United States
Environmental Protection
Agency
EPA/540/S5-91/006
September 1991
&EPA
SVPERFVND INNOVATIVE
TECHNOLOGY EVALUATION
Technology Demonstration
Summary
Design and Development of a
Pilot-Scale Debris
Decontamination System
M.A. Dosani and M.L. Taylor
Metallic, masonry, and other solid
debris that may be contaminated with
hazardous chemicals litter numerous
hazardous waste sites in the United
States. Pplychlorinated biphenyls
(PCBs), pesticides, lead, or other metals
are some of the contaminants of con-
cern. In some cases, cleanup standards
have been established (e.g., no greater
than 10ng PCBs/100 cm2 for surfaces
to which humans may be frequently
exposed). If decontaminated, this debris
could be returned to the site as "clean"
fill or, in the case of metallic debris,
sold to a metal smelter.
This project involves the develop-
ment and demonstration of a technol-
ogy intended specifically for onsite de-
contamination of debris. Both bench-
scale and pilot-scale versions of a de-
bris washing system (DWS) have been
designed, constructed, and demon-
strated. The DWS entails application of
an aqueous solution during a high-
pressure spray cycle, followed by tur-
bulent wash and rinse cycles. The
aqueous cleaning solution is recovered
and reconditioned for reuse concur-
rently with the debris-cleaning process,
which minimizes the quantity of pro-
cess water required to clean the de-
bris.
This Project Summary was devel-
oped by EPA's Risk Reduction Engi-
neering Laboratory, Cincinnati, OH, to
announce key findings of the SITE pro-
gram demonstration that is fully docu-
mented in a separate report of the same
title (see ordering information at back).
Introduction
Numerous sites in the United States
are contaminated with hazardous waste,
and the cleanup of these sites is a top
environmental priority of the decade. Cur-
rently, more than 1200 sites are included
on the National Priorities List (NPL), and
many more have been proposed for in-
clusion on the list. A typical hazardous
waste site contains toxic organic and/or
inorganic chemical residues that are fre-
quently intermingled with remnants of
razed structures (e.g., wood, steel, con-
crete block, bricks) as well as contaminated
soil, gravel, concrete, and sometimes me-
tallic debris (e.g., machinery and equip-
ment, transformer casings, and miscella-
neous scrap metal). Decontamination of
these materials is important to prevent the
spread of contamination offsite and to fa-
cilitate the disposal of the debris in an
environmentally safe manner. Because
most of the contaminated debris at
Superfund sites has no potential for re-
use, the purpose of a debris decontami-
nation system would be to decontaminate
the material sufficiently to permit its return
to the site as "clean" fill or to allow its
disposal in a Subtitle D sanitary landfill or
municipal incinerator rather than a Re-
source Conservation and Recovery Act
(RCRA) Subtitle C hazardous waste land-
fill or incinerator.
Objectives
The project was conducted in two
phases. The objectives of Phase I were:
. to evaluate a hydromechanical
cleaning system, an innovative ap-
proach for decontaminating debris,
Printed on Recycled Paper
-------
. to conduct bench-scale testing with
a portable module for the decon-
tamination of debris, and
. based on bench-scale results, to
develop a pilot-scale experimental
debris decontamination module
(EDDM).
The objectives of Phase II were:
. to continue development of the
EDDM into a proven technology for
removing various contaminants from
debris found on hazardous waste
sites,
. to conduct bench-scale tests to opti-
mize the process,
. to design and construct a transport-
able pilot-scale DWS,
. to field-test the pilot-scale DWS at
two hazardous waste sites where
various types of debris are present,
and
to prepare a conceptual design of a
full-scale DWS.
This report is presented in two volumes.
Volume describes the design and devel-
opment of the Phase I and II pilot-scale
debris decontamination systems and pre-
sents the results of the decontamination
demonstrations conducted at three haz-
ardous waste sites. Volume II contains
copies of the analytical data submitted by
the various laboratories involved in the
project.
