-------







_J
a
o
111
I—
2
/y
t—
U.
O
t/>
UJ
i—
QC
5
£
a
c/3
^
a.

00
N!
UJ
3

























'*• ir!
— • V
•s i?
1*
E
\S
1||
Q| ±5 o »!
sl

£
J_»
§
^
i
0)
3
.2 **
'5


^^
CO
**X
— ' 9
•s s
!*
£

«. I


C CO J=
5: 8 £
§ E. S
8 8 i
3 O M

£
J"~

— -
0, |
M -^
01


01
c.
3
« X
'5


tj
S S
a. to
O* O* 00 O^ (^ O^ O^ COCO
1 1 1 t 1 1 1 II
oooooo o ^° }Q
XXXXKX X XX <
eooNtmocM o^ rxjco
^-•^•ed^trtw in fOfo
FO FO O*> to *<^ O* ^ O OQ O^ ^
»- rU <- vt N- eo ^ «- ro ro ro
v^inNOLncoo* (> \r\ >




o> O^ ^ ^O fM ^* N- o^ «o in KI
d» m m co o r*- o o to m f\J
v- «— v— «— .
T- «- CM «-«-«-«- «- oj r\j <



^.
o
O\ Qs O*- O^ *~ O*1
oooooo **
X X X X X X <
^o r»- m in h- N? ^ ^* *•** **>
"""""" 2 S 22





N-h-coN-trt^co o S-ro \r\
**cMropdrrop^iA o CM S- «-
^>* o ^£ co ^_ o ro^





»^~ ^ ^ r*~ o oo oo ^ Q *^
^OinoOor*.c^ oo ^* *o «~
^ ^-  S CM*~ T
H o o

c/> < vi u- u. < ale — • o v t/> v •
.^- 01 O
,Q t_ *J
ro (D
a> 0) t_
E -^ Q
(U ID (U
a. t— — i

m .Q p












.
I
u
oo
o
S
rtj
S
S
"1
>
•o
pj^
jz
«<
^

u
01
(A
°
O
i
O

. 1^
1 "8
2 %
t. L,
^ «
O 3

ro

-n 5

-g s
U) w
*J 0)
o .c


T3 01

-------
o   The wet/dry and freeze/thaw weathering tests showed small
    weight losses (0.5-1.5%)  at the end of the 12-cycle test
    for the test specimens and their controls.  Unconfined
    compressive strength tests performed after the final
    weathering cycle showed no loss in UCS compared to the
    unweathered samples.

o   Both field- and laboratory-cured samples were used to
    evaluate the effectiveness of the HAZCON process during
    the Demonstration test.  The laboratory formulations  were
    prepared using cement with clean soil, FSA and LAN soils
    without Chloranan.   For bulk density and moisture, the
    values were close to field-processed soils.   For
    uncorifined compressive strength (UCS) FSA was reduced to
    27 psi at 7 days and 38 psi at 28 days.  For LAN the  UCS
    values were 373 psi at 7  days and 539 psi at 28 days.
    These latter values are comparable to the field-treated
    samples.  In comparison,  the clean soil produced the
    highest unconfined  compressive strength of 2000 psi.   The
    permeabilities of the laboratory-formulated  samples were
    in the range of 3.8xlO"8  for LAN to 5.9xlO~9 for clean
    soil and cement.  The values are equivalent  to the field
    results for FSA but a factor of about ten greater for LAN.

It should be noted that due to the difficulties  and time
required to perform the permeability tests, these tests were
run anywhere from 6 to  15 weeks after the soils  were treated.

7.2.?.  CHEMICAL ANALYSES

7.2.2.1  UNTREATED SOIL ANALYSES

The results of the untreated  soil analyses are reported in
Appendix B and summarized in  Table 7.9.  They can be
highlighted as follows:

o   Oil and grease ranged from 1.0% by wt. for DSA to 25.3% by
    wt. for FSA.  The oil and grease levels in the 28-day
    treated cores were  about  40% of the untreated soils as
    would be anticipated from the material balances.

o   Total PCBs ranged from 1.2 ppm for DSA to 52 ppm by wt.
    for LAS.  Most of the PCBs were Aroclor 1260.  However,
    Aroclor 1248 was measured at 19 ppm by wt. in PFA and 25
    ppm by wt. in LAS.   In the other areas, Aroclor 1248 was
    below detection limits of 20 parts per billion by wt.
    Aroclor 1260 was measured in all six plant areas.  These
    values are greater than those reported in the screening
    study, whose samples were collected in May 1987 and are
    reported in Table 4.1.

o   The thirteen priority pollutant metals were analyzed for
    in the untreated soil.  Lead is the predominant metal
    contaminant.  Five other metals of measurable
    concentration were analyzed  for in the 28-day cores  and
    leachates.  These other metals are chromium, nickel,

                               82

-------



00
— J
o
oo

o
UJ
r™

h—
u_
O

oo
oo
	 i
z

_J
o
_


re
o
o>
^


J
CQ
- ro o -t
in N* o

~* O -J- Is-
O >O OJ -J-
r>- N. •«• CM
OJ "O
o o ** ro



ro o ^ Z C\J *
OJ
10
in o o 
OJ (M U -Q
L. L. OJ & 3 E
o o t. Q •- — • 5
"o "o i T3 O It E
O O TJ fO L. 0 fl
t-L. C « 41 j: •— 10
<< ro — _i o z t_>
CO ID
S— . 4->
• ^ CD
Q. ox:


o o
-* CO
»- in




5 o


o in
0 ro





oo tn
r\j tn
«— >o


in o
10 in




^* in
ro










L.
o £
o r-j




o o
000
OJ Z >$•
vt" in"
ro



o o
ic o s

O N-
O O
O O
in o oj
to" ro"
ro


o o o
o o o
OJ O O
-.j- in in
«- o «-

o o
o o
h* OJ Q

in in

o
in
«- o o
•• z z
OJ





1

*•> 03 « o ro *—
oj «- «~ o

O O O CD O
o ro ^0 co rv
o «-«-»- o

"


o o o o o
z z z z z



w cu
i C
O (D
J5 S 5  O d>
Q 0) 1- N
Q 0 0 C
L. — • » Q>
" O t! 0) 4J X
*-• H- h- t— UJ X
ID
"o





































i

V
-------
    cadmium,  copper,  and zinc.   The average of the analyses in
    the six areas shows that lead exists in concentrations
    from 0.3% to 2.3% by wt.  Chromium,  nickel,  and cadmium
    exist at  concentration levels of less than 100 ppm by wt.
    with cadmium the  least.   Copper and  zinc are at
    concentrations of a few  hundred ppm  by wt.  These overall
    levels agree with the values obtained from the screening
    samples.

o   Analyses  for BNA  showed  that all areas contained
    phthalates,  primarily di-n-butylphthalate and
    bis(2-ethylhexyl)phthalate.   Total  phthalate
    concentrations up to 34  ppm  by wt.  in LAS were measured.
    Phenols were also abundant,  with total concentrations over
    400 ppm measured  in the  FSA.  Naphthalene also was
    detected  in  the soil in  all  areas except DSA and LAN with
    concentrations over 100  ppm  by wt.  at FSA.
    Bis(2-ethylhexyl)phthalate was the  primary semivolatile
    detected  in  the screening samples.

o   Volatiles were not detected  in the  soil samples at DSA and
    LFA.  The primary volatiles  detected were toluene, xylene,
    trichloroethene,  tetrachloroethene,  and ethyl benzene.
    Some samples also contained  1,1,1-trichloroethane and
    trans-1,2-dichloroethene.  The maximum volatiles, based
    upon multiple sample averaging, were measured at FSA with
    up to 14  ppm by wt. trichloroethene, 6 ppm
    tetrachloroethene, 26 ppm toluene,  and 91 ppm xylenes.
    The area  with the second largest quantity of volatiles is
    LAS with  maximum  sample  values primarily in  the range for
    each component of 0.3-3.7 ppm by wt.  Toluene was injected
    into the  slurry samples  exiting the  HAZCON unit for DSA,
    LFA, and  PFA, to  produce an  equivalent concentration in
    the feed  soil of  approximately 125  ppm by wt.
    Concentrations of the Injected toluene in the 7-day core
    samples ranged from 1.3  ppm  by wt.  in DSA to 12 ppm in
    LFA.  The toluene in the 28-day core samples ranged from a
    minimum at DSA of 1.3 ppm by wt. to  a maximum at LFA of 24
    ppm by wt.

7.2.2.2  LEACHATE ANALYSES

Results of the leachate analyses, for which data reduction
calculations  are provided in Appendix D, are highlighted as
fol1ows:

o   The TCLP  leachates of the six feedstock soils showed very
    low oil and grease.  For each area,  the treated soil
    leachate  concentrations  were greater than for untreated
    soil leachates, even though  the untreated soil
    concentrations are greater on average by a factor of 2.5.
    The values were from below the detection limits (0.2 mg/1)
    for DSA to 3.7 mg/1 at  FSA.   The leachate results for the
    7-  and 28-day cores appear higher than the untreated soil,
    ranging from 0.6 ppm by wt.  for DSA to 4.1 ppm  for LAN and

                              84
-------
FSA at 7 days, and 1.6 to 10.4 ppm by wt. for LAS and FSA,
respectively at 28 days.

PCBs were not detected in any leachate, whether the soil
was treated or untreated.

