relatively
higher amounts of power. Both the additional pretreatment and power
requirements may significantly influence the logistics support require-
ments in a spill situation. Therefore, it was concluded that reverse
osmosis has only limited application feasibility for the treatment of
hazardous materials.
-------
SECTION VI
DESIGN CRITERIA FOR A MOBILE TREATMENT SYSTEM
Based on the laboratory evaluation of the various hazardous materials
described in the preceding sections of this report, it was concluded
that a flow schematic consisting of chemical reaction, flocculation,
sedimentation, filtration and activated carbon adsorption would provide
the most flexible mobile spills treatment system. These unit treatment
processes were shown to remove the majority of the pollutants evaluated
in this study. It was concluded that RO would not be a suitable treatment
method for many spilled hazardous materials because of their low water
solubility and adverse effect on membranes. RO could however, be
utilized on a selective basis such as for the treatment of water soluble
inorganics. For such selective uses, reverse osmosis equipment may be
mounted on a separate mobile unit that could then be utilized in conjunc-
tion with another pretreatment unit capable of providing chemical and
filtration treatments.
Figure 15 shows a schematic flow diagram of the recommended treatment
system that was utilized for the design of the mobile spill response
vehicle constructed under EPA Contract No. 68-01-0099. The raw waste
pumping system provided on the spill response vehicle, consists of a
submersible pump, an in-line booster pump and sufficient hoses to allow
the deployment of the treatment system as far as 92 m (300 ft) from the
spill site. The raw flow can be controlled manually by an in-line
indicating flow meter and control valve. The maximum hydraulic
design flow capacity of the pumping system Is 12.6 I/sec (200 gpm). The
vehicle is equipped with two other pumps (filter influent and backwash
pumps). Any of the pumps provided on the vehicle can be used inter-
changeably during emergencies. The electrical power to operate the raw
feed system as well as all of the other electrical requirements for the
treatment system are provided by a gasoline fueled generator on the
trailer. Various chemicals such as acid, lime, ferric chloride, chlorine
or potassium permanganate (depending upon what substance is being treated)
can be added in the reactton/flocculation tank to precipitate and/or
flocculate the waste. An in-line pH indicator follows an in-line mixer and
can be used for the adjustment of chemical addition for controlling pM.
Provisions are also made for addition of a polymer to the suction side
of the filter pump as a filtration conditioner. The waste then flows
to a settling tank. The overflow from the settling tank is filtered
through granular media filters followed by granular activated carbon
adsorption. Sludge is removed from the sedimentation tank and stored
-------
«
u
M
i
M
V
Q.
M
-O
i
u
O
u
O)
O
i
M
3
tn
u
3
o>
-------
for ultimate disposal. An additional storage tank is provided for
filter backwash ing operations. Each unit process is capable of being
bypassed if not required for a specific spill.
A portable treatment system must also meet the various limitations imposed
upon transportable equipment. The most significant limitation involved
in the design of this mobile treatment system were size and weight.
A 13*7 m (45 ft) trailer in conjunction with a snub nose tractor met the
required 16.8 m (55 ft) total vehicle length imposed by many states.
Because a significant amount of time is required for wetting activated
carbon (12-48 hrs), the carbon should be transported wet and this increases
the weight of the trailer considerably, since 0.45 kg (1 Ib) of dry carbon
absorbs about 0.45 kg of water. The total treatment vehicle over-the-
road weight selected for this system was 47,600 kg (105*000 Ibs), which
requires special operational permits in most states. Generally, 33,200 to
36,400 kgs (73,000-80,000 Ibs) gross weight is the maximum allowable in
most states under normal circumstances. However, higher weights are
permissible for emergency vehicles. The maximum concentration of hazardous
material in the waste was set at 1% for design purposes. However, it was
recognized that concentrations would in all likelihood be much lower.
The ambient temperature range was defined at 0-100°.
Based on the severe corrosion anticipated for the wide variety of
hazardous materials under study, 316 stainless steel was selected as the
optimum material based on strength, price, ruggedness and durability.
Having established the process flow sheet, the equipment had to be
designed to provide maximum possible hydraulic capacity within the space
and weight constraints of over-the-road equipment mentioned previously.
