AEPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development EPA-600/D-84-036 Mar. 1984
ENVIRONMENTAL
RESEARCH BRIEF
Oil Shale—Potential Environmental Impacts and Control Technology
E. R. Bates, W. W. Liberick, and J. Burckle
Introduction
Since 1973, the U.S. Environmental Protection Agency's
Industrial Environmental Research Laboratory in Cincinnati,
Ohio (lERL-Ci) has performed research related to oil shale
processing and disposal. This research is in support of the
Clean Air Act, the Federal Water Pollution Control Act, the
Resource Conservation and Recovery Act, the Safe Drinking
Water Act, and the Toxic Substances Control Act. Potential
environmental impacts from oil shale development activities
have been identified and potential control technologies are
being evaluated through a combination of laboratory and
field tests on actual oil shale waste streams. This paper
discusses recent results from the oil shale program.
Included are field test results on control of sulfur gases at
Occidental Oil Shale's Logan Wash Site and Geokinetic's
Kamp Kerogen Site; wastewater treatability studies on
retort water and gas condensate at Logan Wash; and
results of laboratory and field testing on raw and retorted oil
shales.
Air Pollutants and Controls
Potentially, the mining and processing of oil shale to
produce a refined shale oil product might produce a variety
of air pollutants, many of which could have a significant
adverse environmental impact if not properly managed.
Such pollutants include those shown in Table 1: fugitive
emissions from mining, transportation, and materials
handling; process emissions from materials preparation
such as sizing for indirect-fired retorting; process emissions
from oil upgrading and storage facilities; fugitive emissions
from waste handling and disposal; process emissions from
utility generation; and process and fugitive emissions from
infrastructure development and secondary pollution
sources.
conversion (direct processes), or adsorption and stripping,
to form a more concentrated stream which is then
processed for sulfur conversion and recovery (indirect
processes). More than 30 such commercial processes exist.
IT Environscience conducted an in-depth evaluation of a
number of emissions reduction systems to determine
applicability for removal of hydrogen sulfide from retort
off-gases (Lovell, etal., 1982). The study found that
because gases from direct-fired retorts have a high content
of carbon dioxide relative to hydrogen sulfide and also
contain large amounts of ammonia and unsaturated
hydrocarbons, they are significantly different from gases
encountered in commercial applications of desulfurization
technologies. Hence, such technologies cannot be simply
applied at full scale through technology transfer; such
transfer must be achieved through an application research
program to minimize risk and maximize success.
Since retort off-gases will be produced in very large
volumes at near atmospheric pressure, many desulfuriza-
tion processes cannot be economically applied. The high
concentration of CO2, andC02/H2S ratios, and the presence
of oxygen, unsaturated hydrocarbons, or organic sulfur
species also make application of a number of desulfurization
techniques impractical.
Of the systems studied, the Stretford, EIC, MDEA (Select-
samineand Adip), Benfield, Diamox, and Selexol appear to
have the potential for the greatest H2S selectivity for
application to direct-fired retorts. Except for the Diamox
process, these systems should be capable of controlling
H2S to about 10ppmy, resulting in an H2S control efficiency
of 99+ %. However,'organic sulfur compounds, principally
COS, are not significantly removed or are only partly
removed by the various processes. Therefore, the presence
of such compounds may lower the overall reduced sulfur
control efficiency to about 98% (Table 2).
Processes for removal of reduced sulfur and carbon dioxide
from gases are known as acid-gas removal or gas-sweeten-
ing processes which involve either adsorption and chemical
When discussing reduction of emissions from oil shale
retorting operations, it is useful to classify the retorting
processes as direct-fired (where the combustion occurs
-------
Table 1. Potential Pollutants from Oil Shale Mining and
Processing
Table 3. Concentration of Major Sulfur Species in Retorted
Off-Gas
Criteria
Pollutants
Paniculate
Sulfur dioxide
Nitrogen oxides
Carbon monoxide
Hydrocarbons
Lead
Hazardous
Pollutants
Asbestos
Arsenic
Beryllium
Mercury
Polycyclic Organics
Radionuclides
Other
Pollutants
Ammonia
Hydrogen sulfide
Trace metals
Reduced organic sul fides
within the retort itself) or as indirect-fired (where combus-
tion to produce the heat required for retorting occurs
outside of the retort). This distinction is important because
it influences the nature of the retort off-gas to be treated
(Table 3) and hence poses distinctly different duty require-
ments which must be considered in the selection of an
emissions reduction technology. The predominant factor in
the selection process is the C02/H2S ratio. Direct-fired
retorts have C02/H2S ratios ranging from 75 to more than
165. Such high ratios require a process that selectively
removes H2S in preference to COz, in order to prevent
excess consumption of reagent capacity by the C02, a nd the
resulting increase in processing costs. Indirect-fired retorts
produce off-gases having a C02/H2S ratio in the range of
4.5 to 5, which permits the use of a nonselective process.
