U.S. Environmental Protection Agency  Industrial Environmental Research FPA-600/7-76-
Office of Research and Development  Laboratory
                 Cincinnati,Ohio 45268     December 1976
      PRODUCTION AND PROCESSING
      OF U.S. TAR SANDS:
      An  Environmental
      Assessment
      Interagency
      Energy-Environment
      Research and Development
      Program Report

-------
                       RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology.  Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields.  The seven series
are:

     1.  Environmental Health Effects Research
     2.  Environmental Protection Technology
     3.  Ecological Research
     4.  Environmental Monitoring
     5.  Socioeconomic Environmental Studies
     6.  Scientific and Technical Assessment Reports (STAR)
     7.  Interagency Energy-Environment Research and Development

This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series.  Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program.  These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems.  The goal of the Program
is to assure the rapid development of domestic energy supplies in an
envirbnmentally—compatible manner by providing the necessary
environmental data and control technology.  Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia  22161.

-------
                                          EPA-600/7-76-035
                                          December  1976
 PRODUCTION AND PROCESSING OF U.S.  TAR SANDS
         AN ENVIRONMENTAL ASSESSMENT
                      by
        N. A. Frazier,  D.  W. Hissong,
     W. E. Ballantyne,  and E. J.  Mazey
                 BATTELLE
           Columbus Laboratories
           Columbus, Ohio   43201
         Contract Number 68-02-1323
               Project Officer

                Eugene Harris
  Resource Extraction and Handling Division
Industrial Environmental Research Laboratory
           Cincinnati, Ohio  45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
           CINCINNATI, OHIO  45268

-------
                                  DISCLAIMER
     This report has been reviewed by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publi-
cation.  Approval does not signify that the contents necessarily reflect
the views and policies of the U.S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute endorsement
or recommendation for use.

-------
                                  FOREWORD
     When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on
our health often require that new and increasingly more efficient pollution
control methods be used.  The Industrial Environmental Research Laboratory -
Cincinnati (lERL-Ci) assists in developing and demonstrating new and
improved methodologies that will meet these needs both efficiently and
economically.

     Factors traceable to the increasing shortfall in U. S. production of
natural crude have rekindled interests in U. S. tar sands as a source of
synthetic fuel.  If U. S. tar sands do become a viable resource base for
syncrude, then their commercial development would create activities and
sources with potential for environmental impacts.  Reported here are the
results of a preliminary study to assess the potential primary environ-
mental impacts of production and processing of U. S. tar sands bitumen.
This research will be especially applicable to research agencies and the
various control agencies associated with energy production.  For further
information contact the Extraction Technology Branch of the Resource
Extraction and Handling Division.

                                     David G. Stephan
                                         Director
                      Industrial Environmental Research Laboratory
                                       Cincinnati
                                    iii

-------
                                   ABSTRACT

      Factors  traceable  to  the Increasing shortfall in U.S. production of
 natural  crude have  rekindled interests in U.S. tar sands as a source of
 synthetic  fuel.   If U.S. tar sands do become a viable resource base for
 syncrude,  then their  commercial development would create activities and
 sources  with  potential  for environmental impacts.  Reported here are the
 results  of a  preliminary study to assess the potential primary environmental
 impacts  of production and processing of U.S. tar sands bitumen.

      With  the possible  exception attributable to chemical differences
 between  tar sand  bitumen and coal, potential environmental impacts of
 producing  tar sands by  mining methods would be similar in type to those of
 mining coal by the  same method and in the same area as the tar sand deposit.

      Processes for  extracting bitumen from the mined tar sand would generate
 solid waste in the  form of spent sand.  Constituents and quantities of
 emissions  to  air  and  water are process dependent but existing control
 technology and good environmental practices are technically applicable.

      A viable in  situ production technology for producing tar sand reservoirs
 has not  yet been  demonstrated.  On the basis of methods tested to date,
 potential environmental impacts of producing tar sands by in situ methods
 would be very similar to those of conventional oil field production.

      Facilities used  to upgrade tar sand oil would pose potential primary
 impacts  of the same type as coking and hydrotreating processes in an oil
 refinery.  Whether  or not new upgrading facilities would have to be con-
 structed or existing  facilities might be used would depend on location and
 size  of  tar sand  deposit.

      Environmentally, in situ production of tar sands would be preferred.
 From  the viewpoint  of resource utilization, production by surface mining
methods,  where economically and technically possible, would be preferred.
Technical and  economic factors will determine if in situ methods, or
possibly underground  methods, are an alternative to surface mining in
environmentally sensitive areas.
                                      iv

-------
                                  CONTENTS

FOREWORD	ill
ABSTRACT	:	  .   iv
FIGURES	   vi
TABLES	vii

     I.  Introduction 	    1
    II.  Characteristics of U.S. Tar Sands	    3
              Geographic Distribution of U.S. Tar Sands 	    3
              Properties of U.S. Tar Sands	    6
              General Environmental Setting of Utah's
                Major Tar Sand Deposits	   11
   III.  Production of Tar Sands	   13
              Production Methods	   13
              Surface Mining	   15
              In Situ Production	   26
    IV.  Extraction and Upgrading of Tar Sand Bitumen	   42
              Processing Operations 	   42
              Potential Environmental Impacts 	   48
     V.  Environmental Comparison of Tar Sand Production
           and Processing Technology	   64
    VI.  Energy Perspective of U.S. Tar Sands	   69

REFERENCES	   72
APPENDICES

     A.  Illustrations of Mining Methods	   76

-------
                                   FIGURES

 Number                                                                 Page

   1  Occurrence of  Petroleum-Impregnated Rocks  and Shallow
      Oilfields in the United States  	   4

   2  Distribution of Tar Sand  Deposits  in Utah	   7

   3  Area Strip Mining with Concurrent  Reclamation	16

   4  Contour Mining 	  17

   5  Steam Drive in Situ Recovery Process  	  31

   6  Illustration of Forward and  Reverse Combustion  in
      Situ Recovery  Processes	33

   7  Diagramatical  Illustration of a Wet Forward
      Combustion in  Situ Recovery  Process	35

   8  Flow Schematic of Process Equipment in Fire Flood
      Recovery Operation Showing Various Emissions 	  39

   9  Flow Schematic of a Water Reuse Processing Facility for
      a Steam Injection Recovery Process 	  41

  10  Flow Sheet for Tar Sands  Extraction System Used at
      GCOS Plant	43

  11  Flow Sheet for Tar Sand Oil  Upgrading System	46

  12  Flow Sheet for Tar Sand Oil  Upgrading System Used at
      GCOS Plant	47

  13  Sulfur  Balance Flow Sheets for  Tar Sand Oil
      Upgrading  System 	  49

A-l  Block Cut  Method	77
A-2  Box-Cut  Method	78
A-3  Slope Reduction:   One  and Two-Cut  Method	  79
A-4   Parallel Fill  Method,  Modified  Slope Reduction  	  80
A-5   Mountain Top Removal Method	81
A-6   Head-of-Hollow Fill	32
A-7   Longwall Stripping 	  83
                                      VI

-------
                                   TABLES

Number                                                                  Page

  1  Deposits of Bitumen-Bearing Rocks in the United
     States with Resources over 1,000,000 Barrels 	    5

  2  U.S. Tar Sands Reserves	    5

  3  Major Tar Sand Deposits in Utah	    8

  4  Characteristics of Utah's Major Tar Sands	    9

  5  Types of Potential Primary Environmental Impacts—
     Surface Mining of Tar Sands	21

  6  Range of Constituents in Produced Formation Water:
     Offshore California	38

  7  Hydrocarbon Emissions "rom Storage Tanks 	   51

  8  Emission Factors for Petroleum Refining Processes	'.  .  .  .   52

  9  Uncontrolled Emissions to Air From Tar Sand Oil
     Upgrading System 	   53

 10  Controlled Emissions to Air From Tar Sand Oil
     Upgrading System 	   54

 11  Water Use Characteristics of Category B Petroleum Refineries ...   56

 12  Maximum Effluent Rates Based on New Source Performance
     Standards for Petroleum Refineries 	   60

 13  Observed Effluent Loadings for Category B
     Petroleum Refineries 	   61

 14  Qualitative Evaluation of Wastewater Flow and
     Characteristics by Fundamental Refinery Processes	62

 15  Environmental Comparison of Potential Tar Sand Production
     and Processing Methods	65

 16  Effect of Production Variables on Utilization of Tar Sand
     Resources	70
                                     vii

-------
                                  SECTION I

                                 INTRODUCTION
     Reported are the results of a preliminary study to determine the
potential primary environmental impact of producing and processing bitumen
in U.S. tar sands deposits.  Bitumen in the U.S. deposits, currently
estimated at 25 to 30 billion barrels, is a potential source of synthetic
crude oil.  Laboratory research and experimental or pilot field projects
on methods of recovering tar sand bitumen have been conducted on an inter-
mittent basis for some 3 decades in the U.S.  Commercial interests in the
deposits have waxed and waned over the years.  Up to the present time, the
U.S. tar sands have not been able to compete with other energy resources
for the capital required for their commercialization as a source of syncrude.

     The United States' increasing dependence on imported natural crude has
been well publicized.  National policies and programs to decrease this
dependence have caused an upsurge in interest in the contribution that U.S.
tar sands, as well as other potential syncrude resources, could make to the
U.S. energy picture.

     At the present time, the only tar sand deposit being commercially pro-
duced on a large scale is the vast Athabasca deposit in Alberta, Canada,
an operation that began about 10 years ago.  However, inplace reserves of
this deposit are about 20 times greater than those of all U.S. deposits as
they are now known.

     Many interacting factors will determine if, when, to what extent, and
at what rate U.S. tar sands are developed in the future.  One of these
factors, the potential primary environmental impact of the producing and
processing the tar sands, is the subject of this report.

     Results of the study are presented in five sections.  In order of
presentation, the contents of these sections relate to:

     (1)  Characteristics of U.S. tar sands deposits, i.e., their
          geographical distribution and properties

     (2)  Probable emission sources and other potential causes of
          primary environmental impact expected to be associated
          with methods of producing tar sand reservoirs and with
          extraction and upgrading of tar sand bitumen

-------
(3)   Assessment of potential primary environmental impacts
     of the production and processing segments of a U.S.
     tar sands industry

(4)   Environmentally preferred components of a tar sand
     operation

(5)   Perspective of U.S. tar sands as a source of syncrude.

-------
                                 SECTION II

                     CHARACTERISTICS OF U.S. TAR SANDS
     Tar sands is a general term applied to deposits of unconsolidated and
consolidated clastic sediments whose interstitial spaces are partially or
completely saturated with highly viscous bitumen or hydrocarbon materials.
Bitumen also occurs in fractures and vugular pores of carbonate rocks but
generally not in sufficient quantities to make them of commercial interest
as a source of syncrude.  Other terms used to refer to or include tar sands
are bituminous sands, asphaltic rocks, and oil-impregnated rocks.

     Although there is no clearcut definition of tar sand reservoirs, they
differ from conventional oil reservoirs by the high inplace viscosity of
the oil or bitumen.  A general rule of thumb is that at reservoir temper-
atures heavy oil will flow to the well bore at very low rates; whereas, for
practical purposes, bitumen in tar sands will not flow at all.  Order of
magnitude values of the viscosity of the bitumen can range from several
hundred thousand to several million centiposes.

GEOGRAPHIC DISTRIBUTION OF U.S. TAR SANDS

     More than 500 occurrences of surface and shallow oil- impregnated rocks,
including tar sands, are known in 22 states (see Figure 1) (1-3) f but their
evaluation as a resource base is not complete.  Major U.S. tar sands, as
shown in Tables 1 and 2, have been estimated to represent a resource base
or inplace reserves of between 18.7 and 30.1 billion barrels.*

     The State of Utah, with some 85 to 95 percent of inventoried U.S. tar
sand resources, has been comparatively active in evaluating its bitumen-
bearing rocks.  Even in the case of Utah, "reserve estimates assigned to
deposits are largely a matter of personal judgment and educated guesswork
buttressed with a certain amount of carefully considered
     Other states are beginning or attempting to begin work to inventory
their deposits in greater detail than they are now known.  Alabama has
recently completed a study(5) of the bitumen-bearing Hartselle Sandstone in
northern Alabama.  Missouri, Kansas, and Oklahoma are initiating or hope to
initiate cooperative efforts in the heavy-oil region of their tristate area,
parts of which -also contain tar sands. (6-10)  Oklahoma is investigating best
* One barrel equals 42 gallons; one gallon equals 3.76 liters.

-------
*'*•'.*
                                H   SOUTH
                                  NEB  RASKA
                4  «   C°LOR,00
                                      K  A N  S  4
                                                                  ^Sli^r^oTfH
                                        0 K L A H '  A
                                                                    &UBAM%EORG I
                                                                 LEGEND

                                                        n  PETROLIFEROUS ROCK
                                                           SHALLOW OILFIELD
                                                           COUNTY REGULAR SHALLOW OILFIELD,
             FIGURE 1. OCCURRENCE OF PETROLEUM-IMPREGNATED ROCKS AND SHALLOW
                        OILFIELDS IN THE UNITED STATES
                        [From Reference (1)]

-------
         TABLE 1.   DEPOSITS OF BITUMEN-BEARING  ROCKS  IN
                      THE  UNITED  STATES  WITH RESOURCES
                      OVER 1,000,000  BARRELS1
State and name of deposit
California:
Edna 	 	 	 	 _ 	 	 	
South Casmatia -. -. -
North Casmalia 	 	
Sisquoc 	 	 	
Santa Cruz ...
McKittrick 	 	
Point Arena . . ... 	 	 	 	
Kentucky:
Kyrock area . . . . 	 	
Davis-Dismal area.. 	 	 	 	 	
Bee Spring area
New Mexico: Santa Rosa . 	 .
Texas: Uvalde 	 	
Utah:
Tar Sand Triangle 	 	 	
P.R. Springs 	 	 . .
Sunnyside
Circle Cliffs
Asphalt Ridge
White rocks
Hill Creek 	
Lake Fork
Raven Ridge
Rimrock 	
Resources
(millions of barrels)
141.4-
46 4
40.0
	 26.0-
10.0
4.8-
	 1.2
	 18.4
	 7.5-
7.6
57.2
	 124.1-
	 10.000.1-
	 3,700.0-
2,000.0-
	 1,000.0-
1,000.0-
	 65.0-
	 300.0-
15.0-
	 100.0-
	 30.0-
166:4

50.0
9.0

11.3

140.7
18.100.0
4,000.0
3,000.0
1,300.0
1,200.0
125.0
400.0
20.0
125.0
35.0
       U.S. total	  18,694.9-    28,863.2
     'Source: Extraction of energy fuels. Federal Council for Science and Technology. Bureau of Mines open file report
   30-73. Wa-.hington, O.C. 1972.

   Reproduced from Reference  (3)      *




                TABLE 2.   U.S. TAR  SANDS RESERVES1




State
Alabama 	
California 	
Kansas 	
Kentucky 	
Missouri 	
New Mexico 	
Ohio 	
Texas 	
Utah 	

Total 	
Largest
published
estimate
(in place)
billion
barrels
0 15 *
.321
05
.084
.0009
	 .057
.0005
.14
29.3

	 30.1 ...


Date of
latest new
information
1973
1963
1964
1951
1935
1942
1941
1962
1973


 1 Compiled by Dr. Frederick Camp of Sun Oil Co. All data represents the work of other investigators. There is no input
of original data from the Sun Oil Co.
 Notice: Only Utah has reserves reported greater than 1,000,000,000 bbls. Only Utah and Alabama report recent exploration.


 Reproduced  from Reference  (3).

 *  1975 estimate for the Hartselle  deposit is 1.18  billion
    barrels

-------
 use  of  tar  or  asphaltic  sands  in  southern Oklahoma and Kentucky is compiling
 existing  information preparatory  to publishing on that state's tar sand
 deposits.(6)

      Utah has  51  deposits(H)  ranging in size from those with little to no
 interest  as a  resource to  the  Tar Sands Triangle group of deposits (see
 Figure  2) with inplace resources  estimated at 12.5 to 18 billion
 barrels.(3,4,12,13)   of  Utah's inventoried 24 to 28 billion barrels of
 inplace bitumen,  some 40 percent  occurs in central Utah in the Uinta Basin
 (see Table  3)  and 60 percent in central southeast Utah.

 PROPERTIES  OF  U.S. TAR SANDS

      Utah's major deposits occur  in sandstone and siltstone.(12)  Ritzma'H'
 summarizes  the lithology of Utah's deposits as follows:

      "Most  deposits,  particularly those of major size, occur in
      sandstone which, with finer  grain size, grades into siltstone
      and, with coarser grain size, grades into grit and conglomerate.
      More than 98 percent  of the  estimated oil in place in Utah's
      deposits  is  contained in  sandstone and siltstone.

      Along  the south flank of  the Uinta Basin, the Argyle Canyon,
      Minnie Maude Creek  and Willow Creek deposits contain notable
      amounts of oil-impregnated limestone in the Green River
      Formation.   The Thistle deposit, also in the Green River,
      contains  considerable heavily impregnated oolitic limestone
      and  coquina.  The Split Mountain deposit occurs in coarse
      crystalline  and vuggy Park City Formation limestones.  The
      Daniels Canyon  deposit occurs in highly fractured quartzite
      and  siliceous limestone.

      In central southeast  Utah, all deposits are contained in
      sandstone, siltstone,  and some conglomerate, except for
      small  amounts of oil-impregnated limestone found in San
      Rafael Swell and Teasdale deposits and localities.

      The Mexican Hat  deposit (San Juan County) occurs in Pennsylvania
      carbonate  rocks, and  the  Rozel deposit (Box Elder County) is
      found  in  oolitic mud  and  salt on the shores of Great Sale Lake".

      Values of  porosity, permeability, and oil and water saturation  exhibit
variations  and  ranges that  might  be expected in tar sand deposits.   Values
of these and other properties  of  major Utah tar sand deposits are  shown in
Table 4.(12>14)

      In the winter of 1975, the Laramie Energy Research Center conducted a
reverse combustion experiment  near Vernal, Utah, in the northwest  Asphalt
Ridge deposit.('  Average characteristics of 22 samples from cores of that
Rim Rock Member (Mesa Verde Formation) test are as follows.

