U.S. patent application number 11/833456 was filed with the patent office on 2008-03-06 for method of quantifying hydrocarbon formation and retention in a mother rock.
Invention is credited to Francoise Behar, Francois Lorant, Elodie Salmon.
Application Number | 20080059140 11/833456 |
Document ID | / |
Family ID | 37882141 |
Filed Date | 2008-03-06 |
United States Patent
Application |
20080059140 |
Kind Code |
A1 |
Salmon; Elodie ; et
al. |
March 6, 2008 |
METHOD OF QUANTIFYING HYDROCARBON FORMATION AND RETENTION IN A
MOTHER ROCK
Abstract
The method according to the invention allows the formation of
oil and the retention phenomenon in the mother rock to be modelled.
Organic matter characterization experiments are used to establish
the molecular model (MM) of the initial sample (E). The thermal
cracking reaction of this molecular model is reproduced by dynamic
molecular simulation computations with a reactive force field (RMD)
and validated by comparison with experimental data. The reaction
mechanism obtained (SR) allows to carry out a kinetic study (C) by
variation of the temperature parameter. The phase equilibria (PES)
of the reaction medium are determined at any time from dynamic
simulation. The successive phase equilibrium assessments at various
progress stages of the cracking reaction allow following the
physicochemical evolution (PC) of the thermal maturation of the
organic sample studied. The free hydrocarbons (liquid and gaseous)
that are not retained in the solid residue can be quantified
throughout numerical modelling of the sample maturation;
representing, in the sedimentary basins, the hydrocarbons that are
not retained in the organic matrix of the mother rock (Q). This
quantity can be used as an indicator or an input value for the
retention threshold in basin models.
Inventors: |
Salmon; Elodie; (Paris,
FR) ; Lorant; Francois; (Thiais, FR) ; Behar;
Francoise; (Paris, FR) |
Correspondence
Address: |
ANTONELLI, TERRY, STOUT & KRAUS, LLP
1300 NORTH SEVENTEENTH STREET
SUITE 1800
ARLINGTON
VA
22209-3873
US
|
Family ID: |
37882141 |
Appl. No.: |
11/833456 |
Filed: |
August 3, 2007 |
Current U.S.
Class: |
703/12 |
Current CPC
Class: |
G01V 99/00 20130101;
G16C 10/00 20190201; G01V 99/005 20130101; E21B 49/00 20130101;
G01N 33/241 20130101; G01V 2210/661 20130101 |
Class at
Publication: |
703/012 |
International
Class: |
G06G 7/58 20060101
G06G007/58 |
Foreign Application Data
Date |
Code |
Application Number |
Aug 4, 2006 |
FR |
06/07.148 |
Claims
1) A method of quantifying formation and retention of hydrocarbons
within a macromolecular chemical system, comprising construction of
a molecular model of the system by means of experimental
characterizations of a sample of the system, comprising: defining a
thermal maturation reaction mechanism for the macromolecular
chemical system by subjecting the molecular model to a molecular
dynamic simulation associated with a reactive force field; and
quantifying free hydrocarbons formed throughout thermal maturation
of the macromolecular system by: a) determining a physicochemical
evolution of a molecular mixture defined by the reaction mechanism,
by calculating phase equilibria, directly from the molecular
dynamic simulation; and b) determining thermal decomposition
kinetics of the macromolecular system by kinetic study from the
reaction mechanism.
2) A method as claimed in claim 1, wherein the molecular model and
the reaction mechanism are validated by comparison of results from
the reactive molecular dynamic simulation with experimental
measurements of the thermal maturation of the macromolecular
chemical system.
3) A method as claimed claim 1, wherein determination of the phase
equilibria is performed by calculating at any time total energy of
each molecule subjected to a force field, then by successive
assessments in order to quantify the hydrocarbon retention in the
macromolecular chemical system.
4) A method as claimed in claim 1, wherein the kinetic study is
carried out by varying temperature of the reaction mechanism.
5) A method as claimed in claim 1, wherein the macromolecular
chemical system represents organic matter of a mother rock of a
petroleum reservoir.
6) A method of simulating genesis of a sedimentary basin, wherein
formation and retention of hydrocarbons within a mother rock
containing kerogen are modelled, comprising: defining by means of
experimental characterizations a molecular model of the kerogen of
the mother rock; defining a reaction mechanism by subjecting the
molecular model to a molecular dynamic simulation associated with a
reactive force field; quantifying evolution of the kerogen to heavy
products and hydrocarbons, by carrying out a kinetic study of the
thermal decomposition of the kerogen from the reaction mechanism;
quantifying hydrocarbon retention within the mother rock by
carrying out assessments on fractions linked to than organic matrix
of the mother rock and on free fractions from phase equilibria
determined by molecular dynamic simulation; and using as the input
data in a basin simulator amounts of hydrocarbons formed and
amounts of hydrocarbons that are not retained in the mother rock,
that can be expelled from the mother rock and migrate to a
petroleum reservoir.
