GEOLOGICAL FORMATIONS

FIELD OF THE INVENTION

[001] This invention is directed to a method and device for rapid, non-destructive oil detection and quantification in subterranean deposits capable of differentiating between oil and water content in cuttings or core samples taken from subterranean or underwater geological formations using Fast- Neutron Resonance Transmission (FNRT) radiography or tomography.

BACKGROUND OF THE INVENTION

[002] The cheapest and fastest way to evaluate a subsurface formation is to analyse drill cuttings retrieved from the well during drilling. A more detailed, but substantially more expensive and slower operation is coring, or retrieving a cylinder of rock for more detailed evaluation. The routine oil-drilling analysis consists of measuring porosity, permeability, and fluid or gas saturation.

[003] More recent techniques of core analysis include X-ray Computed Tomography (CT) and Magnetic Resonance Imaging (MRI).

[004] Nuclear geophysics is a branch of oil, gas, and uranium exploration employed in both nuclear borehole logging and core sample analysis employing thermal neutron computerized tomography to image the distribution of hydrogenous liquids (oil or water) in the core. However, the method cannot differentiate between water and oil content because of its heavy reliance on hydrogen attenuation of the interrogation stream.

[005] De Beer et al. discloses a method of thermal neutron radiography of porous rocks and iron ore to provide images of internal structure of rocks to determine an effective porosity of the object. A Perfect et al. discloses a thermal neutron imaging of hydrogen-rich fluids in geo-materials and engineered porous medium and the Earth Science Reviews disclose a thermal-neutron imaging of hydrogen-rich fluids in geo-materials and discusses the non-destructive visualization of hydrogen- rich fluids within diverse porous media. Luuoj AS noted, these methods suffer from the inability to differentiate between oil and water content in geological samples and lack desired penetration.

[007] Detection of other elements/isotopes, such as U-235, U-238 and Pu-239 and many other elements has been investigated by a related method "Neutron Resonance Spectroscopy" with low energy (eV, keV) neutrons using the GELINA facility.

[008] Fast Neutron Resonance Transmission (FNRT) using neutrons in the MeV range has been applied in the past for detecting low-Z (light) elements, such as H,C, N, and O, in order to determine composition of agricultural products and detect contraband [see Overley JC, IntJ.Appl.Rad and Istopes 36 (1985) 185, Fink C. L., Micklich B. J., Yule T. J., Humm, P., Sagalovsky L. and Martin M. M., Nucl. Instr. & Meth. B99 (1995) 748, Overley J. C, Chmelnik M. S., Rasmussen R. J , . Schofield R. M. S. and Lefevre H. W., Nucl. Instr. & Meth B99 (1995) 728]. A system for detection of explosives in air passenger bags based on this method has also been constructed and tested, [see Overley J. C, Chmelnik M. S Rasmussen R. J, Sieger G. E., Schofield R. M. S. and Lefevre H. W., SPIE, Vol. 2867 (1997) 219; Miller T. G., Van Staagen P. K. and Gibson B. C, SPIE, Vol. 2867 (1997) 215.]. All the above investigators used accelerator-based, nanosecond- pulsed, broad-energy neutron beams for interrogating the objects of interest.

[009] Un-pulsed beam FNRT method was proposed by Randhers-Pehrson and Brenner as taught in US Patent 5,818,054, as well as Chen G. and Lanza R.C. as taught in US Patent 6,693,291 and utilizes a constant wave (CW) accelerator-based mono-energetic neutron beam to scan contraband from various angles. Moreover, detection of diamonds in kimberlite rocks by FNRT has been proposed by Tapper U.A.S and Over G. A. in PCT publication WO0029873.

[0010] FNRT is disclosed by Gokhale P.P and Hussein E.M.A. in Applied. Radiation. Isotopes 48 (1997) 973 using a ²⁵² Cf isotopic source combined with a spectroscopic ³ He neutron detector for detection of explosives. Luui ij ΠΝΚΤ is also employed by Viesti et al. as taught in Nuclear Instr. and Meth. A593 (2008) 592 and employs a tagged ²⁵² Cf neutron source to demonstrate the possibility of employing time- of-fiight spectroscopy for deriving the chemical composition of organic samples such as explosives, paper or drugs. All of the above FNRT applications rely on elemental attenuation in the absence of apriori knowledge of the particular compound being searched. Since FNRT does not differentiate between various possible chemical bonding for an identified element, there is a need for FNRT techniques providing non-destructive, compound selectivity in the identification and quantification of constituents in geological formations.