-------
                                         H
                                                                                UTAH GEOLOGICAL & MINERAL SURVEY       SHEET < a 2
                                                                                            Mip   33.
                                                                                     SupcKcdn Hip Ho, II. "FMiimiwr Lvoiion Mw.
                                                                                     Oil Imtnjnitcd Rodi DtpotiB o{ Uuh". Af(ll l»«
                                                                                        LOCATION MAP

                                                                            OIL-IMPREGNATED QQCK DEPOSITS OF UTAH
                                                                                        April  1973
                                                                                         Howard R. Ritimt
                                                                                         REPRINT JULY, 1974
                                                                                   fined on publWwd, unpublitfwd ind contributed dm
                                                                                 •nd on origin*! field inrntigrtiom of U.G.& MS. perumwl
                                                                              Principrt ioveMigMon. 1965-72: R-L BIAcy, J.L. Bowmin, W,D. Bynt til,
                                                                              J.W. Gwynn, D.C. Mwtn, P.R. feienon. A.R. Pntt, S. Quigtov, H.fl. Riumt
                                                                       WYOMING
                                  A      R  '  "  I        Z"  '   "6     N      A
FIGURE 2.  DISTRIBUTION OF TAR SAND DEPOSITS IN UTAH
                [From Reference (11)]
         AND SOLD |v THE
UTAH GEOLOGICAL AND MINERAL SURVEY
   DONALD T. MCMILLAN, DIRGCYOfl
  103 UTAH GEOLOGICAL SUBVCV BUILDING
      LHUvenUTY OF UTAH
    SAIT UAKI CITT. UTAH. Mill

-------
              TABLE 3.  MAJOR TAR SAND DEPOSITS IN UTAH
                    [From References (4)  and (12)]
                                                              Bitumen
              Deposit                                   In-place  (106bbls)

Uinta Basin
     P. R. Spring                                        4,000 -  4,500
     Sunnyside                                           3,500 -  4,000
     Hill Creek                                          1,160
     Asphalt Ridge                                       1,150
     Argyle Canyon                                         100 -  125
     Raven Ridge                                           125 -  150
     Whiterocks                                             65 -  125

Central Southeast Utah
     Tar Sands Triangle                                 12,500 -  16,000
     Circle Cliffs-East Flank                              860
     Black Dragon                                          100 -  125
     Family Buttes                                         100 -  125
     Circle Cliffs-West Flank                              450

-------
                         TABLE  4.   CHARACTERISTICS  OF  UTAH'S  MAJOR  TAR SANDS

Average Values * '


Permea- Bitumen Water
Porosity bility, sat., sat.,
Deposit I P. v. (») md Z P. V. Z P.V.

Asphalt Ridge 19.6 497 51.4 2.7

N.W. Asphalt
Ridge 22.8 603 45.2 20.2

Circle Cliffs 12.3 228 17.7

Hill Creek 20.2 325 29.7 2.1

P. R. Spring 25.0 1,510 42.5 3.0

Sunnyside 21.3 729 44.8 —

Tar Sand 20.0(c) 207 --
Triangle 19.7 788 70.7
White Rocks

Ranges, []


Compressive Areal No. of
strength, Extent pay
psi (sq. miles) zones

2,491
20-25 2-5

1,598

27.7 1-3

6,555 115-125 6-13
[3]
4,784 240-270 1-13
[13]
7,805 35-90 1-12

3,242(c) 200-230 1
~"*
0.6-0.75 1

« No. of Samples,
Cross
thickness
of pay
(Stracigraphic
range, feet)
M'ftl •
v

10-135



5-310

53-65
(61)
10-102
(7) (39)
15-550

5-300

1000+

() • Average


Overburden
Thickness
(feet)


0-500+



f\l\
Value U '


Gravity
* API


S. 6-17. 5
[4]






Gallons/
Ton

13-27
'

6-26
[4)
0-500+ -11.1-22.4

0-500+

0-500+

0-500+

0-1600+

0-470
[9J
5.5-10.5
[5] (7.9)
5.8-10.3
[37H9.5J
6.2-6.7
[2]
-3.6-9.6
[5] (+4.5)
4.4-12

1.0-21.2
[129]
0.2-30.5
[454J
„

4.9-13.7
[5](9.3)
4.5-31.4
(a)  All reference  (14) values based on samples from cores except as indicated.
(b)  P. V. •• pore volume.
(c)  Surface samples.
(d)  Total liquid saturation, percent by weight.

-------
Porosity, percent
Permeability, millidarcy
Oil  saturation:
  percent of pore volume
  percent by weight
Water  saturation, percent of
  pore volume
              Bitumen
              Inplace
                10.5
               182

                54.0
                 7.6
                16.0
            After Bitumen
              Extracted
                 28.5
                615
     Listed below are ranges in average properties of the Hartselle Sandstone
deposit  at various locations in northwest Alabama.(5)
Porosity,  percent pore volume
Oil  saturation, percent pore
  volume
Thickness,  feet
Barrels per acre-foot
Barrels per acre
Gallons per ton
                                      Range
          6.0
          8.4

          2
         60
        232
          0.8
  23.8
  48.9

  55
  499
  12,802
  7.0
                              Average
      259
    5,270
        3.7
Depths  to  the Hartselle Sandstone range from zero to more than 1,000 feet.
Oil  inplace  is estimated at 1.18 billion barrels over 350 square miles where
the  formation is  thicker than 150 feet.

     Analysis of  oils extracted from the tar sands vary from deposit to
deposit.(4,5,11,16,17)  Sulfur in the Uinta Basin oils (approximately 80
samples) is  on the average about an order of magnitude less than in the oils
from central southeast Utah (approximately 30 samples).  For the former,'
Ritzma(4)  has cited values ranging from 0.14 to 0.87 percent with an average
of 0.4 percent.   In contrast, sulfur in oils from central southeast Utah
range from 1.64 to 6.27 percent with an average of 4.0 percent.  Ranges for
sulfur and nitrogen in extracts from the Hartselle bitumen in Alabama are
1.08 to  1.43 percent and 0.29 to 0.5 percent, respectively.

     Average concentrations of metals in 220 tar ash samples from Unita and
Grand Counties, Utah deposits are listed below in parts per million.
             Chromium
             Cobalt
             Copper
102.92
103.10
109.63
Manganese
Nickel
Zinc
547.47
203.00
211.63
Other metals in the tar ash identified by emission spectroscopy  are  aluminum,
calcium, iron, lead, magnesiuin, silicon, silver, sodium,  titanium, and
vanadium.
                                      10

-------
     Citing Humphreys*, McConvilled8) compared raw bitumen and synthetic
crude product from the Great Canadian Oil Sands, Ltd. commercial operation
at the Athabasca deposits:

                                        Raw            Synthetic
                                       Bitumen       Crude Product
                 Sulfur               4.5 - 5.0%         0.2%
                 Nitrogen             0.5 - 1.0%         0.1%
                 Vanadium             150 ppm            Nil
                 Ash                  1.0%               Nil

Other values noted for metals in the Athabasca bitumen (in ppm) are:

                            Vanadium       210; 250; 290
                            Nickel          82; 100
                            Iron            75
                            Copper           2;   5

GENERAL ENVIRONMENTAL SETTING OF UTAH'S
MAJOR TAR SAND DEPOSITS

     Utah's major deposits are characteristically in relatively inaccessible
and sparsely populated areas and often rugged mountainous terrain with
Asphalt Ridge being the more notable exception.  The Sunnyside deposit is
at elevations of 8,000 to 10,000 feet with elevations of other deposits
ranging from 6,000 to 8,700 feet.  Canyons with 800 to 1,000-foot relief
are commonplace.  The more prominent streams are the Colorado and Green
Rivers but intermittent streams and dry valleys are the general rule in the
immediate environs of the tar sands.

     Deposits underlie Federal, state, private, and Indian lands, but most
of the land is Federally-owned.(13)

     Ritzma(H) summarized the setting of Utah's deposits as follows:

     "Lack of or difficult access to large sources of fresh water will
     hamper exploitation of these deposits as sources of oil in most
     areas.  Water supplies may be available in parts of the Uinta Basin
     to support mining and processing operations on rich, concentrated
     deposits, such as Whiterocks and parts of Asphalt Ridge.  Water
     supply is a serious factor in considering exploitation of the large
     potential reserves of the Tar Sand Triangle and Circle Cliffs.

     The Circle Cliffs deposits are partially within the extended
     boundaries of Capitol Reef National Park and the remainder of
     the deposits is within areas proposed for various scenic, recre-
     ation, and wilderness preserves.  Access to the deposits is severely
     limited.
* Humphreys, R. D., "Some Engineering Aspects of the Tar Sands Project",
  Paper presented at the 75th Annual General Meeting of CIM, Vancouver,
  B. C., April, 1973.

                                     11

-------
The Tar Sand Triangle deposits lie mostly within the Glen Canyon
National Recreation Area and immediately west of Canyonlands
National Park.  Access to the area for development purposes is
severely restricted.

Other conflicts over land use and environmental considerations
are expected to greatly influence development of all of Utah's
deposits, particularly those susceptible to open-cut mining
methods."
                                12

-------
                                 SECTION  III

                          PRODUCTION OF  TAR SANDS
 PRODUCTION METHODS

     Laboratory  research,  field  experiments, and pilot field projects have
 involved  two basic  approaches  for producing  tar sand reservoirs—mining of
 tar  sand  as an ore  for  subsequent extraction and upgrading of the bitumen
 values and in situ  methods  of  reducing viscosity of the tar sand bitumen so
 that it will flow to a  producing well.

 Mining

     When the ratio of  waste (e.g., overburden) to tar sand is below certain
 limits, tar sand reservoirs can  be produced  by surface mining technology
 quite similar to that used  for surface mining of coal and ores.  A break-
 even value for the  overburden  to pay ratio is difficult to determine
 initially^!"' and requires  analysis of several costs including those of
 alternative mining  methods  and equipment, transportation of ore to a surface
 processing facility, the processing facility, transportation and disposal of
 spent sand, and  environmental  protection measures.  The characteristics of
 the  waste material  to be stripped, i.e., whether or not the overburden and
 any  nonproductive strata within  the tar sand deposit require drilling and
 blasting, also affects  the  costs of a surface mining operation.  Richness
 of the tar sand  ore has obvious  importance to determination of the break-
 even ratio of waste to  tar  sand  as does the  market value of syncrude.  In
 his  discussion of commercialization of the Athabasca deposits, McConville(18)
 states that, with current  technology and costs, probable limiting ratios of
 1:1  to 1:1.5 for large  tar  sand  bodies although for small areas the limiting
 ratios may be as high as 2.5:1 to 3:1, depending on richness of the deposit.

     Discussion  in  U.S. literature is almost exclusively concerned with
 surface mining as opposed  to underground mining.  In areas where overburden
would be  too thick  for  an economical surface operation, underground mining
versus in situ methods  would require an analysis of tradeoffs between the
 two methods.  Although  about 90  percent of the bitumen can be recovered from
 the  tar sand mined  by underground methods, some percentage of the tar sand
would not be mined.   Maximum bitumen expected to be recoverable by in situ
production is around 50 percent.  The tradeoff analysis would then center
around what percentage  of the  tar sand ore would have to remain inplace for
an underground operation.   Also, if one of the constraints facing an under-
ground operation is to  prevent or substantially reduce the potential for
surface subsidence  over a mined-out area, then this percentage could be


                                     13

-------
 quite significant and this,  in  effect, might completely offset the higher
 recovery efficiencies from mined  ore.  During  this study, no in-depth infor-
 mation was noted on whether  or  not  underground mining of U.S. major tar sand
 deposits is practical from a mining engineering point of view-

 In Situ Production

      The approach of in situ methods for producing a tar sand reservoir is
 that of reducing viscosity of bitumen so that  vapors or oils will move to
 producing wells where it can be lifted to  the  surface.  Although field pro-
 jects to test various in situ approaches have  been conducted in the U.S. and
 Canada, a commercial operation  for  producing tar sands by in situ methods is
 still in the future.

      Two advantages of in situ  over surface mining methods are that the
 former would eliminate the need for handling and processing vast tonnages
 of bitumen-bearing materials and  for disposing of the resultant spent sand
 waste.  One disadvantage,  from  a  resource  utilization point of view, is that
 recovery efficiency would probably  be no greater than about 50 percent com-
 pared to around 90 percent probable from processing of mined tar sand.  The
 heterogeneous and nonuniform nature of tar sand characteristics also poses
 problems in prediction and control  of in situ  performances and processes.

      As indicated previously, a principal  factor involved in deciding between
 in situ and surface mining methods  is the  ratio of thickness of overburden
 and nonproductive layers in  a deposit to the thickness of the pay sand or
 sands.  Although this ratio  may be  uneconomical for a mining operation, the
 properties (e.g.,  fractures  and joints) of the overburden or perhaps its
 thickness could be insufficient to  confine the pressures associated with
 some in situ methods.   Thus,  at least in concept, there could be some tar
 sands at depths intermediate between those favorable for surface mining and
 those for in situ methods  that  involve a build-up of pressure in the reser-
 voir.   Which of the in situ   methods that  may  ultimately be applied for
 commercialization of U.S.  tar sands is not known within the current state
 of knowledge.   About the most accurate statement that can be made at this
 time is that no single method will  be applicable to all deposits.

      The U.S.S.R.  is reported to  have developed a combination of underground
 mining and in situ methods to produce high viscosity crude.  In commercial
 use since 1972,  the thermal-mining  technology  involves sinking of mine
 shafts to a depth  above the  pay zone, drilling and blasting of passages to
 drilling/production galleries,  and  drilling of inclined .and/or horizontal
 production wells  from the  galleries.  Stream is injected into the reservoir,
 produced oil  and water are moved  to a sump, and oil is separated from the
 water  and pumped  to a central collecting point where it is heated again
 before it is  pumped to the surface.   Production to date is from depths  to
 650  feet with  50  to 60 percent  recovery from a reservoir containing  crude
with viscosities of 15,000 to 20,000 centipoises.
                                      14

-------
SURFACE MINING

     The basic technology of area and contour methods for surface mining of
coal are expected to be applicable to U.S. tar sand deposits in the event of
their commercialization.  The area method  (Figure 3) would apply to deposits
underlying flat terrain and the contour method (Figure 4) and its variations
(see Appendix for illustrations) to hilly  or mountainous terrain.  A con-
siderable body of literature exists on these methods of mining coal with a
central theme of much of the more recent literature concerned with variations
and practices to reduce impacts of mining.  The methods are aptly described
in summary form in References (21) and (22).

     Adaptation of the longwall method for mining deep underground coal to
mining of shallow coal deposits is under investigation.(23,24)  Should this
adaptation of the method, called longwall  stripping(23)} prove to be com-
mercially practical, it too might be a candidate for mining some tar sands
deposits from the surface (see Figure A-7  in Appendix A).

Materials Handling

     Surface mining of U.S. tar sands would involve removal and handling of
substantial tonnages of material including overburden, ore, and any inter-
bedded waste rock.  In that sense, a surface tar sand mine operation would
be essentially analogous to a surface coal mine.  However, once the mined
tar sand is processed (bitumen extracted), large quantities of spent sand
will also have to be handled.  In this sense, the materials handling require-
ments associated with a tar sands mining operation would be different from
those of a surface coal mine operation.  If an extraction plant is located
near the tar sands mine, which would be the expected case in a large
operation, the spent sand presumably would be returned to the mine area for
disposal.

     A perspective of the material that would be handled can be gained by
assuming a U.S. mining operation that supplies tar sand feed to an adjacent
extraction plant supplying an upgrading plant produing 10,000 barrels* per
day of syncrude.  Then, using a tar sand feed containing 20 gallons of
bitumen per ton of sand (approximately mid-range of values in Table 4),
approximately 29,000 tons of tar sand ore  per day would be required if the
ore contained no appreciable water.  After extraction of the bitumen, about
26,000 tons per day of spent sand would need to be returned to the mine or
some other area for ultimate disposal.**   If an overburden to pay ratio of
1:1 is assumed, then an additional 29,000  tons of this material would have
to be handled, giving a total materials handling requirement of 84,000 tons
per day.  Multiple handling of some portion of the material would be required
as mining and reclamation operations proceed.
 * Size of plant used in the subsequent section on extraction and upgrading;
   see Figure 11 in that section.
** In concept, depending on its properties, impurities and distance to a
   market, the spent sand could become available as a material for con-
   struction or other industries using quartz sand as a raw material.

                                      15

-------
                                                            ME  AREA '
                                                           •
                                                         -- V...W,  .„!/,. V-'"'-. ..:...
mwm^&i%$£.
      ORIGINAL SURFACE
      ~   ~CO~AL BED
                                   J-I~-r— STRIPPING BENCH   ——
                    'FIGURE 3.  AREA STRIP MINING WITH CONCURRENT RECLAMATION
                              (Reproduced from Reference 21)

-------

                /I-?: fYV fyf*}.*f3^7^1gK?£*?v*:'-&?&y i-Vf
                P^^C^^^^^^-^^'f^^
                SL-S.-JQjfi-'r^Vf-V 'y'^ft'^**2.fCKvV*ClpCt%*ifcSi&ftJtS\iJifol'^S-ri»<(PK'<
 J. S/Tf PREPARATION
2. DRILLING & BLASTING OVERBURDEN
3. REMOVAL OF  OVERBURDEN
4. EXCAVATING  & LOADING
                            FIGURE 4.  CONTOUR MINING
                      (Reproduced/Modified from Reference 21)

-------
oo
                  (See Appendix for variations and
                  approaches  to  surface  restoration)
                                                                                           TOE
                                           FIGURE 4.   (Continued)

-------
     The operation of Great Canadian Oil Sands, Ltd.  (GCOS) at the
Athabasca deposits has proven the technical feasibility of surface mining,
at least at that location.  That integrated operation can produce 55,000
barrels of oil per calendar day by processing 140,000 tons per day of tar
sand mined at a stripping ratio of 0.5:1.(18)  Other proposed Canadian
mining operations include(18):


Company
Syncrude Canada
Shell Canada and
Shell Explorer
Petrofina Canada,
et al.
Syncrude
Production
(bpcd)*
125,000
100,000

122,500

Tar Sands
Ore
(106 tpy)**
92
75

?

Overburden
or Waste
(106 tpy)
45
29

?

