Dongwon Kim � University of Oklahoma,100 E. Boyd St., Norman, Oklahoma 73019-0628;present address: STX Energy Co. Ltd., 631Namdaemunno 5-ga, Jung-gu, Seoul, SouthKorea
D. W. Kim obtained a B.Sc. degree in geologyin 1988 and an M.Sc. degree in paleontologyin 1991 both from Seoul National University. Heworked as an exploration geologist for KoreanNational Oil Company for 10 years. He also re-ceived his M.Sc. degree and Ph.D. in petroleumgeochemistry in 1999 and 2006, respectively,
Geochemical characterizationof solid bitumen in theChesterian (Mississippian)sandstone reservoir of the Hitchfield, southwest KansasDongwon Kim, R. Paul Philp, andRaymond P. Sorenson
from the University of Oklahoma. He joined STXEnergy in 2007, where he is currently a teamleader of new venture business development,focusing on Indonesia, Colombia, and westAfrica. His interests include petroleum systemsand basin modeling, and their application tothe sedimentary basins of the regions.
R. Paul Philp � University of Oklahoma,100 E. Boyd St., Norman, Oklahoma 73019-0628;[email protected]
R. Paul Philp obtained a B.Sc. degree in chemistryfrom the University of Aberdeen in 1968, aPh.D. from the University of Sydney in 1972, anda D.Sc. degree from the University of Sydneyin 1998. He is currently a professor of petroleumand environmental geochemistry at the Univer-sity of Oklahoma. Current interests include theapplication of organic geochemical techniquesto the characterization of unconventional gasshales, characterization of novel biomarkers andhigh molecular weight hydrocarbons in crudeoils and source rocks, and the use of stableisotopes in environmental problems.
Raymond P. Sorenson � Cimarex EnergyCo., 15 East 5th St., Tulsa, Oklahoma74103-4346
Ray Sorenson received his B.S. degree in geologyfrom Michigan State University in 1972 and hisM.A. degree in geology from the University ofTexas at Austin in 1975. He has worked for
ABSTRACT
Solid bitumen was identified within theMississippian Chestersandstone reservoir in the Hitch field, southwest Kansas. Theadjacent Etzold field has similar reservoir properties but lackssolid bitumen, although the Hitch and Etzold fields werethought to be in pressure communication and have commonsource rocks.Hitch and Etzold crude oils and core extractswerecharacterized geochemically to gain a better understanding ofthe reservoir-filling history and geological and geochemical con-trols on solid bitumen formation in the Hitch reservoir. Basedon a variety of geochemical characteristics, we propose thattheHitch andEtzold oils aremixtures derived fromOrdovicianand Late Devonian–early Mississippian (Woodford Shale)source rocks. The solid bitumen in the Hitch reservoir prob-ably results from mixing of oils having different geochemicalcompositions, which filled the reservoir over an extended pe-riod. No evidence of severe biodegradation or thermal alter-ation is observed. A reservoir-filling scenario is proposed inan effort to explain why the Hitch oils are geochemically moreheterogeneous than Etzold oils. Furthermore, gas de-asphaltingand regional pressure drops as a result of post-Laramide orogenymay have contributed to a phase change in the reservoir fluidand deposition of solid materials by disturbance of thermody-namic equilibrium.
The major significance of this study is related to improv-ing our knowledge on the occurrence of the presence of solid
Texaco (1974–1975), Anadarko Petroleum (1976–2006), and Cimarex Energy (2006–2009), withmost of his career focusing on the geology of theUnited States mid-continent.
Manuscript received June 3, 2009; provisional acceptance July 28, 2009; revised manuscript receivedOctober 2, 2009; final acceptance December 9, 2009.DOI:10.1306/12090909096
AAPG Bulletin, v. 94, no. 7 (July 2010), pp. 1031– 1057 1031
ACKNOWLEDGEMENTS
The AAPG Editor thanks the following reviewersfor their work on this article: Brian J. Cardott,Harry Dembicki Jr., and Kenneth E. Peters.
1032 Hitch Field Bitumens
bitumen in reservoirs that can introduce barriers into the reser-voir, complicating waterflood operations and leading to erro-neous oil-in-place calculations and subsequently lower thanexpected oil recovery.
INTRODUCTION
Solid bitumen is an immovable and highly viscous materialunder reservoir conditions, and originated from the alter-ation of once-liquid petroleum through physical or chemicalchanges. Solid bitumens are commonly identified by a sharplydefined compositional zone in oil reservoirs and are believed tobe related to geological discontinuities (e.g., oil-water contact,unconformity, or shale barrier). The occurrence of solid bitu-men can also be recognized by organic geochemical charac-terization (Larter et al., 1990) because solid bitumen forma-tion is commonly accompanied by compositional changes(e.g., viscosity increases with depth, API gravity decreases to-ward the base of an oil column, and the gas-oil ratio decreaseswith depth).
The occurrence of solid bitumen in petroleum reservoirs iscommon in many petroliferous basins worldwide (Jones andSpeers, 1976;Wilhelms and Larter, 1995; Horstad and Larter,1997;Huc et al., 2000).However, solid bitumen in theAnadarkoshelf or Hugoton embayment has never been previously re-ported. In the Anadarko Basin, bitumen occurrences haveonly been reported along the structurally deformed southernrim of the basin, including the Amarillo uplift, Wichita Moun-tains, or Arbuckle Mountains (Curiale and Harrison, 1981).These areas are 200mi (322 km) or more from theHitch field,and the solid bitumens are unlike theHitch solid bitumen. Theeconomic impact of such bitumen can be extremely significantin a mature area such as the Anadarko Basin. The tar mats de-scribed herein remained hidden for two decades despite nu-merous cores and modern wireline suites, and therefore, addi-tional solid bitumens could possibly be present in other fieldswhere the database is of lower quality. Solid bitumen can in-troduce barriers into the reservoir, complicating waterfloodoperations and leading to erroneous oil-in-place calculationsand lower-than-expected oil recovery (Sorenson et al., 1999).
Several natural processes leading to the deposition of solidbitumen in reservoirs have been proposed in the literature.Three principal processes for tar-mat formation or asphalteneprecipitation most frequently cited in the literature are ther-mal alteration, de-asphalting, and biodegradation. Thermal
alteration (maturation) of preexisting liquid hydro-carbons to form hydrogen-depleted carbonaceousresidues and associated gases can result in deposi-tion of solid bitumen in the carrier bed or reservoir.De-asphalting occurs when a large volume of gasdissolves in crude oil and causes asphaltenes to pre-cipitate. The process may be driven by increasingthe depth of burial to the point that crude-oil crack-ing occurs and produces gas, or by a gas charge in-troduced into the reservoir from another source(Hunt, 1996). Biodegradation, combined withwater washing, gives rise to the formation of solidbitumen as a result of selective and successive re-moval of specific groups of compounds (Huc et al.,2000). Several other formation mechanisms orcombinations have also been discussed in the liter-ature, including low reservoir temperatures (Bolyard,1995), in-reservoir oil mixing (Leythaeuser andRückheim, 1989; Larter et al., 1990), and pressurereduction during reservoir inversion (Hirschberget al., 1984). Solid bitumen can form during pro-duction operations, such as water flooding or CO2
injection during enhanced oil recovery operations(Hwang andOrtiz, 1998). Therefore, detailed geo-logical and geochemical characterization is requiredto define the processes responsible for depositionof solid bitumen in the Hitch reservoir, whichshould in turn lead to the ability to more accuratelypredict additional occurrences of solid bitumen insimilar situations.
REGIONAL GEOLOGY ANDPETROLEUM SYSTEMS
The Hitch and Etzold fields are part of the Shuckfield located in a low-relief northern extension of theAnadarko Basin in the Kansas shelf area (Figure 1).The Shuck field in Kansas covers most of the town-ship T33S, R34W and adjacent areas in T32S,R34W;T33S, R34W; andT33S, R35W. Shuck fieldis the name generally given for any pre-Permian pro-duction (Hugoton and Panoma fields), and theHitch, Etzold North, and Etzold South are water-flood units located within the Shuck field. A gen-eralized stratigraphic column for Pennsylvanianand Mississippian strata, southwest Kansas, from
Montgomery and Morrison (1999), is shown inFigure 2. The Anadarko Basin was a broad shelflikeepicontinental sea that received a thick sequenceof carbonate sediments, interbedded with lesseramounts of clastic sediments from Late Cambrianthrough Mississippian (Johnson, 1989). A north–south-trending paleovalley system developed inthe shallow-marine environment during theMissis-sippian. The basalMississippian Chester sandstone,incorporating the Hitch and Etzold reservoirs, wasdeposited in an incised valley eroded into the under-lyingMississippian limestones of the Ste. Genevieveand St. Louis formations (Severy, 1975). TheHitchsandstone was deposited in a channelized tidal in-let type, whereas the Etzold sandstone was depos-ited in a wider, thinner, sheet of sandwaves, overthe same period. However, the Etzold sandstonewas probably partly separated from theHitch sand-stone by a seaward facies change to a shallow sub-tidal marine mudstone facies (Montgomery andMorrison, 1999; Cirilo, 2002). At the time of dis-covery, the Hitch and Etzold fields were assumedto be in pressure communication. However, con-nectivity between the two sandstone reservoirswas recently suggested to be limited, and the flowof fluid between them was very restricted (L. L.Cirilo, 2003, personal communication).