The 7- and 28-day TCLP leachates for metals showed that
the quantity of metals in the treated soil  leachates (see
Table 7.10) is well  below that for untreated soils.   For
the 7- day and 28- day cores leachates, except for lead,
all the metal concentration levels were near or below
detection limits.  The lead concentrations  were close to
detection limits, at levels of about 2-90 ug/liter except
for FSA, where the concentration ranged from 7-950 ug/1,
averaging 400 ug/1.   For the untreated soils, chromium and
copper were below detection limits and cadmium and nickel
were near the detection limits.   The lead concentration
ranged from 1.5 mg/1 for DSA to  52.6 mg/1 for LAS.  For
zinc, the values ranged from 0.7 mg/1 for DSA to 23.0 mg/1
for FSA.  The solidification process reduced the lead
concentration by a factor of ajjout 500 to 1000.  TCLP data
on the 28-day cores  were equivalent to the  7-day cores.

The leachate concentrations from ANS 16.1 were generally
equivalent to those  of TCLP, while those of MCC-1P were
greater by a factor  of 5-10.  The leachate  concentrations
increased with increased leaching time for  both tests.

For the BNAs, very significant reductions in
concentrations were  obtained between the soil and the TCLP
leachates of the untreated, 7-,  and 28-day  cores.  The
phthalates, as a group, for all  leachates were reduced to
approximately their  detection limit of 0.010 mg/1.  This
may be because the phthalates concentrations were very
significantly reduced from the untreated soils to the
treated soils.  For  the phenols, measurable quantities in
the leachates were observed.  For FSA, where the soil
samples contained about 400 ppm  by wt, leachate
concentrations of 3-4 mg/1 were  measured.  In each area,
concentrations in the untreated  soil, 7-day, and 28-day
core leachates were  approximately the same, even though
the concentrations in the cores  were less than one-half
that in the untreated soil.  The same trend existed for
naphthalene, except  at lower concentrations.  Table 7.11
provides a summary of the leachate concentrations.

The special leach tests, performed only on  28 day cores,
provided results that showed that the leachate
concentrations from  MCC-1P were  equivalent  to TCLP
leachates, but greater than for  ANS 16.1.  For the ANS
16.1 leachates, concentration did not appear to be a
function of leaching time.  However, for MCC-1P the phenol
concentrations in the extracts increased with time,  but
the phthalates and naphthalene did not.
                          85
-------
      TABLE  7.10.   CONCENTRATION  OF  METALS  IN  TCLP  LEACHATES
                                                         *
                             Metal Concentration -  mg/liter

Soil  Location       Pb         Cr           Ni          Cd          Cu          Zn
Soil
DSA
LAN
FSA
LFA
PFA
LAS

1.5
31.8
17.9
27.7
22.4
52.6

<0.008
<0.008
0.27
<0.008
<0.008
<0.008

0.02
0.07
0.11
0.06
0.05
0.07

<0.004
0.02
0.13
0.03
0.01
0.04

<0.03
<0.03
<0.3
<0.08
<0.03
0.13

0.07
1.1
23.0
6.7
1.4
4.8
7-Day Cores
    DSA            0.015    <0.07       <0.15       <0.04       <0.06        <0.02
    LAN           <0.002    <0.07       <0.15       <0.04       <0.06        <0.02
    FSA            0.07      0.02       <0.008      <0.003      <0.03         0.02
    LFA            0.04     <0.07        0.15       <0.04       <0.06         0.04
    PFA            0.01     <0.07        0.15       <0.04       <0.06        <0.02
    LAS            0.14     <0.008      <0.008      <0.003      <0.05         0.04

28-Day Cores
DSA
LAN
FSA
LFA
PFA
LAS
0.007
0.005
0.400
0.050
0.011
0.051
<0.007
0.007
<0.070
0.009
<0.007
0.015
0.020
<0.015
<0.15
0.015
<0.015
0.025
<0.004
<0.004
<0.040
<0.004
<0.004
<0.004
0.023
0.010
<0.060
0.080
0.027
0.055
0.037
0.017
0.037
0.013
0.030
0.258
            Where  the symbol < is used,  indicates values below detection limits of
            quantity shown.  The detection  limits vary between metals and from
            analysis to analysis.

            Where  2 of 3 values were above  detection limits,  three  values were averaged
            assuming the one below detection  limits is zero.   If  only one of three values
            are above detection limits,  the results are reported  as below detection
            Iimits.
                                                   86
-------
          TABLE 7.11
     BASE NEUTRAL/ACID EXTRACTABLES
     IN TCLP LEACHATES
BNA
OSA
    Concentration -  ug/1

LAN     FSA     LFA      PFA
LAS
Untreated Soil

  phthalates       ND      10      ND     10      10      NO
  phenols          ND    1010    2810     ND      ND      ND
  naphthalene      ND      ND      50     ND      ND      10

7-Dav Cores

  phthalates       ND      30      ND     10      20      ND
  phenols          40    1310    3850     30      50     470
  naphthalene      15      ND      60     10      20      ND

28-Dav Cores

  phthalates       ND      10      10     20      30      80
  phenols          ND    1440    2720     80      80     650
  naphthalene      ND      ND      60     ND      ND      ND
ND - Not Detected
                               87
-------
o   For the volatile organics,  the primary compounds detected
    were trichloroethene,  tetrachloroethene,  xylenes,  ethyl
    benzene,  and toluene.   During the Demonstration Test
    operations,  toluene was injected into the slurry mix zone
    of the HAZCON MFU for  DSA,  LFA and PFA as described
    earlier.   For the TCLP leach tests on the untreated soils,
    toluene was  added to the soil at approximately the same
    concentrations as measured  in the 7-day cores.  In the
    TCLP leachates for untreated soils for DSA,  LFA, and PFA,
    toluene concentrations were quite high, 900-5100 ug/1.
    For FSA,  where the feedstock averaged 26  ppm by wt.
    toluene,  the soil leachate  averaged 230 ug/1 toluene.   The
    other primary volatile organics, trichloroethene,
    tetrachloroethene, ethyl benzene, and xylenes were greatly
    reduced in the untreated soil leachates.

    The results  of the 7-day and 28-day core  leachates are
    similar in magnitude to the untreated soil,  even though
    the core concentrations are less than one half of the
    untreated soil.  See Table  7.12 for a comparison of the
    average key  volatile organic components in the TCLP
    1eachates.

    The results  of the two special leach tests showed that the
    leachate concentrations were approximately equal.   They
    were lower by about a  factor of 2 compared to TCLP
    extracts.  VOC concentrations did not appear to increase
    with increased leaching time intervals.

o   Extraction Procedure Toxicity (EP Tox) tests were
    performed on 28-day cores for metals only, for LAN and
    FSA.  For LAN, the leachate metals were predominantly lead
    (Pb) and averaged 0.02 mg/1 for cores averaging
    approximately 0.3% Pb  by wt.  For FSA, where the cores
    contained 1.0% by wt.  lead, the leachate concentrations
    averaged 0.21 mg/1.  These values appear to be larger for
    LAN and less for FSA than the TCLP results.

7.2.2.3  WATER CHEMISTRY

An analysis of the process water, supplied by truck from the
local   fire department, was performed.  The water was low  in
dissolved solids, 340 mg/1, and very low in suspended  solids,
1 mg/1, with a pH of 8.05.  The primary cations detected were
calcium, magnesium,  sodium, and silicon.  Two of  the primary
anions detected were sulfate and chloride at concentration
levels of 35 mg/1.   Details of the water chemistry  analyses
are provided in Appendix B, Tables A-50 and A-51.

7.2.3  MICROSTRUCTURAL STUDIES

Analyses of the untreated  soil and  28-day core  samples were
performed on a microstructural scale.  All analyses were
performed more than  three  months  after soil processing.   All

                               88
-------
          TABLE  7.12.   VOLATILES  IN  TCLP  LEACHATES^
Volatile Organic
     Concentrations - ug/l

DSA      LAN     PSA       LFA
                                     PFA
                                              LAS
Untreated Soil
Toluene
Xylenes
Trichloroethene
Tetrachloroethene
Ethyl benzene
7 Day Cores
Toluene
Xylenes
Trichloroethene
Tetrachloroethene
Ethyl benzene
28 Day Cores
Toluene
Xylenes
Trichloroethene
Tetrachloroethene
Ethyl benzene

915
< 50
< 20
< 40
< 70

380
3.5
< 10
< 20
< 40

370
6
< 9
< 6
< 3

10
7
2.4
< 4
< 7

< 6
6
< 2
< 4
< 7

40
8
2
3
2

245
525
165
19
80

220
340
105
11
60

230
330
100
20
60

5100
< 230
< 95
< 210
< 360

210
5
< 2
< 4
< 7

370
< 6
< 9
< 6
2

1100(b)
< 180
< 76
< 160
< 290 <

350 <
20
< 5 <
< 10 <
< 20 <

670
170
< 9
< 6
< 3

10
35
8
5
7

15
15
5
10
20

50
40
8
10
4
     (a)  < indicates  less than detection limits.  Within one sampling area, the
         detection limit may change between samples.   For these,  the highest
         detection limit is shown.
     (b)  Two values <60 and 2200 ug/l.
                                                 89
-------
samples were studied by scanning electron microscopy (SEM),
x-ray diffraction (XRD),  and optical  microscopy (OM) .   Energy
dispersive x-ray spectrometry (EDXRA) was also performed on
selected samples.  The type of information to be obtained from
each test is:

o   X-ray diffractometry  - crystalline structure of the soil
    and hydration products

o   Energy dispersive x-ray spectrometry - elemental analysis,
    i.e., calcium,  aluminum

o   Microscopy - characterizes crystal appearance,  porosity,
    fractures, and the presence of unaltered soil/waste
    material

The detailed report with  photographs and x-ray diffraction
patterns is included in Appendix C.

The results can be summarized as follows:

o   The untreated soils consist of quartz and clay minerals,
    illite, and in some cases kaolinite.  The filter storage
    area feed was low in  quartz, as would be expected.

o   The solidified samples show crystals of portlandite,
    ettringite, calcium silicates, calcium aluminate,  and
    sometimes gypsum.

o   Abundant pores of various sizes and shapes could be seen
    in all core samples.   Some of the pores include trapped
    air bubbles.  Large cracks also are seen in some of the
    pores.

o   Unhydrated tricalcium silicate was seen in all  core
    samples.  In some cases dicalcium silicate was observed,
    particularly for FSA.

o   Several peaks in each x-ray diffraction pattern could not
    be identified, as known minerals, but were seen both in
    the soil and core samples.  These peaks are likely to be
    the organics.

o   Mixing does not appear to be completely efficient.

o   The cores contain unaltered brownish aggregates that were
    also observed in the untreated soil.

The interpretation of the above observations is presented  in
section 8.1.3.
                               90
-------
7.3  DATA QUALITY ASSURANCE

In Section 7 of the approved Quality Assurance Project Plan
(QAPP), it was indicated that various Quality Control  (QC)
samples would be taken to control  and/or assess data quality.
These are:

o   QC check samples - standard samples of known analyte
    concentration.

o   Laboratory Blanks - deionized  water taken through  sample
    preparation steps.

o   Field blanks -  clean soil samples brought to the field and
    then analyzed in the laboratory to check for field
    contami nati ons.

o   Spiked samples  - samples were  spiked with either known
    contaminants or surrogate standards to confirm analytical
    recoveries and  thus accuracy of the analyses.   Duplicates
    on the spiked samples were also performed.

o   Duplicate samples - duplicate  samples from the field were
    collected and analyzed to confirm soil sample  data.

To verify that correct sampling procedures were used,  EPA sent
a Quality Assurance (QA) team to the field to observe  Radian
Corporation's sampling procedures.   In addition, QA teams went
to Radian's laboratories both in Austin, TX and Sacramento, CA
to observe and correct, if necessary, procedures being used in
the laboratory.  The audits found  Radian's work satisfactory.

The detailed QA/QC  results reported on Radian procedures is
provided in Appendix B.  Overall the QA/QC data indicated that
the measurement data are acceptable and defensible.

The purpose of the  QA/QC program was to fulfill two related
purposes :

o   An organized frame work for sampling and analytical
    efforts.

o   To control data quality within  preestablished  limits to
    ensure that it  was adequate to  achieve the objectives of
    the program.

The following is a  brief summary of the QA analyses:

o   Soil samples

       Blanks--For  metals, the blanks showed levels the  same
       as in uncontaminated soil.   PCBs were not detected in
       the reagent  blank.  Methylene chloride, toluene,
       acetone, and three other VOCs were detected in  the

                               91
-------
   reagent blanks.   Methylene  chloride  is  probably  a
   laboratory contaminant.   Only  toluene  of these  six  was
   considered in the VOC analyses,  and  some errors  in
   these values may exist.   Toluene was  only one  of five
   VOCs looked at for reporting  VOC data.

   Spiked sample results — All  the metals  recoveries and
   almost all the VOC and PCB  recoveries  were within  the
   acceptance range.

   Duplicate sample results--Matrix spike  duplicates  for
   metals showed excellent  repeatability,  with a
   coefficient of variation (CV)  less  than 5%.
   Repeatability for VOCs was  also  acceptable, with a  CV
   of 50%.  The repeatability  for PCBs,  which had  a CV
   within 10%, was  acceptable;  50% is  the  maximum
   acceptable CV.  The repeatabi1itity  for O&G was  good,
   with almost all  CV less  than  5%, but  wide variations in
   sample homogeneity was observed.

Soil  sample leachates

   Blanks — Only acetone was detected in  two samples near
   the detection limits of 5 ug/1 and  was  probably  due to
   laboratory contamination.  Metals and  PCBs were  not
   detected in the  blanks.

   Spiked samples — Matrix spike  recoveries of lead  were
   high; therefore, results for  Pb may  be  biased  slightly
   high.  The recovery of the  other metals was within  the
   acceptable target range of  80-120%  of the theoretical
   value.  Volatile organic matrix and  surrogate  spikes
   were all within  the acceptance criteria.  For  PCBs  the
   surrogate spike  recovery of the samples was below
   acceptance limits; therefore  the detection limits of
   these compounds   (measured by  Method  680) could  be
   slightly greater than the laboratory reported.   PCBs
   were not found in any leachates.

-  Duplicate sampl es —Resul ts  for VOC  and O&G all  met
   acceptance criteria.  Duplicates for untreated metals
   samples were not performed,  but duplicates for metals
   in treated samples showed very low  CVs.

Slurry samples  (7-day cores)

   Blanks —No analytes were detected in the reagent
   blanks.

   Spiked samples —Matrix and  surrogate spike recoveries
   were within  the  acceptance  limits.   Recovery of toluene
   was above the acceptance limits, but impact on the
   overall results  should not  be significant.
                           92
-------
       Duplicate samples--Duplicate samples  analyzed for VOC
       by Method 8240 were within accuracy acceptance limits.

o   Slurry sample leachates

       Blanks — No metals or PCBs were detected.   Acetone was
       detected in some of the  blanks,  but since this is not a
       field contaminant and the levels were so  low, the
       effect on the results is negligible.

       Spike samples--Metals recoveries for  the  nine samples
       were all acceptable.  Matrix spike and surrogate spike
       recoveries for five VOCs were good although a few
       examples of deviation from the acceptance range were
       noted.  For PCBs the recoveries  were  low, indicating
       detection limits may be  higher than reported.  However,
       no PCBs were detected in the leachates.

       Duplicates--Al1  metals and volatile organics met the
       acceptance criteria on coefficient of variation.

o   Core samples and core sample leachates

       Blanks--Chromiurn, lead,  copper,  and zinc  were detected
       in reagent blanks.  However, review of the field data
       shows that the low levels of the analytes make these
       values insignificant.  Eight VOCs were detected in 1 or
       2 of 10 laboratory reagent blanks; therefore low levels
       of those compounds must  be reviewed with  some suspicion
       since toluene, xylene, and ethyl benzene, three of the
       five measurable components of VOCs, are  included.  PCBs
       were not detected in the blank.   Some phthalates were
       detected, but may be laboratory  contaminants.

       Spiked samples — Ten spiked samples for metals were all
       within acceptance limits of 80-120%.   Eight spiked
       samples for VOC were performed and virtually all
       results were within acceptance limits.  All BNA spikes
       met acceptance limits.

       Duplicate samples--Results for metals and VOC were very
       good, within a CV of 10%.

       BNA analyses — The high results for phthalates levels in
       blanks and low naphthalene in the matrix  spikes may
       provide cause for suspicion.  However, phthalates are a
       common laboratory contaminant and low levels in field
       samples may be an error.  An interlaboratory
       performance audit by Radian also left suspicions on the
       phthalates results.  The spiked  sample results for
       phenols met acceptance criteria.

From the above results for blanks, spikes, and  duplicates, the
chemical analyses should be acceptable.  The primary purpose
of the analyses is to observe changes and orders of magnitude

                               93
-------
of the values before and after the HAZCON treatment.   The
deviations noted above should have an insignificant impact.

The physical  tests,  moisture, bulk density,  and unconfined
compressive strength were performed in triplicate for each
sample collected.  The other physical tests,  permeability,
particles size,  and  pH,  were performed only  once on each
sample.  Permeability is the most important  parameter of the
tests performed  once.  The results showed all  the treated soil
permeabilities were  in the 10"° to 10"9 cm/sec range,
which is very low.  Exact numerical values and differences
between samples  is less  important than the observed
consistency of the order of magnitude results.

The QAPP did not include any protocols for the microstructural
analyses, which  were performed by Scientific  Waste
Strategies.  Since these results are only intended to be
qualitative and  no attempt was made to quantify, the trends
reported should  be valid.
                               94
-------
                           SECTION 8

                     DISCUSSION OF RESULTS
The analytical results summarized in Section 7.0, are
discussed and evaluated in Section 8.1. Operating procedures,
a chronology of which is presented in Section 5.2 for the
HAZCON Demonstration Test, are evaluated in Section 8.2

8.1  ANALYTICAL RESULTS

The analytical data consist of physical test results of
untreated soil, treated soil after a nominal 7-day curing
period, and essentially fully cured samples, which were
analyzed more than 28 days after treatment.  The discussion of
the analytical results can be further subdivided into the
fol1owi ng:

o   Physical tests
o   Chemical analysis - primarily soil composition and
    leachate results
o   Microstructural analyses
o   Overall evaluation

8.1.1  PHYSICAL TESTS

The physical tests on the soil and cores consisted of the
fol1 owing:

o   Free moisture - untreated, 7-day, 28-day
o   Undisturbed bulk density - untreated, 7-day, 28-day
o   Particle size distribution - untreated
o   Permeability - untreated, 7-day, 28-day
o   Unconfined compressive strength - 7-day, 28-day
o   Total organic carbon - untreated
o   Oil and grease - untreated, 28-day
o   pH - untreated soil
o   Wet/dry weathering - 28-day
o   Freeze/thaw weathering - 28-day
o   Unconfined compressive strength after weathering - 28-day

8.1.1.1  BULK DENSITY

The treated soil density was about 10-20% greater on average
than  the undisturbed untreated soils.  The bulk density test
for all samples was performed in triplicate.  In general,
except for the untreated soil in DSA, the individual area
results were in tight bands, with a bulk density greater than
1.4 mg/1 particularly for the treated soils.  For FSA, the
                               95
-------
treated soil  results were 5% lower than the untreated soil
results.  Results of the microstructural  analysis presented in
Section 7.2.3 indicate a very porous structure for all  the  7-
and 28-day cores, which may account in part for the relatively
small  bulk density increases.

