890000-oil shale - c g scouten
TRANSCRIPT
OIL SHALE
Charles G. Scouten
Amoco Oil Company
Research and Development Department
Naperville, Illinois
OIL SHALE
Charles G. Scouten
Amoco Oil Company
Research and Development Department
P.O. Box 400, Naperville, Illinois 60566
Prepared for publication in the Handbook of Fuel Science and Technology
J. G. Speight, Editor
Table of Contents
Page I. ORIGIN AND OCCURRENCE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
A. INTROOUCTION.................................................. 1 B. DEFINITIONS AND TERMINOLOGy................................... 4 C. ORIGIN........................................................ 5
1. Sedimentation and Mineralogy............................. 11 2. Production of Organic Matter............................. 27 3. Kerogen Types and Maturat i on. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4. Preservation of Organic Matter........................... 38 5. Marine Oil Shales........................................ 41 6. Lacustrine Oil Shales.................................... 46
D. OCCURRENCE.................................................... 47
II. CHARACTERIZATION, TESTING AND CLASSIFICATION OF OIL SHALES........ 51 A. FISCHER ASSAY: DETERMINATION OF OIL yIELD.................... 51
1. Modified Fischer Assay: The Standard Method............. 53 2. Shortcomings of the Fischer Assay.................... .... 55
B. ALTERNATIVE METHODS FOR EVALUATING OIL yIELD.................. 57 C. ELEMENTAL COMPOSITION AND OIL yIELD........................... 57 D. BITUMEN CONTENT AND COMPOSITION............................... 64
III. PHYSICAL/CHEMICAL PROPERTIES..................................... 71 A. ATOMIC H/C AND N/C RATIOS..................................... 71 B. ALIPHATIC AROMATIC AND CARBON CONTENTS........................ 74 C. NMR DATA...................................................... 74 D. NEWER NMR TECHNIQUES FOR KEROGEN CHARACTERIZATION............. 79 E. ESR RESULTS................................................... 81 F. DENSITY METHODS............................................... 82 G. THERMAL METHODS OF ANALySES................................... 88 H. HEAT CAPACITY AND HEAT OF RETORTING........................... 99 I. MECHANICAL PROPERTIES......................................... 102 J. POROSITY AND PERMEABILITy..................................... 106
IV. THE CHEMICAL STRUCTURE OF OIL SHALE KEROGENS...................... 108 A. KEROGEN ISOLATION............................................. 108
1. Sample Selection and Preparation......................... 109 2. Chemical Methods......................................... 109 3. Physical Methods......................................... 111
B. STRUCTURAL INFERENCES FROM BITUMEN AND SHALE OIL ANALySES..... 115 C. STRUCTURAL ANALYSES OF KEROGENS............................... 118
1. Oxygen Functional Groups................................. 118 2. Oxidation................................................ 120
a. Al kal ine Permanganate............................... 121 b. Chromic Acid Oxidation.............................. 131
3. Depolymerization under Mild Conditions: Heat-Soak/Extraction.................................... 131
4. Micropyrolysis/GC-MS..................................... 135 5. ACid-Catalyzed Hydrogenolysis............................ 138
D. STRUCTURAL MODELS FOR KEROGEN................................. 138 1. Burlingame Model of Green River Oil Shale Kerogen........ 139 2. Djuritic-Vitorovit Model of a Green River
Oil Shale Network....................................... 141
- i -
3. Green River Kerogen Model Proposed by Schmidt-Collerus and Prien.............................. 143
4. The Yen-Young-Shih Multi-Polymer Structure............... 145 5. Characterization of Organic Material in
Rundle Ramsay Crossing Oil Shale........................ 149 6. Generalized Models for Types I, II and III Kerogen....... 161
V. RECOVERY.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 A. OVERVIEW...................................................... 173 B. MINING TECHNOLOGy........ ......... ............................ 176
I. Underground Mining....................................... 176 2. Surface Mining........................................... 179
C. SIZE REDUCTION................................................ 180
VI. RETORTING......................................................... 183 A. ABOVE-GROUND RETORTING PROCESSES.............................. 185
I. N-T-U Process... ... ... ... ..... ..... ... .......... ..... .... 185 2. Gas Combustion Process................................... 190 3. Lurgi-Ruhrgas Process.... .......... ............ ...... .... 195 4. TOSca II Process......................................... 204 5. Shell Pellet Heat Exchange Retorting (SPHER) Process..... 211 6. Shell Shale Retorting Process (SSRP)..................... 213 7. Exxon Shale Retort (ESR) Process......................... 219 8. Chevron STB Retorting Process............................ 220 9. Petrosix Process......................................... 225
10. Moving Grate Processes................................... 229 (a) Allis-Chalmers Roller Grate Process................. 229 (b) Dravo Circular Traveling Grate Process.............. 232 (c) Superior Oil/Davy-McKee Circular Grate Retort....... 236
11. Paraho Process........................................... 238 12. Unocal Oil Shale Processes: Retort A, retort B, SGR..... 248 13. Kiviter and Galoter Processes (U.S.S.R.)................. 263 14. HYTORT Process (IGT)..................................... 273
B. IN SITU RETORTING...... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 1. Laramie True In Situ Retorting Studies................... 285 2. Geokinetics Horizontal In Situ Process................... 289 3. Occidental Vertical Modified In Situ (VMIS) Process...... 294 4. Rio Blanco Modified In Situ Process...................... 300 5. Equity Oil/ARCO BX Project............................... 308
VII. SHALE OIL UPGRADING AND REFINING................................. 312 A. COMPOSiTION................................................... 312 B. UPGRADING STRATEGIES.......................................... 316 C. DEWATERING AND SOLIDS REMOVAL................................. 318 D. ARSENIC AND IRON REMOVAL...................................... 320 E. UPGRADING BY NON-CATALYTIC THERMAL PROCESSES.................. 323 F. CATALYTIC PROCESSING METHODS.................................. 325
1. Sohio ............... :.................................... 326 2. Gulf Shale Oil Upgrading Process......................... 334 3. Catalytic Cracking (Ashland, Chevron).................... 336 4. Amoco Hydrotreating-Hydrocracking........................ 338
VII I. CONCLUSIONS. . . . . . . . . . . . . . . . . . . . • • . . . . • . . . . • . . . . . . . . • . . . . . . . . . . . . 344 REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . 345
i i
I. ORIGIN AND OCCURRENCE
A. INTRODUCTION
Oil shales comprise a truly enormous. and largely untapped fossil fuel
resource. As readily accessible petroleum sources dwindle, utilization of the
oil shale resource to meet world needs for energy and chemical feedstocks will
become both necessary and economically attractive. World-wide oil shale
deposits are estimated to contain 30 trillion barrels of shale oil, but only a
small fraction of this amount is easily recoverable using current technology.
Thus, the utilization of oil shale to replace petroleum will mean finding
economically efficient and environmentally acceptable methods for recovering
the energy-rich organic material locked inside the oil shale's rock matrix and
for upgrading the recovered shale oil. This is a formidable challenge. Oil
shales are complex and intimate mixtures of organic and inorganic materials and
vary widely in their compositions and properties. Some, such as the "oil
shale" of the Green River formation are not even true shales. Today. some
three centuries of research and development lie behind, as we address the
scientific, technological and environmental aspects of the oil shale challenge.
OIL SHALE Page 2
Oil shale technology has a long history. Oil shales were sources of oil
as early as 800 AD and the British oil shale deposits were worked in Phonecian
times.' The use of oil shale was recorded in Austria in 1350. The first shale
oil patent, British Crown Patent No. 330, was issued in 1694 to Martin Eele,
Thomas Hancock and William Portlock, who "after much paines and expences hath
certainely found out a way to extract and make great quantityes of pitch, tarr
and oy7e out of a sort of rock".2 Oil shale utilization on an industrial
scale did not, however, follow immediately. Not until 1838 was the first
industrial oil shale plant put into service at Autun, France. 3
Scotland (1850), Australia (1865) and Brazil (1881) followed.
Soon, plants in
In addition to
being a source of refined shale oil products, it was soon discovered that
torbanite, an especially organic-rich type of oil shale, was useful for
increasing the luminosity of illuminating gas flames. 4 This provided an
important market for the early Scottish oil shale industry and later, as the
Scottish torbanite deposits were depleted, for the early Australian, shale
industry. By the 1870 1 s, Australian torbanite was being exported not only to
Great Britain, but also to the United States, Italy, France and Holland. s The
invention of the Welsbach gas mantle and the advent of low-cost, high-quality
kerosene from American petroleum spelled the end of this period. As the need
for liquid transportation fuels increased, the Australian oil shale operations
consolidated. Elsewhere, oil shale plants followed in New Zealand(1900),
Switzerland (1915), Sweden (1921), Estonia (now USSR, 1921), Spain (1922),
China (1929) and South Africa (1935).6 The high point of this stage of oil
shale development was reached during, or just after, World War II. However,
OIL SHALE Page 3
the oil shale industry in Estonia and neighboring Leningrad Province still
flourishes, most of the mined shale being burned directly in electric power
generating plants with the remaining 10% or so retorted to provide chemical
feedstock and smaller quantities of refined products. 7 The Japanese began in
1926 the commercial production of shale oil from the large Chinese oil shale
deposits at Fushun in Manchuria. Improved retorts were installed at this
complex in 1941 to provide important supplies of liquid fuels for the Japanese
forces during World War II. At Maoming, near Canton in southern China, a
second oil shale project was developed. Shale oil production in the PRe peaked
about 1975 and has since declined as emphasis has shifted to newly-discovered
petroleum supplies. 1 ,7
The near-term future of oil shale is uncertain. Very clearly, this future
will be influenced by international crude oil prices and supplies. Indeed, as
pOinted out by Hutton, et aT., the rise of interest in oil shale during the
late 1970's was due largely to the high prices and tight supply of crude oil.8
With the decline of crude prices and growth of a crude oil surplus, interest in
oil shale and other synfuels waned. How long this will continue is difficult
to predict. At present, commercial oil shale development is hard to foresee,
unless impetus is provided by political or security concerns. However, the
current lull in development activity offers a golden opportunity for scientific
research to attack the many chemical, physical and material problems that were
uncovered or brought into sharper focus during the late period of activity.
Prudence suggests that we sieze this opportunity.
OIL SHALE Page 4
B. DEFINITIONS AND TERMINOLOGV
There is no "easy!! scientific definition of oil shale; the definition is
strictly an economic one: According to Gavin, "Oil shale is a compact,
laminated rock of sedimentary origin, yielding over 33% of ash and containing
organic matter that yields oil when distilled, but not appreciably when
extracted with the ordinary solvents for petroleum."9 Materials containing
<33% ash should be considered coals, however this distinction will be of little
importance in the following discussion. Thus, we will use oil shale to denote
an organic-rich rock that contains little or no free oil.
Shale oi7 is defined as the oil produced from an oil shale on heating.
Three other terms will used extensively, hence their definitions are
important: Bitumen is defined as the organic material which can be extracted
by ordinary organic solvents, such as benzene. toluene, tetrahydrofuran (THF)
and chloroform (CHC1 3 ) or mixtures (generally azeotropes) of such solvents,
such as benzene-methanol (60:40). Kerogen, which comprises the major part of
the organic material, is not soluble in such solvents. 1o Of course, these are
operational definitions and the relative proportions of bitumen and kerogen
depend on the choice of solvent and extraction conditions. Nevertheless, these
are useful definitions provided their limitations are kept in mind. The third
term. kerogen concentrate, refers to the organic concentrate that is produced
by beneficiation or chemical demineralization of an oil shale. Strictly
speaking. this term should refer only to that part of the organic concentrate
that is insoluble in organic solvents. However, in common usage, kerogen
concentrate refers to the total organic material {kerogen + bitumen} obtained
by removing minerals. The common usage will be followed in this discussion,
except where specifically indicated.
OIL SHALE Page 5
C. ORIGIN
Oil shales were formed by the inclusion of materials containing organic
carbon into sediments that eventually become sedimentary rocks. To understand
their origin and formation, we begin by considering the distribution and
cycling of carbon in the environment.
Carbon cycles through the biosphere by photosynthesis and oxidation.
C02
Atmosphere
Photosynthesis
Degradation (oxidation)
Combustion
Organic Carbon
Fossi! fuels
Figure 1. Carbon cycle in the biosphere.
Plants use CO, from the atmosphere to build their cells. Animals eat the
plants and return CO2 to the atmosphere as a product of metabolism. The decay
of dead organisms returns additional CO2 to the environment. The action of
acids on carbonate rocks liberates CO2 into the atmosphere, while rain and the
precipitation of insoluble carbonates remove CO2 • Only a minute fraction of
the carbon that cycles through this marvelously complex cycle is preserved in
the sediments that yield fossil fuels.
Hunt has estimated the amount of carbon present in sedimentary rocks as
hydrocarbons (including N-, $- and O-containing hydrocarbon derivatives that
contain reduced carbon forms) and as oxidized carbon forms (Table 1).11
OIL SHALE Page 6
Table 1. Global carbon inventory_ Carbon, 1018 grams in sedimentary rocks.a
All Sediments
Clays and Shales Carbonates Sands Coal beds thicker than 15 feet
Non-reservoir Rocks
Aspha It Petroleum
Reservoir Rocks
Aspha It Petroleum
Organic Carbon (reduced forms)
Insoluble Organic Matter
8,900 1,800 1,300
15
Soluble Organic Matter
275 265
0.5 1.1
-13,000
Carbonate Carbon (oxidized form)
9,300 51,100 3,900
-64,000
(a) Data from Ref. 11. One barrel of oil contains about 105 grams of carbon.
While these data are only approximate, they do indicate that about one
sixth of the global carbon is present in reduced forms. Unfortunately. much of
this reduced carbon is widely dispersed an not easily recovered. This is
especially true for the very large amount of reduced carbon held in clays and
shales, where the organic matter is almost always a minor -- often a trace -
constituent. Consequently, only a small fraction of the carbon in clays and
shales is likely to be useful as a fuel resource.
OIL SHALE
Table 2. Shale-Oil Resources of the World's Land Areas8 ,109 bblb
Recoverable Known
Range in grade, Resources Known Resources oil yield in gal/ton' 10-100 5-10 10-25 25-100
Africa 10 ne ne ne
Asia 20 2 3,700 ne
Australia and New Zealand 24 ne 84 200
Europe 30 100 200 ne
North America 80 900 2,500 4,000
South America 50 ne 3,200 4,000
Total 214 1,000 9,600 8,000
(a) Data from Duncan and Swanson, Reference 3j Cane, Reference 5.
(b) To convert bbl to m', divide by 6.29.
Page 7
Total Resources c
5-10 10-25 25-100
4,000 80,000 450,000
5,500 110,000 590,000
1,000 20,000 100,000
1,400 26,000 140,000
3,000 50,000 260,000
2,000 40,000 210,000
17,000 325,000 1,750,000
(c) Includes oil shale in known resources and anticipated extensions of known resources.
(d) To convert gal/ton to L/t, divide by 0.2397.
OIL SHALE Page 8
As shown in Table 2, the current oil shale technology is well suited for
processing the richest (organic-rich) oil shale deposits; those that yield over
10 barrels of shale oil per ton of rock. Using the data in Tables 1 and 2, we
can calculate that present oil shale technology can efficiently recover only
0.06% of the potential oil in relatively rich shales (>10 gal/ton) -- and only
about one-billionth of the total carbon in clays and shales! The challenge is
clear; even a marginal improvement in technology could greatly increase this
minute fraction of recoverable shale oil.
Compared to known -- or even anticipated -- petroleum resources, the oil
shale resource is enormous. This is not surprising, since shales are source
rocks for many of the world's crude oils. Moreover, geochemical studies indi
cate that only a small fraction of the carbon in a source rock is typically
released as liquids into migration channels that lead to a petroleum reservoir.
Indeed, many crude oils can be regarded as shale oils produced by natural
heating and upgraded into petroleum during migration from the shale source rock
to the reservoir. Thus, a more useful comparison is between the proven and
anticipated reserves of crude oil, coal and natural gas with the shale oil that
could be recovered using current or improved technology (Table 3).3,12,14
OIL SHALE Page 9
Table 3. World reserves of shale oil, coal, bitumen and asphalt and petroleum.
Shale Oil
Resource in place 4,300
Recoverable 30
Billions of Tons Natural
Coal Bitumen Natural Gas Petroleum
11,000 3,000
663 75
200
80
544
130
(a) Data from References 3,5,12,14 and 77.
Thus, both the coal and oil shale resources dwarf the petroleum resource;
the problem is economical recovery. This is difficult to achieve because of
the widely varying properties of oil shales. The characteristics of some of
the world's best-studied oil shales are summarized in Table 4.14
OIL SHALE Page 10
Table 4. Properties of SOO\e IllflOrtant oil shales from around the world. (Data of Thorne, H. M.; et at. USBM Circ 8126, Reference 14.)
Modified Fischer assay 01 L, L/tC
Oil, lOt X
Water, lOt X Spent shale, lOt X Gas and loss, wt X
Austral ia (Glen
Davls)a
414 30.0 0.7
64.1 4.3
Conversion of organic material to oil d, lOt % 66
Rock characteristics Sp. gr. (at 16-C) Heating value, MJ/Kg f
Ash, lOt X Organic carbon, lOt %
Assay oil $p. gr. (at 16·C)
carbon, lOt % Hydrogen, lOt %
Nitrogen, wt %
Sulfur, lOt %
Ash analysis, lOt X
SiOz
At"" ",0 ,.0 HgO
other oxides
1.60 18.8 51.6 39.8
0.89 85.4 12.0 0.5 0.4
81.5 10.1 3.0 0.8 0.8
3.8
Grall I (Treme!!bereubate)a
156 11.5 6.2
78.4 3.9
59
1.70 8.2
71.4 16.5
0." 84.3 12.0
1.1 0.2
55.8 26.7 8.5 2.8 3.7 2.5
,,,, .... (Nova
Scotia)b
257 18.8 0.8
12.7 2.7
12.6 62.4 26.3
0.88
61.1 30.1 5.0 1.1 1.6
1.1
129 9.7 3.2
84.0 3.1
44
2.03 8.9
70.8 18.8
0.90 84.9 11.4 0.8 0.3
55.1 27.6 9.3 1.7 1.9 4.4
New Scotland South Man- ZeaLand Israel , ...
Garek)b Leba- churia (Ore-nonb (FushLt1)b pukOb
78 307 6.4 24.8 2.2 11.0
88.4 56.5 3.0 7.7
48
60.0 10.6
0.97 79.6 9.8 1.4 6.2
ca 26
ca 45
18.8
0.96 83.2 10.3 0.6 1.5
38
3.0 4.9
90.3 1.8
33
2.29 3.4
82.7 7.9
0.92 85.7 10.7
62.3 26.7
6.1 0.1 1.8 3.0
331 24.8 8.3
57.6
9.3
45
1.46 21.3 32.7 45.7
0.90 83.4 11.8 0.6 0.6
44.2
28.1 20.5 4.6 1.4 1.2
111
8.2 2.2
86.6 3.0
56'
2.22 5.9
12.8 12.3
0.88
55.7 25.1
9.9 2.6 3.1 3.6
Africa (Erm' etO)a
228 17.6 3.0
75.6 3.8
34
1.58 19.1 42.5 43.8
0.93 84.8 11. 1
0.6
61.3 30.5 2.9 1.6 1.7 2.1
Spain
<Puertolleno)a
234 17.6 1.8
78.4 2.2
57'
1.80 12.5 62.8 26.0
0.9<J
0.9 0.3
56.6 27.6 9.1 2.6 2.2
1.9
Swedei'! O:varn- Thailand torp)a (Maesod)b
70 5.7 2.0
87.2 5.1
26
2.09 9.0
72.1 18.8
0.98 85.0 9.0 0.7 1.7
62.4
17.6 10.7 1.2 1.7 6.4
357 26.1 3.8
66.3 3.8
71
1.61 15.4 56.4 30.8
0.88 84.4 12.4 1.1
0.4
60.8
19.9 4.8
3.3 3.8 7.4
Unl ted
States (Colorado)a
122 9.3 1.0
87.5 1.6
70
2.23 5.1
66.9 11.3
0.91 84.6 11.6 1.8 0.5
43.6 11.1 4.6
22.7 10.0 8.0
(a) Average sample. (b) Selected sample. (c) To convert LIt to gal/short ton, multipLy by 0.2397. Cd) Based on the recovery of carbon in oiL from organic carbon In the shale. (e) Carbon content of oil ca. 84 lOt %. (f) To convert MJ/kg to STU/lb, multiply by 430.4.
OIL SHALE Page II
1. Sedimentation and Mineralogy
Oil shales are fine-grained sedimentary rocks in which are dispersed a
minor fraction of organic matter as tiny particles. Formation of an oil shale
requires simultaneous sources of both fine-grained minerals and organics, under
conditions where the organics can be preserved. In addition to these clastic,
detrital components, biogenic and authigenic minerals are also present tn most
shales. Detrital minerals typically include quartz, feldspar and clays (often
including illite, montmorillonite and kaolinite), and sometimes volcanic ash.
Biogenic minerals include amorphous silica and calcium carbonate, usually in
very minor amounts. Authigenic minerals typically include pyrite and other
metal sulfides, carbonates (calcite, dolomite, siderite), chert and phosphates.
Authigenic silica from clay diagenesis is also an important mineral in many
shales, serving to cement together the larger detrital particles. The saline
minerals, trona, nacholite, dawsonite and halite, are often important in oil
shales (e.g.Green River shales) formed in stratified lacustrine environments.
Not surprisingly, halite (NaCl) is usually present in marine oil shales.
The mineralogy of some representative oil shales from around the world is
summarized in Tables 5 and 6.
OIL SHALE Page 12
Table 5. Mineralogy of some oil shales (After H;rus, Reference 16.)
ShaLe Mh MineraL Amorphous Feld· Clay GyP'''' Pyrite Calcite Magnesitea Siderite (Xl matter sHica ard 'W minerals (CaSO, " (FeSZ) (CaC~) (MgC0:3) (FeC03)
(%) quartz 0:> (Xl (Xl 2 HzO (%) (%l <Xl (%) (Xl
Kukersite, Estonia 36.3 47.87 9.0 6.75 13.9 1.1 4.25 56.1
Kohst, H.W.F.P. India 68.7 8 ... 3 12.40 2.47 40.68 0.49 trace 22.68 8.34 3.76
Broxburn, Main 67.4 76.15 16.55 11.30 45.85 0.43 1.76 2.91 2.63 11.24
Kirrmericige, Dorset 37.8 40.89 38.97 5.74 20.68 8.56 4.64 3.51
Ermelo, Transvaal 44.9 47.85 50.13 5.14 29.45 0.24 2.03 1.73 0.24
Tasmanite, Tasmania 79.2 82.05 56.3 6.0 23.75 1.45 1.64
Antlerst, Burma 43.9 46.78 34.33 5.63 27.45 5.49 0.19 trace
Boghead, Autm 65.0 79.2 32.4 n.d. 17.4 1.1 0.7 37.3
P~ef'Ston 1 75.0 83.45 24.6 n.d. 22.9 trace 2.35 5.8 4.15 2.15
P~erston II 66.3 86.77 19.3 n.d. 22.9 0.3 1.35 26.7 12.1 5.1
Middle Dumet 77.6 84.76 26.5 n.d. 54.65 0.3 0.55 4.25 3.65
Newnes, N.S.W. 20.1 20.79 74.0 n.d. 17.9 0.3 0.4 3.3
Cypris shale, Brazi I 65.9 69.50 66.33 n.d. 17.13 0.78 1.24 5.25 0.62
Massive shale, Brazil 72.8 48.5 n.d. 37.2 37.2 0.4 1.4 2.6 1.1
n.d. '" not determined. (., Apparently lIimus did not consider presence of dolomite (Editorial conment).
OIL SHALE Page 13
Table 6. Mineralogy of Selected Australian and Overseas Oil Shale Samples. (After Hutton, Reference 17.)
Satrple Specific Quartz Clay Calcite Siderite Dolomite Pyrite Apatite Others , of Clays Feld$par Gravity Min$l'sls " " II Ch
Condor (Carb.) 1.66 A A T 85 15 T Jolt. Coolon A A T 85 15 Nagoorin (Carb.) 1.60 A A T T 75 " Horwell 1.12 Not determined, less than 1% mineral matter
Alpha 1.10 A A T T 70 20 10 Glen Davis 1.16 A A T F fOO Joadja 1. 17 A A T a5 15
Byfield 1.92 A A T T T B 30 20 50 Condor (Brown) 2.41 A A T T T B 70 20 10 Duaringa (Top) 1.52 A A T T 95 5 Duaringa (Base) 1.60 A A T T 30 55 15 ,-'" 1.66 C A T T ao 20 Hereor!n (Brown) 1.94 A A T T ao 5 15 Nagoorin South 1.65 A A T T 25 75 T RU'ldle (Me Member) 1.41 A A T T 20 70 10 RlI'Idle (Re Menber) 1.62 A A T T T 25 65 10 Stuart (lie Menber) 1. 76 A A T T 15 55 30 Stuart (Ke Member) 1.52 A A C T 5 90 5 ,,""" 1.48 c A T 60 40
Green River 2.01 C T A T T A,
Canooweal 2.50 A A A A T " 25 50 Jut ia Creek 2.02 A A A T 20 20 50 fO Irati * 2.15 A A T T 35 65 5 Kentucky (C HeRber) • 2.24 A A T T 5 5 ao fO Kentucky ($ Shale) '* 2.12 A A T T 5 5 ao fO Paris Basin '* 2.26 A A C T T 25 10 55 10 Hersey River 2.04 A A T T 5 75 20
Coorong! te 1.01 onl y surface sand
Carb."Carbonaceous MC-Munciuran Creek RC-Ramsay Crossing IIC-Hutpy Creek KC-Kerosene Creek Ka-Kaol inite Mo-MOI"Itrrori lIonite ll-lttite en-Chlorite A-abundant C-cOfm'W)n T-trace S-8uddlngtonite (trace) F-Feldspar (traca) Ac-Analcime (coomon) T-trace (much less than 5%.) *-Percentage of total clay minerals No entry indicates not detected
OIL SHALE Page 14
Minerals in the Green River Oil Shales: Mineral assemblages vary from shale to
shale, from top to bottom within a given oil shale deposit; even from sample to
sample within a given stratum. This complicates mineralogy studies. As a
result, comprehensive mineralogy studies of oil shales are few; most studies
have focused on the occurrence or genesis of one particular mineral or mineral
type, while other studies have concentrated on- minerals of economic interest.
This makes it difficult to get a broad view of the mineral assemblages in even
major deposits. The Green River formation has been extensively studied and a
summary of its mineralogy. based largely on the work of Bradley18,19 and
reviews by MiltonZO and SmithZ1 , serves to illustrate the range of minerals
that can be present in a single oil shale deposit:
Table 7. Major minerals in Green River oil shale. a
Ubiquitous
Quartz SiD, III i te KA 1, (A 1 Si, lO, 0 (OHl,
Dolomite CaMg(CO,), Calcite CaCO,
Pyrite, Marcasite FeS, Nachol ite NaHCO,
Soda Feldspar NaA1Si,O, Short ite Na,Ca, (CO,),
Potash Feldspar KA1Si,O, Trona Na, Co, • NaHCO, • 2HzO
Dawsonite NaAl (OH),CO,
(al After Milton, Reference 20.
OIL SHALE Page 15
Table 8. Authigenic SiLicates in Green River Formation (Modified after MiLton, Ref. 20).
Hami of mineral
Clay minerals Kaolinite Stevens i ta
Loughlinite Sepiolite
TaL c
Zeolites Analcite Natrotite Hermotome"wellsite ClinoptiLolite"
Mordenite
Boros; lleates Searles! te Garrels; te leuc9sphenite Reedmergnerite
Other 51 lleates Quartz Orthoclase Albite
Acmite Riebeckite labunts9vite
Vinogradovlte Elpidl te Hatron-catapleiite Biotite Hydrobiotite
f-Qf'muLa
H4AlZSix09 (ALO.06feO.04M9Z.81liO.04)
(Si3.9S A1 0.OZ)010(OH)Z N8 0.04 H16Ns2M93Si60Z4 H4M9ZSi3010 M93(OH)25;4010
NaAlS1Z06 H20 N8ZAlZSi3010.6 H20 (Ba,Ca,KZ)AlZSi6010' 6 H20
NaBSiZ0 6 • 6 H20 (8a,Ca,M9)BZ Si 06(OH)3 CaBaNa3BTi3Si90Z9 NaBSi30S
Si02
KaLSi30a
NaAlSi30a NaFeSIZ06 Ha(FeMg)3 Fe Z(OH,F)(Si4o11>Z (K,Ba,Ha,Ca,Mn,Ti,Hb)
(Si,AlZ(O,OH)7 HZO Ha5TI4AlSi60Z4 • 3 HZO H6HaZZrSI6018 H4 (HaZCa)ZrSi3011 K(Fe,MS)3AlSI3010(OH) (K,H20)(Mg,Fe,Mn)3AtSi3010(OH,HZO)Z
Abundance
Loca It Y abundant
Locally abundant Locally abundant Rare Rare
Widespread Rare Rare
Rare
Locally abundant Rare Rare Rare
Ubiquitous Widespread Widespread Rare Rare
Rare Very rare Very rare Very rare Very rare Very rare
OIL SHALE Page 16
The Green River oil shales were formed in the waters of ancient lakes that
covered much of what is now northwestern Colorado, southwestern Wyoming and
northeastern Utah (Figure 2). There were two, and possibly three. of these
lakes; lake Gosuite in Wyoming. Lake Uinta in Colorado and an unnamed lake in
Utah. These lakes existed for a very long time, some 4-6 million years through
the Early and Middle Eocene. As climatic conditions varied the lakes expanded,
sometimes (at least partially) merging, and contracted. Nevertheless, contin
ued subsidence caused by downward warping of the lake beds enabled the deposi
tion of sediments that are more than 2000 feet thick in some areas (Figure 3).
Green River oil shales have a very fine texture. In a composite sample
from the U.S. Bureau of Mines site at Rifle, CO, Tisot and Murphy23 found that
>99 wt% of the mineral particles were smaller than 44u, 75 wt% were in the
range of 2~20p and 15 wt% was smaller than 2p. Surface area of these particles
was low and mainly external surface. Some mineral micropore structure was
found; pore sizes ranged from 10 to 100 A. Essentially all of the organic
matter was in particles lodged between the mineral grains; it was estimated
that less than 4% of the organic matter was contained within mineral particle
pores.
The lithology of the Green River formation is complicated, in part at
least because fluvial material was deposited by creeks and rivers at the edges
of the lakes; edges that moved as the lake levels rose and fell. McDonald has
extensively reviewed this lithology.23 There are, however, some striking
uniformities. The composition of the organic matter in the oil shale is re
markably uniform, especially within the Piceance Creek Basin on which the
following discussion will concentrate. The most prominent variation going from
top to bottom of the deposit is a decrease in carboxylic acid groups (increased
decarboxylation) in the more deeply buried shales. lateral uniformity is also
OIL SHALE Page 17
Figure 2. The Green River Formation. (Reproduced with permission from
Reference 22. Copyright 1980, Colorado School of Mines.)
10 (I 1(12010 1: ...
-~-' 10 (I 10 20 IOMi "' .. rt !
P/77.1.7il GltEEN RIVER ~ FOII.MATION
s
CLEAR CREEK
'" - .. .::-, " " " \' , ,
\' ,\ " " " \'
" " , " " "'''/,?> ' Y. ' LEGEND PARACHUTE
~ OIL SHALE WITH ;j: CREEK '\ ~ NAHCOLITE MEMBER '" ~--
17'7'7::1 Oil SHALE WITH ~~", '" ,/ /' "'"" DAWSONITE Co ", ,_... .-"'f. ,/ !:"':":! HALITE AND ~ .... _ ...
NAHCOLITE 1 ~. ~BASE LEACHING Y
WYO,
,,10 ---: «- p'
,.,. &<l$ln
~ IO~ IteOllce "IV II: Creek
, S '" I cC::> location of (rau uel;on
o S«(lie in milu , ,
j , I , I 6 ,
UINTA fORMATION
N
SURFACE
Figure 3. South-north cross-section of Colorado's Piceance Creek Basin.
(Reproduced with permission from Reference 22. Copyright 1980,
Colorado School of Mines.)
0 -r CO
'" roo » r ~
7000
;; • z 0
ro~ 5000
'000
OIL SHALE Page 19
remarkable. Many fine details in core samples can be tracked across distances
of 100 mile.s and major features can be tracked over twice this distance. This
uniformity of organic matter is the more remarkable given the great diversity
of the accompanying minerals and suggests that it may be possible to draw some
general conclusions about the conditions of deposition. To explain the lithol
ogy and the chemistry that formed the organic-rich oil shales, Bradl ey25 pro
posed that the ancient lakes were stratified through much of their existence.
Though Bradley later moved away from his proposal, it was developed and greatly
extended by, Smith and his co-workers (Figure 4}.2 1 ,22
---- ---- ---- ----- _____ c"_C"_"""" -__ ~Chemoclin.-"--""'7
Monimolimnion
Figure 4. Stratified lake,
OIL SHALE Page 20
Mineral Deposition in a Stratified lake: The lake history can be divided into
three main periods. During an initial period of 1-2 million years, clay-rich
oil shales (Garden Gulch and Douglas Creek Members in the Piceance Creek Basin,
Tipton Shale in the Gr.een River Basin) were deposited. Following Smith's
argument, Lake Uinta probably began as a normal, shallow lake and deposited
typical lacustrine sediments for a considerable period as it grew, both in
depth and extent. Thermal density differences abetted by low circulation in
the isolated lake probably initiated stratification and hydrolysis of
aluminosilicates began to build up chemical stratification. Using albite as an
example, G'arrels and Mackenzie give the following example: 26
(1 )
2 NaAISi,O. + 2 CO, + II H,O --) AI,Si,O, (OH), + 2 Na' + 2 HCO,' + 4 H,SiO,
Albite Kaol inite
In this reaction, CO2 (from decomposition of organic matter) is consumed
as albite is transformed into the clay, kaolinite, while sodium and bicarbonate
ions are produced. Thus, as the lower, clay-rich sediments were deposited the
concentrations of dissolved sodium carbonates increased in the lower stratum of
water just above. This would increase the density of the lower layer of the
stratified lake. At the same time, CO2 production depleted the oxygen content
OIL SHALE Page 21
of the lower layer providing a reducing environment conducive to preservation
of new organic sediment falling from the oxygenated and productive upper layer.
As HzS was produced by sulfate-reducing bacteria, the precipitation of pyrite
and marcasite would be limited only by the availability of iron in solution.
Eventually, precipitation of the sodium carbonates, such as nacholite (NaHC03 ),
commenced. The resulting transition from normal lacustrine sediments to oil
shale was not abrupt, and there is no evidence for evaporite deposition at this
point. low circulation in the lake is indicated by even deposition of organic
matter in tiny layers (varves). Under these conditions large clastic matter
would be deposited near the lake shore and only airborne particles and the
finer particles would be deposited in the deeper, stratified zone further from
shore. Note that the anoxic lower layer (monimolimnion) in Figure 4 does not
extend to the shoreline. The upper layer (mixolimnion) was oxygenated, hence
capable of depositing normal lacustrine sediments, including the larger clastic
particles" in the shallow water along the water's edge. Of course, as the lake
expanded and contracted with variations in rainfall and runoff the shore moved
and with it the region where lacustrine sediment was deposited.
The s:econd stage in the evolution of the Green River formation, was a
relatively arid period. As the water level dropped, the large lakes separated
into several smaller lakes in what we now know as the Piceance Creek (Colorado),
Uinta (Utah), Green River and Washakie Basins (Wyoming). These isolated lakes
had individual characteristics that are reflected in a series of saline miner
als, largely nacho1ite (NaHCO,), halite (NaC1), dawsonite [NaA1(OH),CO,] and
some nordstrandite [Al(OH),] in the Piceance Creek Basin. Mostly trona
OIL SHALE Page 22
[Na,CO, • NaHCO, • 2 H,O] and shortite [Na,Ca,(CO,),] were deposited during
this period in the Green River and Washakie Creek basins of Wyoming.
In the Piceance basin, the lake was relatively deep and remained strati
fied. As evaporation proceeded, the concentrations of sodium and aluminum ions
in the lower layer increased to the point that precipitation of the nacholite,
which is 3-4 times less soluble than Na2C~, commenced. When both CO2 and
aluminum ions were available dawsonite was precipitated, with occasional
nordstrandite [Al(OH)3] precipitation occurring under conditions of CO2 deple
tion.
In contrast, the Wyoming lakes were shallow and did not remain stratified
as evaporation proceeded to dryness, or nearly so. This exposed the lower
layer of t.he lake to the atmosphere leading to increased oxidation of organic
matter. A.s a result sediments formed during this period contain little organic
material. Also, exposure of the concentrated aqueous NazC03 solution of the
lower layer to atmospheric COz led to trona precipitation, accompanied by
calcite and shortite limited by the availability of calcium. Intermittent rain
falling in this playa lake may have contributed to trona evaporite formation by
dissolving previously-deposited NazC01, thereby exposing it to atmospheric COz.
Under these conditions, trona deposition would no longer be limited by the rate
of COz production from organic decomposition. Deposition of today's thick
trona beds was the result.
The presence and types of authigenic minerals is an important indication
of the conditions present during sedimentation and subsequent burial. For
example, the presence of authigenic sulfides and high levels of organic materi
al together in an oil shale suggests deposition and burial in an anoxic, re-
OIL SHALE Page 23
ducing environment of low Eh (Eh is the oxidation potential of the solution,
referred to the standard hydrogen half-cell).15 Since the typical surface
waters (rain water. flowing streams, ocean surface) are strongly oxidizing,
this probably means an environment of low circulation, such as a lower level in
a stratified lake or sea, where oxidation of the organic matter is minimal.
The fine grain size of the detrital clays generally found in in oil shales is
also consistent with this kind of stagnant depositional environment.
Oxidation-Reduction and Hydrogen Ion Potentials: In general, the deposition of
non-clastic, non-evaporite minerals can be understood in terms of oxidation
reduction (redox, Eh) and hydrogen ion (pH) potentials, subject of course to
the availability of aqueous ionic species. Figure 5 shows a classification of
sediment types based on Eh and pH and the region occupied by natural
environments is shown in Figures 6 and 7. Note that the region of Figure 5 and
the region of Figure 7 where the points are most dense, correspond to only the
center of Figure 6. Most oil shales of interest were formed within a very
restricted region of Eh - pH space, namely the region where pH > 6 and Eh is
below the "sulfate-sulfide fence." This was alluded to in the discussion of
Green River oil shales above and will be referred to often in subsequent
discussions; Figures 5-7 are key to our understanding of oil shale formation.
OIL SHALE Page 24
Figure 5. The relation of Eh and pH to the (non-evaporite) minerals in
sedimentary rocks. (After Pettijohn27 and Krumbein and Garrels28 )
eH
-0.3
pH • 7.0 I
HEMATITE LIMONITE MN OXIDES SIUCA
CHAMOSITE'
0
~~ ORGANIC MATTER
c~ •• 0
o. '" SILICA -4,.~
" ORGANIC PEAT MATTER
w SIDERITE u < w GLAUCONITE -J
RHODOCHROSITE • CHAMOSITE •
~~ >
PHOSPHORITE , --===:: w
<
sU{PJ.flltr;
" • '" " ~ -
w u < w -
- SUt PItIOf:
ORGANIC PEAT MATTER PyRITE PHOSPHORITE
PYRITE SILICA
'* c~o""",'. ," ",." "*'. " ,~p'''.'''o'''. 01 Ih • .. d""~"'{I'Y """ "';'0''',
w < 0 >
• w , J
8.0 I ,
CALCITE
FENCE ftl1 • 0
CALCITE ORGANIC
MATTER '0
'co
CALCITE ORGANIC
MATTER
""f.lct:
CALCITE ORGANIC
MATTER
OIL SHALE Page 25
Figure 6. Eh and pH characteristics of some natural environments. (After
Garrels and Christ, Reference 29. Reprinted with the permission of
Harper & Row) Inc.}
EH
+0.8 ~
-+0.6 ,~
:+0.2 -
0.0 .
-0.2
-0.4 -
-0.6
-0.8
• 1.0 o!;---!2:----4-!---!6~--!:6---1,l,0---IL2-P..,H.,...J14
OIL SHALE Page 26
Figure 7. Measured Eh and pH of some natural environments. (Baas Beeking,
et aI.3o Reprinted with the permission of Pergamon Press.)
! 000
800
600
400
Eh 200
Zero
-200
-400
-~-----~
" , , , .' , .,
Iy,
';,
"
-600,L---~O~--~2~--~----~6~--~8--~~I~O----~12~--pH
OIL SHALE Page 27
2. Production of Organic Matter
Earlier. the terms kerogen (insoluble organics) and bitumen (soluble in
organic solvents) were defined. These are the biogenous materials that are the
major reason for our interest in oil shales. Many oil shales also contain
readily discernable skeletal fragments, fossilized fish bones, insect parts,
diatoms and even bacterial cell walls. These fossils are interesting and are
important indicators of conditions through geological time, but generally have
little effect on the technological properties of oil shales. Oil shales depos
ited in near shore areas generally contain significant amounts of woody terres
trial material that have an important adverse effect on oil generating
potential.
As shown in Figure 1, only a minute fraction of the carbon cycling through
the biospn,ere becomes incorporated into sediments, and of that only a small
part is preserved to become fossil fuels. It is important to understand how
conditions, affect this process because it helps to identify potential oil shale
basins, to locate oil shale deposits within these basins, to predict the geome~
try of oil shale seams and to pinpoint the most valuable areas that have thick
deposits of high-yielding oil shale. Without trying to be too precise, three
oil shale types can be identified by depositional environment: marine shales
that were deposited in salt water are widespread, lacustrine shales that were
deposited in fresh water are extensive as discussed above, while the less
common, coal-associated shales were formed in areas subjected to uplift which
altered the environment from aqueous to swampy. In each case, characteristics
of the oil shale organics were determined by two factors: production and
preservation.
OIL SHALE Page 28
Production of Organic Matter in Aqueous Environments: Photosynthesis by
phytoplankton, especially blue~green algae (Cyanophycaej, diatoms,
dinoflagellates, and in warmer waters by phytobacteria, is ultimately
responsible for producing nearly all of the organic matter in the ocea05. 31 ,32
The relative importance of different life forms to primary productivity in the
aquasphere through geologic time is summarized in Figure 8. 33
A precise quantitative relationship between the fossil record and organic
productivity is not available; many productive species have no hard body parts
that are easily preserved, hence their productivity may not be reflected in the
surviving fossils. Nevertheless, as pointed out by Moore, the fossil record
indicates significant differences between production in marine and lacustrine
environments. 34 These are summarized in Figures 9 and 10, where the forms that
are predomjnantly organic in composition are shown in black, while other forms
(e.g. silicious diatoms) are unshaded.
Photosynthetic productivity is mainly controlled by light, temperature and
nutrient availability. Light seems not to be the controlling factor, except in
deep water, polar regions or turbid coastal regions. 31 The availability of
CO2 , which is essential for photosynthesis, is not limiting in the "eutrophic
zone" -- the top 200 meters, or so, where productivity is highest. Rather,
nutrient availability -- often nitrogen (as ammonium or nitrate) in marine, or
phosphate in lacustrine, environments -- seems to be the most common limit on
productivity.35 The availability of vitamins, especially of vitamin 812 , and
other trace organics is important, and seems to be a limiting factor in some
OIL SHALE Page 29
Figure 8. Importance of phytoplankton types to primary production through
geologic time. (after Tappan and loeblich;33 reprinted with the
permission of the Geological Society of America.)
estimated total amount
(a general lrend) Distribution with respect to type of organisms
Tertiary
• • - e • 2 .,! ... "0 '! ic-----+ '" _"0 _E _"i: .: <'II ';::: 0 II
CretaeoUli
Jurassic o .D ... -uw.,!O> 1-----+ .• --0 -~ - ~ ·-"'--11-41--11-11--1
Triassit: a
Permian
Cambrian
Precambrian
100
ISO
150
'" '00
\\0
600
E
OIL SHALE
Figure 9. Main groups of
reprinted with
upper'
Pali1eo~oic
lowe~
;upper
,lower
Triaute
Devonian
Silurian
Ordovician
Precambrian
" a w ~
microfossils in
the permission
<n w<n aw w~
:r" ~~
<n <n~ :r OW (J J:" a (J"
" -~ ~ "'~ " ~o z '" <n z ::> (J >-
" J:O
Page 30
marine environments. (Moore;34
of Springer-Verlag.)
<n ~
" z <n <n
" a "' 0 :r w w " <n ~
~ ~ 0
'" 0 z 0
<n " N ;; a 0 0 0 ~
,~ 0 " (J 0 " ~ z " ~
(J 0 (J 0 Ow 0 " 0 w
(J" " '" ~
(J ~ ... 15 z 0 -" '" ~
0 " ,,~ :r 0 " <n 0 (J (J 0 <n~ (J ~ '" 0 (J <n
•
OIL SHALE Page 31
Figure 10. Main groups of microfossils in lacustrine environments. (Moore;34
reprinted with the permission of Springer-Verlag.)
« a: Z
if) if) Z w w w W l-
I- <.!> « a: ...J Z « U Z <.!> 0 ...J « ::;: « :::> ...J Q. 0 ...J W co .... « if) Q. Q. a:
Tertiary ;Iower
lCretaceous
Mesozoic urassic
riassic
IP,'rlmian
upper QUS
'Ionian
lower ~Ordovician
mbrian
Precambrian
OIL SHALE Page 32
areas, Productivity in lakes follows the same general pattern as in oceans,
but deposition in the confines of a lake is much more subject to local climatic
variations and periodic changes in clastic input (e.g. volcanic ash) than is
marine deposition.
3. Kerogen Types and Maturation
While most of the organic matter in oil shales was produced in an aqueous
environment. some terrestrial material is often present. Moreover, conditions
of burial affect the nature of the organic matter. The concepts of kerogen
type and kerogen maturation provide a useful framework for understanding and
classifying these effects.
Kerogen types, I, II and III, are determined by the kind of debris that is
deposited in the sediment. As initially deposited in a "recent sediment", each
type of sediment has a (reasonably) characteristic range of composition. As
the sediment becomes buried deeper and/or hotter and for a longer time, the
organic material in the sediment undergoes "maturation" to give oil, gas or a
mixture of the two. The following discussion is based largely upon information
about kerogen type and maturation that has been developed in the context of
petroleum exploration. 11 ,13,38
Type I kerogen is rich in lipid-derived aliphatic chains and has relatively
low polyaromatic and heteroatom contents. The initial atomic H/e ratio is high
(1.5, or more) and the atomic o/e ratio is generally low (0.1, or less). Such
kerogens are generally of lacustrine origin and have very high oil generating
potentials. Organic sources for the Type I kerogens include the lipid-rich
OIL SHALE Page 33
products of ualgal blooms" and the finely divided and extensively reworked
lipid-rich biomass in oil shales that, like those of Green River, were
deposited in stable stratified lakes. The rubbery material, Coorongite, that
results from periodic Botryococcus blooms in the Coorong district of Australia
and the sediments of Big Soda Lake (Nevada), a stratified, saline lake provide
contemporary examples of Type I kerogens. 38 ,39 Oil shales that contain Type I
kerogen include the Autun and Campine boghead shales, and torbanite (Scotland),
as well as the Green River oil shales. Tasmanite is a marine sediment that
contains Type I kerogen.
Type II kerogens include most of the marine oil shales, of which many are
important petroleum source rocks. Atomic HIC ratios are generally lower than
for Type I kerogens, while generally higher o/e ratios reflect more ketones,
carboxylic acids and esters. Organic sulfUr levels are also generally higher,
generally reflecting more thiophenes and in some cases sulfides, as well. The
oil generating potentials of Type II kerogens are generally lower than those of
the Type I kerogens; that is less of the organic material is liberated as oil
upon heating a Type II kerogen (at the same level of maturation, see below).
The organic matter in these kerogens is usually derived from a mixture of zoo
plankton, phytoplankton and bacterial remains have been deposited in a reducing
environment. The Devonian shales of the U.S. and Canada (Chattanooga, Sunbury,
New Albany, Duvernay), the Jurassic shales of Europe (e.g. Toarcian, Paris
basin), and the Paleozoic shales from North Africa contain Type II kerogen.
Type III kerogen is found in coals and coaly shales. Easily identified
fossilized plants and plant fragments are common, making it clear that this
type of kerogen is derived from woody. terrestrial material. These materials
OIL SHALE Page 34
have relatively low atomic H/C ratios (usually < I), relatively high atomic O/C
ratios (0.2 -0.3 or even higher). Aromatic ~nd heteroaromatic contents are
high and ether units (especially of the diaryl type) are important, as expected
for a lignin-derived material. Oil generating potentials are low, while gas
generating potentials are high. No oil shales contain predominantly Type III
kerogen, but many oil shales contain some clastic material of terrestrial
origin. Because this material is easily identified under the microscope. it is
often disproportionately emphasized in petrographic studies. Moreover, because
of its high aromaticity a small contribution of clastic Type III material can
greatly complicate attempts to elucidate the chemical structure and reactivity
of Type I or Type II kerogens. Thus, in oil shale studies it is important to
identify and understand the contribution of Type III kerogen.
After the relatively rapid alterations that take place shortly after the
initial deposition of organic matter in a sediment, the surviving (preserved)
organic matter undergoes additional changes. This is the process of kerogen
maturation, which is responsible for the generation of crude oil and natural
gas. Oil shales contain relatively immature kerogen; that is the kerogen has
not been extensively "cooked", except where exposed to an unusual geothermal
gradient, such as an intrusion of volcanic magma (geologic sill or dike). Over
geologic time, however, further alteration does occur. Thus to understand the
chemical structure and reactivity of oil shale organics, it is necessary to
know not only the origin but also the maturity of the kerogen.
The van Krevelen Diagram provides a simple way of graphically representing
the relationships between kerogen type, kerogen composition and maturation
(-Figure II).
OIL SHALE Page 35
Figure II. The van Krevelen Diagram traces the evolution paths of Types I. II
and III kerogen as a function of increasing burial (increased time
temperature exposure). Note that maturation eventually leads to a
hydrogen- and oxygen-depleted, carbon-rich residue without regard
for kerogen type. (After Tissot and Welte, Reference 13, p 152.
Reprinted with the permission of Springer-Verlag).
, T
9 l' , E 9 • "
'0
0'
o~--~c---__ ~--~----~----~--~~ o 000 0.10 0,15 020 0.25 0.:';0
T"" I
][
][I
---~ •• Atomic mHo OIC
~ Mo,n potl> at oomic e<Xll. {elh" ()Y,ond .1 01,1976) ....... _- Boondar;"s of lile htld a! ktroqtn
....- [voMion paIn. of tile p,jndfl(ll lyPU of ke'<><JM
Age Ind / or f6rm.tion Buin,country Green Ri.e< sholes (PoIt<lCtr>e _ E<K,,"") Ui!\lo. Uloh, USA
AIQOI M'09_ j BoI1yoc<x:cl/$, e1c .. J.VQrious <:>it >IIoles
lOwe, Toore;on st>ote. f'<I<i$.F<"<If'IC",W.Germony
$iIo.ttion .tmI<!. Soh<;l<o,AIQ"'" ond libyQ
Vorio...$ oil "",,"S
UWorC~Ioc_ Ooual<l,eom."""", UIwo. Monnv;~ """te. Albedo, Conod"
1"Dwe. Morwwillt' $IlQIes (Me Iver ,19671 Alb .. Io,Conod<>
• 0
• • • 0
• •
OIL SHALE Page 36
As shown above, maturation involves the loss of hydrogen and oxygen from
the kerogen. Oil shales are generally immature and only the earlier stages of
maturation, termed diagenesis, need be considered. During diagenesis, hydrogen
is lost primarily as methane and other light hydrocarbon gases, water and
hydrogen gas, while oxygen is lost primarily as water and carbon oxides. These
losses are, however, significant in determining the processing characteristics
of the oil shale kerogen in retorting for liquid fuel production.
As maturation proceeds into the "oil generation window", the atomic Hie
ratio decreases due to hydrogen loss as hydrogen-rich liquids that are expelled
from the rock (Figure 12); see Brooks40.41 for discussions of recent progress
in understanding kerogen maturation.
Maturation increases with increased exposure of the organic matter to
time, temperature and pressure. The catalytic effects of minerals in the oil
shale accelerate this process. Thus, maturation involves exposure of organic
matter to a time-temperature-pressure-catalysis history. Similar effects are
seen in "simulated maturation" experiments in the laboratory and in oil shale
retorting to produce shale oils.
OIL SHALE
% WEIGHT OF INITIAL ORGANIC MATTER I SEDIMENTARY ORGANIC MATTER)
..... . ': .. :.'
: : .. " ...
•• SIMPLE MOLECULES (CO!"H!,O,CH4,H!'J
CARBONACEOUS RESIDUE
~
Page 37
A
c
Figure 12. Evolution of hydrocarbons, water and carbon oxides from kerogen
during maturation. (After Durand, Reference 36. Reprinted with
the permission of Editions Technip.)
Legend: I. CO2 + H2 0 2. Oil 3. Wet Gas 4. Methane (dry gas) 5. Resins + Asphaltenes 6. Kerogen 7. Base-solubles (humic and fulvic acids) 8. Acid-solubles (hydrolyzable material, e.g. lipids)
OIL SHALE Page 38
4. Preservation of Organic Hatter
A fundamental problem in basin geochemistry is to understand how organic
stratigraphy data from rock analyses relate to the occurrence and variability
of oil shale beds. Efforts to understand how the preservation of sedimentary
organic matter forms kerogen types I, II and III led to the concept of organic
facies· (Figure 13).42
* The geologic term facies refers to a stratigraphic rock unit, differentiated
from adjacent or associated units by appearance or characteristics that usually
reflect its ori9in.43
The relationship between kerogen type and the origin of the sedimentary
organics was discussed above, as was relationship between organic preservation
and the redox potential of the benthic (bottom) environment. Two additional
facets are needed to round out the background for the organic facies concept:
First. kerogen type strongly correlates with depositional redox potential;
that is, Type I and II kerogens are usually found in situations where the
depositional environment was strongly reducing. Under these conditions, organic
preservation was high and reworking was done primarily by anaerobic bacteria
which attack hydrocarbons, especially aliphatic chains (e.g. lipid-derived waxy
material), much less aggressively than their aerobic counterparts. Further,
anaerobic bacteria feeding on non-lipidic plankton remains are efficient
producers of lipids. Thus. anaerobic bacterial action can actually increase
the content of lipid material that converts easily into shale oil upon heating.
OIL SHALE Page 39
Figure 13. Organic facies mapping depends upon a clear understanding of the
relationships between early diagenetic factors (i.e. the benthic,
or "bottom tl environment) and kerogen type. (After Oemaison, et
a7.~ Reference 42. Reprinted with the permission of John Wiley
Sons, Inc.)
BACTERIAL AEROBIC
OXIDATION
ORGANIC FACIES
OXIC lAEROBIC)
LOW .... ,---
IV III
GAS-PRONE TO
NON SOURCE
SUB-OXIC (DYSAEROBIC)
III
MODERATELY OIL-PRONE
TO GAS-PRONE
ANOXIC (ANAEROBIC)
WATER
I
STRONGLY (cmltlt OIL- PRONE twAtIUISnCS
OIL SHALE Page 40
In contrast, organic degradation in the oxidizing surface waters of a shallow
swamp tends to give coaly material containing highly aromatic Type III Kerogen.
In part, at least, this is because aromatics and hydroaromatics are much less
attractive substrates for bacterial growth than the aliphatic chains of lipids
(i.e. aromatics are less biodegradable). Thus, not only is the initial aromatic
content of terrestrial material generally higher than that of the algal and
bacterial material, but the aromatics are selectively preserved.
Second, Type II kerogens are often associated with situations where the
environment was moderately oxic, but the depositional rate was high. Under
these conditions, the role of high depositional rate was to minimize the time
the organic sediment was exposed to benthic reworking. Of course this means
that organic productivity also had to be high if the organic content of the
resulting oil shale is high enough to be useful. Otherwise, we would have a
little organic diluted in a lot of sedimentary rock!
This leads to to a description of the four organic facies types:
Organic facies I (strongly oil-prone) are typical of the strongly anoxic
environments of stratified lacustrine and marine systems. Organic contents are
often high. The strata are laminated, the absence of bioturbation (mixing)
indicating the absence of benthic worms in the depositional environment.
Organic facies II (oil-prone) is typical of anoxic environments or of the
moderately oxic environment with high depositional rate. Carbon contents are
generally lower than in facies type I, often in the range of 1 - 10% TOe. The
sediments may be partially laminated, showing evidence of benthic worm burrows
and possib1y evidence of bioturbation by other benthic organisms.
OIL SHALE Page 41
Organic facies III (gas-prone) is typical of the mildly oxic conditions in
coal swamps or shallow marine environments. Any planktonic or algal material
deposited in such an environment will usually be degraded quickly by benthic
bacteria and/or worms. Preserved material of aquatic origin is usually thor
oughly bioturbated.
Organic facies IV (non-source) is typical of aquatic environments where
organic matter spends a long time in the oxic zone where aerobic bacteria are
active. Such environments occur even at great ocean depths with circulation,
as well as in shallow, circulating seas with high energy input. low organic
contents in the rocks reflect efficient degradation, even where the initial
organic productivity was high. Extensive bioturbation usually precludes the
observation of laminations (varves).
5. Marine Oil Shales
Marine oil shales are usually associated with deposition in one of two
settings (Figure 14).
The anoxic silled basin shown in Figure 14a can occur in the shallow water
of a continental shelf. High phytoplankton growth rates near the surface will
give a high deposition rate. The sill shields the trough from the circulation
of oxygen-laden water. Under these conditions, the decomposition of organic
sedimentary matter will rapidly deplete oxygen within the confines of the
baSin, thereby providing the strongly anoxic (reducing, low Eh) environment
that is needed for efficient preservation.
The anoxic zone in an upwelling area (Figure 14b) arises from circulation
of an open*ocean current over a cold, oxygen-depleted bottom layer. Mixing of
OIL SHALE
Figure 14a. Anoxic silled basin.
ANOXIC SILLED BASIN
I I
14b. Upwelling region.
t SILL
(CRATON)
ANOXIC ZONE CAUSED BY UPWELLING
( UPWELLING CURRENT-too-
ANOXIC ZONE
Page 42
OIL SHALE Page 43
nutrient-rich current. such as the Gulf Stream, into the CO2 - and light-rich
eutrophic zone gives an environment capable of sustaining very high organic
production. Such environments occur today along the west coasts of Africa and
the Americas, where good fishing is found along with the potential for or9anic
rich sediments.44
Information about the nature of organic matter in marine environments has
resulted from studies of recent deposits and the contemporary oceans. 45 - 47 Only
a small part of primary production in the oceans reaches the bottom. Of an
estimated annual production of 9 x 1019 tons of dry matter, Trask has estimated
that about 2% reaches the floor in shallows and only about 0.02% in the open
sea. 4S The major part of marine primary production is consumed by predators;
most of the rest by microbes. The principal marine microbial scavengers are
bacteria that live free in the water or are attached to organic particles. In
ocean water, organics occur in solution, in colloidal suspension and as partic
ulate matter comprising bodies and body fragments of living and dead organisms.