Stripping or
Waste to Pay
j
Ratio
0.5:1
0.38:1

1.5:1

U.S.  tar sand deposits are much smaller than the Athabasca deposit and thus
a  single U.S. operation the size of any of these Canadian facilities is
h i ghly unlikely.

      The following materials handling operations would probably be character-
istic of a U.S. tar sand surface mine:

      •  Surface clearing

      •  Removal of overburden

      •  Mining of ore (tar sand)

      •  Removal of barren rock or ore that is too low in grade

      •  Construction of haul roads

      •  Transportation of ore to processing facility

      •  Construction of impoundments and drainage diversion
        channels

      •  Transportation of tailings

      •  Surface grading and contouring for reclamation

      •  Rehandling of temporarily-stored waste.

As mining progresses and reclamation begins, different combinations of these
operations may be going on simultaneously.
 * Barrels per calendar day.
** Tons per year.
                                      19

-------
      The size of an operation,  terrain,  mining  method,  rock  properties,
 geology, characteristics of the tar sand,  and distance  to  a  process plant
 would all play a role when choosing type,  size,  and mix of equipment  to  be
 used, e.g., shovels, draglines, bulldozers,  ripping dozers,  pan  scrapers,
 large trucks, bucket-wheel excavators,  conveyors,  and pipelines  for tailings.

 Potential Environmental Impacts

      The types of primary environmental  impacts that could result  from a
 surface tar sand mining operation are shown  in  Table 5.  Of  the  types
 indicated, those that actually  occur, as well as their  duration  and severity,
 will depend on details of mining and material handling  methods employed;
 size of the operation; all elements of  the premining environment;  and the
 degree to which sound environmental protection  measures are  practical during
 planning, mining, and surface restoration  phases of the operation.  Area
 mining is generally considered  to pose  less  potential for  impact than con-
 tour mining but some of the major deposits in the U.S.,  particularly  in
 Utah, are in areas of high relief.

 Air Emissions^—
      Air emissions should be quite similar to those from surface mining  of
 coal except that as tar sands are exposed, volatiles in the  bitumen could be
 an additional emission source.   Severity would  depend on temperature,
 elevation, and whether or not fractures  in the  overburden  have allowed some
 of the volatiles to escape during geological time.

      Materials handling operations can produce  dust, particularly  if  over-
 burden or rehandled waste is dry and unconsolidated.  Haul roads are  a
 source of dust when transporting ore and spent  sand but water or hydroscopic
 materials can be used to reduce dust generated  from this source.   Ore could
 be transported by conveyors which would  eliminate  dust  from  hauling of ore
 to an extraction plant.   Spend  sand can  be slurrieu with extraction process
 water and pumped to a temporary or permanent disposal area which would also
 eliminate the use of haul roads for this purpose.

      Diesel engines emit particulates, sulfur oxides, nitrogen oxides,
 hydrocarbons,  aldehydes,  and organic acids.(25)   Quantities  would  depend
 on fuel  composition,  size and mix of diesel-powered equipment, terrain,
 size  of  mining  operation,  and whether or not ore and/or spent sand is
 transported in  earth-moving equipment.

Water Emissions  and Solid Wastes—
      Surface mining operations  that disturb  existing or expose new surfaces
would increase availability of  solubles  and  suspendable constituents  for
aqueous  transport.   Duration and severity  of impacts on water quality
resulting  from  these  operations would depend on terrain, climate,  details
of mining method  and  environmental  protection practices, geochemistry of
overburden, method  of  transporting  ore and/or spent sand,  surface  drainage
patterns, and geohydrology.

     Excluding possible  effects of  chemical  differences between  tar sands
and coal, types of  impacts  on water quality  of  a surface tar sand  mine would

                                      20

-------
TABLE 5.  TYPES OF POTENTIAL PRIMARY ENVIRONMENTAL IMPACTS--
          SURFACE MINING OF TAR SANDS
Operation or Source of Impact
Site Preparation
Surface Clearing (cleared
area)
Stripping (stripped area)
Tar Sand Extracting (mined
area)
Haul Road Transportation
(construction)
Tailings Disposal
Bitumen in Tailings or Low
Grade Tar Sand Waste
Fines in Tailings
Stripped Waste
Solubles or Water-
Transportation Particles
in Overburden
New Surface
Increases in Surface Slope
From Waste Disposal
Rehandling of Materials:
Backfilling, Grading,
and Recontouring

Increased
Air Emissions
Volatile
Engine Hydrocarbons
Dust Exhaust of Bitumen
X X
X X
X X
X
X X
(X) (X)
X X




X

X X
POTENTIAL IMPACT

Increased Availability of
Aqueous Transportable Materials
Suspended/
Solid Waste Dissolved
Generated Inorganics Organics Solids

X (X) (X)
X (X) (X)
(X) (X)

X
X

XXX
X X
X

X X
X
(X)
(X)
(X)
X
(X)


X
X
X

X
X

-------
                                                  TABLE  5.   (Continued)
NJ



POTENTIAL IMPACT



Surface Changes
Operation or Source of Impact
Site Preparation
Surface Clearing (cleared
area)
Stripping (stripped area)
Tar Sand Extracting (mined
area)
Haul Road Transportation
(construction)
Tailings Disposal
Increased Destruction
Landslide of Existing
Risk Vegetation
X
(X)
(X)

(X)

Alteration
of
Habitats
X
(X)
X

(X)

Topographic Drainage
Changes Diversion
X X
(X) (X)
X
(X)
(X) (X)

Increased
Noise
X
X

X
X
(X)
X
Changes
Ground Water
Regime
Physical Chemical


(X)
(X)




(X)
(X)


Bitumen in Tailings or Low
  Grade Tar Sand Waste
Fines in Tailings
Stripped Waste
Solubles or Water-
  Transportation Particles
  in Overburden
New Surface
Increases in Surface Slope
  From Waste Disposal
Rehandling of Materials:
  Backfilling, Grading,
  and Recontouring
                                                                                                                            (X)

-------
be essentially analagous  to  those  of  a  surface  coal mine  in  the same area.
Surfaces disturbed,  exposed,  or  created by mining operations, are bared  to
the erosive action of precipitation runoff with attendant increases in
suspended solids with the extent being  a function of slope and "looseness"
of the  surface materials.  Good  mining  practices in sloping  terrain provide
for diversions of runoff  upgradient from a mine and/or downgradient
retention basins as  well  as  reducing  the slope  of waste material as mining
progresses.  Permeable materials underlying bare surfaces are also exposed
to the  chemical action of water  on soluble components.  This water can then
return  to the surface downgradient to become a  part of surface water or
enter the groundwater, depending on site specifics.

     The organic phases of tar sand bitumen are more similar to petroleum
than to coal.  Hence, exposed tar  sand  surfaces and unrecovered bitumen  in
spent processed sand, if  exposed to the physical action of runoff, will  be
a potential nonpoint source  of organic  loading  (primarily alkane—or para-
ffin—type hydrocarbons of heavy molecules with lesser amounts of aromatics)
that is not associated with  a surface coal mine operation.

     Sulfur in the bitumen is present as organic compounds which are oil
soluble and should not be leachable.  Also, trace metals  in  the bitumen, as
in petroleum, are present as  oil soluble organic compounds,  e.g., porphyrins
and salts of organic acids.   Although a majority of these compounds are
insoluble in water,  some  could hydrolyze and the metal ion become soluble
in water.  The extent to  which hydrolysis will  occur is pH and rodox
potential dependent, being less  pronounced in alkaline solutions (e.g.,
alkaline surface and ground waters).

     A  potential for impact  on water  quality could develop if either (1)
hydrolyzable components in the bitumen  of tar sand ore wasted at the mine
site or in spent sand wastes  or  (2) their hydrolyzed products become avail-
able for aqueous transport.   However, surface-breached tar sands have been
exposed to surface waters over geologic time and subsurface  deposits are a
part of the geohydrological  regime of an area.  Thus, it  can be conjectured
on one  hand, that surface mining could  reduce potential for  impacts on water
quality because tar  sand  ore  would be removed for processing.  On the other
hand, increased exposure  (change from an anaerobic to aerobic environment)
of tar  sand bitumen  as a  result  of (1)  tar sand benches exposed during
mining  operations, (2) mine-wasting of  low-grade tar sands,  and (3) presence
of unrecovered bitumen on surface  areas of spent sand grains would be sources
of water quality degradation.

     If tar sands in their natural setting (before mining) are now degrading
water quality*, then surface  mining of  tar sands could add to this degra-
dation.  If tar sands in  their natural  setting  are not presently degrading
water quality, surface mining could still result in conditions that have
potential for impacting water quality because of exposure of tar sand bitu-
men.   Evaluation of  this  potential would require field and laboratory
studies.
* No data were noted on water quality in  the environs of  a  tar  sand  deposit.


                                      23

-------
     In the event that bitumen in the tar sand is extracted by a hot water
process with caustic soda added, some components of the bitumen, including
metal compounds and aliphatic and aromatic organic acids, would be expected
to react with the water and dissolve (hydrolyze).  In that case, unrecovered
bitumen in the spent sand should contain very little free water leachable
components.  Any process water wasted with the spent sand could, however,
contain hydrolyzed products.

     Other factors affecting the potential impact on water quality are the
products of weathering resulting from (1) increased area of bitumen exposed
and  (2) wasting of coal and of oil shale known to be associated with some
tar  sands.  Exposure of coal containing pyrite leads to formation of iron
sulfate and in the case of oil shale weathering of organic components leads
to the formation of organic acids.

     Site specifics that will influence potential for water quality impacts
include geochemistry of overburden; pH and redox potential of waters;
climate, particularly preciptation; terrain; and^geohydrology.  Adherence
to environmentally sound mining practices, including burial of wasted coal,
oil  shale, low-grade tar sand, and spent sand as quickly as possible, will
reduce whatever potential that might exist.

     The same can be said for any measures taken to prevent erosive action
of water on bared surfaces and/or to use sedimentation basins,  either of
which would reduce the potential of contaminants being transported as
colloidal or suspended solids or by adhering to suspended inert mineral
particles.

     Based on the preceding discussion, potential sources of effluents that
could be associated with a surface mining operation are precipitation runoff
from the mining area, infiltrated water, and effluents from spent sand
tailings ponds and processing facilities.  Characteristics of effluents
expected from a hot water extraction and from a tar sand oil upgrading
facility are discussed in a subsequent section.  Collectively,  these sources,
within an effluents guideline perspective, would contain the following con-
stituents:

        BOD
        COD
        TSS
        Oil and grease
        PH
        Phenolic compounds
        Ammonia as N
        Sulfur compounds
        Metals
        Solids.

Surface Changes—
     Surface mining of tar sands would change the  topography  of the land
surface through the stripping and excavation of  overburden and ore,  the
construction of haul roads, and the disposal of waste materials.   Vegetation

                                      24

-------
on stripped areas, areas graded for facilities  (e.g., power lines, roads,
and colocated extraction and upgrading plants), and nonstripped areas used
for disposal of solid wastes and for impoundments would be destroyed.
Collectively, topographic changes, denudation of vegetated areas, and noise
from mining (equipment and blasting) can impact wildlife and wildlife
habitats.

     Surface topography could be restored in part with degree of restoration
dependent upon terrain.  Efforts to revegetate areas would meet with varying
degrees of success depending on availability of top soil, mining practices,
terrain, and climate.  More generally, factors determining the degree to
which the area of a tar sand mine can be restored to a premined status
should not differ from those of a surface coal mine in the same area with
one significant exception; namely, reclamation and revegetation of spent
sand waste disposal areas.  Approaches to restoring an area of a tar sand
mine to a premined status will have to address problems of restoring areas
used for disposal of spent sand from a bitumen extraction plant.  Depending
upon specifics of an operation, e.g., present or possible future regulations,
terrain, extraction process, and method of transporting spent sand to a
disposal area, methods of "disposing of spent sand will likely include:

     (1)  Temporary disposal in ponding areas until the sand can be
          disposed of permanently in mined out areas without inter-
          fering with mining operations

     (2)  Permanent disposal behind dams constructed in valleys near
          a mine or processing plant.

     Revegetation of spent sand disposal areas has been the subject of con-
tinuing research at the GCOS operation in Canada.  Attendant problems at
that operation include those of poor settling properties of clay particles
and other fines in the tailing ponds and of minor amounts of bitumen in
the tailings which has precluded recycling of water back to the processing
plant.  Reports are that GCOS has spent at least $1.5 million in research
on this problem and that it would cost up to $1.75 per barrel of recovered
oil to get the fines to settle out of the water completely.(1°)

Ground Water—
     Previously discussed surface sources which might degrade quality of
surface waters would also be potential sources for impacting ground water
quality.  Possible transport mechanisms include infiltration of precipitation
runoff as it moves downgradient after being in contact with the surface
source, precipitation seeping downward through the surface source with
resultant leachate entering the groundwater, and process effluents (from
adjacent processing facilities) that follow either of these two routes.

     In addition to characteristics of the surface sources, potential for
impact would depend on the amount and time distribution of precipitation,
geohydrological and geochemical parameters, and distance the groundwater
travels before it recharges surface streams or is withdrawn for use.  With
regard to the latter, shallow wells (used for watering livestock) in alluvial


                                      25

-------
 aquifers are relatively common in  intermittent or dry stream beds in remote
 and semiarid areas.   Quality of water  in  these aquifers that supply wells
 downgradient from a  nearby tar sand mine  probably would present the most
 immediate potential  for being affected.

      Potential physical effects of surface mining on groundwater are trace-
 able to two basic causes either, both, or none of which could occur at a
 specific mine.  If the mine intersects an aquifer or aquifers, then ground-
 water supply to wells, at least to those  located immediately downgradient,
 could be interrupted or diminished.  Wells supplied from perched water
 tables or alluvial aquifers in intermittent or mostly dry stream valleys
 would be especially  susceptible to this type of an effect.

      The second cause stems from changes  in runoff coefficients resulting
 from alteration in surface characteristics controlling the amount of pre-
 cipitation that infiltrates as a local source for charging an aquifer.

 IN SITU PRODUCTION

      Up to 90 percent of the tar sands deposits in the United States have
 been estimated to be at depths greater than 300 feet.'^)  In situ production
 methods are potentially applicable for recovering bitumen in tar sand
 deposits that are too deep for surface mining methods to be economically
 feasible.   Where used,  these methods would obviate several sources of
 potential environmental problems associated with surface mining, e.g.,
 disturbing large surface areas, removing  overburden, mining and trans-
 portation of ore,  generation of dust, extraction of the bitumen from the
 sand,  and transportation and disposal of  spent sand.

      The technology  for in situ production of tar sands is not as developed
 as that for surface  mining and has not been proven commercially or demon-
 strated to be economically feasible.  The technology is still focused on
 the experimental and field pilot study phases.  In situ production would
 not yield as high a  bitumen recovery as would surface mining.  Recovery
 efficiencies of around 90 percent  can be  expected from surface mining,
 whereas in situ recovery efficiencies probably would range up to approxi-
 mately 50  percent.

 General Methods and  History of
 In Situ Production

      In situ production of a tar sand reservoir would require drilling one
 or more wells  into the pay sand, setting  casing, and perforating the casing
 in  the  potentially productive pay  zone.   Because of the highly viscous nature
 of  the  tar  sand bitumen,  in situ techniques for its recovery would differ
 from  those  generally employed for  conventional oil recovery.  The viscosity
of  the  bitumen must  be  reduced to  a level that will enable it to flow
 through the  sand matrix to the well bore.  This may be accomplished by
injecting hydrocarbons,  solvents,  or emulsifiers into the reservoir to
dissolve the bitumen and  thereby reduce its viscosity, or by adding heat
directly by  the  injection of steam or indirectly by air injection and
                                      26

-------
 combustion  of  part  of  the bitumen in place.   The addition of  liquid solvents
 and  emulsifiers  has generally proven to  be more costly,  slower,  and less
 effective than thermal methods of recovery,  and most  research and  field
 testing has emphasized either direct or  indirect thermal recovery.   One
 reason for  the apparent advantage of thermal techniques  over  solvent
 techniques  is  the rapid decrease in bitumen  viscosity with increases in
 temperature, e.g.,  a 5° API  tar sand oil found  in California  has a  viscosity
 of 38,000 centipoises  (cp) at 140° F,  1050 cp at 210° F,  and  only  57 cp at
 325°  F.<27>

      In situ production of a tar sands reservoir would involve using some
 of the wells to  inject the heat or fluid into the reservoir and other wells
 to produce  the bitumen.   To  improve areal sweep efficiencies, flood  patterns
 would be employed to direct  the injection fluid,  or heat  front, to  central
 producting  wells; some commonly used flood patterns are  line  drive,  5-spot,
 7-spot, and 9-spot.

      Pilot  field projects to develop in  situ methods  for  producing  tar sands
 have  been conducted intermittently in the U.S.  for about  2 decades but, to
 date, no U.S.  or Canadian deposits have  been commercially produced by these
 methods.  In 1959-1960,  Standard Oil of  Ohio tested steam flooding in Utah's
 Northwest Asphalt Ridge deposit;  Shell and Signal Oil (1965-1966) experi-
 mented with steam flooding in Utah's Sunnyside  deposit; and Gulf, in the
 1960's, conducted a fire flood experiment in Utah's Asphalt Ridge deposit.
 Experiments have also  been carried out in Kentucky and western Missouri
 deposits, and  pilot tests to produce viscous oil from California's Vacar
 Tar  Sand were  conducted by American Petrofina in the mid-1960 's using cyclic
 steam injection.
     The most recent field  test      in  the United States was conducted by
ERDA's Laramie Energy Research Center on a 9-spot reverse combustion fire
flood near Vernal, Utah, on a Sohio  lease at Northwest Asphalt Ridge.  The
fire flood burn was started in November, 1975, and was terminated after only
4 weeks because of poor area sweep efficiency.

     At present, there is no evidence of other recent tar sand production
field test projects in the  United States.  A considerable number of pilot
and commercial scale projects have been conducted, however, in several states
for the production of viscous heavy  oils.  Since the early 1950' s, over 100
in situ projects to recover heavy oils  have been reported in the literature.