Phase I: Development and
Testing of Experimental
Modules
During Phase I of the project, a hydro-
mechanical cleaning system, an innova-
tive approach to decontaminating debris,
was developed and evaluated. A bench-
scale, portable module consisting of an
enclosure where debris was placed and a
closed-loop solvent-delivery system was
tested. Based on the bench-scale results,
a pilot-scale EDDM was developed and
field-tested.
A 300-gal-capacity pilot-scale EDDM
was designed, assembled, installed (on a
48-ft semitrailer), and tested at the Carter
Industrial Superfund Site in Detroit, Ml.
This site contained large quantities of dif-
ferent types of PCB-contaminated debris,
including scrap metal, 55-gal metal drums,
tools, equipment, and some furniture items.
Two 200-lb batches of metallic debris
were cleaned in the system. Before and
after treatment, surface-wipe samples were
obtained to determine the contaminant re-
moval efficiency of the system. The per-
centage reduction of PCBs achieved dur-
ing cleaning ranged from 33% to 87%
(average reduction of 58%) for Batch 1
and from 66% to 99% (average reduction
of 81%) for Batch 2.
The surfactant solution in the EDDM
was sampled twice during the actual
cleaning process, and PCS concentrations
of 928 and 420 ng/L were found. Upon
completion of the debris-washing experi-
ment, the cleaning solution was pumped
through a series of particulate filters and
finally through activated carbon. The PCS
concentration was reduced to 5.4 n.g/L
during this treatment. Most municipalities
allow water containing a PCS concentra-
tion of <1p.g/L to be sewered, and this
level was achieved by recycling the pro-
cess water through carbon a second time.
Phase II: Design, Construction,
and Demonstration of a Trans-
portable Debris-Washing
System
Phase II of this project was directed
toward further development of debris
washing into a proven technology for re-
moving various contaminants from debris
found on hazardous waste sites in prepa-
ration for full-scale demonstrations at
Superfund and. other hazardous waste
sites. An initial series of bench-scale tests
were performed in a controlled environ-
ment to optimize the newlydesigned debris
washing system. After the bench-scale
evaluation, a transportable pilot-scale ver-
sion of the DWS was designed, con-
structed, and demonstrated at actual haz-
ardous waste sites.
Based on experience gained during the
Carter site field test, a bench-scale (20
gal of surfactant solution capacity) debris
washing unit was designed, constructed,
and assembled. This system consisted of
a spray tank, wash tank, oil-water separa-
tor, and ancillary equipment (i.e., heater,
pumps, strainers, metal tray, etc.). This
bench-scale DWS was developed to de-
termine the ability of the system to re-
move contaminants from debris and to
facilitate selection of the most efficient
surfactant solution.
During these bench-scale experiments,
surface-wipe samples of the six pieces of
control debris were taken before and after
treatment and analyzed for oil and grease.
Based on the results, a nonionic surfac-
tant solution was selected as the solution
best suited for cleaning oily metal parts
and debris.
As part of the continuing investigation
into the performance of the DWS, the
representative pieces of debris were spiked
with a mixture of spiking material (used
motor oil, grease, topsoil, and sand) con-
taining representative contaminants (DDT,
lindane, PCBs, and lead sulfate) and
washed in the DWS with the selected
surfactant solution. Three trials were per-
formed. Surface wipe samples of debris
from the first two trials were analyzed for
PCBs, lindane, and DDT; the surface wipe
samples from the third trial were analyzed
for total lead.
Table 1 summarizes the quantities of
PCBs and pesticides on the surface of
each piece of debris before and after
cleaning (Trial 2). Table 2 summarizes the
quantities of lead found before and after
treatment (Trial 3). The average overall
reductions of PCBs and pesticides
achieved during Trials 1 and 2 were
greater than 99% and 98%, respectively.
The overall reduction of lead was greater
than 98%.
