NMR in kinetics of hydrocarbon generation

Abstract

Disclosed are methods of characterizing kerogen and its hydrocarbon generation potential using NMR as the primary analytical tool, and using such data to derive the kinetics of hydrocarbon generation and alteration, thus predicting the hydrocarbon potential of source rock in geological setting, which can then be used in petroleum exploration and production.

Claims

1. A method of determining and using hydrocarbon generation potential from kerogen, said method comprising: a) obtaining a sample of kerogen; b) performing elemental analysis on a portion of said kerogen to determine its C, H, N, S and O content; c) performing nuclear magnetic resonance (NMR) analysis on a portion of said kerogen to determine its initial relative abundances of different H and C species; d) pyrolyzing a portion of said kerogen to determine a pyrolysis temperature profile and to produce petroleum fluid and a kerogen residue; e) analyzing the composition of said petroleum fluid; f) performing NMR analysis on said kerogen residue; and g) predicting hydrocarbon generation from said kerogen using the data obtained in steps b-f to determine the hydrocarbon generating potential of said kerogen; and, h) using said hydrocarbon generating potential in formulating and executing exploration and production plans.

2. The method of claim 1, wherein said NMR analysis is solid state NMR.

3. The method of claim 1, wherein said NMR analysis uses 13C NMR.

4. The method of claim 1, wherein said NMR analysis uses 1H NMR.

5. The method of claim 1, wherein said NMR analysis uses 15N NMR.

6. The method of claim 1, wherein said NMR analysis uses both 13C and 1H NMR.

7. The method of claim 1, wherein said NMR analysis is solid-state magic angle spinning (MAS) NMR.

8. The method of claim 1, wherein said NMR analysis is solid state NMR using cross polarization (CP).

9. The method of claim 1, wherein said NMR analysis is solid state NMR using direct polarization (DP).

10. The method of claim 1, wherein said NMR analysis is solid state NMR using both CP and DP.

11. The method of claim 1, wherein said method uses spin counting to calibrate NMR data.

12. The method of claim 1, wherein predicting step uses a network of first order parallel reactions.

13. The method of claim 1, wherein predicting step uses higher order parallel reactions plus sequential reactions.

14. The method of claim 1, wherein predicting step uses the Arrhenius equation.

15. The method of claim 1, wherein said pyrolyzing step produced a pyrogram that can be read to determine S1, S2, S3, and Tmax.

16. The method of claim 1, wherein said identifying step uses gas chromatography or mass spectrometry or NMR or a combination thereof.

17. The method claim 1, wherein NMR provides relative abundances of rigid H, mobile H and C species.

18. The method of claim 1, wherein gold vessel thermolysis of a portion of said isolated kerogen is performed as a double check of the data.

19. A method of determining and using hydrocarbon generation potential from kerogen, said method comprising: a) obtaining a sample of source rock containing kerogen; b) grinding said source rock to produce a powder; c) extracting said powder to produce isolated kerogen; d) performing elemental analysis on a portion of said isolated kerogen to determine its C, H, N, S and O content; e) performing NMR analysis on a portion of said isolated kerogen to determine its initial relative abundances of different H and C species; f) pyrolyzing a portion of said isolated kerogen to determine a pyrolysis temperature profile and to produce petroleum fluid and a kerogen residue; g) analyzing the composition of said petroleum fluid; h) performing NMR analysis on said kerogen residue; and i) predicting hydrocarbon generation potential from said kerogen using the data obtained in steps d-h and using first order parallel reactions or higher order parallel reactions plus sequential reactions; and, j) using said hydrocarbon generating potential in formulating exploration and production plans and using said exploration and production plans in exploring and producing hydrocarbons.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1A-D Exemplary compounds in kerogen wherein FIG. 1A is Algal kerogen, FIG. 1B is Liptinitic Kerogen, and FIG. 1C is Humic Kerogen. FIG. 1D displays chemical features of the compounds in A-C.

