UCRL-94245, Rev. I PREPRINT

is

TW0 to^'MLl

A MODEL OF HYDROCARBON GENERATION FROM TYPE I KEROGEN: APPLICATION TO THE UINTA BASIN, UTAH

J. J. Sweeney A. K. Burnham R. L. Braun

This Paper Was Prepared For Submittal To The American Association of Petroleum Geologists Bulletin

December 18, 1986

This is a preprint of a paper intended for publication in a journal or proceedings. Since changes may be made before publication, this preprint is made available with the understanding that it will not be cited or reproduced without the permission of the author.

DISCLAIMER

This document was prepared as an acroanl of work sponsored by an aaency of Ihe United Stales Government. Neither the United Stales Government nor Ihe University of California nor any of their employees, makes any warranty, express or Implied, or assumes any legal liability or responsibility for Ihe accuracy, completeness, or useful­ness of any information, apparatus, product, or process disclosed, or represents thai Its aw wonld nol Infringe privately owned rights. Reference herein lo any specific commercial products, process, or service by trade aamc, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United Slates Government or the University of California. The views and opinions of authors expressed herein do not necessarily stale or reflect those of the United Slates Government or Ihe University of California, and shall not be used for advertising or product endorsement purposes.

A Model of Hydrocarbon Generation from Type I Kerogen:

Application to the Uinta Basin, Utah *

J. J. Sweeney, A. K. Burnham, and R. L. Braun

Lawrence Livermore National Laboratory

P. 0. Box 808, Livermore, California 94550

ACKNOWLEDGEMENTS

*Work performed under the auspices of the U.S. Department of Energy by the

Lawrence Livermore National Laboratory under Contract W-7405-ENG-48.

Development of the pyrolysis model was made possible through support by the

Department of Energy Oil Shale and Basic Energy Sciences programs. The initial

impetus for this project came from Arthur Lewis. Jean Younker greatly expanded

the ideas of how to couple the geological and chemical models. Discussions

with and suggestions from David Anderson and David Chapman were very helpful.

Reviews and suggestions by Fred Meissner, David Anderson, Douglas Waples, John

Saxby, Andrew MacKenzie, and Ian Kaplan helped improve the manuscript.

Drafting work by Priscilla Proctor is much appreciated.

ABSTRACT

We have developed a computer model that can predict when and how much oil

and gas are generated from a source rock during its burial and later uplift.

Kinetic parameters for the oil and gas generation reactions are obtained from

high-pressure pyrolysis experiments carried out over a wide range of heating

rates and temperatures. In our kinetic model, which strictly applies only to

Green River shale, we use a single activation energy of 52.4 kcal/mole and

different preexponential factors for different products of primary pyrolysis,

which allows us to extrapolate laboratory-derived kinetics to geologic heating

rates. This is in contrast to the use of wide distributions of activation

energies or artifically low apparent activation energies used in some models

of petroleum formation. When extrapolated to geologic heating rates on the

order of 10°C/My, our kinetics show that T , the temperature of the maximum

rate of oil generation, changes by about 15°C when the heating rate is changed

by an order of magnitude. Changes in pressure have relatively minor effects on

the kinetics of oil generation, but are important for gas generation

reactions. We used geophysical data from oil fields in the Uinta Basin of Utah

to develop a model of the thermal history of Green River Formation source

rocks. This time-temperature history was used in our model to predict the

maturation level of the kerogen at a given depth and to predict changes in the

compositional characteristics of the oil. The shape of calculated oil

generation rate curves, as a function of depth in the basin, mimic the shape

of the overpressure curves; this suggests that oil gas generation may be an

important cause of overpressuring. Maturation levels and compositional

characteristics of the oil predicted by our model agree very well with

characteristics of the oil recovered from the basin.

- 2 -

INTRODUCTION

The Uinta Basin in northeastern Utah provides an ideal setting to study

the evolution of kerogen to petroleum. Oil shale rocks of the Eocene-age Green

River Formation outcrop extensively at the southern edge of the synclinal

basin. The same rocks are also found at depths of 3650 m in the deepest part

of the syncline. Mature petroleum is presently being recovered from these

deeply buried rocks, so we know that kerogen has been converted to petroleum

within the basin.

The term "oil shale," as we use it, refers to fine-grained rock that

contains a large amount of organic material. Geological studies by Bradley

(1925, 1931, 1970), Andersen and Picard (1974), Fouch (1975), and Ryder et al.

(1976) demonstrate that the Green River Formation rocks were formed in a

lacustrine environment that began in the mid- to late Paleocene and reached

its maximum extent in mid-Eocene time. Organic and inorganic matter was

deposited in varying proportions in yearly cycles, creating varves that can be

traced over kilometers (Bradley, 1929; Smith, 1974). The organic material is

diagenetically changed from its original form and is mostly in an organically

insoluble form called kerogen. Petrographic studies indicate that most of the

kerogen is amorphous (Bradley, 1925; Robinson, 1969), and that it is most

likely derived from the lipid fraction of lake algae and from terrestrial

spores and pollen (Yen, 1976). It is a classic example of a Type I kerogen in

the classification scheme of Tissot and Welte (1978).

- 3 -

Oil shale from the Green River Formation has been studied in pyrolysis

experiments for many years at Lawrence Livermore National Laboratory (LLNL).

The purpose of these experiments has been to better understand the process of

extraction and to calculate rates and amounts of oil formation for a variety

of pyrolysis conditions. Most recently we have developed a detailed computer

model that can accurately predict the rates and amounts of chemical products

(or species) developed at heating rates, temperatures, and pressures prevalent

in different kinds of oil shale retort processes (Burnham and Braun, 1985).

Development of this model depended greatly on laboratory measurements,

especially those of Burnham and Singleton (1983), which incorporated heating

rates ranging from 1 to 100°C/h, and pressures of 0.1 to 2.7 MPa. The model

works well within this range. We decided to test the fundamental accuracy of

this model by extending the range of application to lower heating rates and

temperatures.

In this paper geophysical and geological data from the Uinta Basin of

Utah are used to develop a time-temperature history of the kerogen-rich Green

River shale. This time-temperature model is then used with pyrolysis model

calculations to characterize the conversion of kerogen to various hydrocarbon

products through time. The end result is a prediction of the present-day

kerogen maturity level for various depths and locations within the Altamont-

Bluebell and Redwash oil fields of the Uinta Basin. We then compare these

results with maturity data based on material extracted from the same oil

fields.

- 4 -

THE KEROGEN TRANSFORMATION MODEL

Although it has been known for many years that pyrolysis of kerogen

yields liquid hydrocarbons, there has been some doubt about whether petroleum

formation could be explained by a pyrolysis process. In the 1960s research

firmly established that petroleum is formed predominantly by thermal trans­

formation of kerogen (e.g., Philippi, 1965), eventually leading to the general

hydrocarbon maturation scheme shown in Figure 1. However, attempts to

establish a quantitative time-temperature relationship for petroleum formation

(e.g., Tissot et al., 1974; Tissot and Espitalie, 1975; Connan, 1974;

Ishiwatarl et al., 1976; Waples, 1978) have led to a persistent conflict

between the 10-20 kcal/mole activation energies for kerogen conversion

reactions determined from geological studies and the 40-60 kcal/mole

activation energies determined from laboratory pyrolysis.

As a simple first-order model, oil generation can be modeled by the

chemical reaction

dV/dt = -kt; where k = A exp(-E/RT), (1)

where V = volume, t = time, E = activation energy, R = gas constant, T =

absolute temperature, and A = a preexponential or frequency factor.

An analytical solution for a sample undergoing pyrolysis at a constant

heating rate, H , is given by Van Heek and Juntgen (1968):

- 5 -

3r = AV exp dt «

AV exp

E RT Hr /

6XP(- Rf^ dT

E_ _ ARl! E_ RT H E e p^ RT;

(2a)

(2b)

Campbell et al. (1978) used this expression to determine an accurate rate

expression for the generation of oil from Green River oil shale. In their work

the rate constant k was found to be

, , -K ~ 0 ml3 -26390/T ,,N k(s ) = 2.8 xlO e . (3)

The exponential factor corresponds to an activation energy of 52.A kcal/mole.

Campbell et al. (1978b) and Burnham and Singleton (1983) have verified the

accuracy of Eq. (3) for Green River oil shale down to 2°C/h and l°C/h, respec­

tively. In addition, results of Saxby and Riley (1984) for Torbanite, another

Type I kerogen, heated at l°C/week (0.006°C/h) are consistent with Eq. (3).

Since the geological heating rate is often approximately constant during

the oil generation phase, we can use Eqs. (2) and (3) for an initial estimate

of the relationship, for Type I kerogen, between laboratory pyrolysis

experiments and the geological parameters. In Figure 2, we show the

calculated rates of kerogen conversion to oil at several typical geologic

—8 heating rates. For a heating rate of 10 °C/h, the temperature of the peak

rate of oil production T is 170°C, whereas T is only 139°C for a heating p p

-10 -9

rate of 10 °C/h. For comparison, a heating rate of 1.14 x 10 °C/h

equals 10°C/million years. Note that a change of 10% in the heating rate

changes T by less than 1°C, and a doubling of the heating rate changes T

by about 5°C. Moreover, even at low heating rates, Figure 2 shows that the

temperature must exceed 120°C to generate any oil.

- 6 -

This extrapolation from laboratory to geological conditions is more sensi­

tive to the value of activation energy used than to the value of the heating

rate at laboratory conditions. Doubling the rate constant k at all tempera­

tures by doubling the preexponential factor [Eq. (3)] causes T at 0.5 x

-9 10 °C/h to decrease only from 149 to 145°C. However, in the laboratory,

this change in k causes T at 2°C/min to shift from 433 to 421°C, which is

outside the range of experimental error. In contrast, if the activation energy

is decreased by 4.7 kcal/mole and the preexponential factor is adjusted so

that T at 2°C/min remains at 433°C [this, in fact, corresponds to the

-9 results of Shih and Sohn (1980)], T at 0.5 x 10 °C/h decreases from 149

to 134°C. This second case describes the approximate uncertainty in activation

energy for the geological extrapolation.

The use of a significantly lower activation energy, such as the 10-20

kcal/mole corresponding to the "rate doubling every 10°C" rule (Waples, 1980),

results in an unrealistic extrapolation. Such low apparent activation energies

are an artifact of either diffusion-controlled reactions or a distribution of

higher activation energies. Waples (1984) recently pointed this out, but he

didn't cite the work of Campbell et al. (1980), who analyzed the need for

distributed activation energies for Green River kerogen pyrolysis. Waples also

did not cite the work of Campbell et al. (1978a), who demonstrated that weight

loss measurements give artificially low values for the activation energy of

kerogen pyrolysis because they measure a combination of water, oil, and gas

release, a problem recently noted by Yukler and Kokesh (1984). We caution,

however, that the use of a single activation energy may not be appropriate for

other kerogen types.