7) A method as claimed claim 2, wherein determination of the phase
equilibria is performed by calculating at any time total energy of
each molecule subjected to a force field, then by successive
assessments in order to quantify hydrocarbon retention in the
macromolecular chemical system.
8) A method as claimed claim 2, wherein the kinetic study is
carried out by varying temperature of the reaction mechanism.
9) A method as claimed claim 3, wherein the kinetic study is
carried out by varying temperature of the reaction mechanism.
10) A method as claimed claim 7, wherein the kinetic study is
carried out by varying temperature of the reaction mechanism.
11) A method as claimed in claim 2, wherein the macromolecular
chemical system represents organic matter of a mother rock of a
petroleum reservoir.
12) A method as claimed in claim 3, wherein the macromolecular
chemical system represents organic matter of a mother rock of a
petroleum reservoir.
13) A method as claimed in claim 4, wherein the macromolecular
chemical system represents organic matter of a mother rock of a
petroleum reservoir.
14) A method as claimed in claim 7, wherein the macromolecular
chemical system represents the organic matter of a mother rock of a
petroleum reservoir.
15) A method as claimed in claim 8, wherein the macromolecular
chemical system represents organic matter of a mother rock of a
petroleum reservoir.
16) A method as claimed in claim 9, wherein the macromolecular
chemical system represents organic matter of a mother rock of a
petroleum reservoir.
17) A method as claimed in claim 10, wherein the macromolecular
chemical system represents organic matter of a mother rock of a
petroleum reservoir.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention relates to a method of quantifying
hydrocarbon formation and retention within a macromolecular
chemical system.
[0003] 2. Description of the Prior Art
[0004] The following documents, mentioned in the description
hereafter, illustrate the state of the art: [0005] Behar F.,
Vandenbroucke M., Tang Y., Marquis F., Espitalie J., 1997. Thermal
Cracking of Kerogen in Open and Closed Systems: Determination of
Kinetic Parameters and Stoichiometric Coefficients for Oil and Gas
Generation. Org. Geochem., 26, 5-6, 321-339. [0006] Burnham, A. K.
and Braun, R. L., 1989. Development of Detailed Model of Petroleum
Formation, Destruction, and Expulsion from Lacustrine and Marine
Source Rocks. Advances in Organic Geochemisrty, 16, 1-3, 27-39.
[0007] Burnham, A. K. and Braun, R. L., 1990. Mathematical Model of
Oil Generation, Degradation, and Expulsion. Energy and Fuel, 4,
132-146. [0008] Faulon, J. L., Prediction Elucidation and Molecular
Modeling. Algorithms and Applications in Geochemistry, Ph. D.
Thesis, Edited by Ecole des Mines, Paris, 1991. [0009] Faulon, J.
L., Stochastic Generator of Chemical Structure (4) Building
Polymeric Systems with Specified Properties, J. Comput. Chem.,
2001, 22, 580-590. [0010] Freund, H., Walters, C. C., Kelemen, S.
R., Siskin, M., Curry, D. J., Xiao, Y., Olmstead, W. N., Gorbaty,
M. L., Bence, A. E., 2005. Predicting Oil and Gas Compositional
Yields via Chemical Structure-Chemical Yield Modeling (CS-CYM).
Organic Geochemistry Challenges for the 21st Century (Vol. 1), 22
IMOG Seville, Spain, 66-67. [0011] Hatcher, P. G., 1988.
Dipolar-Dephasing 13C Studies of Decomposed Wood and Coalified
Xylem Tissue: Evidence for Chemical Structural Changes Associated
with Defunctionalization of Lignin Structural Units During
Coalification. Energy & Fuels 2, 48-58. [0012] Pepper, A. S.,
1991. Estimating the Petroleum Expulsion Behaviour of Source Rocks:
A Novel Quantitative Approach. In Petroleum Migration (Edited by
England W. A. and Feed A. J.), Geological Society, Special
Publication. 59, pp. 9-31. [0013] Pepper, A. S., Corvi, P. J.,
1995. Simple Kinetic Models of Petroleum Formation. Part III:
Modelling and Open system. Marin and Petroleum Geology, 12, 4,
417-452. [0014] Pepper, A. S., Dodd, T. A., 1995. Simple Models of
Petroleum Formation. Part II: Oil to GAS cracking. Mar. Petrol.
Geol., 12, 321-340. [0015] Ritter, U., 2003. Fractionation of
Petroleum During Expulsion from Kerogen Journal of Geochemical
Exploration. 78-79, 417-420. [0016] Ritter, U., 2003. Solubility of
Petroleum Compounds in Kerogen: Implications for Petroleum
Expulsion. Organic Geochemistry. 34, 319-326. [0017] Tissot, B.