(2) FIG. 2 is a pyrogram to determine the amount of pyrolyzable carbon, residual carbon and TOC. Free hydrocarbons are measured by the S1 peak and residual hydrocarbons are measured by the S2 peak. T.sub.max of 472 C. corresponds to the temperature recorded when the S2 peak was achieved. CO, CO.sub.2, and mineral carbon components of the S3 measurements are also displayed. CO.sub.2 is proportional to the amount of a oxygen present in organic matter and provides input for calculating an important index used in determining maturity and kerogen type.

(3) FIG. 3 Exemplary kerogen spectra. From Smermik 2006.

(4) FIG. 4 shows the .sup.13C NMR spectra of four isolated kerogen samples of different maturities.

DETAILED DESCRIPTION

(5) The disclosure provides a novel method, apparatus and system for accurately predicting hydrocarbon generation from kerogen and subsequent alteration. Hydrocarbon generation kinetics are typically derived by open system or closed system pyrolysis followed by product analyses, mostly by Rock-Eval, GC and GC/MS. However, these methods do not provide very accurate C and H mass balances. Thus, we propose to use NMR analysis in this work, thus providing faster turnaround time, as well as improved data for C and H mass balance in kinetics modeling.

(6) The methodology employed can be generally described as follows:

(7) 1. Sample Preparation:

(8) 1.a. Select immature source rock containing a sample of interest, or an analog if target source rock is unavailable.

(9) 1.b. Isolate kerogen from immature source rock by first soxhleting the powdered rock with 90:10 dichloromethane/methanol mixture, then removing minerals with acid digestion.

(10) 2. Initial Characterization:

(11) 2A. Elemental Analysis:

(12) Perform elemental analysis on the isolated kerogen to determine its C, H, N, S and O content. The initial relative abundance of H and C is thus obtained.

(13) Elemental analysis or EA can be by any method known in the art. The most common form of elemental analysis, CHN analysis, is accomplished by combustion analysis. In this technique, a sample is burned in an excess of oxygen and various traps, collecting the combustion products: carbon dioxide, water, and nitric oxide. The masses of these combustion products can be used to calculate the composition of the unknown sample.

(14) Other quantitative methods include: 1) Gravimetry, where the sample is dissolved and then the element of interest is precipitated and its mass measured or the element of interest is volatilized and the mass loss is measured. 2) Optical atomic spectroscopy, such as flame atomic absorption, graphite furnace atomic absorption, and inductively coupled plasma atomic emission spectroscopy, which probe the outer electronic structure of atoms. 3) Neutron activation analysis, which involves the activation of a sample matrix through the process of neutron capture. The resulting radioactive target nuclei of the sample begin to decay, emitting gamma rays of specific energies that identify the radioisotopes present in the sample. The concentration of each analyte can be determined by comparison to an irradiated standard with known concentrations of each analyte.

(15) To qualitatively determine which elements exist in a sample, the methods include Mass spectrometric atomic spectroscopy, such as inductively coupled plasma mass spectrometry, which probes the mass of atoms. Other spectroscopy, which probes the inner electronic structure of atoms such as X-ray fluorescence, particle-induced X-ray emission, X-ray photoelectron spectroscopy, and Auger electron spectroscopy, can also be used

(16) Chemical methods of elemental analysis are also possible.

(17) 2B: Initial NMR Characterization:

(18) Perform NMR analysis of the immature kerogen and determine its initial relative abundances of different H and C species, e.g. aliphatic vs. aromatic H, C with different numbers of bonded H, and correlate the thus determined H and C abundance to elemental analysis results obtained in step 2A.

(19) 3. Thermolysis:

(20) Artificially mature (thermolysis) the immature kerogen/source rock at certain temperatures in a closed vessel (e.g. quartz tube). The sample vessels preferably have adjustable headspace volume. During thermolysis, sample is compacted and encapsulated into a small volume. After thermolysis, the products can be released into the headspace. Each of the gas, liquid and solid products may be measured and identified by sampling from the closed reaction vessel.

(21) We can adjust headspace volumes to investigate the partitioning of petroleum fluids between free space and kerogen matrix (desorbed free species vs. absorbed species in kerogen under different PVT conditions).