The importance of high flow rate in treating a dilute spill can be shown
by the following example: a 19,000 1 (5,000 gal.) chemical spill which
empties into a stream and is diluted to 1000 parts per million creates
19 million liters (5 million gal.) of contaminated water. A treatment
system operating at 12.6 I/sec (200 gpm) on an around the clock basis
will require 17 days to clean the water provided that the clean
effluent water is not reintroduced into the spill contaminated area.
Because of the various physical constraints, it was concluded that a
maximum hydraulic capacity of 12.6 I/sec (200 gpm) could be utilized for
the mobile treatment system. However, flexibility was provided in the
selection of the pumping system so as to operate the treatment system at
lower or higher flow rates when desirable.
Flocculation/Sedimentation System
Large rubber, portable tanks are set up next to the trailer with one
tank inside the other to provide the necessary detention times for
flocculation and sedimentation. A minimum chemical reaction/flocculation
time of 15 minutes and sedimentation time of 60 minutes were considered
necessary for the design of the treatment system. This led to the
50
-------
selection of a 11,350 liter (3000 gal.) reaction/flocculation and 56,775
liters (15,000 gal.) sedimentation tank. Larger contact periods could
be provided at reduced flow rates. In order to conserve valuable space
on a mobile treatment system, specially built collapsible rubber tanks
were selected in place of conventional rigid tanks. A flocculation/
sedimentation system was developed whereby the two 11,350 liters and
56,775 liters tanks could be utilized concentrically. This arrangement
allowed a flocculation time of 15 minutes and a detention time of 60
minutes at a flow rate of 12.6 I/sec (200 gpm). The flocculation tank
is placed at the center of the sedimentation tank. The raw waste is
introduced at the bottom of the flocculation tank. Flocculation is
achieved by hydraulic mixing via the use of two ejectors placed opposite
to each other. The flocculated wastewater flows out of the center into
reaction tank into the 45,^00 liter (12,000 gal.) annular space of the
sedimentation tank through a series of submerged orifices located around
the periphery of the reaction tank. Both these tanks are cylindrical open
top tanks are are supported by staves anchored into the ground. The open
top tanks also permit easy accessibility for manual removal of sludge
and floating materials. A photograph of the completely assembled
flocculation/sedimentation tanks is shown in Figure 16.
A 11.9 I/sec (50 gpm) sludge pump and special suction fittings are
provided for removal of settled and floating contaminants and a 11,350
liter (3000 gal.) rubber stave tank is provided for storage. The sludge
can subsequently be pumped out of the stave tank by the sludge pump into
a tank truck or other container for further treatment and/or final
disposal.
Dual Media Filters
The supernatant from the sedimentation tank is drawn off by the filter
pump through a submerged orifice header ring at the outside tank wall.
A pneumatic level sensor in the sedimentation tank controls filter pump
flow to match raw flow.
After the addition of a filtration conditioner, the sedimented effluent
is pumped through three dual media filters in parallel for removal of
residual suspended solids by the sand-anthracite filter media. Two
in-line turbidimeters monitor the turbidity of the total filter influent
and effluent from each tank. A differential pressure gauge indicates the
degree of filter clogging. The filters may be taken off line individually
and backwashed with air and clean system effluent stored In a 11,350
liter pillow tank. Pictures of the filter pump, filter tanks and plumbing
arrangement are shown in Figures 17 and 18. The tanks are located just
behind the filter and backwash pumps in the center of the trailer. The
filters consist of 61 cm (2k in.) anthracite (coal, 0.9 mm effective
size) over kS cm (18 in.) fine sand (0.45 mm effective size). The
51
-------
Figure 16. Completely assembled
flocculation/sedimentation tank in field
52
-------
Figure 17» Photograph of the mounted
dual media filters
53
-------
Figure 18. Closeup of the filter
plumbing arrangements
-------
filters are designed for a maximum hydraulic loading of 4.8 1/sec/m
(7.0 gpm/ft2). Additional pertinent filter data for each of the three
filters is given below:
Filter diameter 1.17 m (3-5 ft)
Filter area 0.893 m3 (9.62 ft2)
Design filtration rate 2.11 I/sec (33-5 gpm)
Maximum filtration rate 4.22 I/sec (67 gpm)
Design backwash rate 6.3 I/sec (100 gpm)
Maximum allowable differential -
pressure 1.06 kg/m (15 psi)
2
Maximum tank pressure 4.93 kg cm/cm (70 psi)
Depth of sand 45 cm (18 in.)