1969. Revue Inst. Fr. Petrole, 24(4), 470-501. [0018] Ungerer, P.,
1989. State of the Art of Research in Kinetic Modelling of Oil
Formation and Expulsion. In Advances in Organic Goechemistry,
Organic Geochemistry. 16, 1-3, 1-25. [0019] Van Duin A. C. T,
Siddharth D., Lorant F., Goddard III W. A. 2001. ReaxFF: A Reactive
Force Field for Hydrocarbons. J. Phys. Chem. A, 105, 9396-9409.
[0020] The insoluble organic material of the mother rock, also
referred to as kerogen, is a mixture of bio-organic macromolecules
(notably biogeopolymers) having aliphatic and aromatic chemical
structures that evolve in the course of geologic times with the
temperature and the pressure. Thermal maturation of the kerogen in
the mother rock occurs through the agency of two main
phenomena:
[0021] the first one is thermal cracking of the organic matter at
the origin of hydrocarbons; it takes place naturally in sedimentary
basins, generally at a temperature ranging between 80.degree. C.
and 200.degree. C., and a pressure ranging from 200 to 1000
bar,
[0022] the second one is the physicochemical evolution of the
petroleum products within the mother rock that explains the
retention and expulsion of hydrocarbons from the mother rock.
[0023] These two phenomena develop within the same context and they
are juxtaposed. In basin modelling, it is important to be able to
simultaneously calibrate the amounts of hydrocarbons formed and the
amounts of "free" hydrocarbons that can be expelled from the mother
rock and migrate to the reservoir. The process of hydrocarbon
retention in kerogen is a mechanism that controls the free
hydrocarbons/expelled hydrocarbons ratio. If the retention of
hydrocarbons is considered to predominantly occurs in the organic
matter of the mother rock, this retention depends on the
composition of the fluids generated, on the retention capacities of
the kerogen and on the volume ratio of the solid organic matter to
the liquid hydrocarbons. It is thus directly linked with the
physicochemical nature of the kerogen and with the conversion ratio
(Pepper, 1991).
[0024] Kerogen cracking and retention of the products from this
reaction being interdependent phenomena that evolve with the
thermal maturation of kerogen, it appears necessary to develop a
fine analysis of the cracking reaction coupled with the retention
of the products formed in the kerogen in order to be able to
estimate the hydrocarbons retained in the kerogen and the
hydrocarbons available for storage in reservoirs.
[0025] There are known hydrocarbon quantification models that
account for either the thermal cracking reaction, or the
hydrocarbon retention phenomenon, but rarely of both phenomena
simultaneously:
[0026] empirical and mechanistic cracking models are used to
quantify the hydrocarbons formed in the mother rock,
[0027] models of hydrocarbon retention in the mother rock try to
explain the segregation of the hydrocarbons when they are expelled
from the mother rock.
[0028] Thermal Cracking Models
[0029] Two cracking reaction modelling methods are provided in the
literature: empirical models and mechanistic models.
[0030] Empirical models are based on experiments to establish the
global stoichiometric equations that account for the mass balances
observed.
[0031] These stoichiometric equations are coupled with hydrocarbon
formation velocity laws, they correspond to a series of
simultaneous, independent and competitive reactions and they were
developed assuming that the global evolution of the petroleum
potential of a kerogen under maturation is an irreversible kinetic
process (Pepper and Dodd, 1995). The kinetic (E: activation energy,
A: frequency factor) and stoichiometric (Xi: relative contribution
of reaction i) parameters have to be calibrated individually
because the mother rocks do not generate hydrocarbons at the same
rate. Artificial maturation experiments are therefore carried out
in the laboratory under controlled thermal conditions. These
experiments are performed on kerogens or mother rocks and they can
be of different natures. By numerical inversion of the laboratory
data, it is possible to calculate the kinetic and stoichiometric
parameters. These parameters, obtained at high temperature
(300.degree. C.-600.degree. C.) over short times (some minutes to
some days) are then assumed to be extrapolatable for lower
temperatures than those of the experimental conditions such as
those imposed by the geothermal gradients.
[0032] This method is currently the only means allowing providing
information on the formation of hydrocarbons compatible with basin
models. However, this procedure is based on many approximations. In
fact, Behar et al.'s work of 1997 showed that the differences in
the experimental conditions of the pyrolyses carried out (open or
closed medium, in the presence or absence of water or of mineral
matrix, or according to the grain size of the sample) leads to a
kinetic parameters lag and therefore to an uncertainty in the
estimation of the oil window. Similarly, extrapolation of the
kinetic parameters at low temperature involves the nature of the
cracking mechanism does not significantly develop as a function of
temperature.
[0033] Mechanistic models are not based on stoichiometric equations
but on elementary (radical) reactions to simulate the thermal
degradation of complex macromolecules and to reproduce the
distribution of the hydrocarbons formed (Freund et al., 2005).
Elaboration of these models starts by modelling the initial
macromolecule. This modelling is constrained by experimental data
relative to the structural properties of the sample. It is based on
the distribution of the functional groups in the molecule for
establishing the probable structure thereof. Once the initial
macromolecule is defined, elementary reactions are applied to the
structure so as to simulate the formation of the thermal
degradation products. Each elementary reaction has its own kinetic
properties, valid whatever the temperature scale. It is thus
possible to simulate thermal maturation both under laboratory
conditions and under geologic conditions.