(22) If necessary, other materials, e.g. water, minerals, hydrogen, can be co-encapsulated with kerogen/source rock for thermolysis. Thermolysis of kerogen isolate vs. whole rock, with and without water enable studying different aspects/effects of hydrocarbon generation subsurface over geological time.

(23) 4 NMR Analysis:

(24) After thermolysis, the remaining kerogen residue may be analyzed directly by NMR. This will provide information about the hydrogen content of the unconverted kerogen and char-like residue, thus providing how much hydrogen has converted to hydrocarbon fluids.

(25) NMR analyses are performed to determine the abundance changes of H and C in their different chemical environments (e.g. aliphatic vs. aromatic). The overall transformation ratio can be readily and reliably determined by the abundance of C without bonded H (graphite, dead coke), thus a bulk kinetics model can be readily derived. The compound specific H and C NMR signals enable monitoring of composition changes of generated hydrocarbon species (petroleum fluids), which then enables deriving compositional kinetics models.

(26) Unlike current kinetics analysis methods with relatively loose controls on carbon and hydrogen mass balances, kinetics derived from NMR analysis described by this invention have improved mass balance controls on both C and H.

(27) Any method of NMR analysis is possible, including e.g., NMR spectroscopy, Continuous wave (CW) spectroscopy, Fourier transform spectroscopy, Multi-dimensional NMR Spectroscopy and Solid-state NMR spectroscopy. It is known in the art how to obtain high-resolution .sup.13C and .sup.1H or even .sup.15N NMR spectra by solid state NMR, and such may therefore be preferred.

(28) To date, most solid-state .sup.13C NMR studies of kerogen have involved quantifying signal in a range of chemical shift regions and assigning these to specific functional groups. There is an inherent danger in this approach, due to the fact that the NMR signal of some functional groups can be compromised, especially when the cross polarization (CP) technique is used. This issue has been discussed widely in the coal literature, and has led to the greater use of the more quantitatively reliable direct polarization (DP) technique, otherwise known as Bloch decay or single pulse excitation, and which may thus be preferred. Some workers also recommend a simple calibration procedure called spin counting to be very useful for diagnosing NMR quantitation problems in the analysis of organic matter.

(29) 5. Calibration:

(30) For calibration purposes, gold vessel thermolysis can be performed in parallel with NMR samples (undergoing the same thermal stresses). Gold tube thermolysis can be conducted with high confining pressure (mimic of overburden subsurface). This is not doable for quartz tube thermolysis. Thus, the gold tube thermolysis can provide a double check of the accuracy of the pyrolysis data. Eventually, we can use quartz tube thermolysis alone to derive compositional kinetics, only correcting the data by gold tube samples if necessary (e.g. under ultra high pressure conditions).

(31) The petroleum fluids generated inside the gold vessel will be extracted out (e.g. by a supercritical fluid extraction system using carbon disulfide or something similar as solvent). The residue will then be analyzed by NMR for the abundances of different H and C. The abundance change of C without bonded H serves as a double check for the bulk kinetics derived from analysis described in step 4.

(32) The extract from the gold tubes can either be analyzed by NMR for its composition, and/or conventional GC and GC-MS for detailed speciation. If there are any differences resulting from different thermolysis vessels (e.g. between quartz tube and gold tube), such differences will allow correlation of chemical changes occurred under different thermolysis environments (quartz vs. gold tube, both are closed systems, but under different pressures during thermolysis).

(33) 6. Numerical Analysis:

(34) Kerogen maturation and hydrocarbon generation can be described as a redistribution process of hydrogen among hydrogen enriched species (oil and gas) and hydrogen depleted species (coke). The rate of hydrogen redistribution and the resulted concentration changes of different species are governed by kinetics of chemical reaction and, to certain extent by thermodynamics at high maturity stage.

(35) We can model this process using a network of first order parallel reactions, as currently used, or higher order parallel plus sequential reactions, or combinations thereof. Higher order chemical sequential reactions are more challenging to model by traditional kinetics analysis experiments, particularly stoichiometry and mass balance of hydrogen. NMR analysis, however, with direct monitoring of relative abundances of H and C at different chemical environments (structures), allows much tighter control on C and H mass balances, and better numerical solutions for differential equations describing the evolution of different species. A network of reactions will be devised to describe the evolution of different species.

