WATER POLLUTION CONTROL RESEARCH SERIES
14030 EDB 12/71
Water Pollution Potential
of Spent Oil Shale Residues
ENVIRONMENTAL PROTECTION AGENCY • RESEARCH AND MONITORING
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research, Series describes the
results and'progress in the control and abatement of pollu-
tion of our Nation's waters. They provide a central source
of information on the research, development, and demon-
stration activities of the Environmental Protection Agency
through inhouse research and grants and contracts with
Federal, State, and local agencies, research institutions,
and industrial organizations.
Inquiries pertaining to the Water Pollution Control Research
Reports should be directed to the Head, Publications Branch,
Research Information Division, R&M, Environmental Protection
Agency, Washington, D.C. 20460.
-------
Water Pollution Potential
of Spent Oil Shale Residues
by
Colorado State University
Fort Collins, Colorado 80521
for the
ENVIRONMENTAL PROTECTION AGENCY
Grant No. 14030 EDB
December 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.25
-------
EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication. Approval
does not signify that the contents necessarily reflect
the views and policies of the Environmental Protection
Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendations for
use.
ii
-------
ABSTRACT
Physical properties, including porosity, permeability, particle size
distribution, and density of spent shale from three different retorting
operations, (TOSCO, USBM, and UOC) have been determined. Slurry experi-
ments were conducted on each of the spent shales and the slurry analyzed
for leachable dissolved solids. Percolation experiments were conducted
on the TOSCO spent shale and the quantities of dissolved solids leachable
determined. The concentrations of the various ionic species in the
initial leachate from the column were high. The major constituents , SO,
and Na+, were present in concentrations of 90,000 and 35,000 rng/J, in
the initial leachate; however the succeeding concentrations dropped
markedly during the course of the experiment. A computer program was
utilized to predict equilibrium concentrations in the leachate from the
column. The extent of leaching and erosion of spent shale, and the
composition and concentration of natural drainage from spent shale has
been determined using oil shale residue and simulated rainfall.
Concentrations in the runoff from the spent shale have been correlated
with runoff rate, precipitation intensity, flow depth, application time,
slope, and water temperature.
This report was submitted in fulfillment of Grant No. 14030EDB under the
sponsorship of the Water Quality Office, Environmental Protection Agency.
iii
-------
CONTENTS
Section Page
I CONCLUSIONS 1
II RECOMMENDATIONS 3
III INTRODUCTION 5
Production of Shale Oil 5
Oil Shale Activities in the United States .... 5
Future Petroleum Demands 9
Oil Shale Residues 10
Erosion of Spent Shale Piles 10
Stabilization of Spent Shale Piles 10
Purpose and Scope of Report 11
IV WATER POLLUTION CONSIDERATIONS 15
Hydrologic Aspects 15
Surface Runoff Water Quality 18
Percolation Water Quality 20
Exchange Phase - Solution Phase Relationships . . 20
Crystalline Salt Phase - Solution Phase
Relationships 22
V PROCEDURE AND EQUIPMENT 25
Bench Scale Studies 25
Pilot Studies 27
Chemical Analyses 29
VI EXPERIMENTAL DATA AND RESULTS 37
Bench Scale Studies ..... 37
Rainfall Pilot Studies 53
VII DISCUSSION OF RESULTS 71
Physical Tests 71
Pilot Study 73
VIII ACKNOWLEDGEMENTS 77
IX REFERENCES 79
X SYMBOLS AND ABBREVIATIONS 83
XI APPENDICES 87
-------
FIGURES
Page
1 DISTRIBUTION OF OIL SHALE IN THE GREEN RIVER FORMATION,
COLORADO, UTAH, AND WOMING (3) 6
2 CRUSHED RAW SHALE 12
3 SPENT SHALE FROM UOC RETORTING PROCESS 12
4 SPENT SHALE FROM USBM RETORTING PROCESS 13
5 SPENT SHALE FROM TOSCO RETORTING PROCESS 13
6 SEGMENT OF COLUMN AND PORTION OF TOSCO SPENT
SHALE USED IN COLUMN STUDY 28
7 OVERALL LAYOUT OF RAINFALL FACILITY 30
8 CLOSE-UP OF EXCAVATION 31
9 INSTALLATION OF PLASTIC LINER 31
10 SAND DRAIN UNDERLYING SPENT SHALE 32
11 COLLECTION LINE FOR PERCOLATION WATER 32
12 BACKFILLING OF FACILITY WITH 68 TONS OF FRESH TOSCO
OIL SHALE RETORTING RESIDUE 33
13 PARTIALLY BACKFILLED FACILITY 33
14 COMPLETED FACILITY 34
15 COMPLETED FACILITY WITH SIMULATED 1 1/2" RAIN OCCURRING 34
16 ORION SPECIFIC ION EQUIPMENT USED FOR CHEMICAL ANALYSES 35
17 LOGARITHMIC PROBABILITY PLOT OF SIZE DISTRIBUTION
OF TOSCO SPENT OIL SHALE 40
18 LOGARITHMIC PROBABILITY PLOT OF SIZE DISTRIBUTION
OF USBM SPENT OIL SHALE 41
19 VARIATION OF THE PERMEABILITY OF TOSCO AND USBM SPENT
SHALES WITH TIME 42
20 CAPILLARY PRESSURE VERSUS RELATIVE PERMEABILITY FOR
TOSCO SPENT OIL SHALE 45
21 FLOW CHART FOR COMPUTER PROGRAM 49
22 CALCULATED AND OBSERVED VALUES OF Na+, S0= and Mg^
VERSUS VOLUME OF WATER LEACHED . . . . 51
23 CALCULATED AND OBSERVED VALUES OF Ca"*4" AND TDS
VERSUS VOLUME OF WATER LEACHED 52
24 RELATIONSHIP OF TDS IN SPENT SHALE RUNOFF WATER TO
INDEPENDENT PARAMETERS 56
25 SURFACE DEPOSIT ON TOSCO SPENT OIL SHALE 57
26 PERCENTAGE COMPOSITION OF CATIONS IN SURFACE RUNOFF
FROM SPENT SHALE AS RELATED TO INDEPENDENT PARAMETERS 60
27 SEDIMENT YIELD FROM SPENT SHALE FOR THREE HOUR PERIOD
OF SIMULATED RAINFALL 63
28 MOISTURE CONTENT OF SPENT SHALE VERSUS TIME FOR
ONE FOOT DEPTH 67
29 MOISTURE CONTENT OF SPENT SHALE VERSUS TIME FOR
ONE FOOT, SIX INCH DEPTH 68
30 TDS VERSUS CONDUCTANCE FOR SPENT SHALE SURFACE RUNOFF . 74
31 TDS VERSUS me/A OF CATIONS FOR SPENT SHALE RUNOFF ... 75
32 MEASURED CONCENTRATION VERSUS CALCULATED CONCENTRATION
OF CATIONS IN SURFACE RUNOFF FROM SPENT SHALE .... 76
vi
-------
TABLES
No.
I Major Shale Oil Reserves 7
II Empirical Constants for Equation 12 18
III Sieve Analysis of Bureau of Mines Spent Oil Shale
Residue 38
IV Sieve Analysis of TOSCO Spent Oil Shale Residue .... 39
V Physical Properties of the Various Oil Shale Residues . 43
VI Results of the Blender Experiment 46
VII Results of the Shaker Experiment 46
VIII Experimental Results of the Percolation Experiment
Conducted on TOSCO Spent Oil Shale Retorting Residue. 48
IX Definition of Terms Used in Computer Program for
Predicting the Quality of Spent Oil Shale
Percolation Water 50
X Chemical Analysis of Surface Salt Evaporation Deposit . 61
XI Calculated Sediment Yield in 3 Hour Period from
Simulated Storms 61
XII Size Distribution of Sediment in Runoff 64
XIII Jar Test Data for Sediment in Runoff from Oil
Shale Residue 66
XIV Water Balance Data for Simulated Rainfall 66
XV Concentrations of Minor Constituents 69
XVI Carbon and Nitrogen Content of Selected Samples .... 70
XVII Mass of Various Ions Leached Per 100 Grams of
TOSCO Spent Shale 71
XVIII Chemical Analyses of Filtrate from Blender Experiments
Conducted on Surface Soil Samples 72
vii
-------
SECTION I
CONCLUSIONS
1. Leaching tests show that there is a definite potential for high
concentrations of Na*, Ca*"*", Mg*"1", and SO^ in the runoff from spent oil
shale residues. However, with proper compaction, the piles become
essentially impermeable to rainfall. On the other hand, snowfall
eliminates the compaction in the top foot or so, and at least the top
2 feet of the residue becomes permeable to water.
2. Soluble salts are leached readily from spent shale columns.
3. Chemical concentrations of the effluent from spent shale columns may
be predicted by using the relationships developed between soluble and
exchangeable ions in soils which are in equilibrium with a water
solution.
4. Sediment contained in runoff water from spent oil shale residue will
be detrimental to water quality unless removed by settling.
5. Sediment in the runoff water from spent oil shale residue may be
efficiently settled by the addition of small amounts of aluminum sulfate
and/or by long periods of quiescent detention.
6. The chemical quality of surface runoff water from oil shale residue
may be estimated by procedures developed within this report.
7. This list of conclusions is necessarily incomplete until the water
pollution potential of snowfall on spent oil shale residues has been
determined. This work is now (August, 1971) nearing completion.
-------
SECTION II
RECOMMENDATIONS
This project was limited to the study of the quality and quantity of
runoff from spent oil shale residue due to rainfall. However, in the
oil shale area, more than half the total annual precipitation falls as
snow. Annual precipitation is about 12 inches on valley floors and less
than 20 inches on mesa tops. Natural snow that fell during the last of
the rainfall experiments caused changes which indicate that percolation
in the top few feet is going to be much more important in the case of
snow because of the longer contact time. Water quality of runoff from
snow melt is being investigated in another project sponsored by the
U.S. Bureau of Mines.
The problem of water runoff will require close attention with regard to
flash-floods that occasionally occur. In order to handle the runoff from
the areas that drain into the canyons (where the oil shale residue is to
be placed), it is recommended that the water be channeled away from the
canyons containing the spent oil shale residue.
Because this project has demonstrated that, even on a very flat slope,
sediment will be in the runoff, it is recommended that erosion control
measures be studied. One possible erosion control measure would be to
collect the surface runoff in ponds downstream of the residue pile.
This project has demonstrated that this approach is technically feasible,
but the economic practicality was not investigated.
Wetting to the optimum moisture and compaction of the spent shale piles
to 90 percent Proctor density is recommended to insure minimum perme-
ability and maximum stability of the piles.
It is recommended that the top surface of the shale piles not be left
bare and exposed directly to the elements because of surface runoff
leaching, erosion, and possible difficulties in revegetation. Studies
should be made to determine the most suitable and economic methods for
accomplishing this recommendation. One possibility that should be
investigated is the use of local topsoil material to cover exposed
sections of the oil shale residue.
Additional research, similar to that reported herein, should be carried
out on revegetated spent shale residue and/or native surface soils.
The foregoing recommendations are necessarily incomplete until the
water pollution potential of snowfall on spent oil shale residues has
been determined. This work is now (August, 1971) nearing completion.
-------
SECTION III
INTRODUCTION
One of the largest undeveloped natural resources in the United States is
the more than 11 million acres of oil shale land located in Colorado,
Utah, and Wyoming in the Green River Formation (see Figure 1). Shale
strata that yield from a few to about 65 gallons of crude oil per ton of
raw shale are distributed throughout much of the Green River Formation.
The oil shale ranges in thickness from a few feet to about 7,000 feet,
and represents the equivalent of about two trillion barrels of oil in
place, a quantity greater than that of the world's entire petroleum
reserve plus the total petroleum production to date (1). Competent
authorities estimate that an average yield of 75,000 barrels of crude
oil per acre is recovereable by demonstrated mining and retorting methods
(2). This would represent over 82 billion barrels of recoverable shale
oil.
The most extensive known world deposits of shale are listed in Table I.
PRODUCTION OF SHALE OIL ~ Production of oil from oil shale on a
commercial scale dates back to the 1850's when operations were begun in
Scotland and France. In the early 1900's, oil shale industries were
established in New Zealand (1900), Switzerland (1915), Sweden (1921),
Estonia (1921), Manchuria (1929), and later Russia, Germany, Spain, and
South Africa.
By the end of 1961, the principal production of oil from oil shale
had been from deposits in Scotland (about 100 million barrels), Estonia
(possibly 100 million barrels), and Manchuria (more than 100 million
barrels). It is estimated that by the end of 1961, 770 million tons of
oil shale was mined producing about 400 million barrels of oil (4).
Most of the shale-retorting activities mentioned have succumbed sooner
or later to the competition encountered from petroleum derived products.
At the present time there is significant industry only in Estonia and
Manchuria, while Brazil is attempting to establish an oil shale industry.
OIL SHALE ACTIVITIES IN THE UNITED STATES — In the United States,
activities in shale - oil development began about 1916, and over the
past 55 years, many attempts have been made to mine and retort oil shale
from deposits of the Piceance Creek Basin, (note Figure 1).
Since World War II, the major efforts have been devoted to one mining
system, the room and pillar, and to three above ground retorting systems,
those proposed by the U.S. Bureau of Mines, Union Oil Company of
California, and The Oil Shale Corporation. Of late, the U.S. Bureau of
Mines has extensively investigated in situ retorting of oil shale. In
their above ground retorting process the U.S. Bureau of Mines (USBM) used
-------
--. GREAT
DIVIDE
BASIN
NAVAL OIL-SHALE
RESERVES land 3
EXPLANATION
Area underlain by the Green River
Formation in which the oil shol«
is unopproised or low grade
Area underlain by oil shale more
thon 10 feel thick,which yields
25 gallons or more oil per ton
of shale
FIGURE 1: DISTRIBUTION OF OIL SHALE IN THE GREEN RIVER FORMATION
COLORADO, UTAH, AND WYOMING (3)
-------
Table I: Major Shale Oil Reserves (5)
Oil in place,
million bbl
Australia 200
Brazil 342,000
Bulgaria 200
Burman and Thailand 17,100
Canada 34,200
China:
Fushun, Manchuria . 2,000
Other deposits 2,700
England 1,400
Estonia 17,300
France 1,400
Germany (West) 2,000
Israel 20
Italy 34,300
Malagasy Republic 200
New Zealand 200
Republic of the Congo (former Belgian Congo) 103,000
Republic of South Africa 30
Scotland 600
Spain 300
Sweden 2,800
United States 2,000,000
U.S.S.R 6,800
Yugoslavia 1,400
TOTAL 2,570,050
a gas combustion retorting process. The gas combustion retorting process
uses heat produced from the shale mass by combustion of a portion of the
hydrocarbon vapors resulting from shale decomposition on heating.
Crushed oil shale is charged intermittently to the top of a brick-lined
steel shaft through a gas seal. By intermittent discharge of spent
-------
shale from the bottom of the shaft, the raw shale gradually moves down-
ward in the retort, first being preheated by contact with hot combustion
products and hydrocarbon vapors. Coking, or carbon formation on .the
shale also occurs, and as the partially decomposed material moves lower,
some of the remaining organic matter in the shale burns to CO and CC>2 in
a rising stream of hot air. This provides some of the necessary heat for
the retorting process, the balance and major fraction being produced by
combustion of hydrocarbons in a portion of the separated retort gases
recycled to the bottom of the shaft. Blowers furnish air at a controlled
rate into the shale bed at a point about one-third of the distance to the
top of the retort. Spent shale, ideally with very low carbon content,
preheats the recycle gas as it moves through the lowest zone of the
retort and finally is discharged through a gas seal at the base. The
upflowing recycle gas burns to CC-2, CO, and 1^0 in the combustion zone
immediately above the air inlet ports.
In the Union Oil Company (UOC) process, crushed oil shale is retorted in
a vertical bricklined shaft, the heat being furnished by burning the
carbon remaining in the shale after the hydrocarbons have been vaporized.