Except in -regions of a seaweed or plankton "bloom", the dissolved organics
usually predominate. As a result, marine bacteria are most abundant only in
the very upper part of the water column and in the organic detritus at the very
bottom. Even in the oceans, the adsorption of organics onto inorganic detri
tus, such as the silica parts of diatoms, plays an important part in sedimenta
tion.
After the organic sediment reaches the bottom, reworking begins. Bottom
dwelling (benthic) organisms feed on both the sediment and dissolved organics
and, in turn, are fed on by predators (e.g. crustaceans). In this sphere, the
benthic bacteria are largely responsible for the decomposition of organics and
the synthesis of new organics through enzymatic transformations. Moore has
estimated that 60-70% of the sedimentary organic carbon is typically liberated
OIL SHALE Page 44
as CO2 during this reworking, while most of the rest is converted into new
compounds. The result is a very complex mixture.
Cane; has discussed the contributions of various organic compound classes
to oil shales. 48 The compound classes included carbohydrates, 119nin5, humates
and humic acids, lipid-derived waxes and the saturated and polyene acids in
a 1 9a 1 1 i p.ids whi ch can serve as precursors of these waxes, and bi 0109; ca 1
pigments and their derivatives (e.g. carotenoids. porphyrins). Only the latter
three were judged to have sufficient inertness to be major contributors to oil
shale kerogens.
Degens and co-workers have studied the role of proteins, carbohydrates and
humates in marine sediments.49~52 Protein-derived materials included both
original and altered proteins and their decomposition products (amines. amino
acids, amino complexes). Carbohydrates are rapidly hydrolyzed and generally
not important in oil shales. Humates can be important, even in marine shales,
when deposited near shore, though obviously humic material has been found in
marine sediments that were deposited far from land. It is possible that some
humic material may be derived from proteins and/or carbohydrates, perhaps when
these materials are adsorbed on inorganic particulates (e.g. clays, volcanic
ash) in a moderately oxidizing environment.
lipids are produced by phytoplankton and also synthesized from carbohydrates
by microbi.a1 activity.47 With respect to oil shales, the polyene fatty acids
are especially interesting. It is well-known that adverse conditions can lead
to very high lipid production by algae. At low temperature and with limited
oxygen. for example. Chlorella may produce lipids to >75% of their body weight.
Much of these lipids is unsaturated. Abelson and co-workers found that polyene
acids in lipid fats from Chlorella disappeared upon heating, while saturated
acids remained unaltered. 53 Evidently, the polyene acids polymerized. Cane
OIL SHALE Page 45
has extensively studied the role of unsaturated lipid acids in the products
left by Botryococcus braunii blooms in the Coorong of southern Australia. 4s
Polymerization of the unsaturated lipid residue gives "coorongite", a tough,
resilient, insoluble material that, in many respects, resembles kerogen. 150-
and anteiso-fatty acids have been found in some oil shales, but these are
probably secondary products. leo and Parker have found that bacteria are active
in transforming n-alkanoic acids into the branched iso- and anteiso-acids. 54 ,55
Other bacterially-induced transformations include the hydrogenation of oleic
and linoleic acids, decarboxylating and polymerizing alkano;c acids and lipid
hydrolysis. 56 . 58
Carotenoid pigments have been found in many oil shales, and in petroleums
and coals, as well. A recent article by Repeta and Gagosian discusses the
carotenoids and carotenoid transformation products in the waters and sediments
of the Peruvian upwelling and provides leading references to much earlier work
on carotenoids in marine sediments. 59 Studies of the carotenoids isolated from
DSDP (Deep Sea Drilling Project) cores from the Quaternary sediments in the
Cariaco Trench shows that the chemistry of these materials is largely reductive
and traceable over 50,000 - 350,000 years. 60 ,61 This work gives useful insight
into the diagenetic transformations of carotenoids which lead to the observance
of partial1y- and perhydrogenated carotenoids in marine oil shales. 62
The black marine shales formed in shallow seas have been extensively
studied, as they occur in many places. These shales were deposited on broad,
nearly flat sea bottoms, hence usually occur in thin deposits (10-50 m thick),
but may extend over thousands of square miles. The Irati shale (Permian) in
Brazil extends over more than 1000 miles from North to South!.' The Jurassic
marine shales of western Europe (Toarcian, Paris Basin,13 also source rock for
Paris Basin crudes; Kimmeridge, England,64-66 also an important source rock
OIL SHALE Page 46
for North Sea crude 011 67 ), Silurian shales of North Africa and the Cambrian
shales of northern Siberia and northern Europe are other examples of this kind
of marine oil shale.
6. Lacustrine Oil Shales
The lacustrine oil shales of the Green River Formation, which were dis
cussed above, are among the most extensively studied of sediments. However,
their strongly basic depositional environment is certainly unusual, if not
unique. Therefore, it is useful to discuss the characteristics of the organic
material in other lacustrine shales.
A particularly thorough study was made of the lacustrine sequences from
the Permian oil shales of Autun (France) and the Devonian bituminous flagstones
of Ciathness (Scotland).68 Several series of biomarkers were prominent in
extracts from these shales: hopanes, steranes and carotenoids. Algal remains
were abundant in both shales. Blue-green algae, similar to those that contrib
uted largely to the Green River oil shale kerogen, were found in the Devonian
shale, for which a stratified lake environment similar to Green River has been
proposed. 6'9 In contrast, Botryococcus remains were found in the Permian Autun
shale and are presumed to be the major source of organic matter, except for one
sample. No Botryococcus remains were found in this sample and the oil produced
by its retorting was nearly devoid of the straight-chain alkanes and l-alkenes
which are prominent in oils from Botryococcus-derived shales. Evidently,
some, as yet unidentified, algae contributed to the organic matter in this
stratum. Biodegradation cannot be ruled out, but seems unlikely due to the
lack of prominent ;so- and anteisoalkanes. Straight-chain alkanes and I
alkenes were also evident in gas chromatograms of the retorted oils from the
Devonian shale. However, in this case a pronounced hump, which usually indi-
OIL SHALE Page 47
cates polycyclic derivatives, was also prominent. Both extracts and oil from
the Devonian shale were found to be rich in steranes and tricyclic compounds.
Oi- and triterpenoids have been suggested as precursors for the di-and
tricyclic compounds found in many oil shales. 7o - 72 Rock-Eval pyrolysis results
indicate that these shales have high hydrogen indices; the kerogens are all
Type I or Type II, with one of the Devonian samples being clearly Type I.
Other major lacustrine oil shale deposits include the Triassic shales of
the Stanleyvil1e Basin in Zaire and the Albert shales of New Brunswick, Canada
(Mississipian).
D. OCCURRENCE
Oil shale deposits are located throughout the world. 3 • 73 ,74 but it is the
features of the major deposits (Table 9) that are of particular interest in the
present context. Such depOSits are of interest because of their potential to
produce large amounts of oil and (hopefully) having minimal recovery problems.
OIL SHALE Page 48
Table 9. Age and geologic setting of major oil shale deposits.
Average Estimated Type of Grade, Resource, Materi a1 Geological Age Geologic Settjng gal/ton lQ6 barrel s
Green River Oil Shale Tertiary Eocene Stratified lake 25 4,300,000 (22)
Phosphoria (Montana) Oil Shale Permi an Mari ne PI at form (76)
Eastern U.S. (Sunbury, New Albany,
Chattanooga, etc. Oil Shale Devonian Sha 11 ow Sea 10 2,600,000 (75)
Alaska Tasmanite Jurassic Marine 10 Large (3 )
Oil Shale Mississippian Mar; ne PI atform 10 ?
Austria
Australia
Rundle, Stuart Oil Shale Tertiary 65,000 (8)
Toolebuc Oi 1 Shale Cretaceous Sloping Marine Platform 18 365,000 (8 )
Condor Oil Shale Tertiary 107,000 (8)
Dauringa Oil Shale Tertiary 20 63,000
New South Wales Torbanite Permian Marine 33 40 (8,48)
Tasmania Tasmanite Permian lacustrine 42 55 (8,48)
OIL SHALE Page 49
Table 9. Age and geologic setting of major oil shale deposits.
Average Estimated Type of Grade t Reserves, Materi aJ Geological Age Geologic Setting gal/ton lQ6 barrel s
Brazil
Irat i Oil Shale Permi an Marine Platform 21 800,000 (3)
Tremembe-Taubate Oil Shale Tertiary lacustrine 15-18 2,000 (3,74)
Marahu Marahuito Tertiary Boghead, Coaly 42 Small (3,74)
Canada (Albert) Oil Shale Mississippian lacustrine 13-30 150,000 (3)
China (PRC), Fushun Oil Shal e 01 igocene Boghead 15 1,000 (7)
France
Autun Oil Shale Permian Boghead, Coaly 10-18 160 (3)
Severac-le-Chateau Oil Shale Jurassic Marine 10 250 (3)
Germany Oil Shale Jurassic Marine 12 2,000 (3)
Great Brita in
Kimmeridge Oil Shale Jurassic Marine 10-45 1,000 (3 )
Scotland (Lothians) Oil Shale Carboni ferous Lacustrine, Estuary 16-40 400 (3)
Italy (Sicily) Oil Shale Triassic Matine 25 7,000 (3 )
OIL SHALE Page 50
Table 9. Age and geologic setting of major ojl shale deposits.
Average Estimated Type of Grade t Reserves, Material Geological Age Geologic Settjng gal/ton lQ6 barrels
Israel, Jordan, Syria Oil Shale Cretaceous Marine 10-25 120,000 (3 )
Morocco
New Zealand
Orepuki Oil Shale Tertiary 45 8 (3)
Nevis, Otago Central Oil Shale Tertiary Boghead 11-15 300 (3)
Portugal Oil Shale Carboniferous Boghead 30 80 (3 )
Spain Oil Shale Tertiary lacustrine 25 280 (3)
Sweden Oil Shale Cambrian-Ordovician Marine 10-25 3,000 (3)
Union of South Africa
Ermelo Torbanite Permian-Carboniferous Boghead, Marine 25 130 (3 )
USSR (Estonia) Kukersite Ordovician Marine Platform 50 6,500 (3 )
Yugoslavia
Aleksinac Oil Shale Tertiary Lacustrine 45 210 (3)
Zaire (Congo) Oil Shale Triassic Lacustrine 25 100,000 (3)
OIL SHALE Page 51
II. CHARACTERIZATION, TESTING AND CLASSIFICATION OF OIL SHALES
Early Scottish workers estimated the quality of cannel coal by the curl of
its shaving when cut with a knife. Similarly, workers at Fushun, Manchuria
found that rich shale curled when chipped with a glass shard, while lean shale
crumbled. There is still a need for methods to evaluate oil shale quality, and
such qualitative methods are still used, especially in the field. However,
these methods have now been supplemented with a variety of Quantitative methods
employing the latest in laboratory instrumentation to determine chemical and
physical properties. In this section, the more important methods currently
used for ~haracterizing, testing and classifying oil shales will be discussed;
the properties of twelve representative oil shales will be compared and the
methods used to determine these properties will be discussed to illustrate the
general approach.
A. FISCHER ASSAY: DETERMINATION OF OIL YIELD
In the discussion above, the term, "oil yield", was used repeatedly, but
without defining how the values are measured. Related terms include "grade"
(usually in gal/ton), which is synonymous with oil yield, while "rich" and
"lean" refer to shales with higher and lower oil yields, respectively. The
Fischer Assay -- more specifically the "Modified Fischer Assay" (ASTM 03904-80)
is the standard method for measuring oil yield. Because of its importance,
it is appropriate to introduce the Fischer Assay as the first topic in this
section which deals with measuring the properties of oil shales.
The aluminum retort was originally designed by Fischer and Schrader for
assaying coals by pyrolysis. 7! It proved to be superior to existing methods
for analysis of oil shales. 79 In early practice, a sample of -20 g was heated
OIL SHALE Page 52
rapidly to a maximum temperature of 520-550°C, using a gas burner, and the
experiment was concluded 15 minutes after the final drop of distillate was
collected. Stanfield and Frost produced a version holding 100 9 of sample and
containing aluminum discs to minimize local overheating (Figure 15).80
'"
"(> --
, <l,a '"'
w" ,
:; "'6j~J,.'nl 'n __ I' ,., ,
"-; I __ "'~"P<>
0'.' 1(/ ~_ OH~
00 0«' 'l_~
D3904
, o ,
" ,
'0'~' .. . . • . , ,
"" ~,- ; ,"
',,' ""1 "n ~~ IT/a ,., 0 ....
L-, " ,
.i
,·::on .., .... !" ..
Figure 15. Modified retort for Fischer Assay. Capacity is 100 g.
OIL SHALE Page 53
1. Modified Fischer Assay: The Standard Method for Oil Shale Assay
Using the new retort, modifications to the procedure have been made to
improve both accuracy and reproducibility. These include the use of retorts
with standardized heat capacity, standardized heating programs, carefully
controlled condenser and cooling bath temperatures, and continuation of the
assay at a maximum temperature of 500°C (Figure 16) until no more oil is pro
duced. In its present form, the Modified Fischer Assay (ASTM 03904-80) is a
widely used and generally accepted standard method for evaluating oil shales
(Figure 17).81.82
500
400
200
100
o 10 20 30 40 50 60 ELAPSED RETORTING TIME, MINUTES
figure 16. Heating profile during Fischer Assay.
OIL SHALE Page 54
Figure 17. Standardized Fischer Assay apparatus (ASTM 03904.80).
V'== - VfNI 10 (X,jAU51
OV[N
100 ",I C£NIRlfllGE 1lJ6<
COOUNG BATH
OIL SHALE Page 55
The shale to be assayed is crushed to provide a on-pound sample of -8 mesh
material. An aliquot is removed for drying to determine total moisture. The
retort is charged with 100 9 of the crushed oil shale in layers. the layers
being separated by the perforated aluminum discs. The retort is then heated to
SOO°C over 40 minutes, then held at that temperature for 20 minutes, or until
no more oil is collected (10-20 min. longer for shales richer than 20 gal/ton).
The heavier liquid products and some water are condensed in the centrifuge tube
which is maintained at 100°C. Lighter products and the balance of the water
are liquefied in the condenser which is maintained at O°C by a circulating
coolant.
When the heating period is complete) the retort and centrifuge tube are
cooled to room temperature and weighed. The tube is centrifuged to separate
oil and water. The water is determined volumetrically and subtracted from the
liquid product weight to obtain the oil yield. The specific gravity of the oil
is determined and used to convert the weight of oil produced to the oil yield
in gal/ton.
2. Shortcomings of the Fischer Assay
Despite acceptance as a standard method, the Fischer Assay has technical
shortcomings. Gases are not determined directly, but by difference as 'gas and
loss'. Some liquid hydrocarbons are also lost as mist. Thus) the yields of
these important parts of the product slate can only be inferred. To provide a
OIL SHALE Page 56
more com~lete assay, the Toseo Material Balance Assay was introduced in 1974 by
GoodfellOw and Atwood (Figure 18).83
MERCURY SWITCH MANOMETER --\\.
RETORT
THERMOCOUPLE
ICE WATER
CONDENSER
SOLENOID VALVE __ ·! _____ .J
ICE WATER
GAS 80MB
CENTRifUGE TUBE
Figure 18. TOSCO Material Balance Assay apparatus. (Reference 83. Reprinted
with the permission of the Colorado School of Mines.)
OIL SHALE Page 57
An extensive survey comparing the results of Modified Fischer Assay and
several closely related alternatives for Green River oil shales was recently
reported ,by workers at lawrence Livermore National laboratory.B4 Good inter
laboratory precision was found, but these workers strongly recommend the use of
one of the material-balance methods for maximum accuracy. These workers also
conclude that lean shales may give product distributions that are significantly
different from richer samples from the same area.
B. ALTERNATIVE METHODS FOR EVALUATING OIL YIELD
Oil yields depend not only upon retorting temperature, but upon heating
rate. Higher oil yields are generally obtained by rapid heating to an optimum
temperature. The Fischer Assay is not useful for evaluating oil yield under
such conditions. This has led to the development of alternative thermal assay
methods (of which the Rock Eval methodSS is notable), which have recently been
reviewed by Williams. 86 The more important of these, and some non-pyrolytic
alternatives to the Fischer Assay, will be discussed below in the context of
the various instrumental analysis methods.
C. ELEMENTAL COMPOSITION AND OIL YIELD
Most oil shale stUdies have been limited in scope. One of the handful of
studies that both encompassed a wide variety of shales and a wide variety of
techniques was that reported by Robinson and Dinneen at the 7th World Petroleum
Congress. 87 This study provides a starting point for the discussion of how oil
shale properties are determined and the relationships among these properties.
OIL SHALE Page 58
Geological perspective on the twelve oil shales to be discussed Geologically
youngest of the shale samples is the Tertiary Pliocene Sao Paulo shale of
Brazil (1-13 million years old), while the oldest is the Canadian shale of the
Carbonife,rous Period (280-350 million years). Figure 19 shows that the twelve
shale samples represent a wide range of geologic ages and geographic
locations.
GEOLOGICAL PERSPECTtV[ ON Oil SHALES TO BE DISCUSSED
''"' iU1if1i, I~ TEIIT [ARY Jl!ik"" ,,~~" ~'" CRHACOUS ~'" u
0
I """"" .~,,~ ~,~
" • TRIASSIC 1-"" I '"'''''' ,,;;';ii' 1-'~
CARBONIfEROUS
~ 1-'" , S. AFRICA. SPAIN
" . UM"
~ SilURIAN
CO"" • Ii OROO~ICIAII
f-'" CAI1IlIIIAN ~'"
I I'RECAKlR1AH ~""
Figure 19. The twelve oil shales to be discussed represent a wide range of
geologic ages and come from locations around the world. The
table ranks these shales in terms of geologic age. (After Robinson
And Dinneen, Reference 87.)
The twelve samples are also representative of a wide range of depositional
environments, kerogen types and organic facies (Table 10).
OIL SHALE Page 59
Table 10. Geological perspective on the twelve oil shales: Location, depositional environment, geologic period, kerogen type and organic facies type. (After Robinson and Dinneen. Reference 87.)
Kerogen Organic Name Location Geologic Period Environment Type Facies Type
Brazil Sao Paulo Tertiary (Pliocene) Lacustrine I I
Oregon, USA Shale City Tertiary (Oligocene) Marine I I
Colorado, USA Piceance Creek Tertiary (Eocene) lacustrine I I
New Zealand Orepuki Tertiary Assoc. Lignite II I I I
Alaska, Tasmanite Kil igwa River Jurassic Marine I I I
Argentina San Juan Triassic-Permian Marine I I I I
France SL Hilaire Permian Assoc. Coal II III
Australia Cool away Mt. Permian-Carboniferous Assoc. Coal (Probably II, probably III)
Spain Puerto 11 ano Permian-Carboniferous Assoc. Coal (Probably I I , probably III)
South Africa Ermelo Permian-Carboniferous Assoc. Coal I I III
Scotland Dunnet Lower Carboniferous Lagoon-Marine I I II
Canada New Glasgow Carboniferous Assoc. Coal II III
OIL SHALE Page 60
'1"1._ 1""_1_. __ -' _ ___ ......' "-
sample are lacustrine shales that are representative of organic facies Type I
and contain Type I kerogen. The Alaskan, Argentine, Oregon and Scottish shales
are of marine origin, representing organic facies Type II with Type II kerogen.
The remaining samples represent the coal-associated organic facies type III.
However, as we shall see, not all contain primarily Type III kerogen.
At this point, the casual reader may be wondering about the introduction
of two apparently redundant concepts, that of kerogen types I, II and III and
organic facies types I, II and Ill. It is tempting to try to simplify matters
by eliminating one. The South African sample provides a compelling illustration
of the need for both concepts.
Microscopic examination revealed that the organic material of the Alaskan,
Australial1 and South African shales contains mostly algal remains. "yellow
bodies" which are generally accepted as being the remains of algal colonies.
Thus, these shales contain Type II kerogen and are classified as torbanites.
The Alaskan and Australian shales are apparently the result of algal blooms in
the shallow waters of a marine platform or shelf. The Alaskan and Australian
shales show little evidence of woody plant remains (terrestrial input). Thus,
these fit cleanly into organic facies type II, as well as having Type II
kerogen. However, the South African shale deposit is coal-associated and has
appreciable terrestrial input. Presumably. this shale was formed from the
remains of the algal blooms in a stagnant lagoon or stream near a coal-forming
swamp. Thus, the South African sample is an example of a shale that has mostly
Type II kerogen, yet fits into organic facies type III.
OIL SHAL£ Page 61
Organic ~arbon Content and Oil Yield Table 11 summarizes the elemental
compositions and oil yields of the twelve oil shales and data for their
corresponding kerogen concentrates. The organic carbon (TOC) contents vary
from 7.9% for the Canadian shale to 81.4% for the Australian sample. (Strictly
speaking, the Australian sample does not contain 33% mineral matter, hence
should not be termed an oil shale. We will ignore this distinction, for the
purpose of the present ill ustrat i on. ) The oil shales conta i n hydrogen. not
only in organics, but also as water and in the minerals (e.g. as 5i1aoo1s).
Hence, the hydrogen contents of the oil shales are not reported. The kerogen
concentrates were freed of their hydrogen- containing minerals by HC1-HF
treatment and carefully dried. Consequently. the hydrogen contents of the
kerogen concentrates reflect only the organic hydrogen present in the starting
oil shales. (See Appendix A for a description of the calculations used to
extract the reported values from the results of elemental analysis.) Not
surprisingly, the assay oil yields generally increase with the organic carbon
content of the shale. However, the situation is not that simple! Note, for
example that the Colorado and Brazilian shales have nearly identical TOC
contents, but the oil yield from assay of the former shale is much higher
(27.7 gal/ton vs. 17.8 gal/ton). Evidently. factors other than TOC are also
important in determining the quality of an oil shale.
OIL SHALE Page 62
Table 11. Elemental compositions and oil yields of the twelve oil shales. (Data from Robinson and Dinneen t Reference 87.)
Geological Age
Raw oil shale
Organic C, wt% TOC
Total N, wt%
Total S, wt%
Ash, wt%
Mi nera 1 CO2 , wt%
Assay oil, gal/ton
Kerogen concentrate
Organic C, wt%
Organic H, wt%
Ash, wt%
5
Alaska
53.86
0.30
1.30
34.1
0.1
139.0
83.28
11 .48
1.5
6
Argentina
8.88
0.26
0.48
82.6
4.5
14.2
51.66
6.60
34.9
8
Australia
81.44
0.83
0.49
4.4
0.1
200.0
83.58
10.69
1.7
I
Brazil
12.79
0.41
0.84
75.0
0.6
17.8
64.88
8.52
12.5
12
Canada
7.92
0.54
0.70
84.0
2.4
9.4
54.78
5.57
33.4
4
Colorado
12.43
0.41
0.63
65.7
18.9
27.7
66.38
8.76
12.9
OIL SHALE Page 63
Table 11. (Continued) Elemental compositions and oil yields of the twelve oil (Data from Robinson and Dinneen, Reference 87.)
Geological Age
Raw oil shale
Organic C, wt% TOC
Total N, wt%
Total S, wt%
Ash, wt%
Mi nera 1 CO2 , wt%
Assay oil, gal/ton
Kerogen concentrate
Organic C, wt%
Organic H, wt%
Ash, wt%
7
France
22.25
0.54
2.32
66.3
8.4
25.0
70.47
6.45
17.7
2
New Zealand
45.69
0.78
4.79
32.7
0.1
66.2
63.89
7.48
9.9
3
Oregon
25.83
0.51
2.20
48.3
0.2
48.0
47.32
6.00
36.2
11
Scotland
12.33
0.46
0.73
77.8
3.2
22.2
67.61
7.62
18.6
shales.
9
South Africa
52.24
0.84
0.74
33.6
0.9
99.8
81.14
9.12
1.3
10
Spain
26.01
0.55
1.68
62.8
2.3
46.9
61.08
7.04
26.0
OIL SHALE Page 64
o. BITUMEN CONTENI AND COMPOSITION
Bitumen contents of the shales were det~rmined by exhaustive extraction
with benzene. 99 It is interesting to note that there is no simple correlation
between the organic carbon contents and the amounts of extractable bitumen
(Table 12). The Australian shale with the highest organic carbon content gave
one of the lowest bitumen yields, while the Colorado shale with a relatively
low organic carbon content gave one of the highest. likewise, no simple
correlation exists between age and bitumen content. The relatively young
Tertiary shales of ColoradO, Oregon and New Zealand gave higher bitumen yields
per per weight of carbon (9 - 15.5%) than did the older shales (3.4 - 6.2%).
However, the South African shale which is also of Tertiary origin, gave the
lowest bitumen yield, both in absolute terms and on a per weight of carbon
basis (0.3% and 0.7%, respectively). There does appear to be a relationship
between carbonate mineral content and bitumen yield for this limited set of
samples, but such a relationship would not be apparent were the data for a much
larger sample set examined. There are, however, some interesting relationships
between bitumen composition and geologic age and these will be discussed in an
upcoming section.
OIL SHALE Page 65
Table 12. Bitumen contents, extraction ratios, organic carbon contents and
contents of carbonate minerals for the oil shales. 8
Extraction Mineral
Bitumen. Ratio,b TOC, carbonates, wt% wt% wt%
Oregon 4.0 15.5 25.83 0.2
New Zealand 4.3 9.3 45.69 0.1
Colorado 2.7 9.0 12.43 18.9
France 1.4 6.2 22.25 8.4
Scotland 0.7 5.4 12.33 3.2
Canada 0.4 5.3 7.92 2.4
Argentina 0.4 5.1 8.88 4.5
Spain 0.9 3.4 26.01 2.3
Brazil 0.4 3.3 12.79 0.6
Alaska 0.5 0.9 53.86 0.1
Australia 0.5 0.7 81.44 0.1
South Africa 0.3 0.7 52.24 0.9
(a) Data from Robinson and Dinneen, Reference 87.
(b) Extraction ratio = 100 x % bitumen / % organic C.
The benzene extracts were fractionated into asphaltenes (pentane-insolubles)
and pentane-solubles at O·C. The pentane-solubles were further fractionated by
chromatography to obtain n-alkanes, branched- and crelo-alkanes, aromatics,
polars and resins (Table 13).
OIL SHALE Page 66
Table 13. Distribution of components in the bitumen extracted from the oil sha1,
Branched + cyclic Tota 1
n-Alkanes, alkanes, Aromatics, Polars, Resins, i n~, hydrocarbons, wt% wt% wt% wt% wt% wt%
Canada 14.9 31.7 0.5 0.6 40.7 47.1
Scotland 11.4 32.6 2.1 0.1 44.0 46.1
Alaska 4.7 34.5 5.3 0.5 35.7 44.5
Spain 6.9 35.1 2.2 0.1 54.2 44.2
Australia 10.6 28.4 0.9 0.1 40.9 39.9
South Africa 16.5 20.0 0.9 0.1 49.4 37.4
Argentina 0.4 29.8 3.6 0.3 61.5 33.8
France 0.2 23.9 3.5 0.1 56.8 27.6
Colorado 1.1 13.7 0.9 0.1 66.2 15.7
Brazil 1.1 9.6 1.3 0.1 37.4 12.0
Oregon 0.2 7.0 0.2 0.2 43.2 7.4
New Zealand 1.1 1.4 0.2 0.1 9.5 2.7
<a) Data from Robinson and Dinneen, Reference 87.
OIL SH~LE Page 67
It is interesting to note that the bitumen from the Canadian shale (one of
the oldest) contains 47% hydrocarbons and 53% non-hydrocarbons, while the New
Zealand shale (one of the youngest) contains only 3% hydrocarbons the other 97%
being non-hydrocarbons. Evidently, oil shale bitumen, like petroleum, loses
heteroatoms as it matures. What is not clear is whether heteroatom loss to
produce hydrocarbons, or reaction of heteroatom species to produce insolubles,
is responsible for this phenomenon. The observation that the older bitumens
contain less of the heteroatom-rich pentane-insolubles suggests that the latter
may be more important.
The n-alkane fractions were analyzed by GC and the results were used to
calculate the £arbon Rreference indices (CPI)* for the C'4 - C'9 and C25 - C29
ranges (Table 14).89
* Several ways of expressing the odd preference have been proposed. The
original definition by Bray and Evans used the C24 - C34 interval: 89
CPI = 0.5 C25 + C27 + ... + C33 (25 + C27 + ,., + (33
----------------------- + ---------------------
Analogous expressions, that average the preferences in overlapping even
carbon ranges, were used to calculate (PI's in the ranges discussed above.
OIL SHALE Page 68
Table 14. Bitumen CPI values for the C14 - C'9 and C25 - (29 n-alkanes. s7
CPI Values C'4 - C'9 C" - C29
New Zealand 1.04 4.76
Brazil 0.50 4.00
Colorado 1.18 3.17
Australia 0.60 2.03
South Africa 1.00 1.44
Oregon 0.72 1.33
Alaska 1.03 1.28
Scotland 1.04 1.21
Spain 1.03 1.18
Canada 1.04 1.16
France 0.41 0.99
Argentina 1.08 0.96
In the higher MW range, alkanes with an odd preference often indicate
terrestrial input; higher plants produce even-numbered carboxylic acid groups
which give odd alkanes upon decarboxylation. No such preference is observed in
material of marine origin. 91 Because of the abundance of cuticular waxes in
higher plants, the terrestrial contribution usually determines CPI in a mixture
of marine and terrestrial material. Generally, (PI in the higher MW range
decreases with age, as cracking produces lower MW species. Concomitantly, any
initial preference in the lower range is generally overcome by dilution. This
pattern is generally evident in Table 14.
OIL SHALE Page 69
Carboxylic (fatty) acids were isolated from the shales by washing with
mineral acid to remove carbonates, then heating with aqueous KOH for 18 hours
under reflux to saponify esters. The acids were methylated and adducted with
urea to separate the esters of straight-chain acids from those of cyclic and
branched-chain acids. The straight-chain esters were then analyzed by Ge. A
close relationship was observed between the n-alkanes and the carboxylic acids
having one more carbon atom.
Five acyclic isoprenoid compounds expected in the oil shale bitumens were
2,6,10,14-tetramethylhexadecane (C,o, phytane); 2,6,10,14- tetramethylpentadecane
(C'9' pristane); 2,6,IO-trimethylpentadecane (C,.); 2,6,IO-trimethyltridecane
(C'6); and 2,6,IO-trimethyldodecane (C,,). The iso- and eyelo-alkane fraction
chromatograms were obtained using identical injection amounts and conditions
(Figure 20).
The Brazilian and New Zealand bitumens contained only small amounts of the
expected isoprenoids, while the Oregon bitumen contained the most. Phytane:
pristane ratios varied widely. In general the younger bitumens contained more
phytane than pristane. The compounds eluting in the range of 250" - 32S"C
include the steranes and hopanes. Gammacerane, a pentacyclic triterpane with
five six-membered rings (not a hopane), was isolated from this sample of
Colorado bitumen. 92
OIL SHALE Page 70
Figure 20. Isoprenoid compounds in the oil shale bitumens. Chromatograms are
arranged in descending order of phytane:pristane ratios. (From
Robinson and Dinneen, Reference 87. Reprinted with the permission
of Elsevier Publishing Company.)
II.", ) II
O~[(;ON I -l!~
£OtO~'OO
M. . . '~~(HII~A
"-HEW It'l'..o
~ • · ./ l " 1i ru~c[
• 5'Ofl'~D
hA I «iSt:<
.A IA il "H'O~ .I ,M. sP".
Go " "'1 J, , '" ,Ny, SO\II~ .lf~K'
eUltl
Jv.. ""SlUU~ I
" ~ ,~ m '" ..
OIL SHALE Page 71
III. PHYSICAL/CHEMICAL PROPERTIES
The correlations of the oil yield and heating value of oil shales with
their chemical and physical properties are based upon many different kinds of
measurements. These range from simple, qualitative tests that can be performed
in the field to highly sophisticated quantitative measurements using the latest
in laboratory instruments. This section describes some of the more useful
correlations and the measurement methods on which they are based.
A. ATOMIC H/C AND N/C RATIOS
The relationships of the organic hydrogen and nitrogen contents, and assay
oil yields, to organic carbon contents is shown in Table 15. StOichiometry
suggests that oil shales with higher organic Hie ratios can yield more oil per
weight of carbon than those that are hydrogen-poor. Figure 21 shows that this
is generally the case. However, H/e is not the only important factor. The
South African shale with an H/C ratio of 1.35 has an oil yield/TOC of 0.72,
while the Brazilian shale with an H/C ratio of 1.57 gives an oil yield/TOC of
only 0.52. In general, the shales whose kerogen is converted efficiently to
oil, contain relatively low levels of nitrogen. However, the Brazilian and
Colorado shales have very similar nitrogen contents, yet give very different
oil yields.
OIL SHALE Page 72
Table IS. Ratios of hydrogen, assay oil yield and nitrogen to organic carbon.
(Data of Robinson and Dinneen, Reference 87.)
Kerogen Hie Nle x 10' Concentrate (atomic) Oil Shale Oil/TOe Oil Shale (atomic)
Alaska 1.65 Alaska 0.95 Alaska 0.48
Colorado 1.58 Austral ia 0.92 Austral ia 0.87
Brazil 1.57 Colorado 0.85 South Africa 1.38
Australia 1.53 South Africa 0.72 New Zealand 1.46
Argentina 1.53 Oregon 0.70 Oregon 1.69
Oregon 1.52 Spain 0.68 Spain 1.81
New Zealand 1.41 Scotland 0.67 France 2.09
Spain 1.38 Argentina 0.61 Argentina 2.51
South Africa 1.35 New Zealand 0.54 Brazil 2.74
Scotland 1.35 Brazil 0.52 Colorado 2.82
Canada 1.22 Canada 0.44 Scotland 3.19
France 1.10 France 0.43 Canada 5.85
OIL SHALE Page 73
Figure 21. Conversion of oil shale organics as a function of atomic Hie ratio.
(Data from Robinson and Dinneen, Reference 87.)
100
• • ~ • . z 80 0 VJ
.1 a:: w • > z 60 0 0 1 0 z • <: 0 40 a:: 0
20 1.00 1.20 lAO 1.60 1.80
ATOMIC Hlc RATIO
OIL SHALE Page 74
B. ALIPHATIC AROMATIC AND CARBON CONTENTS
A high atomic Hie ratio, as in the Alaskan and Colorado shales indicates
little unsaturation or aromatic. coal-like structure. Conversely, a low atomic
Hie ratio generally indicates high content of material with coal-like aromatic
structure. Infrared spectra of the kerogen concentrates qualitatively confirm
this conclusion, as shown in Figure 22.87
C. NHR DATA
The advent of high-resolution, solid-state NMR techniques has provided the
geochemist a powerful tool for probing the structures of the organic material
in solid oil shale kerogens and coals. Miknis and co-workers93 Maciel and
Dennis,94 Resing and co-workers95 and Hagaman and co-workers96 have used
13C NMR with £ross-Qolarization and magic-gngle 1pinning, the so-called CPMAS
technique, to determine the fractions of aliphatic and aromatic carbon in a
variety of oil shales and the corresponding kerogen concentrates. Typical oil
shale spectra are shown in Figure 23.
About this same time, the sensitivity of quantitative CPMAS NMR results to
the cross-polarization contact time was pOinted out by Vucelic, Juranic and
Vitorovic. 97 The importance of this contribution can hardly be overemphasized.
Thus, to obtain the highest accuracy it is necessary to optimize contact time
for each sample. However, experience has shown that accuracy sufficient for
most purposes, especially for comparisons of closely-related samples, can be
obtained using a contact time of 1-2 milliseconds.
Paramagnetic and ferromagnetic species can also affect the quantitative
accuracy of NMR spectra obtained using the CPMAS technique. The problem of
quantitative reliability in solid-state "c NMR spectra of oil shales was first
OIL SHALE Page 75
Figure 22. Infrared spectra (in KBr) of the kerogen concentrates. Intensity
of the band at 1600 em-' increases with aromatic content. (After
Robinson and Dinneen, Reference 87. Reprinted with the permission
of Elsevier Publishing Company.)
100 80 80 40 20 80 60 40 20
" 0 • 80 e ~ 80 .; 40 0 80 0
60 i 40
• 20 0
0 e
ArganU,.,.
Australia
Brazl!
~ 60 40 ro 0
Canada
80 60 40 ro 0
Colorado
40 Frequency, em·'
• Significant mineral bands appear below 1200 cm-t. For organic structural comparisons use absorptions In regions of: 1600 em,l (aromatic), 1700 em" (carbonyl), and 3450 cm,1 (hydroxyl)
-Low aromatic: Alaska, Colorado, Australia. -Significant aromatic structure: New Zealand, Oregon, Brazil-
tertiary shales. -Carbonyl strongest in Colorado, Brazil, Argentina and France. -Most hydroxy! groups: Brazil, Oregon, Colorado, New Zealand-
younger tertiary shales.
OIL SHALE Page 76
Figure 23. Solid-state "C NMR spectra of oil shales obtained under CPMAS
conditions. (Miknis, et al., Reference 93. Reprinted with the
permission of Pergamon Press.) Note that the samples used in
this work are not identical to the twelve shales discussed in
the previous section, though several are very similar.
a} Australia, Coolaway Mt. 200 gallton Permlan·Carbonlferous Paludal
h) Alaska, Chandalar oeposlt~ 136 gallion Devonian Marine
c) South Afrha, Ermels 68.8 gallton Permlan-Carbonlferous Paludal
d) Color,do, Green River Formation 59.2 gallion Tertiary-Eocene Locustrlne
e) Bmz:ll, Irati Deposit ~ 31.3 gallton Tartlary·Pllocene Locu.tlne
1---530 ppm--t Higher Shielding
~f) South Afrlca, Ermelo
33.3 gallion Permian Carboniferous Paludal
~g) France, St. HUalre
25.0 gallton Permian Paludal
~ h) Scotland, Westwood
22.2 gallion Lower Carboniferous Paludal
!) Auatralla, Glenn Davis Deposit 21.2 gsilion Permlan·Carbonlferous Paludal
~ D Nova Scolia, Slel!arton
• 13.2 gallton Carboniferous Paludal
~530ppm--f Higher Shielding
OIL SHALE Page 77
pointed out by Maciel and Dennis, who found significant differences between the
observed aromaticities of some oil shales and their kerogen concentrates. 94
Paramagnetics were suggested as the cause, but the available data did not allow
confirmation of this hypothesis. Subsequently. these effects have been studied
extensively in the context of coals and products obtained by heating coals. 98 - 99
Treatment with samarium iodide (Sml z ), a mild reducing agent, has been used to
lower the free radical content of Wyodak coal by 75% (I spin/8750 C's down to
1 spin/35,OOO C'S}.100 Nuclear spin-counting experiments revealed that only 58%
of the carbons in the starting Wyodak coal were observed by "c CPMAS NMR. In
contrast, 85% of the carbons were observed after SmI 2 treatment. Application
of this new technique to oil shales may be difficult, due to mass transport
limitations (oil shales have low porosity and permeability), but should be
applicable to kerogen concentrates. Thus, while the usefulness of 13C CPMAS
NMR for oil shale characterization is well-established, the coal experience
suggests that such studies are best carried out in conjunction with character
izations of the paramagnetic and ferromagnetic species present.
From TOC content and the fraction of aliphatic carbon determined by NMR,
the weight% aliphatic carbon in the oil shale or kerogen concentrate can be
calculated. A good correlation was obtained between oil yield and aliphatic
carbon content (Figure 24).101,102 This technique has also been applied to the
evaluation of petroleum source rocks.t03
OIL SHALE Page 78
Figure 24. Aliphatic carbon content vs. oil yields for 22 oil shales and
kerogen concentrates. (After Maciel. et al., References lOl,l02.)
70 ~
"" 0 -3: 60
~
t: 50 0
.0 -IV 0 40 (J ::: IV 30 .<: Q.
« 20 -t: '" -IV 10 Q. Q. «
0 0
I
0
• •
o Kerogen Concentrate • Raw Oil Shale
1 00 200 300 400 500 600 700 800 Oil Yield, (lit)
I I I
50 100 150 200 Oil Yield (U.S. gal/short ton)
OIL SHALE Page 79
D. NEWER NMR TECHNIQUES FOR KEROGEN CHARACTERIZATION
Three other solid-state NMR techniques warrant mention, as they show great
promise for further unraveling the structural features and thermal reactivities
of the kerogens in oil shales and coals.
The tldipolar dephasing" technique introduced by Opella and Frye makes use
of strong dipolar coupling between 13C nuclei and directly bonded 1H nuclei to
selectively relax the 13C NMR signals of the protonated carbons in relatively
immobile rings and chains, but not in methyl groups which freely rotate,104
Thus, allowing dephasing of the 13C polarization for a short time (30-60 ~s),
then refocusing and acquiring the residual signal, enables the observation of a
MAS 13C NMR spectrum due almost entirely to non-protonated and methyl carbons.
However, it has recently been painted out that signals due to mobile methylene
groups can be also observed in dipolar dephasing experiments. 10S Thus, dipolar
dephasing results should be interpreted with some care, and with considerable
structural insight. Nevertheless, the "gipolar-gephasing magic-gngle ~pinning"
(DOMAS) technique is a very powerful tool, especially for probing the aliphatic
components of oil shales, kerogens and coals.
The "yariable-~ngle ~ample ~pinning" (VASS) technique recently introduced
by Sethi, Grant and Pugmire affords detailed structural information through
evaluation of the tensor components of chemical shift anisotropy.106 In the
CPMAS technique discussed above, the sample is spun rapidly (about 3000 Hz, or
-180,000 rpm!) at the "magic angle" of 58"44' to average the components of the
chemical shift tensor and obtain an isotropic chemical shift similar to that
observed in liquid phase spectra. In dOing this, the structural information
contained in the individual chemical shift tensor components is lost. The VASS
OIL SHALE Page 80
technique enables recovery of this information by spinning the sample at angles
other that the "magic angle". The recovered information can then be used to
evaluate the contributions of different types of aromatic carbons: protonated,
substituted and inner. Thus, DOMAS and VASS are complementary techniques.
providing information about the structural details of the aliphatic and the
aromatic components of the kerogen, respectively. Used in conjunction with the
ePMAS technique, DOMAS and VASS make solid-state "e NMR one of the most useful
tools available for characterizing the organic material in solid oil shales.
The third of the newer NMR methods is the dynamic, in-situ 1H NMR technique
pioneered by Lynch and his co-workers at CSIRO.107 The power of the method lies
in the fact that the measurements can be carried out while the sample is being
heated. In contrast, the currently available solid-state "e NMR equipment is
generally restricted to operations near, or below, room temperature. Samples
can be heated, then observed, but not observed during heating. The 1H Signals
observed by Lynch are sensitive to the molecular dynamics of the hydrogen
containing material. Using this technique it is possible to estimate the
fractions of molecular structure that are "rigid" and "mobile" on a time scale
of about 10- 5 second. Oil shale and coal samples typically exhibit dynamic 'H
NMR spectra that are composites containing both rigid and mobile components.
Presumably, crystalline or glassy material contributes to the former, while the
latter, mobile component is due to rubbery, amorphous material. As the sample
is heated, an initial increase in the mobile fraction (ascribed to conversion
of rigid material into mobile) is typically observed, followed by a decrease as
volatile hydrocarbons are evolved. After the initial decrease in the rigid
component, an increase indicates the formation of coke precursors and/or coke.
OIL SHALE Page 81
The kinetics of both changes can be followed by monitoring the dynamic lH
spectrum as a function of temperature and time. This technique has been used
to probe the pyrolysis (i.e. retorting) behavior of a wide variety of oil
shales and coals.
E. ESR RESULTS
flectron ~pin Resonance (ESR, also known as ~lectron garamagnetic resonance
or EPR) methods have been used extensively to monitor the changes in free
radicals that occur when coals are heated. ESR has also been used to study the
process of kerogen maturation, both natural and simulated by heating in the
laboratory -- mostly in the context of petroleum exploration, but also in the
context of oil shale retorting. These studies were surveyed in a recent
article by Silbernagel and co-workers, who also reported the changes in free
radical content and character that occurred upon heating oil shales from the
Green River and Rundle formations, and their kerogen concentrates, for varying
times at relatively low temperatures of 350·C - 375·C.108 Free radical concen
trations generally increased with heating but these increases were shale
dependent, being much larger for the Rundle samples. Changes in 9 values and
line widths were also observed and discussed.
Sousa and co-workers used ESR in conjunction with lhermal slteration index
(TAl) determinations to study the natural maturation of the Irati oil shale of
Brazil. '09 Qualitative agreement between TAl and ESR results were obtained for
stratigraphie column CERI-J for which previous workers had hypothesized an
unusual exposure to paleotemperature at one particular point in the column.
OIL SHALE Page 82
However, the ESR results proved much more sensitive to this local heating than
TAl.
To summarize, the potential of ESR for oil shale studies is clearly great;
the technique is very sensitive to heating. Qualitative ESR methods have been
useful. However, quantitative ESR studies have generally raised more questions
than they have answered. The reason for this may be that the free electron
spins observed by ESR are not directly involved in the chemistry of interest,
but merely reflect exposure to the environment (heat, pressure, structure. etc)
in which this chemistry takes place. If this is the case, then ESR can at best
be expected to provide correlations not a direct probe of the chemistry.
F. DENSITY METHODS
As we shall see in a later section, the difference in density between the
organic and mineral components of an oil shale is important as the basis for
dense medium beneficiation. However, density is also important as a character
ization tool. The relationship between density and oil yield was painted out
by Frost and Stanfield who reported measurements on 32 samples of Green River
oil shale spanning the ranges of 1.67-2.54 g/cm' in density and 10-77 gal/ton
in oil yield. 110 later, Smith derived Equation 2 to describe this relationship
quantitatively:l11.112 A plot of oil yield vs. specific gravity of Green River
oil shale samples from the Anvil pOints area is shown in Figure 25.
OIL SHALE
A(D, - D.) + D.
Where: Dr = Density of the shale rock
A = Weight fraction of organic matter
B = Weight fraction of mineral matter
D. • Average density of the organic matter (gm/cc)
D, = Average density of the mineral matter (gm/cc)
Page 83
(2)
Vadovic has used density separations in another way to estimate the amounts
of hydrogen and nitrogen that are associated with the mineral matter in oil
shales. 113 Vadovic's approach begins with separation into several fractions of
increasing density. A pulverized oil shale sample is separated into the
fractions which sink and float in a medium of specific gravity 1.5. The sink
fraction is freed of the dense medium and dried for analysis, while the float
fraction is subjected to another sink-float separation in a more dense medium.
OIL SHALE Page 84
Figure 25. Specific gravity vs. oil yield for Colorado oil shales. The curve
is calculated from Equation 2. (From Smith, Reference Ill.
Reprinted with the permission of the American Chemical Society.)
OF COLORADO OIL SHALES 2 100r-~;=~~==~;=~;==;~==~~~='-, Ul c o -.
"0 ., :;:: o
.--"0 o :iE
90
80
70
60
SO
40
30
20
y = 31.S63X' - 20S.998X + 326.624 Y = Oil Yield (gallT) x = Specific Gravity of Gros at 60 of
DB = 2.7 glcc DA = 1.0S glcc
Specific Gravity, 60/60 0 F
OIL SHALE Page 85
In this way, fractions having specific gravities of <1.6, 1.6-1.7, 1.7-1.8,
1.8-1.9. 1.9-2.1. 2.1-.2. and >2.2 were obtained. The sink-float fractions
were then analyzed for carbon, hydrogen and nitrogen. For Green River oil
shale from the Colony Mine, a plot of the hydrogen results vs. organic carbon
carbon results for the different fractions gave a straight line (Figure 26).
The intercept of this line at 0% organic carbon (no organic matter) provides an
estimate of the atomic ratio of mineral hydrogen to organic carbon content.
From this ratio and wt% organic carbon, a mineral hydrogen content of 0.08 wt%
(3% of the total H) was estimated for the Colony oil shale. Similar treatments
yielded mineral hydrogen estimates of 0.25 wt% and 0.28 wt% (11% and 15% of the
total H) for Rundle (Australia) and Brazilian oil shales, respectively. This
approach is also applicable to other mineral elements; mineral nitrogen
contents of 0.14. 0.13 and 0.02 wt% (18%. 31% and 4% of total N) were estimated
for the Colony. Rundle and Brazilian Shales. respectively (Figure 27).
OIL SHALE Page 86
Figure 26. Hydrogen analysis for Green River oil shale (Colony Mine, Colorado).
(From Vadovic, Reference 113. Reprinted with the permission of the
American Chemical Society.)
N o u
9
8
7
+ 6 :I:
'" <{
:::: 5 ..J <{ I-o 4 I-:I:
:5 3
2
1
SHALE HIC HMIN
• COLONY 1.55 .10
O'~~~~~~~~--~---L---L--~--~--J o .5 1 1.5 2 2.5 3 3.5 4 4.5 5
100 CORGANIC/12 (ASH + C02)
OIL SHALE Page 87
Figure 27. Nitrogen analyses for Green River, Rundle and Brazilian oil shales.
2.5
2.25
2
'" 1.75
0 (J
+ 1.5 :x:
en <{ - 1.25 ...J <{ I-0 l- I z 0 0 ~ .75
.5
.25
(From Vadovic, Reference 113. Reprinted with the permission of the
American Chemical Society.)
SHALE N/C NMIN
• RUNDLE .017 .16
... BRAZIL .024 .02
• COLONY .Ozg .14
o 10 20 30 40 50 60 70 BO 90 100
1400 CORGANIC 112 x (ASH + C02)
OIL SHALE Page 88
G. THERMAL METHODS OF ANALYSES
Heat is the objective of most interest i'n oil shale and large amounts of
heat must be supplied during retorting to produce shale oil from the rock.
Moreover t when the shale rock is heated its weight decreases as volatile
species are evolved. Consequently, thermal methods of analysis have long been
important in oil shale characterization.
Heating values Not surprisingly, there is a good, linear correlation between
the heating value of dry oil shale and its Fischer Assay oil yield (Figure
28).80
It is also possible to estimate the heating value of a dry oil shale from
its elemental composition. Equations 2 and 3 borrowed from coal science serve
well for this purpose, giving similar results.114 In these equations, Q is the
heating value in BTU/lb., while C, H, 0 and S are contents of the respective
elements in wt%.
BOle Equation115
Q ~ 15,120 C + 49,977 H + 2,700 N + 4,500 S - 4.770 0 (2)
Dulong Equation
Q • 14,544 C + 62,028 (H . 0/8) + 4,050 S (3)
OIL SHALE Page 89
Figure 28. Heating values of dried Green River Oil Shale from the Mahogany
Ledge (Rifle, Colorado) vs. oil yield. (Data of Stanfield, et
aT., Reference BO.b.)
., 7,000 C\l .c
Samples Dried at 221°F <J)
3: 6,000 For One Hour ° C\l a: ° "C 5,000 c
::J 0
° Q. -::> 4,000 I-al .,;
3,000 /&0 ::J C\l ° > / C> 2,000
/0 c /<§! - /C0 C\l .,
::t: 1,000 ° I/) / I/) 0 -CI
00 10 20 30 40 50 60 70 Shale Grade, Gallons Oil/Ton of Shale
OIL SHALE Page 90
Retorting Kinetics using Thermal Methods The kerogen (and most of the bitumen)
in oil shales is not volatile. As oil shale is heated during retorting, these
non-volatile organics crack to give volatile products that are evolved from the
rock. The general area of thermophysical properties of oil shales was reviewed
in 1979 by Rajeshwar, Nottenburg and DuBow.'" An update by Rajeshwar appeared
in 1983. 117
Many workers have used thermogravimetric gnalysis (TGA). which measures
the change in weight as a sample is heated, to study the kinetics of oil shale
retorting. Hubbard and Robinson deduced first order kinetics in a detailed
study of the thermal decomposition kinetics of Green River oil shales. 118
These workers proposed a two-step decomposition pathway (Equation 4).
k, k, Kerogen --------> pyrobitumen --------> Oil + Gas + Char (4)
Further studies on Green River oil shale, using both isothermal and non
isothermal TGA techniques, were made by Nielsen and A11red'19 and Al1red. 12o
Allred plotted the results of Hubbard and Robinson and observed (a) that the
amount of carbonaceous residue reached its ultimate value when oil evolution
was only one-half complete and (b) the weight fraction of kerogen appearing as
oil + gas became constant at that point (i.e. kerogen decomposition was
complete), and (c) the combined amount of oil + gas that subsequently appeared
nearly equal to the amount of bitumen that disappeared (Figure 29).
OIL SHALE Page 91
Figure 29. The rates of Kerogen disappearance and product appearance during
heating of Green River oil shale at 900·C. (After Allred'z, with
data from Hubbard and Robinson. 118 )
IOOr-~~----------------------------------------,
•
80
z w 0 0 <>: w "'60 -' ~ 0 >-~
0 >-4{) z w <> <>:
'" Q
0-J: 20 2 w
"
I I : OIL AND GAS "
I _8_ - -"'---8
_ KEROGEN
I .,A'--I t:('/ I ./
/ J .,( I / I / I / 1/8
If , • 87'"
/ ' ~ ... -.-o .. ().~ BITUMEN . , ..
---
/. <1:): \.'" .,. CARBON RESIDUE ON SHALE
. ,_x~:,--x-x x-----~X
2"8 Y J ••.•.•
".r x..? I 0 .... 0
0 2
~ ···0 !--~~o.::...--:--_~,~--==. . ........... ,-,,()~~.....JJ
4 6 8 10 12 14
TIME. MIN.
OIL SHALE Page 92
Allred's analysis led to the conclusions that the thermal decomposition
involved three first-order steps, the two steps of Equation 2 plus vaporization
of shale oil (Equation 5), with the first two following an autocatalytic rate
law (Equation 6).
" Kerogen ----) Gas pyrobitumen Char
" -- --)
-=-/- 1 • -kt + 7n I
" (Oil + Gas)"q. ----> (Oil + Gas) •• p.
Weight of Oil + Gas Where X = -------------------
Weight of starting Kerogen
(5)
(6 )
Campbell and co-workers121 interpreted their non-isothermal TGA results on
Green River oil shale in terms of a single first order step, as did Herrell and
Arnold in their study of Chattanooga oil shale. 122 Rajeshwar, in a later non
isothermal TGA study, concluded that the data required two first-order
steps.123 First order kinetics were also found by Haddadin and Mizyed124 in a
TGA study of Jordanian oil shales, while Bekri and co-workers125 and Thakur and
Nuttal1 148 found first order kinetics and a two-step organic decomposition
mechanism in studies of Timhadit shales from Morocco. Williams found similar
kinetics in TGA decompositions of Kimmeridge oil shales from Great Britain and
good agreement between the TGA results and those obtained in artificial
maturation. 126 Interpretations of the results of the foregoing studies were
all based upon assumed mechanisms of kerogen decomposition. As pointed out by
Williams, "To ascribe kinetic results to specific chemical and physical
OIL SHALE Page 93
processes, e.g. specific bond-cracking reactions, a more rigorous analysis of
the data ... would be necessary."
The worth of Williams' observation was made clear when Braun and Burnham
reported a new analysis of the kinetics of product evolution from Green River
oil shale. 127 The isothermal retorting data had been earlier obtained in a
quartz fluidized bed apparatus with a flame-ionization detector (from a GC) to
detect evolved organics. 128 This apparatus does not respond to water (unlike
the TGA which records water as weight loss, along with oil + gas), but does
respond to hydrocarbon gases that do not contribute to oil yield. Therefore,
the FLO response was corrected for C1 - C3 hydrocarbon evolution, using rate
data available from Richardson. 128 Nine experiments (three shale samples, each
at three temperatures in the range of 380'C - 540'C), each involving 200-500
individual data points, were fitted, using a non-linear least-squares procedure,
to obtain the kinetic parameters: A, E and reaction order. It was discovered
that the results could be fitted equally well with either two parallel first
order rates or one three-halves order rate (Figures 30 and 31).
OIL SHALE Page 94
Figure 30. Comparison of calculated (----) and measured (-----) oil production
for one 1.51·order reaction (a) Anvil Points sample; (b) Tract Ca,
Sample RB-l; (c) Tract Ca, Sample RB-2. (From Braun and Burnham,
Reference 127. Reprinted with the permission of Butterworth & Co.
(Publishers) Ltd.)
02
0.8 g ";; ;S li o.s
.~ 04 . . "I , u
0.2
0.2
49<\"C
%~.0~O~""0~.~,~.2""~6~2~~"2~"~2~.8",3~.2-t3.'6~4~ Time (miO)
OIL SHALE Page 95
Figure 31. Comparison of calculated (----) and measured (--) oil production
for two parallel first-order reactions (a) Anvil Points sample;
(b) Tract Ca, Sample RB-l; (c) Tract Ca, Sample RB-2. (From Braun
and Burnham, Reference 127. Reprinted with the permission of
Butterworth & Co. (Publishers) Ltd.)
"
540·C
514·C
- 491"C
1.0 ~~~=~~~~-~ c ---
,.,
!fi.o 0.4 0.8 1.2 L6 2,0 2.4 2.8 3.2 3.6- 4.0 Time {min}
OIL SHALE Page 96
There is still some uncertainty regarding the maximum amount of oil that
can be evolved from Green River oil shale under the rapid heatup conditions.
However, there is good evidence that this is at least 110% the Fischer Assay
yield can be obtained for the particles smaller than about 3 mm used in these
studies. A comparison was made of the oil evolution rates predicted by three
models, those of Wallman and co-workers129 and Braun and Burnham127 obtained
under rapid heating conditions and that of Campbell and co-workers '21 obtained
with slow heating. In each case the ultimate oil yield was forced to agree
with experiment: 110% of Fischer Assay for fast heat up and 100% of Fischer
Assay for slow heating. The results show that while the three models agree
well during the first 50% of oil evolution, the last 50% of oil is evolved much
slower than predicted by the slow heatup model (Figure 32). Two possibilities
come to mind as kinetic sources for the "long tail" in oil evolution under
rapid heatup conditions: mass transport of oil from particle interiors bay be
limited or perhaps the rate of kerogen decomposition decreases as conversion
increases. It is not possible from these data to differentiate between these
possibilities.
Both pressure and atmospheric composition can affect the rates of product
evolution during pyrolysis. Suuberg concluded that very high heating rate
(>1000 C/sec.) pyrolysis did not enhance oil yield over Fischer assay for a
rich (50 gal/ton) Green River shale.'49 Little effect of pressure was observed
near or below 1 atm., but higher pressures significantly retarded oil evolution
and altered the product slate to favor lower MW products. Similar observations
were made under slow heating conditions by Burnham and Singleton, who also ob
served that the pressure effect was less pronounced at lower heating rates. 150
OIL SHALE Page 97
Figure 32. Comparison of calculated oil yield at 500°C (expressed as percent
>'" V> V>
'" '-
'" :..c.: u V>
u..
v
'" >-
o
of maximum recovery) as a function of time for three reaction rate
analyses: -- Campbell et a1.121; ---- Braun and Burnham127 ;
.... Wallman et a1.129 (From Braun and Burnham, Reference 127.
Reprinted with the permission of Butterworth & Co. (Publishers) Ltd.)
100 -
80 -
60 '-
401--
20
· · · · · ·
I I ~._._._I._ .. _._._ ............. ;.:.:..:.:.~ .-..
., ... ..... /' ", ,,/
/'/ ..... /
/'1
-
-
-
Time (min.)
OIL SHALE Page 98
It was pOinted out that the residence times of both liquid oil and oil vapor
increase with increasing pressure. Moreover, the low porosity of the oil shale
is likely to retard oil evolution, at least until some products depart genera
ting appreciable porosity. Thus, the lower oil yield and lighter product slate
are both ascribed to longer residence time of the liquid products, which allows
more extensive cracking and char formation.