Based upon a cost evaluation of a model case, the Strerford
process appears to be the most cost-effective approach for
direct-fired retorts (Lpvell, et al., 1982). In Table 4 the
relative costs for using the Stretford, MDEA selective
absorption followed by Stretford or Claus sulfur recovery
systems, and the Diamox process are compared. The
estimated cost for sulfur control using the Stretford process
directly is fifty cents per barrel of oil, which is less than half
th'at of the indirect sulfur removal process.
Based upon the results of this study, EPA funded the design
and construction of a Stretford pilot unit which has been
util ized in three series of tests. These three test series were
conducted at Occidental's Logan Wash retorts 7 and 8
(OXY—6/17/82 to 7/1/82), Geokinetics' Kamp Kerogen
(GKI—9/15/82 to 10/1 /82), and the U.S. Bureau of Mines
coal gasification facility at Twin Cities Research Center
(TCRC—11/5/82 to 11/22/82). The results are summa-
rized in Table 5.
Species
Direct-Fired
Indirect-Fired
WaS
COS
CS,
CH3 - SH
0.1 -0.6 percent v
10O - SOO ppmv
100 - BOO ppmv
50 - 3OO ppmv
1.0 -5.0 percent v
1OO - lOOOppmv
100 - lOOOppmv
50 - 5OO ppmv
Table 4. Relative Costs of Various Gas Desulfurization
Options'
1-Step 3-Step
Stretford Stretford MDEA MDEA
Direct Direct Selective Selective Diamox
Process Process Adsorption Absorption Process
with with with with with
Purge Purge Stretford Claus Claus
Stream Stream Sulfur Sulfur Sulfur
Category Disposal Recycle Recovery Recovery Recovery
Installed
capital
cost 1.0O
Utility
costs 1.0O
Total
operating
costs 1.00
Value of
sulfur
recov-
ered 0.92
Net
annual
cost 1.0O
Cost
effective-
ness 1.OO
1.11
1.51
1.13
O.95
1.15
1.15
1.29
5.90
1.88
0.96
2.OS
2.O4
1.17
9.88
2.15
1.00
2.37
2.34
2.46
16.79
3.90
0.98
4.47
4.54
'Based on a plant processing 10.2 million sm3 of Paraho gas per
day. The most effective or least costly option in each category is
shown as unity (1.OOJ. The relative costs of the other options are
shown as a ratio to the most effective or feast costly system.
Source: Love//, et al, 1982.
Table 2.
Variation of Process Effectiveness with Amount of COS in Raw Gas'
Estimated Overall Emissions
HtS
Remaining
Stretford
MDEA - 1 stage
with Stretford
MDEA - 3 stages
with Claus/
SCOT
Diamox with
Claus/BSPR
(ppmv)
10
10
10
63
COS
Total Sulfur
Total Sulfur
SOa Equivalent
Reduction Reduction Remaining
fwt%)
Nil
60
60
Nil
(wt%)
98.0-99.3
99.2-99.5
99.2-99.5
96.3-97.6
(kg/hr)
19-58
13-29
13-29
70-109
kg/hr
19-58
20-59
20.4-36
75-114
(tonnes/day)
0.45-1.35
0.45-0.90
0.45-0.90
1.81-2.70
(kg/hr)
38-1 15
4O-117
41-72
151-228
(kg/bbl of oil)
0.02-0.05
0.02-O.05
0.02-0.04
0.07-0.11
Overall
Sulfur
Reduction
(wt%)
98.0-99.3
98.0-99.3
98.0-99.3
96.1-97.4
"Based on a Paraho plant producing 8.0003 of oil per day, with COS ranging from 10 to SO ppmv in the raw gas.
Source: Lovell et al., 1982.
2
-------
5. EPA Stratford field Tests
Site
Removal Efficiency
Logan Wash, MIS Retorts 768
Geokinetics. in situ Retorts
U.S. Bureau of Mines (coal gast
2O - 60%
60 - SO+%
75-99.7%
During the OXY test series, the H2S control efficiency
ranged from 20% at startup toa pproximately 60% at the end
of the test. A major field modification of the venturi
absorber was made to improve the gas-liquid contacting;
almost all of the improvement in performance is attributed
to this measure. Analysis of the resulting test data
identified gas/liquid contacting and Stretford solution
chemistry as the targets for further efforts to improve
performance.
In the GKI test series, the H2S removal efficiency ranged
from about 60% to approximately 85%, with brief periods in
the mid to high 90's. During the first phase of the test, the
effectiveness of gas/liquid contact using the venturi and an
inline static mixer was investigated with no noticeable
difference in removal efficiency. During the balance of the
test, various experiments were conducted to evaluate
solution chemistry. First, an increase in the vanadium
content of the Stretford solution improved H2S removal.