- 7 -

Although the rate of kerogen decomposition is central to understanding the

origin of petroleum, many other reactions are important. Continued heating can

convert oil to gas if the oil does not migrate to a cooler reservoir.

Secondary reactions are also important in oil shale processing. We have

recently combined our understanding of the organic pyrolysis reactions into a

detailed chemical model (Burnham and Braun, 1985), summarized in Table 1. The

model was largely developed from the results of Burnham and Singleton (1983),

who determined rates and amounts of oil, gas, and water formation in a

self-purging reactor unlike any of those described in a recent review by

Horsfield (1984). Oil compositions indistinguishable from petroleum were

obtained at moderately low heating rates and high pressures. These experiments

demonstrate that the additional water used in the "hydrous" pyrolysis

technique (Lewan et al., 1979) is not crucial to simulate the chemical aspects

of petroleum formation as long as the intraparticle atmosphere during

pyrolysis is dominated by pyrolysis products and not by an inert pressurizing

agent. A complete description of the development of this model is given by

Burnham and Braun (1985).

The model accurately calculates amounts and rates of oil and gas formed

under a wide range of pyrolysis conditions. It consists of 67 first-order,

non-linear, ordinary differential equations solved by numerical integration.

Numerical integration removes the constant heating rate restriction of Eq.

(2). The current version allows the thermal history to be approximated by

numerous segments of constant heating rate, where the heating rate can be

positive, negative, or zero. This overcomes the need for most of the thermal

history approximations of other models (Waples, 1984). The differential

equations specify the rate of change of each gas, liquid, and solid component

- 8 -

in terms of the chemical reactions and oil vaporization. The generated oil is

divided into 50°C boiling-point interval fractions, allowing (for laboratory

experiments) a direct calculation of liquid-phase residence time before

evaporation. Oil evaporation is not included in the geological case (to

conserve computer time), but this feature allows us to calculate variations in

distillation characteristics of the oil with thermal history. The initial oil

generated is higher in molecular weight than that normally produced, so this

formalism automatically includes, to a limited extent, the pyrobitumen

intermediate of some models. The original model incorporated a kinetic

expression that could calculate the dependence of mineral dehydration on

pressure and heating rate (Burnham and Braun, 1985). However, we eliminated

this part of the model for this application because we thought it was too

empirical to extrapolate to geological conditions.

The model includes rate expressions for gas formation, both during the

original kerogen decomposition and from secondary pyrolysis of the carbona­

ceous residue. The original analysis by Campbell et al. (1980) showed that it

was necessary to use a Gaussian distribution of activation energies (Anthony

and Howard, 1976) for hydrogen and methane generation from char. Burnham and

Braun (1985) adopted a similar form for CO generation from kerogen. Pre­

liminary calculations showed that slightly lower activation energies for gas

generation from kerogen, derived by Campbell et al. (1980) and used previously

(Burnham and Braun, 1985), generate gas before oil by 20°C at typical

geological heating rates. We believe that this is chemically unreasonable and

results from slight errors in the measurements, so we adopted the same

activation energy for all kerogen pyrolysis reactions and adjusted the

preexponential factors so that T for each component was equal to the

- 9 -

experimental value at 2°C/min. One characteristic of the gas generation

kinetics is that CO generation largely precedes oil formation. This is

consistent with the observations of Lawler and Robinson (1967), Robin and

Rouxhet (1978), and Tissot et al. (1978) that carbonyl bands in the infrared

spectrum of Green River kerogen disappear with progressive burial before

significant oil generation.

Oil can be destroyed by both coking and cracking. In our terminology, oil

coking is the polymerization and condensation of hydrogen-deficient oil

components, usually hetero-aromatics, to form a predominantly solid residue

(coke). It leads to an increase in aliphatic content and a decrease of the

hetero-atom content. In contrast, oil cracking is the fission of aliphatic

structures to smaller molecules, ultimately methane. It leads to a

concentration of aromatic components in the oil. There is, of course, overlap

between the types of reactions, but this separation has proved useful.

Correlations have been developed for the dependence of oil density and

elemental composition on the amount of coking and cracking (Stout et al.,

1976; Burnham, 1981; Burnham and Singleton, 1983), but they have not yet been

incorporated into the model.

A few general features of the model are worth noting. Figure 3a shows the

rate of oil and gas generation, partitioned between pyrolysis, coking, and

cracking, for typical geologic heating rates. Note that the temperature of the

maximum rate of gas generation in the absence of oil cracking trails the oil

generation by about 5°C. Oil cracking produces another peak in gas generation

at a temperature about 40°C higher, although the peak would be smaller if oil

migration occurred. These additional gas generation processes may be

responsible for the persis- tence of overpressurized zones. The cumulative

- 10 -

amounts of oil and gas formed are shown in Figure 3b. Note that the maximum

concentration of oil occurs at about 170°C, midway between the maximum rate of

generation of the oil from kerogen and destruction of the oil by cracking to

gas.

A variety of gas-generating processes cause the gas composition to change

during the course of maturation. Figure 4 shows one indicator,

(C_-C )/(C.-C ) in the gas. Without oil cracking, this ratio increases

during nil generation and reaches a maximum at the completion of oil genera­

tion. At higher temperatures, methane generation from char pyrolysis causes

the ratio to decrease. If oil remains to be cracked, the ratio increases fur­

ther. The present version of the model does not allow C -C hydrocarbons

to crack further to methane, so this ratio is probably too high at very high

maturation. Likewise, the model generates substantial quantities of hydrogen

that, in reality, react at high pressures and slow heating rates to form

methane and hydrogenated oil. Work is currently in progress to eliminate these

weaknesses from the model.

Pressure is now recognized to be less important than temperature for

petroleum formation (Waples, 1984). The main effect of pressure in our

pyrolysis model is that it affects the rates and composition of the products

of cracking (Burnham and Braun, 1985). In general, pressure can either

increase or decrease reaction rates, depending on conditions (Montgomery and

Chandler, 1979, Skinner and Wolynes, 1980). The current model assumes that

kerogen pyrolysis is independent of pressure, but the rate of oil cracking

first increases 30-fold to a maximum at 40 MPa and then decreases with further

- 11 -

pressure increase. The functional form of this dependence was fitted to data

given by Fabuss et al. (1964). The ratio of gaseous to light oil products also

decreases as pressure increases according to the results of Voge and Good

(1949). In laboratory experiments, pressure also affects the gas-liquid

partitioning and residence times, thereby changing the amounts and relative

importance of cracking and coking reactions. In the present version of the

model we cannot vary pressure with time, although a new version under develop­

ment can.

Numerous studies have shown that the composition of the first oil to form

is substantially different from the remainder. Coburn and Campbell (1977) and

Burnham et al. (1982) have shown that most of the biological-marker compounds

are evolved as the first 10 to 20% of the oil is generated from the kerogen.

Some of these compounds are in the initial bitumen and some are released by

kerogen decomposition. Figures 5 and 6 show the conversion-dependent

oil-composition indicators that we calculated using data from Coburn and

Campbell (1977), and Burnham et al. (1982). These indicators represent the

sources of biomarker compounds, which are useful for estimating maturity. The

detailed model has an initial oil content made up from 5% of the original

organic matter and allows 5% of the kerogen to decompose by a faster rate

constant. We found that it was not necessary to use the previously assumed

Gaussian distribution of activation energies (Burnham and Braun, 1985) to get

the oil composition to follow the trends of Figures 5 and 6.

- 12 -

The oil composition indicators of Figures 4, 5, and 6 can be used to

estimate kerogen maturity from oil and gas composition data. These oil

indicators are most effective at maturity levels below 50%. Other indicators,

perhaps based on aromatic/aliphatic ratios, would be more useful at higher

maturity. Correlations of these ratios with oil cracking have been presented

by Bissell et al. (1985).

Our next step was to develop a geologic model of the temperature history

of the Uinta Basin in Utah, which we could then use to calculate the

characteristics of the oil and gas generation for selected locations in the

basin. In a later section, the oil generation data are compared with

production history and characteristics of oil sampled from the basin.

GEOLOGIC SETTING AND HISTORY OF THE UINTA BASIN

The Uinta Basin is a structural and topographic depression in the north­

eastern corner of Utah (Figure 7). The topographic low is surrounded by the

Uinta Mountains to the north, Wasatch Mountains to the west, San Rafael Swell

and Uncompahgre Uplift to the south, and the Douglas Creek Arch to the east.

Relief between the lowest part of the surface of the basin and the surrounding

highlands ranges from 900 to 1800 m.

Hydrocarbon production occurs in Paleozoic, Mesozoic, and Cenozoic forma­

tions in the Uinta Basin. Only the Cenozoic part of the basin, in which an

extensive lacustrine environment was in evidence by the beginning of the

Paleocene Epoch, is of concern in this study. During the early Tertiary as

- 13 -

much as 6000 m of lacustrine and alluvial sediments were deposited. The

sediments represent a central core of open lacustrine claystone and carbonate

mud- stones; a marginal lacustrine facies with sandstone, claystone, and

carbonate; and a peripheral alluvial facies of conglomerate, claystone, and

carbonaceous shale (Fouch, 1975). The lacustrine period of deposition waned in

the late Eocene and later deposition was primarily alluvial in character.

Structurally the basin is a simple asymmetric syncline with an

east-west-trending axis near the northern side. Dips on the north limb are 10

to 35 degrees, whereas they are only 2 to 4 degrees on the south limb (Figure

8). In contrast to the structural simplicity, stratigraphic relations are

complex, but marker units can be traced throughout the basin.

Two specific oil fields in the basin were selected for analysis in this

study because of their differing characteristics and because of the availa­

bility of a large amount of data from drilling. The Altamont-Bluebell field

(Figure 7), described by Lucas and Drexler (1975), produces oil and gas from

multiple thin reservoirs in the lower Tertiary section at depths of 2400 to

3600 m. Traps are in fractured sandstone with stratigraphic pinchouts that are

overpressured in the main producing area. The Redwash field is located about

40 miles (64 km) to the east of the Altamont-Bluebell field. Production at

Redwash is from fractured sandstone lenses located in the lower part of the

Eocene Green River Formation at depths of 1500 to 1800 m (Chatfield, 1972).

Both of these fields produce oil and gas derived from lacustrine rocks at

similar stratigraphic levels, but the present depth of burial of these

producing zones varies considerably.

- 14 -

To develop a model of the thermal history of lithologic units in a basin,

the burial history must be known. The key factor in developing a burial

history is to establish basin-wide stratigraphic markers for which good age

control is available. The complex interfingering stratigraphy of the Green

River Formation of the Uinta Basin makes this difficult. Fortunately, a series

of correlation markers was established by Fouch (1975).

