An Assessment of Oil Shale Technologies (Part 7 Of

Home , Mining, Oil shale industry, Petrochemical, Petrosix, Spent shale, TOSCO II process

Page Page Introduction ● .*. ... . * * ,, *,, , **,,.*.,. 119 Demonstration ...... 173 Summary of Findings...... 119 Chapter 5 References...... 175 An Overview of Oil Shale Processing ...... 120 List of Tables Oil Shale Mining...... 123 Surface Mining...... 123 Table No. Page Underground Mining ...... 125 18. Overall Shale Oil Recoveries for Several Processing Options ...... 155 Oil Shale Retorting...... ,128 19. Properties of Crude Shale Oil From True In Situ...... ,...... 128 Various Retorting Processes...... 156 Modified In Situ ...... 131 20. Supply of Finished Petroleum Products Aboveground Retorting ...... 137 in PADDs 2, 3, and 4 in1978 ...... 165 Advantages and Disadvantages of the 21. Technological Readiness of Oil Shale Processing Options...... 153 Mining, Retorting, and Refining Oil Recovery ...... 154 Technologies ...... 168 Properties of Crude Shale Oil ...... 155 List of Figures Shale Oil Refining...... ,..157 Shale Oil Upgrading Processes . ....,...... 158 FigureNo. Page Total Refining Processes ...... 160 19. Oil Shale Utilization ...... 121 Cost of Upgrading and Refining ...... 162 20. The Components of an Underground Mining and Aboveground Retorting Oil Markets for Shale Oil...... 163 Shale Complex...... 122 Shale Oil as a BoilerFuel ...... 163 21. An Open Pit Oil Shale Mine...... 124 Shale Oil as a Refinery Feedstock ...... 164 22. Open Pit Oil Shale Mining...... 125 Shale Oil as a Petrochemical Feedstock .. ...166 23. The Colony Room-and-PillarMining Issues and Uncertainties ...... 167 Concept ...... 126 24, Room-and-Pillar Mining on MultipIe R&D Needs and Present Programs...... 170 Levels ...... 127 R&D Needs ...... 170 25. The Original Mine Development Plan Present Programs...... ,..,...172 for Tract C-b ...... 128 Policy Options ...... **.,*,.. 173 26. True In Situ Oil Shale Retorting ...... 129 R&D...... ,.....173 27. Modified In Situ Retorting...... 132 Page Paqe 28. The RISE Method of MIS Oil Shale 42. The Superior Oil Shale Retort.. , . . . . . 150 Retorting ...... ,. 134 43. The Colony Semiworks Test Facility 29. Initial Modular MIS Retort Near Grand Valley, Colo...... 151 Development Plan for Tract C-a ...... 135 44, The TOSCO II Oil Shale Retorting 30. Present MIS Retort Development Plan System .,, , . . . . . 0 ., . . . . ., .,....,, 152 for Tract C-a ...... 136 45* The Lurgi-Ruhrgas Oil Shale Retorting 31. Retort Development Plan for the . ,0 ,0 ...... 153 Multi Mineral MIS Concept...... 136 46. Refining Scheme Used by the U.S. 32. Preparation of a Multi Mineral MIS Bureau of Mines to Maximize Gasoline Retort ...... 137 Production From Shale Oil ...... 161 33. A Set of Multi Mineral Misreports . . . 138 47* Refining Scheme Employing Coking 34. The Operation of an NTU Retort...... 140 Before an Initial Fractionation, Used by 35, Discharging Spent Shale Ash Froma Chevron U.S.A. in Refining 40-Ton NTU Retort...... , . . 141 Experiments. ., ...... 161 36. The Paraho Oil Shale Retorting System 143 48. Refining Scheme Employing Initial 37, The Paraho Semiworks Unit at Hydrotreating, Used by Chevron U.S.A. Anvil Points, Colo. ● , ● ● ● ● ● ● . ● . . . ● , . ● 143 in Refining Experiments ...... 161 38. The Petrosix Oil Shale Retorting 49. Refining Scheme Employing Initial System . .,..,.,...... ,,..0.,.,0. 145 Fractionation, Used by SOHIO in 39. The Petrosix Demonstration Plant, Prerefining Studies ...... , . . 162 Sao Mateus do Sul, Brazil...... 146 50. The Petroleum Administration for 4U. The Union Oil “B” Retorting Process.. 147 Defense Districts...... 165 41. Block Diagram for the Superior Multi Mineral Concept ...... 149 CHAPTER 5 Technology

Introduction

The mining and processing technologies that could be used to convert the oil

that can be used to convert kerogenf the or- shale to liquid and gaseous fuels; ganic component of oil shale, into marketable ● the upgrading and refining methods that fuels are discussed in this chapter. The char- could be used to convert crude shale oil acteristics of these technologies will influ- to finished products; ence the effects that an oil shale industry will ● the potential markets for shale-derived have on the physical environment, and their fuels; technological readiness will affect the rate at ● the technological readiness of the major which an industry can be established. The steps in the oil shale conversion system; following subjects are discussed: ● the uncertainties and the research and development (R&D) needs that are asso- ● the general types of processing methods and their major unit operations; ciated with each major unit operation; ● and the mining methods that could be used to ● remove oil shale from the ground and the policies available to the Government prepare it for aboveground processing; for dealing with the uncertainties, ● the generic types of retorting methods

Summary of Findings

Oil shale contains a solid hydrocarbon called advanced with other minerals. More oil shale ex- kerogen that when heated (retorted) yields combusti- perience has been acquired with underground ble gases, crude shale oil, and a solid residue called mining, particularly room-and-pillar mining, and spent, retorted, or processed shale. Crude shale oil preparing MIS retorts. Although uncertainties re- can be obtained by either aboveground or in situ (in main, the mining technologies should advance place) processing. In aboveground retorting (AGR), rapidly if presently active projects continue and the shale is mined and then heated in retorting suspended ones resume. vessels. In a true in situ (TIS) process, a deposit is first fractured by explosives and then retorted under- ● TIS is presently a very primitive process, although ground. In modified in situ (MIS) processing, a por- R&D and field tests are being conducted. The prin- tion of the deposit is mined and the rest is shattered cipal advantages of TIS are that mining is not (rubbled) by explosives and retorted underground. needed and surface disturbance from facility siting The crude shale oil can be burned directly as boiler and waste disposal is minimized. The principal fuel, or it can be converted into a synthetic crude oil disadvantages are a low level of technological (syncrude) by adding hydrogen. The syncrude can readiness, low recovery of the shale oil, and a po- also be burned as boiler fuel, or it can be converted tential for surface subsidence and leaching of the to petrochemicals or refined to obtain finished fuels. spent shale by ground water. Some of OTA’s major findings concerning these min- ● MIS is a more advanced in situ method. It is being ing and processing technologies are: further developed on two lease tracts and a pri- ● Limited areas of the oil shale deposits may be vately owned site. The Department of Energy amenable to open pit mining. This technique has (DOE) is providing substantial R&D support. The never been tested with oil shale but it is well- principal advantages of MIS are that large depos-

119 120 ● An Assessment of Oil Shale Technologies

its can be retorted, oil recoveries per acre affected $2.00 more per bbl than refining some of the are high, and relatively few surface facilities are poorer grades of domestic crude. required. However, some mining and some disposal of solid wastes on the surface are required, ● The initial output from the pioneer oil shale in- and the oil recovery per unit of ore processed is dustry will probably be marketed near the oil shale low relative to AGR methods. The burned-out MIS region. Once the industry is established, the shale retorts have a potential for ground water pollution. oil will probably be used as boiler fuel or refined in the Rocky Mountain States. A large industry will ● AGR also has a medium level of technological most likely supply oil to Midwest markets. A 1- readiness. Three retorts have been tested for sev- million-bbl/d industry could completely displace eral months at rates approaching one-tenth the ca- the quantities of jet, diesel, and distillate heating pacity of commercial-size modules. Others are still fuels that are presently obtained from foreign at the laboratory or pilot-plant stages. The princi- sources in the entire Midwest. pal advantage of AGR processing is high oil recovery per unit of ore processed. However, with some ● With the present technical status of the critical mining methods, oil recoveries per acre may be retorting processes, deploying a major oil shale in- lower than with MIS. Surface disturbance is high- dustry would entail appreciable risks of techno- est with AGR because of the extensive surface fa- logical and economic failure. Although much R&D cilities, and because large quantities of solid has been conducted, and development is proceed- waste are generated. ing, the total amount of shale oil produced to date ● The physical and chemical properties of crude is equivalent to only 10 days of production from a shale oil differ from those of many conventional single 50,000-bbl/d plant. Because of its primi- crudes. However, depending on the nature of the tive status, much basic and applied R&D is needed upgrading techniques applied, the syncrude can for the TIS method. The MIS approach and some be a premium-quality refinery feedstock, compar- of the AGR processes are ready for the next stage able with the best grades of conventional crude. of development—either modular retort demonstra- Shale oil is a better source of jet fuel, diesel fuel, tions or pioneer commercial-scale plants. Such and distillate heating oil than it is of gasoline. Al- demonstrations would be costly, but they would though some technical questions remain, the up- substantially reduce the risks associated with the grading and refining processes are well-ad- much larger capital investments needed to create vanced. Refining shale oil may cost from $0.25 to an industry,

An Overview of Oil Shale Processing Converting shale in the ground to finished broken with explosives to create a highly fuels and other products for consumer mar- permeable zone through which hot fluids kets involves a series of processing steps. can be circulated; and Their number and nature are determined by ● AGR processes in which the shale is the desired mix of products and byproducts mined, crushed, and heated in vessels and by the generic approach that is followed near the minesite. in developing the resource. The alternative approaches are: Figure 19 is a flow sheet for the steps common to all three options. How the steps would ● TIS processes in which the shale is left be integrated in an AGR facility is shown in underground, and is heated by injecting figure 20. In the first step, the oil shale is hot fluids; mined and crushed for aboveground process- ● MIS processes in which a portion of the ing, or the deposit is fractured and rubbled shale deposit is mined out, and the rest is for in situ processing. The main product is Ch 5– Technology 121

Figure 19.– Oil Shale Utilization

Mining & Crushing Wastes & or In situ retort b preparation byproducts 7 Treated Re-Use ~ % * Reinfection m ~ water Discharge a ~ o

Y ? - # Wastes & Residues Retorting Treatment # \ Disposal byproducts T A ~ ~ @ 6 ~ Sulfur, NH3, Industrial ● * byproducts markets 7 1 f Wastes & Upgrading< byproducts

9 ‘ransportatonm ‘ef’n’ng l’=% “strbut”n t-+ C==l aMay not be nee~e(j for some in S1 tu methods

SOURCE Off Ice of Technology Assessment raw oil shale with a particle size appropriate that contain sulfur and nitrogen compounds. for rapid heat transfer. Nahcolite ore can be Depending on the extent of the treatment, the one of the various byproducts from this step. upgraded product—shale oil syncrude-can Dust and contaminated water are among its be a high-quality refinery feedstock, compar- wastes, In the retorting step, the raw oil shale able with the best grades of conventional is heated to pyrolysis temperatures (about crude. In the refining step, which may be con- 1,000° F (535° C)) to obtain crude shale oil. ducted either onsite or offsite, hydrogen is Other products are the spent shale residue, added to convert the syncrude to finished pyrolysis gases, carbon dioxide, contami- fuels such as gasoline, diesel fuel, and jet nated water, and in some cases additional fuel. The syncrude, or the crude shale oil, nahcolite and dawsonite ore. The crude shale could also be used directly as boiler fuel. oil may be sent to an upgrading section in After refining, the fuels are distributed to which it is physically and chemically modified consumer markets. Refining also produces to improve its transportation properties, to waste gases and various contaminated con- remove nitrogen and sulfur, and to increase densates. its hydrogen content. (Crude shale oils from some in situ processes may not need upgrad- To protect the environment, contaminated ing before transportation. ) Contaminated air water must be treated for reuse in the oil and water, and in some units refinery coke, shale facility, for reinfection into the ground are the wastes produced along with gases water aquifer source, or for discharge into 122 ● An Assessment of 01/ Shale Technologies

Figure 20.—The Components of an Underground Mining and Aboveground Retorting Oil Shale Complex

FUEL PRODUCTS AND BYPRODUCTS FROM UPGRADING UNITS . Low Sulfur Fuel Oil . Ammorwa ● Liquefied Petroleum Gas ● Sulfur . Coke

UPGRADING UNITS u RETORT UNITS

-— . . w- -. == . - . ,.- . + . . . .. < -– , ,... ., . Shade Convevors, !* .; ;...... : - - ““ -= - . ,—. .s. . Processed . ., /. - L

. . . . . — - Y \ . b, 7 “ . . - ‘ Catchment Dam HEADFRAME DISCHARGE / Surface Runoff STATION Water reused -.=. r~L ~

Roof Upper Bench ---:, .4*! Loading Dnllmg and Blasting x . -’ BOMng ScaImg ,. -1 .-. — ——

gr~~;g;$f~

—. —- - - .“ 1? . ROOM AND PILLAR MINING (Several Hundred Feet or more i u Underground) SKIPTI LOADING STATION

SOURCE C b Shale 01 I Project Defa(led Deve/oprnenf Plan and Re/ated Mater/a/: Ashland 011, Inc and Shell 011 Co , February 1976, p 1-24 surface streams. Contaminated air and proc- bing) and dawsonite (a source of aluminum ess gases must be purified to meet Federal metal). In any case, spent shale from above- and State air pollution standards before they -ground processing must be moistened, com- can be discharged to the atmosphere. The pacted, and revegetated to prevent erosion waste gases from retorting, upgrading, and and leaching. Retorted shale in in situ retorts, refining are potential sources of ammonia and spent shale from surface operations that (for fertilizer and other uses) and sulfur (for is backfilled into underground openings, must sulfuric acid and many other materials). be protected from leaching by ground water. These can be recovered during the treatment steps and sold to industrial processors. In This chapter deals with the processing some portions of the Green River formation, technologies used in mining, retorting, up- the solid residues from mining and retorting grading, and refining. Technologies for treat- will contain nahcolite (which can be used to ing air, water, and solid wastes are discussed produce soda ash and for stack gas scrub- in chapter 8. Ch 5– Technology ● 123

Oil Shale Mining Both MIS and AGR require mining. In the the San Manuel copper mine in Arizona, case of AGR, the mined shale is conveyed to which yields about 50,000 ton/d of ore, About retorts where it is processed to recover shale 70,000 ton/d of 30 gal/ton oil shale would oil. With MIS, the shale may also be retorted have to be mined to support a single 50,000- aboveground, or it may be discarded on the bbl/d plant that used aboveground retorts. * surface as a solid waste. This mine would be substantially larger than Green River oil shale deposits are charac- the San Manuel mine. A 400,000-bbl/d indus- terized by their extreme thickness and by try of aboveground retorts would have to their extensiveness. The richer shale zones in mine about 560,000 ton/d—the equivalent of the Piceance basin, for example, are more 5 Bingham Canyon pits or 11 San Manuel than a thousand feet thick, in places, and are mines. If the same industry used only MIS, 2 about 230,000 to 460,000 ton/d would have to continuous over an area of 1,200 mi . Depos- be mined—the equivalent of four to seven its of this nature could be amenable to either ** Some of the mining surface mining (strip or open pit) or to under- San Manuel mines. ground mining methods (such as room and pil- techniques that could be used to achieve lar), depending on topographical features, ac- these levels of production are described cessibility, overburden thickness, presence of below. ground water in the mining zone, and many other factors. Surface mining may be feasible Surface Mining for very thick oil shale zones that are not deeply buried, especially if their average oil The two principal types of surface mining yield is not high, Because of the thickness of —open pit and strip—both have been widely the overburden, only a limited area of the used to develop coal seams and deposits of Piceance basin and somewhat more of the many other minerals. Their technical aspects Uinta basin and the Wyoming basins is amen- are fairly well-understood for these minerals. able to surface mining. In other areas, However, their feasibilities and effects vary streams have eroded gulleys and canyons with the nature of the ore body. Neither tech- through the shale beds, exposing some of the nique has yet been applied to the oil shales of richer shale zones. Shale that outcrops in the Green River formation. these areas, plus the shale in all deeply Surface mining is economically attractive buried beds, will probably be extracted by for large, low-grade ore deposits because it underground mining, permits high recovery of the resource and Despite the high price of crude oil, oil shale allows sufficient space for very large and ef- is a lean ore compared with many ores that ficient mining equipment. An open pit mine contain valuable metals. Economies of scale could recover almost 90 percent of the oil encourage massive mining installations, re- shale in a very thick deposit. Strip mining gardless of the mining method selected. A could provide even higher recoveries. In con- prospective developer once characterized trast, underground mining would recover less commercial-scale oil shale mines as “prodi- than 60 percent. One of the reasons that in- gious, ” because it connotated a larger size dustry’s bids on a lease for Federal tract C-a than “giant.’” A sense of this can be con- were so high was that the deposit could be veyed by comparing their capacities with those of more conventional mines. At present, *This assumes recovery of 100 percent of Fischer assay oil, the largest surface mine in the United States a recoverv efficiency that h:)s been achieved in tests of the is Kennecott Copper’s Bingham Canyon pit in “rOSCO 11 technology. **This assumes mining of 20 to 40 percent of the shnle in the Utah, which produces about 110,000 ton/d of retort volume, and an oil recovery of 60 percent of Fischer copper ore. The largest underground mine is assay. 124 ● An Assessment of 011 Shale Technologies mined by open pit methods. About five times wards to transmit the pressure without col- as much shale could be recovered by open pit, lapsing. which could have mined the entire deposit, Open pit mining was originally proposed than by underground room-and-pillar mining, for tract C-a by the Rio Blanco Oil Shale Co., which would have been limited to a relatively 2 the lessee. The pit was to have started in the thin zone. northwest corner of the tract, and eventually A mature open pit mine is shown in figure to have covered its entire surface. After a 21 and the steps in its operation in figure 22. few decades, freshly removed overburden In the first step, overburden is drilled and and spent shale from the aboveground retorts blasted loose over a large area above the oil would have been returned to the pit as back- shale zone. The burden is carried by trucks or fill. In the interim, the solid wastes would conveyors to an offsite disposal area. When have been disposed of on a highland to the enough burden is removed to expose the shale northeast of the tract boundaries. This con- beds, the shale itself is drilled and blasted, cept was abandoned when the Department of and is hauled from the pit for processing in the Interior (DOI) refused to allow offtract aboveground retorts. As mining proceeds, a waste disposal. Rio Blanco later switched to huge hole is formed, extending from the top of MIS techniques because the alternative—un- the overburden to deep into the oil shale de- derground mining—would have reduced re- posit. The walls of the pit are under pressure source recovery and threatened profitable from the overburden, and must be angled out- operations. * At present, there are no plans for any open pit mines, although Rio Blanco may reconsider its original plan if offsite dis- 3 Figure 21 .— An Open Pit Oil Shale Mine posal were allowed. In strip mining, overburden is removed with a dragline—a massive type of scraper shovel. When the dragline has filled its scoop, it pivots and dumps the burden into an adjacent mined-out area. One difference between open pit and strip mines is that in strip mining, the burden is simply cast into a nearby area; in open pit, it must be moved far from the minesite to prevent interfering with the development of the pit. Strip mining has not been proposed for any of the Green River deposits. Surface mining of most oil shale deposits is made difficult by the great thickness of the overburden that covers them. In the center of Off site disposal the Piceance basin, for example, the 2,000-ft- thick oil shale zones are buried under about Ov n 1,000 ft of inert rock and very lean oil shale. This does not necessarily preclude surface mining, because the deposits are generally characterized by a favorable stripping ratio—the ratio of overburden thickness to ore- body thickness. The thick beds in the center of the Piceance basin have 1 ft of overburden SOURCE Hear/rigs on 0// Sha/e Leasing, Subcommittee on Minerals, Materials, and Fuels of the Senate Commltee on Interior and Insular Affairs, 94th Cong 2d sess , Mar 17, 1976, p 84 *This change in plans is discussed in vol. 11, Ch 5–Technology 125