TGA for evaluation of oil yield Because of the general acceptance of the
Fischer Assay, TGA has not been generally used for evaluation of oil yield.
Moreover, TGA methods do not readily provide some of the information obtained
from Fischer Assay; the water yield and the distribution of liquid and gaseous
products, for example. TGA is, however, well-suited for comparative studies in
the laboratory; it requires only small samples (1-25 mg) and automated
eqUipment is commercially available. Smith and co-workers,130.131 and
Rajeshwar and co-workers132, 133 have reported the utility of TGA for evaluating
Green River oil shales, while Earnest134 ,135 obtained good correlations between
TGA and oil yield for both Green River and Australian oil shales. Williams
found reasonable correlation between TGA weight loss and oil yield, and good
correlation between Rock Eval peak area and oil yield, for Kimmeridge oil
shales. 136 However, in a more extensive study of Kimmeridge shales, Williams
found that both the ratio of water:oil and the ratio of gas:oil in retorting
products varied with the richness of the shale.'2. These findings led Williams
to recommend caution in the interpretation of oil shale TGA results, especially
when comparing results for shales of widely differing grade.
OIL SHALE Page 99
H. HEAT CAPACITY AND HEAT OF RETORTING
Knowledge of enthalpy changes is important to both modeling and design of
oil shale processes. What is needed are predictive relationships that enable
the calculation of enthalpy changes upon arbitrary changes in the temperature
and composition of the oil shale. Some of the important calorimetric studies
to provide this information were recently summarized by Camp.137
Most of the reported stUdies, including Camp's, deal with Green River oil
shales. Several workers cited by Camp had measured the enthalpies required to
heat oil shale and most had studied shales of varying oil yield.138-142 The
data, especially those of Mraw and Keweshan,142 provided precise estimates of
the heat requirements in retorting. However, the most useful correlations of
heat capacity with temperature and shale grade, those of Carley, were purely
empirical and linear in temperature. 143 Camp criticized these linear relation
ships, noting the pronounced non-linearity at temperatures above 325°C of the
heat capacities of several major minerals in Green River oil shale.
As an alternative, Camp took the approach of summing the heat capacities
of the organics (char in the case of spent shale), volatile products evolved
from the oil shale, the individual minerals plus bound and free water. Using
the resulting non-linear model and published heat capacities144 for a set of
about 20 minerals, excellent fits were obtained to the reported heat capacities
of spent and burned shale over the temperature range of 25° - 600°C. Moreover,
the equation fit the reported heat capacities of raw oil shales of different
grades at temperatures up to 240·C, where decomposition of the Green River oil
shale kerogen is negligible. The heat capacity of the organic fraction was
modeled using published heat capacity data for graphite and sixty organic
OIL SHALE Page 100
compounds and petroleum fractions, for which data were available over a wide
range of temperature, to obtain non-linear Equations 7 and 8 for the heat
capacities of the kerogen and char, respectively. These equations fit the low
temperature data well, but predict lower, more reasonable heat capacities than
the linear equation at the higher temperatures encountered in retorting.
C"",," = 0.2232 + 5.254 x IO·'T - 1.6536 x IO·6T'
= -0.1179 + 4.308 x 10·'T - 1.786 x 10·6T'
Where: Heat capacity = C in kJ/'K kg
Temperature ; oK
(7)
(8)
The resulting model. incorporating non-linear predictions for heat capacities
of both organic and mineral fractions, was then used to fit the measured heat
capacities of several Green River oil shale samples varying in both oil yield
and water content (Figure 33).
Subtraction of the mineral heat capacity from the total heat requirement
enabled estimation of the heat required for kerogen decomposition to volatile
products and char. This heat was found to be substantially larger for shale
samples heated slowly to 350·C"8·'4' than for samples heated rapidly to 500·C
OIL SHALE Page 101
Figure 33. Measured and predicted heat requirements above 25 G C for Green River
oil sha 1 e samples. (From Camp, Reference 137. Repri nted with the
permission of the Colorado School of Mines.)
a. The sample assaying 127 L/Mg (127 liters/106 g) required more
heat than the 147 L/Mg sample, due to its higher content of
hydrated minerals. (Data of Wisej et a1. 139 )
-+ 127 UMg (30.4 gpt)
800 o 143 UMg (34.3 gpt) o 63lJM~ (15.0 gpl)
-;;; - Predlct ons -0 ~ 600
" '" 127 lIMg
" • -'5 400 .,. ~ 63 LfMg
;; • " 200
0 0 200 400 600
T emperatute (0C)
b. Data of Sohns, et 07. 140
.-'"""'--------+ 238 UMg 157 gpl)
800 o 117UMg 28gpt) - Predictions •
0
" -~ 600
~ !l .. 400 ~ m " 200
o 200 400 600
Temperature ("C)
OIL SHALE Page 102
and held for only 2-3 minutes for pyrolysis'" (370 kJjKg of kerogen vs. 275
kJjKg). Of this difference, 50 kJjKg could be accounted for by the larger
sensible heat of the kerogen, relative to that of the evolved products. The
remaining difference, 45 kJjKg of kerogen (12% of the total heat), clearly
reflects an endothermic reaction that occurred to a greater extent during the
long heating at 350·C. Camp suggests that either increased char formation
(coking), or a mineral decomposition reaction not accounted for in the model,
may be responsible.
Heat capacities of liquid products In the previous section, it was mentioned
that Camp modeled the heat capacities of kerogen and char using model compounds
and petroleum fractions. Because heat capacities correlate with many other
properties and are required for heat-balance calculations in process design, as
well as being useful in retorting studies, there continues to be a high level
of interest in this area. Recent articles by Rodgers, Creagh and Prausnitz'4S
and Mraw, et a1. 146 and a recent book by Tsonopoulos, et a1.'47 provide both
a wealth of data and leading references to the field of product
thermodynamics.
I. MECHANICAL PROPERTIES
Minerals playa major part in determining the strength and hardness that
are important in mining and processing (e.g. crushing, grinding) oil shales,
especially those of the more prevalent leaner grades. This is illustrated in
Figures 34 and 35, where both strength and hardness decrease markedly as the
organic content of the oil shale increases. 1s1
OIL SHALE Page 104
Figure 35. Young's Modulus vs. Fischer Assay (Sellers, Reference 151.)
C/)
a. 3 '" o ,... II) :::l -.s2 o :E II)
-0>
§ 1 o >-
Young's Modulus
o~ __ ~ ____ ~ __ ~~ __ ~~ __ ~ __ ~ o 10 20 30 40 50 60
Oil Content, gal.lton
OIL SHALE Page 105
The shear and compressive strengths of Green River oil shale samples from
the Anvil points mine and their implications for room-and-pillar mining were
reported by Agapito. 152 The strength and hardness of oil shales decrease upon
retorting and, not surprisingly, this decrease is greater for richer shales.
Dinneen found that lean oil shale cores « 15 gal/t) retained high compressive
strength (- 90% of initial value both parallel and perpendicular to the bedding
plane) after heating to 950°F.153 Richer samples lost progressively larger
fractions of their strength on heating; a 15 galft sample lost about 75% of
its initial strength, while a 60 galft sample lost> 99% to give a very soft,
almost chalk-like ash. Further heating to 1500°F to decompose the carbonate
minerals caused little further loss in strength. This work has been reviewed
by Baughman. 154
Acoustic wave propagation experiments reflect several phenomena, but do
provide a sensitive probe of the state of the oil shale matrix, which may prove
useful in non-contact monitoring of in-situ and above-ground retorting
processes. Mraz and co-workers have determined the acoustic wave propagation
characteristics of some relatively rich Green River oil samples. 155 Both cores
and rubble samples were studied. The results were interpreted in terms of a
complex set of thermal alterations, including the release of free and bound
water, pyrolysis and release of the organic matter, and re-cementation of the
shale matrix by pyrolysis products. The results were strongly dependent upon
the initial organic content, but only slightly upon temperature. Measurements
on rubble (broken fragments) showed an expected dependence upon the compaction
pressure, and revealed an unexpected anomaly that suggested the presence of
small voids in the heat treated samples.
OIL SHALE Page 106
J. POROSITY ANO PERMEABILITY
Porosity and permeability of shales are generally very low. Because of
this, shales often comprise the cap rock or seal in petroleum and natural gas
reservoirs. However, in retorting, permeability is critical to allow the
escape of products from the rock. Dinneen found that appreciable porosity was
developed by heating Green River oil shale. 153 Tisot and Sohns monitored the
permeability of columns of Green river oil shale particles as a function of
temperature, heating rate and compressive load/strain. 156 Thermoplastic flow
was observed with samples that contained >50 vol% organic matter (>48.5 gal/t).
The loss of strength observed in rich samples led these workers to conclude
that permeability losses under the influence of heating and overburden pressure
could significantly impair in situ retorting of rich zones (> 27 gal/t). In
contrast, a column containing fragments of 27-gal/t shale underwent significant
compression upon heating. but retained its high initial permeability. These
results show that permeability changes upon heating must be considered in
planning an in situ retort operation.
Porosity and permeability have been extensively studied in the context of
petroleum exploration and production. 11 ,13 A recent article by Hall, Mildner
and Borst discusses the methods available for measuring the porosity and perme
ability of very non-porous shaly rocks and describes a new technique that uses
small angle neutron scattering (SANS).157 This technique allows definition of
the size distribution for pores in the range of about 2 - 75 nm and is capable
of detecting anisotropy in pore orientation. Presently, this technique does
OIL SHALE Page 107
not provide information about the connectivity of the pores (topology of the
pore network). As a result, it is not clear that SANS, or the closely related
small angle X-ray scattering (SAXS) technique, is ready to displace the more
traditional methods for measuring permeability but further study of these
methods is clearly warranted.
OIL SHALE Page 108
IV. THE CHEMICAL STRUCTURE OF OIL SHALE KEROGENS
Typical oil shales are non-porous, impermeable rocks containing 80-95 wt%
minerals and only 5-20 wt% organics. Of these organics, only a minor part, the
bitumen, is extractable into organic solvents. By far the major part of the
organics in most oil shales is present as kerogen, an insoluble solid of
variable composition which is usually finely dispersed throughout the mineral
matrix. The difficulty of achieving quantitative and selective reactions of
insoluble organic solids under mild conditions and the scarcity of good methods
for probing the structures of such materials have discouraged attempts to
characterize the organic material in solid oil shales. As a result many of our
ideas regarding the structures of oil shale kerogens have been deri~ed from the
many reported studies of bitumens (oil shale extracts) and thermally generated
shale oils. Nevertheless, over the past twenty-five years several research
groups have begun to probe the chemical structures of oil shale kerogens. In
general the first step in these studies has been isolation of the corresponding
kerogen concentrate. A wide variety of chemical and physical techniques have
been used to characterize the resulting kerogens.
A. KEROGEN ISOLATION
Five excellent reviews of procedures for kerogen isolation are available;
none is very recent.158-162 However, as Durand and Nicase observe at the
outset of their review, the ideal method for isolating kerogen does not yet
exist and the isolated kerogen does not always coincide perfectly with the
concept of kerogen. 158 By far the most common technique for kerogen isolation
OIL SHALE Page 109
involves acid demineralization of the oil shale to produce the corresponding
kerogen concentrate. Recently, a rapid two-step base-acid treatment, and a
combined chemical-physical concentration method offering mild conditions and
high kerogen recovery, have been applied to Green River oil shale. Physical
separations (sink-float, oil agglomeration, froth flotation) have also been
used. All of these methods involve cominution as the initial step and it is
important to select samples that have not been unduly altered by weathering.
1. Sample Selection and Preparation
Because of their low permeability, oil shales do not weather as rapidly as
coal samples. Nevertheless, weathering can alter oil shale kerogens that have
been exposed to the atmosphere. Whenever possible, core samples or samples
taken more than 12 inches below an exposed surface should be used for kerogen
isolation to avoid weathering effects. Storage under nitrogen has not been
found necessary for oil shale samples that are larger than 1/2 inch, or so.
larger pieces of oil shale are crushed in a jaw crusher to pass 8-mesh,
then ground to pass IOO-mesh in a hammer mill or disc mill. Care should be
taken to keep temperatures low during fine grinding, and grinding under an
inert atmosphere (e.g. nitrogen) is recommended to avoid possible oxidation.
Finely ground oil shale should be stored under nitrogen.
2_ Chemical Methods
Acid Demineralization To dissolve minerals, Durand and Nicase recommend a
series of successive treatments at 65-70·C:158
a. Treatment with 6N HCl for 2 hours
OIL SHALE Page 110
b. Washing with distilled water containing a small amount of Hel
c. Treatment overnight with 6N HCI
d. Three washings with distilled water containing a small amount of Hel
e. Treatment for 2 hours with a mixture containing 2N HCI and IS.SN HF
f. Washing with distilled water containing a small amount of HCI
g. Treatment overnight with a mixture containing 2N HCI and lS.SN HF
h. Washing with distilled water containing a small amount of Hel
i. Treatment with 6N HCI for 2 hours
j. Two washings with distilled water
k. Filtration to recover the kerogen concentrate
The initial treatments with Hel and the use of the HC1-HF mixture were
designed to minimize the formation of complex alkaline earth fluorides, such as
ralstonite (NaMgAIF •• H20), while washing with acidulated water is claimed to
avoid the flocculation of clay minerals that would plug the filter. Pyrite is
not removed by this procedure.
In many cases, equivalent demineralization with less chance of organic
alteration has been obtained by carrying out the treatments at 20·C and for
shorter times. 163
Base/Acid Demineralization McCollum at Amoco has investigated the use of base
to dissolve silicates, followed by an acid treatment to dissolve carbonates. 164
This procedure affords a kerogen concentrate with very low ash and has the
advantage of avoiding the long and tedious series of manipulations listed above.
The procedure involves digestion with concentrated aqueous caustic (50-70%) at
150-160'C for 4-6 hours, followed by extraction with IN Hel or other mineral
OIL SHALE Page III
acid. The caustic treatment converts free silica to silicates and alumino
silicates into sodalitic species that are decomposed and dissolved by mineral
acid (aluminum and the base metals are removed as soluble salts, silicon as
silicic aCid). Kerogen concentrates containing less than 10% ash were obtained
from Green River oil shale having 70% - 85% ash. Pyrite is the major mineral
that is not removed by this procedure. About 95% of the organic matter in the
raw shale is recovered without markedly altering the kerogen; the kerogen
concentrate has an Hie atomic ratio of 1.52-1.58 that is similar to that of the
organic fraction of the parent shale (FA - 140 gal/ton). The chemistry appears
similar to that of the molten caustic method developed by TRW and described in
U.S. Patent 4,545,891. However, the temperatures used in the base/acid method
are substantially below those (-300 C) used in the first stage of the molten
caustic method.
3. Physical Methods
Physical methods to produce an organic-rich kerogen concentrate are of
interest for two reasons: exposure of the kerogen to the strong acid and/or
base is avoided, thereby lessening the chance of chemical alteration; secondly,
because such "beneficiation" methods are practiced on a large scale in the coal
and ore-processing industries, they may be useful in a commercial oil shale
operation. Such methods do generally involve the potential for contamination
of the kerogen with materials used to effect the separation. However, in many
cases the potential impact of such contaminants can be limited by using only
one, or a small number, of known and easily identified chemical species.
OIL SHALE Page 112
Sink-Float Methods The sink-float method of Vadovic113 was discussed in an
earlier section. Luts' method involves suspending finely ground oil shale in
an aqueous CaCl z solution and centrifuging the resulting suspension to enhance
separation. 165 The recovered kerogen is then washed with water to remove
CaC1 2 " The density of the CaC1 2 solution can be varied to optimize kerogen
recovery and residual mineral content for a specific oil shale, but is
generally in the range of 1.06 to 1.15 g/mL.
Hubbard and co-workers concentrated Green River oil shale kerogen by
centrifugation in benzene-carbon tetrachloride mixtures. 166 In this work, the
finely ground oil shale was first extracted to remove benzene-solubles, dried,
then subjected to successive sink-float separations in solvent mixtures having
densities of 1.40, 1.20 and 1.15 g/mL. The concentrate that floated in the
1.20 g/mL mixture represented 6% of the starting organics and contained 14 wt%
ash. The concentrate from the last stage contained 9 wt% ash, but represented
only 1% of the starting organic material. In this case the constancy of atomic
H/C ratios and assay oil/Corg ratios suggests that little fractionation of the
kerogen occurred. Similar results were observed by Vadovic.
Thus, laboratory sink-float methods offer mild conditions to minimize
chemical alteration and, in favorable cases, kerogen concentrates with low ash
contents. Also, elemental analyses can be carried out at each step and the
results extrapolated to zero ash to obtain an estimate of the mineral-free
kerogen composition. Disadvantages include high rejection of organics, leading
to low kerogen recoveries, and the possibility of kerogen fractionation along
with mineral rejection.
OIL SHALE Page 113
Oil-Agglomeration Quass 167 relied on selective wetting of kerogen particles by
an oily pasting material, such as hexadecane to separate kerogen from South
African oil shales (for details, see the review by Robinson159 ) This method
was used successfully on Green River oil shale by Smith and Higby,'68 but was
unsuccessful in three of four cases studied by Himus and Basak. 169 Thus, the
oil-agglomeration method does not appear to be generally useful, especially if
not preceded by a detailed characterization of the mineral and organic surfaces
to provide a basis for selecting an optimum, or near optimum, pasting solvent.
Combined Chemical-Physical Treatment From the results of model compound-model
mineral tests, Siskin and Brons identified interactions between acidic clay
_minerals and N-containing organics as being much stronger than other likely
candidates for kerogen-mineral interactions in Green River oil shale.170.171
Ammonium sulfate was selected to serve both as a source of ammonia, a base, and
as a source of acid, bisulfate ion, to generate porosity by attacking carbonate
minerals. These workers found that treatment of 80-100 mesh Green River oil
shale with aqueous ammonium sulfate at 85°C for 72 hr effectively disrupted
kerogen-mineral interactions (Table 16). Nearly all (98%) of the oil shale
organics were recovered in the kerogen liberated by this procedure and 85% of
the mineral matter present in the starting shale was rejected. Nevertheless,
the kerogen fraction still contained 38% minerals.
OIL SHALE Page 114
Table 16. Mineral distributions from aqueous ammonium sulfate enrichment of
Green River oil shale with toluene at 85°C for 72 hours.
wt% of Whole Shale Isolated Minerals Fraction Carbonates Quartz Albite Pyrite Illite Organics
Whole Shale (21.8% organics)
Kerogen Fraction 33.1 (68.2% organics)
Aqueous Phase 37.6 (dissolved)
Mineral Fraction 28.9
Rejection, %a
40.6
3.0
37.6
-0-
93
16.7
3.1
-0-
13.6
81
7.6
1.2
-0-
6.4
84
2.0
1.4
-0-
0.6
30
10.9
3.0
-0-
7.9
73
21.8
21.4
-0-
0.4
1.8
(a) Normalized per cent rejection of the species from the kerogen fraction.
The important role of chemistry is clear in the discussion above. However
differential wetting, a phenomenon that is typically associated with physical
separation methods, was also critical to the success of Siskin's method.
Efficient kerogen recovery was only obtained by adding an organic solvent that
would wet and swell the kerogen, thereby aiding the physical sink-float
separation. Thus, both chemical and physical aspects are important in this new
method for concentrating kerogen under mild conditions.
Toluene was used in the laboratory tests for swelling and floating the
liberated organics. A shale-derived naphtha or distillate would probably be
used in any larger scale application of this method.
OIL SHALE Page 115
B. STRUCTURAL INFERENCES FROM BITUMEN AND SHALE OIL ANALYSES
Inferences regarding kerogen structure drawn from the results of bitumen
and shale oil analyses are generally based on the premise (a) that the bitumen
is analogous to residual monomer in a polymer, i.e. the bitumen represents
units of the precursor that did not become bound into the insoluble three
dimensional macromolecular network of the kerogen. or (b) that the bitumen
comprises units of the kerogen structure that have been cleaved more or less
intact from the kerogen by thermal treatment. In either case, it's assumed
that the kerogen is structurally similar to the extractable bitumen.
Some of the compound types present in oil shale bitumens and the geochem
ical inferences that have been drawn from these results were discussed in an
earlier section. Much more information can be drawn from bitumen composition
in cases where the analysis is comprehensive. In a landmark series of papers,
W. E. Robinson and his co-workers have presented such a comprehensive analYSis
of bitumen from Green River oil shale. Included were reports on paraffins,88
the changes in paraffins with depth,172 steranes173 and other cycloalkanes,7o
aromatics,174 and polar compounds. 175 Additional information was provided by
many other workers.71,176~180 Extensive analyses of bitumens in the British
Kimmeridge oil shale,65-68 Australian oil shales,163,181 French oil shales,179
and Brazilian oil shales63 ,182 have also been reported. There is a remarkable
similarity in the types of compounds found by these analyses (Table 17). Of
course, there are differences in the details of bitumen composition and the
reader is directed to the cited papers for discussions of how these differences
are related to the origin, depositional environment and thermal history of the
particular oil shales.
OIL SHALE Page 116
Table 17. Representatives of major compound types found in oil shale bitumens
and suggested as structural units in oil shale kerogen.
~
n-Paraffins Isoparaffins (2-melhylalkanes)
Pristane
Anteisopsraffins (3-methylalkanes)
Phylane
Carotane
Isoprenoid Paraffins and Isoprenoid Cycloalkanes
A Stefano Series Hopane Series Gammacerane Series
c8 0 00 (:CO ~ &
I", " ,.. ,.. ,.. ,.. ,.. Aromatics
OH
2 0
~o II c,
OH Alcohols Ketones Carboxylic Acids
~ I H
Pyridines Pyrroies
OIL SHALE Page 117
Most oil shales contain only small amounts of bitumen; rarely does the
bitumen content exceed 15% of the total organic matter. Moreover, bitumens are
generally richer in hydrogen and poorer in aromatics and N,S,O-heteroatoms than
the corresponding kerogen. This limits the usefulness of structural inferences
drawn from the bitumen composition. Shale oil, on the other hand, generally
represents a much larger fraction of the organic matter, over one-half in most
cases and in favorable cases even more. Methods for shale oil analysis are
generally similar to those used for petroleum and have been reviewed.183.184
More recent work is presented in References 65, 185 and 186. Structural
inferences drawn from the results of shale oil analyses typically parallel
those from bitumen analyses. However, as a consequence of their thermal
treatment, shale oils are usually richer in aromatics and olefins than the
starting kerogen, and poorer in N- and $-heteroatoms since these become
concentrated in the non-volatile char. Thus, shale oils reflect both the
structure of the starting kerogen and the thermal treatment used to produce the
oil. Consequently, the results of shale oil analyses cannot be relied upon to
give an accurate picture of kerogen structure.
While many useful inferences about kerogen structure have been drawn from
bitumen and shale oil analyses, the limitations outlined above have led to the
development of more direct methods for probing kerogen structure.
OIL SHALE Page 118
C. STRUCTURAL ANALYSES OF KEROGENS
Selective derivatization, sometimes coupled with spectroscopic analyses,
oxidative degradation, mild thermal degradation and pyrolysis-GeMS methods have
been used to probe the chemical structure of kerogens.
1. Oxygen Functional Groups
Only one attempt to characterize the oxygen functional groups in Green
River oil shale kerogen is reported in the literature. Fester and Robinson
used acid demineralization (successive treatments with HCl and HF) to prepare a
Green River kerogen concentrate containing 14 wt% mineral matter, and wet
chemical methods to determine the distribution of oxygen functional groups
(Equations 9-14 and Table 18).'87"88
Carboxylic acids:
2 R-CO,H + Ca(OAc), --------> (R-CO,),Ca + 2 HOAc
Esters:
R-CO,R'
Amides:
R-CONH,
HOAc was steam distilled and titrated with 0.02 N NaOH
Solid was filtered, washed with dilute NaOH (pH -8), dried and analyzed for Ca
0.2N NaOH HCl Ca(OAc), --------------------> -------) -----------) As Above
18-24 hr. Reflux
IN NaOH --------------> R·CO,Na + NH, 3 hr. Reflux
Trap in H,BO, ---------------> Titrate
I I. Acidify ---------------------->
2. Extract with Et,O
IR Analysis for amines
N-Analysis
(9)
(10)
(II)
OIL SHALE Page 119
Aldehydes, Ketones:
5% H2N-OH • HCI/H20 1. Fi lter R-CHO, R2CO ---------------------------) -------------) N-Analysis (12)
pH=7.5-8, 18-24 hr., 20'C 2. H2 0 Wash 3. Dry
NaBH, in 0.1 N NaOH 1. Filter R-CHO, R2CO ---------------------> ------------>
6 hr, Reflux 2. H20 Wash 3. Dry
Alcohols, Phenols:
I. AC20/pyridine 0.2 N NaOH R-OH, Ar-OH ------------------> ------------>
2. Filter Hydrolysis 3. Wash
Quantitative IR Analysis using 5.81' peak (KBr pellet)
1. Neutralize 2. Steam distill
HOAc 3. Titrate HOAc
(13)
(14)
Table 18. Distribution of oxygen functional groups in Green River oil shale
kerogen reported by Fester and Robinson. 187
Oxygen Group
Carboxyl
Ester
Amide
Carbonyl (R-CHO, R2CO)
Hydroxyl (R-OH, Ar-OH)
Ether (by difference)
% of Total Oxygen
15.3
24.7
0.6
1.2
4.7
53.5
later, Robinson and Dinneen applied this procedure to a series of twelve
oil shales from around the world. s7 While directionally correct, the reported
results (Table 19) did not take into account the organic structural changes
OIL SHALE Page 120
which may have occurred during the preparation of the kerogen concentrate or
during the derivatization reactions, nor did they consider the inabilities of
aqueous reaction media to wet and swell the non-porous kerogen concentrates.
Table 19. Carboxyl, ester and hydroxyl oxygen in kerogen concentrates from
around the world. (After Dinneen and Robinson, Reference 87.)
Oxygen Oxygen Oxygen as COOH as COOR as OH
Kerogen (mgJg C) Kerogen (mgJg CJ Kerogen (mgJg C)
New Zealand 64 Argentina 109 New Zealand 42
Brazil 39 Oregon 99 Brazil 33
Oregon 28 New Zealand 78 Oregon 29
Argent ina 19 Canada 62 Colorado 22
Colorado 14 Brazi 1 61 Argent ina 17
Canada 8 Spain 39 Canada 17
Spain 8 South Africa 38 South Africa 15
Scotland 7 Alaska 37 Australia 13
South Africa 5 Scotland 31 Scotland 12
France 3 Colorado 25 France II
Alaska 2 France 20 Spain 10
Austral ia I Australia I Alaska 4
2. Oxidation
Oxidative degradation, one of the primary tools of classical natural
product chemistry, has been widely used to probe kerogen structure. Oxidation
methods for kerogen characterization have been reviewed by Vitorovit. 189
Alkaline permanganate189~194 and chromic acid195~ZOO have been the two most
OIL SHALE Page 121
widely used oxidants, although ozone,201-204 periodate,205 nitric acid,206-Z08
perchloric aCid,209,210 air/oxygen,211 hydrogen peroxide,212,213 electrochemical
oxidation214 (among many other reagents) have also been used.
a. Alkaline Permanganate
Alkaline permanganate oxidations of kerogens have been carried out in two
very different ways. Older work was generally done using the "carbon balance"
method that was developed by Bone and Himus for coal studies. 215 The products
of this exhaustive oxidation are COz. oxalic acid (HOzC-COzH, from aromatic
rings), non-volatile, non-oxalic acids (mostly benzene polycarboxylic acids)
and unoxidized organic carbon (reported values generally include both handling
losses and analytical errors). Because aliphatic material is oxidized mainly
to CO2 , this method is not well-suited to probe the structures of kerogens that
are highly aliphatic. This led to the development of stepwise procedures to
give products that retain more structural information about the starting
kerogen.
The careful development of the stepwise alkaline permanganate method and
its application to a wide variety of oil shale kerogens have been extensively
documented by Vitorovic and his co-workers.189-190 In general. these workers
attempt to minimize unwanted secondary oxidation of the first-formed product by
adding the oxidant (KMnO, in 1% aqueous KOH) in small portions, heating until
the violet (MnO,-) and green (MnO,'-) colors disappear, then separating the
base-soluble oxidation products. The base-insoluble residue is then subjected
to another oxidation step with a fresh portion of permanganate. At 75*-80*C,
the time required to discharge the oxidant color increases from a few minutes
OIL SHALE Page 122
for the initial portions to several hours, until eventually the color is not
discharged after an extended time (12-36 hr.),. Ether extraction yields the
neutral and basic products, acidification with Hel yields a high-MW acid
fraction and a second ether extraction yields the ether-soluble acids. The
latter are then analyzed, either directly or as the methyl esters. using
conventional methods.
In some cases, the acids obtained from stepwise oxidation proved to be of
such high molecular weight that they precipitated upon acidification and were
insoluble in ether. Such acids were difficult to characterize. In these
cases, the precipitated acids were subjected to further stepwise oxidation to
produce the desired ether-soluble acids of lower molecular weight.
The results of stepwise alkaline permanganate oxidations of kerogen con
centrates from four representative oil shales are summarized in Tables 20-23.
While the distributions vary widely. the n-monocarboxylic acids and a,w-dicar
boxylic acids together comprise the major part of the product. Tricarboxylic
acids were much less abundant in the product from the Pumpherston shale. while
isoprenoid acids were abundant (5.74% of total acids) only from the Irati shale.
The Aleksinac and Irati shales yielded more aromatic acids than the Green River
or Pumpherston shales.
OIL SHALE Page 123
Table 20. Compositions of the kerogen concentrates. (After VitoroviC, et aT., Reference 191.)
Green River Aleksinac Irat i Pumpherston (Rifle. CO) (Yugoslavia) (Brazil) (Scotland)
Organics, wt%
C 77 .39 71.87 78.83 75.70
H 10.26 8.73 9.47 10.37
N 3.10 3.21 3.92 4.18
O+S(diff) 9.25 16.19 7.78 9.75
Atomic H/C 1.59 1.46 1.44 1.64
Atomic O/C 0.09 0.17 0.07 0.10
% Organics in 49.61 65.57 51.49 81.98 the Kerogen Concentrate
OIL SHALE Page 124
Table 21. Product yields from stepwise alkaline permanganate oxidations.
(Data of Vitorovit, et aJ., Reference 191.)
Oxidation of kerogen concentrate
Product yield based on starting organics! wt% Neutrals Ether-soluble Precipitated Acids, Total + Bases Acids Acids Aqueous Solutions Yield
Green River 0.86 24.32 51.20 8.12 84.50 (25)
Aleksinac 1.00 27.97 45.57 18.30 92.84 (23)
Irati 1.69 47.85 31.38 8.34 89.26 (31)
Pumpherston 3.73 35.94 22.14 3.33 65.14 (29)
Oxidation of precipitated acids
Product yield based on precipitated acids! wt% Precipitated
Neutrals Ether- so 1 ub 1 e Acids, Acids, Total + Bases Acids Last Step Aqueous Solutions Yield
Green River 11. 22 96.56 5.26 10.13 123.17 (30)
Aleksinac 3.25 72.86 13.35 14.01 103.47 (26)
Irati 7.96 80.21 0.25 9.75 98.17 (25)
Pumpherston 19.24 33.76 12.16 5.44 70.60 (31)
(a) The value in parentheses below each shale name is the number of steps required for complete oxidation of the kerogen or precipitated acids.
OIL SHALE Page 125
Table 22. Acids obtained by alkaline permanganate oxidation. (Data of
Vitorovi c, et al., Reference 191.)
Aliphatic Acids Green River Aleksinac Irati Pumpherston
n-Monocarboxylic C, 0 -C34 (10 -C34 C, 0 -C29 C9 -C36
Monocarboxylic, C15 (15,C17 Branched
a,w-Dicarboxylic C6 -e34 Cs -C33 (6 -C26 C6 -e33
Oicarboxylic. C8 ,C9 CB Branched
Isoprenoid C14 -(17; C14 -C17 ; C14 -C17 ;
C19 ,C20 C19 -e21 ('9-C21
Tricarboxylic C6 -C14 C6 -C15 C4 -C9 C3 -C.
Tricarboxylic, C6 -C9 Branched
Tetracarboxylic Ca -C 14 C8 -(18 (6 ,Ca -e13
Tetracarboxylic, C9 ,C1 0 C10 ,C13 Branched
Aromatic Acids Mono- to Mono- to Mono- to Mono- to Tetracarboxylic Tricarboxylic Tetracar- Tetracarboxyl ic
boxy] ic
OIL SHALE Page 126
Table 23. Distributions of the oxidation product acids. (Data of Vitorovic,
et a 7 .• Reference 191.)
Wt % of Total Acids Aliphatic Acids Green River Aleksinac Irati Pumpherston
n-Monocarboxylic 27.94 18.80 56.81 54.32
Monocarboxylic, 0.02 0.10 Branched
a,w-Oicarboxylic 58.06 55.77 21.04 36.75
Dicarboxylic, 0.04 Branched
Isoprenoid 1.95 0.68 5.74
Tricarboxylic 5.26 6.05 5.26 1.37
Tricarboxylic. 0.10 Branched
Tetracarboxylic 4.08 10.77 3.67
Aromatic Acid~ 2.87 7.93 11.13 3.79
OIL SHALE Page 127
Obviously, much effort has been expended in developing methodology for the
stepwise oxidation procedure and in characterizing the oxidation product acids.
Considerably less effort seems to have gone into defining the susceptibility of
kerogen structural features to oxidation and the precise relationships between
oxidation products and kerogen structure. To quote Stewart, "Permanganate is a
vigorous, and in some cases a drastic, oxidant which has long been used in the
organic laboratory."216 The literature on permanganate oxidation is extensive
and several (none very recent) reviews are available. 216 ¥223 In addition,
Randall and co-workers studied the permanganate oxidations of about seventy
model compounds under the carbon balance conditions. 224 A comparable study to
define oxidation susceptibility under the milder stepwise oxidation conditions
is not available. Therefore, while,a critical review is not possible within
this article, it is useful to outline those features of alkaline perrnanganate
oxidation that seem most likely to be important in kerogen studies.
Generally speaking, alkaline permanganate will oxidize alkylbenzenes,
alkylthiophenes and alkylpyridines -- but not alkylfurans -- to the corresponding
carboxylic acids. This is not true in cases where the aromatic ring bears an
electron donating group (e.g. -OH, -OR, -NH,). In such cases degradation of
the aromatic portion is usually rapid. Condensed aromatics are also attacked;
benzene polycarboxylic acids result (Eq. 9). The need for care can, however,
hardly be overemphasized; even benzene is slowly attacked by hot alkaline
permanganate solutions.
OIL SHALE Page 128
KMn04-,OH-..
(9)
Olefins are rapidly converted into the corresponding glycols, which are
then cleaved to carboxylic acids (Eq. 10). Cyclic olefins yield dicarboxylic
acids. Enolizable ketones are also cleaved, presumably via the enol.
HO OH KMnO, ., OH' I I
R-CH=CH-R' ---------------) R-CH-CH-R' ----) R-CO,H + HO,C-R' (10)
Tertiary and benzylic C-H groups are attacked by a free radical mechanism.
Good yield of tertiary alcohols have been obtained in selected cases. Z25 In
simple alkyl systems, the presence of an alcohol group markedly accelerates the
rate of this reaction, probably by increasing water solubility of the unit that
bears the tertiary C-H. A free radical mechanism was established by isolating
the coupling products from dimerization of the intermediate benzylic radicals
(Eq. II).". A very facile free radical attack has also been observed when a
tertiary hydrogen is gamma to a carboxyl group, as in 4-methylhexanoic acid
(Eq. 12)."7"'9
KMnO, . C.H,-CH,CH,CH3 -------------) (II)
OIL SHALE
.. H3CD H3C 0 0
Page 129
(12)
Primary and secondary alcohols are oxidized to the corresponding acids and
ketones; the rates depending on both pKa of the alcohol and pH, and being
roughly proportional to the concentration of the corresponding alkoxide iOO.230
Evidently, permanganate attacks the alkoxide anion much more rapidly than the
alcohol, itself.
Alkaline permanganate oxidation degrades the porphyrin nucleus, giving
pyrrole-2,4-dicarboxylic acid derivatives under mild conditions. 231 Under
these conditions, the porphyrin sidechains -Me, -Et, -CHzCHzCOzH, -eOCH3 and
-CH(OH)CH, persist in the degradation products, but -CH=CH, and -CHO sidechains
are both oxidized to -COzH.
With this outline in mind, some analysis of the results in Tables 20-23 is
now in order. Clearly aliphatic chains are of major importance in all four oil
shales. The high yield of a,w-dicarboxylic acids and the low yield of aromatic
acids obtained from Green River kerogen suggests that this structure includes a
large number of alkylene bridges joining small, activated aromatic rings that
would be destroyed by oxidation. The appreciable amounts of tri- and tetracar
boxylic acids suggests that some crosslinks in the macromolecular network
involve aliphatic or cycloaliphatic (possibly cyclo-olefininc) units. In this
basic shale, the n-monocarboxylic acids may be present salts, especially of
divalent cations (e.g. Ca2+), though some contribution from alkyl sidechains
cannot be ruled out.
OIL SHALE Page 130
The situation with the Aleksinac shale is similar, but the even higher
yields of tri- and tetracarboxylic acids and the abundance of aromatic acids
suggest that cyclic structures may be more important in this case.
The a,w-dicarboxylic acids are much less abundant in the product from the
Irati shale. Aromatic acids are relatively abundant and n-monocarboxylic acids
are, by far, the major component. These results suggest that relatively large
aromatic clusters with n-alkyl sidechains, and joined by alkylene bridges are
important in the Irati shale.
The Pumpherston shale stands out in giving a very low yield of oxidation
products (65%) and in giving a relatively large amount of neutral and basic
product, -8 wt% based on the starting shale. Presumably, some structures in
this shale were oxidized to CO2 , however, the high atomic HIC and the low yield
of aromatic acids suggest that these were not activated aromatics of the type
found in coals. The high nitrogen content suggests that pyrrolic compounds may
be the source of the CO,. but knowledge of the fate of nitrogen would be needed
to make this more than mere conjecture.
The discussion above shows that oxidation, using mild selective techniques
like the stepwise alkaline permanganate method, can provide a wealth of infor
mation about kerogen structure. However, it should also be clear that much
additional research is needed to define the reactivity of different structural
features and establish clear relationships between oxidation products and the
structure of the starting kerogen.
OIL SHALE Page 131
b. Chromic Acid Oxidation
Chromic acid oxidations of kerogens have been reviewed by Vitorovit189
and the chemistry of chromic acid oxidations in general has been reviewed by
Wiberg232 and by lee,217 The structural information obtained from kerogen
oxidations with chromic acid and other chromium oxidants is usually similar to
that obtained with alkaline permanganate. However, the recovery of organic
carbon in the chromic acid oxidation products is often low. For example? the
total product obtained by Burlingame and co-workers in a stepwise chromic acid
oxidation of Green River oil shale kerogen amounted to less than 10% of the
total organic material, though essentially all of the organic material was
oxidized. 2oo Consequently, the alkaline permanganate procedure appears to be
superior for elucidating kerogen structure.
3. Oepolymerization under Mild Conditions: Heat-Soak/Extraction
While the oxidations discussed above can give reasonably high organic
recoveries in favorable cases, extensive alteration of the structure and some
features are obliterated. This is especially a problem in the case of easily
oxidized nitrogen functionalities. Therefore, while Fischer Assay remains the
standard for oil recovery in retorting and most thermal treatments give oil
recoveries comparable to Fischer Assay, the search for ways to improve thermal
oil recovery continues. Indeed, since the Eele patent of 1694,2 literally
thousands of patents and papers have been published on the subject. One
particularly attractive approach to minimize the formation of intractable
residues has been to heat the shale at a moderate temperature for a long time,
then extract the depolymerized kerogen."· Another approach is to heat the
OIL SHALE Page 132
shale at a low temperature for a time, then increase the temperature to the
conventional retorting temperature, cool. then extract. 233 In either case, the
idea is to depolymerize the kerogen with minimal condensation to intractable
material, then recover even non-volatile oil by extraction.
Until recently, it was not realized that there is an optimum time
temperature window for thermal depolymerization. For example, Hubbard and
Robinson reported that a two-step process (two heating + extraction cycles) was
not necessary for Green River oil shale, since the same conversions could be
obtained by simply heating for a longer period before extraction. 118 Bock and
co-workers found this was not the case. 234 ,235 These workers heated oil shale
under a nitrogen sweep to obtain volatile oils + gases, then extracted the
residue with THF to obtain additional depolymerized, but non-volatile, product.
For example, a sample of Green River oil shale was heated at 400°C for 1 hour
to obtain 40 wt% (based on total organic matter) volatile products, and then
extracted to obtain 44 wt% extractable liquids. The results for other times
and temperatures show that for each temperature in the range of 350°-425°C
there is an optimum heating time, and that heating for 400·C for 1 hour gives
an optimum conversion (Figure 36). Similar results were obtained for Ramsay
Crossing oil shale from the Rundle deposit (Queensland, Australia, Figure 37).
Similar results, but somewhat lower overall conversions were obtained from a
more highly aromatic Eastern U.S. Devonian shale.
OIL SHALE Page 133
Figure 36. Plots of conversion vs. time show that there is an optimum time
temperature window for thermal depolymerization of Green River
oil sha 1 e kerogen. (React i on under 1 atm. Nz , Reference 234).
-DEFINITION OF TIME· TEMPERATURE WINDOW-
100 r---r--,---,--,--,---,--,--, (450°C) ........ --...... (400°C) 0
90..::- • • &, -.. 80
-~ 0. 70 -.! c 60 o te ~ 50 c o o 40 .2 c co e> 30 o
20
10
--..
o
• -o
(375°C)
°o~~~~~~~~~~~~ 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Time (Hr.)
OIL SHALE Page 134
Figure 37. An optimum time-temperature window was also found for the thermal
depolymerization of Ramsay Crossing oil shale kerogen. (Data from
Reference 234).
-Definition of Tlme·Temperature Wlndow-
100r----r--~.----.---.----.---~
(425·C) 90 \ 0
A A
80 -(400·C)
-~ • 70 o o
(3S0·C)
I o,--..,.j
°OL---~---L--~--~----L---~ 2.0 4.0 6.0
Time (Hr.)
OIL SHALE Page 135
After heating at the optimum time, the unextractable organic residue is
not yet converted into the intractable char obtained from conventional thermal
methods. Heating the THF-extracted residue a second time at 400°C for 1 hour,
and again extracting with THF enabled recovery of the balance of the organic
material. Together, the two heat-soak/extraction cycles achieved conversion of
100 wt% (±2 wt%) of the total organic matter in Green River oil shale -- over
90 wt% of which was recovered as liquid products. Moreover, because of the
relatively mild conditions, absence of added reagents and controlled time,
alteration of the recovered organics should be minimal. Thus, the mild heat
soak/extraction method is an attractive alternative to oxidation for providing
liquid products for structural studies.
4. Micropyrolysis/GC-MS
Schmidt-Collerus and Prien used micropyrolysis coupled with GC-MS to
obtain information about the structural units in Green River oil shale
kerogen. 236 In addition to the on-line micropyrolysis/GC-MS studies, larger
samples of kerogen were pyrolyzed to obtain products that were fractionated by
chromatography (ion exchange, Fee1 3 complexation, silica gel) into compound
classes, then by GPC into fractions of increasing molecular weight. These
samples were analyzed by conventional MS techniques. Additional information
was obtained by microscopic and microspectrophotometric studies on the whole
kerogen.
These studies led to the conclusion that Green River kerogen contains two
distinct types of material: a-kerogen, an alginite-like material of low
aromatic content and 6-kerogen with a much higher content of aromatic (probably
OIL SHALE Page 136
polycondensed) material. The latter, representing about 5% of the total. was a
reddish-brown color.
In micropyrolysis, a-kerogen yielded several types of products: normal and
branched alkanes, alkyl naphthalenes and tetralins, alkyl-substituted tricyclic
or phenanthrene derivatives (Table 24). These results led to the important
conclusion that most of the cyclic units (a1icyc1ic5, naphthenics, aromatics)
in the Green River kerogen are small, containing 1-3 rings.
OIL SHALE Page 137
Table 24. Principal fragmentation products from micropyrolysls of Green River
oil shale kerogen. 236
Aliphatic hydrocarbons:
Alicyclic hydrocarbons:
Cyclohexanes
Decalins
Hydroaromatic hydrocarbons:
Dialkyltetralins
Hexahydrophenanthrenes
Dialkylbenzenes
Dialkylnaphthalenes
Alkylphenanthrennes
n-C 10 - n-C 34 Branched C lO-C 36
00- C 5-8 H 17-25
00- C2_5H5_11
C 1-3 H 3-7 + 6 H
~ ~ C 3_4 H 7_9
C 1-3 H 3-7
OIL SHALE Page 138
5. Acid-Catalyzed Hydrogenolysis
Hubbard and Fester used hydrogenolysis in the presence of tin dichloride
to degrade Green River oil shale kerogen Eq. 13).'37 Although the conditions
were severe, the gas make was low and a good yield of liquid products was
obtained for characterization. Under these conditions, most of the heteroatoms
were removed: N, 90% as NH3 ; S, 90% as H,S; 0, 84% as CO, (56%) and H,O (28%).
This led to the conclusion that N, S, and 0 functionalities comprise weak links
in the kerogen structure. This may be so, but seems at variance with other
results. Given the high yields of liquids, this technique looks promising, but
the extreme severity suggests that re-investigation under milder conditions
might be profitable.
Green River 4200 psig H, Kerogen + SnCl, • 2 H,O
Concentrate ----------------)
335'C, 4 hr.
D. STRUCTURAL MODELS FOR KEROGEN
9.3 % Gas 86.6 % Benzene
solubles (13)
The need to put the very large mass of information about kerogen structure
into a compact form that is useful for guiding research and development has led
to models for kerogen structure. These models are not intended to depict the
molecular structure of kerogen, at least not in the sense that the double-helix
describes the structure of DNA, or even in the sense that synthetic polymers
OIL SHALE Page 139
are described in terms of monomers joined to form chains which then segregate
into crystallites of well-defined structure and liquid-like amorphous regions.
Kerogens are not so well-ordered. Instead, the kerogen models attempt to
depict a representative collection of skeletal fragments and fUnctional groups
connected into a three-dimensional network in a way that seems reasonable based
on the available data.
No one analytical technique provides sufficient information to construct a
usefully detailed model of a kerogen structure. Thus, most workers now use a
multiple-technique approach. However, even within one laboratory, the results
of such an approach are diverse -- and often conflicting. This complicates
kerogen modeling. Moreover, the collections of techniques used by different
workers have different strengths, hence tend to emphasize different features of
kerogen structure. Sample-to-sample variation further clouds the picture.
Consequently, it is not surprising that several models have been proposed for
the structure of Green River oil shale kerogen. These will be discussed to
illustrate the state of kerogen modeling. For contrast, the section will end
with a discussion of the characterization and structural modeling of the
kerogen in Ramsay Crossing oil shale from the Rundle Oeposit in Australia, and
generalized kerogen models.
1. Burlingame Hodel of Green River Oil Shale Kerogen
As discussed above, Burlingame and his co-workers used chromic acid
oxidation to degrade Green River oil shale kerogen and mass spectrometry (MS)
as the key tool for characterizing the oxidation products.196~2oo Additional
information was provided by studies of the bitumen, again primarily by MS, and
incorporated into the structural model (Figure 38).
o 0 II II
CH,"" (CH,),('C
METHYLKETO ESTER
ISOPRENOIDAL SUBSTITUENTS
NORMAL HYDROCARBON SU BSTITU ENTS
o II
O-C"Q
AROMATIC R
SUBSTITUENTS ENTRAPPED COMPOUNDS OF UNKNOWN NATURE ~~~ ....
ESTER SUBSTITUENTS
ISOPRENOID ESTER SUBSTITUENTS
,c=:::. KEROG~ NORMAL ESTER SUBSTITUENTS
SUGGESTED SITE OF OXIDATION CLEAVAGE (CrO,)
Figure 38. Burlingame model of Green River oil shale kerogen. 1• 7,1.'
(Reprinted with the permission of Pergamon Journals, Ltd.)
-... o
OIL SHALE Page 141
Key features of this model include very large regions of undefined
structure containing trapped organics of unknown nature and bearing sidechains
linked to the main structure by non-hydrolyzable C-C and hydrolyzable ester
linkages. An ester linkage is shown connecting two such large regions. Also,
the model includes a carbocyclic ring. presumably derived from a Diels-Alder
reaction involving the allyl C·C of phytol (3,7,11,15-tetramethyl-2-hexa-decen
I-al), and the gem-dialkyl site on this ring is identified as a possible
location for oxidative cleavage to give an isoprenoid C16 carboxylic acid. To
understand this model, it is important to recall that the oxidation products
(acids, ketones) upon which most of this structure is based, represented only
0.7 wt% of the total organic carbon.
2. Djuritit-Vitorovit Model of a Green River Oil Shale Network
Vitorovit and co-workers used stepwise alkaline permanganate oxidation
to obtain carboxylic acids in high yield (70% of total organic carbon) from
Green River oil shale kerogen. 193 The more recent studies that were discussed
in a previous section extended and confirmed this work. Based on their
oxidation results a crosslinked macromolecular network structure was proposed
(Figure 39).
UNBRANCHE~\ MONOCARBOXYLIC ACID
00
_ISOPRENOID CHAIN
UNBRANCHED -+-- ALIPHATIC CHAIN
ISO ACID 0 ). 110
o UNSATURATED~O FATTY ACID
/ REGIONS SUSCEPTIBLE TO KMnO, OXIDATION OR ALKALINE HYDROLYSIS t
SIMILAR
~ I I I I -0 ISOPRENOID MONO.
CARBOXYLIC ACID
POLYMETHYLENE BRIDGE
\
Figure 39. Djuricic and Vitorovic model Green River oil shale kerogen
as a crosslinked, three-dimensional macromolecular network composed
mainly of unbranched alkyl and alkylene chains. (Reprinted with
the permission of Pergamon Press. Ltd.) ~ ... N
OIL SHALE Page 143
The most striking feature of this model is the predominance of straight
chain groups in the backbone of the network. The network bears both branched
and unbranched sidechains. The branching points (indicated in the model by
open circles) must be of a type susceptible to oxidative attack, alkaline
hydrolysis, or both in order to yield the observed mixture of mono-, di-, tri
and tetracarboxylic acids. This model accommodates many important experimental
observations, including reversible swelling and gel-like "rubbery" behavior in
the swollen state, but does not satisfactorily account for the aromatic carbons
observed by 13C-NMR or the appreciable Nand S contents found by elemental
analysis.
3. Green River Kerogen Model Proposed by Schmidt-Collerus and Prien
The kerogen model proposed by Schmidt-Collerus and Prien was assembled
from the subunits identified by their micropyrolysis-MS studies (see discussion
above).236 Key features of this model include formulation as a three-dimen
sional macromolecular network and a very uniform hydrocarbon portion comprised
mostly of small alicyclic, and hydroaromatic subunits with few heterocyclic
rings (Figure 40). Long-chain alkylene and isoprenoid units and ethers serve
as interconnecting bridges in this structure. Entrapped species (bitumen)
include long-chain alkanes and both n-alkyl and branched carboxylic acids.
This model provides a useful view of the types and role of hydrocarbon units,
but de-emphasizes heteroatom functional groups and rings. This is not
surprising. The concentration of Nand $ in residues from thermal treatments
of Green River oil shale kerogen is well-known. Hence, groups containing these
elements would not be seen with high efficiency by the micropyrolysis technique
used by these workers.
su su
N!!!j N·Hetero Matrix Subunit -II! Entrapped Acidic Subunit -·11 Entrapped Neutral Subunit
SU Other Matrix Subunits
---
n
----------------
K
/'K o
n
-0 MATRIX SUBUNIT NUCLEUS K = OTHER KEROGEN SUBUNITS
Alkane Bridge (normal + branched) -0- Ether Bridge -'---,4" Methyl Terminal Alkanes
(normal + branched)
~--."" Entrapped Aliphatic Acids (normal + brllnched)
~---,- Entrapped Alkanes (normal + branched)
Figure 40. Small ring systems are highlighted in the Green River kerogen model
proposed by Schmidt-Collerus and Prien.'3. (Reprinted with the
permission of the American Chemical Society.)
o -r V>
j; r m
-0 ., '" '" -... ...
OIL SHALE Page 145
4. The Yen-Young·Shih Multi-Polymer Structure
The structure of Green River oil shale kerogen has been probed by Yen and
his co-workers using a wide variety of techniques, including stepwise alkaline
permanganate and dichromate/acetic acid oxidations, electrochemical oxidation
and reduction (in non-aqueous ethylenediamine/liCl) and X-ray diffraction
techniques.238-243 Based on the results of these studies, Yen concluded that
aromaticity was low ("approaches zero) but isolated carbon-carbon double bonds
were possible, that the structure was largely comprised of 3-4 ring naphthenes,
that oxygen is present mostly as esters and as ethers, that the kerogen
structure comprises a three-dimensional network and that ethers serve as
crosslinks in this network (based on results obtained by Fester and Robinson)
with additional linkages provided by disulfides, nitrogen heterocyclic groups,
unsaturated isoprenoid chains, hydrogen bonding and charge-transfer
interactions (Figure 41).
Using these components as building blocks, Yen constructed a "multipolymer"
network (Figure 42).245 Further discussion of this model was provided by Yen
in References 242 and 246.
The Yen-Young-Shih model is characterized by a very irregular hydrocarbon
structure consistent with their X-ray results that indicated that the kerogen
is amorphous -- lacking any long-range order. This is in sharp contrast to the
very uniform structure envisioned in the model proposed by Schmidt-Col1erus and
Prien (Figure 40).
OIL SHALE Page 146
Figure 41. Components of Green River Oil Shale Kerogen. (After Yen,
Reference 244).
COMPONENTS BRIDGES
cS 0 -S-S- Disulfide
O· -0- Ether
CD Isoprenoids ,0 E -C-O- Ester
~ \
Isoprenoids Steroids -1'=" ®
H N-H 0
0
CV Terpenolds H Heterocyclic 0
I A -C-N- Amide
H
C + CH,-CH-(CH,).-CH + "=15,17,19 R
© Carotenolds R = (CH,),-CH, Alkadiene
+CH,-CHR+ R = (CH,).CH, n = 22,24,26,28 Alkene
OIL SHALE Page 147
Figure 42. The Yen-Young~Shih multipolymer model of Green River oil shale
kerogen. (Reprinted with the permission of Ann Arbor Science
Publishers, Incorporated.)
Circles: Squares:
!IJ E C
S
I C !IJ A 0
[BJ
Essential Components of Kerogen Molecules trapped in Kerogen network
T
I. isoprenoids. S. steroids. T. terpenoids. C. carotenoids.
Bridges:
D. disulfide. O. ether. E. ester. H. heterocyclic. A. alkadiene.
OIL SHALE Page 148
Yen points out that the extractable bitumen molecules could reside, more
or less freely depending on their size, in c~vities (molecular free volume)
within this network.
To account for the observed variations in the products obtained from the
individual steps of stepwise permanganate oxidation, Yen suggested an unusual
"core plus shell II arrangement for the i ndi vidual kerogen part i C 1 es. The core
in this arrangement is a rather loosely crosslinked region containing most of
the alkyl and alkylene chains and the bulk of the kerogen as naphthenic ring
structures. The shell, on the other hand, is more tightly crosslinked and
contains most of the heteroatom functional groups and heterocyclic rings. This
has interesting geochemical implications. Recall that the outside of a
particle of Yen's kerogen concentrate is precisely the part of the kerogen that
was in contact with the mineral matrix in the starting shale. And that the
heteroatom groups tend to interact more strongly with minerals than do the
hydrocarbon chains. Perhaps, soon after deposition and before the kerogen
became rigidly crosslinked, the kerogen components organized themselves into a
micelle-like arrangement with the polar heteroatom groups on the outside and
the non-polar hydrocarbon groups within. Crosslinking could then "lock in"
this arrangement. Organic-mineral interactions in the resulting composite
would then be ideally situated to hinder physical separation of minerals from
kerogen. This picture is consistent with the results of Siskin, et al. on
chemically-assisted oil shale enrichment. 17o ,171 Clearly, confirmation of this
idea by more direct means (perhaps depth profiling by auger or ESCA) would be
desirable to provide a basis for the development of more efficient oil shale
beneficiation and/or grinding technologies.
,
OIL SHALE Page 149
5. Characterization of Organic Material in Rundle Ramsay Crossing Oil Shale
The organic material in oil shale from the Ramsay Crossing seam of the
Rundle Deposit of Queensland, Australia (RXOS) was characterized and modeled by
Scouten and co-workers, using an integrated, multi-technique approach. 163 Acid
demineralization yielded the kerogen concentrate (RXOS-KC) and the chemistry
that accompanied this demineralization was studied. Selective derivatizations
under mild conditions with isotopically labelled reagents followed by solid
state 13C_ and 29Si-NMR analysis enabled a comprehensive study to chemically
characterize the organic functionalities in the kerogen concentrate. Combining
this data with in depth MS and NMR analyses on shale oils produced under mild
conditions from RXOS and variable-temperature X-ray diffraction studies on the
kerogen concentrate led to the development of a detailed structural model of
the organic material. This work is summarized to illustrate both the features
of the Ramsay Crossing kerogen and the power of the closely integrated
characterization-modeling-reactivity approach.
The Ramsay Crossing oil shale used in this work contained 18.69 wt.%
organic material finely dispersed in a mineral matrix. Solid state 13C NMR
employing CP/MAS indicated an aromaticity of 23% which includes contributions
from olefin and carbonyl carbon types. Exhaustive Soxhlet extraction with THF
removed 8.5 wt.% of the organics as bitumen. The as-received moisture content
was >20%. After drying at 50~C in a nitrogen purged vacuum oven, RXOS has
little surface area (16.6 m'/g) or porosity «0.05 cc/g), hence it is very
impermeable to organic solvents and reagents. As a result, organic reactions
which do not dissolve the minerals proceed slowly and often do not go to
completion. To circumvent this mass transport limitation, RXOS was demineral-
OIL SHALE Page ISO
ized, using aqueous HC1/HF at 20~C247 to produce the corresponding swellable
kerogen concentrate (RXOS-KC) prior to chemical reactions for derivatization of
organic functionalities. The 13C NMR aromaticity and the bitumen content of
the kerogen concentrate were unchanged from the raw shale values. However,
elemental analysis indicated that the bitumen had a significantly higher
hydrogen content, was lower in nitrogen and sulfur than RXO$-KC and had an
aromaticity of <1%. Analyses by Ge, MS and GC-MS indicated that aliphatic
carboxylic acids, esters and amides with long paraffinic chains comprised the
major portions of the bitumen. The elemental composition of the ground (-80
mesh) RXOS, and its kerogen concentrate give empirical formulas for the
organics of C'OOH16,Nz.3S0.70X and C'OOH16,N1.8SS0.709.Z, respectively,
indicating that 0.45 N's/lOOC's were lost during the preparation of the kerogen
concentrate.
Hydrolyzable Nitrogen. Nitrogen bases liberated during the acid washings
would be retained in the acid solutions as the corresponding ammonium salts.