Upflow of solid shale is provided by charging it at the bottom through a
hopper in which a large piston moves in and out of the retort. As the
solid shale is forced slowly up through the shaft, it meets a descending
stream of hot gases and hydrocarbon vapors. When its temperature reaches
about 1000°F, decomposition and vaporization of hydrocarbons takes
place. A suction fan in the vapor recovery system provides draft for
moving the gases down through and out of the retort. Because the
incoming shale is cold, oil vapors meeting it are largely condensed,
and the liquid shale oil is drawn off at the bottom of the retort. Near
the top of the shaft, the rising spent shale meets a descending stream
of air which provides oxygen for combustion of carbon on the shale,
generating the heat required for retorting in the zone immediately below.
Burned shale, in a semi-fused or plastic state, is mechanically raked
from the open top of the shaft and discarded.
In the Oil Shale Corporation (TOSCO II) process, crushed oil shale is
preheated in a dilute-phase fluidized bed and then mixed with hot balls
in a rotating drum. The balls and shale equilibrate at a temperature of
above 900°F which is adequate to cause thermal decomposition and
vaporization of the organic constituents of the shale. The vapors are
condensed and fractionated into gas, naptha, gas oil, and heavy residuum;
concurrently, the balls are separated from the processed shale and
recirculated to a ball heater. Heat is recovered from the processed
shale. The processed shale is then moistened to obtain necessary
compactive properties and to control dust (29).
In addition to the preceding approaches, a number of other methods have
been discussed in the literature (7).
-------
FUTURE PETROLEUM DEMANDS — Present forecasts indicate that the domestic
demand for petroleum in 1980 will be about 8 million barrels per day
more than the domestic petroleum supply. This gap can be partially
closed by North Slope crude, or synthetic crudes from coal, tar sands,
and oil shale.
There are four principal alternatives for supplying this excess of
demand over supply for petroleum in the United States (8). These
alternatives are:
1. Increase the rate of discovery of petroleum,
2. Obtain a greater proportion of the oil from known reservoirs by new
or improved methods,
3. Increase the proportion of demand supplied by imported oil,
4. Develop economical processes for producing liquid and gaseous fuel
from alternative sources such as oil shale, coal, and tar sands.
A majority of the studies and forecasts made concerning the first three
alternatives for meeting the greatly increased demand for liquid fuels
agree that they probably will not yield enough oil to meet the entire
increase. As a result, the use of shale oil to supplement crude oil
resources becomes likely as a major future development.
As far as above-ground retorting is concerned, initial operations will
probably occur in the southeastern Piceance Basin of Colorado. A
commercial facility would process 50,000 to 100,000 tons of oil shale
per day, and would produce 40,000 to 85,000 barrels per day of synthetic
crude oil. A full-scale oil shale industry, including up to 10 such
plants, might develop by 1985. A proposed initial plant site is located
in Garfield County, Colorado, in the Parachute Creek escarpment area
north of Grand Valley, Colorado (elevation 5,100 feet). Water reaching
the Colorado River via Parachute Creek has dissolved solids concen-
trations as high as 800 rng/J, during parts of the year (29) .
The nearest U.S. Geological Survey water quality stations on the
Colorado River above and below the oil shale area are, respectively, the
stations near Glenwood Springs, Colorado (6.5 miles upstream from
Roaring Fork River; drainage area 4,650 square miles) and near Cameo,
Colorado (drainage area 8,050 square miles). While the flow of the
Roaring Fork River is significant compared to that of the Colorado
River, its salinity is consistently less. Examination of the records for
the 5 year period of 1964 through 1968 reveals the fact that the salt
load (in tons/day) of the Colorado River increases by a factor of 2.5
between Glenwood Springs and Cameo. In addition, the volume-weighted
average concentration of dissolved solids (residue at 180°C) in the
Colorado River increases from 327 mg/& at Glenwood Springs to 448 mg/£
at Cameo (the time-weighted average concentration increases from 395 mg/Jl
to 609 mg/2,). The salt concentration of the Colorado River at Hoover
Dam is 725 mg/£. Of this, 38% is contributed by natural diffuse sources
(274 mg/fi,).
-------
OIL SHALE RESIDUES — In the event the industrial development of oil
shale occurs, a considerable quantity of retorted oil shale residues
will be produced. The minimum economic-sized shale oil plant will
probably produce in excess of 50,000 tons of processed shale per day.
Some of the waste may be suitable for disposal as mine fill (a maximum
of about 60%), railroad and highway road fill, or even land fill. It
might even be possible to utilize some of the residue as raw materials
for making concrete, cement blocks, and bricks (9). However, the
demand for oil shale residue for these uses is small in relation to the
amount of residue that will be produced. Often the kerogen (kerogen is
the organic matter found in oil shale) content is less than 20 percent
by weight of the discarded materials. This could leave millions of tons
of unattractive barren piles of oil shale residue to mar the natural
beauty of the land if the residue were merely dumped at the mining site.
Dust from the dried out residue could contribute to air pollution and
the runoff water could contaminate water supplies with salts and other
materials.
EROSION OF SPENT SHALE PILES — Erosion would be of particular concern
on the steep slopes of unprotected residue piles. Storms could lead to
the formation of deep gullies on the slopes and alter the pattern of
drainage established from preceding runoffs. Continued erosion would
also expose new surface areas to air and moisture which could lead to
undesirable leaching and the creation of water quality problems. The
effect of revegetation on these potential problem areas is unknown.
If unprotected, a large portion of the sediment from the spent shale
piles might be deposited in stream channels near the disturbed area.
However, sediment would also be carried into large streams, where it
would settle out or move downstream. Thus, entire river basins could
be adversely affected by the spent shale piles if no care was taken to
prevent these potential problems.
Streams carrying heavy loads of sediment may require additional treat-
ment to make them more suitable for domestic and industrial uses.
Recreational use of streams could also be adversely affected by sediment»
and fish habitat could be destroyed (10).
STABILIZATION OF SPENT SHALE PILES ~ Erosion of spent shale piles may
be lessened to some extent through physical, chemical, and vegetative
methods of stabilization (11). Physical methods include covering the
fine tailings with topsoil removed from underneath the shale residue
piles. One possibility for a surface disposal site would be a canyon in
the vicinity of a proposed commercial plant site. The processed shale
could be placed in a series of horizontal layers 1 to 2 feet thick. The
upper surface would be a temporary surface until the last layer is
placed. Each layer could be started a little further back into the
canyon, giving the front surface of the pile (permanent surface) a slope
sufficiently less than the angle of repose to insure frictional
stability.
10
-------
Chemical stabilization involves reacting the residue with a reagent to
form a water and air impermeable crust or layer.
Vegetative stabilization may pose some difficult problems. Wastes are
usually deficient in plant nutrients or may contain material noxious to
plant growth. Tailings and other fine wastes usually must be covered to
a depth of four inches or more with soil and fertilized prior to seeding
(12). However, Kentucky Blue Grass, fertilized at the rate of 150
pounds per acre twice per year and watered at the rate of 1 inch per
week during the 10 week summer season has been grown on the TOSCO Process
waste after conditioning with sawdust (29). The root zone of the
Kentucky Blue Grass penetrated over 11 inches into the processed oil
shale residue.
It is believed that the black color of the oil shale residue must not be
directly exposed to the sun as the result of grass fires and/or over-
grazing. Therefore, the surface color should be more nearly that of the
native topsoil, which could be accomplished by covering with a relatively
thin layer of native topsoil.
PURPOSE AND SCOPE OF REPORT — The purpose of this report is to evaluate
the potential water pollution of rainfall on spent oil shale residues in
order that industry may be properly advised of the hazards before the
fact and as a guide for the federal leasing program.
The spent oil shale residues investigated in this study were from three
pilot plant operations. The processes were: (1) USBM, (2) UOC, and
(3) TOSCO. The oil shale for all three pilot plants came from the
Piceance Basin near Rifle, Colorado, The TOSCO residue was given
considerably more attention because the Colony Development Operation
(Atlantic Richfield Company Operator) is currently engaged in a $17
million field program to investigate the feasability of shale oil
production on a commercial scale.
Given in Figures 2, 3, 4, and 5 are pictures of the raw shale and the
residues from the various retorting processes covered in this report.
The project consisted of three phases of work: (1) Bench scale studies
were used to determine (a) permeability, porosity, and particle size
distribution, (b) the composition and maximum quantity of dissolved
solids leachable by complete slurry treatment; and (c) the composition
and quantity of dissolved solids leachable by simple downward perco-
lation through residue columns. (2) Pilot studies were conducted on the
TOSCO unweathered spent shale to define (a) the composition and concen-
tration of dissolved solids in runoff from a spent shale pile; and (b)
the properties of the residue within the pile before and after rainfall
simulation. (3) Data was interpreted using statistical techniques to
determine the quantitative relationships between the dependent and
independent variables significant to spent oil shale residue leaching.
11
-------
FIGURE 2: CRUSHED RAW SHALE
FIGURE 3: SPENT SHALE FROM UOC RETORTING PROCESS
12
-------
FIGURE 4: SPENT SHALE FROM USBM RETORTING PROCESS
FIGURE 5: SPENT SHALE FROM TOSCO RETORTING PROCESS
13
-------
SECTION IV
WATER POLLUTION CONSIDERATIONS
HYDROLOGIC ASPECTS — Among other variables, the solids leached by the
runoff water from a spent oil shale residue pile will be a function of
the intensity of rainfall, length of overland flow, kinematic viscosity
of the runoff water, and surface slope.
It would be advantageous to combine all of these parameters into one.
By consideration of the overland flow hydrograph, such a parameter can
be derived.
Using the model of laminar flow and the assumptions that (1) surface
tension effects are negligible, (2) flow is two dimensional, and
(3) there is no infiltration, the hydrograph for overland flow may be
developed as follows. The change of velocity with respect to depth is
given by:
dV = g Sine (D-y)
dy v
where
-j- = change of velocity with respect to depth, f t/(sec) (f t) ,
V = velocity at any depth, ft/sec,
y = any depth, ft,
v = kinematic viscosity, ft2 /sec,
D = total depth of flow at lower end, ft,
9 = slope angle (if slope angle is less than 6° , Sin6 may be replaced
by the slope, s, with no loss of accuracy),
g = acceleration of gravity, 32.2 ft/sec2.
Applying the boundary conditions that V = 0 when y - 0, integration of
equation 1 yields
v = g Sine (vD - y2/2)
v
The mean velocity, V, is defined as:
Vdy
. (3)
/ dy
Substituting equation 2 for V into equation 3 gives
15
-------
V
/Ddy
(4)
Completing the integration, V may be expressed as
=
The discharge per unit width, q, is V D or
As with the mean velocity, the mean depth of flow may be defined as
/ D
dx
r dx
o
where L is the total distance of overland flow (in feet) , and x is in
feet and is measured from the upstream end down. Clearly
(x \
— q and D
L J j
therefore,
1/3
, _/ 3vxq '
g Sin9 ~ Lg Sine
/ l
.1/3
, and
D =
3vq rL 1/3
j 6 X
Lg Sin9
Integration of equation 8 gives
= |D
(9)
The equilibrium flow, q , is
The constant 43,200 gives q in square feet per second when i, the
intensity of rainfall, is in inches per hour. Substitution of equations
16
-------
6 and 10 into equation 9 gives
q ^\ 1/3
D = (9.70 x 10~J)
(11)
Therefore, from equation 11, it is seen that all the hydrologic
parameters of concern are contained in the average depth of overland
flow. At 20°C (68°F),
D = (10 4)
10 iL
Sine
1/3
(11A)
In order to use equation 11 a method of determining i must be used.
Many studies have been directed to the development of. empirical
relations for estimating rainfall intensity as required in applying the
rational formula and other similar equations (13). The resulting
relations are usually of the form (14)
•Tm
i -- *2-z (12)
(td + d)n
where
i = rainfall intensity in inches per hour,
td = duration of the storm in minutes,
T = frequency of occurence in years,
c, d, m, and n are regional coefficients and exponents.
(In North America, 5 <^ c <_ 50, 0 < d < 30, 0.1 < m < 0.5, and
0.4 <_ n <_ 1). ~~ ~ — -
The constants for equation 12 are obtained by a graphical or least
squares fitting technique applied to original data. This has been done
by Norton (15) for four Colorado and one Wyoming first order U.S.
Weather Stations. These constants, shown in Table II, allow mathematical
determination of the rainfall intensity for a given storm. Of particular
interest in this study are the empirical constants for Grand Junction,
Colorado.
To verify whether or not the assumption of laminar flow is met, the
overland flow characteristics should be known. The Reynolds number, R,
is used as an index of the occurence of turbulence. The Reynolds
number of overland flow at a section where the depth is D feet and the
mean velocity is V ft/sec, is
» • • (13)
17
-------
Table II: Empirical Constants for Equation 12
Geographical
Location
of Station
Empirical Constants
m
n
Elevation
of Station in
feet above sea level
Pueblo, Colorado 29.9 0.232 8.9
Grand Junction,
Colorado
Denver, Colorado
Cheyenne, Wyoming
Wagon Wheel Gap,
Colorado
6.3 0.265 1.7
17.6 0.299 4.3
27.6 0.199 5.2
6.7 0.248 0.4
0.872
0.721
0.820
0.853
0.707
4712
4845
5331
6156
8390
Substituting equation 5 into equation 13 gives
„ 4 g Sin9 D3
R 3V2 '
Then, substituting equations 9 and 11 into equation 14 gives
R - 9.29 x 10~5 —
(14)
(15)
At 80°F,
R - 10.0 iL .
(16)
Owen (16) states that overland flow becomes turbulent for R greater
than 4,000. Therefore, to insure laminar flow
iL < 400 .
(17)
SURFACE RUNOFF WATER QUALITY — There are two major classifications of
suspended solids in water, settleable suspended solids and nonsettleable
suspended solids. The quantity of settleable solids are important in
determining the size of sludge handling facilities and they can be
determined from a standard Imhoff cone test (17).
The nonsettleable solids are either organic or inorganic. It is
reasonable to assume that the concentration of the dissolved inorganic
solids in the runoff water will be a function of the parameters described
by equation 18,
f(i, AGO, t, v, s, L, p, k)
(18)
18
-------
In which,
C = concentration (mg/Jl) ,
Aco = decrease in surface moisture content since the last rain,
dimension! ess,
t = time since beginning of runoff, (hr) ,
p = the gross density of the surface shale (g/cc) ,
k = permeability of the surface shale (cm2) .
From equation 11, I) = f(i,tv, L, s) . Therefore equation 18 becomes,
f(D, AID, t, p, k) .
(19)
Further refinement of equation 19 may be obtained by making the
following assumptions. _
1. The concentration of dissolved solids varies inversely with D
to some power, N,
--
D
(20)
2. The concentration of dissolved solids varies directly with the change
in surface moisture. This effect of surface moisture is greatest at the
beginning of a storm and diminishes with time. This suggests an
exponential relationship of the form,
(21)
in which K is a constant. The reasoning for the above formulation of
the relationship between C and Ao> is explained in Appendix C.
3. The concentration of dissolved solids in the runoff decreases with
time during a given storm. This implies an exponential relationship of
the form,
(22)
in which y ±s a constant.
Combining equations 20, 21, and 22 with equation 19 yields
DN
f(p, k) exp
KAto - tY
0.435
(23)
19
-------
The coefficients, exponents, and f(p, k) now remain to be determined
experimentally.
PERCOLATION WATER QUALITY — One of the most important factors to be
considered from a water quality standpoint is the increase in dissolved
solids concentration of the percolation water as it percolates through
the spent oil shale. This change takes place by the following
mechanisms:
1. Ion exchange - Ion exchange is the reversible process by which
cations and anions are exchanged between solid and liquid phases and
between solid phases if in close enough contact with each other.
Although ion exchange does not increase the total concentration in
milliequivalents per liter of a solution, it can and does change the
ionic composition of the percolating water.
2. Adsorption and Desorption of Ions - The solid components in the shale
are capable of adsorbing or releasing (desorption) solutes from or to
the soil solution. These processes differ from ion exchange in that an
increase or decrease in concentration of solutes in the solution occurs.
