Introduction
Compositional kinetic and yield analysis describes the rate at
which organic matter decomposes into oil
and gas as well as the yields of
these compounds or compound-classes at various levels of conversion
or maturity. It is needed to determine the yields and timing of
generation of:
Bulk kinetic modeling, i.e., decomposition of kerogen into hydrocarbons without description of products formed, has supplanted empirically-based TTI modeling as the latter does not take into account the variability in kerogen decomposition rates. However, while improving our estimates of the timing of hydrocarbon generation, bulk kinetic data does not provide any description of the products formed nor the timing of generation of the hydrocarbon moities listed above. Furthermore, bulk kinetics constrain all data to a single set of kinetic parameters, which limits its capability to predict early oil generation or the broad dry gas (methane) generation window.
Compositional kinetic data, which measures the rate of kerogen decomposition into specific chemical moieties, provides detailed kinetic parameters and hydrocarbon yields for each of these fractions including dry and wet gas as well as light and normal oil.
Further, these data show that the gas window varies among and within kerogen types, as do the condensate and oil windows. In addition these data explain the formation of early oil, i.e., the oil present in low to early oil window maturity source rocks.
Development of a breakthrough in trapping capability by Humble Instruments has enabled the precise evaluation of C1 through C40+ hydrocarbons using the MACT10 instrument.
Experimental (see Appendix for description of instrument and methodology)
Background
Kerogen, the organic matter that forms oil and gas under increasing
thermal stress, has two reactive components (Cooles et al., 1985). There is oil and gas prone
organic matter as well as an inert carbon residue (Figure 1). While most bulk kinetic and
basin modeling programs only account for gas formation from oil
cracking, there is also significant gas formed directly from kerogen
depending on kerogen composition.
Scheme of Petroleum Generation, Cracking and Expulsion
Modified from Pepper and Corvi, 1995
Figure 1. Diagrammatic illustration of kerogen composition, which leads to oil and gas formation directly from kerogen cracking and gas from oil cracking.
HISI Analytic Technologies' MACT10 was developed for assessment of the oil and the gas portions of kerogen (see Appendix for analytical details). MACT10 data is used to evaluate in chemical detail the products formed from kerogen in terms of both oil and gas formation. Of course, oil cracking must also be accounted for in assessing total gas yields.
The resulting data can be used to assess the yield of various compound classes (Figure 2).
|
|
Figure 2. Compositional yields of resolved hydrocarbons from MACT10 experimental data. |
These yield data for this lacustrine (Type I) sample are not very surprising demonstrating that a high yield of oil is expected. However, for other samples the yield results have sometimes been quite surprising. For example, the Type II kerogen shown in Figure 3 yields over 53% gas!
|
|
Figure 3. A Type II kerogen, which is about 20-30% converted, has a very high gas yield. This explains overpressuring in this formation and inherent "sealing" of the source rock inhibiting expulsion. Without oil expulsion, oil cracking will proceed. |
Timing of Generation of Oil and Gas
Hydrocarbon generation rates are typically calculated from bulk
kerogen kinetics modeled at a specific heating rate. For example,
the timing of hydrocarbon generation can be illustrated by modeling
the generation rate at 3.3° C/my (a reasonable worldwide average
rate) from bulk kinetic data (Figure 4).
Figure 4. Generation rate of "hydrocarbons" from bulk kinetic data at an arbitrary geological heating rate.
As is often the case, however, this organic matter has not experienced temperatures above about 100° C, yet a considerable amount of oil is present in the rock. For example, bulk kinetic data would predict less than 1% conversion at 100° C.
However, compositional kinetic data has a key feature, which does not unnecessarily constrain the data to fit a "bulk" or "average" model. When the same sample above is analyzed by MACT10 to derive compositional kinetic data, the generation rate curves for each fraction are quite different (Figure 5) ( et al., 1999).
Figure 5. Generation rate curves for 4 different hydrocarbon fractions based on MACT10 thermal slicing technology. Note the early oil generation based on the C15+ kinetics, and the broad gas (methane) generation window ( et al, 1999).
The C15+ kinetic data confirm observed results from other analytical techniques such as extraction of immature to early oil window maturity rocks, which yield a high C15+ fingerprint and contain higher amounts of resins and asphaltenes.
These data also show the significant temperatures for oil to gas cracking on the products generated from this kerogen.
How is this possible, when the bulk kinetic data do not show any products extending into these lower temperature ranges? All the products formed during bulk kinetic experiments can only be explained with a single Arrhenius factor and distribution of activation energies. Unfortunately, products formed during kerogen cracking may have vastly different Arrhenius factors that can only be fit from a different distribution of activation energies. Thus, the C15+ fraction is best fit by allowing the Arrhenius factor to vary. The pyrolysis data can be forced to fit underneath the bulk pyrolysis curve by one simple step – fixing the Arrhenius factor on all fractions. However, as the above figures demonstrate, fixing the Arrhenius factor will never allow accurate compositional modeling. Bulk kinetic data, however, is very useful for constraining compositional kinetic data to minimize costs of analysis.