SUMMARY OF THE INVENTION

[0012] According to the teachings of the present invention there is provided, a method for determining constituent content in geological samples including capturing a normalizing, transmission spectrum of unattenuated, incident neutron flux with a neutron detector, the normalizing, transmission spectrum characteristic of neutron count as a function of neutron energy ranging between 0.1 MeV to 10.0 MeV, the flux emanating from a neutron source, bombarding a geological sample with interrogative neutron flux, the neutron flux emanating from the neutron source; capturing a transmission spectrum of the interrogative neutron flux traversing the sample with the neutron detector, the transmission spectrum characteristic of neutron count as a function of neutron energy ranging between 0.1 MeV to 10.0 MeV; and

rendering the transmission spectrum into constituent-specific areal densities in accordance with ri = ^constituent attenuations, wherein ri is a natural logarithm of ratio of transmitted-to-incident neutron flux at energy "i" of the geological sample captured by the detector, the constituent attenuations are a(ATa)i +b(ATb)i, wherein "a" is a constituent-specific, areal density of constituent "a", "b" is a constituent-specific, areal density of constituent "b",

(ATa)i, is a compound-specific, mass-attenuation coefficient of constituent "a" at energy "i", and (ATb)i, is a compound-specific, mass-attenuation coefficient of constituent "b" at energy "i". Luui According to a further feature of the present invention, the sum of constituent attenuations includes c(AT _c )i wherein "c" is a constituent-specific, areal density of constituent "c", and

(AT _c )i is a compound-specific, mass-attenuation coefficient of constituent "c" at energy "i".

[0014] According to a further feature of the present invention, the rendering the transmission spectrum of the neutron flux into a constituent-specific areal density includes employing a method for solving an overdetermined system of linear equations.

According to a further feature of the present invention, there is further provided rendering each of the constituent-specific, areal densities into a corresponding weight fraction.

[0015] According to a further feature of the present invention there is further provided capturing a calibration transmission spectrum of a calibrating neutron flux traversing a calibration sample with a neutron detector, the calibration transmission spectrum characteristic of a calibration neutron count as a function of neutron energy ranging between 0.1 MeV to 10.0 MeV, the flux emanating from a neutron source; and rendering the calibration transmission spectrum of the calibrating neutron flux into compound-specific mass- attenuation coefficients as a function of neutron energy in accordance with (AT _a ); = r, ^{callbratlon sample "a"} /a ^callbratlon , (AT _b )i = n ^{Ciilbratlon sample "b"} lb ^caUbratlon , wherein: (AT _a )i, is a compound-specific, mass-attenuation coefficient of constituent "a" at energy channel "i", r; ^{callbmtlon sample a} i _s the natural logarithm of ratio of transmitted-to-incident neutron flux at an energy "i" of calibration

sample "a", "a ^callbratlon " is a known areal density of calibration sample "a", wherein: (AT _b )i, is a compound-specific, mass-attenuation-coefficient of constituent "b" at energy channel "i", r; cal ⁱ brat ⁱ on sample b _{ig the natura} i logarithm of ratio of transmitted-to-incident neutron flux at an energy "i" of calibration sample "b", "b ^callbratlon " j _{s a} known areal density of calibration sample Luuioj According to a further feature of the present invention, a plurality of the at least one interrogative neutron flux of neutrons traverses the geological sample at different angles relative to an axis of the geological sample.

[0017] According to a further feature of the present invention, there is also provided generating a tomography image from the plurality of interrogative neutron flux of neutrons.

[0018] According to a further feature of the present invention, the bombardment is implemented progressively along the longitudinal axis of the sample.

[0019] According to a further feature of the present invention, the neutron source is implemented as an ion-beam accelerator employing a nuclear reaction of ⁹ Be(d,n) ¹⁰ B, ^u B(d,n) ¹² C, or T-T.

[0020] According to a further feature of the present invention, the neutron source is implemented as an electron accelerator employing a photonu clear reaction of Be(y,n) or D(y,n).

[0021] According to a further feature of the present invention, the neutron source is configured to release the neutrons in short bursts of 1-2 ns.

[0022] According to a further feature of the present invention, wherein the neutron source employs an isotopic tagged source.

[0023] According to a further feature of the present invention, the isotopic tagged source includes ²⁵² Cf or Am-Be.

[0024] According to a further feature of the present invention, there is also provided a spectrum of a non-pulsed neutron source.

[0025] According to a further feature of the present invention, the neutron detector includes organic liquid or solid detector so as to facilitate differentiation between neutrons and gamma-rays using Pulse-Shape-Discrimination (PSD).

[0026] According to a further feature of the present invention, the neutron detector is implemented as a noble-liquid detector. Luuz /j According to a further feature of the present invention, the neutron detector includes a ³ He neutron spectroscopic detector.

[0028] According to a further feature of the present invention, the neutron detector includes solid neutron converters coupled to gas-avalanche electron multipliers.

[0029] According to a further feature of the present invention, the neutron detector is position sensitive having a pixelation size of l-5mm.

