Borehole Logging for Uranium Exploration

Home , Gamma ray logging, Spinthariscope

TECHNICAL REPORTS SERIES No. 212

A Manual

c # INTERNATIONAL ATOMIC ENERGY AGENCY, VIENNA, 1982

BOREHOLE LOGGING FOR URANIUM EXPLORATION

A Manual The following States are Members of the International Atomic Energy Agency:

AFGHANISTAN HOLY SEE PHILIPPINES ALBANIA HUNGARY POLAND ALGERIA ICELAND PORTUGAL ARGENTINA INDIA QATAR AUSTRALIA INDONESIA ROMANIA AUSTRIA IRAN SAUDI ARABIA BANGLADESH IRAQ SENEGAL BELGIUM IRELAND SIERRA LEONE BOLIVIA ISRAEL SINGAPORE BRAZIL ITALY SOUTH AFRICA BULGARIA IVORY COAST SPAIN BURMA JAMAICA SRI LANKA BYELORUSSIAN SOVIET JAPAN SUDAN SOCIALIST REPUBLIC JORDAN SWEDEN CANADA KENYA SWITZERLAND CHILE KOREA, REPUBLIC OF SYRIAN ARAB REPUBLIC COLOMBIA KUWAIT THAILAND COSTA RICA LEBANON TUNISIA CUBA LIBERIA TURKEY CYPRUS LIBYAN ARAB JAMAHIRIYA UGANDA CZECHOSLOVAKIA LIECHTENSTEIN UKRAINIAN SOVIET SOCIALIST DEMOCRATIC KAMPUCHEA LUXEMBOURG REPUBLIC DEMOCRATIC PEOPLE'S MADAGASCAR UNION OF SOVIET SOCIALIST REPUBLIC OF KOREA MALAYSIA REPUBLICS DENMARK MALI UNITED ARAB EMIRATES DOMINICAN REPUBLIC MAURITIUS UNITED KINGDOM OF GREAT ECUADOR MEXICO BRITAIN AND NORTHERN EGYPT MONACO IRELAND EL SALVADOR MONGOLIA UNITED REPUBLIC OF ETHIOPIA MOROCCO CAMEROON FINLAND NETHERLANDS UNITED REPUBLIC OF FRANCE NEW ZEALAND TANZANIA GABON NICARAGUA UNITED STATES OF AMERICA GERMAN DEMOCRATIC REPUBLIC NIGER URUGUAY GERMANY, FEDERAL REPUBLIC OF NIGERIA VENEZUELA GHANA NORWAY VIET NAM GREECE PAKISTAN YUGOSLAVIA GUATEMALA PANAMA ZAIRE HAITI PARAGUAY ZAMBIA PERU

The Agency's Statute was approved on 23 October 1956 by the Conference on the Statute of the IAEA held at United Nations Headquarters, New York; it entered into force on 29 July 1957. The Headquarters of the Agency are situated in Vienna. Its principal objective is "to accelerate and enlarge the contribution of atomic energy to peace, health and prosperity throughout the world".

Permission to reproduce or translate the information contained in this publication may be obtained by writing to the International Atomic Energy Agency, Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria.

Printed by the IAEA in Austria January 1982 TECHNICAL REPORTS SERIES No.212

BOREHOLE LOGGING FOR URANIUM EXPLORATION

A Manual

INTERNATIONAL ATOMIC ENERGY AGENCY VIENNA, 1982 BOREHOLE LOGGING FOR URANIUM EXPLORATION: A MANUAL IAEA, VIENNA, 1982 STI/DOC/lO/212 ISBN 92-0-145082-6 FOREWORD

Intensive worldwide efforts in uranium exploration over the past four years have been stimulating research on and development of new exploration techniques and field equipment. It is widely recognized that borehole logging is one of the most effective ways of estimating uranium resources, and its popularity has grown considerably. Logging for uranium, which once was an auxiliary technique subordinate to chemical analysis and drilled core examination, has become one of the most powerful and efficacious methods of measuring, directly or indirectly, uranium at depth. Borehole logging provides, rapidly and economically, most of the sub-surface information needed by the exploration geologist. This includes in-situ analysis, lithological identification, stratigraphic correlation, and information on density and moisture. Logging is important not only in exploration but in mine development and grade control, in mining operations and in production. Tens of millions of metres will be drilled and logged by exploration firms and governmental organizations in Member States each year in the search for nuclear fuel. The joint NEA/IAEA Group of Experts in Research and Development on Uranium Exploration Techniques, aware of the importance of this particular subject, recommended the preparation of a manual that could be of use to both developing and industrialized countries. Accordingly, the IAEA convened a workshop to discuss problems related to borehole logging and, later, organized an international consultative group charged with the preparation of such a manual. The present text has been prepared taking into account the requirements of both developing countries, which might be at an incipient stage of uranium exploration, and industrialized countries, where more advanced exploration and resource evaluation techniques are commonly in use. While it was felt necessary to include some discussion of exploration concepts and fundamental physical principles underlying various logging methods, it was not the intention of the consultants to provide a thorough, detailed explanation of the various techniques, or even to give a comprehensive listing thereof. However, a list of references has been included, and it is strongly recommended that the serious student of mineral logging consult this list for further guidance. The Agency wishes to express its gratitude to the four consultants, Messrs. J.K. Hallenburg, P.G. Killeen, V.L.R. Furlong and J. Duray, who participated in the meetings and contributed to the report. Special thanks are due to Mr. J.K. Hallenburg, Chairman of the Consultative Group, who not only contributed much original material to the manual but also edited the text for technical content and prepared it for publication. Mr. P.M.C. Barretto, a member of the Group, was the responsible Agency staff member for this project.

CONTENTS

1. INTRODUCTION 1

1.1. How borehole logging fits the uranium exploration program 1 1.2. Natural radioactivity 4

2. THE BOREHOLE: ITS CHARACTERISTICS AND EFFECTS 18

3. LOGGING TECHNIQUES 29

3.1. Common logging techniques 29 3.1.1. Gross-count gamma-ray logging 29 3.1.2. Spectrometric gamma-ray logging 52 3.1.3. Neutron logging 67 3.1.4. Resistance, resistivity, and conductivity 79 3.1.5. Spontaneous potentials 112 3.1.6. Calipers 135 3.1.7. Deviation 138 3.2. Other logging techniques 150 3.2.1. Neutron-activation systems 150 3.2.2. Density 153 3.2.3. Acoustic measurements 169 3.2.4. Induced polarization 176 3.2.5. Magnetic susceptibility 184 3.2.6. Dipmeters 187 3.2.7. Mud logging 190 3.2.8. Use of petroleum logs and petroleum logging equipment .... 192

4. INSTRUMENTATION 194

4.1. Basic systems 194 4.2. Advanced systems 208 4.3. Radioactivity statistics 214

5. FIELD PROCEDURES 216

6. INTERPRETATION 231

6.1. The exploration philosophy 231 6.2. Interpretation of borehole data 232 7. SELECTED REFERENCES 248

APPENDIX A: Calibration procedure for gross-count gamma-ray logging 259 APPENDIX B: Deadtime determination for gross-count gamma-ray logging 266 APPENDIX C: Correction factors for gross-count gamma-ray logging 270 APPENDIX D: Glossary 276

LIST OF CONSULTANTS 279 1. INTRCDUCTiai

The importance of logging in uranium exploration was summarized in a 1976 IAEA report as follows: "Borehole logging rapidly and economically provides most of the subsurface information required by the exploration geologist. This includes in-situ sampling and assaying, lithologic identification, stratigraphic correlation, and the more sophisticated logging pro- grairmes measure petrophysical parameters such as density, and moisture or various types of geological formation. Logging can often greatly reduce drilling costs by obtaining the needed data from less costly noncored holes or from holes previously drilled for other purposes. Generally, logging provides more representative and objective data in less time and at lower cost than is required by descriptive logging, sampling and assaying of cores or cuttings."

The dictionary defines a log as "a record of progress, as in a record of a ship's speed." A borehole log is a record of one or more physical measurements as a function of depth in a borehole. Logs are recorded by means of sondes, probes or tools carrying sensors which are lowered into the hole by a cable. Examples include logs of electrical measurements, nuclear measurements, acoustic measurements and temperature measurements. Measurement of certain parameters are especially useful in a uranium exploration program. The most common logging techniques useful in uranium exploration are described, including the physical basis of each technique, its application, and the analysis of an example log. Less common logging techniques which may become in greater usage in the future are reviewed more briefly.

Finally logging systems (more than one parameter) are discussed including instrumentation, field procedures and interpretation in several specific geologic environments.

1.1. How borehole logging fits the uranium exploration program

1.1.1. A successful exploration program consists of systematic and sequential decision-making actions. In general, these actions are:

1.1.1.1. Preliminary appraisal

A basic decision is made to explore for uranium. Considera- tion is given to the selection of promising areas which may be favorable for uranium occurrence (indicated by existing geological and geophysical information). Detailed geological appraisal of the selected area from existing geologic knowledge is carried

1 out in an attempt to reduce the areas to be examined. Geologi- cal interpretation from aerial photographs is often included in this stage. Geological and radiometric field checks are made for confirmation on the areas selected where possible host environments exists.

1.1.1.2. First phase program

During the first phase of an exploration program several steps should be considered and/or must be followed. Airborne radiometric, ground radiometric, and/or geochemical surface sampling must be considered and carried out if they appear useful. Regional and local geological mapping are very necessary. Stratigraphic drilling should be carefully planned along with the logging of stratigraphic boreholes. All of these efforts serve to detect anomalies. Evaluation and subsequent followup will then refine and define the logical sequence of events.

1.1.1.3. Second phase program

During the second phase, the target (which was defined radiometrically during the first phase) will be detailed. This can use geological mapping, geochemical evaluations (trenching, augering and bedrock geochemistry), wireline logging of scout or exploration boreholes on anomalous targets. Petrographical and mineralogical determinations, and ground geophysical surveys may be used for further target deliniation.

This second phase may establish the presence of uranium mineralization sufficiently interesting to warrant continuation of the exploration program.

1.1.1.4. Third phase program

The third phase is the beginning of the development phase. It will consist of definition drilling and borehole logging of the uranium target. This will also be a refinement of the detailed geological interpretation, further detailed geochemistry, petrology studies and preliminary engineering feasibility studies.

If at this point a decision is made that a viable or reserve is likely, the fourth and subsequent phase is carried out.

1.1.1.5. Fourth phase program

The fourth phase is the main body of die development phase of an exploratory program. It involves detailed drilling and borehole logging on a grid pattern of sufficient density to

2 satisfy standard engineering practices for ore reserve calculation. It will also contain detailed engineering studies of the potential mine area.

It is clearly evident in this idealized exploration sequence of events that borehole logging is used in all phases of an exploration program. It is used in reconnaissance stratigraphic drilling, the subsequent exploratory drilling, in definition drilling of the mineral body and the detailed drilling for ore reserve estimation and engineering studies. Borehole logging is the main technique to be used for gathering and processing data'about the project.

1.1.2. Discussion of borehole logging use

1.1.2.1. Stratigraphic drilling

Stratigraphic drilling is carried out to determine whether a favorable host rock sequence may be present in an area under examination. This is usually carried out in soft rock exploration but is gaining favor in crystalline environments. Gross- count gamma-ray, resistivity and SP logs all assist in determining sedimentary rock characteristics in a stratigraphic hole. In crystalline environments, the main logs are the spectrographic ganma-ray, density, acoustic velocity and perhaps the neutron curves. The probes combine sensors for several of these logs, usually including the gross-count ganma-ray measurement. Many of these probes have large gamma-ray scintillation crystals, and thus are highly responsive to the slightest change in radioactive reading. This response, combined with electrical and. other responses, enables identification of rock boundaries, lithology, rock units and possible discontinuities.

1.1.2.2. Exploratory drilling

The intention of an exploratory hole is to drill into or under a geochemical/radiometric anomaly and intersect the source of the anomaly. Great emphasis is placed on the gamma-ray result and coninonly any positive result can be calculated on the spot to determine the quantity of equivalent U3O8. Often the electrical logs are given casual attention at this stage, however much valuable information can be gained by their use. The resistivity curve is the best suited for developing a cross-section. Except in situations where uranium mineralization is absolutely stratigraphically trapped, it often occurs in association with carbonaceous materials. These carbonaceous materials often have a lower resistivity (very low in the case of graphites) in comparison to the surrounding rock. In these situations there may also be a change in SP response.

3 1.1.2.3. Definition drilling

Assuming the previous exploratory drilling was successful,_ further drilling is undertaken to define the extent of mineralization as well as determining the nature of the host geological environment. Gamma-ray results are calculated on each hole into equivalent U3O8quantities. If drill holes are sufficiently close together (by engineering criteria) hole to hole correlation is made of the amount of uranium mineral in a section, thus allowing volumes to be estimated and approximate ore quantity derived. The confidence in hole-to-hole correlation can be greatly increased if the electrical responses related to the mineralized zone and nearby ones are in agreement. Should there be a significant discrepancy in the electrical and ganma-ray response between holes, there may well be geological inconsist- encies or structural disruptions which would make an interpola- tion between the holes unwise. A continuum may not exist between two seemingly adjoining intersections in adjacent holes. Confidence in the continuity of uranium mineralization between adjacent drill holes can only be achieved after the interpreter has a good working knowledge of the geological environment and for the relative gamma-ray and electrical responses of those rocks.

1.1.2.4. Detailed drilling for ore reserve estimations

Engineering requirements usually dictate the ideal density of pattern drilling carried out during the last phase of drilling preparatory to mining. Actual grid spacing depends upon the size of the ore body, the amount of grade variation, and the method of mining to be used. It generally varies between 5 and 30 metre centers. Large amounts of data are accumulated from these logging operations. For this reason, the log information is best processed by computers.

1.2. Natural radioactivity

Uranium, element 92, is a natural radioactive element which occurs in association with two other radioactive elements, potassium, element 19, and thorium, element 90. These latter can provide interference when we attempt to find uranium by measuring gamna radiation. It is important therefore to be aware of the characteristics of these elements.

1.2.1. Potassium, K

Comprising 2.6% of the earth's crust, potassium is a relatively abundant element in nature. However, only a small fraction (0.0118%) of natural potassium consists of the radioactive

Text continued on p. 7 4 TABLE 1.0. POTASSIUM CONTENT OF VARIOUS FORMATION MATERIALS

Material Potassium Content by Weight(%) (Average) (Range) Sylvite . 54 Potash 44.9 Langbeinite 20 Microcline 16 Kainite 15.1 Camallite 14.1. Orthoclase 14 Polyhalite 12.9 Muscovite 9.8 Biotite 8.7 Illite 5.2 3.51-8.31 Arkose (sandstone) 4.6- 4.4-5.1 Synite 4.53 Glauconite 4.5 3.2-5.8 Granite 4.0 2.0-6.0 Norite 3.3 Grandodiorite 2.90 Shale 2.7 1.6-9.0 Igneous rock 2.6 Graywacke (sandstone) 1.8 1.2-2.1 Diorite 1.66 Basalt 1.3 Sandstone 1.1 0-5.1 Gabbro 0.87 Diabase 0.75 Kaolinite 0.63 0-1.49 Limestone 0.27 0-0.71 Montmorillonite 0.22 0-0.60 Orthoquartzite (sandstone) 0.08 0-0.12 Dolomite 0.07 . 0.03-0.1 Dunite 0-04 Seawater 0.035 Table 1.0 is presented to show the effects of formation potassium content on ganma-ray response. About 0.012% of all natural potassium is radioactive (i.e. 4Qk) giving off a 1.5 MeV gamma ray upon disintegration. Approximatel# y 20% of the gamma-ray amissions frcm shale are caused by the isotope 40k. The remaining radiation frcm shales is generally caused by uranium and thorium series elements. The average sandstone contains about 12% feldspar. Even orthoquartzite sands may contain 10% feldspar. "Although any kind of feldspar may be present the acid feldspars, particularly potash- bearing (orthoclase) varieties are most common" (Pettijohn).

5 TABLE 1.1. DISINTEGRATION SERIES OF URANIUM-238

(Principal members only; isotopes constituting less than 0.2 per cent of the decay products are omitted) Uranium 238 [4.51 X 10'yl

I a 1 Thorium 234 [24.10d] I I /8 I Protactinium 234 [1.14m]

I 0 I Uranium 234 (2.48 X lO'y]

1 Thorium 230 [8.0 X 10

I « 1 Radium 226 |l,C22y] I I « 1 Radon 222 [3.825d] I I « I Polonium 218 (3.05m] I a i Lead 214 [26.8m]

i Bismuth 214 [19.7m] I & i Polonium 214 [1.50 X 10"4s]

1 Lead 210 [22y]

I (8 1 Bismuth 210 [5.02d] I 0 i Polonium 210 [138d] I I Lead 206 [stable]

6 I,0 isotope K. A single ganma ray is emitted with energy 1.46 MeV. Hi is characteristic energy may be used by a gamma-ray spectro- mete 40r to identify 9and measure potassium in rocks. The half-life of K is 1.3 x 10 years. The gamma radiation originating from the K-feldspar content of some rocks may contribute significantly and complicate the measurement of gamma radiation from low concentrations of uranium. However, the gamma radiation from potassium is often useful for lithologic identification, table 1.0.

1.2.2. Uranium, U

The gamma radiation of interest in uranium exploratio 238n measurement originates primarily from the decay product 2 3 5 s of U. The other naturally occurring uranium isotope, U, represents only 0.72% of natural uranium and is therefore relatively unim-

portant excep238 t for neutron-related techniques. The radioactive,

decay of U consists of a series of decays as shown in tabl 238e

1.1. The decay series begin9 s with the long-lived parent U

(half-life = 4.51 x 10 years) 2 ^ and after 14 successive decays reaches the stable end product Pb (lead, element 82). Many of these decay steps in the series result in emission of gamma

rays with energies characteristic of the particula8 r decaying nuclide (see figure 1.1). Unfortunately, U does not itself emit gamma radiation which can be detected and utilized as a measure of the uranium content. Most of the game emission which i211,s useful fo211(r uranium exploration originates from the decay of Pb and Bi (bismuth, element 83), which are the eight and ninth daughter products in the decay series. This is an important fact to remember when utilizing measurements of ganma radiation for uranium determinations. The behaviour in nature of the daughter products in the decay series must be

understood. If the relationship between the daughte 238r products (whose radiation can be measured) and the parent U is known, then the uranium concentration can be determined. This "equilibrium" relationship will be discussed later.

1.2.3. Thorium, Th

232

The thorium decay serie10 s originates with the parent Th

(half-life = 1.39 x208 10 years) and ultimately reaches the stable end product Pb, as shown in table 1.2. Ganma rays characteristic of the particular decaying nuclides are emitted in many of the decay steps in the serie 232s (figure 1.2). Similar- ily to the case of uranium, the parent Th does not itself emit ganma radiation which can be detected and utilized as a measure of the thorium content of rocks. For field measurement

the most 208 useful ganma-ray emitter is the ninth in the decay series, T1 (thallium, element 81). An important fact about the thorium decay series is that the longest half-life of any of

7 TABLE 1.2. DISINTEGRATION SERIES OF THORIUM

Thorium 232 [1.39 X 10'°yl

i Radium 228 [6.7y]

I 0 i Actinium 228 [6.13h] I I 0 I Thorium 228 [1.90y]

Radium 224 [3.64d|

Radon 220 [54.5s; (popularly called thoron) a

Polonium 216 [0.158s]

I a / i ' Lead 212 [10.6h] 0

Bismuth 212 [60.5m] \ (66.3%)0 . ct( 33.7%)

S / \

Polonium 212 [3.0 X 10-'s] Thallium 208 [3.1m] \

Lead 208 [stable]

8 s o> o -c 0.5|- >9 £ 0.4

O< c- 0.3 5 CIDs -js 2 < CQ 2 o 0.2 > > 5 §uj 1r: 5 a. 5 01 0.1 CO in »l!i! S 2 r^ CO CM U> 5«> 01 0 T-^SK 0 1 II £J J l_ - 01 0.5 1.0 1.6 2.0 2.5 3.0 ENERGY (MeV)

Bi-214 Pb-214 Ra-226

FIG.1.1. Characteristic decay energies of the uranium series.

5 5 0) o O.Br S > o 5 s* 0.4 • •2 <0 :o >« psi :o 5 <*> > 0.3- O a 5 °<• Do>

0.2 >ONI !I |6 5 ul r* 5' C O1 0.1 • £

TI-208 Bi-212 Ac-228 Pb-212.

FIG.1.2. Characteristic decay energies of the thorium series. the daughter products is very short compared to those in the uraniu m232 decay series. The longest half-lif 228 e (other than the parent Th) in the series is that of Ra which is about 6 years.

1.2.4. Radioactive equilibrium

Radioactive equilibrium or disequilibrium is an important consideration in all ganma-ray logging measurements. Most of the ganma radiation emitted by nuclides in the uranium decay series are not actually from uranium, but from daughters in the series.

The ganma-ray count rate can be related to the amount of parent, by assuming there is a direct relation between the anount of daughter and parent. This assumption is valid Uien the radioactive decay series is in a state of secular equilibrium.

238 A radioactive decay series such as that of U is said to be a state of secular equilibrium when the number of atoms of each daughter being produced in the series is equal to the number of atoms of that daughter being lost by radioactive decay. When this condition exists, it is possible to determine the amount of the parent of the decay series by measuring the radiation from any daughter element.

The question then is whether the assumption of secular equilibrium is valid for the geologic material being analyzed for its uranium content. Most ganma-ray logging techniques require that secular equilibrium be assumed, until additional information is available.

If the parent element or one or more of the daughter products is lost by any process other than radioactive decay, or if the parent was not deposited sufficiently long ago, the assumption of secular equilibrium is not satisfied. Since each daughter product is an element with its own characteristic physical and chemical properties it may behave differentl 238 y within a given environment. For exanple, in the U decay system there is a gas, radon-222, with a 3.85 day half-life, which can easily migrate. Also, the solubilities of radium, uranium and thorium differ, and preferential leaching may occur.

Radioactive disequilibrium is quite a conmon occurrence in roll-front or sandstone-type uranium deposits. The reason is that uranium may be mobile within the sandstone (usually when it is oxidized), and daughter product formation lags behind. This leads to a distribution of radioelements wherein the daughter

10 211, products (e.g. Bi) are left behind, creating a daughter- excess or parent-deficiency state, with strong ganma ray activity, while at some nearby location there is a (relatively) weakly radioactive uraniferous zone with a daughter-deficiency. For most other rocks, not a great deal is known about the state of radioactive equilibrium.

The degree of disequilibrium will vary with circumstances, and a number of factors are involved. These include the mineralogy of the radioelements and the mineralogy of their surround- ings which may create a reducing or oxidizing environment; for example, the presence or absence of sulfides, or carbonates. Hie degree of disequilibrium is strongly dependent upon climate, topography and surface hydrology. The determination of disequilibrium is also dependent upon sample volume. A small drill core specimen is much more likely to show extreme disequilibrium than will a large bulk sample, such as a gamma-ray logging (i.e. in seme cases the parents and daughters may have moved apart on the scale of a hand specimen, but not on the scale of a cubic metre of rock). The degree of disequilibrium is not easily established with direct field measurements. It can be determined in a number of ways in the laboratory, either by comparing chemical estimates with estimates based on decay product radio- acitivity, or by measuring the radioactivity of different decay products.

Disequilibrium in a uranium deposit or occurrence does not necessarily rule out gamma-ray logging as an evaluation tool. Satisfactory results may still be obtainable by spot-checking the state of disequilibrium of the deposit by laboratory chemical analyses, and applying appropriate corrections to the gamna-ray logs. Ganma-ray logging has been shown to be an effective quantitative evaluation tool in the sandstone-type uranium deposits in the United States, even though these deposits are known to be in radioactive disequilibrium to various degrees.

After a disturbance in any radioactive decay series, equilibrium is re-established in approximately seven half-lives of the longest lived daughter. The thorium decay series can generally be assumed to be in radioactive equilibrium 228 . The longest lived^daughter in the thorium series is Ra with a half- life of 6.7 years. Seven half-lives, totalling less than 50 years, is a geologic "instant", and any redistribution of thorium would be followed by a relatively rapid re-establishment

of secular equilibrium. Thus ,2 32 in practically all geological samples, the amount of parent Th can safely be computed by measuring the ganma-ray activity of the daughter nuclides. In the case of uranium movement, it can reasonably be assumed that

11 231 2 3 8 the isotope *U will move with the parent U (except in rare

circumstances)23 . Thus the longest lived daughter can be taken as °Th with a half-life of 80,000 years, and re-establishment of secular equilibrium should take about half-a-million years.

1.2.5. Units of measuranent The basic unit of measurement in gamna-ray logging is a count rate; i.e. the number of ganma rays counted per unit time.

Fran an exploration and geological viewpoint it is desirable to obtain as direct a measure as possible of the concentration of the uranium in the rock for the purpose of locating uranium mineralization. Before this can be done (by proper calibration of the logging system; described later), consideration must be given to both radioactive equilibrium and the thorium and potassium concentration in the rocks penetrated by the borehole. In the preceding sections, there is a discussion of the fact that . the measurement of ganma-ray intensity for the estimation of uranium concentrations must assume that the respective decay series are in equilibrium. Because, in the absence of cor- roborating evidence, there can be no certainty that equilibrium conditions apply to a measurement, radiometric determinations of uranium abundance are identified as 'equivalent' determinations. If the conditions of equilibrium apply, a given radiometric determination is equivalent to a given element abundance. Hence, in typical rocks, radiometric measurements of uranium concentrations would be expressed as ppm eU or % eU 0e. The same would be true for thorium determinations by spectrometri3 c gamiia-ray logging. The disequilibrium problem does not apply in the case of potassium. Typical radiometric rock analyses for potassium are expressed simply as %K.

If gross-count gamma-ray logging is utilized, in addition to the assumption of radioactive equilibrium, it must also be assumed that the other radioelements thorium and potassium do not contribute significantly to the count rate. If this is not the case, such as in the pegpiatite type deposits of the Bancroft uranium area or the conglomeratic type deposits of the Elliot Lake area in Canada, spectrometric ganma-ray logging must be used.

In later sections on gross-count gamma-ray logging and spectrometric ganma-ray logging, a method of converting the basic count rates to a measure of the grade and thickness of the mineralized zones is described. The logging system must be calibrated to derive a sensitivity (or k-factor) for each gamma- ray probe. The sensitivity is the conversion factor from count rate to grade. The k-factor will have units such as (ppm eU)/

12 (counts per second) or (% eU30 )/(counts per second) or some similar unit of (grade)/(count8 rate). The use of these units and their relationship with the thickness of the zones being analyzed, forms the basis of gamna-ray log interpretation.

1.2.6. Interaction of ganma rays with rocks and the sample volume concept

In borehole logging we are concerned largely with two types of interactions between gamma radiation and matter: Compton scattering, and the photoelectric effect.

These are defined as follows: (a) Compton scattering, in which the gamma rays lose part of their energy to an electron and are scattered at an angle to their original direction. This process predominates for moderate energies in a wide range of materials.

(b) The photoelectric effect, in which the garrma ray gives up all its energy in ejecting a bound electron. This process predominates at low energy, especially in matter with a high atomic number (Z).

In the case of rocks which represent a complex medium, the atomic number is the "effective" or "equivalent" atomic number, Zeq. The equivalent atomic number, Zeq, of a complex medium is defined as the atomic number of the hypothetical element having the same ratio of Compton attenuation to photoelectric attenuation (for a given photon energy) as the complex medium.

A third interaction, pair production, only occurs with gamma rays having energies higher than 1.022 MeV, and even then not in significant proportion, except for large values of Z q of the medium surrounding the borehole. e

The proportion in which these three interactions occur depends on ganma-ray energy and Z q (figure 1.3). e It can be seen that for the range of atomic number, Z,of common rocks (approximately up to Z = 18) about 95% of the gamma- ray interactions with the rock are Compton scattering for the range of gamma-ray energies (see figure 1.3 ) emitted by the natural radioelements and their daughters. If however the uranium content of the rock causes a- significant increase in the Z q of the rock, it is easy to see from figure 1.3 that at lower energiee s the percentage of interactions which are due to Compton scattering decreases and the percentage of photoelectric interaction increases. This photoelectric absorption decreases the

13 ENERGY (MeV)

FIG.1.3. Interaction of gamma-ray energy and atomic number.

gamma-ray count rate and results in a non-linearity in the relation between true or grade and ore grade computed on the basis of garmna-ray count rates. This so called "Z-effect" starts to become an important source of error at about 0.5% U3O8. It can be avoided by counting only gamtia rays of energies which are high enough as to not be affected by the change in Zgq. This may be accomplished with the use of a shielded (i.e. filtered) detector or an electronically set theshold which only permits high energy garmia rays to be counted.

It can be seen from the above discussion that the volume of rocks investigated by the gamtia-ray detector in the probe will depend on the interactions of gamma rays in the rock, and hence also on the physical properties of the rock in the borehole wall. The sample volume is generally defined as the region of rock surrounding the detector from which nearly all (say, 99%) of the gamna rays arriving at the detector originated. The sample volume is often described as a sphere of radius 1 metre or less, centered on the detector. Visualizing the sample volume this way can be somewhat misleading, as it can be construed as mean- ing that any small region within the sample volume contributes the same number of gamma rays to the recorded count rate as any other region of the same volume. The actual situation is shown diagrammatically in figure 1.4. Here, a cross-section along

14 FIG.1.4. Cross-section of a borehole axis showing probe and disintegrating atoms as recorded.

the axis of a borehole, through an infinitely thick homogeneous zone of radioactive material, is plotted showing the borehole, probe and gamna-ray detector. The dots outside of the borehole represent the locations of disintegrating atoms along this section which emitted the ganma rays that subsequently struck the detector and were counted during a given counting period. The probability of a garrma ray from any particular disintegrating atom being counted by the detector decreases rapidly with increasing distance of the atom from the detector. However, at any distance there is always a finite possibility of the gamma ray being detected.

With the above qualifications in mind, a rule of thumb to roughly estimate the radius of the sample volume 3is that the product of the radius (cm) and the densit2 y (g/cm ) of the rock will be approximately 75 to3 120 g/cm . Thus for example, for a rock with density 2.5 g/cm , the sample radius will be roughly 30 to 48 cm.

1.2.7. Radioelement abundances in rocks

The geochemical behaviour of the radioelements controls their distribution in rocks, and at least some fundamental understanding of this behaviour is required for proper planning,

15 TABLE 1.3. RADIOELEMENT CONCENTRATIONS IN DIFFERENT CLASSES OF ROCKS1

U (ppm) Th (ppm) K (%)

Rock Class Mean Range Mean Range Mean Range

Acid Extrusives 4.1 0 8 - 16 4 11 9 1 1 - 41 0 3 1 1 0 -6.2 Acid Intrusives 4.5 0 1 - 30 0 25 7 0 1 -253 1 3 4 0 1 -7.6 Intermediate Extrusives 1.1 0 2 - 2 6 2 4 0 4 - 6 4 1 1 0 01-2.5

Intermediate Intrusives 3.2 0 1 - 23 4 12 2 0. 4 -106 0 2 1 0. 1 -6.2

Basic Extrusives 0.8 0 03- 3 3 2 2 0. 05- 8 8 0 7 0. 06-2.4

Basic Intrusives 0.8 0 01- 5 7 2 3 0. 03- 15 0 0 8 0 01-2.6 Ultrabasic 0.3 0 - 1 6 1 4 0 - 7 5 0 3 0 -0.8 Alkali Feldspathoidal Intermediate 29.7 1 0 9 9 5 -265 0 Extrusives 9 - 62 133 6 5 2.0 -9.0 Alkali Feldspathoidal Intermediate 55.8 0 0 4 -880 0 Intrusives 3 -720.0 132 6 4 2 1 0 -9.9 Alkali Feldspathoidal Basic Extrusives 2.4 0 5 - 12 0 8 2 2 1 - 60 0 1 9 0 2 -6.9

Alkali Feldspathoidal Basic Intrusives 2.3 0 4 - 5 4 8 4 2 8 - 19 6 1 8 • 0 3 -4.8

Chemical Sedimentary Rocks* 3.6 0 03- 26 7 14 9 0 03-132 0 0 6 0 02-8.4

Carbonates 2.0 0 03- 18 0 1 3 0. 03- 10 8 0 3 0 01-3.5

Detrital Sedimentary Rocks 4.8 0 1 - 80 0 12 4 0 2 -362 0 1 5 0 01-9.7 Metamorphosed Igneous Rocks 4.0 0 1 -148 5 14 8 0 1 -104 2 2 5 0 1 -6.1

Metamorphosed Sedimentary Rocks 3.0 0 1 - 53 4 12 0 0 1 - 91 4 2 1 0 01-5.3

•Includes carbonates 'compiled from English language literature by Wollenberg, pers. comm. (1978). from G S C Economic Geology Report 31, 1979 (The proceedings of symposium "Exploration '77", in Ottawa.) interpretation and evaluation of any gamma-ray logging for uranium exploration.

The following is a brief summary of some of the geochemical characteristics of potassium, uranium and thorium. Hie mean radioelement concentrations for a number of different rock types are given in table 1.3.

In the case of igneous rocks, potassium, uranium and thorium all increase in overall abundance with increasing silica content up to and including the pegmatite phase. The relative abundances of the radioelements normally remain fairly constant over this crystallization range. This has useful applications in exploration, since anomalous radioelement ratios could indicate uranium mineralization. The three elements have many geochemical similarities;such as their affinity for oxygen and their relatively large ionic radii. The main difference in geochemical behaviour is an increase in the mobility of uranium under strongly oxidizing conditions which may exist beyond the stage of pegmatite formation or under supergene conditions. This situation arises because uranium has a hexavalent state as well as the quadrivalent state which is characteristic of thorium. However, in the quadrivalent state, uranium behaves similarly to thorium, forming accessory minerals in granitoid rocks concentrated in pegTiatites. There is evidently no great difference of mobility between quadrivalent uranium and thorium or potassium under most igneous rock-forming conditions.

The normal and actual relative concentrations of the radioelements in various types of uranium deposits should be studied, especially those types which are the exploration objective. The five basic types include (a) sandstone deposits, (b) conglomer- ate deposits, (c) vein and palaeo-surface deposits, (d) supergene deposits, and (e) deposits in late stage granitoids.

17 2. THE BOREHOLE: ITS CHARACTERISTICS AND EFFECTS

Borehole geophysical logging, by definition requires the presence of a borehole. This is the means by which a probe gains access to the formation for detailed measurements. The fluids in a borehole, to a great extent, offer the means for the measuring device to contact the formation. They also distort many of the measurements. Therefore, the presence of a borehole, its fluid, its size, its shape and direction are all vitally important to the measurement by sensor from the borehole. See figure 2.1.

The borehole is also a disturbing element to the formation environment. The portions of the formations immediately around the borehole are not the same as those portions farther away from the borehole. The borehole has introduced foreign fluids and stresses into the formation and represents a missing portion of the formation. During the drilling of a borehole, the drill bit may be a rotary type, or a percussive type. It may be a blind or "plug" type, a coring type, or a reaming type bit. It may be a standard or an up-reaming system. In all of these systems portions of the formation are cut away by mechanical force and removed from the borehole. Characteristics of some bit types are shown in tables 2.1 and 2.2.

When the formation is cut away by the drill bit by rotary scraping (drag bit or coring bit) or by hammering (percussion and cone bits) the formation is subject to severe mechanical stress. The stress will be distributed through sane volume, the size of which will depend upon the character of the formation. A slightly consolidated sand formation may confine the resulting strain more closely than a hard formation such as a dense limestone or a basalt.

The work done in cutting away parts of the formation will show up as heat at the cutting interface. The amount of heat will depend upon the rate of cutting and upon the degree of consolidation of the formation. This heat usually is not a serious problem, since one of the purposes of the drilling fluid is to remove it. The drilling fluid may be air, water, a mud mixture, oil, an oil-water emulsion, or a foam. Regardless of the type, the drilling fluid has a great effect upon the formation and upon the logged measurement.

If the formation is porous it will contain fluids. This is the usual situation in sediments where we find hydrocarbons,

18 uranium, coal, evaporites, and water. The fluid may be liquid or gas and is usually, but not always, at a normal hydrostatic pressure. The hydrostatic pressure of a liquid drilling mud is usually adjusted to be slightly greater than that of the formation. The purpose of this is to confine the formation fluids to the formation. This will prevent the entry of the formation fluids into the borehole and will prevent the formation from collapsing. It will also retard deterioration of the qualities of the drilling fluid.

The result of the higher pressure of the drilling mud is to force some of it into the permeable parts of the formations. Figure 2.2 illustrates this.

Water-based drilling muds are composed of water which contains dissolved salts and other compounds and solids which are clays, cuttings, and silts. It is normally circulated down through the central drill pipe, out at the drill bit, around the cutting surfaces, and up the annulus, carrying away cuttings and formation water. At the surface it will flow into a pit where the cuttings will drop out. Additives will be put in at this point, if required. The mud pump picks up the mud and pumps it down the drill pipe.

19 TABLE 2.1. HOLE AND CORE SIZES

SPECIFICATIONS

O.D. OF SET I.D. OF SET BIT SIZE COREBARREL BIT BIT SIZE in irni in nm

XRP XRP 1.275 32.39 0.855 22.48 XRT -.v + XRT 1.160 29.46 0.735 18.67 EX #+ EX, EWX 1.470 37.34 0.845 21.46 EXT * EXT, EWT 1.470 37.34 0.905 22.99 AX #+ . AX, AWX 1.875 47.63 1.185 30.09 AXT AXT, AWT 1.875 47.63 1.281 32.54 BX *#+ BX, BWX 2.345 59.56 1.655 42.04 NX *#+ NX, NWX 2.965 75.31 2.155 54.74 HWX * + HWX 3.890 98.81 3.000 76.20 H -A- H 3.890 98.81 2.875 73.04 EXK * EXK, EWK 1.470 37.34 0.905 22.99 AXK * AXK, AWK 1.875 47.63 1.281 32.54

* Conforms to Canadian Diamond Drilling Association Standards (C.D.D.A.) # Conforms to U.S.A. Standards (D.C.M.A.) + Conforms to British Standards Institute (B.S.I.)

All sizes shown are Mean Dimensions with a tolerance of ±0.005 inch (J.K. Smit & Sons, Toronto, Ontario)

Salts and other dissolved materials were in the original water from which the mud was made, additives may have been used to give the mud particular properties, and dissolved materials are picked up from the formations traversed by the borehole. This liquid portion is called the mud filtrate.