Second, it was suspected that the reaction vessel retention
time was too short to permit the completion of the Stretford
solution reactions, thereby permitting recycling of incom-
pletely regenerated Stretford solution to the venturi scrub-
ber. To verify this effect, the liquid in the reaction tank was
raised to its maximum level with the observed effect of
improved HZS removal. Additionally, it was noted that a 99+
percent HzS removal was obtained whenever the unit was
restarted after an extended down-time and during periods
of low retort off-gas flow through the venturi.
In a subsequent test series, performed on a coal gasification
off-gas at the USBM Twin Cities Research Center, improved
results for given operating conditions were achieved as
predicted from the GKI tests. The only change in operation
from the GKI tests was a reduction in the gas flow rate into
the pilot unit through the addition of a gas bypass
arrangement. Thischange allowed operation at lower liquid
flow rates, while maintaining the required L/G ratio at the
scrubber, and resulted in the increase in residence time in
the reaction tank needed for complete regeneration of the
Stretford solution. Through this change, HZS removal
efficiencies of 75% at the start of the test were increased
under controlled experimental conditions to 99.7%. Still
remaining to be demonstrated is that high removal effi-
ciency under sustained operation is possible on the off-gas
from a direct-fired retort.
In a cooperative program with the Department of Energy's
Laramie Energy Technology Center (LETC), EPA sponsored
tests for the control of oily particulates remaining in the
off-gas from LETC's 150-ton simulated in situ retort after oil
and heat recovery. In September 1980, Monsanto Research
Corporation conducted the tests using EPA's mobile
scrubber pilot unit.
The particle size (weight basis) measured indicated that
about 60% of the particles were less than 15 pm. 50% were
less than 5 fjtn. and the distribution was bimodal with
approximately one-third of the catch reporting to both the
greater than 20 im and to the 1.0 to 2.0 //m fractions.
Qualitative observation of the glass fiber substrates used
for particle collection indicated the presence of a straw-
colored oil material and a black, dust-like material.
The control efficiency as measured by the front half of the
Methods Procedure varied from 67% to 94%. No correlation
of control efficiency with the liquid to gas ratio (L/G) was
found within the relatively narrow experimental range.
Analysis of the "back half" of the Method Strain indicated
the presence of considerable condensible material com-
pared to the "front half," resulting in back-to-front ratios
ranging from 2.7 to 5.5 at the scrubber inlet a nd 3.7 to 22 at
the scrubber outlet. The measured control efficiency for
these condensible materials ranged from 40% to 83%,
much less overall than the front-half materials. These data
indicate the need for careful evaluation of the potential
impact of such fine particulate emissions upon visibility,
since visibility reduction is generally dominated by particles
0.1 to 2 //m in size which tend to be formed by condensation.
Retort off-gas content of ammonia, hydrogen suifide,
carbon monoxide and the Ci - Cs hydrocarbons were
substantial, that is, greater than those measured for
particulate emissions. Emissions characterization showed
that venturi scrubbing had no effect on other pollutants
contained in the emission except for ammonia.
Surprisingly, some 50% to 75% (mass basis) of the
ammonia was removed, primarily because of the rapid
adsorption of the ammonia by the water. This unexpected
result led to consideration of ammonia-bearing retort
wastewater for HjS control. A second test was conductedto
explore the feasibility of this concept. As expected, H2S
removal was achieved at reasonable efficiencies with
operation at appropriate L/G ratios, scrubbing fluid pH, and
NHa/HaS molar ratio. Particulate control efficiency was
also found to be enhanced by some 2% to 5% by the addition
of NH3.
EPA is currently investigating approaches to improve
procedures for testing a variety of chemical systems
controlling both oxidized and reduced forms of sulfur
emissions from both direct and indirect-fired oil shale
retorts. The mobile scrubber research unit is currently
undergoing modifications to incorporate capabilities for
ammonia and caustic scrubbing of the off-gases prior to
their use as fuel. EPA is also planning additional tests to
better define the efficiency of the Stretford system in
removing HzS from retort off-gas.
Wastewater Pollutants and Controls
Over the past several years, EPA has assessed the potential
environmental impact of oil shale development, particularly
in the Western Regions of the U.S. (Colorado and Utah).