Figure 22.—Open Pit Oil Shale Mining

I Jl-.- OVERBURDEN I t 40 -– —“— “—— “

OIL SHALE 1

SOURCE U S Energy Oullook–An Ifrtenrn RePort The National Petroleum Council Washington D C 1972 for every 2 ft of oil shale—a stripping ratio of Underground Mining 1:2. With coal, a stripping ratio of 10:1 is often economically acceptable. A study by the Many underground mining procedures National Petroleum Council indicates that have been proposed for oil shale deposits5-10 open pit mining would be favored for strip- but to date only room-and-pillar mining and ping ratios between 2:1 and 5:1, strip mining mining in support of MIS retorting have been for smaller ratios, and underground mining tested at any significant scale. In room-and- for larger ratios than 5:1.’ pillar mining, some shale is removed to form large rooms and some is left in place, as pil- However, the economic principles of coal lars, to support the mine roof, The relative mining should be applied with caution to oil sizes of rooms and pillars are determined by shale. Removing 1,000 ft of overburden to re- the physical properties of the shale, by the cover 2,000 ft of shale might be possible in thickness of the overburden, and by the theory, but the pit’s boundaries would be so height of the mine roof. Most of the deposits extensive that it would have to be located on of commercial interest are very thick and a very large tract. Furthermore, the large have relatively few natural faults and fis- “front-end” investment in removing overbur- sures. The ore itself resists compression and den many years in advance of retorting would vertical shear stresses. These properties probably make open pit mining of deep depos- allow the use of large rooms, and relatively its uneconomical. Also, strip mining would little shale needs to be left as unrecoverable not be feasible in many parts of the oil shale pillars. region, even those with favorable stripping ratios, because the dragline would have to The U.S. Bureau of Mines (USBM) studied reach to the bottom of a 3,000-ft-thick layer of underground mining of oil shale at the Anvil overburden and oil shale. It is not clear that Points Experimental Station in the late 1940’s 11 such a machine could be built. and early 1950's. The primary purpose of 126 ● An Assessment of 011 Shale Technologies

the mining program was to supply raw shale mining plan proposed for Colony Develop- for the Bureau’s retorting experiments, but ment’s 46,000-bbl/d facility in the Piceance the program was also designed to develop a basin. 12 The rooms are 60 ft wide; the pillars safe, low-cost mining method. The room-and- are 60 ft square; and the mine roof is 60 ft pillar technique was adopted after extensive high. Mining is conducted in two 30-ft-high testing. From their studies, which included benches. The upper bench is mined first by enlarging the rooms until the roof failed, the drilling blastholes into the walls of a produc- investigators concluded that for safe mining tion room, and breaking the shale loose with conditions the rooms should be 60 ft wide explosives. The broken shale is carried by with pillars that are 60 ft on a side. trucks to the crushers where it is crushed to the size range required by the TOSCO 11 Many of the modern mine designs have aboveground retorts. The walls and the roof been patterned after the USBM experimental of the new room are then “scaled” to remove mine. The design depicted in figure 23 is the shale that was loosened by the blasting but

Figure 23.—The Colony Room-and-Pillar Mining Concept

SOURCE R W Marshall Colony Development Operation Room. andPlllar 011 Shale Mlnlng, ” Quarterly 01 the Colorado School o/ Mines, VOI 69, No 2 April 1974. PP 171 184 Ch 5–Technology ● 127 did not fall. Holes are drilled into the roof, coveries would be possible if the pillars were and roof bolts are inserted to assure its integ- subsequently mined out or if the mining were rity, thus protecting the miners from roof conducted on multiple levels. (See figure 24, ) falls. The USBM studies indicated that roof bolts would have to be installed in the access Figure 24. —Room-and-Pillar Mining passageways but not in areas that were ac- on Multiple Levels tively being mined. These production rooms would be vacated long before there was any serious danger of roof falls. & The lower bench is mined next, The sequence is similar except the blastholes are drilled into the floor of the upper bench, and additional roof bolting is not needed, The cy- \ cle of drilling, blasting, loading, scaling, and ~ Pillar roof bolting was designed to produce about . Room 66,000 ton/d of shale. About 60 percent of the , >ill shale in the mining zone was to be removed ~ Mining for processing in the aboveground retorts. Level The rest was to remain in the support pillars. Colony estimated that enough shale is present in the 60-ft seam to support the retorting facility for at least 20 years. The same type of mine was proposed by the Colony partners for development of tract C-b, after considering longwall mining, long-hole blasting, continuous mining, block caving, open pit mining, in situ processing, and many other options. The major advantages of the 13 SOURCE Hearing on 0// Shale Leas/rig Subcommittee on Minerals Materials room-and-pillar method were identified as: and Fuels of the Senate Comm It lee on I n Ierlo r and I n su Iar Affairs 94th Cong 2d sess Mar 17 1976 p 83 it is highly flexible and can easily be modified to accommodate new condi- The mine eventually would have extended tions, new equipment, or technological under the entire surface of tract C-b, as advances: shown in figure 25. However, the concept it can be highly mechanized and high was abandoned when it was discovered that overall production rates can be achieved the shale in the mining zone was highly frac- because many areas can be mined simul- tured—a condition that would have required taneously: large support pillars thus reducing resource the mine openings are relatively easy to recovery to uneconomic levels. All of the ventilate: original partners subsequently withdrew development of access passageways is from the tract, and it is now being developed also a production operation because oil by MIS methods by Occidental Oil Shale and shale would be removed; and Tenneco Oil Co. the mine could be designed to minimize Mining to prepare for MIS retorts is dis- surface subsidence. tinguished from room and pillar both by the disadvantages were the high cost of the volume of the deposit that is disturbed and by roof support system and the relatively low re- the configuration of the disturbed areas. As covery. In this regard, it was estimated that indicated, the underground mines both on the only 30 to 50 percent of the shale in a 75-ft- Colony property and on tract C-b would have thick interval could be recovered. Higher re- been developed in a 60-ft-high mining zone; .

128 ● An Assessment of 01/ Shale Technologies

Figure 25.—The Original Mine Development Plan for Tract C-b

#~– BoUNDARy OF TRACT C-B

...1

/ - :...--. 1------+------J- mlllr--nll-----~”--!n ‘- -- I , i ~I ~- ! I N ~ 1 i 1~------l------il ! II . 11 II - II1------!1: ,II II I 1 -.. .-. -. J . -. J 4111} 1- I t ‘------1 11 l--~ ‘-J: I I RAISE TO I o 20m 0: ~ ‘ uRFACE \ I I ~: I -i II 1--- ’’-----;1 ‘---”----1 ------‘ II Lc!_!_J[ i - ::::::T-- .- ., —— — —. I/_ EXHAUST I ST- / : 11111 -––- ‘- - - –-=---” / “ 7 SHAFT 1. /’ ~ ------.:- .“. 1 I sHAm, .tqw i I 1. , ‘ 1------,1 ‘-’ - Z%i‘ ?!’;’:;‘“i ‘ ] ‘ [....-..-:::-”:::::;T / r - - -... ~ - ~- ; ~-- -- .–. -- SHOP EXHAUST RAISES i 000000[ I I /[ I noooooc.--.._A. I ‘i I------. . . . . j II I I I I ~------II I! 11111 I I I l:%%% 1 L------j= PLANT 8SHA~ PILLAR - - } ‘ --- ”---; +i 1“--- ‘--- ‘ ‘\ I 11 I+w I ‘-

SOURCE C-b Shale 011 Project, Detaded Deve/OPmefrt plan and Relafed MaterlalS, Ashland 011, Iflc , and Shell 011 Co , February 1976, p IV-1 1 present plans for the MIS operations on in which the MIS retorts will be created, plus tracts C-a and C-b call for recovering oil from shafts from the surface for ventilation, drain- the shale in a 750-ft-high zone. However, the age, passageways for the miners, and trans- actual mining required to support this devel- portation of combustion air and retorting opment will take place in relatively confined products. Some of the mining plans are dis- areas. The mine will consist primarily of fair- cussed later in the section on MIS retorting ly small underground openings in the region methods.

Oil Shale Retorting The three generic methods for recovering True In Situ shale oil from oil shale deposits are described below. Brief discussions of some of the more The sequential steps in TIS processing are: significant R&D programs that have been conducted as part of their development are in- 1. dewatering, if the deposit occurs in a eluded. ground water area; Ch. 5–Technology ● 129

2. fracturing or rubbling if the deposit is of the Piceance basin where ground water not already permeable to fluid flow; has dissolved salt deposits to leave large 3. injection of a hot fluid or ignition of a rubble-filled zones. It is estimated to contain portion of the bed to provide heat for py- about 550 billion bbl of shale oil in place. 14 rolysis; and There are interconnected fractures and voids 4. recovery of the oil and gases through in other areas of the formation, but these, in wells. general, have only a very limited permeabili- The principles of TIS processing are illus- ty. The permeability of most of the zones that trated in figure 26. Several types have been appear to have commercial promise is essen- proposed that differ from each other with re- tially zero. Deposits that lie near the surface spect to the methods for preparing and heat- could be fractured by injecting water or ex- ing the deposit. All use a system of injection plosive slurries, but mining would probably and production wells that are drilled accord- be needed to increase the permeability of ing to a prescribed pattern. One that is com- deeply buried deposits. MIS or AGR proc- monly used is the “five-spot” pattern in which esses are more appropriate for the deeper re- four production wells are drilled at the cor- sources. ners of a square and an injection well is In 1961, Equity Oil Co. began developing a drilled at its center. The deposit is heated TIS process for the Leached Zone, which was through the injection well and the products tested in the Piceance basin between 1966 are recovered through the production wells, and 1968. It involved dewatering a portion of For efficient TIS processing, the deposit the zone followed by injecting hot natural gas. must be highly permeable to fluid flow, which Pyrolysis gases and a small amount of oil is true of portions of the Green River oil were swept in the natural gas stream to pro- shales. A good example is the Leached Zone duction wells through which they were

Figure 26. —True In Situ Oil Shale Retorting

1? 11 PRC‘DUC1NGWWQM13

SOURCE B F Grant Retorting 011 Shale Underground—Problems and Posslbllltles, ouarfedy of the Co/orado Schoo/ of M(nes VOI 59, No 3, July 1964, p 40 130 . An Assessment of 01/ Shale Technologies

pumped to the surface. The gas was sepa- of the fracturing techniques used have been rated from the oil, reheated, and reinfected chemical explosives, electricity, and injecting into the deposit. The oil had a much lower high-pressure air and water. These methods pour point, viscosity, and nitrogen content have been used to enhance recovery of con- than oils from aboveground retorts. * These ventional petroleum; oil shale fracturing favorable characteristics may have been re- poses a similar problem. Nuclear explosives lated to the solvent properties of the natural were also proposed in the 1960’s, but were gas, and to the absence of oxygen from the re- not tested because of their potential for harm- torting zone. The quality of the retort gases ing the environment, A nuclear test—Project was also good, partly because of their natural Rio Blanco—was conducted in the Piceance gas component, and partly because during re- basin to fracture sand formations containing torting combustion and the decomposition of natural gas. The test failed. carbonate minerals were minimal. ** The earliest TIS work was by Sinclair Oil Process development was not pursued be- Co., between 1953 and 1966. A thin section of cause too much of the natural gas was lost in shale in the Mahogany Zone of the Piceance the unconfined formation. In 1968, Atlantic basin was fractured by injecting air. The bed Richfield Co. (ARCO) purchased an interest in was ignited, although with difficulty, and a the process and resumed its development. In small quantity of shale oil was collected be- 1971, a revised concept was announced in fore the hot shale swelled and sealed the which superheated steam, rather than natu- fractures. After additional tests, Sinclair con- ral gas, was to be used to heat the deposits. In cluded that the zone’s limited permeability 1977, DOE contracted with Equity to test the would not permit profitable operations. * concept in a 50()-f t-thick seam in the Leached Research on TIS processing began at Zone of the Piceance basin. The seam under- USBM in the early 1960’s. In 1974, the pro- lies about 0.7 acre of surface, and contains grams and personnel were transferred to the about 700,000 bbl of oil in place. It has been Energy Research and Development Adminis- estimated that as much as 300,000 bbl could tration, and in 1978, they moved to DOE. The be recovered over a 2-year period, with about R&D programs have included laboratory ex- half of the oil produced in the first 7 months. ’s periments, computer simulations, pilot-plant Steam injection has begun at the site, and will studies, and field tests. Among the latter continue through 1981. No detailed estimates were tests of electrical, hydraulic, and ex- have yet been released of production rates or plosive fracturing and combustion retorting retorting efficiency. If the process proves to near Rock Springs, Wyo. These revealed be technically and economically feasible, it some of the problems associated with the TIS could be applied in portions of both the Pice- approach. In one early test, for example, ance and the Uinta basins. after inert gas at about 1,300° F (7050 C) was At present, no work is being done in slight- pumped into a fractured formation for a peri- ly fractured formations, but much research is od of 2 weeks, about 1 gal of shale oil was re- being performed to develop methods for in- covered. In another experiment in a zone that creasing their permeability by enlarging nat- contained 7,800 bbl of shale oil in place, it ural fractures and creating new ones. Some was estimated that only 60 bbl of the close to 1,000 bbl that were released were actually *rI’hese properties have economic significance. Pour point is recovered. the lowest temperature at which oil will flow. Oils with high pour points are hard to transport because they solidify at nor- Low oil recoveries are often associated mal ambient temperatures. Viscosity is a measure of a fluid ”s with TIS processing because of the large im- resistance to flow. Oils with high viscosity are expensive to pump. Reducing the high nitrogen content of most crude shale oils consumes hydrogen, which is costly. *rI’he results of a mathematical simulation indicated that **Both of these processes produce carbon dioxide, which is about 18 years of continuous steam injections would be re- a major constituent of gases produced by some above-ground re- c~uired to heat the shale to pyrolysis temperatures within a 30- torts. ft radius of a fracture. Ch 5–Technology ● 131 permeable blocks of shale in the fractured process can be demonstrated for thin depos- formation. These cannot be fully retorted in a its that are covered by less than 100 ft of reasonable length of time. Irregular fractur- overburden. It has been estimated that there ing patterns that can cause the heat carrier are at least 6 billion bbl of shale oil in place in to bypass large sections of the deposit are this type of deposit. It is common in the Uinta another problem. Oil shale that is located in basin,16) and at present, would be most eco- the bypassed regions will not be retorted, nomically developed by strip mining. And, even if all of the shale in a fractured area were retorted, much of the oil would not Geokinetics has been investigating the sur- reach the production wells, but would remain face uplift approach since 1973, and is pres- trapped in the pores of the spent shale or ently burning its 20th retort in Utah. The would diffuse beyond the production wells, to largest retort prior to 1979 measured 130 ft be lost in surrounding areas. by 180 ft by 30 ft thick. A retort about 200 ft on a side and 30 ft thick will be required to To develop methods for improving recov- demonstrate the technical feasibility of the ery, an extensive R&D program has been con- process. By 1982, DOE and Geokinetics hope ducted at the Federal research center in Lar- to have developed a commercial-scale opera- amie, Wyo. Aboveground batch retorts with tion with a production capacity of 2,000 capacities of both 10 and 150 tons of shale bbl/d.17 have been used to simulate in situ retorting, but under more controlled conditions than ex- Modified In Situ ist underground. Oil recoveries of up to 67 percent have been achieved, and an experi- In MIS processing, the permeability of oil ment in a “controlled-state” retort achieved shale deposits is increased by mining some 95-percent oil recovery. * shale from the deposit and then blasting the In 1976 and 1977, the program was ex- remainder into the void thus created. The panded to include field tests of different frac- two-step process is depicted in figure 27. In turing and retorting techniques under cost- the first step, a tunnel is dug to the bottom of sharing contracts with Equity Oil, Talley - an oil shale bed, and enough shale is removed Frac, Inc., and Geokinetics, Inc.** The Equity to create a room with the same cross-section- program has been described. The Talley-Frac al area as the future retort. Holes are drilled program was terminated after the fracturing through the roof of the room to the desired method failed. The Geokinetics process uses a height of the retort. They are packed with ex- fracturing technique called surface uplift in plosives that are detonated in the second which an explosive is injected into several step. A chimney-shaped underground retort wells and detonated to fracture the shale and filled with broken shale results. The access make it permeable to fluid flow, The shale is tunnel is then sealed, an injection hole is ignited by injecting air and burning fuel gas drilled from the surface (or from a higher through one well. It is pyrolyzed by the heat mining level) to the top of the rubble pile. The that sweeps through the bed in the gas pile is ignited by injecting air and burning stream, Oil and gases are pumped to the sur- fuel gas, and heat from the combustion of the face through outlying production wells. It is top layers is carried downward in the gas hoped that the technical feasibility of the stream. The lower layers are pyrolyzed, and the oil vapors are swept down the retort to a *rI’he large retorls resemble the classic Newia-Tex:]s-Utah sump at the bottom from which they are (N’I’U) retort that is described later in this chapter. Because of pumped to the surface. The burning zone the higher percentage of void volume in the rubble, these retorts provided better simulations of hlIS processing thtin of moves slowly down the retort, fueled by the ‘1’1S. “1’he controlled-state retort was much smfiller and was residual carbon in the retorted layers. When equipped for better cent rol of retortin~ renditions. the zone reaches the bottom of the retort, the **Tbe procurement also included a contract with Occidental Oil Shale, Inc., to develop its MIS process, which is discussed in flow of air is stopped, causing combustion to the next section. cease. 132 An Assessment of Oil Shale Technologies