To recover and identify these free bases all of the acid washings were basified
with 50% aqueous KOH under a nitrogen sweep. Any volatile free base evolved
during basification was swept through HCl solutions where it was trapped as the
corresponding hydrochloride. Complete evolution of bases was ensured by
heating the basic mixture until a small of water had distilled into the flask
containing the HCl solution. Solvent extraction of the basic solutions and
evaporation of the HCl solutions to dryness under reduced pressure afforded
0.S2 g of dissolved salts (%C, 0.39; %N, 7.63; %N, 2S.79; %K, 0.21; %C1,
6S.43). Theoretical for NH,C1 (%N, 7.54; %N, 26.18; %C1, 66.28). The very low
carbon content ruled out the presence of major amounts of organic amine
OIL SHALE Page lSI
hydrochlorides. Over 75% of the total ammonia was liberated during the Hel
treatments indicating the presence of primary ami des which are hydrolyzed by
the aqueous acids. The rest of the ammonia liberated during the HF treatment
is due to ammonium ions associated with silicate minerals.248-251
Non-hydrolyzable Nitrogen: Unhindered Basic Nitrogen. Quaternization of
unhindered basic nitrogen compounds (e.g. pyridines) with iodomethane at 50·C
in THF gives the corresponding quaternary ammonium methiodides. Under these
conditions hindered nitrogen bases (e.g., 2,6-disubstituted pyridines) do not
react, or react slowly. Non-basic nitrogen compounds (e.g., pyrroles) do not
react. Thus, methylation with iodomethane {90% 13C enriched was used to
quantify unhindered nitrogen bases. The reaction was followed for 28 days by
"C NMR. Most of the methyl groups were added to oxygen (as esters) appearing
at 51 ppm and to carbon sites at -15 ppm. Nitrogen methyls were added slowly,
but after 14 days 0.1 N-methyls/IOO C's had been added at -40 ppm and this
value did not increase with additional reaction time.
Total Basic Nitrogen - Unhindered Basic Nitrogen"" Hindered Basic Nitrogen.
A sample of O.IN KOH washed (to remove any amine hydrochlorides) RXOS-KC was
treated with anhydrous HCI in methylene chloride at -78·C for times of 5, 10
and 15 min. The number of chlorine atoms added at each reaction time was
obtained by elemental analysis. At the low temperature used, olefin
hydrochlorination to give the corresponding alkyl chloride should be much
slower than protonation (neutralization) of the basic nitrogen groups.
Consequently, the data were extrapolated to zero time to correct for the small
interference of the olefin reaction. This gave 1.0 basic N/s/100 C's as an
estimate of the total basic nitrogens in RXOS-KC. As discussed above, 0.1
OIL SHALE Page 152
unhindered N-bases/IOO C's were identified. The remainder, 0.9 N/s/IOO C's are
ascribed to sterically hindered basic nitrogen functionality.
Non-hYdrolyzable Nitrogen: pyrroles (and Alcohols). Although pyrroles
and indoles are not methylated under PT-O methylation conditions (below), even
highly hindered pyrroles can be quantitatively methylated under more severe
conditions of Qhase transfer £atalyzed - nitrogen methylation (PTC-N
methylation) using methyl tosylate as the methylating reagent, 50% NaOH as the
base and tetrabutylammonium bromide (TBAB) as the phase transfer catalyst.'s,
PTC-N methylation of RXOS-KC with methyl tosylate (90% "C-enriched in the
methyl group) was carried out for 48h under reflux. Analysis by "c NMR (and
confirmed by combustion-MS) indicated the addition of 3.1 methyls/IOO C's;
2.2 O-methyls/IOO C's, 0.3 N-methyls/IOO C's and 0.6 C-methyls/IOO C's. The
2.2 O-methyls/IOO C's correspond to the 1.6 esters observed during PT-D methyl
ation at 51 ppm. The additional intensity corresponding to 0.6 O-methyls/IOO C's
c.a. 56.5 ppm are assigned to derivatives of aliphatic alcohols generally
observed in the range of 55-59 ppm. The assignment of the carbon resonance at
57 ppm to derivatives of aliphatic alcohols is also supported by the results of
'9Si NMR studies of the silylated derivatives. The 0.3 methyls/IOO C's
correspond to derivatives of carbazole-like compounds (0.2 N-methyls/IOO C's),
plus methyl derivatives of pyrroles and indoles (0.1 N-methyls/IOO C's) which
were formed under more severe PTC-N methylation conditions.
Carboxylic Acids and Phenols. The quaternization reaction described
above was carried out in the absence of strong base. Under the influence of a
moderately strong hydroxide base, such as TBAH, acidic O-functionalities,
carboxylic acids and phenols, can be quantitatively methylated to give the
OIL SHALE Page 153
corresponding methyl esters and methyl ethers. This reaction when carried out
in 1:1 THF:CH,OH at 20'C is very selective for O-methylation. PT-O methylation
of RXOS-KC showed by "c NMR to add 2.1 methyls/IOO C's. These methyls were
distributed as 1.6 esters/IOO C's at 51 ppm, 0.3 C-methyls/IOO C's and
0.2 N-methyls/IOO C's.
Ketones. Sodium borohydride is a very mild reducing agent which selectively
reduces ketones in the presence of esters, carboxylic acids and carboxylate
salts. The resulting alcohols can then be converted into the corresponding
trimethylsilyl (TMS) derivatives by silylation. A sample of RXOS-KC was
treated with 0.5N sodium borohydride in diglyme (2-methoxyethyl ether) for 72h
at 20·C. It was then silylated with MSTFA (N-methyltrimethylsilyltrifluoro
acetamide) in THF for I week at 50·C. Analysis of the resulting silyl
derivative by "c NMR indicated the presence of 3.9 TMS groups/IOO C's,
corresponding to 2.5 TMS/IOO C's derived from acids, alcohols and pyrrolic
N-compounds initially present in RXOS, plus an additional 1.4 TMS/IOO C's
assigned to alcohols derived from the corresponding ketones. These carbonyl
carbons, as well as those from acids, amides and carboxylates overlap with the
aromatic carbons in the 13C NMR and must be subtracted from the aromaticity.
Esters. Lithium aluminum hydride is a powerful reducing agent which
rapidly and quantitatively reduces carboxylic acids, esters and ketones to the
corresponding alcohols. Consequently, reduction with LiAIH, in THF for 16h at
20'C, followed by silylation with MSTFA and analysis of the silyl derivative
was chosen to quantify ester fUnctionalities in RXOS-KC. The "C NMR result
indicated the presence of 2.2 TMS/IOO C's arising from the LiAIH, reduction of
OIL SHALE Page 154
esters in RXOS-KC. Since LiA1H4 reduction produces two alcohols per ester
group, it follows that RXOS-KC contains 1.1 esters/IOO C's.
07efins. Olefins rapidly react in the dark with bromine in carbon
tetrachloride at O~C to produce the corresponding dibromides. Interference
from phenol bromination was ruled out because only traces of phenols were
detected during Pt-O methylation. Also, bromination of ketones should be slow
in a nonpolar solvent and in the dark. 8romination afforded a derivative
containing 3.14 Br/IOO C's. It follows that the RXOS-KC contains 1.57 C=C/IOO
C's. Also, 3 olefinic C's/IOO CiS must be subtracted from the total
aromaticity value seen by 13C NMR.
The Distribution of FunctionaliUes in RXOS. Combining the hydrolyzable
nitrogen results with the characterizations on the RXOS-KC enables a
comprehensive description of the distributions of hydrocarbon, and 0- and N
heteroatom functionalities in the solid RXOS (Figure 43). It should be noted
that values for the non-derivatizable functionalities, ethers and
N-alkylpyrroles, were obtained by difference.
Non-Derivatizable FunctionaHties - Analysis of Liquids. The selective
derivatization chemistries described above led to quantification of 76% of the
oxygen (6.95 of the 9.2 O's/IOO C's) and 78% of the nitrogen (1.85 N's of 2.3
N's/IOO C's) present in RXOS. The balance of the oxygen and nitrogen, and
essentially all of the sulfur (0.7 S's/IOO C's) are present in functional groups
which are not easily derivatized. Identification of the bulk of these function
alities was based upon analyses of liquids produced by thermal treatment of
RXOS at 425"C for I h, followed by THF Soxhlet extraction of the residue, which
afforded 86% recovery of the total organic matter as liquids and gas. 234 ,235
OIL SHALE Page 155
Figure 43. Distribution of organic functionalities in Rundle Ramsay Crossing
oil shale.
100 C's -, , MODEll -
77 -
C paraffinic
C=O 4 -------, 3.66: ------..,
C::::::C 3 3.14 : -_____ -1
16 1 -_~:.~_1J
C aromatic
9.2 0's1100 C's-_______ _ lMODEU
2.3
ETHERS
0.6
ALCOHOLS
1.4
METHYL
KETONES
2.2
ESTERS
2.3
CARBOXYLIC
ACIDS ,
AMIDES OA5
--2.24: -
-0.63: -
1 .4 1: -
2 -
2 --
-0 --~42: ----{
Ar-OH <0,0 1: 0.05:
2.3 N's/100 C's ________ _
o 5 R-PYRROlES
PYRROLES 0.3
HINDERED 1 .0
BASIC N
BASIC N 0.1,
AMIOES 0.45
:~_q9..lf!_J O~47: ------' --
-------, 0.31: ----'"I ~:.~~J -
---
OIL SHALE Page 156
High resolution MS and NMR analyses of the liquid products, and of fractions
separated from these products using HPLe techniques, indicated appreciable
amounts of furans, thiophenes, and pyrroles bearing a substitution group on
nitrogen. These results also provided the distribution functions of aromatic
condensation, alkyl chain length, and provided information about the average
pattern of aromatic substitution.
Paraffins (average C23 ) and olefins (average (20) were essentially all
linear, and each comprised about 8% of the HPLe saturates. X-ray analysis of
the RXO$-KC indicated the importance of paraffin-paraffin interactions (gamma
band), and the unimportance of aromatic stacking (no 002 band), in the
secondary organic structure. Also the observation of a sharp band
corresponding to a d-spacing of 4.2A, characteristic of waxy paraffins having
long, linear chains and the disappearance of this band upon heating to 90·C,
reinforces the importance of the paraffin interactions. Alpha- and beta
olefins in approximately equal amounts comprised the bulk of the olefins, with
smaller contributions from gamma- and trisubstituted-olefins. Additional
olefins were found as olefinic side chains on aromatics (20% of the liquids),
and almost all of these olefins were conjugated with the aromatic nucleus.
Condensation of the aromatics in the RXOS liquids averaged 4-6 rings. The
average aromatic nucleus bears one methyl group, remnant of the benzylic
cleavage of ethylene bridges, and one very long side chain about 30 carbons
long. Side chains in RXOS are very linear.
Among the oxygen-containing compounds found in the liquids, there were no
carboxylic acids, alkyl ethers or alcohols. Thermal treatment results in
decarboxylation of acids to form alkanes, cleavage and dehydration of ethers to
OIL SHALE Page 157
form alkanes and olefins and dehydration of alcohols to olefins. Small amounts
of esters were present. Long chain methyl ketones (average 20 C's) were
prevalent. Most of the observed furans were 2- and 3- ring benzologs. Three
ring furans existed as naphthalenofurans.
All of the sulfur in the liquids were present as derivatives of thiophene
benzologs. The 3-r;n9 thiophenes (mostly dibenzothiophenes) comprised about
half of the total, with contributions from 1-, 2-, and 4-rin9 derivatives.
Among the nitrogen derivatives found in the liquids were benzologs of
pyridine and pyrroles, nitriles and N,D-compounds. Indole derivatives
comprised 25% of these N-derivatives - the largest single fraction - followed
by pyridines, quinolines and carbazoles. Aliphatic nitriles and smaller
amounts of aromatic nitriles (1- and 2-ring) were found. The aliphatic
nitriles were essentially linear (average C40 ). These aliphatic nitriles were
probably produced during thermal treatment of RXOS (e.g., by dehydration of
primary amides), as no nitrile absorption was detected in FTIR spectra of RXOS.
Smaller amounts of di-nitrogen compound, and compounds containing both N- and
0- heteroatoms (e.g., 2-pyridone) were also found.
OIL SHALE Page 158
Representative Structural Model of Organic Material in Solid Rundle Ramsay
Crossing Oil Shale. The model with a formula weight of 30,000 daltons (and an
empirical formula of C1ooH160Nz.2SS0.6S09.2Z) was required to accommodate the
large range of heteroatom functionalities and long side chains present in RXOS
(Figure 44). It was not possible to accurately represent the range of compound
types in the bitumen because of its small amount (8.5%) and the need to
maintain a finite model size. Because carboxylic acids comprise the major part
of the bitumen, all the bitumen in the model was represented as carboxylic
acids. The values shown in the dashed boxes in Figure 41 illustrate the good
agreement of the model with the experimental values for the distributions of
hydrocarbon, oxygen and nitrogen functionalities.
~
I.
P'O o
, o
,\ /,
o
0>0
o
~
-.
'ji~' C~
;.0 Q)~\\"
6S1 a6.d
o
o
"
.... 0
! ~
\\ /, '"
o· , '.0
o
.., -. "" c ... '" ... ... V> ... ... C n ... C ... ., ~
" 0 c.
'" ~ 0 .... ... '" '" ... '" ~ '" ~ ... ., ... -. < '" 0 ... "" ., ~
n
" ., ... '" ... ., ~ • • 0
0 . ~ 0
'" c ~ c. ~
'" '" ., " ~ ., '<
'"' ... 0 ~ ~
~
"" 0
~
~
= ., ~
'" o~~·oro
~
31VHS 110
OIL SHALE Page 160
A comparison with models for Green River oil shale (GROS) kerogen serves
to illustrate some of the key features of the RXOS kerogen.
Aliphatic material is the most obvious feature of the RXOS model and RXOS
is more aliphatic than typical samples of GROS. Aliphatics in RXOS are longer
and more linear than those in GROS and are present both as alkylene bridges and
as alkyl side chains. Isoprenoid chains are distinctly less abundant. Some of
the polymethylene chains in RXOS are very long and proximately situated to give
appreciable regions that have the ordering characteristic of paraffin wax
crystals. Thus, secondary structure due to paraffin-paraffin interactions is
important in the RXOS kerogen. No such waxy paraffin interactions are apparent
in the GROS X-ray diffraction results. 246 Hydroaromatic rings also make an
important contribution to the RXOS aliphatics.
The average ring system in RXOS is only slightly larger than that in GROS,
but the size distribution in RXOS is appreciably broader; a significant number
of the larger 4-5 ring systems are present.
Bitumen content of RXOS is lower than that of GROS. The typical biomarker
compounds are present, but are much less abundant than in GROS. Most of the
bitumen is a mixture of carboxylic acids.
The nitrogen content of RXOS is lower than that of GROS (2.3 N's/IOO C's
vs. about 3 N's/IOO C's). Some of the nitrogen in RXOS is present as primary
amides and additional nitrogen is present as aliphatic amines. However, the
major part of the nitrogen in RXOS is present in very stable aromatic nitrogen
heterocycles, derivatives of pyridine and pyrrole.
The organic oxygen content in RXOS is much higher than that in GROS
(9 O's/IOO C's vs. 3 O's/IOO C's) and present in a variety of fUnctional group
OIL SHALE Page 161
types. Carboxylic acids are important and some of the carboxylic acids that
were observed in the kerogen concentrate are initially present in the rock as
amides. Some free carboxylic acids are evidently present in RXOS, as they are
observed in the extractable bitumen. How the balance of the carboxylic acids
are bound in the starting shale is not yet known.
This discussion serves to illustrate the power of the closely integrated
characetrization-modeling-reactivity approach to provide a model of molecular
structure in sufficient detail to be useful as a tool for guiding oil shale
research and interpreting experimental results.
6. Generalized Models for Types I~ II and III Kerogen
Workers at the Institut Francais du Petrole (IFP) have taken a distinctly
different approach to kerogen modeling. The studies discussed above were
directed toward modeling the structure of kerogen in a particular oil shale.
In contrast, the IFP workers have constructed generalized models representative
of the three types of kerogen and the of the asphaltenes from the corresponding
oils, as a function of maturity.254,255 Emphasis in this work was placed on
elucidating the chemistry of maturation for the three kerogen types -- in the
context of the IFP approach to the organic aspects of petroleum geochemistry.
The latest IFP models, those proposed by Behar and Vandenbroucke, repre
sent the kerogen at the beginning of diagenesis sensu stricto (excluding the
early stages of diagenesis which is probably dominated by microbial action), at
the beginning of catagenesis (start of the oil generation window) and at the
end of catageneSiS where late gas begins to be generated. Type I kerogen was
modeled only at he beginning of diagenesis and the end of catagenesis, while
OIL SHALE Page 162
models of Types II and III kerogen were constructed at each stage of maturation
(Figures 45-47). The corresponding asphaltenes were modeled only at the end of
diagenesis/beginning of catagenesis, where they become most abundant (Figure
48). For the asphaltenes, a molecular weight of 8000 amu was chosen.
These models are based on the results of elemental, infrared and 13C-NMR
analyses, pyrolysis (Rock Eval, artificial maturation) and electron microcopy
(fringe analysis) results. The functional group contents that were used in
assembling these models are summarized in Table 25.
The IFP models provide an interesting view of the structural relationships
between the three types of kerogen. However, many oil shales contain Type I
kerogen at a maturation state corresponding roughly to that IFP designates
Stage b. Thus, it is unfortUnate that the IFP workers chose to omit this model
from their publication.
Even though the IFP models are generalized and not intended to precisely
represent the structure of any particular kerogen, some comments about the Type
I models seem in order. First, the emphasis on large 4+ ring aromatics seems
overdone. There is general agreement that most ring systems are small in Type
I kerogen and few of the larger systems are fully aromatized until late in the
maturation sequence. Second, in all the IFP models aromatics are shown in
parallel sheets. There is a common misconception that this is a low-energy
arrangement. It is not. Even pure aromatics, especially those of 1-4 rings
likely to be important in oil shales, do not crystallize in such parallel
sheets, but rather in a "herringbone" arrangement that maximizes edge-to-face
interactions. The highly substituted aromatic units in kerogens are even less
OIL SHALE Page 163
Table 25. Heteroatom functional group contents of the IFP models. Each kerogen type was modeled with an initial collection of 1400-1500 carbons, some of which were lost as the kerogen matures. Thus, the numbers shown dQ not represent the functional group contents based on a constant number of carbon atoms. The contents are given at each modeled stage of maturation: Stage a corresponds to the beginning of dia· genesis, Stage b to the beginning of catagenesis and Stage c to the end of catagenesis. Asphaltenes were modeled only at Stage b, the beginning of catagenesis, where they become most abundant. (Data from Behar and Vandenbroucke, Reference 255.)
Heteroatom Functional Group Content of the Models Aliphatic Aromatic Ketone+ Aliphatic Thio· Thiol +
Kerogen Alcohol Phenol Acid Acid ~ Amide Quinone Ether Furan Pyridine Qhgng Disulfide Amine
Type I-a
-b
Type II-a
-b
-c
Type III-a
-b
-c
9
43
I
26
I
20
5
102
54
2
15
3
4
Asphaltene (at start of catagenesis)
Type I
Type I I
Type II I
2
14
3
I
7
1
48
6
2
14
45
27
16
14
4
5
5
31
6
9
I
48
10
44
2
52
33
9
80
46
19
13
15
14
I
2
2
12
6
7
17
37
5
2
5
I
4
16
8
14
18
9
I
6
6
I
2
22
11
2
5
6
6
1
2
2
I
2
2
26
2
1
OIL SHALE Page 164
Figure 45. Generalized IFP model of Type I kerogen.
(a) At the beginning of diagenesis (start of oil generation window)
OIL SHALE Page 165
Figure 45. Generalized IFP model of Type I kerogen.
(b) At the end of catagenesis (start of late gas generation)
(b)
OIL SHALE Page 166
Figure 46. Generalized IFP model of Type II kerogen.
(a) At the beginning of diagenesis
OIL SHALE Page 167
Figure 46. Generalized IFP model of Type II kerogen.
(b) At the beginning of catagenesis (start of oil generation) (c) At the end of catagenesis (start of late gas generation)
OIL SHALE Page 168
Figure 47. Generalized IFP model of Type III kerogen.
(a) At the beginning of diagenesis
OIL SHALE Page 169
Figure 47. Generalized IFP model of Type III kerogen.
(b) At the beginning of catagenesis (start of oil generation) (c) At the end of catagenesis (start of late gas generation)
OIL SHALE Page 170
Figure 48. IFP models of generalized asphaltenes.
(a) Type I (b) Type II
OIL SHALE Page 171
Figure 48. IFP models of generalized asphaltenes.
(cJ Type III
OIL SHALE Page 172
likely to arrange in parallel sheets. The implications of this kind of ordering
to coal structure (Type III kerogen) are important, as recently pointed out by
Larsen and his co-workers. 2 5 6 The imp 1 i cat ions for oil shale kerogen structure , should be similar, and no less important. Certainly, the X-ray diffraction
patterns of amorphous materials, such as kerogens, are properly interpreted in
terms of the Debye scattering formalism, rather than the diffraction formalism
of Bragg. This argument should also apply to electron diffraction. Perhaps a
re-interpretation using the correct formalism would yield a structure with more
physical meaning. Finally, the trends in functional group distributions do not
seem unreasonable, but it must be remembered that the details of the IFP models
reflect an average over many samples and were not intended to represent any
particular kerogen.
OIL SHALE PAGE 173
V. RECOVERY
A. OVERVIEW
Oil shale utilization will involve several different kinds of technologies.
In any case, the major areas will include mining, size reduction, retorting or
other means of recovering shale oil from the rock, disposal of the spent shale
and upgrading the shale oil into marketable products. A good overview of oil
shale technology through 1980 is provided by Oil Shale Processing Technology,
edited by V. D. Allred.'" The Synthetic Fuels Data Handbook by Baughman is
another extremely useful source of information on many aspects of oil shale
technology.154 It includes extensive discussions, as well as the voluminous
tables one would expect from its title. Chemical aspects of oil shale
retorting and shale oil upgrading are discussed in references 258-260. Other
good sources of information include the proceedings of the symposia held each
year at the Colorado School of Mines,261 papers of the symposia sponsored by
the Institute of Gas Technology262 and the Proceedings of the Australian Oil
Shale Workshops.263 Prien reviewed progress through the mid-1970's.6 The many
papers of workers at the U.S. Department of Energy (nee ERDA, nee U.S. Bureau
of Mines) and the U.S. Geological Survey are also good sources of information
about specific aspects of oil shale technology, as well as oil shale science in
general. The sources listed above do contain some information about foreign
oil shales, but are primarily concerned with technology applicable to the Green
River Formation in the U.S. Much information, unfortunately now dated, about
other shales is contained in the proceedings of a U.N. conference at Tallinn in
1968. Technology used in the Chinese (PRC) oil shale industry has been
reviewed.','·4,'.' A bibliography listing over 5000 oil shale publications
OIL SHALE PAGE 174
(science and technology) up to 1977, and a supplement also containing well over
5000 references to 1981, are available from the U.S. Department of
Energy.266,267
Estimates of required selling prices for synfuels have. through the years,
been very badly off the mark. However, estimating the potential for lowering
shale oil costs by improving the economics of one or another of the various
technologies can be useful in assigning priorities for research and development
efforts. Cost breakdowns have been estimated for shale oil production from
Green River oil shale and Estonian kukersite (Tables 26_27).268,269
OIL SHALE PAGE 175
Table 26. Estimated cost breakdowns for shale oil production from Green River
oil shale. (Data from Reference 268.)
Capital Expense. % Operating Expense. % Scalability
Mining 8 35 Marginal
Shale preparation 10 15 Margi na 1
Preheat-Retorting 7 5 Very good
Combustion 15 12 Very good
Power/heat systems 20 12 Medium
Product upgrading 25 16 Very good
Infrastructure 15 5 Good
Table 27. Estimated cost breakdown for production of shale oil and chemical
by-products from Estonian kukersite. (Data from Reference 269.)
Production Costs, % of 1959 total 1959 1977
Shale feed (Incl. mining. prep. ) 64.5 58.6
Util ities: Electric 5.1 3.9
Heat 12.0 2.2
Water 0.6 0.9
Labor 6.3 2.3
Fixed costs 11.4 6.5
By-product credit -19.6 -24.4
Net production cost 80.4 74.4
OIL SHALE PAGE 176
It is interesting to note how similar these estimates are, even though
their cost bases are obviously different, they differ by years in time (1980
V$. 1987) and were made for shales half a world apart. Yet. in each case the
costs of getting the shale out of the ground and ready to process represent
fully one-half the total cost. It is also interesting to note that substantial
reductions in the cost of the Estonian shale oil were achieved, even though
this industry is about 50 years old. hence should be relatively mature.
B. MINING TECHNOLOGY
Very thick seams of well-consolidated oil shale are characteristic of the
Green River Formation. Around the basin rims (especially of the Piceance
Basin), outcrops are numerous, while considerable overburden covers the rich
seams near the basin center. As a result, both underground mining and surface
(open-pit) techniques may have economic advantages in different locations. In
some locations. in situ methods may have advantages, but even in these cases
some mining will probably be required to provide a void into which rubble can
be blasted. Several general reviews have been made of the mining technology
applicable to oil shale.27o-275
1. Underground Mining
Underground mining was used at the Anvil Points Oil Shale Research Facility
(APF) that was operated from 1944 to 1956 by the U.S. Bureau of Mines,276 then
by others until 1984,277~281 when it was decommissioned. 283 Underground mining
was also used successfully by Union Oil, Mobil Oil and Colony Development, and
OIL SHALE PAGE 177
was planned for Exxon's Colony Shale Oil Project {now terminated).284,285 In
these cases, the mine was of a room-and-pillar design (Figure 49).
'" -
Figure 49. One level of a room·and-pillar oil shale mine. (After East and
Gardner, Reference 276.)
The mine at APF was begun as a three-level mine, but in later operations
only two levels were mined. Z76 Nevertheless, the mining at APF provided a
wealth of information about the safe design, operation and maintenance of an
underground oil shale mine in the unique environment of the Green River
Formation. Variations on this basic design can be tailored to mine the rich
OIL SHALE PAGE 178
strata, while selectively leaving the lean one5. 286 Such designs, however,
leave much of the shale in place, hence are not suited for zones that are both
thick and uniformly rich.
Bulk underground mining methods, such as sublevel stoping with full
subsidence or block caving methods, can provide more complete recovery of rich
oil shale from thick lones. 273 ,287 However, spent shale cannot easily be
disposed of in the mined void. A stoping method with spent shale backfill may
be a compromise that is both reasonably efficient and environmentally
acceptable. longwal1 mining, a method widely used in underground coal mining,
is another alternative. longwall methods have been used for oil shale mining
in the USSR288 and its use in the U.S. has been contemplated. 289 Given the high
cost of mining, it is clear that, even incremental, improvements in oil shale
mining technology could have a major impact on the economics of a proposed oil
shale project. However. it must be realized that mine design is very much site
specific and an improvement applicable at one site may be inappropriate for
another. Moreover, safety considerations dictate a conservative policy with
respect to the amount of rock that must be left in place. Perhaps improvements
in subsurface imaging can be made to enable more precise three-dimensional
modeling of subsurface features using computer techniques. This will require a
much better fundamental understanding of wave propagation in inhomogeneous
media. However the benefits should be both .lowered cost of mine design and
more efficient deSigns that significantly increase the amount of rich oil shale
that can be removed without compromiSing mine safety.
OIL SHALE PAGE 179
From 1981 to 1984, an extensive oil shale fragmentation research program
was conducted at Anvil Points by a consortium comprised of Cities Service,
Getty Oil, Mobil Research & Development, Phillips Petroleum, Sohio Shale and
SunaCD Development. Science Applications, Inc. (SAl) managed the program and
provided technical direction. Los Alamos National Laboratory. and in later
stages Sandia National Laboratory, participated with SAl and the consortium and
shared the experimental work. This program included a wide range of tests,
ranging from single level/single borehole tests to obtain basic fragmentation
data, to multiple level/multiple borehole tests to explore fragmentation as a
function of explosive type, charge placement and detonation sequencing.28o-282
Some of the heavily-instrumented tests carried out late in the program provided
results that were used to verify and refine computer programs that simulate
with remarkable preCision, the behavior of the Green River oil shale under
explosive stress. 282 These results should prove invaluable in designing new,
more efficient mines and in formulating efficient operating procedures for
these mines of the future.
2. Surface Mining
At Fushun, in Manchuria (PRe), there is a very large open-pit mine where
450 feet of low-grade oil shale (15 gal/ton) overlies one of the world's
thickest coal deposits. Oil shale processing has accompanied coal production
since the Japanese initiated large-scale operations at Fushun in 1929. 7,264,265
Surface (open-pit) mining is also practiced on a large scale in the USSR, espe
cially in Estonia.290-292 Surface mining was planned for the (now deferred)
Rundle Project (Queensland, Australia) by a consortium comprised of Esso
OIL SHALE PAGE 180
Exploration and Production Australia, Central Pacific Minerals and Southern
Pacific Petroleum. 285 ,293 Surface {strip} mining has also been envisioned for
the Toolebuc shale, Australia's largest oil shale resource, especially along
the St. Elmo Structure where the shale is 7-14 m thick below a nearly barren
oxidized zone about 18 m thick.s
Surface mining of Green River oil shale has never been practiced on a
large scale, though about 15% of the total reserve is potentially recoverable
by surface mining methods. A detailed engineering economics study was carried
out to explore the potential of open-pit mining of Green River oil shale.294.295
The potential environmental impact of such a large-scale surface mining
operation in the Piceance Basin was examined in a recent U.S. EPA study.296
Surface mining of Eastern U.S. (Indiana, Kentucky) shales was evaluated as one
part of an effort to develop plans for Devonian oil shale utilization. 297 ,298
Surface mining is used extensively in ore recovery and for mining coals,
especially low-rank coals. It is anticipated that technology developed for
these applications will be adaptable to oil shale mining.
c. SIZE REDUCTION
In situ retorting will not require size reduction past that obtained by
blasting to rubblize the shale, however size reduction by crushing and grinding
will be required for above-ground processing. Moreover, size reduction is
expensive and becomes rapidly more so as the target particle size decreases
below about one-half inch. Nevertheless, closely controlled size reduction is
a necessary preliminary to the use of heavy medium cycloning and other of the
OIL SHALE PAGE 181
newer, more efficient methods for oil shale beneficiation. In addition, most
current above-ground retorts have severe particle size restrictions, especially
with respect to fines.
A detailed study of size reduction of Green River oil shale was made by
Salotti and Oatta. 299 Three types of crushers (gyratory, impact, roll) were
studied, using 200-ton samples collected from the R-6 (20.7 gal/ton) and R·5
(24.9 gal/ton) zones of the Rio Blanco Mine (Tract C-a). The authors caution
that extrapolation of the results to leaner shales (-15 gal/ton) should be done
with caution and that extrapolation to appreciably richer shales should be made
only with trepidation. Major findings of this study are summarized below:
1. Power consumption for a similarly-sized feed and product is roughly the
same for all three types of crushers.
2. Variations in feed grade (20-25 gal/ton) have little or no effect on the
performance of the three crusher types for product having a top size of
3/8-inch or larger.
But,
3. When a crusher is producing 3lB-inch or finer top size product, the richer
oil shale tends to have a significantly finer size distribution.
4. Crusher throughput was about 70% of rated capacity.
5. Relative ability to process larger pieces (6-inch x 0) was in the order:
gyratory> impact> roll.
6. Relative percentage of recirculating load present as -3/4-inch material is
in the order: roll> impact> gyratory.
7. Relative production of oversize material is in the order: roll> impact>
gyratory.
OIL SHALE PAGE 182
8. Grinding of these relatively rich shales was roughly equivalent to grinding
a moderately hard limestone, and wear during oil shale grinding is antici
pated to be similar to that suffered during grinding such a limestone.
A similar study of size reduction was made as part of a study of the
beneficiation of an Eastern U.S. Devonian oil shale. 30o
OIL SHALE PAGE 183
V I • RETORTING
Retorting is the process of heating oil shale in order to recover the
organic material as shale oil + gas (less commonly, just as gas). To get
reasonable rates of product recovery. temperatures of 4000-600·C (750 0 -1100*F)
are generally used. Therefore. to avoid unwanted combustion a retort in its
simplest form is a vessel in which the shale can be heated without exposure to
air and from which the product gases and vapors can escape to a collector.
Retorts used in early shale oil processes were just that. Modern retorts are
usually tailored to meet the needs of an integrated oil shale process, hence
are somewhat more complicated. Therefore, this section outlines not only the
major features of the different retorts, but also of the corresponding integ
rated processes that have been developed/improved during the past two decades.
Sha 1 e oil recovery can be carried out above-ground or underground (in-situ).
In-situ processing is attractive because the requirements for mining, hauling,
crushing and grinding the oil shale rock are eliminated or greatly reduced.
Thus, in situ retorting offers the potential for corresponding savings in both
capital and operating expenses. Above-ground retorting, on the other hand,
generally affords better control of retorting conditions that can minimize heat
loss due to carbonate decomposition and lead to a better yield of higher
quality products. Novel processes, using supercritical solvent extraction or
bioleaching to recover shale oil, are still in the early experimental stages,
but appear to have the potential for use either above- or underground. Most
work to date has been on above-ground retorting.
Above-ground retorting processes fall into three broad classes depending
on whether the process heat is generated internally (direct heated retort),
OIL SHALE PAGE 184
externally (indirect heated retort), or have the potential for both. Currently,
only the Kiviter retort used in the USSR for retorting of Estonian kukersite
simultaneously derives major fractions of its process heat from both internal
and external sources. Grouped by their method of heat generation, retorting
technologies include the following:
Above-ground Retorting Process
• Direct Heated
N-T-U
Gas Combustion
• Indirect Heated
Lurgi-Ruhrgas
TOSCO-II
Underground Retorting Process
OXY Modified In-Situ Retort Process
(Occidental Oil Shale, Inc.)
Horizontal Modified In-Situ (LETC)
Retorting Process
LLNL RISE (Rubble In-Situ Extraction)
Laramie True In-Situ Retorting Process
Multi-Mineral In-Situ (MIS) Process
Petrosix (Brazil) Geo-kinetics In-Situ Project
Exxon Shale Retort (ESR) Process Equity Oil BX In-situ
Shell Shale Retorting Process
HYTORT (IGT Hydrogen Atmosphere)
Superior Circular Grate
Allis-Chalmers Roller Grate
Galoter (USSR)
• Combination
Kiviter (USSR)
Union Oil
Paraho
OIL SHALE PAGE 185
A. ABOVE-GROUND RETORTING PROCESSES
Although interest in directly heated retorting has waned, it is useful to
begin the discussion of retorting by examining two directly heated retorts of
relatively simple design.
I. N-T-U Process
Invented in 1923 by Dundas and Howes, The N-T-U retort takes its name from
the N-T-U Company (the letters stand for Nevada-Texas-Utah) which took the lead
in its early development. 301 ,302 The N-T-U batch retort is relatively simple
and inexpensive to construct and proved to be durable in operation (Figure 50).
During the 1920's, the N-T-U Company constructed a 40-ton retort near
Santa Maria, California and the US Bureau of Mines (USBM) constructed a small
N-T-U retort at its Anvil Points facility. This first phase ended about 1930,
but interest was revived by the fuel demands of WWII. In Australia, N-T-U
retorts were used by Lithgow Oil Pty. Ltd. at their facility at Mangaroo.'"
Three 35-ton retorts were constructed, but seldom operated simultaneously due
to lack of feed shale. Nevertheless, nearly 2 million gallons of liquids were
produced during 1944-45. Interest was also revived in the U.S. Upon passage
of The Synthetic Liquid Fuels Act of AprilS, 1944, the USBM constructed two
identical 40-ton N-T-U retorts at Anvil Points to provide design data and
quantities of shale oil sufficient for refining studies. By the time these
retorts were dismantled in 1951, each retort had operated nearly 7000 hours to
process about 18,000 tons of oil shale and produce 6,000 barrels of shale oil.
Inspection showed that the retorts were still in good condition. This early
work was reviewed in 1951 by Cattell, Guthrie and Boyd."4
OIL SHALE PAGE 186
Figure 50. Schematic of the N-T-U Retort. Heat from the downward-moving flame
front drives the retorted oil before it. Gas produced by retorting
can be recycled to provide additional heat. (after Cattell,
Guthrie and Scramm, Reference 304.)
HINGED BOTTOM
OIL SHALE
~~ ~ _COMeUSTIO.
. AIR ". '
•. SPENT, -:. ". SHALE.:',
, .~ ....... .;. '';. J, .
• :COM'euSTION: ~_" ZONE .:. \" . .: ..', .....
. . . . . , .
", ; ,- ... . " .. ~: ': :.' ' .. . ' .... . " ' ..:. : RAW '; ~. :';H~'~'E< . -~ . . . '. " ","
/ . . /
/ .. / I . j
PYROLYSIS PROOUCTS ANO
PRODUCTS OF COM8USTtON
OIL SHALE PAGE 187
In the 1960's, a 10-ton N-T-U retort was constructed at USBM's Laramie
Energy Technology Center (LETC, later the Laramie Energy Research Center (LERC)
of the Energy Research and Development Administration (ERDA) and now the
Western Research Institute). This retort, and the ISO-ton version added in
1968, have been used extensively to study many of the parameters important to
in-situ retorting. The 150-ton retort, located just north of laramie, Wyoming.
has an internal diameter of just over 6 feet and height of 45 feet (Figure 51).
To simulate an in-situ retort, the 150-ton unit is generally charged with
ungraded shale selected to simulate the wide range of sizes obtained by blasting
to produce a rubblized shale bed in an underground retort. Individual shale
blocks weighing up to 10,000 lb. have been retorted successfully. Results from
studies of the retorting of random-sized shale charges have been summarized by
Harack t et al.,305.306 Docktor. 307 and Ruark.308 Oil recovery from shale in
the range of 1/2 to 3-1/2 inches ranged from 80% of Fischer Assay for 30 gal/ton
grade to 87.5% of Fischer Assay for 50 gal/ton grade. Even very large pieces
of shale could be retorted. Harack, et a1. described the results of one run
where 20% of the feed was larger than 20 inches, 10% was smaller than 1 inch
and one very large piece of shale weighed 7500 pounds (Table 28).30'
The N-T-U retorts served well as research tools, but as batch units they
were not well suited for commercial use. This led to a search for a more
efficient and continuous retorting process.
OIL SHALE PAGE 188
Figure 51. The ISO-ton N-T-U retort at the Western Research Institute, Laramie,
Wyoming.
Sit, ,I". tot.h ... ~mt .uh.ne.
Elliftkal .........
(After Cattell, Guthrie and Schramm, Reference 304.)
\ '""''''''' ktn ..
~"',,, ... uelline4 with c .. I'~1o tl>giw 6-.~
-;;;,~;;;; .. ,~;' ~;;;"."ftd !eliot _., W on
-{ 11~~-, .. " ...... ,. jIIIdiIIg- .......... - -. ... ,
/""" ......... --.. .
•• ll"'~'" Ind uUs 10 olluin cWinotoIn .. I "'~I$o' •• :. SIolH", oIeck fo< l'isdo.-.. ol "''''t _'''i,l
OIL SHALE PAGE 189
Table 28. Results summary for a run in the ISO-ton N-T-U retort at the Western
Research Institute. This run included a single block of shale that
weighed 7500 pounds. (Data of Haraek, et al. Reference 305.)
Length of run days 12.25 Operating conditions:
Shale charge tons 178.67 Retort pressure psig 3.0 Air rate (dry sefrn 135
Do sef /ton shale 13,300 Avg. air temp. into retort OF 28 Recycle gas rate (dry) sefm 67
Do sef/ton shale 6,600 Avg. recycle gas temp. into retort of 43 Oxygen content of retorting gas pet 14.5 Space velocity ft' gas/fV bed/min 1. 94
Stack gas rate (dry) sefm 177 Do sef/ton shale 17,400
Gas produced in retort (dry) sefm 42 00 sef/ton shale 4,100
Max retort differential press in. H2 O 0.6 Avg. ambient temp. OF 26 Avg. retorting advance rate in/hr 1. 75 Max bed temp OF 1,600 Bed compaction pet of initial height 5.6
Oil shale properties: Fischer Assay gal/ton 25.4 Water content gal/ton 2.9 Bulk density lb/ft' 80.0 Gross heating value BTU/1 b 2,267
Recovery: Oil gal 2,830 Spent shale tons 125.86 Oil recovery Vol pet of Fischer assay 62.2
Oil properties: Gravity API 25.2 Pour point OF 70 Viscosity SUS at 100°F 79 Hydrogen wt pet 11. 76 Nitrogen wt pet 1.77 Sulfur wt pet 0.76 Carbon wt pet 84.58 Ash wt pet 0.01 Gross heating value BTU/1 b 18,660
Spent shale properties: Fischer assay gal/ton 0 Gross heating value BTU/1b 117
OIL SHALE PAGE 190
2. Gas Combustion Process
Design objectives for the Gas Combustion Process grew out of the USBM work
described above. These included gravity flow to minimize mechanical complexity,
heating of the raw shale by hot gas, and generation of this hot gas by burning
the residual carbon on retorted shale within the retort vessel.
Work began in 1949 on the first version, known as the dual-flow retort. A
second and much improved version, the countercurrent retort soon followed. By
flowing combustion gases up and shale down, oil recoveries above 90% of Fischer
Assay were obtained from 25-20 gal/ton shale at throughputs of 200 lb./hr/ft'.
However, having the combustion zone at the bottom of the retort led to the
spent shale being discharged at high temperature, thereby wasting valuable heat.
Also, handling the hot shale was a problem. The Gas Combustion Process was
developed to alleviate these problems (Figure 52).309
The Gas Combustion Retort, itself, is a marvel of simplicity. No internal
baffles divide the zones and the flow of shale from the retort is controlled by
a movable grate at the bottom. Shale fed into the top of the retort through a
rotating lock is heated by the hot product vapors. This serves to cool the oil
vapors past the dew point into a mist that is swept out into the collector by
the upflowing gases. The preheated shale then moves down into the retorting
zone where, upon further heating, kerogen decomposition yields the oil and gas
products and a carbonaceous residue that adheres to the spent shale. Next the
shale moves into the hottest part of the system where air is introduced for
combustion of the residue and of the hydrocarbons in the recycle gas. In the
lowest part of the retort, heat from the combusted shale is transferred to the
recycled gas as it flows upward. Finally, the cooled, combusted shale is dis
charged from the bottom of the retort:
Figure 52. In the Gas Combustion Retort, the spent shale is cooled by recycled product gas before being
discharged. Subsequent combustion of the hydrocarbons in the recycled gas provides additional
process heat. (After Matlick, et al., Reference 309.)
SHALE
OIL SHALE I-CRUSHING
AND MINE SCREEN
~ GAS
SIZING
-f FINEt DISCARD " SEAL
PRODUCT DUST AND rid 6 ~ GAS \.::1- OIL MIST ~ TYPICAL TEMPERATURE PROFILE
RECYCLE REM,~{A L SHALE I, GAS PREHEATING
BLOWER SHALE OIL ZONE
1------- - --SHALE
R~~TIN~J2NJ;. - --CO BUST!
AIR , ~'" f-_ ZONE -- ----AIR BLOWER
DILUTION GAS SPENT SHALE V COOLING ZONE ... 17 -
~'~--/
500 1000 1500 COOL RECYCLE GAS TEMPERATURE OF SHALE of
GAS SEAL
SPENTtSHALE SOLIDS
OIL SHALE PAGE 192
Table 29. General data SUlJII8ry. USSM evaluation ru'lS on the 150 tpd Gas Coobustion Retort.
Test nurber lli1:1l 26(1-2) 26(3-5 ) 27(1-3) lli1.::il 28(5-6)
length of test hours 120 48 72 72 96 48
Rates and quantities: Shale sile inches 3/S-3 1·2 1·2 4·2 1·2 1·2 Bed height ft and in 9,11 9,11 9,11 9,11 7,2 7,2 Raw shale rate lb/(hr) (sq tt) 299 222 299 350 299 300 Air rate std cu tt/ton shale 3,940 4,230 3,910 3,840 4,010 4,290
Oitutlon gas rate std cu ftlton shale 2,&0 3,800 2,950 3,140 3,100 4,040
Recycle gas rate std cu tt/ton shale 13,340 12,400 12,650 12,660 12,500 10,260
TenperatUf'"es:
Product outlet 'F 162 142 141 143 141 126 Retorted shale out 'F 376 348 356 345 378 447 Raw shale in 'F 40 34 32 30 28 25 Recycle gas 'F 241 247 250 246 224 213 01 lution gas 'F 83 90 92 92 79 86
Air 'F 128 129 131 144 110 98
Yields: Oil vol pet/Fischer assay 82.8 92.3 86.2 86.7 85,1 86.1 G .. std cu ft/ton shale 6,040 6,440 6,000 6,020 6,400 6,090 Retorted shale wt pet of raw shale 81.8 82.9 82.3 82.1 83.3 83.0
llq.,!id water lbltOf'l shale 0.2 5.0 0.9 1.1 4,9 11.2
Miscellaneous:
Retort preS$Uf'e df-op in "?lIft bed 0.90 0.37 0.73 1.02 0.58 0.45 Carbonate deeonposition wt pet 24.9 24.1 26.9 25.6 23.3 23.6
OIL SHALE PAGE 193
Table 30. Material Balances, USBM Evaluation Runs on the 150-Ton Gas COI'ltlustion Retort.
Test nunber ill1.:.2l 26(1-2) 26(3-5) 27(1-3) 28(1-4 ) 28(5-6)
Materials in: Raw shale .b 2,000 2,000 2,000 2,000 2,000 2,000
Recycle gas 'b 986 933 952 946 927 m Oi lution gas 'b 211 286 222 235 230 304 Ai, .b -1!IL 324 -l22.. --l2L -lQL ---ll2..
Total materials in .b 3,499 3,543 3,473 3,475 3,564 3,406
Materials out: Retorted shale 'b 1,636 1,658 1,646 1,642 1,666 1,660 Product 01 1 'b 185 185 178 183 189 184 Offgas 'b 1,644 1,705 1,625 1,631 1,609 1,536
Water in oil .b ---lliL __ 5_ __ 1_ __ 1 _ __ 5_ 11 Total materials out 'b 3,405 3,553 3,450 3,457 3,469 3, 391
Recowry percent 99.0 100.3 99.3 99.5 97.3 99.6
OIL SHALE PAGE 194
Three gas combustion retorts were constructed by USBM at Anvil Points:
6-ton/day and 2S-ton/day pilot units and a ISO-ton/day unit designed to furnish
engineering design data for a commercial unit. Results obtained during this
period were summarized by Matzick and Oannenberg. 310
Several problems of the Gas Combustion Retort were not solved by the USBM.
Oil yields were in the range of 85-90% of Fischer Assay, some 5-10% lower than
desired. The control of oil mist and the design of air-gas distributors were
persistent problems. Finally, the inability to efficiently process shale fines
(smaller than 1/4-inch) was felt a definite drawback.
The first stage in development of the Gas Combustion Process ended in 1955
when work was halted by USBM. Stage II began in 1961 with the passage of
Public Law 87-796 empowering the Secretary of the Interior to lease the
facility, in order to encourage fUrther development of oil shale technology.
After evaluating several proposals, the Interior Department leased the Anvil
Points facility to the Colorado School of Mines Research Foundation (CSMRF).
Under the terms of the lease, CSMRF became lessor, provided administrative and
logistic support, and made the facility available to a consortium that
eventually jncluded Mobil Oil (project manager), Humble Oil & Refining, Pan
American Petroleum Corp., Sinclair Research. Inc., Continental Oil Co. and
Phillips Petroleum Co. Under this agreement, the facility was reactivated in
1964 and was operated for about one year in 1966-67. During this time. nearly
300 runs were made, mostly in the 25- and ISO-ton/day units. Stage II efforts
have been summarized by Ruark, et a1.277.278 and by Lawson, et a1.311
Improvements made in Stage II solved many problems and throughput was just
about doubled to 500 lbs/hr/ft'. However, oil yields remained a disappointing
OIL SHALE PAGE 195
85-90% of Fischer Assay despite efforts to achieve improvement. Nevertheless,
progress was made and many of the results were incorporated into the Paraho
project. No work was carried out on the Gas Combustion Process after 1967, but
the Paraho Process in its directly heated mode is nearly identical, so Paraho
results can be used as a guide to the potential of the Gas Combustion Process.
3. lurgi-Ruhrgas Process
The lurgi-Ruhrgas oil shale process (LR process) grew out of an earlier
Lurgi process for making high-BTU gas from coal fines. Initially, a solid heat
carrier (e.g. sand) was mixed with the coal fines, but later the hot coal char
product was recycled to supply the needed heat. The process was tested on a
variety of coals, and was used commercially in Germany, England, Yugoslavia,
Argentina and Japan. Application to oil shale was a logical extension.
In the lR shale retort, hot solids recycled from a combustor and the raw
shale are fed (ratio 6-8:1) to a screw type mixer (Figure 53).312 Retortin9
takes place in the mixer and in the surge bin (accumulator) that follows. Gas
solid separation takes place in the surge bin and the following cyclone. Most
of the solids are diverted to the lift pipe combustor where burning of the
residual carbon raises the temperature to about 1200·F. The hot solids are
separated from the gases and returned to the retort, which is maintained at a
an optimum retorting temperature in the neighborhood of IOOO·F.
OIL SHALE PAGE 196
Figure 53. Simplified flow diagram of the Lurgi-Ruhrgas oil shale retorting
process shows that the screw mixer section where retorting takes
place is actually a very small part of the system. (After Rammler.
Reference 312.)
lurgi - Ruhrgas Oil Shale Retort
Gas/Solids ·----,l ____ rJ::l ---1 Separation
Bin
Lift Pipe
Cyclone
Oil Shale
Screw Mixer
Air
Surge Bin
Waste Heat 1-___ _
Recovery Flue Gas
Spen° Shale
Gaseous & 011 & Gas Recovery Liquid
Products
Cyclone
OIL SHALE PAGE 197
The use of the mixer provides rapid heat transfer and allows very short
residence times for the product vapors in the hot zone to minimize unwanted
cracking. The forced mixing also reduces the tendency for rich shales to
result, the critical materials problems are confined to a small, manageable
section of the overall facility. Also. it is important that no part of the LR
retorting system operates at high pressure; maximum pressures are a few inches
of water.
The product collector section is designed in stages to provide two, and
preferably three or more, product fractions. This is not to give fractions
defined by boiling range, but rather to avoid oil/water emulsions and for
improved dust control. The first section is an underflow scrubber, where the
product stream is concurrently cooled and scrubbed by injection of the aqueous
product and heavy oil. Fine dust not collected in the cyclone is concentrated
in the heavy oil and the subsequent fractions are essentially dust-free.
For commercial use, provision would be made to recover process heat from
spent shale and heat exchangers would be used to recover additional heat.'"
Also, two or three cyclones may be staged for more complete dust removal from
the product vapors; and scrubbing would be used in all stages to give efficient
cooling and condensation of the product vapors. Electrostatic precipitators
would probably be added to reduce particulate emissions in the flue gas from
the lift-pipe combustor.
OIL SHALE PAGE 198
One of the important features of the LR process is its ability to process
shale fines; the entire oil shale resource can be utilized in a single type of
reactor (Table 31). Good material balances were obtained in pilot plant
operations and oil yields were high from a variety of oil shales; yields were
commonly above 95% of Fischer Assay (Table 32). In favorable cases oil yields
went as high as 110% of Fischer Assay The product oil is 85% - 90% volatile,
though nitrogen and sulfur levels are high, and pour pOints are desirably low
(Table 33). Dust in the heavy oil is troublesome, but lurgi has a patented
process for oil dedusting. Alternatively, the dust-laden heavy oil can be
recycled to the mixer. Thus, The lR retort solved two of the most persistent
problems of the gas combustion retort, namely low oil yields and inability to
process shale fines (i.e. the entire oil shale resource).
Gas products from the lR process are not diluted with combustion gas, so
BTU content is high and the gas should be suitable for reforming to produce the
hydrogen needed in product upgrading.
The aqueous product is only slightly basic and should not require special
materials of construction. It does contain phenols and ammonia that may be
recovered by steam stripping, however further treatment (e.g. biological oxida
tion) may be required before the process water is environmentally acceptable
for mOistening the spent shale.
After removal of the kerogen by retorting and decomposition of some of the
carbonates during combustion, the spent shale is friable (Table 34). Shear
applied in the mixer retort does reduce particle size. On the one hand, these
characteristics make the combusted shale a very good SOx acceptor, while on the
other. the fine particle size means that special care must be exercised in
OIL SHALE PAGE 199
Table 31. Properties of Lurgi-Ruhrgas oil shale pilot plant feeds_ Abil i ty to process fines means the entire oil shale resource can be processed. (Data of Rammler, Reference 312).
TYPE OF SHALE A C D E F
""' __ L ___
Residue " 83.4 84.9 84.4 87.5 91.0 79.7 Gas and 1 ass " 3.8 3.9 4.4 2.3 2 .1 4 .1
Ultimate Analysis Carbonate CO, wt % 18.7 16.4 13.7 9.1 19.6 14.6 Organic Carbon " 15.7 12.9 10.2 7.3 6.7 14.3 Hydrogen " 2.0 1.9 1.3 1.1 1.8 Nitrogen " 0.5 0.3 0.2 0.2 0.6 Total Sulphur " 0.9 3.4 1.0 1.6 2.7
Gross Calorific Value kJ/kg 7400 6150 4770 3520 2830 5690
Grindability Index 30 76 49 90 69 (Hardgrove)
Size Analysis" +5 mm wt % 2.8 0.0 0.0 0.0 0.1 0.1
4 -5 mm " 25.2 0.0 2.7 0.0 2.1 1.4 3 -4 mm " 25.2 0.1 12.2 0.2 7.6 10.0 2 -3 mm " 17.1 16.7 17.5 14.0 13.8 58.2 1 -2 mm 18.3 26.8 23.2 25.8 19.5 29.1 0.8 -I mm 14.0 11.4 10.2 9.7 7.7 0.2 0.5 -0.8 mm 14.0 9.2 7.8 8.5 7.3 0.1 0.315 -0.5 mm 10.4 13.9 8.4 10.0 9.2 0.1 0.2 -0.315 mm 10.4 9.0 5.5 7.6 7.8 0.1 0.1 -0.2 mm 5.2 6.0 5.1 8.4 8.6 0.1 0.063 -0.1 mm 7.0 3.1 3.1 5.2 6.0 0.2
-0.063 mm 7.0 3.8 4.3 10.6 10.4 0.4
Median Grain Size mm 2.05 1.09 1.54 0.97 1.20 2.31
Bulk weight kg/l 1.15 1.25 1.09 1.23 1.18 1.00
Ash Analysis: CaO wt % 24.0 11.5 31.3 25.9 Si02 " 38.4 46.6 29.5 39.1 AI,o. " 13.8 15.5 10.0 11.1 Fe,o. " 6.6 7.5 7.0 4.4 S0, " 9.7 2.6 4.0 4.6 MgO " 2.9 3.4 2.6 5.5 K,O " 2.1 2.0 1.6 1.4 Na,O " 0.7 3.1 0.3 0.2 Others " 1.8 7.8 13.7 7.8
• Crushing to 0-4 mm used in pilot plant work is not typical for larger units.
Oil SHALE PAGE 200
Table 32. Material balances were good in LR oil shale pilot plant operations. Oil yields were high, generally over 95% of Fischer Assay. (Data of Rammler. Reference 312.)
TYPE OF SHALE _B _ _ C _ _ F_
Feed Quantity kg 1000 1000 1000 Moisture wt % 0.4 3.5 5.9 Oil (Fischer Assay) wt % 9.4 5.6 8.0
Output Circulated material discharged kg 266.7 464.3 510.5 Dust in flue gas kg 490.5 330.9 186.2 Oust in make gas kg 9.7 13.5 55.8 Heavy in oil (dust free) kg 62.9 28.8 50.6 Middle oil kg 26.5 18.8 23.5 Gas naphtha kg 10.7 5.4 5.5 Gas liquor kg 24.5 32.9 85.9 Make gas (dry, free of C,.) kg 29.4 27.1 32.8 CO2 from carbonate decomposition in flue gas kg 42.7 42.9 II.I Fixed C burned kg 36.4 35.4 38.1
Totals kg 1000.0 1000.0 1000.0
Yield of oil (% of Fischer assay) 106.5 94.6 99.5
OIL SHALE PAGE 201
Table 33. Inspections of dust-free total oils from the LR retort.· Pour
points of the lR oils are low, but only 85% - 90% is volatile
and the nitrogen and sul fur 1 eve 1 s are high. (Data of Ramml er,
Reference 312.)
T~Qe of Shale _ E_ _F _ _H_
Density (20'C) glcm' 0.985 0.971 0.882 Viscosity (20'C) JO-6""/s 21.5 11.5 6.3
(60'C) JO- 6m2/s 5.1 3.7 2.3 Pour Point 'C -20 -23 +16 Flash Point 'C 103 n.d. +23 Conrad son C wt% 5.3 5.6 2.1 Bromine Number gl100 9 56 n.d. 56.5 H/C Atomic Ratio 1.37 1.49 1.57 Nitrogen wt% 0.89 1.47 0.84 Sulfur wt% 3.25 6.47 0.70 Gross Calorific Value kJ/kg 40,900 40,195 43,340
Initial Boiling Point 'C 130 78 69 5 vol% 'C 175 148 142
10 v01% 'C 200 167 170 20 v01% 'C 232 195 201 30 vol% 'C 260 227 228 40 v01% 'C 290 268 245 50 vol% 'C 318 308 276 60 vol% 'C 340 341 310 70 vol% 'C 370 377 347 80 vol% 'C 405 423 391 90 vol% 'C 450 (88.0%) 450 (85.4%) 440
Loss v01% wt% 1.0 0.9 0.4
* Does not include gas naphtha.
OIL SHALE PAGE 202
above-ground disposal. This should not normally be a problem, since most of
the spent shale would be returned to the mine. Moreover, when moistened, the
spent shale acts like a cement and gives a hard, rock-like mass. 314 This will
reduce the tendency for dusting of the spent shale, hence should simplify
environmentally acceptable disposal.
In summary, the LR process uses relatively simple, inexpensive and
reliable hardware to achieve high oil shale throughput and high oil yields with
the ability to process the entire oil shale resource. Major environmental
concerns, save leachate, have been addressed. 286 On the other hand. the high
nitrogen and sulfur contents of the LR shale oil product will make upgrading
difficult and expensive.
Several suggestions were recently advanced by workers at Monash University
to improve overall efficiency of the lR process when handling oil shale from
the Rundle Deposit in Australia. 115 These suggestions included (a) using a
fluidized bed solid-solid heat exchanger for improved heat recovery from spent
shale. (b) staged drying of the raw shale with the added benefit of raising low
pressure steam, (c) the use of fluidized bed combustion to give better temper
ature control than the lift-pipe combustor and thereby lower heat requirements
for carbonate decomposition, and (d) co-pyrolysis of coal with shale to obtain
higher liquid yields from a retort of given size.
OIL SHALE PAGE 203
Table 34. Inspections of spent shale from the LR retort show that particle
size is low and that the spent shale is much more friable that the
starting shale (i.e. gri ndabil ity is increased) . (Data of Rammler,
Reference 312.)