[0030] There is also provided according to the teachings of the present invention, detection device employing fast-neutron interrogation including a neutron detector; a neutron source configured to direct neutron flux to the detector; and a computer configured to render a received transmission spectrum of the neutron flux by the neutron detector into constituent-specific, areal densities in accordance with ri = ^constituent attenuations, wherein ri is a natural logarithm of ratio of transmitted-to-incident neutron flux at energy "i" of the geological sample captured by the detector, the constituent attenuations are a(ATa)i +b(ATb)i, wherein "a" is a constituent-specific, areal density of constituent "a", "b" is a constituent-specific, areal density of constituent "b", (ATa)i, is a compound-specific, mass-attenuation coefficient of constituent "a" at energy "i", and (ATb)i, is a compound-specific, mass-attenuation coefficient of constituent "b" at energy "i".

[0031] According to a further feature of the present invention, the sum of constituent attenuations includes c(ATc)i wherein "c" is a constituent-specific, areal density of constituent "c", and

[0032] (ATc)i is a compound-specific, mass-attenuation coefficient of constituent "c" at energy "i".

[0033] According to a further feature of the present invention, the computer is further configured to render the transmission spectrum into constituent-specific, areal densities employing a method for solving an overdetermined system of linear equations

[0034] According to a further feature of the present invention, the computer is further configured to render the areal densities into constituent-specific, quantities. According to a further feature of the present invention, the computer is further configured to render a calibration transmission spectrum of a calibrating neutron flux traversing a calibration sample into compound-specific, attenuation coefficients of a calibration substance of known composition and known areal density mass-attenuation coefficients as a function of neutron energy in accordance with (ATa)i = ri calibration sample "a" /a calibration , (ATb)i = ri calibration sample "b" lb calibration, wherein: (ATa)i, is a compound-specific, mass-attenuation coefficient of

[0036] constituent "a" at energy channel "i", ri calibration sample a is the natural logarithm of ratio of transmitted-to-incident neutron flux at an energy "i" of calibration sample "a", "a calibration" is a known areal density of calibration sample "a", wherein: (ATb)i, is a compound-specific, mass- attenuation-coefficient of constituent "b" at energy channel "i", ri calibration sample b is the natural logarithm of ratio of transmitted-to-incident neutron flux at an energy "i" of calibration sample "b", "b calibration" is a known areal density of calibration sample "b".

[0037] According to a further feature of the present invention, the neutron source is implemented as an ion-beam accelerator employing ⁹ Be(d,n) ¹⁰ B, ^u B(d,n) ¹² C, or T-T nuclear reactions.

[0038] According to a further feature of the present invention, the neutron source is implemented as an electron accelerator employing a photonuclear reaction of Be(y,n) or D(y,n).

[0039] According to a further feature of the present invention, the neutron source is configured to release the neutrons in short bursts of 1-2 ns.

[0040] According to a further feature of the present invention, the neutron source employs isotopic tagged source.

[0041] According to a further feature of the present invention, the isotopic tagged source includes ²⁵² Cf or Am-Be. LUU4ZJ According to a further feature of the present invention, the neutron detector includes organic liquid or solid detector so as to facilitate differentiation between neutrons and gamma-rays through Pulse-Shape-Discrimination (PSD).

[0043] According to a further feature of the present invention, the neutron detector is implemented as a noble-liquid detector.

[0044] According to a further feature of the present invention, the neutron detector includes a ³ He neutron spectroscopic detector.

[0045] According to a further feature of the present invention, the neutron detector includes solid neutron converters coupled to gas-avalanche electron multipliers.

[0046] According to a further feature of the present invention, there is further provided a conveyance mechanism configured to convey a geological sample through the detection device.

[0047] According to a further feature of the present invention, the computer is further configured to render a plurality of the at least one interrogative neutron flux of into a tomography.

BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The subject matter regarded as the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention may best be understood with reference to the following detailed description in view of the accompanying drawings in which:

[0049] FIG. 1 is a plot depicting mass attenuation coefficients of silica, oil and water as a function of neutron energy, according to an embodiment;

[0050] FIG. 2 depicts an FNRT irradiation configuration of a rock sample with broad energy spectrum of neutrons and capture of a traversing neutron spectrum with a position sensitive array of neutron detectors, according to an embodiment;

[0051] FIG. 3 A is a schematic view of a detection apparatus, according to an embodiment;

[0052] FIG. 3B is a flow chart of the steps involved in the detection process, according to an embodiment; LUiDJj i . 4 are simulated transmission spectra through a 10 cm thick sandstone sample containing various fractions of water and oil, according to an embodiment;

[0054] FIG. 5 depicts various reconstructed areal density frequency distributions for the sandstone, oil, and water constituents of dry sandstone and dry sandstone with 10% oil from the simulated transmission spectra of FIG. 4, according to an embodiment;

[0055] FIG. 6 are bar graphs depicting reconstructed weight percentages of oil, water constituents of the various simulated transmission spectra of FIG. 4, according to an embodiment;