The Pennsylvanian orogenic episode, includ-ing the Wichita-Amarillo uplift and the rapid sub-sidence of the paleo-Anadarko Basin, resulted inthe formation of a deep asymmetric foreland basin.As a result of the Pennsylvanian orogeny, the Pa-leozoic sedimentary strata dipped steeply towardthe south in the deep Anadarko Basin and had agentle slope in the Kansas shelf area. Petroleumsystem studies suggest that most of the Mississip-pian reservoirs, including Chester sandstones, inthe Hugoton embayment contained oil that origi-nated from the Late Devonian–early MississippianWoodford Shale. The Chester Valley was impor-tant as a focusedmigration pathway, and hydrocar-bon accumulation in the Hugoton embayment isbelieved to be the result of long-distancemigrationfrom multiple source rocks in the deep AnadarkoBasin because potential source rocks in the shelfarea have not entered the oil generation window(Burruss and Hatch, 1989).
Kim et al. 1033
EXPERIMENTAL: SAMPLES AND METHODS
The geochemical and organic petrographic tech-niques applied in this study are outlined in the flowchart shown in Figure 3. A preliminary study usingwireline log and core data was conducted to inves-tigate the areal and stratigraphic distributions ofsolid bitumen in the Hitch field and to assist theselection of samples and types of experimentalmethods. Three types of samples were collectedand analyzed in this study: crude oils, conventionalreservoir core plugs, and potential source rocks.The sampling sites are located at the northern partof the Anadarko Basin in the Oklahoma panhan-
1034 Hitch Field Bitumens
dle, northwestern Oklahoma, and southwestKansas (Figure 1; Table 1). The cores had beenstored in a warehouse without treatment for manyyears and were taken to the laboratory as received.The crude oils were collected directly from eitherpressure-volume-temperature samples, tank bat-tery, or after the reservoir water was separatedfrom the oil by mechanical methods. Each oilsample was stored in a clean glass bottle with aTeflon-sealed cap at room temperature (∼20°C[68°F]).
The remaining and specific details concerningextraction, fractionation, and analytical techniquesare summarized in the Appendix.
Figure 1. Location of the Hitch and Etzold fields and surrounding geologic provinces (left) and sampling localities in a major incisedvalley system in the isopach map of the net basal Chester sandstone, greater than 8% porosity, contour interval 20 ft (6.1 m) (right). Thenorth–south AB cross section on the right map is shown in Figure 11. The Hitch and Etzold fields are part of the Shuck field that coversmost of the township T33S, R34W and adjacent areas in T32S, R34W; T33S, R34W; and T33S, R35W. The Shuck field is the name generallygiven for any pre-Permian production (Hugoton and Panoma fields) and the Hitch, Etzold North, and Etzold South are waterflood unitslocated within the Shuck field.
GEOCHEMICAL CHARACTERISTICS OF THEHITCH AND ETZOLD SAMPLES
Compositional Variation within theReservoir Column
Vertical variations in chemical composition are evi-dent in the Hitch sandstone reservoir. The totalpyrolysates (S1 + S2) determined by Pyran level Ianalysis tend to exhibit maximum values withinthe intervals above the thin cemented or shalylow-permeability layers (Figure 4). (S1 and S2 fromthe Pyran system are basically equivalent to S1 andS2 from Rock-Eval analyses. S1 is the thermal distil-lation component and S2 corresponds to the kero-gen degradation products). In particular, the S2peak maximizes at the base of reservoir sand unitsand increases at several zones within the upper oil-producing column from 6151 to 6237 ft (1875 to1901 m) in the Hitch 8-3 well. This observationmay indicate that several small accumulations ofsolid bitumen occur within the reservoir sandstone
intervals. A few layers with extremely low concen-trations of organic material, for example, at depthsof 6225.8 and 6236.1 ft (1897.6 and 1900.7 m),appear to coincide with calcite cement-rich and lowneutron-density porosity zones. These zones wereidentified by x-ray diffraction (XRD) and wirelinelog character (Figures 5, 6). Although it remainsto be confirmed whether the nearly impermeablezones extend throughout the entire Hitch field,the zones acted as vertical barriers to fluid flowprior to the formation of solid bitumen as a resultof deposition of mud-rich sediment and/or calcitecementation during diagenesis.
Based on the relative size of the S1 and S2 peaks,the oil columns in theHitch reservoir are chemicallymore variable than those in the Etzold reservoir,containing local solid bitumen columns. We specu-late that the large S2 peak in the samples from thesolid bitumen columns was due to abundant heavyhydrocarbons and/or asphaltenes, which is supportedby the coincidence of the S2 peak size and asphaltenecontent of the respective core extracts (Figure 7).
Figure 2. Generalized stratigraphiccolumn for Pennsylvanian and Mississip-pian strata, southwest Kansas, fromMontgomery and Morrison (1999), re-printed with permission from AAPG.
Kim et al. 1035
Bulk Parameters of Oils and Rock Extracts
The bulk chemical properties of the Hitch andEtzold crude oils and core extracts are quite similar,consisting on average of 86.1% saturate, 9.8% aro-matic, and4.1%nitrogen, sulfur, and oxygen (NSO)compounds (Table 2). The relative amounts ofwax components in the produced oils are similarto those of the extractable wax hydrocarbons fromthe corresponding reservoirs in the Hitch andEtzold fields. The produced oils and reservoir coreshave an average wax content of 63.9 mg/g oil and0.36 mg/g rock, respectively, accounting for about7–8% of wax fraction in the total C15+ hydrocar-bons for both oils and core extracts. Althoughthe produced oil is regarded as light oil (40° APIand viscosity of 0.75 cp), the total wax content isrelatively high. The compositional similarity mayrepresent similar source rocks for oil in the Hitchand Etzold fields.
TheHitch oils have relatively higher asphaltenecontent, ranging from 8.7 to 12.6%, twice as high
1036 Hitch Field Bitumens
as that for the Etzold oil (5.6%). Although totalwax content is similar, it may be misjudged by co-precipitation of microcrystalline waxes with theasphaltene fraction (Thanh et al., 1999; Hsiehand Philp, 2000). The iso-octane extract recoveredfrom the asphaltene fraction contains wax compo-nents (>C30) with abundant n-alkanes, similar incomposition tomicrocrystalline waxes and heaviercompounds isolated from the maltene fraction(Figure 8). Subsequent separation of the extract-able wax hydrocarbons into macrocrystalline (C20–
C40) and microcrystalline (>C40) waxes revealsthat the Hitch samples contain three times moremicrocrystalline waxes than the Etzold samples.For example, the Hitch 8-3 oil has a microcrystal-line wax content of 0.3% of the total C15+ hydro-carbons (3.5% of total waxes), whereas the micro-crystalline wax content of the Etzold N2-1 oil isonly 0.1% of the total C15+ hydrocarbons (1.2%of total waxes). Such a difference may be relatedto enhanced wax deposition associated with as-phaltenes in the Hitch reservoir.
Figure 3. Flow chart for analytical procedures used in this study. Pyran FID is a temperature-programmed pyrolysis source rock char-acterization technique; FID = flame ionization detector; NSOs = the polar fraction comprising heteroatomic compounds containingnitrogen, sulfur, and oxygen atoms.
Source of the Hitch and Etzold Oils
The Hitch and Etzold oils and saturate fractionsare characterized by abundant n-alkanes in thelow-molecular-weight region (<C25) with an oddcarbon number predominance, and relatively low
abundance of acyclic isoprenoids (Fowler et al.,1986; Reed et al., 1986). The pristane/phytane(Pr/Ph) ratios are close to 1.5, and the carbon pref-erence index (CPI, in the range of n-C15–n-C21)values are slightly greater than 1. The Hitch andEtzold crude oils have n-C17/Pr and n-C18/Ph ratios
Table 1. Well Locations and Information on Core and Crude Oil Samples Used in this Study
Type
Well Name (Sample ID) State County STR* Geological Age or Formation
Core
Hitch Unit G-8 Kansas Seward 3-33S–34W Mississippian Hitch Unit I-2 Kansas Seward 10-33S–34W Mississippian Clark C 1 Kansas Seward 14-33S–34W Mississippian Hitch Unit 8-3 Kansas Seward 14-33S–34W Mississippian Etzold B 1 Kansas Seward 22-33S–34W Mississippian Etzold Unit 4-3 Kansas Seward 27-33S–34W Mississippian
Crude oil
Eubank North 3-4 (1) and (2) Kansas Haskell 21-29S–34W Mississippian Cavner A 5A Kansas Stevens 33-31S–38W St. Louis Limestone Hitch Cattle 1 Kansas Stevens 18-32S–35W Lower Morrow Hitch Unit 1-11A Kansas Seward 3-33S–34W Mississippian Hitch Unit 4-2 Kansas Seward 10-33S–34W Mississippian Hitch Unit 3-4 Kansas Seward 11-33S–34W Mississippian Hitch Unit 8-3 Kansas Seward 14-33S–34W Mississippian (W-INJ) Hitch Unit 8-2 Kansas Seward 14-33S–34W Mississippian Hitch Unit 9-5 Kansas Seward 15-33S–34W Mississippian Etzold N2-1 Kansas Seward 22-33S–34W Mississippian USA L-1 Kansas Morton 4-33S–40W Upper Morrow USA AA-1 Kansas Morton 11-33S–40W Upper Morrow Hatcher B Kansas Seward 35-34S–34W N/A City of Liberal C1 Kansas Seward 36-34S–34W St. Louis Limestone Interstate 90 Kansas Morton 21-34S–43W Upper Morrow Downs No. A-1 Kansas Seward 11-35S–34W N/A Boles F Kansas Seward 2-35S–35W N/A Dorman A 1 (OK-28) Oklahoma Beaver 23-5N–21ECM Hodges E.L. Addington 3 (OK-41) Oklahoma Texas 36-5N–10ECM Morrow Brown L-1 and L-4 Oklahoma Beaver 21-4N–21ECM Mississippian Smith Trush 1AE Oklahoma Beaver 8-4N–21ECM Chester Ratzlaff 2 (OK-36) Oklahoma Beaver 8-3N–21ECM Chester Joachim 1 (OK-23) Oklahoma Woods 14-25N–13W Red Fork + Mississippian Harmon 2 (OK-16) Oklahoma Major 34-23N–16W Hunton Inman J 1 (OK-17) Oklahoma Major 31-23N–14W Hunton Wyman B 1-28 (OK-33) Oklahoma Woodward 28-21N–18W Chester Gail Moore 1 (OK-10) Oklahoma Dewey 32-19N–18W Cottage Grove
*STR = section, township, range. ECM = east of the Cimarron Meridian.
Kim et al. 1037
of approximately 5 (Table 3). As can be seen inFigure 9, the Ordovician oil is characterized bythe marked odd/even (CPI) predominance ofn-alkanes, particularly in theC15–C20 range, and alsoby the relatively high n-C17/Pr ratio. The UpperDevonian–lower Mississippian Woodford sample
1038 Hitch Field Bitumens
has both a lower n-C17/Pr ratio and CPI, and thesefeatures alone can be used to differentiate the Or-dovician and Devonian oils along with the signifi-cant differences in their asphaltene contents. Themore traditional biomarkers are not significantlydifferent for these Ordovician and Devonian oils
Figure 4. Vertical variations in the amount of hydrocarbons throughout the entire cored section of the Hitch 8-3 well, indicating severalstratigraphic positions of possible low-permeability zones above which asphaltenes appeared to precipitate. The analyzed sandstone iswithin the lower Chester Valley fill as depicted in Figure 2, and the interval is highlighted in Figure 11.
and are not as diagnostic as in other cases. No ver-tical variations in the biomarker distributions wereobserved between solid bitumen and oil leg col-umns, suggesting the solid bitumen is, in all prob-ability, related to the oil (Figure 10; Table 4). Nosignificant differences in the biomarker distributionsof terpanes and steranes in theHitch and Etzold oilswere observed.
Although in general the Hitch and Etzold oilshave Ordovician biomarker signatures, multiplecharging episodes in the Hitch and Etzold fieldsand numerous fields within southwest Kansas andthe Oklahoma Panhandle have partially obscuredthese signatures (Burruss andHatch, 1989).Oils gen-erated from Paleozoic source rocks in the AnadarkoBasin can be classified into several types, each show-ing distinct geochemical characteristics that cor-relate with corresponding source rocks through-out the Anadarko Basin (Burruss and Hatch, 1989;
Wang and Philp, 1997). However, oils from theHitch field appear to be distinct with little similar-ity to crude oils derived from the known sourcerocks in the Anadarko Basin (Figure 9), possiblydue to mixing of input frommultiple source rocks.Nevertheless, the Hitch and Etzold oils show somegeochemical features (CPI, n-C17/Pr, Pr/Ph ratios,asphaltene content) in common with the Ordovi-cian (City of Liberal C1) and Devonian (OK-23Joachim 1) oils, but they probably have little or nocontribution from waxy crude oils derived fromPennsylvanian source rocks. In all probability, theHitch oils appear to be not uniquely sourced fromOrdovician source rocks but instead from a mix-ture of Ordovician, Devonian, and possibly othersource rocks in the region. Ordovician-age oils havewell-defined geochemical characteristics and carbonisotope compositions (Reed et al., 1986; Jacobsonet al., 1988; Fowler, 1992) and are not widely
Figure 5. X-ray diffraction traces of cores from four different intervals. Peaks were qualitatively identified by reference minerals. Q, C, B,and K = quartz, calcite, barite, and kaolinite.
Kim et al. 1039
recognized in the western Anadarko Basin buthave been documented from reservoirs near Hitchin Seward County, Kansas, and have been geo-chemically correlated to source rock in the corefrom the Viola Formation to the south in BeaverCounty, Oklahoma (Beserra, 2008). The MiddleOrdovician Viola shale is composed of up to 90%Gleocapsomorpha prisca (Fowler et al., 2004).The Upper Devonian–lower Mississippian Wood-ford Shale, an organic-rich black shale depositedin a euxinic shallow open-marine environment, is
1040 Hitch Field Bitumens
one of the major sources for hydrocarbons in theAnadarko Basin and a primary source for Hitch oils.The most characteristic feature of the Woodfordoil is a high concentration of asphaltenes and polarcompounds, averaging 23% in the 17.3–26.4% rangefor n-C5 asphaltene and 17% for wax-free asphal-tene. The average asphaltene contents for otherPaleozoic oils in this study are in the range of 0.5–4.5% for n-C5 asphaltene (1.7% for Ordovicianoils). The asphaltene contents in the Hitch core ex-tracts are in the range of 14–25%. The relatively
Figure 6. Composite logof basal Chester sandstonein the Hitch 8-3 well. Theinterval of solid bitumen ischaracterized by low corepermeability and porosityand low amplitude and fastdecay time in the NMR T2distribution (NMR = nu-clear magnetic resonance;T2 = transverse relax-ation time; from Sorensonet al.,1999, used with per-mission of AAPG; seeCoates, 1999). The analyzedsandstone is within thelower Chester Valley fill asdepicted in Figure 2, andthe interval is highlightedin Figure 11. GR = gammaray; SP = spontaneouspotential.
high concentration of asphaltenes in the Hitch oilsimplies that the Ordovician-type oil in the reservoirmixed with asphaltene-rich oil derived from theWoodford Shale to produce a mixed geochemicalsignature. In this particular study, the traditionalbiomarkers, such as steranes and terpanes, are oflimited use in distinguishing Ordovician and De-vonian oils. However, bulk parameters such as as-phaltene content, Pr/Ph, n-C17/Pr, and CPI haveproven to be more useful in distinguishing theseoil types.
Thermal Maturity
Thermal alteration of crude oils in the reservoir isone of themajor contributing factors to the forma-tion of solid bitumen (Rogers et al., 1974; Larteret al., 1990).Maturity assessments were necessary
to clarify whether or not the formation of the solidbitumen in the Hitch reservoir was related to thethermal degradation of the oil. The source rocks andcrude oils in the Anadarko Basin exhibit a widerange of thermal maturity based on vitrinite reflec-tance (Ro) and burial history modeling (Schmoker,1986; Cardott, 1989; Carter et al., 1998). TheHitchand Etzold oils show biomarker ratios (Table 5)corresponding to an early stage of the oil genera-tion window equivalent to Ro values in the regionof 0.6–0.8% (Peters and Moldowan, 1993). In-reservoir oil cracking has been reported to be amechanism responsible for tar mat formation whenthe reservoir temperature reaches 145°C (293°F)or higher (Milner et al., 1977; Huc et al., 2000).The Hitch reservoir had a temperature of 63°C(145.4°F) when it was discovered. Even prior to ero-sion of 2000–5000 ft (610–1524m) of overburden
Figure 7. Plot of the size of the S2 peakand the asphaltene content of the Hitch 8-3core samples, showing a direct relationship.
Table 2. Chemical Parameters by Percentage of Asphaltene (Asp) vs. Maltene (Mal) Fraction, Saturate (Sat), Aromatic (Aro), and Polar
during the early Tertiary, the temperature did notexceed 100°C (212°F) at the time of maximumburial based on the geothermal gradient in theAnadarko Basin, which has been in the range of22–25°C (71.6–77°F)/km (7–8°C [44.6–46.4°F]/1000 ft) since the late Paleozoic (Lee and Deming,1999). Therefore, the Hitch and Etzold crude oilsoriginated from source rocks in the early oil gen-eration zone (Ro = 0.6 to 0.8%), with no evidencefor in-reservoir thermal alteration.
Biodegradation and Water Washing
Thecomplete suiteofn-alkanes in the low-molecular-weight region suggests little or no biodegradationin theHitch and Etzold reservoirs. However, when
1042 Hitch Field Bitumens
oil originates from multiple source rocks or whenmultiple episodes of charging occur, biodegradationmay be difficult to detect (Rooney et al., 1998).Demethylated hopanes (25-norhopanes) are com-monly present in severely biodegraded oils althoughthe origin of the compounds is still somewhat con-troversial (Peters et al., 2005). The 25-norhopanesare generally considered to be the product of heavybiodegradation (Rullkötter and Wendisch, 1982;Peters et al., 1996), but they can also be concen-trated during biodegradation (Noble et al., 1985;Blanc and Connan, 1992). None of the branchedor cyclic fractions from selected Hitch or Etzoldsamples contain 25-norhopanes. Their absence mayindicate the absence of severe biodegradation, butnote that not all severely biodegraded oils contain
Figure 8. High-temperature gas chromatography chromatograms of (a) wax fractions and (b) maltene fractions precipitated withasphaltenes from the Hitch and Etzold oils. Wax fraction includes macrocrystalline (C20–C40) and microcrystalline (>C40) waxes. Thehydrocarbons recovered with iso-octane by the alumina adsorption and Soxhlet extraction from C5 asphaltenes isolated by the classicaln-pentane techniques are similar in geochemical composition to microcrystalline waxes.