Effective size 0.5 mm
Uniformity coefficient 1.5
Quantity of sand 860 kg (1900 Ibs)
Depth of coal 61 cm (24 in.)
Effective size 0.85 - 0.95
Uniformity coefficient 1.7
Quantity of coal 0.425 m (15 ft3) 0.39 ton
The dual media portion of the trailer is a completely assembled system
ready for operation. A backwash system Is provided to flush away the
captured solids and thus the filter is continuously regenerated without
the need for media replacement. The backwash waste is pumped to a
11,350 liter (3*000 gal.) stave tank. From here It can be pumped out
by the sludge pump either into a tank truck for disposal or it may be
reintroduced into the raw feed line and recycled through the system.
The air-backwash system is arranged so that two filters may be left
on line while one is taken off line for backflushing. The flow through
the system will have to be decreased during backwash!ng if reducing
from three to two filters on-line causes a differentia) pressure rise.
The amount of throttlIng-back will depend on how dirty the filters are.
It is recommended to backflush before the filters get too clogged for
the reason that when the clean filter is brought back on line, it will
take a high percentage of the flow at an increased filtration rate.
This condition could lead to unusually high effluent turbidity. One
way to cause balanced filter flows with a new filter on line is to
throttle the filter inlet valve somewhat to cause an "artificial"
headloss across the filter which will tend to balance the flows.
55
-------
Carbon Columns
The filtered effluent flows through three pressure carbon columns
which may be used in parallel or in series. Altogether, they contain
19.6 m^ (700 ft*) of carbon. This volume represents a dry carbon weight
of 8172 kg (18,000 Ibs) of carbon which is the maximum possible weight
that can be accommodated on the trailer because of the overall weight
constraints on the mobile system. The carbon columns are designed for
a hydraulic loading rate of 3.4 1/sec/m2 (5 gpm/ft2). Three 2.1 m
(7 ft) diameter carbon columns with carbon bed depth of 1.8 m (6 ft) are
provided on the mobile treatment system. The selected carbon volume of
19.6 cu m (700 cu ft) on the trailer provides a maximum contact time of
27 minutes for the three columns at a flow rate of 12.6 I/sec (200 gpm).
This carbon contact time was found to be suitable for many of the
hazardous materials evaluated in this study. However, when high contact
times are required, these may be provided by reducing the hydraulic flow
rates through the carbon columns proportionately.
The carbon tanks occupy the back one-half of the trailer as shown in
the photograph of the overall system, Figure 19. The carbon columns
are completely plumbed so that only valve adjustments are necessary to
control the various modes of operation. The trailer is designed so that
it may be transported with the carbon columns full of the wet, drained
carbon (the carbon holds 0.45 kg (1 Ib) for 0.45 kg of water in the drained
condition). However, whether done on the site or back at the homebase,
the carbon will need to be recharged either because of exhaustion, to
prevent undesirable effects of mixing contaminants, or because storage
with contaminants in the carbon could be hazardous. Spent carbon may be
removed using the backwash pump and processed clean water stored in one
of the rubber tanks for this purpose. The clear effluent water Is pumped
through the underdrain system of the carbon column to fluidize the bed
and cause the slurry to drain out of the tank drain fitting.
When the adsorptive capacity of the carbon for the processed hazardous
material is depleted, new carbon may be installed into the tanks in the
field by a slurry pumping system. Depletion time, or breakthrough,
occurs simultaneously for parallel operation and sequentially for series
operation. Thus all of the carbon must be replaced at once for parallel
operation as opposed to one tank at a time for series operation. The
tank plumbing permits a rotation of the flow sequence through the tanks
for series operation. Thus, the leading tank eventually becomes the
second tank in line and finally assumes the third position until
breakthrough and carbon replacement, whereon it again assumes the first
position. The slurry pumping system for carbon replacement utilizes a
dry carbon hopper feeding an eductor through which clean effluent from
the effluent storage tank is pumped to form a slurry. The slurry is
pumped into the carbon column where It is dewatered by the carbon column
underdrains. The water is then returned to the effluent storage tank
completing the closed loop slurry pumping system.
56
-------
Figure 19. Carbon column as mounted on the
mobile treatment system
57
-------
A manual sampling valve in the carbon column inlet line and in each tank
effluent line permits analysis of process water to measure removal
effectiveness and to sense carbon column breakthrough.