Detailed development plans submitted by the prospective
developers indicate that, in the semi-arid region in which
the major development will take place, the industry will be
"water consumptive." This means that oil shale plants will
have to import water to satisfy their process needs;
therefore they will have no wastewater discharges ("zero
discharge"). Because of the scarcity of water, these
facilities will have to reuse water and be very conscious of
optimizing the "partial treatment" of selective wastewater
-------
streams for their "next best use." Developers have
proposed that any "unusable" waste streams should be
mixed with spent shale for moisturizing and ultimate
disposal in the solid waste piles. This concept leads to the
question of what should be the wastewater quality re-
quirements for spent shale moistening, a question which
cannot be answered at this time. EPA is investigating the
wastewaters from various processing technologies (i.e., in
situ, modified in situ, direct and indirect surface retorting)
and will continue to sample wastewaters to further under-
stand potential treatment problems. To date, EPA's most
significant sampling and analytical effort to determine
wastewater treatment efficiency was the field testing at
Occidental's Logan Wash 7 and 8 burns during the summer
of 1982.
These MIS oil shale retorts generate gases and an
oil/water mixture from shale pyrolysis, combustion of
carbonaceous residues, and decomposition of inorganic
carbonates. Off-gases generated exit from the retort bottom
and are brought to the surface for treatment. The retort
oil/water mixture accumulates in the product collection
sump at the retort bottom and is subsequently pumped out
and treated for recovery of the bulk of the shale oil. The
separated gas condensate and retort waters are the
wastewaters which were studied at the Logan Wash field
site.
At Logan Wash, treatability studies were conducted for
three weeks on retort water using filter coalescing,
flocculalion/clarif ication, and steam stripping technologies
(Figure 1). Also, studies were conducted for 14 weeks on
gas condensate wastewater using filter coalescing, steam
stripping, conventional and powdered activated carbon
(PAC) activated sludge treatments, sand filtration, and
granular activated carbon adsorption technologies (Figure
2).
Retort Water
Raw wastewater characterization data collected over the
16 days of retort water treatment tests are summarized in
Table 6. As expected, the raw retort water contained high
concentrations of total dissolved solids (TDS), ammonia,
total Kjeldahl nitrogen (TKN), organics, sulfide, alkalinity,
phenols, chlorides, and fluorides.
Ammonia and alkalinity were readily stripped from retort
water (see Table 7). As expected, removals of these two
Raw Retort
Water
Figure 1. Rotort water treatment scheme. Source: Day. 1983.
4
pollutants increased as the G/L ratio increased. Greater
than 97% ammonia and 47% alkalinity removals were
achieved with G/L ratios equal to or greater than 0.18 kg of
steam per liter of feed water (1.5 Ib/gal). TKN removals
resembling ammonia removals (>99%)at G/L ratios as low
as 27% to 54%, were also achieved and the TKN removals
appear to depend on G/L ratio and feed phenol concentra-
tion. Using the G/L ratios between 0.07 kg/L(0.6 Ib/gas)
and 0.30 kg/L(2.5 Ib/gas), incidental removals of organics
ranged from 0 to 25% for dissolved organic carbon (DOC),
5% to 11 % for soluble BODS, and 16% for COD.
Primarily, retort water was treated to remove oil and
grease, suspended solids, ammonia, and alkalinity. The
filter coalescer, flocculator/clarifier, and steam stripper in
series were used to remove these pollutants. The overall
treatment scheme was very effective for removal of
ammonia and alkalinity. Relatively high sulfide, TKN, and
phenols removals were also achieved. Due to low levels of
oil and grease and suspended solids, the scheme was not
effective in removing these pollutants, nor was it critical
that these low levels be further reduced prior to the next
series of treatment.
Gas Condensate
Raw gas condensate wastewater was analyzed for conven-
tional pollutants during 14 weeks of gas condensate trials.
The results of these analyses are summarized in Table 8. As
expected, the raw gas condensate contained high concen-
trations of ammonia, TKN, organics, alkalinity, phenols, and
sulfide. Analytical results for GC/MS organic compounds,
metal, and DOC fractionation are presented in Reference 2.
Filter coalescing, steam stripping, conventional activated
sludge treatment, sand filtration, and GAC adsorption
comprised the overall treatment scheme for the gas
condensate. The scheme was very effective in removing
ammonia, organics, sulfide, alkalinity, and solids from the
gas condensate (Table 9).
The other treatment scheme utilized coalescing, steam
stripping, and GAC adsorption. The scheme was effective
for removal of ammonia, organics, sulfide, alkalinity, and
solids from the gas cpndensate. However, the performance
of granular activated carbon adsorption was relatively poor
and this scheme was apparently less effective for pollutant
removal than was the scheme which included an activated
sludge system.
In summary, pilot-scale field treatability studies on real-
time oil shale wastewaters from Occidental in situ MIS
retorts demonstrated that retort water had high concentra-
tions of ammonia, TKN, alkalinity, dissolved organics,
phenols, sulfide, and TDS; and gas condensate had high
concentrations of ammonia, TKN, dissolved organics,
alkalinity, phenols, and sulfide. Steam stripping was
effective for removal of ammonia and alkalinity from the
retort water. Steam stripping, activated sludge treatment
(both conventional and PAC), sand filtration, and GAC
adsorption effectively removed ammonia, alkalinity, TKN,
nitrate, soluble COD, soluble BODs, DOC, phenols, sulfide,
and TSS from the gas condensate. Pollutant-removal
efficiencies across individual treatment units for retort
water and gas condensate treatment schemes are pre-
sented in Tables 7 and 9, respectively.
-------
Activated Sludge
Overheat!
Vapor
. flaw Gas
Concfensate
Filter
Coatescer
Light
Oils
Figure 2.
Steam
Steam
Stripper
n
Waste
Sludge
Aeration Basin
(With/Without
Powdered
Activated Carbon)
Sand
fitters
Gravity
Separator
Light
Oils
Return Sludge
Gas cortdensate wastewater treatment schemes.
Source: Day, J9S3.
Granulated.
Activated
Carlson
Columns
Discharge
Table 6. Raw Retort Wastewater Characteristics"
Parameter
Total COO
Soluble COD
Total BOOs
Soluble BODs
Dissolved organic carbon (DOC)
Oil and grease
NHyN
TKN
NOyN
Alkalinity as CaCO3 to pH 4.5
Sulfide-
Phosphorus
Cyanide
Phenols
Fluorides
Chlorides
TSS
VSS
ros
prf
Temperature"
Number of
analyses
performed
4
4
4
4
5
(SJ
6(5j
6(2)
5
614)
3
4
(4J
(S)
7
e
8
e
edj
(10J
(10)
Concentration
flange
3.400 6.OOO
3.10O-5.40O
2.200-4.OOO
1.90O-2.200
1.400-2.300
(50-170)
1.600-3.90O
(1.700-3.700)
1.700-3.000
(2.tOO-2,20O>
3.0-4.8
J2.OOO-J7.0OO
(13.OOO-16.000J
50-130
0.8-2. J
(
-------
Table 7. Pollutant Removal Efficiencies Across Individual
Units for Retort Water Treatment Scheme''"
Filter Flocculation/
Parameter Coalescer Clarification*
Oil and grease 6
Ammonia
TKN
Soluble BODs
DOC
Phenols
TSS 21 0
VSS 20
Alkalinity as
CaCOi to
pH4.S
Fluorides 7
Chlorides 1 1
Steam
Stripper"
97
88
5
4
32
47
'Average removal efficiencies are reported.
"Blanks indicate data not collected.
"Lime dosage at 90 mg/L.
"G/L = O.IBkg/L /J.5 Ib/gal).
Source: Day, 1983.
Solid Waste Environmental Impacts and Control
Analysis of solid waste environmental impacts and controls
for an oil shale facility presents unique problems caused by
the very large volume ol waste produced. A typical 50,000
bbl/day (7,949 mVday) facility fed by 30 gal/ton (103
L/103 kg} share will produce 22-26 million tons/year(20-
24 x 109 kg/yr) of spent shale alone, which over an
operating life of 30 years, would cover an area of 3.5 square
miles (9km1) to a depth of 150 feet (45.7m). (Bates and
Thoem, 1980.) Hence, even though this waste may not be
hazardous, it will require special handling and control to
prevent environmental impacts which would include:
Degradation of surface water quality by runoff.
Degradation of air quality by release of vapors or dust.
Sittation of surface streams by erosion.
Degradation of air quality from auto ignition.
Aesthetic impact.
Mass fail ure of disposal piles threatening life or property.
Degradation of surface and groundwater quality by
leachates.
Technologies to prevent or control most of these problems
have already been developed for other mining industries,
or have been demonstrated on a small scale for oil shale.
Runoff can be collected and treated, infiltrated or evapo-
rated, and if the disposal site is top soiled and revegetated,
Tublt 8. Raw Gas Condensate Wastewater Characteristics
Parameter
Total COD
Soluble COO
Soluble BOOs
DOC
Oil and grease
NHyN
TKN
NO3-N
Alkalinity as CaCOt to pH 4.5
Suttide
Phosphorus
Cyanide
Phenols
Fluorides
TSS
VSS
TDS
prf
Temperature'
Number of
analyses
Performed
2
37(13f
8
33
113)
4O (33)
21
16
27 119)
17
6
14)
(21)
9
8
a
6
(-560)
(-560)
Concentration mg/l
Range
2.0OO-4.10O
1.4QQ-4.10O
(2.0OO-4.20O)
6OO-1.OOO
500-1,400
(1.8-76)
6,100-14.000
{4,800-1 1 .OOO)
1.300-9,700
0.3-3.0
1 '.OOO-37.OOO
f22.OOO-40.OOO)
18-190
-------
Table 9. Pollutant Removal Efficiencies Across Individual Units for Gas Condensate Treatment Scheme''"
Parameters
Oil and grease
Ammonia
TKN
Soluble COD
Soluble BODa
DOC
Phenols
Sulfide
TSS
Alkalinity as CaCOa
topH4.5
Filter Steam
Coalescer Stripper'
28
99
96
56
60
29
97
99
Activated
Sludge
Treatment"
6
59
91
52
93
GAC
Sand Adsorption
Filter Column*
95
70
89
99.5
70
"Average removal efficiencies are reported.