25-norhopanes (Peters et al., 2005). Additionally,Holba et al. (2004) suggested that selective deple-tion of long-chain alkyl aromatic hydrocarbon fam-ilies in oil indicates early-stage anaerobic micro-bial degradation. Long-chain alkyl aromatics (e.g.,alkylbenzenes and alkyltoluenes) are present inhigh concentrations in the Hitch and Etzold oilsand core extracts, but no obvious depletionwas ob-served in core extracts stained heavily with solid bi-tumen. Other biodegradation characteristics in-clude uncommonly low API, high sulfur content,or high viscosity. TheAPI gravities of theHitch oilsare relatively high, in the 38–40° range, with a vis-
cosity of 0.75 cp, and the oils are extremely low insulfur with little unresolved complex mixture evi-dent from the gas chromatography (GC) (Figure 9).Therefore, the Hitch and Etzold oil columns donot appear to have experienced any significantbiodegradation.
DEPOSITION OF SOLID BITUMEN
Based on similar biomarker distributions, we pro-pose that the solid bitumen in the Hitch reser-voir was derived from precipitation of asphaltenes
Table 3. Biomarker Ratios of Saturate Fraction from Crude Oils Used for the Source Identification and Correlation Purpose
Region
Sample ID (Well Name) Pr/Ph* n-C17/Pr
K
n-C18/Ph
im et al.
CPI**
Kansas
Eubank North Unit 3-4(1) 1.68 3.37 2.98 1.10 Eubank North Unit 3-4(2) 1.82 3.16 3.09 1.11 Cavner A 5A 1.42 4.60 4.69 1.09 Hitch Cattle 1 1.53 3.73 4.01 1.09 Hitch and Etzold oils Hitch Unit 1-11A 1.60 4.93 5.29 1.11
Hitch Unit 4-2
1.53 5.21 5.32 1.13 Hitch Unit 3-4 1.69 5.36 6.08 1.09 Hitch Unit 3-4 (N) 1.58 4.84 5.23 1.10 Hitch Unit 8-3 1.57 5.37 5.61 1.12 Hitch Unit 8-2 1.43 5.61 5.35 1.13 Hitch Unit 9-5 1.56 4.39 4.55 1.12 Etzold N2-1 1.48 5.33 5.18 1.13
USA L-1
2.65 2.71 6.98 0.95 USA AA-1 2.05 3.49 6.79 0.98 Hatcher B 1.66 4.44 5.09 1.09 City of Liberal C1 1.24 13.10 9.52 1.17 ISU 90 2.31 3.12 6.75 0.98 Downs No. A-1 1.54 5.49 3.91 1.10 Boles F 1.72 4.64 3.71 1.16
Oklahoma
Dorman A 1 (OK-28) 1.18 0.96 1.03 1.01 E.L. Addington 3 (OK-41) 2.15 3.62 7.29 1.00 Brown L 1 1.51 5.45 4.15 1.12 Brown L 4 1.55 5.42 3.86 1.16 Smith Trush 1AE 1.35 5.20 5.06 1.07 Ratzlaff 2 (OK-36) 1.35 1.84 2.17 1.02 Joachim 1 (OK-23) 1.33 2.19 2.37 1.00 Harmon 2 (OK-16) 1.37 4.03 5.66 0.97 Inman J 1 (OK-17) 1.41 2.72 3.65 0.99 Wyman B 1-28 (OK-33) 1.43 2.00 2.80 1.00 Gail Moore 1 (OK-10) 1.35 2.00 2.38 1.00
originating from the overlying oil column. Amongthe possiblemechanisms described in previous sec-tions, biodegradation and thermal alteration do notappear to be responsible for the formation of thesolid bitumen. Gas de-asphalting and in-reservoiroil mixing combined with a subnormal pressureconditions in the reservoir were more probably re-sponsible for precipitation of the solid materials inthe high-porosity and high-permeability sandstonereservoir.
Gas De-asphalting
Gas injection into a reservoir that contains oil andis undersaturated with respect to gas can cause as-phaltene precipitation (Wilson et al., 1936;Milneret al., 1977).Gas injection results in compositionalchanges, reducing the overall solvent capacity of
1044 Hitch Field Bitumens
the oil, which in turn decreases asphaltene solubil-ity (Wilhelms and Larter, 1994). Evidence sup-porting gas injection as a cause for the formationof solid bitumen in theHitch reservoir can be sum-marized as follows: (1) the Hitch reservoir origi-nally had a small gas cap in the uppermost part ofthe oil-producing column (Figure 11), and core ex-tracts from theHitch upper reservoir containmorelight hydrocarbons than those from the Etzold res-ervoir; (2) no biomarker differences were observedbetween the oil leg and solid bitumen; and (3) theoil-like asphaltenes consist predominantly of ali-phatic moieties.
The origin of gases in the Hitch reservoir is un-clear, but possible sources for gas in the AnadarkoBasin include Pennsylvanian and Permian shaleswith type III kerogen (Rice et al., 1989) or sub-Pennsylvanian shales or carbonate at the mature
Figure 9. The whole-oil chromatogram of the Hitch oil was compared to those from oils derived from four major source rocks in theAnadarko Basin in an attempt to correlate the Hitch oil to other oils and source rocks. In the above chromatograms, the peak labeled * isan internal n-C36D74 internal standard. The n-C20 and n-C30 alkanes are labeled on the Hitch chromatogram, and this chromatogram canbe used as a reference for the other chromatograms that were all run under identical conditions. The Hitch oils are suggested to be notuniquely sourced from Ordovician source rocks but instead a mixture of Ordovician, Devonian, and possibly other source rocks in theregion. CPI = carbon preference index.
Figure 10. Comparison of m/z 191 andm/z 217 chromatograms for branched andcyclic compounds from three different in-tervals of the Hitch 8-3 well and from theupper oil column of the Etzold 4-3 well. Peakidentifications are given in Table 4.
Kim et al. 1045
and postmature stages of hydrocarbon generation(Burruss and Hatch, 1989). The chemical compo-sitions of gas samples from the upperMississippianreservoirs in the Hitch field are similar (Table 6)with relatively high concentrations of C2+ compo-nents and significant quantities of nonhydrocarbongas components (nitrogen, helium, and carbon di-oxide), indicating thermogenic gases and probablyoil-associated gases (Schoell, 1983;Whiticar, 1994).Large volumes of organic-rich, thermally maturesource rocks are absent in the Kansas shelf area.Thus, gases were probably generated from Pennsyl-vanian type III kerogen or older source rocks in thedeep Anadarko Basin during the mature stage ofhydrocarbon generation (Rice et al., 1989). Further
1046 Hitch Field Bitumens
support for the Pennsylvanian origin for these gasescan be found in the comprehensive article by Jendenet al. (1988), which discussed in great detail theorigin of gases in Kansas. Without reiterating allof the information in that article, note that they ob-served that sub-Mississippian gases in the area areparticularly enriched in He with values of 3.86 to3.92% in gases known to be generated from theLowerOrdovician ArbuckleGroup. As can be seenfrom Table 6, gases in the Hitch field have Hecontents that for most part are less than 0.4% andthat along with various other parameters supportthe hypothesis that these gases are derived fromPennsylvanian-age source rocks and not Ordovicianor Devonian source rocks.
Table 4. Identification of Chromatographic Peaks at m/z 191
and m/z 217 Shown in Figure 10
Peak
Structure Assignment
At m/z 191
23* C23 tricyclic terpane 24* C24 tricyclic terpane 25* C25 tricyclic terpane 26* C26 tricyclic terpane (22S and 22R) 28* C28 tricyclic terpane (22S and 22R) 29* C29 tricyclic terpane (22S and 22R) Ts C27 18a(H)-22,29,30-trisnorhopane Tm C27 17a(H)-22,29,30-trisnorhopane 29H C29 17a(H),21b(H)-norhopane 30H C30 17a(H),21(b)-hopane 31H C31 17a(H),21b(H)-homohopane (22S and 22R) 32H C32 17a(H),21b(H)-homohopane (22S and 22R) 33H C33 17a(H),21b(H)-homohopane (22S and 22R) 34H C34 17a(H),21b(H)-homohopane (22S and 22R) 35H C35 17a(H),21b(H)-homohopane (22S and 22R)
*Each of the diasteranes and regular steranes contains four epimers, from leftto right peak in the chromatogram: diasteranes: 13b,17a-20S, 13b,17a-20R,13a,17b-20S, and 13a,17b-20R; regular steranes: 14a,17a-20S, 14b,17b-20R,14b,17b-20S, and 14a,17a-20R.
Table 5. Biomarker Parameters for Thermal Maturity of the
Hitch 8-3 6162.2 ft 0.48 0.59 0.46 0.66 Hitch 8-3 6199.2 ft 0.47 0.61 0.46 0.64 Hitch 8-3 6206.2 ft 0.44 0.60 0.44 0.65 Hitch 8-3 6213.4 ft 0.48 0.60 0.44 0.64 Hitch 8-3 6232.9 ft 0.54 0.58 0.46 0.60 Hitch 8-3 6236.1 ft 0.39 0.56 0.50 0.65 Hitch 8-3 6239.4 ft 0.49 0.59 0.42 0.62 Hitch 8-3 6246.5 ft 0.46 0.59 0.46 0.63 Etzold B-1 6268.2 ft 0.50 0.59 0.45 0.64 Etzold 4-3 6260.1 ft 0.46 0.59 0.45 0.65 Etzold 4-3 6271.5 ft 0.44 0.57 0.45 0.65
*Ts/Tm and 22S/22R: Two maturity parameters measured in the distributionof pentacyclic hopanes at m/z 191 chromatogram: Ts/(Ts + Tm) from C27-18a(H)-22,29,30-trisnorhopane (Ts) and C27-17a(H)-22,29,30-trisnorhopane(Tm) and 22S/(22S + 22R) from C31-, C32-, and C33-homohopanes.