The most important single factor in determining the composition of the
water which percolates through the shale is the concentration and types
of salts found in the shale. These may be divided into:
1. Soluble Salts - These salts are readily soluble in water and are
leached rapidly from the shale.
2. Slightly Soluble Salts - These salts are only sparingly soluble in
water and will leach slowly from the shale.
Previous studies (18) indicate the shale will contain CaSO, as a
slightly soluble salt and MgSO^ or NaSO^ as soluble salts.
All of the above factors govern the composition of the equilibrium
solution, that is, the solution that would result if a solution were
allowed to remain in contact with the shale until no further change in
the resulting solution occurred.
To develop a theory that will enable the prediction of the quality of
the percolating water, it is necessary to consider three phases. These
phases are: ^
1. an exchange phase, consisting of Ca -Mg -Na shale,
2. a crystalline salt phase consisting of the slightly soluble salt,
CaS04' -H- ++ +
3. a solution phase of Ca , Mg , and Na salts.
EXCHANGE PHASE-SOLUTION PHASE RELATIONSHIPS — The relationship between
soluble and exchangeable ions in soils in equilibrium with a solution
can be represented by relationships similar to those employed for
chemical reactions in solutions.
20
-------
For example, for the homovalent exchange of Ca"1""1" and Mg**, the exchange
equation is (19)
[Ca"]
- - - K
[MS/]
and for the monovalent-divalent exchange of Ca"^", Mg++, and Na+, the
exchange equation is
[Na+] [Na+]
jj~ • (25)
L • / _
] + [Mg^] -W [Cap +
In equations 24 and 25, the subscripts a and s refer to the adsorbed
and solution ions respectively, square brackets denote concentration
in moles per liter if in the solution phase, or moles per gram of soil
if in the adsorbed phase. K and K' are equilibrium constants. K is
dimensionless and K1 is -\/moles /liter . If [X] and [Y] are the number
of moles of Na and Mg per gram of soil leaving or entering the complex
(entering the exchange complex -will be considered positive) and B is
the ratio of grams of oven dry soil to the liters of solution contained
in the soil duiring percolation, equations 24 and 25 become
B(l - K)Y2 + [B(Mg + KCa + 0.5 KX + 0.5 X) + Ca + KMg ]Y +
3. cl S S
+ Ca Mg - KMg Ca + 0.5 X (BMg + KXMg ) = 0 (26)
S S S 3. cl S
and
aX4 + bX3 +cX2 + dX + e = 0 (27)
where
a = 0.250 B2K'2 (28)
b = -Kf2(B2Ca + B2Mg + 0.5 BNa ) - 0.5 B (29)
3. SL S
c - K'2(B2Ca2 + 2 BNa Ca +2 B2Ca Mg + 2 BNa Mg + B2Mg2)
a s a a Ba s 6a 6a
- Ca - Mg - BNa +0.25 Kl2Na2 (30)
S S 3. S
d - -2 K'2(BNa Ca2 + 0.5 Na2Ca + 2 BNa Ca Mg + BNa Mg2)
sa sa sa°a s °a
- 2 Ca Na - 2 Mg Na - 0.5 BNa2 - Mg K'2Na2 (31)
sa°sa aeas
e - Kl2Na2 (Ca2 + 2 Ca Mg + Mg2) - Na2 (Ca + Mg ) . (32)
s SL aa a as s
21
-------
The concentrations of Mg , Ca , Ca , Mg , Na , and Na in equations
26-32 are all initial concentrations.
Solution of equations 26 and 27 will give the number of moles of Na
and Mg44 per gram of soil leaving or entering the complex. Because the
total equivalents of Ca44, Mg4*, and Na+ taking place in the exchange
must sum to zero, the number of moles of Ca per gram of soil leaving
or entering the exchange complex, [Z], when a given solution is brought
into contact may be computed from
Z = -(X/2 + Y) . (33)
CRYSTALLINE SALT PHASE-SOLUTION PHASE RELATIONSHIPS — The solubility
of CaSO^ is described by the solubility product
f2 [Ca ][SO, ] = K = 2.4 x 10~5 @ 25°C. (34)
S 4S ^P
K is the solubility product and f represents the activity coefficient
alpcalculated from the Debye-Huckel theory.
If w is the number of moles per liter of Ca++ and SO^ that dissolve or
precipitate (positive w denotes dissolving), equation 34 may be written
as
w2 + (Ca + SO. )w + Ca SO. - K /f2 = 0 . (35)
v s 4s s 4s sp
w may be obtained by using the quadratic formula. The concentration of
Ca and SO. are initial concentrations.
s 4s
Equations 26 and 27 give the relationship between the soil and solution,
and equation 35 gives the relationship between the solution and CaSO^.
By alternately holding the soil phase and the gypsum (CaSO.) phase
constant, the equilibrium solution for this three phase system may be
calculated through an iterative procedure.
To calculate the quality of water percolating through the shale, the
depth of shale is considered to be made up of n segments Ay in length.
The total length of the column, y, is
k=n
y = I A y (36)
k-1
where k designates the particular segment. Assuming the amount of
solution present in each segment, q, , is the same in each segment, the
final concentration of component j, C', in the first aliquot will be
22
-------
where AC, . is the change in concentration of solute j when the aliquot
q^ is passed through segment k. Using equations 26, 27, and 34, and
making n in equation 37 finite, the average concentrations of the
aliquots may be approximated and the quality of the water percolating
through the shale predicted (19).
23
-------
SECTION V
PROCEDURE AND EQUIPMENT
BENCH SCALE STUDIES:
1. Mechanical Sieve Analysis - A representative sample of the various
oil shale residues was selected by proportioning. All samples were
either air dried or oven dried until the moisture content was less than
one percent. This moisture content was small enough not to interfere
with the various analyses.
The 1 1/2", 1", 3/4", 1/2", 3/8", and Nos. 4, 8, 16, 30, 50, 120, and 200
U.S. standard size sieves were used for sieving. A portion of the
sample was placed on the sieves and shaken for ten minutes on a
mechanical shaker. At the end of that time the series of sieves were
removed and the portion of sample retained on each sieve weighed and
recorded. This procedure was followed until the total sample was
sieved.
For that portion of the sample passing the No. 200 sieve, a hydrometer
was used to determine the particle size distribution. Fifty grams of
the fraction passing the No. 200 sieve were placed in a beaker with
250 ml of distilled water and stirred mechanically for one hour. After
stirring, the mixture was transferred to a graduated glass cylinder and
distilled water added until a total volume of 1000 ml was obtained. The
cylinder was then transferred to a constant temperature bath. After the
soil suspension reached a constant temperature, the cylinder was removed,
thoroughly shaken, and returned to the bath. Hydrometer readings were
then taken at the end of 2, 3, 5, 15, 30, 60, and 360 minutes.
2. Porosity - The porosity of the USBM and TOSCO spent shales was
determined by placing 2000 grams of the respective shale in a plastic
cylinder. The walls of the container were tapped with a wooden mallet
until the volume of the sample remained constant.
The sample was then placed under a vacuum for 3 hours , to remove any
trapped air. A known quantity of water was then applied to the shale
through a valve located at the bottom of the cylinder. The porosity,
e , was then calculated from the equation
(38)
where
V =
v
V = the total volume of the sample.
V = the volume of water to fill the voids of the sample, and
25
-------
Because of its size and irregularity, the porosity of the clinkers from
the UOC burned shale was determined using a different procedure. A
container was filled with Ottawa sand, weighed, and the apparent weight
density of the sand determined. The volume of the oil shale residue
sample was determined by calculating the volume displaced by the sample
when it was buried in the container filled with sand. The voids of the
UOC sample were determined as previously described and the porosity
calculated from equation 38.
3. Density - The density of the spent shales was determined from a 50
gram portion of the sample which had passed a No. 40 sieve. A 250 ml
flask was used for a pycometer, and the procedure as outlined in
reference (20) was used to determine density.
4. Permeability - The permeability of the spent shales was determined
using a series of three samples. Each sample weighed 500 grams, and was
placed in a constant head permeameter, and compacted to the same bulk
density as in the porosity determinations. In order to prevent clogging
of the inlet and outlet of the permeameter, 1 centimeter of fine gravel
was placed at both ends of the sample. The head loss through the gravel
was assumed zero because its permeability was many orders of magnitude
greater than that of the spent shale. The samples were vacuum saturated
and placed under a constant head of 100 cm of water.
Darcy's law which is valid for the linear laminar flow regime in porous
media, is
(39)
dy k ^y;
To calculate permeability, equation 39 may be written as
(40)
"F/ "y
where
k is the permeability of the porous medium in cm2,
v is the macroscopic velocity in cm/sec,
y is the absolute viscosity of the fluid in poises, and
-7*- is the pressure drop per unit length due to friction, in dynes per
^ cm2 per cm.
5. Blender, Shaker, and Column Experiments - In order to determine the
composition and concentration of dissolved solids leachable from the
spent shales, three experiments were devised. The first experiment was
the blender experiment which consisted of taking a 100 gram sample of
the shale (which passed the No. 40 sieve) and then mixing it with 250 ml
of distilled water in a blender for 5 minutes. The mixture was then
26
-------
removed from the blender, 750 ml of distilled water was added, and the
suspension was filtered using a vacuum system with a BUchner funnel and
No. 40 Whatman filter paper. The filtrate was then refiltered and the
resulting solution placed in a plastic bottle for storage until the
conductance could be measured and a chemical analysis completed.
The second experiment was the shaker experiment which consisted of
taking a 100 gram sample of shale (which passed the No. 40 sieve) and
placing it in a one gallon container. One liter of distilled water was
added and then the container was shaken manually for 5 minutes. The
mixture was filtered and the solution stored as described above.
To determine the quantities of dissolved solids leachable by simple
percolation, a column experiment was conducted on the TOSCO shale. The
apparatus for the experiment consisted of a plastic column 120 cm in
length and 10 cm in diameter. Taps were inserted into the column at 10,
40, 70, 100, and 120 centimeters from the top of the column so that
percolation water could be removed at these levels. The column was
filled with 12,500 grams of the TOSCO spent shale and compacted to the
bulk density used in the porosity determination. A constant head of
2 cm of tap water was maintained on the top of the shale. Concentrations
in the tap water were subtracted from observed concentrations.
Figure 6 shows a picture of a segment of the column and a portion of the
spent shale as removed from the column after percolation.
PILOT STUDIES:
The most important phase of the study was a pilot study program, using
TOSCO unweathered spent shale, conducted on the Colorado State University
rainfall-runoff facility. The pilot study objectives were:
1. To determine the quality and quantity of total runoff from spent
shale piles using natural and simulated rainfall.
2. To determine the properties of the residue within the piles before
and after rainfall simulation.
Model Characteristics - The model used for the study had the following
characteristics:
1. Approximately 68 tons of TOSCO unweathered spent shale were placed
in a pile 80 feet long, 8 feet wide at the bottom, and 12 feet wide at
the top, with a depth of 2 feet. The surface of the shale had a 0.75
percent slope. This is roughly the maximum permissible slope if
excessive erosion is to be prevented.
2. A four-inch layer of sand was placed below the shale to serve as a
drain for any percolation water.
3. An impermeable plastic barrier was placed below the sand filter and
along the sides of the facility to insure that no percolation losses
occurred.
4. A three-inch perforated plastic pipe was placed in the sand filter to
collect any percolation water and divert it to a 50-gallon drum for
storage.
27
-------
FIGURE 6: SEGMENT OF COLUMN AND PORTION OF TOSCO SPENT SHALE
USED IN COLUMN STUDY
28
-------
5. Artificial rainfall was generated by a system of nozzles spraying
into the air over the facility. The system had the capability of
producing rainfall intensities from about 1/2 inch per hour to over 2.5
inches per hour.
6. Cumulative rainfall mass curves were obtained from a recording-type
raingage. The surface runoff was measured by an H-flume with a standard
float gage. After passing through the flume, the runoff water was
diverted to a small settling basin where it evaporated and seeped into
the soil.
7. Three access tubes for use of a neutron moisture probe were installed
in the middle of the shale at 20, 40, and 60 feet from the upstream end
of the facility.
8. Four thermistors were installed 60 feet downstream for the upstream
end to monitor the temperature of (a) the air, (b) the surface of the
shale, and (c) the shale at depths of one and two feet below the surface.
9. A trailer was located at the downstream end of the facility to
serve as an onsite laboratory.
Shown in Figures 7 thru 15 are pictures taken during the various
stages of construction of the facility,
CHEMICAL ANALYSES — The procedure used to determine the concentrations
of the various constituents is found in the 1965 edition of Standard
Methods for the Examination of Water and Wastewater, (17), except for
the following ions: H+, Na+, Ca4"*", Pb4"1", Cl~, F~, I~, NO^, and Br~.
The concentrations of these ions were determined with the use of the
respective specific ion activity electrodes.
Briefly, measurement of ion activity is accomplished with the electrodes
by determining the potential that is developed between the test sample
and the special filling solution inside the electrode. The Nernst
equation predicts that at 25°C the potential for a monovalent sensing
electrode will change approximately 59 millivolts for each decade
change in ion activity, while for a divalent sensing electrode the change
is 29.5 millivolts, etc. By determining the activity coefficient, f, the
concentration, c, of a particular ion is given by
c = f- (41)
where
a = the activity of a particular ion in moles per liter.
The complete procedure for determining f and c is given in Appendix A.
Figure 16 is a picture of the Orion specific ion activity electrode
equipment used for ionic analysis.
29
-------
FIGURE 7: OVERALL LAYOUT OF RAINFALL FACILITY: Excavation for
oil shale on right and holding pond for runoff water on left.
Large tower in center used to mount floodlights for night work.
30
-------
FIGURE 8: CLOSE-UP OF EXCAVATION: Approximately 3 feet deep x 8
feet wide x 80 feet long.
FIGURE 9: INSTALLATION OF PLASTIC LINER
31
-------
FIGURE 10: SAND DRAIN UNDERLYING SPENT SHALE: Four inch layer of
sand underlying spent shale to serve as a drain of percolation water,
FIGURE 11: COLLECTION LINE FOR PERCOLATION WATER: A 3-inch perforated
pipe in a gravel trench collects infiltration water and transports it
to a 50-gallon drum for storage.
32
-------
\
FIGURE 12: BACKFILLING OF FACILITY WITH 68 TONS OF FRESH TOSCO
OIL SHALE RETORTING RESIDUE
FIGURE 13: PARTIALLY BACKFILLED FACILITY
33
-------
FIGURE 14: COMPLETED FACILITY: Trailer was used as a field
laboratory.
FIGURE 15: COMPLETED FACILITY WITH SIMULATED 1-1/2" RAIN OCCURRING;
Mote vertical pipes in center of shale for insertion of nuclear
moisture probe and rain gage partially hidden by floodlight.
34
-------
FIGURE 16: ORION SPECIFIC ION EQUIPMENT USED FOR CHEMICAL ANALYSES
35
-------
SECTION VI
EXPERIMENTAL DATA AND RESULTS
BENCH SCALE STUDIES
Physical Properties - The results of the mechanical sieve analysis for
the USBM and TOSCO spent shales are given in Tables III and IV. A
mechanical sieve analysis of the UOC burned shale was not possible
because most of the burned shale is in the form of large clinkers as
shown in Figure 3.
The observed data were plotted on logarithmic probability paper as shown
in Figures 17 and 18. In general, 68.2 percent of the plotted points
will fall between the two dashed lines if the data are distributed in
accordance with a geometrically normal frequency distribution. The
method involved in determining the position of the two dashed lines is
described by Ward (21), Both graphs indicate that the size distribution
of the particles may be approximated by a geometric distribution,
especially in the lower ranges of particle sizes (< 0.10 cm). The
smaller particles are of the most concern since they will have the
greatest effect on the other physical parameters such as porosity,
surface area per unit volume, and permeability.