Variation in the Position of the
Gas Window
The gas window is commonly described as reaching 10% of the total
product yield at a thermal maturity of approximately 1.1% Ro for all
kerogens. This does not appear to hold true based on compositional
kinetic data. Thus, the gas window varies as does the oil window
(Figure 6)
( et al., 1996).
Figure 6. Based on methane generation kinetics directly from kerogen, the onset of dry gas generation (10% of total methane yield) varies among and within various kerogen types.
Gas Wetness
These data may be used to evaluate gas wetness, i.e., the ratio of
dry gas to wet gas (C1 / (S C1-C4)). This is useful for assessment
of calorific value (BTU) of products, which has a direct relation
to market value, hence, company revenues.
At temperatures below approximately 140° C, the gas wetness ratio is largely due to gases derived from kerogen cracking (excluding bacterial alteration). Above 140° C, oil cracking will also yield gases.
Our preconceived notions suggest that wet gas is generated before dry gas. However, this is not always true. An example from an onshore Texas source rock suggests that gas is dry, then becomes progressively wetter, and subsequently, drier again with increasing maturity (Figure 7).
Figure 7. An unusual gas wetness profile demonstrated from an onshore Texas source rock.
While most source rocks follow the classical wet-to-dry gas pathway, these data show the need to carefully evaluate actual products formed, rather than using assumed results.
Analytical Services, and Source Rock
Studies Available from
Humble Geochemical Services
Compositional kinetic and yield analysis is an expensive process requiring up to 54 gas chromatographic analysis priced at $15,000.00. Our present approach uses 18 GC analyses plus bulk kinetic analyses and is priced at $7,500.00 per sample.
Data is available for purchase from our compositional consortium and Source Rock Geochemical Library members. Samples include key source rocks from around the world including:
The Source Rock Geochemical Library is a collection of source rocks from around the world and includes over 500 samples representing 135+ different source rocks. Over 200 of these samples have been characterized for kinetic parameters, TOC, Rock-Eval, extract fingerprints, and biomarkers.
For more information on compositional kinetics contact
Appendix
Experimental
Samples are analyzed at multiple heating rates on HISI Analytical
Technologies' Pyrolysis-MACT10. MACT stands
for Multiple Automatic Cryogenic Trapping. This concept is based on
a patent held by Chevron and reported by Tang and Stauffer (1994).
However, HISI Analytical Technologies' MACT10
system includes new technology to trap and resolve all hydrocarbon
gases (methane, ethane, propane, and butane), while reproducibly
releasing hydrocarbons from C5-C35+. The
technique is referred to as pyrolysis "thermal slicing"
gas chromatography, whereby the pyrolysis products from heating the
source rock at a given heating rate, e.g., 3° C/min, are trapped
over distinct temperature intervals. The first pyrolysis products,
i.e., the first products cracked and released from kerogen, are
quite different from those released at higher temperatures.
This is diagrammatically illustrated in Figure 1.
Figure 1. Diagrammatic illustration of pyrolysis-thermal slicing-gas chromatography. This technique allows a precise analysis of the hydrocarbons formed under increasing thermal stress and can be extrapolated to geological conditions to evaluate hydrocarbon generation by compound or compound-class, e.g., dry and wet gas, light and normal oil.
Actual data from the MACT10 is used to calculate the rate of formation of various hydrocarbons and their yields. We use MACT10 data from multiple nonisothermal pyrolysis experiments to calculate the kinetics and yields of:
The gas trapping capability of the MACT10 is required for detailed gas data assessment. During pyrolysis these gases must be held for hours in a trap prior to analysis. In addition higher molecular weight range hydrocarbons must be released in a highly reproducible manner. The resulting product is referred to as Gas Trapper technology (patent pending). Figure 2 demonstrates the very high gas chromatographic resolution of the MACT10 trapping and gas chromatography system.
Figure 2. Pyrolysis (400-424° C slice) of the Brown Shale, Sumatra, using a 3° C/min heating rate. Eight additional thermal slices were taken from this single pyrolysis run. When combined with other heating rates, the rate of decomposition of kerogen into these components can be completed using the Kinetics2000 program.
This technology enables the precise measurement of methane, ethane, propane, and butane gases as well as the liquid components of oil extending from C5-C35+.
Copyright © 1999-2000 Humble Instruments &
Services, Inc.. All rights reserved.
Revised:
