AU602656B2 - Quantitative determination by elemental logging of subsurface formation properties - Google Patents

Description

I

AUSTRALIA

PATENTS ACT 195602656 Form COMPLETE SPECIFICATION

(ORIGINAL)

FOR OFFICE USE Short Title: Int. Cl: Application Number: Lodged: da Complete Specification-Lodged: Accepted: Lapsed: Published: 'This document contains the amendments made under Sertion 49 and is correct for printing.

Priority: Related Art: TO BE COMPLETED BY APPLICANT Name of Applicant: Address of Applicant: SCHLUMBERGER TECHNOLOGY, B.V.

9* KONINGINNEGRACHT 2514 AB THE HAGUE

NETHERLANDS

p Actual Inventor: Address for Service: CLEMENT HACK CO., 601 St. Kilda Road, Melbourne, Victoria 3004, Australia.

Complete Specification for the invention entitled: QUANTITATIVE DETERMINATION BY ELEMENTAL LOGGING OF SUBSURFACE FORMATION PROPERTIES The following statement is a full description of this invention including the best method of performing it known to me:- I

I

-^rs 60.710-CIP2 QUANTITATIVE DETERMINATIONBY ELEMENTAL LOGGING OF SUBSURFACE FORMATION PROPERTIES BACKGROUND OF THE INVENTION 9* The present invention relates generally to investigating earth formations traversed by a borehole. More particularly, the invention relates to methods for determining values for and for further characterizing the attributes of a formation surrounding a borehole by processing well logging data derived by lowering in the borehole one or more apparatus for investigating subsurface earth formations. The ability to assign values for quantify) and further characterize formation attributes formation minerals) permits a wide range of new and improved results to be obtained through logging, including a direct calculation of cation exchange capacity (CEC) and a corrected water saturation (Sw) determination, an oil API gravity log, an improved grain density and porosity determination, an improved understanding of depositional environment, a permeability log, a log of mean grain size, a log of thermal neutron capture cross-section and a log of formation elements not directly measurable, among others.

Quantitative knowledge of the lithological constituents present in a well as a function of depth would be valuable in assessing all aspects of exploration, evaluation, production, and completion. A complete shaly sands lithological description must go beyond simple discrimination between "sands" and "shales" and, for example, establish the quantity of clay minerals in all layers including so-called "clean sands", identify and quantify the non-clay as well as clay minerals prosent, and identify subtle and pronounced changes in depositional or diagenetic facies by characterizing the formation minerals.

Until now, it has been generally accepted that there are no successful techniques available for-taking elemental chemical data and deriving therefrom a quantitative mineralogical analysis of the lithology in question.

Some procedures have been posed for gaining limited knowledge of lithology from chemical data, and different logging tools have been utilized to provide elemental data and indicators, but none of the previously known procedures or tools, alone, or together, have been capable of broadly and accurately permitting quantitative mineralogical analysis from logging. The known techniques and procedures generally only address the derivation of o o oo I i' -2- 60.710-CIP2 particular outputs such as water saturation, porosity, carbon/oxygen ratios, cation exchange capacities, general lithology classifications, etc.

Examples of borehole tools which provide and determine elemental chemical data and yields include natural gamma ray tools, induced gamma spectroscopy tools and high resolution spectroscopy tools. The natural gamma ray tools typically comprise a scintillator and pulse height analyzer which respond to and measure the gamma ray activity due to the decay in an earth formation of the naturally radioactive elements: thorium, uranium and potassium. In the past, the thorium plus potassium content has been used as an indication of clay or shale content. Uranium amounts have been suggested to indicate organic carbon sources and to provide information regarding secondary porosity detection and fractures. See e.g. U.S. Patent No. 4,071,755 issued to Supernaw et al. As detailed by Lock, G. A. and Hoyer, W. "Natural Gamma-Ray Spectral Logging," The Log Analyst, September-October 1971, pp. 3-9, a thorium/uranium ratio may in some instances provide insight into the type of marine environment encountered.

A potassium percentage determination may provide in some instances an indication of potash deposits or micaceous sands.

Induced gamma ray spectroscopy tools typically utilize a pulsed deuterium-tritium accelerator neutron source and sodium iodide detectors, which detect the gamma rays resulting from the interaction of the source neutrons with the formation elements. As disclosed in U.S. Patent No.

3,521,064 issued July 21, 1970 to Moran and U.S. Patent No. 4,055,763 to Antkiw, the spectroscopy tools can be run either in an inelastic or an activation mode and provide elemental yield information on hydrogen, chlorine, silicon, calcium, iron, oxygen, carbon and sulfur. Using various "I ratios of the determined elements, indicators such as fluid salinity, porosity, shaliness, lithology and oxygen activation, among others, may be determined.

High resolution spectroscopy tools are based on the same principles as the induced gamma ray spectroscopy tools except that the accelerator neutron source may be replaced, if desired, by a chemical source, and the detectors utilized are high resolution (such as high-purity germanium) detectors. The high resolution (or enhanced resolution) spectroscopy tools (see Everett, Herron, M. and Pirie, "Log Responses and Core Evaluation Case Study Technique Field and Laboratory Procedures" -3- 60.710-CIP2 SPWLA 25th Annual Logging symposium, June 27-30, 1983 pp. 23-24), may be used to determine both the amounts of the more abundant formation elements such as those determined by the induced gamma ray spectroscopy tools, and the amounts of less abundant elements such as aluminum, vanadium, magnesium, sodium, etc.