(36) For each individual member reaction, its reaction rate constant (k) is described by Arrhenius equation:
k=Ae.sup.Ea/RT
Where A is frequency factor, Ea is activation energy, R is gas constant, and T is temperature in Kelvins.

(37) The whole set of kinetics parameters, including stoichiometry and Arrhenius parameters of each reaction will be determined by non-linear regression with experimental data (e.g. integral of different H and C NMR signals).

(38) Herein we describe an exemplary NMR protocol: Solid-state .sup.13C magic angle spinning (MAS) NMR spectra can be obtained at a .sup.13C frequency of 50.3 MHz on e.g., a Varian Unity-200 spectrometer. Samples are packed in a 7 mm diameter cylindrical zirconia rotor with Kel-F end-caps and spun at 5000100 Hz in a Doty Scientific MAS probe. CP spectra are acquired using a 1-ms contact time and a 0.5-s recycle delay. 10,000-100,000 scans are collected for each spectrum.

(39) DP spectra are acquired using a 6.0-ms (901) .sup.13C pulse. A recycle delay of 90 seconds is used for all samples and 1000 transients collected for each sample. DP spectra are corrected for background signal. Free induction decays for both CP and DP spectra are acquired with a sweep width of 40 kHz. 1216 data points are collected over an acquisition time of 15 ms. All spectra are zero-filled to 8192 data points and processed with a 50-Hz Lorentzian line broadening and a 0.005-s Gaussian broadening. Chemical shifts are externally referenced to the methyl resonance of hexamethylbenzene at 17.36 ppm.

(40) Spin counting experiments are performed using the method of Smernik and Oades. Glycine can be used as an external intensity standard (i.e. the glycine spectrum was acquired separately to those of the samples). For CP spin counting experiments, differences in spin dynamics between the sample and the glycine standard are accounted for using the method of Smernik and Oades, except that a variable spin lock (VSL) rather than a variable contact time (VCT) experiment is used to determine T.sub.1pH.

(41) VCT and VSL experiments are performed as part of the RESTORE procedure [Smernik and Oades] for determining rates of T.sub.1pH relaxation and rates of polarization transfer (TCH). VCT experiments can consist of an array of eight contact times (2, 2.5, 3, 4, 5, 6, 8, 10 ms). The experiments are run in an interleaved fashion, with 32 scans acquired for each contact time, in turn. This is repeated until a total of 4000 scans is acquired. A 0.5-s recycle delay can be employed for all samples.

(42) VSL experiments are performed with three different contact times, 200 ms, 1 and 2 ms. For the 200-ms contact time VSL experiments, ten spin lock times are used (0, 0.3, 0.8, 1.3, 1.8, 2.3, 2.8, 3.8, 4.8 and 5.8 ms), for the 1-ms contact time VSL experiments, ten spin lock times are used (0, 0.5, 1, 1.5, 2, 3, 4, 5, 7 and 9 ms) and for the 2-ms contact time VSL experiments, eight spin lock times are used (0, 0.5, 1, 2, 3, 4, 6 and 8 ms). The VSL experiments are run in an interleaved fashion, with blocks of 32 scans acquired in turn, to a total of 4000, with a 0.5-s recycle delay between scans.

(43) Three spectra are acquired as input spectra for generating RESTORE subspectra; a 1-ms contact time0 spin lock spectrum, a 5- or 6-ms contact time0 spin lock spectrum, and a 1-ms contact time1-, 2- or 3-ms spin lock spectrum. These spectra can be acquired in an interleaved fashion, with blocks of 32 scans acquired in turn, to a total of 10,000-25,000, with a 0.5-s recycle delay between scans.

(44) Proof of principle experiments have been attempted and turned out to be successful. However, data points from these early tests were insufficient for rigorous numerical analysis. Nevertheless, the method was a success, and is an improvement over existing methods due to more accurate and complete data.