The bulk density did not change between 7 and 28 days;  the
change occurred entirely in the initial 7-day period.  Density
decreased with higher oil and grease content, both for  the
untreated and treated soils.  Since the soil  represents only
about  40% of the total weight of the mix of soil, cement,
water, and Chloranan, and the bulk density increase is  10-20%,
the volume of the treated soil more than doubles.
Calculations for volume increase for each plant area is
provided in Appendix D.

The laboratory-prepared solidified formulations on soils from
LAN and FSA,  without the use of Chloranan, showed a lower bulk
density than the field samples, particularly for FSA.  For  FSA
the bulk density was reduced from 1.51 to 1.36 mg/1.  Two
possible contributory causes may be the lack of Chloranan or
the higher moisture content.

8.1.1.2  FREE MOISTURE

The water addition during processing was not tightly
controlled or adjusted for moisture in the feed.  The water
was adjusted by HAZCON based upon visual  observation of the
slump of the concrete mix.  Based upon the material balances
shown in Table 5.2, the total water added per run is the
correct order of magnitude, 40% by wt. of cement.  Comparing
the untreated soil with 7-day treated soil showed an increase
in free moisture content.  Comparing the 7-day material to
28-day showed that, except for DSA, the moisture content
decreased by about 10-15%.  This is probably due to the
continuation of the cement hydration reactions, which would
reduce free moisture  (drying at 60°C).  It appears that the
hydration reactions are 60-80% complete at 7 days, with DSA
and LAS close to 100% complete.

8.1.1.3  PERMEABILITY

The permeabilities of the treated soils were very low,
primarily in the range of 10"8 to 10"9 cm/sec.  This
compares to undisturbed soil permeabilities  of about 10"1 to
10"z  cm/sec, except LAS, which was near 10"5 cm/sec.
Calculations for the  permeability reduction  factors  are
provided in Appendix  D.  The permeability for the laboratory
formulations without  the use of Chloranan were a factor of  10
greater than field cores for LAN and  equivalent to the field
cores for FSA.  The untreated  filter  sludge  was  impermeable,
and the scheduled  tests could  not be  performed.  In  general,
for stabilization/ solidification processes, 10"' cm/sec is


                               96
-------
considered impermeable.  The design of soil  barrier liners for
waste disposal sites target permeabilities of 10"' cm/sec or
less.  The permeability for the two laboratory formulations on
LAN and FSA, without the use of Chloranan, provided results
that were a factor of 10 greater for LAN and equivalent for
FSA to the field cores.  There were not any  differences
between the nominal 7- and 28-day core sample permeabilities.
This is due to the fact that the 7-day as well as the 28-day
permeability tests were run after curing for 30+ days.  The
permeability measurements were very time-consuming, and
difficulties were encountered in performing  the analyses.
Therefore, they could not be performed in the time frame
originally planned.  Tables A-14 and A-23 in Appendix B
provide all the permeability results and the dates on which
the tests were performed.

8.1.1.4  UNCONFINED COMPRESSIVE STRENGTH

The unconfined compressive strength test, which for each
sample was performed in triplicate, produced individual values
that ranged from about 100 psi for FSA to above 2200 psi for
DSA.  Based upon the averages of the three tests, the range
was from 220 psi at FSA to 1570 psi at PFA,  both at 28 days.
In general, compressive strength was markedly lower with
increased oil and grease (O&G).  The average of triplicate
tests at 28 days for the two highest oil and grease areas was
about 220 psi for FSA (25.3% O&G) and 520 psi for LAN (16.5%
O&G).

Decreasing pH with increased O&G is believed to be a result of
the oil and grease concentrations and not a  factor in the
unconfined compressive strengths.  The soil  particle size
distribution may influence core strengths, but insufficient
data exists to confirm.  In addition, with the exception of
PFA, which was a litter coarser, the particle size
distributions were equivalent.

Although it was expected that the 28-day sample tests would
give higher strengths, this was not evident  from the results.
The free water levels discussed above indicate that the cement
hydration reactions were still proceeding, which should have
resulted in increased strengths.  It appears that LAN and PFA
did increase  in strength, but that the others either decreased
or remained unchanged.  An explanation for this unexpected
lack of strength increase is not available.

For the laboratory formulation tests on FSA and LAN, definite
increases in  strength between 7- and 28-day cores of 30-50%
were observed.  These formulations for LAN obtained 7- and
28-day core strengths comparable to the field blocks.
However, for  FSA, the field blocks were many times stronger.
This seems to confirm that the Chloranan helped the
solidification process.  At lower oil and grease levels, the
effect of Chloranan may be less significant.


                               97
-------
8.1.1.5  OIL AND GREASE

Oil  and grease levels ranged from about one percent by weight
for  DSA to 25% for FSA.  Typically the total  organic carbon
(TOC) level  is about 4-5% greater than the level  of O&G,  which
is as expected.   The values for O&G reported  for  the screening
samples are  consistently greater than for the feedstocks  used
in the Program.   This probably can be explained by the method
of obtaining each feedstock for the Demonstration Test,  where
a backhoe was used; some less-contaminated soil was excavated
along with the more-contaminated soils.  For  the  screening
samples,  the most contaminated area in each plant location was
targeted  and only a 40 Ib sample was collected.

8.1.1.6  pH

In general,  the  feedstocks were acidic, except for PFA.
Values as low as a pH = 2.4 were obtained for FSA samples.
There is  some trend toward lower pH values with increased oil
and  grease levels.  However, the original source  of the
contamination would also have an impact,  not  just the
quantity.

8.1.1.7  PARTICLE SIZE

Particle  size distributions on the untreated  soils were  also
measured, except for FSA, which was too sticky for the
screening analyses.  Basically the soils  are  fine, with  about
half the  soil by weight finer than 200 mesh (74 micron).   PFA
was  somewhat coarser than the other soils.

8.1.1.8  WEATHERING

The  two weathering tests, wet/dry and freeze/thaw, are twelve-
cycle tests, measuring weight loss relative to a  control
specimen.  The control specimen is maintained at  72°F in  a
moisture  chamber, when the test specimens are dried or frozen
for  24 hrs.   Both samples then are inserted into  water and
placed in the moisture chamber for 24 hrs.  The results  of
samples from each site location for the wet/dry tests indicate
that the  weight  loss of the test specimens is only slightly
greater than for the control.  However, the loss  differentials
during the freeze/thaw tests appear to be larger.  All weight
losses were  about one percent for both the specimens and
control s.

Unconfined compressive strengths on both  the  test and control
specimens were run.  The results for the  test samples show
that compressive strength was not lost in either the test
specimens or controls.

8.1.1.9  OTHER OBSERVATIONS

Significant  variations in physical properties of the soil

                               98
-------
between each composite within a soil  area were noted.  For
example, oil and grease at LAS ranged from 6.1% to 8.6% and
for LAN from 14% to 18%.  Moisture levels were also somewhat
variable at FSA, LAS,  and LFA.  This  means that individual
grab samples within a  given feedstock are even more variable.
Thus,  physical  test results on untreated soil  based on average
feed properties may not be directly related to the 7-day and
28-day core sample results, which are based more on localized
properties of the solidified soil samples.  Although it is
believed that the overall results are representative for each
plant  area, individual distortions in the data do exist.

In addition, significant variations even within a sample,
split  spoon or  core,  probably exist.   Many of the leaching
tests  (see Appendix D) showed a larger weight of an analyte or
organics group  existed in the leachate than in the solid.
Also,  consistent material balances for the lead and organic
analytes between treated and untreated soil could not be
obtained from the data.  Only for total  oil and grease was
there  a consistency in the concentrations before and after
treatment.  However,  due to the large amount of data
available, definitive  trends in the results exist.  Therefore,
the results obtained  are still valid.

The laboratory  formulations were prepared without Chloranan,
similar to the  method  concrete is typically blended.  For  FSA,
the UCS test values were about one-tenth of those for samples
collected in the field.  In addition, the bulk density was
lower  by almost 10%.   Results for the other laboratory
prepared samples were  equivalent to the  field prepared and
cured  samples.   Thus,  Chloranan improved the physical
properties of a treated soil where the oil and grease levels
were high, about 25%  by wt in the untreated soil.

A grab sample of the  process water showed low suspended and
dissolved solids, with a pH of 8.05.   This water should not
impact the process or  any of the laboratory analyses.

8.1.2   CHEMICAL ANALYSES

The chemical analyses  consist of the  following:

    o     Untreated and treated soil compositions
    o     TCLP leachate analyses
    o     Special leach test analyses

A discussion of the results highlighted  in Section 7.2.2
fol1ows.

8.1.2.1  SOIL COMPOSITION

The composition of the untreated soil was basically as
expected.  The  oil and grease values  were a little low

                               99
-------
compared to the values anticipated based upon the screening
samples and the RI/FS prepared by NUS Corporation for EPA
Region III.  This probably is due to the inclusion of
less-contaminated soil around and below the targeted soil and
to the general  variability within each area.   Samples having
the maximum oil and grease in each of the six plant areas were
targeted for the screening samples.

The PCB concentration levels, with averaged values up to 52
ppm at LAS, were higher than originally anticipated based upon
the RI/FS and screening samples,  which were at less than 25
ppm.  The two Aroclors detected were the same as measured in
the screening samples, but both were at higher concentrations.