     In Canada, in situ research efforts have been conducted by Imperial Oil,
Shell Canada, Amoco Canada  Petroleum, and Gulf Canada.  Imperial Oil has
spent over $15 million^O  on a research and pilot project to test steam-
injection recovery techniques for heavy oil at its Cold Lake facilities.
Shell Canada has conducted  steam injection tests on the Peace River oil
sands deposit, and Amoco is continuing  tests of a 50-bpd pilot plant at tar
sand deposits near Fort McMurray.  Gulf Oil has recently been given approval
by the Alberta Energy Resources Conservation Board for a 50-bpd steam-driven
pilot test at the Wabasco tar sands  deposit.
                                      27

-------
      Most in situ production  development,  to date, has been conducted on the
 heavy oil deposits.   These  heavy  oils are  similar in composition to tar
 sands bitumen but have  some degree  of mobility.  The general rule of thumb
 for distinguishing between  heavy  oils and  tar sands bitumen is that heavy
 oils will flow at a  very  low  rate,  e.g., one barrel per day, whereas tar
 sand bitumen will not flow  at all at reservoir temperatures.  The basic
 technologies employed for producing heavy  oils are, however, potentially
 applicable to the less  mobile tar sands bitumen .  The specific in situ
 production technologies that  are  discussed below are either currently, or
 have been, applied to tar sand and  heavy oil deposits in the United States
 and Canada.

 Specific Methods  of  In  Situ Production

      Most information on  specific processes and field-test results was
 obtained through  a search of  technical journals and reports, reprints of
 presentations at  meetings and symposia, and telephone communication with
 persons familiar  with in  situ production methods.  In most cases, the
 technical content of environmentally-related information was neither
 explicit nor quantitative.  Little  information was available on the
 economics of the  processes  or the analysis of the coproduced streams, e.g.,
 produced water and produced gas.  The characteristics of coproduced streams
 are thus presented only on  a  generic basis.

 Chemical Injection—
      Chemical injection involves  the injection of hydrocarbon-based solvents
 or aqueous alkaline  surfactants into the producing zone to lower the vis-
 cosity of the heavy  oil or  bitumen.  Solvent stimulation using naphthenes
 and aromatics has been  practiced  successfully in California since the mid-
 '60's for the recovery  of heavy oils.  The process served principally as a
 well-bore stimulation technique in  which the solvent was injected into the
 zone around  the well bore to  dissolve the  heavy oils and paraffins.  In one
 test,  after  injecting 500 to  1,000  barrels of solvent, the well was placed
 back on production.(27)   -^he  process was repeated until most of the crude
 oil in the well-bore vicinity has been produced.

      In tests conducted by  the USBM near Bartlett, Kansas, in the early
 1970fs,  explosive fracturing  of a heavy oil reservoir was followed by
 solvent stimulation.  The experiments met  with moderate success; about 12
 percent of the oil contained  within a 3-spot well pattern was produced.

      The application of solvent stimulation or miscible drive displacement
 for  tar sands,  however, has not been successful.  Shell Canada(29) experi-
mented  for several years  on the Athabasca  tar sands and concluded that
viscous  fingering and gravity overlay effects, coupled with the high cost
of  the  naphthenes and aromatic solvents, rendered the process economically
infeasible.   At present,  there appears to  be no activity", either in the
United  States  or  Canada,  directed toward the use of miscible solvents  to
produce  tar  sands.

     Shell Canada developed an alternative to miscible solvent injection
that utilizes  aqueous-based emulsifying fluids.  The emulsifiers were

                                      28

-------
 believed  to be  superior  to  the miscible fluids  in  that  the  oil-loaded
 emulsifying fluid  possessed a viscosity only slightly greater  than  the
 original  emulsifying  fluid  and lower  than the viscosity of  a comparably
 loaded  solvent-based  fluid.  Moreover,  the flow of the  injected  emulsifiers
 occurred  in the wetting  phase path, meaning the bank of tar or heavy oil
 ahead of  the  displacement did not  have  to be moved to obtain penetration
 into the  zone.  Two systems were formulated for use in  the  Athabasca tar
 sands:   (1) an  alkaline  solution of proprietary nonionic surfactants and
 (2) a dilute  solution of sodium hydroxide.   Field  testing indicated that
 the contacted portion of the reservoir  was limited to where high gas or
 water existed,  or  where  a fracture was  created.  The result was  that pene-
 tration and dissolution  of  the tar into the oil was a very  slow  process and
 the concept was abandoned.

 Steam Injection—

     Cyclic Steam  Injection—Cyclic steam injection has found  wide commercial
 application since  its introduction in 1959 for  the production  of heavy oils.
 In this method, steam is injected  into  a producing zone,  followed by an
 alternate period of production from the same well.   The method takes
 advantage of  the high latent heat  of  vaporization  of the steam which, upon
 condensation, is transmitted to the reservoir.   The steam heats  the
 reservoir in  the vicinity of the well bore and  thus reduces the  oil
 viscosity, permitting the oil to flow during the production cycle.

     Development of a cyclic steam injection project entails drilling
 several closely spaced wells into  the reservoir and injecting  anywhere from
 1,000 to  25,000 bbl of 500°  to 700° F steam into each well.(30)  The
 injection process  generally lasts  a few days after which the wells are shut-
 in to allow time for  the reservoir to "soak".   The wells are then opened
 and produced.   The cycle may be repeated several times  before  the oil pro-
 duction rate  diminishes  to  an uneconomical level.

     Cyclic injection has been used successfully in major heavy  oil fields
 in the  United States,  particularly California.   The viscosities  of the oil,
 at reservoir  condition,  are, however, relatively low (less  than  40,000
 cp).(3D

     Imperial Oil, Ltd.(32)  has invested $15 million in a 5,000  bpd project
 at Alberta's  Cold  Lake oil  sands (100,000 cp at reservoir conditions).  Wells
 are drilled in  a 7-spot  pattern and steamed for about a month, after which
 time they are produced on pump for 3  to 4 months.   About 30,000  barrels of
 600° F  steam  are injected into each pattern yielding a  production of 9,000
 to 10,000 barrels  of  oil for a recovery of 20 to 30 percent of the oil in
 place.(26)  Produced  water  is separated from the oil by conventional methods
 and reinjected  into a deep  formation  for disposal.   The produced gas, at
 present,  is separated from  the produced liquids and flared  or  reinjected with
 the steam.  No  information was available on the exact composition of the
produced  gas, although it contains primarily C02 with small quantities of
methane and l^S.   The produced water  is saline,  with 3,000  to  12,000 ppm
                                      29

-------
 NaCl,  and  somewhat  basic with a pH of 8 to 8.5.  The somewhat high pH is
 attributed to  the ion  exchange process utilized to soften the steam feed
 water.

      Imperial  Oil questions  the application of steam cycling as an economi-
 cally feasible method  for producing  the heavy oils.  It was suggested that
 extensive  research  will be required  to more effectively utilize the steam,
 improve  sweep  efficiencies,  optimize selection of candidate production zones,
 and  reduce water influx.  One major  problem is the enormous volume of water
 required for steam  generation and the availability of water.  Imperial is
 currently  working on a project designed to process the produced water to
 make it  acceptable  to  boilers.(32)   Another problem with steam cycling is the
 energy balance; the pilot project requires an equivalent input of one barrel
 of oil for every three barrels of oil produced.

      Cyclic injection  is considered  to be more suitable to reservoirs with
 thick pay  zones and large volume of  oil in place per acre and with wells
 capable  of commercial  production without steam injection.(33,34)  There is
 no evidence of cyclic  injection having been successfully employed on oil
 reservoirs with viscosities  in excess of 100,000 cp.  As this is signifi-
 cantly less than the viscosity of tar sands bitumen (500,000 to 6 million
 cp), nothing can, as yet, be concluded concerning the probable success of
 steam cycling  for tar  sands  production.

      Steam Flooding—Steam flooding  is similar to a conventional water flood
 except that the injected fluid is steam, or steam with an emulsifier.  An
 illustration of a steam flooding recovery process is shown in Figure 5.  Heat
 losses restrict the distance that the steam front can be propagated through
 the  reservoir  and pattern floods are generally used for the development of
 a field, e.g., 5-spot  or 7-spot patterns.

      Steam flooding has found frequent application in reservoirs with high
 viscosity  oils and  relatively good permeabilities.  For oils with viscosi-
 ties up  to 1,000 cp at normal reservoir temperatures, horizontal sweep
 efficiencies are usually high.  However, viscous bypassing may be signifi-
 cant for oils  with  normal reservoir  viscosities greater than 1,000 cp.

     Little work has been conducted  on steam flooding highly viscous heavy
 oil  and  tar sands.  The most noted pilot scale work to date was conducted
 by Shell Canada, Ltd.  in the early 1960's on the McMurray tar sands in
 Alberta, Canada.  A 5-spot pattern of wells was drilled, the production zone
 fractured,  and a caustic and steam solution injected into the four corner
 wells.  Although no information was  available on the actual volumes of
 steam and  caustic injected,or the volumes of oil produced, 0.685 tons  (4 bbl)
 of steam was injected  per barrel of  oil produced.  From the results, Shell
 concluded  that a practical well spacing for a 5-spot pattern would be 4
 acres per  producing well and with this configuration predicted a possible
 overall recovery of 50 to 70 percent of the bitumen in place.  Due to  the
high heat  requirements for steam generation, Shell also concluded that for
a commercial operation to be economically feasible, no more than 0.5 tons
 (2.9 bbl)  of steam  should be injected per barrel of oil recovered.
                                      30

-------
Injection Well
          Production Well
                Steam
Oil and Wa
                 Overburden
.-— •

u-
^

\ Buffer
Sv 1 Zone
^VS. J -N_
\V. Hrnt-rrl r 	 >
^S^«s. IICULLU 1 — 	 	 -\/r
^S-^ Region /
/ -^
— -
ICold Region ^
n
9


          FIGURE  5   STEAM DRIVE IN SITU RECOVERY PROCESS

-------
     Although  Shell  Canada, Ltd.  Is  continuing to conduct field testing of
 steam  injection on a site  in  the  Peace River oil sands deposit, a commercial
 steam  drive  recovery project  has  not been implemented in either the United
 States or  Canadian tar  sands.  The major problems associated with steam
 injection—high energy  and water  requirements and effective communication
 between wells—have  yet to be solved satisfactorily.

     Fireflood—Fireflood  involves in situ combustion to generate heat within
 a  formation  by combustion  of  a portion of the in-place bitumen.  Principally,
 there  are  two  variations of in situ  combustion-r-forward and reverse.

     The forward combustion process, illustrated in Figure 6, requires the
 drilling of  air injection  and oil production wells.  Ignition is started at
 the  air injection well  and the combustion front propagates through the
 formation  in the direction of air flow toward the producing wells.  The
 temperature  of the dry, burned region increases from the temperature of the
 injected air at the  sand face to  the maximum temperature at the burning
 front  (600°  to 900°  F).  Injected air captures heat from the burned zone
 and  moves  it toward  the burning front.  Immediately ahead of the burning
 front,  water and light  components of the crude bitumen are vaporized and
 driven toward  the producing well.  The residual bitumen and coke provide
 the  fuel to  sustain  the combustion process.

     Forward combustion has been  applied primarily for secondary and tertiary
 recovery of  heavy oil with viscosities typically ranging from 100 to 1,000 cp
 at reservoir conditions.   The process has been demonstrated, however, to
 suffer  a major drawback, particularly with the heavier more viscous oils in
 low  permeability reservoirs.   The oil and water vapors, swept ahead of the
 burning front,  contact  the unheated portions of the reservoir where they
 cool and condense.   The condensed liquids become very viscous and tend to
 plug the pores in which they  are  deposited thus restricting flow to the
 production wells.

     In a  reverse combustion  process, also illustrated in Figure 6, ignition
 is affected  in the production well and the combustion front propagates
 through the  tar  sand  toward the air injection well, i.e., counter to the
 direction  of air  flow from the injection well.  A portion of the bitumen
 is vaporized and  carried with  the air stream to the producing well; the
 remainder  is burned as  fuel or left as residual coke in the sand.  Reverse
 combustion has not been utilized  to the extent that forward combustion has,
although two factors  are cited as advantages over the forward combustion
process for  tar sands application^^):

     (1)  The vaporized fluids are directed through the hot,
          burned-out  zone  and are, therefore, less likely to
          condense and plug the pore spaces.

     (2)  The oil produced is of  higher quality as a result
          of the  thermal cracking of the bitumen.
                                      32

-------
                Injection
                  well
                                     Production
                                       well
Foword  combustion
p
4
4


-------
The  process,  however,  is  sensitive  to  the air flux, or rate of injection of
air, which  insures  that the burning does not reverse direction and behave as
forward  combustion.

     The most recent reverse combustion experiment was conducted by ERDA/
LERC at  the Northwest  Asphalt Ridge tar sand deposit near Vernal, Utah,
in late  1975.  Combustion tube experiments indicated that 40 percent of the
in-place bitumen would remain in  the pay zone as coke and 10 percent would
be burned.  On the  basis  of a 50  percent recovery and a 75 percent sweep
efficiency, it was  estimated that 800  bbl of oil could be recovered from
the  9-well  line-drive  pattern.  However, only 55 bbl of upgraded bitumen
and  170  bbl of water were recovered.(14,15)

     Forward  Combustion-Water Flood—This process (see Figure 7) involves
instituting a conventional forward  combustion drive until a portion of the
reservoir has reached  a temperature of about 1,500° F, after which water is
injected with air into the formation.  The water serves to lower combustion
temperature,  and the generated steam transfers the heat into the formation
more rapidly, accelerating the recovery process.  This feature is regarded
as being particularly  attractive  for the recovery of tar sands bitumen.(30,35)

     To  date, most  of  the research  on  forward combustion and water flooding
has  been conducted  by  the Amoco Production Company in the Athabasca tar
sands.   Field tests, using conventional and reverse combustion, were com-
menced in 1958 to produce bitumen from the Gregoire Lake area.  Shown below
are  the  properties  of  the tar sands and the bitumen.

     •   Specific gravity  of bitumen 1.08

     •   Viscosity (200° F) 1,000  cp (50° F) 2,000,000 to 5,000,000 cp

     •   Hydrogen/carbon ratio 1.44

     •   Sand  minus  200 mesh

     •   Porosity 35 to 40 percent

     •   Permeability most tar zones, 200-300 millidarcies, clean
         sand  from tar  zones, several darcies, silt zones, few
         millidarcies

     •   Saturations 0  to  90 percent bitumen (remainder water,
         little or no gas  saturation)

     •   Bitumen content on weight basis 0 to 18 percent.

     In  1966,  water was injected  concurrently with air in a forward com-
bustion  drive.  The process was labeled the Combination Forward Combustion
and Water Flood.  Tests performed on a 1/2 acre 5-spot pattern yielded a
32 percent removal  of  the bitumen in place and it was estimated that  55
percent  of the oil  heated to 150° F or more was recovered.  Following this
                                      34

-------
INJECTION
  WELL
                         PRODUCTION
                           WELL
                    OVERBURDEN LAYER
    yA Water & Air
         HEATED
          SAND
, COMBUSTION"^
  ZONE
               \ HYDROCARBON^'
                   VAPORS
 COLD REGION
'OF RESERVOIR
    FIGURE 7.  DIAGRAMATICAL ILLUSTRATION OF A WET FORWARD
              COMBUSTION IN SITU RECOVERY PROCESS
              (From Reference 14)
                          35

-------
successful operation, an expanded pilot with 12 patterns over 35 acres was
developed.  Additional tests are currently being conducted.(30)

     Forward combustion with water flooding has the advantage over dry
forward and reverse combustion in that most of the coke is left in the
reservoir as residue, resulting in a less viscous, upgraded produced oil.(37)
Recovery efficiencies are also considered to be greater because of improved
heat transfer from the steam.  In addition, the required volume of injected
air is generally lower with the combined water flood.  A summary(38) Of
results from 24 dry forward combustion drives indicated an injected air/
produced oil ratio of 6 to 44 mscf/bbl as opposed to a range of 1 to 6
mscf/bbl for two wet combustion drives.

     Wet combustion, in addition to water requirements of approximately one
barrel of water per mscf of injected air(3°), poses another possible problem
for producing tar sands.  Condensation of the vaporized oils and water could
result in plugging of reservoirs with low permeabilities.(37)

Potential Environmental Impacts

     To date, there is no commercial in situ production of tar sands and a
viable production technology is yet to be demonstrated.  Information per-
taining to environmental effects is estimated on the basis of a few pilot
studies conducted on tar sands and the application of similar technology
for the production of heavy oils and is limited at best.  Primary environ-
mental impacts should, however, be similar to those of conventional
petroleum production and any environmental problems could be addressed in
a similar manner.

Air Emissions—
     Gaseous byproducts would constitute the major atmospheric discharge
associated with in situ combustion.  The volumes of gases produced with the
oil may vary considerably depending on the reservoir conditions, the type
of production mechanism, and the characteristics of the bitumen in place.
Gas to oil ratios ranging from 5 to 75 mcf/bbl have been reported from in
situ combustion recovery operations.(38)

     Generally, the produced gas can be expected to contain varying amounts
of sulfur compounds (S02, H2S, etc.), nitrogen, carbon dioxide, carbon
monoxide, oxygen, and hydrocarbons.  A review(38) of 24 different wet and
dry in situ combustion recovery projects on heavy oil reservoirs showed the
average range of concentrations of the produced gases to be:

     •  Oxygen, 2.5 to 3.5 percent

     •  Carbon dioxide, 10 to 17 percent

     •  Carbon monoxide, 0 to 2 percent

     •  Hydrogen sulfide, 0 to 2 percent

     •  Methane,  0 to 2 percent; and the balance nitrogen.

                                     36

-------
 No information was found on particulates, SC>2, or nonmethane hydrocarbon
 content.

      An analysis(15»37) of the gas produced from the ERDA/LERC reverse
 combustion test near Vernal, Utah, revealed the following approximate
 compositions:

      •  Carbon dioxide, 9 percent

      •  Carbon monoxide, 3 percent

      •  Hydrogen, 1 percent

      •  Methane, 1 percent

      •  Nitrogen, 80 percent.

 There was no evidence of hydrogen sulfide.

      Generally, the gases probably would be very lean and lack sufficient
 heating value to justify collection and processing for marketing.   They
 might be either vented from the water knockout and separation tanks to the
 atmosphere, or combusted further to completely oxidize the carbon  monoxide,
 hydrogen sulfide, and methane.  The ERDA/LERC pilot facility utilized a  gas
 treater with a continuous pilot flame to insure complete combustion of the
 off gases.(37)

      Other air emissions associated with in situ recovery projects are the
 exhaust discharges from diesel or gasoline-powered equipment and dust
 generated from vehicles traveling the access roads.