From the information gathered by the tools disclosed above, as well as other tools known in the art including electrical resistivity tools, sonic exploration tools, and other nuclear tools such as the gamma-gamma (formation density tool), or neutron-neutron (neutron i orosity tool) tools, many attempts have been made to comprehensively evaluate and interpret lithology, including systems for two-mineral interpretation and shaly sands interpretation. Some systems such as SARABAND and CORIBAND S. (registered trademarks of Schlumberger Technology Corporation, described respectively in Poupon, A. et al., "Log Analysis in Formations with Complex Lithologies", J. Pet. Tech. (Aug. 1971) pp. 995-1005 and Poupon, A. et al.

"Log Analysis of Sand-Shale Sequences A Systematic Approach" J Pet.

Tech. (July, 1970), correct porosity and resistivity logs for borehole and mudcake effects and then correct for the influence of clay, and/or shale content, and the effects of light hydrocarbons, etc. before computing porosity, matrix density, water saturation, movable hydrocarbon saturation, etc.

Other techniques for shaly sand interpretation include the Waxman-Smits approach in which clay conductivity is used for a determination of water saturation. Clay conductivity is expressed in terms of cation exchange capacity (CEC), per unit volume, Qv. However, as shown in Burck, Lockhart, "A Review of Log and Core Methods for Determining Cation Exchange Capacity/Qv", Transactions of the Eighth European Formation Evaluation Symposium (London, England March 14-15, 1983), unless constant minerology and salinity are assumed, conventional logging cannot provide a satisfactory determination of Qv. Moreover, the Wasman-Smits approach cannot be said to provide a comprehensive evaluation and interpretation of lithology.

Another approach to lithology evaluation has been to analyze formations through core analysis. Thus, core analysis has been used to determine CEC or Qv. A summary of the different core measurement techniques is provided in the aforementioned Burck article including both destructive (pulverizing) and non-destructive techniques. In addition, core

K

4 1--1 i -4- 60.710-CIP2 o

A..

A.

analysis has been utilized in conjunction with logging to correlate radioactive elements to cation exchange capacity. In U.S. Patent No.

4,263,509 issued on April 21, 1981 to Fertl et al., it was suggested that the cation exchange capacity determined by the laboratory testing of a cored borehole could be correlated to a function of the natural gamma rays detected by logging the said borehole. Natural gamma ray logging operations in subsequent boreholes within the same geological region would then provide, in conjunction with the predetermined function, an in situ estimation of the depth related cation exchange capacity of the subsequent borehole. Such a technique is of limited utility, however, because cation exchange capacity is being correlated to elements which generally have little global relation to the clay minerals which dictate cation exchange capacity.

Core analysis has also been used by geochemists in the analysis of depositional environments. One analysis technique is called "factor analysis" and is extensively described in Joreskog, Klovan, J.E. and Reymont, Geological Factor Analysis, Elsevier Scientific Publishing Company (-Amsterdam, the Netherlands 1976). Factor analysis is a technique which can be used in geochemistry to take multiple data sets of variables such as elemental concentrations and to correlate and anticorrelate the variables such that the subject rock or formation can be described with a good degree of certainty by a small number of independent factors which can be identified. Factor analysis has been used in the past to correlate elements to desired outputs such as aerosol sources and air pollution. Thus, the detection of an increase in the abundance of the element lead would indicate increased local usage of fossil fuels. In such a correlation, score analysis is utilized to determine how the magnitude of the factors changes from sample to sample.

Factor analysis was used in conjunction with regression analysis in Tardy, Yves, Element Partition Ratios in Some Sedimentary Environments.

Sci. Geol. Bull. 28, 1, p. 59-95 (Strasbourg, 1975), to classify a formation and to solve for the distribution of trace elements among the classified fractions of a rock. Thus for example, in a particular core sample set, by factor analysis, forty variables were correlated such that four groups (rock fractions) were identified: detrital, sulfide, phosphate (apatite) and organic carbon. Through the use of regression analysis, the distribution in ppm of the trace elements among the four groups was determined. Also, by L -1

J\

5 *4.

C

S.

S

S..

S

'S

analysing results from twenty-one other sets of shale and sandstone core samples, a study of the occurrence of trace elements in identified principal rock fractions was accomplished with the resulting conclusions that environmental conditions such as weathering, deposition and diagenesis might be determinable from a determination of trace elements in the rock formation.

While the interpretation of logging results and of core data have provided many useful outputs to help describe and 10 evaluate lithology, no techniques have been provided which can permit a comprehensive and accurate analysis of a formation by determining from initial log inputs the values for or quantities of the formation attributes such as formation minerals.

It is therefore an object of the invention to provide methods for taking log data as input and providing values for the attributes of the formation under investigation.

It is a further object of the invention to provide methods for taking log data as input and providing a mineralogical 20 analysis including both a quantitative determination and a characterisation of the minerals in the formation under investigation.

SUMMARY OF THE INVENTION According to this invention there is provided a method for 25 determining the concentration at a given depth in a formation of each of a selected group of minerals wherein said group comprises one or more of kaolinite, illite, and feldspar, and wherein at least one mineral of said group is distinguishable by degree of crystallinity, said method comprising the steps of: determining by borehole logging the concentration of each of a selected group of elements at said depth wherein said group of elements comprises iron, potassium and at least one trace element; and applying an element/attribute transform to the determined concentration of elements at said depth, wherein the components of said transform are related to the concentration of each of said elements in *J S p. 5 S S I; _L141v' i -6 each of said minerals, to thereby determine the concentration of each of said minerals at said depth, including said mineral distinguished by its degree of crystallinity.