[0056] FIG. 7 are bar graphs depicting weight percentages of oil, water constituents derived from a simulated transmission through a 7.8 cm thick sand sample, according to an embodiment;

[0057] FIG. 8 depicts experimental neutron time-of flight spectra transmitted for various sample types, according to an embodiment;

[0058] FIG. 9 are plots depicting experimentally determined mass attenuation coefficients for dry sand, car engine oil and water, according to an embodiment;

[0059] FIG. 10 depicts various plots depicting experimentally determined areal density frequency distributions of sand, oil and water based on mass attenuation coefficients of Fig. 9; according to an embodiment; and

[0060] FIG. 11 depicts several bar graphs of experimentally determined weight percentage of oil and water in sand samples employing the areal densities of FIG. 10; according to an embodiment.

DETAILED DESCRIPTION

[0061] In the following detailed description, the details are set forth in order to provide a thorough understanding of the invention. It should be understood by those skilled in the art that the present invention may be practiced without these specific details. Furthermore, it should be appreciated that well-known methods, procedures, and components have been omitted, the figures are not necessarily drawn to scale all for the sake of clarity. Additionally, reference numerals repeated in multiple figures are done so to depict corresponding or analogous elements. Luuozj AS noted, the present invention relates to the detection and quantification of non-gaseous- hydrocarbon and water content in geological samples without distortion from water content using Fast-Neutron Resonance Transmission (FNRT). It should be appreciated that such FNRT techniques can also be applied to the identification of other non-gaseous materials.

[0063] FNRT is a method that exploits characteristic structure resonances in the neutron attenuation coefficients of the constituents of an inspected object and can therefore be exploited to determine their identity and proportion within an interrogated object.

[0064] FIG. 1 depicts mass attenuation coefficients of various constituents of a sandstone core sample containing silica, oil, and water as a function of neutron energy ranging between 1-10 MeV obtained from evaluated nuclear data file (ENDF) 2015.

[0065] As shown, attenuation coefficients of each of the three substances exhibit different behaviours with neutron energy due to resonances of the most abundant elements in the constituents; carbon in oil, oxygen in water, and oxygen plus silicon in silica. Accordingly, the dips and peaks at specific energies are characteristic of each of constituent and carries composition information of the sample.

[0066] FIG. 2 depicts an irradiation configuration 10 of fast neutrons 20 bombarding a geological sample 24 of sandstone 21 containing oil 22 and water 23 and a fast-neutron detector 25 configured to receive at least one neutron flux. It should be appreciated that in a certain embodiment neutron flux 20 is implemented as a poly-energy beam whereas in another embodiment neutron flux 20 is implemented as a plurality of mono-energy beam directed at sample 21 at various angles to enable detection of a collective poly-energy transmission spectrum.

[0067] The geological sample 21 can be, inter alias, an oil-drilling core or a drilling cuttings. Without diminishing in scope these parameters will be discussed in regards to sandstone and nongaseous hydrocarbon, sandstone and water, plus sandstone, water, and oil, however, this detection process can be applied to samples including less than or more than three constituents or having different constituents.

[0068] In a certain embodiment, fast-neutron detector 25 is pixelated between l-5mm.

[0069] A transmission spectrum of inspected sample 21 is rendered into constituent content in accordance with a transmission relationship in which the ratio R; of transmitted-to-incident neutron flux at energy "i":

Ri = ε ^χ ρ[-(μ?ρ ₅ ^χ + μ?ρ ₀ ^χ + μ Ρ _ν *)]

Wherein μ-', μ°, μ™ are compound-specific, mass-attenuation coefficients characterizing the extent in which the fast neutron beam in the MeV energy range is attenuated as it passes through a particular compound. Compound-specific mass-attenuation coefficients are unique to each compound and are determined experimentally after a normalization and calibration spectra have been generated as will be further discussed.

[0070] p _s x, p ₀ x, p _w x are compound-specific areal densities averaged over trajectory A over a thickness "x" of sample 21

[0071] Lateral non-uniformity in sample within an area subtended by a detector pixel composition may be corrected using the methods taught by Hampel U. and Wagner M. found in Meas. Sci. Technol. (2011) 22, 115701 and are hereby incorporated by reference.

[0072] Since the spectrum has many discrete energies "m", a transmission equation is generated for each energy level. The natural logarithm r; of transmission equation R; provides a set of "m" linear equations where the areal densities are the unknown variables of interest at each channel or energy level " i". This over-determined system of "m" linear equations of with three unknowns; the compound-specific areal densities for sandstone " p _s x", oil "p ₀ x", and water "p _w x" spanning sample height "x". In a certain embodiment, the linear equations are solved using a least squares solution. In another embodiment, Bayesian minimization is employed as described by Mor I. et al in "Reconstruction of Material Elemental Composition Using Fast Neutron Resonance Kaaiograpny" available in Physics Procedia 69 (2015) pages 304 - 313 and is hereby incorporated by reference. It should be appreciated that other methods of solving such over-determined systems are included within the scope of the present invention

[0073] Once a solution for the three areal densities is found in a given detector pixel, the areal- density-ratio of oil or water to that of sandstone can be determined to yield the weight fraction of oil or water in the traversed core, independent of sample thickness or shape.