Water Emissions—
      The most  potentially significant quantity of water from a volume and
 possible contamination standpoint would be water coproduced with the oil.
 Some water almost always accompanies produced oil but may vary in  quantity
 from less than 1 percent to over 99 percent of the production stream. No
 information was found on the composition of the water produced from the  tar
 sands production processes.  However, since petroleum and tar sand reservoirs
 can occur in the same areas, constituent species in the coproduced water
 probably would be similar to those from conventional petroleum production,
 an example of  which is given in Table 6.  Shown in Figure 8, a flow schematic
 of some of the surface equipment employed in a fire-flood recovery operation,
 are constituents to be expected in air, solid waste, and water emissions.

      Coproduced waters in the conventional onshore petroleum production  are
 not known to present any major environmental problems as the common practice
 is to reinject them into another subsurface formation.  Subsurface disposal
 of coproduced  waters has been practiced in some in situ tar sands  pilot
 projects  and is currently being practiced at Imperial Oil of Canada's Cold
 Lake steam-injection pilot operation.
                                      37

-------
TABLE 6.  RANGE OF CONSTITUENTS IN PRODUCED FORMATION
          WATER:  OFFSHORE CALIFORNIA^)
   Effluent
 Constituent                       Range, mg/1

 Arsenic                           0.001 - 0.08
 Cadmium                           0.02 -  0.18
 Total Chromium                    0.02 - 0.04 .
 Copper                            0.05 - 0.116
 Lead                              0.0 - 0.28
 Mercury                           0.0005 - 0.002
 Nickel                            0.100 - 0.29
 Silver                            0.03
 Zinc                              0.05 - 3.2
 Cyanide                           0.0 - 0.004
 Phenolic Compounds                0.35 - 2.10
 BOD5                              370 - 1,920
 COD                               340 - 3,000
 Chlorides                         17,230 - 21,000
 TDS                               21,700 - 40,400
 Suspended Solids
      Effluent                     1-60
      Influent                     30 - 75
 Oil and Grease                    56 - 359
 (a)  Some data reflect treated waters for reinjection,
      [From Reference (39)]
                      38

-------
       Produced
       Oil and
       Water From
       Fire Flood
                          I
           Off Gas
                C02
                CO
               H2S
            CH4, C2H6
                 N2
                 02
u>
VD
 Water Cooled
Heat Exchanger
                       3-Phase
                      Separator
                               Produced Sand
                                  to Pit  -
Oil With
  Water
Emulsion
 Water
 Eaulsion
Separator
                      Produced
                       Water
                                             Dissolved Solids
                                             BOD, COD, Slightly
                                             Acidic, Suspended
                                             Solids, Oil,  and Grease
                                                                                            Oil
                                                                                          Storage
                                                                                                                     Air
                                                                                                          To Injection
                                                                                                         [Disposal Well
                      FIGURE 8.   FLOW  SCHEMATIC OF PROCESS  EQUIPMENT IN FIRE  FLOOD
                                   RECOVERY OPERATION  SHOWING VARIOUS EMISSIONS

-------
      In  in  situ  tar sands production, water would be used primarily for
injection into the reservoir  (steam flooding and wet combustion processes)
and  for  cooling  the produced  fluids.  In some regions in the United States,
particularly where major tar  sands exist, water available for these purposes
might be a  problem.  If so, it could be necessary to treat and reuse, rather
than dispose of  the produced  water by injection.

      Treatment of the water prior its reuse for water injection in a wet
combustion  process would involve neutralization, filtration, and flotation;
treatment for reuse for steam injection would likely require an additional
water-softening  step (ion exchange).  Figure 9 is a schematic of a water
reuse processing facility for steam injection recovery.  Potential for
environmental impact would be low and somewhat comparable to that of
injection disposal.

Solid Wastes—
      Produced sand and drill  cuttings would constitute the major solid wastes
associated  with  in situ production.  The produced sand should not be signi-
ficant if gravel packing or other sand filtration methods are applied in
the  well bore.   The sand that would be produced could be separated from the
bitumen  and probably discharged into a pit that would later be backfilled.
This procedure is commonly employed on conventional onshore production.

      The cuttings produced from the drilling operation could also be disposed
of by procedures commonly employed in the oil production industry—burying in
a pit and back-filling at the completion of the drilling operations.  The
volume of cuttings would be relatively small; the drilling of a 600-foot well
would produce approximately 30 cubic yards of rock cuttings.

      Disposal of drilling muds should pose little to no potential for
environmental impacts because these muds are customarily collected and
reused.  Cable tool or air drilling operations do not employ drilling muds.

Surface  Changes—
      In  situ production of tar sands offer much less potential for surface
impacts  than surface mining.  Some vegetation would likely need to be cleared
and  the  ground graded to accommodate equipment installations, but only
relatively  small areas would  be affected.  Roads, pipelines, and the appro-
priate separation and treatment equipment would have to be constructed and
electric power and possibly telephone lines erected.  Surface operations  can
be conducted so  as to avoid or minimize interference with the concurrent
land usage  such  as livestock  grazing, agricultural cultivation, recreation,
and  wildlife habitat.

      At  the completion of the production operation, all equipment and the
concrete pads could be removed and the wells plugged in accordance with
recommended procedures.  Restoration of a site should not be an extensive
operation because the extent  of disruption to the land surface would be
minimal.
                                      40

-------
Produced
Oil and
Water
I
  Water Cooled
 Heat Exchanger
                          Produced
                          Sand
                                           Produced
                                           Water


Filter



Water
[Ion-Exchange)


Steam
Generator


                                                                                                         Injection
                                                                                                            Well
                    FIGURE  9.   FLOW SCHEMATIC OF A WATER REUSE  PROCESSING  FACILITY
                                FOR A  STEAM INJECTION RECOVERY PROCESS

-------
                                  SECTION IV

                 EXTRACTION AND UPGRADING OF TAR SAND BITUMEN
PROCESSING OPERATIONS

     Extraction refers to the separation of the bitumen, or tar sand oil,
from the mined tar sand.  Upgrading refers to the on-site processing of the
tar sand oil to produce a synthetic crude oil product suitable for shipment
to a petroleum refinery.  The final processing into products such as gasoline,
heating oil, etc. will almost certainly be done not at the mine site but
rather at a petroleum refinery.  The complex processing operations necessary
to produce these final products would not be economical on the small scale
of a U.S. tar sand facility.  Even the existing and planned Canadian tar
sand facilities, which are much larger than U.S. facilities will be, produce
only a synthetic crude oil.

     Most of the operations required in the extraction and upgrading of tar
sand oil are identical to or similar to operations conducted in petroleum
refining.  This fact has been used in analyzing the potential emissions from
these sections of a tar sand facility.  It is also because of this similarity
that the extraction and upgrading sections are discussed generally together
here.  However, separate discussions are provided where appropriate.

     The processing operations selected for the emission analysis were based
on a consensus of the one existing and several planned Canadian tar sand
facilities.  Material balance and emissions data are based on a facility
producing 10,000 bbl/day of synthetic crude oil product, which is about the
maximum size anticipated for U.S. tar sands.(40)  These quantities can, of
course, be scaled proportionally to any other facility size.

Extraction Facilities

     A flow sheet for the extraction system used at the Great Canadian Oil
Sands (GCOS) facility is shown in Figure 10.(18.19)  This system is based
on the hot water extraction process first developed by Dr. K. A. Clark(41»42)
but also piloted by  several  companies.   The  extraction sequence includes the
following steps:

     (1)   Pulping of the oil sand with water, steam, and dilute
          caustic

     (2)   Separating the bitumen from the sand by skimming and
          froth flotation
                                      42

-------
                                                  Makeup Naphtha
                                                    from Coker
                Makeup
                 Water
Pond Water Return
       Tar Sand Feed*!
  2,480 T/D Bit.
 26,710 T/D Min.
Conditioning
   Drum
                                  Froth
                                 Settler
Oversize
Steam f o
i £
Caustic n ¥
..... w> 	 i
Addition c
•H C Cr-t\i k
3 B rrotn
=! ^ 	 	 _. Middlings

Separation 	
Cell \. Sand ^>
* Assume 20 gallons ^s. .s'
of bitumen per ton ^v^X^
of tar sand



T^ JO
4J
O
fe

1 ' r» ^
	 *-v» Scavenger [-J
M-JH^I -fnoQ Ce l!s

1 r


Settling Pond






1

\







f
Tail
;
I
                                                                                          Tar Sand Oil
                                                                                          to Coker
                                                                                          2,160 T/D Bit.
                                                                                            65 T/D Min.
                                                                                       300 T/D Bit.
                                                                                    25,790 T/D Min.
           FIGURE 10.  FLOW SHEET FOR TAR SANDS EXTRACTION  SYSTEM USED AT GCOS PLANT
                         (QUANTITIES SCALED DOWN TO  10,000 B/D SYNCRUDE PRODUCT
                        FROM UPGRADING SYSTEM)

-------
      (3)   Mixing  the  bitumen  froth with a diluent  (naphtha)

      (4)   Centrifuging  to  remove water and fine solids.

      In addition  to the froth and sand removed from the separation cell, a
 "middlings"  stream is also withdrawn.  This  stream contains mostly water but
 also  some  suspended fine mineral and bitumen particles.  A portion of this
 stream is  returned for  mixing with the conditioning drum effluent to dilute
 it  properly  for pumping.   The balance of the middlings, which is called the
 "drag stream", is withdrawn to purge the system of very fine mineral matter.
 The drag stream is treated by scavenging, which involves froth flotation
 using air.   The scavenger  froth is combined  with the settling cell froth,
 and the tailings  from the  scavenger cell are combined with the separation
 cell  tailings.  The tailings  go to a settling pond.

      Note  that, based on Figure 10, the overall bitumen recovery in the
 extraction section is about 87 percent.

      Other extraction methods that have been proposed for tar sands include
 cold  water separation and  anhydrous solvent  separation.

      Another alternative for  separating the  bitumen from tar sands is
 retorting, a high temperature "destructive distillation" similar to that
 used  for oil shale.   This  alternative has not been given much attention in
 the literature.   In Europe, a Lurgi Ruhrgas  process plant used in coal and
 oil shale  retorting experiments has been used in an experiment to distill
 about 40 metric tons  of American tar sand feed stock.(43)  A major reason
 that  it might be  considered for U.S. tar sands is that it could reduce the
 water consumption, and  Utah's tar sands are  located in areas that are
 generally  short of water.  The major disadvantage of retorting is that,
 because of the high temperature required, the thermal efficiency (i.e.,
 oil recovery efficiency) will be lower than  for extraction.  Also, the
 quality of the oil product will be inferior  because of increased degradation
 of  the  hydrocarbons by  the high temperatures and increased contamination of
 the oil with trace species contained in the  sand.  On the other hand, the
 lower water  consumption should be reflected  in a lower water pollution
 potential.   The major solid waste, i.e., the spent sand, should be
 essentially  unaffected.

      All the planned  Canadian tar sand facilities as well as the operating
 GCOS  include hot  water  extraction.  For that reason, this process was
 chosen  for the emissions analysis given here.  The fuel quantities shown
 on  the  flow  sheet in  Figure 10 are based on  10,000 bbl/day of synthetic
 crude oil product from  the upgrading section (discussed below).  The
 quantities were scaled  down from the material balance data for the GCOS
plant but are based on  20  gallons of bitumen per ton of tar sand feed  (see
Table 4).
                                      44

-------
Upgrading Facilities

     The raw  tar  sand  oil  will  be  upgraded,  probably  on-site,  to make it
more like a typical petroleum crude  oil.   This will involve

     •  Removing  the small amount  of mineral matter remaining
        in the oil

     •  Decreasing the density,  viscosity, and carbon/hydrogen
        ratio of  the oil

     •  Removing  some  of the sulfur  and metals from the oil  (most
        U.S.  tar  sand  oil  contains more sulfur than most petroleum
        crudes do).

To accomplish this, the GCOS plant plus all  but  one of the proposed Canadian
plants  use a  coking process followed by hydrotreating of the coker liquids.
The only exception to  this sequence  is the proposed Shell Canada, Ltd. plant,
which is to use vacuum flashing, solvent  deasphalting, and hydrotreating.(19)
One reason for the attractiveness  of coking  as the primary upgrading step is
that the mineral  matter (sand)  in  the feed cannot plug a coker and is
readily removed with the coke.

     A  flow sheet for  the  upgrading  system is shown in Figure  11.  A fluid
coker is used in  this  system.   The naphtha and gas oil produced by the coker
are hydrotreated  separately.  Off  gas from the coker  and the hydrotreaters
is freed of H2S and then used to generate the hydrogen required and to meet
other fuel needs  of the plant.   The  H2S recovered from the gas is converted
to elemental  sulfur in a   Glaus  sulfur plant.  The coke produced by the
coker is used to  generate  steam, part of  which is used as such and part of
which is in turn  used  to generate  the electrical power required by the plant.
The hydrotreated  naphtha and gas oil are  recombined to form  the synthetic
crude oil product.

     The fuel quantities shown  on  the flow sheet in Figure 11  were scaled
down from the material balance  data  for the  proposed  Syncrude  Canada, Ltc.
plant.(19)  Note  that  the  overall  oil yield  for  the upgrading  section is
about 82 percent.

     For comparison, a flow sheet  for the upgrading system of  the GCOS plant
is shown in Figure 12.(19)   The  primary differences between  this flow sheet
and the previous  one are that a  delayed coker is used instead  of a fluid
coker and three liquid streams  are taken  from the coker instead of two.
Delayed coking produces more coke  than fluid coking,  which is  reflected in
a lower liquid product yield.  The overall oil yield  for this  flow sheet is
only about 78 percent.  The GCOS plant produces  more  coke than is required
for its fuel  needs.(^)  In order  to bring the coke production more in line
with the needs and thus to increase  the product  yield, the future facilities
are planning  to use fluid  coking.  One of the planned facilities  (Home Oil/
Alminex) plans to use  a variation  of fluid coking known as Flexicoking, in
which most of the coker produced is  converted into a  fuel gas.
                                      45

-------
                       Steaa
                                                                                            Naphtha to
                                                                                          •^—Extraction
                                                                                     202 B/D System
Tar Sand Oil
12,200 B/D
(2,160 T/D Bit.)
(  65 T/D Min.)
                                                                                               Sulfur
                                   359 T/D Coke
                                    65 T/D Min.
Synthetic
 rude
10,000 B/D
(1,532 T/D)
               FIGURE 11.  FLOW SHEET FOR TAR SAND OIL UPGRADING SYSTEM
                             (FUEL QUANTITIES  SCALED DOWN FROM  SYNCRUDE,
                            CANADA FLOW SHEET)

-------
                                                        Gas
12,800 B/D
(2,280 T/D)
                                                                                     Synthetic
                                                                                     Crude
'*   10,000 B/D
    (1,440 T/D)
                              580 X/D
         FIGURE  12.   FLOW SHEET FOR TAR  SAND OIL UPGRADING SYSTEM USED AT GCOS PLANT
                      (QUANTITIES SCALED  DOWN TO 10,000  B/D OUTPUT)

-------
      The  flow  sheet based on fluid coking  (Figure 11) should be more repre-
 sentative of future tar sand facilities and hence was used in the calculations
 presented here.

 POTENTIAL ENVIRONMENTAL IMPACTS

 Air  Emissions

      The  extraction and upgrading facilities include a number of sources of
 emissions to the atmosphere.  Some of these emissions can be roughly quanti-
 fied, whereas  others can be discussed only qualitatively.  Some of the
 emission  estimates to be discussed are based on the similarity of these
 process operations to those used in petroleum refineries.

 Sulfur Oxide Emissions—
      The  major potential sources of sulfur oxide emissions are the tail gas
 from the  Claus sulfur plant and the flue gas from the steam boilers.  A
 Claus plant normally recovers about 95 percent of the sulfur fed to it, and
 the  other 5 percent is present as SOX (primarily 802) in the tail gas.  Tail
 gas  treatment  processes are available that can increase the overall sulfur
 recovery  to 99.5 percent, thus reducing the tail gas emission by a factor
 of 10.  An even more severe problem is the flue gas from the steam boilers,
 which are fired with the coke produced by the coker.  If the feed to the
 coker contains 4 weight percent sulfur, the coke produced will contain
 about 5 weight percent sulfur.(45)  Combustion of such a high sulfur coke
 results in considerable emissions of SOX in the flue gas.  Flue gas desul-
 furization (FGD) processes are available, or at least being developed, that
 can  remove up  to 90 percent of the SOX from flue gas.  Actually, the same
 S02  recovery process could be used for both the boiler flue gas and the
 Claus plant tail gas.

     To estimate the magnitude of these emissions, sulfur balances were made
 for  a tar sand oil (to the upgrading section) containing 4 weight percent
 sulfur.   This  is the average concentration for the large deposits in south-
 east Utah.  The sulfur content of the syncrude product was taken as 0.27
 weight percent, based on the Syncrude Canada data(19)} and that of the coke
 was  taken as 5 weight percent.  Then, based on the flow sheet in Figure 11,
 of the 86.40 tons/day of sulfur entering the upgrading section with the feed,
 4.14 tons/day  will leave in the syncrude, 17.95 tons/day will go into the
 coke, and 64.31 tons/day will go to the Claus plant as H2S.  Two sulfur
 balances  from  that point on are shown in Figure 13.  Without a flue gas
 desulfurization system, the total emission will be about 42.3 tons/day of
 S02-  With a flue gas desulfurization system treating both the flue gas and
 the  Claus plant tail gas, the total emission will be about 4.4 tons/day of
 S02-

     Another option for controlling sulfur oxide emissions from the steam
boilers is to  combust the coke in a fluidized bed with limestone added to
 tie  up the sulfur.  This method would be as effective as flue gas desulfur-
 ization in reducing the SOX emissions but would have the following
disadvantages  relative to FGD:
                                     48

-------
    Uncontrolled
               3.22 T/D S as SO,,
           64.31 T/D S as H S
                                  Sulfur
                                   Plant
                    61.09 T/D S
                                     '  17-95  T/D S as SO,
           17.95 T/D S in coke
                                  Steam
                                  Boilers
                                             Emission » 21.17  T/D S = 42.30  T/D SO,
   Controlled
                   64.31 T/D S as
                   4.21 T/D S  as SO-
                  17.95 T/D  S as SO,
17.95  T/D S in coke
 Steom
Boilers
                               80.04 T/D S
                                                 19.95 T/D S  as SO,
                                                       2.22 T/D S as SO,
                                             Emission •= 2.22 T/D S = 4.44 T/l) SO,
    FIGURE 13.   SULFUR BALANCE FLOW SHEETS  FOR TAR SAND OIL
                  UPGRADING SYSTEM

                             (10,000  B/D Syncrude Product)
                                      49

-------
      •  The Glaus  plant  tail  gas  problem  could not be solved
         simultaneously.