**20S/20R and bb/aa: Two maturity parameters measured from the distributionof C29-regular steranes in the m/z 217 chromatogram: (20S/20R = 20S/(20S + 20R) and bb/aa = bb/(bb + aa) isomerization at C14 and C17 positions).
Reduction of Temperature and Pressure
Asphaltenes can precipitate in a reservoir as a resultof reduced pressure and temperature (Hammamiet al., 2000). The occurrence of several small solidbitumendeposits at near-zero-permeability bound-aries, perhaps resulting from gravitational settlingof asphaltene particles through the oil column, sup-ports pressure reduction as a feasible mechanismfor asphaltene precipitation (Wilhelms and Larter,1994). TheHitch reservoir had a reservoir pressureof 1920 psi (13.2MPa) and a temperature of 63°C(145°F) at discovery (Sorenson et al., 1999). Recentstudies suggest that the current pressure and tem-perature in the Hitch field are extremely subnor-mal compared to normal pressure and temperature
gradients for foreland sedimentary basins like theAnadarko Basin (Lee andDeming, 2002; Sorenson,2005), supporting pressure reduction as a possibleformation mechanism for the solid bitumen in theHitch reservoir.
In-Reservoir Oil Mixing
Geochemical characteristics of the Hitch field oiland core extract sample indicate the possible mix-ing of input frommultiple source rocks resulting inasphaltene precipitation in theHitch reservoir. Oilmixing as a cause for the asphaltene precipitation,and subsequent solid bitumen formation,was testedby conducting someverybasic andpossibly over sim-plistic simulation experiments.Despite the simplistic
Figure 11. North–south cross section of Mississippian strata along the axis of the Chester incisement, showing the position of solidbitumen (dark green) and gas cap (pink) in the Hitch reservoir with variation of porosity and permeability in the Hitch 8-3 well (modifiedfrom Sorenson et al., 1999, used with permission of AAPG). The Ste. Genevieve Limestone subcrops below Chester strata throughoutmost of Seward County, although deeper incisement along the channel axis has locally exposed the St. Louis Limestone, as depicted inFigure 2. Other analytical results from the highlighted Hitch 8-3 well are shown in Figures 4 and 6. See Figure 1 for the location of the ABcross section. The logs on the cross section are gamma ray (track 1), shallow, medium and deep resistivity (track 2), and neutron-densityporosity (track 3). The vertical exaggeration is approximately 20:1. The cross section red shading occurs where density porosity is greaterthan neutron porosity on a limestone scale and is indicative of sandstone and/or gas effect. The cross section blue curves are coreanalysis data. Permeability is in track 2, increasing to the right on a logarithmic scale. Porosity is in track 3, on the same scale as theunderlying neutron-density porosity curves. To the right of the cross section, deep pink is the core porosity and bluish purple is the corepermeability. Both measurements are abnormally low within the bitumen zone not due to rock properties but instead to the failure of thecore cleaning process to remove all of the bitumen.
Kim et al. 1047
nature of these experiments, the resulting data areuseful in the sense of acting as useful guidelines forfuture studies in this area and as developing simu-lation studies that may more closely resemble pro-cesses occurring under reservoir conditions.Artificialmixing of oils, at standard temperature andpressure,produced from the Anadarko Basin was used to de-termine the amounts of heavy organic materialsprecipitated when individual oils are mixed basedon a simple mass-balance approach using relativeconcentrations of individual hydrocarbons.
We observed that when two oils were mixed,the amount of solid material precipitated aftertreatment with n-pentane increased up to 58 wt.%(Figure 12). The increase in the proportion of theprecipitated organic material from the mixed oilsrelative to the sum of individual oils varies depend-ing on geochemical characteristics of two individ-ual oils, but all mixed oils showed an increase inthe precipitated material. The precipitated solidmaterial was dominated by the asphaltene fractioninstead of paraffin wax in most of the mixing ex-periments. The precipitated asphaltenes increased
1048 Hitch Field Bitumens
in the range of 19–69 wt.% (average 39%), whereasparaffin waxes increased from 2 to 45 wt.% (aver-age 20%). For example, when the asphaltene-richoil (OK-23, Devonian–Mississippian WoodfordShale) was mixed with paraffin-rich oil (City ofLiberal C1, Ordovician source rock; Dahdah andWavrek, 1997), possibly similar to the mixing thatoccurred in the Hitch reservoir, asphaltene precip-itation increased by 29 wt.% (12.8 to 16.4 mg/g oil)and the precipitated wax increase was 10 wt.%(5.1 to 5.6 mg/g oil).
Applying these mixing experiments to large-scale field casesmaybe questionable to some extent,but the results suggest that oil mixing contributesto the formation of solid bitumen in the Hitch res-ervoir. As suggested in the literature (LeythaeuserandRückheim, 1989; Larter et al., 1990;Mansoori,1997), paraffinic crude oils have less capacity todissolve asphaltene and wax fractions, leading toflocculation of the asphaltene particles in paraffiniccrude oils. In summary, precipitation of solid bitu-men in the Hitch reservoir appears to be inducedby mixing of oils that have different geochemical
Table 6. Geochemical Composition of Gases Produced from Upper Mississippian Sandstone Reservoirs in Seward County, Kansas*
*The gas data were provided by Anadarko Petroleum Corporation, and the analytical method is not available.
compositions, especially paraffinic and asphaltene-rich crude oils, possibly frommultiple source rocksfilling the reservoir over an extended period, alongwith an additional gas charge(s).
RESERVOIR FILLING HISTORY
Detailed knowledge of the spatial heterogeneity ofcrude-oil composition within a reservoir providescritical information on reservoir fill direction, type,and maturity of the source rocks that contributedthe petroleum charges (Leythaeuser andRückheim,1989; Hillebrand and Leythaeuser, 1992). In thisstudy, a possible reservoir-filling scenario was pro-posed to explain the observed variations in oil com-position and their resulting oil fingerprints.
The Hitch oil field is located in the northernpart of an incised-valley system and is structurally
higher than the Etzold field. Both reservoir unitsconsist of a sequence of fluviodeltaic sandstonesthat are frequently withmudstones and calcareousshales. Thus, the permeability and porosity of thereservoir rocks are variable, and the oil fields maybe vertically and laterally separated by permeabil-ity barriers, probably in place before the formationof the solid bitumen. We assume that the Hitchand Etzold fields were in pressure communicationto some extent at an early stage of reservoir filling.Oil expelled from source rocks at modest burialdepth in the Anadarko Basin began to migrate tothe Kansas shelf area and initially filled only the toppart of the structure in the Hitch field (Figure 13).As the basin continued to subside, newly generatedoil charging the reservoir became successivelymoremature. Later, oil expelled from other potentialsource rocks in themarginal part of the basin beganto mix with previously emplaced oil in the Hitch
Figure 12. Comparison of the amount of solid material precipitated between single and mixing oils. The expected amount was calcu-lated by the sum of the precipitated solid materials when the individual oils were not mixed.
Kim et al. 1049
and Etzold reservoirs. The later-arriving oil appearsto occupy all pore space in both reservoirs, but theearlier-arriving oil was dominant in the upper sand-stone of the Hitch field. Although most geochem-ical characteristics are similar throughout the field,suggesting the same source for the Hitch and Etzoldoils, compositional variations within the individualreservoir units were recognized in the C15+ hydro-carbons. The progressive trap filling with ineffi-cient mixing process of reservoir fluids may leadto compositional gradients observed in core extracts
1050 Hitch Field Bitumens
and compositional heterogeneities preserved in theHitch reservoir as discussed below. According to thefirst in–last out hypothesis proposed by Wilhelmset al. (1996), extracted oil from crushed rock in se-quential extraction is considered to be the first crudeoil that entered the sandstone reservoir. Oil ex-tracted from whole cores is the later charge to thereservoir. Little difference in the distributions ofC15+ hydrocarbons between the Hitch and Etzoldwhole-core extracts is observed, whereas a signifi-cant difference was found in the composition of
Figure 13. Proposed model for the fillinghistory of the Hitch and Etzold fields. Oilfilling of the stacked reservoirs was amultiple-charge process.
whole-core and crushed core extracts in the Hitchfield (Figure 14). In addition, the extracted oilrecovered from the lower Hitch and Etzold reser-voirs displays a fairly homogeneous compositionbased on chemical properties and biomarker distri-butions, which is more similar toOrdovician oil thanthat from the upper column of the Hitch reservoir.Accordingly, the upper column of the Hitch fieldreservoir may have received multiple charges of oil,including asphaltene-rich Woodford oil, paraffin-rich Ordovician oil, and gases, probably in chrono-logical order. The early oil charge from WoodfordShale was limited primarily to the upper Hitchreservoir and may remain as a trace amount in theEtzold reservoir.
We recognize that the above represents onepossible filling-history scenario, which in our opin-ion closely reflects the geochemical and geologicaldata available. However, the compounds used toreach these conclusions and shown in Figure 14may not possibly be genetically significant but sim-
ply change as a result of thermal maturation andpossibly mild biodegradation. If that was the case,the filling history may possibly be related to mixingof charges from the same source rock as it experi-enced a changing geothermal history.
Timing of Solid Bitumen Formation
Two scenarios can be postulated for the relativetiming of gas de-asphalting and pressure reduction.Pressure drop in reservoirs throughout the mid-continent occurred during the early Tertiary Lara-mide orogeny (Figure 15). However, whether gasesmigrated and filled the Hitch reservoir before orafter the orogeny is uncertain. The first scenario isbased on geochemical observations that the Hitchfield contains more associated gas than the Etzoldfield. The Hitch reservoir oil may have containedsome natural gas in solution and was saturated withgas under the reservoir temperature and pressureconditions during the early stage of oil accumulation.