Preferably the element/attribute transform is the inverse of a starting matrix the components of which are numerical values representing the concentration of each of said elements in each of said minerals.

BRIEF DESCRIPTION OF THE DRAWINGS Additional objects and features of the invention will S. become more apparent upon consideration of the following detailed description of preferred embodiments when taken in •co conjunction with the accompanying drawings wherein: FIG. 1 is a flow diagram representing the invention for determining mineral quantities and further characterising the :e formation minerals from logs.

FIG. 2 is a flow diagram representing a modified way of deriving an element-minerals transform matrix.

FIG. 3 is a log depicting the absolute percentage of minerals in an examined borehole.

FIG. 4 depicts a porosity log, a matrix density log, and a refined porosity log.

25 FIG. 5 is a cross-plot of the invention-derived CEC measurements with those of the laboratory-derived CEC measurements.

FIG. 6a through 6h are cross plots of laboratory-derived mineral quantity measurements versus log-derived mineral quantity measurements.

FIG. 6i is a cross plot of laboratory-derived fines measurements versus log-determined fines measurements.

DETAILED DESCRIPTION Turning to FIG. 1, a method is provided in flow diagram format for taking logging data input from a borehole, and g klerefrom determining

C!

R 'I -7- 60.710-CP2 values for and further characterizing formation attributes as a function of borehole depth to provide valuable information for exploration, evaluation, production, and completion of the borehole and oil field.

The invention provides an element-mineral transform matrix (1) which transforms the concentration of elements into the weight percentage of the minerals. is a column matrix of mineral abundances. is a column matrix of elemental concentrations. is the inverse of an end member composition matrix as explained hereinafter.

While the invention is described as using an element-mineral transform matrix, those skilled in the art will recognize that the invention simply requires a transform. Thus, simultaneous equations may be used in S* lieu of a matrix.

The specific manner in which the values of the components of the transform (such as matrix 1 are determined forms no part of the 'invention, which can be successfully practiced using values quoted hereinafter in Tables 1 to 5. However, for the sake of illustration a brief description of ways of formulating the transform will be given.

The inventor hereof believes that the chemical composition of "sedimentary minerals of the same degree of crystallinity, at least, is uniform, S" irrespective of their geographical origin. Thus, for example, it appears that all medium-crystallinity illites anywhere in the world have the same chemical content. This hypothesis, called geochemical uniformity, is contrary to established geochemical theory, in which the chemical composition of minerals, and especially clays, is considered to be extremely ,variable. It now appears that much of the variability seen in laboratory measurements can be ascribed to sources other than variation in chemical ,4 content, in particular to substantial errors (frequently 50%) in clay analyses and to the use of mineral 'standards' which are in reality mixtures of several different mineral phases.

Thus, as depicted in Figure 2 at 200, a standard 'starting' matrix relating say 12-15 elements to a similar number of minerals known to occur on a widespread basis in formations of interest may be constructed, using the constant chemical formula for each mineral selected for inclusion.

8 The 12-15 elements shown in Fig. 2 are for the sake of illustration only. A lesser or greater number of elements and corresponding number of minerals may be selected.

Another consequence of the geochemical uniformity concept is that minerals distinguishable by degree of crystallinity can be quantified knowing the concentration of selected trace elements in the minerals. Trace element concentrations appear to reflect the degree of structural disorder, ie., degree of S 1crystallinity, in certain clays. Trace elements can be 0 included, along with minerals distinguishable by degree of crystallinity, when constructing the starting matrix. The 49*9 r resulting output when applying the related element/attribute transform to inputs including the concentration of the selected trace elements at a given formation depth will include the quantity of the minerals distinguishable by degree of crystallinity.

Excellent results in deriving mineral concentrations, where at least one mineral is distinguished by its degree of Scrystallinity, have been obtained using as little as a third order starting matrix, ie. one with three elements and a similar number of minerals, where at least one of t1-e three elements is a trace element, and two of the three minerals are the same mineral with different crystallinity. When using a third order matrix, it has been determined that an advantageous 25 selection of elements are iron, potassium, and the trace element thorium. Corresponding minerals might be well ordered and poorly ordered kaolinite (distinguished by their degree of crystallinity), and one of illite and feldspar.

The quantities of each element in a given formula can be derived in a variety of ways known to those skilled in the art, such as measurement, regression analysis or maximum entropy spectral analysis (see for example the papers by I. Barrodale and R.E. Erickson in Geophysics, vol 45, no.3, March 1980, pp.

420-432 and pp. 433-446). For the elements and minerals identified in Tables 1 to 5 below, values given therein can be p1j used. The results of existing factor analyses performed as c3', mentioned below for boreholes previously studied may be helpful -q E r Ir t 'i 8A in identifying minerals to be incorporated, together with the associated elements and the values to be incorporated in the matrix.

The choice of elements and minerals making up the starting matrix thus derived may be modified for a given oilfield, as at 210, on the basis of existing knowledge of the field. In most fields the geology, for example major mineral constituents, are already known to some degree. Thus it may be possible to select additional or substitute minerals and elements to be incorporated in the matrix for application in that field. This selection of tI @555

S.

4* *r S -9- 60.71G -CIP2 elements and minerals may then be further modified in light of data already available, for example to incorporate elemental concentrations which can be derived from existing logs, and because of practical considerations, such as difficulty of running specific logging tools in prevailing borehole conditions.

SObviously, factor analysis may be used in this process inasmuch as the basic data is available or obtainable.