(45) A set of petroleum source rock samples of different thermal maturities were used. Kerogen isolate was prepared via acid digestion of the source rocks. Bitumen was extracted from source rock and kerogen using dichloromethane as solvent.

(46) Solid-state .sup.13C and .sup.1H magic angle spinning (MAS) NMR measurements were performed on a Bruker DSX-300 spectrometer operating at a magnetic field strength of 7.05 T (.sup.1H frequency=300 MHz) using a 4.0 mm Bruker MAS probe. During the measurement, the sample was undergoing magic angle spinning at a rotational speed of 5 kHz. Quantitative .sup.13C NMR spectra were obtained using a direct polarization method with high power 1H decoupling at 10 kHz MAS. In order to remove signal background, a double acquisition sequence called Elimination of Artifacts in NMR Spectroscopy (EASY) was utilized.

(47) FIG. 4 shows the .sup.13C NMR spectra of four isolated kerogen samples of different maturities. As maturity increases from sample A to D, relative intensities of aromatic carbon signal increase while those of aliphatic carbon decrease. The aromatic fraction, f.sub.Ar.sup.Ker, is listed in Table 1. Such quantitative measurements of the changes of aromatic and aliphatic carbon over a given thermal history (temperature and time) allow derivation of the kinetics of the transformation of kerogen to petroleum fluids, and the subsequent alterations of generated petroleum fluids.

(48) .sup.1H NMR analysis can differentiate rigid .sup.1H signal and mobile .sup.1H signal in a given sample. Rigid .sup.1H signal is typically very broad due to dipolar interaction, whereas mobile .sup.1H is much narrower due to averaging out dipolar interactions. Once generated, the majority if not all of the heavier petroleum fluid is absorbed in the kerogen matrix. The true kerogen fabric is rigid and produces rigid .sup.1H signal, while the petroleum fluids absorbed in the kerogen fabric are mobile and produce mobile .sup.1H signal. For these four kerogen samples, the percent of mobile .sup.1H signal is summarized in Table 1. This stands out as a distinctive advantage of NMR based hydrocarbon generation kinetics analysis over conventional kinetics analyses. Conventional compositional kinetics analyses employ tedious and error prone chemical separation procedures, e.g. solvent extraction, filtration, to separate and determine the amount of generated petroleum fluids versa residual kerogen.

(49) Bulk H:C ratio can be readily obtained from NMR analysis. The H:C ratios for these four kerogen samples are summarized in Table 1. Over all, as the kerogen goes through earlier oil window to late gas window, the H:C ratio decreases, fraction of aromatic carbon increases, and the mobile .sup.1H signal decrease, consistent with observations from conventional kinetics analysis experiments.

(50) TABLE-US-00003 TABLE 1 Percent of mobile (non-rigid) kerogen measured using 5 kHz .sup.1H MAS NMR, aromatic carbon fraction (.sub.Ar.sup.Ker) from 10 kHz .sup.13C MAS NMR, and the H:C ratio measured with ssNMR Sample .sup.1H Mobile % .sup.13C.sub.Ar.sup.Ker H:C A 39.6% 0.60 0.95 B 20.2% 0.75 0.73 C 13.4% 0.80 0.65 D 9.3% 0.87 0.55

(51) The following references are incorporated by reference in their entirety for all purposes: Petsch, et el., A solid state 13C-NMR study of kerogen degradation during black shale weathering, Geochimica et Cosmochimica Acta, Vol. 65, No. 12, pp. 1867-1882 (2001), available online at http://works.bepress.com/cgi/viewcontent.cgi?article=1007&context=steven_petsch Smernik R. J., et al., Assessing the quantitative reliability of solid-state 13C NMR spectra of kerogens across a gradient of thermal maturity, Solid State Nuclear Magnetic Resonance 29 (2006) 312-321, available online at http://www.geo.unizh.ch/mschmidt/downloads/Smernik2005.pdf. Smernik, R. J., Oades, J. M., Geoderma 96 (2000) 159. Smernik, R. J., Oades, J. M., Geoderma 96 (2000) 101. Smernik, R. J., Oades, J. M., Eur. J. Soil Sci. 54 (2003) 103.