The base of the Tertiary sequence of the Uinta Basin begins with the Upper

Cretaceous to Paleocene North Horn Formation (see Figure 8). The overlying

Paleocene to Eocene Green River Formation consists of the central lacustrine

facies, which interfingers with the marginal and peripheral facies of the

Colton and Wasatch Formations. Markers identified by Fouch (1975) are (1) the

lower marker, (2) the Paleocene-Eocene boundary, (3) the top of the carbonate

marker unit, (4) the middle marker, (5) the Mahogany oil-shale bed, and (6)

the upper marker.

The Green River Formation is overlain by the Uinta and Duchesne River

Formations. These formations are fluvial deposits that mark the demise of

lacustrine conditions and the inception of Laramide tectonic activity

(Andersen and Picard, 1974). The Duchesne River and upper parts of the Uinta

Formation have been eroded away from some parts of the basin and are exposed

in places near the Altamont-Bluebell and Redwash oil fields. The Oligocene

Bishop Conglomerate and Oligocene to Miocene Browns Park Formation, which

overlie the Duchesne River Formation, occur in the area surrounding the Uinta

Mountains, but are largely eroded from the Uinta Basin. The Bishop

Conglomerate is considered by Hansen (1984) to be correlative in age to the

upper part of the Duchesne River Formation. The Bishop probably formed a wide

continuous bajada around the Uinta range, reaching the centers of the

- 15 -

surrounding basins during a long period of crustal and climatic stability

beginning in the Oligocene (Hansen, 1984).

The beginning of Tertiary time has been placed at 66.4 million years ago

(My), with the Paleocene-Eocene boundary at 57.8 My (Decade of North American

Geology Time Scale, Palmer, 1983). Dating using K-Ar ages of biotite in tuff

(Mauger, 1977) suggests an age of about 45.0 My for the Mahogany oil-shale

unit, with the base of the Uinta Formation at about 44.0 My. Mauger (1977)

estimates the uppermost part of the Uinta Formation to be about 41.0 My.

Andersen and Picard (1974) estimate that the uppermost part of the Duchesne

River Formation—the Starr Flat Member—may be as young as Oligocene. Hansen

(1984) considers part of the Bishop Conglomerate to be equivalent to the Starr

Flat Member. Tuff in the Bishop Conglomerate has been dated at 29.0 My

(Hansen, 1984). The period of stability that began about 29 My extended up

into the Miocene, when renewed uplift resulted in downcutting and deposition

of the Browns Park Formation. Tuffs in the upper part of the Browns Park are

8-12 My in age (Hansen, 1984). Uplift and erosion have continued into the

Quaternary.

From the stratigraphic and age data, we have developed a simple geologic

model of the basin history. The time events chosen for the model are as

follows:

Event Time Period (My)

Renewed Uplift

Period of stability and peneplanation

Deposition of Duchesne River Formation

and Bishop Conglomerate

Deposition of Uinta Formation

Deposition of Mahogany oil shale

Paleocene-Eocene boundary - 16 -

10

30

41

44

45

57.

-

-

-

-

8

present

10

30

41

The model assumes that deposition rates were uniform throughout, and thus

represents a time averaging of the actual, more complicated depositional his­

tory. Ages used in the model are not solidly constrained by geologic data, so

the numbers used are what we consider a best estimate. We consider this model

to be a working model that can be modified as further study warrants.

The next step in determining the burial history of the basin is to estab­

lish the history of the thickness and burial depth of the stratigraphic units

of the model. Present burial depths are determined from borehole lithologic

correlations, but they do not represent the maximum depth of burial because a

considerable amount of uplift and erosion of the basin has occurred. Deter­

mination of the amount of overburden removed in the basin is a crucial part of

the analysis.

BASIN ANALYSIS AND DEVELOPMENT OF THE BURIAL HISTORY

Previous studies most pertinent to this work are those of Reed and

Henderson (1971), Tissot et al. (1978), and Anders and Gerrild (1984). Reed

and Henderson analyzed crude oil from nine fields in the Uinta Basin, inclu­

ding Redwash and Pariette Bench, for alkane and elemental compositions. They

found strong evidence for stratigraphic control of crude oil composition and

concluded that the oil shales have not been the dominant source-rocks of the

petroleum in the reservoir. The study by Tissot et al. (1978) showed that the

lower part of the Green River Formation is in the principal stage of oil

generation and is responsible for most of the crude oils produced. Anders and

Gerrild (1984) compared variations in the compositions of organic material

from the Altamont-Bluebell area with location, depositional environment, and

thermal maturation.

- 17 -

An analysis of the historical stress regime for the Altamont Field was

computed by Narr and Currie (1982), who found evidence that fractures

developed in the basin only after burial to the maximum depth, and that

fracturing continued throughout the subsequent period of uplift and erosion.

Pitman et al. (1982) found the Green River Formation at the Pariette Bench

Field to be thermochemically immature and not the source of production. They

suggest that the oil found at Pariette Bench migrated through a network of

fractures from the Bluebell-Altamont area. Anders and Gerrild (1984) also

concluded that migrated oil is present in the Pariette Bench area as well as

in other parts of the basin.

These findings show apparent contradictions that are related to the fact

that oil in various fields may have different sources. The Altamont-Bluebell,

Redwash, and Pariette fields contain Green River Formation rocks at different

present-day burial depths and have different burial histories. The intent of

this study is to integrate information about geological and geochemical

evolution of the rocks in each field to evaluate the oil generation potential.

Knowledge of the oil generation potential at each field will then allow us to

assess the possibility that migrated oil contributes to the field production.

The complete burial history of a basin is determined from the maximum

depth of burial and the thickness of each stratigraphic unit through time. The

time-thickness relation can be determined from the density or porosity of the

unit as a function of burial depth and from the burial history using a method

known as backstripping (Sclater and Christie, 1980; Steckler and Watts, 1978).

Geologic evidence indicates that a considerable part of the stratigraphic

section has been eroded from the Uinta Basin. The Duchesne River Formation and

the upper part of the Uinta Formation are exposed at the surface in some parts

- 18 -

of the basin (Stokes, 1963). Outcrops of the Browns Park Formation occur as

much as 1825 m above the basin floor. Narr and Currie (1982) take this

difference to be representative of the amount of sediment removed from the

basin. Narr and Currie (1982) analyzed fluid inclusion data from core samples

from the Altamont field, estimating that 510 to 2890 m of overburden was

removed. Tissot et al. (1978) use 1780 m for the amount of erosion at the

Shell Brotherson 1-23-B4 well, Altamont field, but they do not say how they

determined that value. Pitman et al. (1982) used reconstructed thicknesses of

the Uinta and Duchesne River Formations to estimate a maximum of 1000 m of

overburden removal in the Pariette Bench Field, to the southeast of the

Altamont Field.

These large differences in estimates of the amount of removed overburden

will have a major influence on estimates of the maximum depth of burial and

the maximum temperatures attained by kerogen-bearing rocks in the basin. The

alluvial nature of the Uinta and overlying formations makes it extremely

difficult to reconstruct the overburden thickness for various parts of the

basin. Because of the widespread availability of downhole acoustic logs (DHAL)

in the basin, we decided to estimate the amount of removed overburden from

shale compaction data, as demonstrated by Magara (1978).

Interval velocity in shales is related to their density and hence

porosity. Because of the irreversible character of shale compaction, shale

density (or porosity) provides a "memory" of the maximum depth of burial that

can be determined by averaging data from many DHAL records in a basin.

Typically, as Magara (1978) did for the Cretaceous shales of Canada, a normal

compaction curve of the basin is determined from an area of little or no

erosion. This normal compaction curve is then used with the log data from

another area to estimate the amount of erosion. In the Uinta Basin erosion has

- 19 -

taken place everywhere, so we had to obtain a normal compaction curve by

averaging the slopes of the compaction from a number of DHAL records.

We used sections of low resistivity (less than 10 ohm-m) on electric logs

to identify the shale-rich layers and then selected the corresponding interval

velocity from the DHAL log. We purposely chose locations where detailed litho-

logic data were available from previously published work (Narr and Currie,

1982; Pitman et al., 1982; Tissot et al., 1978; Reed and Henderson, 1971;

Fouch, 1981; and Fouch and Cashion, 1979), and then obtained a series of logs

that provided data over a depth interval of at least 1525-2440 m. The

locations of these wells are shown in Figure 7.

Overpressures at depth are seen in most logs from the Uinta Basin.

Figure 9 is a representative transit time-depth plot from borehole data. Note

that below depths of 5000 to 6000 ft (1500 to 1800 m), interval transit time

begins to increase with depth and then varies up and down in a cyclic fashion.

Only the portion of the curve above 6000 ft (1800 m) depth has been used to

determine the erosion values. The average value for the slope of the curve,

0.000096 ft"1, is within the range of 0.000085 to 0.000147 ft"1, which

Magara (1978) obtained for Cretaceous shales in western Canada. (Magara

obtained lower slope values for the youngest rocks.)

The results of the above calculations for 13 wells in the Uinta Basin are

given in Table 2, as well as the values of the amount of removed overburden

Z corresponding to a plus-or-minus one-standard-deviation change in slope. Uo

These numbers show that the uncertainty in the estimate of Z Q due to

uncertainty in slope is about + 300 m. Note from the table that lower values

of Z n o are obtained from wells in Redwash field (Chevron Redwash, Chorney

South Redwash, and #13 Broadhurst). The highest value of Z is from the UD

- 20 -

Pariette Bench #5 well. Figure 10 shows the variation in Z n o throughout the

Altamont-Bluebell area. Values of Z n D range from 1800-2000 m except for Gulf UD

Verl Johnson (1598 m) and Duchesne Tribal (2252 m).

Estimates of Z n Q from fluid inclusion data by Narr and Currie (1978) are UD

also listed in Table 2 for comparison with this study. The values of Z B for

the Shell Christensen and Shell Brotherson wells determined by Narr and Currie

roughly agree with values determined here. Other values of Narr and Currie are

either very low or very high compared with our estimates. Tissot et al. (1978)

use a value of 1780 m for overburden removed at the Shell Brotherson 1-23BA

well in Sect. 23, 2S, AW; it agrees well with values of 1875 m (Sect. 3, 2S,

AW) and 1796 m (Sect. 11, 2S, AW) given in Table 2.

The assumption that we and Magara (1978) make is that the zero-depth

interval transit time for water-saturated mud is 200 ys/ft [corresponding to

a velocity of 1.52 km/s (5000 ft/s)]. This corresponds to typical velocities

for water-saturated fine-to-medium silt and compacted terrigenous mud

(Carmichael, 1982). In the case of the Uinta Basin, shales were deposited in a

lacustrine environment during the period of the Green River Formation and in

an alluvial or overbank environment during later periods. The zero-depth

interval transit time to use for alluvial or overbank muds is not known, but

it is most likely that, as a result of later burial, a normal compaction trend

similar to lacustrine or shallow marine environments will be obtained within

depths of a 100 m or so. At most the zero-depth interval transit time may vary

by 10 ys/ft above and below 200 ys/ft. This translates to an uncertainty

in the erosion estimate of about 150 m.