The results for the priority pollutant metals were as
expected.  Lead was the major contaminant with concentrations
up to 2.3% by wt. measured.  After reviewing  the soil
analyses, only the major metal contaminants were carried
forward to analyze in the 28-day cores and leachates.  These
were chromium,  nickel, zinc, copper, and cadmium.  In general
the levels of contamination in the soil agreed with those
obtained during the screening tests.

The results for the base neutral/acid extractable (BNAs)
showed primarily phthalates, phenols, and naphthalene.
Phenols differ from other BNAs in that they have greater
solubility in water and therefore, may be more readily
Teachable.  Other BNAs reported in the RI/FS  but not detected
in the screening analyses, such as fluorene,  pyrene, and
f1uoranthene, were only occasionally detected in these
analyses.  The phenols were not checked for on the screening
samples.  Phthalate concentrations in the 28-day core samples
(ND to 2.15 mg/kg) were very low compared to  the untreated
soil (12.15 to 34.2 mg/kg).  This may be caused by
base-catalyzed hydrolysis reactions, which is a reasonable
possibility, or possibly other reactions caused by the
Chioranan.

The volatiles  in the  FSA samples were considerably greater
than anticipated, averaging 150 ppm by wt.  It had not been
expected to observe volatile levels above 30 ppm for any of
the plant areas.  The screening test results showed LAS  to
have the highest concentration of volatiles,  with much less
reported for any of the other plant areas.  For LFA, PFA, and
DSA, toluene was injected into the slurry mix to produce a
final concentration of 125 ppm by wt., based upon the feed
soil.  The maximum value reported in the core samples was 24
ppm in 28-day  cores at LFA, with values down to  1 ppm.   Since
the soil  is only 40%  of the total core weight, this maximum
value is equivalent to 60 ppm on a soil weight basis.
Therefore, the majority of the toluene was lost.  This
indicates  that the toluene either vaporized off  in the HAZCON
mixing screw,  possibly due to poor injection or  improper

                              100
-------
mixing, or during sample preparation in the pulverization
step, or both.  However, sufficient toluene remained to
provide valuable information for leaching test analyses.

Attempts to relate VOC and BNA component concentrations before
and after solidification were not successful.   There was such
variability in the grab samples that in some cases it appeared
that concentrations were greater in the cores  than in the
untreated soil.  However, for oil and grease,  approximate
material balances could be obtained; see Tables 7.1-7.6.

8.1.2.2  TCLP LEACHATE ANALYSES

Oil and grease in the leachates from the untreated soils were
near the detection limit of 0.2 ppm by wt.  in  the range of
<0.2 ppm at DSA to 3.7 ppm at FSA.   These results are lower
for each plant area than for the 7- and 28-day core leachates,
where the results were in the range of 1-10 ppm by wt.  (mostly
2-4 ppm by wt.) .  Since these values are all so close to the
detection limits for the laboratory procedure, it may not be
proper to differentiate between them.  However, it is also
possible that the treatment process tended  to  agglomerate the
oil and grease and after crushing the solid core for
performing the leach test, some O&G globules were released
into the leachate.  The leachate concentrations for the
special leach tests were less than  for TCLP.  Also, the oil
and grease concentrations in the treated cores are about 40%
of these of the untreated soils.

The immobilizing of the priority pollutant  metals was
accomplished by the solidification.  Except for lead and zinc,
virtually all the metals were reduced to their detection
limits in water.  For lead the values were  just above
detection limits, ranging up to 400 ug/1, which is well below
regulatory levels of about 5 ppm by wt.  However, lead  was the
predominant metal contaminant with  measured soil
concentrations ranging from 0.3 to  2.3% by  wt., so the  lead
reduction was dramatic.

Except for phenols, the BNAs (semivolati1e  organics) were
reduced to near their detection limits of 10 ug/1 in the
leachates of both untreated and treated soil samples.  This is
a reduction of more than one thousandfold,  compared to  the
contaminated soil.  However, the phenols were  not reduced to
the same extent.  For LAN and FSA,  with concentrations  in the
soil or cores ranging from 5-400 ppm by wt., the reduction
factor is of the order of 10-100.  For FSA, TCLP leachate
concentrations of 3-4 mg/1 were observed for both treated and
untreated soils.  Therefore, the HAZCON process did not appear
to reduce the Teachability of phenols.  Results of the  MCC-1P
and ANS 16.1 leach tests provided equivalent results.  In
addition, the quantity of phenols in the leachate increased
with time.
                              101
-------
The volatiles in the leachates ranged from 100 to 1000
micrograms per liter (for the higher soil  and core
concentrations).  These are reduction factors of about 100 on
average compared to the soil.  However,  the concentrations in
the untreated soil, 7-, and 28-day core  leachates are
approximately the same.  Therefore,  at least in the
concentration range investigated,  up to  150 ppm by wt. (FSA),
the solidification did not appear  to impact the leachate
results.  The results of the special leach tests showed
leachate concentrations less than  for TCLP leachates, and the
concentrations did not appear to be  time-dependent.

Calculations for total VOCs, toluene, total BNAs, phenols, and
lead showed in many instances, except for  lead, greater
quantities of the analyte(s) in the  leachate than in the
soil.   This indicates that great variability in the
concentration of these analytes may  exist  within a given
sample.  Concentrations in the test  specimens used for the
three  leach tests were possibly much greater than in the
material used for determining concentration levels in the
total  samples.  This could also account  for, in some cases,
the greater migration potential (leaching  potential) of the
organic analytes in the treated soil compared to that in the
untreated soil,  even though average  concentrations of the
analytes in the  cores are less than  one-half those in the
untreated soil.   However, since many leaching analyses have
been performed,  the general observations for organics, that
the leachate concentrations for the  treated soils are about
the same as the  untreated soil, is valid.

The special leach tests, MCC-1P and  ANS  16.1, were performed
on one sample set taken from 28-day  cores  for LAN, FSA, PFA,
and LAS.  In general, for a given  time frame, the results from
MCC-1P were greater than from ANS  16.1,  which may be due to
MCC-1P being performed at 40°C vs. ambient conditions for
ANS 16.1.

The MCC-1P leachate concentrations for VOC and BNAs were
equivalent to TCLP extracts.  For  lead the values were greater
and increased with leach time.  For  oil  and grease, the TCLP,
MCC-1P, and ANS  16.1 leachate concentrations appear equal.

A brief comparison test between EP Tox and TCLP for metals was
performed on samples from LAN and  FSA.  The results for EP Tox
were greater for LAN and about the same  for FSA.  This is
expected for basic solutions  (these  leachates are basic due to
the cement components).

8.1.2.3  SPECIAL LEACH TEST ANALYSES

Leach tests ANS  16.1 and MCC-1P were two procedures originally
developed for the  nuclear  industry for the leaching of low
level  radioactive  wastes, but have been  applied with
modifications to hazardous wastes.  These tests utilized solid

                              102
-------
cores simulating the solidified wastes,  as  compared to TCLP
and EP Tox where the solid cores are first  ground to a
powder.  MCC-1P simulates a relatively static groundwater
flow, with the samples in contact with one  leachate for time
periods of 3,  7, 14, and 28 days at a temperature of 40°C.
ANS 16.1 simulates a moving groundwater  regime with the solid
core specimen  placed in a new fresh leachate, so that the
boundary concentrations of the analytes  are kept below
saturation level.   Samples are collected after 1, 3, 7, 14
days with the  total  leach time being 28  days. A diffusion
coefficient might  then be calculated.

It should be noted that experience with  these tests on
hazardous wastes is  limited.  Comparisons to the treated soil
TCLP results have  been made, but the significance of the
differences is unclear.  Some differences would be expected
due to the diverse ratios of solid to extract used in each
test.  The nature  of these two procedures,  that of simulating
leaching from  a solidified mass, makes it illogical to perform
these tests on untreated soils.  Therefore, the results of
these tests are compared to the treated  soil TCLP tests only.

It was anticipated that TCLP would be the most severe leaching
test, providing the  highest contaminant  concentrations in the
extracts.  However,  the extract concentrations for the metals
for MCC-1P were greater than for TCLP, 0.3-0.7 mg/1 versus
0.01-0.06 mg/1, and  approximately equivalent for VOCs, BNAs,
and oil and grease.   This indicates that at least for the
28-day cores from  the HAZCON process the increased surface
area for leaching  for the TCLP test is balanced by the
increased time and temperature of MCC-1P.  The differences
between MCC-1P and ANS 16.1 extracts may be due to the higher
leaching temperature, 40°C versus 22°C.

Since the leaching times range from 1 to 28 days in these
tests, the effect  of time was reviewed.   For the organics,
time did not appear  to be related to leachate concentrations
except for phenols in the MCC-1P test.  Since phenols are
soluble in water,  this is not unexpected.  The concentration
of lead appears to increase with time for both leach tests,
although the trend for MCC-1P is less definitive.

8.1.3  MICROSTRUCTURAL STUDIES

Microstructural and  microchemical analyses  are proven methods
for understanding  the mechanisms of structural degradation  in
materials similar  to those in this Demonstration Test.  The
literature is  replete with examples of SEN  and XRD analyses of
soils, cement, soil-cement mixtures, and each of these mixed
with various inorganic and organic compounds.  However, there
have been relatively few studies of the  microstructure of
complex waste/soil mixtures like those resulting from
stabilization/solidification procedures.  Consequently, in

                              103
-------
some cases interpretation of microstructural  observations may
be difficult.  The microstructual  report is intended to be
complete in its reporting,  yet conservative in its
conclusions.   Many observations will  become more useful as the
understanding of stabilization/solidification technologies
improve.