T~~e of Shale _ E_ --L... _H _
Organic C wt% 0.3 1.3 2.0 Carbonate CO, wt% 15.8 19.3 12.3 Hydrogen wt% 0 0.2 0.2 Total S wt% 0.3 1.5 2.5
Size Analysis > 5 mm wt% 0 0 0
4 5 mm wt% 0 1.1 0 3 4 mm wt% 0 2.9 2.9 2 3 mm wt% 3.1 12.6 7.1 1 2 mm wt% 7.7 2.1 8.8 0.8 I mm wt% 3.3 5.2 2.0 0.5 0.8 mm wt% 3.0 5.7 4.7 0.315 0.5 mm wt% 6.9 8.3 8.0 0.2 0.315 mm wt% 7.9 8.2 6.5 0.1 0.2 mm wt% 5.4 11.8 14.6 0.063 0.1 mm wt% 4.4 10.7 11.1 0.045 0.063 mm wt% 3.8 7.9 28.0
< 0.045 mm wt% 54.5 23.5 6.3
Median Size mm 0.32 0.68 0.55
Grindability of the 150 124 n.d. circulated material (Hardgrove)
OIL SHALE PAGE 204
4. TOSCO II Process
Development of this process by Tosco Corporation (formerly The Oil Shale
Corporation) began in 1955 as an attempt to overcome the shortcomings of the
USBM gas combustion retorting process. Exploratory work proceeded through a
25 ton/day pilot plant built in 1957 at Golden, Colorado and a 1000 ton/day
semi-works plant constructed at Parachute Creek, Colorado in 1965 to detailed
engineering design work completed in 1968 for a 66,000 ton/day commercial
facility the Colony project. To implement the recommendations of the 1968
study, a second phase of developmental work was carried out in the 1000 ton/day
plant for acquisition of the data necessary for construction of the commercial
facility. The developmental phase results have been reviewed by Whitcombe and
Vawter. 316 Initially. the commercial facility was a joint venture of Tasca
(40%) and Arco (60%), with Arco as the operator, but in 1980 Exxon acquired
Areo's interest and assumed the role of operator. Construction of the facility
and the associated town of Battlement Mesa, began in 1980. However, faced with
increasing project cost estimates, ultimately reaching about $7 billion (nearly
twice the original estimate, due in large part to high interest rates) and
lower crude oil prices, the project was terminated early in 1982. 317 ,318 At
that point, Exxon acquired Tosco's 40% interest in the Colony project.
In the Tosco process, hot ceramic balls are mixed with smaller oil shale
particles in a rotating drum retort. After retorting, the ceramic balls are
separated from the spent shale and reheated in a separate ball heater using gas
as the fuel (Figure 54).".
Raw shale from eNsIler
shale separator
Clean gas to atmosphere
flue
Ball elevator
Clean
R~ .... -;;";,,;V surge h<
lneluding
Hydroearbon incinerator
flue
flue To spent shale disposal
generator drum
'rum
Figure 54. The Tosco oil shale retorting unit decoup1es the heat generation
and kerogen pyrolysis steps, thereby simplifying process control.
(Reprinted with the permission of the Center for Professional
Advancement, New Brunswick, NJ.)
Cfean gas 10 atmosphere
OIL SHALE PAGE 206
The feed streams to the Tasca retort are I/2-inch ceramic balls heated to
about 1270'F (690'C) and the oil shale crushed to pass a 1/2-inch screen and
preheated by contact with hot flue gases from the ball heater. The streams are
rapidly mixed under an inert atmosphere. Heat transfer is rapid and at the
retort/s exit the shale and ceramic balls are essentially the same temperature
and the shale is fully retorted.
Retorting in the Tosco process is carried out in a rotating drum that is
mechanically a simpler device than the screw mixer of the lurgi process, but
like the Lurgi retort is only a small part of the total facility (Figure 55).
However, unlike the Lurgi process heat for the Tosco retort is provided by gas,
not combusted shale. The independence of gas flow and shale flow effectively
decouples retorting from heat generation and gives the Tasca retort unusual
latitude for processing shales of widely varying grades. Oil yields from the
Tasca retort generally exceed Fischer Assay. An oil yield nearly 108% of the
Fischer Assay was obtained from one 7-day test in the semi-works plant.
Product vapors from the retort are passed through a separator to remove
fines, then into a fractionator that yields heavy oil, distillate, naphtha and
product gas streams. Because combustion is separated from retorting in the
Tosco process, the product gas is a high-BTU stream that may be used as plant
fuel or reformed to give hydrogen for liquid product upgrading.
OIL SHALE PAGE 207
Figure 55. The Tasca retort is a rotating drum that comprises only a small
part of the total facility. (Reprinted with the permission of The
Center for Professional Advancement, New Brunswick, NJ.)
Rotary Pyrolysis Drum and Seals
Inlet seal Discharge seal Inle ts
---«.::: Accumulator
Retort II Support w
Retort support , ~ ;e!i '-rollers I -, JI
7
OIL SHALE PAGE 208
The ceramic balls and spent shale discharged from the retort are fed into
a cylindrical screen (trammel) inside the spent shale accumulator housing. The
ceramic balls, being larger than the holes in the screen, pass though the
accumulator and are recycled to the ball heater by a bucket elevator. The
spent shale falls into the accumulator at -900°F and passes into a rotating
heat exchanger where its heat is used to raise steam in water-filled tubes.
The spent shale is further cooled in the moisturizer where is water content is
increased to 12-13% to reduce dusting before disposal.
One commercial-size Tasca retort would process about 11,000 tons of oil
shale/day to produce 4,500 barrels of crude shale oil (Table 35). Properties
of a typical oil produced from Mahogany zone Green River oil shale are similar
to those of a low-sulfur crude -- with three very important exceptions: olefin
and nitrogen contents are high, which make the oil very unstable with respect
to sludge and sediment formation (Table 36). like other oils from Green River
shales, the Tasca oil also has a high content of arsenic that is a powerful
poison for the catalysts needed for efficient upgrading.
OIL SHALE PAGE 209
Table 35. Projected products from a single 11,000 ton/day Tosco II Retort.
(Data of Whitcombe and Vawter, Reference 316.)
Shale Feed, 20 gal/ton
Pipelineable shale oil
Ammonia
Sulfur
Coke
II ,000 tons/day
4,SOO bbl/day
150 tons/day
177 tons/day
836 tons/day
Table 36. Inspection data for a typical Tosco II crude shale oil. (Data of
Baughman, Reference IS4,)
°
Gravity, °API 21.2 Pour Point, <OF 25* Carbon, wt% 8S.1 Hydrogen, wt% 12.6 Nitrogen, wt% 1.9 Sulfur, wt% 0.9 Arsenic, ppm Nickel, ppm Vanadium, ppm Iron, ppm
Viscosity, SUS 100°F 106 212°F 39
Distillation S vol% at
10 vol% at 20 vol% at 30 vol% at 40 vol% at 60 vol% at 70 vol% at 80 vol% at
200°F 27soF 410°F SOO°F 620°F 775°F 8S0°F 920°F
The 2S"F pour point is for oil that has been "heat treated" under conditions described in U.S. Patent 3,284,336.
OIL SHALE PAGE 210
In summary, the Tasca II Retorting Process achieved several important
technical goals: Oil yields were high; consistently at or above Fischer Assay.
Shale fines could be processed, making it possible to utilize the entire shale
resource. The gas by-product was not diluted with combustion gases, hence had
a high value for heating or hydrogen production. The process design afforded
both good operability and an unusually high degree of flexibility in control
and in ability to handle shales of varying richness. However. operability and
flexibility were obtained only at the cost of expensive hardware, making the
Colony Project very capital-intensive. This last factor, coupled with the high
interest rates of the early 1980's and uncertain crude oil prices ultimately
led to the demise of the Colony Project that would have used Tosco II retorting
technology. Finally, even though flue gas heat was recovered in preheating the
shale and spent shale heat was used to raise steam. the Tosco II process was
relatively heat inefficient because of the concurrent flows of heated balls and
cooler shale.
OIL SHALE PAGE 211
5. Shell Pellet Heat Exchange Retorting (SPHER) Process
Workers at Shell Development addressed the Tasca drawbacks of high capital
cost, mechanical complexity and heat inefficiency in their design of the Shell
Pellet Heat Exchange Retorting (SPHER) Process. While never progressing past
the experimental stage, a brief discussion of this process is worthwhile, since
it was an important conceptual bridge in the evolution of the newer fluidized
bed processes. The SPHER design grew out of Shell's experience with fluidized
beds in refinery processes, such as riser transport and catalytic cracking in
dense beds, the former operating at relatively high and the latter at relatively
low superficial gas velocities. The key features of the SPHER process are
countercurrent flows of shale and the heat exchange pellets for improved heat
efficiency and the use of fluidized beds for low capital cost.319·322
As conceived, the SPHER process uses two loops for circulation of the heat
transfer pellets or balls (Figure 56). In the cool ball loop, balls fall from
the preheater into a countercurrent fluidized bed for recovery of heat from the
spent shale. A pneumatic riser transports the balls to the top of the cool
ball loop, where they rain down through the up· flowing shale in the preheater.
In the hot ball loop, after heating in a ball heater (riser/lift·pipe combustor
or Tosco-type) the heated balls fall through a dense bed of shale fluidized by
superheated steam in the retorting vessel. Segregation of the two ball loops
allows the size of the different elements and the size and material of the
balls to be tailored to each specific task, while high throughput and
mechanical simplicity are maintained throughout. Projected thermal efficiency
of the SPHER process was 67%, giving it a 4% advantage over Tosco II and a 9%
advantage over the Paraho process.
OIL SHALE PAGE 212
Figure 56. The SPHER process was designed with fluidized beds for low capital
cost and mechanical simplicity, and countercurrent flows of shale
and heat transfer pellets for high thermal efficiency.
RAW SHALE
SPENT ~HAL.t
COOL BALL LO'OP
I
I
PREHEATER
HEAT RECOVERY VESSEL
• •
-'JI/L BALL HEATER
l~~~~~ FUEL GAS
HOT BALL LOOP
OIL SHALE PAGE 213
While preliminary Shell economics indicated that shale oil from the SPHER
process would be about 15% less expensive than Tosco shale oil, due primarily
to lower capital cost, the retorting operation was still relatively complex and
required heat transfer pellets. To obtain reliable fluidization in the SPHER
beds was found to require grinding to -1/16 inch size. Such grinding would be
expensive, even using staged grinding with oversize recycle, especially for
hard, relatively rich shales. such as those from Green River and Jordan. Also,
agglomeration of small particle of relatively rich shale would seriously impair
the operability of the fluidized beds. Balanced against these factors are the
benefits of fast heat-up and low vapor residence time in the retort. both of
which contribute to increased yields of high-value liquids. Other workers at
Shell and at Exxon have realized the capital cost advantages of fluidized bed
retorting and have begun to study ways of further reducing costs and improving
operability in fluidized bed retorting.
6. Shell Shale Retorting Process (SSRP)
A recent report outlines current the current design basis for oil shale
retorting at Shel1. 323 ,324 Concentrating on the retort. itself, van Wechem and
his co-workers set out the following objectives for retort design:
• Achieving low capital cost by minimizing solids recycle means the use
of simple reactor designs and minimum reactor volumes throughout
• Rapid particle heat-up to minimize coking that limits the conversion
of kerogen to volatile products
• Maximum oil recovery, which means vapor residence times of seconds to
minimize cracking reactions that produce gas and thereby reduce
the amount of liquid products that can be recovered
OIL SHALE PAGE 214
• Scale-up capability to achieve economies of scale without undue
recourse to multiple trains, while maintaining a solids residence
time of 5-8 minutes to allow complete retorting
• Flexibility with respect to shale grade - capability to process both
rich and lean shales from a wide range of deposits with reliable
operability and efficiency
• Capability to process shale fines to enable use of the entire resource
Consideration of these factors led to the selection for SSRP of a staged,
cross-flow fluidized bed, operating in the bubbling dense bed regime.
In a gas-solid system, rapid heat-up to a temperature where retorting
proceeds rapidly will require the use of small particles. due to the low
volumetric heat capacity of the gas. Based on the results of Forgac, 2-3 mm
particles would give acceptable average heating rates of lOO-600°C/min.325
This requirement for small particle size is expected to lead to a substantial
penalty in both capital and operating costs for size reduction in SSRP vs.
processes that can use larger particles.
Retorting kinetics vary with shale type; due in large part to differences
in heat and mass transport within the individual oil shale particles. USing
the model of Wallman (Equation 14).'26 the Shell workers concluded that staged
solids flow would be required to obtain the desired kerogen conversion and
selectivity towards oil.
OIL SHALE PAGE 215
(14)
k, KEROGEN --------) LIGHT HYDROCARBONS + BITUMEN
~~~;;::;;~;~~~-------------- ------------------r:k~-----:']-~--------- -------
Within Bed
Bulk Vapor Phase
LIGHT HYDROCARBONS RELEASED
HEAVY OIL RELEASED
Transport
LIGHT OIL +
GAS
Transport & Cracking ""
HEAVY OIL +
GAS
COKE +
GAS
GAS
Transport
H2 0, CO, CO2
Using this model, and data reported by Wilkins321 and generated in-house,
calculations led to a total vapor residence time (extra-particle + bulk vapor)
of 1-2 seconds to maintain cracking to a level of about 10% of total product
oil at normal retorting temperatures. To achieve such a short vapor residence
time is not easy. A moving fixed bed would have to have a very low gas flow to
avoid entrainment of the finer shale particles. In turn, this would require
low bed depth and a very large bed area that could only be met with parallel
trains. Increasing the particle size would allow increased gas flow, but would
also increase particle heat-up time, thereby increasing the bed area reqUirement,
and lead to increased coking and lower product yields.
OIL SHALE PAGE 216
Because of the need for a relatively long solids residence time (8-12 min)
to achieve complete retorting, selection among the various fluidized bed designs
was driven by ability to maximize particle holdup (1 - £). Whereas a dense bed
offered particle holdup in the range of 0.45 - 0.55 and a turbulent bed holdup
would be about 0.4, holdup in a riser bed design would only be in the range of
0.1 - 0.2. Moreover, it was found desirable to minimize turbulence to narrow
the distribution of particle residence times, hence minimize the reactor volume
required to achieve 99% retorting of the average particle. Thus, for both its
higher holdup and its narrower distribution of residence times, the dense bed
was favored. However, gas fluidized beds are inherently unstable and tend to
have a large degree of backmixing associated with the circulation patterns
around rising gas bubbles. Beds with large height/diameter ratios (L/D)
restrict this circulation, hence tend to lower backmixing. To obtain desirable
L/D, while keeping the bed height (hence vapor residence time) at an acceptably
low value, would require multiple trains or horizontal staging. For its lower
capital cost, the Shell workers chose the latter option with steam as the
fluidizing gas (Figure 57).
In order to facilitate close-coupling the various stages within a Single
housing, the use of rectangular bed sections (instead of the usual circular bed
cross-section) was suggested. This suggestion emphasizes the importance Shell
attaches to a Single train/single unit design in minimizing capital cost. The
report indicates that a single unit capable of processing 100,000 tons of oil
shale to produce 50,000 barrels of crude shale oil per day could be constructed
using the SSRP design.
OIL SHALE PAGE 217
Figure 57. Principle of the horizontally·staged fluidized bed selected for the
Shell Shale Retorting Process. (After Voetter, et 01., Ref. 324.)
.. I I I I
OIL + GAS
'> X .. /
SPENT SHALE
SHALE
.:.:.2 r -t t t t t t t t
STEAM FOR FLUIDIZATION
OIL SHALE PAGE 218
Other important features of the horizontally-staged design include the
following:
• The fluidizing gas has the primary function of keeping the solids fluidized;
its flow can be restricted to that duty.
• Product vapors leave the reactor along a short path, thereby minimizing
unwanted cracking.
• A large bed area can be made available without incurring a severe penalty
from backmixing due to turbulence.
• linear vapor velocity can be kept low enough to minimize fines carryover,
without sacrificing vapor residence time.
• Different sections could be operated at different conditions selected for
optimum product yield/quality. Presumably, some heavy product might be
recycled to one or more sections for additional cracking.
• Products from the different sections could be collected separately by
partitioning the gas space without compromising solids staging.
Heat for SSRP would be provided by burning the carbon on the spent shale.
Three types of fluid bed combustors (FBC) were indicated as candidates. A
bubbling dense bed FBC, similar to those in use for coal combustion, but with a
gas velocity somewhat below the 2 m/sec typical for coal, was one option. It
was pOinted out that this would require a large bed height and a large shale
inventory, due to the limited amount of air available for combustion. At the
other extreme, a dilute phase riser FBC would offer high shale throughput and
no shortage of air, but would require a tall reactor to provide sufficient
residence time for complete combustion. As a compromise, the use of a fast
OIL SHALE PAGE 219
circulating fluid bed combustor (FFBC), similar to those now being commissioned
for coal combustion, was suggested. The FFBe combines good lifting power
making gravity flow of the hot combusted shale possible with good contacting
and burn-out, while maintaining a moderate reactor size.
It is not clear from the cited reports how much experimental work backs up
the SSRP design. However Shell, like other major oil companies, has extensive
experience in the design, construction and operation of very large fluidized
beds (e.g. for catalytic cracking). As with other fluidized bed retorts, the
success of the SSRP design may well rest on its ability to control fines. The
available reports show that Shell is well aware of this problem, but past the
obvious measures of controlling gas velocities and installing high-efficiency
cyclones for fines removal, give little hint of the envisioned fines management
strategy.
7. Exxon Shale Retort (ESR) Process
Recent work at Exxon also led to a fluidized bed retort. As reported by
Bauman and co-workers, research on the Exxon Shale Retort Process (ESR) began
with exploratory work in 1980, and by 1982 had progressed to the point that a
decision was made to proceed to the pilot plant stage for the development of a
proprietary oil shale retorting technology.'" A 5-ton/day pilot plant was
constructed at Exxon's Baytown laboratory and operated from July 1984 until
September 1986, processing both Colorado and Rundle (Australian) oil shales.
EVidently, this facility, which included a fully redundant computerized data
acquisition and control system, was constructed with an eye to the installation
of a second, larger pilot plant at a later date. However. such plans were not
discussed.
OIL SHALE PAGE 220
According to the report, fluidized bed designs were selected for both the
retort and the combustor. It was claimed that during 26 months of operation,
this pilot plant logged 3800 hours of operation and completed 73 yield periods,
including a continuous operating period of 1040 hours on feed. It was also
claimed that conditions affording oil yields over 100% of Modified Fischer
Assay were identified. No product information or further details of the
process were provided.
8. Chevron STB Retorting Process
The Chevron Staged Turbulent Bed (STB) Oil Shale Retort represents a very
different approach to fluidized bed retorting from that taken by the workers at
Shell and Exxon. To approach plug flow performance in an open fluidized bed
reactor requires minimizing the backmixing associated with circulation around
rising bubbles. Also, to avoid generating shale fines in the retort little
turbulence is desirable. For these reasons the Shell and Exxon workers chose
to avoid the turbulent bed regime. In contrast, the STB retort operates in the
turbulent flow regime and staging by restricting the flow at intervals is used
to approach plug-flow conditions for the solids.33o.331
Solids move down through the STB retort against a countercurrent of the
fluidizing gas (Figure 58).
Like the Shell and Exxon retorts, the STB is a small'particle retort, hence can
process the entire oil shale resource. However, top size of the shale feed to
the STB retort is 1/4 inch (6.4 mm). Grinding to < 1/4 inch is more expensive
that the crushing required for lump shale processes, but is considerably less
expensive than fine grinding (to <2 mm) required by more conventional fluidized
bed processes.
GAS 1 SOLIDS SEPARATION • FLUE G AS AND FINES
OIL :J. (HOT)
... SPENT SHALE WATER STB
RETORT
RAW SH ALE FEED - RECYCLE - SHALE --- COM BUSTOR -- COARSE -
STRIPP ING GAS - RETORTED ~ SHALE
RETORTED SHALE FINES I-
figure 58. Process flow sheet for the Chevron STB retorting process. Pre-heat
and heat recovery sections are not shown. (After Tamm, et aI.,
Reference 330. Reprinted with the permission of the American
Institute of Chemical Engineers.)
o -r
OIL SHALE PAGE 222
Calculations based on the results of Wallman and co-workers indicated that
retorting of even the largest particles should be complete in about 4 minutes
at 500'C (932'F).326,332 Experimental studies at this residence time (4 min.)
revealed a broad maximum in oil yield in the temperature range of 480·-500·C
(896'-932'F). In this temperature range, yields of C5 + oil were 100% (±3%) of
Fischer Assay_
Locally, the bed of solids in the STB appears to be fluidized, but the
superficial gas flow is well below that required for fluidization of the larger
particles. However, rapid local mixing and good solid-solid heat transfer help
avoid local overheating. This minimizes cracking and coking reactions that
lower oil yields.
The STB retort is flexible with respect to fluidizing/stripping gas; it
can be operated with steam, a recycle gas stream, or a mixture of the two. The
superficial gas veloCity is in the range of 0.3-1.5 m/sec (1-5 ft/sec) at the
bottom of the retort, but increases up the retort as product vapors add to the
gas volume.
A combination of thermal shock as particles enter the bed, the removal of
kerogen, and turbulence in the bed, causes some breakage to generate fines
within the retort. Particles smaller than about 200 mesh are elutriated with
the product vapors, but most are recovered before the oil is condensed. These
fines are rich in carbon, hence are sent to the combustor for recovery of their
fuel value.
Testing for development of the STB concept was carried out in variety of
units, including a I-ton/day pilot plant and a 320-tin/day semi-works unit at
Chevron's Salt Lake City, UT refinery. The semi-works unit was commissioned in
OIL SHALE PAGE 223
1983, operated for about 2 years, and has now been demolished. The nominal
operating conditions show that throughput of the STB retort is very high -- in
the range of 1-2.5 tons/hr/ft2 (Table 37). Properties of oil produced from
Anvil Points shale in the STB are similar to those of oils from other processes
-- nitrogen and Arsenic contents are notably high (Table 38). It was reported
that oil properties were essentially invariant with shale grade over the entire
range of grades (14-38 gal/ton) studied in the STB pilot plant."o
Table 37. Nominal operating conditions for the STS retort. 330
Shale Throughput
Retort Temperature
Residence Time in Retort
Stripping gas velocity
Residence Time in Combustor
Combustor Outlet Temperature
Recycle Shale:Raw Shale Ratio
10-25 tonnes/hr/m2 (2000-5000 lb/hr/ft2)
477-510'C (890-950'F)
2-8 min
0.3-1.5 m/sec (1-5 ft/sec)
1-5 sec
594-816'F (1100-1500'F)
2:1 to 5:1
OIL SHALE PAGE 224
Table 38. Properties of a typical oil from the STB Retort Pilot Plant. Oil
from 27 gal/tom (93 L/tonne) shlae from the Anvil Points Mine.".
Specific Gravity 0.934
Carbon, wt% 85
Hydrogen, wt% 11
Nitrogen, wt% 2.1
Oxygen, wt% 1.2
Sulfur, wt% 0.6
Arsen ie, ppm 20
Viscosity, cSt at 100'F 22
cSt at 130'F 12
Pour Point, OF 19
Ramsbottom Carbon, wt% 3.5
Distillation, ASTM 0-1160
Vol% Distilled 'C
IBP/S 146/196
10/20 220/273
30/40 322/371
50 411
60/70 452/489
80/90 529/ ---
95/EP - --/534
OIL SHALE PAGE 225
9. Petrosix Process
The Petrosix Process was developed in Brazil, especially for processing
oil shale from the Irati Formation. A large demonstration plant (1,600 ton/day)
with a 18-foot diameter retort has logged over 82,000 hours of commercial oper
ation and has processed over 4,600,000 metric tons of Irati shale to produce
about 2 million barrels of shale oil since 1972.333-335 An even larger unit,
with a 36-foot diameter retort, designed by Stone & Webster, is scheduled for
start-up in mid-1988.334·336 Thus, the Petrosix technology has emerged into
the commercial stage. Although the process was developed for the Brazilian
Irati shales, the Petrosix technology has also been considered for use with the
Eastern U.S. Devonian shales. 337
The Petrosix process is indirectly heated. In the demonstration plant, a
hot recycle gas stream is further heated by gas and injected into the retort to
heat and retort the shale (Figure 59). Fluidized bed combustors would probably
be used in the commercial Petrosix plant.
Recognizing the high capital and operating costs of fine grinding, the
Brazilian workers have developed Petrosix to handle large pieces of shale up to
six inches in one dimension. However, to obtain the higher throughput possible
with faster heat-up. secondary crushers and screens are used to provide shale
in the size range of -2 inches+I/4 inch. Fines that pass the I/4-inch screen
are briquetted and added to the retort feed.
OIL SHALE PAGE 226
F;gure 59. Simpl ified flow diagram of the Petrosix retort. (Reprinted with
the permission of the Pace Company, Consultants and Engineers.)
Oil SHALE FEED SEAL GAS-<"ill
FEED HOPPER HIGH BTU GAS
PYROLYSIS VESSEL
DISCHARGE
SHALE DISTRIBUTOR
CYCLONE
ELECTROSTATIC PRECIPITATOR
IIIIIIIIII~-n-\
HOT
HEAVY SHALE OIL
COMPRESSOR HEAVY
00000 I"'-'-GA,,;;:,S---I-SHALE OIL
HEATER MECHANISM
o-\'lA./C'IAo'}a COOL GAS
RETORTED SHALE FACILITIES
SEAL SYSTEM
WATER:-1O'I....!.._-.!'...J--£r--__ RETORTED SHALE SWRRY TO DISPOSAL
CONDENSER
SEPARATOR
LIGHT SHALE
OIL
WASTE WATER
OIL SHALE PAGE 227
Crushed shale is fed into the retort at the top, through a feeder designed
to prevent horizontal segregation. The shale moves downward through drying.
heating and retorting zones against an upflowing stream of heated recycle gas.
The retorted shale then moves down into the lowest section of the retort, where
it is cooled by an unheated recycle gas stream before being discharged through
one of the hydraulically sealed spent shale hoppers.
Gases and product vapors are carried out of the top of the retort by the
recycle gas stream and pass successively through a cyclone for fines removal,
an electrostatic precipitator for collection of heavy oil mist, heat exchangers
where the light oil and water are condensed, and into the gas treatment section
for HzS removal and recovery of light naphtha and lPG. Non-condensible gases
are compressed and part is used to cool the spent shale, part is heated and
injected into the retorting zone, and the balance is used as fuel. Sulfur is
recovered in a conventional Claus plant.
Atmospheric and vacuum distillation are used to recover distillate from a
composite of the heavy and light oils. The vacuum bottoms will be used as fuel
within the oil shale complex. Most oil shale facilities planned for the U.S.
have included on-site hydrotreating to stabilize the crude shale oil before
pipelining. In contrast Petrosix plans to pipeline a composite of distillate
and naphtha to a refinery for hydrotreating. The hydrotreated shale oil will
then be refined along with petroleum crudes.
Relatively little information is available on the properties of Petrosix
shale oil, however some properties of oil produced from Irati shale. in the
demonstration plant are given in Table 39.
OIL SHALE PAGE 228
Table 39. Some properties of Petrosix oil retorted from Irati shale. (Data of
Bruni~ Reference 338.)
Density, 'API
Pour Point, "F
Anil ine Point, of
Viscosity at lOO"F, cs
Sulfur, wt%
Nitrogen, wt%
Paraffin, wt%
Diolefins, wt%
19.6
25
86
20.76
1.06
0.85
0.02
15.0
The 36-foot diameter (II-meter) commercial retort is designed to retort
260 tons of oil shale/hour to produce 2600 bbl. of crude shale oil and 52 tons
of sulfur per day. A commercial facility, using 20 such retorts to produce
50,000 bbl/day of shale oil, was originally planned. Projected cost of this
facility would have been $2.2 billion (1982 U.S. dollars), however, in 1982 the
project was scaled back to a single retort in response to the world petroleum
situation. During the mid 1980's, construction of the single retort was slowed
to free Brazilian funds for offshore petroleum exploration and the unit is now
scheduled to come on-line during 1990. Cost of the single unit will be about
$57 million (1986 U.S), and it is being constructed at Sao Mateus do Sol, in
order to share existing facilities with the demonstration plant. 336
OIL SHALE PAGE 229
10. Moving Grate Processes
In the earlier discussion of the Shell fluidized bed retort (SSRP), it was
mentioned that both short vapor residence time (to minimize cracking) and long
solids residence time (to obtain complete kerogen conversion) could be achieved
in a moving bed, provided the bed height were small and the bed area were very
large. The Allis-Chalmers, Oravo and Superior Oil moving grate retorts were
designed to meet these criteria. In each case, a support grate moves through
zones where the oil shale is loaded to form a dense bed, preheated, retorted,
cooled and discharged. There are, however. differences in grate design, degree
of bed agitation and the way gas flows are controlled to remove products and
generate heat. Because these three processes are so closely related in concept
they will be discussed as a group.
<aJ Allis-Chalmers Roller Grate Process
This process grew out of experience in design and construction of large
iron ore and cement plants on the scale of 10,000 tons/day, about one-fifth the
size needed for a 50,000 bb1/day shale oil retorting plant. In this process,
the shale is conveyed along a straight-line path by a series of closely-spaced
slotted rollers (Figure 60).339·340
Raw shale crushed to 1-3/4 x 1/4-inch is fed into the roller grate. Fines
(-I/4-inch) are screened from the feed, agglomerated and fed on top of the bed
of crushed shale. Bed depth can be up to three feet. The rollers impart a
mild tumbling motion to the shale particle. This has two important benefits in
retorting: by exposing new surface to the hot gas sweep it speeds heat-up and
it causes fines to quickly migrate down through and out of the bed, thereby
preventing pockets of fines that would cause gas channeling.
00
RETORT GAS ~ ,.... GAS WITH '-'-' TO HEAT RECOVERY HEAVY OIL
~ I I HEAT 1 " I
I HEAT EXCHANGER I I AUXIL!~~oY AND
+ rnMR"S~/ FUEL TOH.,
r.., 0
RETORT I AETORT !l lv' COOLING I COOLING II ~\ __ .1.. ___ .1.. •• L_ .• L ___ ~ .- -- ._-'-
~ ~
LIGHT OIL ~ ~
I AIR AIR
F;gure 60. F10wsheet of the A11is·Cha1mers oil shale retorting process.
(Reprinted with the permission of Allis-Chalmers Corporation.)
u
CO OL SPENT SHALE
OIL SHALE PAGE 231
In the preheater, the shale is heated by off-gas from the retorting zone.
This dries the shale and liberates some light oil. Next the shale moves into
the first retorting zone where it is further heated by recycled non-combustible
retort gases in the temperature range of 900 0 -!OOO°F (482°-538°C). The heating
gas flows down through the shale bed and slotted rollers, sweeping liberated
oil into the preheat zone where it is partially condensed. In the second
retorting lone, a 1200°F (649°C) gas stream completes the retorting process to
give a heavy oil that is taken through a heat exchanger and condenser.
Sealing between the retorting and combustion zone is accomplished by the
use of solid rollers (instead of slotted ones), drag plates at the entrance and
exit of the sealing lone, and careful maintenance of equal overbed pressures.
In the combustion zone, air flow is upward through the grate and shale bed
in order to keep the grate temperature as low as possible. The two cooling
zones are kept separate, since the hotter gases from the first cooling zone are
combined with the hot off-gas from the combustion zone and passed through a
heat exchanger where their heat is transferred to the non-condensible retort
gas stream used in the retorting section.
The Allis-Chalmers process was tested in a process development unit (PDU)
18 inches wide and 8 feet long. In this unit, less than 0.2% of crushed Green
River oil shale was carried out as dust, while about 1% was captured in the
windbox. Size analysis showed that little size reduction occurred during
retorting. However, the oil collection system of the PDU was not designed for
efficient capture of light oils. As a result it is not possible to draw firm
conclusions regarding the overall performance of this system. 340
OIL SHALE PAGE 232
.. _., _____ .. _ .. __ ....... ~ ...... "' .. , I"
material-handling equipment and processes for the mining and mineral
industries. The retort design is similar to traveling grate machines supplied
by Oravo for the sintering or pelletizing of iron ore. The grate is a
continuous chain of wheeled pallets that can accommodate a bed of shale up to
95 inches thicK as it travels around a circular track. The shale would be
crushed and screened to give a -I-inch +1/4-inch shale feed. Unlike the Allis
Chalmers roller grate, the shale bed is not agitated on the Dravo grate. This
minimizes the generation of fines during retorting. Some of the fined
generated in crushing could be burned if needed to heat the recycle gas
streams. The remainder and oily fines collected throughout the process would
be agglomerated and fed as pellets or briquettes onto the top of the shale bed.
This capability enables processing of the entire resource and avoids the
expense of environmentally acceptable disposal of oily fines.
Retorting takes place in four zones {Figure 61).341.344 In the first
zone, the shale is heated by oxygen-free gases produced by combustion of gas
recycled from the heat recovery section (natural gas is used for start-up and
additional natural gas is added if needed to maintain temperature). The upper
20-30% of the bed is retorted in this zone.
A major part of the process heat is generated in the second zone, which is
fed with a mixture of recycle gas and air to burn the carbon in the spent
shale. As the combustion front moves down through the shale bed, the retorting
front moves ahead of it to retort the middle portion of the bed. The amount of
air is controlled to limit combustion and prevent breakthrough of oxygen-rich
gas into the product collection system. In Zone 3, oxygen-free gases are used
Gas
Figure 61. Simplified flow schematic of the Dravo Traveling Grate Oil Shale Retorting Process.
(Figure courtesy of Dravo Engineering Companies, Inc.)
Air Air
Gas
Oil Recovery
Oil
o -r
~ '" N W W
OIL SHALE PAGE 234
retorted adiabatically. Finally, oxygen~free gases cool the bed to below 250°F
before the grate is tipped to dump the spent shale into a water-sealed discharge
hopper.
The Oravo process was tested in lab units and in an integrated 300 ton/day
pilot plant (Figure 62). The use of separate windboxes in the pilot plant
allowed demonstration of a key feature of the Dravo process t isolation of the
process steps by varying gas flows to control local pressures above the various
windboxes.
In the oil recovery section, oil and water are condensed from the retort
off-gas, using direct contact quench towers or air heat exchangers. Heavy oil
mist is recovered by electrostatic preCipitation. Desalting and deashing of
the raw shale oil is accomplished using (Petrolite) technology developed for
petroleum applications. Overall oil recoveries in the range of 95% - 100% of
Fischer Assay are reported, depending upon the grade and type of shale fed.
Western (Colorado) and Eastern (Kentucky Devonian Sunbury and Cleveland High
Grade) U.S. shales'4','42 and Australian shales'4' have been tested in the
Oravo 300 ton/day pilot plant. In each all case, excellent operability was
reported.
The quality of shale oil obtained from the Dravo retort is very similar to
that obtained from the Paraho retort operating with direct heating. Pour point
is high, as are the contents of nitrogen, arsenic, iron and nickel. In several
cases, samples of raw shale oil were sent to Gulf Oil for upgrading via 2-stage
hydrotreating. 342 Details of the Gulf results were not discussed by Dravo.
Oil SHALE
FEED CHUTE BLEED TO INCINERATOR
INERT GAS FROM INCINERA TOR
CyCLONE
SHALE OIL STORAGE TANK
NATURAL
G"
COMBUSTION AIR FAN
". COOLER OEMISTER POT "N .--, /
::::::;;:l ~~NT QlE
TO INCINERATOR lINERT GAS GENERATOR)
Figure 62. Flowsheet for the integrated 300 ton/day Dravo pilot plant. (Re·
printed with the permission of Dravo Engineering Companies, Inc.)
OIL SHALE PAGE 236
(c) Superior Oil/Davy-HcKee Circular Grate Retort
This retort was developed as a joint effort of Superior Oil (subsidiary of
Mobil Corporation) and DavY-McKee, another firm well-established in the
material handling and mineral processing field. The retort is not designed
just to recover fuel; it is but one part of the Superior Multi-Mineral approach
to efficient oil shale utilization. 346 ,347 The retort design is based on Oavy
McKee's mineral processing technology.348
The distinctive feature of the Superior Oil/Davy-McKee retorting process
is the way crushed shale is fed onto the grate in three layers; the smallest
particles (+1/4 inch. the -1/4 inch material is not retorted, but may be burned
in a separate unit) are fed directly onto the slotted grate, intermediate sized
particles form the center layer and the largest particles are placed on top of
the bed (Figure 63). As a result, the largest particles that heat up slowest
are exposed to the hottest recycle gases for the longest time. This minimizes
the residence time needed to achieve complete retorting, and has the added
benefit of minimizing overheating which would be detrimental to subsequent
recovery of soda ash and aluminum trihydrate. Laboratory tests carried out in
an adiabatic batch reactor were followed by testing of the retorting step in a
250 ton/day pilot plant.'4' Product quality from Colorado oil shale retorted
in the Superior retort seems somewhat inferior to that from the Paraho retort.
OIL SHALE PAGE 237
Figure 63. A cross-sectional view of the Superior Oil/DavY-McKee Circular
Grate Retort shows the distinctive layering that puts the largest
shale particles on top of the bed where they are heated quickest.
This view also shows how simple water troughs provide effective
seals that are tolerant of both thermal distortion and dimensional
S~.t.U 8(0.
variance in construction. (From Weichman, Reference 347. Reprinted
with the permission of the Colorado School of Mines.)
EOJIVAL£NT TO AN .Ol ....... rlc SECTION OF SOLIDS
STATIONARY
""""
Ott. M.ST AND "'-- OVCT -1f-t1- RECYCLE GAS
TO OIL REMOVAL
OIL SHALE PAGE 238
11. Paraho Process
Above were discussed several new approaches to oil shale retorting that
resulted from attempts to overcome the shortcomings of the USBM Gas Combustion
Retort: inability to process fines and hence to process the entire resource,
and oil yields substantially below Fischer Assay. In contrast, the Paraho
retort and retorting process seem best described as evolutionary - the result
of careful refinement of the Gas Combustion concept. The design philosophy was
outlined in 1981 by Pforzheimer, who listed his criteria for a good oil shale
retort: 350
• Counter-current flow of raw shale and heat transfer medium
• Gravity feed of raw shale through the retort and out
• lumps in - Lumps out (minimal crushing and no fine grinding)
• Direct heating - Low BTU product gas used as plant fuel or for electric
power generation
• Simple configuration
• It works! (Author's note: In 1981 Paraho already had demonstrated ability
to operate their retort for long times with yields approaching
100% of Fischer Assay on a variety of oil shales.)
A comparison of these 1981 criteria with those recently set out by developers
of the Shell Shale Retorting Process (SSRP)3'3,3'4 provides some insight into
the profound changes a few years have wrought in oil shale R&D.
Paraho was a small privately held, publicly reporting, corporation formed
in 1971 to commercialize oil shale technology. Organization of the Paraho Oil
Shale Demonstration began in 1973 after Sohio and Cleveland Cliffs Iron Company
had surveyed available technologies and concluded that Paraho's was the most
promising. Eventually, seventeen companies {Sohio, Cleveland Cliffs Iron,
OIL SHALE PAGE 239
Southern California Edison, Kerr-McGee, Gulf, Sun Oil, Shell, Amoco, Exxon,
Mobil, Webb Venture, Davy McKee, Texaco, Marathon, AReO, Phillips, Chevron)
participated in this Paraho managed project located at Anvil Points, CO. 350
Based on the results, a 10,000 bbljday commercial facility, the Paraho-Ute
Project, was planned for a location on on the White River near Vernal, Utah.
Sohio led a group providing financial backing. However, the demise of the
Synthetic Fuels Corporation removed the possibility of loan guarantees, thereby
increasing financial risk. As a result, Sohio and its partners withdrew their
support for the project and late in 1985 Paraho Development Corp. filed for
protection under Chapter 11.351 It has since emerged from bankruptcy as the
New Paraho Development Corporation and is developing non-fuel uses for oil
shale.352.353 The Paraho operations are an important part of the oil shale
story: the retort did work, Paraho addressed many of the pressing environmental
issues in spent shale disposal, and much of the readily available information
about oil shale upgrading and refining comes from work on 100,000 barrels of
Paraho shale oil that was refined at Sohiols Toledo refinery under a U.S. Navy
contract and from a variety of studies on comparable Paraho oil that were
carried out elsewhere in smaller units.
The Paraho technology, a modification of a commercially-available lime
burning kiln, is described in U.S. Patent 3,726,247.354 The key component in
this design is the discharge grate which ensures even solids flow down the
retort. An artist's rendition of the Paraho oil shale retort appeared in
Sohio's 1973 Annual Report (Figure 64).
OIL SHALE PAGE 240
Figure 64. Cutaway view of the Paraho retort. (From the 1973 Annual Report,
Standard Oil Company of Ohio. Reprinted with permission.)
Raw shale
Rotating spreader
Refractory
Collectors
Distributors
Distributors ~H+
Distributors
Moving grate
Shale moved through grate
Gas/air I ~IJ Jl~ I""
Retorted shale to disposal
Gas/air
OIL SHALE PAGE 241
The Paraho semi-works retort used in the demonstration project was a
refractory-lined open tubular retort. Coarsely crushed shale is fed into the
top of the retort by a rotating feeder (to prevent size segregation) and flows
down against the gas flow. This retort was designed so it could be operated
with either direct- (Figure 65) or indirect-heating (Figure 66).'"
With indirect heating, the Paraho retort is similar to the Gas Combustion
retort in concept, but oil yields are substantially better, approaching 100% of
Fischer Assay (Table 40). With indirect heating, gross oil yields are similar,
but thermal efficiency is lower because more carbon is rejected with the spent
shale. No facility for external combustion of the spent shale was provided at
the Paraho semi-works stage.
It should, however, be noted that the performance data in Table 40 were
obtained with Colorado shale. Performance with a 19 gal/ton Utah shale was
markedly poorer. 356 In particular, oil yield was 92.8% of Fischer Assay when
retorting the Utah shale with direct heating.
Feed shale
Rotating spreader
Collecting tubes
Distributors
Moving grates
Retorted shale
____ + Product gas
Oil/gas separator
Shale oil
Recycle gas blower
Figure 65. Paraho semi-works retort operating in the directly heated mode.
(From Jones, Reference 355. Reprinted with the permission of the
Colorado School of Mines.) ~ ~
'" .. '"
Feed shale
1" ____ Product
gas
Gas & oil mist Oil/ gas separator
Collecting TUo'eSJ
Moving mo,tQO
Retorted shale
Shale oil
Recycle gas blower
Figure 66. Paraho semi-works retort operating in the indirectly heated mode.
(From Jones, Reference 355.
Colorado School of Mines.)
Reprinted with the permission of the
OIL SHALE PAGE 244
Table 40. Product yields from the Paraho retort in directly- and indirectly
heated modes. (Data of Jones, Reference 355.)
Raw Shale, Fischer Assay, gal/ton
Yields: Shale Oil, gal/ton
Volt of Fischer Assay
Oil Qua 1 ity:
Gravity, "API
Pour Point, of
Viscosity, SUS at 130"F
Ramsbottom Carbon, wt%
Water Content, wt%
Solids Content, B.S. wt%
Gas Properties, Vol% (Dry Basis): H,
N,
0,
CO
C 's ,
H,S
NH,
Heating Value, HHV, BTU/scf
Directly Heated
28.0
27.2
97
21.4
85
90
1.7
1.5
0.5
2.5
65.7
-0-
2.5
2.2
24.2
0.7
0.6
0.7
0.4
2660 ppm
2490 ppm
102
Indirectly Heated
28.0
27.2
97
21.7
65
68
1.3
1.4
0.6
24.8
0.7
-0-
2.6
28.7
15.1
9.0
6.9
5.5
2.0
3.5
1.2
885
OIL SHALE PAGE 245
Water and energy requirements for the Paraho process have been discussed
extensively.357,358 Water requirements were of particular concern, since water
is a precious commodity in NW Colorado and NE Utah (also in many other shale
rich areas). The retort section of the Paraho process is a net producer of
water. albeit sour water, in the direct mode. Water in the shale is recovered,
as is the water of combustion. With the retort in this mode, the overall
process requires about 2.1 gallons of water required for each gallon of shale
oil. The majority of this water goes to the cooling tower with smaller amounts
required for refining, power generation, dust control and vegetation uses.
Scale-up issues and the prospects for commercialization were discussed in
several articles as the Paraho technology advanced. 35o ,359,36o An engineering
design study for the proposed Paraho-Ute commercial retort was carried out by
Davy-McKee. 361 Basically, the commercial design consisted of 40 rectangular
sections, each roughly equal in area to the Paraho semi-works unit, and with
the rotating feeder of the semi-works unit replaced by 320 stationary rock
hoppers fed by a moving conveyor (Figure 67).350 The result was mechanically
and structurally complex (Figure 68) and, as a result, did not offer the
economies of scale expected in a project of this magnitude.
OIL SHALE PAGE 246
Figure 67. Cutaway view of the 10,000 bbl/day commercial retorting module
proposed for the Paraho-Ute Project. (From Pforzheimer, Reference
350. Reprinted with permission of the Colorado School of Mines.)
- . ~---
1. Shale Feed System 2. Produc:t OiliGas ColleeUon 3. Top AlfIRecycie Gas 4. Middle AlrlReeyc::le Ga. 5. Bottom Recycle Gas 6. Moving Orates 7. PrOCftSsed Shale
to Reclamation
UPPER AlG OIST
MIDAIG OIST
RECYCLE GAS INLETS
Davy McKee INGINf •• , ANI) (ONS'_UCIOltS
HOPPERS
>\ .... .-__ OfFTAKES
Figure 68. A cutaway view of one 400 ft' Paraho retort sub·module shows its
internal complexity. Eight such units are combined to form the
commercial retort module. (From Greaves, Reference 361. Reprinted
with the permission of the Colorado School of Mines.)
OIL SHALE PAGE 248
12. Unocal Oil Shale Processes: Retort A, retort B, SGR
Unocal (nee Union Oil) has been active in oil shale research and develop
ment since its acquisition of Colorado oil shale properties in the 1920's. The
three closely-related Unocal processes are known as the Retort A, Retort Band
SGR (steam-gas recirculation) oil shale processes. All three processes move
shale upward by means of a ~rock pumpff. a reciprocating mechanical piston that
alternately loads oil shale from a feed hopper, and rams the loaded shale up
into the bottom of the cone-shaped retort (Figure 69).362 In all arrangements,
the rock pump piston operates totally immersed in relatively cool product oil.
Technical feasibility of the rock pump concept was demonstrated during the
1940's in a 2-ton/day pilot retort. This work was followed by construction and
operation of a 50-ton/day unit and scaled up to a semi-works unit at Unocal's
Parachute Creek site in Colorado during the late 1950's.363 This semi-works
unit, termed "Retort A" had a feed piston diameter of 5.5 feet affording a
shale pumping rate of 1200 tons/day (Figure 70). Activities through the 1960's
were at a low level, due to unfavorable economics for shale oil, but were
resumed in the early 1970's leading to Retort B, an indirectly heated version
that gave higher oil yield and a high-BTU gas product.'64,,6s An attempt to
Improve thermal efficiency by burning the carbon rejected with spent shale from
Retort B led to the steam-gas recirculation (SGR) concept.'66 In 1981, Unoca1
began construction of an integrated 10,000 bbl/day shale oil facility (mine,
Retort B, upgrading facility) at Parachute Creek. 367 This facility was placed
in service in 1983 and, with continuing development, is the only commercial
shale oil facility now operating in the U.S.'68 The development of these
processes was reviewed in 1981 by Barnet, who also describes the environmental
studies that accompanied retorting development. 369
OIL SHALE PAGE 249
Figure 69. The rock-pump is a reciprocating piston that alternately loads oil
shale from a feed hopper, and rams the loaded shale up into the
bottom of the retort. (After Hartley and Brinegar, Reference 363.)
FEED CYliNDER
STEP I STep 2
OSCillATING FEED HOPlpfl~
STep 3 STEp ..
OIL SHALE PAGE 250
Figure 70. Retort A incorporated the "rock pump" into an internally heated,
solids upflow design. A limited amount of air is blown into the
top of the retort to support combustion of the spent shale in the
uppermost part of the moving shale bed. (Printed with the
permission of Unocal, Inc.)
r-----l(:=lo--AIR
SHALE a:;::=:=::;J FEED ----,
.. SPENT 7 /''--1",;:>1--')
SHALE COOLING
AND DISPOSAL
lOW BTU GAS
Oil
OIL SHALE PAGE 251
As solids are pumped upwards through the conical Retort A, air is blown in
to burn the carbon on the spent shale. The amount of air is limited to confine
combustion to the spent shale in the uppermost part of the moving shale bed.
but is sufficient to complete combustion within the retort. The hot combustion
gases continue downwards, retorting the shale in the central part of the cone
and picking up product vapors. As the gases laden with product vapors contact
the cooler shale in the lower part of the bed, most of the heavy oil condenses
and trickles down through the shale bed; the balance forms a mist. The cooled
stream of combustion gases, light product vapors and mist exits the retort
through the slots at the bottom and pass into the light product recovery
section. A disengaging section surrounds this part of the retort cone and is
sealed by product oil. As the retorted shale rises above the retort cone, it
forms a freestanding pile. Spent shale falls off the pile by gravity and falls
down discharge chutes through the dome wall into a vessel where it is cooled by
a water spray. The steam generated during cooling helps to strip products from
pores in the spent shale.
Condensation of the product oil within the retort is thermally efficient
and significantly reduces the need for external heat exchangers and condensers,
and for external cooling capacity, for product recovery. Also, a filtering
action, that gives the product oil a relatively low solids content, occurs as
the condensed oil trickles down through the incoming shale.
Rich shales tend to become plastic and agglomerate during retorting. This
agglomeration leads to uneven gas and solid flows, or even plugging in extreme
cases, in retorts where solids flow is gravity-driven. Because there is always
a positive mechanical force available to move the shale up through the retort,
OIL SHALE PAGE 252
anything more than local agglomeration within the bed is avoided. A slowly
revolving rake is provided to break up any agglomerates that do form. Also,
conditions can be selected so that retorting takes place near the top of the
bed where interparticle pressures are low, hence the tendency to agglomerate is
lowest. Thus, agglomeration is not generally a problem, even with very rich
shales.
In Retort At with once-through air as combustion gas, peak temperatures at
the top of the retort reached 2000-2200'F. As a result, cracking limited the
oil yields to about 75% of Fischer Assay. Moreover, heating value of the gas
product was low, about 120 BTU/scf, because of dilution with nitrogen and CO,
from combustion and carbonate decomposition.
Efforts to improve oil yield and heating value of the gas product led to
the development of the second generation concept, "Retort Bn and the Unishale B
process (Figure 71). In this case the heat required for retorting is provided
indirectly, using a recycle gas stream heated to 950-1000'F (510'-540'C) in an
external fUrnace. Rundown oil Yields are high, essentially 100% of Fischer
Assay, and the C4 + oil yields are significantly above Fischer Assay. Moreover,
the gas product has a high heating value of over 800 BTU/scf.
In Retort B, the space above the freestanding pile of spent shale is
enclosed by a dome to exclude air and heat is provided by a recycle gas stream
that is heated in an external fUrnace. Spent shale falls off the retort by
gravity and down discharge chutes through the dome wall into a vessel where it
is cooled by a water spray. Steam generated during cooling helps to strip
products from pores in the spent shale. The cooled shale is then moistened
before being discharged.
RECYCLE GAS
"" HEATER IIIH Hill 1I t J
~ VENTUR I RAW - SCRUBBER
SHALE ¥ Oll~WATER
~RETOR~ SEPARATOR
~ J RE TORT MAKE GAS GAS TREA liNG
r~ r--J\ r- ~- ~C_w.
~ .
d-l--WATE R . .
SEAL .. ,: .. . ,
I ~y .
~ RETORTED SHALE TO - - ,
I !-DISPOSAL --
TO
-MAKEUP WATER
RUNDOWN OIL PROD UCT
Figure 71. In the Unishale B Process, Retort B incorporates the rock-pump, but
is indirectly heated. (Printed with the permission of Unocal, Inc.)
OIL SHALE PAGE 254
As in Retort A, the Retort B shale oil and the gas stream containing
product gas, vapors and mist exit at the bottom of the retort through slots
that lead into the disengaging section. Gases from the disengaging section are
scrubbed and cooled. Oil and water collected as mist and condensate are
separated. The oil is returned to the disengager, while the water is used to
moisten spent shale. Part of the scrubbed gas is compressed and heated before
being recycled to the retort. The balance is processed by compression and
scrubbing to remove heavy ends and sweetened using the Unisulf Process to
remove H2 S. The sweetened gas is then suitable for use as plant fuel.
However, hydrogen requirements for shale oil upgrading are substantial and the
made gas is rich in hydrogen (Table 41). As a result, fractionation to recover
this hydrogen may be desirable.
Retort B produces a very high quality shale oil (Table 42). Pour point, a
very important parameter if the oil is to be pipe1ined, is notably low and the
Conradson carbon value also has a very desirable low value. Treatment of this
crude shale oil involves water washing (2 stages) to remove solids, removal of
chemically bound arsenic to a level of 1 ppm (using a proprietary absorbent),
and stripping of light ends to stabilize the oil. Unoca1 plans to upgrade the
crude shale oil by hydrotreating to produce a syncrude that is a premium
quality feedstock for a conventional refinery. Thus, the high quality of the
Retort B shale oil will eliminate the need for the coking step used to remove
refractory material in upgrading Retort A shale oil.
OIL SHALE PAGE 255
Table 41. Unocal Retort B make gas properties (dry basis). (Data of Dhondt,
Reference 364.)
Mol%
H2 25
Methane 24
C2 I s 10
C3 I S 8
C4 's 5
C5 IS 2
C.-plus I
CO 5
CO2 16
H2S 4
Total 100
Heating value, Gross BTU/sef 980 (37 MJ/m3)
OIL SHALE PAGE 256
Table 42. Crude shale oil from the Unocal Retort B has both a low pour point
and a low Conradson carbon value. (Data of Dhondt, Reference 364.)
Gravity, 0 API
ASTM 0-1160 Distillation, of
IBP
10%
50%
90%
Max
Nitrogen, wt%
Oxygen, wt%
Sulfur, wt%
Water (Karl Fischer), wt%
Arsenic, ppm
Conrad son Carbon, wt%
Pour Point, ·F
Heating Value, Gross M BTU/gal
22.2
150
390
770
1010
1095
1.8
0.2
0.8
0.2
50
2.1
60
142
OIL SHALE PAGE 257
Shale for the Unocal processes is crushed in two stages to <2-inches. No
fine grinding is required and little of the mined shale is discarded.
Water is precious in the area of the Green River Deposit. Therefore, it
is important that water requirements be low. Unocal estimates that Retort B
process water requirements will be only 1-2 bbl/bbl of shale oil. Water needs
will go up somewhat if an SGR module (vide infra), or a shale oil upgrading
facility, is added, but should still be relatively low. By way of comparison,
the water required for producing synthetic oil from coal would amount to 6-8
bbl/bbl oil, and the water requirements of an oil-fired stearn generating plant
are about 10 bbl/bbl oil. Thus, shale oil production has relatively low water
requirements.
From a nominal 34 gal/ton Green River oil shale, the Unishale B Process
recovers 87% of the available energy as oil and gas, but rejects 13% as coke.
Some of the oil and gas produced must be burned to provide heat for retorting.
If the retorted shale were burned to recover its energy. more oil and gas would
be available for sale and thermal efficiency would rise. The Steam-Gas
Recirculation (SGR) concept is Unocal's way of achieving this goal of higher
thermal efficiency. In the original SGR concept, carbon on the spent shale was
gasified with steam to produce a hot synthesis gas stream that provided heat
for retorting. Carbon conversion was excellent, but carbonate decomposition
was extensive leading to a high CO, content (up to 60 mol%) in the recycle
gas. 370 Further work led in 1977 to SGR-3, an add-on that replaces the recycle
gas heater for Retort B to avoid these shortcomings. 366 In the SGR-3 process,
carbon on the spent shale is burned in a vessel isolated from the retort by a
steam seal to prevent dilution of the made gas by flue gas (Figure 72). As a
RAW SHAle
SEALING - f-- SHAM H I WAS VENTURI
1000 TOW ER :'-
SCRUaBER
Oil/WATER SEPARATOR
• -(---
p C-O -~ - I~ \--cl-~
, :> i_ t G
"U~ ~ 0
~ u
10 V .d I[J)4 w ~
~ % ~ • 0 • w w ~ ~ w ~
~ U Z ~ ~
~ '" U z :> w 0 ~ :>
0 - ~
0 ~ ~ ~ 0 '" 0 ~ ~
• '" '" 0 ~ " • , L
• L • .~ " :> • w
'" w w:> w L U w U ~ • ~ ~U w ~ 0 w w
~ -. :> '" 0 0 0- ~
~ • '" '" .u ~ ~ 0 L ~
Figure 72. SGR-3 process flowsheet. (After Duir, et a7., Reference 366.)
OIL SHALE PAGE 259
result, the product slate obtained using SGR-3 is identical to that given above
for Retort B. Heat from combustion is recovered from the hot flue gases to
heat the recycle gas stream for retorting and to raise steam for other process
heat needs, including most of the power plant requirements.
Typical operating conditions for SGR-3 include a temperature of 1550-F at
the combustor outlet (Table 43). At this temperature, endothermic carbonate
decomposition will consume an appreCiable fraction of the heat produced, but
apparently such a high combustor temperature is necessary to heat the recycle
gas stream to the lOOO°F required for efficient retorting. The combustor temp
erature is, however. maintained well below the fusion/clinkering temperature of
1800°F by varying the flows of air and flue gas recycle to the combustor. Flue
gases from the combustor are very low in SOx because of scrubbing action of the
basic calcium-containing components. but treatment to lower the high NOx levels
will probably be required to meet emission standards.
Table 43. Typical SGR-3 pilot plant operating conditions for 36 ga1/ton Green
River oil shale. (Data of Duir, et al., Reference 366.)
Shale feed rate, tons/day
Shale size, inches
Pressure - Retort B top, psig
Recycle gas temperature at retort inlet, OF
Combustor outlet temperature, OF
Retort recycle gas flow, scf/ton
Air to burning zone of combustor, scf/ton
Flue gas to burning zone, scf/ton
Flue gas to combusted shale cooling zone, scf/ton
3.0
+114-1
3.0
1,000
1,550
13,500
14,000
12,000
11,000
OIL SHALE PAGE 260
Table 44. SGR-3 flue gas composition (dry basis). (Data of Duir, et al.,
Reference 366.)
N, 71 mol%
CO, 28 mol%
0, 0.5 mol%
SO, 5 ppm
NO, 300 ppm
A comparison shows that the SGR-3 add-on substantially improves thermal
efficiency over that achieved with Retort B alone (Table 45). Not shown in
Table 45 are the higher capital and operating costs, and the added complexity
of operation, associated with the SGR-3 add-on. These factors will, at least
partially, offset the higher thermal efficiency of SGR-3.
OIL SHALE PAGE 261
Table 45~ Thermal efficiency of SGR-3 is substantially higher that that of
Unishale B. (Data of Duir, et al., Reference 366.)
Retort B Yield M BTU
Inputs:
Raw shale, 34 gal/ton 1.0 ton
Purchased power 21.6 kWh
Oi ese 1 fuel b O. 3 gal
Total Input:
Outputs: c
6,350
221'
41
6,612
4,650 Shale oil product
Pipe 1 ine gasd
Sulfur
32.86 gal
o o 0.0008 UK ton _--,7_
Total output 4,657
Thermal efficiency 70 % (100 x energy output/input)
(a) Power plant fuel.