Figure 27. —Modified In Situ Retorting

STEP A STEP B GAS TO ---– –~ PURIFICATION ~~, ~BLOWER RECYCLE GAS . .COMBUSTION — AIR .—

OVERBURDEN OVERBURDEN / — ,- d:——. F--- ~7— 7– 7— BLASTHOLES ,/,/ /, ’ <-b-, .< , /‘, , / //, ,/ ‘/ ‘/ ,/ ,!’ , /// , , /,/ 1:/ (![, ‘,,/ /, // ,’/ , / {’/// k, . J nv’ ] //,/1‘ ‘ i \ RUBBLED SHALE

LEAN SHALE ‘/////Al REMOVED FOR

//////l I OR DISPOSAL

/ P-——————=——————”’

SOURCE T A Sladek Recent Trends In 011 Shale — Part 2 Mlnlng and Shale 011 Extraction Processes Mjnera/ /rrdus/rles Bu//e/lrr VOI 18 No 1 January 1975, p 18

Occidental Oil Shale, Inc., a subsidiary of Occidental Petroleum (Oxy) has demonstrated the MIS process, 18 Oxy’s work began in 1972 on the D. A. Shale property along Logan Wash—a private tract on the southern fringe of the Piceance basin about 50 miles northeast of Grand Junction, Colo. To date, six retorts have been burned at the site, with varying degrees of success. The first retort was about 32 ft on each side and 70 ft high. About 60 percent (1,200 bbl) of the shale oil in place was recovered, comparable to the best results from USBM’s experiments at Lara- mie, The next two retorts were slightly larger, and performed similarly, The fourth retort was of nearly commercial size—120 ft on each side and 270 ft high. It contained about 32 times as much shale as the first retort and yielded about 25 times as much oil. The fifth was the same size but was designed for a much lower void volume. The rubbling

Phofo credif OTA staff step did not evenly distribute the void volume, and the performance was poor. Only 40 per- Head frame at tract C-a cent of the oil was recovered. The burning of Ch 5–Technology ● 133 retort 6 began in 1978, but the sill pillar—the wall mining or mechanical underreaming, layer of unbroken shale that caps a retort— could be used in other areas. It might be pos- collapsed into the rubble pile. Operations sible to operate mechanical underreaming were disturbed while the top was resealed, machines by remote control from the surface, but retorting was eventually completed, and thus reducing or even eliminating the need about 40 percent of the oil was recovered. for miners to work underground. None of DOE will share the costs of retorts 7 and 8, these approaches has been tested in any oil which are scheduled for 1980 and 1981, shale deposit. The oil shale at the Logan Wash site is not Other MIS techniques that use vertical- considered to be of commercial quality be- burn patterns have also been developed. For cause of its low kerogen content. In 1976, Oxy example, the firm of Fenix and Scisson de- acquired access to the higher quality re- signed two systems for underground mining, sources of tract C-b by exchanging its tech- rubbling, and retorting in the vertical mode. z’) nical knowledge for a half interest in the To date, they have been tested only with a lease. In 1978, Ashland Oil Co., Oxy’s partner computer model. DOE’S Lawrence Livermore in the C-b Shale Oil Venture, withdrew and Laboratory has also developed an MIS tech- left Oxy in full charge. In 1979, Tenneco Oil nique called rubble in situ extraction (RISE) Co. purchased a half interest in the lease for with the aid of computer simulation and lab- $110 million and is proceeding to develop the oratory experiments in pilot-scale above- tract in cooperation with Oxy, If present -ground retorts.21-23 (See figure 28. ) Several plans are followed, Oxy’s technology could be levels are mined, and a portion of the deposit used to produce about 57,000 bbl/d by 1985. is removed at each. The remaining shale is Oxy’s technique uses a vertical burn con- broken with explosives. Sufficient broken figuration— that is, the combustion zone pro- shale is then removed so that there is a total gresses vertically through the shale bed. It is void volume of 20 percent in the retort area. also theoretically possible to advance the The rubble is then ignited at the top, and re- burn front horizontally, in much the same torting proceeds as in the Oxy system. way as it is done in TIS processing. A crude The RISE approach was originally pro- version of this approach was implemented in posed by Rio Blanco Oil Shale Co. for tract Germany during World War II, when a few C-a, but the firm is now going ahead with its MIS retorts were created by digging tunnels own process, which has benefited from tech- into oil shale deposits and then collapsing nical information acquired under a licensing their walls into the void volume. These opera- arrangement with Oxy. Livermore’s modeling tions were short-lived because oil recoveries studies and laboratory experiments are con- were extremely low, and they were very hard tinuing. to control. 19 The initial plans for developing tract C-a by Horizontal MIS might be practical if a tech- MIS methods are shown in figure 29. They nique could be developed to remove large sec- would have involved five precommercial re- tions of oil shale strata. One possibility is to torts of increasing size. The present plan, use solution mining: the injection of fluids into which was adopted after purchase of Oxy’s the formation to dissolve soluble salts from MIS technology, is shown in figure 30. In the among the oil shale layers, The result would new development plan, a small pilot retort be a honeycomb pattern of voids that could (retort “O”) will be followed by two demon- then be distributed throughout the area to be stration retorts of increasing size. The largest retorted by injecting and detonating an explo- (retort “2”) will be close to commercial scale. sive slurry. This method would be limited to Several options of different size are being areas like the Leached Zone or the Saline considered for retort 2, with one option a re- Zone that contain significant concentrations tort with dimensions 60 ft by 150 ft by 400 ft of soluble salts. Other methods, such as long- high, The method by which the shale is rub- 134 An Assessment of Oil Shale Technologies

Figure 28.—The RISE Method of MIS Oil Shale Retorting . — --- —-- —--- —- ——— ~“” ~~~--.—-~ --- -—-—-— ‘- .—.— =.

/ / -/5 OIL OUT /

A. Isometric

I 1 I

. . . . . ? .“ .“.’.s+ I 1 Development proceeds simultaneously on several levels. At each Level A level, horizontal drifts are driven the full width of the retort block, and a vertical slot is bored to provide void volume for blasting. About 200/. of the broken shale is removed after each blasting operation. The rest is left in the retort volume. (1 . . . . : Sublevel Level B 4 Starting slot ...... ~Leve’c‘“’”””””””””””””’!!;:- B. Elevation

SOURCES: 01/ Shale Retorting Technology, prepared for OTA by Cameron Engineers, Inc., 1978; and C. K. Jee, et al., Rewew and Analyws of Od Shale Techno/ogy— Vo/ume 3. Mod/tied In Situ Tec/rrro/ogy, report prepared for the U.S Department of Energy by Booz-Allen & Hamilton, Inc , under contract No, EX-76-C- 01-2343, August 1977, p. 6. Ch 5–Technology ● 135

Figure 29.— Initial Modular MIS Retort Development Plan for Tract C-a

COMMERCIAL FIELD 2 1 ,, .’ ~., ,. - .- r c’ ‘r) / ‘> t -“

SOURCE RIO Blanco 011 Shale Project Revised Deta~/ed De~e/OP~eflt P/an Tracf C a Gulf 011 Corp and Standard 011 Co (Indiana), May 1977 P 125 bled, and the fraction of the shale that is volume before the next higher layer is mined from each retort area are two of the blasted. Through this technique, Rio Blanco major differences between the Rio Blanco ap- hopes to obtain uniform size distribution in proach and Oxy’s technique. Oxy has tested the rubble. This is believed to be a key techni- several rubbling methods, including drilling cal requirement for efficient MIS retorting. blastholes up from a room at the bottom of the Rio Blanco also proposes to mine twice the retort, drilling down from the top, and boring volume (40 v. 20 percent) as is contemplated a central shaft the full length of the retort by Oxy. and then blasting the surrounding shale into the shaft. In Rio Blanco’s approach, a room will be created at the bottom of the future re- Another MIS method that is still being de- tort, blastholes will be drilled from the sur- signed is the integrated in situ technology pro- face into the roof of the room, and explosives posed by Multi Mineral Corp. for recovering will be detonated sequentially at different shale oil, nahcolite, alumina, and soda ash levels, The shale above the room will thereby from oil shale deposits in the Saline Zone of be blasted loose in layers, with each layer of the Piceance basin. This zone underlies the rubble falling to the bottom of the retort Leached Zone, is relatively free from ground 136 ● An Assessment of Oil Shale Technologies

Figure 30.— Present MIS Retort figure 31, If technical and economic feasibili- Development Plan for Tract C-a ty is indicated during the test program, Multi Mineral would proceed to a commercial-scale \ Down-Hole facility that would produce 50,000 bbl/d of shale oil, 10,000 ton/d of nahcolite, 1,000 ton/d of alumina, and 20,000 ton/d of soda ash. The Multi Mineral approach resembles the

Escape% Shaft RISE technique in that mining and rubbling are conducted at several levels along the Ventilation Shaft height of the retort, It departs from RISE in Off-Gas Shaft that, after rubbling, the broken shale in the retort is removed from the bottom, crushed I’HF- Access Shaft and screened, and returned to the top in a continuous operation. This part of the retort development plan is shown in figure 32. The rubbled shale will contain free nahcolite and

+1 I I Ill II II II z SeDarator %= Figure 31 .— Retort Development Plan for the Multi Mineral MIS Concept

CROSS CUT

-- (LEVEL B) .-

SOURCE The Pace Co Consultants and Engineers Inc Cameron Synthe(lc Fue/s Report VOI 16 No 3 September 1979 p 28 water, and contains extensive resources of nahcolite and dawsonite in addition to oil shale. Development is hindered because the zone is deeply buried (about 2,000 ft) below the surface of the basin. To reduce the costs of its R&D program, Multi Mineral has proposed to use an 8-ft-diameter shaft that was drilled by USBM in 1978 through the Leached Zone and into the Saline Zone. In the first phase, mining and mine safety methods will be tested, and about 8,000 tons of nahcolite and 11,000 bbl of shale oil will be produced. The nahcolite will be used for stack-gas scrubbing tests in a powerplant. In the sec- (LEVE~D; ond phase, a retorting module will be used to -- produce up to 3,000 ton/d of nahcolite, 10,000 bbl/d of shale oil, 200 ton/d of alumina, and STOPE FLOOR SOURCE B Welchman, Sa//ne Zone 0// Sha/e Deve/opmen/ by [he /nlegrated 3,000 ton/d of soda ash. The modular retort /n S/fU Process Multi Mineral Corp Houston Tex December 1979. and the access shafts and drifts are shown in p9 Ch 5– Technology ● 137

Figure 32.— Preparation of a Multi Mineral residual carbon, and soda ash and aluminum MIS Retort oxide—the products of the thermal decomposition of dawsonite. The last three products remain in the retort rubble. In retort 2, the re- < HOLE NO. 1 sidual carbon is being gasified by injecting a mixture of steam and air. The heat from the gasification reaction is carried to retort 3. In HOLE NO. 2 ➤ retort 1, the hot gasified shale is being cooled by passage of cold recycle gas. The heat thus BACKFILLED n recovered is conveyed to retort 2. After the I shale in retort 1 is cooled, the soda ash and the aluminum oxide can be leached out with water. The leachate is then pumped to the surface and processed to recover its mineral SHAFT values.

Temperature control is the key to the entire operation. Oil alone could be recovered with a combustion-type method, such as Oxy’s, but because of the high temperatures, the decom- 2200’ position of the dawsonite would produce com- \)/ LEVEL pounds insoluble in water. The gas-recycle \ heating method would avoid this problem by maintaining lower retorting temperatures. The operation also depends on the ability of explosive rubbling techniques to produce broken shale that can be fed to conventional crushing equipment. Overall, the method is very interesting because of its potential for w \ simultaneously recovering fuel and minerals from deposits that may not be accessible with SOURCE B We(c hman Sa/~ne Zone O() Shd/e Deve/oprnent by fhe /nlegrd(ed In SJtu Process M u It I M I neral Corp H o uston Tex December 1979 any other approach. However, too little is D 15 known about the various steps to permit a nahcolite that is still associated with the thorough evaluation at this time. large blocks of shale. Crushing will liberate most of the associated nahcolite, and screen- ing will separate it from the shale particles. Aboveground Retorting The shale will be backfilled to the top of the Hundreds of retorts have been invented in retort; the nahcolite is transported to the sur- the 600-year history of oil shale development. face for processing. The result would be a re- Most were never brought to the processing tort that is filled with uniformly sized oil shale stage but some were tested using laboratory- particles. With this configuration, Multi Min- scale equipment, and a few at pilot-plant and eral hopes to avoid the channeling and by- semiworks scales. None has been tested at a passing problems that may occur in TIS and scale suitable for modern commercial opera- MIS processing. tions. This section summarizes the technical In a commercial-scale facility, the retorts aspects of seven retorting systems that offer would be operated in sets of three, as shown the promise of being applicable in the near in figure 33, In retort 3, the oil shale is being future. One obsolete technology that is the pyrolyzed by injecting hot gases to produce basis for several of the modern systems is shale oil (which is removed to the surface), also discussed.

l–< ... -1 138 ● An Assessment of 01/ Shale Technologies

Figure 33.— A Set of Multi Mineral MIS Retorts

EXCESS GAS RETORT 1 RETORT 2 (RETORT 3 (HEAT RECOVERY) (CARBON RECOVERY) (PYROLYSIS) IIt COLD RECYCLE GAS I ‘;

TK======--====-=------=="*----=---*==------; I il ,1 II IIi

OIL I ‘1t 1, 1, 1, ;1 Ii 1, I ,1t

------a

SOURCE B Welchman, Sa//rre Zone” 0(/ Shale Deve/oprnerrf by the Irrtegrafed In SIfu Process, Multl Mineral Corp , Houston, Tex December 1979, p 11

Although aboveground retorts differ wide- They include the Nevada-Texas-Utah ly with respect to many technical details and (NTU) and Paraho direct processes de- operating characteristics, they can be scribed below, and also USBM’s gas grouped into four classes: combustion retort and the Union “A” retort. Class 2 retorts produce a spent Class 1: Heat is transferred by conduc- shale low in residual carbon and low-Btu tion through the retort wall. The Pump- retort gas. Their thermal efficiencies are herston retorts used in Scotland, Spain, high because energy is recovered from and Australia were of this class, as is the retorted shale. However, recovery the Fischer assay retort that was devel- efficiencies are relatively low—about 80 oped in the 1920’s. It is a laboratory to 90 percent of Fischer assay. device for estimating potential shale oil yields. Its oil yield is the standard to ● Class 3: Heat is transferred by gases which the retorting efficiencies of all that are heated outside of the retort ves- other retorts are compared. Because sel. Retorts in this class are also called conduction heating is very slow, no mod- indirectly heated. They include the Para- ern industrial retorts are in class 1. ho indirect, Petrosix, Union “B,” and Su- Class 2: Heat is transferred by flowing perior retorts discussed below, and also gases generated within the retort by the Union SGR and SGR-3, the obsolete combustion of carbonaceous retorted Royster design, the Soviet Kiviter, the shale and pyrolysis gases. Retorts in this Texaco catalytic hydrotort, and others. class are also called directly heated. These retorts produce a carbonaceous Ch. 5–Technology ● 139

spent shale and a high-Btu gas. Thermal oil. Three units that were operated in Austra- efficiencies are relatively low because lia during World War 11 produced nearly energy is not recovered from the residu- 500,000 bbl. As noted previously, USBM used al carbon, but oil recovery efficiencies two units to simulate in situ retorting. are high, from 90 to over 100 percent of The operating sequence for the NTU retort Fischer assay. is shown in figure 34. The unit is loaded with ● Class 4: Heat is transferred by mixing crushed oil shale and sealed. The gas burner hot solid particles with the oil shale. is lighted, and air is blown in. Once the top of They include the TOSCO II and Lurgi- the shale bed is burning (step A), the fuel gas Ruhrgas retorts described below, and is shut off but the air supply continues. As the also the Soviet Galoter retort. Class 4 re- air flows through the burning layer, it is torts achieve high oil yields (about 100 heated to approximately 1,500° F (815° C). percent of Fischer assay) and produce a This hot gas heats the shale in the lower lay- high-Btu gas. The spent shale may or ers and induces the pyrolysis of the kerogen. may not contain carbon, and thermal ef- The oil and gases produced are swept down ficiencies vary, depending on whether through the cooler portions of the bed to the the spent shale is used as the heat car- exhaust port (step B). The solid product from rier. The retorts are sometimes referred the conversion of the kerogen (residual car- to as indirectly heated, as in class 3, be- bon) remains on the spent shale and is con- cause they also lack internal combus- sumed as the combustion zone moves down tion, and produce a similar gas product. the retort, providing fuel for additional com- Several other conversion methods have been bustion and thereby heat for additional pyrol- developed that cannot easily be placed in ysis. When all the carbon is burned from the these classes. These include microwave heat- upper layers of the bed, the four zones shown ing, bacterial degradation, gasification, and in step C are formed. The top layer contains circulation of hot solids slurries. Although burned and decarbonized spent shale. The some of these processes have potentially val- second layer is burning, releasing heat for py- uable characteristics, they will not be dis- rolysis in the third layer. In the bottom layer, cussed in this section because they have not the shale is being heated but is not yet at py- yet been proposed for near-term commercial rolysis temperatures. application. As time passes, the top layer expands downward and the lower three zones move The Nevada-Texas-Utah Retort uniformly down the retort. When the leading edge of the combustion zone reaches the ex- The NTU retort is a modified downdraft haust port, oil production ceases, air injection gas producer similar to units used in the 19th stops, and the retort is emptied. The dumping century to produce low-Btu gas from coal. It of spent shale ash from the 40-ton NTU at is a vertical steel cylinder, lined with refrac- Laramie is shown in figure 35. The entire tory brick and equipped with an air supply cycle, from ignition to dumping, takes about pipe at the top and an exhaust pipe at the bot- 40 hours. tom. The top may be opened for charging a batch of shale; the bottom for discharging the NTU retorts are simple to operate, and re- spent shale after retorting. The unit was de- quire no external fuel except for small veloped and tested for oil shale processing by amounts of gas to start the retorting. They the NTU Co. in California in 1925, and was can process a wide variety of shales with oil also tested by USBM at Rifle, Colo., from 1925 recoveries ranging from 60 to 90 percent of to 1929. Two units with nominal capacities of Fischer assay. They are unsuitable for mod- 40 tons of raw shale were built and operated ern commercial applications because they by USBM at Rifle between 1946 and 1951. are batch units with high labor costs and They produced more than 12,000 bbl of shale small capacities. Over 600 150-ton retorts — .