A discussion  of the results presented in Section 7.2.3
fol1ows:

o   The two predominant clays seen in the samples, illite and
    kaolinite, are nonexpandable clays.   Therefore,  the
    organic compounds in the soil  will  not be adsorbed within
    the layers of the clay minerals.   Any organics adsorbed
    will be on the outside surface and  therefore will  be more
    loosely held.

o   The crystalline phases seen in the  core samples,
    portlandite, ettringite, calcium silicates,  calcium
    aluminate, and gypsum,  are the principal  components of
    Portland  cement.  Portlandite  and ettringite are the major
    hydration phases of portland cement.  Thus,  the  major
    bonding agent is portland cement, and SEM observations
    agree with this conclusion.

o   An abundance of pores of all sizes  and shapes were
    observed  in the core samples.   Some  of the pores were due
    to trapped air bubbles.  The high concentration  of the
    pores may be due to incomplete mixing.  Crystals of
    portlandite, calcium aluminate hydrate, and  ettringite
    grew into these pores.   These  crystals are thus  more
    easily accessible to percolating water.

o   The quantities of unhydrated tricalcium silicate and
    dicalcium silicate were much greater than usually observed
    in hydrated portland cement with moderate water-to-cement
    ratios.  The water-to-cement ratio  of 0.4, typical of
    cement, is used in the processing operation.  The high
    concentration of unhydrated silicates could  result from
    inefficient mixing, resulting  in insufficient water
    transported to these grains.

o   Mixing does not appear to be highly efficient.  Among the
    facts indicating this are:  1) the  brownish  aggregates do
    not undergo disaggregation 2)  there  are significant
    amounts of unhydrated Portland cement clinker in the
    samples,  3) there are globules of dark colored material  in
    the core samples visible to the naked eye, and 4) there
    are many pores, including air bubbles.

o   Two factors suggest that soil  components passed through
    the process unchanged.  They are the presence in the cores
    of apparently unaltered brownish aggregates  and peaks in


                               104
-------
    the x-ray diffraction pattern for both the soil  and the
    cores, which could not be identified with any expected
    mineral constituents of the soil.

In summary, some of the waste in the soil appears to pass
through the process unaltered, and thus, it appears  that
encapsulation is a major part of the mechanism of
solidification/stabilization.  In addition, the mixing
operation is not highly efficient, resulting in incomplete
hydration of the cement and high porosity.

8.2  OPERATIONS

For the 5-cu-yd test runs, HAZCON utilized four people to
operate the unit.  Their functions were as follows:

    o    1 man - add feed soil and water to feed hopper
    o    2 men - move and watch slurry mixer, smooth surface
         of blocks, and control Chloranan flow
    o    1 man - at the control panel and supervise  operations

In addition, EPA provided support services to excavate the
feedstock, screen it, and bring it to the HAZCON MFU.  The
soil was brought to the unit in a 2 1/4 cu yd (soil  level to
surface of bucket) front-end loader, which fed it very slowly
to the feed hopper at approximately the rate the MFU processed
it.  Therefore, at least two support people were required full
time to provide feed to the HAZCON unit.  Also a cement supply
truck with operator/driver was on hand full time.  For the
25-cu-yd run, three additional people were required  to operate
the cement pump and control the feed to the large pits.

Some operating difficulties were encountered in all  of the
runs, causing down times on the order of one minute  to two
hours.  The shutdown periods related to momentary screw
pluggages due to lack of water, oversized feedstock  (stones
greater than 3 inches), interruptions of Chloranan feed, or
emptying of the cement feed hopper, which required refilling
from a supply truck.  For two runs, DSA and PFA, only four
blocks were prepared due to a combination of lack of daylight,
need for more feed, and operating difficulties.  On  two
occasions the soil feed screw jammed so badly that the MFU had
to be brought to the decontamination area for a complete clean
out.  Direct addition of water by hose to the feed screw
facilitated the movement of the soil and reduced the
likelihood of jamming.  This was a water feed stream that was
not measured directly by the in-line flowmeter.  Only
approximate consumptions could be provided.

In generally observing the operations, the consistency of the
slurry was quite variable due to variations in water
feedrate.  The consistency of the mixed product varied from a
very thin slurry to almost a dry powder.  In addition, the

                              105
-------
variability of the slurry mix may have prevented the Chloranan
from achieving its fully claimed benefit in counteracting the
presence of organics.  As indicated in Section 8.1.3,  if
insufficient water is provided,  components of the cement, such
as the tricalcium silicate was incompletely hydrate, as
observed in the x-ray diffraction analyses.  This will  result
in a weaker and less durable block, as compared to a block
from fully mixed components.

A review of the material balances shows that for the 5-cu-yd
runs, the total mass of the blocks was consistent with  the
calibrated rates for cement and  soil  adjusting for outage
time.  Material balances also indicate the following:

o   The soil was only 40% of the total weight of the blocks.
o   The cement-to-feed ratio was approximately 1:1.
o   The Chioranan-to-feed ratio  on a  weight basis ranged from
    0.05 to 0.09, which is less  than  the target value  of 0.1.
    The lowest value was for FSA.  No apparent reduction in
    physical properties were observed.
o   The water-to-feed ratio was  approximately 0.4 on a  weight
    basis, which is appropriate  for concrete preparation.
o   The volume of undisturbed feedstock to produce the  5 cu yd
    of slurry to fill the five 1-cu-yd molds was approximately
    2 cu yd.  (The front-end loader held 2 1/4 cu yd,  filled
    level to the top of the bucket, which produced 4 to 5
    blocks.)  The approximate bulk density of the screened
    soil after transporting to the unit was about 20-30% less
    than in the undisturbed samples analyzed and reported in
    Tables 7.1-7.6.

For the extended length run at LAS, approximately 22-24 cu yd
of slurry was produced to fill the three 1-cu-yd molds  and the
two large excavations.  The excavations were 8x16 ft and were
filled to a depth of 2-2 1/4 ft.  As  shown in Table 5.1,
approximately 52,300 Ibs of the  total feed (soil + cement +
Chloranan + water) was used.  However, based upon the  bulk
density of solidified soil of 1.7 g/ml (106 lbs/ft3),  for
the total weight processed only  about 18 cu yd of slurry would
have been produced.  Therefore,  it appears that 10,000-15,000
Ib more total feed material was  processed.  Since the  cement
and water rates are constantly measured and the soil is based
upon the initial calibration, it is quite likely that  the
missing mass, compared to the HAZCON  Monitoring Worksheet, is
contaminated soil.  Six front-end loader loads of soil  were
processed, which is approximately 13-14 cu yd.  If the
screened feed soil had a bulk density of 1.2-1.3 g/ml  (75-80
lb/ftj), this would indicate a total  soil usage of about
30,000 Ib, which is over 10,000  Ib greater than that recorded
on the HAZCON Monitoring Worksheet, thus providing sufficient
feed to fill the pits.

Also shown on the Monitoring Worksheet, two-thirds of the
cement was used before starting  the slurry flow to the second

                              106
-------
large pit; this value should have been only slightly more than
50%.  This ratio tended to be confirmed by the fact that 130
of a total of 255 gallons of Chloranan was used only 10
minutes after the second large pit was started.

Based upon this reasoning, the overall ratio of cement-to-soil
was about 2:3 and probably even lower for the second large
pit.  In addition, the Chioranan-to-soi1  feed rate is about
0.08 versus that listed in Table 5.2, 0.119, which is based
upon the HAZCON Monitoring Worksheet.  It does not appear that
any loss in physical  properties resulted  from this apparent
reduction in cement feed.  (See Appendix  B, Table A-14.)

8.3  MEETING OF SITE PROGRAM OBJECTIVES

Information relating to each of the program objectives in
Section 1.2 was obtained.  This information can be summarized
as follows:

o   Immobilization of site contaminants — The priority
    pollutant metals were very satisfactorily immobilized,
    with leachate concentrations reduced  to below 0.1 ug/1 for
    soils containing metals up to 23,000  mg/kg.  Untreated
    soil extract concentrations were typically 20 to 50 mg/1.
    Immobilization of organics, VOCs, BNAs, and PCBs was not
    observed.  TCLP leachate concentrations for organics in
    untreated and treated soils were approximately equal.
    This occurred even though, on average, the treated soil
    composition is lower in contaminants, due to the addition
    of cement, water, and Chloranan.

o   Technology effectiveness as a function of oil and grease
    concentrations—The soil samples taken were in the range
    of 1-25% by wt. oil and grease (O&G).  Higher O&G soils
    tended to have more VOCs and BNAs.  Greater VOCs and
    phenols (BNA components) led to higher leachate
    concentrations.  The most direct impact of higher O&G is
    reduced unconfined compressive strength.  In addition,
    FSA, which had the greatest O&G content, appeared to have
    a slightly greater permeability than  the other test
    samples.

o   System performance—The operational performance of the
    system was erratic, with many short and long shutdowns.
    The system shutdowns ranged from 1-2  minutes up to 2-3
    hours.  Many shutdowns occurred for each soil feedstock.
    The slurry consistency was quite variable, ranging from a
    powdery mix to a very thin slurry.  However, for the
    extended duration run, operation outages were less and the
    consistency of the product slurry improved.  The impact of
    consistency on physical or chemical properties of the
    solidified blocks was not observed.
                              107
-------
Long-term integrity--Measurements  of long-term integrity
of the solidified  masses cannot be directly performed.
However,  indications  of potential  difficulties could be
inferred  from the  microstructural  analyses.  These
observations showed a porous structure,  with incomplete
cement hydration and  organic globules existing in the
solidified cores.

Remediation costs--Remediation costs were prepared based
upon using HAZCON's MFU and the operating conditions used
for the Demonstration Test.  The results showed that the
cleanup cost was $205 per ton of contaminated soil under
the ground rules defined.
                           108
-------
                           SECTION 9

                           ECONOMICS
9.1   INTRODUCTION

A cost analysis addresses two main categories,  capital  costs
and operating and maintenance costs.