(b) Fuel for mining and spent shale disposal.
SGR-3 Yield M BTU
1.0 ton 6,350
5.9 kWh 60'
0.3 gal 41
6,451
34.04 gal 4,816
519
0.0008 UK ton _--,7,-
5,342
649 scf
83 %
(c) Includes all facilities to produce salable products.
(d) 800 BTU/scf gross heating value.
OIL SHALE PAGE 262
In summary, Unocalls shale upflow retorting technology has several strong
points:
• Oil vapors liberated from the shale are quickly carried down away from the
retorting zone and cooled. This limits polymerization and cracking that
lead to undesirable gas and heavy oil.
• Condensation of oil within the retort is thermally efficient and reduces
the need for expensive external heat exchangers and condensers.
• The high heat capacity of the Retort B recycle gas and high gas-solid heat
transfer rates enable very high shale throughput.
• Positive mechanical force provided by the rock-pump, coupled with process
control that limits retorting to the region near the top of the retort where
pressure between shale particles is least, enables reliable retorting of
rich shales that tend to agglomerate.
• The Unocal technology has been proven over many years of operation at
scales that include laboratory. pilot plant and commercial units.
OIL SHALE PAGE 263
13. Kiviter and Galoter Processes (U.S.S.R.)
Oil shale provides a vital energy source for the Baltic region, Estonia
and lenningrad. About 75% of the oil shale mined is burned directly to raise
steam for electric power generation, while the balance is retorted. Kukersite,
the richest Estonian oil shale, is the preferred feed. The two processes are
complementary - Kiviter handles coarse shale, Galoter the fines. Relatively
little technical information on these processes is available in English, but
some details were presented by Soviet workers in 1976 at a seminar sponsored by
Resource Sciences Corporation371 ,372 and a review of the Kiviter process by
Soviet workers includes a list of leading references. 373 Baughman reviewed both
processes in 1978374 and Cieslewicz reviewed Estonian-Russian work in 1971. 375
The older of the two Soviet processes, termed the Kiviter Process, handles
coarse oil shale lumps of 1-5 inch size in a single multi-function retort that
is generically similar to the direct-fired Paraho and Gas Combustion retorts in
the U.S. Shale is gravity-fed downward through the Kiviter retort. However,
relatively rich kukersite (like very rich Colorado shale) becomes plastic on
slow heating and the resulting agglomeration can lead to uneven gas and solids
flows and/or plugging of the retort. Gas flow patterns selected to avoid these
problems, the retort design to accommodate these flows, and the water-sealed
spent shale discharge with mechanical shovel unloading are distinctive features
of the Kiviter retort which is covered by a U.S. patent (Figure 73).376
OIL SHALE PAGE 264
Figure 73. The Kiviter Process uses a single, multi-function retort. Rapid,
cross-current gas flow through the thin annular semi coking region
gives very short residence times for the evolved product vapors.
(After Doilov, et al., Reference 376. Printed with the permission
of Williams Brothers Engineering Company.)
OIL SHALE SEMICOKING CHAMBER
OIL VAPORS COLLECTING B EVACUATION CHAMBER
RECYCLE GAS INLETS
RECYCLE GAS INLETS
GAS BURNERS
OIL E
CHARGING DEVICE
OIL VAPORS B GAS
HEAT CARRIER PREPARATION CHAMBER
GAS BURNERS
RECYCLE GAS FOR COOLING SPENT SHALE
SPENT SHALE DISCHARGE OEVICE
OIL SHALE PAGE 265
As shale is fed into the top of the Kiviter retort, it enters a relatively
long and thin annular space (1-1.5 m thick x 10-12 m tall), the "semicoking
zone", where it is rapidly heated by hot (700'-900'C) gases that flow rapidly
outward through the annular shale bed from the "heat carrier preparation
chamber" in the center of the retort. Because the bed is thin, the shale can
be heated rapidly and the product vapors can be removed with very short gas
residence times. The rapid heating and efficient removal of product vapors
enable reliable retorting of all but the richest Estonian kukersites.
The retorted shale, which the Soviets term "semicoke n, then moves down
into the gasifying zone for gasification at 900 0 e, using a steam-air mixture.
This mixture is injected into the very bottom of the retort and helps to cool
the spent shale as it travels upward. However, the steam-air mixture cannot
provide enough heat to sustain gasification, so additional heat must be
supplied by gas burners built into the side of the retort.
Now the curious name Hheat carrier preparation chamber" becomes more
understandable. Part of the hot gas for retorting is the hot synthesis gas
stream flowing up from the gasification lone, while the balance is generated by
gas burners in the central chamber. Thus, the functions of the heat carrier
preparation chamber include generation of hot flue gas, mixing this flue gas
with the hot synthesis gas, and distributing the resulting hot gas mixture to
give even heating of the annular shale bed and rapid product vapor removal.
As the spent shale moves further down toward discharge, it is cooled both
by the upflowing steam-air mixture in the center of the retort and by cool
recycle gas streams injected around the periphery. Obviously, there are large
temperature gradients in this region, hence the geometry of the discharge
OIL SHALE PAGE 266
chutes and cooling gas injection points and flows must be carefully designed to
achieve efficient heat recovery. Few details concerning these key aspects were
presented.
large amounts of gas for cooling the spent shale, and for gasification,
are injected into the lower part of the Kiviter retort. This gas must travel a
long path before passing through the annular retorting zone at the retort top.
Even though care is taken to maintain bed permeability, the bottom of the
retort operates under appreciable pressure. Also, reliably maintaining an even
flow of shale through the retort was apparently as troublesome for the Soviet
workers as for other retort developers. To provide sealing against pressures
up to 1,000 mm-H20 effective solids flow control, the Kiviter retort uses
discharge chutes that project down into a water trough that is fitted with a
reciprocating mechanical shovel (Figure 74). Spent shale flows down the
discharge chutes until it piles up to the chute exit. The mechanical shovel is
pivoted to sweep shale from the semicircular bottom of the trough onto a
sloping drain tray. The shovel is lifted for its return past the discharge
chute, then lowered for another sweep. The arrangement is claimed to work
well. It does seem that a simpler, and probably more reliable, mechanical
shovel would result if the discharge chutes were made vertical and the shovel
were made double-acting that is drain trays were provided on both sides of
the seal trough and the shovel were made to sweep spent shale on both forward
and reverse pivot strokes. However, other methods for controlling solids flow
have proven better suited for continuous operations on a large scale (e.g. the
non-mechanical steam seal used in Unocal's SGR-3 system).
OIL SHALE PAGE 267
Figure 74. The Kiviter discharge section provides sealing against pressures up
to 1,000 mm~H20 and a pivoted mechanical shovel that both unloads
the spent shale and controls solids flow through the retort.
(After Ooilov, et .1., Reference 376.)
SPENT SHALE DISCHARGE
•
1 •• •••••• •• 1
SPENT SHALE CONVEYOR
MECHANICAL SHOVEL
N"-- WATER LEVEL
OIL SHALE PAGE 268
to even heating in the Kiviter retort. For this reason, the feed shale is
screened to remove shale particles smaller than I-inch (25 mm).
With a Baltic shale (presumably kukersite), the Kiviter oil yield is about
85% of Fischer Assay_ Gas production is also significant, but the Kiviter gas
is diluted with nitrogen from the combustion air and CO2 produced by combustion
and carbonate decomposition (40-50% carbonate decomposition). As a result, it
is a low-BTU gas suitable for use in firing the retort or for providing heat
for upgrading/refinery operations, but not for pipelining (unless blended with
a high-BTU gas). Overall thermal efficiency of 74% is claimed for operations
with a 50 gal/ton Baltic shale (enriched at the mine). By-products also
generate significant credits; a high-quality cement is produced from the spent
shale from selected retorts.
Kiviter retorts capable of processing 250-300 tons of shale per day are
routinely operated, generally in banks of 10 retorts connected to a common
product collection system and sharing a common upgrading facility. A prototype
1,000 ton/day Kiviter retort was under construction in 1981. Presumably, this
retort has been placed in service, but no operational details are at hand.
The Galoter Process is similar to the Lurgi-Ruhrgas, Tasca II and Shell
(SPHER, SSRP) processes in using hot spent shale as the heat carrier.3ll Dried
oil shale (smaller than I-inch, IIO'C) is mixed with hot spent shale (800'C)
in a screw mixer, then passed into a 50QoC rotary kiln, similar to Tosco II
(Figure 75).
x
IX WASTE HEAT
BOILER
I--
TO ELECTROSTATIC PRECIPITATOR 6 STACK
I y ~ /' ASH
SEPARATOR r-...~50<>C Y:
" /' HEAT CARRIER SEPARATOR .----t':-'0
DRYl"'- . OIL SHALE SEPARATOR OIL VAPORS TO
CONDENSATION SECTION
DUST ( ;.; s~~t:&~' ~ \-E HO°C 8000C REMOVAL t,',i
•£.""0 CHAMSER,'i===':;' 7 ""
~' AS~; DRtER\L ~ MIXEC"-1l_~~C::'i5"OO",.-,Ch-,
HEAT j ........ \ t.! EXCHANGER "REACTOR' ( ~
LU __ --tJ~H\ ,--, / SPENT , SHALE
600° C
t
Figure 75. Flowsheet shows that the Galoter Process shares many features with
the lurgi-Ruhrgas and Tosco II processes. (After Resource Sciences
Corporation, Reference 371. Reprinted with the permission of
Williame Brothers Engineering Company,)
OIL SHALE PAGE 270
From the rotary kiln, the retorted shale and product vapors pass into a
gas-solid separator, from which the vapors are sent to the product recovery
section and a part of the spent shale is discarded. The balance of the spent
shale is fed into an air-blown riser combustor where burning of residual carbon
raises the solid temperature to BDObe. The hot shale stream is used to provide
heat for retorting, while the hot gases are used to raise steam and then to dry
the wet incoming shale.
Oil yields 85-90% of Fischer Assay and an 82% overall thermal efficiency
are claimed for Galoter process operations with 40 gal/ton wet Baltic shale.
It should, however, be realized that the comparison with Kiviter is unfavorable
to Galoter for two reasons. First, shale for the Kiviter retort is processed
at the mine to remove sand and rock, while shale for the Galoter retort is not
cleaned. Second, the reported Galoter results were obtained with a shale that
contained appreciably more water (12.4 wt% VS. 9.0 wt%); removing this extra
water will lower overall thermal efficiency. Also, the gas product from the
indirectly heated Galoter retort is not diluted with combustion gases, hence
has a high heating value of 1170 BTU/scf and is suitable for pipelining.
The Galoter process was tested extensively in a 500 ton/day plant. As of
April 1975, this plant had operated for 68,000 hours, processed 1,400,000 tons
of shale and produced 186,300 metric tons of oil and 2.5 billion scf of gas.
Based on these results, a larger Galoter retort was designed to process shale
at the rate of 3,300 tons/day. A Galoter complex would include four of these
larger retorts, together with facilities for gas treatment and upgrading shale
oil to produce finished products. The first unit of this complex was scheduled
for completion in 1978. Eventually, Soviet workers envision shale processing
OIL SHALE PAGE 271
complexes that contain both Galoter and Kiviter retorts for more efficient
processing of the entire mined resource. In such complexes t the low-BTU gas
from the Kiviter retorts would serve primarily as plant fuel, while the high-BTU
Galoter gas would be pipelined (perhaps as a blend containing a small amount of
excess Kiviter gas).
The shale oils produced by Kiviter and Galoter retorts are reportedly of
essentially the same composition (Table 46),371
These results show that oils from the Baltic shales are much richer in
oxygenates, especially phenols, than a typical Green River shale oil which is
almost devoid of phenols. Also, the Hie ratio of the Baltic shale oil is much
lower than the 1.55-1.60 that is typical of retorted Green River shale oils.
In both respects, the Baltic shale oil is much more similar to shale oils from
retorting of Devonian shales from the Eastern U.S. Peterson and Spall have
reported a detailed comparison of Estonian and Green River shale products. 377
OIL SHALE PAGE 272
Table 46. Typical properties of crude shale oil produced from Baltic shale in
either Kiviter or Galoter retorts. (Data presented by Resource
Sciences Corporation, Reference 371.)
Density at 68"F, glcm'
Viscosity at 167°F, °Engler
Pour Point, of Coking Value, wt%
Phenols, wt%
Calorific Value, Gross, BTUjlb
Distillation: Temperature. °C 190 (IBP) 200 250 300 360
Elemental Composition (dry basis)
Carbon
Hydrogen
Sulfur
Oxygen + Nitrogen
Atomic HIC Ratio
I. 01
4.5
5 8
28
17,010
Vol%
1 6
21 45
83.3
10.0
0.7
6.0
1.43
OIL SHALE PAGE 273
14. HYTORT Process (IGT)
The Institute of Gas Technology (IGT) has been a leader in exploring the
use of a hydrogen atmosphere in retorting (hydrotorting). Initially. the
objective was gasification at high temperatures, however most of the products
obtained below the carbonate decomposition temperature were liquids. 378 Next,
IGT devised and patented a process for gasifying the shale liquids. 379 More
recent hydrogen retorting studies at IGT have focused on obtaining high liquid
yields and improved product quality, especially from Eastern shales.38o-391 In
addition to IGT, Texaco was active in this area during the 1960'5392-400 and
Phillips Petroleum studied process response for hydrotorting Indiana New Albany
shale in a recirculating loop reactor system. 401 Phillips also participated
with IGT. Bechtel and Hycrude Corp. in process design studies aimed at the
commercial use of the IGT technology with Eastern U.S. shales.386.387 Work at
Mobil Research & Development has resulted in the development of a Rapid Heat-Up
(RHU) Assay, especially designed to evaluate Eastern sha1es. 402
Most retorting studies have used Fischer Assay oil yield as a benchmark.
Some of the more efficient fluidized bed processes have achieved oil yields of
110%, or so, of Fischer Assay. However, Fischer Assay recovers only a small
part of the organic matter in many of the world's oil shales. Thus, potential
oil yield is several times Fischer Assay in these cases, notable among which
are the Devonian shales of the Eastern U.S. Hydrogen retorting affords greatly
enhanced oil yields in many such cases (Figure 76).
OIL SHALE PAGE 274
Figure 76. Hydrogen retorting dramatically increases oil yield over Fischer
Assay for many of the world's oil shales. (From Rex, et aI.,
Reference 388. Reprinted with the permission of the Colorado
School of Mines.)
SWEDEN - BILlINGEN SWEDEN-NARKE
SICILY
..l.
MONTANA HEATH-FORMATION CANADA-ONT ARlO
FEDERAL REPUBllC OF I GERMANY AND LUXEMBOURG
JORDAN-El lAJJUN BRAZIL-LOWER IRATI
MOROCCO AUSTRAUA-CONDOR
CANADA-NEW BRUNS¥nCK THAIlAND
AUSTRAUA-RUNDlE WESTERN U.S. -COLORADO BRAZIL -PARAlBA VALLEY
SOUTH AFruCAN TORBANITE
LOTHIANS-SCOTlAND Q CONVENTIONAL,
THERMAL RETORTING 1:::::L_-L_-L_..L._..L._L-_L-_L----1
100 200 300 400 500
OIL YielD, % OF ASCHER ASSAY
OIL SHALE PAGE 275
The quality, in comparison to Fischer Assay oil, of the oils obtained by
hydrogen retorting depends markedly upon the type of shale (Table 47).403 The
oil produced from Green River oil shale by the HYTORT process is very similar
to the Fischer Assay oil; the main differences in elemental composition are a
somewhat lower oxygen content. Nitrogen content was slightly higher and this
seems to a general phenomenon in hydrogen retorting. In sharp contrast, the
oils from hydrogen retorting of the two Eastern shales were quite different
from their Fischer Assay counterparts. Moreover, the two Eastern shales were
quite different in their response to hydrogen (though oil yield increased in
both cases). In both cases, the HYTORT oils had appreciably lower atomic Hie
ratios than the Fischer Assay oil. However, the sulfur content was about the
same and the oxygen content only slightly lower for the New Albany HYTORT oil,
while both sulfur and oxygen contents were much lower for the Sunbury HYTORT
oil. The reason for this behavior is not known.
OIL SHALE PAGE 276
Table 47. Elemental analyses of shale oils and hydrotreated shale oils. (Data
of Netzel and Miknis, Reference 403.)
Colorado Shale Oil Carbon Hydrogen Sulfur Nitrogen Oxygen H/C Atomic Ratio
Kentuckv New Albany Shale Oil Carbon Hydrogen Sulfur Nitrogen Oxygen H/C Atomic Ratio
Kentucky Sunbury Shale Oil Carbon Hydrogen Sulfur Nitrogen Oxygen H/C Atomic Ratio
Elemental Analysis, Fisher Assay A B IGT ·HYTORT
85.08 11.44 0.73 1.80 I. 31
84.91 9.83 1.02 1.52 I. 72
84.36 9.96 1.33 1.33 2.12
84.30 11.42 0.63 I. 90 1.50
1.61 (Avg.)
1.40
I. 41
84.97 10.08 1.61 1.56 2.00
(Avg.)
84.59 9.98 1.59 1.48 2.36
(Avg.)
84.57 11.40 0.62 2.13 0.98 1.61
85.46 9.42 1.52 2.12 1.61 1.31
85.45 9.56 0.99 2.12 1.22 1.33
Wt% IGT·HYTORT
Hydrotreated
88.27 11. 70 0.05 0.40 0.12 1.58
87.30 12.60 0.06 0.17 0.28 1.72
OIL SHALE PAGE 277
Most hydrogen production involves the production of synthesis gas, which
is further processed to produce hydrogen. In coal liquefaction, synthesis gas
in the presence of liquid water has been shown superior to hydrogen alone for
improving reaction rates and oil yields. Hydrogen retorting was compared to
retorting in a mixture of synthesis gas and steam for three Eastern shales.
The oil yields and organic carbon conversions achieved with the synthesis gas
mixture were approximately those that would have been obtained by hydrogen
retorting at the hydrogen partial pressures used. However, the synthesis gas
mixture yielded oils with appreciably higher Hie ratios and API gravities, and
lower nitrogen contents (Table 48).404 Sulfur contents were, however, higher
for the oils produced with synthesis gas.
OIL SHALE PAGE 278
Table 48. Comparison of oil yields and properties of oils produced by hydroretorting and syngas retorting of Eastern shales. (Data of Punwani, et al., Reference 404.)
SYNGA~ HYDRQGE~' ~Y~GAS HYQBOGEN SYNGA~ HYDROGEN" Maximum Temperature,·C 649 654 568 569 502 487
'F 1201 1209 1055 1056 935 909 Total Pressure, atm 69.4 28.4 68.9 28.3 68.9 28.4
psig 1006 402 998 401 998 403 Hydrogen Partial Pressure,
atm 24.3 28.4 24.3 28.3 24.3 28.4 psig 350 402 350 401 350 403
Shale Oil Ultimate Analysis, wt%
Carbon 83.30 84.67 83.89 84.81 83.99 84.61 Hydrogen 9.86 9.62 10.16 9.63 10.02 9.97 Sulfur 1. 79 1.28 1.52 1.23 1.50 1.39 Nitrogen 1. 70 2.07 1.64 2.05 1.80 1.93 Ash 0.0 0.0 0.0 0.0 0.0 0.0 Oxygen (by difference) 3.35 2.36 2·79 2.28 2.69 2.10
Total 100.00 100.00 100.00 100.00 100.00 100.00 C/H Ratio 8.45 8.80 8.26 8.81 8.39 8.49
Specific Gravity [15.6/16.6'C (60/60'F)) 0.947 0.978 0.945 0.986 0.960 0.967 API Gravity, 'API 17.9 13.2 18.2 12.0 15.9 14.8 Oil Yield
lITonne 89.2 100.5 90.06 103.4 92.1 97.6 Gal/Ton 21.4 24.1 21.6 24.8 22.1 23.4
Organic Carbon Conversion, %*** 69.4 78.6 72.9 75.6 68.1 72.9
* Average of three tests ** Average of two tests *** Elemental conversions based on solids and ash balance
OIL SHALE PAGE 279
Early work at IGT (1972-1979) focused on moving bed hydroretorting, the
HYTORT process. Tests were made with selected shales at rates up to 1 ton/hr.
Additional progress was made during 1980-1983 by the joint Hycrude-IGT-Bechtel
Phillips Petroleum effort, especially in reducing hydrogen consumption. 38 6. 38 7
Based on these results, an engineering design study was made for a commercial
HYTORT facility (Figure 77).389 Possible environmental impact of such a HYTORT
facility vs. other alternatives for retorting Eastern shales has been examined
for a central Tennessee 5ite. 405
The HYTORT effort served to bring into focus the advantages of hydrogen
retorting for Eastern oil shales. However, the HYTORT moving bed technology is
mechanically complex and its large, high-pressure vessels are expensive to
construct. Moreover, fines are not readily processed in the moving bed system,
hence are usually discarded. Therefore, the emphasis of more recent work at
IGT has shifted to a pressurized fluidized bed hydroretorting (PFBH) concept.
The PFBH process will use a vertically-staged fluidized bed (Figure 78).40'
The PFBH is expected to afford higher throughput, hence lower reactor capital
costs, than the moving bed HYTORT process. Moreover, for Eastern shales the
use of smaller shale particles in the PFBH should boost oil yields, while
decreasing gas yields and having little effect on overall carbon conversion
(Figure 79). Thus, not only should the PFBH process more shale per capital
dollar than the moving bed, it should also give about one-third more shale oil
per unit weight of shale. As a result, the PFBH is expected to substantially
reduce the cost of producing shale oil from Eastern U.S. Devonian oil shales.
ISO
HIGH PRESSURE ..-HYDRAUU(: LIFT
100
RAW SHAlE
'-
so SLURRY MIXING
\ ON
DEWATERING
lA SEPARATOR p~f:!:,'f';: ,:,.:JfP ff'." , ( r\ .' PA$WE ,I
. SOlIDS ' ' """'"
,- , SH",", OUENCH )
PREHEAT TOWER )
ZONE "-
~ ~ Ii
FEED-PRODUCT
SEAl. ZONE ~ANGER
OU""'" 1',';;:,':::'1 TOWER ='
."''''' ZON'
'" GAS-O';:-
S""'" m .. ,+- SEPARATOR COOLING FIRED~EATER
ZONE
I MASS FLOW DISCHARGE , CONVEYOR DISCHARGE \ •. __ .! RAW OIL
.~
"'T("'T J Sf'ENT SHALE
LOCKHOf'PERS
WATER SURGE
, "'W ROOUCT GAS
RE(:Y(:LE GAS
C<ON< atr- HIGH PRESSURE TAN' Y Y SPENT SHALE FEEDER
T+ Y I--
G FINES DISCHARGE f IMoe
Figure 77. Retort area flowsheet for a commercial HYTORT facility, (Printed
with the permission of the Institute of Gas Technology,)
OIL SHALE PAGE 281
Figure 78. The PFBH Retort is a vertically-staged fluidized bed unit for
retorting hydrogen-deficient oil shales, such as the Eastern U.S.
Devonian shales, in a hydrogen atmosphere at high pressures.
(Printed with the permission of the Institute of Gas Technology.)
FEED SHALE
RAW SHALE PREHEAT
SHALE HYDRO
RETORTING
SPENT SHALE
COOLING
FLUIDIZING GAS-----'
PRODUCT OIL AND GAS
'-----_ SPENT SHALE
OIL SHALE PAGE 282
Figure 79. Decreasing particle size substantially increases oil yield and
reduces gas yield, while having only a small effect on the overall
~ -z 0 Vl 0:: w > z 0 u z 0 CD 0:: <C u
carbon conversion, in hydrotorting of an Eastern oil shale.
(Printed with the permission of the Institute of Gas Technology.)
90
80
70
60
50
40
30
20
10
o 0.0
• •• • • TOTAL
LIQUID
0 0
(0)
INDIANA NEW ALBANY
H2 35 atm
HEAT -UP RATE: 20°F/min
(0)
00 0
GAS
0.1 0.2 0.3 0.4 0.5 0.6
EQUIVALENT DIAMETER, inches
0
0.7
OIL SHALE PAGE 283
Other advantages of the PFBH vs. the moving bed HYTORT process stem from
the characteristics of the fluidized bed. The efficient gas-solid contacting
and good heat transfer characteristics of the fluidized bed should improve
thermal efficiency and the need to briquette beneficiated shale (required for
moving bed hydroretorting) will be eliminated.
Additional work is underway to develop a fluidized bed gasification system
that will utilize the carbon content of the fines to produce hydrogen. 407 ,408
However 1 the shale processing facility will require process heat and steam, and
the carbon on spent shale is usually the energy source that meets these needs.
Therefore, even if technically successful, the impact of gasification as a
hydrogen source may not be dramatic. However, if chemistry can be found that
will enable the use of synthesis gas to increase the rate and/or lower the
overall pressure requirement for hydrotorting, the impact of gasification may
be very important.
Finally, it should be pointed out that the molecular chemistry responsible
for the striking yield enhancements often obtained by hydrogen retorting are
not yet well understood. Nor, for that matter, are the reasons why shales that
are remarkably similar in elemental and mineral composition sometimes respond
very differently to hydrogen pressure. 408 Perhaps, a better understanding of
the molecular structure of the kerogen, and of kerogen-mineral interactions, in
the Eastern shales might help to clear up some of these mysteries.
OIL SHALE PAGE 284
B. IN SITU RETORTING
In situ retorting involves heating the shale in-place to produce shale oil
and gas. Because the shale remains underground, the requirements for mining,
transporting, crushing and grinding the shale rock are eliminated or greatly
reduced. True in situ retorting also eliminates the costs of an above-ground
retort. Thus, in situ retorting offers the potential for corresponding savings
in both operating and capital costs. However, making an in situ retort work is
generally difficult because most oil shales have very low porosity and almost
no permeability. Without permeability, getting combustion air into the oil
shale formation, or oil and gas products out of the formation is not possible.
Permeability can be induced, for example, by blasting or by injecting high
pressure air or water. However, maintaining such induced permeability is
difficult because shale oil tends to fill the void space in the bed and because
oil shale swells (exfoliates) upon heating. Attempts to induce and maintain
permeability have led to the two main in situ retorting approaches: "true in
situ" methods that involve blasting and/or other fracturing techniques, but no
mining; and "modified in situ" methods that involve mining part of the shale
to generate free underground space followed by blasting to generate a permeable
zone of "rubblized" oil shale for retorting. Bo~h true and modified in situ
methods use a moving flame front to generate heat for retorting, hence are
similar in basic principle to the above-ground N-T-U Retort described earlier.
In a few areas of the Green River Formation, leaching of water-soluble minerals
affords a porous zone of high permeability. One such case is the "leached
zone", located below the rich "mahogany zone" and extending across much of the
Piceance Creek Basin (see Figure 3). The BX In Situ Oil Shale Project was an
attempt to take advantage of this natural permeability.
OIL SHALE PAGE 285
1. laramie True In Situ Retorting Studies
In the early 1960's, Laramie workers began to lay the foundation for
development of in situ retorting technology for Green River oil shale. Data
obtained from the above-ground 10- and 150-ton N-T-U Retorts at Laramie provided
basic information about the ignition of rubblized shale beds and the use of a
moving combustion front to provide heat for retorting. 304 *308 A series of nine
field experiments to explore true in situ retorting were carried out in the
Green River Basin near Rock Springs, WY (Figures 80_81).410w415
The major object of these experiments was to demonstrate that fracturing
of the oil shale formation could induce sufficient permeability to support
underground combustion. A variety of techniques - electrical, hydraulic and
explosive - were evaluated for inducing permeability (Table 49).
Table 49. Types of research carried out at Rock Springs Sites 1-9.
Fracturing Research In Situ Steam Site Electrolinking Hydraulic Explosive Combustion Injection
I X X X
2 X X X
3 X
4 X X X X
5 X X
6 X X X
7 X X X
8 X X
9 X X X
OIL SHALE PAGE 286
Figure 80. The laramie in situ retorting experiments were carried out near
Rock Springs, WY in the Green River Basin, which lies north of the
Piceance Creek Basin where most oil shale activity has been focused.
(From Burwell, et a7., Reference 413.)
r ,OG
i--l~ I ~ JJV//
, EVANSTOth
L_v/&~
UTA H
5,.1_, ... lIn 10 1) 10 20 30
"'"'" 1
Jill!!Q. ~ GIIEEH R!V£R fORMA.TION
N G
liNN IIIVER
RIFLE •
OIL SHALE PAGE 287
Figure 81. layout of the Rock Springs exploratory in situ retorting sites.
(From Burwell, et al., Reference 413.)
Sile I ~
lOO O,-=:::;:'Ci"=::::i'OO ~-: =j
Seole, fee!
OIL SHALE PAGE 288
The major objective of the laramie studies was met: sufficient permeabil
ity was induced for combustion to be initiated, and for the combustion front to
be maintained and moved through the rubblized shale bed. Although post-burn
coring established that significant amounts of shale were retorted in the cases
where jn situ combustion was sustained, oil recoveries were poor. In part,
this was due to recurring problems with pumps that plugged with shale debris.
However, non-uniform combustion and inadequate product containment also contrib
uted to low oil recovery in these exploratory studies.
In parallel with, and following, the test burns, Laramie workers carried
out an extensive program of environmental studies.416-419 These studies
identified the disposition of the produced retort water as a key environmental
concern. This water is produced during thermal retorting and is derived from
dehydration of shale minerals, as well as from combustion. The intrusion of
ground water may also contribute to the water recovered with the oil. Retort
water is odoriferous, yellow to brown in color and contains high levels of both
organic and inorganic dissolved constituents. It is usually quite basic, with
a pH in the range of 8-9.5. The amount of retort water is large; about equal
to the volume of oil produced, in favorable cases. However, testing revealed
that the retort water is not particularly toxic, and indicated that standard
purification techniques should be adequate for its environmentally acceptable
disposal.
OIL SHALE PAGE 289
2. Geokinetics Horizontal In Situ Process
The Geokinetics Horizontal In Situ Retorting Process (HISP) is a true in
situ process, designed to retort shallow shale seams with no mining.420-422
Because no mine construction is involved, a key feature of HISP is its very low
initial capital cost. In 1974, HISP development began as a joint effort of
Geokinetics Inc. and Aminoil USA. In 1976, ERDA (later DOE) joined the effort.
In 1978, Aminoll withdrew, leaving Geokinetics and DOE as the remaining
participants. In 1984, DOE withdrew and Geokinetics completed the project at
its own expense.
In the process. the soil and subsoil are first removed using earthmoving
equipment, then a pattern of holes is drilled through the overburden and shale
seam to be retorted (Figure 82). Explosive charges are then loaded and
detonated sequentially to create the jn sjtu retort. The blasting pattern is
precisely designed so that the initial blast lifts its rubble some 10-20 feet
at the surface.423~426 While this rubble is still aloft, about 0.5 second
after the initial blast, the subsequent rows of charges fire at approximately
0.1 second intervals to laterally displace rich shale into the void. The
result is a disruption of the surface (spalling) only at one end of the retort.
Offgas well are then drilled at the disrupted end and air injection wells at
other. Oil production wells are associated with sumps to facilitate oil
collection (Figure 83). Observation, thermocouple and gas sample wells
complete the required drilling. The subsoil is then replaced and thoroughly
tamped to seal the retort before ignition.
Figure 82. Geokinetics blasting sequence. (a) Site preparation involves removing soil and subsoil, then drilling blastholes. (b) First blast initiates vertical uplift. (c) Subsequent charges displace rich shale laterally into created void space, without disrupting overburden. (d) Retort at the end of the blasting sequence. (After lekas, Reference 420.)
"'''"'''''''''''''-..,.., .... "' .. , .. _ .............. _n'., " .. '''' .... , .. ~ .. O;'_.~ ...... _ ........... _"""'.,_".".' .. "'d •.
f, ..... ,q~' , ... ~ .,_ "', • .,." ...... ' ..... _
o -r
Figure 83. The subsoil is replaced and thoroughly tamped to provide a good seal, before the Geokinetics
Horizontal In Situ Retort is placed in operation. (After Lekas, Reference 420.)
Off"'''-'' to cleanup system
Gas-ol.lt pipe
Oil pump
10il '" stor ... I' I Oil pump (oil to sto'· ... I\
i from blowers
Air inl.t i i
~ ___________________________ 3ro' ____________________________ ~
Side view
10'
OIL SHALE PAGE 292
Geokinetics field studies were carried out at the Seep Ridge site dubbed
"Kamp Kerogen" located south of Vernal, UT. A total of twenty-eight retorts
were designed. Twenty were blasted and burned; six were blasted, but not
burned; and one retort was abandoned prior to blasting. The final five,
Retorts 24-28, were of commercial size.
Oil recoveries from Retorts 25 and 26 were 59% and 51% of Fischer Assay,
respectively, where superficial gas velocities were maintained at or above 0.8.
Due to substantially lower air injection rates, oil recoveries and production
rates from the other retorts were much lower.
Quality of the oil produced by HISP will vary from retort to retort, due
to variations in shale composition, and also varies with time during production
from a given retort. However, a composite analysis of oils from Retorts 27 and
28 shows that oil quality is quite good (Table 50). In particular, the metals,
asphaltenes and residuum contents are desirably low.
OIL SHALE PAGE 293
Table 50. Properties of crude shale oil from the Geokinetics Horizontal In
Situ Retorting Process. 1 (Data of Lekas, Reference 420.)
Gravity 25-26 'API
Viscosity: at 100 'F at 140 'F
Flash Point (ASTM 093-23):
as & W (Maximum):
Ash:
Pour Point:
Aspha Henes:
Elemental Analysis: Carbon: Hydrogen: Oxygen: Nitrogen: Sulfur:
Metals: Iron: Arsenic: Vanadium: Nickel:
Heat of Combustion (Gross):
Distillation (ASTM 01160):
12-16 cSt 6-8 cSt
180-200 'F
1.0 Wt%
0.015-0.030 Wt%
70-80 'F
0.5 - 1.5 Wt%
83.0 - 84.7 Wt% 11.8 - 11.9 Wt% 0.9 - 1.6 Wt% 1.5 1.6 Wt% 0.6 1.0 Wt%
87 8 I 6
740 ppm II 3
58
19,000-19,500 BTU/1b
Vol % IBP 10 30 50 70 90 FBP
'F 160 255 420 470 520 - 580 600 - 675 775 - 790 900 - 920 980 - 1150
(I) Data derived from analysis of raw shale oil samples produced from HISP Retorts #27 and # 28 during Geokinetics' development program.
OIL SHALE PAGE 294
3. Occidental Vertical Modified In Sjtu (VHIS) Process
The true in situ retorting processes discussed above involved drilling,
but no underground mining. In contrast, the modified in situ methods discussed
below involve mining 15-40% of the shale to create void space within the
formation, then blasting to rubblize the remaining shale to fill the resulting
retort. The retort is ignited at the top and burned with a downflow of air
(Figure 84}.427 w 431
In 1972, Occidental began field development of VMIS retorting at its Logan
Wash oil shale mine north of DeBeque, Co. Early work at Logan Wash included
the construction and processing of three small retorts and one commercial-sized
retort. The small retorts were about 30 feet square by 72-113 feet in height.
Retorts IE and 2E used a vertical raise in the center of each retort, with a
single room at the bottom to provide void volume. Three horizontal rooms were
spaced vertically to distribute void volume in Retort 3E, which improved yield
relative to IE and 2E. Retort 4, with two vertical slots parallel to each
other across the retort width, was an attempt to scale up the vertical void
concept. The lower yields obtained from Retort 4 were ascribed to limitations
imposed by rock mechanics that led to non-uniform flows and bed stability
problems.
Retorts 5 and 6 were commercial-sized units that were developed under a
1976 cooperative agreement with US OOE.430-432 Retort 5 used a single vertical
slot mined across its width to provide void volume, while Retort 6 was a scale
up of the Retort 3 design with three horizontal rooms (Figure 85). The latter
design again afforded a superior oil yield.428-429
OIL SHALE PAGE 295
Figure 84. Occidental Oil's Vertical Modified In Situ (VMIS) Retort. (After
McCarthy and Cha, Reference 427. Reprinted with the permission of
the Colorado School of Mines.)
Pillar
Air & Recycle """._ .... Or Steam
- -~ -... -- ,~' - "-
:".~ ........ -.:::-> " -..-~" ~ -
- - - - -- -
"::"- '~, -...........- ",,-
Combustion Zone Front -'
Movement
Retorting Zone
. ~'. . '.
0"
l;a~
t
Pillar
OIL SHALE PAGE 296
AIRL~~~
ILL PILLAR . ___ -'---:OS;;;;;;;;;:;;j UPPER SHALE LAYER
UPPER INTERMEDIATE LEVEL
~~~~~~~~ __ ~~INTERMEOIATESHAlELAYER LOWER::~~~1?~=====
INTERM
"-"t_. 165'
PRODUCT LEVEL 100' x 100'
LOWER SHALE LAYER
BULKHEAD
Figure 85. Isometric view of the Oxy VMIS Retort 6 before blasting, (After
Ricketts, Reference 428. Reprinted with the permission of the
Colorado School of Mines,)
Retorts 7 and 8 were constructed under the second phase of the Occidental-
DOE cooperative agreement. 432 These were identical, commercial-sized retorts,
measuring 165 x 165 feet across x 241 feet tall. They were constructed side- by
side, using the Retort 6 design, and burned at the same time in order to obtain
reproducibility data not available from single retort tests (Figure 86),433'43.
Retorts 7 and 8 afforded even higher oil yields than Retort 6 (Table 51).434
OIL SHALE PAGE 297
Figure 86. Isometric view of Oxy VMIS Retorts 7 and 8, with partial height
Retort 8X in the background. (From Ricketts, Reference 433.
Reprinted with the permission of the Colorado School of Mines.)
-- , "
R8
" ,
, ,
R8X
•
• -,
"
-R7
,
, , , , / / ,
, / . ,
,
OIL SHALE PAGE 298
Table 51. Summary of the retort operating parameters and results for Oxy
Retorts 7 and 8. (After Stevens and Zahradnik, Reference 434.)
Superficial Gas Velocity, ft/min
Air/Steam Ratio, Average
Fischer Assay of Target Rubble, gal/ton
Oil-in-place in Target Rubble, bb 1
Produced Oil, bbl
Yield, % of Fischer Assay
Void Rock Mined, tons
Yield, bbl oil/ton of rock mined
Retort 7
0.54
80/20
17.3
143,200
92,326
64.5
111,604
0.83
Retort
0.54
80/20
16.9
140,100
106,200
75.8
114,042
0.93
8
17.1
283,300
198,526
70.1
225,646
0.88
It is interesting to note that dividing the amount of oil produced in
Retorts 7 and 8 by the amount of shale mined gives a value of 0.88 bbl/ton.
This value compares very favorably with the value of 0.41 bbl/ton that would
have been obtained if the mined shale had been retorted under conditions
affording 100% of Fischer Assay_ In a commercial operation, the mined shale
would probably be processed in such an above-ground retort. In this scenario,
for maximum overall efficiency the bulk of the shale would be retorted using
the cost-effective underground method, while surface retorting of the mined
shale would provide nearly complete resource utilization.
The distance between Retorts 7 and a was 160 feet. To assess the effect
of a shorter inter-retort distance, the partial height Retort ax was
constructed 50 feet behind Retort 8. Retort 8 had the 165 x 165 foot cross
section of Retorts 7 and 8, but was only 63 feet tall. It was blasted first,
so that blasting of Retort a could provide information about the deformation,
OIL SHALE PAGE 299
stresses and fracturing that would occur at the shorter distances called for in
the commercial development plan. No specific problems associated with the
50-foot inter-retort distance have been mentioned, and Retort 8 performance
does not seem to have suffered. However, rock structure is highly site
specific and it would be unwise to assume that such a short distance would be
acceptable in any other site.
4. Rio Blanco Modified In Situ Process
The Rio Blanco Oil Shale Company (RBOSC) was originally a 50/50 joint
venture of Gulf Oil Corporation and Standard Oil Company (Indiana), and is now
a division of Amoco Corporation. It was formed after winning a bid in January
1974 for Federal Oil Shale lease C_a. 437 Site preparation for modified in situ
(MIS) development on Tract C-a began in late 1977. Two retorts were designed,
constructed, then successfully rubblized and burned to demonstrate the RBOSC
MIS technology. The MIS retorting phase was completed during the first part of
1982. Following successful completion of the RBOSC effort, an extensive study
of above-ground retorting based on Lurgi technology, was made at the Gulf Oil
Research Center at Harmarvil1e, PA. Because Tract C-a is suitable for open-pit
mining, current RBOSC interest has shifted to the above-ground technology. The
Tract C-a operation has now been suspended until RBOSC can obtain off-tract
land for disposal of overburden and spent shale. During suspension, RBOSC is
maintaining the leasehold and continuing environmental monitoring work. while
proceeding with internal oil shale research and development programs at the
Amoco Research Centers.
Figure 87 is an isometric view showing the mine and the two experimental
retorts; Retort 0 is to the right of the much larger Retort 1.43' At 400 feet
tall, Retort 1 is the tallest in situ retort ever built.
OIL SHALE PAGE 300
Figure 87. Rio Blanco Oil Shale Company's experimental Retort 0 is to the
right of the much larger Retort 1 in this isometric view, that
a 1 so shows the 1 ayout of the three-l eve 1 mi ne. (From Berry, et
aI., Reference 438. Reprinted with the permission of the Colorado
School of Mines.)
Sicst Holes
Servjce~Escape Shaft
Instrument Shaft
Exhoust Shaft
Off.Ga. Shaft
Production Shaft
. , le .... el
level
Separator
, level
Romp to Shoft Bottom
OIL SHALE PAGE 301
Key features of the RBOSC mine include the production shaft which was
constructed first using conventional shaft-sinking equipment, the exhaust shaft
which was then raise-bored to the surface to establish a conventional ventil
ation system for the mine, and the service/escape shaft which was also raise
bored to the surface. A separate circuit for supplying air and steam to the
retorts through the surface-drilled blastholes, and for piping offgases back to
the surface through the offgas shaft, was sealed off from the working areas of
the mine. a unique feature of the RBOSC design is the underground oil-water
separator that is a mined cavity, that minimizes the capital cost of large
above-ground tankage.
The main production level is the G-level, 850 feet below the surface. The
G-level provided access to the bottoms of the retorts before bulkheads were
installed to seal off the mine from the retorts.
Retort 1 extends through the Upper Aquifer of the Piceance Basin in which
the pressure is several hundred feet of water. Initially a ring of surface
wells was drilled to dewater the area around the production shaft; later the
sub E level was mined and drain holes were drilled into the aquifer. After
water production rates as high as 3600 gal/min., water rates stabilized near
1100 gal/min and only 10-30 gal/min was entering Retort I at its ignition.
Retort 0 was not instrumented. In contrast, Retort 1 was heavily instru
mented with over 150 thermocouples and many pressure taps drilled into the
rubble to provide a clear picture of conditions inside the operating retort.
The RBOSC proprietary rubblizing procedure that affords high void-volume,
hence high permeability, is felt to be the single factor most responsible for
the success of the process. The first step in this procedure is to drill blast
OIL SHALE PAGE 302
holes from the surface to the full depth of the planned retort and undercut a
room that will serve as a product drain is then mined (Figure 88). In Step 2,
the lower part of each blast hole is loaded with explosive that is detonated to
blast a segment of roof down into the room at the bottom. Part of the rubble
is mucked, leaving the rubble at its angle of repose. The load, blast and muck
sequence is repeated until the desired void space is obtained. In Step 3, the
remaining roof rock is rubblized without mucking, until the entire retort is
constructed with only a small free space at the top to serve as a feed gas
distributor.
The RBOSC rubblizing procedure was found to give a relatively small average
particle size with a random size distribution that is conducive to high, even
and maintainable permeability. This was confirmed by extensive cold flow tests
with Freon tracer gases. These tests indicated an overall sweep efficiency of
85% for both retorts. Sweep efficiency in Retort 1 could be investigated in
more detail, because of its instrumentation. In this case, it was found that
sweep efficiency in the top two-thirds was 100%, while that in the lower third
was only 60%. leak rates were measured under stable pressure conditions and
void volumes were calculated from pressure changes with the effluent valves
closed. Cold flow pressure drop as a function of flow rate could be calculated
from the Ergun Equation, using the experimental void volume and a calculated
effective mean particle size.
Figure 88. Construction of RBOSC's high-void retort involves three steps. (From Berry, et al., Reference
438. Reprinted with the permission of the Colorado School of Mines.)
V Explol,ve C~O'ge
Undue",! ROQm
Step 1: Undercut Working room
and raise-bore blastholes.
Step 2: Rubblize lower part of
retort and remove excess rock.
Step 3: Blast upper part to complete
preparation of the retort.
~ "' w
'" w
OIL SHALE PAGE 304
Ignition of the retorts was achieved within 28 hours, using gas-fired
burners lowered downhill. Durability of the burners, a problem in some
other in situ studies, was demonstrated by repeated shut-down/re-ignition
cycles. It was found that the higher pressure drop in ignited rubble tended to
direct flow to unignited rubble, and thereby to promote even spreading of the
f1 arne front.
Stock
Scrubber
Incinerator
Water, Oil to Surface
• Go,
Underground Sepofolor
Flo re
I Air
• steom-L .- Boder
R e 10 rt
Figure 89. Block flow diagram of Rio Blanco production at Tract C-a. (From
Berry, et al., Reference 438. Reprinted with the permission of
the Colorado School of Mines.)
OIL SHALE PAGE 305
Air and steam were fed to the ignited retort through the same boreholes
used for blasting (Figure 89). Products flowed to the underground separator?
from which oil and water were pumped separately to the surface. The philosophy
for operation was to maximize the flame front advance within the safe operating
limits of the available equipment. Consequently, the advance rate decreased
with time, due to the increasing pressure drop from the top of the bed to the
front. Even though some problems that limited advance (air leakage, sulfur
emission limits) were encountered, the front advance averaged about 3 ft/day
and were considerably higher during the early stages (Figure 90).
Oil production, which totaled 24,444 bbl for Retort I, includes liquid oil
from the separator room, the C6 + condensables from the exit gas, and any oil
mist found as an aerosol in the exit gas. The first liquid oil appeared about
15 days after ignition. and thereafter oil production tended to mirror the
flame front advance (Figure 90).
Oil recovery in the RBOSC tests was very good, 68% of Fischer Assay from
each retort (Table 52). The maximum yield value in Table 52 was calculated
using a computer model from Lawrence Livermore National Laboratory which takes
into account the shale grade, particle size, air/steam ratio and front advance
rate. 439 The maximum calculation assumes 100% sweep efficiency. However, it
will be recalled that efficiency was only about 85% for these retorts. Thus,
the experimental values are in reasonable accord with expected behavior.
Figure 90. Flame front advance and oil production in RBOSC Retort 1. (From Berry, et a7., Reference 438.
Reprinted with the permission of the Colorado School of Mines.)
400 -,...~ 40
300 30
-- 24,444 '-c ---0 ........ -- ,;,; • 0 200 ,; 20
0- ;' .... " -C ,,-
0 ",-~
'" u..
",-" 100 _/ 10
.... -",-
/ /
",-
0 0 J u I Aug Sept 0, • Nov De,
1981
• .D
"' '-~
." ~ , ~
> 0 u ~
'" 0
o -r
'" l>
'" m
w 0
'"
OIL SHALE PAGE 307
Table 5Z. Oil Yields and operating parameters for Rio Blanco Modified In Situ
Retorts 0 and I. (Data of Berry, et al., Reference 438.)
Air/Steam Ratio
Front Advance Rate, Avg. ft/day
Actual Oil Yield, bbl
Fischer Assay of Available Rubble, gal/ton
Maximum Calculated Oil Yield, % FA
Actual Oil Yield, % FA
Retort 0
SO/50
2.7
1,876
17.3
70
68
Retort I
70/30
3.0
24,444
21.6
79
68
In summary, the Rio Blanco effort led to a simple design that approaches
laboratory results in oil recovery efficiency. The high void-volume retorts
are suitable for MIS retorting at high burn rates. The entire retort can be
developed from a single mine level (e.g. the G-level); no access to the upper
part of the retort is needed for ignition. Safe and very effective ignition
was achieved using downhole burners. In addition, it has been demonstrated
that successful retorting can be achieved in the dewatered Upper Aquifer. Even
though current Tract C-a interest has shifted, the success of the RBOSC-MIS
effort suggest that the technology may be applicable to other depOSits situated
even deeper beneath the surface than that at Tract C-a.
5. Equity Oil/ARCO BX Project
In the "leached zone" of the Piceance Basin, dissolution of water-soluble
minerals has led to a relatively rich oil shale of high natural permeability.
The Equity Oil BX Project took advantage of this peremeability for in situ
retorting, using injected superheated steam to heat the oil shale. 44o ,441 The
project was carried out on a 1000-acre fee property jointly owned by Equity Oil
OIL SHALE PAGE 308
and Atlantic Richfield (AReO) and located in Rio Blanco County, CO, about
midway between Tracts C-a (Rio Blanco) and Cob (Occidental). At the project
site, the leached zone averages 540 feet thick with shale assaying 24 gal/ton.
The project comprised a pattern of eight injection wells. five production
wells and three observation wells covering an area of 0.7 acres (Figure 91).442
Initial plans were to inject steam at 1000'F, 1500 psig and a total pattern
rate of 974,000 lb steam/day (2,784 bbl/day water). This was to be maintained
over a 2-year period to produce a total of 650,000 bbl of shale oil (100% of
Fischer Assay). Because of equipment limitations and mechanical problems,
these injection targets were not met. As a result, very little shale oil was
actually produced. The initial oil observation was made only after a year of
steam injection and 18 months of steam injection yielded 46 barrels of crude
shale oil.
Not only the quantity, but also the quality, of permeability seems to be
important. The permeability of the leached zone shale is not uniform; it is
appreciably lower in the vertical direction, than in the horizontal direction
(bedding plane). The effects of this anisotropic permeability are readily
apparent in Figure 92, which shows that efficient heating was confined to the
two steam injection zones. These results do not preclude using the Equity Oil
steam injection scheme, but does mean that more complicated and expensive
injection equipment would be needed for distribution of the steam into multiple
injection zones to provide uniform heating of the shale formation.
OIL SHALE PAGE 309
Figure 91. The Equity Oil/ARCO BX project included sixteen wells drilled in a
pattern covering 0.7 acres. (Dougan and Docktor, Reference 442.
Reprinted with the permission of the Colorado School of Mines.)
<:)N
50 fEET (15.2m) I I
SCALE
1¥\2····· '\!J
@
WELL PATTERN lll: II'< SHU Oil SHAU HOl(CI
o INJECTION WEll o PRODUCTION WEll o TEMPERATURE OBSERVATION WEll
OIL SHALE PAGE 310
Figure 92. The temperature profile vs. depth for BX Observation Well No.2,
located near the center of the pattern, shows that efficient
heating was confined to the two steam injection zones. (Dougan
and Docktor, Reference 442. Reprinted with the permission of the
. ~ W o
Colorado School of Mines.)
300,----------------------------------------.
500
700
900
1100
1300
TEMPERATURE OBSERVATION WELL NO 2
JUNE 29. 1980
INJECTION ZONE
INJECTION ZONE
1500~--~----~--~----~----~--~----~--~ 50 100 150 200 250 300
TEMPERATURE. OF
350 400 450
OIL SHALE PAGE 311
In summary, the BX Project attempt to take advantage of the leached zone's
natural permeability for in situ retorting by superheated steam injection did
not result in high oil recoveries. In large part, the BX results were traced
to equipment limitations. However, the results also showed clearly that the
anisotropy of leached zone permeability is a factor to be reckoned with in any
future work along these lines.
OIL SHALE PAGE 312
VII. SHALE OIL UPGRADING AND REFINING
Crude shale oil, sometimes termed retort oil, is the liquid oil condensed
from the effluent in oil shale retorting. Crude shale oil typically contains
appreciable amounts of water and solids, as well as having an irrepressible
tendency to form sediments. As a result, it must be upgraded to a synthetic
crude oil (syncrude) before being suitable for pipelining or substitution for
petroleum crude as a refinery feedstock. However, shale oils are sufficiently
different from petroleum crudes that processing shale oil presents some unusual
probl ems.
A. COMPOSITION
Shale oils, especially those from Green River oil shale, have particularly
high nitrogen contents -- typically 1.7 - 2.2 wt% vs. 0.2 - 0.3 wt% for a
typical petroleum. In many other shale oils (including those from Eastern U.S.
shales) nitrogen contents are lower than in the Green River shale oils, but
still higher than those typical of petroleums. Because retorted shale oils are
produced by a thermal cracking process, olefin and diolefin contents are high.
It is the presence of these olefins and diolefins, in conjunction with high
nitrogen contents, that gives crude shale oils their characteristic instability
towards sediment formation. The sulfur contents of shale oils vary widely, but
are generally lower than those of high-sulfur petroleum crudes and tar sand
bitumens.
Figure 93 shows the nitrogen and sulfur distributions, as a fUnction of
boiling pOint, in raw shale oils produced from the Utah part of the Green River
Formation. 356 Surprisingly, the sulfur content is relatively constant in all
the fractions, while nitrogen is concentrated in the higher boiling fractions.
OIL SHALE
3.0 z w
'" 0 0: .... 2.0 z ...J ;:: 0
1.0 .... ~ 0
I-' 3
O+--------.--------r-----~
gj 1.0 u. ...J ::> (J) 0.5 <f!.
o 30 60 90
I-' 3 0 +----,-----,-----, o 30 60 90
VOLUME % DISTILLED
PAGE 313
Figure 93. Nitrogen and sulfur concentrations in raw Green River shale oil, as
a function of distillation range. The shaded band in each case
shows the range of data from oils produced by the Union-B and
Paraho-DH (directly-heated) retorts. (From Lovell, Reference 356.
Reprinted with the permission of the Colorado School of Mines.)
In addition to the olefins and diolefins mentioned above, The Green River
shale oils contain appreciable amounts of aromatics, polar aromatics and
pentane-insolubles (asphaltenes).356 The distribution of compound types as a
fUnction of distillation range is shown in Figure 94, where the concentration
of polar aromatics and pentane-insolubles in the hig,her-boil ing fractions
parallels the nitrogen concentration in these fractions.
OIL SHALE
100
~ 75 -t":~,, => ..J o > W 50 > I-<l: ..J
i 25 => u
o
PAGE 314
AROMATICS'
30 60 90 MID-VOLUME %
Figure 94. Distribution of compound types in Crude Green River shale oil, as
a function of distillation range. (From Lovell, Reference 356.
Reprinted with the permission of the Colorado School of Mines.)
Oxygen contents are higher than those typically found in petroleum, but lower
than those of crude coal liquids. Crude shale oils also c.ontain appreciable
amounts of soluble arsenic, iron and nickel that cannot be removed by
filtration. The compositions of some representative crude shale oils are
shown in Table 53.
OIL SHALE PAGE 315
Table 53. Compositions of some crude shale oils.
Paraho-DHa Occidental MISb Oravoc CSRd Colorado Colorado Kentucky JUlia Creek
Gravity, "API 20.4 22.9 12.0 10.1 Pour Point, of 85 65 15 -Il Viscosity at IOO"F 213 SSU 15.9 cSt
122"F 44.9 SSU 21. 94 cSt 210"F 5.27 cSt
BS&W, Wt% 0.05 0.2 Conradson Carbon, Wt% 2.98 (r) 1.36 6.9
Elemgntal Analysis Carbon, Wt% 83.68 84.85 83.19 83.0 Hydrogen, Wt% 11. 17 12.27 9.84 9.5 Nitrogen, Wt% 2.02 1. 51 1.14 1.3 Oxygen, Wt% 1.38 0.65 1.97 1.3 Sulfur, Wt% 0.70 0.64 3.86 4.9 Ash, Wt% 0.014 0.092
Metals Arsenic, ppm 28 27.5 100 26 Iron, ppm 53 45 IlO 55 Nickel, ppm 2.4 6.7 3 Vanadium, ppm 0.17 0.42 153 Sodium, ppm 0.3 Il
Hydrocarbon Tyge Saturates, Wt% 11.4 C" = 51.6% Aromatics, Wt% 54.7 H" = 12.4% Polars, Wt% 14.6 Hot "" 3.2% Asphaltenes (c7 ), Wt% 0.89 0.34
Bromine Number 23.6 57
gi~tilhtion {ASTM D-Il60l, "F IBP 90 376
5% 255 467 439 10% 324 570 477 30% 454 670 598 50% 572 712 714 70% 675 820 836 90% 834 953 981 95% 895 EP 972 % Recovered 97 87
(a) Reference 469. (b) Reference 465. (c) Reference 345. (d) Reference 476.
OIL SHALE PAGE 316
B. UPGRADING
Upgradjng, or partial refining. to improve the properties of a crude shale
oil may be carried out with different objectives, depending on the intended use
for the product:
• Stabilization to produce a pipelinable oil that can be transported to a
distant refinery
• More complete upgrading to produce a premium refinery feedstock with low
nitrogen, low sulfur and essentially no residuum
• Upgrading to produce chemical feedstock streams
• Complete refining of the crude shale oil or selected fractions to produce
finished end products {e.g. gasoline, diesel, jet fuel}
It is difficult to generalize regarding shale oil processing. Not only do
the shale oil properties vary, refineries vary widely. For example. there are
about 300 fluid £atalytic £racking (FCC) units in free world refineries -- and
these use more than 260 different cracking catalysts! Therefore, several of
the reported large-scale studies have been selected to illustrate the major
features of shale oil upgrading and refining. These studies have generally
used one of three approaches:
• Thermal Conversion (Visbreaking, Coking) followed by Hydrotreating
• Hydrotreating followed by Fluid CatalytiC Cracking
• Hydrotreating followed by Hydrocracking
Much of the recent information about oil shale upgrading and refining has
come from studies sponsored by the U.S. Department of Defense and Department of
Energy, and carried out under contract by petroleum refiners (Table 54).
OIL SHALE PAGE 317
Table 54. Shale oil upgrading and refining studies sponsored by the U.S.
Department of Defense (DOD) and Department of Energy (DOE).
Processing Program Crude Shale Oil Soyrce Upgrading Contractor
• 1975
Shale I (000) Paraho 1 Gary Western443
• 1975 - 1976
Aviation Turbine Fuels Exxon Research from Synthetic Crudes & Engineering444w446 (DOD)
• 1978
Shale 11 (DOD) Paraho II Sohio (Toledo)447,448
Advanced Catalytic Processes (DOE) Paraho Chevron449-454
• 1978 - 1981
Process Methods Paraho IIjOxy #6 Ashland45So457
Process Methods Paraho IljOxy #6 Suntech455.456,458
Process Methods Paraho I1jOxy #6 Amoco459·461
• 1979 -1981
Catalyst Development Occidental Retort #6 Amoco462w464
• 1980
Shale III Geokinetics HRI - Suntech
OIL SHALE PAGE 318
C. DEWATERING AND SOLIDS REMOVAL
Crude shale oil usually contains emulsified water and suspended solids.
Therefore, the first step in upgrading is usually dewatering/desalting.
Sikonia and his co-workers at UOP found that a conventional two-stage
electric desalter reduced the water content of a crude Occidental MIS shale oil
to 0.05 wt%, but was not able to achieve the 10 ppm toluene-insoluble solids
level desired for hydrotreating. 46s The 100-500 ppm solids left in the feed
necessitated special measures to prevent or alleviate fouling and pressure-drop
problems in fixed-bed reactors.