After completion of the bench-scale de-
bris-washing experiments, the cleaning
solution was neutralized to a pH of 8 and
then pumped through a series of particu-
late filters and finally through activated
carbon. During this treatment, the PCS,
lindane. and DDT concentrations were
reduced to <2.0, 0.03, and 0.33 ^g/L, re-
spectively. The concentration of lead was
reduced to 0.2 mg/L after treatment.
Design, Fabrication, and
Demonstration of Pilot-Scale
DWS
Based on the results obtained from
bench-scale studies, a Phase II SOOgal
capacity pilot-scale DWS was designed
and constructed. The process flow dia-
gram of the pilot-scale system is presented
in Figure 1.
The pilot-scale DWS was assembled in
a warehouse in Cincinnati, OH, and sev-
eral tests were conducted. After the
warehouse testing, the DWS was disas-
sembled, loaded onto a 48-ft semitrailer,
and transported to the Gray PCS site in
Hopkinsville. KY, which was selected for
the field demonstration. The entire DWS
was reassembled on a 25-ft x 24-ft con-
crete pad. A temporary enclosure (ap-
proximately 25 ft high) was built on the
concrete pad to enclose the DWS and to
protect the equipment and the surfactant
solution from rain and cold weather. The
Gray PCS site contained between 70 and
80 burned out transformer casings and
other large amounts of scrap metal. The
demonstration took place in December
1989, and ambient temperatures were at
or below freezing during the entire opera-
tion.
-------
Table 1. Summary of Bench-Scale Results of Controlled Debris Analyzed for PCBs and Pesticides (Trial 2)
Controlled
Debris Contaminant
Lindane
Metal 4.4' DDT
PCB- 1260
Lindane
Metal 4.4' DDT
PCB- 1260
Lindane
Metal 4.4' DDT
PCB- 1260
Lindane
Brick 4.4' DDT
PCB- 1260
Concrete Lindane
Block 4,4' DDT
PCB- 1260
Lindane
Plastic 4.4' DDT
PCB- 1260
. U indicates that the target compound
Pretreatment Posttreatment . Percent Average
fog/WO cm*) fag/lOO cm") Reduction Reduction
11,800
9320
1770
8180
7540
f 780
6750
5840
7450
5870
5660
7220
6440
6670
7390
70,300
8400
7620
was not detected at
Table 2. Summary of Bench-Scale Results of Controlled
Controlled
Debris Contaminant
Metal Lead
Metal Lead
Metal Lead
Brick Lead
Concrete
Block Lead
Pretreatment
fag/1 00 cm *)
876
414
450
508
474
0.73 U
2.32
2.0 u
0.37 u
4.8
2.79
0.41
2.61
2.0 u
3.49
10.5
4.1
397
389
66.7
52
223
35
this level.
Debris Analyzed for Lead (Trial 3)
Posttreatment Percent
(\ig/100 cm *) Reduction
6.0 99.31
6.0 98.55
<3.0 >99.33
<3.0 >99.4 (
<3.0 >99.27
700
99.97
299.89
700
99.94 >99.91,
99.84 for metal
99.99
99.95
>99.86
99.94
99.87 99.80
99.66
93.83
94.11 94.39
95.24
99.49
97.34 98.22
97.84
Plastic
Lead
446
<3.0
>99.33
-------
•— Sl«p 1 • Sprty Cyd«
•ftfft. sup 2 - W»h Cydt
»•• sup 3. Rinit Cyd«
DE F»t«r
• " i W«Ur Traimwd Sltp
Q Pump
Activated Ctrbon
Figure 1. Schematic of Pilot - Scale Debris Washing System.