[0034] The advantage of this approach is that it minimizes the
uncertainty on the reaction velocities extrapolated to the geologic
conditions. On the other hand, the complexity of these models makes
elaboration of the reaction mechanism difficult. Finally, and above
all, the very large number of reactions in these models is
incompatible with current basin simulators.
[0035] Retention Models
[0036] Physical models have been provided in order to estimate the
proportion of expelled hydrocarbons as a function of the kerogen
conversion rate. Ungerer's expulsion model (1989) sets a threshold
corresponding to a conversion rate for which the hydrocarbons
formed in the mother rock are expelled. In this connection, Pepper
(1991) considers correlating "the petroleum expulsion efficiency"
(PPE) with the initial petroleum potential of the mother rocks
considered. These models reproduce more or less accurately
(according to the mother rock type) the quality of the hydrocarbons
of the reservoir. They consider the mother rock in two states only,
before and after expulsion of the hydrocarbons, without taking into
account the qualitative evolution of the mother rock or the
expulsion kinematics.
[0037] Later, Ritter (2003) provided a retention model based on the
solubility of the hydrocarbons in kerogen. He established an
empirical relation between the swelling ratio and the Hildebrand
solubility parameter, for each type of organic matter. This
relation defines a retention ratio for each group of compounds.
Finally, this model confirms the fractionation sequence observed in
nature, except for branched aliphatic hydrocarbons. The polymer
solubility theory and this model thus do not totally explain the
hydrocarbon composition differences between mother rock extracts
and reservoir oils, observed in petroleum systems. Similarly, this
model does not explain the great accumulation of aliphatic
hydrocarbons in coals. This model has two limitations: the first
one is that the values of the swelling ratio are filed according to
their chemical class but they are not normalized. Thus, the model
does not respect the mass conservation principle. This generates
too high retention thresholds and the sum of the compositions is
above 100%. The second drawback is due to the fact that the
swelling phenomenon involves swelling of the organic matrix. Now,
there is little chance that this swelling occurs in rocks subjected
to high overpressures.
[0038] In conclusion, the expulsion and retention models provided
to date involve possible mechanisms and they are developed with
more or less assumptions, which leads to more or less realistic
approaches. In fact, the structure of kerogen and the nature of the
effluents vary with the mother rock maturity, therefore thermal
cracking and expulsion are indissociable processes.
[0039] The method according to the invention allows quantification
of the formation and the retention of hydrocarbons in a mother rock
from a new type of simulation. This simulation type is based on a
dynamic molecular modelling technique coupled with a reactive force
field. As in the case of mechanistic models, this approach requires
as the starting point a "molecular" representation of the structure
of the kerogen. The method according to the invention does not
require writing hundreds of a priori reactions: the reaction
mechanism is not an input datum, it becomes a result of the dynamic
simulation. As in the case of radical mechanisms, this new
technique is applicable in any thermal regime.
SUMMARY OF THE INVENTION
[0040] The invention relates to a method of quantifying the
formation and the retention of hydrocarbons within a macromolecular
chemical system, comprising construction of a molecular model of
the system by means of experimental characterizations of a sample
of the system, and comprising:
[0041] defining a thermal maturation reaction mechanism for a
macromolecular chemical system by subjecting the molecular model to
a molecular dynamic simulation associated with a reactive force
field;
[0042] quantifying the free hydrocarbons formed throughout the
thermal maturation of the macromolecular system by:
[0043] a) determining a physicochemical evolution of the molecular
mixture defined by the reaction mechanism, by calculating phase
equilibria, directly from the molecular dynamic simulation; and
[0044] b) determining the thermal decomposition kinetics of the
macromolecular chemical system by kinetic study from said reaction
mechanism.
[0045] The present invention comprises a method of modelling the
thermal reactivity of macromolecular chemical systems whose
molecular structure is unknown or not well known.
[0046] In particular, it can be used within the context of basin
modelling where the method allows parameterization of the
proportion of hydrocarbons formed and possibly retained in the
organic matrix of the mother rock.
[0047] In the description hereafter, what is referred to as
"macromolecular chemical system" or "macromolecular system" is a
system consisting of at least one organic macromolecule, for
example the organic matter of the mother rock.
[0048] The "reaction mechanism" is understood to be all of the
chemical reactions describing the thermal maturation of each
macromolecule. A reaction mechanism is associated with a set of
quantitative and qualitative data that characterize the reagents,
the products and the velocities of the chemical reactions.
[0049] A "force field" is a set of parametrized equations that
describe the various contributions to the total potential energy of
a chemical system (notably Van der Waals, Coulomb, torsional
energies).
[0050] In the method according to the invention, the molecular
model and the reaction mechanism are advantageously validated by
comparison of the results from the reactive molecular dynamic
simulation with experimental measurements of the thermal maturation
of the macromolecular chemical system.