Capital  costs include both depreciable and nondepreciable cost
elements.   Depreciable costs include  direct costs for site
development, capital equipment,  and equipment installation as
well as  indirect costs for engineering services prior to unit
construction, such as feasibility studies and consultant
costs, administrative tasks such as permitting, construction
overhead and fee, and contingencies.   Nondepreciable costs
include  start-up costs including operator training,  trial or
test run expenses, and working capital.   Operating and
maintenance costs include variable, semivariable and fixed
cost elements.   Variable operating cost  elements include raw
materials,  utilities, and residual water disposal costs.
Semivariable costs include unit  labor and maintenance costs,
and laboratory  analyses.  Fixed  costs include depreciation,
insurance,  and  taxes.

The above  cost  element breakdown, however, is based  upon a
permanently sited hazardous waste cleanup device.  The  HAZCON
MFU as employed at the Douglassvi11e  Superfund  site  is  a
transportable unit that will not be installed at a fixed
site.  Thus, it assumes some different cost elements that will
impact on  a cost analysis from the more  frequently encountered
permanent  installations.

In general, the cost for a transportable hazardous waste
remediation facility falls into  three categories:  capital
costs including all  costs that can be amortized over the
service  life of the  unit; mobilization/demobilization costs
associated  with start-up and shutdown at a given site,  that
can be amortized over the duration at the site; and  operating
cost to  operate and  maintain the system.  Capital costs can be
subdivided  into direct, indirect, and nondepreciable cost
elements.   Mobilization/demobilization costs can be  accrued as
semivariable operating and maintenance costs.  Operating costs
include  variable utility and raw material costs, semivariable
labor and  maintenance costs, and fixed costs such as
depreciation, insurance, and taxes.

In addition, for a mobile unit,  several  capital cost elements
defined  for the permanently sited unit should be redefined
into a different cost element category.   These  include  the
direct costs for site development and the direct costs  for

                              109
-------
engineering studies,  which on a site-specific basis become
mobilization/demobilization costs.   These factors are not
included here because of the complexity of the analysis and
planning in this area.   Total site  cleanup is the
responsibility of others,  with the  HAZCON technology used for
only part of the total  remediation.

Based on the above,  an  overall cost  element breakdown, as
illustrated in Table  9.1,  can be developed.

9.2  COST ELEMENTS

A detailed discussion of each of the cost elements defined in
Table 9.1 is provided in the following subsections.  Since
this cost analysis is being prepared based upon the
Demonstration Test,  the cost to process each ton of
contaminated soil will  be  distorted  to a greater value.  Cost
projections for a commercial installation will be included in
the Applications Analysis  Report.   In addition, not every
expense that might be encountered  in a site cleanup is
included.  Items such as permitting  and site preparation were
omitted due to their  complexity to  predict costs.

9.2.1  CAPITAL COSTS

9.2.1.1  DIRECT COSTS

Equipment Fabrication/Construction  and/or Purchase --

The costs for the design,  engineering, materials and equipment
procurement, fabrication and installation of the HAZCON MFU
(including vehicle),  are included  as direct costs.  The costs
include all the subsystems and components installed but do not
include the costs of the vehicles  for the transport of the
accessory equipment,  described in  Section 3.2.  Waste
preparation equipment is not included as it can be assumed to
be rented or provided by the site-responsible party.
Pretreatment or posttreatment of the soil is assumed not to be
required.  For the items calculated  as a fraction of direct
costs, the total of the HAZCON MFU  and associated tanks,
pumps, etc., is the capital cost value used.

The capital cost value of the MFU was provided by HAZCON and
is $100,000.  It is assumed that the equipment has a 10 yr
life.  In addition, storage bins or tanks for cement,
Chloranan, and fuel are required,  as well as an air blower for
transferring cement,  pumps for the  liquid, and associated
piping and controls.   This was assumed to be an $80,000 cost
to the project, whether the equipment was purchased new or
used or it was sold or discarded at project completion.  These
factors are site specific and major variations could occur
between sites.
                              110
-------
9.2.1.2  INDIRECT COSTS

Administrative/Permitting --

Administrative costs associated with regulatory compliance
issues could be numerous and varied.  The costs that are being
accrued under this cost element are directed to the overall
non-site-related regulatory activities in establishing federal
and state permit requirements,  preparing initial  permit
applications, and supporting permit application information
throughout the permit issuance  process.   Once the final
permits are issued, recordkeeping,  inspection,  survey response
to permitting agencies, and additional reporting  activities
may be required.  These costs include the preparation of
technical support data, sampling/analysis,  and  quality
assurance project plans by in-house engineering personnel,
RCRA/TSCA permit forms (if applicable);  time, travel, and per
diem for consultant and in-house staff interfacing with
Federal EPA officials; and in-house administrative and
clerical staff.  For the cost estimates  developed in this
analysis, these factors are not included.

For this cost analysis, administration costs are  taken as 10%
of direct costs (HAZCON MFU and tanks, pumps, etc.) on an
annual basis.  It includes office expenses  such as supplies,
telephones, reproduction equipment, and  furniture, but not
salaries (included elsewhere).

Contingency --

A contingency cost, approximating 10% of the direct capital
cost on an annual basis, is allowed for  unforeseen or poorly
defined cost definitions.  For  the  HAZCON process this is a
minor factor.

9.2.1.3   NONDEPRECIABLE COSTS

Operations Procedures/Training  --

In order to ensure the safe, economical, and efficient
operation of the unit, the creation of operating  procedures
and a program to train operators is necessary.   Costs that may
accrue include:  preparation of a unit health/safety and
operating manual, development and implementation  of an
operator training program, equipment decontamination
procedures, and reporting procedures.  These documents must be
site-specific.  They can be related to basic documents, the
preparation costs for which can be  amortized over the life of
the equipment.
                              Ill
-------
               TABLE  9.1.   COST  ELEMENT  BREAKDOWN
CAPITAL COST

  Direct

     o  Equipment Fabrication/Construction or Purchase

  Indirect

     o  Administrative/Permitting
     o  Contingency

  Nondepreciable

     o  Operations Procedures/Training
     o  Initial Start-up/Shakedown
     o  Working Capital

OPERATING & MAINTENANCE COSTS

  Variable

     o  Raw Materials - Cement
     o  Fuel
     o  Power
     o  Water
     o  Chemicals - Chloranan

  Semi vari able

     o  Labor
     o  Maintenance
     o  Equipment Rentals and Consumables
     o  Analytical Services
     o  Mobilization/Demobilization

           Site Preparation/Logistics
           Transportation/Setup
           On-Site Checkout
           Working Capital
           Decontamination/Demobilization

  Fixed

     o  Depreciation
     o  Insurance
     o  Taxes
                              112
-------
It was assumed that three days of training would be required
for all  HAZCON personnel, site support personnel,  health and
safety officer,  and sampling technician.   The cost includes
salaries and living expenses.   These are  the only  costs
included in this category.

Initial  Start-up/Shakedown  --

After the unit is brought to a site it must be initially
started  and  operated to check out the mechanical  and
technical integrity of the  equipment and  its controls.   This
cost is  assumed as part of  site mobilization.

Working  Capital  --

Although the unit is a transportable system, it will  require a
supply of maintenance materials attributable to a
nondepreciable capital cost.  Maintenance materials account
for approximately one-half  of  the total  maintenance cost,  and
3-month  inventories are usually maintained.  This  cost  as  used
in Table 9.2 is assumed as  10% of maintenance costs.

9.2.2  OPERATING & MAINTENANCE COSTS

9.2.2.1   VARIABLE COSTS

Variable operating cost elements for this unit include  fuel,
power, water, chemicals, and process waste disposal.   They are
defined  as variable operating  cost elements because they can
usually  be expressed in terms  of dol1ars-per-unit  flow  of  soil
treated  and as such, these  costs are more or less  proportional
to overall facility utilization during specific site
operations.  It is also assumed for the tabulation of costs
that there are no process waste byproducts.

Fuel --

The fuel requirement for the unit includes diesel  fuel  to
power the vehicle part of the  MFU.  In addition, fuel is used
for supporting vehicles - backhoe, front-end loader,  etc.   It
is estimated by HAZCON that their equipment will consume 4
gph.  It is assumed that the supporting equipment  consumes an
additional 4 gph.  The cost of diesel  fuel is about $1.00/gal.

Power --

The power requirement for the  unit includes the electrical
requirements for the trailers, sampling equipment, auxiliary
lighting, etc.  It is assumed  that the daily average  power
consumption is 5 kw with the cost of electricity $0.04/kwhr.

Water --

Water use is based upon the water content of the feedstock to

                              113
-------
bring the cement-like final  slurry to about 20% by wt.  water.
In addition,  some water is used for decontamination.   The
total average water usage rate is 14 gpm at a cost of $0.80
per 1000 gal Ions.

Chemi cals - -

The HAZCON proprietary additive is Chloranan, which inhibits
the effects organics have on the solidification of cement.  It
is utilized in a ratio to the contaminated soil of 1:10.
HAZCON indicates the delivered cost for cement is $50/ton and
Chloranan is  $3.00/gal (Chloranan density is 9.0 Ibs/gal).

Decontamination Water --

If the unit is not operated  24 hours per day, the unit needs
to be cleaned with high pressure water or steam to prevent
plugging.  Costs will accrue for the containment and  disposal
of this waste stream.  An additional 10% water consumption was
assumed.