"Blanks indicate data not collected.
°G/L = 0.19 kg/L (1.6 Ib/gal) average.
'Hydraulic retention time =16 hours, sludge age = 32 days.
'Contact time = 19 minutes.
Source: Day. 1983
runoff will be of a quality equivalent to that from undis-
turbed areas. Emission of vapors can be prevented by
removal of volatiles from wastewater disposed with the
shale and by cooling the shale prior to disposal. Fugitive
dust can be controlled with water sprays or chemical
binders during placement, and by vegetation after reclama-
tion. Technology to revegetate spent shale, through the use
of top soil covers, irrigation, water harvesting techniques,
fertilization, and use of selected plant species has been
demonstrated in small-scale field studies. Good reclama-
tion techniques will substantially improve the appearance
of the disposal sites and will control erosion and prevent
generation of wind blown fugitive dust. Presently EPA
studies are assessing the auto ignition potential of car-
bonaceous retorted shales and fine grained raw shale
wastes. Preliminary indications from these studies are that
these materials have about the same or less potential for
auto ignition than do bituminous coals. Good engineering
design similar to that used for earth-filled dams might
prevent mass failure of disposal piles if moisture movement
within the piles can be controlled. The extent and nature of
moisture movement within the spent shale disposal pile
remains unknown and is the subject of much controversy.
When precipitation falls on a disposal site, some runs off,
some evaporates, and some infiltrates. Of the moisture
which infiltrates, most, and perhaps all, will be transpired
by plants on the reclaimed disposal site. However, any
moisture which infiltrates too deeply to be transpired will
become net infiltration into the retorted shale. The quantity
of this net infiltration into the disposal pile in conjunction
with the hydraulic properties of the retorted shale are
particularly significant because they determine the quantity
of leachate produced, and the potential for a portion of the
disposal pile to become saturated and fail. Care must also
be taken in spent shale pile location and design to prevent
groundwater, springs, and streams from infiltrating the
pile.
In cooperation with the oil shale industry, EPA is currently
sponsoring laboratory studies at Colorado State University
to determine the hydraulic properties of spent oil shales;
soon to be initiated are studies of codisposal of wastewaters
with retorted shale. Field determinations of the quantity of
net infiltration for western Colorado disposal sites will also
be initiated. Current laboratory studies are focused on
determining the nature of leachate produced from various
retorted and raw shales, and the permeability and water
holding capacity of retorted shales for various compactive
efforts and loading conditions.
Retorted oil shales will be placed in disposal sites at
relatively low moisture contents, generally.between 5%
and 20%, as required for dust control, and to achieve
desired compaction. As water penetrates into the pile from
initial irrigation or seasonal precipitation, a portion of the
water will be in storage behind the wetting front. The water
thus held in storage is not available to extend the wetting
front deeper into the pile. Given the huge size of proposed
spent shale disposal piles and the relatively low precipita-
tion in western Colorado and Utah, this water holding
capacity will be a signiificant factor influencing moisture
movement in the shale pile. Table 10 shows water holding
capacities for three retorted shales. Water holding capaci-
ties are quite large but are inversely related to initial
compaction and to loading pressures. These values are
important, not only as a measure of sorptive capacities for
net infiltration, but also as an indication of whether the
bottom of a disposal pile will saturate from loading
pressures given an initial moisture content and compactive
effort.
In addition to waterholding capacity, the hydraulic conduc-
tivity or permeability is obviously important in assessing
potential moisture migration in a retorted shale pile. An
apparatus for measuring hydraulic conductivity for various
compactive efforts and loading pressures is illustrated in
-------
Table 10. Water Holding Capacity of Retorted Shales (Expressed as Weight % Water/Dry Solid/
Sample
LURG1
No Compaction fAsh}
No Compaction
1.30g/cc(Asfij
1 .45 g/cc (Ash)
1.60g/cclAsh)
1.60g/cc
TOSCO II
No Compaction
l.30g/cc
f.45g/cc
1.60g/cc
HYTORT
t.30g/cc
1.45g/cc
1.60g/cc
14.7 psi
0.10 MPa
73.6
27.5
62.4
60.2
47.2
20.7
48.0
42.2
36.0
34.6
35.2
31.0
30.5
44. } psi
0.30 MPa
62.0
27.6
62.3
56.7
46.3
20.2
45.8
42.0
33.8
33.5
33.7
27.6
30.3
73.5 psi
0.51 MPa
64.5
26.9
62.2
55.6
45. 8
19.8
45.9
41.9
32.9
32.1
32.6
253
28.9
147 psi
1.01 MPa
63.1
25.3
62.0
55.5
44.4
19.8
43.8
41.6
32.1
30.8
31.8
23.8
25.4
200 psi
1.38 MPa
59.5
15.5
61.7
55.2
437
19.O
44.7
41.4
30.5
30.5
31.0
23.2
24.6
Source: McWhoner, 1982a.