Before the cleaning process began, the
transformer casings (ranging from 5 gal to
100 gal in size) were cut in half with a
metal-cutting partner saw. A pretreatment
sample was obtained from one-half of each
of the transformer casings by a surface-
wipe technique. The transformer halves
were placed into a basket and lowered
into the spray tank, which was equipped
with multiple water jets that blast loosely
adhered contaminants and dirt from the
casings. After the spray cycle, the basket
of casings was removed and transferred
to the wash tank, where the debris was
washed with a high-turbulence wash. Each
batch of debris was cleaned for a period
of 1 hr in the spray tank and 1 hr in the
wash tank. During both the spray and
wash cycles, a portion of the cleaning
solution was cycled through a closed-loop
system in which the oil/PCB-contaminated
cleaning solution was passed through an
oil/water separator, and the cleaned solu-
tion was then recycled into the DWS. Af-
ter the wash cycle, the basket containing
the casings was returned to the spray
tank, where it was rinsed with fresh water.
On completion of the cleaning process,
posttreatment wipe samples were obtained
from each of the transformer pieces to
assess the post-decontamination PCS
levels. The average PCS concentrations
on the internal surfaces of the transformer
casings before and after cleaning are
summarized in Table 3. The before-treat-
ment concentrations ranged from 0.1 to
98 ng/100 cm2. The after-treatment analy-
ses showed that all the cleaned trans-
formers had a PCS concentration lower
than the acceptable level of 10 ^g/100
cm2.
After treatment of all the transformers
at the site, the surfactant solution and the
rinse water were placed in the water treat-
ment system, where they were passed
through a series of particulate filters, then
through an activated-carbon drum, and
finally through an ion-exchange column.
The before- and after-treatment water
samples were collected and analyzed for
PCBs and selected metals (cadmium,
copper, chromium, lead, nickel and ar-
senic).
The water treatment system reduced
the PCS concentration in the water to
below the detection limit. The concentra-
tions of each of the metals (except ar-
senic) were reduced to the allowable dis-
charge levels set by the city of Hopkinsville.
On receipt of the analytical results, the
treated water was pumped into a plastic-
covered 1 0,000-yd3 pile of contaminated
soil at the site.
During this site cleanup, 75 transform-
ers were cleaned in the DWS. All of them
are now considered clean and acceptable
for sale to scrap metal dealers or to a
smelter for reuse.
Demonstration at the Shaver's
Farm Drum Disposal Site
In August 1990, a second demonstra-
tion of the DWS was conducted at the
Shaver's Farm drum disposal site near
Chickamauga, GA, where 55gal drums
containing varying amounts of a herbicide,
Dicamba (2-methoxy-3,6 dichlorobenzoic
acid), and benzonitrile (a precursor in the
manufacture of Dicamba) were buried.
EPA Region IV had excavated more than
4000 drums from one location on this 5-
acre site when this demonstration oc-
curred.
The pilot-scale system was transported
to this site on a 48-ft semitrailer and as-
sembled on a 25x24 ft concrete pad. The
temporary enclosure used at the Gray site
was reassembled to protect the equip-
ment from rain. Ambient temperature at
the site during the demonstration ranged
from 75° to 105° F.
The 55-gal herbicide-contaminated
drums were cut into four sections, and
pretreatment surface-wipe samples were
-------
obtained from each section. The drum
pieces were first placed in the spray tank
of the DWS for 1 hr of surfactant spray-
ing, then in the wash tank for an addi-
tional hour of surfactant washing, and fi-
nally in the spray tank for 30 min of water
rinsing. The drum pieces were then al-
lowed to air-dry before the posttreatment
surface-wipe samples were taken. Ten
batches of 1 to 2 drums per batch were
treated during this demonstration.
Tables 4 and 5 summarize the surface-
wipe concentrations of benzonitrile and
Dicamba, respectively, on the internal sur-
faces of the drums before and after clean-
ing. Pretreatment concentrations of
benzonitrile in surface wipe samples
ranged from 8 to 47,000 ng/100cm2 and
averaged 4556 |j.g/100cm2; posttreatment
samples ranged from below detection limit
to 117 m)/100cm2 and averaged 10 \ig/
100 cm2. Pretreatment Dicamba values
ranged from below detection limit to 180
H.g/100cm2 and averaged 23 ng/1oocm2;
posttreatment concentrations ranged from
below detection limit to 5.2 ng/100cmz
and averaged 1 ng/100cm2.