If the particle size distribution is geometrically (or log normal)
distributed, then the geometric mean of the particle size distribution,
M , in cm, can be calculated from the results of the sieve analysis using
tne following equation:
log M = E p log X (42)
o
where p is the decimal fraction of the total sieve sample retained
between 2 sieves whose geometric mean size is X cm. If Xi represents
the size of the upper sieve and X£ represents the size of the lower
sieve, then X = /x7xT"~, The geometric standard deviation of the
particle size distribution, a , dimensionless, is calculated using the
equation ^
log a = (Z p log2 x )1/2 (43)
o &
where
x - X/M
g g
Knowing the permeability, geometric mean size, geometric standard
deviation, and porosity, the shape factor, (dimensionless), of the
37
-------
Table III: Sieve Analysis of Bureau of Mines Spent
Oil Shale Residue
Sieve
U.S. s tandard
1.50 in.
1.00 in.
0.75 in.
0.50 in.
0.345 in.
No. 4
No. 8
No. 16
No. 30
No. 50
No. 120
No. 200
Hydrometer
Summation
Opening
(mm)
38.10
25.40
19.05
12.70
9.52
4.76
2.38
1.19
0.595
0.297
0.125
0.074
0.0346
0.0336
0.0268
0.0157
0.0077
Weight retained
in grams
0
398
1,834
2,567
1,715
2,441
1,807
1,422
1,181
1,061
1,660
1,546
533
381
446
38
12
19,042
Percent
retained
0.00
2.09
9.63
13.48
9.01
12.82
9.49
7.47
6.20
5.57
8.72
8.12
2.80
2.00
2.34
0.20
0.06
100.00
Cumulative
percent finer
100.00
97.91
88.28
74.80
65.79
52.97
43.48
36.01
29.81
24.24
15.52
7.40
4.60
2.60
0.26
0.06
0.00
media may be calculated from the following equation (23):
(k)(36 KT)(1 - e)2a
(j)2 = s
e3 M2
g
where
36 ±s a pure number, Q <_ $ <_ 1. (<|> = 1 for spheres),
T = tortuosity. The theoretical value of T for fully saturated
unconsolidated porous media that are isotropic is 2 (24),
(44)
38
-------
Table IV: Sieve Analysis of TOSCO Spent
Oil Shale Residue
Sieve
U.S. Standard
No. 8
No. 16
No. 30
No. 50
No. 120
No. 200
Hydrometer
Summation
Opening
(mm)
2.38
1.19
0.595
0.297
0.125
0.074
0.0461
0.0346
0.0336
0.0268
0.0157
0.0077
Weight retained
in grams
0
567
390
588
1,170
784
1,134
1,125
2,043
2,882
287
68
11,038
Percent
retained
0.00
5.14
3.53
15.33
10.60
7.10
10.28
10.20
18.50
26.10
2.60
0.62
100.00
Cummulative
percent finer
100.00
94.86
91.33
86.00
75.40
68.30
58.02
47.82
29.32
3.22
0,62
0.00
K = a dimensionless constant that depends on the shape of the cross-
section of flow and theoretically, 2 < K < 3. Ward (25) has
determined experimentally that K has a value of 2.36 ± 0.11 for
unconsolidated porous media. K is exactly 3 for a cross-section
formed by closely spaced parallel plates and is exactly 2 for a
circular cross section.
The permeability experiments were run over a period of one week, with
readings taken at various time intervals. The observed data were plotted
as shown in Figure 19. As the graph indicates, the permeability of the
spent oil shale residues appears to decrease with time and this decrease
may be approximated by
k (cm2)
(USBM)
(45)
39
-------
10
Percentage
40 60
T
98%
i i i i I i i i I I I I i I I I I I I | i I i I i I I I I I I I i I I i I I i 11
2.0 1.5 1.0 .5 0 .5 1.0 1.5 2.0
Number of standard deviations
FIGURE IT: LOGARITHMIC PROBABILITY PLOT OF SIZE DISTRIBUTION OF
TOSCO SPENT OIL SHALE
40
-------
Percentage
2% 5 10 20 40 60 80 90
98%
E
u
a>
a>
"o
+-
w
&
2 1.5 I 05 0 0.5 I 1.5 2
Number of standard deviations
FIGURE 18: LOGARITHMIC PROBABILITY PLOT OF SIZE DISTRIBUTION OF
USBM SPENT OIL SHALE
41
-------
to
x 1.5
je
1.0
1.0
k=Q58§xffi±+2J7x|0.9
I I
1-0 I.I 1.2 1.3 1.4 1.5
3 4
t,
2
10 205Q
FIGURE 19: VARIATION OF THE PERMEABILITY OF TOSCO AND USBM SPENT SHALES WITH TIME
-------
and
k (cm2)
1Q
-10
(46)
in which
JL/ Z
= the number of half days from the start of the permeability test.
< CB
Tlie initial
Equations 45 and 46 are valid only for 1 < t^ /2
permeability, at t]/2 = 0, was determined by averaging the permeability
of three respective samples immediately after saturation.
A summary of the physical properties of the various oil shale residues
is given in Table V.
Table V: Physical Properties of the Various
Oil Shale Residues
Oil Shale Residue
geometric mean size, cm
geometric standard deviation
particle shape factor
bulk density, g/cc
solids density, g/cc
porosity
permeability, cm2
maximum size, cm
minimum size, cm
USBM
0.205
8.05
0.0526
1.44
2.46
0.41
3.46 10"9
<3.81
>0. 00077
TOSCO
0.007
3.27
0.097
1.30
2.49
0.47
2.5 10-10
<0.476
>0. 00077
UOC RAW
1.80
2.71 2.34
0.33
It is interesting to note that the eventual permeabilities of the USBM
and TOSCO residues are 63 and 91% respectively of their initial
permeabilities.
Additional permeability experiments were conducted on the TOSCO spent
shale because it was to be used in the pilot study. In particular, the
permeability of the TOSCO spent shale and its relationship to the degree
of saturation was determined.
To determine this relationship, it is first more convenient to express
relative permeability as a function of the capillary pressure, pc, and
43
-------
then using certain approximations obtain the desired relationship. For
P > PV,> this relationship is described by
K
rw
(47)
in which
p is in dynes/cm2,
p, is a parameter called the bubbling pressure and is the approximate
capillary pressure at which the nonwetting phase becomes continuous
throughout the media, dynes/cm ,
K is the relative permeability (dimensionless) and is the ratio of the
effective permeability for the wetting phase at some particular
saturation to the permeability at saturation, k.
A plot of the data from the capillary pressure - permeability experiment
is given in Figure 20. In Figure 20, y is the weight density of the
fluid in dynes/cm3. The equation of the line for p < p is
K
rw
205 19'5
P,
(48)
c
From Figure 20, the saturation value of p /y is 135 cm.
Using the relationships developed by Corey (26) and assuming the residual
saturation for this type of media is zero, equation 48 becomes
Krw " s' (49)
in which S = the degree of saturation, dimensionless. The saturation
moisture content of the TOSCO oil shale residue is 38% by weight (47%
by volume). S = 1 and Krw = 1 when the moisture content is 38% by
weight. The value of the exponent in equation 49 (3.34) indicates that
the pore size distribution is almost uniform [a value of 3 indicates a
completely uniform pore size distribution (26) ] .
Blender and Shaker Experiment - The chemical analyses of the blender
and shaker experiments are given in Tables VI and VII respectively. The
data given in each table represents the average of three samples.
The results indicate that the concentration of dissolved solids in the
filtrate from the blender experiment is slightly higher than that from
the shaker. However, allowing for differences in composition due to
sampling, it could be concluded that the blender and shaker experiments
yield a filtrate of approximately the same concentration and composition.
44
-------
1.0
0.8
0.6
04
0.2
O.I
0.08
0.06
0.04
0.02
0.01
0.008
0.006
0.004
0.002
0.001
1—I I I I I I—9
10
K
K
rw xp
= 19.5
i 11 I i
20 40 60 80 100 200 400 600
pr/r(cm)
c
FIGURE 20: CAPILLARY PRESSURE VERSUS RELATIVE PERMEABILITY FOR TOSCO
SPENT OIL SHALE
45
-------
Table VI: Results of the Blender Experiment
Sample
Raw
USBM
UOC
TOSCO*
pH
8.15
7.78
9.94
8.40
Conductance
ymhos /cm
@ 25°C
310
1,495
11,050
1,750
TDS (mg/A)
103°C 180° C
277 146
1,091 1,001
10,011 9,702
1,262 1,115
600° C
134
892
9,680
1,058
Concentrations (mg/A)
j ,i. ,i-i,
K+
24
72
625
32
Na
48
225
2,100
165
Ca
10
42
327
114
I 1
Mg ECO.,
J
1.0 75
3.5 38
91 28
27 20
Cl
2.2
13
33
7.6
_
so4
79
600
6,230
730
Table VII: Results of the Shaker Experiment
Sample
Raw
USBM
TOSCO*
pH
8.41
7.82
8.43
Conductance
ymhos/cm
@ 25 °C
300
1,320
1,640
TDS (mg/A)
@ 103°C
270
970
1,121
Concentrations (mg/A)
4.
10
4.
Na
206
i i
Ca
102
• . _
Mg HC03
31 20
Cl
5.8
_
so4
775
NO- concentrations for the blender and shaker were 5.6 and 5.1 mg/A respectively.
-------
Column Study - The first leachate from the column of TOSCO spent shale
took 2 weeks to move through the column after the water was first
applied. For the following 28 days, volumes of leachate were collected
at various time intervals until a total of 4.6 liters had been percolated
and collected. All samples were collected at the bottom of the column
because it was impossible to obtain a sample from the various taps that
had been installed in the column, (see Figure 6).
The volume and conductance of each leachate sample collected was deter-
mined. The major constituents and their concentrations for the first
eight samples were determined. These results are given in Table VIII.
A computer program was developed to calculate the concentration and
composition of the percolation water by the procedure described in
Section III. A simplified flow chart of the entire program is shown in
Figure 21, and Table IX gives the definitions of the terms used in the
program. An entire listing of the program as written for a CDC 6400
computer may be found in Appendix D.
The quantities required for the calculation are:
1, The initial concentration in moles per liter of the ions in the
solution which is to be percolated through the shale.
2. The concentrations of the exchangeable ions in the shale complex in
moles per gram of shale.
3. The concentrations of the soluble ions in moles per liter at a water
content of B, where B is the ratio of grams of oven dry shale residue to
liters of solution contained in the soil during percolation.
4. One half the summation of the product of the concentration of the
solution anions, excluding SO^, in moles per liter times the valence
squared, and the values of the exchange constants K and K' as found in
equations 24 and 25.
The following assumptions were made:
1. K4" was present in negligible amounts fsee Tables VI and VII),
2. Cl~ did not enter into the reactions,
3. CaSO^ was the only moderately soluble salt present,
4. Uniform physical and chemical properties of the shale,
5. The chemical reactions are adequately described by equations 24 and
25.
The experimental values and theoretical values versus volume leached are
shown in Figures 22 and 23. Although some segments of the calculated
curves deviate considerably from measured values, the overall agreement,
and in particular, the equilibrium values, is quite good.
The bottom row of figures in Table VIII are the expected eventual final
steady state concentrations that would be expected after a very large
quantity of water had passed through the spent oil shale residue. In
other words, one would not expect concentrations less than these,
regardless of how much prior leaching had occurred. It is clear that
47
-------
Table VIII: Experimental Results of the Percolation Experiment
Conducted on TOSCO Spent Oil Shale Retorting Residue
Volume of
leachate
sample (cc)
254
340
316
150
260
125
155
250
650
650
650
760
Total volume
of leachate
(cc)
254
594
910
1,060
1,320
1,445
1,600
1,850
2,500
3,150
3,800
4,560
00*
Conductance
of sample
(ymhos/cm @ 25° C)
78,100
61,600
43,800
25,100
13,550
9,200
7,350
6,825
5,700
4,800
4,250
3,850
1,800
Concentration (mg/£) of sample
Na+
35,200
26,700
14,900
6,900
2,530
1,210
735
502
-
-
-
-
86
Ca^
3,150
2,145
1,560
900
560
569
585
609
-
-
-
-
64
M ++
Mg
4,720
3,725
2,650
1,450
500
579
468
536
-
-
-
-
118
3E
S°4
90,000
70,000
42,500
21,500
8,200
5,900
4,520
4,450
-
-
-
-
740
Cl
3,080
1,900
913
370
205
138
138
80
-
-
-
-
11
oo
* These are extrapolated values and obviously were not actually observed,
are probably accurate to within ±6%.
These extrapolated values
-------
4-
|START I
|ACCEPT INITIAL VALUES|
1-
IF L=l GO TO 2
IF L=2 GO TO 10
2-
SCA=SCA+BCA
SMG=SMG+BMG
SNA=SNA+BNA
SS04=SS04+BS04
ANA=ANA
ACA=ACA
AMG=AMG
U=UUU
STOP
10-
|ENTER EQUILIBRIUM CYCLE|
3-
CALCULATE X-EQ. (27)
CALCUALTE Y-EQ. (26)
J=J+1
L=2
ACA=ACA(J)
ANA=ANA(J)
AMG=AMG(J)
AS04=AS04(J)
J=J-1
GO TO 3
9-I PRINT OUT CONCENTRATION OF IONS
SMG=SMG-BY
SNA=SNA-BX
SCA=SCA+B(X/2+Y)
ACA=ACA-(X/2+Y)
ANA=ANA+X
AMG=AMG+Y
J=0
SCA=SCA
SMG=SMG
SNA=SNA
SS04=SS04
SCA1=SCA+Z
SS041=SS04+Z ,
UH=[2(SCA1+SS041+SMG+0.25SNA)+U] '
11=1
8-
5-
CALCULATE W-EQ. (35)
IF 11=2 GO TO 7
J=J+1
ACA(J)=ACA
AMG(J)=AMG
ANA(J)=ANA
IF (J-I)
<0 GO TO 1
=0 GO TO 9
>0 GO TO 9
6-
AACA=SCA+W
SCA2=SCA+W
SS042=SS04
TJH= [2 (SCA2+SS042+SMG+0.25SNA)+U]
11=2
GO TO 5
1/2
|EXIT EQUILIBRIUM CYCLE
IF[ABS(SCA2-AAASCA)GT,10"5GO1T03
7-
IF [ABS(SCA2-AACA)GT.10"51 GO TO 4
I
SCA2=SCA+W
SS042=SS04+W
AAASCA=SCA+W
Figure 21: Flow chart for computer program.
49
-------
Table IX: Definition of Terms Used in Computer Program for
Predicting the Quality of Spent Oil Shale Percolation Water
Computer Symbol
Definition
SCA
SNA
SMG
SCL
SS04
ANA
ACA
AMG
CK1
CK2
CKS
BETA
BMG
BNA
BCA
BS04
uu
L, II, J
Concentration in moles/liter of Ca
water
in the applied
Concentration in moles/liter of Na in the applied
water
I j
Concentration in moles/liter of Mg in the applied
water
Concentration in moles/liter of Cl in the applied
water
Concentration in moles/liter of S0~ in the applied
water
Concentration in moles/gram of soil of exchangeable
Na on the shale
Concentration in moles/gram of soil of exchangeable
Ca"1"'1" on the shale
Concentration in moles/gram of soil of exchangeable
Mg"*"*" on the shale
Exchange constant in equation 25, K1
Exchange constant in equation 24, K
Solubility product of gypsum
The moisture content of the shale during percolation
I I
The concentration in moles/liter of soluble Mg
The concentration in moles/liter of soluble Na
I I
The concentration in moles/liter of soluble Ca
The concentration in moles/liter of soluble SO,
One half the summation of the product of the
concentration of the solution anlons, excluding
SO,
in moles/liter times the valence squared
Parameters to direct program
50
-------
4>
CO
ja>
o
E
c
o
0.3
0.2
O.I
0
1.5
1.0
o 0.5
c
QJ
I
O
2.0
1,5
1.0
0.5
0
I I
Observed
Calculated
Calculated
- \
V*—Calculated
0 400 800 1200
Mg-
SCu
3600
Volume leached (ml)
FIGURE 22: CAI£ULATED AND OBSERVED VALUES OF Na+, SO," AIO) Mg++ VERSUS
VOLUME OF WATER LEACHED
51
-------
150
10
b
100
.1 80
c
O
s
O
60
40
20
0
bserved
T.D.S.