9.2.2.2  SEMIVARIABLE COSTS

Labor --

Operating costs for personnel for the HAZCON unit is  based
upon 3 shifts per day, 21 shifts per week and totals  17
people; this  includes 12  process operators, 4 supervisors, and
1 overall coordinator.  In addition, there are support
personnel for operating the  contaminated soil moving
equipment, a  site safety  and health office, the project
manager, office personnel, and a part-time sampling
technician.  This totals  21  people.

The labor and living expenses for the HAZCON personnel  was
provided by HAZCON.  These costs range form $17.50/hr to
$50/hr for salaries with  overhead and $85/day for expenses.
The support personnel costs  were based upon actual costs
Incurred by EPA/Env1response for the Demonstration Test.
Salaries with overhead range from $20 to $60/hr with  living
expenses (except local hires) at $120/day.

Maintenance --

Maintenance materials and labor costs are extremely difficult
to estimate and cannot be predicted as functions of a few
simple waste  and facility design characteristics because  a
myriad of site-specific factors can dramatically affect
maintenance requirements.  Annual maintenance cost will be
assumed as 10% of capital cost.

Analyses --

In order to ensure that the unit is operating efficiently  and

                              114
-------
meeting environmental standards, a program for continuously
analyzing waste feed and treated solids is required.
Initially sample sets will be taken daily, and less often as
operation efficiency improves, approximately once per week.
based upon the expected need to perform many of the laboratory
tests described for the Demonstration Test, the cost of a
sample set is estimated to be $5000.

Mobilization/Demobilization --

As discussed in Section 9.1, the following costs will accrue
to the HAZCON unit at each specific site.   The costs are site-
specific and may vary widely depending on  the nature and
location of the site.  They include site preparation/
logistics, transportation/set-up,  construction supervision,
on-site checkout, site-specific permitting/engineering
services, working capital, and decontamination/
demobilization.  Site preparation  is assumed to be by others
and no costs are included.  The other costs listed above are
included elsewhere.  It is assumed that mobilization is three
days of salaries plus living expenses for  all personnel.

Site Preparation/Logistics -- The  costs associated with site
preparation/logistics include advanced planning and
management, detailed site design and development, auxiliary
and temporary equipment and facilities, water conditioning,
emergency and safety equipment, and site staff support.  Soil
excavation, feedstock preparation, and feed handling costs are
also included.  This may be performed by other than HAZCON,
but still comprises part of the site remediation costs.  Due
to the temporary and transient nature of the setup at
Douglassvi11e, the costs incurred  by EPA for the test are not
directly used because they would be misleading.  Costs for
advanced planning, detailed site design and development, and
water conditioning if needed, are  assumed  to be part of the
site prime contractor's expenses,  and are  not included.

Transportation/Set-Up --  The cost of transportation and
set-up includes disassembly of the unit at its presen*
location and transport to a new location.   The HAZCON unit is
integral with the vehicle automotive function.  Auxiliary
process equipment is transported on separate flatbed trucks.
The costs for transporting the unit to the Douglassvi11e site
are not included.

On-site Checkout --  Once the unit has been set up, it is
necessary to shakedown the system  to ensure that no damage
occurred as a result of disassembly, transport, and
reassembly.  This cost is shown as initial startup and is
assumed to be 10% of direct costs  calculated on an annual
basis.
                              115
-------
Working Capital  -- Fuel  inventory,  Chloranan,  and cement
storage facilities will  exist at each site,  and as such are
semivariable operating costs specific to the site-specific
mobilization/demobilization cost element breakdown.   It is
assumed that all these raw material  items will  be fully
consumed and therefore,  no additional charge to the  project is
added.  The storage facilities for  these materials are
included as part of the  direct capital  costs.

Decentamination/Demobilization -- With  the completion of
activities at a specific site, the  unit must be decontaminated
and demobilized before  being transported to its next
location.  Costs that will accrue to this cost  element include
field labor and supervision, decontamination equipment and
materials, utilities, security, health/safety activities, and
site staff support.  The demobilization costs included are
based upon three days for all personnel.

9.3  OVERALL COST EVALUATION

A primary purpose of this economic  analysis is  to estimate
costs for a commercial-size remediation.  It was assumed for
this analyses that part  of the Douglassvi11e site would be
remediated.  Due to the  short-term  nature of the Demonstration
Test and the fact that labor and chemical costs dominate the
remediation costs, actual costs for the test were not directly
used.  However, since HAZCON used a small-scale, continuous,
commercial unit, the capacity, on-stream factor, and chemical
usage during the Demonstration Test was the starting basis  for
a commercial cleanup estimate.  The results of the analysis
are presented in Table 9.2.

The results of  the analysis show that the cost per ton of soil
processed is $206.   In comparison to the costs of a future
commercial unit, which would be larger, have an improved
on-stream factor and probably use less chemicals, the cost  per
ton of soil processed is very high.  The lower values can be
obtained based  upon  HAZCON's recommendations for reducing
chemical consumptions 25-50% for attaining an on-stream  factor
of 90%,  and for use  of a new 2300 Ib/min batch processing unit
and might reduce these costs by 50%.  This type of analysis
will be  included in  the Applications Analysis Report.  Since  a
70% on-stream factor with  high chemical consumption was
actually seen at Douglassvi11e, PA,  the costs for this level
of operating efficiency were calculated.  It should be noted
that not all the expenses  encountered in a site cleanup  are
included, such  as  site preparation  and  permitting.

As can be seen  from  the results, 90% of the costs consist of
raw materials  (cement and  Chloranan) and labor.
                               116
-------
                   TABLE  9.2.   ESTIMATED  COST
CAPITAL COST

     Pi rect
       Equipment Costs, $                       100,000
       Chemical Storage, $/ton                     2.25

     Indirect.  $/ton
       Administration (10% Direct Costs)           0.32
       Contingency (10% Direct Costs)              0.32

     Nondepreciable. $/ton
       Operator Training (3 days)                  0.84
       Working  Capital

OPERATING AND MAINTENANCE COSTS

     Variable,  $/ton
       Cement                                     50.00
       Chloranan                                  66.67
       Fuel ($1.00/gal-diesel)                     1.29
       Electricity ($0.04/kwhr)                    0.03
       Water ($0.80/1000 gal)                      0.08

     Semivariable. $/ton
       Salaries and Living Expenses               64.70
       Equipment Rentals and Consumables          10.36
       Analytical Services                         6.50
       Maintenance (10% Direct Costs)              0.32
       Mobilization/Demobilization                 1.66

     HAZCON and Support Fixed
       Site Preparations
       Insurance and Taxes (10% Direct Costs)      0.32
       Depreciation (10 yrs.)                      0.18

       TOTALS,  $/ton                             205.84
                              117
-------
                       TABLE 9.2. NOTES

1.   The demonstrated HAZCON MFU capacity is 300 Ib/min.

2.   It is assumed that 35,400 tons  of soil  at the
    Doug!assvi11e site was processed.  This value derives from
    continuously  operating the  demonstrated unit for 6 months
    at an on-stream factor of 90%.   The actual  on-stream
    factor used  was 70%.

3.   Utilities Consumption Estimates
    1,200 max installed KVA
    Water - 14  gpm
    Diesel Fuel  - 8 gph

4.   Chemical  Consumption
    Cement    -  1:1 to contaminated  soil
    Chloranan -  1:10 to contaminated soil

5.   Labor Estimate - 4 shifts per week - includes overhead
    HAZCON -  1  Manager, $50/hr; 4 Shift Supervisors, $35/hr;
               12 Operators, $17.50/hour;  10% OT
    Support Personnel  - 1 Manager,  $50/hr;  4 Shift
                        Supervisors, $45/hr; 12 Operators,
                        $40/hr; 10% OT
    Others -  1  SSHO, $50/hr; 1  Project Manager, $60/hr;
             Laboratory Technician  (part-time), $40/hr;  Office
             Manager,  $40/hr; Secretary, $20/hr
    Living Expenses (motel, food, car rentals,  etc.) -
             HAZCON, $85/D; Support Personnel - Managers and
             Supervisors  $120/D,  Operator  - local - no charge;
             Others -  $120/0

6.   Rental and  Consumable Supplies
    Health and  Safety consumables and instruments - $450/D

         Office  space, office supplies, portable sanitary
         faci1i ti es -  $50/day
         Sampling Materials - $40/set
         Heavy  Equipment  Rental - front-end loaders, backhoes,
         steam  cleaner, drill rig,  etc. -  $1000/day

7.   Site preparation costs, since it would  be so interrelated
    with the  overall planning and costs of  the Prime
    Contractor  for the entire site,  are not included.

8.   The costs taken as a  fraction of capital and maintenance
    costs are prorated to the actual time  on site.

9.   Operator  training assumes 3 days of training for HAZCON
    operators,  site preparation operators,  the Health and
    Safety Officer, and sample technician.
                              118
-------
                           REFERENCES
1.  Remedial Investigation  Report/Feasibility Study of
    Alternatives, Berks Associates  -  Douglassvi1le Disposal
    Site - Union Township,  Berks  County,  PA., EPA Work
    Assignment No. 59-3651,  Contract  No.  68-01-6699, June
    1986.  Prepared by NUS  Corporation  for  Region III.

2.  Letter Report, S. Sawyer  to  P.  de Percin, Laboratory
    Analyses of Screening Samples,  July 6,  1987.

3.  Demonstration Plan - HAZCON/Douglassvi11e SITE Program,
    August 31, 1987, EERU Contract  No.  68-32-3255.  Submitted
    to P. de Percin, EPA HWERL,  Cincinnati,  Ohio, August 31,
    1987.
                             «U.S. GOVERNMENT PRINTING OFFICE:! 9 89 -6 its-16 3^7066
                              119
-------