140 ● An Assessment of 011 Shale Technologies

Figure 34.—The Operation of an NTU Retort

AIR SUPPLY — AIR SUPPLY AIR SUPPLY BLOWER BLOWER BLOWER

rGAS BURNER: ON GAS BURNER: OFF rGAS BURNER: OFF h- SHALE IGNITES: /%- ECOMBUSTION ZONE; SPENT SHALE ZONE: ED RESIDUAL CARBON BURNS } ALL RESIDUAL CARBON IN AIR STREAM BURNED PYROLYSIS ZONE: COMBUSION ZONE: KEROGEN CONVERSION RESIDUAL CARBON } WITHOUT FURTHER \ BURNING COMBUSTION \ N!!iN PYROLYSIS ZONE; KEROGEN CONVERSION N BURNER GAS AND 1 SHALE OFF-GAS ‘+ HEAT REMAINING PREHEAT ZONE: SHALE SHALE HEATED BY ) OFF-GAS TO BELOW l\ II PREHEAT ZONE; PYROLYSIS TEMPERATURE { ( ~1~1 \\ y

1- BURNER FLUE GAS LOW-BTU GAS AND OIL MIST Yt LOW-BTU GAS AND SMOKE, AND FUMES TO SEPARATION EQUIPMENT OIL MIST

STEP A STEP B STEP C

SOURCE T A Sladek, Recent Trends In 011 Shale– Part 2 Mining and Shale 011 Extract Ion Processes, M(nera/ /ndustr/a/ Elu//ef/n VOI 18 No 1 January 1975 p 5

would be needed to produce 50,000 bbl/d of to the gas combustion technology but it is crude shale oil. In contrast, plants using some more likely to be commercialized.-It was de- of the continuous technologies described be- veloped by Development Engineering, Inc., low would require only about six retorts for (DEI) in the late 1960’s, and was tested for the same production capacity. limestone calcining in three cement kilns in South Dakota and Texas. Each kiln had a ca- The Paraho Direct and Indirect Retorts pacity of 330 ton/d and was comparable to the largest gas combustion retort tested by An NTU retort becomes a Paraho direct re- the six oil companies. tort when it is turned upside down, made continuous, and mechanically modified. The Par- After verifying the solids-flow character- aho retorts are based on USBM’s gas combus- istics of the Paraho technology in the lime- tion retort which, in turn, evolved from the stone kilns, DEI leased the Anvil Points site NTU retorts tested at Anvil Points in the late from the Federal Government in May 1972, 1940’s. The gas combustion process was and began a 5-year program to develop the tested in 6-, 10-, and 25-ton/d units by USBM technology for oil shale processing. Funding between 1949 and 1955, and by a consortium was obtained from a consortium of 17 energy of six oil companies between 1964 and 1968. companies and engineering firms. In return The Paraho direct retort is similar in design for a contribution of $500,000, each company Ch 5– Technology ● 141

Figure 35.— Discharging Spent Shale Ash From a 40-Ton NTU Retort

SOURCE U S Department of Energy 142 ● An Assessment of 0il Shale Technologies was guaranteed a favorable royalty arrange- Figure 36.— The Paraho Oil Shale Retorting System ment in any subsequent commercial application of DEI’s technology. The Paraho Develop- RAW SHALE ment Corp. was formed to manage the proj- 1 OIL MIST ect. * Two retorts, a pilot-scale unit 4.5 ft in SEPARATORS diameter and 60 ft high and a semiworks unit ---- -—- 10.5 ft in diameter and 70 ft high, were in- MIST FORMATION stalled and tested after August 1973. They AND PREHEATING were used to produce over 100,000 bbl of - — ---- crude shale oil, some of which was used for ltl- OIL RETORTING OIL refining and end-use experiments by DOE (?ZONE w PRODUCT ru and the Department of Defense. Maximum GAS ELECTROSTATIC throughput rates reached about 400 ton/d in I-- — - --- I t PRECIPITATOR F 1 the semiworks unit. Both direct and indirect COMBUSTION RECYCLE GAS ZONE BLOWER heating modes were tested. - i ----—- 4 RESIDUE The Paraho retort is shown in figure 36, COOLING and the Anvil Points semiworks unit in figure AND GAS 37. In its direct mode, the retort is very simi- PREHEATING AIR BLOWER \ --- —-- lar to the older gas combustion design. How- ● 4 t * A A ever, significant differences exist in the shale /v feeding mechanism, in the gas distributors, ~ SpEED CONTROLLER‘y and in the discharge grate. During operation, RESIDUE raw shale is fed to the retort through a rotat- A. Direct Heating Mode ing distributor at the top. It descends as a moving bed along the vertical axis of the re- RAW tort. As it moves, it is heated to pyrolysis tem- SHALE peratures by a rising stream of hot combus- OIL MiST tion gases. The oil and gas produced are SEPARATORS swept up through the bed to collecting tubes and out of the retort to product separation MIST FORMATION equipment. The retorted shale retains the re- AND sidual carbon. As the shale approaches the PREHEATING ——-—- — [ ~OIL burner bars, the carbon is ignited and gives RETORTING OIL ZONE off the heat required for pyrolyzing addition- L — STACK al raw shale. Passing beyond the burner bars, GAS ELECTROSTATIC --—- —- t < HEATER PRECIPITATOR the shale is cooled in a stream of recycle gas 0 HEATING T and exits the bottom of the retort through the i RECYCLE GAS — BLOWER discharge grate. It is then moistened and sent ------t to disposal. RESIDUE COOLING PRODUCT GAS The retort may also be operated as a class AND GAS 3 retort. The configuration would resemble ~REHEATIN$ AIR BLOWER —— —-- directly heated operations except that air c?- would not be injected and the offgas steam would be split into four parts after oil separa- ‘)1( tion. One part would be the net product gas. RESIDUE Another would be sent through a reheating B. Indirect Heating Mode furnace and then reinfected into the middle of SOURCE C C Shlh, et al Techno/ogma/ Ovefv)ew Reporfs for Eight Sha/e 0// Recovery Processes report prepared for the U S Environmental Pro- *“ Paraho” is from the Portuguese words “para homen”’— tection Agency by TRW under contract No EPA.600/7.79-075 March for mankind. 1979, pp 20 and 23 Ch 5– Technology ● 143

Figure 37.— The Paraho Semiworks Unit at Anvil Points, Colo.

SOURCE Paraho Development Corp 144 ● An Assessment of Oil Shale Technologies the retort. A third would not be reheated but tion plant near the city of Sao Mateus do Sul would be reinfected through the bottom of the in the State of Parana. In the second stage, retort to cool the shale before discharge. The the full commercial plant would be built on fourth would be used for fuel in the reheating the same site, to include 20 Petrosix units, furnace. All heat for kerogen pyrolysis would each about 33 ft in diameter. Brazil is not be provided by the reinfected gases, and no committed to either stage, but if financing is combustion would occur in the retort vessel obtained in 1980, the first stage could be com- itself. pleted by 1983 and the second by 1985. In addition to the shale oil, the plant would To date, the Paraho retorting technology also produce sulfur and liquefied petroleum has been tested at about one-twentieth of the gases. Preliminary plans are also being pre- size that would be used in commercial plants. pared for two additional commercial-scale Paraho would like to test a full-size module plants south of the present demonstration producing about 7,300 bbl/d at Anvil Points, plant in the State of Rio Grande do Sul. and has submitted a proposal to DOE for such a program. Both direct and indirect heating Except for mechanical differences, the modes would be tested. Permission has been Petrosix retort is very similar to the Paraho obtained from the Department of the Navy to indirect retort described above. This similari- use large quantities of shale from the Naval ty is not surprising in view of the shared engi- Oil Shale Reserves for the program. An envi- neering heritage of the two systems. One ronmental impact statement (EIS) is being operational difference is that the Petrosix prepared for the project. Paraho has re- spent shale is discharged into a water bath sponded to a DOE Program Opportunity No- and pumped in a slurry to a disposal pond. tice for a $15 million engineering design study Paraho shale is discharged dry, with only a of modular oil shale retorting, and would little water added prior to disposal. base the design of the Anvil Points module on results of the study. Outcome of the procure- Little information has been released about ment has not yet been announced. the demonstration program. Oil characteristics have been described, but these have little The Petrosix Indirectly Heated Retort relevance to the processing of Green River shale. It can be predicted that the retort The Petrosix process was developed for the should have high oil recovery efficiencies and Brazilian Government by Petrobras, the na- produce a retort gas with high heating value. tional oil company, with the assistance of The spent shale would be carbonaceous. In Cameron and Jones, Inc., a U.S. engineering the demonstration plant, recovery of energy firm, Russell Cameron, later president of from the spent shale was not possible be- Cameron Engineers, worked on the gas com- cause of the slurry disposal method. In any bustion program at Anvil Points, as did John commercial plant, it is possible that the shale Jones, later president of Paraho. The system would be burned in a separate unit to pro- is depicted in figure 38. Figure 39 is a photo- duce heat for pyrolysis. graph of the demonstration retort that has been built and tested in Brazil. The retort is The Union “B’ Indirectly Heated Retort 18 ft in diameter, and has a capacity of 2,200 ton/d of Irati oil shale. It was built in 1972, The Union “B” retort is a class 3 retort that and tested intermittently until quite recently. evolved from the Union “A,” a class 2 retort, In 1975, a U.S. patent was issued for the proc- by the Union Oil Co. of California. Two other ess, A 50,000-bbl/d plant is planned by the systems, the SGR and SGR-3, have also been Brazilian Government, to be developed in a proposed by Union, Union describes them all two-stage project. The first stage would in- as continuous, underfeed, countercurrent re- volve construction of a 25,000-bbl/d facility torts. The “B” has not been field tested, but about 5 miles from the site of the demonstra- the “A” was tested in Colorado in the 1950’s Ch. 5–Technology ● 145

Figure 38. —The Petrosix Oil Shale Retorting System

=ALGAS<”’LSHALEFEED HIGH BTU GAS FEED HOPPER PRODUCT FOR PURIFICATION

DISTRIBUTOR ELECTROSTATIC PRECIPITATOR CONDENSER ● CYCLONE I SEPARATOR COMPRESSOR HEAVY SHALE OIL HEAVY PYROLYSIS HOT VESSEL v 9 SHALE OIL \ 00000 d ‘As WASTE WATER DISCHARGE HEATER MECHANISM D \ COOL GAS

SEAL SYSTEM

WATER RETORTED SHALE SLURRY JftJ - TO DISPOSAL

SOURCE 0// Sha/e Refer//ng Technology, prepared for OTA by Cameron Engineers, Inc , 1978 at up to 1,200 ton/d. Figure 40 is a sketch of withdrawn from the top of the pool. The gases the Union “B” design. It incorporates most of are split into three streams. One is reheated the design features of the “A.” and reinfected to induce additional kerogen During operation, shale is fed through the pyrolysis; one is used as fuel in the reheating bottom of the inverted-cone vessel. The re- furnace; and one is the net product. The shale torting process thereafter resembles that of a is discharged from the top of the retort and continuous NTU retort. Hot gases enter the falls into a water bath in the retorted shale top of the retort and pass down through the cooler. From there it is conveyed to disposal. rising bed, causing kerogen pyrolysis. Shale The key to all of Union’s retorting systems oil and gas flow down through the bed. The oil is the solids pump that is used to move the oil accumulates in a pool at the bottom, which shale through the retort. In the “B” design, seals the retort and acts as a settling basin the solids pump is mounted on a movable car- for entrained shale fines. Oil and gas are riage and is immersed in the shale oil pool. 146 ● An Assessment of 0il Shale Technologies

Figure 39.— The Petrosix Demonstration Plant, Sao Mateus do Sul, Brazil

SOURCE Cameron Engineers, Inc Ch 5– Technology . 147

Figure 40.—The Union Oil “B” Retorting Process Ii?

‘. “’ P-4=.+.’.’ AA-+--d 1

= OIL LEVEL

RETORTED SHALE / DISCHARGE TO WATER SEAL

RETORT * GAS \\\\\l\ll j?{! .h\\\\\\lllt 7

T r —

- ROCK PUMP M r m * SHALE OIL k!% m rI

TiT\\ \ \~\\\ ‘\\ \\~. \ \’ \l\\ . ~. . .1 \\\\ \“.’\\\\\\\\\\\\~.\

A. The Retorting System

SHALE IN SHALE IN SHALE IN SHALE IN I

B. Hock Pump Detail

SOURCE 0/1 Strale Reforf/ng Technology, prepared for OTA by Cameron Engineers Inc .1978 .

148 . An Assessment of Oil Shale technologies

The pump consists of two hydraulic assem- A block diagram for the Superior approach blies that act in sequence. (See figure 40.) is shown in figure 41. In step 1, the mined While the cylinder of one assembly is filling oil shale is crushed to pieces smaller than 8 with oil shale, the other is charging a batch of inches and screened. About 85 to 90 percent shale into the bottom of the retort. When this of the nahcolite in the shale is in the form of operation is completed, the carriage moves distinct, highly friable nodular inclusions that horizontally on rails until the full cylinder is become a fine powder during the crushing under the retort entrance. A piston then operation. This is screened from the coarser moves the oil shale in this cylinder upwards shale in step 2. The shale is then further into the retort, while the other fills with fresh crushed to smaller than 3 inches and is fed in shale from the other feed chute. The carriage step 3 to a doughnut-shaped traveling-grate then returns to its original position and the retort, which includes sequential stages for process is repeated. heating, retorting, burning, cooling, and discharging the oil shale feed. The retort is The “B” achieves high oil yields, and the sketched in figure 42. In the heating section, retort gas has a high heating value, although oil is recovered by passing hot gases through much of it is consumed in the reheating fur- the moving bed. In the carbon recovery sec- nace. The mechanical nature of the rock tion, process heat is recovered by burning the pump is troublesome because its moving residual carbon. In the cooling section, inert parts would be subject to wear during opera- gases are blown through the bed of spent tion. However, the pump is immersed in the shale, cooling the shale and heating the gases shale oil pool, which should provide adequate for use in the heating section. After dis- lubrication. Union appears to be satisfied charge, the cooled spent shale is sent to other with its reliability. units in which alumina is recovered from cal- In 1976, Union announced plans to build a cined dawsonite and soda ash from calcined demonstration plant on its private land in Col- nahcolite. The alumina is shipped to offsite orado. The “B” retort was to be used to pro- aluminum refineries; the soda ash to glass duce 7,000 bbl/d of shale oil from 10,000 ton/d plants; and the nahcolite to refineries and of oil shale. Later, the announced capacity powerplants for use as a stack-gas scrubbing was increased to 9,000 bbl/d. The demonstra- agent. tion plant, called the Long Ridge project, Superior chose the traveling-grate retort would be the first step towards a 75,000-bbl/d because it allows close temperature control, commercial-scale plant. Air quality permits important to maintaining dawsonite’s solubil- have been obtained for the modular plant, ity during the burning stage. Similar, simpler which could be completed before 1985. Con- devices have been used to sinter iron ore for struction has not begun because Union is steelmaking and to roast lead and zinc sul- awaiting financial incentives from the Feder- fides. Superior’s process has been tested for al Government. oil shale in pilot plants in Denver and Cleve- land. A commercial-scale plant would con- The Superior Oil Indirectly Heated Retort sume 20,000 ton/d of oil shale to produce The Superior retort is unique among the 10,000 to 15,000 bbl/d of shale oil, 3,500 to aboveground retorts discussed in this section 5,000 ton/d of nahcolite, 500 to 800 ton/d of because it is designed for recovery of sodium- alumina, and 1,200 to 1,600 ton/d of soda ash. bearing minerals in addition to shale oil. As In the early 1970’s, Superior proposed to discussed in chapter 4, the minerals nahcolite build a commercial-size demonstration plant and dawsonite occur in substantial quantities on its 7,000-acre tract in the northern Pi- in portions of the Piceance basin. They are ceance Basin. The deposit was to be devel- potential sources of sodium bicarbonate, oped by deep room-and-pillar mining on sev- soda ash, and aluminum. eral levels. The single retort was to produce Ch 5–Technology 149

Figure 41 .— Block Diagram for the engineering design and cost estimate for a Superior Multi Mineral Concept commercial-scale plant that would contain six TOSCO II retorts, and convert 66,000 ton/d of oil shale into about 46,000 bbl/d of shale oil. In 1969, ARCO joined Colony, and a 14EH---TI I I second semiworks program began to test I SPENT SHALE SHALE scaleup procedures and to evaluate environmental controls. The program continued until

STEP 2 STEP 3 STEP 4 April 1972. Between 1965 and 1972 the semi- SHALE ASH ALUMINA & NAHCOLITE > OIL E works plant converted 220,000 tons of oil RECOVERY SODA ASH RECOVERY RECOVERY shale into 180,000 bbl of shale oil. In 1974, the 1968 cost estimate, which had been updated in 1971, was further revised to ● ● ● ● NAHCOLITE OIL SODA ASH ALUMINA incorporate operating data from the latter SOURCE 01/ Shale Retorting Technology, prepared for OTA by Cameron Engl part of the semiworks program and to ac- neers, Inc .1978 count for additional pollution controls. The 11,500 bbl/d of shale oil, plus the byproducts resulting cost estimate was about three times described above. Although the tract’s re- as large as the previous estimate. This cost sources are extensive, its L-shaped configu- escalation raised doubts about commercial ration does not favor large-scale develop- feasibility, and the project was postponed in- ment. Superior therefore proposed to ex- definitely. SOHIO and Cleveland Cliffs subse- change portions of the tract for adjacent land quently withdrew from the Colony group. controlled by the Bureau of Land Manage- Early in 1974, Tosco, ARCO, Ashland, and ment (BLM). Approval was delayed by BLM Shell purchased a lease for tract C-b from the review and by preparation of an EIS, a draft Federal Government. Initial development of which was recently released. In February plans for the tract involved a plant similar to 1980, BLM denied the exchange because the that proposed for Colony’s private tract. two tracts involved were not considered to These plans were also affected by the cost es- have equivalent value. The decision is open to calations, and in 1976 suspension of opera- review. tions on tract C-b was granted by the Govern- ment. In 1977 Tosco and ARCO withdrew The TOSCO H Indirectly Heated Retort from the tract. Shell withdrew in 1977. Ash- land, the remaining partner, teamed with The TOSCO II is a class 4 retort in which Oxy to develop the tract using MIS tech- hot ceramic balls carry heat to finely crushed niques. oil shale, It is a refinement of the Aspeco process developed by a Swedish inventor. Colony’s semiworks retort, which is about Patent rights were purchased by The Oil one-tenth of commercial scale, is shown in Shale Corp. (later Tosco) in 1952. Early devel- figure 43. The TOSCO II retorting system is opment work was performed by the Denver sketched in figure 44. Raw shale is crushed Research Institute, including testing of a 24- smaller than one-half inch and enters the sys- ton/d pilot plant. In 1964 Tosco, Standard Oil tem through pneumatic lift pipes in which the Co. of Ohio (SOHIO), and Cleveland Cliffs Iron shale is elevated by hot gas streams and pre- Co. formed Colony Development, a joint ven- heated to about 500” F (260° C). The shale ture company, to commercialize the TOSCO II then enters the retort, a heated ball mill, and technology. Between 1965 and 1967, the contacts a separate stream of hot ceramic group operated a 1,000-ton/d semiworks balls. As the shale and balls mix, the shale is 0 0 plant on its land near Grand Valley, Colo., heated to about 950 F (510 C), causing re- next to the site of Union’s semiworks opera- torting to take place. Oil vapors and gases are tions. In 1968, Colony prepared a preliminary withdrawn. The oil is condensed in a fraction- 150 . An Assessment of Oil Shale Technologies

Figure 42.—The Superior Oil Shale Retort

COOL RECYCLE GAS HOT S+LE HEATING RECYCLE + COOL RECYCLE ‘As -~ ~--- GAS / 1- -- 1

SHALE FEED

SOURCE. 0// Sha/e Retorturg Tec/rno/ogy, prepared for OTA by Cameron Eng!neers, Inc , 1978 (after Arthur G McKee & Co j ator. Some of the gases are burned in a heat- increasing energy recovery, and freeing the er to reheat the ceramic balls to about 1,2000 valuable retort gases for other uses. F (650° C). At the retort exit, the retorted shale and the cooled ceramic balls pass over The retort’s chief disadvantages are its a trommel, a perforated rotating separation mechanical complexity and large number of drum. The shale, which has been thoroughly moving parts. The ceramic balls are con- crushed during retorting, falls through holes sumed over time. The need for a fine feed size in the trommel. It is cooled and sent to dispos- results in crushing costs that are higher than al. The larger balls pass over the trommel those of systems that can handle coarser and are sent to the ball heater. feeds. On the other hand, all crushing operations produce some fine shale that could be Oil yields exceeding 100 percent of Fischer screened from the feed to coarse-shale re- assay have been reported for the TOSCO II torts and converted in auxiliary TOSCO II technology. However, overall thermal effi- units. Disposal of TOSCO II spent shale ciencies are low because energy is not recov- presents some problems because it is very ered from spent shale carbon, and much of finely divided and contains carbon. the product gas is consumed in the ball heater. Tosco has patented processes to burn the At present Tosco is participating in two retorted shale as fuel for the ball heater, thus projects that are committed to its proprietary Ch 5–Technology ● 151

Figure 43. — The Colony Semiworks Test Facility Near Grand Valley, Colo.