The solid portion of a water-based mud is comprised of clays, cuttings, sand, and silt picked up by the passage of the drill bit through the' formation. Clay and other, solids may be added to maintain viscosity and other properties. This is called the mudcake.

20 TABLE 2.2. DIAMOND CORE DRILLING, DIMENSIONS, WEIGHTS (Courtesy of Longyear Drill Rods)

Size P.P., I.D. Weight inches ran inches , ran lb per 10 ft kg per : : J metre ; A 1 5/16 33 27/32 21 28 4.2 E 1 5/8 41 1 1/8 29 38 5.6 B 1 29/32 57 1 13/32 36 46 6.8 N 2 3/8 • 60 2 . 51 49 7.3

Flush-couples Casing RX 1 7/16 36 . 1 13/16 30 18 2.7 EX 1 13/16 46 1 5/8 41 18 2.7 AX 2 1/4 57 2 51 29 4.3 BX 2 7/8 73 2 9/16 65 47 7.0

NX 3 1/2 89 3 3/16 81 60 : 8.9 HX 4 1/2 114 4 1/8 105 90 13.4

Diamond Coring Bits Size Core Diameter Hole Diameter inches rrm inches mm AQ, AQ-U 1.062 27.0 1.890 48.0 BQ, BQ-U 1.432 36.4 2.360 59.9 NQ, NQ-U 1.875 47.6 2.980 75.7 HQ 2.500 63.5 3.872 98.3 PQ 3.345 85.0 • : 4.827 122.6 BQ 33.5 60.0 3 NQ3 45.0 75.7 HQ3 61.1 96.0

PQ3 83.0 122.6

When the drilling mud is forced into the formation by the hydrostatic pressure of the drilling mud, the restricted permeability of the hole wall will separate the mud into its two components. The mudcake will be deposited upon the wall of the hole at a permeable zone. The mud filtrate will invade the formation.

The thickness of the mudcake will depend primarily upon the water-loss of the mud. Low water-loss muds will generally have thin, tough mudcakes and are considered to be a better quality mud. Mudcakes may be a millimetre thick to several centimetres. It is probable that many of the hole blockages encountered are really mudcake and not caves or ledges.

21 FIG.2.2. Mud invasion of permeable areas.

FIG.2.3. Areas of the invaded zone. The depth of invasion of the formation by the mud filtrate will depend upon the permeability and porosity of the formation. A low porosity and/or a high permeability formation, together with a high water-loss mud will result in a deep invasion by the filtrate. A higfi porosity, a low permeability, and/or a low water-loss mud will restrict the depth of invasion by the mud filtrate. Invasion depth may be a few centimetres with a low water-loss mud to several metres with a high water-loss mud.

As the mud filtrate invades the formation it will displace the formation (interstitial) water. If there is hydrocarbon in the formation, the invading fluid may not displace all of the hydrocarbon. This is because the hydrocarbon will often preferentially wet the sand grains. If the invading fluid is less saline than the formation water and there are clays in the formation, there will be a tendency for the clay to adsorb and absorb the filtrate. This may be accompanied by swelling. This is the so-called heaving shale. It will swell and slough off and will result in permeability changes. If the invading fluid is more saline, the clay will not swell.

Since the invading mud filtrate differs electrically, physically, and chemically from the interstitial fluid, it will change the character of the formation (and of any cores taken). The mud filtrate has a resistivity, R^f,which may be different from that of the formation water. This is one cause of the electrochemical component of the spontaneous potential. The movement itself will generate a diffusion potential or electrofiltration potential. Rjnf is probably different from R , the formation water resistivity. Therefore, the invaded zonw e resistivities, R and R ,will be different from the undisturbed formation resistivityxo , R^z . Refer to figure 2.3. It is sometimes useful to use another fluid than water in the borehole. If ground water is not plentiful or if the permeability is low, the hole may be drilled with air. In this case, a large compressor furnishes air to the drill pipe to blow the cuttings out of the hole and cool the drill bit. A water truck and driver or an expensive mud system can be saved. For- mation damage may be reduced. Very often, reverse circulation (up the drill pipe instead of the annulus) will be used. This gives a Higher velocity for chip removal.

If possible, an air-drilled hole should be logged with radioactive tools and an induction log. Acoustic velocity tools are useless and electric logs are unsatisfactory. Electric logs are often attempted with "scratcher" or contact electrodes. These are invariably noisy.

23 When a small amount of ground water is present and the permeability is low, foam is often used. This is a water-based mud, mixed with a soap or a detergent or other foaming agent, and foamed with air. High air velocities are not needed to get cuttings out of the hole with foam. The characteristics of the foams vary greatly, so it is difficult to predict how an electric log will act. Spontaneous potential curves are often unreliable. Resistivity curves may be surprisingly good. Radio- active logs operate almost as if the foam were a liquid mud. Do not wait until the foam collapses. In general, resistivities of foam will be low. Invasion will be negligible.

Problems with swelling clay and damaged permeability sane- times are combatted with an oil-based, emulsion, or a brine mud.

Oil-based muds and oil-phase emulsions have electrical characteristics similar to an air-filled hole, except for radioactivity logs. Radiation will be attenuated as with a water- based mud. Electric logs are less reliable than in air holes. The induction log will be excellent and acoustic velocity systems are useable. Water-phase emulsions are as unpredictable as foams. In general, electric logs are unsatisfactory in oil-phase emulsions but useable in water-phase emulsions. Radioactive logs should be handled as in water-based muds. Induction logs and acoustic velocity logs are useable. Brines usually affect the recording of electric logs, induction logs and neutron logs.

In general, when a drilling program, and especially the mud program is set up, the logging is not considered. In mineral and engineering logging projects there usually is no mud program. These are serious mistakes. The main purpose for drilling a hole in mineral, engineering, and scientific projects is to get geophysical log data. Even in a petroleum hole, the data gathering is of major importance. A mud program must be designed with logging in mind. Muds must be used to extend the life of the borehole and to increase the ease and reliability of interpretation. Money and time are wasted otherwise. Expensive equipment may be damaged and faulty information may be gathered. It is false economy to try to save money on drilling but cut down the usefulness of the logs and risk lost probes.

The borehole itself will affect measurements of most logging systems because it is part of the measured volume. This is true of the omnidirectional tools. The directional and sidewall tools have usually been designed to minimize or eliminate borehole effects.

24 An omnidirectional probe, unless it is deliberately centered, is assumed to be against the wall of the hole. This is probably a valid assumption, except when the borehole is very nearly vertical (<2° slant) and the logging speed is high (> about 6-9 metres per minute, 360 to 540 metres per hour) in water or mud.

If an omnidirectional probe is centered in the borehole, the borehole is a symmetrical part of the measured volume of the tool. In this case, the effect of the borehole is usually fairly easy to calculate.

The borehole is not a significant portion of the measured volume when it is the same diameter as the sonde. When the diameter of the borehole is increased, the borehole becomes a larger and significant part of the measured volume, until the borehole finally fills the entire measured volume of a centered tool and the detector is measuring nothing but borehole parameters. If the probe is eccentered the borehole has no effect if it is the same diameter as the sonde. This case is identical to the centered case. As the borehole diameter is increased, however, with the probe remaining in contact with the wall of the hole, there is always sane formation within the measurement volume. The end case, as the diameter of the borehole approaches infinity, is that the borehole wall becomes a flat surface with the sonde against it and the borehole fills only one half the measured volume (approximately), with formation filling the other half. Therefore, the eccentered situation is usually preferred. These cases are illustrated in figure 2.4.

It is vital that borehole, mud, and drilling information be noted on the log heading. Items such as mud resistivity, bit sizes, rig type (rotary, core, cable tool, etc.), mud additives, casing sizes and materials,date, and temperature must be noted on the heading.

Often the borehole is cased to prevent its collapse. If it is to be logged over a period of time and then abandoned, the casing will have a minimum of effort and money expended upon it. It will often be plastic, not centered nor cemented. If the hole will be used for production of oil, gas, water, uranium, sulfur, or evaporites, the casing will usually be steel and will frequently be centered and cemented in place. These two situations can have a great difference of effects upon any logging done in and through the casing; figure 2.5.

Casing may be made of PVC (polyvinylchloride), ABS (acrylic resin), polyethylene, polypropylene, fiberglass-polyester, fiberglass-epoxy, aluminum, or steel. The chlorine in PVC and the boron in the borosilicate glass of fiberglass have large

25 FIG.2.4. Effect of centered and eccentered sondes.

Sonde and \ Casing EccenteredI

Sonde Centered Casing fQ) Eccentered V y/ J

Sonde Eccentered Casing Centered Sonde and Casing Centered

FIG.2.5. Centered and eccentered sondes and casings.

26 cross-sections to thermal neutrons and will reduce the counting rates of the neutron porosity tools and the Delayed Fission Neutron (DFN). Steel and aluminum have a severe attenuating effect upon gamma rays. Correction charts and factors are available. Steel will prevent the use of a magnetic deviation tool. An inertial device must be used. Any casing will prevent recording an electric log, but only steel and aluminum casing will prevent the use of an induction log.

All casing and drill-pipe have joints which will additional- ly affect radioactivity logs. These will appear as periodic attenuations every 3, 6 or 9 metres, or 10 or 20 feet.

Drill collars (the weighted section of drill-pipe just above the bit), if they are used, have severe effects upon all radioactive logs because of their thickness. Very often the driller does not know the actual thickness. If he does, it will probably be the new thickness (and it may be worn). It should be measured by the logger. Otherwise, drill collar corrections should be handled the same as casing corrections.

It is worth while to determine if drill collars were used during drilling or if a "pull-down" was used. Most of the small, shallow, mineral type, portable, rotary drilling rigs use a "pull-down". With this, part (or all) of the weight of the drilling rig is put on the top end of the drill pipe by a mechanical or hydraulic system. It is a fast, efficient system. But, the drill-pipe is flexible and this system results in crooked holes. These holes can give problems, especially if the driller has been in a hurry.

Drill collars are usually used in holes deeper than 300 metres. These holes are usually straight and nearly vertical, especially if stabilizers were used. Straight holes may give strange patterns on a deviation plot.

27 3. LOGGING TECHNIQUES There are three general kinds of logging techniques in use in uranium exploration and development at this time. These are radioactivity logging, electric logging and mechanical logging. All three categories have methods which detect naturally occurring manifestations of the type and those \\foich induce or cause various phenomena.

Ganma-ray logging is the prime example of the measurement of natural radioactivity. Neutron and density logging are good examples of the measurement of caused phenomena.

Electric logging is one of the oldest types of borehole logging and a tremendously useful one. The measurement of spontaneous potentials is a good example of measurement of natural potentials or voltages. The resistivity measurements on the other hand, are all the result of causing currents to flow in the formations.

The mechanical logs are many and varied. Probably some of the types, such as magnetic susceptibility, should not be included. But it is a "catch-all" category. It includes acoustic methods, deviation, caliper, and magnetic measurements in this text.

3.1. Common logging techniques This group of logging techniques is a collection of the types most commonly used at this time in mineral logging and especially in uranium work. Included in this group are some methods which have.been in use for a long time. Others, such as the spectrometric gamma-ray methods are relatively new.

This group covers gross-count ganma-ray, spectrometric gamma-ray, neutron porosity, resistivity, spontaneous potential, caliper, and deviation methods.

3.1.1. Gross-count ganma-ray logging

3.1.1.1. Physical basis

Gross-count gannia-ray logging is one of the more extensively used means to discover and evaluate uranium mineralization. The logging measurement exploits the radioactivity of uranium and its decay products. However, uranium itself does not emit ganma rays, it decays by alpha particle enmission. The daughters of 21Ituranium are the source of the ganma rays. Among the daughters, Bi and

28 Spectra from: KUT Water Factor/Borehole Size Model U.S. Department of Energy Grand junction, Colorado, USA Accumulation Time: 30450 sec. (8.46 hours) Model-. 1.52 m (5 ft.) thick, ore and concrete blend 11.43cm (4.5inch)air filled borehole Probe outer diameter: 5.4cm (2.l25inch) Probe shell thickness: 3.2 cm (0.125 inch)SS . Detector, size. No I (Tl), 1.91 cm * 5.1 cm (0.75 inch x 2 inch)

2I"B«23b U)

2O0T1(Z32 Th)

I 1,300

GAMMA-RAY ENERGY (keVl

FIG.3.1. Typical gross-count gamma-ray spectrum.

211, Pb contribute the bulk of the gamma rays recorded by a gross- count gaima-ray logging system. A typical gross-count gamma-ray spectrum is shown in figure 3.1.

Potassium, uranium, and thorium do not contribute equally to the gross-count ganma-ray log. Specifically, 1% K contributes a count rate that is equivalent to the count rate from the uranium series for a grade of 1.8 x 10eU 0 . Similarly, the count rate from 1 ppm eTh translates to a3n apparen8 t uranium grade of 0.4 x 10eU 0 . (1 ppm eU is equivalent to 1.18_x 10-"% eU 0 .) For3 example8 , in a typical sedimentary environment con- tainin3 8 g 10 ppm thorium and 2% potassium the contribution to the gross-count ganma-ray response would equivalently be 0.00041% eU 0 and 0.00036% eU 0 , respectively. 3 8 3 8 For gross-count gamna-ray logging, the basic relation (Scott, et al.1961) is

G T = kA Y The symbol "A" represents the corrected area under the gamma-ray log curve in units of count rate-length; Gy is the average equivalent radiometric uranium grade in a mineralized zone of length T along the borehole. The constant of proportionality, k (k- factor), is determined by the calibration of the system and has the units of equivalent uranium grade per count rate.

29 The units for the area A are specified herein as count rate- length. This is a departure from past practice in which the units for area are only count rate. The specification of area under the log in two dimensions, count rate-length, makes the calibration factor independent of the units used for length. In other words, using the proper dimensions for area, count rate- length, results in a universal calibration factor. This calibration factor will be the same for any unit of length.

The equivalent uranium grade (percent eU 08 or eU) is calculated from the measured gamma-ray activity. 3 The lower case letter "e" indicates that this unit is associated with a measurement of the uranium daughter gamma-ray activity. This daughter activity is a measure of the equivalent parent uranium concentration (assuming secular equilibrium).

The radiometric (count rate) determination (gamna-ray log) does not measure uranium directly, while the chemical determination (core assay) does. They may be different because of disequilibrium. The disequilibrium factor is discussed later in appendix C.

The calibration procedure (see appendix A) consists of a dynamic (or static) measurement of the count rate by a specific instrument, under a set of known standard conditions, in a model or a suitable field borehole, having a known grade-thickness product (GT). The area (or average count rate in the middle of a mineralized GT zone) is determined from the log. One must take care and choose the same unit for length as that of the grade- thickness product assigned to the model or field borehole. The area, A, under the log response can be used to determine k. Since Gy T is known, then

k = A where k has the units of %eU 0e/bount rate or ppm eU/count rate, whichever is used. For a stati3 c measurement in a horizontal zone that is effectively infinite (much more than 1 metre in thickness and diameter) the relation used to determine k is

N G, the average radiometric grade, is known and where N is the average deadtime corrected count rate. The calibration factor has the same units as above.

30 POWER HIGH SUPPLY VOLTAGE

ARMORED CABLE

RATE CHART METER RECORDER

ANALOG/DIGITAL RECORD CONVERTER

PRE AMP ODOMETER

NAI

VPROBE FIG.3.2. Schematic of basic instrumentation for gross-count gamma-ray logging.

In most systems, a deadtime correction must be made to the log response before the area A can be determined. The deadtime is the time it takes for the entire system to process the electrical signal representing the detection of one event and be ready to process the next. Methods for determining the deadtime are given in appendix B. Basically, the equipment necessary to acquire a gross- count gamma-ray log is a probe, surface signal conditioning and recording instrumentation, and armored cable that connects the probe to the surface electronics for signal and power transmission and usually a winch. The natural gamma-ray radioactivity sensed by the detector, usually a Nal(Tl) crystal, is converted into an electrical signal by a photomultiplier tube, amplified and sent via the cable to the surface equipment where it is recorded. A schematic diagram of a typical gross-count system is shown in figure 3.2. The typical surface record is a calibrated pen deflection on a strip^chart as a function of the position of the probe, as it is winched up the borehole. Some of the newer analog equipment may also make a digital record con- sisting of printout of the number of (gamma-ray) counts per second for each successive depth interval along the borehole. The same spectrum as figure 3.1 is presented again in figure 3.3 on a linear scale in Order to show the low energy region of the spectrum better. The probe shell and all other material surrounding the gamma-ray detector will completely attenuate the response below about 50 keV. Probes whose electronic (energy) threshold is

31 1.000-

GAMMA-RAY ENERGY (keV)

FIG. 3.3. Spectrum of Fig.3.1 on a linear scale.

less than the probe shell cut-off threshold are referred to as gross-count probes. Those probes whose electronic (energy) threshold lies above the probe shell cut off are referred to as total or integral count probes. Probes that add additional material between the probe shell and the detector for the purpose of attenuating the low energy portion of the spectrum (e.g. the region below 400 keV) are called filtered probes.

Most of the gross-count response lies below 400 keV. In the spectrum shown, 87% of the gross-count response comes from the ganma-ray count rate below 400 keV. Compton scattering of most of the primary gamma rays has reduced the primary ganma-ray energies substantially.

3.1.1.2. Application

The gross-count ganma-ray log is a relatively quick, simple quantitative measurement. The log is inexpensive compared to the cost of coring and laboratory analyses. The core samples the borehole, the log samples the formation surrounding the borehole. Furthermore, the ganma-ray log samples a much larger volume of the formation than drill core. Comparison of core assays with the log should consider that they do not sample the same volumes. The ganma-ray log can be made in cased boreholes, below or above the water table. Because it correlates with the SP log, it is sometimes substituted qualitatively for the SP

32 log. Besides its application to uranium, the gamma-ray log is useful in the exploration for coal (often low garnia-ray activity) and is used as an index to shaliness.

The gross-count gamma-ray log is useful to explore a large region by logging widely spaced drill holes. It is an asset in the beginning exploration program because of the lithologic and stratigraphic identification and correlation it provides. The gross-count gamma-ray log is the primary means by which to evaluate a uranium mineral body. During production it is used in mine grade-control.

A major drawback to the gross-count gamma-ray log is that it is not a direct measure of uranium. In addition to the effect of disequilibrium there are other causes which result in a (spurious) ganma-ray activity not representative of the degree of uranium mineralization. Radium, a daughter of uranium, has a chemical affinity for iron. In time, the bismuth and lead activity build up, due to the precipitation of radium salts on the casing, giving a false indication for uranium on the gaimia-ray log. During drilling radon gas (another uranium daughter) may also lead to false indications of uranium.

Like any quantitative measurement, the gamma-ray system is calibrated under a certain set of standard conditions. When the logging conditions encountered in the field, in the borehole, and the formation differ from the standard calibration conditions, corrections must be applied to the log values (see appendix C). Only through proper understanding of the measurement and care about procedure can the log be made quantitative.

The gross-count gamna-ray log is limited in its useful sensir tivity to radiometric grades of about 0.01% el^Oa (85 ppm eU) and above. Although this can be overcome by increasing the detector sensitivity, there is reluctance to do this because the probe may become overloaded (electronically) in ore-grade zones. The recommendation for low mineralized regions is to employ a large crystal, spectral gamma-ray system.

The most important procedure, apart from the logging itself, is to record the data concerning the operation of the equipment and the conditions under which the log is run in a notebook, directly on the strip chart, or by some other recording device. Such data as the identifying number of the probe used, its calibration factor, deadtime, and appropriate borehole size/water correction factor, the logging speed and strip chart scales, the identifying hole number and location, its nominal size, total depth and depth of the borehole fluid, and so on, are some of the basic information about the logging job that will be

33 needed after the job is completed. This is covered further in section 5.3.

The logging operation is done while the probe is being winched up the borehole at a constant speed. Typical speeds range from 1.0 m/min to 10 m/min and more. Excessive logging speeds may cause dynamic distortions of the log (including a depth offset) acquired by an analog system. This is not as great a problem with digital systems. It is useful with an analog system to scan the formation as the probe is lowered in- to the borehole, to get a rough idea of the mineralization present and what full-scale deflection is appropriate to use for the strip-chart recorder. The best guidance in setting full- scale is to choose a scale suitable for the unmineralized parts along the borehole. It must be sensitive enough to show the differences due to small lithological changes, but not interfere excessively with the other curves. Mineralized zones will probably be off-scale, in this case, but a repeat section at a less sensitive scale will retain the detail and, at the same time indicate the proper full-scale deflection. The depth scale factor may nominally be set at 100:1 (1 metre of borehole to 1 cm of strip chart) or 50:1 (% metre of borehole to 1 cm of chart).

While logging, it is useful to estimate the grade of the mineralized zones. One can, neglecting the deadtime correction, use the thick zone approximation, G = kn. The lower case letter "n" is used to emphasize that the count rate is not corrected for deadtime. (Some may recall that this approximation is G = 2kn for % foot calibration factor or G=10kn for a 10 cm calibration factor.) It must be emphasized that this is an approximate expression for on-the-spot use. It is invalid for thin zones (zones less that about 1 metre thick).

The interpretation of thin, dipping beds has been given some attention recently. The relation

1 Gy T = kA (la)

is still valid. But care must be taken to determine the area under the log of a thin bed by using intervals that are at least as small as the true thickness, T, of the bed. When such suit-

able depth intervals 1are chosen, the thickness in (la) is an effective thickness, T . It is related to the true bed thickness by the cosine of the dip angle, 0:

Ti = cos 9

Gy T = kAcosQ (2)

34 GT = k • A . cos ©

Figure 3.4 is a composite of logs taken in thin, dipping bed models at Grand Junction, Colorado. On the left in figure 3.4 is a schematic diagram of the models indicating the respective dip angle 6 (unlike the figure, the actual thin zones do not project onto each other). On the right is a composite of logs taken in each of the 50 mm thick zones. Each zone has a grade of 0.251% eUaOs (a GT product of 0.013% eU 0 metres). The logs were made with air in the borehole. After 3makin8 g the deadtime correction and choosing a depth sampling interval of 5 mm, the areas Al through A4 were determined. Multiplying each area by the respective cosine of the dip angle makes the areas, A2, A3, and A4 equal to Al. The product of the calibration factor and the area does produce the GT of 0.013% eU30 metres for the model (within the uncertainty of the measurement)8 . Also, note that the count rate at the peak of the anomaly does not remain constant with angle.

Keep in mind that most reserve calculations are made assuming horizontal zones. The dip angle is used to correct the result of the reserve calculations. In such a case, there is no need to use the cos0 factor for each log input to the reserve calculation.

35 FIG.3.5. Artificial gamma-ray anomaly.

COUNT RATE (ARBITRARY SCALE, x 1000) 0.0 1.0 2.0 3.0 I i • ! i I I i i i |

COUNT RATE VALUES

Ni 20

N2 1 16

Nj 568 ^

N4 1866 0.15m

Ns 2890 y

3040 N6

Nt 3344

Ns 1632

N» 424

Nio 100

Nil 10

SUM 14,010

DEPTH AREA » SUM X INTERNAL

A • 14,010 X 0.15

A • 2101.5 count rata-metar

FIG.3.6. Area of Fig.3.5 computed by the total-area method.

36 COUNT RATE (ARBITRARY SCALE, x IOOO) 0 1.0 2.0 3.0

E, 1460

Ez 590 = 1460 (E.»E,) 2050

COUNT RATE VALUES

1.38 (E | + Ez) 2629

N, 2780

NZ 2950

N3 3270

N« I 980 SUM 13,809

AREA = SUM x INTERVAL

A = 13809 x 0.15

A = 2071.4 count rate -meter

DEPTH

FIGS. 7. Area of Fig.3.5 computed by the tail-factor method.

In an exploratory drilling program, the region under consideration may have already been drilled for gas, oil, or coal. The ganma-ray logs made for these purposes may be reported in API units (Belknap, et al. 1959). Such logs may provide some useful information for uranium exploration (approximations only). As- suming a proper API calibration and that the measured activity originated from the uranium series in equilibrium

5 1 API unit = 2x 10~ % eU 0 3 8 or. 1 API unit = 0.17 ppm eU

Garrma-ray logs of formations having low gamma-ray activity (such as oil and gas gamma-ray logs) may have significant count-rate contributions frcm potassium and the thorium series. The use of a spectral gamma-ray system is recommended to determine such contributions. API logs require deadtime and hole size corrections and have been designed for low level detection. They are completely unsuited to any use where the radiation level is more than about 5 times the shale level. Low grade uraniferous zones are often seriously out of equilibrium, also.

37 COUNT RATE (ARBITRARY SCALE, « 1000) 0.0 1.0 2.0 3.0 I ' ' I i I ' • • • I ' • ' • I ••

GRADE (%eU30e) 1 0.10 15m 0.18 T 0.12

0.25

0.05

DEPTH

FIG.3.8. How the anomaly of Fig.3.5 was constructed.

3.1.1.3. Analysis In this section there are examples of the quantitative analysis of gross-count ganma-ray logs. Figure 3.5 is an artificial gamma-ray anomaly constructed to show how the area is computed by a total-area method, figure 3.6, and the tail-factor method, 3.7. In each case, area is calculated by approximating the gamna-ray log by rectangles having length in count-rate units and a width in uniform length units. The tail-factor method is convenient for hand integration becuase it approximates the area under the tails of an anomaly. A step-by-step calculation using the tail-factor method is given in appendix A and in the section that follows. Figure 3.8 shows how this artificial gamma-ray anomaly was constructed (the same way that an actual log is comprised) by a sequence of mineralized zones. In this construction, each of the zones was 15 cm thick.

Selected portions of the logs shown in figures 3.9 and 3.10 exhibit fairly typical anomalies. The logs are a typical suite: ganma-ray, SP, resistance and thermal neutron logs. The depth scale of these logs was originally 240 to 1 for presentation purposes. Quantitative analysis, however, is more conveniently

38 Hole no.: RD-5 Date: 2 June 1979 Location: State Wyoming County Sweetwater Sec. 12 T 24N R 94W Elev: 6864 ft Drilled depth 799 ft Logged depth 799 ft Fluid Driscos Gel Drill bit size: 4Va inch Casing; type, depth none Gamma: scale 500 cps/div baseline 0 k 1 214x10' d.t. 0.96 us SP: scale 20 mV/div baseline 3 div to right Resistance: scale 1000/div baseline 5'/2 R Neutron: scale 500 cps/div baseline 7'/i R, type thermal spacing 17 inches Remarks: k units, % eUsOs/cpS; water level at 16 feet: chart full scale = 10 inches

0 500 cp_s 500 cps

FIG.3.9. Analysis of a portion of log shown in Fig.3.6.

39 Hole no.: RD-10 Date: 30 August 1979 Location: State Wyoming County Sweetwater Sec. 2 T 24N R 94W Elev: 6700 ft Drilled depth 600 ft Logged depth 599 ft Fluid Driscos Gel Drill bit size: 5 in/0-600 ft Casing: type, depth none Gamma: scale 100cps/div baseline 0 k 1.124 x 10"5 d.t. 0.96 /us SP: scale 20 mV/div baseline 5Vi div to right Resistance: scale 1000/div baseline 5R Neutron: scale 500 cps/div baseline lVi R, type thermal spacing 17 inches

Remarks: k units, % eU30d/fcpS; water level at 2 feet; chart full scale = 10 inches

100 cps 500 cps

FIG.3.10. Analysis of a portion of log shown in Fig.3.7.

40 done with logs on a 50 to 1 scale. The use of the English system of units is of little consequence to the analysis except when dealing with depths along the borehole and the choice of the tail-factor value. Use of the calibration factor (defined in appendix A) having units of %eU 0account per second (cps), allows the analysis to be applicabl3 e in any system of units, metric or English. It is most important, however, that the analyst be consistent in the choice of units. This area determination is done according to the tail-factor method for the reasons outlined earlier. The tail factor, 1.38, has been empirically determined for 0.5 foot (15.2 cm) intervals. In principle it should scale inversely with the interval. For example, a one-foot tail factor would be 2.76 (=1.38 x 1.0/0.5). Similarly, a 15 cm tail factor is 1.40 (=.1.38 x 15.2/15.0). However, because the value 1.38 was empirically determined for 0.5.foot (15.2 cm) intervals, inverse scaling to intervals smaller than 0.5 foot may result in unacceptable errors.

The total-area and tail-factor methods (figures 3.6 and 3.7, respectively) can be easily implemented and refined by automated data processing techniques. It is recommended, however, that manual methods be thoroughly understood before more sophisticated schemes are employed.

The eleven steps of the basic analysis procedure are listed below. This procedure will be used in the analyses that follow.

1. Determine the half-amplitude point Ei, on the flank of the upper boundary, and the corresponding depth. 2. Determine the half-amplitude point, on the flank of the lower boundary, and corresponding depth. 3. The difference between the half-amplitude point of the lower boundary and Ei is the thickness, T, of the anomaly, if the anomaly is thicker than 1 metre. 4. Beginning at Ei, determine the deflections at regular intervals. E2 is the first interval at or below (in depth) the lower boundary of the anomaly. Note that the intervals do not need to be regular. However, each count-rate must be multiplied by the appropriate width. 5. Multiply Ei and E2 by the appropriate tail factor. 6. Apply the deadtime correction to each count-rate value. 7. Determine the area. 8. Apply correction factors to the deadtime corrected area. 9. Multiply the corrected area by the calibration factor to obtain the average grade-thickness product, GT. 10. Divide GI by the thickness, determined in step three (above), to obtain the average radiometric grade. 11. Multiply the average grade by the Z ff correction factor, if known, to get the true average radiometrie c grade.

41 In the examples that follow, the log deflections are measured (and indicated in the figures) in units of grid deflection. The depth is measured in feet. A 0.5 foot tail factor will be used where appropriate. The corrections are made relative to the standard calibration condition. The gross-count probes used in obtaining these logs were calibrated in Grand Junction using the N3 standard model. The calibration conditions are listed in appendix A.

Example 1

The character of this anomaly, taken from the log in figure 3.9, is similar to the response in a thick-zone calibration model except that this thick zone is not as uniform as the calibration model. In 3.11, the Ei point is chosen after first determining the maximum deflection nearest the upper boundary. The point Ei is one-half the maximum deflection. In this example, Ei occurs at a depth of 312.8 feet. Near the lower boundary, the small shoulder on the log is chosen as the maximum deflection. The lower boundary of the anomaly is the point one-half of this maximum deflection. The depth at this half-amplitude point is noted at 321.0 feet. The difference between the upper and lower half- amplitude points is the thickness, T (8.2 feet), of the anomaly.

42 Sheet !_ Borehole Number _ Dots Logged 2 .lung 1979 WORK SHEET GROSS-COUNT GAMMA-RAY LOG

«at»>

Logging

Operator , Unit No. Proa* 7? , x toerag> logging iipeeil ..60 ffiet/minUK—calibration factor! %eU(Os/cp»

Borehole Pei«otlo •.imm»_686l!_Fee • • t M-^nnrniimip | tgUIBWB|» • •

T«ol 799 Feet drilling field Oriscos Gel F1

Bit .I»./1.P»K A 5/8 inches/0-799 feet r..i.,/j.r.i. none

Log Boeellne m_0 Deflection 500 cps/div t.n scole Ji£00_c£s_ Clwft 5 ffft/dWUim Remarks

Correction Factor* 0-96 Borehole

Del lection n N Deflection n N Deflection n N

1 .Sit 970 971

0.95 475 475

E. + E» 1 i.i.6

I .M(C|+Eg) 1995

3.00 1500 1502

3.72 1860 1863 3.60 1800 1803

3.21 1605 1607

3.05 1525 1527 2.92 11(60 11.62

3.11 1555 1557

3.16 1580 1582

3.H 1570 1572

3.28 16<<0 161)3

3.1.7 1735 1738 1.18 1590 1592 3.00 1500 1502

2.77 1385 1387

2.37 1185 1186

1.68 840 841

SUM 26,361

FIG.3.12. Example 1: deflection worksheet. WORKSHEET GROSS-COUNT GAMMA- RAY LOG 3tieet 1— Borehole Nat<„ RD-5 DM* Lo«(e*2 June 1979

Laaer boundary wi.o feet

llpr» tuuMhiry (rt 312.8 feet

Thickness (T1 8.2 feet

Tail-factor _Li2

Sum. 1.38 (E, + E?) + ZN 26,361 _cps

Summation interval 0.5 feet

Area 13,181 cps-feet

Corrections

Siie/water Casing Moisture

1.13 f 1.0 X 1.0 - 1.13

Corrected Area 1'(.895 CDS-feet

Dif«qiiilihriuffl .0,87

Disequilibrium corrected area 12,958 cps-feet

_5 Calibration fnrtor 1.2H * 10 »eU,0Q/cP3

Average grade-thickness 0.157 ^eU^Ofl-feet

0.019 Average grade (Gy) % eUsOB

Zeff

Zen corrected grade 0.019 % eU3Oe

FIG.3.16. Example 2: sum of tail and deadtime corrected "counts".

44 FIG.3.14. Example 2: analysis of detail.

Note that a bed boundary always coincides with an inflection point on the curve.

Beginning at the upper boundary Ei, the deflections are read off the log at regular intervals. The point E2 is the first interval beyond the lower boundary, half-amplitude point. The deflections are entered on the worksheet, figure 3.12, and are converted to count-rate (counts per second) by multiplying each deflection value by the deflection scale conversion (500 cps/ division). Next, the count-rate values are corrected for deadtime using the formula (2B) given in appendix B.

The sum of the tail "counts" and the deadtime corrected "counts" are added and entered in the worksheet shown in figure 3.13. The "counts" were sumned every 0.5 foot so the summation interval is 0.5 foot. The area is the product of the sum and the summation interval.

The borehole is 4 5/8 inches diameter, filled with water. It was assunmed to be bit-size because a caliper was not run. There was no casing in the borehole so the casing correction factor is 1.0.

45 »«• 1

WM. Uyrminn W.efwatpr s— 2 r 24N » 94U

Logging np...!.. Il.» a. 182^ D.„k. y. _75 A»erage lowing tpeejfel) fget/lflinutB Ccllbrotloi toctor 1.211 rt 10 % • u,0,/est

Borehole rkMtiM 6700 feet — Total 6°° Drilllne llnnPriStg? Gel Fluid lewl 2 feet ail «i.»5.0 inch/0-600 feet none

Log nulla. nm 0 n«flMti.,n —k. 4500 cps rkni a.m 5 feet/division

Correction Foctors 0-96 »« ^liui 1.145 1.0 uol'loo 1.0 Biux>illlMu> 1.09 7 It l.o

Deflection n N Deflection n N Deflect loi n N

E| 1.09 545 545 Ez 0.96 480 480

e, + e2 0.81 405 405 0.62 310 310 I.S8(t|+E,) 0.42 210 210

0.63 315 315

0.72 360 360 SUM 28.186

<1.81. 420

1.15 575 575

1.53 765 766

2.56 1280 1282

3.98 1990 1994

6.29 3145 3115

8.51. 1.270 1.288

7.54 3770 3784

6.19 3095 3104

1..36 2180 7185

2.65 1325 1327

2.18 1090 1091

1.76 88a 8S1 1 .42 710 710

FIG.3.18. Example 3: deflection worksheet.

46 WORKSHEET GROSS-COUNT GAMMA- RAY LOG

2 - •• -1 Rmkafe Ifur RP-10 Oat* Loao* 30 August 1979

Lower boundary 387.2 feet

l|ppf» ^."rtn-Y lnt 386-1 feet

Thickness (T1 feet

Toil-factor not used

0 Sum- ,(E, + E7) + £N 28.186 epc

Summation interval 0.25 feet

Area 7047 cps-feet

Corrections

Size/water Casing Moisture

1 .145 , 1.0 . 1.0 1.145 Corrected Area 8069 cps-feet

Disequilibrium corrected area 8795 cps-feet

rnlihrotinn fnMnr 1 .214 x 10"5 ieU^/cps

Average grade-thickness 0.107 SeU^Og -feet

Average grade (Gy) 0.097 <*, aU30„ z.„ '-0

Z«ff corrected grade 0-097 %bUjO0

FIG.3.16. Example 2: sum of tail and deadtime corrected "counts".

47 FIG.3.17. Example 3: anomaly taken from Fig.3.9.

After the corrections have been made, the correcte-5 d area is multiplied by the calibration factor (1.214 x 10 % eUaOs/cps) to obtain the average radiometric grade-thickness product, GT. This grade-thickness product is then divided by the thickness, T (8.2 feet), giving an average grade of 0.018% el^Oe.

Example 2

Shown in figure 3.10 at a depth of about 387 feet is a good example of a thin zone. In the analysis of thin zones, the summation interval chosen must be small enough to accurately determine the area. The best procedure for this example, shown in detail in figure 3.14, is to determine the area from background to background, the total-area method. Beginning with the choice of the peak as the first deflection point, the other deflection points are chosen every one-half of a small division on either side of the peak. (This corresponds to a 3 inch or 0.25 foot summation interval.) Include the complete anomaly.

The thickness, T, of this thin zone was estimated to be the distance between the half-amplitude points on each flank of the anomaly. It was determined to be approximately 1.1 feet. If the curve inflection points can be detected, it is much more accurate

48 Sheet -J of —5 BoreMe Number RP-5 Date Lotted 2 41106-1379

Deflect Ion n N Deflection n N Deflection n N

2.24 1120 1 121

1 870 871

e, + E! 1992

I.M lE|tE > 2 2749

2.94 11(70 1472

3.37 1685 1688

4.32 2160 2164

4.16 2080 2084

3.88 1940 1944

3.73 1865 1868

4.32 2160 2164

14.16 2080 . 2084

3.88 1940 1944

3.73 1865 1868

3.72 1860 1863

3.68 1840 1843

3.21 1605 1607

2.63 1315 1317

2.04 1020 1021

SUM 29,682

FIG.3.18. Example 3: deflection worksheet.