Many of the shale units used to determine Z n o came from the Uinta UD

Formation, much of which formed in an alluvial environment. We calculated

- 21 -

normal compaction trend slopes for units occurring only above the Green River

Formation. The average slope was 3.811 x 10 ft-1 (a = 1.397 x 10~5).

which leads to estimates of Z^ that are generally about 300 m greater than

those of Table 2. The uncertainty in this slope value is greater because there

is more variability in the interval velocity data for shallower well depths

and there are fewer thick shale units in these alluvial sediments. The net

result of using the slope value of Table 2 is a possible underestimation of

the overburden removed, which in turn leads to lower values for maximum burial

depths and consequent estimates of lower hydrocarbon maturity levels.

The calculated erosion values are next added to the known present depth of

a given marker horizon to obtain the maximum depth of burial. We chose to use

a simple bulk-average porosity function to describe the compaction of the

sedimentary layers with burial. Several different porosity functions were used

in the backstripping calculations, including the case where no correction was

made for compaction. Because water depths were never greater than 100 m, no

corrections for water loading were used. The various backstripping

calculations resulted in very similar values for the heating rate at the

critical range (for hydrocarbon maturity estimates) of temperatures greater

than 110°C. The maximum difference in heating rate calculated using the

various porosity functions was only 0.1 x 10 , which has very little effect

on the maturity calculations (see Figure 2).

Detailed analysis of temperature data by Chapman et al. (1984) provides an

estimate of 25°C/km for the present day geothermal gradient in the Uinta

Basin. We here make the assumption that the geothermal gradient from the

Tertiary to the present has been constant, and we ignore localized effects on

thermal gradient such as overpressuring, lithologic variation and hydrothermai

- 22 -

circulation. We choose a value of 10°C for the long-term average surface

temperature. From these assumptions and the time-burial depth data corrected

for compaction, we have constructed time-temperature-depth plots for the well

locations of Table 2. A representative burial history and temperature is shown

in Figure 11 along with values of heating rates for the base of the Eocene

during the separate time intervals.

A change of one standard deviation in the value of Z n Q results in 6 to UD

_9 8°C change in T and a change of about 0.1 x 10 °C/h in the heating

rate. Uncertainty in the porosity function has no effect on T , but leads

_9 to an uncertainty in the heating rate of 0.1 x 10 °C/h. The geothermal

gradient varies throughout the basin (Chapman et al.f 1984) and the present-

day gradient may be different from past gradients; thus, its uncertainty is

difficult to assess. Combining a ± 5°C/km uncertainty in geothermal gradient

with a ± 300-m uncertainty in Z^, a net uncertainty in T of ± 35°C

OB J max

is obtained. This range of temperatures would certainly have a profound affect

on hydrocarbon maturity; such differences can be evaluated by comparing the

geologic model with hydrocarbon maturation data. From the pyrolysis modeling

we know that a doubling of the heating rate changes T by about 5°C; thus,

it is obvious from the analysis of the uncertainty in the geologic data that

maximum temperature is a far more critical variable than heating rate in the

assessment of hydrocarbon maturity.

The bulk of the Type I kerogen contained in post-Cretaceous rocks of the

Uinta Basin occurs in the Green River Formation between the Mahogany shale and

the base of the Eocene. This stratigraphic interval is of prime concern for

hydrocarbon generation. The kerogen-bearing units at the Altamont-Bluebell

field have been heated to maximum temperatures of 149 to 175°C (derived from

data in Table 2). Maximum heating rates occurred during the period of 41 to

- 23 -

30 My and range from 0.51 to 0.74 x 10~ °C/h. The maximum temperature

attained by the base of the EoCene in the well from the Redwash field (Figure

12) was much lower, 111°C, with a lower heating rate between 41 to 30 My of

-9 0.16 x 10 °C/h. Conditions at Pariette Bench (based on data from Table 2)

were intermediate to the two other fields, with a maximum base Eocene

_9 temperature of 132°C and a heating rate of 0.35 x 10 °C/h.

According to our calculations, the Shell Brotherson well (Figure 11) is

well within the oil generation "window" because, at the heating rates shown,

maximum oil generation occurs at about 149°C (see Figure 2). The Chorney Oil

South Redwash well (Figure 12) shows temperatures much too low to be in the

oil production window. The fact that oil is found in the Redwash area suggests

that (assuming our modeling is correct) migration of oil has taken place.

In Figure 13 an oil generation curve, based on heating rates calculated

from the geologic basin model, is superimposed on a drilling log. The peak of

the oil generation curve closely corresponds to the depths of oil shows or

recovery. This indicates that the higher value of activation energy used in

the calculations is applicable to the petroleum formation process. The maxi­

mum oil-recovery depth of Figure 13 is about 300 m above the depth of the

maximum calculated oil generation. Maximum recovery, of course, depends on the

TOC (total organic carbon) of the source rock and characteristics of the

reservoir. Maximum recovery and maximum generation rate will not necessarily

occur at the same depth. Figure 14 is a more quantitative comparison of the

predicted oil generation with the actual generation of hydrocarbons in the

Altamont Field. The shape of the shaded area of Figure 13, indicating over-

pressuring in the well, corresponds to the rise of the rate of oil generation

curve. This supports the theory (Spencer, 1986) that overpressuring in this

area may be caused by volume changes accompanying oil and gas generation.

- 24 -

These preliminary results are encouraging. They indicate that the geo­

logic model is reasonable and consistent with extrapolations of simple labora­

tory retort models of kerogen evolution. In contrast, the parallel reaction

scheme of Tissot and Espitalie (1975) predicts multiple oil generation peaks

(Figure 15) at both laboratory and geological heating rates, even though only

one is observed. Next, we use the detailed geochemical model and predict

hydrocarbon compositions as well as generation rates. We then compare the

modeled composition and generation rate data with published data for the oil

fields to establish the validity of our chemical and geological models and our

preliminary conclusions.

APPLICATION OF THE DETAILED GEOCHEMICAL MODEL

We ran the pyrolysis model, with the complete thermal history from several

wells, to calculate oil formation and degradation at various depths. Table 3

summarizes the results for several of the wells listed in Table 2. The

conversion of kerogen ranges from 7 to 100%. The elemental composition of the

remaining unextractable organic material, plotted in Figure 16, indicates that

the material follows, as expected, the maturation curve of Type I kerogen

shown by Tissot et al. (1978). The calculated boiling point distribution of

the liquid oil for several selected samples is shown in Figure 17. The oil

becomes noticeably more volatile with increasing maturity. This is caused by

three effects: the oil generated from kerogen is more volatile than the

initial bitumen, coking tends to remove heavier oil components, and cracking

converts heavy components to light ones. An interesting observation is that

essentially all the generated cokable oil (30% of the total) has coked. This

accounts, in part, for the difference in the aliphatic content of petroleum

and most laboratory pyrolysates. Another observation is that although very

- 25 -

little of the oil has cracked to gas, very few high-boiling-point components

remain in two of the oil samples. For the other oil samples cracking is

probably less extensive. In this case biomarker/normal alkane ratios can be

related to the extent of kerogen conversion. This provides an additional basis

for comparing the geological and chemical model calculations.

At present, there is no consensus on the best method to assess the extent

of kerogen conversion in organic-rich geologic materials. Advantages and

disadvantages of the use of indicators such as vitrinite reflectance, thermal

alteration index, and pyrolysis temperature are discussed by Anders and

Gerrild (1984). There is also no standard method in practice of analyzing

hydrocarbon composition; thus, different investigations have different means

of representing compositional data. Because our model keeps track of mass

balance for various species, we can calculate or estimate most of the param­

eters used to quantify maturity level. This ability allows us to compare our

model's predictions with a wide range of published data on hydrocarbon

maturity characteristics.

At this point we will use published gas chromatography and Rock-Eval

results from samples recovered from the Uinta Basin to estimate kerogen

maturation levels at different depths and compare these estimates with

maturation levels predicted by our kinetic model. We used gas chromatograms to

calculate the maturation indicator ratios shown in Figures 5 and 6 and thus

obtain an estimate of maturity. From our modeling, the values of liquid oil

and oil ungenerated (Table 3) are roughly equivalent (for Green River shale)

to S and S of the Rock-Eval apparatus. Thus our model results can be

- 26 -

used to calculate the transformation ratio ((S /(S + S ) to be compared

with published values measured from samples recovered from boreholes in the

basin. Below, we compare our calculated values of maturation with estimates

based on recovered samples for individual wells in both the Altamont-Bluebell

and Redwash oilfields.

Three published studies (Reed and Henderson, 1971; Tissot et al.f 1978;

and Anders and Rerrild, 1984) give analyses of hydrocarbons at various depths

from different lithologic facies in the Uinta Basin. Gas chromatograms of the

oil are given by Reed and Henderson and Tissot et al. We used the cumulative

and instantaneous isoprenoid/normal alkane ratios [pristane/CC,., + C.Q) 1 / IB

and phytane/(C,^ + C1Q)] and a normal alkane ratio [C.-./CC, + C1Q)] 1/ lb 1/ 16 IB

(see Figures 5 and 6) to compare these results with our calculations of

predicted maturity in the form of percent of oil generated.

The data of Tissot et al. (1978) are compared with the results of our

model in Table 4. The composition ratios needed for comparison [such as

C.^/(C, . + C._)], were estimated using published gas chromatograms in 1/ 16 lo

the article. Data from the published paper are listed on the left side of the

table, data calculated by our model or interpolated from Figure 18 are listed

on the right side.

The Shell Brotherson 1-11B4 well, which we modeled, is in the same sec­

tion as the Shell Brotherson 1-14B4 and 1-23B4 wells used by Tissot et al.

Using 1796 m for ZnQ, our results agree well with Tissot et al. for all four UtJ

depths sampled. A range of values of Z was used to compare the

2782-m-depth sample of Shell Brotherson 1-23B4 with our model (last two lines

of Section A, right side in Table 4). Our estimate of maturity from Tissot et

al. is 30 to 60% for this well at a depth of 2782 m. The low value of Z_D, OB

- 27 -

1547 m, results in a 19% maturity, whereas the high value of Zno, 2142 m, UD

results in a 73% calculated maturity. Both of these calculated values are out

of the range by roughly equal amounts, suggesting that the value of 1796 m for

Z n Q is a good determination for this section. UD

In Section B of Table 4 we compare the Shell Murdock well data of Tissot

et al. at two different depths with our calculations for the Shell Brotherson

1-11B4 (next section east) and Shell Christensen (next section north) wells at

the same depths. In each case, the calculated values (56, 78, 99, and 100%)

are consistent with the (40 and 60%) values estimated from the oil

composition. In this case, it is difficult to make any conclusions, except

that the Z n D of 1987 m from the Christensen well may be slightly high. Ud

Wells from the Redwash area are compared in Section C of Table 4. Exact

locations of the wells used by Tissot et al. are not given in their paper, so

we compared their data with a range of values for Z n D in the Redwash area. Ud

For the shallower wells (1698 m), the calculated maturity of 6 to 7% is within

the range of 0 to 20% estimated from the oil composition. Values of maturity

from our calculations at the deeper levels (2649 m) are smaller than those

from actual samples from the area. The highest value of Z used, 1803 m, UD

results in a maturity level of 24% at 2649 m, which is less than the 30 to 60%

maturity indicated by the data from Tissot et al. This discrepancy can be

explained by assuming that the oil sampled from the field had migrated from a

more mature source area.