The matrix thus obtained is an end member composition matrix as discussed in relation to equation The required element-mineral transform matrix is then obtained by inversion of the matrix depicted at 220.

Additional modifications may be made at 260 based upon comparison of other logging measurements, as at 250, with the results obtained by applying the element-mineral transform matrix, as at 230, in the manner hereinafter described to elemental concentrations obtained by logging at foe0 240; thus, an evaluation of thermal neutron capture cross-section derived as mentioned hereinafter could be compared with a direct measurement of that parameter a good degree of correlation would help confirm that appropriate i :...'selections of minerals and elements had been made. If any mineral in the starting matrix were evaluated as having zero concentration in a particular S. field, it could be concluded that that mineral was not significant in that field, and the matrix could be simplified, for that field, by removal of that mineral.

The modified end-member matrix would be inverted, as at 220, and re-tested S if desired, leading to the definition of a working element-mineral transform matrix.

Another possible way of constructing the required transform involves the use of factor analysis and regression analysis of measurements of the element and mineral content of formation samples. Full details of this approach are given in published European Patent Specification No. 151 073.

Since the transform derivation is not a part of this invention, no further Sdetails need be given here.

Once the element-mineral transform matrix (operation) is in place, the well under study, or other wells, especially those expected to have similar geological characteristics, e.g. in close proximity, may be extensively investigated. First, a decision must be made at 65 as to which logs are necessary. The decision regarding which tools to run downhole depends upon both which elements have been chosen such that their concentrations are 60.710-CIP2 required as input into the element-mineral matrix; and what final results or indicators are desired, e.g. oil API, grain density etc. Thus, for example, if the elements include potassium, iron, and sulfur, and the final results desired include an oil API log as well as a coal determination log, a natural gamma radiation tool together with an induced gamma spectroscopy tool and an enhanced resolution gamma spectroscopy tool could provide logs of the required elements. Thus, the natural gamma radiation tool would provide the potassium concentrations, while the induced gamma spectroscopy tool could provide concentrations of iron, sulfur and silicon, the last being required as discussed hereinafter for the coal determination. Finally, the enhanced resolution spectroscopy tool would provide a log of vanadium concentration required for an oil API log.

|i After determining which logs will be required to provide a quantification and further characterization of the minerals as well as the desired output results, the logs are run at 70 and elemental concentrations are determined by preprocessing the log information. Thus, the output spectrum acquired by an induced gamma spectroscopy tool. can be processed *set according to a least squares fitting technique to provide an iron yield according to the teachings of the commonly-owned U.S. Patent No. 3,521,064 owe issued July 21, 1970 to Moran. Then, to take the iron yield from the gamma eae spectroscopy tool and output an iron concentration, the iron yield is divided by the iron plus calcium plus silicon yields and that ratio is used to cross-plot core and log data from which a yield-to-concentration algorithm may be developed for iron. A similar algorithm is developed for the aluminum yields by plotting the log information against the core data for aluminum. To determine potassium concentrations, the natural gamma ray spectrum may be Kalman filtered and divided into five contiguous windows for processing according to the teaching of commonly-owned U.S. Patent No. 3,976,878 issued August 24, 1976 to P. Chevalier et al. and an article by Ruckebusch, 1G. "An Application of Kalman Filtering in Nuclear Well Logging," IEEE International Conference in Acoustics, Speech Signal Processing, Vol 3, 1982.

For each investigated station in the borehole (or along the entire length thereof), the elemental concentrations determined by the processing of the tool data gathered are input into the determined element-mineral transform matrix, with the resulting determination at 75 of the quantitative -11- 60.710-CIP2 amounts of the minerals at each depth. Additionally, by using the logs to determine concentrations of whatever elements may be observed, the determined elemental concentration information may be combined with the mineral quantity information to characterize the minerals at 80. Thus, while a formation may be said to contain certain amounts of kaolinite and illite, it may be desirable to determine whether or not the located kaolinites or illites are typical. For example, non-marine illite typically contains a magnesium concentration of If the illite can be characterized as having a magnesium concentration of a marine depositional environment is strongly indicated.

Mineral quantification 75 and characterization 80 may be used as feedback to help in the construction of the element- mineral transform matrix. Thus, for example, if, due to the mineral characterization and processing of the derived information, it is believed that a mineral such as *8 .illite should be divided into a marine and a non-marine illite, the element- *mineral transform matrix must be expanded to include an additional mineral and element.

*While mineral quantification and characterization are desirable goals in themselves, the present invention permits a broad range of new and a. improved determinations to be made on the investigated formation using mineral quantification and/or characterization as a starting point. As *depicted in FIG. 1 by the dotted line around 75 and 80, and discussed above, an improved depositional environmental analysis 82, a production risk analysis 84, a coal determination 86, grain density measurements and refined porosity indications 88, direct CEC calculations and corrected water saturation readings 90, and a vanadium content and oil API log 92 can all use the mineral quantification and/or characterization method invention as a starting point. The mineral quantifications provided by the invention may also be used in determining formation parameters such as thermal neutron capture cross-section, grain size and permeability. Details of these techniques for using the results provided by the invention are given in the afore-mentioned European Patent Specification. Since these techniques do not form any part of the present invention no further details need be given here. Those skilled in the art will recognize that other new and/or improved results will be made possible due to the method invention disclosed above.

-12- 60.710-CIP2 Although the derivations of values for these formation properties are described in that Specification as separate steps of mineral quantification followed by derivation of formation property value, it will be apparent to those skilled in the art that these two steps can be combined, the coefficients being incorporated into the appropriate values for each mineral in the element-mineral transform matrix.

Referring back to FIG. 1, teklmethod i.e.t for quantifying and

O

6 S o^ 9 Eo V% asIos /q characterizing formation minerals\was practiced on two boreholes.

For the first of these boreholes the mineral components, quartz, feldspar, kaolinite and illite could be fairly well-characterized by only three elements: iron, potassium and one of aluminum or thorium. The concentrations of these elements in each sample were then transformed at step 75 into weight percent of each mineral by constructing according to step an appropriate matrix 1 as the inverse of the matrix whose components were the concentration of each element (with aluminum chosen as the third element) in each mineral. Thus, (using multiple linear regression on measured elemental concentrations in samples) the relationships between the minerals and the elements for inclusion in the matrix to be inverted were determined as follows: TABLE 1 Re.r *0 a S a 45 6 eec.

.5 S S S S

O

.0 A 00 S. Kaolinite Illite Feldspar Al 19.0 12.0 9.7 Fe 0.14 10.6 0.05 0.35 Thus, by determining the elemental concentration of aluminum, iron and potassium from logs, the minerals in a formation adjacent the boreholes could be quantified. It should be appreciated by those skilled in the art that the above matrix is by way of example only for the particular investigated borehole, and that different minerals, elements, or matrix percentages may be used as described below. Moreover, those skilled in the art will also appreciate that the size of the matrix could be expanded or contracted in practicing the disclosed invention.

L i i I -13- 60.710-CIP2 An expanded table of relationships between minerals and elements for inclusion in the matrix and that has been found to be particularly useful has been determined as follows: TABLE 2 Non- Porous Al Fe K Si Ca H La Kaolinite 19 X X 22 X 1.7 Illite 12 7.3 4 23 X .9 Feldspar 9.7 X 12/7 30 X X X Quartz X X X 47 X X X S. Calcite X X X X 40 X X r. Smectite 11 1 .5 26 2 3.9 Kaolinite 19 X X 22 X 1.7 I Each numerical value in Table 2, except those values in the S.t* lanthanum column, represents the concentration of a given element in a given mineral as a percentage figure. The lanthanum values are expressed 0 ale in ppm. The double entry where the column for potassium and row for feldspar meet designates values for K-Feldspar and total Feldspar respectively.

S.The values in Table 2 are small by comparison with the largest number in each row. Specifically, these values are less than or equal to of the largest number in each row.

and designate poorly ordered and well ordered kaolinite respectively, kaolinite which is generally distinguishable by a low and high degree of crystallinity.

The error criteria for Table 2 is shown in Table 3, where each entry represents a plus and minus value. These values in effect define a range for each Table 2 entry. By way of example, the number 2 in Table 3 where the row for illite and aluminum meet, means that (when looking back at the same row and column entry in Table 2, which is a 10% to 14% aluminum 12 plus and minus 2) has been determined as the range of percentage of aluminum usually found in illite. This and the other range of r -14- 60.710-CIP2 values defined by Tables 2 and 3 together, has proven to be useful in constructing the deterministic model taught herein.

TABLE 3 Non- Porous Al Fe K Si Ca H La Kaolinite 1 1 .6 Illite 2 4 1 2 .4 Feldspar 2 1/2 1 Quartz 1 Calcite 1 Smectite 2 4 1 3 2 1.7 Kaolinite 1 1 1 .1 Also proven useful when the order of crystallinity of a mineral, such as kaolinite, is used as a distinguishing factor between minerals being quantified, is the conclusion that the concentration of other trace elements, such as thorium, uranium, scandium and vanadium, appear in differing quantities in such minerals as a function of degree of crystallinity.

Table 4 shows column values (expressed in ppm), for each of the above recited trace elements, which have proven useful in developing an appropriate element/attribute transform to determine the quantity of the minerals listed in Table 2 which includes kaolinite distinguished by degree of crystallinity. As illustrated, this table may be thought of as an extension Sof Table 2, 4 additional columns.

1 c. IC__ _WLI ;i i i I n 60.710-CIP2 TABLE 4 TABLE 2+

V

120 220

X

X

X

150 260 0* .9 9 9* *0 9 9090 0 S 9** 0 0 9.

9.

0 99 The respective error figures to append to Table 3 are shown below in Table TABLE TABLE 3 *0 0 90 0@ S 0* *0 T. 4 A p &QQ4- d B n g Arlprlyrrn- rn'nu hn licn 4 -ill ;n certain circumstances, for the two investigated boreholes as w 1 or other boreholes in formations expected to have s ia ogical characteristics, the transform matrix resulti g r, Table 1 above provides a way of determining m" quantities from elemental concentrations available 4, "f ¥f-iddentites of the elements dictated which tools should be run down-hole in the boreholes to gather information. Thus, it was determined at step 65 that a natural gamma ray tool could provide a potassium concentration, an induced gamma ray spectroscopy tool, the iron concentration, and an enhanced resolution spectroscopy tool, the aluminium 1 -16- 60.710-CIP2 concentration. In addition, because other end results were desired such as a direct CEC calculation and corrected water saturation reading, a grain density and refined porosity log, etc., it was decided to run additional tools such as a formation density tool which would provide a bulk density log, and a deep propagation tool which would provide a resistivity log.