Treatment System Layout
System layout will be dictated in many cases by the space available and
terrain conditions at the spill site as well as the accessibility of
driving the trailer close to the site. A raw feed pumping system on the
trailer includes 92 m (300 ft) of hose to transport the contaminated
water to the treatment system. Hose lengths totaling 92 m are also
available for discharging the treatment system effluent into the
receiving stream, it is essential that a plan for the various required
locations for the raw feed submersible pump be devised before the
final treatment system location is selected. In that way a central
location can be selected which will allow pumping from most if not all
of the planned locations without moving the treatment system. Much time
and effort is lost through needless moving of the system.
The trailer location must also take into account the hazardous conditions
in the immediate vicinity of the spill site. These precautions should
include provisions for operator safety from contamination emitting
directly from the spill site and from the treatment system itself since
it functions as a reconcentrator of the spill and as such is a hazardous
location. In addition, the electrical generator is powered by a
gasoline fueled internal combustion engine. The separator should not
be operated in areas where there is danger from fires or explosions.
A recommended plan view layout of the treatment system is shown in
Figure 20. The required site is approximately 15 m x 30 m (50* x 100').
There are sufficient hose provisions on the trailer to accommodate the
layout shown. It is preferable that the entire site be level and free
from obstructions. In particular, the reaction/sedimentation tanks rely
on being level for proper function. In this regard it is recommended
that, if feasible, the selected treatment site be bulldozed so that the
tank can be located on very level ground, i.e. slope less than 5 cm in
8 m (2 In. in 25 ft). Problem in supporting the large tank occurs when
the slope exceeds 15 cm in 8 m (6 in. in 8 ft).
The pillow tanks should also be placed on smooth and level ground to
facilitate filling and emptying. The trailer does not have.to be on a
level surface to function properly, but it is preferred. Front-back
inclination can be corrected by the landing gear on the trailer. Side
to side inclination can be corrected by skimming under the wheels and
landing gear with planks. The ground level at the trailer should not
be more than ten feet above the ground level at the reaction-sedimentation
and effluent storage tanks because of suction limitations on the trailer-
mounted, self-priming pumps which draw from these tanks.
58
-------
§
o
o
10
"a.
4)
O
V
QC
V
L.
o>
OS)
w
59
-------
In assembling the system, care must be taken to allow a minimum of three
feet clearance all around the trailer for access to control valves and
sampling points. Ready access should be provided to the control platform
which is on the right forward portion of the trailer. When laying hard-
wall hoses in areas where there is vehicle traffic, either the hoses
should be buried or straddled with planks which are pinned to the ground.
This will prevent crushing of hoses and the resultant flow limitation.
The above described treatment system provides the necessary response
vehicle that can be activated for a wide variety of spill situations in
the shortest possible time. Operating experience with the vehicle and
predetermined treatment procedures for specific chemicals will insure
effective spill response and treatment.
Engineering Drawings, Specifications, and Operations and Maintenance
Manuals
As stated in the Introduction of this report, the detailed engineering
drawings and equipment procurement specifications for the fabrication of
the system, as well as a set of manuals containing complete operation,
repair and maintenance instructions for each component of the system,
can be made available from EPA's Industrial Environmental Research
Laboratory, Edison, New Jersey 08817.
60
-------
SECTION VII
REFERENCES
1. Wilder, I. and LaFornara, J., "Control of Hazardous Material Spills
in the Water Environment - An Overview", paper presented before the
Division of Water, Air and Waste Chemistry, American Chemical
Society, Washington, D.C., September 1971.
2. Dawson, G. W., Shuckrow, A. J., and Swift, W. H., "Control of
Spillage of Hazardous Pollution Substances", Battelle Memorial
Institute, FWQA Report 15090 F02 10/70 (Contract 14-12-866),
November I, 1970.
3. Fornwalt, H. J. and Hutchins, R. A., "Purifying Liquid with Activated
Carbon", Chemical Engineering, April 11 and May 9, 1966.
4. Standard Methods for the Examination of Water and Wastewater, 13th
Edition, American Public Health Association, Inc., New York, N.Y.
1971.
5. Methods for Chemical Analysis of Water and Wastes, Environmental
Protection Agency, Analytical Quality Control Laboratory, Cincinnati,
Ohio, 1971.
6. Analytical Methods for Atomic Absorption Spectrophotometry,
Perkin-Elmer Corp., 1971.