- 28 -

In Table 5 we compare recovered oil samples analyzed by Reed and Henderson

(1971) with results from our modeling. The procedure is the same as for Table

4, except that in this case the wells used for the model are farther away from

the wells used by Reed and Henderson, so there is more extrapolation involved.

In the Roosevelt area (Section A), agreement is not very good for the

shallower (2227 to 2230 m) depth. A value of Z n Q (or a higher value for UD

geothermal gradient) higher than 1980 m would be needed for the model to

calculate the same maturity suggested by oil composition. However, for the

deeper sample, 3019 to 3043 m, agreement is good for a Z of 1598 m. UD

Migration may play a role at the shallower depths in this area.

In sections B, C, and D of Table 5 calculated maturity levels are con­

sistently lower than maturity levels estimated from the published oil

composition. At Pariette Bench, the model predictions using the (probably

high) value of 2673 m for Z n D are still too low (9 to 12%) compared with 20 UD

to 40% for the sampled oil. This suggests that the oil in place is migrated

oil, as has also been suggested by Pitman et al. (1982) and Anders and Gerrild

(1984). Similarly, the highest values of Z n o do not produce high enough UD

maturity levels in the Duchesne and Redwash areas, and thus the oil there also

is probably migrated oil.

For comparison, we applied the method of petroleum formation analysis

described by Waples (1980) to the burial history curves for the Shell

Brotherson (Figure 11) and Pariette Bench (Table 2) wells. The

time-temperature index (TTI) for the Shell Brotherson well we calculated is

57.8 for the 2557-m level and 101.1 for the 2782-m level. These indexes

- 29 -

correspond roughly to maturity levels of 45 and 55%, respectively. These

estimates are slightly high when they are compared with the data of Tissot et

al. in Table 4. We calculate a TTI of 211.3 (100% maturity) for the base of

the Eocene (2250 m) in the Pariette Bench well using the Waples method—this

is much higher than levels indicated in Table 5. These differences could be

due to excessively high values of Z n D but they more likely indicate that the Ub

Waples method overestimates the maturity level for this type of kerogen.

Higher values for the geothermal gradient would produce higher calculated

maturity levels. In the Redwash area (Table 5, Section B) at the 1676-m level

with an assumed value for Z n Q of 1800 m, a geothermal gradient of 33°C/km UB

(an increase of 8° C/km over the value used in the model) would produce a

T of 124°C. The corresponding maturity level calculated with our model is

about 30% (from Figure 18) and would agree well with measured values of 20 to

40%. However, the lower values of geothermal gradient work best for the Shell

Brotherson wells in the deepest part of the basin where migration is less

likely. There is no geologic evidence (such as magma intrusion at depth) for

large changes in geothermal gradient over such a small area, and it is

unlikely that the magnitude of change required could be accounted for by

hydrologic circulation. Furthermore, the nature of fracturing in the basin

(Narr and Currie, 1982, Anders and Gerrild, 1984) is conducive to migration of

hydrocarbons. Thus, we conclude that migration, rather than large changes in

Z n Q or geothermal gradient, is the most likely explanation for the

difference between modeled predictions and measurements of maturity levels

indicated in Tables 4 and 5.

- 30 -

The data of Table 3 can be used to make another type of calibration curve.

In each model, the percent of oil ungenerated, percent of liquid oil created,

and the atomic H/C ratio are calculated. From this, we can determine the

transformation ratio, or production index [S /(S + S )]. In the

Rock-Eval pyrolysis method, S corresponds to the free hydrocarbons that are

released between 90 and 300°C in flowing helium. Hydrocarbons generated during

kerogen pyrolysis between 300 and 600°C correspond to S„. In Table 3, the

liquid oil produced is equivalent to S., and the oil ungenerated is

equivalent to S . (This neglects the contribution of gas to S , so the

calculated transformation indices are probably about 10% too high.)

Transformation ratios for various depths in the wells in the Altamont-Bluebell

field have been determined by Anders and Gerrild (1984). Figure 18, which

compares the amount of kerogen converted with the transformation and H/C

ratios for a range of T , was prepared from Table 3 and is the basis for max

comparison of our geologic-geochemical models with published data. In each

case, we assumed that the value of lnn (which determines T ) is the OB max'

determining variable for hydrocarbon maturation, and that small changes in

heating rate (due to different burial histories and different values of

T ) are negligible. max' a

Anders and Gerrild (1984) sampled five wells from the Uinta Basin and

compared various maturity indicators with total organic carbon (TOC) determi­

nations and parameters such as stratigraphic facies. In Table 6 we compare our

determinations of the transformation ratio (from values of Z n D with a OB

25°C/km geothermal gradient and using Figure 18) with those determined from

sampled material by Anders and Gerrild.

- 31 -

Our predictions show good agreement with real data for the Dustin #1 well

(compared using a Z Q B of 1868 m determined for the Ute Tribal B-7 well) for

depths shallower than 2926 m. Our model predicts higher values of trans­

formation ratio for greater depths. It is impossible for maturity to decrease

with increased depth. This suggests that most of the oil at 3353 and 4054 m

must have either migrated or cracked. However, our model calculations show

that cracking of oil to gas is negligible here, so the discrepancy must be

caused by migration. Results are similar for the Daniel Uresk well (compared

using a Z of 2225 m). Hydrocarbon samples from the Wosco Ex-1 and CEDAR UD

RIM #3 wells are very immature, in complete agreement with our modeling. Our

model of these depths of the Ute Tribal 1-16 well predicts maturity levels

that are low for Z n D = 1870 m (Ute Tribal E-l) and high, but close, for UD

Z = 2252 m (Gulf Duchesne). A value of about 2100 m of overburden removed UD

(assuming the geothermal gradient used is correct) gives an optimum match of

maturity for the area of the Ute Tribal 1-16 well.

SUMMARY AND CONCLUSIONS

Using geophysical log data and present-day values of the geothermal

gradient, we have developed a model of the thermal history of stratigraphic

marker horizons for selected wells in the Uinta Basin. We incorporated

thermal-history data into a pyrolysis model to predict the amounts of hydro­

carbons produced and indicators of their maturity level as the kerogen evolved

thermally. For a geothermal gradient that is spatially and temporally con­

stant, the controlling variable in the thermal history model, in this case, is

the amount of overburden removed by erosion.

- 32 -

Predicted maturity levels of the evolved hydrocarbons in the basin agree

well with measured maturity levels. We feel that this agreement gives us con­

fidence in the general applicability of a laboratory-based pyrolysis model to

geological processes. These results are consistent with the assumption that

the source rock in the Uinta Basin is the Green River oil shale. However, we

cannot rule out the possibility that oil generated from kerogen in underlying

sediments such as the North Horn Formation has also contributed to oil

production in the basin. Although the present model is valid only for Type I

kerogen, it seems certain that similar models could be developed for Type II

and Type III kerogens. We emphasize the following conclusions:

The value of activation energy (52.A kcal/mole) used in the

pyrolysis model results in good predictions of amounts and

compositions of evolved hydrocarbons in the deepest part of the

basin.

Discrepancies between predicted and measured maturity levels in the

Redwash and Pariette Bench areas are probably related to migration.

The estimated values for removed overburden of approximately 1800 m

in the Shell Brotherson 1-11B4 well area result in very good

agreement between predicted and measured maturity levels, and this

value of 2nD is constrained by about ± 150 m by the model UD

(assuming a constant geothermal gradient).

- 33 -

If reasonable values for overburden removed are limited to being

less than 21t>0 m and greater than 900 m, the geothermal gradient

would have had to be about 10°C/km (40%) greater in the Duchesne

River, Redwash, and Pariette Bench areas to account for the

difference be- tween predicted and measured maturity levels with no

migration of hydrocarbons.

In the process of this investigation we feel that we have gained much in­

sight into the natural kerogen conversion process as well as the nuances of

the pyrolysis model. By incorporating the ability to vary pressure in the

pyrolysis model (in future work) and adding calculations of pore volumes and

pressures, the true potential for geological application of the model can be

realized. Utilization of a refined pyrolysis model with more detailed geo­

logic data and models would enable us to study the phenomena of migration and

overpressuring in detail. Refinement of these techniques and study of other

geologic basins could eventually lead to the use of the pyrolysis model in

quantitative evaluations of the thermal history of sedimentary basins and the

associated hydrocarbon resource potential.

_ 34 _

REFERENCES

Anders, D. E., and P. M. Gerrild, 1984, Hydrocarbon generation in lucastrine

rocks of Tertiary age, Uinta Basin, Utah—organic carbon, pyrolysis yield,

and light hydrocarbons, in J. Woodward, F. F. Meissner, and 0. L. Clayton,

eds., Hydrocarbon Source Rocks of the Greater Rocky Mountain Region, Rocky

Mountain Association of Geologists, Denver, pp. 513-529.

Andersen, D. W., and M. D. Picard, 1974, Evolution of Synorogenic clastic

deposits in the intermontane Uinta Basin of Utah, in W. R. Dickinson, ed.,

Tectonics and Sedimentation, SEPM Sp. Publ. 22, pp. 167-189.

Anthony, D. B., and 3. B. Howard, 1976, Coal devolatilization and

Hydrogasification: American Institute of Chemical Engineers Journal, V.

22, p. 625.

Bissell, E. R., A. K. Burnham, and R. L. Braun, 1985, Shale oil cracking

kinetics and diagnostics, I&EC Process Design & Development, V. 24, p. 381.

Bond, G. C, and M. A. Kominz, 1984, Construction of tectonic subsidence

curves for the early Proterozoic Miogeocline, Southern Canadian Rocky

Mountains: implications for subsidence mechanisms, age of breakup, and

crustal thinning: G.S.A. Bull, V. 95, no. 2, pp. 155-173.

Bradley, W. H., 1925, A contribution to the origin of the Green River

Formation and its oil shale: AAPG Bull, V. 9, pp. 247-262.

- 35 -

Bradley, W. H., 1929, The varves and climate of the Green River epoch: USGS

Prof. Paper 158-E, pp. 87-110.

Bradley, W. H., 1931, Origin and microfossils of the oil shales of the Green

River Formation of Colorado and Utah: USGS Prof. Paper 168, p. 58.

Bradley, W. H., 1970, Green River oil shale—concept of origin extended:

G.S.A. Bull, V. 81, pp. 985-1000.

Burnham, A. K., 1981, Chemistry of shale oil cracking, in H. C. Stauffer, ed.,

Oil Shale, Tar Sands and Related Materials: American Chemical Society

Symposium Series 163, Washington, DC, p. 39.