All site activities described in this docu-
ment were governed by EPA-approved
Health and Safety and Quality Assurance
Plans.
Conclusions
Field-test results obtained with the pilot-
scale DWS in demonstrations at two Re-
gion IV hazardous waste sites showed
the unit to be both transportable and rug-
ged. Extreme high and low temperatures
had little effect on the operation of the
equipment. The system successfully re-
moved PCBs from transformer casing sur-
faces and herbicides and pesticides resi-
dues from drum surfaces.
The cleaning solution was recovered,
reconditioned, and reused during the ac-
tual debris-cleaning process; this mini-
mized the quantity of process water re-
quired for the decontamination procedure.
The water treatment system was effective
in reducing contaminant concentrations,
with the exception of arsenic and possibly
Dicamba, to below the detection limit.
Planned progression of this EPA-devel-
oped technology includes design, devel-
opment, and demonstration of a full-scale,
transportable version of the DWS unit.
The full report was submitted in fulfill-
ment of EPA Contract No. 68-03-3413 by
ITEP, Inc., under the sponsorship of the
U.S. Environmental Protection Agency.
TABLE 3. Results Obtained During Field Demonstration of DWS at Gray PCB Site
Average PCB Concentration of Surfaces fog/lOO cm 'l)
Number
1.
2.
3.
4.
5.
6.
7.
6.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
Before
Average
19.7(1^=10)
9.9 (N=6)
6.6 (N=4)
4.1 (N=6)
4.0 (N=8)
2.0 (N=4)
2.8 (N=2)
23.5 N=5)
8.3 (N=4)
5.2 (N=4)
9.4 (N=4)
48.8 (N=4)
12.3 (N=2)
16.7 (N=2)
18.5 (N=4)
11.3 (N=2)
24.8 (N=4)
8.4 (N=5)
8.3 (N=4)
24.0 (N=3)
18.6 (N=8)
25.0 (N=4)
8.6 (N=4)
6.8 (N=8)
Cleaning
Range
< 0.1-94.0
4.8- 17.0
5.0 - 9.9
<0. 1- 12.0
<0. 1 -28.0
<0.1 - 7.8
1.4 - 4.3
<0. 1 -70.0
2.9 - 23.0
<0.1 - 9.7
<0.1 - 17.0
2.3 - 98.0
9.6 - 15.0
8.7 - 25.0
8.1 - 27.0
8.6 - 14.0
1.1 - 80.0
<0.1 - 19.0
<0.l- 18.0
13.0 - 45.0
<0. 1 - 44.0
12.0 - 35.0
1.5 - 18.0
<0. \ - 31.0
After
Average
1.5 (N=10)
1.5 (N=6)
1.4 (N=4)
0.8 (N=6)
<0. 1(N=8)
2.9 (N=4)
3.9 (N=2)
1.3 (N=5)
3.1 (N=4)
1.9 (N=4)
3.0 (N=4)
1.1 (N=4)
5.1 (N=2)
*01(N=2)
-------
Table 4. Results of Surface Wipe Samples Analyzed for Benzonitrile, 2,4-Dichlorophenol, 2, 6-Dichlorophenol, and 1,2,4-Trichlorobenzene
During Field Demonstration of DWS at Shaver's Farm Site (\ig/100cm*)
Benzonitrile
2,4-Dichlorophenol
2,6 Dichlorophenol
1,2,4-Tricholorbenzene
Batch Sample Pre-
Number Number treatment
\
2
3
4
5
6
7
8
9
10
\
2
1
2
1
2
J
2
f
2
1
2
J
2
J
1
2
1
2
180'(50)t
130"(50)
125
90
43
28
4400
2700
4700
2200
10*(5)
8 "(5)
200
320
1400
3000
3500
22*©
1400
Post- Pre-
treatment treatment
ND§
ND
117
7.8"(5)
ND
ND
ND
HO
10* (5)
7.9'(5)
ND
ND
ND
W(5)
28
ND
7'(5)
ND
ND
ND(50)
ND(50)
34
43
ND
ND
NM
HA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
16*(5)
14*(5)
NA
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Pre-
treatment
ND(50)
ND(50)
ND
ND
ND
ND
NA
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Post- Pre-
treatment teatment
ND
ND
ND
ND
ND
ND
NA
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND (50)
ND (50)
ND
ND
ND
ND
NA
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Post-
treatment
ND
ND
ND
ND
ND
ND
NA
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
. Estimated result less than 5 times detection limit.