7. Barnes, R. A., Atkins, P. F., Jr., and Scherger, D. A., "Ammonic
Removal on a Physical-Chemical Wastewater Treatment Process",
EPA Report No. EPA-RZ-72-123, prepared for Office of Research and
Monitoring, USEPA, Washington, D.C., 1972.
8. King, P. H., Yeh, H. H., Warren, P. S., and Randall, C. W.,
"Distribution of Pesticides in Surface Waters", Journal AWWA,
Vol. 61, No. 483, September 1969.
9. Lotse, E. G., Graetz, D. A., Chesters, G., Les, G. B., and Newland,
L. W., "Llndane Adsorption by Lake Sediments", Environmental
Science Technology, Vol. 2, No. 353, May 1968.
61
-------
10. Schwart, H. G., "Adsorption of Selected Pesticides on Activated
Carbon and Mineral Surfaces", Environmental Science and Technology,
Vol. 2, No. 353. May 1968.
11. Huang, J. C. and Llao, C. S., "Adsorption of Pesticides by Clay
Minerals", Journal Sanitary Engineering Division, Proceedings of
the Chemical Society of Civil Engineers, 96:SA5, 1957 (Oct. 1970).
12. Morris, J. C. and Weber, W. J., "Adsorption of Biochemically
Resistant Materials from Solution, 2", USPHS Environmental Health
Series 999-WP-33, U.S. Government Printing Office, Washington,
D.C., 1966.
13. Aly, 0. H. and Faust, S. D., "Studies on the Removal of 2,4-D and
2,4-DCP from Surface Waters", Proceedings of the 18th Industrial
Waste Conference, Purdue University, p. 6, 1963.
14. Robeck, G. G., Postal, K. A., Cohen, J. M., and Kreissl, J. F.,
"Effectiveness of Water Treatment Processes in Pesticide
Removal", Journal AWWA, Vol. 57, No. 181, February 1965.
15. Hartung, R. and Klinger, G. W., "Concentration of DOT by Sedlmented
Polluting Oils", Environmental Science and Technology, Vol. 4,
No. 407, May 1970.
16. Babers, F. J., Journal, American Chemical Society, Vol. 44, 46666,
1955.
17. Bowman, M. C., Acree, F., Jr., and Corbett, M. K., Journal ,
Afr. Food Chem., Vol. 8, 406, I960.
18. Chamberlain, N. S., and Griffin, A. E., "Chemical Oxidation of
Phenolic Wastes with Chlorine", Sewage and Industrial Wastes",
24:750, 1962.
19. Shapiro, H. and Frey, F. W., "The Organic Compounds of Lead",
Inter-Science Publisher, New York, N.Y., 1968.
20. Mason, D. G., Gupta, M. K., and Wilmoth, R. C., "Treatment of
Ferrous Iron Acid Mine Drainage by Reverse Osmosis", paper
presented at the Fourth Symposium on Coal Mine Drainage Research,
Pittsburgh, Pennsylvania, 1972.
62
-------
SECTION VIM
APPENDIX A
ANALYTICAL APPARATUS AND PROCEDURES
63
-------
ANALYTICAL INSTRUMENTS AND APPARATUS
pH meter: Beckman Model 552
B»ckman Instruments Incorporated
Fullerton, California
Analytical Balance: Type H10
Mettler Instruments Corporation
Heightstorm, New Jersey
Spectrophotcmeter: Coleman Model 14
Coleman Instrument Company
Maywood, I 11i no i s
Atomic Absorpt ion
Spectrophotometer: Perkin Elmer Model 403
Perkin Elmer Corp.
Norwalk, Connecticut
TOC Analyzer:
Turbid imeter:
Beckman 915 TOC Analyzer
Beckman Instruments Co.
Fullerton, California
Hach Model 2100A Turbidimater
Hach Chemical Co.
Ames, Iowa
Gas Chromatograph: Barber-Col man Series 500C
Barber-Col man Co.
Rockford, I 11inois
61*
-------
ANALYTICAL PROCEDURES
The following analyses were performed according to Standard Methods for
the Examination of Water and Wastewater, (2) (Standard Methods) and
Methods for the Chemical Analysis of Water and Waste (16) (WQO).
CHLORINE - P. 110 Standard Methods; iodometric titration for total
chlorine.
pH - P. 276, Method A, Standard Methods, pH meter measurement using a
Beckman Mode) SS2 pH meter.