Burnham, A. K., and R. L. Braun, 1985, General kinetic model of oil shale

pyrolysis: In Situ, V. 9, no. 1, pp. 1-23.

Burnham, A. K., and M. F. Singleton, 1983, High-pressure pyrolysis of Green

River oil shale, in F. P. Miknis, ed., Chemistry and Geochemistry of Oil

Shale, ACS Symposium Series 230, American Chemical Society, Washington,

DC, p. 335.

Burnham, A. K., J. E. Clarkson, M. F. Singleton, C. M. Wong, and R. W.

Crawford, 1982, Biological markers from Green River kerogen decomposition:

Geochemica et Cosmochimica Acta, V. 46, pp. 1243-1251.

- 36 -

PampbeU, J. H., G. Gallegos, and M. Gregg, 1980, Gas evolution during oil

shaj.e pyrolysis. 2. Kinetic and stoichiometric analysis: Fuel, V. 59,

pp. 727-732.

Campbell, J. H., G. 3. Koskinas, and N. D. Stout, 1978a, Kinetics of oil

generation from Colorado oil shale: Fuel, V. 57, no. 6, pp. 372-376.

Campbell, J. H., G. 3. Koskinas, N. D. Stout, and T. T. Coburn, 1978b, Oil

shale retorting: effects of particle size and heating rate on oil

evolution and intra-particle oil degradation: In Situ, V. 2, no. 1, pp.

1-47.

Carmichael, R. S., ed., 1982, Handbook of Physical Properties of Rocks,

Volume II, CRC Press, inc., Boca Raton, FL, 345 pp.

Chapman, D. S., T. H. Keho, M. S. Bauer, and M. D. Picard, 1984, Heat flow in

the Uinta Basin determined from bottom hole temperature (BHT) data:

Geophysics, V. 49, no. 4, pp. 455-466.

Chatfield, 3., 1972, Case history of Redwash Field, Uinta County, Utah, in R.

E. King, ed., Stratigraphic Oil and Gas Fields, AAPG Memoir No. 16, pp.

342-353.

Coburn, T. T., and 3. H. Campbell, 1977, Oil Shale Retorting: Part II,

Variation In Product Oil Chemistry During Retorting of an Oil Shale Block,

Lawrence Livermore National Laboratory, Livermore, CA, Report UCRL-52256,

pt 2.

- 37 -

Connan, J., 1974, Time-temperature relation in oil genesis: AAPG Bull, V.

58, no. 12, pp. 2516-2521.

Fabuss, B. M., J. 0. Smith, and C. N. Satterfield, 1964, Thermal cracking of

pure saturated hydrocarbons: Advances in Petroleum Geochemistry and

Refining, V. 9, no. 1, p..157.

Fouch, T. D., 1975, Lithofacies and related hydrocarbon accumulations in

Tertiary strata of the western and central Uinta Basin, in D. W. Boyard,

ed., Symposium on Deep Drilling Frontiers in the Central Rocky Mountains,

Rocky Mountain Association of Geologists Special Publication, pp. 163-173.

Fouch, T. D., 1981, Distribution of rock types, lithologic groups, and

interpreted depositional environments for some lower Tertiary and upper

Cretaceous rocks from outcrops at Willon Creek-Indian Canyon through the

subsurface of Duchesne and Altamont oil fields, southwest to north central

parts of the Uinta Basin, Utah, USGS Oil and Gas investigation map, Chart

OC-81.

Fouch, T. D., and W. B. Cashion, 1979, Preliminary chart showing distribution

of rock types, lithologic groups and depositional environments for some

lower Tertiary, upper and lower Cretaceous and upper and middle Jurassic

rocks in the subsurface between Altamont oil field and San Arroyo gas

field, north central to southeastern Uinta Basin, Utah, USGS open-file

report 79-365, 2 sheets.

Hansen, W. R., 1984, Post-Laramide tectonic history of the eastern Uinta

Mountains, Utah, Colorado, and Wyoming: The Mountain Geologist, V. 21,

no. 1, pp. 5-29.

- 38 -

Horsfield, B., 1984, Pyrolysis studies and petroleum exploration, in 3. Brooks

and D. Welte, eds., Advances in Petroleum Geochemistry, Volume 1, Academic

Press, pp. 247-298.

Hunt, J. M.f 1979, Petroleum Geochemistry and Geology, W. H. Freeman and Co.,

San Francisco, CA, 617 pp.

Ishiwatari, R., M. Ishiwatari, I. R. Kaplan, and B. G. Rohrback, 1976, Thermal

alteration of young kerogen in relation to petroleum genesis: Nature, V.

264, pp. 347-349.

Lawler, D. L., and W. E. Robinson, 1967, Fatty acids in n-alkanes in Green

River oil shale: changes in depth: American Chemical Society Division of

Fuel Chemistry Preprints, V. 2, no. 5, part 2, pp. 480-486.

Lewan, M. D., 3. C. Winters, and 3. H. McDonald, 1979, Generation of oil-like

pyrolysates from organic-rich shales: Science, V. 103, pp. 897-899.

Lucas, P. T., and 3. M. Drexler, 1975, Altamont-Bluebell: a major fractured

and overpressured stratigraphic trap, Uinta Basin, Utah, in D. W. Bolyard,

ed., Deep Drilling Frontiers of the Central Rocky Mountains, Rocky

Mountain Association of Geologists Symposium, pp. 265-273.

Magara, K., 1978, Compaction and Fluid Migration—Practical Petroleum Geology,

Flsevier, New York, pp. 11-25

- 39 -

Mauger, R. L., 1977, K-Ar ages of biotites from tuffs in Eocene rocks of the

Green River, Washakie, and Uinta Basins, Utah, Wyoming, and Colorado:

Contributions to Geology, University of Wyoming, V. 15, no. 1, pp. 17-41.

Meissner, F., 1982, Basin Geochemistry--Case Histories, notes from AAPG

Geochemistry for Geologists School, Houston, TX, October 4-6, 1982.

Montgomery, J. A., and D. Chandler, 1979, Trajectory analysis of a kinetic

theory for isomerization dynamics in condensed phases: Journal of

Chemical Physics, V. 70, p. 4056.

Narr, W., and J. B. Currie, 1982, Origin of fracture porosity—example from

Altamont Field, Utah, AAPG Bull, V. 66, no. 9, pp. 1231-1247.

Palmer, A. R., 1983, The decade of North American geology 1983 geologic time

scale: Geology, V. 11, pp. 503-504.

Philippi, G. T., 1965, On the depth, time and mechanisms of petroleum

generation: Geochemica et Gosmochimica Acta, V. 25, pp. 1021-1049.

Pitman, J. K., T. D. Fouch, and M. B. Goldhaber, 1982, Depositional setting

and diagenetic evolution of some Tertiary unconventional reservoir rocks,

Uinta Basin, Utah: AAPG Bull, V. 66, no. 10, pp. 1581-1596.

Reed, W. E. and W. Henderson, 1971, Proposed stratigraphic controls on the

composition of crude oils reservoired in the Green River Formation, Uinta

Basin, Utah: Advances in Geochemistry, Pergamon Press, Oxford, pp.

499-515.

- 40 -

Robin, P. L., and P. G. Rouxhel, 1978, Characterization of kerogens and study

of their evolution by infrared spectroscopy: carbonyl and carboxyl

groups: Geochemica et Cosmochimica Acta, V. 42, pp. 1341-1349.

Robinson, W. E., 1969. Kerogen of the Green River Formation, in G. Eglington

and M. T. J. Murphy eds., Organic Geochemistry, Springer-Verlag, New York,

pp. 619-637.

Ryder, R. T., T. D. Fouch, J. H. Elison, 1976, Early Tertiary sedimentation in

the Western Uinta Basin, Utah: GSA Bull., V. 87, pp. 496-512.

Saxby, J. D.f K. W. Riley, 1984, Petroleum generation by laboratory-scale

pyrolysis over six years simulating conditions in a subsiding basin:

Nature, V. 308, March 8, pp. 177-179.

Sclater, J. G., P. A. F. Christie, 1980, Continental stretching: an

explanation of the post mid-Cretaceous subsidence of the Central North Sea

Basin: Journal of Geophysical Research, V. 85, no. 87, pp. 3711-3739.

Shih, S. M., and H. Y Sohn, 1980, Nonisothermal determination of the intrinsic

kinetics of oil generation from oil shale: Industrial Engineering

Chemical Processes Design and Development, V. 19, p. 420.

Skinner, J. K., and P. G. Wolynes, 1980, General kinetic models of activated

processes in condensed phases: Journal of Chemical Physics, V. 22, p.

4913.

- 41 -

Smith, J. W., 1974, Geochemistry of oil-shale genesis in Colorado's Piceance

Creek Basin: Rocky Mountain Association of Geologists Guidebook, pp.

71-79.

Spencer, C. W., 1987, Hydrocarbon Generation as a Mechanism for Over-

pressuring in the Rocky Mountain Region, AAPG Bull., (in press).

Steckler, M. S. and A. B. Watts, 1978, Subsidence of the Atlantic-type

continental margin off New York: Earth and Planetary Science Letters, V.

41, pp. 1-13.

Stokes, W. L., 1973, Geologic Map of Utah, College of Mines and Mineral

Industries, University of Utah, 2 sheets.

Stout, N. D., G. J. Koskinas, J. Raley, S. D. Santor, R. J. Opila, and A. J.

Rothman, 1976, Pyrolysis of oil shale: the effects of thermal history on

oil yield, in Proceedings 9th Oil Shale Symposium, Colorado School of

Mines, Golden, Colorado.

Tissot, B., G. Deroo, and A. Hood, 1978, Geochemical study of the Uinta

Basin: formation of petroleum from the Green River Formation: Geochemica

et Cosmochimica Acta, V. 42, pp. 1469-1485.

Tissot, B. P. and D. H. Welte, 1978, Petroleum Formation and Occurrence,

Springer-Verlag, New York, 538 pp.

- 42 -

Tissot, B. and J. Espitalie, 1965, L'evolution thermique de la matiere

organique des sediments; applications d'une simulation mathematique

potential petrolier des bassins sedimentaires et reconstitution de

l'histore thermique des sediments: Reveue de l'Institut Francais de

Petrole, V. 30, pp. 743-777.

Van Heek, K. H., and H. Juntgen, 1968, Determination of reaction-kinetic

parameters from nonisothermal measurements: Ber. Bensenges. Physik.

Chem., V. 72, p. 1223; available as English Translation in Lawrence

Livermore National Laboratory Report UCRL-TRANS-10974 (1975).

Voge, H. H., and G. M. Good, 1949, Thermal cracking and high paraffins:

Journal of American Chemical Society, V. 71, p. 593.

Waples, D. W., 1980, Time and temperature in petroleum formation-application

of Lopatin's method to petroleum exploration: AAPG Bull, V. 64, pp.