[0051] Determination of the phase equilibria is preferably
performed by calculating at any time the total energy of each
molecule subjected to the force field, then by successive
assessments in order to quantify the hydrocarbon retention in the
macromolecular chemical system.
[0052] The kinetic study is preferably carried out by varying the
temperature of the reaction mechanism.
[0053] In an embodiment, the macromolecular chemical system
represents the organic matter of a mother rock of a petroleum
reservoir.
[0054] The invention also relates to a method of simulating the
genesis of a sedimentary basin, wherein the formation and the
retention of hydrocarbons within a mother rock containing kerogen
are modelled, and wherein the following stages are carried out:
[0055] defining by means of experimental characterizations a
molecular model of the kerogen of the mother rock;
[0056] defining a reaction mechanism by subjecting the molecular
model to a molecular dynamic simulation associated with a reactive
force field;
[0057] quantifying the evolution of the kerogen to heavy products
and hydrocarbons, by carrying out a kinetic study of the thermal
decomposition of the kerogen from said reaction mechanism;
[0058] quantifying the hydrocarbon retention within the mother rock
by carrying out assessments on the fractions linked to the organic
matrix of the mother rock and on the free fractions from the phase
equilibria determined by molecular dynamic simulation; and
[0059] using as the input data in a basin simulator the amounts of
hydrocarbons formed and the amounts of hydrocarbons that are not
retained in the mother rock, that can be expelled from the mother
rock and migrate to a petroleum reservoir.
BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Other features and advantages of the method according to the
invention will be clear from reading the description hereafter of a
non limitative embodiment example, with reference to the
accompanying figures wherein:
[0061] FIG. 1 shows the principle of the method according to the
various space and time scales of the phenomena modelled;
[0062] FIG. 2 illustrates the sequence of the stages of the method
for reaching evaluation of the hydrocarbons formed and retained in
the mother rock;
[0063] FIG. 3 illustrates the lignite molecular characterization
protocol;
[0064] FIG. 4 illustrates the result of the reactive dynamic
simulations according to the invention on three molecular models of
lignite fragments;
[0065] FIG. 5 illustrates the lignite thermolysis mechanisms;
and
[0066] FIG. 6 is a cross-sectional view of an oil development
field.
DETAILED DESCRIPTION
[0067] The method according to the invention numerically models the
thermal maturation of the organic matter of the mother rock or of
any other macromolecular system, and the retention of the products
from this reaction in the residual organic matrix, and to possibly
extrapolate the results to the basin scale.
[0068] FIG. 1 shows the principle of the method as a function of
the various space scales (D(m): Distance in meter) and time scales
(t(s): time in second). The initial kerogen (or any other
macromolecular structure) sample (E) is experimentally
characterized, which allows determination of a molecular model of
the structure (MM (.ANG., 10.sup.-15 s)). This molecular model is
the input datum of the dynamic simulations coupled with a reactive
force field (RMD (.ANG., 10.sup.-15 s)). These simulations model
the thermal cracking reaction of the molecular model under the
given conditions, and they are validated by comparison with the
experimental thermal maturation data. At several stages of the
simulated reaction, phase equilibrium calculations (PES (.mu.m,
10.sup.-15 s)) are carried out in order to establish a
physicochemical assessment of the reaction medium. The free
hydrocarbons and those retained in the solid residue are estimated
on the molecular scale from all of these calculations, then
extrapolated to the geologic conditions (Q (Km, Ma)).
[0069] The method can be broken down into three major stages:
[0070] 1---Determination of the nature and of the molecular
structure of the macromolecular chemical system (kerogen for
example) [0071] 2-Reactive dynamic molecular simulation [0072]
3-Thermodynamic study and kinetic study
[0073] These stages are diagrammatically shown in FIG. 2; from
laboratory data on sample (E), a molecular model (MM) of the sample
is established and introduced in a reactive dynamic simulation
(RMD). The results of the reactive dynamic simulation are validated
(V) by comparison with the experimental data. The reactive dynamic
simulation (RMD) allows construction of a reaction mechanism (SR)
and description of the physicochemical evolution of the retention
(PC) from the phase equilibria determination. From the reaction
mechanism, the thermal decomposition reaction kinetics is
determined, notably by thermal cracking (C). Coupling the kinetic
study and the physicochemical study allows quantification of the
hydrocarbons produced and the "free" hydrocarbons (Q).
[0074] First Stage: Determination of the Molecular Structure and of
the Nature of the Organic Matter--Elaboration of a Molecular
Model
[0075] The first stage of this method determines the structure and
the nature of the macromolecular system (notably organic matter of
the mother rock) that will be used as initial data for modelling
the thermal maturation reaction. For example, many organic matter
characterization techniques, such as elementary analysis (AE),
Fourier transform infrared spectroscopy (FT-IR), nuclear magnetic
resonance spectroscopy (NMR) and X-ray diffraction (DRX), can be
used to quantify the various functions and to know the spatial
distribution of the molecular structure. Similarly, indirect
characterization techniques such as the study of pyrolysis products
are also an efficient means of determining macromolecular
structures. Using molecular models such as Signature (random
macromolecular generator, Faulon, 1991 and 2001) associated with
the experimental characterization data can allow obtaining a more
realistic macromolecular system. The first stage of the method
according to the invention allows to establish a molecular model of
the macromolecular system.