' ' ' 1 L
0,1
en
0.06
i
O
0.04
0.02
0
bserved
calculated
iiit
0 400 800 2000
Volume leached (ml)
3600
FIGURE 23: CALCULATED AND OBSERVED VALUES OF Ca AND TDS VERSUS
VOLUME OF WATER LEACHED
52
-------
even these minimum possible values would be in excess of the maximum
permissible allowed for drinking water, namely 250 mg/£ for SOT and a
total dissolved solids (T.D.S.) concentration of 500 mg/fc. On the other
hand, the concentrations naturally occurring in the ground water in this
area are unknown. It is worth noting in this connection that 10% of the
salinity of the Colorado River at Hoover Dam is contributed by natural
point sources with an average salinity of 3,220 mg/£. Therefore, it is
not inconceivable that the natural ground water salinity in the oil
shale area may have a salinity as great as the percolation water from
oil shale residue. Examination of the records for the Colorado River
near Cameo, Colorado, for the water years (a water year begins October 1
the previous year and ends September 30 of the water year) 1964 through
1968 showed that the maximum annual dissolved solids concentration
observed occurred at the minimum annual flow rate observed, in general.
Low flows are usually composed primarily of ground water inflow, and
therefore low flow quality generally is representative of groundwater
quality. The values observed are as follows:
Date
September 1-30, 1966
January 1-17, 1968
March 1-23, 1965
December 13-26, 1966
December 1-31, 1963
Mean Dissolved Solids
Discharge Concentration in mg/Ji
(cfs) (residue at 180°C)
1,700
1,398
1,350
1,234
1,153
o*
691
875
876
956
970
1,600*
* extrapolated value (not observed)
Apparently the ground water salinity in the Colorado River Basin above
Cameo is somewhere between 970 and 1,600 mg/£ typically. Comparison
of these latter 2 figures with those on the bottom row of Table VIII
indicate that the ultimate effect on ground water salinity may not be
significant. On the other hand, the data in Table VIII show that the
initial effect of oil shale development (by surface retorting) on ground
water salinity may be substantial.
RAINFALL PILOT STUDIES
A total of 13 experiments were conducted on the rainfall-runoff facility.
Of these, 10 of the storms were simulated and 3 were natural. The total
water applied amounted to over 44 inches in a 130 day period or about
three years of precipitation for the oil shale area.
The inplace density of the surface of the shale (top 3 inches) for the
first 9 experiments was 86 lbs/ft3. For the last 4 experiments the
53
-------
shale was mechanically compacted to a surface density of 101 lbs/ft^.
The overall inplace density in both instances was about 55 lbs/ft3.
As developed in Section IV, the concentration of dissolved solids in the
runoff water may be described by equation 23. The values of the
exponents, constants, and various relationships were determined as
follows.
First, plots of total dissolved solids versus time for each simulated
run were made. The values of concentration at time equal zero were
extrapolated from each graph and plotted on semilog paper against Ato,
the change in moisture content from the saturated value that occurred as
the result of solar drying prior to the rainfall event. By selecting
the value of concentration in this manner, all independent parameters
except Aw were constant. The data yielded the following relationship:
4.5Aoi
exp 035
(50)
Secondly, the values of concentration were extrapolated_ to time «» on a
reciprocal time plot and these values plotted against D on log-log
paper. The plot gave a straight line of the form
C « — (51)
D
Next values of concentration observed after runoff equilibrium had been
reached (when runoff rate equals rainfall rate) were plotted versus time
on semi-log paper. A straight line was obtained when
(52)
035
Various relationships involving p, k, and concentration were tried and it
was found that C could best be related to p and k by
C « £ (53)
Combining equations 50, 51, 52, and 53 yields
.l/3\
P
C « -^- exp
kD2
4.5Au) - t
0.435
(54)
54
-------
Data from all simulated runs were then plotted as shown in Figure 24,
and a least squares analysis performed to give the relationship:
10~9 P . - ,,,,.
— exp - - (55)
in which C is expressed in mg/d of inorganic dissolved solids. In
Figure 24, r is the coefficient of correlation. In general, 0 <_ |r| <^ 1.
When r = 0, there is no correlation, and when |r| = 1 there is perfect
correlation. The value of r = 0.968, means that 96.8% of the observed
variance is explained by equation 55. In equation 55,
It is expressed in cm2,
D2 is expressed in cm ,
p is expressed in grams /cc, and
t is expressed in hours.
The permeability for the different compactions was calculated from
equation 44 using in-place values of porosity.
As with the bench top studies, the major constituents in the surface
runoff from the spent shale pile were Ca , Mg , Na , SO" and HCO~.
The composition of the runoff water also varies with time. In general,
at the beginning of a run, Na constituted the greatest portion of the
cations while near the end of the run, the runoff water was essentially
Ca"14" in cation concentration. The SO^ ion constituted the major part of
the anion concentration, with HCOI concentration essentially constant at
0.3 to 0.4 me/A.
Moisture content also affected composition. Drying of the shale surface
causes movement of water from the interior to the surface by capillary
action. On reaching the surface, the water evaporates leaving behind
a white deposit that is clearly visible on the black surface. This
deposit is dissolved during the rainfall with the result that both
concentration and composition of dissolved solids in the runoff water
vary with time and depend on the amount of drying prior to the rain.
Figure 25 is a picture of this deposit. From equation 55, it would
appear that compaction increases the concentration of dissolved solids
in the runoff because compaction increases p and decreases k. The rate
at which the deposit is formed therefore is clearly dependent on the
rate at which capillary action can carry the very concentrated solution
from the pores within the shale residue to the surface, because the
material can be evaporated more rapidly than it can be transported to
the surface by capillary action. If the reverse were true, compaction
would have no effect on runoff concentrations, because evaporation
would be the rate determining step. The maximum rate of capillary
movement can therefore be estimated from evaporation rates. In the area
where these experiments took place, the maximum evaporation rate from a
fresh water surface is about 9 inches per month (in July) . The
interstitial or microscopic velocity in the pores corresponding to this
rate is /T/e times the
55
-------
-2.0
+0.8
FIGURE 24: RELATIONSHIP OF TDS IN SPENT SHALE RUNOFF WATER TO
INDEPENDENT PARAMETERS
56
-------
FIGURE 25: SURFACE DEPOSIT ON TOSCO SPENT OIL SHALE
57
-------
aforementioned evaporation velocity, or I 9!_*• v2
37 inches per
\ 0.345
month or about 1 inch per day for fully saturated oil shale retorting
residual.
It should also be noted that the parameter t in equation 55 is the time
from the beginning of runoff from a given storm. While equation 55
indicates that if a storm were of infinite duration, the runoff concen-
tration would approach that of rainfall (all simulated and natural
rainfall concentrations were subtracted from observed runoff concen-
trations) , it does not mean that eventually, after several rainfalls,
very little would be leached. In fact, there was no observable decrease
in the leaching characteristics of the outdoor oil shale residue from
the first to the last experimental run, as Figure 24 clearly indicates.
Of course all the rainfall tests were conducted during one summer, and
many years of weathering might show an eventual decrease in leaching
characteristics.
If one substitutes equation 11 into equation 55, one obtains
1.14 x 1Q"8 p(Sin6)2/3 |4.5Aco - t1/3 \ ,,,.,.
^ exp -7- (55A)
or at 60°F,
IP"4 Q(Sin9)2/3 (4.5*0 - t1/3
kOLOiL)2/3 6XP l °'435
(55B)
From equation 55A, it would appear that the maximum concentrations in
the runoff will be found when compaction is greatest, slopes are steep,
drying has been extensive, runoff has just begun, the shale residue has
a low permeability, runoff water temperature is high, rainfall intensity
is low, and length of overland flow is short. However, it should be
noted that equation 55A is based on data from only one slope and only
2 different compactions. Further work should be done on the effect of
compaction and slope angle. The quantities having the greatest effect
on runoff dissolved solids concentrations are the extent of drying
before rainfall and time since runoff began. Next in effect are the
bulk density and permeability of the residue. Least effective (but
still very important) are slope of residue surface, runoff water
viscosity, rainfall intensity, and length of overland flow.
In order to get a feel for runoff concentrations, perhaps an example
calculation is in order. Assume: runoff water temperature = 60°F,
1 year storm frequency, 60 minute storm duration, Grand Junction
(Colorado) data applies, p = 1.63 g/cm3, slope « 0.1, AID - 0.202, runoff
has just started, k «= 6.11 x 10"11 cm2, and iL » 400 inch-feet per hour.
58
-------
From equation 12, i = 6.3 T°'265/(t + 1.7)°'721 = 6.3/(61.7)°'721 - 0.321
inches per hour, so that the length of overland flow is 1,247 feet.
Substituting these values into equation 55B, one obtains
(10"4X1.63X0.1)2/3 _ [(4.5X0.202)1
--exp
f(4.5)((
O.t
(6.11 x 10-n)E (10X400) ]2/3 [ °'435
18,300 mg/A.
Using equation 16, equations 11A and 55B can be put in terms of the
Reynolds number, R:
D * <10~4>
1/3
feet (11B)
and
c - 10"4 P(sin9)2/3 k.SAca - t1/3 ]
c -- _____ exp | Q^35 |
Ostensibly, the maximum value of R for which equations 11B and 55C are
valid is 4,000. Substituting this value into equations 11B and 55C,
one obtains
- „ L feet (110
(sin8)
and
c - 3.95 x IP"7 P(8in9)2/3 /4.5Aa - t1/3
C - exp
Figure 26 shows the variance of composition. To determine the percentage
of any of the major cations present in the runoff water, only the change
in moisture from saturation, Ato, and t, the time since beginning of
runoff need to be known. For the previous example ,4.5Aw - t^-'-* = 0.9,
so the composition is 9% Ca"*", 78% Na+, and 13% Mg44". From Figure 31,
18,300 mg/Jl is equivalent to 260 me/fi. of cations. Therefore, the
concentrations are 23 me/J, Ca44", 203 me/A Na+, and 34 me/A Mg44". The
concentration of SO^ then would be roughly 260 me/H.
To determine the composition of the deposit left on the shale surface a
sample of this white deposit was dissolved in a liter of water and an
59
-------
30 xlO"9 , ~~XA A~A A
X "A A "A X A
FIGURE 26: PERCENTAGE COMPOSITION OP CATIONS IN SURFACE RUNOFF FROM
SPENT SHALE AS RELATED TO INDEPENDENT PARAMETERS
60
-------
analysis made for the major constituents. The results are presented
in Table X:
Table X: Chemical Analysis of Surface Salt
Evaporation Deposit
Conductance
ymhos/cm
@ 25°C
28,500
Concentration (me/A)
„ 4- „ -H- _ ++
Na Mg Ca
580 80 10
SOI
740
As seen from Table X, the resulting solution was composed basically of
Na and SO? with Mg and Ca present in smaller quantities. Therefore,
it seems reasonable to assume that the initial runoff will be composed
primarily of Na+ and SO?, with Mg"*"1" present in greater concentrations
than Ca""".
Sediment - The yield of sediment from the spent shale piles is a complex
process responding to all variations that exist in precipitation,
vegetation, runoff, and topography. This study is aimed only toward a
discernment of gross sediment yields from such piles.
An Imhoff cone test was used to determine the amount of sediment
transported by the runoff water from the shale. Table XI gives the
amount of sediment transported in a three hour period by each storm.
Table XI: Calculated Sediment Yield in Three Hour
Period from Simulated Storms
Storm Sediment,
Ib
1 45.0
2 20.0
3 50.5
4 73.8
5 153
6 170
7 20.0
8 74.6
9 136
10 121
Accumulated
sediment (Ib)
45.0
65.0
115.5
189.3
342.3
512.3
532.3
606.9
742.9
863.9
Sediment Yield
Ib/(ft2)(hr)
0.0150
0.0067
0.0143
0.033
0.083
0.0484
0.0062
0.028
0.0454
0.0643
61
-------
To obtain an insight into how sediment yield might vary with the various
parameters, a modification was made of the extensively used equation
developed by Musgrave (28). This equation states
w, .1.75 1.35 T0.35
" 2 30 S
where
E' = rate of erosion (tons/acre),
2i_n = maximum 30 minute rainfall intensity of 2 year frequency
c2m
(30 + d)n
s = slope, and
L = length of slope (ft).
Assuming sediment yield varies inversely with relative moisture content,
a) ((J = w/io ), equation 56 becomes
1C f S
.1.75 1.35 0.35
E « I *— L (57)
in which E = sediment yield (Ib/hr-f t2) .
A plot of this relationship is given in Figure 27. Such a plot may be
used to give gross estimates of sediment yield from a particular storm.
In the aforementioned example, o> = 0.115 = 0.363, and
r 0.317
,1.75 s1.35 L0.35/u) . (0.321)1'75 (P.!)*'*5 (1,247)°-35 . 2>Q6 x ^
r U.363
Extrapolating Figure 27, one obtains a sediment yield of 0.1
lb/(hr)(ft2). The equation of the straight line in Figure 27 is
0.243 i°-" s°'764 L'/- (57A)
To obtain an idea of the size distribution of the sediment, a sample of
the sediment in the settling basin was analyzed. The results are listed
in Table XII.
The size distribution of the sediment in the runoff is arithmetically
normal with a mean size of 0.0317 millimeters and a standard deviation
of 0.0109 millimeters. Comparing this information and the data in
Table XII with the data in Table V, it appears that particles larger
than the geometric mean size of the oil shale retorting residue are not
62
-------
CM
•o
Q>
V)
I III I I I I I I
II I I I I I I
I III I I I I I
£ icr2 -
io-3
jl.73 gl.35
(in)1-75 (ft)c
FIGURE 27: SEDIMENT YIELD FROM SPENT SHALE FOR 3-HOUR PERIOD OF
SIMULATED RAINFALL
63
-------
Table XII: Size Distribution of Sediment in Runoff
Size (mm)
0.074
0.037
0.026
0.015
0.0074
0.0015
% Finer
100
68
30
6
2
1
subject to erosion by rainfall, at least under the conditions of the
test runs. On the other hand, approximately 1/2 of the residue (by
weight) would appear to be subject to erosion by runoff from rainfall.
The settling velocity of the mean size particles appears to be roughly
0.05 cm/sec (Stokes1 law range) at 10°C. Stokes law is
VQ = (g/18)[(s - l)/v]d2 (58)
s s
where any consistent set of units can be used. In the cgs system, v
is the settling velocity in cm/sec, g is the acceleration (the
acceleration due to gravity is 981 cm/sec2), ss is the specific gravity,
v is the kinematic viscosity in cm2/sec, and d is the particle diameter
in cm. At 85°F, the settling velocity of the TOSCO runoff sediment is
vo = 104d2 (59)
S
Therefore, in order to remove 99% of the sediment from the runoff,
without the aid of chemical coagulants, the overflow rate of the settling
basin must be less than about 10~^ cm/sec. On the other hand, if the
overflow rate is 10" ^ cm/sec, about 97% of the sediment will be removed
without the aid of chemical coagulants. In a basin of 12 feet deep,
with this latter overflow rate, approximately 10 hours retention time
will be necessary to remove 97% of the sediment in the runoff.
To determine whether or not this settling time can be reduced by the use
of a coagulant, a standard jar test was run on the sediment and
A12(SO^)3 • 14H20 used as a coagulant. The results of this test are
presented in Table XIII. These results indicate that Al2(SO^)3 • 14HoO
would make a good coagulant when applied in a concentration of 2.5 mg/fc
and at a pH of 8.3. Under these conditions, the apparent settling
velocity of the suspension is roughly 10~2 cm/sec.
64
-------
To account for the water, a water balance was made on each run. These
results are given in Table XIV.
The volume of water applied is the average rainfall intensity multiplied
by the time of application times the area of application. The volume
of runoff water was determined from the hydrograph recorded by the
stage recorder. If no allowance is made for evaporation, the volume of
water stored may be calculated by subtracting the amount of runoff water
from the amount of water applied. The observed volume of water stored
was obtained by multiplying the area of application by the average
rainfall intensity times the observed time for the surface of the shale
to become saturated. (At best, this observed time would be within ±10%).