49 WORKSHEET GROSS-COUNT GAMMA-RAY LOG

BoreMIe fta*erjiIL£__ Dote L««»*2 .lijne 1979

Lower boundary i\i 7 fpet Upper boundary (al E|) 707.0 feet Thickness tri s.7 feet

Tail-factor 1.38

Sum. 1.18 (Ei + E7) + XN 29.682 eps

Summation interval Q.5 feet

Area i*i.8iii cps-feet

Corrections

Size/water Cosing Moisture

i.n « i.o x 1.0 = l.n Corrected Area 16.770 cps-feet

Disequilibrium 0-87 Disequilibrium corrected area U.59Q cps-feet

Calibration metar 1.21 fr x to'^eujOg/cps

Average grade-thickness Q.177 %eu^08-feet

Average grade (Gy) P-03' % eUjOg

Z««r corrected grade o.Mi % eU,Os

FIG.3.19. Example 3: sum of tail and deadtime corrected "counts". FIG.3.20. A typical logging system.

to use them to find the bed thickness. Figures 3.15 and 3.16 give the results for each step in this example. The average radiometric grade is determined to be 0.097% eU 0 . Note that if the tail-factor method is employed in this example3 8 , the resulting area under the anomaly is about 20% less than that determined by the total-area method.

In sunmary, choose a simulation interval that is significantly less than the thickness, T. Begin the total area method with the choice of the first deflection point on the peak of the anomaly, working outward along each flank to include the complete anomaly. This will usually be on the order of a total depth interval of 1 metre plus the bed thickness. It is an aid to work with a log recorded at an expanded scale.

51 Example 3 •

Here, the problem is to determine the grade-thickness above a predetermined cutoff grade. Figure 3.17 is an anomaly taken from another log. The cutoff, 0.015% eUaOs, corresponds to a count- rate of 1335 cps or a deflection of 2.67 divisions, calculated from the relationship G = kn. Determine the area under the anomaly between about 707 and 711 feet, which is the amount above the cutoff grade.

This is a thick zone so the tail-factor method is used. The maximum deflections nearest either flank, but above the cutoff, are chosen. The half-amplitude point, E:, is determined and the difference in depth between Ei and the other half-amplitude value is the thickness, T. The remainder of the analysis is similar to the first example. Step-by-step results are given in figures 3.18 and 3.19. The grade-thickness above cutoff is 0.177% eU 0 feet. 3 8

Figure 3.20 shows a logging system (surface equipment only) which is in common use at this time. It primarily records the gross-count garmia-ray curve. It does, however, also record a single point resistivity curve, an SP curve, and other radioactivity curves, with the addition of the proper downhole tools.

3.1.2. Spectrometric gamma-ray logging

In areas where the radioelements thorium and potassium contribute a significant proportion of the natural gamma-ray activity of the rocks, the gross-count garnma-ray log cannot be used to measure the uranium content. Spectrometrie gamma-ray logging equipment utilizes the information contained in the energy spectrum of the gamma rays to separate out and determine the contribution of each of the three radioelements. The required instrumentation is necessarily more complex and expensive than for gross gamma-ray counting. Also, it becomes even more important to have a good foundation knowledge of natural radioactivity, radioactive equilibrium, and the interactions of gamma rays in rocks. These points have been discussed in the introduction.

The spectrometric gamma-ray logging equipment must be calibrated in model holes in a similar fashion to the gross-count logging equipment. However, numerous additional calibration factors must be determined besides the sensitivity or K factor. Once the system is calibrated and these factors applied, the resulting corrected count rates for the three radioelements can be handled in a similar way to gross-count logs for computation of the thickness and grade of uraniferous zones.

52 gamma-ray energy (MeV)

FIG.3.21. Gamma-ray energy spectra obtained in model boreholes using a 32-mm X 12-mm sodium iodide detector: (a) spectrum from a model borehole containing the radioelement Th (thorium); (b) from a U (uranium) model; (c) from a K (potassium) model. These spectra were obtained with a 256-channel spectrometer, each channel being about 12 keV wide. The shaded areas are the traditional K, U and Th spectral windows centered at 1.46, 1.16 and 2.62 Me V, respectively. After stripping, all gamma rayswith energies between the upper and lower cutoff energies for the Th window are attributed to the thorium decay series, and similarly for uranium and potassium. Generally, the "contents"of these three windows are stripped and then plotted beside the total-count window (i.e. all gamma rays with energies greater than some predetermined minimum such as 0.4 MeV) against depth to give the 4-channel spectral log.

3.1.2.1. Physical basis of spectrometric ganma-ray logging

The principle natural gamma rays causing the energy spectra were shown in figures 1.1 and 1.2. The potassium spectrum is a single gamma-ray peak at 1.46 MeV (not shown). The relative activities or intensities are those obtained under conditions of radioactive equilibrium. Disequilibrium may introduce differences in the relative peak heights depending on whether any given daughter emitter has been removed or deposited. A ganma- ray spectrometer has the ability to measure the energy of a

53 — decreasing energy

FIG.3.22. Schematic presentation of the interplay between the K, U and Th windows.

ganma ray before it is counted. This principle is used to dis- criminate those gamma rays associated with each of the three radioelements in three separate energy regions or windows of the spectrum and record the counts in separate memories or counters.

Due to the inherently imperfect energy measuring capabilities of scintillation detectors and scattering of gamma rays and degradation of ganma-ray energies in the source rock, the measured spectra for uranium and thorium decay series will look more like those shown in figure 3.21 (center and bottcm). The measured potassium spectrum will resemble that shown in figure 3.21a,

Figure 3.21 also indicates commonly chosen K, U and Th energy windows for a differential spectrometer 14 . These are

1.36-1.56 MeV (I< window, 1.4 21

(U window, 1.76 Me 20V peak of Bi), and.2.4-2.8 MeV (Th window, 2.614 MeV peak of ®Tl). Sometimes the upper limit of the K window coincides with the lower limit of the U window, and some workers raise the upper limit of the U window to 2.4 MeV. It can be seen that there is interference between the three spectra; some gamma rays originating from the Th decay series will be counted in the U and K windows, some gamma rays originating from the U decay series will be counted in the K window and to a small extent in the Th window, and finally some small portion of the K gamma rays may be counted in the U window. To determine the "k-factor" or sensitivity for the individual radioelements K, U and Th, the gamma-ray counts due to each individual radioelement must be determined. This means the above-mentioned interference must be taken into account and the gamma-ray count rate in each window mast be corrected. This is accomplished by the procedure known as spectral stripping.

54 The stripping factors

The determination of the stripping factors using model boreholes will be described below. Figure 3.22 is a schematic re- presentation of the interplay between the three radioelement windows K, U and Th and identifying the stripping factors, as listed below:

Stripping Factor Used to Strip Off

a Th gamma rays in U window B Th ganma rays in K window Y U gamma rays in K window a U gamma rays in Th window (usually small) b K ganma rays in Th window (zero) g K ganrna rays in U window (approx. zero)

For many purposes, only the first three stripping factors are used and a, b and g are assumed to be zero. For high uranium concentrations the upward stripping factor "a" is necessary. It has a value of approximately 0.05, but the actual value is dependent on factors such as detector size and resolution, window widths and energy settings.

The stripping factors are applied to the digitized gamna-ray spectral log as follows:

Th = ^ ftiv Duvuy-CI ; ^ m^s (3.1) + U(l-bB) _ Th(gB-a U-J) + K(ba-g) U = -^"VftP T Duyi-up ; ' ft/ Q 2)

K _ Th(ay-B) + U(aB-y) + K(l-aa) s D (3.3) where Th, U and K are the gamma-ray intensities recorded in the thorium, uranium and potassium windows of the spectral log, respectively, in counts per second. Th^, U and Kg are the- § corrected (stripped) gamma-ray intensities m those same windows. The term in the denominator, D is given by

D = 1 - gy - a(a-gB) - b(B-ay)

As stated above, the factors b and g are negligibly small for borehole logging detectors and may generally be eliminated from the above equations, giving

55 Th 111 aU (3.4) s = ~D

U = U ~ ctTh (3.5) s D = Th(ay - S) + U(ag - -y) + K(1 - aa) (3.6) D where D is now given by

D = 1 - aa

The C- factors or sensitivities

To avoid confusion it has been suggested that the term "k-factor" be used to refer to total-count or gross-count sensitivities, while "C-factors" be used to refer to the sensitivities of the individual window counts. Thus the conmonly used k-factor is a sensitivity for total count if all the counts are due to uranium alone. When thorium and/or potassium are present, the relation GT = kA, where A refers to area under the gross- count or total-count log, is invalid. The count rate in the individual spectral windows must be related to the grade-thickness product for each radioelement potassium (K), uranium (U) and thorium (Th). Thus we have in the spectral windows:

(3.7)

(3.8)

G T C Th Th ^Th (3.9) where the Ai refer to the areas under the stripped window count log. The C factors or window sensitivities, will be in units of "grade/count rate".

The stripped ganma-ray intensities from equations 3.4, 3.5 and 3.6 are used in equations 3.7, 3.8 and 3.9 to compute the concentrations of potassium, uranium and thorium in a similar fashion to the quantitative interpretation of gross-count logs explained in section 3.1.1.

Determination of the calibration factors

The stripping factors and C-factors are determined by making measurements in model boreholes containing ore zones of known K, U, and Th concentrations.

56 The model holes are logged in the same way as field holes using the same logging parameters and the same hole size, casing, etc. as used in the field. Correction factors must be applied to adjust for any parameter which is different from field conditions .

The count rates obtained by logging the model holes, and the known grades and thickness of the ore zones are then used to compute all of the above calibration factors.

Depending on the material in the ore zones, two methods may be used: (a) If the ore zones contain more than one radioelement, each will contribute gamma-ray counts to the windows as described earlier. In this case at least three model holes are required to solve for the calibration factors. The count rates obtained, and the radioelement concentrations of the ore zones are then used as input to a regression analysis (least squares) computed program which solves for the stripping factors and the sensitivities or C-factors.

(b) If the model boreholes contain monoelemental ore zones, the

stripping factors and sensitivities may be determined : simply, as follows (instead of by regression analysis ) record the spectrometric gamma-ray log through the ore zones (say 1.5 m thick) in each model borehole. The areas under these curves are used to compute the stripping factors and the sensitivities from the following relations:

Stripping Ratios

Computed only U Area u _K Area from Th models " Th Area p Th Area

Computed only r_K Area a Th Area from U models U Area U Area

Computed only _Th Area g _U Area from I< models ' K Area K Area

Sensitivities (using C - GT/A)

Computed only , _(%K) x (1.5) from K models 'K K Area Computed only , _(ppm U) x (1.5) from U models V U Area Computed only , _(ppm Tli) x (1.5) from Th models Th Th Area

57 | GAMMA RAY INTENSITY Ccounts/s) 1

0 400 0 15 0 15 0 15

FIG.3.23. Four-channel log from model hole with thorium ore zone.

If the object is to measure uranium only, the minimum require- ment for monoelemental model holes would be one thorium model to obtain a and one uranium model to obtain a and CJJ. In the field, these values would be used with equation 3.5 to strip the uranium window log and with equation 3.8 to compute the grade- thickness product of the mineralized zone.

Consider the 4-channel log shown in figure 3.23 obtained in a model hole containing a thorium ore zone. These four curves represent the response of the total count, potassium, uranium and thorium channels. Although the amounts of uranium and potassium in the thorium ore zone are known to be negligible, the uranium and potassium channels show deflections which are due to gamma rays from the thorium source. The data from this thorium ore zone could be used to derive a, 3, and Gj^ from the relation given above. It is easy to see from this spectrometric gamma- ray log that the interference of thorium in the uranium channel is quite large. In field logs, false radioelement indications, such as these, must be removed by stripping of the spectral log. The stripped logs may then be used with the C-factors to obtain grade-thickness products. It should be noted that the areas under the log anomalies will have units of count rate x distance (depth). Thus, if count rate is in counts per second and depth is in metres, the units of A will be "metres-counts per second".

58 When A is multiplied by the C-factor in units of grade/count rate (e.g. ppm eU/count per second) the result will be the grade- thickness product in ppm eU-metres.

3.1.2.2. Application

Spectrometric gamma-ray logging is a relatively recent development and it is usually not available in most conmercial multiparameter logging systems. Because of this, the rather specialized instrumentation for spectrometric ganma-ray logging is discussed in this section along with its application, rather than in the later section on instrumentation.

Gamma-ray spectrometry or pulse height analysis The essential details of a scintillation counter and its method of operation can be understood by reference to figure 3.24. Gamma rays which interact in the crystal cause tiny flashes of light (scintillations) the intensities of which are

59 Detector pulses Level In comparator

Logic signals •Out ^ > To counter or ratemeter _J FIG.3.25. Schematic diagram of a single channel analyzer.

[Scintillation | Preamplifier

FIG.3.26. Four-window spectral logging system.

60 FIG.3.27. Block diagram of typical borehole spectrometer system.

proportional to the energy deposited in the crystal by the ganma rays. The photomultiplier tube, which is^optically cou- pled to the crystal, usually with transparent grease, converts the scintillations to corresponding electrical signals which can be amplified and sorted in the subsequent electronic circuitry.

The technique of sorting pulses according to their amplitudes is known as pulse height analysis. The most elementary method is to use a single discriminator circuit (a "threshold" or "integral" spectrometer) which allows all pulses above a certain preset amplitude (i.e. energy) to be counted and rejects all others, or vice versa. A slightly more sophisticated arrangement uses two such discriminators which can be preset to levels (a "differential" spectrometer). Pulses having amplitudes which fall between the two levels (an energy window) are counted while all others are rejected. This arrangement is known as a Single Channel Analyzer (SCA) and is shown in figure 3.25.

Spectrometric logging equipment which utilizes a SCA can have only one output recorded at a time. This means for complete information from the windows relating to U, and Th and K, several separate runs in the borehole must be made. However, most conmercial spectral logging systems have three or four windows, arranged as shown in the block diagram of figure 3.26. Three of the single channel analyzers were set to cover the pulse amplitudes (i.e. energy bands) corresponding to those of potassium, uranium and thorium, while the fourth was set to cover an energy range encompassing all three, the "total count".

61 Most of the readily available gamma-ray spectral logging systems evolved from surface portable ganuia-ray spectrometers which were adapted to the borehole environment by the manufacturers .

The main components of a typical system are shown in the block diagram of figure 3.27. Some of the important points re- garding the equipment are given below.

Data recording

An analog recording system will sometimes be less expensive to purchase than a digital system, and perhaps will be somewhat more reliable under field conditions (although this is rapidly changing). A digital system eliminates the expense and errors of manual digitization of the data and generally makes the data processing easier, if any but the most basic processing techniques are to be used. The processed results from a digital system will generally be more accurate than those from an analog system. A hybrid recording system wherein the signal is passed through a ratemeter, as in an analog system, and then digitized eliminates the tfedious and error-prone step of manual digitization required by true analog recording. This system shares many of the disadvantages of pure analog recording, however, and should not be confused with the true digital system where the counts are sunned for a preselected period (say, 1 second), and then the sum is written on tape along with depth, etc.

The detector

To evaluate high-grade uranium deposits, the scintillation crystal (detector) must be quite small, perhaps as small as 1 by 1 cm, to avoid swamping the electronics with gamma-ray pulses. A small detector .(say, 2.5 by 2.5 cm) is also best for delineat- ing thin zones and resolving complex sequences of zones of dif- erent grades. For lower-grade deposits and occurrences, a large" detector, perhaps 2.5 by 7.5 cm, will be needed. For stratigraphic logging in a non-uraniferous environment, still larger detectors may be required in order to detect enough ganma rays to make the results statistically reliable. Detectors on the order of 5 by 25 cm are commonly used in the standard 11 cm dianeter boreholes drilled in exploration for roll-front type uranium deposits in sandstones. In most hard-rock areas, where smaller diameter holes are the rule, large detectors of this type are impractical. In this case, special high-density detector materials (which are still experimental) instead of the standard sodium iodide crystals may provide the answer. For spectral ganma-ray logging, larger detectors will generally give higher-quality spectral information.

62 '37qs single energy

FIG.3.28. Calculation of detector resolution.

One of the standard measures of performance of a scintillation detector is the resolution of the 662. keV single energy peak of the isotope 137Cs. This is a measure of the sharpness of the photo peak and is by implication a measure of the ability of the detector to resolve two closely spaced peaks. Figure 3.28 illustrates the way that detector resolution is calculated and specified as a percentage. A detector of dimensions 2.5 cm x 7.5 cm would conmonly have a resolution of about 10 percent or slightly less.

The winch

Hand-cranked winches are often favored in difficult terrain for their relative reliability and portability; but they also have severe disadvantages. The tedium involved in attempting to crank the winch at a constant low speed for hour after hour is considerable, and tends to lead to operator inattention, resulting in undesirable speed fluctuations. A motor-driven winch is heavier and requires a generator for power; but the convenience of operation and the improved accuracy of the results make the motor drive a highly desirable feature of the winch.

The depth counter

Experience indicates that winch-mounted depth counters of the capstan-and-pressure-roller type tend to be inaccurate and unreliable. Such devices should be checked regularly. This may be accomplished by marking off the cable at regular intervals of say, 50 metres (by using a tape measure), and then comparing the cable marks against the depth counter readings as the probe is

63 lowered down the borehole and brought back up again. If significant errors (more than a few cm) are found, the problem is either with the depth counter or due to cable stretch. This latter, however, is not very likely. A well-head pulley depth counter assembly with an optical encoder should give improved accuracy. Calibration, as described above, can help compensate for the systematic depth errors in any depth measuring system, and allows a cable-stretch correction to be applied if required.

Energy calibration and spectral stabilization

As mentioned earlier, the spectrometer is a pulse height measurement device which accepts detector pulses from the main amplifier and counts them whenever a detector pulse falls between amplitude limits set on two panel controls. The most serious problem with spectrometric logging is the difficulty in determining the energy limits to which the calibrated panel controls actually correspond. In the case of portable four-channel spectrometers for exploration work, a fifth channel is usually incorporated, preset to cover the energy peak from an isotope such as the 662 keV peak of 137Cs, to allow for manual energy calibration before logging the borehole. The other window limits also have preset controls not accessible on the instrument front panel. As a probe moves along the borehole, ambient temperature changes will affect the properties of the detector and the electronics in the probe. As a result, the spectrum can expand or contract drifting out of "energy calibration", perhaps to the extent that the energy peaks will move entirely out of the windows set up for them (figure 3.28). Various spectral stabilization techniques are being used to compensate for this shift, such as the monitoring of an internal calibration source in the detector by an additional window. However, most existing techniques have some limitations; for example, the latter technique runs into difficulties at high count rates where the instrumentation cannot distinguish between gamma rays from the internal source and the natural gamma rays from the rock.

Field procedure

Upon arrival at the borehole, the winch (drawworks) and pulley (sheave wheel) are set up, and the probe is attached to the cablehead and lowered into the hole to allow it to come into thermal equilibrium with the fluid in the hole. The spectrometer, interface unit and chart recorder are then connected and readied for logging. If a system with spectral stabilization is being used, the energy calibration would be carried out next. The best procedure for a spectral logging system without spectral stabilization is to leave the probe in a water-filled section of the borehole (if possible) for some time before bringing

64 it up to calibrate the spectrum (using a 137Cs source, generally) . Allow at least 20 minutes for a 3 cm diameter probe and 30 minutes for a 4 cm diameter probe to approach the temperature of the water in the borehole. Larger probes require correspondingly longer times to equilibrate. The probe power should be on during this equilibration period. Of course, because the temperature varies with depth along the borehole, the calibration may still shift during the logging, and only a stabilized system can compensate for this shift. In an air-filled hole, the equilibration period would be very long and spectral stabilization is virtually a necessity. The energy calibration usually consists of placing a small 137Cs source (or other source provided by the manufacturer) against the probe at the detector, and adjusting the voltage or gain control on the front panel or on the interface unit until a ratemeter dial on the spectrometer indicates a maximum. This centers the calibration window (a narrow window set at 0.662 MeV) on the 137Cs peak, and thus the spectrometer windows for K, U(21"Bi) and Th(rtsTl) are properly set. The procedure usually takes less than 30 seconds. The probe may then be returned to the hole, and the depth counter set properly to begin logging the hole.

A constant logging speed is important with both digital and analog systems, otherwise the problem of making a good interpretation becomes difficult or impossible. Although it is important to log slowly for accuracy, it is even more important to use a short sampling interval. A sampling interval of about half the detector length or less (down to a practical minimum of 1 or 2 cm) should be used for best results. (For an analog system this is the digitizing interval of the analog record.)

Few analog logging systems offer chart speeds which are fast enough to give good results. An analog system that can't produce 5-10 cm of chart recording for each metre of borehole should probably be modified. Of course, this sort of ratio (1:20 or 1:10) may only be required for detailed logging of interesting zones, but it should be available for that purpose for best results.

Logging a hole

Most spectrometric logging equipment conmonly employs analog chart" recording and the following discussion refers to such a system.

In a new area it is probably a good idea to carry out some experimentation with the equipment using various combinations of ranges, time constants, logging speeds and chart speeds,. It is also useful to adopt two sets of parameters. The first set of

65 parameters can be considered the "reconnaissance" settings. The second set can be considered the "detailed" setting.

Since the selection of ranges, time constants and chart speeds are different on different equipment, it is not possible to recommend any specific values for these instrument settings. However,a typical"reconnaissance" setting with a 2.5 x 7.5 cm detector might consist of approximately the following values. Logging speed 4 m/min; chart speed 2 cm/min (this gives 2 m of depth per cm of chart paper, i.e. 200:1 ratio); time constant of 2 seconds. The range would be set to magnify information between peaks on the log for possible lithological information. This usually means that anomalous zones go off scale on the log, but these are captured later on the "detailed log". If a single pen chart recorder is being used, this reconnaissance log would be a log of the total-count channel.

Although it is generally accepted procedure to log holes from bottom to top, it is often useful to log on the way down with the total-count channel. This reconnaissance log locates the anomalous zones which must be logged in detail, in order to do a quantitative analysis of the mineralized zones. The detailed log is run at a slow speed, say 0.3 m/min and the range is adjusted to ensure that the entire anomaly is recorded for later digitizing. The chart speed must be the same as before, in order to expand the depth scale on the chart. If the logging speed/chart speed is sychronized, a different ratio than the previous 200:1 must be used in order to expand the depth scale. This is necessary to accurately digitize the anomaly on the chart for the grade-thickness computations. Separate runs will be required for each window if a single pen recorder is used, whereas digital recording or use of a multi pen chart recorder will provide all the required spectrometric information in one run. It is very important that the depth and count-rate scales be determined and marked on the strip chart recording for later interpretation and application of the calibration data previously obtained in model boreholes. One method of ensuring that the count-rate scale is accurately known is to switch the instrument to "accumulate" mode with the probe stationary in the hole and the chart recorder running. In "accumulate" mode the count rate in each channel can be accurately determined (the statistical size of the sample is increased), and related to the trace on the chart by counting for an extended period of time. The zero setting should also be checked to insure that zero counts are indeed recorded as zero on the strip chart.

One of the main problems which can be encountered in logging is the varying speed of either a motorized or a manual winch. With a constant chart recorder speed, the depth scale on the

66 chart will vary with the winch motion. Thus, the depth scale expands and contracts as the winch motion is adjusted to maintain a reasonably constant 4 m/min. This would be less serious if all four channels are recorded at once, but when the log of each channel is run separately, the varying scales are difficult to match and correlation of peaks in the varying channels becomes tedious. Ideally, for analog chart recording the chart drive should be synchronized to the winch motion, and a fiducial marker should be provided to insure that depths marked on the chart coincide with the depth of the probe. The chart recorder should be bi-directional, or at least have the capability of reversing the deflection direction and moving the zero to the opposite side of the chart. This will permit logging both up and down the hole and at the same time maintain increasing gamma-ray intensity as a positive deflection when zero depth is at the top of the log trace.

3.1.2.3. Analysis of a spectrometric ganma-ray log

Figure 3.29 shows a typical raw (digital) ganma-ray spectral log in a mixed uranium-thorium environment. From the total- count log (3.29a) it is impossible to tell which radioelement series is causing the anomalies. From the thorium channel it appears that there is significant thorium in the four distinct radioactive zones between about 52-56 m, but little above or below those zones. However, quantitative information is still lacking. In figure 3. 30 the log has been stripped on a point by point basis according to equations 3.4, 3.5 and 3.6 (and smoothed slightly to reduce the statistical noise). Now it is clear what radioelements are responsible for the various anomalies. Using the techniques described in section 3.1.l| along with equations 3.7, 3.8 and 3.9 it is possible to determine the concentration of each of the three radioelements in any anomalous zone. The virtual elimination by the stripping process of all potassium anomalies in the example is especially impressive. It now is clear that about half of the apparent uranium anomaly at 55 metres in the unstripped log was due to thorium gamma- rays.

3.1.3. Neutron logging

3.1.3.1. Introduction

Neutron-neutron (neutron porosity) logging is an effective technique designed to measure the porosity of sedimentary formations. It is based on the ability of hydrogen, which is present in sedimentary formations primarily as water, to moderate (slow down) high-energy neutrons to low energy. The neutron population can be interpreted in terms of the formation's water

67 COUNTS/S 0 T.C 10000 o0 K . 500 0 U 400 0 Th .200 F

E i1 H 0. Ill 0 V ® ® © © FIG.3.29. Typical raw gamma-ray spectral log.

COUNTS/S T.C. 10000 0 . . K . . 500 0 U 400 0 Th 200

£ x i- Q. I UJ Q \ ® ® © © FIG.3.30. Log of Fig.3.29 as stripped and smoothed.

content, mechanical, elemental and chemical composition. Since most of the hydrogen in a formation is contained in fluids, neutron porosity logging is used to measure the relative amount of fluid in the formation. There are, of course, certain factors that can lead to aberrations in the results. As the technique is, to a certain extent, sensitive also to chlorine and trace elements, correction for these elements is required.

68 Similarly, the water present in clays can give a misleading picture of the formation's porosity. Nevertheless, such aberrations can be used to characterize and quantify particular conditions, especially when neutron logging data are correlated with information yielded by other porosity measurement techniques. Neutron-based techniques can also be effective in nor- malizing the results obtained with certain other techniques, such as DFN (delayed-fission neutron). In addition, neutron cross-sections and capture ganma-emission data can be used for elemental and lithological determinations. Neutron porosity logging circuitry is similar to other types of circuitry for radiation detection, the main difference being the type of detector used. The most conmonly used device for neutron logging is the helium-3 detector. Most casing materials are quite transparent to neutrons. Difficulties in interpretation may result, however, from non-centered casing or the use of PVC materials for the casing.

3.1.3.2. Principles

There are virtually no free neutrons in the earth's geological formations. One can therefore assume that any neutrons detected in logging are the result of the neutron logging process. This means that there is rio background component to be considered.

All neutron logging devices contain a neutron source. There are three general types. The first type source is isotopic and undergoes spontaneous fission with the emission of neutrons. The major representative of this type source in logging is califomium-252 (2!iCf). It has a 2.63 year half-life (T>) and an output of about 0.12 neutrons per second per Becquerel at 2.3 MeV average energy (6 MeV maximum) It is a moderate ganma emitter. It emits about 2 x 10b n/s per microgram.

The second type source is the alpha-beryllium source. It is an isotopic alpha-particle source. The alpha particles are reacted with beryllium (sometimes lithium or boron) to get neutrons. Some of the isotopes used in this reaction are listed in Table 3.1. The most common one, at this time, is 21tlAm. It has a reasonably long half-life (433 years) and a low energy ganma output. Thus, decay corrections and shielding problems are not. difficult.

The third type of source is the linear accelerator or neutron generator. This type source usually uses the deuterium- tritium reaction

2H + 3H ->- "He + n (14 MeV)

69 TABLE 3.1. CATALOG OF NEUTRON SOURCE MATERIALS

RADIOACTIVE MATERIAL ACTIVE HALF-LIFE NEUTRONS NEUTRON OTHER GAMMA PHOTONS GAMMA ISOTOPE PER CURIE''* ENERGY-MeV EMISSIONS PER CURIE-'-" RADIATION Per Second (Ave.) (Max•) (Type) (MeV) Per Second Rh-'irr'Cj-' PoIonium-beryl1ium 210Po 138d 2.5x106 4.2 10.87 a 5.3 0.0001 6 Y 0.80 1.54xl0

Californium 252Cf 2.65y 4.4x10 9 2.3 6 a 6.11 y 0.2-1.8 2.4xl010 0.15

Anericium-beryllium 211 'Am 458.ly 2.2x106 4.0 11 a 5.44(13.8%) 5.48(85%)

y 0.017(48.19%) 13.7xl09 0.026(3.52%) lO.OxlO8 0.043(.08%) 22.2x10° 0.060(48.19%) 13.7xl09 0.099(.026%) 7.4xl06 0.001 2.8.10'0(total)

Radium-beryllium 226Ra l,620y 1.5x107 3.9 13.08 a 4.8(94%) 4.6(6%)

•Y 0.19 0.2-2.2( daughters) 1.5x10" 0.974

Plutonium-beryllium 239Pu 24,400y 2.2xl06 4.5 10.74 a 4.89(75%) 4.85(25%)

Y 0.017(99.36%) 10.7xl0e 0.039 (.07%) 7.4x105 0.053(.24%) 25.9xl05 0.060* 0.1 (.18%) 20.25xl05 0.124(.09%) 9.25x105 0.384(.05%) 5.6x10s 0.011 10.8xl08(total)

See Glossary, appendix D. * Americium daughter build up with decay of plutonium. The accelerator sources may be operated in a pulsed or a continuous mode. Outputs range from 107 to 1013 neutrons per second, depending upon the tube design and power.

Regardless of which type process or reaction is used to generate the neutrons, they enter the formation at high energy (velocity). Within the formation the neutrons will undergo collisions with the formation atoms and lose part of their energy to them.

The energy transfer with hydrogen is extremely efficient. Therefore the primary moderator in the formation is hydrogen. It is six times more likely to thermalize neutrons than the next closest, common geological element, carbon.

Moderation will also depend upon the likelihood of a collision, or cross-section, I. I is the "target" size of an atom. It is a probability function and is in barns. One bam is 10 2h square centimetres.

Primarily the neutron population at thermal and epithermal energies is a function of the amount of hydrogen in sedimentary formations and the original or source neutron flux rate. (Thermal energies are <0.03 eV and epithermal energies are <100 eV). The systems using this reaction are called n - n systems.

The thermal neutron population also depends upon the occurrence of thermal neutron reactions. Since there are many of these, the thermal population may be depleted by other materials in the formation (e.g. chlorine, boron, gadolinium, and other high cross-section isotopes).

When a neutron reacts with an atomic nucleus garrma energy is emitted. The likelihood of absorption and reaction is a function of the reciprocal of the neutron velocity. Thus, thermal neutrons are most likely to be absorbed.

The ganma energies resulting from neutron capture are characteristic of the element (or isotope) reacting. Therefore, spectrographic detection of capture ganrnas provides a means of elemental identification. This is a very important technique in mineral logging. The number of capture gamnas is a measure of the thermal neutron population, or the amount of hydrogen present. These are n - y systems.

Gamma detection (for porosity determination) must take the chlorine and trace element concentration of the formation and borehole into account.

71 Also, at thermal energies neutrons will easily cause fission in uranium-235 and thorium-232. This is used in the prompt fission neutron (PFN) and delayed fission neutron (DFN) techniques. These will be discussed in a later section.

3.1.3.4. Detectors

Since a neutron has no charge and no electromagnetic phenomena associated with it, the detector must be somewhat different from gamma and charged-particle detectors. These neutron detectors employ materials which have large cross-sections to neutrons and all react with neutrons to emit an alpha particle.

In logging, there are three major types of detectors in use. The early neutron porosity equipment (some still in use today; Schlumberger GNT series, for example), and all of the spectrographic equipment (elemental analysis) make use of gamiia-ray detectors. The response of these must take the formation "chemistry" into account. These are the n - y systems.

A second type detector is of the scintillation type, using a material which has a higih neutron cross-section, such as a lithium compound.

A third class are the tube or counter type detectors. The most common of this type, and the most camion in present day logging equipment is the helium-3 detector. Sometimes the counters are lined with uranium to increase their efficiency. These last detectors are used in the n - n systems. All of these reactions are sensitive to epithermal as well as thermal neutrons. The relative sensitivity to epithermal neutrons can be increased by surrounding the detector with cadmium. Cadmium has an extremely high cross-section to thermal neutrons, but a much lower one to epithermal neutrons. There- fore, it will absorb thermal neutrons and pass most of the. epithermal neutrons. All of the currently used neutron detectors have pulse outputs, with one pulse representing the passage of one neutron through the system. Efficiencies are usually high and backgrounds are negligible. Therefore, in spite of often low counting rates, reliable measurements may be made with neutron detectors. With the ganma detectors, the natural gamma background must be taken into account.

3.1.3.5. Neutron-porosity systems

As a neutron diffuses through a sedimentary formation, the hydrogen in the formation is the most efficient agent for slow-

72 APPARENT NEUTRON POROSITY - %

APPARENT NEUTRON POROSITY - % (FROM CHARTS OBTAINED USING LIMESTONE TEST BLOCK DATA)

FIG.3.31. Estimated neutron correction for formation chemistry effects (neutron-neutron logging, water-filled holes).

ing the neutron. There are few chances of absorption of the neutron before it reaches epithermal energy (1 to 100 electron volts). Thus, the population of epithermal neutrons is a function of the efficiency of moderation of the neutrons. This type detection has a rather low counting rate. Therefore, it is not used as much as a mixture of thermal and epithermal detection. However, it makes an extremely good, linear system.

When the neutrons have been slowed further, to thermal energies, the population will also depend upon the cross-section of other materials in the formation. These materials will absorb thermal neutrons and deplete their population. A moderate correction for formation composition must be made with

73 100 200 300 400 500 600 700

NORMALIZED COUNTING RATE

FIG.3.32. Neutron porosity index in limestone as a function of normalized counts per second. Normalized to 40 cps in water. 1 CiMiAm-Be source. (Century Geophysical Corporation, 1978-12-16)

74 Thermal Neutron D ^ Population \

/ Operating fIhermalSv \ Range / Neutron J Density / ' 1 -— Distance from Source w

FIG.3.33. Thermal neutron population.

thermal neutrons. Figure 3.31 shows this type of correction for a mixture of thermal and epithermal neutron detection. Figure 3.32shows a typical family of correction curves for a thermal- epithermal system.

The detection of the capture gamma rays has the greatest amount of variance,when it is used for porosity measurements. The capture ganma-ray intensity will depend upon the thermal neutron population, the average capture cross-section of the formation, the borehole fluid amount and type, and the sensitivity of the detector to natural gannia rays. Typical ganma detection curves for neutron porosity equipment are shown in figure 3.33.

As the neutrons diffuse away from the source, they are slowed down at a rate which will approximately depend inversely upon the atomic weights of the formation atoms. Thus, there will be a Gaussian distribution of thermal neutrons around the detector, forming a spheroidal shell whose radius will depend inversely upon the porosity of the formation. In a 35 percent porosity sandstone, about 90 percent of the signal for thermal neutron detection comes from a depth of less than 0.25 metres. There- fore, the peak of the thermal neutron distribution will lie about 10 to 13 centimetres from the source. As the porosity is decreased to zero (for a carbonate formation) this distance (of the peak) will increase approximately 2.5 times that or to 25 to 30 centimetres. Since the detectors on most modem logging tools are on the order of 38 to 50 centimetres from the source, they read the variation on the far slope (figure 3.33). The counting rate is an inverse function of the porosity.

75 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 API UNITS

FIG.3.34. Response of GNT-F + G in limestone; basic curves at standard spacing. (Courtesy of Schlumberger Well Services, Inc.) FIG.3.35. General Formula for counting rate, CR = a + b log Q.

100 • |,_ 200 -I- 300 M i _ J , ' I Ml I iLTj

FIG.3.36. Overlay method of scaling field logs approximately.

1134 Most porosity logging tools are omnidirectional devices which ride the wall of the hole by gravity. One must be careful, with such tools, that the probe does not float away frcm the hole wall because of the Bernoulli effect. This is a danger with any hole whose deviation is less than about two degrees.

3.1.3.6. Interpretation

Neutron-porosity systems are usually interpreted (for porosity or percent water) by means of empirically determined charts. Examples of these are shown in figures 3.32 and 3.34. The general form for the counting rate, CR can be closely approximate by:

CR = a + b log <}> where a is the intercept at 100 percent water, b is the slope of the curve (both will also be function of hole size) and is the water-filled porosity. This is illustrated in figure 3.35.

The overlay method is a method for approximately scaling field logs which can be used with a fair degree of success. It is valid on any type of neutron-porosity log. If we pick two deflections which represent known apparent porosities, a logarithmic scale can be laid between them and the values for any other deflection can be determined. One of the values may be shale, which has an apparent porosity of about 45%. The other may be a formation where cores have been taken or it may be an estimated porosity. The logarithmic scale may be any length which will fit on the log. An example is illustrated; figure 3.36. Lay the scale on the log and read off apparent porosities. The accuracy will be as good as your determination of the two reference points. At worst, the relative porosities will be accurate.

In general, the neutron porosity tool is a good system to measure porosity. It must be remembered, however, that it will "see" the bound water in clay or shale as apparent porosity. Sensitivity of the neutron-porosity systems to the water content of clay, which is not effective porosity, makes it necessary to correct for the presence of clay. To do this, determine the apparent porosity, <(>a from the neutron curve, making a correction for the rock type. Then determine the apparent porosity in a pure shale, (jig^. This latter will be on the order of 0.45 to perhaps more than 0.80. Use the gamma-ray curve to determine the actual clay or shale content, in the formation of interest: Y - Y„ V = - * sh y u ~ Y ' sh 's

78 where y is t±ie gamma-ray reading in the horizon being examined, Ys is the ganma-ray reading in a clean sand (shale-free), and

Ysh is the gamma-ray reading in a pure shale. Multiply the apparent shale porosity su by the shale content, Vg^ and sub- tract the product from the apparent porosity,

n = ^a ~ ^sh^sh

Because of the sensitivity to the formation rock composition, the neutron-porosity tool is an excellent one to use in cross- plotting for lithology determinations.

In mineral exploration a borehole is often drilled with air and is water-filled only to record a resistivity curve. Since both the resistivity and neutron curves are porosity curves, the neutron porosity can be used in place of the resistivity. The neutron tool is also effective in cased holes, especially when the casing is steel. Thus, the cost of a water truck may be saved.

In hard-rock logging, porosity is not a significant factor. But, the neutron response is sensitive to the thermal neutron cross-sections of the various materials. Thus, it can effectively be used as a lithology measurement in hard-rock environments. One should use a fairly long spacing in this case.