916-926.

Waples, D. W., 1984, Thermal models for oil generation, in J. Brooks and D.

Welte, eds., Advances in Petroleum Geochemistry, Volume 1, Academic Press,

pp. 7-68.

Yen, T. G. and G. V. Chilingarian, eds., 1976, Oil Shale, Elsevier, Amsterdam.

Yukler, M. A., and F. Kokesh, 1984, A review of models used in petroleum

resource estimation and organic geochemistry, in J. Brooks and D. Welte,

eds., Advances in Petroleum Geochemistry, Volume 1, Academic Press, pp.

69-114.

- 43 -

Table 1. Chemical Reactions and Rate Expressions that Define the Model.

For a Detailed Explanation of the Model and Parameters See

Burnham and Braun (1985).

Reaction Rate Expression1

Kerogen Pyrolysls and Bitumen

100CH1.50N0.025°0.05 "* 5-3CH1.56N0.021°0.01

+ 74.2CH1>56N0 0 2 L 0 n > 0 1

+ 14-7CH0.63N0.056°0.02

+ 0.3C0 + 1.0H20

+ 0.6CH,

+ 3.6CH

Initial oil (bitumen)

5% by kb= 7xl013e-26390/T

95% by k = 2.8xl013e-26390/T

o

k = above o

. Q in12 -26390/T k = 9x10 e x

+ 1.0H,

+ 1.3C0,

kh= 5 x l 0 1 2 e - 2 6 3 9 0 / T

h

i by k1= l x l 0 ^ e - (22000±2200) /T

| by k2= 7x l013e-(26390±1800)/T

O i l Coking

100CH0.99N0.038°0.02 "* 95CH0.63N0.0«°0.02

+ 3CHA+ 11H2+ 2CHx

Oil Cracking

Oil. -> Oil,. + char + gases

Secondary Char Pyrolysis

100CH0.63N0.05°0.02 - 9*-5CH0.23N0.03°0.02

Tertiary Char Pyrolysis

+ 5.5CH^+ 6.4H2+ (2.2NH3)

k,-3 . 2 x l 0 1 0 e - 1 7 6 2 0 / T

(1 + 2xlO~V, ) H2

100CH0.23N0.03°0.02 " 100CH~0.1N0.03°0.02+ 8-°H2

. n „ in10 -26390/T k .= A.'2.11x10 e c,i 1

k5= 3.5xloV(25210±2285>/T

kt, 3.lxl013e-(39000±4090)/T

Dolomite Decomposition

MgCa(C03)2 -» MgO + CaC03+ C02 - _ in10 -29090/T k .= 2.5x10 e

d

First-order reaction unless otherwise noted. Activation energies with ± values use distributed activation energy theory. All rate constants in s"1 and pressures in pascals.

- 44 -

Table 2. Calculated Values of ZQB in Meters for Well Locations in the Uinta Basin. A Range of Values is Calculated for a Range of One Standard Deviation of the Slope Parameter, b.

Well and Location A f 0 (us/ f t )

Using b Range

1552-2150

1711-2370

1614-2336

1939-2687

2302-3189

1376-1906

1547-2142

1608-2228

1486-2058

1579-2188

1700-2355

1617-2241

1519-1681

Values Determined by Narr & Currie

(1982)

—

1733

510

-

-

2850 1338 756

1968

-

-

339 583

-

-

Energy Res. Gp. Broadhurst #13, 7S22E 113.41

Shell Christensen 1-33A5, 1S5W 107.01

Shell Tenneco-Brotherson 1-3B4, 2S4W 110.86

Gulf Duchesne Co Unit 1, 3S4W 98.44

Davis Oil Pariette Bench #5, 9S18E 86.22

Gulf Verl Johnson #1, 1S2W 120.96

Shell Brotherson 1-11B4, 2S4W 113.63

Chevron Ute Tribal 6-7, 2S3W 111.09

Chorney Oil S. Redwash Fed. 1-23, 8S23E 116.18

Shell Miles #1, 1S4W 112.28

Gulf Ute Tribal 1-21, 1N2W 107.44

Texaco Ute Tribal E-l, 3S6W 110.73

Chevron Redwash 250, 7S24E 128.34

1802

1987

1875

2252

2673

1598

1796

1868

1726

1834

1974

1878

1410

- 45 -

Table 3. Calculated Compositional Fractions for Selected Depths in Particular Wells. Input Data Came from Table 2 and Thermal History Determinations Similar to Figures 11 and 12.

Depth (m)

1.Energy Res. Gp. 1433 Broadhurst #13

2.Energy Res. Gp. 1756 Broadhurst #13

3.Davis Oil 927 Pariette Bench #5

A.Shell Brotherson 2003 1-11B4

5.She11 Brotherson 2643 1-11B4

6.Energy Res. Gp. 2758 Broadhurst #13

7.Davis Oil 2246 Pariette Bench #5

8.Shell Brotherson 3164 1-11B4

9.Gulf Duchesne Co. 2747 Unit 1

10.Shell Christensen 3453 1-33A5

11.Shell Brotherson 4004 1-11B4

-̂Present-day burial depth

T 0 U

max H/C 0/C Ungen-(°C) erated

91 1.48 0.044 0.933

99 1.48 0.042 0.930

100 1.48 0.042 0.930

105 1.47 0.041 0.925

121 1.42 0.037 0.787

124 1.39 0.036 0.670

133 1.07 0.034 0.271

134 1.02 0.034 0.225

135 0.95 0.033 0.176

146 0.51 0.029 0.000

155 0.44 0.025 0.000

Oil Oil Liquid Coked Cracked Oil

0.022 0.00 0.045

0.023 0.00 0.048

0.023 0.00 0.047

0.024 0.00 0.051

0.066 0.00 0.147

0.101 0.00 0.229

0.220 0.00 0.506

0.234 0.003 0.538

0.249 0.004 0.571

0.302 0.024 0.674

0.302 0.075 0.623

- 46 -

Table 4. Comparison of Hydrocarbon Maturity of Wells Studied by Tlssot et al. (1978) with Model Calculations. Maturity Levels are Estimated from the Compositional Ratios Using Figures 4 and 5.

A.

B.

C.

well Name I

Shell Brotherson 1-14B4(2S,4W)

Shell Brotherson 1-23B4(2S,4W)

Shell Brotherson 1-23B4

Shell Brotherson 1-23B4

Shell Murdock (2S.5W)

Shell Murdock

Standard of Calif. Redwash 132

Redwash 164

Wells from

Depth1(m)

2298

2210

2557

2782

2982

3552

2649

1698

Tlssot et

Compi

Prlstane C17* C18

0.B4

1.26

0.58

0.18

0.23

0.09

0.32

1.46

al. (1978)

osltion Ratios

r Phytane 17 C17 + C1B C16 + C18

0.69 1.2

0.95 1.3

0.58 0.5

0.36 0.6

0.20 0.6

0.05 0.5

0.14 0.6

1.90 0.6

Estimated Maturity Level (% Oil Generated)

0-25

0-20

15-35

30-60

40

60

30-60

0-20

1

Model Well

A. Shell Brotherson 1-11B4 (2S.4W)

Shell Brotherson 1-11B4

Shell Brotherson 1-11B4

Shell Brotherson 1-11B4

Shell Brotherson 1-11B4

Shell Brotherson 1-11B4

B. Shell Brotherson 1-11B4

Shell Chrlstensen (1S.5W)

Shell Brotherson 1-11B4

Shell Chrlstensen

C. Chevron Redwash 250 (7S.24E)

Broadhurst #13 (7S.23E)

Chevron Redwash 250

Broadhurst #13

'todels from this Studv

Depth1(m)

2298

2210

2557

2782

2782

2782

2982

2982

3552

3552

2649

2649

1698

1698

*«<•>

1796

1796

1796

1796

2142

1547

1796

1987

1796

1987

1410

1803

1410

1803

for Comparison

T max (deg)

112

110

119

125

133

118

129

134

144

149

HI

121

88

98

Calculated H/C

1.40

1.40

1.36

1.28

0.99

1.36

1.14

0.93

0.52

0.46

1.40

1.35

1.45

1.44

Calculated Maturity Level (% Oil Generated)

11

9

20

36

73

19

56

78

99

100

10

24

6

7

Present-day depth In well.

Table 5. Comparison of Estimated Hydrocarbon Maturity In Nells of Reed and Henderson (1971) with Model Calculations. Maturity Levels are Estimated from the Compositional Ratios Using Figures 4 and 5.

A.

B.

C.

0.

Well Name

Roosevelt Field Area

Roosevelt

Roosevelt

Redwash Area (7S,22E

H. V. Stagecoach

N. V. Gypsum Hills

Parlette Bench Area

Parlette Bench River Junction

Duchesne Area (3S.5W

Duchesne Co.

Flat Mesa

Indian Ridge

Data of Reed and Henderson (1971)

Composition Rati

Present-Day Prlstane Phytane Depth(m) C17* C18 C17* C18

(IS,IE)

2227-2230

3019-3034

to 8S, 23E)

1547-1550

1608-1686

(9S. 18E)

1478-1500 1322-1325

to 6S, 4N)

1562-1593

2688-2774

2245-2476

0.64

0.30

0.35

0.25

0.24 0.34

0.47

0.08

0.03

0.43

0.27

0.27

0.24

0.26 0.26

0.29

0.04

0.01

.OS

C17 C16* C18

0.61

0.58

0.59

0.57

0.59 0.58

0.72

-0.53

0.52

Estimated Maturity Level (« Oil Generated)

15-25

30-40

20-35

25-50

20-40 20-40

15-25

80

80

Models

Present-day Model Nell Depth(m)

A. Roosevelt area

Gulf Verl Johnson (1S.2N)

Gulf Ute Tribal (1N.W) Gulf Verl Johnson (1S.2N)

Gulf Ute Tribal (IN,2*)

B. Redwash Area

Chevron Redwash 250 (7S.24E)

Broadhurst #13 (7S.22E)

Chevron Redwash 250 Broadhurst #13 Gulf Duchesne (3S.4N)

C. Parlette Bench Area (9S,

Parlette Bench Parlette Bench

2225

2225 3018

3018

1524

1524

1676 1676 1676

18E)

1494 1323

D. Duchesne Area (3S,5* to 4S.4W)

Texaco Ute Tribal (3S.6W)

Gulf Duchesne (3S.4W)

Texaco Ute Tribal Gulf Duchesne

Texaco Ute Tribal Gulf Duchesne

1585

1585

2688 2688

2377 2377

from this Study for

1598

1974 1598

1974

1410

1803

1410 1803 2252

2673 2673

1878

2252

1878 2252

1878 2252

106

115 125

135

83

93

87 97 108

114

no

97

106

124 134

116 126

Comparison

Calculated H/C

1.42

1.38 1.28

.87

1.46

1.44

1.46 1.44 1.41

1.39 1.40

1.44

1.42

1.30 0.93

1.38 1.26

Calculated Maturity Level (% Oil Generated)

8

14 37

82

0

6

0 7 8

12 9

7

8

33 78

15 42

Table 6. Comparison of Hydrocarbon Maturity of Wells Studied by Anders and Gerrild (1984) with Model Calculations.