Sullivan and Stangeland described dewatering of a crude shale oil produced
by the Paraho semi-works retort (indirectly- heated mode).454 As-received. the
crude shale oil was an emulsion containing 6 wt% water and about 0.5 wt% fine
solids. Heating to 170°F broke the emulsion; most of the water and suspended
solids settled out after standing at 170'F for 6 hours. Passage through a 15"
filter removed little additional material, however, a O.45~ filter retained
252 ppm fine solids (Table 55).
OIL SHALE PAGE 319
Table 55. Properties of the dewatered Paraho-IHa crude shale oil that was
upgraded at Chevron's Salt Lake City refinery. (Data of Sullivan
and Stangeland, Reference 454.)
Gravity, °API Pour Point, of Total Nitrogen. Wt% Sulfur. Wt% Arsenic, ppm Iron, ppmb
Carbon, Wt% Hydrogen. Wt% Atomic H/e Ratio Oxygen, Wt% Chloride, ppm Ash, Wt% (ASTM 0-486)
Filter Residue Ash (0.45p filter): Total solids. ppm Ash, ppm
BS&W (sediment +water), Vol% Bromine Number Average Molecular Weight Viscosity, cSt
122'F 210'F
Acid Neutralization Number, mg KOH/g Base Neutralization Number, mg KOH/g pH Maleic Anhydride Number, mg/g Heptane Asphaltenes
(includes any fines), Wt%
ASTM 0-1160 Distillation, 'F IBP/5 10/30 50 70/90 EP % Overhead (excludes trap) % In Trap % In Flask
(a) Produced in the indirectly-heated mode.
20.2 90 2.18 0.66
28 70 84.30 11.29 1.60 1.16
< 0.2 0.03
252 194
0.1 51
326
25.45 5.54 2.3
38 9.2
40.6
0.17
386/456 508/659 776 871/995 1022 94
1 5
(b) This iron was not removed by filtration through a 0.45p filter.
OIL SHALE PAGE 320
D. ARSENIC AND IRDN REMOVAL
If not removed, the arsenic and iron in shale oil would poison and foul
the supported catalysts used in hydrotreating. Because these materials are
soluble, they cannot be removed by filtration. Several methods have been used
specifically to remove arsenic and iron. Other methods involve hydrotreating;
these also lower sulfur. olefin and dialefin contents and thereby make the
upgraded product less prone to form gum.
Workers at Atlantic Richfield (ARCO) found that contacting the crude shale
oil with hydrogen at 800'F and 1400 psig (visbreaking conditions) resulted in
precipitation of arsenic and selenium, and also lowered the pour point of the
011.466 A crude shale oil containing 17.4 ppm arsenic and having a pour point
of 7S e F was processed to obtain an upgraded oil with 2 ppm arsenic and a pour
point of ·30'F.
Dhondt has described Union Oil's treatment to remove solids and arsenic. 364
Two-stage water washing removes suspended solids, which are recycled. Arsenic
is removed by a proprietary, nickel-containing absorbent at 288 8 _343°C in the
presence of hydrogen at moderate pressure.467 ,468 It is claimed that this
absorbent will lower arsenic content from 500 ppm to less than 1 ppm, and that
the absorbent can pick up arsenic to about 80% of its own weight.
For commercial-scale processing (73,000 bbl) of crude Paraho shale oil,
Sohio workers installed a guard bed containing layers of high-surface 1/16-inch
alumina extrudate loaded between layers of low-surface 1/4 and 1/2-inch alumina
balls (Figures 95 and 96),469 No active metal was deposited on the guard bed
packings and the bed was only operated at 420'F. As a result, little or no
arsenic was removed by the guard bed. However, iron was removed as iron pyrite
OIL SHALE PAGE 321
which remained in the guard bed. Also deposited in the guard bed was a black
pitch material, whose formation may have been caused by inadvertent recycle of
sulfuric acid to the shale feed tank during start-up of the acid treater used
as a final polishing step for the jet and diesel fuels. Build-up of pyrite and
pitch was sufficient to cause bed plugging after 24 days on-stream.
Figure 95. Sohio Hydrotreating unit used to process 73,000 barrels of crude
Paraho shale oil into military transportation fuels. (Robinson and
[vin, Reference 469. Reprinted with the permission of Butterworth
n
1
HED DRUM
Publishers.)
HYOROTREATER REACTOR
HYDROGEN RECYCLE
MAKE·UP HYDROGEN CONDENSATE
I ( ~ )T I-
I- V--HIGH o! PRESSURE
SEPARATi~ COAlESCER
:,:\:f I FOUL CONDENSATE
'1 .. 1 'Q L"~ ~~~~riNA BED
"""""'."""~'"
Off GAS
! •
~
STRIPPER COLUMN
-@ ..
CRUDE SHALE SPITlER BOTTOMS LIQUID RECYCLE
=:J. CRUDE SHALE Oil
OIL fIl TER
SPUTTER COLUMN
~ REflUX DRUM
l )
n JET FUEL r--' STRIPPER
r"-Of. STRIPPER
Y
•
GASOLINE RANGE STOCK
JET FUEl
Of.
HEAVY fUEL Oil
OIL SHALE PAGE 322
Figure 96. The guard bed used in Sohio's shale oil hydrotreating study was
packed with alumina extrudate and alumina balls. (Robinson and
Evin, Reference 469. Reprinted with the permission of Butterworth
Publishers.)
OUTLn
OUTLET SCRHN
o SAMPle LOCATIONS
, , o 000 ,
MANHOLE
T Of EXTRUOATES REPLACED AFTER fIRST PlUGGAGE
, 'J, SAllS I , , ,
~~------ 22't, Of ALUMINA EXTRUOAHS ------• .11
DIRECTION Of FLOW <::;========
NOTE: GUARD BED WAS MOUNTED VERTICAllY OURING RUN
TRASH BASKETS INLH
OIL SHALE PAGE 323
E. UPGRADING BY NON-CATALYTIC THERMAL PROCESSES
Thermal conversions, coking and visbreaking, are conceptually simple, 000-
catalytic methods for lowering the high pour 'point and viscosity of raw shale
oils, in order to make the oil more suitable for hydrotreating that is needed
to remove nitrogen and sulfur. Coking also separates suspended solids.
Visbreaking is a mild thermal treatment that lowers viscosity and pour
point, to make the shale oil pipelinable. As noted above, it also precipitates
arsenic, but does little to reduce the contents of nitrogen, sulfur or olefios.
In visbreaking, the oil is heated to 420 0 -480·C for a short time (seconds to
minutes) under hydrogen, during which some product is cracked to gas, along
with the desired cracking. Capital costs for visbreaking are low, but energy
consumption is high for the benefits obtained. As a result, visbreaking is not
a preferred method for upgrading.
Delayed coking followed by hydrotreating was used in the upgrading of
8,505 bb1 of crude Paraho shale oil at the Gary Western Refinery in 1975.443
Hawk and co-workers used coking, followed by severe hydrotreating, to produce
experimental quantities of military jet and diesel fuels in the hydrogenation
pilot plant at USBM's Bruceton, PA faci1ity.47. Delayed coking was studied at
the bench- and pilot plant scales at Chevron451 ,453 and a bench-scale study was
carried out at Phillips Petroleum. 471 Delayed coking was also used to process
about 3400 bb1 of Occidental MIS crude shale oil at Chevron's Salt Lake City
refinery, but in this case the shale oil (13% - 19%) was co-processed with the
refinery's normal petroleum residuum. 453
OIL SHALE PAGE 324
In delayed coking, the oil is heated to 480'C and fed to one of two coke drums
that are typically 10 feet in diameter and 100 feet tall. Pressure in the
drums is sufficient to prevent vaporization, hence coking reactions take place
in the liquid phase. Liquid is continuously withdrawn and coking is allowed to
proceed until the coke drum is nearly full. Feed is then switched to the
second drum and the first is cooled and emptied. Because no catalyst is
involved, coking is very tolerant of metals and solids. However, in the Gary
Western test about 20% of the shale oil feed was converted to low-value gas and
nearly 30% was converted to coke. Thus, the yield of high·value transportation
fuels amounted to only one-half the shale oil fed to the coker. Moreover,
because of its high impurity content the shale-derived coke was not suitable
for making carbon electrodes and could only be used for fuel. In the USBM and
Phillips studies, considerably lower coke makes were achieved, but the coke
makes of 27% and 13%, respectively, still represented a substantial loss of
valuable oil. Consequently, delayed coking cannot be regarded as a preferred
method for shale oil upgrading. Nevertheless, the Gary-Western, USBM and
Phillips refining tests did demonstrate that shale oil can be processed into
transportation fuels, using conventional refining technology with suitable
adjustments of operating parameters.
Two more advanced coking technologies, Exxon's Fluid Coking and Flexicoking
processes warrant mention. Although no reports were found of their application
to shale oil upgrading, they have been successfully used to upgrade the heavy
bitumen from Canadian tar sands, as well as conventional petroleum residua. 472
OIL SHALE PAGE 325
Fluid coking is carried out with steam as a fluidizing gas that also
serves to strip liquid products from the coke. Coke particles are continuously
withdrawn from the cqker and fed to a burner, where they are fluidized in air
and partially burned and heated to about 620'C. The hot coke is returned to
the coker where it heats the incoming feed, leading to deposition of more coke.
Only about 5% of the coke is actually burned to raise heat, so the "fluid coke"
is withdrawn as a product. Compared to delayed coking, fluid coking offers the
advantages of being a continuous and more rapid process. the shorter pyrolysis
time leading to higher liquid yields than delayed coking.
Flexicoking combines fluid coking with gasification of the fluid coke.
The product gas can be used as fuel or reformed to produce hydrogen. Coking,
combustion and gasification are all high-temperature operations with high heat
demands. By integrating all three operations into a single close-coupled unit,
Flexicoking achieves very high thermal efficiency, while retaining the high
liquid yield capability of fluid coking.
F. CATALYTIC PROCESSING METHODS
Hydrotreating is more flexible and less destructive than coking as a way
to remove nitrogen, sulfur, oxygen, arsenic and metals. (Here we discuss the
hydrotreatment of crude shale oil, not the hydrotreatment of sync rude that
would be used to produce finished products for market.) One approach has been
to distill the .crude shale oil, then to hydrotreat the fractions. Shaw et al.
suggested that a more reasonable processing approach would be to hydrotreat the
whole (dewatered/desalted) shale oil and then to process at least the 900'F
product in a conventional refinery.444 This approach was followed by Chevron
OIL SHALE PAGE 326
in their 1977 pilot plant study of Paraho shale oil upgrading,451,453 and Sohio
in refining 73,000 barrels of Paraho oil at Toledo (Figures 94, 95).447,448,469
1. Sohio
The problems caused in the Sohio test by pyrite fines accumulation in the
guard bed were discussed above. The Sohio hydrotreater also accumulated fines.
but did not experience a pressure drop problem during the 24-day test period.
However, catalysts activity declined significantly, as shown by the whole
product nitrogen plot (Figure 97). After completion of the shale oil test. the
top 10% of the catalyst bed was replaced with fresh Shell 324 Ni/Mo catalyst
and the reactor was placed back in service for petroleum hydrotreating. The
reactor was still in service with the same catalyst four years later! These
results demonstrate the need for an effective guard bed to remove arsenic.
Table 56. Summary of products from Sohio refining of crude Paraho shale oil.
(Data of Robinson and Evin, Reference 469.)
Gasoline (including butanes)
Jet Fuel: JP-5
JP-8
Diesel Fuel, Marine (DFM)
Residual Fuel Oil
Total
Barrels
8.473
9.546
490
18.939
37.220
74.939
Volume %
11.96
13.73
25.90
50.91
102.50
% Nitrogen
0.067
0.220
0.430
0.220
OIL SHALE PAGE 327
Figure 97. Increased nitrogen content of the product reflects deactivation of
the catalyst during the 24 days needed to process 73,000 barrels of
crude Paraho shale oil in Sohio's Toledo refinery. Data from the
WHOLE PRODUCT NITROGEN WT.%
0.40
0.30
0.20
0.10
Sohio pilot plant study are included for comparison. (Data of
Robinson and Evin, Reference 469. Reprinted with the permission
of Butterworth Publishers.)
10
----@--- TOLEDO DATA
- -@ - - PilOT PLANT DATA
/'
• ®
/'
®---/' .
/'
•
•
--- -®.--
40 50
CUMULATIVE FEED 11000 BBlS. SHALE OIll
60 70 80
OIL SHALE PAGE 328
Table 57. Properties of raw and upgraded Paraho shale oil. The Sohio oil was
from the refinery test,448 while the LETC oil was from Holmes' study
in a bench-scale unit.473
Paraho Sohio LETC Crude Hydroprocessed Hvdroprocessed
Carbon, Wt% 84.0 86.5 86.1
Hydrogen, Wt% 11.4 13.1 13.3
Nitrogen. Wt% 2.19 0.43 0.42
Sul fur, Wt% 0.66 0.02 0.05
Oxygen, Wt% 2.0 0.12 0.27
HIC Atomic Ratio 1.63 1.82 1.85
Gravity, 'API 21.4' 33.8' 34.9
Pour Point, 'F 85' 85' 70
Average Molecular Weight 311 265 295
Aromatic Carbon, r£ar 21 10 9
Aromatic Hydrogen, Har 5.8 10.8 3.0
Simulated Distillation (GC) Cut Point. 'C Weight % of Shale Oil
38 nil nil 0.3 93 0.2 0.2 0.8
149 0.7 2.8 3.8 204 3.7 7.5 8.2 260 8.9 12.9 15.0 316 11.9 14.0 15.3 371 14.8 18.3 16.4 427 16.3 15.7 15.0 482 17.7 12.4 14.2 538 13.1 5.4 8.3 Residuum 12.7 10.7 2.7
( a) Data from Sohio.
OIL SHALE PAGE 329
Extensive hydrodenitrogenation (HON) studies were carried out by Holmes at
the laramie Energy Technology Center (now Western Research Institute) in a
small continuous unit, but under conditions that closely approximated the Sohio
refinery conditions, except that Shell 514 support balls were used in the guard
bed in the place of SOhlO/S alumina extrudate and no residuum was recycled to
the hydrotreater. 185 ,473,474 Adsorption chromatography on alumina and silica
gel separated the starting Paraho crude shale oil and the Sohio and LETC
upgraded oils into six fractions each, and the nitrogen content of each
fraction was determined (Table 58).
Fourteen nitrogen compound types were identified and quantified in the
fractions, using a combination of infrared, high-resolution mass spectroscopy
and differential potentiometric titration. Because both the Sohio and LETC
hydroprocessing removed about 80% of the total nitrogen, and nearly all the
remaining nitrogen was characterized the nitrogen compound distributions in the
two products can be compared and correlated with hydroprocessing conditions.
The extent of removal can be compared with the overall nitrogen removal of 80%
to evaluate the relative susceptibility of each compound type to HDN. Nitrogen
types that were more than 80% removed were relatively susceptible to HDN. while
those less than 80% removed were relatively resistant to HDN.
Little difference was found in the susceptibility of different nitrogen
base types to HDN; weak bases were only slightly more difficult to remove than
very weak or non-basic species (Figure 98). Within experimental error, removal
of nitrogen bases was the same for both hydroprocessing conditions.
OIL SHALE PAGE 330
Table 58. Weight distributions and nitrogen contents of the fractions obtained
by chromatography of the hydroprocessed shale oils. (Oata of
Holmes, References 473 and 474.)
Hydrocarbon
Pyridine I
Pyrrole
Pyrrole/Arylamine
Pyridine II
Amide/Pyridine III
Total Sample Recovery
Hydrocarbon
Pyridine I
pyrrole
Pyrrol e/Aryl amine
Pyridine II
Amide/Pyridine III
Recovery, % of Total N
Paraho Crude Shale Oil
31.7
16.9
3.56
5.47
19.2
20.8
97.6
Sohio
Wt% of Sample
81.6
2.84
0.48
2.54
2.88
0.45
91'
Nitrogen, Wt% of Fraction
0.003
1.22
1.10
3.92
4.11
4.53
100
0.0006
4.51
0.34
4.65
5.79
5.99
102
LETC
77.6
7.71
2.43
1.03
2.91
1.09
93'
0.0035
1.68
3.62
4.21
3.89
3.29
98
(a) Low recovery in these cases is probably due to evaporation of the lighter
components from the hydrocarbon fraction during removal of chromatographic
solvent.
OIL SHALE PAGE 331
Figure 98. Removal of nitrogen bases was the same for both Sohio and lETC
hydroprocessing conditions. Weakly basic nitrogen was only
slightly more difficult to remove than very weak and non-basic
nitrogen. (Holmes, Reference 473. Reprinted with the permission
*' <0 > 0 E '" a: c:
'" 0> 0 ~
.1:: Z
of the Colorado School of Mines.)
100
80
60
40
20
o L..-_
Sohio
~ LETC
Weak Base
I
I Uncertainty
Weak Base
II
Very Weak Base
Nitrogen Base Types
Nonbasic
OIL SHALE PAGE 332
Significant differences were, however, observed in the distributions of
the individual compound types and ascribed to differences in HDN susceptibility
and hydroprocessing conditions (Figure 99). With the exceptions of very weak
bases of unknown structure (G) and N-alkylcarbazoles (H), the susceptibility of
nitrogen compounds to HDN was similar under both the Sohio and LETC conditions.
It is most difficult to remove hindered alkylpyridines/quinolines (A), basic
alkylpyrrolesjindoles (F), and non-basic alkylpyrrolesjindolesjcarbazolesjben
zocarbazoles (I). The less hindered alkylpyridines/quinolines/acridines were
removed in amounts proportional to the overall nitrogen removal, while the weak
base II compounds of unknown structure (D) were only slightly more difficult to
remove. The easiest nitrogen species to remove were the non-hindered
alkylpyridinesjquinolinesjhydropyridines (C), alkylhydroxypyridines (E) and
alkylcarboxamides/diazaaromatics (J). The very weakly basic compounds of
unknown structure (G) were apparently removed more effectively under the tETe
conditions and the reason is not understood, but the amounts were small.
The buildup of N-alkylcarbazoles during Sohio processing was ascribed to
residuum recycle, since these heavy compounds become concentrated in the
residuum.
Figure 99. Removal of nitrogen compound types by hydroprocessing. Compound types are identified in the
text. (Holmes, Reference 473. Reprinted with the permission of the Colorado School of Mines.)
100
80
,"-20
--40
B
I Sohio
~ LETC
c o E F G
Nitrogen Compound Types
I Uncertainty
H
OIL SHALE PAGE 334
2. Gulf Shale Oil Upgrading Process
The shale oil upgrading process developed by Gulf Oil is often described
as two-stage hydrotreating. but would be more properly termed a three-stage
hydrotreating process (Figure IOO). Data have been reported on upgrading of
both Western (Paraho, Union-B) and Eastern (Oravo/Kentucky shale) shale
oi15. 345 ,475
The pretreating section of the Gulf process is designed to remove arsenic.
iron, trace metals and residual solids, and to stabilize the oil before it
enters the main hydrotreater. The first pretreating reactor is designed to
remove most of the arsenic, iron and solids with minimal hydrogen uptake. It
operates at low hydrogen pressure with a low-cost disposable catalyst, and is
duplicated so that one reactor can be on-stream while the other is being
recharged with fresh catalyst. The second pretreating reactor stabilizes the
oil, removes a substantial part of the sulfur. The main hydrotreating reactor
contains a catalyst selected for high hydrodenitrogenation activity. and
produces an effluent that typically contains 7S0 ppm nitrogen. This effluent
is fed to the flash separator, from which the 6S0°F- vapors enter a vapor
hydrotreater. Using about IS% of the total hydrogen, this final stage yields
,37soF- naphtha that typically contains < 4 ppm nitrogen, hence can be used as
reformer feed and 37S' - 6S0'F distillate that contains 10-IS0 ppm nitrogen.
The 6S0'F+ separator bottoms typically contain 1000-2000 ppm nitrogen and is a
good cracking feedstock.
Figure 100. Flow diagram of the Gulf Shale Oil Upgrading Process. (Jones, et al., Reference 345. Reprinted
with the permission of the Colorado School of Mines.)
r-------------! PRETREATING I
1 1 1
OAT OIL: 1
1
~ J 1
I ~ 1 1 1 1 1 · · • 1
1 • • ~ 0 0 1 1 • • • u u u 1 1 < < < • • • • « • I
1 1 1 r 1 l---c 1 1 1 1
1 1
1 1 1
1 1 1 1 ____ - ________
r-------I BULK I
HYDROTREATIHO I 1 I I
1 1
1 1 6 1 1
I 1 1 1 g I 1 u
1 < 1 • • 1 1 • < I 1 , 1 1 1 1 Y I 1
1 1 1 1
r--------l FLASH VAPOR
I HYDAOTRU TlNO I 1 1 1 1 1 I 1 I 1 1 1 1 1 1
~ I 1 1 • 1 o. I • 0 <. 1 >u I % < I •• 1 ~«
• 1 1 ...--c 1 1 r-; 1 1 1
i~ I I I
1 : <. J< 1 ••
I • b 1 L _ ______ I
I 1 1 : 1 1--- ______ 1
SVNCAUOE
OIL SHALE PAGE 336
3. Catalytic Cracking (Ashland, Chevron)
Fluid catalytic cracking (FCC) of shale oils has been studied by Moore
and his co-workers at Ashland Petroleum. 4s7 Pilot plant FCC studies were made
by Chevron. 449 ,451,453 Sikonia and co-workers at UOP have carried out small
scale studies of hydrotreating as a pretreatment for FCC.465 Both bench-scale
and pilot plant FCC studies of shale oil were carried out at Gulf. 345
In the FCC process, relatively small (40 - SOp) catalyst particles are
fluidized by upflowing shale oil vapors in the reactor vessel, which is
maintained at 950' . 1000'F (Figure 101). As the cracked vapors exit the
reactor, they pass through cyclones, where the catalyst particles are separated
from the products and recycled. Along with the desired cracking reaction, some
coke is deposited on the catalyst surface. Therefore, catalyst is continuously
withdrawn from the cracking reactor and transported by an air stream into a
combustor where the coke is burned. When returned to the cracking reactor the
hot, regenerated catalyst supplies heat for vaporization of the feed.
Shale oils are rich in high-molecular weight, waxy paraffinic material.
Thermal cracking lowers molecular weight, but yields straight-chain products of
low octane number. Fluid catalytic cracking not only lowers molecular weight,
but also causes isomerization to produce branched products with higher octane
numbers. As a result, the so-called "cat cracked" shale naphtha is a more
desirable feedstock for hydrotreating to make gasoline blend stock, than is the
naphtha from thermal cracking or coking of shale oil.
OIL SHALE PAGE 337
FLUE_--, GAS
,.......-- REACTOR PRODUC T
AIR
. " , I,::,: .... :. .... ,. ' ..
.' " .' ,' ... . '.'
mJIo--CYCLONES
REGENERATOR II
I ...... --DIP LEGS REACTOR ~Y-1
.''-:.:. : ,," ..... .: .. : ", "~'.
'", '. ';" ",,' .... .
t" STRIPPING j """ GAS _~
STANDPIPES -1 TRANSFER
GRID
OIL FEED
'\..--,,- LI NE S ---;~J (RISERS)
Figure 101. Fluid catalytic cracking unit.
Hydrotreating of the raw shale oil to remove basic nitrogen that would
poison the acidic cracking catalyst, and would also promote unwanted coking, is
required as an FCC pretreatment. 465 Workers at Gulf found that hydrotreated
syncrudes from Union-B (Colorado shale) and Dravo (Kentucky shale) retorts
performed well as FCC feeds in a MAT test, and the atmospheric tower bottoms
from a Paraho syncrude gave good results in an FCC pilot plant test.'4S
OIL SHALE PAGE 338
4. Amoco Hydrotreating-Hydrocracking
Under an Air Force contract, Tait. Miller and Hensley at Amoco investigated
all three of the upgrading routes outlined at the start of this section. 459 ¥464
Of these, hydrotreating followed by hydrocracking offered the most flexibility
and was the only approach to efficiently maximize jet fuel production. This is
an important consideration, since Western U.S. shale oils characteristically
have high contents of straight-chain, waxy paraffins and low contents of the
aromatics that are undesirable in jet fuel. (Aromatics set the smoke point in
jet fuel.) Coal liquids, on the other hand, are characteristically highly
aromatic and well suited for processing into gasoline blend stocks. Thus, its
molecular structure makes shale oil a prime source for jet fuel.
As part of the Air Force contract, the Amoco workers developed catalysts
that are capable of direct upgrading of a whole shale oil into high yields of
JP-4 jet fuel boiling range material in a single-stage process. In contrast to
the multi-step processes discussed above, the Amoco catalyst has the ability to
sequentially saturate, denitrogenate, and crack, the whole shale oil in the
presence of contaminants such as ammonia, water and basic organic nitrogen
compounds, while maintaining high selectivity towards the jet fuel boiling
range materials.
The catalyst development studies were carried out on an Occidental MIS
crude shale oil that contained 1.32 wt% nitrogen, 0,64 wt% sulfur, 1.33 wt%
oxygen. 26 ppm arsenic and about 60 ppm iron. Although it was recognized that
an effective guard bed would be required for commercial operation, none was
used in the development study. nor was the oil pretreated to remove arsenic or
iron.
Screening of existing catalysts revealed that a proprietary catalyst
containing cobalt, chromium and molybdenum could effectively remove nitrogen
OIL SHALE PAGE 339
and effect moderate conversion of the feed into the JP-4 boiling range. By
screening catalysts in which the metals loading was systematically varied, it
was found that optimal metal oxide loadings were 1.5% cobalt oxide. 10%
chromium oxide and 15% molybdenum oxide. The high loading of chromium oxide
actually lowered denitrogenation activity, but was necessary to impart high
temperature stability.
A study of different supports revealed that 20% silica/alumina was better
than alumina, and that 50% ultrastable molecular sieve/alumina was more active
yet. Next, a series of eight supports containing 50% ultrastable sieve, but
having varying surface area (SA), pore volume (PV) and average pore diameter,
were prepared. These variations were achieved by modifying the alumina
component, since modification of the sieves, themselves, could destroy their
inherent cracking activity. Evaluation of catalysts prepared from these
supports revealed that optimum activity was obtained with a catalyst having a
narrow pore size distribution and APD of about 7 nm. Variations in product
yields and hydrogen consumption, as a function of support are summarized in
Figures 102 and 103.
Activity testing was plagued by several upsets, yet the optimized catalyst
proved very durable. 464 Based upon the temperature response factor calculated
from data obtained during the activity tests, an expected catalyst life of
about 4.5 months was estimated for a constant JP-4 yield of 75% and reactor
temperatures increasing from 775' to 800·F (686· to 700·K). Because of the
upsets encountered during testing. this was judged to be a minimum life
expectancy and a catalyst life of six months was considered probable.
OIL SHALE PAGE 340
Figure 102. The effect of ultrastable molecular sieve concentration in the
support on the product distribution from single-stage hydrotreating
hydrocracking of Occidental MIS crude shale oil. (Data of Tait,
Miller and Hensley, Reference 461.)
oor---------------------------~
+"' c: 60 (I) JP·4 (.) ... (I) Co +"'
~ c: 40 0 .-+"' (.) (t) ... u.. Diesel
20 Gas oil
o~~--__ ~~------~------~ 20 30 40 50
Wt percent sieve in alumina
OIL SHALE PAGE 341
Figure 103. Correlation of JP-4 jet fuel yields with total hydrogen consumption
for catalysts containing 0%, 20%, 30% and 50% ultrastable molecular
sieve (left to right) in the support. (Data of Tait, Miller and
Hensley, Reference 461.)
80r-----------------------------------~
70
60
... C Q) 50 u ... Q)
c. ... 40 ~ ..r , c.. 30 '"")
20
10
1300 1400 1500 1600 1700
Hydrogen consumption, SCFB
OIL SHALE PAGE 342
Separate distillations of the product were carried out to produce JP-4 and
JP-S jet fuels. The analytical data indicate that the samples would meet all
specifications, with perhaps one exception (Table 59). The pour point of -40 G F
(233'K) for the JP-S fraction is somewhat higher than the JP-S specification of
-5S'F (223'K). A slight lowering of the boiling point range would lower the
pour point to give a product that meets all JP-B specifications.
The results demonstrated that the single-catalyst system was capable of
hydrocracking a whole shale oil containing large amounts of nitrogen. However,
analysis of kinetic data from the studies above led to the development of a
dual-catalyst system that affords the same high product quality, plus higher
throughput. 464 Using the dual-catalyst system with recycle of the 520'Ft
distillate (#2 fuel oil) afforded a 91 wt% (lOS vol%) yield of JP-4 boiling
range material as the only liquid product. With recycle operation, the gas
make increased slightly, based on fresh feed, from 6.1 wt% to 7.9 wt%. As a
result. the hydrogen consumption also increased on a fresh feed basis from
ISOO scf/bbl to 2060 scf/bbl. However, because of the higher yield obtained,
the hydrogen consumption dropped on a JP-4 basis from 2320 scf/bbl of JP-4 to
1920 scf/bbl for the recycle operations.
Thus, the Amoco work led to novel catalyst systems capable of directly
upgrading a whole shale oil into high yields of jet fuels in a single step.
First a single-catalyst hydrocracking system was developed. Then an even more
active dual-catalyst system was developed with a catalyst specifically designed
for high denitrogenation activity at high nitrogen content preceding the
hydrocracking catalyst. The development of these catalysts is significant,
since their use could eliminate the need for complex, multi-step, multi-catalyst
processing of whole shale oils.
OIL SHALE PAGE 343
Table 59. Properties of the Occidental MIS crude shale oil feed, the composite
product and jet fuel fractions from the hydrocracked shale oil.
(Data of Tait, Miller and Hensley, Reference 464.)
Co/Cr/Mo on Alumina, 1800 psi Hz, 0.55 LHSV, 790°F, Hz Consumed = 1400 scf/bbl
Occidental MIS Composite Crude Shale Oil Feed Product JP-4 JP-8
Gravity, • API 12.8 39 49.4 (49-57) 43.4 (37-51)
Weight, % 100 76 61
Pour Point, of 60 -5 -85 -40
Aromatics, Vol % 16.0 (25.0) 18.0 (25.0)
Olefins, Vol% 1.0 ( 5.0) 2.5 ( 5.0)
Nitrogen, ppm 13,200 1.1 0.7 1.1
Carbon, wt% 85.82 85.99 86.10
Hydrogen, wt% (AtomiC H/C = 1.67) 14.17 14.00 (13.6) 13.86 (13.6)
Sulfur, wt% 0.64
Oxygen, wt% 1.33
Distillation, 0-2887 IBP, OF -47 22 250 10% 203 190 322 (367) 20% 268 238 (266) 353 30% 321 276 390 40% 372 312 413 50% 410 346 (365) 436 60% 446 377 461 70% 487 408 489 80% 547 440 520 90% 624 480 (482) 564 EP, OF 789 553 (608) 622 (626)
Values in parentheses limits.
are maximum (minimum for hydrogen content) specification
OIL SHALE PAGE 344
CONCLUSIONS
The oil shale challenge is clear. Sooner or later, we will need to make
use of the world's abundant oil shle resource -- if not for fuel, then for
chemical feedstock -- to replace dwindling petroleum supplies. The timing will
probably be dictated more by political cosiderations than by purely technical
factors. As a result, our need to use oil shale may come upon us suddenly,
giving no time for the development of new technology. We must be ready.
Technology is now available that could be used to provide liquid fuels and
feedstock from oil shale, but is expensive. A rapid shift to such technology
could have dire and far-reaching consequences. Continued research and
development is needed to provide further improvements in oil shale science and
technology, so that when the need arises we can use oil shale economically and
in an environmentally acceptable manner.
OIL SHALE PAGE 345
REFERENCES
I. Russell, P. L. 19th Oil Shale Symp. Proc.; Gary, J. H., Ed.;
CSM Press: Golden, CO, 1986; 94-97.
2. Eele, M.; Hancock, T.; Portlock, W. U.K. Patent 330; 1694.
3. Duncan, O. C.; Swanson, V. E. U.S. Ge01. Surv. Gire. 523, 1965.
4. Silliman, D. Amer. J. Science and Arts. 1869, 4-8, 86-92.
5. Cane, R. F. 12th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press:
Golden, CO, 1979; 17-25.
6. Prien, C. H. In Oil Shale; Yen, T. F.; Chilingarian. G. V., Eds.;
Elsevier: Amsterdam, 1976; 235-267.
7. Schmitz, H. Ho; Tolle, A. Erodel, Kahle, Erdgas, Petrochem. 1987,
40, 245-252.
8. Hutton, A. C.; Saxby, J. 0.; Ellis, J. 20th Oil Shale Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1987; 24-40.
9. Gavin, J. M. Oil Shale; Government Printing Office: Washington, 1924.
10. The term "kerogen" was coined by Crum-Brown in 1912, specifically to
denote the insoluble organic matter in oil shale. Steuart, O. R. In,
The Oil Shales of the Lothians, Geol. Surv. Hem. Scotl. Part III;
H. M. Stationery Office: London, 1912.
11. Hunt, J. M. Petroleum Geochemistry and Geology; W H Freeman & Co.:
San Francisco, 1979.
12. Ward, C. R. Coal Geology and Coal Technology; Blackwell: Melbourne, 1984.
OIL SHALE PAGE 346
13. Tissot, B. P. j Welte O. H. Petroleum Formation and Occurrence, 2nd Ed.;
Springer-Verlag: Berlin: 1984.
14. Thorne, H. M. U.S. Bur Mines Inf eire. No. 8216, US Government Printing
Office: Washington, 1964.
15. Demaison, G. P.; Moore, G. T. AAPG Bull. 1980, 64, 1179-1209.
16. Himus, G. W. "Observations on the Composition of Kerogen Rocks and the
Chemical Composition of Kerogen" In Oil Shale and Cannel Coal; Institute
of Petroleum: london, 1951; Vol. 2, 112-135.
17. Hutton, A. Co; Korth, J.; Ellis, J.; Crisp, P.; Saxby, J. O. 20th Oil
Shale Symp. Proc; Gary, J. H., Ed.; CSM Press: Golden, CO, 1987; 10-23.
18. (a) Bradley, W. H. U.S. Geol. 5urv. Prof. Paper 168, 1931.
(b) Bradley, W. H. U.S. Geol. Surv. Prof. Pap. 15B-E, 1929.
19. (a) Bradley, W. H. Geol. Soc. Am. Bull. 1970, 81, 985-1000.
(b) Bradley, W. H. Ibid., 1973, 84, 1121-1124.
20. Milton, C. Wyoming University Contrib. Geol. 1971, 10(1) 57-63.
21. Smith, J. W. Chapter 12 in Geochemistry and Chemistry of Oil Shales
Miknis, F. P.; McKay, J. F., Eds.; ACS Symposium Series 230; American
Chemical Society: Washington, 1983; 225-248.
22. Smith, J. W. Oil Shale Resources of the United States, Hineral and
Energy Resources Series, Vol. 23, No.6; CSM Press: Golden, CO, 1980.
23. Tisot, P. R; Murphy, W. I. R. U.S. Bur. Hines Rept. Invest. 6284, 1963.
24. McDonald, R. E. "Eocene and Paleocene Rocks of the Southern and Central
Basins B, In Geologic Atlas of the Rocky Mountains Region; Rocky
Mountain Association of Geologists: Denver, CO, 1972, 243-256.
25. Bradley, W. H. AAPG Bull. 1925, 247.
OIL SHALE PAGE 347
26. Garrels, R. Hj Mackenzie, F. T. Evolution of Sedimentary Rocks;
W. W. Norton: New York, 1971.
27. Pettijohn, F. J. Sedimentary Rocks, 2nd Ed; Harper & Row: New York, 1957.
28. Krurnbein, W. C.; Garrels, R. M. J. Geol. 1952. 60, 1-33.
29. Garrels, R. H.; Christ, C. L. Solutions, Minera1s and Equilibria;
Harper & Row: New York, 1965.
30. Baas Becking, L. G. M.; Kaplan, I. R.; Moore, D. J. Geol., 1960,
68, 243-284.
31. Menzel, D. W. "Primary Productivity, Dissolved and Particulate Matter and
the Sites of Oxidation of Organic Matter." In The Sea. Marine Chemistry;
Goldberg, D., Ed.; Wiley: New York, 1974; Vol. 5, 659-678.
32. Sorokin, J. I. Rev. Ges. Hydrobiol. 1971, 56, 1-48.
33. Tappan, H; Loeblich. A. R., Jr. "Geologic Implications of Fossil Phyto
plankton Evolution And Time-Space Distribution" In Symp. Palynology of
the Late Creaceous and Early Tertiary; Kosanke, R. H.; Cross, S. T., Eds.;
Geological Society of America: 1970; 247-340.
34. Moore, L. R. "Geomicrobiology and Geomicrobiological Attack On Sedimented
Organic matter" In Organic Geochemistry; Eglinton. G.; Murphy, M. T. J.,
Eds.; Springer-Verlag: 8erlin, 1969; 265-303.
35. Strickland, J. D. H. "Production of Organic Matter In The Primary Stages
of the Marine Food Chain" In Chemical Oceanography; Riley, J. P.;
Skirrow, G. Eds.j Academic Press: New York, 1965; Vol. I, pt 12, 478- 595.
36. Durand, S., Ed. Kerogen; Editions Technip: Paris, 1980.
37. Eglinton, G. j Murphy, M. 1. J. Organic Geochemistry; Springer-Verlag:
8erlin, 1969.
OIL SHALE PAGE 348
38. Broughton, A. C. Trans. Proc. R. Soc. S. Aust. 1920, 44, 386.
39. Ormeland, R. Private communication.
40. Brooks, J., Ed. Organic Maturation Studies and Fossil Fuel Exploration;
Academic Press: London, 1981.
41. Brooks, J.; Welte, O. Advances in Petroleum Geochemistry; Academic Press:
London, 1984.
42. Demaison, G.; Holck. A. J. J.; Jones, R. W.; Moore, G. T. Proe 11th World
Petroleum Congress; Wiley: Chichester, 1984; Vol. 2, 17-29.
43. Bates, R. L., Jackson, J. L., Eds. Glossary of Geology, 2nd Ed.;
American Geological Institue: Falls Church, VA, 1980.
44. Debyser, J.; Deroo, G. Rev. Inst. Fr. Petrol. 1969,24(1), 21-48.
45. Trask, P. D. Origin and Environments of Source Sediments of Petroleum;
Gulf Publishing: Houston, 1932.
46. Bader, R. G.; Hood, B. W.; Smith, J. B. Geochim. Cosmochim. Acta 1960,
19, 4.
47. Bordovsky, O. K. Nar. Geol. 1965, 3, 3-114.
48. Cane, R. F. "Origin and formation of Oil Shale" In Oil Shale;
Yen, T. F.; Chilingarian, G.V., Eds.; Elsevier: Amsterdam, 1976.
49. Degens, E. T. "Biogeochemical Alterations in the Early Stages of
Diagenesis" Proc. 6th International Congress on Sedimentology,
Amsterdam-Brussels, 1963.
50. Degens, E. T.; Prashnowsky, A.; Emery, K. 0.; Pimenta. J. Neues Jahrb.
Geol. U. Palaentol. Honatasch 1961, 8, 413-426.
51. Degens, E. T.; Reuter, J. H.; Shaw, K. N. F. Geochim. Cosmochim. Acta
1964, 28, 45-66.
OIL SHALE PAGE 349
52. Degens, E. T.; Mopper, K. "Factors controlling the distribution and early
diagenesis of organic material in marine sediments." In Chemical0cean
ography; Riley, I. R.; Chester, R., Eds.i Academic Press: New York, 1975.
53. Abelson, P. H.; Haering, T. C.; Parker. P. L. "Fatty Acids in Sedimentary
Rocks" In Advances in Organic Geochemistry 1963; Colombo, U.; Hobson.
G. 0., Eds.; Pergamon Press: London, 1964; pp 169-174.
54. Leo, R. F.; Parker, P. l. Science 1966, 152, 649-650.
55. Cranwe 11, P. A. "Alkyl Esters in Recent Sediments of Two Product i ve Lakes"
In Advances in Organic Geochemistry 1981; Bjorpy, M.; et a1., Eds.;
Wiley: Chichester, 1983; pp 299-308.
56. Schonbrunner, J. J. Biochem. Z. 1940, 304, 26-36.
57. ZoBell, C. E. Annual Report for 1944, API Project 34A; American
Petroleum Institute: Washington, 1944.
58. Rosenfeld, W.O.; Archiv. Biochem. 1948, 16, 263-273.
59. Repeta, D. J.; Gagosian, R. B. "Carotenoid Transformation Products in the
Upwelled Waters Off the Peruvian Coast" In Advances in Organic Geochem
istry 1981; Bjorpy, M.; et al., Eds.; Wiley: Chichester, 1983.
60. Watts, C. 0.; Maxwell, J. R. Geochim. Cosmochim. Acta 1977, 41, 493- 497.
61. Watts, C. D.; Maxwell, J. R.; Kjosen, H. "The Potential of Carotenoids
as Environmental Indicators" In Advances in Organic Geochemistry 1975;
Campos, R.; Goni, J., Eds.; Enadamis.: Madrid, 1977; pp 391-414.
62. Kimble, B. J.; Maxwell, J. R.; Philp, R. P.; Eglinton, G.; Albrecht, P.;
Ensminger, A.; Arpino, P.; Ourisson, G. Geochim. Cosmochim. Acta 1964,
38, 1165-1181.
OIL SHALE PAGE 350
63. Costa Neto. C. "Geochemistry of Brazilian Oil Shales" In Geochemistry
and Chemistry of Oil Shales Miknis, F. P.i McKay, J. F., Eds.; ACS Sym
posium Series 230; American Chemical Society: Washington, 1983; pp 13-35,
and references cited therein.
65. (a) Williams, P. F. V.; Douglas, A. G. "Kimmeridge Oil Shale. A Maturation
Study" In Organic Maturation Studies and Petroleum Exploration;
Brooks, J., Ed.; Academic Press: london. 1981; pp 255-269.
(b) Williams, P. F. V.; Douglas, A. G. Fuel 1985, 64, 1062-1069.
66. Williams, P. F. V. Fuel 1987, 66, 86-91.
67. Baird, R. A. AAPG Bull. 1986, 70, I-II and references cited therein.
68. Hall, P. B.; Douglas, A. G. "The Distribution of Cyclic Alkanes in Two
Lacustrine Deposits" In Advances in Organic Geochemistry 1981;
Bjorpy, M.; et al., Eds.; Wiley: Chichester, 1983; pp 576-587.
69. Donovan, R. N.; Scott, J. J. Geol. 1980, 16, 35-72.
70. Anders, D. E.; Robinson, W. E. Geochim. Cosmochim. Acta 1971, 35, 661-678.
71. Bendoraitis, J. G. "Hydrocarbons of Biogenic Origin in Petroleum - Aromatic
Triterpenes and Bicyclic Sesquiterpenes" In Advances in Organic Geochem
istry 1973j Tissot, B.; Bienner, F., Eds.; Editions Technip: Paris,
1974; pp 209-224.
72. Philp, R. P.; Gilbert, T. 0.; Friedrich, J. Geochim. Cosmochim. Acta
1981, 45, 1173-1180.
73. Duncan, O. "Geologic Settings of Oil-Shale Deposits and World Prospects"
7th World Petroleum Congr. Proc.; Elsevier: Amsterdam, 1977; Vol. 3,
pp 659-667.
74. Jaffe', F. C. Colo. Sch. Mines Min. Ind. 8ull. 1962, 5(2), I-II.
OIL SHALE PAGE 351
75. Vyas, K. C. "Syncrudes From Eastern Oil Shales" In Proc. 14th Oil
Shale Symp.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1981; 82-90.
76. Claypool, G. E.; Love, A. H; Maughan, E. K. AAPG Bull. 1978, 62, 98-120.
77. Willmon, G. J. Proc. 12th World Petrol. Congr.; Wiley: New York,1987;
Vol. 4, 3-13.
78. Fischer, F.; Schrader, H. Angew. Chem. 1920, 23, 172.
79. Davis, J. D.; Galloway, A. E. Ind. Eng. Chem. 1928, 20, 612.
80. (a) Stanfield, K. E.; Frost, I. C. US Bur. Hines Rept. Invest. 4417,
1949. (b) Stanfield, K. E.; Frost, I. C.; McAuley, W. S.; Smith, H. N.
U.S. Bur. Hines Rept. Invest 4825, 1951.
81. Heistand, R. N. Energy Sources 1976, 2, 397.
82. Hubbard, A. B. US Bur. Mines Rept. Invest. 6616, 1965.
83. Goodfellow, L.; Atwood, M. T. Colo. Sch. Hines Quart. 1974, 69(2), 205-219.
84. Singleton, M. F.j Koskinas, G. J.; Burnham, A. K.; Raley, J. H. "Assay
Products from Green River Oil Shale" Lawrence Livermore National Labor
atory Report 1986, UCRL-53272, Rev. I.
85. Espitalie, J.j laPorte, J. L.j Madec, Hoi lePlat, P.; Paulet, J.j
Boutfeu, A. Rev. Inst. Fr. Pet. 1977, 32, 23-42.
86. Williams, P. F. V. Fuel 1983, 62, 756-771.
87. Robinson, W. E.; Dinneen, G. U. Proc. 7th World Petrol. Congr.;
Elsevier: Amsterdam, 1967; 669-680.
88. Cummins, J. J.; Robinson, W. E. J. Chem. Eng. Data 1964, 9, 304-307.
89. Bray, E. Eoj Evans. E. O. Geochim. Cosmochim. Acta 1961, 22, 2-15.
90. Reference 13, pp 100-109.
91. Koons, C. B.; Jamieson, G. W.; Cierszko, L. S. AAPG Bull. 1965, 49, 301-304.
OIL SHALE PAGE 352
92. Hills; I. R.; Whitehead, E. V.; Anders, D. E.; Cummins, J. J.; Robinson,
W. E. Chem. Commun. 1966, 752-754.
93. Miknis, F. P.; Maciel, G. E.; Bartuska. V. J. Org. Geochem. 1979, 1,
169-176.
94. Haciel, G. E.; Dennis, L. W. Org. Geochem. 1981, 3, 105-107.
95. Resing, H. A.; Garroway, A. N.; Hazlett, R. N. Fuel 1978, 57, 450-454.
96. Hagaman, E. W.; Schell, F. H.; Cronauer, D. C. Fuel 1984, 63, 915-919.
97. Vuclelic, D.; Juranic, N.; Vitorovic, D. Fuel 1979, 58, 759-764.
98. Botto, R. Eo; Wilson, R.; Hayatsu, R.; McBeth, R. L.; Scott. R. G.;
Winans, R. E. AC5 Oiv. Fuel Chem. Prepr. 1985, 30(4), 187-192.
99. Dereppe, J.-H.; Horeaux, C. Fuel 1987, 66, 1008-1009.
100. Huntean, J. V.; Stock, L. H.; Botto, R. E. Energy & Fuels 1988, 2, 108-110.
101. Haciel, G. E.; Bartuska, V. J.; Hiknis, F. P. Fuel 1979, 58, 155-156.
102. Haciel, G. E.; Bartuska, V. J.; Hiknis, F. P. Fuel 1978, 57, 505-506.
103. Miknis, F. P.; Smith, J. W.; Maughan. E. K.; Maciel, G. E. AAPG Bull.
1982, 66, 1396-1401.
104. Opella, S. J.; Frye, H. J. J. Amer. Chem. Soc. 1979, 101, 5854-5856.
105. Soderquist, A.; Burton, O. J.; Pugmire, R. J.; Beeler, A. J.;
Grant, D. H.; Durand, B.; Hue, A. Y. Energy & Fuels 1987, 1, 50-55.
106. Sethi, N. K.; Grant, D. H.; Pugmire, R. J. J. Hagn. Res. 1987, 71, 476.
107. Parks, T. J.; lynch, L. T.; Webster, D. S. Fuel 1987, 66, 338-344, and
references cited therein.
108. Silbernagel, B. G.; Gebhard, l. G.; Siskin, H.; Brons, G. Energy & Fuels
1987, 1, 501-506.
OIL SHALE PAGE 353
109. Sousa, J. J. F.; Vugman, N. N.; Mangrich, A. S. Chern. Geol. 1987,63, 17-20.
110. Frost, I. C.; Stanfield, K. E. Anal. Chern. 1950, 22, 491-492.
III. Smith, J. W. Ind. fng. Chern. 1956, 48, 441-444.
112. Smith, J. W. US Bur. Mines Rept. Invest. 7248, 1969.
113. Vadovic, C. J. "Sink Float Procedures for Shale Characterization" In
Geochemistry and Chemistry of Oil Shales Miknis, F. P.; McKay, J. F .• Eds.;
ACS Symposium Series 230; American Chemical Society: Washington, 1983;
385-396.
114. For a review of equations for calculating heating value (often termed
"calorific value" in the coal literature), see: Elliott, M. A.; Yohe, R.
In The Chemistry of Coal Utilization; Elliott, M. A., Ed.; Wiley: New
York, 1981; 2nd Supplementary Volume; pp 26-27.
115. Boie, W. fnergietechnik 1953, 3, 309.
116. Rajeshwar, K.; Nottenburg. R.; DuBow, J. J. Hater. Sci. 1979, 14,
2025-2052.
117. Rajeshwar, K. Thermochim. Acta 1983, 63, 97.
118. Hubbard, A. B.; Robinson, W. E. U.S. Bur. Mines Rept. Invest. 4744, 1950.
119. Allred, V. D.; Nielson, G. I. Chern. fmg. Progr. Symp. Ser. 1965, 160, 54.
120. Allred, V. D. Chern. fng. Progr. 1966, 62, 55.
121. Campbell, J. H.; Koskinas, G. H.; Stout, N. D. Fuel 1978, 57, 372.
122. Herrell, A. Y.; Arnold, C, Jr. Thermochim. Acta 1976, 17, 165.
123. Rajeshwar, K. Thermochim. Acta 1981, 45, 253-263.
124. Haddadin, R. A.; Mizyed, F. A. Ind. fng. Chern. Process Oes. Oev. 1974,
13, 332.
OIL SHALE PAGE 354
125. Bekri, 0.; Baba-Habib, H.; Cha, C. Y.; Edelman, M. C. Proc. 16th Oil
Shale Symp.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1983; 345-364.
126. Williams, P. F. V. Fuel 1985, 64, 540-545.
127. Braun, R. L.; Burnham, A. K. Fuel 1986, 65. 218-222.
128. Richardson, J. H.; Huss, E. S.; Ott, L. L.; Clarkson, J. E.; Bishop, H. 0.;
Taylor, J. R.; Gregory. L. J.; Morris, C. J. Lawrence Livermore National
Laboratory Report 1982, UCID-19548.
129. Wallman, P. H.; Tamm, P. W.; Spars, B. G. In Oil Shale, Tar Sands and
Related Materials; ACS Symposium Series 163; Stauffer, H. C., Ed.;
American Chemical Society: Washington. 1981; 93-113.
130. Smith, J. W.; Johnson, D. R. Proc. 2nd Internatl. Conf. on Thermal
Analysis 1969, 1251-1268.
131. Johnson, O. R,; Young, N. B.; Smith, J. W. "Thermal Analysis on Oil
Shales: Determination of Potential Oil Yields and Oawsonite, Nacholite
and Norstrandite Content" U.S. Department of Energy, Laramie Energy
Research Center Report 1978, LERC/RI-77/6.
132. Rosenvold, R. J.; Rajeshwar, K. Thermochim. Acta 1982, 57. 1.
133. Rajeshwar, K.; Rosenvold, R. J.; DuBow, J. B. Thermochim. Acta 1983,
66, 373.
134. Earnest, C. M. Thermochim. Acta 1983. 60. 171-180.
135. Earnest, C. M. Thermochim. Acta 1982, 58. 271-288.
136. Williams. P. F. V. "The Analysis of Oil Shale and Shale Oil" In
Petroanalysis '81; Crump, G. B., Ed.; Institute of Petroleum: london,
1982; 326-341.
OIL SHALE PAGE 355
137. Camp, D. W. "Oil Shale Heat Capacity Relations and Heats of Retorting and
Dehydration" In 20th Oil Shale Symp. Proe.; Gary, J. H" Ed.; CSM Press:
Golden, CO, 1987; 130-144.
138. Shaw, R. J. "Specific Heat of Colorado Oil Shales" U.S. Bur. Mines Rept.
Invest. 1947, 4151.
139. Wise, R. L.; Miller. R. c.; Sohns. H. W. "Heat Contents ad Some Green
River Oil Shales" U.S. Bur. Mines Rept. Invest. 7482 1971.
140. Sohns, H. W.; Mitchell. l. E.; Cox, R. J.; Barnet, W. I.; Murphy,
W. I. R. Ind. Eng. Chem. 1951, 43, 33.
141. Cook, E. W. Colo. Sch. Mines Quart. 1969, 65, 133.
142. (a) Mraw, S. C.; Keweshan, C. J. Fuel 1986, 65, 54-57. (b) Proc. 2nd
Australian Workshop on Oil Shale, Brisbane, 6-7 Dec. 1984; pp 147-152.
143. Carley, J. F. "Heat of Kerogen Decomposition and Improved Enthalpy
Temperature Relationships for Raw and Spent Colorado Oil Shales".
Lawrence Livermore National Laboratory Report 1975, UOPKK-7S-28.
144. Robie, R. A.; Hemingway. B. S.; Fisher, J. R. "Thermodynamic Properties
of Minerals and Related Substances at 298.15K and I Bar (lOS Pascals)
Pressure and at Higher Temperature". U.S. Geol. Surv. Bull. 1452, 1979.
145. Rodgers, P. A.; Creagh, A. L.; Prausnitz, J. M. Fuel 1988, 67, 134-142.
146. Mraw, S. Co; Heidman, t. J.; Hwang, S.-C.; Tsonopoulos, C. Ind. Eng.
Chem. Proc. Des. Dev. 1984, 23, 577.
147. Tsonopoulos, C.; Heidman, L. J.; Hwang, S.-C. Thermodynamic and Mass
Transport Properties of Coal Liquids; Wiley & Sons: New York, 1986.
148. Thakur, D. S.; Nuttall, H. E., Jr. Ind. Eng. Chem. Res. 1987, 26, 1351.
149. Suuberg, E. M.; Sherman, J.; Lilly, W. D. Fuel 1987, 66, 1176-1184.
OIL SHALE PAGE 356
150. Burnham, A. K.; Singleton, M. F. "High Pressure Pyrolysis of Green River
Oil Shale" In Geochemistry and Chemistry of Oil Shales; ACS Symposium
Series 230; Miknis, F. P.; McKay, J. F., Eds.; American Chemical Society:
Washington, 1983; 335-351.
151. Sellers, J. B.; Haworth, G. R; Zambast P. G. Trans. Soc. Min. Eng.,
AINE June 1972, 252, 222·232.
152. Agapito, J. "Pillar Design in Competent Bedded formations." Thesis 1485,
Graduate School, Colorado School of Mines, 1972.
153. Dinneen, G. U. "Constitution of Green River Oil Shale" presented at the
U.N. Symposium on the Development and Utilization of Oil Shale Resources.
Tall inn, 1968.
154. Baughman, G. L. Synthetic Fuels Data Handbook, 2nd Ed.; Cameron
Engineers, Inc.: Denver, CO, 1978.
155. Mraz, T.; DuBow, J.; Rajeshwar, K. Fuel 1983, 62, 1215-1222.
156. Tisot, P. R.; Sohns, H. W. U.S. Bur. Nines Rept. Invest. 7576, 1971.
157. Hall. P. L.; Mildner, D. F. R.; Borst, R. L. Appl. Phys. Letters 1983,
43, 252-254.
158. Durand, So; Niease, G. "Procedures for Kerogen Isolation" In Kerogen;
Durand, B., Ed.; Editions Technip: Paris, 1980; 35-53.
159. Robinson, W. E. "Isolation Procedures for Kerogens and Associated Soluble
Organic Materials" In Organic Geochemistry; E911ntoo, G.; Murphy, M. T. J.,
Eds.; Springer-Verlag: Berlin, 1969; 181-195.
160. Saxby, J. D. :Chemical Separation and Characterization of Kerogen From
Oil Shale" In Oil Shale; Yen, T. F.; Chilingarian, G. V., Eds.;
Elsevier: Amsterdam, 1976; 103-128.
OIL SHALE PAGE 357
161. Forsman, J. P. "Geochemistry of Kerogen" In Organic Geochemistry;
Breger, I. A., Ed.; Pergamon Press Limited: Oxford, 1963; 148-182.
162. Forsman, J. P.; Hunt, J. M. "Insoluble Organic Matter (Kerogen) in Sedi
mentary Rocks of Marine Origin" In Habitat of Oi7; Weeks, l. G., Ed.;
The American Association of Petroleum Geologists: Tulsa, 1958; 747-778.
163. Scouten. C. G.; Siskin, M.i Rose, K. D.; Aczel, T.; Colgrove, S. Go;
Pabst, R. E. Proe. 4th Australian Workshop on Oil Shale, Brisbane,
Dec. 3-4, 1987; 94-100.
164. McCollum, J. D. ACS Div. Petrol. Chern. Prepr. 1987, 32(1}, 74.
165. Luts, K. Brennstoff-Chem. 1928, 9, 217-218.
166. Hubbard, A. B.; Smith, H. N.; Heady, H. H.; Robinson, W. E. U.S. Bur.
Nines Rept. Invest. 5725, 1952.
167. Quass, F. W. J. Inst. Petrol. 1939, 25, 813-819.
168. Smith, J. W.; Higby, l. W. Anal. Chern. 1960, 32, 17-18.
169. Himus, G.; Basak, G. C. Fuel 1949, 28, 57-64.
170. Siskin, H.; Brons, G.; Payack, J. F., Jr. ACS Div. Petrol. Chem. Prepr.
1987, 32(1}, 75-78.
171. Siskin, Ho; Brons, G.; Payack, J. F., Jr. Energy & Fuels 1987, 1, 100-108.
172. Robinson, W. E.j Cummins, J. J.j Dinneen, G. U Geochim. Cosmochim. Acta
1965, 29, 249-258.
173. Anderson, P. C.; Gardner, P. M.; Whitehead, E. V.; Anders, D. E.;
Robinson, W. E. Geochim. Cosmoehim. Acta 1969, 33, 1304-1307.
174. Anders, D. Eo; Doolittle, F. G.; Robinson, W. E. Geochim. Cosmochim.
Acta 1973, 37, 1213-1228.
OIL SHALE PAGE 358
175. Anders, O. E.; Doolittle, F. G.; Robinson, W. E. Geochim. Cosmochim.
Acta 1975, 39, 1423-1430.
176. Gallegos, E. J. Anal. Chern. 1971, 43, 1151-1160.
177. Gallegos, E. J. "Biological Fossil Hydrocarbons in Shales" In Oil Shale
Yen, T. F.; Chilingarian. G. V., Eds.; Elsevier: Amsterdam,
1976; 149-180.
178. Eglinton, G.; Maxwell, J. R.; Murphy, M. T. J.; Henderson, W.; Douraghi
Zader, K. Abstr. Ann. Heeting Geol. Soc. Amer. 1966, 59-60.
179. Robinson, W. E. "Origin and Characteristics of Green River Oil Shale" In
Oil Shale; Yen, T. F.; Chilingarian, G. V., Eds.; Elsevier: Amsterdam,
1976; 61-79.