3.1.4. Resistance, resistivity, and conductivity

3.1.4.1. Introduction

If a material which contains free charged particles is subjected to a voltage gradient the particles will flow and transport their charges. The material will conduct an electrical current. The impedance to this flow is called electrical resistance. Resistance is a function of the geometry of the flow and the resistivity of the material. Resistivity is a fundamental property of a material.

The current flow in a sediment is primarily within the fluids in the pore spaces. Thus, the determination of resistivity is indicative of the character of the pore space and material. Resistivity of the material and the interstitial fluid can be related to the pore volume.

Resistivity curves are useful for correlation, stratigraphy evaluation, porosity calculation, depositional history, and determination of formation water character.

79 Wire Voltage drop, V in the wire

1' ©n

Cross-sectional area

Voltage drop, V1i n the block

FIG.3.37. Diagrammatic presentation of Ohm's Law and its application.

Various tools have been designed for specific types of measurements; shallow, deep, symmetrical, and surface. Some are focussed and others are not. Each system has advantages and disadvantages.

3.1.4.2. Physical description

If a material contains free charged particles, such as ions or electrons, and is subjected to a voltage gradient or a voltage field, the charged particles will move under the influence of the field and will transport their charges through the field and do work. The rate of particles flowing is called the current. The impedance to their flow is the resistance.

Such a situation is encountered if one connects the two terminals of a battery (a voltage source) with a wire (a material containing free electrons)(figure 3.37).There will be a voltage drop or voltage gradient, V, along the wire, measured with a voltmeter. The flow of electrons, A (current), can be measured. If the voltage drop is in volts and the current is in amperes, the relationship, resistance, ft in ohms is

ft = V/A

80 TABLE 3.2. RESISTIVITIES OF GEOLOGIC MATERIALS

Material Resistivity Range, ohmmetres Sulfur 109 to 1016 Petroleum Biotite Sylvite Mica Quartz Muscovite Halite -10" to 101" Calcite 107 to 1012 Cinnabar 106 to 1010 Sheelite Limonite Anhydrite 10" to 1010 Sphalerite 105 to 107 Bauxite 102 to 106 Bituminous Coal 10 to 106 Basalt 103 to 105 Gabbro Diabase 103 to 10" Gneiss 102 to 10" Sub-bituminous Coal Hematite 10-3 to 106 Dolomite 1 to 10" Limestone 80 to 6 x 103 Lignite . • 5 to 103 Conglomerates up to 3 x 103 Siderite 10 to 103 Sand up to 103 Slate Marl up to 600 Siltstone up to 300 Peat Argillite Braunite 0.01 to 100 Ilmenite Marcasite Shale up to 15 Pyrolusite 1 to 10 Chalcopyrite 10"3 to io-1 Sulfides Pyrite io-" to 10"1 Magnetite 10"" to io-2 Bornite 10"6 to lO"2 Galena 10"5 to io-2 Pyrrhotite . 10"5 to 10"" Graphite io-6 to 10-" Native Metals 10-6

81 This is Ohm's Law. If another piece of material, such as carbon, rock, or metal is put into the circuit, as in figure 3.37, the resistance of that piece of material may be determined.

A conversion factor, R

n = R(L/A)

the resistivity, is a fundamental property of the material and is in ohmmetres. Thus, if we can determine the resistivity, R, we will have a means of identifying a property of the material:

R = fi (A/L)

This property is independent of the geometry of the material. It is of the same nature as the color, smell, taste, density. The reciprocal of R is conductivity and is in mhos per metre. If a general geometric factor, K is substituted for the particular one, (A/L)

R = Kfi

This equation is the fundamental relationship used for resistivity measurements.

Within the formation, the flow of current is primarily within the fluid filling the pore space of a sedimentary formation. Notice that most of the common rock materials, such as quartz, calcite, and other sedimentary materials have extremely high resistivities. Their resistivities range from more than 10 ohmmetres to over 1012 ohmmetres. The major exceptions are the clay minerals (see table 3.2).

The fluids filling the pore spaces are usually solutions of salts and other compounds (mostly ionizable) in water. These compounds are often dissociated and, therefore electrically conductive.

When the pore water and the quartz sand are pictured as resistors in parallel, it can be seen that virtually all of the electrical current will flow in the medium with the lower resistivity, the pore water solution, rather than in the sand grains. If the water, for example, has a resistivity of one ohmmetre and the sand grains 10 ** ohnmetres, 1011* times as much current will flow in the fluid as in the sand, itself.

Thus, the resistivity measurement in a "clean" sand (or any other shale free sedimentary formation) is primarily a function of the electrical effects of the pore space (figure 3.38).

82 FIG.3.38. Resistivity measurement in "clean"sand.

Shale in the formation is a solid material filling part or all of the pore space in a sand. But, it is very conductive compared to the sand material. In fact, it has a conductivity or resistivity comparable to that of the pore fluid. There- fore, when a formation contains shale, the shale content will have to be taken into account in any analysis for actual porosity.

Once the resistivity of a sedimentary material has been determined, it may be used to examine the character of that material.

The ratio of the resistivity of the formation, R , to that of the fluid filling the pore space, Rw, is called the Formation Resistivity Factor, F, or simply, the Formation Factor:

F = R /R o w If the.formation is shale free, has a granular porosity, and is completely water saturated, then the Formation Factor, F, is t-i _ ,-m _ a F - ad) = ,m

83 FIG.3.39. Determination of the formation factor.

\ 10 \ \ Carbo lates \ \ a Sands 1.0 \ \ m 0.1 \i 0.5 1.0 2.0 3.0

FIG.3.40. Determination of the values of a and m for the formation factor. If the formation is not completely water saturated, the amount of saturation, S , is the fraction of the pore space filled with water: 1

sw =

These expressions can be used in several ways. The apparent porosity of the formation, under several conditions, can be calculated, The comparison of the deep and shallow values of apparent porosity can give the amount of oil, gas, or air in the formation. Cross-plotting with another porosity device can help identify lithology, correct porosity, and water resistivity,

If a shallow device, such as a short normal, single point, or a microdevice is used, the resistivity of the flushed zone,

R^q, may be measured. Since the flushed zone is filled with the fluid portion of the drilling mud,:the mud filtrate (see chapter Borehole Effects and Mid), the resistivity value, R^, of this fluid may be measured on the surface, eliminating one unknown. In this case:

The value of F or of the apparent porosity.in the shallow zone may also be determined from any of the shallow porosity devices, such as the Neutron-Porosity, Density, or Acoustic Velocity systems.

3.1.4.3. Instrumentation

The circuit shown in figure 3.41 is a eingle point resistance measuring circuit. A single point system will always have one downhole electrode, combining both the current and the measure circuit. The return electrodes are at a great distance away.

The downhole electrode may be separated as shown in figure 3.42into a current electrode, e^ and a measure electrode, e^. If the measure circuit uses a separate electrode from the current circuit, the system is known as a normal array or a Normal Resistivity circuit. The return electrodes are still at a great distance.

85 FIG.3.41. Single point resistance measuring circuit.

FIG. 3.42. Separation of current and measure electrodes.

FIG.3.43. Grounding resistance of an electrode. A lateral or Inverse circuit will always have three or four electrodes downhole, not at great distances from each other.

An alternating current source is usually used because the electrical current flow causes chemical reactions to take place at the electrode surface. These reactions interfere with the flow of current, but are short-term-reversible. Therefore, the periodic reversal of the flow of an alternating current tends to nullify the effects of the chemical reactions.

There are some minor exceptions, but the principle of all resistivity systems remains the same; an electrical current is caused to flow through the subject material, the voltage drop along the current path is measured, and the resistance or resistivity is calculated. The electrode configuration determines the measurement geometry.

3.1.4.4. Single point resistance and resistivity

If a spherical electrode is placed in a homogeneous medium, as shown in figure 3.43,the resistance of the medium is called the "grounding resistance," of the electrode. The grounding resistance of a spherical electrode is

n = R ,o e 4TT dg/2

where R is the resistivity of the medium surrounding it and de is the electrode diameter. Thus, the grounding resistance is directly proportional to the resistivity, R of the medium.

The sphere of investigation.of the single point system contributing 90 percent of the measurement has a diameter, d^t

= 90 w~ Qjr ~ d~7r) °- Air de/2

Thus, a 3 centimetre diameter electrode will receive 90 percent of its signal, or measurement from a sphere 30 centimetres diameter in a hcniogeneous medium. Anything outside this sphere contributes to only 10% of the signal. Anything outside about five times this distance is essentially an infinite distance away and its contribution is negligible. It.becomes apparent that the contribution of the borehole volume can be a major factor in the single point resistance measurement, whether the electrode is centered in the borehole or eccentered. This is shown in figure 3.44.

87 Centered Eccentered

FIG.3.44. Single point investigation, centered and eccentered electrodes.

—®-I A

6 c\ (b)

FIG.3.45. Single point resistivity circuit (a) and (bj.

A similar procedure can be followed to determine the distance to the return electrode, assuming the influence of the position of the return electrode to be less than one percent. Solving for dp, in this case it is 100 times the electrode diameter. Therefore, with a 3 centimetre diameter electrode, the return electrode position will have less than one percent influence upon the measurement, if it is more than 3 metres away from the measure electrode.

The total circuit, including the high impedance measuring circuit,is shown in figure 3.45a and b.

A spherical electrode is difficult to incorporate into a downhole tool, mechanically. Therefore, the electrode is usually made cylindrical. However, except for the expression for

88 the grounding resistance, the effects are the same as for spherical electrodes.

The expression for the grounding resistance, fleoo of a cylindrical electrode is

fteoo = (R/4rrLe) In (2Le/d£) where Le is the length of the electrode and In is the logarithm to the base e (the natural logarithm). It is identical to the

spherical electrode when Lg = 0.82dg.

Notice in the above equation that both the length and diameter of the electrode must be known in order to solve for the resistivity, R. It is good practice to scale the log in resistivity and adjust the circuit to correct to resistivity. This is done routinely for all other arrays.

The combination of an extremely shallow depth of investigation, a linear dependence upon resistivity, and a response symmetry make the single point resistance curve quite useful for many purposes:

R = ftK s Le = 4tt ZLe In

Several things are indicated by these equations:

1. The response of the Single Point system is symmetrical, and therefore excellent for correlation. Single Point signatures are quite representative of formation character. Bed boundaries are at curve inflection points.

2. The resistance deflection, the resolution, and the curve smoothness will all depend upon the electrode length. In- creasing the electrode length will result in a decreased deflection, and increased curve smoothness and depth of investigation, and a decreased resolution.

3. The shallow depth of investigation might be used to exam-

ine the flushed zone around the borehole and determine Rxo

4. The response of the Single Point system is extremely responsive to borehole conditions. A larger borehole or a lower resistivity mud will greatly lower and smooth the curve. ' / '

89 FIG.3.46. Single point resistivity departure and hole size correction.

The determination of R^ will depend upon being able to separate the effects of the borehole from the effects of the formation. This relationship has the appearance of a pair of parallel resistors:

1 = _L + (1 ~ b) R R R m xo Figure 3.46 shows the probable relationship, for an eccentered tool, of the value of "b" as a function of hole size and formation resistivity (Departure Curves). Notice the great variation of reading with hole size. Also notice the greater linearity at low resistivities and the dependence upon mud resistivity.

Figure 3.47 shows the current distributions around the Single Point electrode under various conditions of resistivity. Notice that the current distribution can depart radically from a radial one. Consequently, the equipotential surfaces will depart severely from the spherical one described by the conversion constant. This problem is also illustrated by the departure curves in the flattening of the curves at the resistivities greater than R . m

90 FIG.3.47. Current distributions around a single point electrode under various resistivity conditions.

91 This last problem, that of a changing current distribution around the current electrode (the geometry), is the major im- pediment to using the Single Point quantitatively. There is an extreme loss of resolution when the formation resistivity is more than twice the mud resistivity. Therefore, the major usage of the Single Point curve has been qualitative. It can be quantitatively used if care is taken. It is a poor curve for coals, limestones, saline muds, and large hole diameters.

3.1.4.5. Return electrode

So far the influence of the return electrode has been ignored. It has been assumed that it has a negligible grounding resistance.

Figure 3.41 shows that the measurement of the resistance includes not only the variable grounding resistance of the measure electrode, but also the constant grounding resistance of the return electrode. They are in series, therefore their grounding resistances must be added. If the curve is to be used quantitatively the return electrode grounding resistance must be kept extremely low or must be subtracted from the final resistance value.

There are two major options for handling the return electrode problem. First, we may make the electrode a short length which results in a significant grounding resistance, but hold it constant. Then this value need only be subtracted from the total reading on the log. This means that it must be kept in a constant environment, such as the mud pit. Freezing, drying, or nonexistent mud pits are a problem. A metal stake can be used, but this is a substantial problem in the arid climates of many mineral projects, because of the high resistance of the surface layers.

It is common practice to return the resistance or resistivity current circuit to the cable armor. This puts the return electrode into the borehole fluid. Frcm figure 3.45a it can be seen that this will result in a return electrode which is normally extremely long. Thus, the grounding resistance of such an electrode is very low; so low as to be negligible. For example, 100 metres of 0.48 centimetre cable will have a grounding resistance of about 10 percent of the mud resistivity (in ohms).

Using the cable as a return electrode is an excellent idea, except when the sonde approaches the surface of the earth, the surface of the mud in the hole, or a thick, high resistivity zone (such as a thick, massive limestone or coal). In this

92 FIG.3.48. The Delaware Effect.

case, as the sonde approaches the surface the return electrode becomes shorter until its grounding resistance is no longer negligible nor constant. This is called the Delaware Effect and is present in other arrays as well as the Single Point. Its magnitude is proportional to the mud resistivity. Figure 3.48.

3.1.4.6. Normal resistivity

The problsns inherent in the Single Point system, the shallow investigation, sensitivity to hole size, and the difficulty of quantization have resulted in it not being used in the petroleum logging business, .except for special purposes. In its place, Normal resistivity systems of several different spacings or arrays have been employed. Although many of the problems of the Single Point are present in the Normals, the Normals have been extensively investigated and "departure" curves are available from any of the major logging contractors for many different arrays.

A Normal resisitivity device is one where the current electrode, A, is much closer to the measure electrode, M, than either is to the return electrodes, N and B. The measure return, N, and the current return, B, are at almost infinite distances away from the measure electrode, M, and the current electrode, A.

The resistivity measured by a normal device with a spacing AM in a homogeneous medium is VM - R = fiK = -Ji- 4TT AM AA vdiere Vj^ is the potential of the measure electrode and A^ is the current through the current electrode.

If the Normal device is in an infinite, homogeneous medium the resulting equipotential surfaces (which are always at right angles to the current) are spheres. These shapes are subject to the same variations and distortions as with the Single Point system. The departure curves for the Normal devices have been published and verified. The depth of investigation (a function of ffi) usually is great enough that the hole diameter and resistivity contrast must be large to be significant. Be- cause of the substantially larger volume of investigation of the usual Normal device, compared to the Single Point, the resolution of the Normal is proportionally less.

The A electrode and the M electrode are physically separated on the sonde body; thus, within the measurement volume. Because of this, a bed boundary (or any other change has little effect upon the signal value while the boundary is between the two electrodes. The resulting curve shapes are shown in figure 3.49. Notice that there are distortions at the bed boundary which make it extremely difficult to determine the location of the bed boundary accurately. In nonpetroleum logging (more than petroleum logging) this is a detriment to the use of the Normal systems because of the greater interest in thin beds. Whenever a Normal device is run, a Single Point probably should accompany it.

Since the depth of investigation can be controlled and determined easily, a conmon practice in petroleum logging is to run a combination of two or more Normal spacings. This allows one to assess the effects of the invasion of the borehole fluids into the permeable formation. This is illustrated in figure 3.50.

Besides the bed boundary effects_of the Normal device, the response in thin beds (thinner than AM ) makes its use difficult

95 FIG.3.50. Depth of investigation determined by spacing of electrodes.

in nonpetroleum logging. Figure 3.49 illustrates this problem. One can see that it is possible to mistake one thin bed for two beds. And, if the bed thickness is exactly equal to AM, there will be a very ambiguous deflection.

Departure Curves for the 16" Normal and 64" (40 centimetres and 162 centimetres) Normal are shown in figures 3.51 and 3.52.

3.1.4.7. Lateral resistivity

Lateral devices are devices where the A electrode is closer to the B electrode than it is to the M electrode. Ihis is shown

96 R64"/Rm

FIG.3.51. Departure curve for 64-inch Normal.

in figure 3.53. This will result in a deep penetration device. On the other hand, the response curves shown in figure 3.54 illustrate the extreme asymmetrical nature of the curve. For this reason, the Lateral or Inverse systems are seldom used.

The response of the Lateral (or any 4 electrode device) is

R = KI 4, x J_1 ! ' ^ IT BM ¥ BN Because of the deep depth of investigation of the Lateral device, it was widely used with early electric log suites for determining RT or Rq. Figure 3.52 illustrates how to pick the best value of resistivity from this curve.

97 Rl6'/Rm

200 300

FIG.3.52. Departure curve for 16-inch Normal.

FIG.3.53. Resistivity electrode arrays.

98 Thin Resistive Bed

FIG.3.54. Lateral response.

Because of the difficulty of using the Lateral curve, it is probably best to use it only as a last resort.

3.1.4.8. Focussed resistivity systems

The correct readings of the nonfocussed devices depend upon correcting for the departure of the measurement from linearity (or upon making the resolution so gross that interfering effects are negligible). It is also possible to stabilize the measurement electronically and retain resolution. This class of systems is gaining importance in the mineral business. It is the focussed system. These are also known as guard logs and Laterologs (Schlumberger trade mark).

99 FIG.3.55. Current flow from a single point electrode in a homogeneous medium. Electrode length is long compared with its diameter.

FIG.3.56. Long electrode broken to from 3-electrode guard log system.

An equipotential surface around a single point electrode is an ellipsoid of revolution. The length of the major axis of the ellipsoid depends upon the length of the electrode. If the length of the electrode is extended (figure 3.55) the flow of current near the center of the electrode is nearly parallel and that portion of the equipotential surface becomes very nearly a cylinder.

100 FIG.3.57. Block diagram of 7-electrode guard log. IP <£h

i— —o ^

FIG.3.58. Block diagram of 8-electrode guard log.

. The electrode can be separated into three pieces and a measurement made from the center piece. In a homogeneous medium the current will flow parallel into the formation and most of the measurement will be from a cylinder whose height is the length of the two end pieces or guard electrodes. Furthermore, if the guard electrodes are electronically kept at the same potential as the measure electrode, the size and shape of the measure cylinder will remain constant, even in conditions of nonhomogeneity.

101 Resistivity Ohnmetres 10 20 30 40 60 70 80 90

Single Point response True Shale resistivity Mud Limestone Guard Log response

FIG.3.59. Guard log response compared to the single point resistivity.

In a typical system an electronic circuit compares the potential of the measure electrode at constant current with the guard electrodes potential and forces the guard electrodes to the measure potential by adjusting the guard current. As a result, the geometry of such a system is stable and constant. Such a system is shown in figure 3.56. The response of signal to resistivity is linear over a wide range. The depth of investigation can be deep, and the resolution is symmetrical and fine.

There are several variations of this device in use.

The guard electrodes may be separated far from the measure electrode and monitor electrodes may be used to monitor the potential differences. This is a seven electrode system (Laterlog 7) which is popular in petroleum logging. It is used because of its deep penetration and because it can allow the recording of a Spontaneous Potential curve. This device is shown in figure 3.57.

The eight electrode system and the dual guard log systems have current returns which are not at an infinite distance (as they are in the seven electrode system). This results in a shallower investigation (figure 3.58).

The beam or current sheet of most of the guard log or focussed systems is thin enough and deeply penetrating enough that borehole effects are minimal or negligible. Therefore, these devices often measure, very nearly, the value of R-p. They need ,

102 &ED THICKNESS - (FT)

n o 70 70 m r>

n> 0 70 1 TO

BED THICKNESS - (FT)

FIG.3.60. Guard log correction for adjacent bed effects.

a minimum amount of correction. The use of these devices in petroleum work is to obtain a deeply penetrating, synmetrical response when the value of R,^ is much lower than Rt. Neither the nonfocussed devices nor the Induction Log is at its best in this situation. A high ratio of R^/Km is a frequent occurrence in uranium exploration.

103 I

FIG.3.61. A modern logging truck with a focussed resistivity system. (Courtesy of Century Geophysical Corp.)

In mineral logging the focussed devices are used to obtain a high, thin bed resolution, while retaining a symmetrical response and minimal borehole effects. Because they can be designed for a wide range of linear response, they make an excellent quantitative resistivity tool in fresh water sediments, coals, limestone, and hard rock environments. A typical response to a thin bed is shown in figure 3.59.

Figure 3.60 shows the "Adjacent Bed Effect" of one type focussed tool. When the resistivity contrast between a thin bed and the adjacent beds is great, there is danger of exceeding the design capability of the guard electrode control circuits. Then, a distortion is present which must be taken into account. Most uranium (mineral) type systems can satisfactorily handle a ratio of resistivity to that of the adjacent bed of 2000 or more. The upper portion of the logging probe (sonde) in figure 3.61 is a focussed resistivity device.

104 3.1.4.9. Indue tiori logs

The Induction logging systems were developed for petroleum logging in oil-based muds. They need no electrical contact with the formation and can be used effectively in oil, water, or gas-filled boreholes. They depend upon magnetic fields to induce electrical currents proportional to the formation conductivity in the surrounding formation.

If a magnetic field moves across a conductive medium,it will induce an electrical current in it. Therefore, if a coil of wire, with the solenoid axis parallel to the borehole, has an alternating current flowing in it, there will exist an alternating magnetic field which will move into and out of the formation surrounding the borehole. The magnetic field will induce a current in the formation which will flow around the borehole (around the solenoid). This circular current will set up a second magnetic field which will be 180° out of phase to the first field. A second solenoid on the axis of the borehole will have a current induced in it from the second magnetic field. With suitable circuitry,the effect of the first or primary field can be eliminated. See figure3.62. The primary magnetic field strength will be proportional to the primary current. The induced current will be proportional to the magnetic field strength of the primary field and to the conductivity of the formation. The secondary field strength, and thus the signal will be proportional to the induced current in the formation. Therefore, with a constant coil current, the signal is a function of the formation conductivity.

By properly placing additional primary (transmitter) coils and secondary (receiver) coils, and adjusting the polarity and ampere-turns, the primary magnetic field and the response of the tool can be shaped. The five and six-coil induction log tools will sense the conductivity of a torus of formation around the borehole. Thus, the induction log used in petroleum work is virtually independent of borehole and invaded zone characteristics.

The induction tool is important in oceanography. The resistivity and thus, the salinity of ocean waters may be measured quickly and accurately with a simple two-coil device (nonfocussed). Figure 3.63 shows a comparison between a two-coil and a six-coil focussed device. A two-coil laboratory device measures the conductivity of small fluid samples with lessened danger of contamination from electrode materials.

] 05 FIG.3.62. Principle of induction logs. (Courtesy of Schlumberger Well Services, Inc.)

106 0048 ohms E 21 ohms m /m m'/m ' TWO-COIL SONDES FOCUSSING SONDES TYPE 6F | .03 .» 0 50 I 9 B 20 tkmi m3/m

- EXAMPLE OF COMPUTED INDUCTION LOGS FOR DIFFERENT VALUES OF RATIO SPACING "I".

FIG.3.63. Comparison of two-coil and six-coil devices. (Courtesy of Schlumberger Well Services, Inc.)

107 Wenner © © © © Schlumberger

Schlumberger © © © ©

Double dipole © ©

t— a-i ©H- b —©1 h- c H

FIG. 3.64. Surface resistivity arrays.

3.1.4.10. Surface systems

The principles outlined in the Physical Description section of this chapter apply equally well to surface resistivity measurements. There are several differences, however. Firstly, the arrays used, and thus the sample volumes examined by surface methods are usually much larger than those in downhole methods. They are on the order of tens of metres to kilometres. Secondly, surface measurements examine a half volume rather than a full one. Thirdly, resolution is poorer with surface methods because of the larger volumes. There are other differences, too, but in general, the similarities are greater than the differences.

Since the arrays used in surface resistivities are large, four-pole soundings or surveys are usually made: The Wenner, the Schlumberger, and the polar-dipole arrays. This is because the return electrodes, to be essentially at an infinite distance (as in downhole arrays) would have to be many kilometres away. The types of cormion arrays are pictured in figure 3.64.

The Wenner and Schlumberger arrays are similar. They both use four colinear electrodes. The outer pair are the current electrodes (A and B) and the inner pair (M and N) are the measure or voltage electrodes. With the Wenner array all spacings, AM, MN, NB are equal. With the Schlumberger array the MN

108 spacing is much smaller than the other spacings. When a resistivity sounding is made, the Wenner array is expanded about the midpoint (to increase the depth of investigation) keeping the spacings equal. Only the two outer or current electrodes are moved outward with the Schlumberger array.

With the polar dipole system, the current array (A and B electrodes)and the measure array (M and N electrodes) spacings are both small compared to the distance between the current and measure arrays.

As any of the arrays are expanded a deeper sample is examined (a sounding is made) and the sample becomes larger.

The array data (current, voltage, and spacings) are converted to apparent resistivity with the assumption that the sampled volume is uniform. The relationship to calculate apparent resistivity is the same as for the downhole arrays:

R = K(V/A) = Kft

With surface arrays, however, the factor 2TT is used indicating a half volume, limited by the earth's surface. Thus the value of K is

AM BM AN BN

In field practice the readings of the surface measurement are converted to an apparent resistivity plotted as a function of apparent measurement depth (see figure 3.65). Known features which will interfere with the readings are then identified. These features are metal fence lines, power lines, grounding electrodes, galvanic electrodes, pipe lines, cased wells and so on.

The remaining resistivities should be correlated with any downhole resistivity measurements. (This, incidently, is a very valuable step which is seldom planned for and not often done.) One of the major interpretation problems of surface resistivity is the proper identification of the subsequent layers or vertical anisotropy. These can be quickly and accurately identified and evaluated with a borehole resistivity curve. Along with this correlation, a distribution of resistivities which will fit known field data is picked from models or from a borehole resistivity log to match the calculated apparent resistivities.

Surface methods are best suited to covering large areas in a short time, but have a lower resolution than with borehole

109 resistivity measurements. The methods are particularly valuable when the terrain or land owner problems make access by drilling equipment difficult. This is especially true when they are combined with well-chosen borehole methods. The combination is ideal to map subsurface channeling and many other multilayer- ed environments, rather than only using it to find specific mineral horizons. Surface resistivity methods combined with downhole measurements are valuable for engineering uses because of the complete picture they furnish. Surface resistivity surveys are often combined with induced polarization measurements. Only minor circuit modifications are required to allow one system to measure both parameters. The Induced Polarization measurements furnish further data about mineralization and redox interface trends.

3.1.4.11. Calibrations

Calibrations of resistivity systems fall into two categories; the primary calibration used to verify or determine the array geometry (conversion factors and departures) and the secondary or field calibrations.

110 FIG.3.66. Calibration of normal resistivity.

The secondary calibrations are usually quite simple and often consist of only a precise resistor or network to verify or to set the current value. The value of the calibrating resistance is usually calculated from the resistivity expression using the conversion or K factor:

Q = R/K

The resistor is of a low value compared to the measure circuit impedance. Therefore, it is usually permanently placed in the measure circuit. The current circuit is connected across the calibration resistor and the current required for a full scale deflection is noted. This current is then set when the circuit is returned to the normal or logging position. This is shown in figure3.66. With modern, stable circuitry and good current regulators, this step is probably superfluous. Almost any change to the circuit will be negligible.

The primary calibration is quite different, however. It is poorly understood by most users and designers. This step is long and complex and involves many stages. A primary calibration will accomplish these things:

1. It will determine if the calculated K factor is correct. This verifies the geometry of the system. This is usually done in a homogeneous medium of known resistivity. It is most easily done in a large tank of water of known salinity and temperature. The tank must be apparently infinite in size. These systems may be scale modelled effectively.

2. It will determine the depth of investigation. This, too, verifies the calculated or assumed geometry. It is accomplished in the same type tank as the one above. High resistivity boundaries (plastic sheets) can be inserted at known distances and the influence on the measurement can be noted.

Ill 3. Primary calibrations will verify the geometry. This is further verification of the first"two steps and can be done in the same tank. A voltage probe returned to an infinite distance or a two-electrode gradient probe can be used to determine the shape of the equipotential surfaces in a homogeneous medium.

4. The range of measurement must be determined. The same tank again may be used for the range measurements. Known amounts of sodium chloride (verified with a resistivity tester) should be added to the tank water. The plotted results will indicate the measurement range of the system.

5. The calibration will determine the departure from the ideal geometry. This is a difficult phase. Known resistivities and sizes of borehole, formation and adjacent formations in many combinations must be used and the response of the system determined. This may be done in the tank with artificial formations, with a resistance network, or/and with computer modeling. The resulting departures from theoretical or homogeneous situations must be known for a realistic evaluation of the formation character. These departures will occur at some point with any system.

As with the other types of logging systems, the calibration phase is at least as lengthy and exacting as the design of the system itself. It is also a step or a phase which is usually neglected.

The calibration of the resistivity (and other) systems is a phase which should be better documented and publicized by logging equipment manufacturers, designers, and operators. The usual assumption by the user is that the operator or contractor has made the measurements carefully. This may or may not be true.

3.1.5. Spontaneous potentials

3.1.5.1. Introduc tion

Spontaneous Potentials (SP) are the natural potentials which occur in the earth. They occur because of imbalances of ion species caused by four mechanisms. These mechanisms are diffusion, adsorption, redox reactions, and electrofiltration action.

The SP curve is used in petroleum logging to differentiate between porous, permeable formations, and shales. The electrochemical component (the deflection opposite a sand) is used to

112 calculate the value of the formation water resistivity, Rw (to facilitate resistivity porosity determinations). The SP curve can show oil/gas/water contacts.

In nonpetroleum logging the SP curve is used as a lithology log to distinguish between sand, shales, and shaly sands.This employs the electrochemical component. It is used to calculate Rw for legal purposes. It is also possible to use the redox component in a trend mapping technique. The electrofiltration potential component is largely an interfering potential, but it can be used to spot fluid movement into or out of the borehole.

Other forms of spontaneous potential measurements are the surface potential measurements (redox), the Selective SP, the Static SP, differential SP, pH measurements, and redox logging.

The SP in a borehole can only be measured in an open, uncased hole. Both metallic and nonmetallic casing prevent the measurement of a borehole SP.

3.1.5.2. Physical description

Spontaneous potentials arise from an unbalanced distribution of ions caused by one or more mechanisms. The primary mechanisms appear to be diffusion, adsorption, oxidation-reduction and electrofiltration. Any or all of these processes can act to unbalance the ion concentrations and result in potential differences and resulting current flows. There are other effects which may contribute, but mostly they are interfering or noise effects. Some of these will be covered later in the section.

When a borehole traverses the formations, the combinations of the potentials in the formations will cause electrical currents to flow in the borehole mud. The SP electrode assumes a voltage due to the potential drop in the mud column, caused by this flow of current. This voltage is measured with respect to a distant, constant, identical electrode by a high impedance voltmeter. This illustrated in figures 3.67 and 3.68.

In surface methods, the potentials themselves or the potential drop through the formation may be detected.

The mechanisms giving rise to the spontaneous potentials operate to separate anions from cations, deplete or enrich solutions of anions or cations, deplete or enrich solid portions of the formation with one of the ion types, or change concentrations which may give rise to any of these situations.

113 FIG.3.67. Equivalent circuit of the electrochemical SP.

Reference Electrode

Measure Electrode

FIG.3.68. Spontaneous potential system. FIG.3.69. Electrochemical component of SP.

The potential differences set up within the formations can usually be considered to be across an interface between the invading mud filtrate and the formation water of a sand and relative to the adjacent shale. These potential differences result from various combinations of the four major mechanisms. The resulting currents flow through the mud column causing a potential drop along the mud column, as shown in figure 3.69. The sum of the potentials is ESSp:

= I(r + r + r ) ssp sh m'

114 and the potential change, as the electrode moves from a sand to a shale is the electrode potential with respect to the reference electrode. E = I r sp m where I is the current in the mud column and rs, rsh, and rm are the resistances of the sand, shale, and mud, respectively.

These equations can be combined:

E = E (rs + ^^ + ^^ ssp sp rm

If the borehole could be plugged so that no current could flow in it, then the value of rm would approach infinity:

Limit Esp = ESSp

m This principle was made use of in an early Static SP method and the Selective SP method.

If the beds are thick and the resistances of the sand and shale are low, ESp and ESSp will nearly equal, because rs and rsh will be small compared to rm. But, in resistive beds the values of ESp and ESSp will be different, because rs and rsh will no longer be negligible.

The position of the bed boundary with relation to the slope of the SP curve will depend upon the ratio rs/rsh. The bed boundary will always coincide with the inflection point of the curve. Several examples of this effect are shown in figures 3.70 and 3.71.

Since the SP system measures the ohmic drop in the mud column, if the formation contributes nothing to the potential (as with a massive, impermeable limestone), then the SP reading within that formation will be merely the potential gradient between the two nearest contributing formations. The SP curve will usually be a straight line through the limestone. This is illustrated in figures 3.72 and 3.73.

3.1.5.3. Ins trumentation

In principle, the instrumentation of the SP logging system is quite simple. It should consist of a measure and a reference electrode which are identical, chemically inert and stable, a reference position which is representative of the formation conditions, connecting wires, and a high impedance voltmeter.

115 Inflection Inflection Points Shale Points Shale

R R Sand Rs = Rsh Sand s » sh low resistivity lew resistivif

Shale Shale

SSP "SSP

Shale Shale

Sand Sand s sn

Shale Shale

Rm ff >Rw

FIG. 3. 70. Relationship of bed boundary and SP curve.

116 Inflection Shale Inflection Shale Point Point

Limestone Rs " Rsh " ^n Xs^sh^

Inflection Shale Point \ Rs > Rsh « «tn Sand

Inflection Shale Inflection , Shale Poin Point

"Is sh^ Kl Limestone Rs < Rgh >> Sand

SP bed boundary effects

FIG.3.71. SP bed boundary effects. FIG. 3. 72. SP curve in nonpermeable limestone.

Sale • St als

pe nn^al le iraine /

Shal: «*»•«»* sam<>Aws - •st mwrutaM v. saweBasasameaasJii.'. BHTOSSSiBaiBiMB-aw.k'' ^^ msmn m^m ct „ Shal;

FIG.3.73. SP curve in various limestones. The SP electrodes mast have equal potentials. These should "buck" each other out, resulting in a measurement of only the changes of potential of the measure electrode due to formation potentials. The SP potential measured by a practical system is

r m E = E -7 ^ ^ c- +(E ~ E sp ssp (rg + rsh + rm) e ref where Ee is the measure electrode potential and Eref is the reference electrode potential.

High impedance measurements must be made for two reasons. First, the environment must be disturbed as little as possible. Some current must be drawn from any system on which we make a measurement of potential. This results in a lowering of voltage depending upon the amount of current drawn and the impedance of the source:

E = E - I r measured source meter source A high impedance measuring system draws negligible current, making the second term nearly zero. Secondly, the high measurement impedance makes any other circuit impedance (such as electrode impedance and cable resistance) negligible. Thus, these other circuit impedances can change greatly without appreciably affecting the measurement.

3.1.5.4. Electrochemical potentials

The electrochemical component of the spontaneous potentials is the only portion used in petroleum logging. It is the por~ tion of the SP curve used to determine lithology and to calculate the value of Rw.

The electrochemical or diffusion-absorption potential is evaluated as the deflection opposite a permeable formation, compared to the potential of the adjacent shale potential. It is the sum of the voltages or potential differences developed across the liquid junction between the formation water in a sand and the mud filtrate in the invaded zone, the interface between the sand and shale, and the shale-mud interface. It depends upon the following factors:

1. The chemical and mineralogic composition of the rock com- posing the sand. These factors will affect the apparent salinity of the sand and may vary in both amplitude and polarity.

119 <°CI (10) (38) (66) (93) (121) (149)

STATIC SP - MILLIVOLTS

FIG.3.74. Equivalent formation water resistivity from SSP. 2. The composition and concentration of salts in solution in the electrolyte filling the pore spaces. This effect decreases with increased concentration and may change polarity with multivalent ions.

3. The degree of saturation of the rock with electrolyte.

4. The density of the rock.

5. The grain size of the rock. Increases of the last three will result in departures from a simple situation.

6. The cation exchange capacity (CEC) of the clays.

7. The salinity amount and type of the mud.

3.1.5.5. Diffusion-adsorption potentials (electrochemical potentials)

Diffusion of ions across concentration interfaces within a sand or at the contact of the formation water and mud filtrate will cause ion imbalances because of different ion mobilities. This is the diffusion effect.

Many materials, such as clay and extremely fine quartz (silts and muds), have electrical properties (because of their large surface areas) which will bind ions of one charge (usually the negative charge or anion) to the material surface. This results in a depletion of the solution of that charge and an excess of that charge associated with the solid material. This process is the Nemst diffusion layer process and the resulting potential is the shale (or Nemst) potential. It is most pre- valent in the clays of shales, but can occur in fine grain quartz materials in silts and mudstones, and is the adsorption effect.

The adsorption function can be combined with the diffusion function, and usually is, since these two effects occur together, The combined potential, the diffusion potential, plus the adsorption potential is the SP deflection opposite a sand, E^a- Thus, the change of potential, E^ for dilute solutions of NaCl

^A = °DA 1o§ (W vAiere I^f and R^ are the effective resistivities of the mud filtrate and the connate water, respectively. The effective resistivity will depend upon the ion types present, since this equation assumes only Na+ and Cl~ ion types. Figure 3.74 illustrates the electrochemical or diffusion-adsorption potential phenomenon.

121 TABLE 3.3. CONVERSION FACTORS FOR SPECIFIC ION TYPES TO NaCl AND NaCl TO SPECIFIC ION TYPES

Ions Factor, Factor, Conversion Conversion to NaCl from NaCl

+ Na 1.00 1.00

+ K 1.00 1.00

++ Ca 0.95 1.05

++ Mg 2.00 0.50 so4~ 0.50 2.00 cr 1.00 1.00 HC0 ~ 0.27 3 3.70 co3- 1.26 0.79 When ion types are known, use this table to convert to R or R J r ' TaT Y

The notation of effective resistivity is made necessary because the assumption is made in all of the charts and relationships that the solution is dilute and only NaCl in water.