[0076] Second Stage: Reactive Dynamic Molecular Simulation
[0077] The cracking reaction of the molecular model of the
macromolecular system is reproduced by dynamic molecular simulation
calculations coupled with a reactive force field (RMD). These
simulations are compared with experimental thermal maturation data
(quantitative data (conversion, mass balance, elementary analysis .
. . ) and qualitative data (product characterization (by gas
chromatography coupled with a mass spectrometer GC-MS, magnetic
resonance NMR, Fourier transform infrared spectroscopy FT-IR, X-ray
diffraction DRX . . . ) from pyrolyses in open or closed media, or
other thermal maturation experiments depending on the simulation
conditions). The similarity between the experimental and numerical
results validates both the molecular model of the initial sample
and the reaction mechanism provided by the simulation.
[0078] Dynamic simulations are three-dimensional molecular
modellings allowing representation of a set of molecules in a force
field describing the energy status of the medium. In a given
volume, a system of molecules having n atoms and m bonds will turn
into a system keeping a number n of atoms but evolving towards a
number m' (different from or equal to m) of bonds.
[0079] Van Duin et al. (2001) provide a reactive force field called
"Reaction Force Field" (ReaxFF). This force field is an
intermediate approach between the quantum models and the
non-reactive force fields.
[0080] The two particular features of this field are that its
energy function is based on the order of the bonds between atoms,
and that there is a continuity between the binding energies and the
non-binding energies. The result is a rather complex energy
function that incorporates the partial energy contributions
allowing to describe the various types of bonds, as well as the
cleavages and the formation of bonds:
E.sub.total=E.sub.bond+E.sub.over+E.sub.under+E.sub.val+E.sub.pen-
+E.sub.tors+E.sub.colj+E.sub.vdw+E.sub.coulomb with: [0081]
E.sub.bond=binding energy [0082] E.sub.under=undercoordination
energies [0083] E.sub.over=overcoordination energies [0084]
E.sub.val=valence energy [0085] E.sub.pen=penalty term (associated
with E.sub.over, E.sub.under in the case of allenes) [0086]
E.sub.tor=torsional energy [0087] E.sub.conj=conjugation energy
[0088] E.sub.vdw=van der Waals energy [0089] E.sub.coulomb=Coulomb
energy
[0090] A parametrized analytical function relating the energy to
the order of the bonds involved corresponds to each energy term.
For example, the binding energy between two atoms i and j is
written as follows:
E.sub.bond=-D.sub.eBO.sub.ijexp(p.sub.be,1(1-BO.sub.ij.sup.p.sup.be,1))
where BO.sub.ij is the order of the bond between i and j, D.sub.e
and p.sub.be,1 are parameters calibrated for different atom pairs.
According to Pauling's principle, the bond order itself depends on
the relative distance between the atoms. In the case of pluriatomic
systems, calculation of BO.sub.ij between two atoms takes account
of the atomic environment. Van Duin et al. (2001) developed a
calculation mechanism relating the bond orders to the relative
positions of the atoms in space while taking account of all of
these interactions.
[0091] To date, ReaxFF allows carrying out dynamic simulations on
systems consisting of at least one of the following atoms: C, H, O,
N, S, Si, Pt, Zr, Ni, Au, V, Bi, Ti, Mo. The parameters associated
with the different energy terms are calibrated by numerical
inversion of both the experimental data (generally reaction
enthalpy) and the theoretical data calculated by means of quantum
methods.
[0092] Concerning the dynamic simulation as such, ReaxFF comes in
form of a software including both the force field (in form of an
auxiliary file) and the molecular dynamics engine. This engine
takes up the basic principles of dynamic simulation wherein the
motion of the atoms is described by the Newtonian mechanics in the
three-dimensional space: { F -> i = m i .times. d 2 .times. r
-> i .function. ( t ) d t 2 = - d E -> d r i .function. ( t )
E -> = f .function. ( BO ij ) = f ' .function. ( r 1 , .times. ,
r n ) ##EQU1## with: F.sub.i=force exerted on atom i [0093]
r.sub.i=trajectory of atom i.
[0094] Thus, dynamic simulations with force field ReaxFF allow
reproduction of the intra and intermolecular mechanisms of chemical
reactions.
[0095] FIG. 4 illustrates the main parameters of the reactive
dynamic simulations on three lignite fragments. First (SI), one or
more molecules (N.sub.0 atoms and M.sub.0 atomic bonds) are
constructed in three dimensions in a box of predetermined
dimensions. It is possible to vary the volume parameters of the
box, the pressure and/or temperature parameters during integration
of the model as a function of time. A succession of states of the
system (SF) evolving as a function of time in the force field
(N.sub.0 atoms but Mt (.noteq.M.sub.0) bonds) is obtained.