SOURCE Colony Development Corp 152 ● An Assessment of 011 Shale Technologies

Figure 44.—The TOSCO II Oil Shale Retorting System

FLUE GAS PREHEAT SYSTEM TO ATMOSPHERE STACK 4 —

WATER VENTURI BALLS WET SCRUBBER

RAW FOUL WATER To CRUSHED SEPARATOR z FOUL WATER SHALE -h ~ STRIPPER u CERAMIC a NAPHTHA BALL : F SLUDGE” HEATER v < - - HYDROCARBON ~ GAS OIL TO GAS 01 L VAPORS HYDROGENATION

n HOT BALLS BOTTOMS OIL TO : ACCUMULATOR DELAYED COKER Q ~ u. w ~

2 w t- 3 -1 BALL CIRCULATION E SYSTEM STACK

VENTURI WET SCRUBBER I HOT FLUE GAS FLUE GAS * MOISTURIZER SLUDGE PREHEAT SYSTEM HOT cnnu CTFAM+ I 4 SCRUBBER PROCESSED (INCLUDES INCINERATOR) WATER STACK SHALE VENTURI WET SCRUBBER

r+)t COOLER SLUDGE MOISTURIZER“’++ MOISTURIZED PROCESSED

COVERED PROCESSED SHALE CONVEYOR- ‘+$%spOsAL SOURCE 0// Shale Retorflng Technology prepared for OTA by Cameron Engineers, Inc , 1978.

retorting technology. The Colony project has The Lurgi-Ruhrgas Indirectly Heated Retort been mentioned previously. The other venture, the Sand Wash project, is being devel- The Lurgi-Ruhrgas retorting system uses a oped on 14,000 acres of land in the Uinta class 4 retort in which hot retorted shale car- Basin leased from the State of Utah. Unlike ries pyrolysis heat to oil shale. The process the Colony project, which has been sus- was developed jointly by Ruhrgas A. G. and pended pending resolution of economic uncer- Lurgi Gesellschaft fur Warmetechnik m. b. H., tainties, Sand Wash is proceeding towards two West German firms that have been in- commercialization in compliance with the due volved in synfuels production for decades. diligence requirements of the State leases. A The process was developed in the 1950's for plant similar to the proposed Colony facility is low-temperature coal carbonization. A 20- contemplated, but no schedule has been an- ton/d pilot plant was built in West Germany, nounced for its construction. At present, to test a variety of coals, oil shales, and work consists of monitoring the environment, petroleum oils. European shales were tested and preparing to sink a mine shaft. in the late 1960’s, and Colorado oil shale in Ch 5–Technology ● 153

1972. The plant has since been scrapped but Except for the mixer, the process is mechani- smaller retorts are available. cally simple and has few moving parts. It should be capable of processing most oil Coal processing plants using the Lurgi- shales, if they are crushed to a fine size. The Ruhrgas technique have been built in Japan, two major problems with the system appear West Germany, England, and Argentina. to be accumulation of dust in the transfer There are also two Yugoslavian plants, each lines and dust entrainment in the oil. Dusty built in 1963, with a capacity of 850 ton/d of crude oil is not a severe problem because the brown coal. The Japanese plant is also of dust is concentrated in the low-value residual commercial size, and uses the process to product when the oil is subsequently refined. crack petroleum oils to olefins. No large-scale As with the TOSCO II process, requirements oil shale plants have yet been built. for a fine feed material will result in high The retorting system is shown in figure 45. crushing costs. These costs would be partial- Finely crushed oil shale is mixed with hot re- ly offset by the ability to process the fine frac- torted oil shale in a mechanical mixer that re- tion from any crushing process, including sembles a conventional screw conveyor. Re- those used to prepare shale for coarse-shale torting takes place in the mixer, and gas and retorts. shale oil vapors are withdrawn. Dust is removed from these products in a cyclone sepa- In 1972, Lurgi proposed to develop its re- rator and oil is separated from the gas by con- torting technologies with Colorado oil shale. The program was not funded, but Lurgi’s in- densation. Retorted shale leaves the mixer terest in commercializing the technique has and is sent to a lift pipe where it is heated to continued. In recent years Lurgi has been about 1, 100° F (5950 C) in a burning mixture working with Dravo Corp. to interest U.S. of fuel gas and air. The hot retorted shale is firms in using the technology. At present, at then sent back to the mixer, and the process least one company—Rio Blanco—has ob- is repeated. tained a license to investigate the use of High oil yields have been reported for the Lurgi-Ruhrgas retorts. Present plans call for retort, and the product gas is of high quality. constructing a modular Lurgi-Ruhrgas retort that will be close to commercial size (2,200 ton/d). It will be used to retort the shale that Figure 45. —The Lurgi-Ruhrgas Oil Shale Retorting System will be mined during the preparation of MIS WASTE HEAT retorts on tract C-a.

-OLIDS1 ~-~ FLUE GAS wAsTE RECOVERY Advantages and Disadvantages of the Processing Options

OIL GASEOUS ANO The greatest advantage of TIS processing LIQUID PRODUCTS is that mining is not required, and spent shale CONDENSER DUST is not produced on the surface. The technical, economic, and environmental problems asso- MIXING SCREW TYPE RETORT SOLIDS ciated with above-ground waste disposal are SURGE BIN LIFT thereby avoided. MIS does involve mining and PIPE aboveground waste disposal, although to a TO WASTE lesser extent than with AGR. However, the MIS waste is either overburden or raw oil shale. Both materials are found naturally ex- posed on the surfaces of deeper canyons in ~ AIR - FUEL the oil shale basins. Although raw shale has (!! Reqwed) low concentrations of the soluble salts, it does SOURCE 0// .Shale Retorf/ng Technology, prepared for OTA by Cameron Engt neers Inc 1978 contain soluble organic materials that could 154 . An Assessment of 011 Shale Technologies

be leached from the disposal piles. It should retorts. It is expected that yields from MIS re- be noted that the presence of spent shale un- torts could be increased by injecting steam or derground has the potential to cause environ- hydrocarbon gases (as is done in Equity’s TIS mental problems because soluble salts could process), but it is doubtful that recoveries can be leached by ground water. Therefore, envi- reach those of carefully controlled above- ronmental controls will also be needed for -ground retorts. On the other hand, the pres- TIS and MIS methods. ent low efficiencies of MIS operations are partially compensated for by their ability to Surface Facilities convert very large sections of an oil shale deposit, by their ability to process shale of a TIS has another theoretical advantage in lower grade than would be practical for AGR, that the required surface facilities are min- and by their lower cost of preparing the shale imal, consisting only of wells, pumps, gas for retorting. cleaning and product recovery systems, oil storage, and a few other peripheral units. It is difficult to compare overall recoveries These facilities would probably resemble from MIS and AGR without making numerous those for processing of crude oil and natural assumptions about the operating characteris- gas. MIS requires more facilities to support tics of both systems. To make a rough com- the mine and the waste disposal program. parison, it could be assumed that AGR (with Aboveground processing, which uses the room-and-pillar mining) and MIS were to be largest number of facilities, causes the most applied to two 30()-f t-thick deposits with iden- surface disruption. tical physical characteristics. The net recoveries from several development options are summarized in table 18. The highest recovery Oil Recovery (100 percent) is for full-subsidence mining in The advantage of AGR is that the condi- conjunction with AGR processing. It should tions within the retorts can be controlled to be noted that full-subsidence mining, for achieve very high oil recoveries—approach- either MIS processing or AGR, would result ing or even in some cases exceeding the yields in extensive surface disturbance and could achieved with Fischer assay retorts. Retort- increase risks to the miners. Subsidence mining efficiencies are usually lower for MIS ing has never been tested in oil shale. Its po- processing and much lower for TIS because tential for surface disturbance could be re- of the difficulty in obtaining a uniform dis- duced by backfilling the mined-out areas with tribution of broken shale and void volume, spent shale from surface processing, The re- which in turn, makes it difficult to maintain torted shale in burned-out MIS retorts would uniform burn fronts and leads to channeling also reduce the severity of subsidence. and bypassing of the heat-carrier gases. The The three generic approaches to oil shale few estimates of the retorting efficiencies of processing also have various other advan- TIS operations that have been published have tages and disadvantages with respect to not been encouraging. (USBM achieved recov- water needs, environmental effects, financial eries of 2 to 4 percent in its field tests. ) MIS requirements, and social and economic im- retorting has not exceeded 60-percent recov- pacts. These aspects are discussed in the re- ery of the potential oil in the shale within the spective chapters of this report. Ch 5–Technology ● 155

Table 18.–Overall Shale Oil Recoveries for Several Processing Options

First-stage Overall shale Case retorting technology First-stage mining method Second-stage processing oil recoverya 1 AGR Conventional room-and-pillar mining on three levels 60-ft None 36% rooms, 60-ft barrier and siII pillars. 2 MIS Conventional MIS mining. 40% of shale left behind in barrier None 29% pillars 20% of the disturbed shale IS removed to the surface to provide void volume in the retorts 3 MIS As in case 2 AGR processing of the mined shale 41 %, 4 AGR As in case 1 Barrier and siII pillars collapsed and 74% retorted by MIS 5 MIS Full subsidence b None 48% 6 MIS Full subsidence b AGR processing of the mined shale 68% 7 AGR Full subsidence b None. 100%

aAssurnlng AGR reco~ers 100 ~ercenl of the potential 011 In the shale retorted MIS assumed to recover 60 percent of the potential 011 bEntlre deposit IS mned SOURCE Off Ice of Technology Assessment

Properties of Crude Shale Oil Crude shale oil (also called raw shale oil, One of the most important factors is the con- retort oil, or simply shale oil) is the liquid oil densing temperature within the retorting sys- product recovered directly from the offgas tem—the temperature at which the oil prod- stream of an oil shale retort. Synthetic crude uct is separated from the retort gases, The oil (syncrude) results when crude shale oil is lower this temperature, the higher the con- hydrogenated. In general, crude shale oil centration of low molecular weight com- resembles conventional petroleum in that it is pounds in the product oil, composed primarily of long-chain hydrocarbon molecules with boiling points that span The properties of crude shale oil from several aboveground and MIS retorting proc- roughly the same range as those of typical 24 petroleum crudes. The three principal differ- esses are listed in table 19. It is important to ences between crude shale oil and conven- note that the oils that are characterized were tional crude are a higher olefin content (be- produced in small-scale test runs under con- cause of the high temperatures used in oil ditions that may not be representative of shale pyrolysis), higher concentrations of oxy- those that will be encountered in other areas, gen and nitrogen (derived from oil shale kero- and with larger processing systems. The oils gen), and, in many cases, higher pour point from commercial-scale facilities in other and viscosity, parts of the oil shale region may have properties that are quite different, The physical and chemical properties of crude shale oil are affected by the conditions The properties of the oil produced by dif- under which the oil was produced. Some ferent AGR processes vary widely, but the retorting processes subject it to relatively differences between these oils and the in situ high temperatures, which may cause thermal oils are much more significant. In situ oils are cracking and thus produce an oil with a lower generally much lighter, as indicated by their average molecular weight. In other processes higher yields of material with relatively low (such as directly heated retorting) some of the boiling points, and would produce more low- lighter components of the oil are incinerated boiling product (such as gasoline) and less during retorting. The result is a heavier final high-boiling product (such as residual oil). In product. Others may produce lighter prod- general, the low yields of residuum make ucts because of refluxing (cyclic vaporization shale oils attractive as refinery feedstocks in and condensation] of the oil within the retort. comparison with many of the heavy conven- 156 . An Assessment of Oil Shale Technologies

I I I I

U20.Omcommo ml.no WCu;ko< I I I I m.

Ill Ill I I I I

Ocomlkmco.com-mOooooomm moo Illlil I I I I -f=lnu?amr=r=mmo

U30mo.-cn mcrl.mb- +-1--”oh I I I I co-

Iy I I I I

I I

m I I

a 1 I

I I

CNm I Ill% Ill

u- --

CcJm r=cDo-1 Lnr= Ar=- Cu =iXl -ok“’ I I I a3l- I I I

I I I z I I I

I I I I I I

mm (o-w m Lnm o-lam ‘1”””l-ok I:. I I I I I I z: co ...... , ., ...... , . . . $:::: . . . .,. . . . . ;~ ,...., ,. ,Q . . . .-LLILILLLILUILLULU Ch 5– Technology ● 157 tional crudes that are currently being proc- ferent major fuel product from an equivalent essed in the United States. For example, in amount of refining. situ oil from Oxy’s MIS process contains 1 to Among the negative characteristics of most 4 percent of material with a boiling range of over 1,000° F (5350 C), compared with 20 per- crude shale oils are high pour point, high cent for crude from Alaska’s North Slope, viscosity, and high concentrations of arsenic and 18 percent for Arabian Light crude.25 26 and other heavy metals and of nitrogen. The Other foreign crudes, such as Kuwait and pour point and viscosity are of economic im- Arabian Heavy, contain even larger residuum portance because transporting viscous oil fractions. Oils from some AGR processes con- that has a high pour point is difficult and costly, thus suggesting the need for pretreatment tain as much as 30 percent of residuum As shown in table 19, in which, although higher than the contents of in prior to marketing. * situ crudes, is substantially lower than that situ oils with their relatively low pour points for many conventional crudes. and viscosities could be marketed without pretreatment but they would retain their high Coal-derived liquids are often regarded as nitrogen contents. This would reduce their alternatives to shale oil feedstocks. However, value as refinery feedstocks and boiler fuels. syncrudes from coal have a much higher yield High concentrations of arsenic and other of gasoline and low-boiling distillates than shale oil, with little or no material boiling at metals are a disadvantage because they poison refining catalysts, especially in hydro- temperatures above 850° to 1,000° F (4550 to genation units. They must be removed prior to 535° C), The coal liquids would be well-suited catalytic processing, and a variety of physi- for gasoline production because their higher cal and chemical methods have been devel- concentrations of lower boiling constituents 28 29 when refined would yield the desired light oped for this purpose. It should be noted naphtha fractions. Shale oil, on the other that the concentrations of heavy metals in hand, has a much higher concentration of crude shale oil will vary with the location of high-boiling compounds, and would favor pro- the deposit from which the oil is recovered. duction of middle distillates (such as diesel Oils from some sites may be relatively free of fuel and jet fuels) rather than naphtha, Shale such contaminants. oil and coal-derived syncrudes should, therefore, be regarded not as a competitive or sub- *Some convention:]] (Iru(ics h:)~[’ ({)mpilr[]t)l} h]gh pour stitutable feedstocks but rather as comple- points. For example, the Alt[~mon I crude from Ut:)h hi]s ;{ pour mentary feedstocks, with each yielding a dif- point Of about 100° F (35° (;).