Using the elemental concentrations determined by the logging tools and known data processing techniques, and the element-mineral matrix previously derived, the quantities of the minerals in the formation were Sdetermined. It will be apparent to those skilled in the art that the operatives Sdescribed herein such as matrix inversion, application of the resulting transform matrix to log measurements, and derivation of additional information using mineral quantities determined with the invention, may be accomplished, for example, by means of an appropriately programmed *general purpose computer or by spoecial-purpose circuitry. FIG. 3 depicts a t log of the mineral concentrations by well depth as generated and quantified by the instant invention. Thus, at depth 1580, a shale zone is depicted with 48% kaolinite, 25% illite, 0% feldspar, the remaining 27% assumed to be quartz.

From the mineral quantities and elemental concentrations, the minerals were further characterized in accord with step 80. For example, S* kaolinite and illite were further characterized according to their vanadium content as described hereinafter. Also, the kaolinite and feldspar elemental S* compositions were determined to be similar to look-up values. However, the illite composition appeared to be substantially different than the "average S illite" (given by Weaver, C.E. and Pollard, Developments in Sedirmentology; Elsevier Scientific Publishing Company, (New York, 1975)).

In fact, the illite appeared to be somewhat similar to an atypical illite from weathered biotite which has concentrations of K 2.7% and Fe 9.0% and often indicates a mixed layer component. In support of the conclusion that a mixed layer component might be present, it is of note that Weaver and Pollard have noted the likelihood of the presence of impurities and poorly characterized mixed-layer components in "pure" illites. Moreover, the X-ray diffraction patterns for the "illite" analyzed in the sidewall core samples also had revealed some mixed-layering with a swelling clay, presumably smectite. Finally, it was seen that an "illitic factor" in an r-mode factor analysis of core samples was heavily loaded with divalent cations Fe, Mg, Ca i. -17- 60.710-CIP2 which may represent a mixed layer component. Thus, the minerals were characterized as generally being of typical composition, with the exception of the illite, which was of atypical composition and possibly represented a mixed-layer component.

From the mineral quantification and characterization, a depositional environmental analysis, and a production risk analysis, as well as a grain density and refined porosity log, a CEC and corrected water saturation log were accomplished.

Turning to FIG. 3, the percentages of minerals in an investigated formation are seen. The mineral percentages were determined according to ;n AU -3aos' S.

the invention as described abav From the mineral quantification and i characterizations, a depositional environmental analysis was accomplished.

It is known to those skilled in the art that in a deltaic system, the dominant clay type frequently changes from kaolinite in the alluvial environment to S illite in the more marine environment. As seen in FIG. 3, the lower depths (1600 to 1750) show a much higher kaolinite to illite ratio than the 1530 to 1600 depth interval. Thus, it was apparent from the derived clay types and abundances in the investigated borehole that a regressive deltaic environment was indicated, with more alluvial, kaolinitic sediments being overlain by more marine, illitic sediments.

A production risk assessment was also made viable by the mineral quantifications and characterizations. In the sand zones (high quartz) of the I investigated borehole, a small percentage of kaolinite was present.

This amount and type of clay represented a very low risk for most production techniques. The major risk involved was the possibility that the kaolinite could be displaced from its physical location in the formation by rapidly flowing fluids, thereby causing the kaolinite to be lodged in pore throats, with a resultant reduction of the permeability of the formation. Such rapid fluid flow was therefore to be avoided during hydrocarbon production.

A second risk in the hydrocarbon production concerned the two uppermost sand bodies (1520 to 1530 and 1550 to 1570) in the formation which may be seen in FIG. 3 to be bounded by iron-rich illitic clays. An acidizing treatment of these sand zones, which would be utilized to dissolve the clays, would very likely release iron ions from these bounding clay S'I'-4 bodies. This iron would then precipitate as an iron oxide gel, thereby Sreducing the permeability of the reservoir sands. Such an acidization i i F j r p.

I

-18- 60.710-CIP2

S

600 S

S*

S

S*

S S 9* 5 0

OOSO

Se 0 5 treatment therefore was to be avoided during production of the uppermost *sand bodies.

The mineral quantifications were then utilized to produce a refined porosity log. The grain densities of the different minerals were determined from whole core data. After mineral quantities were calculated at according to the method invention, the matrix density was determined by the equation: P (2) Pmatrx,d P mwm,d m=1 where pm is the grain density of the m'th mineral and wm is the percentage of the m'th mineral in the formation at depth d. The electron density which is convertible to bulk density pb was determined by the litho-density tool (see commonly owned U.S. Patent No. 4,048,495 to Ellis) over the depth of the borehole. Porosity was then calculated as a log according to P -P b,d P matrix.d (3) P fluid matrix,d Pfluid taken as equalling unity. The improved porosity log is seen in FIG. 4, with 160 indicating the porosity log without improvement, and 162 indicating a matrix density log determined according to equation which is used to provide an improved porosity determination log 165 according to equation In a manner similar to the determination of matrix density, a CEC log of the formation was derived according to S. S

SS

0* S S0 CEC CEC d mr -m.d rn1 where CECm is the cation exchange capacity of the m'th clay mineral, and qmd is the quantity of the m'th clay mineral at depth d. The measured CEC was then converted into a log of water saturation using equations known in the art for this purpose.