[0096] Third and Fourth Stages: Thermodynamic Study and Kinetic
Study
[0097] The dynamic simulation results allow establishing a reaction
mechanism and they can be interpreted or developed according to two
approaches:
[0098] physicochemical approach of the reaction by studying the
phase equilibria: thermodynamic study,
[0099] kinetic approach of the decomposition reaction for
quantifying the formation of the products predominantly formed or
the dissociation of the initial model kerogen: kinetic study.
[0100] Physicochemical study (PC): at each time interval, the
dynamic simulation calculates the total energy of each molecule of
the force field. From these energies, it is possible to sort out
the molecules according to their physical state (liquid, solid or
gaseous). The physical evolution of each molecule can then be
monitored during simulation of the reaction. The successive phase
equilibrium assessments (PES) allows following the physicochemical
evolution of the thermal maturation of the sample studied. This
thermodynamic analysis allows study and qualification of the
thermolysis products that are independent of the "residue" (one or
more molecules in solid phase and of high molecular mass) of the
dependent products.
[0101] Kinetic study (C): the kinetic study of the reaction
requires dynamic calculations at different temperatures. At each
temperature, quantitative assessments of the appearance or
disappearance of predominant molecules are established in order to
calculate the reaction velocity (k: rate of appearance or of
disappearance of a chemical species). From the values of different
reaction velocities of each chemical species at the different
temperatures, it is then possible to calculate the kinetic
parameters (activation energy Ea and pre-exponential factor A) and
to deduce therefrom the thermal decomposition kinetics of each
molecular species in the temperature range studied.
[0102] Finally, from a double study (physicochemical and kinetic)
of the mechanisms provided by the dynamic simulations, the free
hydrocarbons (liquid and gaseous), not retained in the solid
residue, can be quantified throughout the numerical maturation of
the sample. They represent, in sedimentary basins, the hydrocarbons
likely to be expelled from the mother rock (Q) and they can thus be
an input parameter for basin simulators.
[0103] Example: Study of Lignitic Coal
[0104] The Australian brown coal of this example essentially
consists of lignite molecules (L). Some characteristics of this
coal and of related chemical structures are given in Hatcher's
work, 1988.
[0105] Definition of the Initial Chemical System: Molecular
Model
[0106] First, from experimental organic matter characterization
measurements carried out on a sample of this coal (NMR analysis,
infrared spectroscopy IR, atomic composition, pyrolyses in closed
circuits PYR--FIG. 3), the chemical functions are identified and
counted, then, by comparison with the lignite structures provided
by Hatcher (Hatcher, 1988), one deduces that the coal can be
correctly represented as an agglomerate of a form of lignite that
is modelled by a macromolecule of about 250 atoms. In this example,
the kerogen model is manually constructed. For more complex
kerogens, simulators such as Signature (Faulon 1999, 2001) can be
used to randomly organize known units of the structure.
[0107] Thermal Reactivity Modelling: Reactive Dynamic Simulation
(RMD)
[0108] Once the initial chemical system defined, the thermal
reactivity of all or part of the chemical structure deduced in the
previous stage is then modelled. A molecular dynamic simulation
algorithm associated with a particular force field, called ReaxFF
(van Duin et al., 2001), and allowing prediction of, among other
things, the thermal cracking of organic matter under given
conditions, is therefore used. Examples of such simulations are
shown in FIG. 4. In these examples, we have modelled the thermal
cracking of three typical fragments of the lignite structure shown
in FIG. 3. For each one of the three lignite fragments (L)
selected:
[0109] Number of units: 15
[0110] Volume: 3723.2 .ANG..sup.3
[0111] Density: 1.4 kg.L.sup.-1
[0112] Number of units: 17
[0113] Volume: 5661.2 .ANG..sup.3
[0114] Density<1.4 kg.L.sup.-1
[0115] Number of units: 21
[0116] Volume: 4870.9 .ANG..sup.3
[0117] Density<1.4 kg.L.sup.-1
[0118] For each fragment, these simulations are carried out as
follows:
[0119] a volume containing n units of the same fragment is
constructed with the algorithm,
[0120] after an initialization stage (atom velocities balancing and
calculation of the system density at experimental P and T), the
reactive dynamic simulations are carried out under different
thermal conditions and at constant volume. In FIG. 4, temperature T
is above 2000 K. Although the physical duration modelled is finally
of the order of some picoseconds, the calculation times can be very
long (some hours to some days CPU).
[0121] At the end of a computation, in the final systems (SF), the
volume contains new molecules from the various thermal
decomposition mechanisms that have affected the initial lignite
fragments (initial systems SI). The software allows identification
of and counting of all of the molecules present in the volume at
any time. Fine analysis of these data thus allows both prediction
of the composition of the cracking products of the initial
fragments and determination of the mechanisms at the origin
thereof.