Wells from Ander;

Well Name

Dustin #1(2S.3W)

Daniel Uresk C4S.1W)

WOSCO EX-H9S.20E

Cedar Rim #3 (3S.6W)

Ute Tribal 1-16 (4S.7W)

» and Gerrild (1984) Results from this Stud' Transformation Ratio Estimated from Figure

Depth Measured Model Depth T (m) SL/(S1+ S2) Well (m) (

m**}

2591 2701 2926 3353 4054

1524 2091 2377 2926 3374

) 610 853

1402 1951 2256

2134 2286 2377

0.13 0.19 0.70 0.61 0.61

0.07 0.17 0.43 0.12 0.07

0.09 0.01

0.05 0.07 0.08

0.16 0.17 0.18

Ute Tribal 6-7(2S,3W) ZQB=1868 m

Gulf Duchesne(3S,4W) Z0B=2252 m

Pariette Bench(9SF18E) Z0B=2673 m

Ute Tribal E-1(3S,6W) ZQB=1878 m

Ute Tribal E-1(3S,6W) Z 0B=1878 m

Gulf Duchesne (3S,4W) ZOB=2252 m

2591 2701 2926 3353 4054

1524 2091 2377

2926 3374

610 853

1402 1951 2256

2134 2286 2377

2134 2286 2377

121 124 130 141 158

104 119 126 139 151

92 98

92 106 113

110 114 116

120 123 126

1

17. Sl

(sx+ s2)

0.18 0.26 0.57 0.92 1.00

0.11 0.14 0.35 0.88 1.00

0.05 0.05

0.05 0.06 0.07

0.06 0.09 0.10

0.14 0.22 0.36

- 49 -

1

DIAGENESIS

CATAGENESIS

METAGENESIS

Organic Debris

f

Kerogen

1 Thermal Degradatia

1

n

Cracking

Carbon

Initial Bitumen

11

Oil & Gas

1

Methane

Oil Reservoir

,

Migration

Figure 1—General scheme for hydrocarbon maturation in a geological set­

ting. We define initial bitumen as soluble organic matter that is never in­

corporated into kerogen. Both kerogen and initial bitumen generate oil and

gas. High-molecular-weight intermediates between kerogen and oil are possi­

ble, but they are not explicitly distinguished from oil in this picture.

- 50 -

u o

E O > >• (0 +•>

15 < c CO

a. c o o

•o O

10-10oC/hr

0.5 X 10 9 °C/hr

100 120 140 160

Temperature °C

10'8 °C/hr

180 200

Figure 2—Oil generation rate curves for the conversion of Type I kero-

gen using the results of Campbell et al. (1978). Tp is the temperature cor­

responding to the maximum rate of oil generation. Slower heating rates lead to

lower values of T .

- 51 -

1.0 (a)

ID

tc C o

£0.5

"S N

"5 E k. o z

— with oil cracking • — without oil cracking

Gas from kerogen pyrolysis and oil coking

Gas from S

s "Char" p y r o l y s i s \ V * f c — - ^ - '

-15 -14

Figure 3a~Rates of oil (10 kg/kg-s) and gas (10 mole/kg-s)

_9

generation at a heating rate of 1.2 x 10 °C/h. The amount of gas pro­

duced from oil cracking depends on how much oil does not migrate.

- 52

7.0

c o

c 0)

O 0)

> 3 0.5 E o 0) N

r—r (b)

with oil cracking without oil cracking

I Initial bitumen •

_L

Gas H _L JL -L

700 120 740 760

Temperature, °C

780 200 JUJ 220

Figure 3b--Cumulative generation of oil and gas using the same param­

eters as Figure 3a.

- 53 -

0.6

. 0.4

0.2

1 1

-

--

-

-

-

i i

T 1 1 | 1 1 1 1 r i r-

Wilh oil cracking-^^^ U

-•"tHJOO4-^ Without oil cracking^

i t

R

H A \ -

•

-

[ L

' ' 70 90 110 130 150 170 190 210 230

Temperature, C

Figure 4—Changes in the ratio (C_- C.)/(C.- C ) with tempera­

ture for the cases with and without oil cracking.

- 54 -

0 20 40 60 80 100

% of Oil Produced

Figure ^--Maturity indicators, in the form of isoprenoid/normal alkane

composition ratios, calibrated to percent oil generated.

- 55 -

1.4

1.2

1.0

- 0.8

/ 0.6

0.4

0.2

-

- u

\

-

C

/ Se C IB

r0--^^ -0—s^=ft=

c,6+c,.

i i

—n— n

• i

i i

-

-

.

-

-a -a^ ° -

• i

40 60

% of Oil Produced

Figure 6--Calibration of the normal alkane ratio 0,^/(0, . + C 1 0) 1 / 16 lo

with percent oil generated.

- 56 -

Figure 7--Map of the Uinta Basin showing locations of the Altamont-

Bluebell and Redwash oil fields. Line A-A1shows the location of the cross-

section of Figure 8.

- 57 -

sw ALTAMONT FIELD

0 5 10 15 20 kilometers

Figure 8—Cross section from south to north arms* thP central part nf the

1

1

0

2000

4000

6000

8000

10.000

12.000 -

10

Shell-Tenneco-Brotherson •t al Unit No 1-3B4 2S4W

Duchesne Formation

Uinla Formation

Graen River Formation "

Upper Marker /

Mahogany • " / Shale

Middle Marker

Top ^. Carbonate

Eocene

Paleocene

20 40

Al in

1 1 '

K>V

jfc

f "& o

fi1

ct

K H

1 . 1 60 B0

lii/fl

1 100

1

1

200

-

-

Figure 9—Interval transit time-depth plot for the Shell lennaco-

Brotherson 1-3B4 well in the Altamont-Bluebell field. Stratigraphic location

markers are for present-day locations in the well.

- 59 -

R6W

TIN

T1S

T2S

T3S

T4S

I

I R5W I R4W I R3W ' R2W I R1W \ '_

1987 1 8 3 4 A'«a™>nt

•o- <> O 1875

^1796

<>2252

__ Duchesne fc23

I I

Bluebell Field | <> 1974

1 Field

<>1598

<h1868

a UTAH

1

-MAP AREA

1

1 |

i

Du

ches

ne

C

ou

nty

Uin

ta

Co

un

ty

1

Figure 10—Locations of wells in the Altamont-Bluebell field area used in

this study (see also Figure 7). The calculated amount of removed overburden

(ZQB) in meters (from Table 2) is listed next to each well.

- 60 -

Figure 11—Burial history and temperature curve, corrected for compac­

tion, for the Shell Brotherson 1-11B4 well.

- 61 -

Figure 12—Burial history and temperature curves for the Green River

Formation at the Chorney Oil South Redwash well in Redwash oil field.

- 62 -

Depth

3000

12.000

Oil Generation Rate - Arbitrary Units

- | 1 1 1 1 1 1 1 1 r—

Duchesne Formation

- — T 69

Shell Brotherson 1 - 1 1 B 4 « M I I

Uinta Formation

j —— Oil show * —— Oil show

\ Mahogany Shale T m M - 116^

Oil general ion curve

—— Oil ihow — Maximum oil recovery

- O i l .

- Base of Eocene T _ „ * 149c

y Jk—DST, •"•«- "hydrostatic" *? mud prenurecurve

N— Zone of overprenuring

0.7 0.9 0.11 0.13 0.15

Fluid Pressure Gradient psi/ft

Figure 13—An o i l generation curve for Shel l Brotherson 1-11B4 wel l

( locat ion 2S, 4W) superimposed on the l i t ho l og i c log showing o i l production

depths and pore f l u i d pressure data. The o i l generation curve is based on

heating rates determined from Figure 11. Data from Meissner (1982) and Fouch

(1981).

- 63 -

Saturated HC/TOC (mg/gl

0 20 40 60 80 100 120 140 160 ISO 200

3000

4000'

T " ~T T " T " " T "T" T "

Altamont oil field Uinta Basin

Calculated Oil Generation

Curve

J L 40 60

Percent Oil Generated

80

Figure 14—Milligrams of saturated hydrocarbons per gram T0C of samples

recovered from Altamont field, plotted versus maximum depth of burial—

assuming 1800 m of erosion (after Tissot et al., 1978). Superimposed on the

plot is the cummulative percent of oil generated (assuming no cracking or

migration), calculated using our kinetic and thermal history models.

- 64 -

*

1.0

0.8 £ re a.

« 0.6

O 0.4

L

T—r~\—i r—r

0 I I I I—-T ' ' L_l I I I I I I I I I I l_

300 320 340 360 380 400 420 440 460 480 500

Temperature, °C

1.0

£ 0.8

S 0.6

o s 04

Z 0.2

50 70 90 110 130 151) 170 190 21C 230 250

Temperature. ' C

Figure 15—Normalized rate of oil generation calculated using the kinetic

parameters of Tissot and tspitalie (1975) in Eq. (2b). The multiple peaks come

from their assumption of six parallel oil-generating reactions.

- 65 -

0 0.05 0.10 0.15 0.20

Atomic Ratio O/C

0.25 0.30

Approximate iso-values of vitrinite reflectance

— — Boundaries of the field of kerogen

-« Evolution paths of principal kerogen types

Figure 16—Van Krevelen diagram showing calculated changes in elemental

composition of kerogen from the wells modeled and listed in Table 3. The

iso-reflectance lines are not calculated values (taken from Tissot and Welte,

1978) and are only intended for reference.

- 66 -

fe

Shell Brotherson 3164 m

Broadhurst #13 2758 m

Broadhurst #13 1433 m

100 200 300 400

Distillation Temperature, °C 500 600+

Figure 17—Boiling point distributions calculated by the model for

selected wells listed in Table 3.

- 67 -

100 -

80 -•o 0)

« o 60 u c 8> o I 40 5?

2 0 -

80

-

-

-

-

—

•

-

-

.

I I I

- D - C D - - D - -H/C ratio-/

% Kerogen converted -N

I \ I

I I I I

^4j &^

1 On J\ / ^

/ \ r/ V.

^ > ^ A S1/(S1+S2).

I I I 1

1.0

0.8

0.6

0.4

Q2

0

CM

C/5 _

+ eo~

co"

-

-

1 2.0

| 1.8

•

1.6 1

1.4 1

1.2 1

1.0 1

0.8 1

0.6 1

0.4 1

0.2 1

ol 100 120 140

Temperature,0 C 160

Figure 18--Calibration of transformation ratio [Si/CSi + s2)l

a n d

H/C ratio with the percent of kerogen converted and maximum temperature,

T , attained by the kerogen. Data points are from the model runs of max

Table 3. The arrows show the direction of increasing maturity for each curve.

- 68 -