      •  The sulfur value would be recovered as a mixture of
         calcium-sulfur compounds  instead  of as elemental
         sulfur.

      •  The particulate  matter emission via the flue gas would
         be greater.

      There may be  some other  emissions of sulfur oxides if process heaters
 are fired  with liquid products or intermediates.  Emission factors for this
 will be  given in a later section.  Information on the extent to which such
 fuels will be used is not available.  Most of the fuel needs of the plant
 will probably be met with the cleaned fuel gas.

 Hydrocarbon Emissions From  Storage Tanks—
      Storage tanks will  be  required for the tar sand oil feed, the naptha
 and gas  oil intermediates,  and the syncrude product.  The emissions of hydro-
 carbons  from these tanks would depend primarily on the type of tanks used.
 Following  EPA petroleum  refinery  guidelines, storage facilities required
 would depend on volatility  of stored hydrocarbons (Federal Register 39FR9308,
 March 8, 1974):

      •  Nonvolatile (vapor  pressure •''1.5 psia)
            Cone roof tanks

      •  Moderately volatile (vapor pressure >1.5 but <11 psia)
            Floating roof tanks

      •  Volatile (vapor  pressure  >11 psia)
            Pressure facility.

      To  estimate the magnitude of  these emissions, some calculations were
 made using the equations developed by the American Petroleum Institute and
 published  by the EPA.(46)   f^g results of these calculations are shown in
 Table 7.   The calculations  are for a single tank for each material stored,
 with the size of the tank based on the flow rate of the stream involved.
 The  data shown for floating-roof  tanks apply for the best design of such
 tanks  and  indicate that  the emissions from such tanks are only about 3 per-
 cent  of  those from fixed-roof tanks.  As indicated in the footnote, some
 types  of floating-roof tanks  can have emissions three times greater than
 the  "best  design"  values.

Miscellaneous Emissions—
     There  are a number  of  other  emission sources in the extraction and
upgrading  sections  that  can be roughly quantified based on the similarity
of many of  the operations involved to operations in petroleum refineries.
This can be  done by using the published emission factors for the pertinent
refinery operations, and these are given in Table 8.(47)  The estimated
emissions based on  these emission  factors are shown in Tables 9 and 10.
                                     50

-------
                      TABLE  7.   HYDROCARBON EMISSIONS FROM  STORAGE  TANKS
Material Stored
Tank Size, 103 bbl
Tank Dimensions, ft
Vapor Pressure at 70 F, psia
Specific Gravity
Turnover Rate, E input/yr/B capacity
Enr'-sions for Fixed-Roof Tanks, 'lb/day
Breathing loss
Working loss
Total
Emissions for Flcating-Roof Tanks, lb/day
Standing loss
Withdrawal loss
Total
Tar Sand Oil Feed
100
122 D x 48
5.2
1.011
30

718
4,000
4,718

121
11
132
Naphtha
20
63 D x 36
6.6
0.751
13

330
920
1,250

66
_i
67
GC.S Oil
50
86 D >: 48
0.3
0.919
13

53
120
173

2
_3
6
Syncrude Product
100
122 D x 48
3.5
0.860
13

626
2,700
3,326

68
_4
72
Total






1,727
7.740
9,467

258
19
277
(a)   Correspond  to  Reid vapor pressures of 7.0 psi for tar sand oil feed, 10.5 psi  for naphtha, 0.4 psi for gas  oil, and
     6.0 psi for syncrude product.

(b)   Data for proposed Syncrude, Canada plant.
(c)   Froa "Supplement No. 1  for Compilation of Air Pollutant Emission Factors", 2r.d Edition, U.S. EPA, July, 1973.
     Other parameter values  taken froa this source were tank outage - 50% of tank helghc, csily temperature variation -
     15 F, wind  velocity = 10 mi/hr.
(d)   For welded  tanks.  To calculate emissions for completely riveted tenks, multiply by the following factors:
                                                       Single Se_al            Doub:.s 3as3

                                Pontoon Roof              2.89                   2.44
                                Pan Roof                  3.11                   2.89

-------
                                 TABLE 8.   EMISSION  FACTORS FOR PETROLEUM REFINING PROCESSES
Ui





Particulate Sulfur Oxides, Carbon
Type of Process
Boilers and process heaters
lb/103 ft gas burned
10/103 bbl oil burned
Fluid cckir.g units
Uncontrolled, lb/10 bbl fr feed
With ESP, ib/103 bbl fr feed
Compressor engines, lb/10 ft gas burned
Blow Down Systems
Uncontrolled, lb/10, bbl ref cap
Controlled, (c>lb/10J bbl ref cap
Process Drains
Uncontrolled, lb/103 bbl wastewater
Controlled,^) lb/103 bbl wastewater
Cooling Towers, lb/10 gal cooling water
Miscellaneous losses, lb/10 bbl ref cap
Pipeline valves and flanges
Vessel relief valves
Pur.? seals
Co-.pressor seals
Others*6)
(a) s - refinery gas sulfur content (lb/100
Matter

0.02
840

523
6.85
Neg

Neg
Neg

Neg
Neg
Neg

Neg
Neg
Neg
Neg
Neg
ft ). Factors
(b) S - fuel oil sulfur content (weight percent). Factors
of 336 Ib/bbl (0.96 kg/liter).
(c) Vapor recovery system or flaring.


as S02

Monoxide Hydrocarbons

2s Neg
6,720 S(b> Neg

NA
NA
2s(a)

Neg
Neg

Neg
Neg
Neg

Neg
Neg
Neg
Neg
Neg
based on
based on



Neg
Neg
Neg

Neg
Neg

Neg
Neg
Neg

Neg
Neg
Neg
Neg
Neg
complete combustion
complete combustion



0.03
140

Neg
Neg
1.2

300
5

210
8
6

28
11
17
5
10
of sulfur
of sulfur


Nitrogen
Oxides,
as NO

0.23
2,900

Neg
Neg
0.9

Neg
Neg

Neg
Neg
Neg

Neg
Neg
Neg
Neg
Neg
to S02-
to SO. and




Aldehydes

0.03
25

Neg
Neg
0.1

Neg
Neg

Neg
Neg
Neg

Neg
Neg
Neg
Neg
Neg

assumed fuel




Ar.conla

Neg
Neg

Neg
Neg
0.2

Neg
Neg

Neg
Neg
Neg

Neg
Neg
Neg
Neg
Neg

oil density


(d) Vapor recovery or covers on oil/water separators.
(e) Direct air blowing, sampling, etc.







    \^y   — — — — ~ —           ^f -
    Source:  Compilation of Air Pollutant  Emission Factors, 2nd Edition, AP-42,  pp 9.1-3 to 9.1-5, U.S.  EPA, April,  1973.
    Conversion forlorn:   kfi/Utrr - 0.002853 (Ib/bM),  kg/m3 - 16.02  (lb/ft3), kR/lUor - 0.1198 (Ib/pnl).

-------
Ui
OJ
            TABLE  9.  UNCONTROLLED EMISSIONS TO AIR FROM TAR SAND OIL UPGRADING SYSTEM

                                 (10,000 bbl/day Syncrude Product)
Emissions, Ib/day
Process Operation
Steam plant
Sulfur plant
Storage tanks (fixed roof)
Fluid coker
Blow down systems
Process drains
Cooling towers
Valves, flanges, seals, etc.
Subtotal
Process heaters
Compressor engines
(a) Based on 70" of mineral
P Articulate
Matter
2,450(a^

—
6,381
—
—
8,831

—
matter in coke feed
Sulfur Oxides,
as S02
71,733
12,850
—
—
—
—
84,543


Carbon
Monoxide
359
-------
             TABLE  10.   CONTROLLED EMISSIONS  TO AIR  FROM TAR SAND OIL UPGRADING  SYSTEM
                                        (10,000 bbl/day Syncrude Product)
Process Operation
Steam plant
Sulfur plant
Storage tanks
Fluid coker
Blow down systems
Process drains
Cooling towers
Valves, flanges, seals, etc.

Sulfur
Particulate Oxides,
Controls Matter as SO
Flue gas desulf (FGD)
system — 8,872
Tail gas recycle to FGD
system 0
Floating roof — —
Electrostatic precipitator 84
Vapor recovery or flaring
Vapor recovery or covers
on separators --
Ends sions , Ib /day
Nitrogen
Carbon Glides,
Monoxide Hydrocarbons as NO.
(a) (a) (a)
n b*l
— » ZOJ
61
106 Cc)
855

Aldehydes Acmonia
_-
  Subtotal
Process heaters
Compressor engines
                                                     84
                                                               8,872
1,372
(a)  Depends on  type of flue gas desulfurization system used.

-------
Table 9 gives the uncontrolled emissions and Table 10 the controlled
emissions.  These tables also include  the data from the previous sections
on sulfur oxide emissions and hydrocarbons storage losses.

     The miscellaneous emissions include (Table 9):

     •  Hydrocarbons from blowdown  systems, process drains,
        cooling towers, and leakage from valves, flages, and
        seals

     •  Particulate matter from fluid  coker and steam plant

     •  Nitrogen oxides from the steam plants.

There are also emissions from process  heaters and compressor engines.
Emissions from the process heaters depend strongly upon whether gas or oil
is used to fire the heaters, as can be seen from the emission factors in
Table 8.

     The emission control methods considered here include (Table 10):

     •  A flue gas desulfurization system for the steam boiler
        flue gas and the Glaus plant tail gas

     •  Floating-roof storage tanks

     •  An electrostatic precipitator  on the fluid coker

     •  A vapor recovery or flaring system for process blowdown

     •  Vapor recovery systems or covers on the oil/water separators.

These methods result in overall emission reductions of about 90 percent for
sulfor oxides, 88 percent for hydrocarbons, and 99 percent for particulate
matter.  These figures do not include  the emissions from process heaters and
compressor engines, which could not be estimated.

     In order to estimate the emissions from two sources (process drains and
cooling towers), information was required on the cooling water circulation
rate and wastewater flow rate.  Lacking any more specific data, average
data for Category B petroleum refineries were used.(48)  These data are
shown in Table 11.  A Category B refinery is one which includes cracking-
type processes like coking but not more complex operations such as lube oil
or petrochemical production.  This concept will be discussed further in the
following section on water emissions.

     A source of hydrocarbon emissions not shared by petroleum refineries
is that of evaporation from the surface of the settling pond used for the
tailings.   This emission cannot be quantified but will be less than  the
flow rate of bitumen to the pond (300  tons/day in Figure 10), since  some of
this bitumen will be returned to the process with the pond water return.
                                     55

-------
        TABLE 11.  WATER USE CHARACTERISTICS OF CATEGORY B
                  PETROLEUM REFINERIES
Cooling Water Circulation Rate, gal/bbl crude            1,450

Wastewater Flow, gal/bbl crude

  Refineries with no once-through cooling water

    Median                                                24
    Range                                                4-89

  Refineries with some once-through cooling water

    Median                                                174
    Range                                               7-6,861
Source:  Brown & Root, Inc., "Economics of Refinery Wastewater
         Treatment", API Publication No. 4199, pp V-2 and V-9,
         August, 1973.

Conversion factor:  kg/liter = 0.002853 (Ib/bbl).
                                56

-------
Water Emissions

     The extraction and upgrading facilities include a number of operations
that will generate wastewater at various levels of contamination.  Like any
other modern plant, a new tar sand facility would be expected to use modern
technology for reducing the wastewater flow and for treating the wastewater
prior to discharging it.  The wastewater problems of a tar sand facility
will be very similar to those of a petroleum refinery.  The final pollutant
emission rates from the facility will depend upon the extent to which the
wastewater is treated.  Lacking other information, one can roughly estimate
the emission rates for a tar sand facility by assuming that such a facility
will have to restrict its emissions to about the same level as a modernly
controlled petroleum refinery of similar size and complexity.  These concepts
will be elaborated on in the following sections.

Sources of Wastewater—
     Based largely on the experience of petroleum refineries(49)s one can
say that the sources of wastewater from tar sand oil extraction and upgrading
facilities will include the following:

     (1)  Cooling Water Slowdown.  As will be discussed in the
          following section, the facility will almost certainly use
          an evaporative, recirculating cooling water system.  In
          such a system, a small amount (usually 0.5-2 percent(45,49))
          of the circulating water must be withdrawn to purge dis-
          solved solids from the system.  This "blowdown" contains a
          very high concentration of dissolved solids (usually 0.2-
          0.4 percent(49)) an(j small amounts of various species added
          during treatment of the water or picked up during its use.

     (2)  Boiler Feed Water Blowdown.  The normal practice in plants
          requiring steam is to recycle the steam condensate for use
          as boiler feed water.  When this is done, a small amount
          (typically about 5 percent(49)) of this water must be
          withdrawn to purge dissolved solids from the system.  This
          blowdown is similar in composition to the cooling water
          blowdown discussed above.

     (3)  Sour Water.  Sour water, containing primarily H2S and NH3,
          will be generated in the fractionation operations that
          follow the coker and the hydrotreaters.  In addition to
          H2S and NH3, the condensate from the coker fractionator
          will contain phenols and perhaps cyanides.  Sour water can
          also come from process knockout drums.

     (4)  Storm Water Runoff.  The runoff water from paved process
          areas and tank areas will be oily, whereas that from
          utility areas will contain solids but not oil.
                                     57

-------
      (5)  Pump and Compressor Cooling Water.  Some water used in
          cooling pump pedestals and glands and compressor jackets
          will become contaminated with oil.

      (6)  Tank Bottom Draws.  The water periodically drained from
          tanks will be oily and in some cases also sour.

Water Segregation and Reuse System—
     In order to minimize the wastewater treatment costs and the final
effluent rate, the facilities will employ modern technology for segregating
various effluent streams and reusing water.  The facilities will almost
certainly use a recirculating cooling water system with tooling towers
rather than once-through cooling water.  Steam condensate will be reused as
boiler feed water.  Segregating the effluent according to composition will
increase the efficiency of the wastewater treatment system, since each
effluent will receive only the treatment it requires.  For example, only
oily waters will go through the API oil-water separators.

Wastewater Treatment Processes—
     A combination of wastewater treatment processes will be used in a tar
sand  facility.  These processes probably will be selected from the following:

      •  Primary Treatment Processes
           Equalization  (to dampen surbes in flow and loadings)
        -  API oil-water separators
           Sour water strippers

      •  Secondary Treatment Processes
        -  Dissolved air flotation
        -  Aerated lagoons
        -  Activated sludge treatment
        -  Chemical coagulation and sedimentation
        -  Filtration.

Tertiary treatment processes, such as  carbon adsorption, ion exchange, and
reverse osmosis, probably will not be  used.

Estimated Effluent Rates—
     The operations conducted in the extraction and upgrading sections of a
tar sand facility, and hence the wastewater problems, will be very similar
to those of a petroleum refinery.  Therefore, one can use information on
petroleum refineries as a basis for roughly estimating the effluent rates
for a tar sand facility.  To be more specific, one can use data  for
Category B refineries.  This category  includes cracking-type or  petro-
chemical production.  Two types of data for such refineries can  be used—
effluent regulations or effluent data—for actual refineries.

     In considering effluent regulations, it is most meaningful  to consider
the New Source Performance Standards (NSPS), since these apply to  new
facilities such as the tar sand facilities will be.  These standards  specify
the maximum effluent rates, in pounds  per thousand barrels of feedstock,  for


                                      58

-------
various effluent characteristics  (BOD, COD,  etc.).  The standards depend on
the size and complexity of  the refinery.  The complexity is expressed in
terms of a "process configuration" factor, which is six for the flow sheet
of interest in which all  the  feedstock goes  through a coker.  For a Category
B refinery of this complexity and the size range of interest here (up to
24,900 bbl/day), the NSPS are given  in Table 12.(50)  Also included in
this table are the corresponding maximum  effluent rates (in pounds per day)
for a tar sand facility producing 10,000  bbl/day of synthetic crude oil
product.

     For comparison with  the  NSPS, some data on observed effluent loadings
for existing Category B refineries are given in Table 13.(51)  This table
also shows the wastewater treatment  processes used at these refineries.
For all the effluent characteristics on which data were available, the
observed effluent loadings  span a fairly  wide range (a factor of at least
2.5) and (except for ammonia) go both above  and below the 30-day average
NSPS.

     Some qualitative evaluations of the  wastewater-related characteristics
of the individual processes used in  the extraction and upgrading sections
are shown in Table 14.(51)  This table indicates the extent to which these
processes contribute to the flow rate and various effluent characteristics
of the total wastewater.

Solid and Miscellaneous Wastes

     The most abundant solid  waste material  produced by the extraction and
upgrading sections will be  the mineral matter sent as tailing to the
settling pond.  For a tar sand facility producing 10,000 bbl/day of
synthetic crude oil product,  about 25,790 tons/day of mineral matter will
be rejected in this manner  (see Figure 10).  This mineral matter probably
will be returned to the mine.

     Another 65 tons/day of mineral  matter will go the steam boilers in the
coke.  About 20 tons/day of this material will be recovered from the boilers
as bottom ash, and about 45 tons/day will go into fly ash in the flue gas.
If only cyclones are used,  only part of this fly ash will be recovered.  If
high efficiency electrostatic precipitators  are used, over 99 percent of
the fly ash will be recovered.  If a. flue gas desulfurization system is
used, some collection of particulate matter  will be done ahead of the FGD
system.  The mineral matter recovered as  bottom ash or fly ash will be
relatively easy to dispose  of because it  will be inert and dry.