Figure 3. Table 11 presents permeability values for several
given initial moisure contents and compactrve efforts as
these samples were subjected to increasing loading
pressures. Generally, the hydraulic conductivity decreases
sharpjy with increased in -rial compaction bul decreases
onry slighlly with increased loading pressures. This indf-
cates that the desired permeability values should be
achieved by initial compaction at the time of placement
rather than depending upon secondary compaction through
loading.
Inf/ovf
Top Load
Plate
Bottom Load
Plate
Load Spring
Cage Arm
Dial Gage
Indicator
Tension Rods
Load Ceil
Hydraulic Jack
Permea meter
Body
Overflow
Piston
Porous Stone
Sample
Bottom Plate
Flgurt 3. Schematic of pffme*metef, Saufca. McWftafter.
1982a.
As previously discussed, net infiltration, as well as water
holding capacity and permeability, are influenced by engi-
neering design and will be important in determining the
quantity of leachate produced. Also of interest is obtaining
an esfimete of quality of lea chare from either raw sriale
storage piles or retorted shale disposal sites. Standardized
tests such as the RCRA EP (acetic acid) or ASTM (water)
shaker tests as well as various column leaching tests have
been usedfor this purpose. Tables 1 2 and 13 present major
ion concentrations in leachate from raw shale, retorted
shale, and when available, field )ysimeters. The apparatus
for producing leachate from constant rate injection into a
dry packed column is illustrated in Figure 4. Column
leachate data varies depending upon the quantity (pore
volumes) of water that has passed through the material.
Data presented are for initial leachate produced and for
leachate produced after approximately one pore volume
(PV) has passed through the material. (Note that the
malarial has been contacted by one more pore volume than
shown since the column must be wetted with one complete
pore volume before any leachate is eluted.)
The data in Tables 12 and 13 indicate that none of the
laboratory methods employed provides a good indication of
what happened under field conditions. Generally, the
ASTM and RCRA shaker tests seem to correlate more
closely with the column-produced leachate after one or
more pore volumes, while the initial column-produced
leachate correlates more closely with the field data.
Logically, this might be expcted since the RCRA and ASTM
tests contact the shale with much more water than occurs
under field conditions. Generally, all the laboratory methods
are indicative of what wilt be leached but do not provide a
good indication of the concentration of species in solution.
Hence, the results of laboratorty leaching tests on raw or
retorted oil shales should ba interpreted with caution
because results form actual disposal may be significantly
different. Presently column tests are being conducted with
the material being packed into the columns in a pre-wetted
state rather Irian dry as it is felt this may better reftect field
conditions. Results from this procedure will be reported in
subsequent papers.
-------
Table 11. Summary of Hydraulic Conductivity Measurements (Constant Head)
Material
TOSCO II
LURGI-RG-I
•LURGI-ULG-I
•LURGI-RB
PARAHQ
"PARAHO
•TOSCO U
Water Density
% g/cm3
20.8 1.56
11.0 1.39
11.4 1.29
26.6 1.41
10.4 1.33
dry 1.81
dry 1,22
20.0 1.67
10.0 1.36
dry 1.09
dry 1.46
50 psi
0.34 MPa
6.8 x 10'e
5.6 x 70~5
6.7x10'*
1.0x10'"
3.0 x 70~6
8.3 x 10'*
1.2 x 10's
4.7 x 70~7
9.7 x 10"
4.6 x W~'
2.8 x 10'*
Permeability
100 psi
0.69 MPa
6.2x10'*
4.6 x 70~5
4.0 x 7CT5
6.5 x 10~7
1.9 x 70"5
4.2 x 70"7
7.7x70""
cm/s
200 psi
1.38 MPa
5.6x10'*
3.9 x JO"5
2.5 x 1 0'*
6.5 x 10'7
2.0 x 7CT5
4.5 x 70~7
2.5 x 10"
•Leaching columns (Constant Rate Injection).
Source: Me Whorter, 1982a.