/ Numbers in parenthesis indicate the minimum detectable concentration of the analyte.
§ None detected in excess of the minimum detectable concentration of 5 \ig/100crr? unless otherwise specified.
A Not analyzed.
Table 5. Results of Surface Wipe Samples Analyzed for Dicamba, 2,4-D, and 2,4,5-7 During Field Demonstration of DWS at Shaver's Farm Site
fag/I 00cm2)
Dicamba 2.4-D 2,4,5-7
Batch
Number
4
5
6
7
8
9
10
Sample
Number
1
2
1
2
1
2
f
2
1
2
1
2
1
2
Pre-
treatment
1.9
3.4
ND
ND
ND(2.7)
ND(2.7)
7.3*(2.7)
15
55
13
1.7
ND(2.7)
41
180
Post-
treatment
0.63*f0.27)*
ND
ND
2.6
ND
ND(2.7)
1.8
2.3
5.7*(2.7)
0.62*(0.27)
0.63'(0.27)
ND
0.30*(0.27)
0.34*(0.27)
Pre-
treatment
ND§
NM
ND
ND
ND(12)
ND(12)
ND
ND(12)
ND(12)
ND
ND
ND(12)
ND(12)
ND(12)
Post-
treatment
ND
NA
ND
ND
ND
ND( 12)
ND
ND
ND(12)
ND
ND
ND
ND
ND
Pre-
treatment
ND
NA
ND
ND
ND(2.0)
ND(2.0)
ND
ND(2.0)
ND(2.0)
ND
ND
ND(2.0)
ND(2.0)
ND(2.0)
Post-
treatment
ND
NA
ND
ND
ND
ND(2.0)
ND
ND
ND(2.0)
ND
ND
ND
ND
ND
' Estimated result less than 5 times detection limit.
* Numbers in parenthesis indicate the minimum detectable concentraiton of the analyte.
§ None detected in excess of minimum detectable concentration of\Dicamba at 0.27: 2,4-D at 1.2; and 2.4.5-7 at 0.20 unless otherwise specified.
A Not analyzed.
6U.S. GOVERNMENT PRINTING OFFICE: 1991 •
-------
M.A. Dosani and M. L Taylor, are with IT Environmental Programs, Inc., Cincinnati OH
45246.
Naomi P. Berkley is the EPA Project Manager (see below).
The complete report, entitled "Technology Evaluation Report: Design and
Development of a Pilot-Scale Debris Decontamination System,"
consists of two volumes:
"Volume /"(Order No. PB91-231456AS; Cost: $19.00, subject to change)
discusses the development, demonstration, and evaluation of the debris
decontamination system.
"Volume II" (Order No. PB91-231464AS; Cost: $35.00, subject to change)
contains copies of the analytical data submitted by the various laboratories
involved in the project.
Both volumes of this report will be available from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22 16 f
Telephone: 703-487-4650
The EPA Project Manager can be contacted at:
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Envirnmental Research
Information
Cincinnati, OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA PERMIT NO. G-35
Official Business
Penalty for Private Use $300
EPA/540/S5-91 /006
l HHERL
JOHN KING
SAIC CO
ONE SEARS DR
PARAMUS
NJ 07652
------- |