The effective resistivit+ y is what the resistivity would be if the ion types were Na and Cl~. Obviously, this is not always true. Therefore, a correction must be made. Table 3.3 shows how the various ion types compare to a solution of NaCl. Fig- ure 3.75 shows the relationship between water resistivity and effective water resistivity.

The empiric family of curves in figure 3.76 arises because of the temperature effect upon the mobilities of the ion species. Below 0.1 ohmmetre a spread is due to decreased relative activity due to concentration. Above 0.1 ohmnetre++ , th++ e mobilities of the calcium and magnesium ions, Ca and Mg , dominate and cause a spread due to temperature.

After the effective value of Rw has been calculated (Rwe)> it must be converted to the proper value for the particular mixture of ion types. Figure 3.76 is representative of the

122 FIG.3.75. Rw versus (Rw)e for solutions of several salts. ground waters found in the Gulf Coast, Midcontinent, and Rocky Mountain areas. If the ion type is known, figure 3.75 may be used.

The mobility of an ion is a function of the absolute temperature. Therefore, one must correct all measurements to the same temperature. The chart in figure 3.77 shows the variation of a sodium chloride solution resistivity as a function of temperature at constant concentrations.

n a The value of I<£>A i- clean sand and a shale, for NaCl solutions is -70.7 millivolts at 75°F. Note that the value of K is negative. Thus, if the resistivity of the mud filtrate, Rrof,is higher than Rwe, the value of the logarithm function will be positive. Thus EQA opposite a sand will be negative. If, on the other hand, the mud filtrate is more saline than the connate water, the logarithm function will be that of a ratio less than one and will be negative. This, multiplied by the negative value K[)a indicates a positive SP deflection opposite a sand

123 R - OHMS m2/m

2 RWe - OHMS m /m

FIG.3. 76. Rw versus (Rw)e at various temperatures.

when saline mud. is used. Simiarly no SP deflection indicates Rmf = Rwg. These last two situations commonly occur in nonpetroleum logging where salinities are low and where formation waters are often surface runoff. At 100 parts per million total dissolved solids (which is not uncommon for Rocky Mountain surface water) the amount of dissolved salt (NaCl) is only 1 gram in 10 kilolitres of water. Thus, it takes only a small additional amount of salt to change the SP curve dramatically. Literally, positive, negative, and zero SP deflections can be encountered in one hole.

124 RESISTIVITY - OHMS m2/m

RESISTIVITY - OHMS m2/m

FIG.3.77. Variation of a sodium chloride solution as a function of temperature.

The value of may be calculated from the SP curve in the following manner (refer to figure 3.74).

1. Draw a straight line from shale deflection to shale deflection across the sand deflection on the SP log (a shale line).

2. Measure the distance, in millivolts, from the shale line to the deflection in the "cleanest" portion of the sand.

3. Solve the equation for R f/R , or use figure 3.74. m we 4. Find the value of

a. If a mud tester and mud press are available, measure R and correct for formation temperature (figure 3.77). f

125 2 Rmf or Rmc ~ OHMS m /m .01 .02 .08 .1 .2 .5 1 2 5 10

7 - - V / / / / / r / // / / r / / * A— / / y / / / / f / / / A / V r/ A / / / / / / / / / V / .01 .02 .OS .1 .2 .6 12 S 10 Rmf or Rmc - OHMS in2/m

FIG.3. 78. Rm -V?mf Rmc relationships (empirical chart).

b. If no press is available, find the average value of Rmf from the value, using figure 3.78. Correct for formation temperature.

5. Divide R^ by the ratio found in step 3 to obtain R . wg 6. Return to figure 3.77 to obtain a correction value of R at formation temperature.

7. The value of the total dissolved solids in terms of eNaCl can be obtained from figure 3.77.

126 Static S.P. Clean sand

Rt „ _ p Rt „ PSP

FIG.3. 79. Plot of the reduction factor, a, for shaly sands.

3.1.5.6. Shaly sands

When a sand contains a shale the SP will be reduced. The amount of reduction will be a function of the amount of shale in the sand and upon the cation exchange capacity (CEC) of the shale. As the amount of shale is increased the potential will approach that of a pure shale.

The value of a, the reduction factor is _ Pseudo-static SP in shaly sand Static SP in clean sand. The value of a may be obtained from figure 3.79 or may be calculated:

a lo T? = §10 • log^u) where u = (R^/R^) and q = (R,./R ) • (p/l~p) sh where p is the proportion of shale, p may be determined from the ganma-ray curve or from a crossplot. With the gamma-ray

127 \

FIG.3.80. Redox potentials of the SP curve.

curve the shale content, V ^ (by volume of total formation), may be estimated by

Y s V ~ Sh Y' s ~ Y's ,h are where y j-) and Ys the gamnna-ray readings, in any units, of s the closest pure shale and clean sand, respectively and Y is the ganma-ray reading of the point in question.

By combining relationships, the value of ^ in a shaly sand may be found from

^wa _ /t •> i iogs- - a--) iogv

The value of R^ is the apparent water resistivity calculated shaly sand deflection, uncorrected.

3.1.5.7. Redox potentials, figure 3.80

The redox component of the SP curve is seldom used in the petroleum industry. Occasionally, it is used to indicate the

128 presence of "pyrites," but little beyond this. In mineral logging, many important types of mineral depositions are accompanied by or caused by reduction-oxidation (redox) reactions. These include some depositional mechanisms involving uranium, copper, silver, iron, and petroleum. Many compounds change solubility when undergoing a redox change. Some compounds change electrical and magnetic character with redox changes.

A redox reaction is one where an ion, such as a metal ion changes valence. It is said to be oxidized if it loses electrons and becomes more positive. It is said to be reduced if it gains electrons and becomes more negative. Oxidation of a material will always be accompanied by something being reduced. The reactions can be abbreviated in equation form with the so- called half-reaction. Thus, iron is oxidized from the ferrous state to the ferric state and the reaction is shown thus

+++ Fe++^=±Fe + e"

The double arrow indicates the reaction is reversible. e~ is the resulting electron. It represents the release of energy and it can do work. There are, of course, other parts to the reaction, but they are not needed to illustrate the process. It must particularly be remembered that something else is being reduced by gaining the electron. The reduction process requires energy and work must be done to accomplish it.

These redox/electrical interactions occur and are made use of in many common places. Ferric iron (hematite) is reduced to metallic iron in a reduction furnace. Carbon supplies electrons to the ferric iron and is oxidized in the process. A lead-acid battery starts an automobile by oxidizing plumbous oxide to plumbic oxide. Chromium is plated on an automobile bumper by reducing chromic salts with electrons (electrical current).

In the earth, redox reactions may take place through several mechanisms. Meteoric water, containing dissolved oxygen,can percolate into a permeable outcrop. Any uranium compound in dilute form will be oxidized by the water-oxygen solution. In general, uranium compounds are much more soluble when in an oxidized state than when in a reduced state. The oxidized uranium compounds will remain dissolved until they encounter a reducing medium, such as a reduced sandstone. The uranium compounds will decrease in solubility and precipitate out in that zone. The sandstone will be altered by oxidation in the process. More-oxidized solution of uranium will dissolve the compounds at the back part of the deposit and carry them forward. They will shortly encounter more reducing material and precipitate. The process occurs repeatedly or continuously, concentrating the mineral as the cell moves slowly downstream.

129 Occasionally a highly reducing zone is reached, such as a sulfide zone, a zone containing organic trash, humates, organic or sulfurous gases, and probably, reducing bacteria. Then the uranium compounds are concentrated in that zone. A change of permeability often appears to be associated with this zone.

This process is only one of many variations of the redox process causing mineral deposition, leaching, and often concentration. Similar mechanisms occur near petroleum deposits, resulting in halos of minerals, many of which are radioactive. Most uranium deposits seem to require redox reactions to hold the mineral in place, even though these may not be roll-front type cells.

Surface measurements of spontaneous potentials for the location of sulfide ore bodies have been known since before the beginning of the twentieth century. Practice was to use two stable electrodes, such as copper in saturated copper chloride solution in porous pots connected through a sensitive voltage indicator (a potentiometer). A sulfide ore body lying across the redox interface of the surface oxidation and the reduced formation below will have strong currents associated with it. The sulfide will oxidize at the top and the highly conductive sulfide will concentrate the resulting current flow. While there is disagreement as to the exact process, the mechanism appears to be a redox reaction. As a result, voltage drops (ohmic drops) can be read on the surface.

Potential differences are associated with redox deposits because of imbalances of ion species. These potential differences appear on the SP log as base line shifts which can be compared and mapped. They correlate well with the redox interface trends, as determined from hematite staining in the drill cuttings and cores.

A base line shift to the negative or left indicates the SP electrode is approaching a more oxidized zone. A shift to the positive or right shows the zone is less oxidized or more reduced. The downhole potential is compared (with the SP system) to the potential of the reference electrode at the surface (usually in a highly oxidized environment).

The contribution of the redox component is the potential, Vox: V = (RT/ZF) In (k C /C ) ox ox rd R, T, Z, and F are the gas constant, absolute temperature, valence, and Faraday's number, respectively (as with the electrochemical component). k is the reaction rate constant and CqX

130 and Cyj are the concentrations of the more and less oxidized solutions, respectively.

Much more work needs to be done on this type of reaction. It is quite possible that many of the SP perturbations which have been blamed on other things actually originate with the redox potentials. This certainly appears to be true of the base line drift of the SP curve as it approaches the oxidized surface formations. Examples of base line drifts which have been correlated with redox situations have been observed from a few millivolts to more than 90 millivolts. The mapping technique does appear to be a valid and useful one, even though the explanation is incomplete.

Redox reactions will also appear in coal and in sulfides, caused by the dissolved oxygen in the drilling mud. One patent has been issued describing a method for deliberately causing this type oxidation by introducing highly oxidizing material to the mud and then logging with the SP.

3.1.5.8. Electrofiltration potentials

Electrofiltration, electrokinetic, Helmholz, or streaming potentials are caused by the mechanical flow of a solutio1 n across a restricting or semipermeable barrier. Helmholz Law states this potential to be V^:

VH " "I!" 4P where £ is the dielectric constant of the liquid, R is the electrical resistivity, £ is the zeta potential (the potential between the fixed and mobile layers of fluid), y is the viscosity and AP is the pressure drop (differential pressure) (see figure3.81). Wyllie found the electrokinetic potential or streaming potential V , to be s

vs = kpy where P is the differential pressure and k and y are constants depending upon the system and temperature. This agrees with the Helmholz relation for special cases.

It was believed at first that this component was the major contributor to the SP curve. It has since been demonstrated that the streaming potential is usually negligible in oil field logging. It may not be negligible in nonpetroleum logging. In| this case we log in freshly drilled holes filled with rather poor mud. Invasion rates are probably high, resistivitie1 s are high and viscosities are low. Referring to Helmholz Law

131 V-TJT (r.-r)

E = P tH 4 TO K

FIG.3.81. Helmholz-Law.

equation, these make x, and R high and p small, resulting in a possible high value of V^.

The streaming or electrofiltration potentials will cause erroneous results in the calculation or R . If the SP curve is used quantitatively, the filtration potentiaw l must be considered. If possible, the value of the mud resistivity must be kept low, the viscosity high and the differential pressure low.

3.1.5.9. Electrode and probe effects

So far, not much has been said about the potentials which can arise from the electrode materials and the probe body, and their interactions with the fluids in the borehole and formation. It has been assumed, generally, that the electrode and probe are inactive electrically. This is not true.

Most electrode metals will react with the surrounding solutions. While they are reacting, metal ions are going into solution, transporting charges, leaving the electrode with a negative potential. As the surrounding solution gains more and more metal ions, the rate of reaction decreases. When the atmosphere around the electrode becomes saturated, the reaction stops. Thus, when a piece of metal is put into a water solution of NaCl (for example), if the potential of the metal is measured, with respect to a stable reference electrode, the potential will change. It will start high (positive) and drop rapidly or slowly depending upon the metal.

132 Some metals will not react. There are two reasons for this. The noble metals (gold, platinum, etc.) are chemically, relatively inert and make good electrodes. Other metals, such as stainless steel, aluminum, tantalum form inert, electrically insulating coatings which protect the surface. Some coatings, such as copper alloys and compounds, have undesirable electrical properties. Some metals react rapidly and continuously. Sodium is an extreme example. It reacts violently. Some metals change their mode of reaction depending upon their thermodynamic state. Nickel alloys fall into this category. Some metals are mechanically not suitable. Mercury is an example. Thus, if an electrode is to be made of a bare metal, it must be chosen carefully. It must have one major thermodynamic equilibrium state, it must have reactions which are electrically reversible, it must have a limited reactivity and solubility, it must be mechanically competent, and it must not be costly. There are few metals or alloys which come close to these criteria.

Lead is close to being an ideal material. It is cheap, easily worked, mechanically reasonable. It has two states of thermodynamic equilibrium (but one is not likely to change to the other under usual conditions), the solubilities of its reaction products are usually low, its reaction products are bi- symmetrical, and they will cling to its surface. Therefore lead is used with reasonable success as an SP electrode material.

Silver is another possibility for an electrode material, but is not often used. It has more desirable properties than lead, it is stronger mechanically and more stable thermodynamically. Other electrodes can be designed which are stable and should be investigated. These are the impolarizable or saturated electrodes .

A lead SP electrode is presumed to be identical electrically and chemically to the surface reference electrode. In general, this is approximately true. The reference electrode should be in the mudpit, which contains the same fluid as the borehole. Both electrodes are made of the same material.

As soon as the electrode (measure or reference) is put into the mud, reactions begin to put lead cations into solution around the electrode. This environment saturates very quickly (several seconds) because of the low solubility of lead compounds and, since the mud is viscous, remains saturated. The reaction or solution stops and the electrode stabilizes electrically. Therefore, the two electrodes very, quickly reach approximately the same potential. This is shown in figure 3.82.

133 CI-. ci- ci-

ci- M* M*M+C|- CI-CIM*

STATIONARY ELECTRODE MOVING ELECTRODE OF METAL M IN OF METAL M IN A CHLORIDE- RICH A CHLORIDE- RICH ENVIRONMENT ENVIRONMENT

FIG.3.82. Stationary and moving electrodes in a chloride-rich environment.

As soon as the measure electrode is moved however, it leaves behind the reaction products, the saturated atmosphere, and reaction rate increases. The potential of the electrode becomes more negative due to motion through the borehole mud. If the electrode is moved at a constant rate the reaction attains a constant rate and the electrode porential becomes pseudostable, but at a different potential than the reference electrode. This effect is typically on the order of -5 millivolts.

If the electrode bumps the wall of the hole, the clinging reaction products are scraped off, leaving a bare reactive metal surface. This will result in a sharp, momentary voltage change. This is a common problem and can ruin a log completely. It may be several tens of millivolts.

Alloy solder is often used to repair an electrode. It will cause the electrode to exhibit a different potential, due to the presence of other metals. When the probe rotates, periodic changes will be seen on the SP curve.

A serious effect is bimetallism. If part or all of the probe insulation is removed (i.e. when bare metal is exposed at a tool shoulder through wear), the steel (or brass, or aluminum) housing will have a much different potential than the zinc of

134 FIG.3.83. SP curve mirroring the resisitivity curve.

the cable armor. This will cause large currents (several amperes) to flow through the mud and the metal of the probe. A d.c. lateral curve will then be superimposed upon the SP curve. This lateral may be 100 or more millivolts in amplitude and will cause the SP curve to be nearly a mirror image of the resistivity curve (see figure 3.83).

Since the SP which is measured is the ohmic drop in the mud column, any bare housing section will "by-pass" the mud column. It will result in a reduction and distortion of the SP curve.

3.1.6. Calipers

3.1.6.1. Introduc tion

Calipers or borehole calipers are used to determine the condition of the borehole, provide correction for other measurements, and determine borehole volume for cementing purposes.

There are several types of calipers available. Single arm calipers are often run with other services. Multiple, ganged-arm devices are most common. Multiple, independent- arm devices are gaining favor because of the better quality information. Acoustic devices, similar to a sonar, are occasionally used and may soon be found incorporated in other tools. .

3.1.6.2. Method

In general, the measurement of borehole diameter is made with one or more arms against the wall of the borehole. If one arm is used the tool body is assumed to be against the wall

135 opposite the arm. The angle between the arm and tool body is measured. This angle, 0, is a function of the borehole diameter : d, = d + L sin 0 h s where d is the tool diameter and L is the length of the caliper arm.

There are several methods of measuring this function. The most cotimon method is to measure the displacement of the arm through a rack and pinion system which also transmits a spring force to the arm to hold it out. This mechanism is further used to extend and retract the arm for use or storage. The position of the rack is sensed with a linear or rotary potentiometer which will give an electrical signal which is a function of sin 0. The signal can be made linear with either circuit parameters, computer programs, or log scales. An analog system usually has a nonlinear scale. This type system is often found on sidewall tools, such as the formation density tool.

Three arm caliper systems are very common. A three point measurement, with arms at 120 degrees, is quite stable in a borehole. A three arm caliper has a tendency to measure the minor diameter of a noncircular borehole.

Systems with three or more independent or semi-independ- ' ent arms have been designed and used in petroleum logging. These give much better values for hole diameter and shape. Four aim calipers are common with opposite arms ganged. Birdwell runs a 12 arm caliper.

3.1.6.3. Uses

Single arm borehole calipers are of restricted use, so they are designed for special purposes. One usually sees them on tools designed primarily for other purposes, such as the formation density tool. Here, it indicates the quality of the density curve by giving an idea of the character of the wall of the hole; if it is smooth or rough, to-gauge or caved, and the mudcake thickness. For these purposes it is quite good. There is always an uncertainty whether the single arm caliper is measuring "the major or minor axis of the borehole, or in between, or if the hole is circular or oval or ellipsoid in cross-section. Too, there is always an uncertainty about whether the measured feature is present on the opposite wall where the other measurement is being made. Therefore, these should never be used to calculate borehole volume. The problem is illustrated in Figure 3.84.

136 FIG.3.84. Possible caliper measurement situations.

The three-ganged arm caliper is someutiat better than the preceding ones, because it tends to give a more realistic measurement. There is less uncertainty as to the position of the tool and the shape of the hole.

If the caliper has four arms, the assumption can be made that they are measuring the major and minor axis. This is not always true but the error in this assumption is small. If the caliper has more arms, the actual shape of the cross-section can be taken as the major and minor axis, respectively. In this case, borehole volume is, V^:

V^ = TT a b h where a is the maximum reading, b the minimum and h is the depth interval. If the borehole is caved asynmetrically the volume, V^ can be approximated by assigning a regular geometric shape to the cave, in addition to the regular shape of the borehole. Thus V = V + V, , . hu cave borehole

If the borehole is to be cased and the casing cemented, there are several ways the amount of cement can be determined. If a computer is not available, the best method is probably to solve the problem completely on a short length basis. If V is the volume of cement

V Z (A 1 c = h " V

137 where i is a convenient depth interval, Aft and Ap are the cross- sectional areas of the borehole and the outside of the casing respectively. Computer programs, of course,have much more flexibility. If only an analog trace is available, a planimeter may be used to determine the average borehole diameter. Then the volume of the hole is

2 V, = n r Ad h ave and the cement volume

2 2 V = n (r - r ) Ad c ave p

where r and r are the averag& e radius of the borehole and the ave p outside of the casing, respectively.

3.1.7. Deviation

3.1.7.1. Introduction Borehole deviation measurements fall generally into two categories, those that use the earth's magnetic field and those that use an inertial reference. The earth's magnetic field is stable enough and well enough known that it can be measured to reference the borehole direction to a geographical direction. It can be measured in a horizontal plane or three direction vectors downhole. Inertial measuranent systems are much more complex than magnetic systems. Their use is restricted to holes which are steel cased.

Borehole deviation measurements are made independently or in combination with other systems.

There are several types of deviation tools available. The surface readout, continuous systems appear to be superior in every way to single shot and photographic methods.

Deviation measurements are used to correct other measurements, define deposit locations, satisfy legal and contractual requirements, and spot problans. 3.1.7.2. Station type deviation systems

The simplest directional tools are the so called "one-shot" tools. The use of these is mainly restricted to drill pipe to satisfy contractual obligations. They may be a photographic type using a magnetic or an inertial sensor system, or they may employ an acid etch of a container.

138 FIG.3.85. Typical multishot image.

The multishot photographic systems were used for many years in the petroleum and mineral business. There are several minor modifications of the multishot system. However, except for the less-used inertial systems they all photograph a magnetic compass through a tilt indicator (a plumb or a ball in a watch- glass or a bubble type "spirit" level) so that the two images are on one diagram showing degree and direction of tilt of the tool. Figure 3.85 shows a typical image. Photographs are made at depth intervals or "stations" in the borehole which are chosen and recorded manually by the operator. At the station the probe is controlled either by a timer (with the downhole instrument having self-contained batteries) or by electrical control and power from the surface. Figure 3.86 is a diagram of this type tool. The operator will insert a pattern of de- liberate bad or inconclusive shots in order that he may identify

139 I Cable Elastic Susp. (\ Clock*

^ Camera o Plumb ^ _ - Compass GlQ Lights I'M' Batteries

Space

Weight

* not used in surface controlled equipment

FIG.3.86. Photographic deviation tool.

Typical Continuous Deviation Tool Circuitry

2-Axis, 1 G o Accelerometers o

3 Plane 9 Fluxgate

Fluxgate and Accelerometers Fixed with Respect to Tool

FIG.3.87. Fluxgate and accelerometers fixed with respect to tool.

140 the position of any measurement in his notes. The film is developed after the survey has been run and the patterns are matched to the manual notations. The information is read and calculated later. Some modern multishot tools may be small and compact and incorporate a high degree of damping. Thus, they may be run in combination with other devices by merely pausing for two or three seconds during logging to take a station reading, or they may be operated on the way downhole before the other curves are recorded.

These devices have two serious drawbacks. Firstly, the film must be developed and the calculations made long after the tool comes out of the hole. Secondly, the limited available time in the borehole and the limited film capacity require long intervals between stations which can cause major inaccuracies.

3.1.7.3. Continuous type

Surface readout tools were developed to circumvent the drawbacks of the photographic tools. There are two major types of these tools in use.

One type, the Humphrey system uses a 2-plane fluxgate mag- netometer to read the 2 horizontal direction vectors of the earth's magnetic field. A resultant signal from this operates a servo system to continuously orient a cage to North - South. The cage contains two plumbs which indicate the N-S component of tool tilt and the E-W component. These two values define the position of the tool in space, unambiguously. The position of the tool is assumed to coincide with the axis of the borehole. This type is a tremendous improvement over the photographic systems. Since the signals are sent to the surface continuously, any station length may be used, but usually is short (1 metre). The readings may be taken while the tool is moving. Therefore, the interval between measurements may be made so small as to introduce negligible error. The signals usually are processed at the site. A finished plot of borehole deviation is often available shortly after the survey is finished. Figure 3.87.

Another system in cannon use uses a 3-plane fluxgate to read the direction of the earth's magnetic field to determine the tool position within that field. Also, two fixed-axis accelerometers read the earth's gravitational field to define the tilt of the tool. All five of these signals are digitally encoded, sampled periodically, and transmitted to the surface. A surface computer solves the orientation problem and plots hole deviation on-site, within a few minutes after the finish of the measurement. One instrument of this type is made by Owl Technical

141 3 Bearing 3 D D.l 3 Closure 3 ^.3 i True Vertical Dpth. .? 8 9.3 I

FIG.3.89. Sample readout of Century Geophysical system. FIG.3.90. Inertial deviation tool sensor.

Associates, Inc. Samples of the readout of it and the Century Geophysical system are shown in figures 3.88 and 3.89 respectively.

3.1.7.4. Inertial type

None of the magnetic field type tools can be used within a steel pipe, such as casing. Steel or magnetic mineral bodies near a borehole can distort the earth's field seriously. These problems can be overcome by using an inertial device, such as a gyroscope, to furnish a continuous, spatial position reference. Figure 3.90 shows a portion of this type tool diagrammatically. Inertial tools can be handled in exactly the same way as magnetic tools. They have several drawbacks, however, which limit their use severely. They are expensive and delicate. They are subject to precession because they maintain a fixed position in space, regardless of the motion of the earth. They are subject to drift caused by bearing and skin friction. And, the small size required by logging tool diameter makes design difficult.

3.1.7.5. Calculations

Calculation of the drift or deviation of a borehole from deviation surveys can be done in one of two manners. In the mineral industry, it is customary to use a tangential method; a linear extrapolation from each station or calculation point of the two horizontal components of position. This method is excellent and simple if the distance between calculations or

143 FIG.3.91. Positional error in deviation calculation when using the tangential method and long stations.

144 Tension \Weigh t

L. J FIG.3.92. Effect of pulldown or drill collars on the borehole.

stations is short or if the borehole deviation is small and variable. Figure 3.91 shows the effect of station length upon a hole deviation determination. With the linear extrapolation, the angle of deviation and direction of deviation are assumed to be constant from one station or calculation to the next. The vector components of one station are added to the next, with the assumption that they are unchanged.

A better method is employed in the petroleum industry when the slant angle of the hole is changing. This method, the "radius of curvature" method assumes the hole is curved if the upper and lower station slant angles are not the same. Thus, the error is minimized.

Hole deviation methods allow one to locate where the bottom of the hole is. This is important in nonpetroleum'logging where a pull-down method instead of drill collars is often used to put weight on the drill pipe. A pull-down puts force on the top of the flexible drill pipe causing it to deviate because the drill pipe will bend slightly. When drill collars are used, the weight is directly on the bit and the drill pipe is under tension. Figure 3.92 illustrates this problem.

Hole deviation methods also allow calculation of the true vertical thickness of a formation. This is necessary in order to properly calculate reserves. Neglect of a 17 degree deviation will cause an overestimation of reserves by five percent. Figures 3.93, 3.94and 3.95 are charts showing the corrections fran a deviation measurement.

145 MEASURED DEPTH - (FT)

100 500 1,000 5,000 10,000 50,000 -1,000 r - - 1,000 r-~ / H— / - cos- = 7' V / ' - 500 r / / n / / -500 _ b y /' / / y / / / \ / / o / V/ / / / / b = COS •« c / / / / / / / / Dlip acamin -c - b r • / / / / / '/ / OS - / / / / / / / / / / 1/ / / / * / — A $ - 100 - 100 r y / • -v / / - A XV * -A 7 tL -/ / A / - A- - 50 r r/ r / / -50 // / / / / / / / / / / / / / ,/ / / / r/ / / / / / ^ / / / / / t / / / / / / / / / / / A / / / / / */ y / / / / / / r / / / / / / / V / / / - 10 / -10 y J / V J j A- fM JA— y- * / 7" —/ r / y • - 5 t / — 4 / / / -r 7 y / / / / ' / / / / / / / / y / / / / / / / / / / / DRIFT ANCLE - / / / / / / / / / / / / | -1 / / / / / / / -1 100 500 1,000 5,000 10,000 50,000

MEASURED DEPTH - (FT)

FIG.3.93. Depth correction to find true vertical depth of nonvertical holes. MEASURED DEPTH - (FT)

100 500 1,000 5,000 10,000 50,000

MEASURED DEPTH - (FT)

FIG.3.94. Possible horizontal displacement of nonvertical boreholes from their surface location.

147 FORMATION VERTICAL THICKNESS - (FT.)

1 2 3 5 10 20 30 50 100 200 300

FIG.3.95. Chart of correction of deviation measurement.

Hole deviation measurements are difficult in boreholes which are nearly vertical. In this case, the movement of the fluid past the tool will cause it to float away from the hole wall because of the Bernoulli Effect. When this happens the deviation device does not measure the hole drift, but only the probe position within the borehole. Figure 3.96 illustrates this. This will happen at hole deviations of less than about two degrees, depending upon the logging speed, mud viscosity, and tool weight. One should insist that a decentralizer be used on any deviation tool.

148 High . velocity ^fluid V Stationary fluid

Borehole

FIG.3.96. Flow of the borehole fluid around a moving sonde in a nearly vertical borehole.

FIG.3.97. A modern computer logging system. (Courtesy of Century Geophysical Corp.)

149 Field computer systons, which are beginning to be used in seme areas, make the foregoing calculations and corrections possible at the logging site, within a few minutes after the completion of the survey. Such a systan is shown in figure 3.97. This type systsn is not readily available yet, but will be in the near future.

3.2. Other logging techniques

This class of logging techniques includes many methods which are not used to any great extent in uranium exploration and development. They are, however, excellent methods and should be seriously considered. Some of them are run occa- sionaly. These are the density, induced polarization, and magnetic susceptibility. Others, such as the acoustic systems and the high resolution dipmeter are available from both mineral contractors and oilfield contractors. They can be very valuable and should be investigated. The mud log should be used. The value of mud resistivity should be on every log heading, when water-base or foam mud was used.

3.2.1. Neutron-activation systems

Whenever a neutron interacts with an atom there is an exchange and radiation of energy. The detection of that energy can be used to identify the reaction atom specifically. These reactions usually (but do not always) involve thermal neutrons, because of the high probability of thermal neutron capture by most atoms.

There are three major categories of this type tool: direct uranium measurement, the elemental analysis, and the lifetime or decay systems (this latter is used for petroleum only).

3.2.1.1. Prompt and delayed fission neutron methods

The thermal neutrons which 235result from moderation in the formation can react easily with U. When the neutron is absorbed, the atom becomes unstable and will decay into several smaller fragments. It will undergo fission (that is, uranium-235 is a fissile material). When the fission of uranium-235 takes place, several neutrons are emitted within a few microseconds. These neutrons will moderate to epithermal energy where they may be detected by the prompt fission neutron technique (PFN) . The fission fragments are also unstable and decay, emitting more (delayed) neutrons. The average half-life of these fragments is less than 22 seconds.

150 Circuitry . Cuicuitry Circuitry ca —11 Source )etector 1

Low Voltage 5 kV Power Power Belt 125 kv 110 KV Pulse Pulse Power Power Thermal )etector 2 Neutron Detector Epithermal Gamra Neutron Detector Detector

IRT DFN PFN Mnhil OFN

FIG.3.98. Prompt and delayed fission neutron systems.

The neutrons from both of these reactions can be detected when they have reached thermal energies. This is done with the delayed fission neutron technique (DFN).

There are two DFN systems in use. One was designed by Mobil and uses a pulsed generator as a source of neutrons. The other was invented by Eberline and Shreve, run by Kerr-McGee and improved by IRT Corporation. It use9 s a californium-252 isotopic source of neutrons, of about 5 x 10 neutrons per second. The IRT version (development funded by the U.S.D.O.E.) carries the source on a moving belt to irradiate the formation for a longer time, while moving the probe continuously. The PFN and the

Mobil DFN both use electronic generator sources of 14 MeV neu 17 - trons. At about eight MeV the neutrons can react with the 0 in the formation and borehole causing a fission and a production of neutrons. These must be taken into account on both of these systems as a background reading. Uranium-238 is also fissile at about eight MeV. However, since the relative composition of uranium is fixed in most natural deposits, the neutrons from uranium-238 aid detection; rather than hinder it. The IRT syste2 m does not have either of these reactions, since it uses a Cf source of 2.3 MeV neutrons, with a maximum energy of six MeV. All three systems appear to be workable systems. Figure 3.98 shows the three systems diagrammatically. However, all three use complex processes'and therefore, have complex corrections to make. All three are. sensitive to the formation

151 porosity and borehole fluid. The 1RT system has a slight advantage in that it does not have the oxygen-17 background to contend with. However, it must use a larger source to compensate for the loss of the uranium-238 contribution.

The generator type sources can be pulsed and can be turned

' off; the californium-252 source cannot be. On 7 the other hand,

the neutron generato14 r has a life of 10° to 10 pulses or ap-

proximately 5 x 10 neutrons17 . The califomium-252 will have a half-life of about 4.2 x 10 neutrons.

The PFN system pulses the neutron generator 100 t9 o 200 times a second for a total neutron output of about 10 neutrons per second. Each pulse is a few microseconds long. After the neutron pulse is over and the original or source neutrons have reacted and disappeared, an epithermal neutron detector is gated on and the population of prompt epithermal neutrons due to the fission reaction is measured. The original generator neutrons will die away in 50 to 100 microseconds. Any neutrons remaining will be a function of the amount of fissile material in the formation.

The Mobil DFN system uses a similar generator, but it is

only pulses d two to ten times a second for a total output of about 10 neutrons per second. Detection is delayed until all primary and prompt fission neutrons are gone. Detection times are 100 to 500 milliseconds. Substantially higher counting rates are obtained than with the PFN system.

The IRT DFN system exposes a portio9 n of the formation with a californium-252 source of about 10 neutrons per second for a period of time. The source is then moved mechanically to another exposure location and a detector reads the delayed neutrons at the first location.

Handling of all three systems is done by computers in the truck. The general procedure is to run a conventional ganma- ray log. The PFN or DFN is then run through any anomaly noted on the log, regardless of size. A neutron log is also run to assess the porosity of the formation . With the PFN and Mobil DFN a correction is made for oxygen-17 background count.. This is subtracted from the total (corrected) count. The remainder is a function of the uranium (and thorium) content, borehole size, borehole fluid type, and formation porosity. Some trace elements, such as boron and gadolinium will cause errors. How- ever, at present no correction is made for these. Their presence is uncertain, but evidence appears to indicate amounts present which are large enough to cause errors.

152 Sensitivities of 100 parts per million uranium (0.01 percent by weight) seem to be practical for the DFN systems. The lower sensitivity for the PFN system is unknown.

Elemental analysis

If the neutron source is combined with the gamma-ray spectrograph, the number of ganma photons, resulting from neutron capture by atoms, can be quantitatively measured, as a function energy.

Capture gamma photons result from the absorption of neutrons by atoms. The absorption takes place easily at thermal energies of the neutrons. A gairma ray of specific energy will be emitted. Since the energy of a thermal neutron is low (less than one electron volt) the emitted ganma-ray energy is essentially characteristic of the isotope involved in the absorption. Therefore, identification of the capture gamma energy and intensity provides means of identifying the type and abundance of the isotope present.

The neutron activation elemental analysis system can use, to advantage, the high resolution of the germanium detectors. However, a fair amount of work, especially in the laboratory, is done with the lower resolution sodium iodide detectors.

This technique can use any source of neutrons. There are, however, several advantages to the pulsed neutron source. Be- sides the advantage of being able to turn it off, a pulsed neutron source of 14 MeV neutrons can cause reactions with iron and oxygen isotopes at about eight MeV. Since these are both abundant geological constituents, a knowledge of their abundance may be valuable.

At present, this technique is slow and requires strong sources. It is, however, one of the most promising of the new mineral techniques. It has, for instance, given trace element analysis in coal, overburdens, and fluvial muds which are quite good. It is being run commercially in gold and silver, with excellent sensitivity and reasonable accuracy.

3.2.2. Density

3.2.2.1. Introduction

The physical principles which permit ganma rays to be used for density measurements have been known for many years. How- ever, it has only been since about 1960 that the solutions of the practical problems have allowed the successful use of the

] 53 density log. These problems were the expression of the process in physical terms to allow an orderly approach, the progress of technology permitting the development of viable sources,detectors, solid state circuits, and the determination of character istic density ranges of geologic materials.

Density tools are the most common scattered ganma-ray systems. These systems measure the bulk density of the formation, which can be used to derive formation porosity rather accurately. It is an excellent method for evaluating coal because of the 2:1 contrast in density between common sediments and the coal. In combination with the acoustic velocity tool it can be used to determine elastic parameters more reliably than from cores because of the probability of mechanical damage to cores. Density can be effectively cross-plotted with several other systems to determine lithology, corrected porosity, gas content, water content, and several other parameters.

The general quality of most directional, sidewall measurements is excellent, if proper care is taken. Calibrations are imperative for quantitative measurements. The initial calibrations are elaborate and complex.

Some measurements use gamma-ray excitation in one form or another. However, at this time these are mostly restricted to surface techniques.

Circuitry and detectors are similar to gross-count and spectrographic gamma-ray systems. An isotopic source of gamma rays is usually carried in the tool.

3.2.2.2. Density logging

Density logging makes use of gamma rays from a source and scattered back to a detector from the electrons within the formation.

All scattered ganma-ray systems carry a source of ganma rays. Isotopic sources are usually used because of their high yield, compactness, ease of maintenance, and simplicity. Most isotopic sources have gamma-ray energies of less than two MeV.

Density logging uses the same type circuitry as the gross- count gamma systems. Density detectors are standard gamma-ray detectors. They may be G-M tubes, but are usually sodium iodide (thallium activated) crystals and photomultipliers.

Discriminator levels are often set to preferentially respond to the single scattered energies and higher. However, the

154 discriminator may be set to respond to all Compton scattered energies above the housing cutoff energy. Response to single scattered energies is probably to be preferred, but requires much more careful settings and more stable circuitry.

Studies of the depth of investigation indicate that about 50 percent of signal of the density systems originates within an ovaloid volume whose major diameter is slightly larger- than the source to detector spacing and whose minor radius is about one- third the S-D spacing.

The maximum scattering angle within the formation is about 112 degrees. The average energy will be near that of the 112 degre13 e scattering energy. This energy is about 480 keV if a Cs source, at 663 keV primary energy, is used.

If the gamma-ray intensity, I, is examined in a direction outward, normal to the probe axis and in the direction of greatest intensity, it will decrease exponentially. The relationship is

I = I e^px

137 With Cs source of 663 keV garrina rays, 90 percent of the scattering will take place within 16 centimetres of the source. Thus, the density system is a shallow investigation system.

Source type and strength must be chosen with care, as they must meet several conflicting criteria. A gamma ray will interact with the orbital electrons kfoich the gamma photon passes. The photon will act much as a particle and be scattered, absorbed, or disrupt the normal electron pattern. The photon will impart some or all of its energy to the electron.

If the photon energy is low, the photon may eject an electron from its orbit, but in doing so it may lose all of its energy to the electron and cease to exist. The free electron is a photoelectron. This reaction depends upon the atomic number, Z,of the atom. The higher Z atoms require less energy to lift the outer electrons from atomic orbit. Thus, the photoelectric reaction is a function of atomic number and a function of the reciprocal of the photon energy. Both are nonlinear.

If the photon energy is high and the atomic number is high, the gamma photon may cause a pair production of the electron. An electron and a positron will be produced at a threshold energy of 1.02 MeV. The positron will rapidly recombine, with the emission of two gamma-ray photons of 0.51 MeV each.