Sulfate - P. 286, WQO; Turbidimetric method using Hach Sulfaver.
TOC - P. 221, WQO; Beckman Model 915 TOC analyzer.
Chloride - P. 29 WQO; Mercuric nitrate titration using Hach chemicals.
Ammonia Nitrogen - P. 226 Standard Methods; Direct nesslerization using
Hach colormeter.
Sulfate - P. 286 WQO; Turbidimetric method using Hach Sulfaver.
The following analyses were performed according to the methods listed in
Analytical Methods for Atomic Absorption Spectrophotometry, Perkin
Elmer Corp. (17), equivalent to the mentioned WQO method. The digestion
procedure described on pages 86-89 of Methods for Chemical Analysis of
Water and Waste was used for all metals (16). Al 1 metals were
measured by atomic absorption spectroscopy.
Copper - P. 106, WQO
Lead - P. 110 WQO
Mercury - Perkin Elmer Corp.
The following analyses were performed according ot the procedures listed:
Phenol - Hach Catalog No. 10, P. 52. Chemical removal of interference
followed by treatment with 4 - aminoantipyrine for color development.
Color development part is based on the following procedures:
Standard Methods, P. 504 and 507, or WQO, P. 232.
Nitrate Nitrogen - Hach modification of the cadmium reduction method
(Standard Methods, 13th Ed., P. ^58). This method measures nitrate and
nitrite nitrogen. Nitrite N must also be measured and subtracted to
obtain Nitrate N.
65
-------
Cyanide - P. 23 Hach Catalog No. 10; A rapid method which eliminates the
need for distillation and pretreatment. Cyanide is measured
colorimetrically.
Nitrite Nitrogen - P. 42 Hach Catalog No. 10; Hach modification of
the diazotization method using pre-packaged chemicals.
Acetone - Column - 0.6 cm x 1.8 m (i" x 6') stainless steel column
containing Poropak Q.
Column Temperature - 200°C
Carrier Gas - He)ium
Detector - flame ionizatlon detector at 235 C.
Injector - 215°C
Quant I tat Ion by peak height using standards of acetone in water.
Acetone Cyanohydrin - Column 0.6 cm x 1.8 m stainless steel column
Poropak Q.
Column Temperature - 1?5°C and 180°C
Carrier Gas - He)ium
Detector - flame ionization detector at 200°C
Injector temperature - 180°C
Quant I tat ions using peak area determined by
disc integrator
Acrylonitrile - Column - 0.6 cm x 1.8 m stainless steel column containing
Poropak Q.
Column temperature - 160°C and 170 C
Carrier Gas - Helium
Detector - flame ionization detector at 170 C and 200°C
Injector - 170° - 200°C
Quantitations using peak area determined by disc
Integrator.
Method - Column - 0.6 cm x 1.8 m stainless steel column containing
Poropak Q.
Column temperature - 140 C
Carrier Gas - Helium
Detector - flame ionization detector at 170 C
Injector - U5°C
Quantitation - using peak area determined by disc
integrator.
66
-------
DDT - Column - 0.6 cm x 1.8 m (i" x 6') glass column containing
DC-200 on ABS.
Column temperature - 210°C.
Carrier Gas - Argon
Detector - Electron capture (Nicked-63) detector
operated at 20V and 2?0°C.
Injector - 200°C.
Quant I tat ion using peak area determined by disc
integrator.
Extraction Procedure for DDT
Extraction of the pesticide was performed with three 15 ml portions of
pesticide grade hexane (Burdick & Jackson). Extractions were performed
on a 200 ml sample aliquot using a separatory funnel. The extract plus
the hexane used for rinsing was dried over anhydrous sodium sulfate and
reduced in volume using a warm water bath and a stream of air which had
been dried and filtered through activated carbon. The hexane extract
was generally reduced to a volume of about 1 ml.
The DDT in the extract was measured by gas chromatography using a Ni
electron capture detector operating at 20 volts and 280 C. A i" x 6'
coiled glass column containing 10% DC-200 on Anakrom ABS was used at
210 C. Injector temperature was 220 C. Argon was used as the carrier
gas at a flow rate of about 100 cc/min. Elution time for p,p'-DDT
varied from 16.5 to 19 minutes depending on the gas flow rate for the
series of runs. The calibration curve was linear to about 35 vg p,p'-DDT
and usable to 60 yg. Most of the samples analyzed contained less than
30 yg p,p'-DDT per injection.