As a check on the ability of the disclosed method invention to convert elemental log information into a CEC log via the derived element-mineral i II~ -19- 60.710-CIP2 Smatrix and resulting mineral quantification, the logging-derived CEC measurement was compared against the laboratory measured CEC of sidewall core for many depth points along the sidewall cored well. Such a comparison is seen in FIG. 5. While the correlation of .87 is good, and the absolute CEC differences for each depth are not great, some discrepancies are evident. It is likely that the discrepancies arise from depth shift and from error in the laboratory measurements due to the difficulty in obtaining reproducible laboratory results because of laboratory effectiveness problems 4 in cation replacement, removal of excess cation, as well as possible analytical interferences. Another cause of the scatter arises from the fact that the formation illite was shown to have varying degrees of mixed layering resulting in a non-unique CEC for illite over depth. Nevertheless, as seen in I FIG. 5, the agreement is relatively good.

*1 o. A test of the suitability of the invention was how well the mineral weight percentages (as determined in the lab) could be reproduced from the S* log concentration determined for the elements potassium, iron and aluminium for the investigated wells. FIGS. 6a-c show mineral weights percentages measured by X-ray diffraction crossplotted against those derived for a borehole from which the element-mineral matrix was derived. The figures clearly show good agreement, except for very low concentrations which may simply reflect the large relative error in X-ray diffraction techniques at those levels. FIGS. 6d-f show, for a second borehole, mineral 9 :weight percentages measured by X-ray diffraction crossplotted against those derived using the element-mineral transform matrix previously derived from the first borehole. While good agreement is found, clearly there is more *scatter, especially for the illite as seen in FIG. 6e. Some of this scatter might f be explained by depth shift and the mixed layering nature of the illite.

However, another factor which might account for more error was the fact that the X-ray dif-raction analysis technique was changed and was perhaps less accurate.

As a further test of the validity of the invention, FIGS. 6g and 6h show the X-ray diffraction measurements of quartz plotted against the residual from the geochemical model. unity minus the sum of derived kaolinite, illite, and feldspar percentages for the two boreholes. Again, there is good agreement; not only are very shaly sections well separated from the r- c~ i Ic i

J

i :6 *so* Do .10.

0000 Does a 00 0 0 S S

S

0 S. S

S

60.710-CIP2 high quartz sand zones, but less shaly sands are also well quantified in quartz.

Finally, it may be expected that a correlation would exist between the sum of the derived kaolinite plus illite fractions and the measured fines fraction because the fines (less than 20 micron) fraction of the formation is dominated by the clay minerals, as indeed are most sedimentary sequences.

However, this estimation should be expected to underestimate the measured fines since it does not include fine quartz and other minerals which may be present. FIG. 6i is a cross-plot of the measured versus the derived fines fraction and shows that the instant invention gives an excellent characterization of the shale volume fines present. The fines fraction as derived by the invention provides an excellent measurement of the shale volume, the well logging parameter Vsh. The correlation coefficient between measured and derived percent fines is 0.98. Moreover, as aforementioned, the CEC measurements also correlated well.

Thus, the method invention proved to be capable of producing an accurate characterization and quantitative description of the major lithological components of the formation.

There has been described and illustrated herein, methods in accordance with the present invention for determining from log data the quantities of minerals (values of attributes) in a formation and the characteristics of those minerals, as well as methods for further using the method inventions to derive new and improved results. However, while particular embodiments of the present invention have been utilized and described, it is intended that the invention be broad in scope and that the specification be read likewise. Thus, while the terms "logs" and "logging" are utilized, they are not intended to be limiting in any manner. The terms are intended to encompass the gathering of data at a single depth station, multiple stations or during continuous movement of the tool. and to include all data processing and data transfer techniques known in the borehole logging arts. Also, while much of the invention has been described as utilizing a "transform matrix", other techniques such as simultaneous equations may be utilized in lieu of the matrix to accomplish the stated objectives. Therefore, it will be apparent to those skilled in the art that other changes and modifications may be made to the invention as described in the specification.

1 1 F.

Claims (29)

1. A method for determining the concentration at a given depth in a formation of each of a selected group of minerals wherein said group comprises one or more of kaolinite, illite and feldspar, and wherein at least one mineral of said group is distinguishable by degree of crystallinity, said method comprising the steps of: determining by borehole logging the concentration of each of a selected group of elements at said depth wherein said group of elements comprises iron, potassium and at least one trace element which appears in differing quantities in said at least one mineral as a function of degree of crystallinity; and applying an element/attribute transform to the determined concentration of elements at said depth, wherein the components of said transform are related to the concentration of each of said elements in each of said minerals, to thereby determine the concentration of each of said minerals at said depth, including said mineral distinguished by its degree of crystallinity.

2. A method as set forth in claim 1 wherein the element/attribute transform is the inverse of a starting matrix the components of which are numerical values representing the concentration of each of said elements in each of said minerals.

3. A method as set forth in claim 2 wherein the group of elements includes aluminum and the numerical value representing the concentration of aluminum in kaolinite is in the range of 18% to

4. A method as set forth in claim 2 wherein the group of elements includes aluminum and the numerical value representing i i- L- 22 the concentration of aluminum in illite is in the range of to 14%. A method as set forth in claim 2 wherein the group of elements includes aluminum and the numerical value representing the concentration of aluminum in feldspar is in the range of

7.7% to 11.7%. 6. A method as set forth in claim 2 wherein the group of S 10 elements includes aluminum, the group of minerals includes smectite and the numerical value representing the concentration of aluminum in smectite is in the range of 9% to 13%. A method as set forth in claim 2 wherein the numerical *see value representing the concentration cf iron in illite is in the range of 4% to 10.6%.

8. A method as set forth in claim 2 wherein the group of minerals includes smectite and the numerical value representing the concentration of iron in smectite is in the range of 0% to o.o

9. A method as set forth in claim 2 wherein the numerical value representing the concentration of potassium in illite is S 25 in the range of 3% to A method as set forth in claim 2 wherein the group of minerals includes potassium feldspar and the numerical value representing the concentration of potassium in potassium feldspar is in the range of 11% to 13%.