155 ATOMIC NUMBER

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 50 50 i i

i 1i 1 1/ / I/ /

1t f\ sy /

6 fi 4 A r

(a)

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

ATOMIC NUMBER

FIG.3.99. Compton scattering and photoelectric absorption atomic attenuation coefficients versus atomic number: (a) 0.3 MeV gamma-ray genergy, (bj 0.1 MeV gamma-ray energy; narrow beam geometry. For total attenuation (in barns) add component values.

156 ATOMIC NUMBER

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 50 —I— 1—7 50

25 30 35 40 45 50 55 60 65 70

ATOMIC NUMBER

157 TABLE 3.4. CATALOG OF COMMONLY USED RADIOACTIVE GAMMA-RAY SOURCE MATERIALS

RADIOACTIVE SYMBOL HALF-LIFE GAMMA RADIATION LEAD HVT**I)ENSITY** RADIATION** MATERIAL (relative) (inches) (g/cm3) (MeV) Iodine-131 1 3 1 j 8.05d 0.21 0.14 4.93 60.250(2.8%),0.335(9.3%) 0.608(87.2%),).815(0.7%)

y0.080(2.2%),0.163(0.7%) 0.284(5.3%),0.364(80%) 0.637(9%), 0.722(3%)

Iridium-192 192Ir 74.5d 0.50 0.19 22.4 80.1(1.5%), 0.257 (7%) 0.537(41%), 0.673(48%)

Y Numerous energies from 0.136 to 1.159 (0.136) 0.210, 0.205, 0.282, 0.296 0.308, 0.316(32%), 0.374 0.440, 0.468, 0.589, 0.604 0.613, 0.745, 0.783, 0.885

Cobalt-60 s "Co 5.27y 1.24 0.49 8.9 60.306 Y 1.17(100%), 1.33(100%)

Cesium-137 137Cs 30±3y 60.51(92%), 1.17(8%) Barium-137 137Ba 2.6 min 0.32 0.25 3.5 (CsCl) Y0.662

Radium-226 22 6Ra 1,620y 0.825 0.56 18.7 a 4.8(94%), 4.6(6%) yO.19 (With high contamination from daughters, 0.2-2.2)

Krypton-85 asKr 10.27y 0.30 0.17 .004 60.15 (65%), 0.695(99+%) Y0.54 (65%)

** See Glossary, appendix D. D •Ut

S ^^ Ganma rays are scattered away from their path through the formation

FIG.3.100. Scattering of gamma rays from their path through the formation.

In between these two types of reactions the photon may eject an orbital electron and retain part of its energy. The amount retained will depend upon the energy of the photon. It also will depend upon the angle of scattering. A 90 degree scattering will have a Gaussian distribution of energies centered about a value of approximately half that of the original photon. This reaction is nearly linear with respect to atomic number. This is Compton or elastic scattering.

When a source is chosen for a density system, it must be of a high enough energy to not cause an appreciable number of photoelectric reactions (because of the extreme nonlinearity of these). A lower limit of the scattered energies, for linear, quantitative results is about 300 keV. The upper limit for primary energy is 1.04 MeV. Further considerations are a high enough specific activity that the source is small and essentially a point source,and a long enough half-life that frequent corrections need not be made (this is not as important a consideration in computer systems as in analog systems).Figure3.99.

The number of photons scattered within the formation from the source and back to the detector will depend upon the number of orbital electrons present and the source strength and energy. The number of the photons arriving at the detector, however, will depend upon the number scattered away from the path of the photons and away from the detector whether the system is a back- scattering one or a transmission one. This is illustrated in figure 3.100.

The nucleus of an atom contains virtually all of the mass of the atom. This is the combined mass of the protons and neutrons in the nucleus. Approximately 0.1 percent of the mass of an atom is due to the electrons. On the average, for light elements (Z < 25) the number of neutrons in the nucleus is nearly

159 equal to the number of protons. The atomic number of an element, Z is the number of protons (there is one orbital electron for each proton), and 1the atomic mass, A is the number of neutrons plus protons. The ratio of Z/A is, on the average, about 0.5. Thus measuring the number of electrons will indicate the mass of the material. The major exception to this is hydrogen, which has a Z/A ratio of 1.0.

It is relatively easy to calculate the ratio of any compound whose composition is known. The actual ratio is equal to the sum of the partial ratios. Thus, the Z/A ratio is atomic number of each element Z/A = Z(mole volume) atomic weight of each element

In addition to being easily calculable, Z/A ratios have been determined for a great number of formation minerals. These are listed in Table 3.4. For many purposes it is sufficiently accurate to use a Z/A ratio of 0.5, which is how the tool is calibrated. If a Z/A ratio of 0.5 is assumed, the reading of the density tool will be p , apparent density. The apparent density will differ from thea real bulk density by an amount proportional to the real Z/A ratio compared to 0.5. Thus, apparent bulk densities, p may be converted to real bulk density, p^ a _ 0.5 Pb - z7A Pa If water or petroleum is present in any quantity in the formation and/or borehole, the apparent densities of the materials, especially the water or petroleum, must be used for any calculations. This is because of the large percentage of hydrogen in both of these compounds. Therefore, an apparent density of 1.11 g/cm3 must be used for water instead of the real density, 1.0 g/cm^.

The density system is sensitive to natural gamma radiations as well as the radiation scattered from the source. It is imperative that a density tool be designed to minimize the effects of natural ganma radiation while enhancing the response to scattered gamma radiation. In general, a well-designed density tool will employ shielding and the detector design to have as low a sensitivity to natural gamma rays as possible. In order to have effective counting rates with an insensitive detector it is necessary to use a large source (high output). With cesium-137, probably 100 millicuries is the lowest strength reliable source which should be considered. In sidewall probes, one to three curies is preferable, even with a relatively short spacing.

160 Dead-time effect corrections are as important as they are with gross-count gamma-ray logging. They will correct substantial nonlinearities. If a computed log is presented scaled in density units, the relationship between counting rate and density must be stated on the log heading or scale. .

3.2.2.3. Single spacing

The single spacing density systems have two major variations. The omnidirectional systems are the original and simplest type. The sidewall, directional systems are more complex, but can be used more effectively.

The omnidirectional tools are usually standard gamma-ray tools with a spacer and a gamma-ray source a fixed distance from the detector. Garnna rays are radiated in all directions and the detector is sensitive in all directions. Ctnnidirectional systems must have long source-to-detector spacings and large source strengths in order to minimize borehole effects. Typical spacings are 30 to 50 cm. They are so sensitive to borehole effects however, that it is difficult to use them to calculate density. These systems should only be used in small diameter holes. The possibility of.substantial natural gamma-ray and borehole effects must always be kept in mind. Omnidirectional systems are useful vfaere sidewall systems cannot be used. For example, when the tool has to be lowered through small diameter drill pipe there usually is not room for a sidewall system. These systems are excellent for rapidly locating coal depths (figure 101).

161 FIG.3.102. Sidewall density tool.

The sidewall density tools usually employ an eccentering device, such as one or more bowsprings or spring loaded arms causing the tool to ride against the hole wall. The ganma-ray source is collimated in the direction against the wall of the hole. Figure 3.102.

This type tool responds primarily to formation density. The ratio of formation response to borehole response is the front-to -back ratio and is, ideally, infinite. The actual ratio will depend upon the tool design, tool diameter, type shielding material, and source energy. A well-designed, two inch diameter tool can easily have a ratio of 200. Neglecting the small borehole effects, the response of a sidewall tool (apparent density, Pa) is

b(c p = f + e" " Yn) a where y is the counting rate due to natural garrma radiation, c is the totan l counting rate (corrected for dead-time effects), b is a constant of proportionality and will depend upon tool design and source characteristics, and f is a constant which is a function of the fixed background radiation due to the front-to- back ratio.

One sees, occasionally, density tools using 10 to 100 microcuries of radium. This is done to circumvent license problems. However, such a system is generally useless quantitatively.

162 In actual practice, the response of the density tool may deviate from a pure exponential relationship at high densities because of the increased likelihood of photoelectric reactions. At low densities, the deviation from exponential will be due to increased amounts of hydrogen (and the resulting high Z/A ratio) in the sedimentary formations. Because of these departures from a logarithmic relationship the response of density tools may be represented by a section of parabolic arc, rather than an exponential relation. This takes the form for apparent density, p : a 2 p = a + be + dc a where c is the counting rate and a, b, and d are constants which depend upon the tool designs and settings. This relation is best determined by a least squares fit and is valid only for a limited range. One special form of the single spacing sidewall tool is built for locating coal bed boundaries. This tool uses an extremely short spacing, 5 cm or less. It is usually of small diameter. In order to obtain enough shielding between the source and detector and from the borehole, a low-energy gamma- ray source, such as americium-241 (60 keV) is used. As a result of the short spacing (shallow investigation) and the high amount of photoelectric absorption in the sand or shale, the bed boundary response is excellent. However, these systems are only suitable for locating bed boundaries in smooth wall, good borehole conditions. No attempt should be made to use this system quantitatively.

3.2.2.4. Compensated density , Figure 3.103

It is feasible, by using various combinations of spacings and, sometimes, gating, to eliminate many of the problems inherent in the single spacing tools. These are the compensated systems and are usually a single source and dual detectors.

If the response of a short spacing (15 to 20 cm ) is compared to the response of a long spacing system (30 to 50 cm ) as the thickness of mudcake is increased, the mudcake is a larger proportion of the total volume of the short spacing than of the long spacing. Therefore, as the mudcake thickness increases the counting rate of the short spacing will change more rapidly than that of the long spacing. If this response is plotted, the results are shown in figure3.104. Mud cake may be detected, its thickness determined and its effect upon the measurement compensated for. Irregularities or rugosity of the borehole wall may be compensated for in the same way. Also, if the short spacing detector is made identical to the long spacing detector the effect of the natural gamna radiation would be the same in the

163 CD ^-Mudcake spaci r Lon g op

FIG.3.103. Dual-spacing compensated density system.

0 . . Low density mudcake / Higji or cave _ / density M .5 / / ^-nudcake 8w.

i—sri U W / 1' Density, short spacing

FIG. 3.104. Response, compensated density system.

two. Thus, the effect of natural gamma radiation (i.e. in a uranium bed) could be determined. The compensated density log usually includes a monitor curve which gives a warning when the system is not able to compensate.

3.2.2.5. Uses of density measurements

The density of the formation material is of importance directly. The low density range of coal, for example, is de- finitive of its presence. Likewise, higji density readings can be indicators of massive, nonporous formation material or extra

164 dense contaminants. In table 3.5 densities of various geological materials are listed. It can be seen that the ranges of densities are narrow enough to allow an identification of many lithologies. And, the density measurement is especially valuable in hard-rock environments.

The density systems were originally adopted in petroleum logging as a measure of porosity. It is an excellent one because it allows an accurate porosity determination even in the presence of shale. The reason is that the shale density in a deep hole is very near that of the sand grain density. Table 3.6 shows Gulf Coast shale densities as a function of depth.

The relationships for fractional porosity, <|> are

= Pf + (1 - ) Pg

p = p + + V p b * f Vs sh sh where p^, p^, p , p , and are the bulk density, density of g the fluid, grain material, sand, and shale respectively. If the apparent density is used, the others must be apparent or uncorrected densities.

While density readings can be used effectively by themselves, their value is greatly increased when they are used with other curves, as in cross-plotting.

3.2.2.6. Calibrations

Calibrations of the density systems are important and critical. Literally, no quantitative use of a density system is possible without calibrations. With them the results are precise and the possibilities are endless.

Density calibrations are done in blocks of known density. These blocks must be simple, predictable materials. The block primary material and the trace elements must be determined by chemical analysis and taken into account to determine a real density. Then the Z/A ratio and an apparent density determined. Several blocks of densities covering the range expected should be chosen. The materials conmonly chosen are aluminum, magnesium, and water.

Aluminum has a density of 2.70 grams per cubic centimetre. Its Z/A ratio is 0.4818, so its apparent density is 2.60 grams per cubic centimetre. Aluminum, like many commercial metals is seldom pure. It will have various impurities and alloyed metals.

165 TABLE 3.5. ELEMENTS, ATOMIC NUMBERS AND WEIGHTS, VALENCE, AND THERMAL NEUTRON CAPTURE CROSS-SECTIONS

Atomic Atomic Thermal Neutron Number Weight Capture Cross- Element Symbol (Z) (A) Section (Barns) Hydrogen H 1 1.008 0.33 Helium He 2 4.003 0.0000 Lithium Li 3 6.939 0.0376 Beryllium Be 4 9.013 0.009 Boron B 5 10.811 758.86 Carbon C 6 12.011 0.0034 Nitrogen N 7 14.007 0.075 Oxygen 0 8 15.999 0.0002 Fluorine F 9 18.998 0.0098 Neon Ne 10 20.183 0.0038 Sodium Na 11 22.898 0.400 Magnesium Mg 12 24.120 0.0625 Aluminum Al 13 26.982 0.232 Silicon Si 14 28.086 0.1638 Phosphorous P 15 30.974 0.190 Sulfur S 16 32.065 0.49 Chlorine CI 17 35.457 33.338 Argon • Ar 18 39.946 0.668 Potassium K 19 39.102 2.152 Calcium Ca 20 40.080 0.455 Scandium Sc 21 44.96 25.000 Titanium Ti 22 47.90 6.401 Vanadium V 23 50.95 4.800 Chromium Cr 24 52.01 3.074 Manganese Mn 25 54.94 13.300 Iron Fe 26 55.47 2.514 Cobalt Co 27 58.94 37.5 Nickel Ni 28 58.71 4.264 Copper Cu 29 63.54 3.812 Zinc Zn . 30 65.38 1.06 Gallium Ga 31 69.72 1.147 Germanium Ge 32 72.60 1.470 Arsenic As 33 74.92 4.300 Selenium Se 34 78.96 12.0 Bromine Br 35 79.916 2.798 Krypcon Kr 36 83.80 31.0 Rubidium Rb 37 85.48 0.7 Strontium Sr 38 87.63 1.2 Yittrium Y 39 88.92 1.3 Zirconium Zr 40 91.22 0.18 Niobium Nb 41 ' 92.91 1.2 Molydenum Mo 42 95.95 2.7 Technetium Tc 43 97.00 22. Ruthenium Ru 44 101.1 2.6 Rhodium Rh 45 102.91 150. Paladium Pd 46 106.4 8.

166 Atomic Atomic Thermal Neutron Number Weight Capture Cross- Element Symbol (Z) (A) Section (Bams) Silver Ag 47 107.70 96.00 Cadmium Cd 48 112.41 7,000.00 Indium In 49 114.82 190. Tin Sn 50 . 118.70 0.62 Antimony Sb 51 121.76 5.7 Tellurium Te 52 127.61 4.7 Iodine I 53 126.91 6.20 Xenon Xe 54 131.30 35. Cesium Cs 55 132.91 28. Barium Ba 56 137.36 1.20 Lathanum La 57 138.92 8.9 Cerium Ce 58 . 140.13 0.73 Praseodymium Pr 59 140.92 3.90 Neodymium Nd 60 144.27 46. Prometheum Pm 61 145.0 Samarium Sm 62 150.35 5,600. Europium Eu 63 152.0 1,400. Gadolinium Gd 64 157.26 46,000. Terbium Tb 65 158.93 46. Dysprosium Dy 66 162.50 930. Holmium Ho 67 164.94 65. Erbium Er 68 167.20 173. Thulium Tfti 69 168.94 127, Ytterbium Yb 70 173.04 37. Lutetium Lu 71 174.99 112. Hafnium Hf 72 178.49 105. Tantalum Ta 73 180.95 21. Tungsten W 74 183.86 19.2 Rhenium Re 75 186.2 86. Osmium Os 76 190.2 15.3 Iridium Ir 77 192.2 440. Platinum Pt 78 195.09 8.8 Gold Au 79 197.0 98.8 Mercury Hg 80 200.61 380. Thallium T1 81 204.39 7.0 Lead Pb 82 207.19 0.17 Bismuth Bi 83 208.99 0.034 Polonium Po 84 209. 0.003 Astatine At 85 210. Radon Rn 86 222. 0.72 Francium Fr 87 . 223. Radium Ra 88 226. 20. Actinium Ac 89 227. 510. Thorium Th 90 232.04 7.4 Protractinium Pa 91 231. 260. Uranium U 92 238.03 2.7

167 TABLE 3.6. TYPICAL GULF COAST SHALE DENSITIES

Depth Bulk density (feet) (metres) (g/cm3) 0 0 1.8 2000 600 2.20 4000 1200 2.34 6000 1800 2.44 8000 2400 2.52 10000 3000 2.57 12000 3600 2.60 16000 4800 2.63

The real bulk density of this block (of aluminum and alloyed metals) is ^(-weighypercen. t . (Bulk density of the component element.) _ -[Weight percent Z/A Z (Bulk density of the - L 100 • 03 component element. The total Z/A ratio will be J

Z/A = Z Weight^gercent ^ (z/A of the component element .)J

Magnesium typically has large amounts of aluminum, copper, and silicon in it. But, the procedure is the same as for aluminum. Water is an excellent material to use for the low density end, because it can usually be obtained with negligible impurities ( < 500 parts per million). It does have a Z/A ratio of 0.55 and an apparent density of 1.111 grams per cubic centimetre. It can have salts dissolved in it to raise the density. Sodium chloride, for example, can be used to raise the apparent density of the water solution to more than 1.2 grams per cubic centimetre.

168 If the front to back ratio of the density tool is greater than 200, no correction need be made for borehole effects. If it is less, however, calibrations must be made with air and water-filled boreholes.

Gases can occur in significant amounts in conmercially cast magnesium and aluminum. Also, they often are dispersed so fine- ly they are virtually undetectable. If possible, the block should be forged after casting and before machining. Finally, a machined sample may be taken from the borehole space for density determinations (and chemical assaying). If possible, the entire block should be weighed accurately and the density calculated.

The calibration block (or tank, in the case of water) must be large enough that the effect of external materials is negligible. If we assume that the source is cesium-137, at 660 keV energy the radius of the block must be greater than 16.4 centimetres, for 0.1% accuracy.

Once three reliable points have been determined for the tool response, a curve of counting rate (or logarithm of counting rate) can be plotted against density (apparent). The curve, on semilog paper should be nearly a straight line with a slight droop at the high and low density ends.

3.2.3. Acoustic measurements

3.2.3.1. Introduction

Acoustic measurements include the sonic absorption, acoustic velocity, and acoustic amplitude systems. They were originally developed as porosity devices. However, later useage has demonstrated their usefulness in other areas, such as determining cement bonding of casing, determination of formation elastic parameters, and lithology determination in sediments and hard- rock environments.

The acoustic systems all put mechanical energy into the formation, usually via the borehole, and usually in the form of acoustical frequency pulses. The absorption tool puts a continuous mechanical wave into the formation. The. transmission velocity, the wave types, and the attenuation, and the driving energy all depend upon the mechanical properties of the media through which the mechanical energy travel. The most conmon form of the system measures the wave velocity or travel time.

169 TABLE 3.7. FORMATION AND MINERALS SONIC VELOCITY CATALOG (Compressional wave)

Material Matrix Velocity m/s (average value)(range) Dunite 7978 (7408 - 8780) Gabbro 7189 (6400 - 7189) Hematite 7100 Dolomite 6927 (6773 - 7620) Norite 6609 (6218 - 7000) Diabase 6838 (6628 - 6928) Anorthosite 6710 Calcite 6555 (6417 - 6706) Aluminum. 6260 Anhydrite 6096 Albitite 6070 (6020 - 6161) Granite 6000 (5697 - 6513) Steel 6000 Limes tone 5862 (5715 - 6401) Langbeinite 5862 Iron 5852 Gypsum 5806 (5.751 - 5806) Serpentine 5700 Quartzite 5542 (5300 - 5800) Quartz 5532 (5486 - 5570) Sandstone 5347 (3048 - 5944) Casing (steel) 5334 Basalt 5301 Polyhalite 5301 Shale (1793 - 5080) Aluminum tube 4999 Trona 4694 Halite 4572 Sylvite 4115 Copper 3871 Camallite 3658 Cement (wide variation) 3208 (3208 - 3658) Anthracite coal 2903 (2540 - 3387) Concrete (wide variation) 3200 (2438 - 3658) Bituminous coal 2540 (1800 - 3048) Sulphur 2499 Lignite 1905 (1000 - 2177) Lead 2191 Water, 200.000 ppm NaC1.15Ib£/in2 1689 150,000 ppm NaCl, 15'lbf/in2 1638 100,000 ppm NaCl,15lbf/in 1585 Pilre 1472 Glacial Ice 3499 Rubber (Neoprene) 1600 Kerosene, 15 lbf/in2 1420 Oil 1280 Methane, 151bf/in^ 488 Air, 15lbf/in2 335

170 3.2.3.2. Description

A mechanical wave or pulse will exist as a compressional wave in a liquid or a gas. A compressional or pressure wave CP- wave) is one where the molecules of the transmitting medium move parallel to the direction of wave propagation. In a rigid solid, however, the energy can propagate as a P-wave or a shear wave (S-wave). An S-wave is one where the molecules of the medium move normal to the direction of energy propagation. P-waves and S-waves travel at different velocities within the same medium. These velocities will depend upon the mechanical properties of the medium.

The travel time of the P-wave is ATpi

% AT = (p/gE ) p b where p is the formation bulk density, g is the gravitational acceleration, and E^ is the bulk modulus of the formation.

The shear wave travel time, ATg, is similarly related to the density and elastic properties:

/3 ATS = (p/Eb* Esg)^ where Eg is the shear modulus.

The interval travel time, AT, is usually in microseconds per foot on field logs in the U.S. The velocity, V,may be used instead of interval travel time:

6 V = 10 /AT

3.2.3.3. System types The simplest form of acoustic velocity tool is a single transmitter, single receiver tool. When this tool is centered in the hole, the acoustic pulse travels from the transmitter to the formation through the mud as a P-wave. Hie pulse will enter the formation and travel as a P-wave and an S-wave. This discussion will be confined to the pressure (compressional) wave and the shear wave.

Each material has its own characteristic narrow range of velocities (see Table 3.7). The P-wave and S-wave will travel through the formation at the average velocity for that mixture of materials. The waves will radiate back into the hole. They

171 travel through the mud from the wall of the hole and are detected by the receiver. Therefore, the total time, AT^, for a centered tool is

ATrp = ATe + AT + AT T fnl m 11f0 = AT +2 AT. m t where AT-Q and ATf2 are the travel times through the borehole fluid at the transmitter and receiver, respectively, and AT is the average velocity through the formation.

With the single receiver system, the AT through the mud remains an unknown quantity because of the uncertainty of the tool position. In order to overcome this problem, two receivers are commonly used. Then, electronically the travel time at the first receiver is subtracted from that at the second:

AT, = AT + 2AT T 2AT T m2 f " ml " f

AT AT = m2 " ml AT = AT / d T where d is the distance between the two receivers. This way, if the tool is positioned properly, the travel time through the mud is cancelled out. If the distance between the receivers is divided into the travel time, the interval travel time, AT is obtained.

When the two receiver system is used, the assumption is made that the transmitter and both receivers are the same distance from the borehole wall. This is not always true. The probe can be canted in the borehole, or the borehole diameter may be irregular. The Borehole Compensated Sonic system (BHC) was developed to overcome this difficulty. This systan uses two transmitters and two receivers. One transmitter is above the receivers and one below. The pulse from each transmitter is handled as with the two receiver instrument. But, then the responses due to the two transmitters are averaged: AT, + AT ? AT = —

The acoustic tools use two main types of recording systems. The original and still the most widely used system records only the travel time of the first arrival of the P-wave. This is usually detected by means of an amplitude discriminator. The second type system records the full wave train arriving at the

172 receiver. Detection and sorting into component waves may be done optically (photographic recordings) or it may be done digitally (sensing relative time frames, apparent frequency, and phase relationships). The first type is usually recorded in log or strip-chart form. The second method recording will depend upon the system and the use intended.

If the full pulse is sent to the surface in analog equipment, the pulse is usually used to modulate the amplitude or intensity of a triggered oscilloscope. One will see examples of several types of recording. The initial pulse will trigger the oscilloscope. The sweep rate will be picked to have a definite time scale. Sometimes the initial portion of the sweep,' from the triggering pulse until just before the first arrival, is suppressed. This allows the interesting detail to be expanded without losing data. The recording is made on a strip film which is drawn past the oscilloscope screen as a function of depth and logging speed.

If the system is a digital one, sampling of the wave may be done downhole and an encoded signal sent up, an analog signal sent up, an analog signal may be sent to the surface and reduced at the surface, or separation of the P-wave and S-wave may be done downhole. In any of the three digital options the first arrival of the P-wave is determined by discrimination. The S-wave, however, arrives later, because it usually had a lower velocity, and is added vectorially to the P-wave. Therefore, the S-wave can be detected by a sharp change in amplitude, an apparent frequency and phase change, and shift.

There are several important modifications of this system for special uses. The amplitude of the P-wave can be recorded. This is usually done in analog log form but may be one of the digital variations. The resulting log is a function of the attenuation of the signal due to scattering, reflection, refrac- tion, or compression losses.

Laboratory equipment is used routinely to determine the acoustic velocities in cores. These methods have been standard for reference for many years. The principle is the same as downhole logging, but the geometry is different, of course. Helander and Myung feel, with good reason, that the in-situ (or log) measurements probably are more reliable than the laboratory measurements; see references. Surface seismic techniques are valuable in combination with the acoustic velocity logs. You are referred to a standard text on surface seismic techniques.

173 3.2.3.4. Uses

The initial use of the acoustic velocity tool was as a porosity measuring device.

Since the individual components of a formation have characteristic travel times. (AT), and the travel time through the formation is the sum of all of the component travel times, we can say that the measured travel time, AT (Wyllies Time Average) is

AT = <}>ATf + (1 - )AT where $ is the fractional porosity, ATf is the travel time in the pore fluid and AT is the travel time in the rock matrix: A more general case involvem s mixtures:

AT = VrATf cf + V sA T s + V s,h A T s,h + V a AT a v^iere s is sand, sh is shale and a is any other component and V is the fractional volume, Vf is the same as <\>. Note that these two equations resemble the corresponding equations for the density closely.

The P-wave is usually used for porosity determinations, although either the P-wave or S-wave could be used. When the wave mode is not specified, one is safe in assuming the data are for the P-wave. This determination is a good one if three factors are kept in mind: First, the acoustic velocity measurement is a shallow one. Secondly, the acoustic wave porpagation is sensitive to the presence of clay. Thirdly, any gas, or other "compressable" material will cause severe attenuation.

We have already seen that the travel time (or velocity) of a mechanical pulse within the formation depends upon the elastic moduli of the formation material. This is an excellent use for this curve and has the potential of saving large coring costs. The density curve is used with the P-wave and S-wave information. The relationship is

F/A = E e/l c o where F is the force applied, A is the cross-sectional area, Ec is the modulus of compression or Young's Modulus, e is the elongation (positive or negative) and 1 is the original length. This can also be expressed in terms of acoustiG c velocity:

E = ZE (1 +P) c S where P is Poisson's Ratio:

174 2 _ %(V /V ) - 1 P S (V /VV )2- 1 •p'D SC ' Poisson's Ratio is the ratio of transverse contraction to elongation.

The Shear Modulus, Eg,is the modulus of the change of shape with transverse force. The shearing stress, S,is

S = (F/A) = 0Eg where F is the applied force, A is the unit area and 0 is the shear angle. From the logs Eg is

Es = PV where p is the bulk density and Vg is the shear wave velocity.

The Bulk Modulus, E^,is the modulus of the change of volume with external force: v - v P = E — bk v o where AP is the change in external force or pressure, v is the final volume and is the original volume. From the logs Vq

2 Eb = pVp - 4/3 Eg The shear wave does not exist in a liquid, so in a liquid

Eb = Py A very important use of acoustic velocity logs is in correcting the values used in seismic surveys. This is called a Seijs'mic Reference Service when it is combined with other services, such as a geophone measurement of surface shots. This allows a much more accurate and detailed assessment than either one alone.

Another acoustical method which is important is the Cement Bond Log. The impedance match between steel casing and drilling mud is poor. Therefore, this interface is a highly reflecting surface to high frequency mechanical pulses. But, if cement is bonded to the pipe in the annulus, the interface impedance match with the steel is much closer. As a result, when the cement is bonded properly, energy is lost from the pipe to the cement and formation and less is available at the receiver. Therefore, a drop in amplitude indicates good bonding.

175 Other methods which will only be referenced are acoustic holography and standard acoustic methods for mine-face logging, high resolution seismics for location of coal, and sonar methods for underwater use. One final technique is fracture location. Because of the sharp discontinuity affected by a fluid-filled fracture, reflection from a fracture is high. This usually results in a great loss of amplitude and cycle skipping.

Cycle skipping is caused by a pulse amplitude which is low enough that the discriminator rejects the first arrival and detects the second. The result on the log is an anomalously long travel time.

3.2.4. Induced polarization 3.2.4.1. Introduction

Induced polarization methods are mainly surface methods. However, downhole induced polarization methods have shown usefulness in locating sulfide mineral deposits, for correcting surface methods, and responding to alteration in clays and shales. Their greatest usefulness is when they are used in combination with other methods, surface and downhole. Induced polarization measurements are generally used at this time in combination with resistivity measurements.

3.2.4.2. Explanation

To some extent resistivity measurements are frequency sensitive under many circumstances. They are complex measurements and contain reactive components. The reactive portion is normally small, but measurements can be designed to enhance the detection of those electrical properties of the formations which change with frequency. These are called "induced polarization" methods (IP) or complex formation resistivity measurements.

The IP methods are based on the electrochemical phenomenon of overvoltage. The technique establishes and detects double layers of electronic conducting materials when an electrical current is caused to flow across the interface (figure 3.105).

If af direct current is caused to flow through a formation, a voltage drop across the formation can be measured. The ratio of this voltage to the current is the resistance of the volume under investigation and can easily be converted to resistivity. If the current is interrupted, the voltage will drop rapidly to a lower value and then will decay slowly over a period of time from milliseconds to hours (see figure 3.106).

176 FIG. 3.105. Basis of IP effect.

FIG.3.106. Typical chargeability gate definition on a multigate time domain IP receiver.

If the current is an alternating one, the earth will appear to have a reactive component and the resistance (more correctly "impedance") measurement will vary with frequency.

All naturally occurring sulfides of metallic luster, some oxides, and graphite give marked IP responses when present in sufficient volume. This response can occur even at low concentrations of these materials and vfoen they are in the form of discrete, non-interconnected particles. Refer to table 3.8. Induced polarization is the only method currently available which can be used to directly detect disseminated sulfides, such as the "porphory type" or bedded copper deposits, and bedded lead-zinc deposits in carbonate rocks. Certain clays and platey minerals, such as serpentine, sericite, and chlorite sometimes exhibit so-called "membrane" polarizations. This phenomenon can be used as an indicator of alteration. See figures 3.107(a) and 3.107(b).

177 Ul o z «t o aUJ. 2 a UJ N < s oCC z

0.01 0.1 1 10 100 FREQUENCY. Hz

FIG.3.107(a). Frequency effect of some clay-bearing sandstones (after Sill, 1964).

0.001 0.01 0.1 .12 FREOUENCY.H z

FIG.3.107(b). Frequency effect of a porphyry copper.

178 TABLE 3.8. IP RESPONSES OF VARIOUS ROCKS

Rock Frequency Metal Effect % per Factor niho 'm Decade

Rock with concentrated sulfides > 10 > 103 High-magnetic iron ores 20 to 10,000 Porphyry copper ores (2% to 10% sulfides) 5-10 30 to 3000 Rocks with a trace of sulfide mineralization 2-5 1 to 100 Volcanic tuff 2-4 1 to 300 Sandstone 1-3 1 to 30

Basalt 1-2 h to 7 Granite 0.1 to 0.5 < %

FIG.3.108. Principle of the time domain technique.

179 Most IP measurements are made by measuring voltages from overvoltage effects in boreholes (or at the surface), due to currents flowing between electrodes and through the formation. This is the electrical IP (or EIP). Scintrex also has a technique which measures the magnetic fields associated with these currents and call it the magnetic IP (MIP) method. This latter has the advantage of not requiring electrical contact with the ground for the measuring system, thus opening the way for moving vehicular measurements.

There are three different techniques which are used in general field practice for both the EIP and MIP. These are the time domain, frequency domain and phase shift techniques. These

techniques usuall2 y yield equivalent results if currents are less than 10 yA/cm . Frequencies of less than 0.125 Hz are usually used to eliminate electromagnetic coupling effects due to conductive formations.

3.2.4.3. Time domain

With the time domain technique, an interrupted, alternating square wave current of low frequency is passed through the formation (see figure 3.108). The IP voltage measurement is made during the off period between current pulses.

The relationship between the over-voltage, E (which is the extra voltage needed to charge the particle), andc the current density, J, is

E = a - b logioJ c where a and b are experimentally determined constants.

The IP response, or chargeability, M

M'(t, ) = E ^ p is the ratio of the average secondary voltage, E , or overvoltage effect, to the primary voltage, E , inmediately befors e the current is cut off. P

3.2.4.4. Frequency domain

The frequency domain technique usually uses a low frequency, square wave current but without the off-time between current pulses. The apparent resistivities are determined at several different frequencies (figure 3.109).Frequencies of 0.01 to 10 Hz are usually used.

180 Avmv

6 ks- h \ 4 K

3 I v v Ttf*t Domain: T s i,Z4or6MCOflds. «wntcft Mioeutft. N 1 \ \ —T— •-T-H V 1 4 u 0 50 100150 t, sec Fr«qu«4icy Domain: T • jandl • 001.03.10w JO Mi.

INDUCED POLARIZATION DE- CAY CURVES FOR (1) PLASTIC (2) LIMESTONE, (3) DOLOMITE AND (4) SANDSTONE

FIG.3.109. Principle of the frequency domain technique.

When variable frequency measurements are made, results are given in terms of "metal factor" (M.F.) or percent change in resistivity, P, per decade of frequency change:

p - Ri ~ Rz x 100% "\1 Ri x R 2 5 P x 10 M.F. =

l|Ri R2 where Ri and R2 are the apparent resistivity measurements made at frequency 1 and frequency 2.

The IP is also sometimes represented by the percent frequency effect, PFE: Rjr - Rl 0£ PFE = \ Rr

181 PHASE. MILLIRADIANS

FIG. 3.110. Phase shift method in clay-bearing sandstone.

FIG.3.111. Frequency response of a phase shift method.

3.2.4.5. Phase shift method, Figures 3.110 and 3.111. # Another method, which is really equivalent to the chargeability (M) and the frequency effect (PFE) is the phase shift method. It is expressed in milliradians, <|>,

182 where Vq is the quadrature voltage (or reactive component) and Vj is the resistive component voltage. This is often called the metal conduction factor (MCF): V = v j v f I + q Scintrex has a method which measures the relative phase shift of the 1st and 3rd harmonics, instead of the shift of the fundamental. They claim an accuracy of ± 0.15%.

3.2.4.6. Operation

To make EIP measurements, a receiver is connected to two impolarizable electrodes in contact with the formation. On a typical downhole system, a current electrode is at the bottom of the array and the cable armor is the current return. The measurement or potential electrodes are installed at intervals of 10, 40, 80, or 160 feet on the sonde and/or cable. These may be one or two electrodes, depending upon the system and spacing desired. The depth of investigation is a function of the measure electrode spacings.

For directional logging, the current electrodes may be placed on the surface. They are usually deployed on a north- south line, then on an east-west line. The resulting downhole measurements can determine the quadrant in which the target lies. Since frequency effects are being measured, the IP measurements can be affected by the cable length (in an analog system) and by the spacing of the electrodes. 3.2.4.7. Surface methods

Surface methods are similiar to downhole methods and evaluations. Spacings are much greater, and thus the currents are higher. Surface Induced Polarization/Resistivity measurements are excellent in combination with downhole measurements for the location and evaluation of sulfide mineralization and redox trends. The surface-downhole combination is one the explora- tionist should seriously investigate. See figure 3.112. Refer to "Electrical Methods in Geophysical Prospecting" by George V. Keller and Frank C. Frischknecht.

3.2.4.8. Applications

The primary use of IP methods has been for the detection of sulfides. These semimetallic conductors exhibit strong IP

183 20W I2W 4W -1 I

n =1 Resistivity

(OHM-m) n=3

n = 1 — RPS. n-2 (Degrees) n=3 mam Anomaly location 3Hz transmitted square wave FIG. 3.112. Dipole length 200ft. IP: lines of equal resistivity frequency effect and phase shift.

effects which are a function of the current, spacing and the sulfide quantity present within the measured volume. Most of the work so far has been performed in igneous and metamorphic environments. Some interest, lately, has been evident in sedimentary environments for uranium and coal evaluation.

The IP technique appears to be an excellent one for uranium, because of the alteration products and (often) associated sulfides. In a sedimentary environment the altered feldspathic minerals on the barren side of the geochemical cell can result in low level, but anomalous IP responses. On the reduced side of the cell, sulfide minerals are common.

In hydrothermal and other metamorphic environments, the uranium can be indicated by the often associated sulfide minerals and graphites.

In coal, the IP appears to give a response which is a function of the sulfide mineral quantity and perhaps the carbon content.

3.2.5. Magnetic susceptibility

3.2.5.1. Introduc tion

Magnetic susceptibility logging is not a very widely used technique at this time but it has promise. Much of the pro-

184 blem is related to instrument design. Susceptibility measurements are a means of identifying certain compounds and especially certain states of many compounds. The technique could be important in uranium exploration and in hard-rock logging.

3.2.5.2. Magnetic susceptibility

Magnetic susceptibility, x> of a material is a measure of how that material differs, magnetically, from a vacuum or free space. It can be expressed mathematically by

B = (uQ + X) H where B is the magnetic flux density, H is the magnetic intensity, u0 is the magnetic permeability of free space, and x is the magnetic susceptibility of a material. Magnetic susceptibility is a property of a material and can be used for identification of the material and of its chemical state.

Susceptibility is usually expressed as a component of the permeability, V^jOf a material:

ym = (v0 + x)

It is also encountered in the relative magnetic permeability, K:

Jlagnetic susceptibility is usually measured by putting an alternating magnetic field in the formation and measuring the resulting field strength. This type of instrument is very similar to the induction log instrument, but it measures the field component which is out-of-phase (rather than in-phase).