[0074] Alternatively, by multiplying each pixel areal density by a pixel area produces the mass of each component in a volume defined by pixel area and sample height "x" and integrating over all pixels produces the total weight of oil, water, and dry rock in the entire interrogated sample, from which the average weight fractions of oil and water in the sample is determined.

[0075] Three-dimensional images of oil or water distribution in a sample can be generated from multiple views derived from irradiating the sample from 36-180 angles and applying computer implemented tomography reconstruction techniques.

[0076] As noted above, values of compound- specific mass attenuation coefficients as a function of neutron energy must be determined experimentally for a given system using calibrated standards of pure dry rock, oil and water.

[0077] A compound-specific mass- attenuation coefficient (AT _b )i for compound "b" at neutron energy "i" is generated as r, ^{callbratlon sample "b"} lb where r, ^{callbratlon sample "b"} is the natural logarithm of ratio transmitted-to-incident neutron flux at energy "i" for the calibration sample and "b" is a known areal density of constituent "b" also for the calibration sample. In an analogous manner, compound-specific mass-attenuation coefficients are calculated for each of sample constituents.

[0078] The method measures the average liquid/dry-sample weight ratio in the path traversed by the fast neutrons regardless of the object shape, thickness or distribution. Neutron scans can be implemented in a time period between 2-10 minutes thereby advantageously allowing screening of relatively long core samples of about three meters in a certain embodiment Luu/yj ine fluid-weight fractions of the sample are determined independently thereby advantageously providing an oil-to-rock weight-ratio estimation independent of water content. In contrast, conventional thermal neutron radiography water and oil content estimations are necessarily correlated to each other because the dominant attenuation by hydrogen is common to both hydrocarbons and water and therefore reduces or negates constituent specificity.

[0080] Fast neutrons emitted from the source are usually accompanied by high energy gamma-rays in MeV range that can advantageously augment constituent content information through gamma-ray radiography or tomography.

Detection Apparatus

[0081] Fig 3 depicts an embodiment of a detection apparatus 30 including a neutron source 26 from which a broad-energy, fast-neutron flux emanates from nuclear reactions, a conveyer 27, position sensor 25 and a computer 28 linked to detector 25 and neutron source 26, according to an embodiment.

[0082] Computer 28 is configured to render transmission data into constituent content estimates as described above. In a certain embodiment, computer 28 is also configured to control neutron source 26 and conveyer functionality. As shown, interrogated sample 24 like drilling cores or cuttings are conveyed on conveyor 27 while fast-neutron source 26 collimated to produce a neutron beam is viewed by spectroscopic fast- neutron imaging detector 25 having a pixelation size of between 1-5 mm.

[0083] Neutron source 26 is implemented as a nanosecond pulsed accelerator, such as a pulsed cyclotron producing 1-2 ns bursts of 12 MeV deuterons impinging on a thick beryllium target. The neutrons are produced by the Be(d,n) reaction and its useful spectrum extends up to neutron energy of 12 MeV, in a certain embodiment.

[0084] In another embodiment, neutron source 26 is implemented as a broad-spectrum neutron source is a pulsed T-T neutron source, capable of providing a flat neutron spectrum up to 9 MeV. LUU83j in another embodiment, neutron source 26 is implemented as a variable-energy quasi-mono- energetic beam is employed so that continuous neutron energy variability is achieved by directing the neutron beam to detector 25 from different angles.

[0086] Isotopic neutron sources, such as ^{^} Ci and "'Am-Be can also be used by applying the tagging Time of Flight (TOF) technique disclosed by Viesti et al. found in Nuclear Instr. and Meth. A593 (2008) 592, and also disclosed by Scherzinger J. et al disclosed in Applied Rad. and hot. 98 (2015) 74. Both of these are disclosures are hereby incorporated by reference in their entirety.

[0087] Neutron detector 25 is implemented in a certain embodiment as a high position-resolution linear array or a two dimensional array of fast-neutron detectors, such as liquid or plastic organic scintillators operated in integrating mode as set forth in WO2006048871 and is hereby incorporated by reference.

[0088] Alternatively, a two dimensional array of fast- neutron detectors is configured to operate in the event counting mode as taught by Schossler et al. in the article "Time and Position Sensitive Single Photon Detector for Scintillation Readout' available in the Journal of Instrumentation 7(02) (2012) C02048, and also taught by Brandis M. in Ph.D. thesis "Development of Gamma-Ray Detector for Z-Selective Radiographic Imaging" at Hebrew University of Jerusalem, November 2013.