     A small amount of waste  material is  produced by the amine treatment
operation used for removing H2S from off  gases.  This is a purge stream
containing primarily amines and various organic compounds produced from
amines.  The amount of this purge stream  is  about 1.6 pounds per ton of
sulfur recovered from I^S.^)  Therefore, based on the sulfur balance
previously discussed, a facility of  the size considered here (10,000 bbl/
day)  would produce about 100  pounds  per day  of this material.
                                      59

-------
              TABLE  12.  MAXIMUM EFFLUENT RATES BASED  ON NEW SOURCE PERFORMANCE
                          STANDARDS  FOR PETROLEUM REFINERIES(a)
Maximum Total Effluent Rate for 10,000
Maximum Unit Effluent Rate,
pounds/1000 bbl feedstock^
Effluent Cnaracteristic
BOD5
CODj^
TSS
Oil and grease
Phenolic compounds
Ammonia as N
Sulfide
Total chroaiura
Kaxavalent chromium
pH (diaensionless)
For Any Average
One Day 30
5.75
41.16
3.97
1.69
0.0417
6.55
0.0367
0.0833
0.00714
of Daily Values for
Consecutive Days
3.07
20.83
2.48
0.92
0.0198
2.98
0.0169
0.0486
0.00317
bbl/day
For Any
One Day
70.2
502.2
48.4
20.6
0.508
79.9
0 . 44S
1.016
C.087
of Synthetic Crude Product
pounds, dsy^c'
Average of Daily Values for
30 Consecutive Days
37.5
254.1
30.3
11.3
0.242
36.3
0.206
0.593
0.039
Within the range 6.0 to 9.0
(a)  Based on Federal Register, 40 (98),  May  20, 1975.  Standards  for Category B refinery with process
     configuration = 6.
(b)  Feedstock is tar sand oil to upgrading.  Values apply for up  to 24,900 bbl/day.
(c)  Based on 12,200 bbl/day tar sand oil to  upgrading.
Conversion factors:  kg/liter •* 0.002853  (Ib/bbl), kg/m3 - 2.853  (Ib/bbl), kg/day - 0,4536 (Ib/day).

-------
     TABLE 13 .   OBSERVED EFFLUENT LOADINGS FOR CATEGORY B PETROLEUM REFINERIES
Refinery Number
Wastewater Treatment Processes Employed
Aerated lagoon
Activated sludge
Dissolved air flotation
Equalization
Filtration
Oxidation pond
Polishing pond
Effluent Loadings, lb/1000 bbl feed
BOD,
R
COD
TSS
Oil and grease
Phenolic compounds
Ammonia as N
Sulfide
12345

XX X
X
X X
X
X
X
X X

2.8 4.4 2.1 3.6 1.3

13.8 24 34 25.0 13.8
8.7 12 3.0 1.5
0.8 3.2 1.4 1.0
0.001 0.145 0.13 0.018 0.002
1.7 0.05
0.07 0 0.010 0.005
Source:  Development Document for Effluent Limitations Guidelines  and  New  Source
         Performance Standards for the Petroleum Refining Point Source Category,
         EPA 440/l-74-014a, p 104, April, 1974.

Conversion factor:  kg/liter = 0.002853 (Ib/bbl).

-------
                  TABLE 14.   QUALITATIVE EVALUATION OF WASTEWATER FLOW AND CHARACTERISTICS.
                             BY FUNDAMENTAL REFINERY PROCESSES
to
Production Process
Wastewater Feed and
Characteristic Product Storage
Flow
BOD
COD
Phenol
Sulfide
Oil
Emulsified oil
pH
Temp
Ammonia
Chloride
Acidity
Alkalinity
Suspended
solids
XX
X
XXX
X

XXX
XX
0
0
0

0

XX
Distillation
XXX
X
X
XX
XXX
XX
XXX
X
XX
XXX
X
0
X
X
Coking/Thermal
Cracking
X
X
X
X
X
X

XX
XX
X
X
0
XX
X
Hydrotreating
X
X
X

XX

0
XX

XX
0
0
X
0
           X - Minor  contribution, XX -  moderate  contribution, XXX - major contribution, 0 - no
           problem.
           Source:  Development  Document for  Effluent  Limitations Guidelines and New Source
                    Performance  Standards  for the Petroleum Refining Point Source Category,
                    EPA  440/l-74-014a, p 18,  April,  1974.

-------
     If a flue gas desulfurization process is used, a small amount of
waste material may be produced as a result, depending on the nature of the
FGD process.  For example, the Wellman-Lord FGD process, which would be
well suited to this application, produces a purge stream by which certain
oxidation products (e.g., sodium sulfate) are removed from the system.  In
this case, processing steps can be included so that only a solid waste
material is produced.  Although the quantity of this sulfate material is
small, its high water solubility makes disposal difficult.
                                      63

-------
                                 SECTION V

              ENVIRONMENTAL COMPARISON OF TAR SAND PRODUCTION
                         AND PROCESSING TECHNOLOGY
      Information on tar sand production and processing technologies and their
potential sources of primary environmental impact have been described in pre-
ceding sections.  Drawing from that information, the environmental pros and
cons  of the production and processing technoligies are discussed below and
summarized in Table 15.

      Surface mining methods for producing tar sands would pose greater
potential for environmental impact than underground methods.  This greater
potential is traceable to problems of restoring larger surface areas dis-
turbed by surface mining to a premined condition, the greater availability
of aqueous transportable materials and the restoration of areas used for dis-
posal of spent sand.

      All mining methods pose greater potential for environmental impact than
in situ methods.  This is because tar sands produced by mining methods would
require an intermediate process to extract the bitumen from the mined tar
sand  ore.  Thus, the potential impacts of bitumen extraction process would
either occur at or in the region encompassing the mining area in the case of
a colocated extraction plant or at some other location in the unlikely event
that  the tar sand ore is transported considerable distances from the mine.

      Environmental advantages of wet versus retorting (dry) extraction pro-
cesses are not as clear cut as in the case of production methods.  Both
would generate solid waste and some process and/or cooling water effluent
with  amounts dependent upon provisions for reusing water.  Process effluents
from  wet extraction could contain hydrolyzed components of tar sand con-
stituents which would have to be traded off against the potential air
emissions from retorting.

      Tar sand oil extracted from the mined tar sand or produced by in situ
methods would go to an upgrading facility.  In this sense, the potential
environmental impacts attributable to upgrading facilities would probably be
somewhat independent of production method.  However, data on composition of
water, gas, and tar sand oil produced by the various in situ methods are
extremely meager.  The tar sand oil might be transported to an existing
coke/hydrotreating facility for upgrading or a new facility constructed.
Several factors will determine which of these options is selected, not the
least of which is the location and size of the resource base.
                                     64

-------
                       TABLE  15.    ENVIRONMENTAL  COMPARISON OF  POTENITAL  TAR  SAND  PRODUCTION
                                         AND PROCESSING METHODS
           Operation or
              Process
                                  Status of Tar Sand Technology
                                        Major Potential Technical
                                        Disadvantages or Environ-
                                             mental Impacts
                                                    Major Potential Technical or
                                                         Environmental
                                                           Advantages
           PRODUCTION

           (1)   Surface Mining
One commercial operation in
Canada;  others planned or
proposed.
Small scale open-pit projects
in U.S,
No large-scale demonstration
of technology in U.S.
Oi
Host potential for surface changes
  during mining.
Least potential for complete restoration
  of surface to preraining  condition.
Require removal and transportation of
  tar sands.
Disadvantages include those of extraction
  (see below).
Most potential for increasing availability
  of aqueous transportable materials.
Generates  most solid waste (mine waste plus
  spent sand).
Greatest materials handling requirements
  (solid waste plus tar sand).
Basic surface mining technology
applicable.

High to highest bitumen recovery.
           (2)  Underground        Applicability of available
                Mining            underground mining technology
                                  not yet investigated
                                  Require removal and transportation of tar
                                    sands.
                                 Create potential for surface subsidence.
                                 Disadvantages include those of extraction
                                   (see below).
                                             Can draw from available  underground
                                               mining technology.
                                             Potential for underground disposal of
                                               spent  sand.
                                             High bitumen recovery (longwall taethod
                                               of mining).
                                             Least potential (among mining methods)
                                               for surface changes during mining.
                                             Best potential (among mining methods)
                                               for minimizing permanent surface
                                               changes if no subsidence and spent
                                               sand returned to mine.
                                             Eliminates, stripping  operation and
                                               resultant solid wastes.

-------
                                               TABLE  15.     (Continued)
Operation or
   Process
                        Status of Tar Sand Technology
                                           Major  Potential Technical
                                           Disadvantages or Environ-
                                               mental  Impacts
                                                      Kajor Potential Technical or
                                                            Environmental
                                                              Advantages
(3)  tongwall
       Stripping
(4)  In Situ
       Production
       From Surface
Applicability to tar sands not
yet investigated.
No commercial production to date
from tar sands.

Pilot projects have been conducted
but _ln_situ_ technologies for
Conner dally producing tar eanda
are not yet demonstrated.
Require removal and transportation of
  tar sand.
Disadvantages include those of  extraction.
If used after contour mining would probably
  increase potential for environmental
  impact.
Nonuniform nature of tar eand character-
  istics poses technical problems.
Bitumen recovery expected to be  less  than
  that for surface mining.
Availability of water unless coproduced
  water treated and reused  '(depending on
  location end uethod)
Applicability of method for mining
  of coal under investigation.
Total potential for environmental  im-
  pact probably less than that  of
  other methods ol mining same
  amount of sands.
If used after contour mining, could
  recovt-r greater percentage of tar
  sand t.eposit.

Basic technologies include those of
  producing hcpvy oil reservoirs.
Produce tar sands that are too  deep
  to be produced economically by
  mining methods.
Eliminate generation of solid waste
  of mining and spent sand from
  extraction process.
Some upgrading of tar sand bitumen
  could occur underground.
Coproduced waters could be disposed
  of by subsurface treated and  reused
  in water-short areas.
Disturb much less surface area  than
  surface mining.
Eliminate potential sources of  dust
  and noise in surface mining operation.
(5)  In Situ
       Production
       Underground
Applicability to tar sands not
yet investigated.
Disadvantages probably would be similar
  to those of surface in situ
  production.
                                                                                                           U.S.S.R.  reported to be  producing
                                  (20)
  heavj oil deposit by this method
Other advantages probably would be
  similar to those of surface in^situ
  production.

-------
                                                TABLE  15.     (Continued)
Operation or
   Process
                        Status  of Tar Sand Technology
                                          Major Potential Technical
                                          Disadvantages or Environ-
                                               mental  Impacts
                                                     Major Potential Technical or
                                                           Environmental
                                                             Advantages
EXTRACTION OF
BITUMEN FROM MINED
TAR SAND
Wet Process:
  One commercial operation in
    Canada;  others  planned or
    proposed/
  Small scale or bench tests in
    U.S.
Retorting:
  Applicability to  tar sands
    generally unevaluated
    exoertmentally*
Wet Process:
  Generation  and  disposal of solid
    wastes (spent sand).
  Availability of water if process
    water not reused.
  Leached constituents in waste
    .•ater effluents.
Retorting:
  Availability of water for once-
    through cooling  systems.
  High temperature could result
    in more contaminants in syncrudc
    than wet  processes.   ^
Wet Process:
  Could result in less  contaminants
    in syncrude than retorting.
Retorting:
  Could be  less potential for water
    pollution than in wet processes.

-------
     As indicated above, development of a tar sand industry using in situ
methods of production poses the least potential for environmental impact.
From the viewpoint of maximum utilization of a tar sands resource base,  pro-
duction of tar sands by surface mining would be preferred.   In situ methods
or possibly underground mining might be possible,  however,  in lieu of sur-
face mining in environmentally sensitive areas where technical and economic
factors permit a choice of production method.
                                     68

-------
                                  SECTION VI

                     ENERGY PERSPECTIVE OF U.S. TAR SANDS
     Factors restraining  development  of U.S.  tar sands relate to or stem
 from:

     •  Need for exploration  of  the deposits  to acquire data for
        evaluating  their  commercial attractiveness and for developing
        a proven production and  processing  technology

     •  Lack of proven production and processing technologies,
        particularly  in situ  technology, applicable to U.S.
        deposits

     •  Envisioned  problems,  project  delays,  or cancellations,
        because of  possible or probable problems with environmental
        acceptability of  a tar sands  industry.  Major percentage of
        deposits underlie or  are in environs  of national parks and
        monuments,  pristine wilderness areas, and arid/semiarid
        mountainous terrain.

     •  Net energy  to be  gained  (energy balance)

     •  Competitive position  of  tar sands vis-a-vis other energy
        sources (e.g., teritiary recovery of  oil, shale oil,
        liquefaction  and  gasification of coal, and deep water
        offshore oil) for manpower, equipment manufacturing and
        construction  resources,  and available capital.

     •  U.S. tar sands, as now known, represent a comparatively
        small resource base.

     •  Economic incentives and  risks, investment climate, product
        price, and  governmental  policies.

     Given a resolution of restraints favorable for development of tar sands,
a "ball park" perspective of  U.S. tar sands is that they might potentially
represent 7 to 10 billion barrels of  syncrude, depending on how one specu-
lates on the values and combinations  of the production variables involved.
Production variables  include  those shown in Table 16 as well as the per-
centage of tar sand reservoirs produced by each of the methods.

     Ten billion barrels  of syncrude  from tar sands would be approximately
3-1/4 times the amount of natural domestic crude produced in 1975 and 6-3/4

                                      69

-------
      TABLE  16.   EFFECT OF PRODUCTION VARIABLES ON
                    UTILIZATION OF TAR SAND RESOURCES

Production
Method
% of In Place
Assumptions Bitumen Produced
Surface Mining
Underground Mining
                  (b)
100% of Ore  Recovered
 90% Extraction Efficiency

 90% of Ore Recovered
 90% Extraction Efficiency
                                                    (a)
                                                             90
                                                             81
Continuous or 55% of Ore Recovered
Conventional 90% Extraction Efficiency
Longwall 80% of Ore Recovered
90% Extraction Efficiency
In Situ 50% Sweep Efficiency^
70% Displacement (d)
75% Sweep Efficiency
70% Displacement
90% Sweep Efficiency
70% Displacement
50
72
35
50
63
 (a)  Extraction efficiency: % of bitumen recovered from ore processed at
     extraction plant.

(b)  Underground mining of  tar  sands seldom mentioned in literature.  Ore
     recovery based or experience from underground mining of coal.

(c)  Sweep efficiency: % of ore.  body's total volume from which bitumen
     is removed.

(d)  Displacement:   % of bitumen in affected volume driven  to production
     wells.
                                70

-------
times the amount of natural crude imported by the U.S. in 1975  (based on
values in Monthly Energy Review. FEA, PB-242769-12, December, 1975.  Even
if all of the 30-billion barrel tar sand resource base were commercially
producable, domestic tar sand deposits would not approach the 600-billion
barrel^-*) resource base of the oil shale deposits of the Green River
Formation in western U.S.  Consensus is that something in the neighborhood
of 7 to 10 years would be required to attain large-scale production of U.S.
tar sands, assuming a favorable resolution of the restraints listed pre-
viously.
                                       71

-------
                                  REFERENCES
  (1)  Ball and Associates, "Surface and Shallow Oil-Impregnated Rocks and
      Shallow Oil Fields in the United States", U.S. Bureau of Mines,
      Monograph 12 (1965).

  (2)  "Energy Alternatives—A Comparative Analysis", Prepared by the Science
      and Public Policy Program, University of Oklahoma, Norman, Oklahoma
      (May, 1975).

  (3)  "Energy From U.S. and Canadian Tar Sands:  Technical, Environmental,
      Economic, Legislative, and Policy Aspects", Report Prepared for the
      Subcommittee on Energy of the Committee on Science and Astronautics,
      U.S. House of Representatives, U.S. Government Printing Office
      (December, 1974).

  (4)  Ritzma, Howard R., Utah Geological and Mineralogical Survey, "Utah's
      Oil-Impregnated Sandstone Deposits, A Giant Undeveloped Resource",
      Paper Presented to Rocky Mountain Section, American Association of
      Petroleum Geologists, Annual Meeting, Salt Lake City, Utah (February
      28-March 3, 1973).

  (5)  Chapin, Hebert S. Jr., et al., Geological Survey of Alabama, "Petro-
      liferous Rocks (Mississippian Age) of North Alabama", In Preparation
      (September, 1975).

  (6)  Personal Communications.

  (7)  Heath, Larman J., et al., "Solvents and Explosives to Recovery Heavy
      Oil, Bartlett, Kansas", U.S. Bureau of Mines, Technical Progress
      Report No. 60 (September, 1972).

  (8)  Wells, J. S., and K. H. Anderson, "Heavy Oil in Western Missouri",
      Bulletin, American Association of Petroleum Geologists, Vol. 52,
      No. 6, (1968) pp 1720-1731.

  (9)  Searight, Walter V., "Asphaltic Rocks in Western Missouri", Missouri
      Geological Survey and Water Resources, Information Circular No. 13
      (1957).

(10)  Arnold, M. D., and A. Herbert Harvey, University of  Missouri,  Rolla,
      "Evaluation of Thermal Methods for Recovery of Viscous Oils in
      Missouri and Kansas", U.S. BM Contract No. G0133100, Rinal Report
      (June,  1974).
                                      72

-------
(11)  Ritzma, Howard R., Compiler, "Location Map—Oil-Impregnated Rock
      Deposits of Utah", Utah Geological and Mineralogical Survey Map
      33 (April, 1973).

(12)  Campbell, Jack A., "Oil-Impregnated Sandstone Deposits in Utah",
      Mining Engineering, Vol. 27, No. 5 (May, 1975).

(13)  Glasset, J. M., "Surface Mining of Utah Tar Sands", Paper Presented
      at First Rocky Mountain Fuels Symposium, Brigham Young University
      (January 31, 1975).

(14)  Cupps, Cecil Q., et al., Laramie Energy Research Center, "Field
      Experiment of In Situ Oil Recovery From a Utah Tar Sand by Reverse
      Combustion", Paper Presented at AICHE Meeting, Los Angeles,
      California (November 20, 1975).

(15)  Carlson, F. M., "Field Experiment of Underground Reverse Combustion
      in a Utah Tar Sand", Paper Presented at the First Rocky Mountain
      Fuels Symposium, Brigham Young University (January 31,  1975).

(16)  Wood, R. E., and H. R. Ritzma, "Analysis of Oil Extracted From Oil-
      Impregnated Sandstone Deposits in Utah", Utah Geological and
      Mineralogical Survey, Special Studies 39 (January, 1972).