At this time, most susceptibility measurements are done on surface samples. The reason is that temperature and pressure induced instrument changes have been difficult to handle. These are not insurmountable problems, however.

The importance of this method lies in the facts that: 1. the susceptibility of a mineral is characteristic of the mineral, 2. iron, which exhibits large susceptibilities and changes of susceptibility, is frequently associated with sedimentary mineral deposits, and 3. changes in susceptibility take place when many minerals are oxidized or reduced.

185 TABLE 3.9. MAGNETIC SUSCEPTIBILITIES

Material Temperature Susceptibility °C 10"® ergs

Aluminium, Al 18 0.65 230 0.64 Aluminum Oxide AI2O3 -0.098

Calcium Carbonate CaC03 -0.382

Carvon, diamond, C 2020 -0.5-0.409

Carbon, graphite, C 20 -3.5 300 -2.7

Chromium, CR 18 3.6

Chronic Oxide, Cr203 18 25.5

Copper, Cu 18 -0.086

Cupric Sulfate, CuS0./5H20 5.9

Cupric Sulfide, CuS 17 -0.20

Cuprous Sulfide, Cu2S 18 -0.18

Ferric Chloride, FeCl3 20 86.2

Ferrous Chloride, FeCl2"4H20 19 60.1

Magnesium Carbonate, MgC03"3H20 -0.525

Nickel Hydroxide, Ni(0H)2 48.3

Nickel Sulfate, NiS0i/7H20 19 16.0

Silicon Dioxide, Si02 -0.493

Uranium, U 18 2.6

Uranium Dioxide, U02 17 7.5

Uranium Oxide, U308 15 0.95

Uranium Trioxide, U03 16 1.08

Vanadium Oxide, V02 13 3.73

Vanadium Petoxide, V205 15 0.85

Vanadium Trioxide V203 15 13.9

186 These methods should be examined more closely. They probably will be a useful method in hard-rock environments. Table 3.9 is a table of magnetic susceptibilities of many minerals.

3.2.6. Dipmeters

3.2.6.1. Introduction

Dipmeter systems make use of any high resolution measuring system which is directional and viiich responds to lithological changes. They have been developed using the SP, short normals, laterals, calipers, hole rugosity measurements, and (the most successful) focussed resistivities. Since the focussed resistivity device is the major one in use now, the discussion will be confined to it. The same principles apply to all others.

Dipmeter measurements are used to determine formation dip and are extremely valuable for evaluating depositional sequences, history, and stratigraphic features.

3.2.6.2. Principle

Dipmeter measurements are made at several places around a circle, in a plane normal to the axis of the borehole. Typi- cally, these will be spaced at 120° (3 arm) or (HRD) 90° (4 arm). The position of the arms in the borehole is known with respect to magnetic north. The assumption is made that the same anomaly will appear all around the circle (which coincides with the borehole wall). Thus, the relative depths of the anomalies of the three or four readings can be used to calculate the dip of the formation with respect to the borehole axis.

A hole deviation device is run simultaneously with the dipmeter. The calculation of the hole deviation defines the position of the borehole axis. Thus, the true dip angle and direction of the formation can be determined, with respect to true north and to a horizontal plane.

High resolution dipmeters (HRD) can also be used to locate fractures and fracture types.

3.2.6.3. Computation

Dipmeter computation methods can be manual, but the mathematics, vAiile relatively simple,is rather long and involved. Typical methods can be found in the referenced literature. Many modern methods use computer correlation and computation, and usually require digitized data on magnetic tape.

187 FIG.3.113. Standard presentation of formation dips and borehole inclination for the arrow, or tadpole, plot.

3.2.6.4. Presentation

Dipmeter data, like borehole deviation information, must be reduced and presented in a comprehensive form before they can be utilized. There are several popular formats. These are tabular printouts, vector (or "tadpole") plots, cylindrical, linear polar and azimuth frequency plots. Each has its use and one should specify the plot or plots best suited to his needs. Of course, with the data on magnetic tape there is usually no problem replotting the information in another format.

Usually tabular data list the depth and thickness (or top and bottom depths) of a formation, the formation dip angle and direction, and (often) the borehole angle and direction. A tabular listing should also indicate the degree of confidence (or correlation coefficient) of the computation.

The vector plot presents a grid containing a vertical depth scale and a nonlinear horizontal degree-of-dip scale. Small arrows point in the direction of dip (azimuth). A small ball at the base of the arrow indicates the degree of dip. This presentation has the advantage of emphasizing patterns (figure 3.113). This is the standard method of presenting dipmeter data. Other types of presentation are shown in figure 3.114.

The azimuth frequency polar plot can be quite useful. It is constructed using a polar plot and plotting the number of thickness units of dip contained in a particular segment.

188 FIG.3.114. Examples of dipmeter plots.

189 3.2.6.5. Uses

There are many uses for the dipmeter including detection of traps, determination of fluid movement, estimation of drilling target depth, location and type of fractures, and many features of structural geology and sedimentary geology.

The obvious use, determining formation dip, is the original and most frequent use of the dipmeter. In mineral exploration formation dip information is needed to calculate how deep a mineral-bearing horizon will be at: the next hole and the direction to it. The target depth, d, is

d = s + (D tan 6) - e where D is the distance to the next hole, 8 is the angle of dip in the direction to the next hole and e is the elevation of the horizon of interest in the first hole and s is the surface elevation of the next hole.

Formation dip indicates the probable present direction of fluid flow in the formation. Amount and change of amount of dip give indications of the proximity of basin axis and edges. In both petroleum and mineral logging, dip indicates the direction to go for the next hole; up dip for oil or gas, down dip for water, up dip for altered zones, down for unaltered, along the strike for trends of sedimentary minerals. This is a simplified generalization, of course.

In combination with the other logs, such as the resistivity and Spontaneous Potential curves, the dipmeter can give excellent pictures of the occurrences, locations, and even orientation of faults, fractures, and stratigraphic features.

The most valuable use of the HRD system is that its patterns give clear, detailed and accurate pictures of the stratigraphy and depositional history of the formations. The presence, the type, and the details of many features, such as simple, multiple and compound faults, cross-bedding, and many others are clearly depicted on the HRD plots.

3.2.7. Mid logging

3.2.7.1. Introduc tion

Mud logging is a measurement technique used extensively in the oil field and which should be seriously considered in mineral exploration and development. The technique involves

190 measuring mud characteristics before and after circulating through the borehole,to control the quality of the mud for drilling purposes and to gain some knowledge of subsurface conditions through analysis of changes in the mud characteristic due to materials picked up during drilling.

3.2.7.2. Uses

Control of mud quality for drilling purposes is gaining favor in the U.S.A. in uranium development and exploration work. Often, the saving of time and money through a well-designed mud program (including mud logging) is enough to pay for the program itself. More important than designing a mud program to improve drilling (which is extremely important), it should be designed to improve logging. Logging is almost the sole purpose for drilling the mineral borehole. Improvement of logging with a good mud program should be a major concern.

Drilling mud quality control can reduce swelling shales and alleviate the problem of stuck drillpipe. It can reduce the number of "bridges" encountered by a logging sonde going into the hole, thus reducing the amount of time needed for logging, improving depth measurements and reducing frequency of damaged cable and lost equipment. Reduction of mud water-loss will reduce the depth of mud filtrate invasion into the formation. This will improve log quality and reduce formation damage. It very often results in the holes remaining useable for a long period of time. Reduced invasion and control of mud filtrate resistivity can improve SP and resistivity logs greatly. Control of mud viscosity can improve the accuracy of the SP log.

There are many other things which can be done to improve drilling mud which will also improve both samples and logs. It is strongly suggested that the services of a good mud engineer be retained before the field phase of the project. He should be further consulted after the program has entered into the field phase and his recommendations adhered to.

During drilling, passage of the drill bit through the borehole results in cuttings, clays, formation water and gas being mixed with the drilling mud and causing deterioration of the mud and formations. These can be monitered by logging the mud char- acterestics during drilling, resulting in a more trouble-free drilling operation. In addition, valuable formation information may be gained, such as anomalous radioactivity, changes of salinity, changes of shale and clay content, the presence of sand, clay, or

191 carbonaceous material, formation pH, the redox state of the, formation, and gas presence and type. Virtually all resistivity and potential measurements from a borehole must be normalized to mud parameters. Alpha and radon measurements can only be obtained by monitoring mud conditions. These are all important parameters which can be as valuable to evaluating a mineralized formation as the logs or cores themselves.

Mud logging can be elaborate or simple. An elaborate mud program should be handled by a mud logging contractor. Simple programs can be handled by a few in-line measuring instruments. These instruments include ganma-ray or alpha monitors, resistivity (salinity) meters, flowmeters, viscosity meters, pH meters, mud density meters, gas chromatographs, and drilling time records. All of these instruments are available cornier- ially and should definitely be considered for definite purposes.

3.2.8. Use of petroleum logs and petroleum logging equipment

3.2.8.1. Introduction

Petroleum logs are often a valuable source of information for uranium exploration. One must be careful, however, because petroleum logging equipment and mineral logging are designed for two different purposes.

3.2.8.2. Sources

There are many sources of petroleum logs and logging equipment. When petroleum logs are run in an unknown area, the logs should be recorded to the surface, to aid in the detection of mineralization. When exploring in a new area it is worthwhile to check for the existence of any oilwells and resulting logs; log libraries in many big cities in petroleum producing areas are a good information source. The latter information sources should be carefully checked at the beginning of any uranium project.

3.2.8.3. Characteristics of petroleum equipment and logs

Petroleum logging equipment is designed for the environments in which petroleum is' found.. This means that the tools are large, have heavy housings and are designed for different sensitivity ranges than those we need for uranium exploration.

A petroleum gamma-ray curve is recorded with a large, high sensitivity detector. The equipment usually has a long and often variable deadtime. Recordings of API units give little

192 clue to counting rates. Therefore, a counting rate of 3 times shale is as high as the curve may be safely used quantitatively. About 5 times shale intensity is often the counting rate limit of oil-field equipment.

Most petroleum tools have long spacings. Therefore, compared to mineral logs, detail is lost. Many, tools, such as the induction log have limited ranges, since they are designed for more saline environments. High resolution bed detail and boundary location are not important in petroleum work. There- fore, they are net accurate. On the other hand, virtually all petroleum logs are well calibrated and documented, compared to mineral logs. Make use of petroleum logs, but be careful of them.

193 4. INSTRUMENTATION

4.1. Basic systems

4.1.1. Radioactivity detectors

The detector is the primary component which differentiates one radioactivity logging system from another. The type detector also determines many of the interpretation details and methods. Therefore, a fair amount of study must be devoted to them. Since the general characteristics of radiation detectors are similar for all of the radioactivity logging methods, these have been combined in this general section, rather than in each specialized section.

All detectors used in logging make use of ionization or excitation produced in a detection medium as a result of the absorption of all or part of the energy of the photon or nuclear particle. An X-ray and gamma-ray photon must undergo a photoelectric or Compton interaction which transfers all or part of its energy to an electron. The electron, in turn, produces a track of ionization or excitation. Neutrons react negligibly with electrons (since neutrons have no electromagnetic field). But, they interact by collision with a nucleus which will produce a charged particle, which then produces the ionization or excitation.

An ionizing particle will lose energy in a gas by collision or other interaction. The energy lost will produce pairs of charge carriers. An electrostatic field between two electrodes in the gas will cause the charge carriers (ion pairs) to move, resulting in a current in the circuit. Radiation is detected and measured by measuring the current in the chamber circuit. This is the ion chamber (figure 4.0) . It is slow and inefficient because of the low field strengths and low gas density used.

If the field strength of the chamber is increased sufficiently, electrons can accumulate enough energy after ionizing collisions to ionize other impacted gas molecules. The number of carriers will be increased by a factor of two with every collision. Therefore, the chamber current is higher in a proportional counter. The pulse amplitude of a proportional counter is proportional to the energy of the incident particle.

The proportional counter was used in early ganma-ray spectroscopy. The greatest use of proportional counters in logging equipment is in neutron detection. One popular form is the helium-3 detector. The helium-3 detector is a proportional detector filled with iHe gas. The pressure of the gas is made

194 FIG.4.0. Ion chamber circuit. FIG.4.1. Proportional counter circuit.

as high as possible to increase the density and make collision by a neutron more likely. Pressures of 4 to 10 atmospheres are camran. Also, helium-3 has a large cross-section to thermal and epithermal neutrons. Sometimes the detector is lined with uranium-235 to further increase the sensitivity to slow neutrons. Uranium-235 has a large cross-section to slow neutrons and also is fissile with radioactive fragments. ( Figure 4.1.)

The pulses from a helium-3 detector are small and require external amplification, but the detectors are extremely efficient and compact.

The helium-3 detector can be made into an epithermal (only) detector by surrounding it with sheet cadmium. Cadmium has a tremendous cross-section to thermal neutrons. Thus, only the epithermal neutrons pass the cadmium for detection.

If the gradient of the electric field in the chamber is increased, the probability increases that light photons will be emitted. These photons may produce further electrons through photoionization, causing avalanche effects throughout the chamber. In this case the pulse height is constant and is independent of the energy of the primary ionizing particle. This is a Geiger-MQller or G-M detector. The avalanche effect is responsible for many of the desirable features of G-M detectors. Pulses are of high amplitude (as much as 100 volts) and of uniform amplitude. This eliminates some of the circuitry needed for other types of detectors.

When the slower moving positive ions of an ion pair in the G-M detector reach the cathode shell, their impact will cause emission of light and a repetitive avalanche causing the tube to destroy itself. Therefore, external and internal quenching mechanisms are designed in. These are current limiting, voltage dropping methods and gases for filling which will combine rapidly with the ions created by the process.

195 FIG.4.2. Effect of increasing the voltage across a detector tube (plateau).

One result of the avalanche process is that the G-M tube is insensitive to another pulse for about 300 microseconds (dead time). Thus the counting rate is severely limited. This effect can be reduced by using a bundle of tubes which can divide the dead-time by the number of tubes.

G-M detectors are a great improvement over earlier detectors. Their sensitivity is reasonable (about 1 percent). The required circuitry is simple; no ultra stable power supplier nor ampli- fiers are needed. And the signal to noise ratio is excellent.

If the voltage across a detector tube is increased (see figure 4.2) the counting rate will increase at first. This is the proportional range and is described in the preceding para- graphs. With a further increase of voltage there will be a level, above which the counting rate will not increase with voltage. This is the G-M range and is commonly called the plateau. If the voltage is increased above the plateau range the avalanche process will be repeated continuously.

Hie operating voltage of the tube should be set a small amount below the center voltage of the plateau. This will allow fairly substantial power supply changes without affecting the counting rate sensitivity. Setting the voltage slightly below the center point allows a greater range of voltage variation with changes due to higher temperature downhole.

Scintillation detectors were developed during World War II. They are an outgrowth of a device called a spinthariscope used by Lord Rutherford. A modem scintillation detector uses one of many types of scintillator material and flashes of ligfrt from radioactive events are detected by a photosensitive device.

196 Most geophysical ganma-ray detectors use a sodium iodide, thallium activated crystal as a scintillator. Radon (alpha) detection uses zinc sulfide.

The photoemission from the scintillator is usually chosen to be in the visible range. This is detected by a photomulti- ' plier tube (PM tube).

The PM tube has a photosensitive cathode (one which gives up photoelectrons upon impingement of light photons), a series of dynodes (a secondary electron emission surface) of succes- sively higher potential, and a collection anode. Figure 4.3 shows a PM tube diagram. Each photoelectron is accelerated through a field of 100 to 200 volts to the surface of the first dynode. There, it has enough energy to produce five or more secondary electrons. These are, in turn, accelerated to the

197 next dynode where a further multiplication takes place. Typical PM tubes in logging equipment will have 9 to 11 dynodes and will produce an amplification of 105 to 107.

The path lengths of individual electrons are not all the same, so the spread of transit time will lengthen the original pulse, slightly. The pulse widths at the anode of the EM tube can be from 250 to 500 nanoseconds long. They are, however, substantially shorter than those of even a bundle of G-M tubes. Thus, scintillation devices are capable of much greater counting rates than G-M tube systems.

Because of the higher counting rates of scintillation detectors, the statistical variation of the rate (the probable error) is lower. The size of the detector is much smaller than G-M detectors. Because of the high density of the scintillator material the efficiency is greater than for G-M detectors (10 percent to 15 percent). On the other hand, scintillation detectors are mechanically fragile, they have severe temperature limitations, they require stable power supplies, and the pulse outputs require amplification. In general, though, they are an improvement over most other detector types.

The light output of a scintillator is a function of the path length of the particle. Thus, the scintillator is sensitive to the particle energy. The energizing particle must be completely absorbed, however, for the measurement to be meaningful. Therefore, the scintillator must be as large as possible and as dense as possible.

In logging equipment scintillation detectors exhibit a pseudoplateau. If the exciting voltage of a photomultiplier is increased the counting rate will, at first increase. This is because the primary ganma rays and their small amplitude variations will be increasingly detected. When the voltage has been increased enough to bring the main body of pulses above the discriminator level further voltage increases will result in only counting rate increases. These additional counts will be due to the Compton scattering within the formation, producing gamma photons of lower energy. At lower gamria energies, however, more are absorbed by the steel housing of the logging tool and eventually none will get through to the detector. Further voltage increases will not result in substantial counting rate increases. This is the "plateau" of a scintillation detector. Further increases of PM voltage will bring the circuit noise above the discriminator level and there will be a sharp counting rate increase, independent of ambient radioactivity. The length of the pseudoplateau is a function of the system noise level.

198 Free charge carriers can be generated in solids by ionizing radiation in a manner analogous to gas ionization. Ioniz- ing particles, such as ganma-ray photons, will create hole/ electron pairs in solids. An electric field across the solid will cause the pair to migrate in the electrode directions. Semiconductive materials which will not trap the electrons or holes quickly are used. Reversed biased p-n junctions of semi- conductors are used in this manner. The semiconductor detectors are, in effect, solid state ionization chambers. Hie semiconductor detectors have extremely good energy resolution for gamma rays. Therefore, they are excellent for spectrographic work.

The resolution of semiconductor detectors is dominated by the noise of the semiconductor. The efficiency will depend upon the leakage current. Both of these factors are temperature dependent. Therefore, these detectors must be operated at cryogenic temperatures (some germanium detectors must be stored at cryogenic temperatures, also). The commonly used solid state detectors are silicon and two types of germanium (lithium drifted and intrinsic). The lithium drifted germanium detector must be stored at cryogenic temperatures.

Their use has been hindered by the need for low operating temperatures, however. Liquid nitrogen, when used as a coolant must evaporate in order to cool. This problem has been overcome by using a heat sink of melting propane. It makes a viable downhole spectrographic detector of ganma rays.

Germanium detectors are being operated experimentally in neutron activation tools and conmercially as a direct uranium tool.

4.1.2. Instrumentation and instrumentation effects

Radioactivity instrumentation is generally similar for most types of systems, Aether they are natural ganma-ray, scattered gamna-ray, neutron, or spectrograph. Therefore, a general discussion of radioactivity instrumentation is covered, rather than to repeat it for each type of use.

Detectors have already been examined. They usually have a pulse output, whether the rate or the amplitude is to be measured. Therefore, the associated circuitry is designed to handle pulses, rather than steady state signals.

Radioactivity circuitry falls into two major categories: analog and digital. There are, also combinations or hybrids. Each type of system has its advantages and disadvantages. In

199 spite of the newness of the full digital systems, they are extremely popular and their use is growing rapidly. Most units in use today are analog systans. So, both types will be covered.

Analog and digital systems are not interchangeable. In order to properly handle the data one must know which system or combination of systems was used. This is information which should be required on the log heading. It is not always apparent from the appearance of the log nor the description of the operator.

4.1.2.1. Preamplifer

The first stage of either the analog or the digital system may be a preamplifier. Not all detectors (i.e. G-M detectors) require preamplification. The purpose of this is to provide a low noise linear amplification for low level signals. It is usually found as close to the detector as it is possible to place it. It may have any amplification factor from one to more than 1000, depending upon the need. It may be considered part of the detector.

4.1.2.2. Discriminator

A discriminator is part of both types of systems. The main purpose of the discriminator is to eliminate circuit noise. In spectrographic circuitry the discriminator is used in multiplic- ity to separate channels of different pulse heights.

In all radioactive systems the setting of the discriminator level is critical. A discriminator operates by passing pulses of amplitude above a preset level. Hie level may be adjustable or it may be fixed by design.

When a discriminator is used with a G-M detector, set the operating voltage of the detector within the plateau and then set the discriminator level below the uniform pulse height. This can be judged by varying the discriminator level or detector voltage until changing it does not change the counting rate. The level should be set as high as possible without affecting the counting rate. Figures 4.4 and 4.5.

When the circuit uses a scintillation detector or a proportional counter, the setting of the discriminator level in a gross-count system can be very important. If a pulse height analyzer is not available, the pseudoplateau can be used. The discriminator level should be lowered (or detector voltage raised) until there is negligible'change in the counting rate. Then it should be reversed until just before the counting rate

200 FIG.4.4. Plateau setting.

FIG.4.5. Effect of a discriminator.

is affected. This should be performed with a moderate counting rate, not a low or "background" rate and with a wide range of gamma energies, not a monoenergetic source. A range of energies can be obtained by scattering the gamma rays in a bulk of material.

If the discriminator level is set too high in a gross-count system, the level is apt to be at a point where the counting rate can change rapidly with small power supply changes. This will invariably result in erratic sensitivity of the system. If it is set too low, noise will be admitted. In a density system the level must be below the peak of the single scattered gamna rays. With spectrograph^ systems of any kind, the initial discriminator setting is a design problem and should not be changed.

From the output of the discriminator, the components will depend upon whether the data handling is analog or digital.

The type detector and size, discriminator settings, and operating voltage will determine the counting rate for a given amount of radioactivity. Therefore, these things should be determined and set initially for any design of probe. Once they

201 are determined the probe or system must be calibrated, then they should be kept constant. If one setting is changed, a calibration must be performed afterward. A specified pulse height, with a specified monoenergetic source, at a fixed distance, should be obtained at the discriminator input. Amplification should be adjusted to obtain the correct, specified pulse height whenever the probe is checked. It should be checked regularly. Hie discriminator level must then be checked and, if necessary, reset after the input pulse height has been set. These steps will insure a minimum change of sensitivity.

4.1.2.3. Pulse shaper, analog

Following the discriminator in an analog system, or incorporated in it will usually be a pulse shaping system. If the tool is a gross-count tool, the pulses will be formed to a uniform shape.

4.1.2.4. Output amplifier, analog

There will finally be a power amplifier. The purpose of this is to put the pulses on the cable to the surface so there will be a minimum of distortion in the cable. This stage is usually the tool output in a simple analog circuit.

4.1.2.5. Derandomizer, Figure 4.6

One class of problems with radioactivity measuring systems is caused by the randomness of the occurrence of the radioactive events, and thus the electrical signal pulses, resulting probable error and dead-time losses. The cable will increase the dead-time of the system substantially. A short pulse (five microseconds or less long) put on 1000 metres of logging cable will have a dead-time of about 20 microseconds by the time it is taken off at the top.

One way to alleviate dead-time problems is to derandamize the signal pulses and make them periodic before putting them on the cable. Then, there will be no dead-time losses until the pulse rate reaches an easily determined rate of overlap. At this point the system ceases to transmit pulses completely.

There are several methods of accomplishing derandomizing. One of the better ones is to use a storage register to store the random pulses. Then they may be read out at a periodic rate. This type system usually is inserted before the final amplifier. It can result in real deadtimes as short as the design of the downhole equipment will permit.

202 FIG. 4.6. Comparison of random and derandomized pulses.

Some systems, such as count rate divider circuits only offer a small improvement in dead-time effects. They are, however, offered as improving or shortening the apparent dead-time and are somewhat effective, particularly when high counting rates are encountered.

4.1.2.6. Cable

The logging cable is, by far, the most detrimental component in an analog system. Unless there is a full derandcmizer, an analog system must be recalibrated any time a change is made to the cable. If, for example, a cable is 1000 metres long and 30 metres are cut off during reheading, there will be about a five percent decrease in system dead-time at the surface.

With an analog system, especially one without a downhole derandomizer, any time a tool is changed, or a cable is altered, or any surface change is made a calibration must be performed. In a simple analog system, the dead-time measurorient must include the whole system. Ihis is not true of a fully derandomized or a digital system.

4.1.2.7. Cable termination

There are some types of cable termination which will improve the dead-time characteristics of the cable. They will not be covered except to point out that the use of one should be noted

203 on the log heading. They should be used only with analog systems.

4.1.2.8. Pulse shaper and ratemeter

A straight-forward, simple analog systan will have a pulse shaper and ratemeter. The pulse shaper is to insure that all of the pulses are of uniform width and amplitude.

The pulse shaper is usually one of the major problem points with commercial equipment. The pulse width is usually adjustable. They frequently are not set properly, resulting in nonlinearities of the following ratemeter. Pulse widths out of the shaper must be less than the incoming pulse width and uniform. If the pulse width from the shaper is longer than the incoming pulse, pulses may be lost because of the deadtime of the trigger circuit. This circuit must be checked on any new equipment.

FIG.4.7. Ratemeter circuit.

The ratemeter circuit (figure 4.7) uses the uniform pulses frcm the pulse shaper and charges a capacitor, C,through a diode pump. Pulses arriving at the capacitor will charge it up. Thus, the voltage across the capacitor will increase to the maximum set by the pulse amplitude, V. The capacitor is discharged by a parallel resistor, R, at a time constant

T CR seconds where C is in farads and R in ohms. The voltage across the capacitor will then be a function of the rate at which pulses are put into the circuit. The voltage is read with a higjv impedance voltmeter or recorder.

Digital ratemeters are becoming more popular and probably represent a significant advancement over the analog ratemeter. They usually are an accumulating register which is read every tenth second or every second. This type ratemeter should be buffered so as to not lose counts while it is reading out. A multiple buffer is preferable but not often available. Digital

204 ratemeters usually have an analog output for operating a strip chart recorder and a digital output for operating a printer.

The smoothing of the irregular signal in an analog rataneter is determined by the time constant, x,of the ratemeter. Thus, the standard deviation a, if N is the counting rate,is

o ="\|2(RCN) = ^[2tN~

This is the same fractional standard deviation that would be given by a counter counting over a time, T=2RC. Increasing RC would reduce the standard deviation and make the recorded curve smoother, but the response would be more sluggish and important detail (resolution)'could be lost.

The time constant T puts severe restrictions upon the use of an analog ratemeter circuit for logging purposes. The smoothness (readability) of a log curve of a radioactivity measurement will depend upon the percentage standard deviation. The response time of the curve going across a bed boundary or through a deflection will also depend upon the value of T. Logging speed must be kept slow enough that the bed boundaries and the details within the anomaly will not be skewed. Skewing will take place with an analog system only. It will occur in the direction of logging. And it will be equal to L, L = T seconds x logging speed 60 seconds per minute This, of course places a severe restriction upon logging speed. It requires logging slowly through critical zones. The normal 6 metres per minute results in an upward displacement of 20 centimetres with a 2 second time constant. The normal recommended 1.5 metres per minute speed and one second time constant through an anomaly will give about 2.5 centimetres displacement.

Standard practice with an analog system should be to use as short a time constant setting as possible; certainly no more than two seconds and preferably one second. Limit logging speed to no more than 6 metres per. minute. When nearing an anomaly of any kind, positive or negative, slow down to one or two metres per minute. Reduce the time constant if the anomaly is positive or increase it if the anomaly is negative. At 1000 counts per second a T of 0.1 second is quite satisfactory. At 10 counts per second, the a will be ± 20 percent even with T = 5 seconds.

4.1i2.9. Digital readouts

At the output of the surface pulse amplifier, the pulse signal may be used to operate one or more digital systems.

205 These systems use the analog pulses from downhole and present a digital readout, recording, or data processing. These mixed systems are still analog systems.

4.1.2.10. Digital displays

In its simplest form, the digital section takes the pulses from downhole and counts them with a conmercial timer/sealer. Usually the count will be for one second at a time, so the readout will be in pulses per second or counts per second. This is then displayed on a digital display panel.

Note that the one second digital display is not the same thing as the ratemeter reading on a one second time constant. The ratemeter response is due 63 percent to the second just past; 24 percent to the one before that, 7 percent to the second one before, 4 percent to the third one before and so on. The digital counter, on the other hand examines one separate second only. Everything else is ignored. Thus, the counter reading of the digital readout only corresponds to the reading which contributes 63 percent to the ratemeter reading. The analog and counter readings will closely resemble each other, but will seldom be identical.

There is no lag or skewing of an anomaly introduced by the digital readout. It reads the total number of pulses counted during the second displayed. Therefore, a peak reading of the digital system may or may not coincide with the peak reading of the analog system, depending upon the logging speed.

The less expensive counting devices cannot count while dis- playing a count. All of the data available during display is lost. A better counter has a buffer memory which allows the system to display and count at the same time. The extra cost is well spent. A digital printer and/or a tape (magnetic or perforated tape) recorder can record the output of the counter. This will record the count every second, along with the depth (from a depth encoder). Or, the printout and recording can be triggered by the depth encoder. When the counter output is triggered by the depth encoder, some convenient interval, such as ten centimetres, is picked for the readout interval. This allows mineral grade calculations to be done more easily. With this type system most of the data are discarded, even when the counter has a buffer memory. This is because logging a half-foot of hole at five feet per minute requires six seconds. Thus, five out of six readings or 83 percent of the data are discarded.

206 Some conmercial circuitry incorporates interrogation or scanning type digital systems. This allows the digital recorder to accept, in sequence, the outputs of a number of panels. These can be depth encoders, other counters, digital ratemeters, analog to digital converters, etc. Thus, the whole logged output may be recorded. Recording of the data on a tape allows it to be played back at any time m the future. This has many advantages which will be covered in the chapter on Computer Handling.

4.1.2.11. Microprocessors and computers

A full digital data processing system may also be used. This can be either (or both) a microprocessor system or a computer system. Some of the functions of these systems can be handled on-site or later at the base (or both). It also allows the presentation of log data in terms directly useable in the field, such as percent uranium, barrels per square metre, percent porosity, and any other useful parameter which can be put into equation or table form. One must ascertain, however, exactly what processing has been done to the data. Insist that the contractor explain exactly in writing and note in the log heading what processing has been done.

The major differences between the microprocessor approach and the digital computer approach are the amount of memory available and the fact that the microprocessor system is usually dedicated to a single set of procedures. These differences are disappearing quite rapidly.

4.1.2.12. Digital data processing

A digital data processing system is an extension of the digital readout and recording systems just described. These systems use the information m real time or from the recorded tape to process the data to a more useable form.- The most common system of this type, at this time, is a microprocessor which accepts the gross-count gaimia-ray counting rate each interval and the depth readout and performs the following calculation for percent uranium grade, G:

G = 2KNF where K is the system conversion factor, N is the dead-time corrected counting rate, and F if the borehole correction factor (water factor, casing factor, etc.).

The value of G in the above is influenced by the radiation from 0.5 metre in every direction and is the average for 1 metre.

207 It is also possible to use a computer at the output of the cable. The computer accepts the signals from downhole and records them in raw form on magnetic tape (or sane similar device). Then the computer reads the recorded data and processes it to finished form for a printout, a log, or some other finished form. The taped data are retained for further use. The amount of processing and the final form of the data are only limited by the capabilities of the computer and the programmer.

These are fine systems. But one must remember two things:

1. the raw data must be retained, and 2. the actual processing method is important and must be known.

4.1.2.13. Full digital systems The sensor signal can be digitized immediately at the output of the sensor. The digital signals can be handled by digital circuitry downhole and encoded for transmission to the surface. This has many advantages. Digital signals are of the on-or-off form, with a large difference between the two states. A great amount of signal attenuation and distortion can be tolerated without destroying the information content of the signal. Thus, distorting components, such as a logging cable, have little or no effect upon the signal content. Sampling methods and encoded transmissions can increase the effective information density. Digital systems allow the use of other averaging systems. Be- sides the data for any point we have a complete recording of data before and after the point. Thus, we have available a whole range of symmetrical processes for averaging and otherwise handling data which do not distort the data and which are not easily available to analog processes. Finally, digital systems handle data in a form which does not have to be converted for computer handling, and usually when the data is faulty it is obvious.

4.2. Advanced sys terns

Throughout this text the mechanics of data processing have largely been ignored. This was done purposely to emphasize the importance of the mechanism in the formation and that in the detector system. The type of data handling and data processing also have an effect upon the finished information.

4.2.1. Analog systems

The original logging systems were all analog systems. The term analog comes from the fact that these systems use electrical signals which correspond proportionally to or are analogs of

208 the parameter being measured. The SP signal, for example, is a direct current analog whose amplitude varies directly as do the potentials at the electrode. The signal at any place in the system is the analog of the phenomenon being measured. This type system has many advantages. It is simple in concept. A person who knows little about information theory can design and use most analog logging systems. This system usually requires few active components. Analog systems have the big advantage that they are usually straightforward in concept. They have some disadvantages, too. They must be designed, built, main- tained, and used with great care and precision. Any small error or nonlinearity anywhere in the system will appear in the output. Often this is not noticed because it may be of the same order of magnitude as the signal itself. Components can change gradually with time, temperature, pressure, moisture and cause corresponding changes in the output. Many elaborate "feedback" schemes have been devised to .help overcome these gradual drifts. Even so, we have millions of feet of bad logs because the systems was "out of calibration" and the observers were not aware of it. Analog systems usually handle signals continuously. Each analog processing system (in general) must be dedicated to one signal. The dynamic range over which the circuits can vary is quite limited, usually 1000 to 1. Occasionally, in an excep- tional system, it may be 10,000 to 1 or more, but the cost of this extension of range is great. The drawbacks of analog systems are especially apparent with radioactivity measuring circuits. Averaging is done only on the basis of past time. Sampling adjustments (time constant, range, etc.) are usually left to the discretion of the operator. He may be poorly trained or simply lack enough data to make a viable choice. Design variations have evolved to successfully overcome many of these problems. Multiplexing of signals on a time shared basis and other methods have been quite successful.

Bandwidth is the range of the frequencies a system can handle. , It is a direct measure of the amount of data which can be transmitted. Bandwidth restricts and distorts every analog signal directly. A device, like a cable, does not simply transmit up to a certain frequency and then quit. The higher the frequency is, the more the signal is attenuated. And, pulse type signals contain many high frequency components.

Modern analog systems avoid some of the random (radioactive) signal problems by derandomizing the signal. Suppose we have a system which will transmit 50,000 pulses per second with a reasonable amount of design effort (a 50 kHz bandwidth). With a straightforward periodic pulse system (one where the pulses are uniformly separated) 50,000 pulses per second could be transmitted. Any higher frequency would be severely attenuated. But,

209 ONE CHANNEL OF AN ADVANCED ANALOG SYSTEM, WITH DIGITAL DATA HANDLING AND PROCESSING

FIG.4.8. Block diagrams of typical analog circuits. one channel of 50 kHz could be transmitted, or two of 25 kHz, ten of 5 kHz, etc. If the signal were a random one, as it would be with an unprocessed gamma-ray system, the limit would be 16,667 pulses per second, if a 50 percent dead-time correction is used as the upper limit of favorable probability.

A typical analog circuit is shown in block form in figure 4.8.

4.2.2. Digital systems

Digital systems handle the signal with combinations of pulses which represent samples of the full signal. If an adequate number of samples is taken, a small number of sample pulses can be used to represent and recreate the full signal. The smaller number of pulses is easier to transmit.

If a digital encoding is used; for example, a BCD 1248 code, and sampling is done each tenth of a second, then rates of 999,999 pulses per second, each, are theoretically possible. In actual practice we could realize something less than this, but still, quite substantial.

A code, like the BCD 1248 code represents a number by pulses separated in time or space or both. BCD stands for binary- coded-decimal. BCD 1248 indicates that there are four channels of binary (on-off) information combined to represent a decimal number. A pulse on channel 1 has a value of one; a pulse on 2 has a value of two; a pulse on channel 3 has a value of four; and a pulse on channel 4 has a value of eight. The sum of 1 plus 2 plus 4 plus 8 is 15. If we consider no pulses on a channel as zero, then there are 16 combinations of pulses and channels; more than enough to represent any decimal number. Thus, if any number is picked at random, it can be represented by a BCD 1248 code. The number 176,295, for example, in BCD 1248 would be 1 7 6, 2 9 5

channel 1 1 1 1 1

channel 2 2 2 2

channel 3 4 4 4

channel 4 8 1 7 6 2 9 5

The first pulse on channel 2 represents 100,000. The second group of pulses on channels 1,2, and 3 represent 70,000. The

211 ONE CHANNEL OF A HYBRID LOGGING SYSTEM, ANALOG DOWNHOLE AND COMPUTER DATA HANDLING

MULTICHANNEL ANALOG SYSTEM

AN EXAMPLE OF A FULL DIGITAL DATA HANDLING, DATA PROCESSING SYSTEM

FIG.4.9. Block diagrams of typical advanced analog, hybrid, and digital systems. third group of pulses on channels 2 and 3 represent 6,000 and so on. By sending a sequence of pulses on 4 channels the number 176,295 can be represented unambiguously with 11 pulses. In- stead of 176,295 pulses only 11 are needed. The trade-off is in time, of course. The signal on an analog system is continuously representing 176,295. On the digital system of our example that number is read only every l/10th of a second. If the analog signal accidently drops in amplitude 10 percent then it will erroneously read 158,665. If the digital signal drops by 10 percent it still represents 176,295. As long as the digital signal can be detected it is interpreted as the correct number.

Digital systems have disadvantages. The circuitry is complex. Even a simple digital system contains many active components. The old vacuum tube gamma-ray probes had seven active components (vacuum tubes). In a digital probe one chip (of many) has 10,000 active components (transistors) in it. The signal at any place in the circuit must be decoded to determine if it is of the correct form, so it is almost impossible for a technician to repair the system in the field.

Digital systems can use other averaging systems than time. Data points are known and available symmetrically about the measure point. (This, of course could be accomplished with a suitable analog recording.) Distance averaging can be used and higher logging speeds do not distort the signatures of the logs as severely as they do in analog systems. A big advantage to a digital system is that the information is in a form easily used by a computer. Digital field systans are often designed to re- lieve the operator of routine duties where the probability of making an error is high. They keep track of scales, zeros, and changes. They denand calibrations, they remind operators of options, and they force correct procedures. On-site computers can present data in meaningful parameters, instead of arbitrary ones which must be converted. They also preserve data in unprocessed form to allow further or later processing.