As indicated by the data, the volume of water stored is small (2.0%).
This is also indicated by the change in moisture within the shale as
determined by use of a neutron moisture probe. The moisture content of
the shale as monitored for the first 43 days of the experiment is given
in Figures 28 and 29.
Thermistors located 60 feet downstream indicated temperatures within the
shale remained relatively constant between 20-24°C throughout the
duration of the experiments. However, the dark color of the spent shale
caused temperatures as high as 77°C to be measured on the surface.
Temperatures this high could be lethal to germinating seeds.
Minor Constituents - Chemical analyses for minor constituents were
performed on various samples. The maximum concentrations of these
constituents and the sources are given in Table XV.
Organic Analyses - Several samples were selected and analyzed for total
nitrogen and carbon. These results are given in Table XVI. The samples
were prepared by taking a 100 ml portion of the respective sample,
evaporating it at 80°C, and then analyzing the residue.
65
-------
Table XIII: Jar Test Data for Sediment in Runoff from Oil Shale Residue
Sample
1
2
3
4
5
6
Alum
mg/2.
0.5
1.0
1.5
2.0
2.5
3.0
pH
8.92
8.53
8.32
8.37
8.26
8.18
First
Floe
(min)
none
none
10
2
1.5
1.5
Description of water after given time
3 min.
dense
dense
smoky
smoky
fair
fair
5 min.
dense
dense
smoky
smoky
good
good
10 min.
dense
dense
smoky
fair
excellent
excellent
30 min.
dense
dense
fair
good
excellent
excellent
1 hr.
dense
dense
good
excellent
excellent
excellent
8hr.
dense
smoky
fair
excellent
excellent
excellent
36 hr.
fair
excellent
excellent
excellent
excellent
excellent
Table XIV: Water Balance Data for Simulated Rainfall
Run
1
2
3
4
5
6
7
8
9
10
Total
Volume of water
applied (ft3)
179
16Q
350
480
780
333
117
377
666
288
3,730
Volume of water
runoff (ft3)
173
148
350
500
755
320
120
352
658
279
3,655
Volume of water stored (ft3)
Calculated
+ 6
+12
0
-20
+25
+13
- 3
+25
+ 8
+ 9
+75
Observed
+ 9
+14
+10
+20
+19
+18
+ 9
+31
+16
+ 7
+153
-------
28 -_
» I I I I I I I i
60* Downstream
40'Downstream
c
c
20' Downstream
02468 10
14
18
22
26
30
34
38
42
46 50
Time (Days)
FIGURE 28: MOISTURE CONTENT OF SPENT SHALE VERSUS TIME AT 1-FOOT DEPTH
-------
00
60' Downstream
8
40* Downstream
20' Downstream
0
J L_JL
I I I
I I I
02468 10
14 18 22 26 30
Time (Days)
34
38
42
46
50
FIGURE 29: MOISTURE CONTENT OF SPENT SHALE VERSUS TIME .AT 1-FOOT,
6-II'CH DEPTH
-------
Table XV: Concentrations of Minor Constituents
Ion
*!**
B.++
Br~
Cl"
co=
F~
Fe^
l"
K+
Mn**
N0~
Pb"^
P°4
Zn++
Maximum
concentration observed
(mg/i) *
2.5
4.0
<0.1
3,000
21
<0.1
<0.l
3.4
1.7
0.16
1,100
<0.1
186
<0.1
35
2.5
Source
TOSCO
RAW
TOSCO
UOC
UOC
TOSCO
TOSCO
TOSCO
TOSCO
TOSCO
TOSCO
Test
column (first leachate)
blender
column (first leachate)
blender
blender
column (first leachate)
column (first leachate)
column (first leachate)
column (first leachate)
column (first leachate)
column (first leachate)
Maximum concentrations in actual runoff would be expected to be less
than 4% of these values. Therefore, none of the ions listed in this
Table XV would be present in runoff in concentrations greater than
allowable for human drinking water. In other words, the total maximum
concentration of all these ions in actual runoff would be less than
174 nig/A.
69
-------
Table XVI: Carbon and Nitrogen Content of
Selected Samples
Sample
TOSCO Spent Shale
BOM Spent Shale
Raw Shale
Sample from column after 1,320
ml of water had been
leached through spent shale*
Sample from column after 3,150
ml of water had been leached
through spent shale*
Sample taken 1/2 hour after
simulated 1" rainfall began
on first compaction*
Sample taken 1/2 hour after
simulated 1" rainfall began
on second compaction*
Values found,
% by r weight
Carbon
10.2
8.2
14.7
0.72
1.12
1.79
0.76
Nitrogen
0.38
0.28
0.39
0
0
0.79
0.44
Total residue,
mg/A
-
-
-
11,900
4,907
1,219
2,458
*TOSCO spent shale
70
-------
SECTION VII
DISCUSSION OF RESULTS
PHYSICAL TESTS
As shown in Figure 19, the permeability of both the TOSCO and USBM
residues decreased with time. The reason for the decrease in permea-
bility might be attributed to one or more of three phenomena:
1. movement of fines
2. swelling
3. precipitation of CaCO., or some other cementing compound.
To determine if the movement of fines might be the cause of the decrease
in permeability, each sample, on which the test was run, was divided in
half and a sieve analysis run on each section. The sieve analysis
showed an almost completely homogeneous sample with respect to size
distribution. Therefore, it was concluded that there was no significant
movement of fines in the samples. However, it is conceivable that a
short migration of the fines might block interstices and cause a
reduction in permeability. This latter possibility might not show up in
the 2 sieve analyses made.
These same samples were also closely examined under a microscope for
any noticeable deposition of a cementing compound. None was observed.
Therefore, the most likely reason for the decrease in permeability is
swelling of the shale. Because the chemical analyses of all experiments
indicate the presence of Na+, the swelling could be due to the hydration
effects of the sodium ion.
In order to put the results of the shaker, blender, and column
experiments on a comparative basis, the mass of the various ions leached
per 100 grams of TOSCO spent shale were determined. These results are
given in Table XVII. As indicated by the data, the amounts of the
various ionic species leached per 100 grams of shale residue are very
similar.
Table XVII: Mass of Various Ions (mg) Leached
per 100 Grams of TOSCO Spent Shale
Ion Shaker Blender Percolation
Ca^
Mg**
Na+
so"
Cl~
Total
102
31
206
775
5
1,119
114
27
165
728
8
1,042
64
40
258
675
18
1,055
71
-------
Also, for comparison with the bench-scale study results, blender
experiments were conducted on six surface soil samples, and chemical
analyses made of the major constituents found in 3 samples of surface
soil obtained in Parachute Canyon near Grand Valley, Colorado. These
results are given in Table XVIII. Figure B3 shows the approximate
locations of the surface soil samples. These soil and talus slope
samples were obtained April 3, 1970, and are as follows:
1. Surface soil east slope
2. from cut below #1, 10 feet of colluvial material (colluvial soils
are soils that contain sharp angular fragments of the rock from
which they originated)
3. nearly level surface soil
4. from cut on road, colluvial material about 10 feet below the surface
5. about 10 feet below surface colluvial material exposed in gravel pit
6. surface soil in pasture.
The data indicate a great deal of variation in the concentration and
composition of each ion found in the surface soil. However, the total
concentration of the ions is of the same order of magnitude as the
total concentrations determined in the blender experiments on the TOSCO
and USBM spent oil shale retorting residues. The composition of the
surface soil samples is shown in Figure Bl.
Table XVIII: Chemical Analyses of Filtrate from Blender
Experiments Conducted on Surface Soil Samples
Sample Conductance Concentration (me/£)
Number ymhos/cm — — -
@ 25°C Ca Mg Na+ K+ S0~ HCO~ Cl~ Total
1 865
2 1,610 9.70 3.08 1.33 0.12 12.80 0.72 0.24 +0.47
3 87
4 700 0.72 0.25 5.55 0.10 3.87 0.86 2.06 -0.17
5 1,870
6 148 0.85 <0.01 0.41 0.13 0.29 0.92 0.19 <-0.01
This seems reasonable because the local surface soil, for the most part,
is "weathered" shale. The weathered oil shale, because of surface
exposure over a long period of time, has lost most of its organic
content; consequently it is similar to the spent shales. However, a
direct comparison between surface soil and oil shale residue would only
have meaning if surface soil tests were performed on the runoff facility.
The results in Table XVIII can be compared with those listed in Table
VI.
72
-------
PILOT STUDY
As described in the preceding section, the concentration of TDS in the
runoff is given by equation 55, and the cationic composition in the
runoff is given in Figure 26.
To verify these results,, equation 55 and Figure 26 were used to predict
the concentration of Ca , Mg , and Na in the runoff water from the
three natural rainfall events (the first 3 tests) that occurred during
the course of this study. This was done as follows.
First a plot of conductance vs TDS was constructed for the assimilated
data, as shown in Figure 30. Thus, from the simple measurement of
conductance, an estimate of the TDS of a sample can be determined.
Next, a graph of TDS versus total me/A of cations was constructed as
given in Figure 31. Therefore, knowing the TDS as determined from
Figure 30, the me/A of cations can be estimated.
Using the value_of intensity as determined by the recording rain £age,
an estimate of D~ was made from equation 11. Using this value of D, the
quantity CD"2 k was calculated and the percentage composition of the
P
cations in the surface runoff determined from Figure 26. Because the
HCOo concentration is relatively constant (about 0.35 me/A), the S0=
concentration was calculated from anion-cation balance.
The results of the measured concentrations versus the predicted values
are given in Figure 32. The figure indicates the described procedure
may be used to give a reasonable prediction of runoff quality by
knowing only the conductance of the runoff water and the intensity of
rainfall.
The results of the column study indicated that the soluble salts would
be leached by percolation through the spent shale. This result could
not be verified in the pilot study because no percolation occurred
during the rainfall simulation.
In general, only minor fluctuations were observed in the moisture
content of the shale below the 9 inch depth. The greatest moisture
change occurred at the station located at the downstream end of the
facility. This should be expected because a greater portion of the
runoff passes this point.
73
-------
IT)
CVJ
O
.C
0>
O
u
•o
c
o
o
3200
2800 -
2400 -
2000 -
1600 -
Conductance -~ )=50 + l.028T.D.S.(mg/i)
1200 -
800 -
400 -
0 400 800 1200 1600 2000 2400 2800 3200 3600
TD.S. (mg/£)
FIGURE 30: TDS VERSUS CONDUCTANCE FOR SPENT SHALE SURFACE RUNOFF
74
-------
T.D.S.(mg/l)»t.7 + 70.5(meq/ICation$
r * 0.996
6 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72
Milhequivalents per Liter of Cations
FIGURE 31: TDS VERSUS MBft/L OF CATIONS FOR SPENT SHALE RUNOFF
75
-------
0
3456
Observed values (meq/i)
8
FIGURE 32: MEASURED CONCENTRATION VERSUS CALCULATED CONCENTRATION OF
CATIONS IN SURFACE RUNOFF FROM SPENT SHALE
76
-------
SECTION VIII
ACKNOWLEDGEMENTS
This paper Is based on research supported by the Environmental Protection
Agency. Five percent of the total project cost was provided by Colorado
State University. The extensive cooperation of the Union Oil Company,
the U.S. Bureau of Mines, and the TOSCO organization was indispensable
in making this study possible.
The authors of this report are Dr. John C. Ward, Gary A. Margheim, and
Dr. George O.G. Lof. Doctors Ward and Lof are Professors of Civil
Englneerina at Colorado State University. Mr. Margheim was a graduate
research assistant in the Sanitary Engineering Program of the
Department of Civil Engineering at Colorado State University during the
course of this project.
The first draft of the final report was sent to the following people for
review:
Don K. McSparran
Colony Development Operation (TOSCO)
Atlantic Richfield Company Operator
Harold E. Carver, Project Manager
Oil Shale Department
Union Oil Company of California
Gerald U. Dinneen, Research Director
Laramie Enerqy Research Center
Bureau of Mines
The authors wish to thank these gentlemen for their reveiw and many
valuable comments and suggestions.
The objective of this research project was to investigate the water
pollution potential of spent oil shale residues. Such research projects,
intended to study the causes, control, and prevention of water pollution,
are reguired by Section 5 of the Water Pollution Control Act, as amended.
This project of EPA was conducted under the direction of the Pollution
Control Analysis Section, Ernst P. Hall, Chief, Dr. James M. Shackelford,
Project Manager, and Fred M. Pfeffer, Project Officer.
77
-------
SECTION IX
REFERENCES
1. U.S. Bureau of Mines, Mineral Facts and Problems, Bulletin 585,
(1960), pp. 573-580.
2. Colorado State Board of Immigration, Mineral, Oil and Shale
Resources (1924), p. 13.
3. Duncan, Donald C., and Vernon E. Swanson, "Organic Rich Shale of
The United States and World Land Areas," Geological Survey Circular
523, (1965), p. 12.
4. Ibid (3), p. 5.
5. Throne, H. M., K. E. Stanfield, G. U. Dinneen, W. R. Murphy, "Oil
Shale Technology: A Review," Bureau of Mines Information
Circular 8216. (1964), p. 24.
6. Project BRONCO, A Joint Government-Industry Study of Nuclear
Fracturing and In-Situ Retorting of Oil Shale. Clearinghouse for
Federal Scientific and Technical Information, National Bureau of
Standards, U.S. Department of Commerce, Springfield, Virginia,
October 13, 1967, p. 4.
7. Prien, Charles H., "Oil Shale and Shale Oil," Paper presented at
Second Oil Shale and Cannel Coal Conference, Royal Technical
College, Glasgow, Scotland, July 3-7, 1950.
8. Ibid (6), p. 2.
9. Gambs, G. C., "Piower Plant Ash - A Neglected Asset," Mining
Engineering, Vol. 19, (1967), pp. 42-44.
10. Boccardy, J. A., and W. M. Spaulding, "Effects of Surface Mining on
Fish and Wildlife in Appalachia," Resource Publication 65.
Bureau of Sport Fishries and Wildlife, Washington, D. C.,
Government Printing Office, (1968), p. 4.
11. Dean, K. C., H. Dolezal, and R. Havens, "New Approaches to Solid
Mineral Wastes." Mining Engineering. Vol. 22, (1968), pp. 59-62.
12. Schmehl, W. R., and B. D. McCaslin, "Some Properties of Spent Oil-
Shale Significant to Plant Growth," Research Report to Colony
Development Company, Denver, Colorado, (1969), p. 11.
13. Linsley, R. K., Jr., M. A. Kohler, and J. L. Paulhus, Hydrology
for Engineers, McGraw-Hill Book Comapny, Inc., 1958, pp. 267.
79
-------
14. Fair, G. M., J. C. Geyer, and D. A. Okum, Water and Wastewater
Engineering, Vol. 1 - Water and Wastewater Removal, John Wiley and
Sons, Inc., 1966, Chapters 7 and 15.
15. Norton, Thomas N., "Cattle Feedlot Water Quality Hydrology,"
Masters Thesis, Colorado State University, March, 1969, p. 37.
16. Owen, W. M., "Laminar to Turbulent Flow in Wide Open Channels,"
Proceedings ASCE. 9 separate No. 188, April, 1953.
17. American Public Health Association, Inc., Standard Methods for the
Examination of Water and Wastewater, Twelfth Edition, New York,
1965.
18. Ibid (12), p. 18.
19. Margheim, G. A., "Predicting Quality of Irrigation Return Flows,"
Masters Thesis, Department of Civil Engineering, Colorado State
University, Fort Collins, Colorado, December, 1967.
20. The Asphalt Institute, "Soils Manual for Design of Asphalt Pavement
Structures," Manual Series No. 10 (MS-10), 2 edition, April, 1963,
p. 217.
21. Ward, J. C., "Stream Flow Quantity and Quality Correlations and
Statistical Analyses," Research Report No. 3, University of
Arkansas, Engineering Experiment Station, 1963, pp. 19-32.