TETRETHYL LEAD - Column - 0.6 cm x 1 .8 m stainless steel column
containing 20% Apiezon L on Chromsorb W (AW - DMCS)
Column temperature - 160 C.
Carrier gas - Helium
Detector - Hydrogen flame ionization at 185 C.
Injector - 170°C.
Samples were injected directly on the column.
Quantities of tetraethyl lead were determined from
peak height. Standards were prepared using pesticide
grade ethyl acetate as the solvent.
67
-------
TETRAMETHYLLEAD - Column - 0.6 cm x 1.8 m (*" x 6') stainless steel
column containing 20^ ApiezonoL on Chromosorb W (AW-DMCS)
Column temperature - 70 C and 80 C for samples.
Detector - Hydrogen flame detector at 180°C.
Injector - 150 C.
Samples were injected directly. Quant Station was by
peak area using a disc integrator. Standards were
prepared using pesticide grade 0 - Xylene as the
solvent.
68
-------
APPENDIX B
BENCH SCALE TEST PROCEDURES
-------
CARBON COLUMN TEST PROCEDURE
The test equipment is shown in Figure 21.
The lead column is utilized as a surge chamber for the positive dis-
placement piston type pump. The pump accurately meters the test
solution at the desired rate. The second column is filled with
Filtrasorb carbon (approximately 450 grams) prior to each test. The
third column is utilized as a siphon break to prevent loss of liquid
from the second column.
1. Prepare the solution to be tested.
2. Prepare carbon for testing by boiling, to remove entrapped
air in the carbon pores.
3. Load the carbon as a wet slurry. Keep the column agitated
with a long rod as it is being filled to prevent the forma-
tion of air pockets.
4. Start pumping test solution at predetermined rate
(5-10 gpm/sq ft).
5. Take periodic column inlet and effluent sampler for analysis.
6. Plot effluent pollutant concentration -vs- time or pumped
volume. This is the breakthrough curve.
7. Examine curve for time of break through, shape to determine
adsorption characteristics, adsorption capacity.
8. Adsorption capacity is calculated by shaded area shown below,
expressed as a fraction of the total area. This fraction times
the mass of pollutant pumped over the mass of carbon utilized
is the adsorption capacity.
c Co
O
§
C
O
n Break
ough Curve
Volume
End of Test
70
-------
SUR
COL
O '
GE
CARBON
COLUMN
UNDER
TEST
UMN
t
W
FEED
PUMP
'
T
/
t
i
1
i
i
_L
1
SIPHON
BREAK
DIFFERENTIAL
PRESSURE
MEASUREMENT
ALL COLUMNS 6' x I' ID PYREX GLASS
Figure 21. Laboratory carbon column test apparatus
71
-------
9. Vary the depth of the carbon bed to determine the effect of
different contact-time.
10. Vary the pump speed to determine the effect of hydraulic
loading (1/sec/m2).
PROCEDURE FOR CARBON ISOTHERM TEST
1. To each of six 250 ml erlenmeyer flasks, add one of the following
weights of pulverized filtrasorb 400 activated; carbon: 0 gm,
0.2 gm, 0.5 gm, 1.0 gm, 2.5 gm, 5.0 gm. (Experience may show
that other weights of carbon are preferable for the solution being
tested. One of the flasks, however, should always contain no
carbon.)
2. Add 100 ml of solution to each flask, stopper with a foil lined
cork and vigorously agitate on a Burel1 Shaker for one hour.
3. Immediately after shaking is completed, remove the carbon by
filtration through a 0.1*5 micron filter. If TOC or COD analyses
are to be performed on the filtrate, be sure to wash the filters
prior to use.
k. Perform the desired analysis on the filtrate. TOC or COD analyses
are usually done; although analysis for a specific compound may
also be done.
f* f*
5. Plot loo o- against log C.. where C is the weight (mg) of material
- n o
in the flask containing no carbon CM is the weight (mg) of material
remaining in solution in the flask containing M mg of carbon, and
M is the weight (mg) of carbon in the flask. (It is easiest to
plot CO~^M -vs- CM on log-log paper.) Draw a straight line
M
through the points and extrapolate the line to Cn; the value of
C -C
0 M at C, is an estimate of the amount (mg) of material that
CM
adsorbed by a unit weight (mg) of carbon being fed a solution
containing C mg/100 ml material. This is known as adsorption
capacity.
