it was believed that DCR treatment of the stock-
DCR treatability testing
piled soil could eliminate most, if not all, of the
Following these initial sample characteriza-
tions, 100-g aliquots of soil-amended Basin F
organic constituents that are currently leaching
sludge from above the liner (sample 24216) were
into the water being collected in the primary and
removed for treatment with various quantities of
secondary sumps at the site. However, stabiliza-
natural and hydrophobized CaO. For these
tion of the higher-molecular-weight chlorinated
scoping studies, the samples were mixed by hand
pesticides might be improved by elevated con-
in a laboratory fume hood. No attempt was made
centrations of liquid-phase organics. Therefore,
to quantify any vapor emissions, because the third-
DCR treatability tests were undertaken with and
party laboratory GC/MS analyses scheduled for
without the addition of a benign food oil additive
the samples would allow quantification of vola-
[Archer Daniels Midland (ADM) Regular Soy Oil
tile constituent removal.
Soap Stock]. It was determined in a wettability
After several iterations, it was determined that
study that addition of the soy oil significantly
the high levels of sulfate (and possibly urea) in
improved the water repellency of the resultant
the waste (Table 2) were interfering with the CaO
product, so its addition was implemented as a
hydration inherent to the DCR process. As a re-
standard procedure for all further tests. Under
sult, uneconomically high levels of CaO reagent
the limited time constraints of this project, how-
were required to obtain an acceptable product.
ever, no attempt was made to optimize the quan-
To circumvent this problem, additional tests were
tity (10% of soil weight) or type of food oil addi-
undertaken to see if the waste could be pretreated
tive, and this remains a study area that should be
with lime milk [hydrated Ca(OH)2] to tie up the
pursued.
sulfate (for example, as gypsum, CaSO4 2H2O)
and also remove the urea. These tests were very
Independent laboratory testing
successful, and it was found that addition of 30%
Following the initial scoping studies completed
lime milk slurry in H2O [Ca(OH)2 at approxi-
in the first week after the sample was received,
mately 1013% of contaminated soil weight] was
0.7 kg-scale batch preparations were undertaken
sufficient to tie up all the sulfate as gypsum. Then
to generate sufficient DCR-treated material for
independent, third-party laboratory testing. Un-
with the subsequent addition of dry CaO for the
der the extremely short time-frame required to
DCR reaction, it proceeded without delay and at
obtain initial laboratory results by 1 April 1995,
economically feasible reagent addition levels. The
little or no additional reagent optimization test-
lime milk pretreatment also proved to be an ex-
cellent way to remove ammonia by freeing up
ing was possible, and the initial DCR product
any ammonium originally present as ammonium
prepared for outside laboratory analyses was
generated from 55% CaO addition to lime-milk
sulfate and catalyzing the hydrolysis of urea to
pretreated soil plus 10% ADM soy oil. The total
CO2 and ammonia. In field applications, this ap-
CaO reagent requirement (including that used to
proach could be used to release and reclaim am-
generate the lime milk slurry) for this sample was
monia through a specially designed emissions con-
trol capture system on the transportable treatment
64.7%. This material and a sample of the raw
unit (TTU) prior to DCR processing.
untreated waste soil (sample 24216) were pack-
The DCR process (unlike other pozzolan-based
aged for shipment and transferred under chain of
stabilization approaches) achieves its greatest tech-
custody to the U.S. Army Corps of Engineers En-
nological advantage with wastes containing very
vironmental Laboratory in Hubbardston, Mass-
achusetts, on 20 March 1995.
high (>1030%) levels of organics (e.g., diesel oil
The following analyses were requested with a
residuals and waste oils) because such materials
one-week turnaround time for the results:
facilitate transfer of the contaminants of concern
onto the hydrophobic CaO reagent before the dis-
FID GC (Diesel Range Total Petroleum Hydro-
persion by chemical reaction takes place. From
carbon Screening, EPA Method 8015M)
the data presented in the previous section, it was
ECD GC (EPA Method 8081 Pesticide and PCB
apparent that there would not be sufficient free
Analyses)
organic (oily) phase in the sludge/soil matrix for
optimum DCR processing. Based on the Basin F
GC/MS (EPA Method 8270 Total Waste
waste pile leachate data provided by GeoTrans*
Analyses)
GC/MS (EPA Method 8260 Total Volatile Or-
ganic Analyses)
* K. Swingle FAX to J. Payne, 12/29/94.
7
Previous Page