[0122] Assessment of the Reactive Dynamic Simulation Results:
Achieving a Reaction Mechanism
[0123] In a third stage, the results of the reactive dynamic
simulations are assessed in order to extract therefrom a general
coal thermal decomposition mechanism. In the example given, this
work shows (FIG. 5) that the major part of the experimentally
observed pyrolysis products is obtained by means of two main
cracking mechanisms affecting the lignite. These two reactions
explain both the formation of water and of methane, through a
defunctionalization process (mechanism 1--M1), and the
depolymerization in stages (mechanism 2--M2) of the lignite, at the
origin of the constituents of the C.sub.14+ fraction illustrated in
FIG. 3.
[0124] In conclusion, a general pyrolysis mechanism for the coal
studied should have the form as follows:
[0125] the lignite undergoes depolymerization in fragments,
[0126] the fragments undergo defunctionalizations and give
molecules such as molecules of water (H2O), methane (CH4) and other
hydrocarbons.
[0127] Kinetic Study
[0128] In order to quantify the thermolysis of lignite, a kinetic
study is carried out from the reaction mechanism predicted by
reactive dynamic modelling.
[0129] In general terms (Tissot 1969, Braun and Burnham 1989 and
1990, Pepper and Corvi 1995), the primary cracking models
correspond to a series of simultaneous reactions, independent and
competitive, describing the mass transfer between the source
kerogen and the hydrocarbons. A reaction mechanism example is given
below. A velocity law of order 1 that depends both on time and on
temperature corresponds to each reaction of the mechanism. Thus,
the amount m of hydrocarbons formed in the course of time is
expressed by: { m = X inf .times. i = 1 N .times. .times. X i
.times. q i d q i d t = k i .function. ( 1 - q i ) n ( 1 ) ##EQU2##
with [0130] q.sub.i=conversion ratio of reaction i [0131]
X.sub.i=relative contribution of reaction i [0132]
X.sub.inf=petroleum potential [0133] k.sub.i=velocity constant of
reaction i [0134] n=order of the reaction [0135] X.sub.i and
k.sub.i obey the following laws: i = 1 N .times. .times. X i = 1 (
2 ) k i = A i .times. exp .function. ( - E i RT ) ( 3 ) ##EQU3##
with A.sub.i=frequency factor and E.sub.i=activation energy for
reaction i.
[0136] Thermodynamic Study of the Molecules
[0137] From the reaction mechanism of the dynamic simulations, it
is also possible to sort out the molecules formed according to
their molecular weight and to determine their thermodynamic state.
Simulators using the Monte Carlo method allow theoretical study of
the phase equilibria in a molecular mixture. Using this method or
an equivalent method on the molecular mixture (products and
reagents of the lignite thermolysis) allows monitoring of the
thermodynamic evolution of the reaction with time. This
thermodynamic analysis allows studying and quantification of the
thermolysis products that are independent of the "residue" (one or
more molecules in solid phase and of high molecular mass) of the
dependent products.
[0138] The potentialities of molecular modelling allow producing
complex molecular models (case of kerogen), to predict the
thermodynamic properties at equilibrium of these chemical
structures, and to calculate parameters involved in the retention,
such as viscosity, diffusion, density, phase state, electrostatic
charges distribution, binding and repelling forces, to predict the
thermal reactivity of these structures according to pressure and
temperature, and to determine the chemical nature of the products
formed.
[0139] The method according to the invention uses a reactive
dynamic simulation, which allows prediction of the evolution of the
kerogen to heavy products and hydrocarbons, as well as to perform
assessments on the fractions linked to the organic matrix and on
the free fractions, thus to quantify the retention.
[0140] The method intended for modelling the thermal reactivity of
complex chemical systems was applied for quantifying the formation
and the retention of hydrocarbons within a complex
carbon-containing system such as the organic matter contained in a
mother rock. FIG. 6 shows an example of an oil exploration system.
Zone G represents the gas window (generally temperature above
150.degree. C. and pressure above 53 MPa). Zone represents the oil
window (temperature generally ranging between 80.degree. C. and
150.degree. C., pressure ranging between 24 MPa and 53 MPa). Zone
IT is the thermal immaturity zone (temperature below 80.degree. C.
and pressure below 24 MPa). The "free" hydrocarbons migrate from
the mother rock (RM) to the reservoir zone (RR). In the absence of
cap rock (RC) at the surface, the petroleum (P) can seep at the
surface. Boreholes are drilled through the rock layers to take the
crude from the reservoir rock (RR).
[0141] This modelling can also find applications in other
particular spheres in order to understand and/or quantify a
multiscale process that can be simulated through experiment and
explained by complex and/or unknown chemical mechanisms. Non
limitative examples thereof are the thermal behavior of heavy
crudes during steam or air flooding in a reservoir, and
hydrocracking of asphaltenes during refining.
* * * * *