[0089] In another embodiment, detector 25 is implemented as a noble-liquid scintillating detector as described by Breskin et al. in the article "A Novel Liquid Xenon Detector Concept for Combined Fast-Neutrons and Gamma-Imaging And Spectroscopy" available in the Journal of Instrumentation, vol.7 (2012) C06008. All of these teachings are hereby incorporated by reference in their entirety.

[0090] It should be appreciated that detector 25 can also be implemented as a hybrid converter or gas-avalanche detector, or any other fast-neutron imaging detector, in other embodiments. LUimj ΠΝΚΤ requires a precise measurement of the incident and transmitted neutron energy spectrum to establish the neutron energy attenuation.

[0092] In a certain embodiment, Time of Flight (TOF) neutron spectroscopy is employed in which a nanosecond pulsed broad-energy neutron beam or a tagged neutron source is used. In another embodiment, fast- neutron spectroscopy using unfolding methods, as is known in the art.

[0093] In another embodiment, neutron telescope spectroscopy is employed.

[0094] FIG. 3A is a flow chart 40 of processing steps employed in the detection and quantification of constituents in a geological sample, according to an embodiment.

[0095] Prior to any sample irradiation, in step 41, transmission spectra are normalized to a flat spectrum using 2- 10 ⁸ incident neutrons in the absence of a sample to establish a baseline transmission spectrum for comparison with future transmission spectra. The detector provides number of neutron counts as a function of either neutron energy or time-of-fiight. It should be appreciated that other quantities of neutrons providing such functionality are also included with the scope of the present invention.

[0096] In step 42 during calibration, a geological calibration sample of known constituent content is irradiated with a neutron flux.

[0097] In step 43 a spectrum of neutrons is captured by a detector after traversing the calibration sample.

[0098] In step 44, the spectrum of neutrons is used to generate compound-specific attenuation coefficients as a function of neutron energy to be used later when determining constituent content of geological samples as set forth above.

[0099] Sample analysis commences in step 45 with the bombardment of a geological sample with a neutron flux.

[00100] In step 46 a spectrum of neutrons of the flux captured by the detector after traversing the sample is rendered into a transmission spectrum as a function of neutron energy. A precise measurement of incident and transmitted neutron energy spectrum is achieved through Time of Flight (TOF) neutron spectroscopy when a nanosecond pulsed broad-energy neutron beam or a tagged neutron source is used. In another embodiment, fast-neutron spectroscopy using unfolding methods, as is known in the art. In another embodiment, neutron telescope spectroscopy is employed.

[00101] In step 47, the spectrum of neutrons is used to generate constituent-specific areal densities as explained above.

[00102] In step 48, constituent-specific content estimations are generated from the areal densities as explained above.

[00103] In step 49, the content estimations are displayed either graphically on a display screen or textually, or both in accordance with the embodiment.

Method Simulations

[00104] Computer implemented Monte-Carlo simulations were performed to validate the proposed method using a series of GEANT 4 Monte-Carlo as taught by Agostinelli et al, Geant4-A Simulation Toolkit available Nucl. Instr. and Meth. A 506, 250-303).

[00105] These simulations were performed for several configurations of formation materials containing various quantities of oil, water, and a mixture of oil plus water. A simulated incident neutron beam with uniformly distributed energy spectrum ranging from 2-10 MeV, impinged on a core sample. Various quantities of uniformly distributed oil, water and a combination of oil and water were added to the sample. The simulated core materials were dry sandstone (DS-bulk density 2.3 g/cc, 10 cm thick) and sand having a bulk density of 1.55 g/cc a sample thickness of 7.8 cm. The latter material was simulated in order to compare to experimental results which were carried out with sand and will describe later in this document. Luuiuoj in order to test the influence of counting statistics on the reconstructed, simulated values, the Monte-Carlo simulation of the 10 cm thick sandstone sample was performed for 3 different numbers of incident neutrons: N= 10 ⁷ , 10 ⁶ and 10 ⁵ neutrons.

[00107] Simulations of neutron transmission through pure standard materials for dry sandstone , oil, and water were performed with 10 ⁸ incident neutrons to determine their respective mass attenuation coefficients

[00108] The simulated attenuated neutron spectra were normalized to a flat spectrum using 2- 10 ⁸ incident neutrons without a sample in the neutron beam to establish a baseline transmission spectrum for future transmission spectra.