Table 12. Raw Shale
Location
Test
Pore Volume
pH
EC fitS/cm at 25°C)
HCO3
C03
TDS
Cl
SO*
F
Mg
Na
C«
K
Leachates by Leaching Method (mg/L)
C-b
RCRA ASTM Field (meant
7.67 8.57 7.7
4.350 850 7.000
2,461 141 177
4.9 2.2
5.884 710 6.450
1.3 2.8 14.2
154 315 3.680
2.4 7.8
252 38 218
36 76 1.045
1,092 50 451
3.5 14 7.8
ASTM
8.22
2.500
150
1.1
2.430
84.5
1.480
1.5
173
117
350
7.0
C-a
Column*
Initial
8.17
39,000
183
1.16
-
2.433
25.224
18.5
6.465
4,660
430
53
Column Field (mean)
.73 PV
8.08 7.5
5.800 20.000
86.9 384
0.45
30.080
16 163
4.110 20.500
5.8 10.4
680 4.490
236 880
472 633
4.0 7.8
"Column leaching by constant rate injection method initial values and after O.73 pore volumes.
Source: McWhorter, 1982b.
-------
Table 13.
Shale
Paraho
Retorted Shale Leactiates by Leaching Method (mg/Li
Parameter field RCfiA
Pore Volume
ASTM
Column
Initial
Column
0.918
PH
9.57
9.27
12.O5
11.55
12.35
at25°C)
HCOj
COi
SO*
Cl
F
Mg
Na
Ca
K
Parameter
TOSCO II Pore Volume
PH
EC tjjS/cm
at 25°C>
HCO3
CO,
SO*
c/
F
Mg
Na
Ca
K
LURGI* Parameter
Pore Volume
PH
£(fjS/cm
at 25° C>
HCOa
CO,
S04
Cl
f
Mg
Na
Ca
K
21.100
--
--
1 2,350
526
1J.3
7.7
5.591
421
834
Field
£-9
ro,ooo
--
._
3O.270
..
13
156
10.270
463
110
Field
,.
_*
_„
..
..
,.,
..
_.
-•
4,600
•2,723
217
226
29
..
484
37
724
6.5
RCRA
772
5.7TO
3,325
229
22
_.
81
131
1,872
3.9
RCRA.
8.67
5.650
2.940
59
880
19
..
430
55
1.479
11
2,800
5
236
536
7
73.5
0.5
745
266
3f
ASTM
8.69
2.550
191
_.
1,130
JO
20.2
35
545
31
8
ASTM
11.85
4.270
6.9
210
2,290
17
6.3
0.4
275
713
64
9.230
15.3
232
3.840
49
21
1.6
1,500
610
140
Column
Initial
3.24
35.080
619
46
25.0OO
178
27
628
1O.095
545
89
Column
Initial
12.24
59,500
10.S
775
34,000
2.250
26.4
3.5
18,770
535
1,464
6.250
4.8
4B3
2,045
14
11.4
1.7
285
670
38
Column
1.03
9.21
S.18O
188
13
2,470
13
29.5
60
945
83
11
Column
0.621
11.93
4.250
7
2O3
2,070
15
7.6
0.3
325
575
150
'Several Lurgi shales hava been tested and results differ slightly. Example provided is Lurgi shale provided by ftio Blanco Off Shale.
Source: Me VWiorfer. 1982a, for laboratory data. Bates S Tbotm, 1980, lor Paraho fteW data. Metcaff & Eddy Engineers, 1975, tor TOSCO
field data.
to
-------
Outflow
Column Top Plate
Tension Springs
. Perforated Top
Plate
"filter Disc
Column Body
>i» Filter Disc
Perforated
.X Bottom Plate
Transducer fort
References
Lovell, R. J., S. W. Dylewski, and C. A. Peterson, IT
Environscience, Control of Sulfur Emissions from Oil Shale
Retorts, EPA-600/7-82-016. U.S. EPA, Cincinnati. OH
45268. 1982.
Day, Duane R., Desai, Bharat O., and Liberick. Walter W.,
The Treatabiiity of Wastewaters Produced During Oil Shale
Retorting, 16th Oil Shale Symposium Proceedings, Colo-
rado School of Mines Press, Golden, CO, August 1983,
pages 512-533.
Bates. E. R. and Thoem, T. L, Editors, 1980, Environmental
Perspective on the Emerging Oil Shale Industry, EPA-
600/2-80-205a, U.S. EPA, Cincinnati, OH 45268, pageSS.
McWhorter, D. B., 1982a, Leaching and Hydraulic Proper-
ties of Retorted Oil Shales, unpublished report, U.S. EPA,
Cincinnati. OH 45268.
McWhorter, D. B., 1982b. Quality and Quantity ofLeachate
from Raw Mined Colorado Oil Shale-Interim unpublished
report, U.S. EPA, Cincinnati, OH 45268.
Metcalf & Eddy Engineers. 1975. Water Pollution Potential
from Surface Disposal of Processed Oil Shale from the
TOSCO II Process, A Report to Colony Development
Operation at Atlantic Richfield Company, Operator, Volume
1, page A4-6, mean values for twelve samples.
Inflow
Figure 4. Schematic of teaching column—constant rate
injection tests. Source: McWhorter. J982a.
------- |