11. A method as set forth in claim 2 wherein the numerical value representing the concentration of potassium in feldspar is in the range of 5% to 9%. S12. A method as set forth in claim 2 wherein the group of minerals includes smectite and the numerical value representing 23 the concentration of potassium in smectite is in the range of 0% to

13. A method as set forth in claim 2 wherein the group of elements includes silicon and the numerical value representing the concentration of silicon in kaolinite is in the range of 21% to 23%.

14. A method as set forth in claim 2 wherein the group of 10 elements includes silicon and the numerical value representing the concentration of silicon in illite is in the range of 21% @too to

15. A method as set forth in claim 2 wherein the group of elements includes silicon and the numerical value representing the concentration of silicon in feldspar is in the range of 29% to 31%.

16. A method as set forth in claim 2 wherein the group of elements includes silicon, the group of minerals includes quartz and the numerical value representing the concentration of silicon in quartz is in the range of 46% to 48%.

17. A method as set forth in claim 2 wherein the group of 25 elements includes silicon, the group of minerals includes smectite and the numerical value representing the concentration of silicon in snectite is in the range of 23% to 29%.

18. A method as set forth in claim 2 wherein the group of elements includes calcium, the group of minerals includes calcite and the numerical value representing the concentration of calcium in calcite is in the range of 39% to 41%.

19. A method as set forth in claim 2 wherein the group of elements includes calcium, the group of minerals includes smectite and the numerical value representing the concentration L' of calcium in smectite is in the range of 0% to 4%. V S 0iy I j- 24 A method as set forth in claim 2 wherein the group of elements includes nonporous hydrogen and the numerical value representing the concentration of nonporous hydrogen in kaolinite is in the range of to 2.3%.

21. A method as set forth in claim 2 wherein the group of elements includes nonporous hydrogen and the numerical value representing the concentration of nonporous hydrogen in illite is in the range of to 1.3%.

22. A method as set forth in claim 2 wherein the group of e' elements includes nonporous hydrogen, the group of minerals includes smectite and the numerical value representing the concentration of nonporous hydrogen in smectite is in the range of 2.2% to 5.6%.

23. A method as set forth in claim 2 wherein the group of elements includes lanthanum, the group of minerals includes poorly ordered kaolinite and the numerical value representing the concentration of lanthanum in poorly ordered kaolinite is in the range of 50 ppm to 110 ppm.

24. A method as set forth in claim 2 wherein the group of elements includes lanthanum and the numerical value 25 representing the concentration of lanthanum in illite is in the Srange of 15 ppm to 55 ppm. A method as set forth in claim 2 wherein the group of Selements includes lanthanum, the group' of minerals includes smectite and the numerical value representing the concentration of lanthanum in smectite is in the range of 10 ppm to 50 ppm.

26. A method as sec forth in claim 2 wherein the group of elements includes lanthanum, the group of minerals includes well ordered kaolinite and the numerical value representing the AL concentration of lanthanum in well ordered kaolinite is in the range of 5 ppm to 25 ppm. 0-, t i I,

27. A method as set forth in claim 2.wherein the group of elements includes thorium, the group of minerals includes poorly ordered kaolinite and the numerical value representing the concentration of thorium in poorly ordered kaolinite is in the range of 22 ppm to 32 ppm.

28. A method as set forth in claim 2 wherein the group of elements includes thorium and the numerical value representing o the concentration of thorium in illite is in the range of 11 ppm to 23 ppm. Oe o

29. A method as set forth in claim 2 wherein the group of elements includes thorium, the group of minerals includes smectite and the numerical value representing the concentration of thorium in smectite is in the range of 20 ppm to 40 ppm. A method as set forth in claim 2 wherein the group of elements includes thorium, the group of minerals includes well ordered kaolinite and the numerical value representing the concentration of thorium in well ordered kaolinite is in the range of 25 ppm to 65 ppm.

31. A method as set forth in claim 2 wherein the group of elements includes uranium, the group of minerals includes 25 poorly ordered kaolinite and the numerical value representing S* the concentration of uranium in poorly ordered kaolinite is in the range of 4.7 ppm to 8.7 ppm.

32. A method as set forth in claim 2 wherein the group of elements includes uranium and the numerical value representing the concentration of uranium in illite is in the range of ppm to 6.5 ppm.

33. A method as set forth in claim 2 wherein the group of elements includes uranium, the group of minerals includes smectite and the numerical value representing the concentration /j of uranium in smectite is in the range of 0 ppm to 4 ppm. L 26

34. A method as set forth in claim 2 wherein the group of elements includes uranium, the group of minerals includes well ordered kaolinite and the. numerical value representing the concentration of uranium in well ordered kaolinite is in the range of 2 ppm to 6 ppm. A method as set forth in claim 1 wherein said group of minerals includes well ordered kaolinite and poorly ordered kaolinite.

36. A method as set forth in claim 35 wherein said at least one trace element is selected from the group consisting of lanthanum, uranium, thorium, scandium and vanadium. 0

37. A method as set forth in claim 1, substantially as herein described with reference to and as illustrated in the accompanying drawings. DATED this 23rd day of July, 1990 SCHLUMBERGER TECHNOLOGY B.V. S 25, By Its Patent Attorneys GRIFFITH HACK CO. Fellows Institute of Patent Attorneys of Australia. L