22. Fair, G. M., J. C. Geyer, Water Supply and Wastewater Disposal,
John Wiley and Sons, Inc., New York, Seventh Printing, October,
1966, pp. 111-113.
23. Ward, J. C., "Turbulent Flow in Porous Media," Closure, Journal of
Hydraulics Division, ASCE, Volume 92, No. HY4, Proc. Paper 4859,
July, 1966, pp. 110-121.
24. Corey, A.T., Fluid Mechanics of Porous Solids, Colorado State
University, Fort Collins, Colorado, 1965, p. 48.
25. Ibid (23), pp. 110-121.
26. Ibid (24), pp. 50-57.
27. Margheim, Gary A., Dissertation, Colorado State University,
Department of Civil Engineering.
28. Musgrave, G. W., "Quantitative Evaluation of Factors in Water
Erosion - First Approximation," Journal Soil and Water Conservation,
Volume 2, No. 3, (1947), pp. 133-138.
80
-------
29. "The Environmental Aspects of a Commercial Oil Shale Operation,"
by J. S. Hutchins, W. W. Krech, and M. W. Legatski, AIME
Environmental Quality Conference, June 7-9, 1971, pages 59-68.
81
-------
SECTION X
SYMBOLS AND ABBREVIATIONS
a = subscript referring to absorbed phase, also ionic activity
B = ratio of grams of oven dry soil to liters of solution contained
in the shale during percolation, grams/liter
C = concentration, mg/£
c = coefficient in equation 12, also concentration in moles/liter
C = initial concentration of component j, moles/liter
C' = final concentration of component j, moles/liter
C = concentration of component j in segment k, moles/liter
I) = depth of flow at lower end of watershed, ft
D = mean depth of flow, ft
D = depth of flow at a distance x downstream, ft
x
d = parameter in equation 12
E = sediment yield in equation 57, lb/(hr)(ft2)
E* = rate of erosion, tons/acre
e = 2.71828... (base of Napierian logarithms)
f = activity coefficient, dimensionless
g = acceleration of gravity, 32.2 ft/sec2 or 981 cm/sec2
i = intensity of rainfall, in/hr
i - maximum 30 minute rainfall of 2 year frequency, in/hr
£* j \J
j = subscript referring to jth component
K = constant used in equation 21 (4.5), equilibrium constant in
equation 24, and constant characteristic of the cross-section
of flow through porous media (2.36) in equation 44, dimensionless
in all 3 cases
K1 = equilibrium constant in equation 25, / moles/liter
K = relative permeability (ratio of the permeability at a given
rw saturation to the permeability at 100% saturation), dimensionless
K » solubility product for CaS04 » 2.4 x 10"° (moles/liter)2 at 25°C
k « permeability, cm2 or subscript indicating particular segment
L - total length of overland flow, ft
M « geometric mean size of the particle size distribution, cm
O
m «= exponent in equation 12
N » exponent in equation 20 - 2
n = exponent in equation 12
p » pressure, dynes/cm2, and fraction by weight of sample retained
between adjacent sieves, dimensionless
p = bubbling pressure, dynes/cm2
p - capillary pressure, dynes/cm2
c
q = discharge per unit width, ft2/sec
q - equilibrium discharge per unit width, ft2/sec
a - volume of solution present in kth segment, liters
*\r
83
-------
q = discharge per unit width at a distance x downstream, ft2/sec
R - Reynolds number, dimensionless
S = degree of saturation, dimensionless
s = slope (dimensionless) or subscript denoting solution phase
T = tortuosity of porous media (2), dimensionless, and frequency of
occurrence, years
t = time since beginning of runoff in hours (equation 18)
t, = duration of storm in minutes (equation 12)
a
t,/p = time in half days, days/2
v = macroscopic velocity in porous media, cm/sec
V = velocity of overland flow, ft/sec
V = bulk volume of shale, cm^
V = void volume of shale, cm
V ++ =
w = moles per liter of Ca or SO, that precipitate
X = moles of Na per gram of shale entering exchange complex,
moles/gram, and the geometric mean size of the sieve openings
between adjacent sieves, cm
x = a horizontal coordinate, feet
x = geometric deviation, dimensionless
I |
Y = moles of Mg per gram of shale entering exchange complex,
moles/gram
y = total depth of shale and a vertical coordinate, feet
Z = moles of Ca per gram of shale entering exchange complex,
moles/gram
e = porosity, dimensionless
6 = slope angle, degrees or radians
V = viscosity, poises = gram/(cm)(sec)
v = kinematic viscosity, ft2/sec and cm2/sec = stokes
£ = exponent in equation 47
p = density, grams/cc
CT = geometric standard deviation, dimensionless
O
<|> = shape factor, dimensionless
u) = moisture content by weight, dimensionless
u = relative moisture content (ratio of moisture content to moisture
content at saturation), dimensionless
w = saturation moisture content by weight, dimensionless
s
D the bar over a symbol indicates the mean value
« proportional to
> greater than
< less than
I, summation
A finite difference
-------
» infinity
d derivative
/ integral
°C centigrade degrees
cc cubic centimeter
cm centimeter
ft foot
g gram
hr hour
in. inch
£ liter
Ib pound
In base e logarithm
log base 10 logarithm
me milliequivalent
mg milligram
ml milliliter
mm millimeter
No. number
sec second
IDS total dissolved solids
TOSCO The Oil Shale Corporation
UOC Union Oil Company
USBM United States Bureau of Mines
85
-------
A.
B.
C.
D.
SECTION XI
APPENDICES
Chemical Analysis with Specific Ion Activity
Electrodes
Figure Al: Conductance Versus Calculated
Ionic Strength
Figure A2: Ionic Strength Versus Activity
Coefficients
Soil Data . . ,
Figure Bl:
Figure B2:
Figure B3:
% Composition of Cations and Anions
in Filtrate from Blender Experiments
Conducted on Surface Soil Samples
Moisture Calibration Curve
Approximate Location of Surface
Soil Samples
Experimental Data ,
Table Cl: Experimental Test Conditions
Computer Program ,
Page No.
89
91
94
95
95
96
97
99
99
111
87
-------
APPENDIX A
CHEMICAL ANALYSIS WITH SPECIFIC ION ACTIVITY ELECTRODES
Direct potentiometric measurement of ion activity is based on the fact
that definite energy levels exist between two different states of the
same matter; and that these differences are proportional to the relative
populations of the atoms or ions involved. In electrolyte solutions,
these energy level differences are measured as an electrical potential.
The Nernst equation of classical thermodynamics expresses this potential
for a given activity of an ion relative to a standard state as follows:
E = E0 - [RT/(zF)]ln a (Al)
where
E = potential observed for any given activity of a particular ion,
volts
E = potential of the standard state (a = 1 mole/liter), volts
o
R = universal gas constant = 8.314 joules/(°K)(mole)
T = absolute temperature in degrees Kelvin
z = number of electrons transferred in the reversible reaction
= equivalents per mole (for specific ion measurements, z takes the
sign of the ion and is simply the valence of the ion)
F = Faraday constant = 96,500 coulombs per equivalent
and
a = the activity of the particular ion to be determined, moles per
liter.
Measurement of ion activity is accomplished with the electrodes by
determining the potential that is developed between the test sample and
the special filling solution inside the electrode. The Nernst equation
predicts that at 25"Centigrade the potential for a monovalent sensing
electrode will change approximately 59 millivolts for each decade change
in ion activity, while for a divalent sensing electrode the change is
29.5 millivolts per decade change in ion activity, etc.
In order to determine the concentration of a particular ion it is
first necessary to prepare a working curve on semi-logarithmic graph
paper by plotting the electrode potentials given by a standardizing
solution on the linear axis. The log axis represents the activity of
the particular ion in solution.
The following procedure was developed to measure the concentration of
a particular ion in a solution. The concentration of a particular ion
is given by
a - fc (A2)
89
-------
where
a = the activity of a particular ion, moles/liter
c = the concentration of a particular ion, moles/liter and
f = the activity coefficient, dimensionless.
Because the electrode measures activity, it is necessary to determine
the activity coefficient in order that the concentration may be obtained.
Because the activity depends upon the total ionic strength, it is
necessary to relate the ionic strength to some easily measured parameter.
During the course of this project, an empirical relationship between
ionic strength y and conductance of aqueous solutions contacted with oil
shale retorting residues was determined. For y in moles/liter, the
relationship is
y - (2.5 x 10~7) (conductivity)1'57 (A3)
where the conductivity is in ymhos/cm at 25°C. A plot of equation A3
along with the experimental data from which it was determined is shown
in Figure Al.
The activity coefficient may be calculated from the Debye-Huckel formula,
, . 0.5 z2 /\i ,.,,
logiof --- 7—7^ (A4)
1 + vy
in which z denotes the valence of the ion. Equation A4 is valid for
y ^ 0.1 moles/liter.
Knowing the above relationships, it is a simple matter to estimate the
activity coefficient by simply measuring the conductance of a given
solution. When the analysis is complete, the true ionic strength may be
calculated from
U = Z c z2 (A5)
where c. is the concentration of the ith type of ion, z. is its valence,
and the summation is carried out for all types of ions, both positive
and negative, in the solution (n is the number of different ions in
solution) .
If the agreement between the assumed ionic strength and the calculated
ionic strength is good, no adjustment in the activity coefficient is
necessary. If there is a large difference between the assumed and
calculated ionic strength, the activity coefficient must be adjusted
until the agreement is tolerable.
90
-------
10
-I
CO
£
o
E
o>
o
o
10s
/i=2.50X|0~7 (Conductivity)1-57
I I I I I I I I
200
400
I03
4000
Conductance (/^mhos/cm)
at 25° C
FIGURE Al: CONDUCTANCE VERSUS CALCULATED IONIC STRENGTH
91
-------
Using the subscripts 1, 2, and 3 to denote, respectively, singly, doubly,
and triply charged ions, one can rewrite equations A4 and A5 respectively
as follows:
- log f.
0.5
1 +
(A6)
and
yc = | [I GI + 4 Z
9 £ c3] .
(A7)
Using equation A2, equation A7 may be rewritten as
~ n I f~ " Si-t * .c "
Clearly f;L =
1/4 _ £ 1/9
2 V °3
(A8)
, so equation A8 may be rewritten
a1 4 S a7 9 £ a«
_Ji + —L + ^.
(A9)
Simultaneous solution of equations A9 and A6 gives the true value of the
ionic strength y providing that the activities of all the ions in
solution have been determined. Short of a complete chemical analysis,
equation A3 is most useful. The usual case, however, is that some of
the activities are known and some of the concentrations are known. In
this latter case, equations A7 and A9 may be combined to give
y = yc + ya (A10)
where y is a constant for a given solution and y is a function only of
f- for a given solution. Consequently, simultaneous solution of
equations A10 and A6 gives the true value of the ionic strength y. The
range of plotting values of f
for which equation A6 is valid is 0.76 to
1.
A suggested systematic procedure for obtaining the true value of y with
as little effort as possible is as follows:
(1) use equation A3 to estimate y,
(2) use equation A6 to calculate f1,
(3) calculate y using equation A7,
(4) estimate y using equation A9,
(5) estimate y using equation A10,
(6) if the y obtained in step 5 is the same as the y obtained in step 1,
no further work is necessary,
92
-------
(7) if the condition in step 6 does not hold, then the y obtained in
step 5 should be used to calculate a new value of fn using equation
A6, and steps 4 through 7 repeated until equations A10 and A6 are
both satisfied by the same values of f. and y.
An alternate graphical procedure would be to plot equation A10 on a
graph on which equation A6 is already plotted. Because the plot of
equation A6 is the same for all solutions, a master graph could be made
using 3 cycle semi-log graph paper with log y plotted versus f . At
any rate the correct values of y and f- are those where equations A6
and A10 intersect on the graph. A master graph such as that mentioned
above is Figure A2. Also plotted on the same graph are values of
fl - f2 and fl ' f3 '
93
-------
0.3
o
o>
2
*»»
OT
O
o
0.001
0.0009
0.0008
0.0007
0.0006
0.0005
0.0004
O.OOO3
0.0002
0.0001
f2= ff , Divalent Activity Coefficient
0.4 0.5 0.6 0.7 0.8
0.9
I - 1 - 1 - 1 T I
*3=f? lflfl , Trivalent Activity Coefficient
0.2 0.4 0.6 0.8 1.0
1.2
0.65 0.70 0.75 0.80 0.85 0.90
f|, Monovalent Activity Coefficient
0.95
1.4
0.01
0.009
0.008
0.007
0.006
Figure A2: Activity Coefficients as a Function of
Ionic Strength. (Equation A4)
94
-------
APPENDIX B
SOIL DATA
,- ,
YWYV
/WXA/VW
Co** or HCO;
Figure Bl: % Composition of Cations and Anions in Filtrate
from Blender Experiments Conducted on Surface Soil Samples.
Triangles are cation compositon (height of triangle indicates %K ), and
circles are anion composition. Numbers are sample conductance in
umhos/cra at 25°C. Straight line is composition of runoff from oil
shale retorting residue. Arrows indicate the direction in which
composition changes during a given rainfall.
95
-------
c
3
o
a>
>
1.20
1.10 -
1.00
0.90
0.80
0.70 -
0.60-
0.50
0)
o:
0.40
0.30
0.20
0.10
Model
Sealer 200R (Troxler)
Probe I04A
Source Am. Be.
Standard count 45,655 per minute
Special Acess tubing
l.902o.d., 0.049 wall
Seamless steel
Porosity 0.54
Bulk density I.l3g/cc
Date 3-10-70 (G.A.M.)
0 2 4 6 8 10 20 30
Moisture (percent by dry weight)
FIGURE B2: MOISTURE CALIBRATION CURVE
96
-------
vD
-g
:
W <-o->- E
S
Figure B3: Approx-
imate Location of
Surface Soil Samples
(Garfield County,
Colorado).
Explanation on
following page.
Each N x E square
is 36 square miles
(1 inch = 4.5 miles)
LCQUN
-v
-------
The oil shale mines at N = 8.1, E = 6.6 (8,000 feet) and at N = 8.6,
E = 4.9 (7,300 feet) should be noted. The Bureau of Mines Experiment
Station (N = 7.8, E = 6.8) appears to be at an elevation of about 5,700
feet. The coordinates of the TOSCO offices and experimental plant
are roughly N = 9.2, E = 4.8 and N = 9.1, E = 5 respectively. The
coordinates at which the 6 samples were taken were approximately as
follows:
Sample Coordinates
Number N E
I 974 479-
2 9.4 4.9
3 8.8 4.9
4 8.9 4.8
5 8.4 4.8
6 7.4 4.9
98
-------
APPENDIX C
EXPERIMENTAL DATA
The experimental data for the experiments on the CSU rainfall-runoff
facility are summarized below. It is of interest to note that the
dissolved solids removed during a given test, F, in lb/(hr)(103ft2) was
directly related to Ao> by the following empirical equation:
log1Q F = - 1 + 3.7Ao> (Cl)
However, equation Cl is strictly limited to the CSU rainfall-runoff
facility and can not be applied elsewhere. Therefore, it is simply a
qualitative expression that shows that F is directly related to Aw, and
is in fact the basis for the formulation of equation 21.
The test conditions are summarized in Table Cl.
Table Cl: Experimental Test Conditions
Tests
1-6
7-10
Table V
V
g/cm3
1.39
1.63
1.30
e
0.442
0.345
0.47
ws
0.317
0.212
0.38
k x 1010,
cm~
1.42
0.611
2.5
W
0.115-0.256
0.11 -0.165
i,
in./hr
0.46-2.25
0.40-2.12
p is the residue bulk density, e is the residue porosity by volume,
ws is the saturation moisture content by weight, k is the residue
plrmeability, a) is the observed moisture content by weight (Aw - iog - u) ,
and i is the rainfall intensity.
In the following tables, the total me/A = me/A cations - me/A anions.
Also the total mg/A = mg/A Na+ + rag/A Ca^ + mg/A Mg++ + mg/A
mg/A HCO-.