Shale Oil Refining Shale oil has been successfully refined in of the economic aspects of shale oil utiliza- oil shale operations in Sweden, Scotland, tion, and justified continued efforts aimed at Australia, West Germany, the U. S. S. R., and its recovery. In recent years, refining R&D other countries, although on a relatively has been revived because refiners now con- small scale and under unusual economic consider the availability of shale oil to be a dis- ditions. In the United States, the initial refin- tinct possibility. There is also a need to per- ing research was conducted by USBM at the form more precise and up-to-date economic Petroleum and Oil Shale Experiment Station analyses. at Laramie, Wyo. It was coordinated with the early development of the gas combustion re- To date, refining studies have been con- tort at Anvil Points, Colo. The results of this ducted on the upgrading of crude shale oil to work, plus the findings of other investigators, a transportable product, and on the total re- allowed a preliminary assessment to be made fining of shale oil into finished fuels. The dif- 158 . An Assessment of Oil Shale Technologies

ference between these operations lies in the stocks. These studies differ in their approach nature of the desired final product. As dis- to the analysis of refining requirements. cussed previously, some crude shale oils have Studies of existing refineries must consider pour points and viscosities that make trans- the equipment that is in place, and must allow porting them difficult and expensive. In some for the limited flexibility of this equipment for situations, economic considerations may dic- processing a feedstock that is different from tate that the crude shale oil be partially re- the one for which the refinery was designed. fined (upgraded) near the retorting site to im- Studies of specially built refineries, in con- prove its transportation characteristics. In trast, need not be biased in this manner, and other instances, a developer may desire to ob- can draw upon any processing technique that tain a complete array of finished fuels from is available within the refining industry. an integrated processing facility located near However, both types of studies must make the minesite. In this case, a total-refining assumptions about feedstock characteristics facility would be considered rather than a and desired product mixes. These will vary more simple upgrading plant. with the location of the refinery, the nature of the market it serves, and the type of retorting To date, upgrading experiments have been facility that supplies its feedstocks. The op- carried out largely at the bench scale, and in relatively small pilot plants. Theoretical timal refining conditions for one set of assumptions will probably not be applicable to studies and computer modeling have also another set. For example, a refining method been used to evaluate the expected perform- to maximize gasoline production from TOSCO ance of three types of upgrading processes: II shale oil would not maximize diesel fuel thermal, catalytic, and additive. Thermal processes include visbreaking (a relatively production from Oxy’s in situ oil. mild treatment) and coking (a severe treat- Numerous computer studies and bench- ment). Mild thermal treatment will reduce scale refining investigations have been con- pour point and viscosity, but the oil will retain ducted for a wide range of shale oil feed- its initial amounts of nitrogen and sulfur. In stocks and operating conditions. The results contrast, severe thermal treatment reduces of these studies can be extrapolated, with pour point, viscosity, and sulfur content and some degree of caution, to predict the also causes the nitrogen compounds to con- performance of commercial-scale refineries. centrate in the heavier products. The proper- However, refining tests both in pilot plants ties of the lighter products will thus be con- and in commercial-scale facilities, because of siderably improved. much higher costs, have focused on only a In catalytic processes, the shale oil is re- few feedstocks, and have been conducted for acted with hydrogen in the presence of a cat- particular sets of operating constraints. In alyst. Viscosity is reduced, and the nitrogen general, each large-scale study has dealt only and sulfur are converted to ammonia and hy- with oil from aboveground retorts or with oil drogen sulfide gases that can be recovered as from in situ operations, but usually not with byproducts. In additive processes, blending both types of feedstocks. The conclusions of agents are added that reduce the pour point all studies are highly dependent on the com- and allow the crude to be transported by bination of feedstock and refining conditions pipeline. Such pour point depressants have assumed. Caution must be used when apply- been added in several instances with success, ing the results to different conditions. but the technique is not yet highly developed. Total refining studies have focused either Shale Oil Upgrading Processes on the needs of existing refineries that would have to be modified for processing shale oil, The treatment techniques that can be used or on those of newly built facilities that could to improve the transportation properties of be designed specifically for shale oil feed- crude shale oil are briefly described below. Ch 5–Technology ● 159

Visbreaking Catalytic Hydrogenation This technique involves heating the crude In these processes, the crude shale oil is shale oil to approximately 900° to 980° F reacted with hydrogen in the presence of a (480° to 525° C) and holding it at this tem- catalyst. The sulfur in the oil is converted to perature range for from several seconds to hydrogen sulfide, the nitrogen to ammonia, several minutes. The product is then cooled, the oxygen to water, the olefin hydrocarbons and the gases evolved during the heating are to their paraffin equivalents, * and long-chain removed. There is little reduction in the con- molecules to smaller molecules. The hydro- tents of nitrogen, sulfur, and oxygen. There- genation reactions can take place in a fixed- fore, the principal improvements are reduc- bed reactor through which a mixture of oil tions in pour point and viscosity. This tech- vapors and hydrogen is passed, in a fluidized- nique is simple but energy-intensive. It could bed reactor, or in an ebulliating-bed reactor. reduce the pour point of crude shale oil from The latter technique is being promoted by Hy- 30 about 850 F to about 400 F (300 to 40 C). drocarbon Research, Inc., as the H-Oil process. Plants in several foreign nations have Coking successfully used it for heavy petroleum feedstocks and for residuum fractions from a va- This process involves heating the oil to riety of crudes. In this process, a mixture is about 900° to 980° F (480° to 525° C) and injected into the bottom of a reactor at a high then charging it into a vessel in which ther- enough velocity to cause catalyst ebulliation mal decomposition occurs. If the vessel is a (a boiling motion). This movement reduces the coke drum, the process is called delayed cok- possibility that the bed will become plugged ing. The coke—the solid product from ther- by coke and by the liquid tars that are formed mal decomposition—is allowed to accumulate during the coking process. It also allows spent until it fills about two-thirds of the drum’s vol- catalyst and coke to be removed, and fresh ume. The feed is then switched to another catalyst to be added, so as to keep the bed ac- drum while the coke is cleaned out of the first tively stocked. one. Catalytic hydrogenation produces up- In the fluid coking process, hot oil is graded products of the highest quality, but it charged into a vessel that contains a fluidized is relatively expensive. The use of fixed-bed bed of coke particles. The particles become reactors would probably be confined to the coated with oil, which then decomposes to treatment of streams from an initial frac- yield gases and another layer of coke. The tionation step; fluid-bed or ebulliating-bed gases are withdrawn from the vessel. The processes could be used for either frac- coke is also withdrawn continuously, at a tionator products or for the whole shale oil. rate sufficient to maintain an active stock of coke within the bed. The flexicoking process, developed by EXXON, combines conventional fluid coking with gasification of the product coke. The ad- ‘Olefins are unsaturated (lower ratio of hydrogen to carbon) hydrocarbon compounds having at least one double bond. They vantage is that energy is recovered from the are the source of synthetic polymers such as polyethylene and coke. The process is used in the refining in- polypropylene used in the manufacture of fibers and other ma- dustry, but tests would have to be performed terials. Paraffins are saturated hydrocarbons (equal ratio of hydrogen to carbon) having only single bonds. Methane, to determine if it would be suitable for the ethane, propane, and butane are some of the paraffin hydro- coking characteristics of crude shale oil. carbons. 160 ● An Assessment of 0il Shale Technologies

Additives that have occurred in the proposed configurations of shale oil refineries since the 1950’s: Chemicals may also be added to crude the earlier studies placed much more empha- shale oil to improve its transportation proper- sis on gasoline production. For example, early ties. Pour point depressants have been suc- designs by USBM called for the extensive use cessful in some instances, but this does not of middle-distillate cracking and reforming to mean that they would always work. A chemi- yield gasoline, The ratio of gasoline to distil- cal that is suitable for one type of oil may not late yields was nearly 3 to 1.31 Most of the work at all with oil from another retorting refinery configurations that have been pro- process. Furthermore, pour point depres- posed more recently indicate a gasoline-to- sants are disadvantageous because they only distillate ratio of about 1 to 4.” change the physical characteristics of the oil and not its chemical properties. Thus, their The third factor—equipment and operat- cost can be offset only if they save transpor- ing constraints—has become increasingly im- tation costs. portant in recent years. The modifications to convert a conventional refinery to shale oil Conventional petroleum crudes are other feedstocks might not be economically justifi- potential blending agents. Since shale oil (at able unless the refiner could be assured of an least in Colorado) will be produced in an area adequate supply of shale oil. The economic that also contains petroleum reserves, and desirability of building a refinery specifically even active crude oilfields, the possibility ex- for shale oil would be thoroughly scrutinized. ists that the light petroleum crudes could be Modular retorts, or even a few pioneer com- mixed with crude shale oil to form a trans- mercial plants, would not produce enough portable blend. The feasibility of this concept shale oil in the mid-term to justify a new refin- is unclear because, in general, the blend ery unless the refiner was assured that the would not be as valuable as a refinery feed- operations would continue until his invest- stock on a per-unit basis as would the petro- ment cost could be recovered. For this rea- leum alone. However, the decrease in unit son, the most recent studies have stressed value would be offset by the increased vol- modifying existing facilities to make them ume. In the case of a refinery that does not suitable for processing shale oil, rather than have a reliable supply of crude, this could be building new ones. In some cases, this entails a significant advantage. only minor changes to installed equipment, in others, the adaptation of an existing facility Total Refining Processes by adding new units. The three primary factors that affect the The basic unit operations in crude oil refin- design of a refining system are: ing are: . the characteristics of the crude shale oil ● coking, feedstock; hydrotreating, ● the desired mix of finished products; and distillation, ● the constraints imposed by the equip- hydrocracking, ment and operating practices of the pro- ● catalytic cracking, and posed refinery. ● reforming. The first factor probably will have the lowest The various refining schemes that have been effect because, except for the higher nitrogen proposed for shale oil cannot easily be gener- and arsenic contents, the characteristics of alized because, depending on both the de- crude shale oil are not widely different from sired product mix and the possible operating those of conventional petroleum. The second conditions, many configurations could be de- factor—the product mix—is much more sig- signed that would achieve the same results. nificant. This is evidenced by the changes The one selected will largely depend on the Ch 5– Technology 161 availability of equipment and the individual Figure 47. —Refining Scheme Employing Coking economics of the particular refinery. Before an Initial Fractionation, Used by Chevron U.S.A. in Refining Experiments One significant difference among the various configurations described in the literature is the relative arrangement of distillation and thermal or catalytic treatment. Two general approaches have been investigated: 1. distillation of the whole crude into its components, then catalytic or thermal treatment; or I T“ 2. catalytic or thermal treatment of the RpJY whole crude, then distillation.

In the first approach the properties of the SOURCE G L Baughman The Ref(nfng and Markeflrtg of Crude Sha/e 0{/ finished products are better controlled. In the second, the net load on successive processing Figure 48.— Refining Scheme Employing units is reduced, and, in general, the overall Initial H ydrotreating, Used by Chevron U.S.A. yield of high-value hydrocarbon products is in Refining Experiments increased. Most of the refining research to date has been focused on the second ap- c, proach. Three versions are shown schemati- c, G (iog,aaunal cally in figures 46 through 48. The scheme in $ c 3 1S04WT w REFORMER d w~ ~ v I K 6 b Figure 46.— Refining Scheme Used by the cc 0 HYDROTREATER P- > i Fti U.S. Bureau of Mines to Maximize Gasoline 1 $ } Production From Shale Oil !5

Gas z 660”F< > I lt-d3 Ga90hm w Gasohw REFORM ER ~ cc $ 0. cmdO Shale “ft- cm 7. SOURCE G L Baughman. The Reflrrlng and A4arkeflr?g of Crude Sha/e 0//

-T”’ue’Fu@l Od able into finished fuels. The disadvantage of t- coking is that there might not be a ready market for coke in the vicinity of the refinery, particularly if it were located in the oil shale re-

SOURCE G L Baughman, The Refining and Markefing of Crude Sha/e 0// gion. Another refining scheme that was investi- figure 46, which was used by USBM at Anvil gated by Chevron is shown in figure 47. It 33 Points in the late 1940'S, had a gasoline-to- uses a fixed-bed catalytic hydrotreater to up- distillate ratio of 3 to 1. Figure 47, a con- grade the shale oil before distillation. No coke figuration that was investigated by Chevron is produced because most of the heavier com- U.S.A. during pilot-plant runs on Paraho ponents of the crude shale oil are upgraded shale oil,34 had a gasoline-to-distillate ratio of into lighter and more valuable liquid fuels 1 to 4. Both of these systems used an initial during the hydrogenation process. This meth- coking step to upgrade the feedstock and to od may be more costly than the coking ap- supply a product stream more easily refin- proach, but that can only be determined by 162 ● An Assessment of Oil Shale Technologies operating experience and evaluating the eco- pumps and explosions in processing units. nomics. Ash, or particulate matter, must be removed to prevent its deposition in pipes, heat ex- Chevron also investigated the possibility of changers, and catalyst beds. Recent studies substituting a fluidized catalytic cracker for have shown that heating the crude oil to the hydrocracker in figure 48. A similar ap- about 165° F (75° C) then letting it stand for proach was used by SOHIO in processing 85,000 bbl of Paraho shale oil at its Toledo about 6 hours allows the water and solid mat- refinery. 35 The chief difference between the ter to separate from the oil. As noted, arsenic and other metals poison the hydrotreating SOHIO runs and the Chevron experiments is catalysts. A variety of processes have been that SOHIO used an acid/clay treatment to upgrade the distillation products into jet fuel developed for their removal; consequently, and marine diesel fuel for military applica- their presence no longer presents a technical tions. The residuum fraction from the column problem. ARCO has patented several catalytic techniques and methods for heat treat- was used for fuel in the refinery. ing the oil in the presence of hydrogen. Re- The approach in which the crude oil is cent studies by Chevron U.S.A. have shown fractionated before hydrogenation or other that an alumina guard bed preceding the hy- treatment is shown in figure 49. This scheme drotreater will effectively remove both arsenic and iron from shale oil. 39 Figure 49.— Refining Scheme Employing Initial Fractionation, Used by SOHIO in Prerefining Studies The quantity and quality of the fuels produced will be determined by both the configuration of the equipment and the operating conditions used in the refining step. The fuels produced by USBM at Anvil Points in the 1940’s were quite satisfactory.’” However, some of the fuels from the Gary Western refining run in 1975 failed to meet certain military specifications, principally those for stability. 41 This has been attributed to the application of refining techniques unsuitable for shale oil feedstocks, specifically inadequate hydrogenation. Subsequent refining tests at SOHIO’s Toledo refinery show that, with appropriate refining, fuels can be produced that are of superior quality and that can meet all applicable specifications. 42

SOURCE. G L Baughman, The Refining and Market/rig of Crude Shale 0// Cost of Upgrading and Refining was used by SOHIO during the prerefining The most recent estimates of the cost of up- studies carried out before 10,000 bbl of Para- grading crude shale oil to a transportable re- ho shale oil were refined at the Gary Western 36 finery feedstock have been prepared by Chev- refinery in Fruita, Colo. During the actual ron U.S.A.43 The retorting complex that was refinery run, a combination coker/fraction- considered had a capacity of 100,000 bbl/d. ator was used rather than the separate units Conventional hydrotreating was the upgrad- shown in the diagram. ing technique evaluated. Chevron considered The shale oil must also be treated to re- two possible locations for the upgrading facil- move excess amounts of water, ash, and ity: a newly built unit at the retorting site; and heavy metals such as arsenic. Water must be a unit to be added to an existing refinery at removed because it can cause cavitation in some distance from the retorts. In both cases, Ch 5–Technology 163 the estimated cost for upgrading 1 bbl of was estimated to range from $8.00 to $10.00/ crude shale oil was $6.50, in first-quarter bbl of crude shale oil. For a 50,000-bbl/d 1978 dollars. The product would be a high- refinery located in a remote area of the Rocky quality syncrude suitable as feedstock for Mountains (e.g., near the retorting facility), most refineries in the United States. the refining cost would be approximately It is more difficult to estimate the costs of $10.00 to $12.00/bbl.* These are somewhat higher than the costs for refining a high-qual- the total-refining option for converting crude ity conventional crude oil because of the addi- shale oil into finished fuels. This is because in tional amounts of hydrogen that would be addition to the properties of the crude oil con- needed to reduce the nitrogen content of the sideration must be given to the location of po- shale oil crude. tential refinery sites, the availability of refining equipment, the proximity and stability of Another study compared the cost of shale potential markets, the ease of product distri- oil refining with those of refining Wyoming bution, and other factors. Chevron consid- sour crude oil and Alaskan crude. It was as- ered some of these in its analysis of refining sumed that a refinery in the Rocky Mountain costs, although in a relatively generalized region was modified for these feedstocks. The manner. Three total-refining options and two increased costs to refine crude shale oil, refining capacities and refinery locations rather than the other crudes, was in the were considered. For a l00,000-bbl/d refin- range of $0.25 to $2.00/bbl.44 45 ery located in an urban area in the Rocky Mountains (e.g., Denver), the refining cost *Costs are in first-quarter 1978 dollars.

Markets for Shale Oil Crude shale oil has three major potential Shale Oil as a Boiler Fuel uses: as a boiler fuel, as a refinery feedstock, and as a feedstock for producing petrochemicals, The output from a mature oil shale in- Shale oil will most likely first be used as a dustry will probably be used for all three pur- boiler fuel because of the relatively small poses. However, the relative importance of capital investments and very short leadtimes the three markets will change with time as that would be required. Because of Govern- the industry develops. In the mid-1980’s, ment regulation, the current trend in the utili- when shale oil first becomes available in sig- ty industry is to replace oil- and gas-fired nificant amounts, its most likely use will be as boilers with coal-fired units, thus freeing the boiler fuel, with only a small quantity di- natural gas for domestic consumers. In some rected to nearby refineries that could be mod- areas, it will be a two-stage transition, with ified to accommodate the feedstock without the gas first replaced by oil, and the oil later large capital expenditures. As more shale oil by coal. During the transition, there maybe a becomes available, its use as a refinery feed- market for about 50,000 to 80,000 bbl/d of stock will increase as conventional petroleum crude shale oil near the oil shale region. In becomes scarcer. At a later date, when the addition, the refining industry has a small but market for boiler fuels declines, shale oil will significant demand for boiler fuel because re- begin to be used for petrochemical produc- fineries are also changing from natural gas to tion. oil. Therefore, refineries located near the oil —

164 ● An Assessment of 011 Shale Technologies shale region are likely to be near-term shale ies have dealt with four general types of oil customers. In the long run, the largest refining facilities: market for shale oil boiler fuel is likely to be 1. in the Great Lakes States. Transportation dis- a new refinery just for shale oil; tances will probably preclude its use in the 2. a new refinery for a mix of shale oil and other two major markets for boiler fuels—the conventional crude; east and west coasts. 3. an existing refinery modified for, and dedicated to, shale oil; and 4. an existing refinery processing a mix of Shale Oil as a Refinery Feedstock shale oil and conventional crude. There has been relatively little research on The first two approaches are precluded for the refining of crude shale oil because build- the foreseeable future because, as long as re- ing an oil shale plant takes about 5 to 7 years, fined products continue to be imported, the whereas a refinery operator can evaluate a United States will have excess refining ca- new potential feedstock in a few days, devel- pacity. At least through the 1980’s, shale oil op a feasible refining strategy within a mat- will most likely be refined in existing refin- ter of weeks, conduct the necessary pilot- eries, either by itself or as a blend with con- plant refining studies in a few months, and ventional petroleum, modify the refinery to accommodate the new The shale oil produced by demonstration feedstock in less than 3 years. Thus, neither facilities will probably be processed in local the developer nor the refiner has any induce- refineries* or in more distant refineries ment to study shale oil refining until they are owned and operated by the energy companies sure that the oil will, in fact, be forthcoming. that participate in the oil shale programs. Another reason is that until quite recently, The much larger output from a commercial- shale oil was not considered a highly desir- ize industry will be more widely distributed; able refinery feedstock. During the 1950’s thus it will have to compete with other feed- and 1960’s, the refining industry tended to stocks, at least regionally. Recent studies maximize gasoline yields at the expense of have indicated that the Midwest is the most middle and heavy distillates. Because shale likely market area for large quantities of oil is a good source of heavier distillates, not shale oil. This includes the States in the Petro- of gasoline, it was not highly regarded. How- leum Administration for Defense District 2 ever, most projections indicate that gasoline (PADD 2), as shown in figure 50. There will demand will peak in the early 1980’s and then be secondary effects in other districts, as a decline slightly, even though total demand for consequence of the supplies of shale-derived refined products will continue to grow.46 This fuels in PADD 2, because the conventional pe- will be the result of the increasing efficien- troleum that it displaces will become avail- cies of gasoline engines in automobiles and of able for use elsewhere. a greater use of diesel engines in automobiles The quantitative impact on the supply and light trucks. Also, because the current situation in PADD 2 can be determined by re- supplies of conventional petroleum are be- fering to table 20, which indicates how the coming more like shale oil with respect to district’s supplies of finished fuels were di- their distillate yields, the refining industry is vided among domestic and foreign sources in being forced to adopt techniques that would 1978. As shown, the district consumed about be equally suitable for shale oil. 518,000 bbl/d of medium and heavy distillates For these reasons, shale oil’s desirability is (jet fuels, diesel fuel, and distillate fuel oil) increasing, and its potential availability as a from foreign sources. According to Chevron’s premium feedstock has encouraged the refin- *’I’hese could include the Gary Western refinery in Fruita, ing studies that have been conducted by Colo., the Little America refinery in Rawlins, Wyo., the Chev- Chevron and other organizations. These stud- ron refinery in Salt Lake City, Utah, ond others. Ch 5–Technology ● 165