(17)  Gwynn, John Wallace, "Instrumental Analysis of Tars and Their
      Correlations in Oil-Impregnated Sandstone Beds", Unitah and Grand
      Counties, Utah, Utah Geological and Mineralogical Survey, Special
      Studies 37 (October, 1971).

(18)  McConville, L. B., "The Athabasca Tar Sands", Mining Engineering,
      Vol.  27, No. 1 (January, 1975).

(19)  Cameron Engineers, Inc., Synthetic Fuels Data Handbook (1975).

(20)  "Mining of Viscous Crude Claimed Viable", Oil and Gas Journal, Vol.
      74, No. 1 (January 5, 1976) pp 46-47.

(21)  Grimm, Elmore C., and Ronald D. Hill,  "Environmental Protection in
      Surface Mining of Coal", U.S. EPA-670/2-74-093 (October, 1974).

(22)  University of Oklahoma, "Energy Alternatives - A Comparative Analysis",
      Report Prepared for the Council on Environmental Quality, Contract No.
      EQ 4AC034 (May,  1975).

(23)  Moomau, Henry F., et al.,  Potomac Engineering and Surveying,
      "Feasibility Study of a New Surface Mining Method, Longwall
      Stripping", U.S.  EPA Report 670/2-74-002, Contract No.  68-01-0763
      (February, 1974).
                                      73

-------
 (24)  Roe, K.  F.,  et  al.,  "Shallow Cover Coal Mining and the Environment",
      Report Prepared by Kansas Institute for Mineral Resource Research,
      University of Kansas for Ozarks Regional Commission, Contract No.
      TA  73-9(N) (September 1, 1974).

 (25)  "Compilation of Air  Emission Sources", U.S. EPA, Office of Programs,
      Research Triangle Park, Publication No. AP-42 (April, 1973).

 (26) ' Personal Communication, Mr. Russel Powell, Imperial Oil of Canada,
      Ltd. (January 28, 1976).

 (27)  Extraction of Energy Fuels Panel, Extraction of Energy Fuels, Prepared
      for the  U.S. Bureau  of Mines, NTIS Report No. QFR30-73 (September,
      1972).

 (28)  Heath, L. J., F. S.  Johnson, and J. S. Miller, "Solvents and Explosives
      to  Recover Heavy Oil", USBM Technical Progress Report No. 60 (September,
      1972).

 (29)  Doscher, T.  M., "Technical Problems in In Situ Methods for Recovery
      of  Bitumen From Tar  Sands", 7th World Petroleum Congress, Vol. 3
      (1967) pp 625-632.

 (30)  Mungen,  R.,and  J. H. Nicholls, "Recovery of Oil From Athabasca Oil
      Sands and From  Heavy Oil Deposits of Northern Alberta by In Situ
      Methods", 9th World Petroleum Congress, Vol. 5 (1975) pp 29-41.

 (31)  Both, R. C., "Cyclic Steam Project in a Virgin Tar Reservoir", Journal
      of  Petroleum Technology (May, 1967) pp 585-591.

 (32)  "New Cold Lake  Pilot Onstream", Oil Week (October 13, 1975).

 (33)  Harmsen, G.  J., "Oil Recovery by Hot-Water and Steam Injection",
      8th World Petroleum Congress, Vol. 3 (1972).

 (34)  White, P. C., "Tar Sands and Liquid Fuels From Coal", Reprint of
      Presentation at the Oil Daily Forum, New York City, New York (June
      10, 1974).

 (35)  Dietz,  D. N., "Wet Underground Combustion:  State of the Art",
      Journal of Petroleum Technology (May, 1970) pp 605-617.

 (36)  Utah Tar Sand Test Yields Small Amount of Oil, Oil and Gas Journal
      (January 12,  1976) p 26.

 (37)  Personal Communication, Frank Carlson, ERDA/Laramie Energy Research
      Center  (January 25,  1976).

(38)  Farouq,  Ali,  S.  M.,  "A Current Appraisal of In Situ Combustion Field
      Tests",  Journal of Petroleum Technology (April, 1972) pp 477-486.
                                      74

-------
(39)   U.S.  Environmental Protection Agency, "Draft Development Document for
      Effluent Limitations Guidelines and New Source Performance Standards
      for the Oil Gas Extraction Point Source Category" (October, 1974).

(40)   Personal Communications, U.S. Environmental Protection Agency.

(41)   Clark, K. A., "Research Council of Alberta Report 8", Annual Report
      1922, Edmonton, Alberta, Canada (1923) pp 42-58.

(42)   Clark, K. A.,  Transactions of the Canadian Institute of Mining and
      Metallurgy,  Vol. 47 (1944) pp 257-274.

(43)   Rammler, Roland W., "The Retorting of Coal, Oil Shale, and Tar Sand
      by Means of Circulated Fine Grained Heat Carriers as a Preliminary
      Stage in the Production of Crude", Quarterly Journal of Colorado
      School of Mines, Vol. 65, No. 4 (October, 1970) pp 141-167.

(44)   Ternan, M., B. N. Nandi, and B. I. Parsons, "Hydrocarcking Athabasca
      Bitumen in the Presence of Coal:  Part I:  A Preliminary Study of the
      Changes Occurring in the Coal", Canadian Mines Branch Research Report
      No. R-276 (October, 1974) pp 1-2.

(45)   Nelson, W. L., Petroleum Refinery Engineering, 4th Edition, McGraw-
      Hill (1958) p 134.

(46)   U.S.  Environmental Protection Agency, Supplement No. 1 for Compilation
      of Air Pollutant Emission Factors, 2nd Edition (July, 1973).

(47)   U.S.  Environmental Protection Agency, Compilation of Air Pollutant
      Emission Factors, 2nd Edition, AP-42 (July, 1973) pp 9.1-3 to 9.1-5.

(48)   Brown and Root, Inc., "Economics of Refinery Wastewater Treatment",
      API Publication No. 4199 (August,  1973) pp V-2 and V-9-

(49)   Beychok, M. R.,  Aqueous Wastes From Petroleum and Petrochemical
      Plants, John Wiley and Sons (April, 1973).

(50)   Federal Register, Vol.  40, No. 98 (May 20, 1975).

(51)   U.S.  Environmental Protection Agency, "Development Document for
      Effluent Limitation Guidelines and New Source Performance for the
      Petroleum Refining Point Source Category", EPA 440/l-74-014a (April,
      1974) p 104.

(52)   Battelle estimate based on contacts with process vendors.
                                      75

-------
                APPENDIX A
      ILLUSTRATIONS OF MINING METHODS
 (Figures A-l to A-6 Reproduced/Modified
from Reference 21.  Figure A-7 Reproduced/
      Modified from Reference 23)
                    76

-------
TOP Of RIDGE
^ 	 HIGHWAU 	 -^
CUT 7 I CUT 5 CUT 3 CUT)
—U- "(-<•- -+*— -*
CUT 2
— —*-
CUT 4
— -*•
CUT 6
	 OUTCROP BAKRIER- — 	 	 '"^
HOUOW
                   PROCtDUBt.
                   I.5CAIP FROM 1OP Of HIC.HWALl 1O OUTCROP BARRIER.
                    REMOVE AND STOKE (OPiOtL
                   1 REMOVt ANO DISPOSE OF OVtRBURDEN fROfA CUT i.
                   3.PICK UP COAL. 1EAVINO AF  UASI A IS FOOT UNDISTURBED
                    OUTCROP BARRIER
                   4MAKE SUCOLS1VE CUTS AS NUMBERED
                   S-OV£RBU«DtN IS MOVED IN  THE DIRECTION. AS SHOWN BY
                    ARROWS, tND PLACED IN THE ADJACENT PIT
                   ^.COMPLETE BACKFILL ANO GRADING TO THE APPROXIMATE
                    ORIGINAL CONTOUR.
                       RIDGE TOP
                         -BARRIER

                   Stripping phase
                        RIDGE TOP
HOUOW
                                                   OMPACTED
                                                  ClAY
       Backfilling phase
                                         BARRIER
      FIGURE  A-l.    BLOCK CUT  METHOD
                               77

-------
00
             PROCEDURE;
              t. SCALP FROM POINT A TO POINT B
              2..MAKE CUT A C D
              5. PLACE SPOIL TKOM A c o IN D E
              4. ESTABLISH ROAD AT POINT  E ON FILL BENCH
              S. SOIID  BENCH - POINT C TO f-O
-------
                                lit STEP (27- EXAMPLE)
 PROCEDURE:
   1. SCALP FROM TOP OF 2nd CUT HIGH WALL IO TOE OF Fill.
   2. REMOVE SPOIL FROM Ut CUT AND PUSH DOWN SLOPE.
   3. SPREAD SPOIL AND COMPACT IN LAYERS
     UNTIL STORAGE ANGLE ISBEACHED.
   4. LEAVE AT LEAST IS' BARBIE*.
   I. PICK UP COAL.
                                                                          TOE
                                                                         OF FILL
                                2nd STEP (27 • EXAMPLE)
                                SECOND CUT AN3 SPOIL
 PROCEDURE:
   1. REMOVE AND STACK SPOIL FROM 2nd CUT.
   2. PICK UP COAL.
   3. AUGER IF PERMITTED.
    DIVERSION
      OUCH
                                  SUP (27° IXAMP1EI
                      FINAL OKADIKC (ONE AND TWO CUT METHOD)
               HIGHWAll
                                                REDUCED SLOPE
                                                  IOWIR HALF
     SEAM
                           c
                            - BARRIER
PROCEDURE:
  I. COMPACI SUITABLE SPOIL IN AND ABOVE AUGER HOLES.
  2. PUSH STACKED SPOIL AGAINST H1OHWALL.
  3. SLOPE BENCH IO SPECIFIED GRADE.
  4. AT LEAST 15' OF BARRIER IS LEFT INTACT.
                                                                         TOE
                                                                        OF FILL
FIGURE  A-3.    SLOPE  REDUCTION:   ONE  AND  TWO-CUT METHOD
                                    79

-------
                                        l«t STEP (27e EXAMPLE)
        —PIT - 50•-*
  HtCUT
"-Pit.ioo
           I. SCALP FROM TOP OF 2«d CUT HIGHWALL TO TOE Of FltL
           2. REMOVE SPOIL FROM Itl CUT AND PUSH OOWNH.OPE.
           3. SPREAD SPOIL AND COMPACT EN 3' LIFTS OR LAYERS
            UNTIL MAXIMUM DEPTH IS REACHED FOR THAT DCGfftE OF
            ORIGINAL GROUND SLOPE.
           4. LEAVE AT LEAST 15' BARRIER.
           S. PICK UP COAL.
                                      2nd SUP (27° tXAMPlE)
                                                           ANGLE Of
                                                            REPOSE
         PROCEDURE:
          1. REMOVE AND STACK SPOIL. FROM 2nd CU1.
          2. PICK UP COAL-
          3. AUGER IF PERMITTED.
         DIVERSION
         0IICH
                                        3rd STEP (17" EXAMPLE)
            -~- HIGHWAlt
          PROCEDURE:
            I. COMPACT SUITABLE SfOll IN AND ABOVt AUGflt HOIES
            2. PUSH SPOIL fROM Jnd CUT AGAI^SI HIOMWAll.
            3. StOPE BiNCH TO SPfCIFIED GRADE.
            4. AI IfASI IS' OF ZAKRItR IS LEFT INIACI
            5. ROAD ON EOO6 Ot Fill EINCH IS NOT DISTURBED.
                                                             ANGlt
                                                            OF REPOSE
                                                                  TOE
                                                                 OF Fill
FIGURE  A-4.    PARALLEL  FILL METHOD,  MODIFIED  SLOPE  REDUCTION
                                            80

-------
                                         MOUNTAIN TOP
                                                               FIRST CUt
                                                                    HOSSOM
                                                                                                         SECOND CUT


                                                                                                     MOUNTAIN TOP-
                                                                                        ORIGINAl
                                                                                        GROUND
                                                                                        SlOPt.
                                                                                                                                  ItOSSOM
09
                     (RUSH DAM
                                        lOURTH CUT


                                     MOUNTAIN TOP,
                                       tARRIER
                                                     DIVERSION
                                                     DITCH
top after final grading end topsoiling
                         DIVERSION
                          DITCH
                                     FIGURE  A-5..   MOUNTAIN  TOP REMOVAL METHOD

-------
                                          MOUNTAIN TOP
                                          BtINO ([MOVED
               I                          I     9v'LI
               L«—CROWNfD Fill BENCH—•"Ls—BtNC
 lopsoii—^^p^'^*^,*^t?5i*as*?^^iY—

0 ~JZ?r;jJ^?/*'•'''  ' "'  ' P'  ','"'  i, -'"'^         I
c£~Z?; • ''fti!;ti'/i'; •'•'•'•Vi:vV;/.v.vV/.ig;!J?'        IAIIRAI
o^r?,'v;y-i:/'',i'-'v'.:iJ''.'.'^'-':''-'^'^:'           OKAIN
wfM'rfwfttMtt^&ff'^'
                                            HIGHWAU
                  ROCK FILLED .{FRENCH DRAIN),
                  NATURAL DRAIN WAY
              15CAIP INTdtf Alt* THAT Will K COVIf ED WITH fill. IIMOVI AND STORC TOfSOfc,
              {.CONSTRUCT FIUNCH DSAINS IN IHl HOUOW WATER COUIIIS
              S.6UK.D THE F1U IN COMPACTED LAYERS.
               IACE OF Fill NO Sttms THAN 3:1.
              4.CONSTIOCT «OWNIO TERRACES IVEIY 30 mi.
               APPROXIMATELY 20 FEET WIDE.
              S-CtNtlt OF COMPUTED FIU BENCH IS CROWNED
               fOW'RD THE HIGHWAU, SO THAT WATU
               WIU FLOW ONIO fXCAVAHD BENCHIS.
              4.BOHO JILT CONItOL SIRUCIURES BtLOW HOLLOW FIU.
 FIGURE   A-6.    HE/D-OF-HOLLOW  FILL
                                  82

-------
Hill. MID VAILEY

 ISOLATtO	ELEVATION
        VAWOUS TYPES OF TERRAIN APPLICABLE TO
        LONGWALL STRIPPING SYSTEM
 / ELT    L'^:^} p—j™ij™^ ^k$&°
                                                                                           BtCr.rll LCD KO /^-,
                                                                                           ORIGJNAL CONlOU't
                                                                      PLAN VIEW OF LONGWALL STRIPPING SYSTEM
                                                     1
                                                                              SURFACE
       CHOSS-SECTIOM VIEW OF LONGWALl  STRIPPING
       SYSTEM ALOI>'G HifiiiWALL.

 SECTION  fl-A

TYPICAL CROSS-SECTION VIEW OF
LONGWALL  STRIPPING SYSTEM
                            FIGURE  A-7.    LONGWALL  STRIPPING
                                                   83

-------
                                   TECHNICAL REPORT
                            (Please read Instructions on the reverse
       DATA
       before completing)
 1. REPORT NO.

   EPA-600/7-76-035
                              2.
                                                           3. RECIPIENT'S ACCESSION-NO.
 4. TITLE AND SUBTITLE
   PRODUCTION AND PROCESSING OF U. S. TAR SANDS

   An  Environmental Assessment
               5. REPORT DATE
                December 1976 issuing  date
               6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
   N. A. Frazier,  D.  W.  Hissong, W. E. Ballantyne,
   and E. J.  Mezey
                                                           8. PERFORMING ORGANIZATION REPORT NO.
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
                \
   Battelle
   Columbus Laboratories
   505 King Avenue
   Columbus.  Ohio 43201      	
               10. PROGRAM ELEMENT NO.

                 EHE623
               11. CONTRACT/GRANT NO.
                 68-02-1323
 12. SPONSORING AGENCY NAME AND ADDRESS
    Industrial Environmental Research Laboratory  -  Cin.,OH
    Office  of Research and Development
    U.  S. Environmental Protection Agency
    Cincinnati.  Ohio 45268	
               13. TYPE OF REPORT AND PERIOD COVERED
                  Final	
               14. SPONSORING AGENCY CODE
                EPA/600/12
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
      Tar  sands is a potential source of synthetic  fuel for the U- S.  If, when, to what
 extent and  at  what rate U. S. tar sands are developed in the future is dependent  to
 a large extent upon the environmental impact of  the  producing and processing of tar
 sands.  Reported here are the results of a preliminary study to assess the potential
 primary environmental impacts of production and  processing of U. S. tar sands bitumen.
 Currently there are two basic ways for producing tar sands—mining and in-situ.   Pro-
 ducing tar  sands by mining methods would be similar  to those of mining coal.  Solid
 waste in  the form of spent sand would have to  be dealt with, but existing technology
 can control it if good environmental practices are followed.  Currently there is  no
 in-situ production technology but it is believed that environmental impacts would be
 similar to  those of conventional oil field production.  Facilities used to upgrade
 tar sand  oil would pose environmental impacts  same as coking and hydrotreating  pro-
 cesses in an oil refinery.  Environmentally,  in-situ production of tar sands would
 be preferred.   From the viewpoint of resource  utilization, production by surface
 mining methods where economically and technically  possible  would be preferred.
 Technical and  ecomomic factors will determine  if in-situ methods  are an alternative
 to surface  mining in environmentally sensitive areas.
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                               COS AT I Field/Group
 Tars, Bituminens,  Oil Sands, Producing
 Wells, Mining,  Wastes, In-Situ Combustion,
 Emmission,  Sandstones, Siltstones,  Over-
 burden, Refining,  Coking, Surface Mining,
 Dust, Volatility,  Viscosity, Reserves,
 Reservoirs
  Tar Sands,  Synthetic
  fuels, Hydrotreating,
  Tar Sands  Triangle,
  Great  Canadian Oil Shale,
  Ltd, Particulates,
  Asphaltic  Rocks, Oil
  Impregnated.Rocks
     8G
     81
    13B
 3. DISTRIBUTION STATEMENT

  RELEASE TO PUBLIC
  19. SECURITY CLASS (This Report)
   UNCLASSIFIED
21. NO. OF PAGES

    92
                                              20. SECURITY CLASS (Thispage)
                                                UNCLASSIFIED
                             22. PRICE
EPA Form 2220-1 (9-73)
84
                                                    • U.b. GOVERNMENT PRINTING OFFICE: '977-757-056/5555 Region No. 5-11

-------