180. Vandenbroucke, M. "Structure of Kerogens as seen by Investigations on
Soluble Extracts" in Kerogen; Durand, B., Ed. Editions Technip: Paris,
1980; 415-443.
181. Hutton, A. C.; Kantsler, A. J.; Cook, A. C. Austral. Petrol. Explor.
Assoc. J. 1980, 20, 44-67.
182. Cardoso, J. N.; Chicarelli, M. I. 'The Organic Geochemistry of the
Paraiba Valley and and Mara. Oil-shales" In Advances in Organic
Geochemi stry 1981; Bjor~y, M.; et a 1., Eds.; Wil ey: Chi chester, 1983;
pp 828-833.
183. Uden. P. C.; Siggia, S.; Jensen, H. S.; Eds. Analytical Chemistry of
Liquid Fuel Sources (Ady. Chern. Ser. 170); American Chemical Society:
Washington, 1978.
184. Fenton, M. D.; Henning, H.; Ryden, R. L. 'The Chemistry of Shale Oil and
Its Refined Products' In Oil Shale, Tar Sands and Related Haterials;
Stauffer, H. C., Ed.; American Chemical SOCiety: Washington, 1981; 315-325.
OIL SHALE PAGE 359
185. Holmes. S. A.; Thompson, t. F. QNitrogen-Type distribution in Hydrotreated
Shale Oils" In 14th Oil Shale Symposium Proceedings Gary. J. H., Ed.;
Colorado School of Mines Press: Golden, Co, 1981; 235-248.
186. Regtop, R. A.; Crisp, P. T.; Ellis, J. Fuel 1982, 61, 185-192.
187. Fester, J. I.; Robinson, W. E. Anal. Chem. 1964,36, 1392-1394.
188. Fester, J. I.; Robinson, W. E. "Oxygen Functional Groups in Green River
Oil-Shale Kerogen and Trona Acids ff In Coal Science (Adv. Chern. Ser. 55);
American Chemical Society: Washington, DC, 1966; pp 22-31.
189. Vitorovit, D. UStructure Elucidation of Kerogen by Chemical Methods ff
In Kerogen; Durand, B., Ed.; Editions Tecnip: Paris, 1980; 301-338.
190. Robinson, W. E.; Heady, H. H.; Hubbard, A. B. Ind. Eng. Chem. 1953,
45, 788-791.
191. Vitorovit, D.; Ambles, A.; Djordjevit, H.; Bajc, S. ACS Div. Petro
Chem. Prepr. 1987, 32(1), 37-42.
192. Down, A. L.; Himus t G. V. J. Inst. Petro1eum 1941 t 27, 419-445.
193. Djuricic, M. V.; Murphy, R. C.; Vitorovic, D.; Biemann, K.
Geochim. Cosmochim. Acta 1971, 35, 1201-1207.
194. Ambles, A.; Djuricic t M. Vo; Ojordjevic, Lj.; Vitorovic, D. "Nature
of Kerogen from the Green River Shale based on the Character of the
Products of a Forty-Step Alkaline Permanganate Oxidation" In Advances
in Organic Geochemistry 1981; Bjor~y, M.; et aI., Eds.; Wiley:
Chichester, 1983; pp 554-560.
195. Hoering, T. Co; Abelson, P. H. Carnegie Inst. Washington Yearbook 1965,
64, 218.
196. Burlingame, A. L.; Simoneit, B. R. Science 1968, 160, 531-533.
OIL SHALE PAGE 360
197. Burlingame, A. L.; Simoneit, B. R. Nature 1969, 222, 741-747.
198. Burlingame, A. l.; Haug, P. A.; Schnoes. H. K.; Simoneit, B. R. "Fatty
Acids Oerived from the Green River Formation Oil Shale by Extraction and
Oxidation - A Review" In Advances in Organic Geochemistry 1968; Schenck,
P. A.; Havenaar, 1'1 Eds.; Pergamon Press: Oxford, 1969; pp 85-129.
199, Burlingame, A. L.; Wszolek, P. C.; Simoneit, B. R. "The Fatty Acid
Content of Tasmanites" In Advances in Organic Geochemistry 1968; Schenck,
P. A.; Havenaar, I., Eds.; Pergamon Press: Oxford, 1969; pp 85-129.
200. Simoneit, B. R. T.; Burlingame, A. L. "Ketones Derived from the Oxidative
Degradation of Green River Formation Oil Shale Kerogen" In Advances in
Organic Geochemistry 1973; Tissot, B.; Bienner, F .• Eds.; Editions
Technip: Paris, 1974; pp 191-201.
201. Bitz, M. C.; Nagy, B. Proc. Natl. Acad, Sci. U.S. 1966, 56, 1383-1390.
202. Bitz, M. C.; Nagy, B. Anal. Chem. 1967, 39, 1310-1313.
203. Van den Berg, M. L. J.; leeuw, J. W. de; Schenck. P. A. "Constitutional
Analysis of Kerogens and lignite by Ozonation" In Advances in OrganiC
Geochemistry 1973; Tissot, B.; Bienner, F., Eds.; Editions Technip:
Paris, 1974; pp 163-178.
204. Van den Berg, M. l. J.; Mulder, G. J.; leeuw, J. W. de; Schenck, P. A.
Geochim. Cosmochim. Acta 1977, 41, 903.
205. Nekrasov, V. I.; Urov, K. E. Tr. Tallinsk. Politekhn. lnst., Ser. A,
I971, 311, 79.
206. Robertson, J. B. Proc. Roy. Soc. Edinburgh 1914, 34, 190.
207. Punga, V.; Fomina, A. S.; Oegtereva, Z. A. [ZV. Akad. Nauk. Est. SSR,
Ser. Khim. i Geo1. 1973, 19, 195 and references cited therein.
OIL SHALE PAGE 361
208. Robinson, W. E.; Lawlor, O. L.; Cummins, J. J.; Fester, J. I. U.S. Bur.
lIines Rept. Invest. 6166, 1963.
209. McGowan, C. W.; Diehl, H. Fuel Processing Technol. 1985, 10, 169-179.
210. McGowan, C. W.; Diehl, H. Fuel Processing Technol. 1985, 10, 181-193.
211. Proskuryakov, V. A.; Vakovlev, V. I.; Otrodnykh, T. F.; Lukin, A. J.
Okislenie Uglevodorodov i Kaustobiolitov 1975, 43 and references cited
therein.
212. Dancy, T. E.; Giedroyc, V. J. Inst. Petroleum 1950, 36, 607-623.
213. Kinney, C. R.; leonard, J. T. J. Chem. Eng. Data 1961, 6, 474.
214. Wen, C. S.; Yen, 1. F. "Electrolytic Oxidation and Reduction of Oil Shale"
In Science and Technology of on Shale; Yen, T. F .• Ed.; Ann Arbor
Science Publishers: Ann Arbor, MI, 1978; pp. 83-101.
215. Bone, W. A.; Horton, l.; Ward, S. G. Proc. Roy. Soc. (London) 1930
127A, 480-510.
216. Stewart, R. "Oxidation by Permanganate" In Oxidation in Organic Chemistry;
Wiberg, K. B., Ed.; Academic Press: New York, 1965; pp 1-68.
217. lee, D. G. The Oxidation of Organic compounds by Permanganate Ion and
Hexavalent Chromium; Open Court Publishing Company: La Salle, Il, 1980.
218. Arndt, O. Manganese Compounds as Oxidizing Agents in Organic Chemistry;
Open Court Publishing Company: la Salle, Il, 1981.
219. Chinn, l. J. Selection of Oxidants in Synthesis; Marcel Dekker: New York,
1971.
220. Augustine, R.; Trecker, D. Oxidation; Marcel Dekker: New York,
Volume 1, 1969; Volume 2, 1971.
OIL SHALE PAGE 362
221. Waters, W. A. Mechanisms of Oxidation of Organic Compounds; Wiley: New
York, 1964.
222. Stewart, R. Oxidation Me~hanisms; W. A. Benjamin: New York, 1964.
223. Wiberg, K. B. Surv. Progr. Chern. 1963, 1, 211-248.
224. Randall, R. S.; Senger, Hoi Grocock, C. M. Proc. Roy. Soc. (London)
1938, 165A, 432-452.
225. Eastman, R. H.; Quinn, R. A. J. Arner. Chern. Soc. 1960, 82, 4249.
226. Waters, W. A. Trans. Faraday Soc. 1946, 42, 184.
227. Kenyon, J.; Symons, M. C. R. J. Chern. Soc. 1953, 2129.
228. Kenyon, J.; Symons, M. C. R. J. Chern. Soc. 1953, 3580.
229. Wiberg, K. B.; Fox, A. S. J. Arner. Chern. Soc. 1963, 85, 3487.
230. Stewart, R.; Van der Linden, R. Discussions Faraday Soc. 1960, 211.
231. Nicolaus, R. A. Rass. Med. Spero (Suppl. 2) 1960, 7, 1.
232. Wiberg, K. B. "Oxidation by Chromic acid and Chromyl Compounds" In
Oxidation in Organic Chemistry; Wiberg, K. B., Ed.; Academic Press:
New York, 1965; pp 69-184.
233. Freeman, N. H. U.K. Patent 291,475; 1928.
234. Bock, J.; McCall, P. P.; Robbins, M. t.; Siskin, M.
U.S. Patent 4,458,757; JuliO, 1984
235. Bock, J.; McCall, P. P.; Robbins, M. Lo; Siskin, M.
U.S. Patent 4,461,696; Jul 24, 1"984
236. Schmidt-Collerus, J. J.; Prien, C. H. ACS Div. Fuel Chern. Prepr. 1974,
19(2), 100-108.
237. Hubbard, A. S.; Fester, J. I. Ind. Eng. Chern. 1958, 3, 147.
OIL SHALE PAGE 363
238. Young, D. K.; Shih, S.; Yen, T. F. ACS Oiv. Fuel Chern. Prepr. 1974,
19(2), 169-174.
239. Young, D. K.; Yen, T. F. Geochim. Cosmochim. Acta. 1977, 41, 1411.
240. Barakat, A. 0.; Yen, T. F. ACS Oiv. Fuel Chern. Prepr. 1987, 32(1),4347.
241. Young, D. K.; Shih, S.; Yen, 1. F. "Mild Oxidations of Bioleached Oil
Shale" In Science and Technology of Oil Shale; Yen, T. F., Ed.; Ann Arbor
Science Press: Ann Arbor, MI, 1976; pp 65-71.
242. Barakat, A. 0.; Yen, T. F. Fuel 1987, 66, 587-593.
243. Barakat, A. 0.; Yen, T. F. Energy & Fuels 1988, 2, 105-108.
244. Yen, T. F. "Structural Aspects of Organic Components in Oil Shales" In
Oil Shale; Yen, T. F.; Chilingarian, G. V., Eds.; Elsevier: New York,
1976; pp 129-148.
245. Yen, T. F. ACS Oiv. Fuel Chern. Prepr. 1974, 19(2), 109-114.
246. Yen, T. F. "Structural Investigations on Green River Oil Shale Kerogen"
In Science and Technology of Oil Shale; Yen, T. F., Ed.; Ann Arbor
Science Press: Ann Arbor, MI, 1976; pp 193-203.
247. Silbernagel. 8. G.; Gebhard. l. A.; Siskin, Mo; Brons, G. B.
Energy & Fuels 1987, 1, 501-506.
248. Cooper, J. E.; Evans, W. S. Science 1983, 219, 492-493.
249. Taylor, R. W.; Burnham, A. K.; Smith, G. S.; Sanborn, R. H.;
Grgeory, l. S. ACS Oiv. Fuel Chern. Prepr. 1985, 30(3), 338-348.
251. Cooper, J. E. Anal. Chern. 1986, 58, 1571-1572.
252. Riley, K. W.; Moss, T. D.; Orban, H.; Quezada, R. A. Fuel 1987, 66, 323-5.
253. Wang, N.-C.; Teo, K.-E.; Andersen, H. Can. J. Chern. 1977, 55, 4112-4116.
254. Tissot, B.; Espitalie, J. Rev. lnst. Fr. Petro 1975, 30, 743-777.
OIL SHALE PAGE 364
255. Behar, F.; Vandenbroucke, M. Rev. Inst. Fr. Petro 1986, 41, 173-188.
256. Cody, G. D., Jr.; Larsen, J. W.; Siskin, M. Energy & Fuels 1988, 2,
340-344.
257. Allred, V. D., Ed. Oil Shale Processing Technology; The Center for Pro
fessional Advancement: New Brunswick, NJ, 1982.
258. Stauffer, H. C., Ed. Oil Shale, Tar Sands and Related Haterials (ACS
Symp. Ser. 163); American Chemical Society: Washington, DC, 1981.
259. Miknis, F. P.; McKay. J. F., Eds. Geochemistry and Chemistry of Oil
Shales (ACS Symp. Ser. 230); American Chemical Society: Washington, DC,
1983.
260. Newman, S. A. Shale Oil Upgrading and Refining; Butterworth: Woburn, MA,
1983.
261. Proceedings of the first five symposia gathered into one volume, and
those of subsequent symposia (the 20th Symposium was held in 1987) as
single volumes, are available from the Colorado School of Mines Press:
Golden, Colorado 80401.
262. Papers presented at the symposia on Synthetic Fuels from Oil Shales and
Tar Sands (1st, 1979; 2nd, 1981; 3rd, 1983) are available from the
sponsor, the Instutute of Gas Technology, lIT Center, 3424 South State
Street, Chicago, Il 60616.
263. Proceedings of the Australian Workshops on Oil Shale (1st, 1983; 2nd,
1984; 3rd, 1986; 4th, 1987) are available from the Secretary, Australian
Workshops on Oil Shale, CSIRO Division of Energy Chemistry, Private Mail
Bag 7, Menai, NSW, 2234, Australia. .
264. Whitworth, K. World Coal 1979, 5(10), 26-29 and 34-37.
OIL SHALE PAGE 365
265. Wyllie, R. J. M. World Hining 1979, 32(11), 100-104.
266. Oil Shales and Tar Sands: A Bibliography (TIO-3367); U.S. Department of
Energy, Technical Information Center: Oak Ridge, TN, 1977.
267. Grissom, M. C., Ed. Oil Shales and Tar Sands: A Bibliography (TID-3367,
Suppl. 1); U. S. Department of Energy, Technical Information Service:
Oak Ridge, TN, 198!.
268. Voetter, H.; Poll, I.; van Wechem. H. M. H. "A Discussion of Process and
Project Design in Oil Shale Retorting" 20th Oil Shale Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1987; pp 122-129.
269. Mashin, V. N.; Serebryannikov, N. Do; Purre, T. A. "Oil Shale Processing
Costs" 10th World Petrol. Congo Proc. 1980, 3, 311-314.
270. Kauppila. T. A. "Surface Mining in U.S.A. II In Oil Shale Processing Tech
nology; Allred, V. D., Ed.; The Center for Professional Advancement:
New Brunswick, NJ, 1982; 1-21.
27!. Kauppila, 1. A. "Underground Mining in U.S.A." In Oil Shale Processing
Technology; Allred, V. D., Ed.; The Center for Professional Advancement:
New Brunswick, NJ, 1982; 23-54.
272. Jee, C. K.; White, J. D.; Bhatia, S. K.; Nicholson, O. Review and Anal
ysis of on Shale Technologies. Vol. 1. Oil Shale Deposits, Hining
Hethods and Environmental Concerns NTIS FE--2343-6(Vol. 1), Aug. 1977.
273. Hoskins, W. N.; Upadhyay, R. P.; Bills, J. 0.; Sandberg, C. R. III;
Wright, F. D. "Technical and Economic Study of Candidate Underground
Mining Systems for Deep, Thick Oil Shale Deposits." Final Report:
Contract S0241074, Oct. 1976. NTIS PB--262525.
274. Farris, C. B. Mining Engineering (New York) 1980, 32(1), aa-ii.
OIL SHALE PAGE 366
275. Crookston, R. B.; Weiss, O. A. "Oil Shale Mining - Present and Future!!
In Symposium Papers: Synthetic Fuels from Oil Shale II, Nashville, TN,
Oct. 26-29, 1981; Institute of Gas Technology: Chicago, 1982.
276. East, J. H., Jr.; Gardner, E. O. Oil Shale Mining, Rifle, Colorado,
1944-56. U.S. Bur. Mines Bull. 611, 1964.
277. Ruark, J. R.; Sohns, H. W.i Carpenter, H. W. U.S. Bur. Mines Rept.
Invest. 7303, 1971.
278. Ruark, J. R.; Sohns, H. W.; Carpenter, H. W. U.S. Bur. Mines Rept.
Invest. 7540, 1971.
279. Jones, J. B., Jr.; Heistand, R. N. "Recent Paraho Operations" 12th Oil
Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1979; 184-194.
280. Dick, R. D.; Young, C. i Fourney, W. L. "Summary of the Oil Shale Fragmen~
tation Program at Anvil Points Mine, Colorado" 17th Oil Shale Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1984; 225-242.
281. Fourney, W. l.; Dick, R. D. Review of Oil Shale Consortium Fragmentation
Program at Anvi1 Points Mine, Colorado Los Alamos National Laboratory
Report LA-IOS38-MS, 1985.
282. Parrish, R. L. Instrumented Cratering Tests S8-1 and 58-2 at the Anvil
Points Mine Sandis National Laboratory Report SAND-84-0818, 1986.
283. Virgona, J. E. "Decommissioning of the U.S. Department of Energy Anvil
Points Oil Shale Research Facility" 19th Oil Shale Symp. Proc.; Gary,
J. H., Ed.; CSM Press, Golden, CO, 1986; 161-176.
284. Crookston, R. B. Underground Space 1978, 2(4), 229-241.
285. Willmon, G. J. 12th World pterol. Congo Proc. 1987, 4, 3-13.
OIL SHALE PAGE 367
286. Rio Blanco Oil Shale Project, "Detailed Development Plan, Tract C-a."
Gulf Oil Corporation and Standard Oil (Indiana), 1976.
287. Rajaram, V.; Utter, S.; Hooker, V. "Geotechnical and Environmental Design
of a Mining Research Facility in Deep, Thick Oil Shale Deposits" In Rock
Mechanics (Proc. 19th Symp. Rock Mechanics, Lake Tahoe, NV, 1 May 1978):
Kim, Y. 5., Ed.; University of Nevada: Reno. 1978.
288. Kurapei, G. A.; Cheshko, Yu. E. "New Technology of Shale Recovery in
Lenningrad Deposit Mines" Ugol' 1977, (12),23-27.
289. Trent, R. H.: Dunham, R. K. "Longwall Mining of Oil Shale" 10th Oil
Shale Symp. Proc. Gary, J. H., Ed.; CSM Press, Golden, CO, 1977; 89-98.
290. Timofeev, N. N.; Vesiloo, P. I. "Prospects of Developing Recovery
Operations in Open Pits" Ugol' 1977, (12), 40-44.
291. Anonymous "Open Pit Operations in the USSR: Shale Oil Processing Devel
opments" Coal Miner Sept. 1978, 3(3), 77,78,80,81.
292. Cieslewicz, J. W. "Selected Topics of Recent Estonian-Russian Oil Shale
Research and Development" Colo. Sch. Mines Quart. 1971, 55(1), 1-154.
293. Siwinski, R. E. Proc. 4th Australian Workshop on Oil Shale Brisbane,
3-4 Dec, 1987: 72-77.
294. Adams, R. C.; Banks, C. E.; Bradley, W. S.; Brannick. L. L.; Christian,
W. G. "Technical and Cost Evaluation of candidate Large-Scale Open-Pit
Oil Shale Mining Methods in Colorado. Volume I. Technical and Economic
Study of an Integrated Single-pass System for Open-pit Mining of Deep
Oil Shale DepOSits." Final Report: Contract S0241046, July 1976.
(NTIS PB--274435)
OIL SHALE PAGE 368
295. Adams, R. C.; Banks, C. E.; Bradley, W. S.; Brannick, L. L.; Christian,
W. G. "Technical and Cost Evaluation of Candidate Large-Scale Open-Pit
Oil Shale Mining Methods in Colorado. Volume II, Parts 1 and 2. Technical
and Economic Study of an Integrated Single-pass System for Open-pit
Mining of Deep Oil Shale Deposits." Final Report: Contract S0241046,
July 1976. (NTIS PB--274520)
296. Lappi, R. L.; Carey. D. I.; Pelofsky. A. H.i Bates, E. R.; Martin, J. F.
"Overview of Environmental Impacts of large-Scale Surface Mining of Oil
Shale: Piceance Basin, Colorado" 15th Oil Shale Symp. Proe.; Gary, J. H.,
Ed.; CSM Press: Golden, CO, 1982; 231-251.
297. Eitel, G. L. "Update on the American Syn-crude/lndiana Shale Oil Project"
18th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO,
1985; 216-224.
298. Addington. R. R.; Humphrey, J. P.; Piper, E. M. "American Syn-Crude's
Adaptation of the Petrosix Process" Symposium Papers: Synthetic Fuels
from Oil Shale and Tar Sands, Louisville, KY, May 17-19; Institute of
Gas Technology: Chicago, IL, 1983; 497-530.
299. Sal ott; , C. A.; Datta, R. S. !!Size Reduction of Green River Oil Shale"
16th Oil Shale Symp. Proe.; Gary, J. H., Ed.; CSM Press: Golden, CO,
1983; 394-412.
300. Datta, R. S.; Salotti "Beneficiation of Eastern Oil Shale" Proceedings
Eastern Oi1 Shale Symposium, Lexington, KY, Oct. 11-13; Institute for
Mining and Minerals Research, University of Kentucky: Lexington, KY,
1982; 227-234.
301. Dundas, R. C.; Howes, R. T. U.S. Patent 1,469,628; 1923.
OIL SHALE PAGE 369
302. Wallace, G. W. U.S. Patent 1,536,696; 1925.
303. Mapstone, G. E. "Wartime Shale Oil Production at Mangaroo, N.S.W." In
Oil Shale and Cannel Coa7, V. 2; Sell, GO) Ed.; Institute of Petroleum:
London, 1951; pp 489-499.
304. Cattell, R. A.; Guthrie, B.; Schramm, L. W. "Retorting Colorado Oil Shale
- A Review of the Work of the Bureau of Mines, U.S. Department of the
Interior" In Oil Shale and Canne1 Coal, V. 2; Sell, G., Ed.; Institute
of Petroleum: London, 1951; pp 345-398.
305. Harack, A. E. and others "Some results from the Operation of a ISO-ton
Oil Shale Retort' U.S. Bur. Hines Tech. Rept. No. 30, 1971.
306. Harack, A. E. and others "Prel iminary Design and Operation of a ISO-ton
Oil Shale Retort" Annual Meeting of AIMf. Denver, CO, 1970.
307. Docktor, L. "Retorting Random-Sized Lean Oil Shale in a Batch-Type Retort
- An Engineering Analysis' 65th Annual Meeting of Aliff, New York, 1972.
308. Ruark, J. R. and others "Description and Operation of the N-T-U Retort on
Colorado Oil Shale" U.S. Bur. Mines Rept. Invest. 5279, 1956.
309. Matzick, A. and others 'Developement of the USBM Gas Combustion Oil
Shale Retorting Process· U.S. Bur. Hines Bull. 635, 1966.
310. Matzick, A.; Dannenberg, R. O. "USBM Gas Combustion Retort for Oil Shale
- A Study of Process Variables" U.S. Bur. Hines Rept. Invest. 5545, 1960.
311. lawson, J. E. and others "Gas Combustion Retorting Performance in a Large
Demonstration Retort" lOath Annual Meeting of AIHE, New York, 1971.
312. Rammler, R. W. "The lurgi-Ruhrgas Process for the Retorting of Oil Shale'
In Oil Shale Processing Technology; Allred, V. D., Ed.; The Center for
Professional Advancement: New Brunswick, NJ, 1982; pp 83-105.
OIL SHALE PAGE 370
313. Weiss, H. J. "The Retorting of Oil Shale by the Lurgi-Ruhrgas (LR)
Process" 6th [IASA Resources conference on World Oil Shale Resources
and Their Potential for Development; Golden, CO, June, 1981.
314. Watson, G. H.j Ziemba, E. A.; Bissery, P.; Namy, 0.; Griffis, R. l.;
Nicholson, D. E. "The Filling of Oil Shale Mines with Spent Shale Ash.
Ash Characteristics and Grout Development" 15th Oil Shale Symp. Proe.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1982; 397-410.
315. Potter, O. Eo; Fryer, C.; Christiansen G.; eunico, P.; Swaine, Go; Tam,
K. C.; Preslmaier, R.; Hoskin, C. M. "New Variations on the Old Theme of
Shale Oil Recovery" Proe. 2nd Australian Workshop on 011 Shale, Brisbane,
6-7 Dec. 1984; pp 188-193.
316. Whitcombe, J. A.; Vawter, R. G. "The Tasca II Process and the Colony
Project" In Oil Shale Processing Technology; Allred, V. 0., Ed.; The
Center for Professional Advancement: New Brunswick, NJ, 1982; pp 153-168.
317. Anonymous Oil & Gas J. Hay 10, 1982, 86.
318. Anonymous Oil & Gas J. Har 22, 1982, 215-216.
319. Gwyn, J. E.; Roberts, S. C.; Hinds, G. P., Jr.; Hardesty, O. E.; Johnson,
G. L. "Shell Pellet Heat Exchanger Retorting -- SPHER Intensive, Energy
Efficient Process for Retorting Oil Shale" 13th Oil Shale Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1980; 35-45.
320. Gwyn, J. E.; Roberts, S. C.; Hardesty, O. E.; Johnson, G. l.; Hinds, G.
P., Jr. "Shell Pellet Heat Exchange Retorting: The SPHER Energy-Efficient
Process for Retorting Oil Shale" In Oil Shale, Tar Sands and Related
Materials (ACS Symp. Ser. 163); Stauffer, H. C., Ed.; American Chemical
Society: Washington, 1981; pp 167-181.
OIL SHALE PAGE 371
320. Gwyn, J. E.; Roberts, S. C.; Hardesty, O. E.; Johnson. G. L.; Hinds, G.
P., Jr. ACS Div. Fuel Chem. Prepr. 1980, 25(3), 59-69.
322. Gwyn, J. E. U.S. Patent 4,081,968; 1978.
323. van Wechem, H. M. H.; Poll, I.; Voetter, H. "The Basis of Reactor Selection
for the Shell Shale Retorting Process" In Proc. 4th Australian Workshop
on Oil Shale, Brisbane, 3-4 Dec. 1987; 165-173.
324. Voetter, Ho; Poll, l.; van Wechem, H. M. H. ffA Discussion of Process and
Project Design in Oil Shale Retorting" 20th Oil Shale Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1987; 122-129.
325. Forgac, J. M. AIChE Spring Natl. Meeting 1984, 576, 78.
326. Wallman, P. H.; Tamm, P. W.; Spars, B. G. ACS Div. Fuel Chem. Prepr.
1980, 25(3), 70-81.
327. Wilkins, E. S.; Nuttall, H. E.; Thakur, 0. S.; Proc. 2nd World Congr. on
Chem. Eng., Montreal, 1981.
328. Bauman, R. F.; Mitchell, W. N.; Eakman, J. M.; Koveal, R. J. "Exxon Shale
Retort (ESR) Process Pilot Plant Development Program" In Proc. 4th
Australian Workshop on Oil Shale, Brisbane, 3-4 Dec. 1987; 198-203.
329. Schmalfeld, I. P. "The Use of the Lurgi-Ruhrgas Process for the
Distillation of Oil Shale' Eighth Oil Shale Symp. Proc.; Gary, J. H.,
Ed.; Colorado School of Mines Press: Golden, CO, 1975; 129-145.
330. Tamm, P. W.; Bertelsen, C. A.; Handel, G. M.; Spars, B. Go; Wallman, P. H.
Energy Progress 1982, 2(1), 37-42.
331. Tamm, P. W.; Langlois, G. E. U.S. Patent 4,199,432; 1980.
332. Wallman, P. H.; Tamm, P. W. j Spars, B. G. "Oil Shale Retorting Kinetics"
in Oil Shale, Tar Sands and Related Materials lACS Symp. Ser. 163);
Stauffer, H. C., Ed.; American Chemical Society: Washington, 1981; 93-113.
OIL SHALE PAGE 372
333. Rezende, J. "Present State of Development of the Petrasix Process" In
OJ] Shale Processing Technology; Allred, V. 0., Ed.; The Center for
Professional Advancement: New BrunswicK, NJ, 1982; pp 121-135.
334. Piper, E. M.; Iva. O. C. "The Petrasix Project in Brazil - An Update"
19th on Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO,
1986; 98-101.
335. Piper, E. M.; Iva, O. C. "The Petrosix Project in Brazil - An Update"
Eastern Oil Shale Symposium, lexington, KY, 1986; 7p.
336. Piper, E. M. Personal communication, June 1988.
337. Eitel, G. L.; Domahidy, G. "Update on the American Syn-Crude/lndiana
Project" 18th on Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press:
Golden, CO, 1985; 216-224.
338. Bruni, C. E. "Demonstration Plant for Retorting Irati Oil Shale"
United Nations on Shale Symposium, Tall inn, 1968.
339. Faulkner, B. P.; Weinecke, M. H.; Cnare, R. F. "The Allis-Chalmers Roller
Grate Retort Process for Eastern Oil Shale" Symposium Papers: Synthetic
Fuels from Oil Shale and Tar Sands, Louisville, KY, May 17-19; Institute
of Gas Technology: Chicago, IL, 1983; 545-561.
340. Faulkner, B. P.; Weinecke, M. H.; Cnare, R. F. "Results of the Processing
of a Western Oil Shale on the All i s-Cha 1 mers Roller Grate Retort System."
16th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1983;
255-266.
..
OIL SHALE PAGE 373
341. Forbes, F.; Kinsey, F. W. "The Dravo Traveling Grate Process for Oil
Shale Retorting" Proceedings Eastern Oil Shale Symposium, lexington,
KY, Oct. 11-13; Institute for Mining and Minerals Research, University
of K~ntucky: Lexington, KY, 1982; 235-240.
342. Forbes, F.; Kinsey, F. W.; Colaianni, L. J. MOravo Process in the Means
Oil Shale Project" Proceedings Eastern Oil Shale Symposium, Lexington,
KY, Nov. 13-16; Institute for Mining and Minerals Research, University
of Kentucky: Lexington, KY, 1983; 337-344.
343. Forbes, F.; Kinsey, F. W.; Colaianni, L. J. "Thou Shalt Not Waste - Total
Resource Utilization Using the Dravo Retorting Process" In Proc. 2nd
Australian Workshop on Oil Shale, Brisbane, 6-7 Dec. 1984; 182-187.
344. ·Process Description" Report: Oravo Engineering Companies, Inc.
345. Jones, W., II; lyzinski, D.; Miller, J. B.j Cugini, A. V.; Antezana, F. J.
"Gulf Shale Oil Upgrading Process Technology" 17th Oil Shale Symp. Proe.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1984; 133-142.
346. Weichman, 8. "The Superior Process for Development of Oil Shale and Asso
ciated Minerals" 7th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press:
Golden, CO, 1974; 25-43.
347. Weichman, B. "Oil Shale Is Not Dead" 9th Oil Shale Symp. Proe.; Gary, J.
H., Ed.; CSM Press: Golden, CO, 1976; 71-84.
348. Boyda, R. B. "The Superior Oil/Davy McKee Circular Grate Retort" Sympo
sium Papers: Synthetic Fuels from Oil Shale and Tar Sands, Louisville,
KY, May 17-19; Institute of Gas Technology: Chicago, IL, 1983; 479-487.
OIL SHALE PAGE 374
349. Knight, J. H. j Fishback, J. W. "Superior's Circular Grate Oil Shale
Retorting Process and Australian Rundle Oil Shale Process Design" 12th
Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1979; 1·16.
350. Pforzheimer, H., Jr. "Update on Paraho" Symposium Papers: Synthetic
Fuels from Oil Shale and Tar Sands, Nashville, TN, May 17-19; Institute
of Gas Technology: Chicago, IL, 1981; 315-322.
351. "Paraho Files for Protection under Chapter 11" Platt's Oi1gram News
Service 25 Oct. 1985, 63(206), 4.
352. "Paraho Removes Chapter 11 Yoke" Platt's Oi1gram News Service
8 Apr. 1986, 64(149), 4-5.
353. "New Paraho is Beginning a Two-Year $2.5 Million Test using Shale Oil to
make. an Extra-Tough Asphalt" Platt's ai/gram News Service 24 June 1987,
65(121), 5.
354. Jones, J. B., Jr.; Reeves, A. A. "Retorting of Solid Carbonaceous
Mater; alII U. S. Patent 3,736,247 (ass igned to Paraho Corporat i on); 1971.
355. Jones, J. B., Jr. "Paraho Oil Shale Retort" 9th Oi I Sha I e Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1976; 39-48.
356. Lovell, P. F. 'Production of Utah Shale Oils by the Paraho DH and Union-B
Retorting Processes" 11th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM
Pres,s: Golden, CO, 1978; 184-192.
357. Kunchal, S. K. "Energy Requirements in an Oil Shale Industry Based On
Paraho's Direct Combustion Retorting Process" 8th on Shale Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1975; 21-29.
OIL SHALE PAGE 375
358. McKee, J. M.; Kunchal. S. K. "Energy and Water Requirements for an Oil
Shale Plant Based on Paraho Processes" 9th on Shale Symp. Proc.;
Gary, J. H., Ed.: CSM Press: Golden, CO, 1976: 49-64.
359. Pforzheimer, H., Jr. "The CUrrent Status of Paraho' s Oil Shale Deve 1 op
ment" Symposium Papers: Synthetic Fuels from Oil Shale and Tar Sands,
Atlanta, GA, 3-6 Dec. 1979: Institute of Gas Technology: Chicago, IL,
1979: 469-477.
360. Pforzheimer, H .• Jr. "Paraho - Outlook for Commercial Development" 9th
Oil Shale Symp. Proc.; Gary, J. H., Ed.: CSM Press: Golden, CO, 1976;
65-70.
361. Greaves, M. J. "Scale-Up of the Paraho Process for Retorting Oil Shale"
Symposium Papers: Synthetic Fuels from Oil Shale and Tar Sands, Nashville,
TN, May 17-19; Institute of Gas Technology: Chicago, IL, 1981: 323-343.
362. Berg, C. U.S. Patent 2,501,153; 1950.
363. Hartley, F. L.; Brinegar, C. S. 5th World Petroleum Congress Proceedings
Section II; Fifth World Petroleum Congress, Inc.: New York, 1959: 37-47.
364. Dhondt, R. O. "Union's Retort B Technology" Symposium Papers: Synthetic
Fuels from Oil Shale and Tar Sands, Atlanta, GA, 3-6 Dec. 1979: Institute
of Gas Technology: Chicago, IL, 1979; 439-448.
365. Snyder, G. B.; Pownall, J. R. "Union Oil Company's Long Ridge Experimental
Shale Oil Project" 11th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM
Pre$s: Golden, CO, 1978: 158-168.
366. Duir, J. H.; Deering, R. F.; Jackson, H. R. Hydrocarbon Processing
May 1977, 147-150.
OIL SHALE PAGE 376
367. Ouir, J. H.; Griswold, C. F.; Christolini, B. A. Chem. Eng. Progr.
February 1983, 45-50.
368. MacArthur, O. Personal communication.
369. Barnet, W. I. "Union Oil Company of California Oil Shale Retorting
Processes" In Oil Shale Processing Technology; Allred, V. D., Ed.; The
Center for Professional Advancement: New Brunswick, NJ, 1982; 169-187.
370. Hartley, F. L. "Oil Shale: Another Source of Oil For the United States"
presented to the Oil Daily's Third Annual Synthetic Energy Forum, New
York, June 10, 1974.
371. Resource Sciences Corporation "U.S.S.R. Oil Shale Presentation"; Tulsa,
OK, 3 Dec. 1975.
372. Prescott, J. H. Chem. Eng. 2 Feb. 1976, 66-67.
373. Yefimov, V. M.i Volkov, T. M.; Petukhov, E. F.; Rooks. 1. K. "Thermal
Processing of lump Oil Shale: The Kiviter Process" In Oil Shale Processing
Techpology; Allred, V. D., Ed.; The Center for Professional Advancement:
New Brunswick, NJ, 1982; 67-81.
374. 8aughman, G. L. Synthetic Fuels Data Handbook, 2nd Ed.; Cameron
Engineers, Inc.: Denver, CO, 1978; pp 76-80 and 90-92.
375. Cieslewicz, W. J. "Selected Topics of Recent Estonian-Russian Oil Shale
Rese,arch" Quart. Colo. Sch. Hines 1971, 66{l}. This entire issue is a
wide-ranging survey of published oil shale work from the U.S.S.R.
376. Doilov, S. K.; Efimov, V. M.; Ioonas, R. E.; Nazinin, N. A.; Piik, E. E.;
Raad, K. E.; Roox, I. K.; Serebryannikov, N.D.; Shaganov, J. Y.;
Ananiev. L. S.; Voikov, A. S. U.S. Patent 4,007,093; 1977.
OIL SHALE PAGE 377
377. Peterson, E. J,; Spall, W. D. "Physical and Chemical Characterization and
Comparison of Solids, liquids and Oils from Estonian and Green River Oil
Shales" Los Alamos National Laboratory Report LA-9722-MS, 1983.
378. Weil, S. A.; Feldkirchner, H. l.; Tarman, P. B. "Hydrogasification of Oil
Shal.e" ACS Div. Fuel Chern. Prepr. 1974, 19(2), 123-146.
379. Linden, H. R. U.S. Patent 3,703,052 (assigned to the Institute of Gas
Technology): 1972.
380. Schara, F. c.; Tarman, P. S.; Feldkirchner, H. l.; Weil, S. A. "Process
Shale to SNG" Hydrocarbon Processing. April 1977.
381. Schara, F. C. "The Application of the IGT Hydroretorting Process to Eastern
Shale" Energy Topics - A Supplement to IGT Highlights, May 9, 1977.
382. Feldkirchner. H. L.; Janka, J. C. "The Hytort Process" Symposium Papers:
Synthetic Fuels from Oi7 Shale and Tar Sands, Atlanta, GA, 3-6 Dec.;
Institute of Gas Technology: Chicago, IL, 1979: 489-524.
383. Janka, J. C.; Dennison, J. M. "Devonian Oil Shale - A Major American
Energy Resource" Symposium Papers: Synthetic Fuels from Oil Shale and
Tar Sands, Atlanta, GA, 3-6 Dec.; Institute of Gas Technology: Chicago,
IL, 1979: 25-116.
384. Weii, S. A.; Rue, O. M. "laboratory Studies on Hydroretorting Eastern
Shales" Symposium Papers: Synthetic Fuels from Oil Shale and Tar Sands,
Nashville, TN, 26-29 Oct.; Institute of Gas Technology: Chicago, Il,
1981: 217-227.
OIL SHALE PAGE 378
385. Roberts, M. J. Hydroretorting of Eastern Devonian and Western Eocene
Shahs" Symposium Papers: Synthetic Fuels from Ofl Shale and Tar Sands,
Nashville, TN, 26-29 Oct.; Institute of Gas Technology: Chicago, IL,
1981; 277-300.
386. Schora, F. c.; Tsaros, C. L.; Janka, J. C. "Update on the Hytort Process"
Symposium Papers: Synthetic Fuels from OiT Shale and Tar Sands,
Nashville, TN, 26-29 Oct.; Institute of Gas Technology: Chicago, IL,
1981; 345-365.
387. Schara, F. Coj Janka, J. C.; Lynch, P. A.; Feldkirchner, H. "Progress in
Commercialization of the HYTORT Process" Proceedings Eastern Oil Shale
Symposium, lexington, KY, Oct. 11-13; Institute for Mining and Minerals
Resaarch, University of Kentucky: lexington, KY, 1982; 183-190.
388. Rex, R. C., Jr.; Feldkirchner, Ho; Janka, J. C.j Schara, F. C. "Selection
of a Commercial Plant Design for the Hytort Process" Symposium Papers:
Synthetic Fuels from on Shale and Tar Sands, louisville, KY, May 17-19;
Institute of Gas Technology: Chicago, IL, 1983; 531-543.
389. Rex, R. C., Jr.; Janka, J. C.; Knowlton, T. M. "Cold Flow Model Testing
of the Hytort Process Retort Design" 17th Oil Shale Symp. Proc.j Gary,
J. H., Ed.; CSM Press: Golden, CO, 1984; 17-36.
390. VYU, K. C. "Syncrude from Eastern Oil Shale" 14th Oil Shale Symp. Proc.;
Gary, J. H., Ed.; CSM Press: Golden, CO, 1981; 82-90.
391. Feldkirchner, H. L. "Hydrogasification of Oil Shale" Final Report (No.
GRI·79j0064j prepared by the Institute of Gas Technology for the Gas
Research Institute, October 1980.
OIL SHALE PAGE 379
392. Schlinger, W. G. "Hydrotorting - Use of Hydrogen for Improved Recovery of
Shale Oil" Quart. Colo. Sch. Hines 1967, 62(3), 699-706.
393. Schlinger, W. G. 'Hydrotorting Oil Shale" ACS Div. Petro Chem. Prepr.
1966,
394. Eastman, 0.; Schlinger, W. G. U.S. Patent 3,117,072; 1964.
395. Schllnger, W. G.; Eastman, O. U.S. Patent 3,224,954; 1964.
396. Schlinger~ W. G.; Jesse, D. R.j Tassoney, J. P. U.S. Patent 3,617,469;
1971,
397. Schlinger, W. G.; Jesse, D. R.; Tassoney, J. P. U.S. Patent 3,617,470;
1971,
398. Schlijnger, W. G.; Jesse, D. R.; Tassoney, J. P. U.S. Patent 3,617,471;
1971.
399. Schlinger, W. G.; Jesse, D. R.; Tassoney, J. P. U.S. Patent 3,617,472;
1971.
400. $chlinger, W. G.; Jesse, O. R.; Tassoney, J. P. U.S. Patent 3.565,784;
1971.
401. Ewert, W. M.; Scinta, J. "Hydrogen Retorting of Indiana Shale in a Loop
Reactor System" Symposjum Papers: Synthetic Fuels from Oil Shale and Tar
Sand~, Louisville, KY, May 17-19; Institute of Gas Technology: Chicago,
IL, 1983: 395-406.
402. Audeh, C. A. ACS Div. Petro Chem. Prepr. 1984, 29(1), 19-25.
OIL SHALE PAGE 380
403. Netz~l, D. A.; Miknis, F. P. "Hydrocarbon Type Analysis of Eastern and
West~rn Shale Oils Produced by the IGT HYTORT and Fischer Assay
Proc~dures" Symposium Papers: Synthetic Fuels from Oil Shale and Tar
Sand$, Nashville, TN, 26-29 Oct.; Institute of Gas Technology: Chicago,
IL, 1981; 229-250.
404. Punw~ni, O. V.; lau, F. S.; Roberts, M. J.; Hu, W. C. S.; Rex, R. C., Jr.
"Sta~us of Current Research for Hydroretorting of Eastern Oil Shales"
19thion Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO,
1986\ 82-93.
405. Rex" R. C., Jr. j Hu. W. C. S. "A Comparison of Hydroretorting and Thermal
Retort i n9 Facil it ies for the Process i"9 of Reastern Oil Shales"
Proc~edings 1986 Eastern Oil Shale Symposium, LeXington, KY, Nov. 19-21;
Instntute for Mining and Minerals Research, University of Kentucky:
Lexihgton, KY, 1986.
406. Punw~ni, O. V. Personal communication.
407. Lau,; F. S.; Rue, O. M.; Punwani, O. V.; Rex, R. C., Jr. ACS Div. Petro
Che~. Prepr. 1987, 32(1), 103-109.
408. lau,; F. S.; Rue, O. M.; Punwani, O. V.; Rex, R. C., Jr. "Fluidized-Bed
Gasi!fication of an Eastern Oil Shale" Proceedings 1987 Eastern Oil Shale
Sym~sium, leXington, KY, Nov. 18-20; Institute for Mining and Minerals
Res~rch, University of Kentucky: leXington, KY, 1987; xxx-xxx.
409. Roberts, M. J.; Punwani, O. V.; Hu, W. C. S.; Rex, R. c., Jr. IIEffects of
Tem~rature and Pressure on the Hydroretorting yields of Three Eastern
Oil Shales" Presented at AIChE National Meeting, Miami, 2-7 Nov. 1986.
OIL SHALE PAGE 381
410. Bur~el1 E. l.; Carpenter, H. C.; Johns, H. W. U.S. Bur. Mines Oil Shale
411.
412.
,
pro~ram Technical Progress Report - 16, 1969.
Bunte11, E. L.; Sterner, T. E.; Carpenter, H. C. J. Petrol. Teehnol.
1971, 22, 1520-1524.
cam*bell, G. c.; Scott, W. G.; Miller, J. S. "Evaluation of Oil Shale
Fratturing Tests Near Rock Springs, Wyoming" US Bur. Mines. Rept.
Inv1st. 7397, 1970.
413. Bur~el1, E. lo; Sterner, T. E.; Carpenter. H. C. US Bur. Mines Rept.
Inv'st. 7783, 1973. I
414. lon~, A., Jr.; Merriam, N. W.; Mones, C. G. "Evaluation of Rock Springs ,
Sit~ 9 In Situ Oil Shale Retorting Experiment" 10th all Shale Symp.
Pro~.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1977; 120-135.
415. law1or, D. l.; Latham, D. R.; Fausett, D. W.; Asplund, R. O. "Post-Burn
Stu1Y of the Rock Springs Site 9 In Situ Retorting Experiment" 12th ail
Shale Symp. Proe.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1979; 221-227.
416. Far~ier, O. S.; Virgona, J. L; Phillips, T. E.; Poulson, R. E.
"Entironrnental Research For In Situ Oil Shale Production" 11th Oil Shale
Symk. Proe.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1978; 81-99.
417. Wi11 iams, S. E.; Gauger, W. K.; Farrier, O. S. "Microbial Interactions
Wit~ Aqueous Effluents" 12th Oil Shale Symp. Proe.; Gary, J. H., Ed.;
CSMI Press: Golden, CO, 1979; 115-121. I
418. Ski~ner, Q. D.; Moore, T. S.; Asphund, R. 0.; Sexton, J. C.; Farrier, D. S.
"TO*iCologiCal EvalUation of an In Situ Oil Shale Process Water" 12th Oil
Shale Symp. Proe.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1979; 122-138. !
OIL SHALE PAGE 382
420. lekas, M. A. "Developing the Geokinetics/Department of Energy Horizontal
In Situ Retoritng Process - Final Report Contract No: DE-FC20-78LCI0787"
NTISI: Springfield, VA, 1985. (NTIS DE 86001051).
421. Lekas, J. M. "The Geokinetics Horizontal In Situ Retorting process"
422.
423.
424.
425.
426.
427.
Sym sium Papers: Synthetic Fuels from Oil Shale and Tar Sands, Nashville,
TN, 6-29 Oct.; Institute of Gas Technology: Chicago, IL, 1981; 251-259.
Lekap. M. A. "The Geokinetics Horizontal In Situ Retorting Process"
14t~ Oil Shale Symp. Proe.; Gary, J. H., Ed.; CSM Press: Golden, CO,
19811; 146-153.
Hattrri, K.; Takahashi, M. U.S. Patent 4,175,990; 1970.
zargr, D. P. U.S. Patent 4,205,610; 1980.
Brit~on, K. "limitations of Commercial Explosives and Blasting Caps and
the+ Effect on In Situ Blast Design" 18th on Shale Symp. Proe.; Gary,
J. ~., Ed.; CSM Press: Golden, CO, 1985; 109-122.
Brit~on. K. R. C. "Surface Uplift Blasting for Shale Oil Production" In !
Mec~niCS of Oil Shale; Chong, K. P; Smith, J. W., Eds; Elsevier Applied
Scie ce Publishers: New York, 1984.
McCa thy, H. L; Cha, C. Y. "The Oxy Modified In Situ Oil Shale Process
Deve opment and Update" 9th on Shale Symp. Proc.; Gary, J. H., Ed.; I
CSM fress: Golden, CO, 1976; 85-100.
428. Ric tts, T. E. ·Occidental's Retort 6 Rubblizing and Rock Fragmentation
Prog am" 13th on Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press:
Gold n, CO, 1980; 46-61.
OIL SHALE PAGE 383
429. Camp ell, J. H.; Peters, P.; Libicki, S. P.; Gregg, M. l.; Clarkson, J. E.
"Ana ysis of the Operation of Occidental's Modified In Situ Retort 6"
14th Oil Shale Symp. Proc.; Gary. J. H., Ed.; CSM Press: Golden, CO,
1981 99-117.
430. Loue s, R. A. "Final Report for the Period November 1, 1976 through April
30, 979, Occidental Vertical Modified In Situ Process for the Recovery
of 0·1 from Oil Shale - Phase I" Prepared for the U.S. Department of
Ener y under Contract No. DE-FC20-78LCI0036, 1979.
431. Nels n, R. M. "Final Report for the Period June I, 1979 through August
31, 981, Occidental Vertical Modified In Situ Process for the Recovery
of 0·1 from Oil Shale - Phase 2" Prepared for the U.S. Department of
Ener y under Contract No. DE-FC20-78LCI0036, 1982.
432. Romi , B. A. "Progress on Phase II of the DOE/OSSI Cooperative Agreement"
14th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO,
1981 91-98.
433. Rick tts, T. E. IIRubblization of Occidental's Retorts 7 and 8" 15th Oil
434.
435.
Shal Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1982; 341-360. ,
Stev~ns, A. L.; Zahradnik, R. "Results from the Simultaneous Processing
of Mpdified In Situ Retorts 7 and 8" 16th Oil Shale Symp. Proc.; Gary,
J. HI., Ed.; CSM Press: Golden, CO, 1983; 267-280.
BiCk~l, T. C. "Analysis of Occidental Vertical Modified In Situ Retorts
7 an 8" 16th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press:
Gold n, CO, 1983; 281-295.
OIL SHALE PAGE 384
436.
,. .... , .
Rick tts, T. E. "Improving Flow Uniformity in Vertical Modified In Situ
nil, ba19 .... l?Qt?de~_.ll.'tP"'O;i,1 S~a.t'" 'VttlQ_>n€rQ(' : .. G<,r'i __ J > _H _ ~ .. Erl_.!.._CSf!1. __
Summ ry," "Volume 2: Mining, Processing and Support Facilities," Rio
Blan 0 Oil Shale Project, Gulf Oil Corporation and Standard Oil Company
(Ind ana), May, 1977.
438. Berr, K. l.; Hutson, R. l.; Sterrett, J. S.; Knepper, J. C. "Modified In
Situ Retorting Results of Two Field Tests" 15th on Shale Symp. Proc.;
Gary J. H., Ed.; CSM Press: Golden, CO, 1982; 385-396.
439. Brau, R. L.; Chin, R. C. "Computer Model for In Situ Oil Shale Retorting:
Effe ts of Input-Gas Properties U 10th on Shale Symp. Proc.; Gary, J.
H., d.; CSM Press: Golden, CO, 1977; 166-179.
440. Doug n, P. M.; Reynolds, F. S.; Root, P. J. liThe Potential for In Situ
Reto ting of Oil Shale in the Piceance Creek Basin of Northwest Colorado"
6th il Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO,
1970 57-72.
441. Jaco s, H. R.; Marzinelli, M. J.; Udell, K. S.; Dougan, P. M. uLaboratory
Mode ing of In Situ Retorting of Oil Shale from the 'Leached Zone' of the
Para hute Creek Formation by Superheated Steam Injection" 13th Oil Shale
Symp Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1980; 62-72.
442. Doug n, P. M.; Docktor, L. "BX In Situ Oil Shale Project" 14th on Shale
Symp Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1981; 118-127.
OIL SHALE PAGE 385
443. Bar ick, H.; Kunchal. K.; Switzer, 0.; Bowen, R.; Edwards, R. "Final
Rep rt: The Production and Refining of Crude Shale Oil into Military
Fuels" Office of Naval Research. Applied Systems Corporation Navy
Con ract NOOOI4-75-C-0055; August, 1975.
444. Sha, H.; Kalfadelis, C. 0.; Jahnig, C. E. "Evaluation of Methods to
Pro uce Aviation Turbine Fuels from Synthetic Crude Oils Phase I" Air
For e Aero Propulsion laboratory, AFAPL-TR-75-10; Volume 1, 1975.
445. Kal adelis, C. D. "Evaluation of Methods to Produce Aviation Turbine
Fuels from Synthetic Crude Oils -- Phase II" Air Force Aero Propulsion
Lab ratory, AFAPL-TR-75-10; Volume 2, 1976.
446. Taylor, W. F.; Kaufman, J. l.; Brown, E. C.; Cunningham, A. R. "Evaluation
of to Produce Aviation Turbine Fuels from Synthetic Crude Oils
Air Force Aero Propulsion Laboratory, AFAPL-TR-75-10;
Vol e 3, 1977 and Addendum, 1978.
447. , C. G.; Robinson, E. T. "The Commercial-scale Refining of Paraho
e Shale Oil into Military Fuels. Final Report for the Period 10 Nov.
- 10 Mar. 1979" Navy Energy and Natural Resources R&D Office,
Rep rt No. NOOOI4-79-C-006J; July 1979.
448. Robinson, E. T. "Refining of Paraho Shale Oil into Military Specification
Fuels" 12th Oil Shale Symp. Proc.; Gary, J. H., Ed.; CSM Press:
CO, 1979; 195-212.
449. Sullivan, R. F.; Stangeland, B. E. "Converting Green River Shale Oil to
sportation Fuels" 11th Oil Shale Symp. Proc.; Gary, J. H., Ed.;
CSM Press: Golden, CO, 1978; 120-134.
OIL SHALE PAGE 386
450. Sull'van, R. F.; Stangeland, B. E. "Catalytic Hydroprocessing of Shale
Oil 0 Produce Distillate Fuels" ACS Div. Petro Chem. Prepr. 1977,
23(1 , 322-342.
451. Sul1 'van, R. F.; Stangeland, B. E.; Rudy. C. E.; Green, D. C.; Frumkin,
H. A. "Refining and Upgrading of Synfuels from Coal and Oil Shale Using
Adva ced Catalytic Processes: First Interim Report" U.S. DOE Report
No. e-2315-25; April, 1978.
452. Frum in, H. A.; Owens, E. J.; Sutherland, R. B. Chem. Eng. Progr. 1979,
75(9 , 64-72.
453. Su1l'van, R. F.; Stangeland. B. E.; Frumkin, H. A. "Refining Shale Oil"
Pres nted at the API Refining Department, 43rd Midyear Meeting, Toronto,
Onta io, Canada; 10 May 1978.
454. sun 'van, R. F.; Stangeland, 8. E. "Catalytic Hydroprocessing of Shale
Oil 0 Produce Distillate Fuels" In Refjning of Synthetic Crudes; Adv.
Chern Ser. 179; Gorbaty. M. l.; Harney, B. M., Eds.; American Chemical
Soci ty: Washington, 1979; 25-51.
455. Jet uel Looks to Shale Oil -- 1980 Technology Review Air Force Aero
Prop Ision Laboratory, AFWAL-TR-81-2063, 1981.
456. Jet uel from Shale Oil -- 1981 Technology Review Air Force Aero
Prop Ision Laboratory, AFWAL-TR-81-2135, 1981.
457. Moor, H. F.; Sutton, W. A.; Turrill, F. H.; long, R. P.; Johnson, C. A.i
Hett'nger, W. P., Jr. "Jet Fuels from Shale Oil via Extractacracking n
In 5 ale Oil Upgrading and Refining; Newman, S. A., Ed.; Butterworth
Publ shers: Woburn, MA, 1983; 223-241.
OIL SHALE PAGE 387
458. edock, J. P.; Reif, H. E.; Macris, A. "Manufacture of Aviation and
459. :~A. 'M. '; H;msrey~- ~: -l~'-ntva'i'uatlofl of-HYdr6cracf(,fii~r:CataTYsfs£for--·
rsion of Whole Shale Oil into High Yields of Jet Fuels" Air Force
Propulsion Laboratory, AFWAL-TR-81-2098, 1981.
460. Tai, A. M.; Hensley, A. L. ACS Div. Petro Chem. Prepr. 1977,
23(1), 322-342.
461. Tai , A. M.; Hensley, A. L. "Jet Fuels from Shale Oil by Single-stage
Hyd cracking" ACS Div. Fuel Chem. Prepr. 1982, 27(2), 187-200.
462. Tait, A. M.; Miller, J. T.; Hensley, A. L. AIChE Summer Heeting Paper
No. 16d, Cleveland, 1982.
463. Tait, A. M.; Miller, J. T.; Hensley, A. L. AIChE Annual Heetfng Paper
No. 5d, Los Angeles, 1982.
464. Tait, A. M.; Miller, J. T.; Hensley, A. L. "Direct Hydrocracking of
Shal Oil" In Shale Oil Upgrading and Refining; Newman, S. A., Ed.;
Butt rworth Publishers: Woburn, MA, 1983; 73-97.
465. Siko ia, S. G.; Board, J. G.; Wilcox, J. R.; Hilfman, l. "Shale Oil
ding for FCC Conversion" In Shale Oil Upgrading and Refining;
n, S. A., Ed.; Butterworth Publishers: Woburn, MA, 1983; 29-48.
466. Curtin, D. J. U.S. Patent 4,029,571; 1977.
467. Alle, S. K.; McArthur, D. P.; Whitehead, K. E. "Upgrading of Petroleum
Resi ues, Bitumen and Shale Oil" Proceedings of the Eleventh World
Pet leum Congressj Wiley: Chichester, 1984; Volume 4, 221-230.
468. Youn, D. A. U.S. Patent 4,046,674; 1977.
OIL SHALE PAGE 388
469. Rob; son, E. T.; Evin, C. G. uConvnercial-Scale Hydrotreating of Shale Oil"
In 5 ale Oil Upgrading and Refining; Newman, S. A., Ed.; Butterworth
Publishers: Woburn, MA, 1983; 49-72.
470. Haw, C. 0.; Schlesinger, M. D.; Ginsberg, H. H.; Hiteshue, R. W. USBM
Rept. Invest. 6548, 1964.
471. Mon ornery, O. P. Ind. Eng. Chem. Prod. Res. Oev. 1968, 7. 274-282.
472. Speight, J. G. The Chemistry and Technology of Petroleum; Marcel Dekker:
New York, 1980: pp 329-330.
473. Hal es, S. A. uNitrogen Compound Distribution in Shale Oil: Comparison
of ench-Scale and Commercial Hydroprocessing" In Shale Oil Upgrading
and Refining; Newman, S. A., Ed.; Butterworth Publishers: Woburn, MA,
198 : 159-181.
474. Hal es, S. A.; Thompson, L. F. "Nitrogen-Type Distribution in Hydrotreated
Sha e Oils: Correlation with Upgrading Process Conditions" 14th Ofl
Sha e Symp. Proc.; Gary, J. H., Ed.; CSM Press: Golden, CO, 1981;
235 248.
475. lyz nski, D.; Jones, W, II "Upgrading Eastern Shale Oil Via Gulf Shale
Oil Hydrotreating Process" Proceedings Eastern Oil Shale Symposium,
lex ngton, KY, Nov. 26-28; Institute for Mining and Minerals Research,
Uni ersity of Kentucky: Lexington, KY, 1984: 289-295.
476. Atk ns, A. R.; Fookes, C. J. R.; Muridan, A.; Stephenson, l. "Upgrading
Str tegies for Julia Creek Shale Oil" Proc. 3rd Australian Workshop
on il Shale, Brisbane, 15-16 May 1986: 229-234.