For more information, see "Purifying Liquids With Activated
Carbon" in April 11, 1966 Chemical Engineers by Fornwalt and
Hutchins(3).
72
-------
REVERSE OSMOSIS BENCH TESTS
A schematic of the RO test apparatus utilized is shown in Figure 22.
It consists of a feed inventory tank with a cooling coil to adjust
the temperature of the feed solution, a 10 micron cartridge filter, a
high pressure pump (Moyno), a half size B-9 membrane and the necessary
valves, controls and instrumentation.
1. Make desired solution in tap water in the feed water tank.
2. Adjust cold tap water flow in the cooling water line to get
temperature of feed solution between 55 F. and 65 F.
3. Turn on high pressure pump with back pressure regulator by-pass
open.
4. Bring pump to desired operating pressure with pressure
regulator.
5. Both the concentrate and permeate are recirculated to the feed
tank and recombined.
6. Record feed water temperature, inlet pressure, outlet pressure,
concentrate flowrate, feed flowrate, inlet pH, concentrate pH,
product flowrate, product pH, and sample volume. (Measure
flowrate by bucket and stopwatch.)
7. In order to simulate higher feed water recoveries than achiev-
able with the system, waste the permeate until the desired
recovery level is achieved. This has the effect of brine
recycling, or simulating the effect of already having passed
through one bank of RO membrances and then passing to the
next bank. For example, wasting half of the original feed
solution volume simulates having already recovered 50% of the
product water.
8. Let unit run for one to four hours. Periodically record the
data mentioned in Step 6.
9. Sample product water, feed water, and brine for chemical analysis,
10. Adjust the recorded flow data for temperature (68°F) and
pressure (400 psi) and calculate the percent recovery salt
rejection capability through the membrane in the following
manner:
% Recovery = Ratio of product flow to feed flow x 100
% Rejection = (1-Ratio of product concentration to feed concen-
tration) x 100.
73
-------
3
w
ID
u.
s.
a.
I/I
I/I
O
U
.c
U
CO
CM
«
U
3
o>
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-76-109
3. RECIPIENT'S ACCESSION"NO.
4. TITLE AND SUBTITLE
DEVELOPMENT OF A MOBILE TREATMENT SYSTEM FOR
HANDLING SPILLED HAZARDOUS MATERIALS
5. REPORT DATE
July 1976 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Mahendra K. Gupta
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORG-\NIZATION NAME AND ADDRESS
ENVIREX INC. (A Rexnord Company)
Environmental Sciences Division
5103 W. Beloit Road
Milwaukee, Wisconsin 53201
1O. PROGRAM ELEMENT NO.
1BB041; ROAP 21AVN; Task 021
11. CONTRACT/SKKKKNO.
68-01-0099
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report documents the results of a laboratory test program undertaken to define
the treatment processes for the development of a modular transportable treatment unit
for an on-site handling of spilled hazardous materials in aqueous solutions. The
hazardous materials evaluated during this study were selected based on the priority
ranking system developed by EPA. Nine materials evaluated for treatment by chemical
reaction, clarification and activated carbon adsorption were: acetone cyanohydrin,
acrylonitrile, ammonia, chlorinated hydrocarbons, chlorine, methanol, phenol,
tetraethyllead (TEL) and tetramethyllead (TML). Several additional materials listed
in the report were evaluated for treatment feasibility by reverse osmosis.
The results of the laboratory tests indicated that the unit treatment processes of
chemical reaction, flocculation, sedimentation, granular media filtration and
activated carbon adsorption would form the most suitable and versatile system for an
on-site removal and treatment of hazardous materials. A 12.6 I/sec (200 gpm) mobile
treatment system consisting of the above-mentioned processes was built based on the
design data outlined in this report. This treatment vehicle is now ready and avail-
able for response to an actual or test spill and should be used at the earliest
possible application opportunity.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
*Water Treatment
*Decontamination
*Activated Carbon Treatment
*Chemical Removal (Water Treatment)
Flocculating
Precipitation (Chemistry)
Water Pollution
Hazardous Material Spill
Cleanup; Hazardous Mate-
rial Spill Control; Haz-
ardous Polluting Substanc
Spills; Hazardous Chemi-
cal Spills; Hazardous
Spills Treatment Trailer
13B
3. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
85
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
75
AUSGPO: 1976 657-695/5456 Region 5-1
------- |