Figure 14. Polar plot of peak height ratios for whole-core and crushed rock extracts from each reservoir unit. Peaks were selected frombranched and cyclic compounds in the range of n-C18 to n-C21 on the gas chromatogram.
Kim et al. 1051
Normal-phase equilibrium calculations show thata decrease of pressure above the bubble-point pres-sure could induce asphaltene precipitation, but be-low the bubble-point pressure gases are liberated,leading to pressure reduction and increasing thesolubility of the asphaltene in the oil (HirschbergandHermans, 1984; Edmonds et al., 1999). As gascomes out of solution, the oil becomes undersatu-rated and pressure is reduced. An additional pressuredrop could result from the regional uplift leadingto asphaltene precipitation in the Hitch reservoir.
The second scenario is based on regional tec-tonics. Regional uplift in the earlyTertiary Laramideorogeny removed 2000–5000 ft (610–1524 m) ofoverburden throughout themid-continent (Sorenson,2005). Uplift of asphaltene-rich oils in the Hitchreservoir containing asphaltene-rich oils may haveresulted in deposition of asphaltenes due to a de-crease in solvent capacity (Trindade et al., 1996).The pressure drop in the reservoir could have forcedadditional oil and gas to migrate northward and fill
1052 Hitch Field Bitumens
the preexisting fields as a result of the increasingpressure difference between source regions and res-ervoirs. Gaseous components move faster and aremoredependent on pressure differences than liquidhydrocarbons, which made it possible for gases tomigrate far from source rocks in the deepAnadarkoBasin and accumulate in traps in the Kansas shelfarea. Based on this scenario, gas de-asphaltingwouldbe the major factor for asphaltene precipitation,and the formation of solid bitumenwouldhave beena relatively recent phenomenon in the Hitch reser-voir. However, it is not clear why only the Hitchreservoir received gases since the Hitch and Etzoldfields adjoin each other.
CONCLUSIONS
Several geochemical techniques and concepts wereused to determine possible origins of solid bitumenin the Hitch reservoir. Based on the results from
Figure 15. Changes in pressure and temperature of the Mississippian reservoir in southwest Kansas through geological time. 1000 psi =6.9 MPa.
this integrated study, the following conclusionswere reached.
1. Oils extracted from the Hitch reservoir indi-cate spatial variations in composition consistentwith an upper and lower oil-producing columndivided by a solid bitumen layer, which is en-riched in asphaltenes and microcrystalline wax.Porosity and permeability are highly variablethroughout the reservoir because of the occur-rence of thick impermeable solid bitumen andlithologic heterogeneities.
2. The GC fingerprints and biomarker parametersof the Hitch and Etzold oils indicate similarsource rocks and thermalmaturity. Core extractsfrom the upper Hitch reservoir, especially solidbitumen, are enriched in asphaltenes, light hy-drocarbons, and gaseous components.
3. The reservoired oils in the Hitch field are at ma-turity levels equivalent to Ro of 0.6–0.8%, indi-cating that thermal alteration was not a majorfactor in the formation of solid bitumen.
4. The primary source of the Hitch and Etzold oilsappears to be Ordovician, although most of theoils also have a significant Late Devonian tolower Mississippian Woodford Shale source.
5. Expecting that a singlemechanism is responsiblefor the deposition of solid bitumen in the Hitchreservoir is unreasonable. Gas injection fromdeeper rocks and regional pressure drops as a re-sult of post-Laramide orogeny may have con-tributed to asphaltene precipitation.Depositionof solid organic materials in the Hitch reservoiralso appears to be induced bymixing of oils thathave different geochemical characteristics, espe-cially the addition of gases and paraffinic crude oilto asphaltene-rich oil, frommultiple source rocksfilling the reservoir over an extended period.
6. Possible reservoir-filling scenarios were sug-gested in this study, but the origin and timingfor gas charging of theHitch reservoir are uncer-tain. Clearly, the Hitch field oils are more het-erogeneous in geochemical composition thanthe Etzold field oils because ofmultiple sources.
Finally, deposition of solid bitumen is enig-matic in the Hugoton embayment because the
giant field has been known for production of gasand light oil for many years. Reservoirs in the Hu-goton embayment having oil and reservoir condi-tions similar to the Hitch field have the potentialfor solid bitumen formationwithin their reservoirs.
APPENDIX: EXPERIMENTAL METHODS
Preparation Procedures forGeochemical AnalysisCleaned core samples were finely pulverized to 100–200mesh size (45–90 mm) by amechanical rock grinding machinefor organic geochemical characterization. Uncrushed cores(whole rock) were used for sequential extraction and Ro mea-surements. The geochemical methods used in this study in-clude (1) programmed pyrolysis (Pyran level I-flame ionizationdetector [FID]), (2) GC and high-temperature gas chroma-tography (HTGC), and (3) gas chromatography–mass spec-trometry (GC-MS).
For screening purpose, all rock samples were analyzed byPyran I-FID. About 5–8 mg of crushed rock sample was py-rolyzed under a temperature program (40–600°C [104–1112°F] at 25°C [77°F]/min) in a furnace. The pyrolyzedproducts were transferred directly into an FID using He ascarrier gas, and information on the quantity and type of hy-drocarbons was obtained from the calculation of parameterswith reference to standards of known composition.
For GC and GC-MS analyses, approximately 20–30 g ofcrushed rock was extracted with a 1:1 mixture of chloroform(CHCl3) and iso-octane using a hot Soxhlet extractor for 36 hr.Asphaltenes were isolated from crude oils and core extractsusing a combined method of alumina adsorption and sequen-tial Soxhlet extraction as described by Thanh et al. (1999).An alternative method for the removal of asphaltene fraction,the so-called n-pentane technique, was also used to examinecompositional differences in the maltene fraction between thetwo methods and to characterize possible coprecipitates withasphaltenes (Gürgey, 1998). Some selected core samples wereextracted using sequential extractionmodified from the meth-ods suggested by Price and Clayton (1992) andWilhelms et al.(1996). About 20–30 g of a single whole-core (uncrushed)plug was first extracted with hot iso-octane in the Soxhlet ex-tractor, and then crushed core was sequentially extracted inthe Soxhlet unit. The first extract was obtained using hot iso-octane for 24 hr, and for the second extract, a mixture ofchloroform and methanol (95:5 v/v) was used for 6 hr. Theresidual rock after extraction was finally suspended in carbondisulfide in a continuous ultrasonic bath for 40 min to removeall extractable hydrocarbons from the core sample.
Following the separation of asphaltenes, the non-asphaltenefraction was fractionated using open-column chromatogra-phy. Saturate hydrocarbons were recovered by percolationwith n-hexane as eluant on a preactivated alumina column.Aromatic hydrocarbons were adsorbed on the alumina inthe presence of n-hexane and desorbed by a mixed solventof 70:30 (v:v) hexane-dichloromethane. The polar fraction
Kim et al. 1053
consisting ofNSOcompoundswas finally eluted from the chro-matographic column using a 95:5 (v:v) dichloromethane-methanol solution. Molecular sieving using silicalite (UOP-S115) is an effective method to rapidly isolate branchedand cyclic compounds from saturate hydrocarbon fractions(West et al., 1990). Solvents were evaporated to dryness ona rotary vacuum evaporator, and then each fractionwas trans-ferred to 4 mL vials, blown dry under a flow of nitrogen, andweighed.
Waxes were concentrated using a modified wax precipi-tation technique described by Burger et al. (1981). Nonwaxfractions were dissolved in a prechilled (−18°C [−0.4°F])mixture of acetone:petroleum ether (3:1 v/v), and wax frac-tions were recovered by hot iso-octane. Following the waxseparation, microcrystalline waxes containing hydrocarbonsabove C40 were separated from total wax fractions (>C20)by the addition of excess n-pentane. The soluble materialsin n-pentane are called macrocrystalline waxes (C20–C40).
Gas ChromatographyThe saturate and aromatic fractions were analyzed by GCusing a Hewlett Packard 5890A GC and a Varian 3300 GC,respectively. The HP-5890A GC had an on-column injectionsystem and was equipped with a 30 m (98 ft) �; 0.32 mm(0.012 in.) (i.d.) J&W Scientific DB-1 fused silica capillarycolumn with a 0.1-mm film thickness of dimethylpolysilox-ane. The temperature program for the saturate fractions wasinitially 40°C (104°F) with a 1.5-min holding time and thenincreased at 4°C (39.2°F)/min to a final temperature of 310°C(590°F), whichwas held for 31min. The injector and detectortemperatures were 300 and 310°C (572 and 590°F), respec-tively. The Varian GC had two detectors, FID and flamephotometric detector (FPD), equipped with a J&W ScientificDB-5 fused silica capillary column (30 m [98 ft] �; 0.32 mm[0.012 in.] with a 1.0 mm film). Wax fractions were analyzedusing a CarloErba GC 8000 HTGC equipped with an on-column injector and a 25-m (82-ft) �; 0.32-mm (0.012-in.)(i.d.) �; 0.1 mm SGE HT5 fused-silica capillary column. Theoven temperature was programmed from 40 to 370°C (104 to698°F) at 4°C (39.2°F)/min, with the flame ionization detec-tor temperature set at 380°C (716°F). Helium was used as acarrier gas at a flow rate of 4 mL/min.