Analysis of Simulated 10 cm Thick Sandstone Samples

[00109] FIG. 4 shows the simulated transmission neutron spectra for the 10 cm thick Dry

Sandstone (DS) core containing various per weight percentages of oil or water; (DS), DS+10 oil,

DS+20 oil, DS+10% water and DS+10% oil+5% water. The total number of incident neutrons impinging on the sample was 10 ⁷ . The simulated detected number of neutrons ranged from 1.3- 10 ⁶ to 3- 10 ⁶ counts. As shown, the spectrum is dominated by the shape of the DS spectrum; nevertheless, the proportions of the various peaks are different for each composition. A total number of neutrons impinging on sample was 10 ⁷ .

[00110] Reconstruction of simulated sample composition followed the procedure of Mor et al, available in Physics Procedia 69, 2015, 304-311 , employing a Bayesian analysis. The procedure of Mor et al is hereby incorporated by reference in its entirety. The analysis was carried out using the software WinBUGS by Lunn et al., WinBUGS-A Bayesian modeling framework, available in Statistics and Computing 10, 2000, 325-337, an interactive Windows version of the BUGS program (Bayesian inference Using Gibbs Sampling) developed by the Medical Research Center (MRC) and Imperial College of Science, Technology and Medicine, UK.

[00111] The Bayesian analysis provides a probability distribution of the areal density for each constituent and indicates the likelihood of its presence in the inspected material at the most probable areai aensity. Substances which are likely to be present in the inspected sample have probability distributions that are Gaussian in shape and exhibit standard deviations within a few percent of the mean, whereas the frequency distributions of areal density for substances which are not likely to be present in the inspected sample, have probability densities that peak at or around zero and are skewed to the extent that the standard deviation is of the order of 30 % of the mean or higher as shown in the plots of FIG. 5.

[00112] FIG. 5 shows the resulting Bayesian reconstruction in arbitrary units (AU) as a function of areal density of simulated areal density distributions of dry sandstone, oil and water for two simulated configurations. The distributions in the left column are various dry sandstone (DS) core samples while those in the right column are reconstructions for cores containing 10% oil by weight.

[00113] As evident from the distributions, the sample on the left does not contain any liquid, while the distributions on the right indicate that a significant amount of oil (standard deviation: 3.5%) is present and suggest a small likelihood for water-(standard deviation: 38%) as well. Table 1 provides a summary of reconstructed areal densities and their standard deviations for all sandstone core configurations.

Table 1: Summary of reconstructed areal densities of sandstone core configurations N=10 ⁷ neutrons per simulation

Configuration DS [g/cm ² ] Oil [g/cm ² ] Water [g/cm ² ]

Pure DS 22.99±0.06 (23*) 0.01+0.01 (0*) 0.02±0.07 (0*)

DS+10% oil 22.8310.12 (23*) 2.2210.04 (2.3*) 0.15+0.28 (0*)

DS+20% oil 22.97+0.12 (23*) 4.58+0.04 (4.6*) 0.07+0.05 (0*)

DS+10% water 23.20+0.11 (23*) 0.07+0.05 (0*) 2.16+0.07 (2.3*)

DS+(10% oil+5%H ₂ 0) 23.05+0.13 (23*) 2.3+0.05 (2.3*) 1.12+0.08 (1.15*)

*Simulation input value

[00114] The ratio of reconstructed areal density of oil (or water) to that of DS yields the weight percentage of each fluid in the core. Fig 6 shows the percentages by weight of oil and water in each configuration. Luui 13 j i e influence of counting statistics on the reconstructed values of fluid content is shown in Table 2.

Table 2 Reconstructed % fluid in sandstone vs. number of incident neutrons per simulation No. of neutrons —► 10 ⁷ neutrons 10 ⁶ neutrons 10 ⁵ neutrons

[00116] The absence of fluids in dry sandstone is always determined with high accuracy. However, for samples containing fluids, both the accuracy and precision of the reconstructed fluid content deteriorate when the number of incident neutrons is reduced below 10 ⁶ . Based on this study, it is preferably to examine such samples with at least 10 ⁶ incident neutrons integrated over the entire spectrum.

Analysis of a Simulated 7.8 cm Thick Sand Sample

[00117] Further experimentation was performed on sand samples using the above described detection apparatus for 7.8 cm thick, dry sand sample having a bulk density 1.55 g/cc and also for mixtures of dry sand containing 10% or 20% oil and 10% water by weight; each sand sample was contained in a box and approximates an analysis scenario encountered when analyzing drill cuttings or oil sands on site. Luui isj ine total number of incident neutrons for each case was 10 ⁷ . Table 3 summarizes the results obtained in this simulation and FIG. 6 shows the percentages by weight of oil and water in each configuration.