Figure 50.— The Petroleum Administration for Defense Districts

Includes Alaska and Hawaii

SOURCE Natmna/ At/as, Department of the Intenor

Table 20.–Supply of Finished Petroleum Products in PADDs 2, 3, and 4 in 1978 (thousand bbl/d)a

PADD 2 Midwest PADD 3 Texas and New Mexico — PADD 4. Rocky Mountains Foreign Domestic Foreign Domestic Foreign Domestic sources sources Total sources sources Total sources sources Total Gasoline. 976 1,540 2,516 422 604 1,026 33 219 252 Light ends 224 236 460 276 388 664 12 25 37 Jet fuel 81 131 212 50 71 121 7 29 36 Kerosine 17 27 44 19 27 46 0 2 2 Distillate fuel oil 420 672 1,092 194 279 473 13 113 126 Residual fuel 011 150 173 323 237 340 577 4 38 42 Petrochemical feedstocks. 18 26 44 211 298 509 0 1 1 Special naphthas and still gas 60 98 158 114 165 279 2 14 16 Others 138 232 370 102 146 248 4 31 35 Total 2,084 3,135 5,219 1,625 2,318 3.943 75 472 547 ap,ADO = pe/rOleum ,4dmlfllSlfa10fl tor Defense Dlstrlcl

SOURCE G L Baughman The Refmmg and Markef~ng of Crude Sha/e 01/ studies, refining will convert about 74 per- tricts. The same size industry would produce cent of a crude shale oil feedstock to similar about 170,000 bbl/d of gasoline, which would distillates.” A l-million-bbl/d industry would be equivalent to about 17 percent of the dis- yield about 740,000 bbl/d of medium and trict’s gasoline currently obtained from for- heavy distillates. If it were marketed in PADD eign sources. 2, this production would completely displace the foreign supplies and free an additional An alternative marketing strategy would 222,000 bbl/d of the fuels for use in other dis- be to sell the output from a major shale oil in- 166 ● An Assessment of 011 Shale Technologies dustry in PADD 4—the Rocky Mountain re- foreign crude. Recent marketing studies have gion—and to supply any surplus fuels to adja- identified several large refineries in this area cent districts such as PADD 2 or PADD 3 that, with only minor modifications would be (Texas and New Mexico). As shown in table able to handle crude shale oil. 48 Furthermore, 20, the Rocky Mountain district consumes rel- some of the refineries only have access to the atively little distillate fuel (about 20,000 heavier petroleum crudes at present, and bbl/d) from foreign sources. A l-million-bbl/d their production is being limited by their ca- shale oil industry could easily displace this pacity to process the large quantities of resid- entire supply. The surplus production (about uum from the distillation of these fuels. Shale 720,000 bbl/d) could displace about 95 per- oil, with its relatively small yield of bottoms cent of the supply of foreign-derived distil- fractions, would help alleviate this problem. lates in both PADD 2 and PADD 3. The gasoline derived from the shale oil could supply Shale Oil as a Petrochemical Feedstock about 67 percent of the total gasoline demand in the Rocky Mountain States. Three principal factors must be considered As indicated previously, the capabilities of in evaluating the suitability of shale oil sup- the refineries in the Midwest and the Rocky plies for producing petrochemicals: Mountain States will strongly affect the will- ● the yields of petrochemicals from shale ingness of the refiners to accept shale oil oil feedstocks; feedstocks. In some cases, shale oil could not the ability of existing and future petro- be accommodated without significant invest- chemical plants to process the shale oil; ments of capital. However, the receptivity of and refiners to shale oil will also be influenced by ● the logistics of supplying the shale oil to the reliability of other feedstocks such as for- the plants. eign petroleum. The area in which the supplies of crude are most uncertain is the north- Because shale oil is produced by pyrolysis, its ern tier of States, which includes Montana, olefin content is approximately 12 percent, North and South Dakota, Minnesota, and which is appreciably higher than convention- Wisconsin. These States have historically de- al crudes. Together with its fairly high hydro- pended on refinery feedstocks from Canada, gen content, these characteristics make shale but, in recent years, a significant reduction in oil, and its hydrogenated derivatives, appro- these supplies has led the refiners in this priate feedstocks for petrochemical produc- area to look elsewhere. The result has been tion.” 50 Steam pyrolysis has been used to the present interest in building a pipeline to process crude shale oil, and the yields of transport crude from Alaska and from for- olefin products have been comparable with eign nations into the area. An alternative—a those from many conventional crudes. Shale pipeline from the oil shale region—could also oil syncrudes, with even higher olefin yields, be built as the oil shale industry developed. are considered to be premium petrochemical feedstocks. These conclusions are based on The area that covers Iowa, Missouri, Illi- laboratory studies under carefully controlled nois, Indiana, Michigan, and Ohio also does conditions. The feasibility of marketing shale not have an adequate indigenous supply of oil to the petrochemical industry depends on crude. However, unlike the northern tier, the ability to replicate these conditions in these States have good pipeline systems with commercial chemical plants. adequate access to both foreign and domestic crude supplies. The feasibility of marketing Historically, the primary feedstock for pet- shale oil in this area will largely be deter- rochemical plants has been natural gas liq- mined by the cost differential between it and uids from the gulf coast. Because crude shale other crude supplies and by the differences in oil is quite different from these liquids, it the reliability of its supply versus that of would be difficult to switch traditionally de- Ch 5–Technology ● 167 signed petrochemical plants to shale oil. How- United States, the petrochemical industry is ever, the production of domestic natural gas concentrated on the gulf coast. This concen- and its associated liquids is declining, and the tration will continue into the foreseeable fu- petrochemical industry is shifting to heavier ture. Therefore, it will be necessary to either feedstocks such as naphtha and gas oils. The move the shale oil to the coast, or to build supplies of these feedstocks are uncertain a new petrochemical complex in the Rocky and irregular. The availability of naphtha has Mountain region. In the latter case, the half- been affected by a growing demand for its finished products from the new plant would use as a gasoline blending agent in response still have to be transported to the coast for to the phasing out of tetraethyl lead. Gas oils final conversion to commercial chemicals. are also being used more frequently for home The former approach is more likely but is im- heating fuels, which causes seasonal varia- peded by the lack of a product pipeline sys- tions in their availability. Because of these tem between the oil shale region and the gulf supply uncertainties, new petrochemical coast, and by the high cost of alternative plants are being designed to be highly flexible modes of transportation. with respect to feedstocks. As using heavier feedstocks becomes more common in the in- In summary, tests have shown that crude dustry, shale oil may become a highly re- shale oil and its derivatives could be used to garded raw material. produce petrochemicals. However, these materials cannot be considered to be viable feed- The use of shale oil for petrochemical pro- stocks in the near future because existing duction is hampered by the distance between chemical plants are generally unable to proc- the oil shale region and the petrochemical ess them and there is ‘no economical transpor- plants. While refineries and oil-fired boilers tation link between the oil shale region and are distributed fairly uniformly across the the existing petrochemical plants.

Issues and Uncertainties The technological readiness of the major Germany.5z In these operations, the lignite is mining and processing alternatives is summa- covered by 900 ft of overburden, and strip- rized in table 21. Estimated degrees of read- ping operations will soon extend to 1,600 ft. iness are shown as judged by DOI in 1968,51 This is comparable to the oil shale deposits, and as they appear under present conditions. which are covered by a maximum of about There are significant differences between the 1,800 ft of overburden. two evaluations because much R&D work has been conducted in the interim, and because Nevertheless, uncertainties remain with two new processing methods—MIS retorting respect to the effects of shale stability and and concurrent recovery of associated min- strength on mine design, mine safety, and re- erals—have since entered the picture. As source recovery. The effects of large inflows shown, room-and-pillar, open pit, and MIS of ground water, such as have been encoun- mining methods are regarded as reasonably tered on tract C-a, could pose severe opera- well-understood. Open pit mining has not tional difficulties, especially with under- been tested with Green River shales, but it is ground mines. In all mines, the logistics asso- highly developed for other minerals such as ciated with moving many thousands of tons of copper and iron ores. It was evaluated on raw material and solid wastes could present paper for application to the shales on tract some formidable problems. Materials-han- C-a. Some highly relevent experience has dling systems exist that could be applied, but been obtained from the operation of large- they have yet to be tested in commercial oil scale lignite (a form of coal) mines in West shale operations. —. .

168 An Assessment of Oil Shale Technologies

Table 21 .–Technological Readiness of Oil Shale Mining, Retorting, and Refining Technologies

Technological readiness Unit operation In 1968 In 1979 Developments in the Interim Remaining areas of uncertainty Mining Room and pillar Medium Medium Mine design studies. Experience of Colony, Rock mechanics Ground water Deep shales Mobil, and the six-company program at Anvil Logistics. Points MIS Unknown Medium Oxy experience at Logan Wash Mine design Rock mechanics Ground water Deep shales studies. Logistics Open pit Low Medium Established procedure for other minerals Rock mechanics Logistics Reclamation Large-scale mines in West Germany. Mine design studies for tract C-a, Other Low Low Mine design studies No field experience All areas Retorting Aboveground Medium Medium Pilot studies for Lurgi-Ruhrgas Semiworks Effects of scaleup on stream factor and recov- programs by Colony and Paraho Detailed ery. Characteristics of emissions streams designs by Colony and Union. Reliability of peripheral equipment Materials handling. TIS Low Low Field work and lab tests by USBM. Field tests Stream factor and recovery. Characteristics of by Equity, Geoknetics, and Talley-Frac. emissions streams Rock mechanics Deep National lab support shales. MIS Unknown Medium Oxy experience, Design studies for tracts C-a Effects of scaleup on stream factor and recov- and C-b. DOE and USBM design studies ery Characteristics of emissions streams Modeling and lab support by national Reliability of peripheral equipment Use of laboratories. low-Btu gas Rich shales. Mixed shales, Fracturing Rock mechanics Deep shales, Multimineral Unknown Medium Pilot plant studies by Superior USBM lab Effects of scaleup on stream factor and recov- tests of product recovery methods. Nahcolite ery Characteristics of emissions streams scrubbing tests. Reliability of peripheral equipment, Materials handling Integration of recovery steps Underground waste disposal Marketability of byproduct minerals. Upgrading and refining...... High High Studies and refinery runs by SOHIO/Paraho, Effects of retorting conditions on crude Chevron, and others. Marketing analyses characteristics Cost effectiveness of alter- by Oxy and DOE. nate processes Use of pour point depressants Effects of metals

SOURCE Ofhce of Technology Assessment

AGR is regarded as having a medium level Although understanding has increased of readiness, as it was in 1968. The under- since 1968, TIS must still be regarded as standing of its technical aspects has been im- being in the conceptual stage. Many uncer- proved since then by field tests in the Pice- tainties remain, especially with respect to ance basin, but the largest tests conducted to economics and environmental effects. date have been at the semiworks scale— about one-tenth of commercial size. Their re- MIS retorting, a new concept since 1968, sults do not permit accurate cost projections has advanced to a medium level of technologi- for commercial-scale plants. Particular prob- cal readiness, approaching that of above- lems are noted with respect to the effect of -ground retorts. This progress is largely a re- scaling up the semiworks design to commer- sult of Oxy’s development efforts in Colorado, cial size. The on-stream factor—the fraction but additional understanding has been ob- of the time that the retorts could be expected tained through simulations by USBM, DOE, to operate at design capacity—is unknown. and the national laboratories, most notably The reliability of some associated systems the Lawrence Livermore and Los Alamos (emissions controls, product recovery de- Laboratories. The remaining uncertainties vices, materials-handling equipment) is also are similar to those for aboveground retorts, questionable. except the materials-handling problems may Ch 5–Technology ● 169 be less substantial with MIS methods. The conventional petroleum, and the fact that all major uncertainties relate to the use of the of the components of a facility must work as large quantities of low-Btu retort gas for pow- designed—not just the well-established ones. er generation and to the application of the With the present state of technological technique to very rich or very deep shales. knowledge, it is not clear that an oil shale The multimineral concept was also largely plant would perform as desired, nor that the unknown in 1968, although the presence of oil would be sufficiently cheap to compete sodium minerals was recognized. At present, with its currently designated competitor—im- the aboveground system of Superior Oil is reported petroleum. Even though the future of garded as having medium technological read- oil shale looks brighter, few companies are iness. Its uncertainties are much the same as willing to build large-scale plants immediate- for other AGR methods, with the addition of ly. They prefer to follow conventional engi- the potential difficulties with integrating the neering practice by proceeding to an inter- systems for recovering mineral byproducts mediate step—the so-called modular demon- with those for recovering oil and gas. The stration retort. marketability of the nahcolite, soda ash, and The modular retort is the smallest unit that alumina has also not been well-established. would be used in commercial practice, A com- The MIS concept proposed by Multi Mineral mercial-size oil shale facility would use sev- Corp. is not specifically evaluated in the eral of these retorts in parallel to obtain the table. It would share the same uncertainties desired production rate. The capacity of a as conventional MIS, with additional po- module varies with the developer. For Oxy, a tential problems introduced by the need to modular MIS retort might have a capacity of evacuate a large underground retort, and be- only a few hundred barrels per day of shale cause the oil shale resource is deeply buried. oil; a commercial facility would have several Upgrading and refining systems were given dozen of these retorts operating simultane- a high rating in 1968, which has been im- ously in modular clusters, each producing proved by additional study. No major prob- several thousand barrels per day. A modular lems are anticipated, although more needs to Lurgi-Ruhrgas retort would have a capacity be known about the feasibility of using pour of about 2,200 bbl/d. One or two such units point depressants, the effects of retorting might suffice for the mined shale from an MIS conditions on the characteristics of the crude operation; a facility that used only AGR might oil, and the effects of metals (such as arsenic) need a dozen comparable units. A commer- on refining catalysts. The major uncertainty cial-sized plant that used Paraho retorting in the distribution area—the existence of an might have seven or eight individual retorts, adequate pipeline system to the most likely each producing 7,300 bbl/d. A comparable markets—is not directly related to the nature plant might use seven Union “B” retorts, each of the refining technology and is therefore not producing about 9,000 bbl/d. Colony prefers indicated. to bypass the modular demonstration phase and to proceed directly to a commercial-size It should be noted that mining and retorting facility, claiming that the TOSCO 11 technol- (the major subprocesses having only medium ogy is ready to be scaled up to such a capacity levels of technological advancement) require for demonstration purposes. only about 35 percent of the capital investment that is needed to establish an AGR com- Regardless of these differences, the sev- plex. The remaining 65 percent is distributed eral “next steps” that are proposed by the among upgrading units, byproduct recovery developer’s have two points in common: the systems, utilities, sidewalks, and other well- production units must be large enough to sim- established items. Yet developers hesitate to ulate actual commercial practice, and the commit themselves to oil shale plants. The equipment must be operated long enough to reasons given are an uncertain regulatory obtain reliable data on its performance under climate, an uncertain future for the price of a complete spectrum of operating conditions. 170 . An Assessment of Oil Shale Technologies

R&D Needs and Present Programs R&D Needs TIS Retorting Mining TIS is the most primitive of the processing methods. It has some potentially valuable fea- Room-and-pillar mining and mining in sup- tures but these cannot be evaluated because port of MIS operations have been tested with of a lack of information. The potential im- oil shale, and an extensive body of informa- pacts on surface characteristics and ground tion has been assembled. To date, however, water quality are especially unclear. The all of the field tests have been conducted in R&D needs include: one area—the southern fringe of the Pice- ● ance basin. In each case, the deposit was development of less expensive drilling techniques; reached through an outcrop along a stream ● course. The limited area in which mining development of efficient and cost-effec- tive fracturing and rubbling techniques; tests have been conducted is unfortunate be- ● cause the characteristics of deposits in other development of methods for determin- areas are quite different. They may be more ing the success of a rubbling program or less favorable to mine development. For through, for example, surveys of the per- example, there is little ground water in the meability increase that has been achieved; southern fringe. In contrast, the deposits on ● tracts C-a and C-b, nearer the basin’s center, development of ignition methods and of lie within ground water aquifers, and the in- methods for maintaining a uniform burn front; flow of water into the mine shafts has been a ● problem on tract C-a. Similar problems were study of the effects of heat-carrier com- encountered at the USBM shaft in the north- position, rate of injection, and temperature on product recovery; ern part of the basin. Work on this site was ● also impeded by the presence of highly frac- study of the effects of creep* on retort stability and product recovery; tured zones, which are not common on the ● southern fringe. determination of the effects of ground water infiltration on retorting; and Similar surprises could be avoided in fu- ● evaluation of the long-term potential for ture projects if the suitability of the candi- surface subsidence. date mining techniques for developing deposits throughout the oil shale region were better Valuable information is being obtained in the understood, especially in these areas where Equity and Geokinetics projects, but these re- near-term development is likely. This infor- sults are applicable only to specific types of mation could be obtained through coring and oil shale deposits—the Equity process to rock mechanics studies, mathematical simu- shale in the Leached Zone; the Geokinetics lation, and experimental mining. Field testing method to thin, shallow beds. Additional field of mining methods would be expensive, work in other types of deposits would aid in although overall costs could be minimized evaluating the potential of TIS methods for through developing a single site that could be large-scale production. To minimize the cost used to test many mining alternatives. A and duration of these tests, they could be sup- plemented with initial theoretical studies and single shaft or adit, for example, could be laboratory programs. used to test room and pillar, longwall, block caving, and other methods. It should be noted that open pit mining, because of the necessity for costly and large-scale operations, would probably not be amenable to testing in a lim- *Creep is the gradual change in the shape of a solid object in- ited field program. duced by prolonged exposure to stress or high temperatures. Ch 5–Technology ● 171

MIS Retorting Some of these needs could be addressed by further laboratory-scale and semiworks test- MIS is more highly developed, but more ing. Others could be estimated by theoretical testing is needed before its potential applica- calculations and modeling. All of them and es- tion to other areas of the Green River forma- pecially the need for reliability studies, will tion can be determined. The R&D needs are eventually have to be addressed in full-scale similar to those for TIS. They are: retorting modules, either alone or as part of a ● the development of better rubbling tech- commercial-size complex. niques; ● the development of improved remote Upgrading, Refining, and Distribution sensing procedures for permeability, fluid flow, and temperature; Because crude shale oil is sufficiently ● the development of methods for creating similar to conventional petroleum crude, no and maintaining a better burn front; substantial problems are anticipated in the ● evaluation of the effects of heat-carrier refining area. R&D on the effects of heavy characteristics; metals on refining catalysts and of retorting ● evaluation of the effects of creep and conditions on oil properties could be con- subsidence; and ducted in the laboratory, provided that re- ● examination of the effects of ground torts were operating that could supply a prod- water infiltration, retort geometry, and uct resembling the crude oil that will be pro- particle-size distribution. duced in commercial operations, Many of these needs will be addressed in the In the upgrading area, the major need is re- MIS programs on tracts C-a and C-b and the lated to the feasibility of using chemical addi- USBM shaft. tives to depress the pour point of crude shale oil. The necessary R&D could be conducted in Aboveground Retorting the laboratory or in a pilot refinery, again assuming the availability of a representative Several candidate retorting processes crude shale oil. have been tested in Colorado, but only for relatively short periods, and in small-scale R&D is also needed to determine the opti- facilities, More R&D is needed, and particu- mum distribution pattern for the finished lar emphasis should be given to: fuels, which will vary with the size of the industry, the location of the facilities, the need evaluating the effects of scaleup on the for various fuels and feedstocks, and the flow patterns of solids and fluids within availability of a transportation system. R&D the retort vessel; is needed to determine optimal plans for like- determining the reliability and effec- ly combinations of these factors. Work has tiveness of peripheral equipment such begun in this area, and more work could be as solids-handling systems, pollution conducted at relatively low cost, since it is controls, and product separators; theoretical rather than experimental in examining the effects of heat-carrier nature. However, it will not be possible to characteristics on product recovery and define an optimum pattern for the actual equipment reliability; and future industry until the sites of the produc- determining the reliability of mechanical tion facilities are designated. components such as Union’s rock pump; Tosco’s retort vessel, separation trom- The System mel, and ball elevator; the Lurgi-Ruhrgas screw conveyor; and the raw shale dis- All manufacturing and processing plants tributors and spent shale discharge potentially suffer from a lack of systems reli- grates of all retorts that use gas as a ability. Because of the scale of operations and heat carrier. the need for the coordinated performance of .—. .—.—— .