Gas Chromatography–Mass SpectrometrySelected branched/cyclic and saturate fractions were ana-lyzed using a FinniganMAT triple-stage quadrupolemass spec-trometer (TSQ-70) linked to a Varian 3400 gas chromato-graph. The gas chromatograph was equipped with a 60-m(197-ft) �; 0.32-mm (0.012-in.) (i.d.) J&W Scientific DB-5fused silica capillary column with a 0.25-mm film thickness.The temperature of the GC oven was initially 40°C (104°F)for 1.5 min and then increased at 4°C (39.2°F)/min to a finaltemperature of 310°C (590°F) and held for 31min. The initialinjector temperaturewas 40°C (104°F)without hold time andthen increased at 169.8°C (337.64°F)/min to 310°C (590°F)and held isothermal for 99 min. The temperatures of both the
1054 Hitch Field Bitumens
detector and transfer line were 310°C (590°F). Analyses wereundertaken in the full-scanmodeor single ionmonitoringmodeat 50–600 Da/s and ionization energy of 70 eV, and data werecollected on the Digital workstation (DEC station 5000/25)using ICIS software. Component identification was made bya comparison of retention time with samples of known com-position and coinjection with a standard sample. Relative ra-tios of certain biomarkers and biomarker groups were ob-tained from peak areas.
Petrographic AnalysisThe XRD analysis was performed on finely powdered coresamples from the Chesterian sandstone sections in the Hitch8-3 andEtzold B-1wells; 20 samples from6147.2 to 6247.4-ft(1873.6 to 1904.2-m) interval and four samples from 6246.1to 6289.8-ft (1903.8–1917.1-m) interval, respectively. Pow-dered samples mounted on glass slides with methanol anddried at room temperature were analyzed on a Rigaku auto-mated diffractometer equipped with a graphite monochrom-eter using ëKa copper radiation. The focused monochromaticbeam was obtained with a filament intensity of 30 mA at40 kV. Scans were performed from 3 to 70° 2è at 0.05°/s. Thex-ray diffractometer charts were obtained with Jade software,and the accuracy of measurements is ± 0.01 °2è or ± 0.02 A(as determined by Robert Turner).
Pellets of dispersed organic matter were prepared fromwhole-rock and kerogen concentrate samples containing or-ganic matter for Ro measurement and visual kerogen analysis.Thewhole-rock or kerogen concentrate was placed in 1.75-in.(4.4-cm) ring form with a mixture of Buehler Epoxide epoxyresin and hardener (5:1 v/v). The pellet was polished with0.05-mmWendt Dunnington alumina slurry. The organic pel-lets of whole rocks and kerogen concentrates were examinedusing a Leitz MPV compact microscope (Leitz Ortholux IIPOL-BK) with both a white light and 100-W mercury lampas the ultra-violet light source. Reflected light was used to ob-serve the distribution of solid bitumen in the rock and mea-sure the reflectance of phytoclasts in the kerogen concen-trates. A total magnification of 500�; with oil immersion(Cargille Type B, ne (23°C [73.4°F]) = 1.5180) was usedfor reflectance measurements.
REFERENCES CITED
Beserra, T., 2008,Oil classification and exploration opportunityin theHugoton embayment, western Kansas, and Las Ani-mas arch, eastern Colorado: AAPG Search and DiscoveryArticle 10146, 21 p., http://www.searchanddiscovery.net.
Blanc, P., and J. Connan, 1992, Origin and occurrence of 25-norhopanes: A statistical study: Organic Geochemistry,v. 18, no. 6, p. 813–828.
Bolyard, D. W., 1995, Influence of composition and tem-perature on hydrocarbon migration through Morrowfluvial reservoir, Las Animas arch, Colorado (abs.):AAPG 1995 Annual Convention, Official Program, v. 4,p. 11A.
Burger, E. D., T. K. Perkins, and J. H. Striegler, 1981, Studiesof wax deposition in the trans-Alaska pipeline: Journal ofPetroleum Technology, v. 33, p. 1075–1086.
Burruss, R. C., and J. R. Hatch, 1989, Geochemistry of oilsand hydrocarbon source rocks, greater Anadarko Basin:Evidence for multiple sources of oils and long-distancemigration, inK. S. Jonhson, ed., Anadarko Basin Sympo-sium 1988: Oklahoma Geological Survey Circular 90,p. 53–64.
Cardott, B. J., 1989, Thermal maturation of the WoodfordShale in the Anadarko Basin, in K. S. Johnson, ed.,Anadarko Basin Symposium 1988: Oklahoma Geologi-cal Survey Circular 90, p. 32–46.
Carter, L. S., S. A. Kelley, D. D. Blackwell, and N. D. Naeser,1998, Heat flow and thermal history of the AnadarkoBasin, Oklahoma: AAPG Bulletin, v. 82, no. 2, p. 291–316.
Cirilo, L. L., 2002, Transgressive estuarine fill of an incisedpaleovalley, upperMississippianChesterian series, Shuckfield area, Seward County, Kansas: Master’s thesis, Uni-versity of Houston, Houston, Texas, 194 p.
Coates, G., 1999, NMR logging: Principles and applications:Houston, Gulf Publishing Co., 233 p.
Curiale, J. A., and W. E. Harrison, 1981, Correlation of oiland asphaltite inOuachitaMountain region ofOklahoma:AAPG Bulletin, v. 65, no. 11, p. 2426–2432.
Dahdah, N. F., and D. A. Wavrek, 1997, Characterization ofhigh molecular weight paraffins in Ordovician SimpsonGroup reservoirs (Oklahoma and Texas)—Implicationsfor advanced recovery technologies, in K. S. Johnson,ed., Simpson and Viola groups in southern mid conti-nent Symposium 1994: Oklahoma Geological SurveyCircular 99, p. 203–208.
Edmonds, B., R. A. S. Moorwood, and R. Szczepanski, 1999,An unified framework for calculating solid depositionfrom petroleum fluids including waxes, asphaltenes, hy-drates and scales: Fluid Phase Equilibria, v. 158–160,p. 481–489, doi:10.1016/S0378-3812(99)00138-7.
Fowler, M. G., 1992, The influence of Gloeocapsomorphaprisca on the organic geochemistry of oils and organic-rich rocks of Late Ordovician age from Canada, in M.Schidlowski, S.Golubic,M.M.Kimberly,D.M.KcKirdy,and P. A. Trudinger, eds., Early organic evolution: Impli-cations formineral and energy resources: Berlin, Springer-Verlag, p. 336–356.
Fowler, M. G., P. Abolins, and A. D. Douglas, 1986, Mono-acyclic alkanes in Ordovician organic matter: OrganicGeochemistry, v. 10, no. 4–6, p. 815–823.
Fowler, M. G., L. D. Stasiuk, M. Hearn, and M. Obermajer,2004, Evidence forGloeocapsomorpha prisca in Late De-vonian source rocks from southern Alberta, Canada: Or-ganic Geochemistry, v. 35, p. 425–441, doi:10.1016/j.orggeochem.2004.01.017.
Gürgey,K., 1998,Geochemical effects of asphaltene separationprocedures:Changes in sterane, terpane, andmethylalkanedistributions in maltenes and asphaltene co-precipitates:Organic Geochemistry, v. 29, p. 1139–1147, doi:10.1016/S0146-6380(98)00134-X.
Hammami, A., C. H. Phelps, T. Monger-McClure, and T. M.Little, 2000, Asphaltene precipitation from live oils: An
experimental investigation of onset conditions and re-versibility: Energy and Fuels, v. 14, p. 14–18, doi:10.1021/ef990104z.
Hillebrand, T., and D. Leythaeuser, 1992, Reservoir geo-chemistry of Stockstadt oilfield: Compositional hetero-geneities reflecting accumulation history and multiplesource input, inC. B. Eckardt, J. R. Maxwell, S. R. Larter,and D. A. C. Manning, eds., Advances in organic geo-chemistry 1991: Oxford, Pergamon Press, Organic Geo-chemistry, v. 19, nos. 1–3, p. 119–131.
Hirschberg, A., and L. Hermans, 1984, Asphaltene phase be-havior: A molecular thermodynamic model, in Charac-terisation des Huiles Lourdes et des Residues Pétroliers:Paris, Éditions Technip, v. 40, p. 492–497.
Hirschberg, A., L. N. J. de Jong, B. A. Schipper, and J. G.Meijer, 1984, Influence of temperature and pressureon asphaltene flocculation: Society of Petroleum Engi-neers Journal, v. 24, p. 283–293.
Holba, A. G., L. Wright, R. Levinson, B. Huizinga, and M.Scheihing, 2004, Effects and impact of early-stage anaer-obic biodegradation on Kuparuk River field, Alaska, inJ. M. Cubitt, W. A. England, and S. R. Larter, eds.,Understanding petroleum reservoirs: Toward an inte-grated reservoir engineering and geochemical approach:Geological Society (London) Special Publication 237,p. 53–88.
Horstad, I., and S. R. Larter, 1997, Petroleum migration, al-teration, and remigration within Troll field, NorwegianNorth Sea: AAPG Bulletin, v. 81, no. 2, p. 222–248.
Hsieh, M., and R. P. Philp, 2000, Characterization of highmolecular weight biomarkers in crude oils: Organic Geo-chemistry, v. 31, no. 6, p. 1581–1588.
Huc,A.Y., P.Nederlof, R.Debarre, B.Carpentier,M.Boussafir,F. Laggoun-Défarge, A. Lenail-Chouteau, andN. Bordas-Le Floch, 2000, Pyrobitumen occurrence and formationin a Cambro-Ordovician sandstone reservoir, Fahud saltbasin, north Oman: Chemical Geology, v. 168, p. 99–112, doi:10.1016/S0009-2541(00)00190-X.