Table 3: Summary of reconstructed areal densities of 7.8 cm sand samples with a neutron count of N=10 ⁷ per simulation

Configuration S [g/cm ² ] Oil [g/cm ² ] Water [g/cm ² ]

Pure Sand (S) 12.09±0.04 (12.09*) 0.006±0.005 (0*) 0.009±0.008 (0*)

S+10% oil 12.0010.05 (12.09*) 1.21±0.02 (1.21*) 0.03±0.03 (0*)

S+20% oil 12.2210.05 (12.09*) 2.4110.02 (2.42*) 0.02+0.02 (0*)

S+10% water 12.18+0.06 (12.09*) 0.036+0.023 (0*) 1.15+0.04 (1.21*)

S+(10% oil/5%H ₂ O) 12.09+0.07 (12.09*) 1.22+0.03 (1.21*) 0.6+0.05 (0.605*)

* Simulation input value

[00119] FIG. 6 are bar graphs depicting simulated weight percentage of oil and water in 7.8 cm thick sand sample for dry sand (S), S+10% oil, S+20% oil, S+10% water and S+(10% oil+5% water).

[00120] As shown, the results of reconstruction agree very well with the known content values of the sample.

[00121] FIG. 7 are bar graphs depicting simulated weight percentages of oil, water constituents derived from a 7.8 cm thick sand sample.

Experiment

[00122] An experimental procedure was performed in which neutrons were produced by a cyclotron-produced 12 MeV deuterium beam impinging on a 3 mm thick Be target. The beam was pulsed at a pulse repetition rate of 2 MHz frequency and pulse width of 1.7 ns. Average beam current was approximately 2 μΑ. The useful part of the neutron energy spectrum ranges from ca. 1 MeV up to 10 MeV as taught by Brede et al in Nucl. Instr. and MethA9S9, 21 A, 332-344. Neutron spectroscopy was performed by the time-of-flight (TOF) method using a cylindrical 2"x2" liquid scintillator detector (NE213 type) positioned at 11.5 m from the target. Luuizjj ¾mce a large non-pixelated detector was used, the samples were homogenously mixed and of constant thickness. They consisted of thin-walled, rectangular metal boxes 7.8x10.9x9.0 cm ³ filled with dry sand having a bulk density of 1.55 g/cc, pure used motor oil having a viscosity of 10W40 and a density of 0.857 g/cc, pure water, or mixtures of dry sand containing 10.4%, 22.4% oil and 11.1% water by weight. The sample thickness traversed by the neutron beam was 7.8 cm. The distance between the sample and the detector was 65 cm. The acquisition time for the runs ranged from 1000 to 1200 s/run.

[00124] Pure samples of dry sand, oil and water were used to determine the mass attenuation coefficients of each substance required for reconstruction.

[00125] FIG. 8 shows experimental TOF spectra of sample free, dry sand (DS), a mixture of DS+10.4% by weight motor oil, DS+22.4% oil and DS+11.1% water. The sharp peak on the left side of the spectra is the gamma-flash due to gamma-rays emitted from the target. The total number of detected neutrons (after subtracting the gamma-ray peak counts from the total) was 6- 10 ⁷ and 1.7-3· 10 ⁷ without and with samples respectively. As expected, the TOF spectra transmitted through the samples are dominated by the silica features. The spectra of DS+10.4% oil and DS+11.1% water appear to be quite similar in most parts of the spectrum, except for the peaks around channels 450 and 730. The latter correspond to carbon resonances at around 3 MeV.

[00126] FIG. 9 depicts experimental transmission TOF spectra of dry sand, pure oil and water used to determine the experimental, compound-specific mass attenuation coefficients. Although the sharp resonances are less pronounced here than in the ENDF data of FIG. 1, the general features are similar.

[00127] FIG. 10 shows the reconstructed experimental areal density distributions of sand, oil and water in each sample. The distributions of the pure sand, sand +22.4% oil and sand+11.1% water follow quite well the expected behavior and areal densities that peak at or around 0.0 g/cm ² and are skewed for the non-existent component), however the water distribution of the sand +10.4% oil sample snows an unexpected pattern, as if a significant amount of water is present in the sample, although the large error of ± 24% in this result compared to typically 5-8% uncertainty for an existing component may indicate that the reconstructed water density value might be consistent with zero, within statistical errors.

[00128] Table 3 summarizes the results obtained in the experiment and FIG. 11 shows the empirical percentages by weight of oil and water in each configuration.

Table 3: Summary of reconstructed areal densities of sand samples Configuration Sand [g/cm ² ] Oil [g/cm ² ] Water [g/cm ² ]

Pure Sand 12.04±0.07 (12.04*) 0.01+0.008 (0*) 0.018±0.015 (0*)

S+10.4% oil 11.05±0.19 (12.04*) 1.11±0.073 (1.3*) 0.55±0.13 (0*)

S+22.4% oil 11.2410.24 (12.04*) 2.4810.12 (2.7*) 0.33±0.21 (0*)

S+ll.1% water 12.6510.157 (11.9*) 0.07310.055 (0*) 1.29±0.11 (1.32*)

*Expected value based on measured weights and volumes of the dry and liquid saturated samples

[00129] While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes, and equivalents will now occur to those of ordinary skill in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.