99
-------
TEST:1
Ps - 1. 39 grams/cc
i = 0. 54 in/hr
Dissolved Solids mtlHequlvalent/1 / mg/1
100
-------
TEST: 2
pg - 1. 39 grame/cc
1 -0.46 in/hr
k s 1.42 x 10"10cm4
Solids mtlllequivalent/1/ mg/1
101
-------
TEST: 3
Ps - 1. 39 grams/cc
S " 31. 7
r\s « 0.442
"J = 23. 0$
Auu = .08T
t =1.00 In/hr
k = 1.42 x 10"l°cm2
Sample
At
Hours
since
runoff
begin
'onduct-
ance
imhos/cmlSedlment
PH
251
Na
Dissolved Solids mtlUequivalent/1 / mg/1
Ca
Mg
SO.
HCO,
Total
3-2
0.0833
4.4
7. 74
3-3
0.250
470
4.1
7. 88
3-4
0.583
248
3.5
7. 96
3-5
1. 083
163
3.3
8. 21
3-6
1.583
125
2.9
8. 19
3-7
2.583
106
2.8
8.46
3-8
3.583
88
1.6
8.52
102
-------
TEST: 4
Ps - 1.39 grama/cc
cus - 31.7 %
"J - 16. 1%
AUJ = .156
i =1.70 in/he
k = 1.43 x 10"10cm2
Dissolved Solids milliequivalent/1/ mg/1
* Hefere to station 30 feet downstream
** Refers to station 50 feet downstream
103
-------
TEST: 5
ps = 1. 39 grams/cc
ujs = 31.7"(a
«\3= 0.442
u>= 11.5 %
Aix> « .202
1 = 2.25 in'/hr
k = 1.42 x 10"l°cma
Solids milliequlvalent/l/mg/1
* Refers to station 30 feet downstream
•* Refers to station 50 feet downstream
104
-------
TEST: 6
p « 1. 39 grams/cc
LUS« 31.7%
18.7*
• -130
i = 0. 94 in/hr
k = 1.42 x t
^0.43
26.2
,+0.08
117.!
6-7
2.00
125
6.8
8.15
0. 30
6.9
0.71
14.2
0.16
S
1.9
0.77
37.0
0.41
25. 0
-0.01
85.0
6-8
2.25
•or
121
3.8
8.20
4.8
13.8
1.9
31,7
25, 0
77,2
6-9
3.00
115
3.4
8.07
0.56
26.9*
0.38
23.2
. 05
67.0
105
-------
Sample
TEST:7
fta - 1.63 grome/cc
uua = 21.2 *
»\s = 0.34S
uu = 15.4
£10 = 8.056
i = 0.40 in/hr
k = 6.11 x 10 cm
At
Hours
since
runoff
begin
Conduct-
ance
imhos/cmpediment
g/1
pH
251
Na
Dissolved Solids mtlUequlvalent/1 /
Ca1
Mg
SO.
HCO,
Total
7-1
16.3
X1-19
7-3
0. 333
JO. 1
1560
3.2
8. 11
232
7-4
0.583
855
2. 3
8.13
7-5
1.08
487
2.0
8.36
7-6
1. 58
372
2.0
8.42
7-7
2. 08
323
1.5
8.22
7-8
2.58
278
2.0
106
-------
t M.20 in/hr _
k =6.11 x 10 cm
Solids milliequivalent/l/ mg/l
107
-------
TEST:9
Ps- 1.63 grams/be
i = Z. 12 in/Ur
Solids milliequtvalent/l/ mg/l
108
-------
TEST: 10
Ps = 1.63 grams/ce
s « 21.2 #
l\a « 0. 345
uu = 16.5$
Aiw = O.OV?
i = 1.72 in/hr _
k =6.11 x 10 cm
Sample
At
Hours
since
runoff
begin
Conduct-
ance
.imhos/cm|Sedin)ent
25 "t
PH
251
Na
Dissolved Solids milliequivalent/1 / mg/1
CaT
Mg
SO.
HCO,
Total
10-1
473
12. 0
6.26
0.57
s
13.2
2.87
59.5
0.69
3. 35
8.4
161
0.42
X
25.6
-fO. 46
267.7
10-2
0. 150
643
7.8
8. 07
1.56
s
19.0
.7.65
368
,0.37
22. 6
-0.21
544.5
10-3
0. 350
322
6.1
8.02
. 70
16.1
p. 37
132
22.6
-0. 14
210.1
10-4
0.600
0. 15
211
5.6
8.15
3.4
1.09
S.
21. 8
0.41
S
25. 0
+0. 01
S,
99. 1
10-5
0.850
0. 11
203
5. 3
8.21
2.6
1.05
21.0
0.26
3.2
1. 00
S
48.0
0.36
S
22.0
+0. 06
S
96.8
10-6
1. 350
J3. 08
189
3. 1
8. 19
1.8
1.02
20.5
0.20
2.4
0. 78
S
37.1
10-7
1.85
179
2.7
8.24
v<0. 05
0.00
18. 0
0.11
1.3
0.68
32.6
0.42
25.6
-0.09
S
77.5
109
-------
APPENDIX D
COMPUTER PROGRAM
This is the computer program used to predict the concentration of Mg
804, Na , Ca , and total dissolved solids (see Figures 22 and 23)
in the percolation water (for a CDC 6400 computer), as discussed in
Section VI.
Ill
-------
PROGRAM MA IN ( IMPUT , WJTPUT . T APF5 - (M^UT , TAPE6=OUTPUT>
DTMF.NSTON C< 1] > «RRf ) 0 ' *R I ' I 0 ) . R I ( 4 ) .R?f'4) .DCA(50> .D'-ntSO) .HNfl(50)
PEAD(5'.?0) SCA.SM(J.SNA.SSO/'»SCL«f:Kl «CK?.CKS
lo TORMAT (ft^io.g)
READ (5. 31) ACft.AMG. AM A. BMft.BCa . BMA . PSO4 .iJU.Ullli
3!
MM=0
J=0
SSCA=SCA
SSMG=SMG
55504=5504
SANA=ANA
SCA=SCA
SNA=SNA
SMG=SMG
U=UU
L^2
13 GO TO (201.202) «L
202 5CA=SCA+BCA
SMG=SMG+BMG
SNA=SNA+BNA
5504=5504+6504
ACA=SACA
AMG=SAMG
U=U»UUU
13 CALL EQUIL (SCA.SMG.SSOA. SNA. ACA.AMG. ANA, CKS»U.CK1 .CK2, BETA)
J=J»1
OCA(J)=ACA
OMA(J)=ANA
DMG'J>=AMG
IF(J-I) 12.10.10
10 WRITE»6-87) MM.SCA-SMQ.SSOA «SNA
87 PORMAT(lH0.13.20Xv-SCA*?6X^SMG-;i27X-*SSn4^26X«SNA<>/(4F30. 10) )
MM=MM*1
J=0
SCA=SSCA
SMG=SSMG
SNA=SSNA
5S04=SSS04
u=uu
201 J=J+1
IF(MM-IO) ?3-?3.SJ
23 COMTrNUF
L=l
ACA=OCA(J)
AMG-DMG' J)
ANA=DNA(J>
J=J-1.
GO TO 13
51 CONTINUE
FMD
112
-------
SUBROUTINE EOlHL(SCA.SNG.SS04.SNA.ACA.AMG«ANA.CKS«U»CK1«CK2.BETA)
DIMENSION C(l1).RR(10>-RI(10).Rl(4).R?(4)»DCA(50)»OMG(50)»DNA(50)
IFLAG=0
C (?)=-! 000. 0*BETA*CK2**2*(BETA»ACA*BETA»AMG*0.5*SNA>-0.5»BETA
C<3)=1000.0»BETA»CK2**2*<8ETA*ACA**2«-2.0*SNA»ACA*2.0*8ETA»ACA»AMG-»
72.0»SNA»AMG+BETA»AMG*»2)-SCA-SMG-BETA*ANA+250.0*CK2»*2«SNA«»2
Cf4)=-2000.0»BFTA»CK2*»2*(SNA»ACA»»2+2.0*SNA*ACA*AMG*SNA*AMG«»2)-A
ftNA*(2.0»SCA+2.0*SMG+0.5*BETA«-AMA)-JOOO.O*CK2**2*SNA»*2*(AMG+ACA)
C<5)=1000.0*CK2*»2«SMA»*2*(ACA»*2+2.0»ACA*AMG*AMG*»E>-ANA*»2*»25X*Rim*/<2F30.10> )
IF(ABS(RHI)) .GT.l .E-05) GO TO 50
Rl (N)=RRd)
50 CONTINUE
IF(N.LT.l) GO TO 500
IFdFLAG.GT. 0) GO TO 200
IFLAG = 1
XMIN=100.0
DO 70 1=1. N
70 XMIN = AMIN1 (XMIN.RUI))
X=XMIN
HOLD = X
GO TO 300
200 XMIN=100.0
DO 90 1 = 1 'N
IF(ABS(R1 (I)-HOLD) .LT.XMIN) GO TO 80
XMIN •= XMIN
GO TO 90
80 XMIN=ABS(R1 (I)-HOLD)
M=I
90 CONTINUE
X=R1 fM)
WRITE(6,48)X
48
HOLD=X
300 CONTINUE
A = 8ETAM1 .-CK1>
B=SCA+BETA*(AMG+CK1^ACA-0.5*CK]»X+0.5*X) *CK1*SMG
V=SCA«AMG-SMG*ACA*CK1 »0.5»BErA»i(*AMG*0.5»CKl*X*SMG
ARG= B**2-A.*AttV
Y=(-B+SQRT(ARG)
SMG=SMG-BETA*Y
SNA=SNA-BETA*X
ACA=ACA-(X/2.0»Y)
AMG=AMG»V
ANA=ANA+X
z=o.
4 SCA1=SCA*Z
UH=SQRT(2.»(SCA1*SS041*SMG*0.25»SNA)*U)
11 = 1
113
-------
ion 9B=SCA*SS04
CC=SCA»SS04-CKS*.EXP(9.336*UH/n . +UH)
Z=<-8B+SQRT(8B*BB-4.*CC) >»0.5
I^UT.EQ.21 GO TO 102
101 AACA=SCA+Z
SCA2=SCA+?
UH=SQRT (?.•
11 = 2
no TO 100
102 SCA2=SCA»Z'
IF(A8S(SCA2-AACA) .GT. 1 .F-05) GO TO
SCA=SCA+Z
17 IF(ABS=0.
1 = 1-1
9 CONTINUE
10 AXR=0.8
N3=l
A|_P1R =
ALP1 I=AXI
M=l
GOT099
M BET1R=TEMR
BET1T=TEM1
AXRiO.85
ALP2R=AXR
ALP2I=AXI
M=H
GOT099
1?- BET?R = TEMR
BET2I-=TEMI
AXR=0.9
ALP3R=AXR
ALP3I=AXI
M=3
60T099
13 B£T3R=TEMR
BET3T-TEMI
114
-------
TE2=ALP1I-ALP3T
TE5 = AI.P3R-ftLP?P
TE6=ALP3J-ALP2I
/TEM
5-TF.l"TE6)
TE7=TE3+1.
TE1=TE9*8ET2R-TE10K-BET2I
TE2=TE9»-BET2T*Tpi 0*BET?R
TE13=TEl-BETlR-rE7<;8ET3RME10»eET3I
TEM=SG)RT
ir(TEl)113. 113.112
113 TE4=SQRT (.5»(TEM-TE1)
GO TO HI
112 TET=SQRT (.5*(TEM*TE1»
jraE2> no. 200.200
110 TE3=-TE3
200 TEA=.S*TE2/TE3
111 TE7=TE\3+TE3
T£9=TE13-TE3
TE1=2.»TE15
TE2=2.*TE16
TE7=TE9
TE8=TE10
E05 TEM
TE3=(TE1»TE7+TE2»TE6) /TEM
AXR-ALP3R»TE3*TES-TE«»TE6
GO TO 99
15 N6=l
38 TF(A8S (HELL)»A8S (BELL > -1 -E-l 0 ) 18. 18. 16
16 TE7=ABS
-------
ALP1R=ALP?R
ALP1 I=ALP2I
ALP2R=ALP3R
ALP2I=ALP3I
ALP3R=ALP4R
ALP3J=AIP4T
BETtR=BET2R
BETH=BET2J
8ET2R=8ET3R
BET2r=BET3T
BET3R=TEMR
BET3I=TEMI
IF 14. 18,18
18 N4=N^+1
ROOTR(N4)=ALP4R
ROOT I (N/O
N3=0
37 RETURN
30 IFfABS (ROOTnN41)-l.E-5)]0.1001
31 GO T0'32»10) ,L
32 AXR=ALP1R
AXT=-ALP1T
ALP1I=-ALP1 T
M=5
GO TO 99
33 8ET1R=TEMR
BET1T=TEMJ
AXR=ALP2R
AXI=-ALP2I
ALP2I=-ALP2T
M=6
GO TO 99
34 BET2R=TCMR
BET2T=TEMI
AXR=ALP3R
AXI=-ALP3I
AUP3I=-ALP3T
L=2
M=3
99 TENR=COE<1>
TEMT=O.O
1301001 = 1. Nl
TE1=TEMR*AXR-TFMT*AX T
100 TEMR= TEl+COEd + M
HELL=TFMR
BELL=TEMI
42 IFfN4)102-103.10?
102 001011=1. N4
TEM1=AXR-ROOTR( J)
TEM2=AXI-ROOTI (T)
TEl=T£M1<»TEMl+T£M2«rEM2
TE2=
-------
1
Accession Number
5
2
Subject Field &. Group
05E
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
Sanitary Engineering Program, Department of Civil Engineering, Colorado State
University, Fort Collins, Colorado 80521
Title
WATER POLLUTION POTENTIAL OF SPENT OIL SHALE RESIDUES
10
Authors)
Ward, John C.
Margheim, Gary A.
L8f, George 0. G.
16
Project Designation
EPA, WQO Grant No. 14030EDB
21
Note
A companion report to this report entitled,
"Water Pollution Potential of Snowfall on Spent
Oil Shale Residues," is in preparation.
22
Citation
Water Pollution Control Research Series, Water Quality Office, Environmental
Protection Agency, 1971
23
Descriptors (Starred First)
*Salinity, *0il Shale, *Colorado River, *Colorado, *Sodium Sulfate, *Rainfall
Simulators, Capillary action, Snowfall, Erosion Control, Soil Temperature,
Water Analysis, Soil Chemistry, Soil Water Movement, Overland Flow, Rainfall
Intensity, Porous Media
25
Identifiers (Starred First)
*Water Quality Hydrology, *Ion Activity Electrodes, Specific Conductance,
Soil Evaporation, TOSCO II Process, Piceance Basin, Parachute Creek
27
Abstract
Physical properties, including porosity, permeability, particle size distribution,
and density of spent shale from three different retorting operations, (TOSC04 USBM, and
UOC) have been determined. Slurry experiments were conducted on each of the spent shales
and the slurry analyzed for leachable dissolved solids. Percolation experiments were
conducted on the TOSCO spent shale and the quantities of dissolved solids leachable
determined. The concentrations of the various ionic species in the initial leachate
from the column were high. The major constituents, S0£ and Na+, were present in concentrations
of 90,000 and 35,000 mg/£ in the initial leachate; however the succeeding concentrations
dropped markedly during the course of the experiment. A computer program was utilized to
predict equilibrium concentrations in the leachate from the column. The extent of leaching
and erosion of spent shale, and the composition and concentration of natural drainage
from spent shale has been determined using oil shale residue and simulated rainfall.
Concentrations in the runoff from the spent shale have been correlated with runoff
rate, precipitation intensity, flow depth, application time, slope, and water temperature.
This report was submitted in fulfillment of Grant No. 14030EDB under the
sponsorship of the Water Quality Office, Environmental Protection Agency.
extractor
onn C. Ward
Inxtitution
Colorado State University
WR:t02 (REV JULY IB6»)
WRSIC
SEND TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C- 20Z40
* CPO: 1969-359.33g
-------
|