172 An Assessment of Oil Shale Technologies

many components, it is certainly possible that nologies in the various types of oil shale oil shale plants will have significant prob- deposits. lems. On the other hand, they may not be more severe than in other, more conventional In situ processing has been given the major industries. R&D programs, such as mathe- emphasis throughout the program, and much matical simulations and industrial engineer- of the technical R&D will be conducted in the ing studies, would help to eliminate some of “Moon Shot” project that will address the the uncertainties regarding the expected per- second goal. Initial support of AGR will focus formance and reliability of oil shale systems. on designing the retort module and on surface Basic data on the lifetimes of equipment, and underground mines to support single operating characteristics, and other factors plants and an industry of 1 million to 3 million could be obtained from those minerals proc- bbl/d. The decision to proceed with construc- essing and refining plants that oil shale facil- tion of AGR modules will be determined by ities will resemble. However, because of the the economic outlook for shale oil in unique character of oil shale operations, pre- mid-1980. DOE will consider a cost-shared dictions from these studies will be tentative. program if industry has not announced firm It will only be possible to define performance plans to proceed without Federal participa- characteristics after large-scale oil shale tion. The program will also include resource plants are operated at their maximum pro- characterization studies that will help to duction capacities. delineate the portions of the oil shale basins where the different types of development Present Programs technologies would be most applicable. Other studies will include assessments of air, Some of the current R&D programs for water, land, and socioeconomic impacts; of individual retorting technologies were de- occupational safety and health; and of meth- scribed previously. An effort that has not ods for increasing the efficiency of water use. been discussed in detail is DOE’s integrated research, development, and demonstration These efforts should substantially advance program for oil shale.53 Its major objective is the understanding of the technological as- to provide the private sector with the techni- pects of oil shale development. The budget of cal, economic, and environmental informa- over $387 million for fiscal years 1980 tion needed to proceed with the construction through 1984 should be adequate to address of pioneer commercial plants. Its specific most of the R&D needs identified in the previ- goals are: ous section. This budget includes about $126 million for developing and operating a com- ● by mid-1981: to provide technical de- mercial-scale MIS retort, and half of the esti- signs, cost data, and environmental in- mated $200 million cost of an AGR module. 54 formation for construction and operation of at least one AGR module; The demonstration facilities are especially ● by 1982: to design at least one commer- important to the acquisition of firm engineer- cial-size MIS retort that could be used on ing and economic data. Unfortunately, only the Federal lease tracts or in other loca- one in situ technology and one aboveground tions; and retort will be tested, and it will be difficult to ● by 1985 to 1990: to remove the remaining evaluate fully the effects of resource charac- technical uncertainties that impede com- teristics on the feasibility of alternate mining mercial-scale use of the alternate tech- methods. Ch 5–Technology ● 173

Policy Options R&D The two general approaches to funding such demonstration programs are discussed Some of the remaining technical uncertain- below. Selecting an option will depend on the ties could be alleviated with additional small- desired balance between information genera- scale R&D programs. These could be con- tion and dissemination, Federal involvement, ducted by Government agencies or by the pri- timing of development, and cost. vate sector, with or without Federal participation. If full or partial Federal control is de- Private Funding sired, the programs could be implemented through the congressional budgetary process If left alone, the industry would develop in by adjusting the appropriations for DOE and response to normal market pressures and op- other executive branch agencies, by provid- portunities, and the Government’s expense ing additional appropriations earmarked for and involvement would be minimized. How- oil shale R&D, or by passing new legislation ever, the Government would not be assured of specifically for R&D on oil shale technologies. access to the technical, economic, and environmental information that it needs to formu- Demonstration late future policies and programs, although some of this information could be obtained Full-scale demonstrations will be needed to through third-party reviewers or through li- accurately determine the performance, reli- censing arrangements. Another disadvantage ability, and costs of development systems un- is that industry may not risk even the relative- der commercial operating conditions. In gen- ly modest investment of a modular program eral, potential developers would prefer to fol- until economic and regulatory conditions low conventional engineering practice and clearly favor development. For example, approach commercialization through a se- Union and Colony have announced that they quence of increasingly larger production will not proceed until Federal incentives are units. Union, Colony, and Paraho have pro- provided and regulatory impediments re- gressed through this sequence to the semi- moved. Industry may eventually proceed, but works scale of operation—about one-tenth of perhaps not in time for the resource to con- commercial size. tribute substantially to the Nation’s fuel supplies within the next decade. If this conservative approach were continued, the next step would be a modular Government Support demonstration facility. Although such a plant would cost several hundred million dollars, it The alternatives are full Government fund- would provide the experience and the techni- ing of demonstration facilities, indirect fund- cal and economic data needed to decide on ing through incentives to industry, and a the commitment of much larger sums to com- sharing of the costs with industry. The op- mercial-scale operation. Union has expressed tions are discussed in detail in chapter 6 and its preference for this path; Rio Blanco and summarized in chapter 3. In brief, Federal Cathedral Bluffs are following it. Colony re- ownership would provide the Government gards a pioneer commercial plant as the best with the maximum amount of experience and facility for proving the TOSCO II technology. information. It would also maximize Govern- 174 ● An Assessment of Oil Shale Technologies

ment intervention and the commitment of A single module on a single site.—This op- public funds, and it might discourage private tion would provide comprehensive informa- developers from proceeding with independ- tion about one process on one site. Either ent demonstrations. Also, industry and Gov- underground or surface mining experiments ernment would design, finance, and operate a could be performed, but probably not both. demonstration project in very dissimilar The costs would be small overall but large on ways. The Federal experience with a Govern- a per-barrel basis, because there would be no ment-owned facility may have little relevance economies of scale. Some of the shale mined to the problems that would be encountered by could be wasted because the single retort a private developer with the same production might not be able to process all of it economi- goal, The Government’s experience would cally. If the site could subsequently be devel- therefore provide little guidance for evaluat- oped for commercial production (e. g., a pri- ing oil shale as a private investment oppor- vate tract, a potential lease tract, or a can- tunity. didate for land exchange), the facility would Incentives programs could involve tax have substantial resale value. Otherwise, it credits, purchase agreements, price sup- would be valuable only as scrap. ports, or other types of support, either singly Several modules on a single site.—This or in combination. They could be structured program might consist of an MIS operation, to encourage the participation of specific coupled with a Union retort for the coarse types of firms and could be combined with portion of the mined oil shale and a TOSCO II regulatory changes, and possibly land ex- for the fines. As with the single-module op- changes or additional leasing, to control both tion, either surface or underground mining the growth and the nature of the ultimate could be tested, depending on the site, or commercial industry. They would cost less possibly both if the plant had a sufficiently than Government ownership. They would also large production capacity. The total costs tend to provide the Government with less in- would be larger than for the single-module formation and with no operating experience. program, but unit costs would be lower. For However, disclosure requirements could be example, a three-module demonstration plant inserted into the leases or the incentives legis- would cost about 2. I times as much as a lation as a prerequisite for project eligibility y. single-module facility; a six-module plant The cost sharing of demonstration facilities about 3.7 times as much. Different technol- would entail intermediate expenditures of ogies could be combined to maximize re- public funds and intermediate levels of infor- source utilization, and detailed information mation. The receptivity of industry to such could be obtained for each. However, all of proposals would depend on how much the the information would be applicable to only Government would intervene in designing and one site. If many modules were tested, the operating the projects. If industry responded, demonstration project would be equivalent to the Federal investment that would be needed a pioneer commercial plant, except that a for a single Government-owned plant could be true pioneer operation would probably not spread over several projects, thereby in- use such a wide variety of technologies. creasing the total amount of information generated. Single modules on several sites.—Several technologies might be demonstrated, each at Program Alternatives a separate location. For example, an underground mine could be combined with a Demonstration will require designing, TOSCO II retort on one site; a surface mine building, and operating full-size production with a Paraho retort at another. Total costs units, either as separate modules or incorpo- could be large, as would unit costs, which rated in pioneer plants. would be comparable with those of the single- Ch. 5–Technology 175 module/single-site option. The principal ad- methods would be selected that would be vantage would be that different site charac- most appropriate for the characteristics of teristics, processing technologies, and mining the site and the nature of its oil shale de- methods could be studied in one comprehen- posits. The maximum amount of information sive program. would thus be acquired, in exchange for the maximum amount of investment. Each project would resemble the several-module/single- Several modules on several sites.—For site option; the collection would constitute a each site, a mix of mining and processing pioneer commercial-scale industry.

Chapter 5 References

‘T. Ertl, “Mining Colorado Oil Shale, ” Quarter- ‘2P. W. Marshall, “Colony Development Opera- ly of the Colorado School of Mines, vol. 69, No. 3, tion Room-and-Pillar Oil Shale Mining, ” Quarterly July 1965, pp. 83-92. of the Colorado School of Mines, vol. 69, No. 2, ZRio Blanco oil Shale Project, ~etailed ~eVe~oP- April 1974, pp. 171-184, ment Plan: Tract C-a, Gulf Oil Corp. and Standard “C-b Shale Oil Project, Detailed Development Oil Co. (Indiana), March 1976, pp. 7-2-1 to 7-2-20. Plan and Relatea’ Materials, Ashland Oil, Inc., and ‘J. B. Miller, President of Rio Blanco Oil Shale Shell Oil Co,, February 1979, p. IV-2. Co., letter to G. Martin, Assistant Secretary for “Oil Shale R, D, & D—Program Management Land and Water Resources, Department of the In- Plan, Department of Energy, Washington, D. C., terior, Oct. 26, 1979. June 25, 1979, p. C-3. ‘U.S. Energy Outlook—An I~terim Report, The 151bid. National Petroleum Council, Washington, D, C., ‘bIbid., p. C-2. 1972. “Ibid. ‘H. E. Carver, “Oil Shale Mining: A New Possi- 18R. D. Ridley, “Progress in Occidental’s Shale bility for Mechanization, ” Quarterly of the Col- Oil Activities,” Eleventh Oil Shale Symposium Pro- orado School of Mines, vol. 60, No. 3, July 1965. ceedings, Colorado School of Mines Press, Golden, “J. J. Reed, “Rock Mechanics Applied to Oil Colo., November 1978, pp. 169-175. Shale Mining,’” Quarterly of the Colorado School “D. R. Williamson, “Oil Shales: Part 5—Oil of Mines, vol. 61, No. 3, July 1966. Shale Retorts, ” Mineral Industries Bulletin, vol. 8, ‘R. O. Bredthauer and T. N, Williamson, No. 2, March 1965. “Mechanization Potential in Oil Shale Mining, ” ‘(’J. L, Ash, et al., Technical and Economic Study Quarterly of the Colorado School oj Mines, vol. 61, of the Modified In Situ Process for Oil Shale, re- INO. 3., July 1966. port prepared for the U.S. Bureau of Mines by Fe- 8P. T. Allsman, “A Simultaneous Caving and nix and Scission, Inc., under contract No. Surface Restoration System for Oil Shale Min- SO241O73, November 1976. ing, ” Quarterly of the Colorado School of Mines, “J. H, Campbell, et al., “Dynamics of Oil Gen- vol. 63, No. 4, October 1968. eration and Degradation During Retorting of Oil ‘W. N. Hoskins, et al., “A Technical and Eco- Shale Blocks and Powders, ” Tenth Oil Shale Sym- nomic Study of Candidate Underground Mining posium Proceedings, Colorado School of Mines Systems for Deep, Thick Oil Shale Deposits, ” Press, Golden, Colo., July 1977, pp. 148-165, Quarterly of the Colorado School of Mines, vol. 71, 22R. L. Braun and R. C. Y. Chin, “Computer Mod- No. 4, October 1976, pp. 199-234. el for In-Situ Oil Shale Retorting: Effects of Gas In- ‘“C. E. Banks and B. C. Franciscotti, “Resource troduced Into the Retort, ” Tenth Oil Shale Sym- Appraisal and Preliminary Planning for Surface posium Proceedings, Colorado School of Mines Mining of Oil Shale, Piceance Creek Basin, Col- Press, Golden, Colo., July 1977, pp. 166-179. orado,” Quarterly of the Colorado School of 2tJ. H. Raley, et al., “Results From Simulated, Mines, vol. 71, No. 4, October 1976, pp. 257-286. Modified In Situ Retorting in Pilot Retorts, ” IIJ. H. East and E, D. Gardner, “Oil Shale Min- Eleventh Oil Shale Symposium Proceedings, Col- ing, Rifle, Colorado, 1944 -1956,” Bulletin 611, U.S. orado School of Mines Press, Golden, Colo., No- Bureau of Mines, 1964. vember 1978, pp. 331-342. —

176 ● An Assessment of Od Shale Technologies

“G. L. Baughman, Synthetic Fuels Data I-Iand- ings, The Colorado School of Mines Press, Golden, book, 2nd Edition, Cameron Engineers, Inc., Den- Colo,, November 1978, pp. 120-134. ver, Colo., 1976, ‘%upra No. 28, 25R. D, Ridley, “A Marketing Perspective for ‘%upra No. 29, Shale Oil,” paper presented at the meeting of the ‘OH. C, Carpenter, et al., “A Method for Refining Commercial Development Association, Marcos Shale Oil,’* Industrial and Engineering Chemistry, Island, Fla,, 1979. vol. 48, pp. 1139-1145. zbR. F, Sullivan, et al., “Refining Shale Oil, ” “Applied Systems Corp., Compilation of Test Re- paper presented at the 43rd Midyear Meeting of sults, report prepared for the Office of Naval Re- the Refining Department of the American Petro- search under contract No. NOO014-76-C-0427, leum Institute, Toronto, Ontario, May 10, 1978. 1976, ‘71bid. ‘2 Supra No. 35, 2W.S. patent Nos. 3,804,750: 3,876,533; and ‘]H, A. Frumkin, et al., “Cost Comparisons—Al- 4,029,571. ternative Refining Routes for Paraho Shale Oil, ” 2YR. F. Sullivan, et al., Refining and Upgrading of paper presented at the 86th National Meeting of Syn~uels From Coal and Oil Shale by Advanced the American Institute of Chemical Engineers, Catalytic Processes, report prepared for the De- Houston, Tex., Apr. 1-5, 1979. partment of Energy by Chevron U.S.A. under con- %upra No. 25. tract No. EF-76-C-01-2315, 1978. “’Pace Company Consultants and Engineers, 1{)W. L. Nelson, “Visbreaking as Pour Point Re- Inc., Shale Oil Refining Analysis, report prepared ducer,” The Oil and Gas journal, Nov. 10, 1969, p. by the Pace Co. for the Department of Energy and 229. Occidental Oil Shale, Inc., 1978, “J. D, Lankford and C. F. Ellis, “Shale Oil Refin- %upra No. 25. 7 ing, ” Industrial and Engineering Chemistry, vol. ‘ Supra No, 43. 43, No. 1, January 1951, pp. 27-32. ‘“Purvin and Gertz, Inc., Markets for Crude %upra No. 29. Shale Oil in Central U. S., report prepared for Oc- ‘] Supra No. 31. cidental Oil Shale, Inc., 1977. ‘%upra No. 29. “yIbid, “E. T. Robinson, “Refining of Paraho Shale Oil 5’]C. F. Griswold, et al., “Light Olefins From Hy- Into Military Specification Fuels, ” Twelfth Oil drogenated Shale Oil,” Chemical Engineering Shale Symposium Proceedings, The Colorado Progress, September 1979, pp. 78-80, School of Mines Press, Golden, Colo., August “Prospects for Oil Shale Development—Colora- 1979, pp. 195-212. do, Utah, and Wyoming, Department of the Interi- ‘“H. Bartick, et al., The Production and Refining or, Washington, D, C., May 1968, p, 47. of Crude Shale Oil Into Military Fuels, report pre- 52M. A. Weiss, et al., ShaJe Oil: Potential Econo- pared for the Office of Naval Research by Applied mies of Large Scale Production, Workshop Phase, Systems Corp. under contract No. NOOO14-75- The Massachusetts Institute of Technology, Ener- C-0055, 1975. gy Laboratory report No. MIT-EL-79-031-WP, July “R. F, Sullivan and B. E. Strangeland, “Convert- 1979, pp. 6-7. ing Green River Shale Oil to Transportable ‘] Supra No. 14, at pp. 1-6. Fuels, ” Eleventh Oil Shale Symposium Proceed- 5%upra No. 14, at pp. viii-ix.