Hunt, J. M., 1996, Petroleum geochemistry and geology:New York, W.H. Freeman and Co., 743 p.
Hwang, R. J., and J. Ortiz, 1998, Effect of CO2 flood on geo-chemistry of McElroy oil: Organic Geochemistry, v. 29,no. 1-3, p. 483–503.
Jacobson, S. R., J. R. Hatch, S. C. Teerman, and R. A. Askin,1988,MiddleOrdovician organicmatter assemblages andtheir effect on Ordovician-derived oil: AAPG Bulletin,v. 72, no. 9, p. 1090–1100.
Jenden, P. D., K. D. Newell, I. R. Kaplan, andW. L. Watney,1988, Composition and stable isotope geochemistry ofnatural gases fromKansas, mid-continent, U.S.A.: Chem-ical Geology, v. 71, p. 117–147, doi:10.1016/0009-2541(88)90110-6.
Johnson, K. S., 1989, Geologic evolution of the AnadarkoBasin, inK. S. Johnson, ed., Anadarko Basin Symposium1988: Oklahoma Geological Survey Circular 90, p. 3–12.
Jones, H. P., and R. G. Speers, 1976, Permo-Triassic reservoirof PrudhoeBay field,NorthSlope,Alaska, in J. Braunstein,ed., North American oil and gas fields: AAPGMemoir 24,p. 23–50.
Larter, S. R., K. O. Bjørlykke, D. A. Karlsen, T. Nedkvitne,T. I. Eglinton, P.E. Johansen,D. Leythaeuser, P.C.Mason,A.W.Mitchell, andG. A.Newcombe, 1990,Determina-tion of petroleum accumulation histories: Examples fromthe Ula field, Central Graben, Norwegian North Sea, inA. Buller, ed., North Sea oil and gas reservoirs: II: London,Graham and Trotman, p. 319–330.
Lee, Y., and D. Deming, 1999, Heat flow and thermal historyof the Anadarko Basin and the western Oklahoma plat-form: Tectonophysics, v. 313, p. 399–410, doi:10.1016/S0040-1951(99)00210-3.
Lee, Y., andD.Deming, 2002,Overpressures in theAnadarkoBasin, southwesternOklahoma: Static or dynamic?: AAPGBulletin, v. 86, no. 1, p. 145–160.
Leythaeuser, D., and J. Rückheim, 1989, Heterogeneity of oilcomposition within a reservoir as a reflection of accumu-lation history: Geochimica et Cosmochimica Acta, v. 53,no. 8, p. 2119–2123, doi:10.1016/0016-7037(89)90330-X.
Mansoori, G. A., 1997, Modeling of asphaltene and otherheavy organic depositions: Journal of Petroleum Scienceand Engineering, v. 17, p. 101–111, doi:10.1016/S0920-4105(96)00059-9.
Milner, C. W. D., M. A. Rogers, and C. R. Evans, 1977, Petro-leum transformations in reservoirs: Journal of Geochem-ical Exploration, v. 7, p. 101–153, doi:10.1016/0375-6742(77)90079-6.
Montgomery, S. L., and E. Morrison, 1999, South Eubankfield, Haskell County, Kansas: A case of field redevelop-ment using subsurface mapping and 3-D seismic data:AAPG Bulletin, v. 83, p. 393–409.
Noble, R., R. Alexander, and R. I. Kagi, 1985, The occurrenceof bisnorhopane, trisnorhopane, and 25-norhopanes asfree hydrocarbons in someAustralian shales:OrganicGeo-chemistry, v. 8, no. 2, p. 171–176, doi:10.1016/0146-6380(85)90035-X.
Peters, K. E., and J. M. Moldowan, 1993, The BiomarkerGuide: Interpretingmolecular fossils in petroleumand an-cient sediments: Englewood Cliffs, New Jersey, PrenticeHall, 363 p.
Peters, K. E., J. M. Moldowan, M. A. McCaffrey, and F. J.Fago, 1996, Selective biodegradation of extended ho-panes to 25-norhopanes in petroleum reservoirs. Insightsfrommolecularmechanics:OrganicGeochemistry, v. 24,no. 8/9, p. 765–783, doi:10.1016/S0146-6380(96)00086-1.
Peters, K. E., C. C. Waters, and J. M. Moldowan, 2005, Thebiomarker guide: II. Biomarkers and isotopes in petro-leum systems and Earth history: Cambridge, UnitedKingdom, Cambridge University Press, 1155 p.
Price, L.C., and J. L.Clayton, 1992, Extraction ofwhole versusground source rocks: Fundamental petroleumgeochemicalimplications including oil-source rock correlation: Geo-chimica et Cosmochimica Acta, v. 56, p. 1213–1222,doi:10.1016/0016-7037(92)90057-P.
Reed, J. D., H. A. Illich, and B. Horsfield, 1986, Biochemicalevolutionary significance of Ordovician oils and theirsources, in D. Leythaeuser and J. Rullkötter, eds., Ad-vances in organic geochemistry 1985: Oxford, PergamonPress: Organic Geochemistry, v. 10, p. 347–358.
1056 Hitch Field Bitumens
Rice, D. D., C. N. Threlkeld, and A. K. Vuletich, 1989, Char-acterization and origin of natural gases of the AnadarkoBasin, in K. S. Johnson, ed., Anadarko Basin Sympo-sium 1988: Oklahoma Geological Survey Circular 90,p. 47–52.
Rogers, M. A., J. D.McAlary, and N. J. L. Bailey, 1974, Signif-icance of reservoir bitumens to thermal-maturation stud-ies, Western Canada Basin: AAPG Bulletin, v. 58, no. 9,p. 1806–1824.
Rooney, M. A., A. K. Vuletich, and C. E. Griffith, 1998,Compound-specific isotope analysis as a tool for charac-terizing mixed oils: An example from the West of Shet-lands area:OrganicGeochemistry, v. 29, no. 1–3, p. 241–254, doi:10.1016/S0146-6380(98)00136-3.
Rullkötter, J., and D. Wendisch, 1982, Microbial alterationof 17a(H)-hopane in Madagascar asphalts: Removal ofC-10 methyl group and ring opening: Geochimica etCosmochimica Acta, v. 46, no. 3, p. 1543–1553.
Schmoker, J.W., 1986, Oil generation in the Anadarko Basin,Oklahoma and Texas:Modeling using Lopatin’smethod:Oklahoma Geological Survey Special Publication 86-3,40 p.
Schoell, M., 1983, Genetic characterization of natural gases:AAPG Bulletin, v. 67, no. 12, p. 2225–2238.
Severy, C. L., 1975, Subsurface stratigraphy of the Chesterianseries, southwest Kansas: Master’s thesis, University ofColorado, Boulder, 120 p.
Sorenson, R. P., 2005, A dynamic model for the PermianPanhandle and Hugoton fields, western Anadarko Basin:AAPG Bulletin, v. 89, no. 7, p. 921–938, doi:10.1306/03010504045.
Sorenson, R. P., S. P. Kelly, and D. Cantwell, 1999, Tar matformation within the Hitch oil field, Seward County,Kansas, in D. F. Merriam, ed., Transactions of the1999 AAPG Midcontinent Section Meeting: KansasGeological Survey Open-File Report 99-28, p. 156–165.
Thanh, N. X., M. Hsieh, and R. P. Philp, 1999, Waxes andasphaltenes in crude oils: Organic Geochemistry, v. 30,no. 2/3, p. 119–132, doi:10.1016/S0146-6380(98)00208-3.
Trindade, L. A. F., R. P. Philp, A. M. P. Mizusaki, R. L. A. desSantos, E. Tchouparova, andM. S.Djafarian, 1996, Geo-chemical characterization of waxy oils from the DomJuao oil field, Reconcavo Basin, Brazil: Proceedings ofthe 5th LatinAmericanOrganicGeochemistryCongress,Cancun, Mexico, October 1996, p. 281–283.
Wang, H. D., and R. P. Philp, 1997, Geochemical study ofpotential source rocks and crude oils in the AnadarkoBasin, Oklahoma: AAPG Bulletin, v. 81, no. 2, p. 249–275.
West, N., R. Alexander, and R. I. Kagi, 1990, The use ofsilicalite for rapid isolation of branched and cyclic alkanefractions of petroleum: Organic Geochemistry, v. 15,p. 488–501.
Whiticar, M. J., 1994, Correlation of natural gases with theirsources, in L. B. Magoon and W. G. Dow, eds., The pe-troleum system, from source to trap: AAPGMemoir 60,p. 261–283.
Wilhelms, A., and S. R. Larter, 1994, Origin of tar mats in
petroleum reservoirs: II. Formation mechanisms for tarmats: Marine and Petroleum Geology, v. 11, p. 442–456, doi:10.1016/0264-8172(94)90078-7.
Wilhelms, A., and S. R. Larter, 1995, Overview of the geo-chemistry of some tarmats from theNorth Sea andU.S.A.:Implications for tar-mat origin, in J. M. Cubitt andW. A.England, eds., The geochemistry of reservoir: GeologicalSociety Special Publication 86, p. 87–101.
Wilhelms, A., I. Horstad, and D. Karlsen, 1996, Sequentialextraction—A useful tool for reservoir geochemistry?:Organic Geochemistry, v. 24, p. 1157–1172, doi:10.1016/S0146-6380(96)00100-3.
Wilson, R. E., P. C.Keith, andR. E.Haylett, 1936, Liquid pro-pane: Use in dewaxing, deasphalting and refining heavyoils: Industrial Engineering Chemistry, v. 28, p. 1065–1078, doi:10.1021/ie50321a022.