A typical digital systan is shown in figure 4.9.

In general, analog systems are good for portable use, experimental use, and for single measurements. Digital systems are ideal for production and other high volume work. Hybrid systems (combinations of analog and digital systems) and digital systems are effective for permanent records. Digital systems with on- site computers at this time are probably the best commercial logging systems ever designed.

213 4.3. Radioactivity statistics

Radioactive events are completely random events. Thus, we must handle the data and the analysis of radioactive logs with statistical mathematics.

Since the occurrence of events is random, we will not be able to arbitrarily pick a reading, N,and expect it to be the unchanging value representative of that steady state sequence of events. The next one we pick, even though nothing has been changed except time, will be different. We may, however, sample n number of times, and take the mean, m. The more samples we take (the larger n is) the more confidence we can have in the value of m. The mean is

m = (Z Nx )/n

If we take one particular rate reading, N, it will not be equal to m because of the randcm variation of the signal. Ap- proximately two-thirds of the time the value of N will fall between the limits,

m = m i a or, expressed as a percent

J -2 + (mi ) x 100 w „

(2m T)

Due to the random nature of the signal pulses there is always a finite chance that one pulse may occur so close to the preceding one that they cannot be separated by the circuit. The length of time during which a succeeding pulse cannot be detected is called the system dead-time, u. The longer the dead time is, the more likely it will be that there will be a loss of pulses. The dead-time is determined by one or more of several methods (see references).

If the real or correct counting rate is N and the recorded counting rate is n then the corrected counting rate is

N = n / 1 - yn and n = N / 1 + yN

It is evident from these relations that it is desirable to have as high a counting rate as possible to reduce the percent

214 variance. Also it is desirable to have y as short as possible in order to make as small a correction as possible.

The dead-time correction is a probability correction. That is, within the limits, the probability of N being correct is greater than it being incorrect. The upper limit on the expression is about 50 percent correction.

215 5. FIELD PROCEDURES

5.1. Mobilization

5.1.1. General

Obviously, a description of details of mobilization is difficult, since each situation has many unique features. There are, however, some good practices which are corrmon to all situations.

5.1.2. Planning

Probably the foremost item in mobilization is good planning. One must decide well in advance what information he will log for. Probably a good method is to list the parameters one wishes to know, then list the ways of obtaining these parameters. And, finally, eliminate those which are impractical or unnecessary. Table 5.1 shows a list of logging methods with some of the uses of these methods. The type of vehicle must be chosen with care. If a contractor is used, this may be chosen for you. However, your opinion is a major factor even with a contractor.

The type of vehicle must be chosen to fit the job, or terrain, or both. If one were going to have to scale the moun- tainside, he would not choose, for example, a 2 wheel-drive highway vehicle. The obvious choice, in this case, is a light, simple, backpacked unit. Many of the choices are not so obvious. Table 5.2 lists some of the types of equipment available and the types terrain where they are used. Equally important is choosing the proper type of logging equipment. One would not choose a simple, single pen analog recording system when trying to determine formation elastic moduli, for example. He would choose a full-wave analog recording or a digital sampling system. In general, a digital handling and processing system is better, more complete and more flexible than an analog system. However, digital systems are not always available (especially at this writing). A backpack system has a definite place and function. But, it will not take the place of a full logging unit.

Plan for emergencies. Consider a "worst-case" situation and plan for it. Usually, the better planned a project is the smoother and more trouble-free it will proceed.

5.1.3. Transport

Once the type equipment has been chosen, the method of transportation must be considered. This, of course, will be part of

216 TABLE 5.1. TYPES OF LOGS AVAILABLE

LOG TYPE MEASUREMENT USES Natural Gamna- Total radiation of Sand/Shale Ray, Gross-Count gamma rays. sequences, (G.C.G.R.). effective mineral grade, correction lithology.

Spontaneous Imbalances of ion Sand/Shale, Potential (SP). concentrations in permeability, fluids. relative or absolute salinity, total dissolved solids.

Single Point Relative resistance Sand/Shale Resistance. of formations; sensitive approx. lith. to formation fluids. stratigraphy, correlation.

Focussed Absolute resistivity Lithology, Resistivity of formations; deep Rt stratigraphy, penetration pius Rw, porosity. symmetrical detailed response.

Neutron Porosity. Neutron moderation, Lithology, capture porosity, cross-section of Resistivity thermal neutrons. substitute.

Delayed or Fission neutrons. Direct Prompt Fission measurement of Neutrons. uranium content.

Density. Scattered ganma Formation bulk ray; electron dens i ty,poros i ty, density. lithology.

Deviation. Gravitational, magnetic Hole Deviation, field strengths Bed location, and directions. depth &thickness.

High Resolution Focussed resistivity Determine Dipmeter. 4 meas. at 90°. Bed Dip and Fracturing Geol. history Gamna-Ray Characteristic energy Determine U Spectrograph. peaks of ganma thorium ratio, radiation. eval. K content depo. history quant, uranium.

217 TABLE 5.2. VEHICLES AVAILABLE FOR LOGGING

Vehicle Terrain Conments Type

Passenger Hard surface Low clearance. automobile, roads, improved Suspension not suitable standard, dirt roads. for a rough surface. Gets rear-wheel stuck easily because it drive. has 1 drive i^ieel. Limited space.

Same, but Same terrain. Low road clearance. with a Suspension not suitable limited slip for a rough surface. rear-end or Handles better in mud, front-wheel snow, ice, and sand. drive. Limited space.

Same, Hard surface, Better road clearance, 4-wheel improved and and suspension. drive unimproved roads. Useful under bad conditions.

Pick-up Hard surface Good road clearance. Get truck or roads. Improved stuck easily and handle carry-all, dirt roads. poorly on ice because of standard. Unimporved dirt light load on drive roads. wheels. Good space features and good load capacity.

Pick-up Improved or Good road clearance. truck or unimproved roads Handle well in rough carry-all and rough terrain, esp. with power 4-wheel surface. steering. Use weight in drive the pick-up. Good capacity

Truck, Improved or Extra drive wheels and larger. unimproved, load capacity improve rough surfaces. handling in difficult terrain.

"Swamp- Soft surfaces. Excellent in swamps, bogs, buggy" sand and snow. Special order. Difficult on hard roads.

Snow- Medium to hard Personal transportation. mobiles snow.

218 Vehicle Terrain Conments Type

Snow Soft to hard Room for equipment and machines. snow. shelter.

Helicopters. Any. Sensitive to wind. Space limited.

Fixed-wing Any. Need suitable landing aircraft. space.

Trailers. Improved roads. Can provide room for Must be pulled. instruments and shelter, can be left on-site.

Air- Smooth surface. Potentially very useful. cushion Hard or soft. Not suitable for dusty or vehicles. dry, soft sand locations.

Tracked Rough, hard or Can cause surface damage Vehicles. soft surfaces. by providing access to poor terrain.

Hand truck Rough surfaces. Should be considered. or cart. Must be pulled.

Back-pack. Can be taken Space limited. Can be anywhere a man tiring. can go.

Honda Rough, soft Personal transportation. tricycle. terrain.

Motorcycle Smooth to medium Personal transportation. hard terrain.

Boats. Water. Very versatile.

Air boats. Shallow water, Versatile. Space limited. swamps.

219 the choice of equipment in many cases. The method of transport tation, however, includes packing and protection of the equipment. Probably more instrument failures originate with damage during transport than with any other single cause. Shock loads in a tool-rack in a truck, for example, can easily reach lOOOg (gravity), if the tool is not clamped tightly to the tool-rack, and if the tool-rack is not shock-mounted.

Small units, such as portable gear, backpack units, or modules must be cushioned while being carried in an automobile or truck. These units are usually built assuming ligjht g loading and can be damaged by bouncing in a truck or sliding in the trunk of an automobile. Truck-mounted gear must also be packed properly. These units are designed to be attached rigidly to the truck while in transport. Methods must be supplied for fastening modules in cabinets and tools in racks so they cannot move.

Skid mounted equipment for transport by truck, boat, or aircraft must be well secured. Not only must the equipment be tightly fastened to the unit, but the unit must be completely immobilized in the vehicle. Most equipment damage originates from the intense shock of a unit bouncing against a non-resilient surface. A steel sonde, dropped 15 cm onto a concrete surface, will undergo a lOOOg acceleration (shock).

5.2. Checking

Before any packing for transport can be done, the equipment must be thoroughly checked. It is useless to take logging equipment to the field if it will not work.

5.2.1. Batteries

Batteries should be checked and replaced and charged (if they are rechargeable). If the unit has lead-acid cells, they should be left on a trickle charge while they are not being transported or used. Nickel-cadmium cells should be periodically charged. Non-rechargeable cells, such as carbon-zinc cells, should be left out of the unit until it is used. They should be removed immediately after use. It is probably wise to store and carry any type battery separately from the unit to. prevent spillage and corrosion.

Check power supplies of generator or line-powered equipment. Make sure voltages are correct and frequencies are proper and stable. Service the engine of an engine-driven power plant.

220 An operational check, including calibrations, should be performed on all equipment. A calibration check is especially important with any analog equipment. Analog system sensitivities can change with time. The system may operate, but not properly. This is true of digital equipment also, but is not as likely to happen.

All parts should be assembled in one area to make sure everything is on hand. Spare parts should be included. Chart paper and O-rings are difficult to find in the middle of a desert. Pens and ink are important. Make a list of equipment, supplies, and parts, and check it.

5.2.2. Packing

Pack and store all parts with final use in mind. Arrange pieces so they may be unpacked as needed; so they may remain packed if they are not needed.

Weather covers are very important for any except enclosed truck-mounted systems. They must be complete and weather-proof and tied firmly in place.

Keep temperature extremes in mind. Liquids can freeze. Bottles and cans can bulge or rupture from pressure in the hot sunlight. Engine oil should suit the climate.

Pack equipment in their boxes and in the vehicle so they will not be damaged or scattered during transit. Also, pack them so it will be difficult to pilfer them.

5.2.3. En route

Pick roads and/or routes to ease the problem of transport. The equipment you have is expensive and delicate. It is of no use to anyone if it is inoperative. Most damage to logging equipment occurs en route. Hold down the speed on rough roads. Do not drop, backpacks during breaks. Keep all equipment clean and dry. Some equipment can be damaged by heat, direct sunlight or cold. Try to foresee these problems.

Keep land damage to a minimum. Most of the land you will cross and work cti belongs to someone else. Treat it with respect. If land is torn up or if trash is scattered around, someone else will have to undo it and you will be blamed.

Give yourself plenty of time to travel and arrive ahead of time.

221 5.3. Arrival at location

5.3.1. Site inspection

Finally, when arriving at the location, the site should be inspected carefully. Determine ahead of time where the unit will be set up. Levelness and protection from the sun, wind, and weather can be very important. It is also desirable to avoid the mud as much as possible for personal comfort and cleanliness, and for trouble-free operation. It is wise, too, at this point, to check with the Project Geologist and the driller to determine any unusual occurrence, such as drilling problems. In a multi-hole location find out if you are in the right place. Further, it is wise person- ally, technically, and politically to let people know you are ready, on time. Find out if the Project Manager has any specific requirements or measurements to be made.

Get information for the log and log heading. Log headings should be filled out as completely as possible now, to save time later. The information to be included on the log heading is

1. Hole identification (number) and owner 2. Location of hole 3. Date 4. Logging operator's name and truck number 5. Curve types run on this log 6. Sensor types and sizes 7. Tool serial numbers, types and diameters 8. Surface equipment serial number and types (serial numbers are not important for digital equipment nor advanced analog equipment) 9. Source types, strengths serial numbers 10. Spacings 11. Calibration readings, date of calibrate and calibration model location 12. Mud type, resistivity and top of fluid in the hole 13. Scales used (depth and deflection) 14. Zero positions (if possible, record these on the log) 15. Bit sizes and where each size was used, if there was more than one size 16. Depth drilled 17. Maximum depth logged (this may be on the log) 18. k-factors 19 Deadtimes 20. Water or hole factors 21. Time constants or averaging time 22. Other services run.

222 5.3.2. Prelogging check All of the gear should be assembled for the first log. Again, an operational check should be made, in detail. This is the time to make prelogging field calibrations. Pens should be set and zeroed with analog systems; digital systems should be brought alive and programs loaded. Recorders should be checked for sufficient paper, film, or tape. Power plants should be running smoothly; voltages and frequencies should be normal and steady. Hydraulic systems and brakes should be checked and operational. If not, now is the time to make corrections.

You must be ready to log when the drilling is complete.

5.4. Logging The logging of the hole should be almost an anticlimax if preparations have been proper. Since the equipment is calibrated and is operating properly, no further time need be spent on that. The logging unit should be moved into the preselected location and the boom or sheave-wheel assembly should be put into place.

5.4.1. Positioning

The sheave assembly or truck boom should be positioned so that the probe and cable will be against one wall of the hole. Hie probe should not be centered in the hole, unless central- izers are used on the probe.

Put out the surface electrode or electrodes, if appropriate. An SP return electrode should go into an earthen mud-pit but not into a metal one. If the mud-pit is not available, the electrode should be put in the mud or cuttings near the hole. As a last resort, the electrode may be put into a shallow depression near the borehole and water poured over it. This, however, is a poor last choice.

Set the engine speed if a hydraulic system is being used and lower the probe until the zero point on the tool is at ground level. Then, set the depth indicators to zero. If it is difficult to set the zero level, it may be easier to put the tool zero at a known height above ground. In this case, the depth measuring system should be set at a minus value to reach zero at ground level.

Check the tool operation again, and turn on the recorder without recording or other indicator so the progress may be monitored going downhole. In a difficult location it may be

223 good practice to make a recording while going down, in case of later problems preventing a log.

5.4.2. Downhole trip

Start downhole at a moderate speed (i.e. 30 to 40 m per minute) and monitor the curves frequently. Also, monitor the depth indicator, the line-speed indicator, the line-tension indicator, and particularly, the resistance or resistivity curve. This last is a good indicator of probe movement. If the probe is moving downhole and not stopped, the resistance curve or fine resolution resistivity curve almost never stops moving. If the resistivity stops moving, check immediately for a stopped probe. Frequently monitor the line on the winch drum. If the tool stops moving and the winch continues, serious snarls and damaged cable can result. At best, it will result in lost time. Par- ticularly note any radioactive anomalies on a uranium log. Note the depth and deflection amounts. This information will be needed while logging. On the way down, with analog equipment,, decide what scales would be best and set the sesitivity. This is not necessary with computer or full digital equipment.

While on the way down with an analog system, check the pen depth settings on zeroes. Each pen must be adjusted to a scale distance from the zero or measure point which corresponds to the relative sensor positions on the sonde. Be aware too, that these settings change with different depth scales. Also, be aware that many things, such as starting the ink flow or setting the pens, may bend them out of position. Zeroes should be checked by pressing the zero switch or turning off the probe power. These procedures are not needed on a computer system.

When you are near the depth to which the hole was drilled (15 metres above, or so), slow the probe's descent.

5.4.3. Bridges

If the probe is stopped on the way down, move the probe up and down with the winch. Never put such slack cable into the hole. At shallow depths (less than 50 m) do not allow more than 1 metre of slack cable. If there is too much slack, the cable on the winch will become stationary and will snarl. If the probe stops at a depth greater than about 50 metres, no more than 2 metres should be allowed. If the probe drops through with too much slack, the weak point may be broken and the sonde lost. Allow the probe to stand on the obstruction with a little slack cable. Watch the resistivity reading. There is a possibility the probe may be moving slowly. If it is, let it move, allowing it about a metre of slack at a time.

224 If the probe is not moving, note the depth at which it has stopped by taking ail slack out of the cable. Then lift the probe 3 or 4 metres and let it down at about 15 to 20 metres per minute. If the probe stops at a greater depth, repeat the process.

If the probe stops at the same depth, put about 1/2 metre of slack in the line and lock the winch. Then move the probe manually up and down about 1/3 metre at a time to try to liquefy any mudcake of the "bridge".

Usually, these procedures will get through the obstruction. If they do not, after 20 or 30 minutes of trying, call the drill rig back. Do not lift the probe and let it drop free, or "spud". This damages probes. The probe is not a drilling tool! Also, if the tool does break through, the weak point in the cable head may easily be broken, with a resulting lost probe. Rig time is cheaper than a damaged or lost probe and a "fishing" job. Note the depths of the obstructions or bridges for caution on the way up if you do get througjh, or for the driller's information if you do not.

5.4.4. Logging out

When the probe touches the bottom of the hole, there will be a number of indications. Theengine/tirawworks will change sound, the weight indicator will change readings, the resistance or resistivity curve will stop moving (this is probably the best indication), and the logging cable may become slack. (The slack will be difficult to detect in a deep hole.)

Upon touching bottom, stop the winch immediately and lift the probe off bottom. Note the depth at which the probe lifts off and mark it on the log.

The bottom of the hole will almost never be at the depth to where the hole was drilled. There are several reasons for this:

1. driller's measurements are coarser than logging depth measurements, 2. errors will appear in both the driller's and logger's depth measurement, 3. holes will fill in because of settling of cuttings.

The total loss of hole, however, should seldom exceed 1 or 2 metres. If it is more, consistently, ask the driller to circu- late longer.

225 Many drilling contracts use the logged depth of the hole as the depth of hole for payment. Therefore, there are more reasons than just technical for having good depth measurements.

Leave the probe 3 to 4 metres off the bottom while final adjustments and pen settings are made. This virtually eliminates the possibility of getting stuck at the bottom.

Logging is usually done coming out of the hole. There are several reasons for this:

1. hole conditions, hazards, and anomalies can be noted on the way in, 2. tool operation and scales can be noted on the way in, 3. most important, the tight cable coming out insures reliable depth measurements.

Set the pens at the correct position on an analog system. This means that the electrical zero must be checked for each circuit. It also means that the pens must be positioned according to the positions of the sensors on the probe and the depth scale used. The pens must be separated the same number of scale units apart that the sensors are separated in actual units. The pens must be checked, since it is easy to bend them out of position.

Lower the probe to the bottom, with about 1 metre of slack. Immediately turn on all systems and start, logging, coming out. Make sure the pens are writing. Check the logging speed. Be sure the paper is moving. Make sure the sonde is moving.

If a digital system is being used, the foregoing procedures may or may not be required, depending upon whether or not real- time logging is being done and upon the type of recorder being used. Usually, a verification of operation is all that is needed.

Upon nearing an anomaly with an anolog system, slow down to about 1.5 to 2.0 metres per minute. It is important to get the proper shape of the entry and exit curves, even on the primary trace. The amount of distortion will depend directly upon the time constant and logging speed. Neither of these is a factor with digital systems.

With an analog system, log completely through the anomaly with about a metre or two of extra trace at background level. Then, shut off the pens and drop the probe back to about 2 metres below the anomaly. Change scales so the rerun trace will cover about two thirds of the span, turn on the pens, and log

226 through the anomaly again, at a logging speed no more than 1.5 metres per minute and at as short a time constant as possible. Note both the scale and time constant on the log. These do not apply to a digital system.

When completely through the anomaly, record about 2 metres additional. Then record the zero for 1 metre. Note directly on the log all scales, scale changes, bias changes, shifts, hole problems, logging speed changes, and comments. With a full digital system, the only notes need be conments, as the computer usually handles the rest.

Near the surface, slow the logging speed and stop the winch at the surface at the original zero point. If there is a depth discrepancy at the zero, note it on the log. If there is time, determine the cause of any errors or problems and relog the hole.

5.5. Comnunications

5.5.1. Ins tructions

Communications can take many forms on an exploration or development project. One will get instructions, both written and verbal. He needs to keep in touch with his base, especially in wild country, and needs to confer with the others on the project.

When instructions are received, they will probably be for the type logs needed, locations of drill rigs, and instructions for getting to the location. Write down these instructions and, if possible, read them back to the sender. Be sure you under- stand them.

It is vital to keep in communication with the Project Geo- logist and the drillers. It is a prime necessity for efficient work. This is easily done by 2-way radio. But, lacking a radio, a logger should check often to see if there have been changes or additions.

If possible, use a radio system. These can save time and money. Remember though, if the radio is occupied by idle chatter it is available neither for business nor emergencies.

Finally, always keep in touch with your base. Let than and the Project Geologist know where you are, and when you will be through. Not only can this eliminate unnecessary trips, but much of the land being explored, even in the U.S., is wild and is dangerous. And, you will often be alone.

227 It is important to set up communications procedures in advance, whether they are simple or complex. If your procedures are known, any disruption will result in someone checking. This may save your life.

5.6. Auxiliary equipment

Be sure you have the equipment to do the job. This means it must operate properly.

Power is important. It operates your equipment, gets you to and from the location, in and out of the hole, it keeps you warm or cool, and it makes communication possible. It is the most important feature, after the instruments themselves.

Tools are required. These can mean the difference between getting a job done and not. Tools should include the elementary hand tools to make emergency repairs. They include items to get out of trouble, such as shovels, tow chains, and tire chains. If the truck has a winch, make sure it works and is not tangled.

Backpack units should also include tools. They will not be as elaborate as for a truck unit, but they are as important.

Double check the reliability of your transportation. This is as important for a pair of shoes as for a helicopter. Vehicle batteries, water, oil, and fuel should be checked. Air- craft should be inspected. Tracked vehicles are especially vulnerable. Carry spares (i.e. tires, oil, fan belts) for any kind of vehicle. It is not a bad idea to have a spare pair of boots in the vehicle. Be sure you have emergency equipment. Put together a kit containing dried or canned food, compass, saw, matches, clothes, and a knife. It may be reasonable to carry a small pistol or a rifle in wild country. Put this in a bag and carry it with you. However, be sure to check with the land owner and for legal prohibitions.

5.7. Information

5.7.1. General

There are many items of information required for a trouble- free operation. Also, the purpose of logging is to gather information. Therefore, this is an important topic, even if some items are repeated.

228 5.7.2. Preliminary

Since many or most locations are remote from the home base and often from sources of supply, good planning involves ' obtaining good information. Make notes. Do not trust your memory.

5.7.3. Directions and instructions

Obtain detailed directions to the location, unless you have been there many times before. It is wise to record the instructions as they are given. It is also wise to check the location on a map.

Ask if there are any road problems; bridges out, mud, snow, sand, water, washes, construction. Locate sources of help in case of trouble. Ask about terrain problems at the location. Check the weather. Ask about the need for special tools or vehicles. Find out if there are any problems with local people, locked gates, closed roads, animals, or authorities. Find out whom to contact at the location. Ask about hole depths and the number of holes to be logged. Determine if there are any land problems. Check that you have permission for access. Find out which logs are to be run and when they expect you.

5.7.4. En route

Check maps and landmarks en route. This will help to prevent getting lost. Also, check the time. Determine if you are on schedule. If you are much behind schedule, it is not only an inconvenience, but people, may be searching for you. Allow plenty of time.

5.7.5. At the location

At the location get all pertinent information and record it immediately. This includes many items: 1. depth now 2. target depth 3. anomaly location 4. estimated time to and at target depth 5. hole size, bit size 6. hole problems 7. what.is desired in the logs 8. driller's name 9. geologist's name 10. hole number 11. where will the Project Geologist be

229 12. where will the drilling crew go next 13. where is the next hole 14. what is the geology 15. are there nearby logs.

Examine the cores and cuttings for normal and unusual features. Examine the mud. Check the drill bits for size. If the hole is cased check the casing thickness. If you are to log through the drill-pipe, find out the size (inside and outside diameter). Make notes of all of this. Much of it should go on the log heading.

5.8. Leaving

Fill in the log heading and attach it to the log. This heading is as important as the log itself. Do not hesitate to put data and comments on the heading.

Leave the logs with an authorized person only. This is valuable information and must not be available to the wrong people.

Plug the borehole, unless you know it is to be left open. You are usually the last one to use the hole. Plugging it will prevent a person or an animal getting hurt. It will also prevent erosion. Police the area. Pick up trash and throw it into a container for that purpose, or into the mud pit. Leave so as to minimize land damage.

Notify someone you are leaving. Check with the driller and/ or the geologist, if someone is still on location.

Lock all gates and check and note the time. If you are expected at a particular time and are behind schedule, let someone know so they will not hunt for you.

230 6. INTERPRETATION

6.1. The exploration philosophy

The geophysicist or geologist whose job is to interpret the borehole logs for a particular area or project must have previously: 1. an exploration theory for uranium mineralization which does or may apply to that location; 2. an idea of the rock characteristics within which it will be found; and 3. a fair estimate of the depth where the mineralization is likely to be encountered. To obtain the maximum information from the logs, in other words, the interpreter must be familiar with the regional geology and likely structural conditions, and be aware of the intrinsic geophysical responses of those formations.

Generally, borehole logging is carried out after prior surface geological mapping, radiometric surveys, geochemical surveys, perhaps aerial and ground radiometric and electrical surveys, in many cases surface trenching to provide better rock exposure. From these various sources the log interpreter should be able to piece together a reasonable idea of the underlying rock characteristics. At this point a few stratigraphic holes may be desired for verification. However, they generally are not needed, unless the area is very poorly known.

The primary intention of borehole logging in this book is for locating uranium. Whether a hole has been cored, or drilled by the open hole method, a geological interpretation can often be made of the core or chips to give the geologist a reasonable idea of the lithology and mineralization. However, if the hole exists, it should be logged at least to verify core or cuttings (chips) data. Working together with the geologist, the log interpreter can assign to the various geologic formations representative background levels of radiation encountered, as well as the character of responses from the spontaneous potential and resistivity portions of the log and verify the assumed geological sequences. Unfortunately, interpretation of borehole logs often is done in isolation, with neither the geologist or interpreter interfacing with each other. The most worthwhile results can only be obtained by a truly co-operative effort using ALL the information available at that time and in close cooperation with other professional colleagues.

Generally uranium mineralization in a drill hole has what is termed a "signature"; a characteristic log response. By

231 putting together the gamma-ray, spontaneous potential, neutron- neutron and resistivity logs, one can determine a signature of that particular anomaly. It then follows that when drilling adjacent holes, these signatures or logical extrapolations of than are looked for. This is correlation between drill holes. It is rare to encounter a totally featureless formation. There- fore, with a number of adjacent drill holes, these signatures can, and should be, traceable from one hole to another. Where there may be large vertical changes in location of matching signatures in adjacent holes, one should suspect the presence of features such as faults, unconformities or others.

There is nothing magical about interpretation. Hard work and experience gradually develop a close feel between the response levels found in borehole logging, the signatures and the intrinsic characteristics of rocks.

6.2. Interpretation of borehole data

6.2.1. General

The following illustrations and text have been chosen as examples of some of the various types of geologic environment in which uranium has been found. The diagrams are simplified but based on existing geological knowledge of deposits found in various parts of the world. The illustrations which will be examined are:

6.2.2. Sedimentary environments

1. Calcrete deposits 2. Roll-front types 3. Stratiform deposits

6.2.3. Metamorphic environments

1. Unconformity vein type deposits (Athabasca, Canada) 2. Unconformity vein type deposits (N. T. Australia)

6.2.4. Granitic environments

1. Vein type deposits in granites

6.2.5. Hole to hole correlations

1. In roll-front deposits 2. In granites deposits 3. In stratiform deposits 4. Unconformity vein type deposits

232 FIG. 6.1. Calcrete deposits.

6.2.2.1. Calcrete deposits, Figure 6.1.

Calcrete deposits usually occur along the axes of mature valleys. Uranium bodies generally occur within the calcrete. A typical section is sequentially made up of overburden, followed by alluvium, the calcrete, the earliest alluvium, altered basement and finally basement. Mineralization is described as cynatite which forms a thin film on voids and fractures within the porcelaneous calcrete. Generally the cynatite is covered by a thin shell of silica.

The gamma-ray profile illustrates a gradual increase of gamma-ray intensity downward through the geologic column to the highest values near basal portion. A slight anomaly is noted in the underlying.alluvium. The remainder of the gamma-ray signatures are typical of the underlying altered granite and granitic basement.

The resistivity rises sharply within the calcrete, as would be expected from a porcelaneous, calcium-rich rock. Beneath the calcrete, the resistivity falls markedly in the lower alluvium and weathered granite, and then rises generally in the unweathered granite.

A slight negative anomaly is noted coincident with the high resistivity and high gamma-ray response. This is likely due to

233 an increase in resistivity. The remainder of the spontaneous potential graph is of little value.

The neutron-porosity measurement is reflecting mostly the variations of water content in this case. There is a change of deflection between the overburden and the alluvium because of the change in clay amounts (and, thus in the amount of bound water). Below the water table, the amount of free water (water saturation, Sy) increases, causing a decrease in deflection. In the calcrete, the above factors enter, plus additional water of crystallization. The altered granite has some water, but the basement granite only has about 1%. Thus, the granitic counting rate values are high.

6.2.2.2. Roll-front deposits, Figure 6.2

The lithology in these fresh water sediment basins is typically claystones and sandstones. The illustration represents

234 the upper and lower tails of a "C" shape roll front or geochemical cell. Mineralization is generally uraninite. Often pyrite is found above and around the front of the roll. The roll-front type deposit will always be bounded below by an impermeable layer; in this case, calcite, but it may be claystone. Often there is a restricting layer above, also.

The ganma-ray profile is fairly featureless and shows mostly the sand-claystone sequence. The ore zones (the upper and lower tails) show as excellent peaks. When the cell is bounded by claystone or shale, the position of the peaks within the claystone or not indicates the relative proximity to the main body of the cell.

The resistivity has values characteristic of each of the rock units with a maximum occurring coincident in the calcite cement at the base of the roll. Although this is a fairly common sequence it should be noted that the adjacent formations can also be low resistivity shales. The resistivity curve is excellent for determining the exact locations of the various layers.

The SP values are fairly normal for each of the rock types. The SP is "reversed", in this case indicating that the drilling fluid is more saline than the formation water. This may be due to solution of the calcium material when drilling through the calcite cemented layer.

The neutron-porosity curve is rather unimpressive at the scale shown here. It is reflecting changes in water content between sands, clays and limestones. It resembles the resistivity curve closely and could be substituted for it.

6.2.2.3. Stratiform deposits, Figure 6.3a and 6.3b.

The lithology of a stratiform deposit is a sequence of sandstones, siltstones and mudstones. Mineralization tends to aggregate in sandstone units which have some carbonaceous material. Seme scattered sulfide is often present. Mineralization is largely cynatite in the oxidized zone and uraninite in the unoxidized zone.

The profile of the ganma ray is relatively inactive reflecting the small radiation changes from the sands to silts to mudstones. There is a good anomaly where uranium mineralization is present. The first high on this profile occurs in the weathered zone above the water table where the uranium probably has been largely oxidized and removed because of weathering.

235 (a) I I

Siltltona. Uranium ORE ZONE K -r sc Smndntona "« :h7 h WATER TABLE

Sandttonm with Uranium f*Mutphid9M and earbonacaoua malarial 1 ORE ZONE

IIVTIH-KIITlia

(b) Ovmrburdtn Ipottibiy smttft• '-.:;A

Smnditon* Mud t ton*

Mmdwton'w- tUrmntnitm.Vanadium. Coppir with humie^ mat» rial I one ZONE

i•[iTin-Btii m H

FIG. 6.3. Stratiform type deposits: (a) general; (bj Colorado Plateau.

236 Beneath the water table is a significant anomaly which coincides with sandstones containing carbonaceous materials and some sulfides.

The sandstones encountered beneath the water table have high resistivities because of the probable low total dissolved solid content of the connate water. Within the ore zone are both conductive and resistive bands. These may be related to the changing amounts of carbonaceous material and sulfide. The low resistivity band does not appear to be shale or clay. Beneath the ore zones the resistivity curve reflects the siltstone.

This SP curve is relatively featureless except for a negative deflection, coincident with the highest gamna-ray peak and may be caused by both the carbonaceous material and the sulfides present.

Again, the neutron-porosity curve is reflecting the variation in water content. The overburden and siltstone show small amounts of bound water above the water table. There is a sharply lowered deflection at the water level in the borehole and further, again at the water table. The curve is essentially unaffected by the mineralization in the ore zone. Note that the curve is available where there is no water in the borehole, and it is almost unaffected by casing. The counting rate in an air- filled borehole will be approximately twice that in the same formation, with a water-filled borehole.

6.2.3.1. Unconformity vein type deposits (Athabasca, Canada), Figure 6.4 Mineralization in unconformity vein type deposits is located at the unconformity and in fractured and altered rocks imnedi- ately above and below the unconformity. Mineralization is often high grade uraninite. Most often it is associated with organics or carbonaceous material. The rock surrounding the highest grade mineralization has usually been altered to clays. The lithology is made up of overburden or glacial till, Athabasca Sandstone, a zone of hematite-rich sandstone (low grade mineralization) followed by more highly altered and bleached sandstone (with increasing grades). At the unconformity, the uranium grade is generally at its highest and co-coincident with clays. Downward, the clay content decreases and passes into a altered graphitic metapelites and eventually to altered basement.

The gamma-ray curve shows the overburden and top sandstone to have little difference in radiation. Approaching the bottom portion of the sandstone, the ganma-ray deflection increases and generally reaches the highest response just above or at the unconformity. The amplitude then decreases through the altered

237 Athabaaca Sandatona

Athabaaca SmrfKMi LOW GRADE ORE Athabaaca Sfditona HIGH GRADE ORE Unconformity _ Rogotith

Aitarad-fCiay) Zona LOW GRADE ORE Faulty Graphitic Matapaiita

Attarad Baaamant

itir MltltlU •IITHI-MHTIM

FIG.6.4. Unconformity vein type deposits.

zone to near background when reaching the graphitic metapelites, and falling to background in the altered basement.

These sandstones typically have high resistivities. As the alteration increases (and the clays increase) the resistivity markedly decreases. Near the unconformity, and coincident with the highest clay content, the resistivity becomes very low. A small, high resistivity peak is coincident with a fault at the bottom portion of the mineralized zone and the remainder of the profile decreases to background for those typical rocks.

Little SP variation is seen in the unaltered sandstone. A slight negative peak is evident in the hematitic portion, and a second negative peak is coincidental with the unconformity. Sometimes this can be attributed to the sulfides present, but is probably due to the clay content at the unconformity. A positive peak is coincidental with a resistivity peak on the fault zone and is probably an indication of permeability. The remainder of the curve has little variation.

The neutron-porosity curve shows the normal contrast between the sands and the clay-rich ore zone. Note the slight changes in the casing and the hematitic zone because of the increased scattering due to density.

238 Smndatam

Schist

Minor Qrmphitic Schist

IEU1IOI-IEUIIOI

FIG.6.5. Unconformity vein type deposits (N.T. Australia).

6.2.3.2. Vein type deposits (Australia), Figure 6.5.

The geological characteristics of this sequence are made up . of a series of schists with viable content of mica, biotite and feldspar, carbonaceous schists, pegmatites, dolomites and basement schists. The mineralization, which is uraninite, occurs in the graphitic schists. Generally there are no ancillary minerals present except for gold.

The gamma-ray backgrounds are generally low except where graphitic schist is encountered. A broad band of deflection is present in the upper scist. The pegmatites have some potassium present and thus have a slightly higher background levels. The minor graphitic bands below the ore zone are slightly anomalous. The dolomite has a typically extremely low radioactive background. A slight increase in the background is encountered in the basement schists.

The schists all have a low resistivity backgrounds with significant lows occurring in the graphitic schist which are probably caused by the conductivity of the graphite. There are the usual resistivity highs in the pegnatites and the dolomitic rocks. Approaching the basement schists, the resistivity decreases.

239 E3 Atkmlinm Two Mica Granite

FIG. 6.6. Vein type deposits in granites.

A negative deflection is encountered in the upper part of the series and a slight positive deflection occurs in the pegfiiatites. A second low occurs over the minor graphitic zone. A fairly uniform background is encountered for the remainder of the profile, because of the high resistivities in these formations .

In this type environment the neutron-porosity curve con- trasts show the change in water content between the flakey schists and the more massive pegmatites. There is also some contrast due to different capture cross-sections.

6.2.4.1. Vein type deposits in granite, Figure 6.6.

Granite terrain is relatively homogeneous with the only changes being either lamprophye or aplite dykes, breccia zones, and perhaps mylonite zones. Mineralization is usually uraninite with some sulfides. Some vein types though, have pitch- blende, coffinite, and secondary uranium minerals with calcite, pyrite and hematite.

The gamma-ray profile exhibits a fairly high uniform level in the granites as would be expected from a prospective granite.

240 Gamma-ray anomalies occur in the mylonite zone and in the silicified breccia zone. The remainder of the profile is fairly uniform. Granite exhibits a fairly uniform resistivity profile except for the low reading in the mylonite zone (due to the presence of clay) and a high coincidential with a silicified breccia zone. The profile remains uniform for the remainder of the log. The lower granite zone reads higher than the (apparently) similiar upper one because the borehole is bottomed within that granite.

A negative SP peak occurs at the mylonite zone and a positive one occurs in the silicified breccia zone. The low associated with the mylonite zone is probably due to the clays and the high associated with the breccia zone is probably due to permeability.

In the vein type environment, even with a water-filled hole, most of the deflections of the neutron-porosity curve are high because of the massive character of the formations and the fact that they contain little water. The fragmented and brecciated zones contrast greatly because of their water content. The dykes may show a small contrast due to different capture cross- sections.

6.2.5.1. Roll-front deposits - Hole to hole correlation, Figure 6.7 There are certain characteristic signatures and values of the gamma-ray and electrical logs in roll fronts which are typical and can be related to characteristics of the roll front. Some of the effects on the electric logs are hidden because of the neglect of the mud resistivity measurement in uranium logging. These effects are related to the alteration of the sand as the geochemical cell passes through the sand.

Hole number 1 is back in the remote barren portion of the geochemical cell. The gamna-ray curve shows two anomalous peaks, one in the shale or clay above the sand and the other in the shale, clay, or limey clay below the sand. The uranium mineral has been leached out of the sand, leaving little radioactivity behind, except that associated with a clayey sand and that in the almost zero permeable clay above and below. The passage of the cell has resulted in alteration of the feldspathic mineral in the sand and thus the resistivity is rather low and the SP is between a shale and sand potential. The formation water contains small amounts of very soluble metal compounds which lowers the resistivity and makes the SP more negative. The SP baseline is shifted toward the negative in the sand because the sand is oxidized. A neutron curve would show a normal shaly (clayey) sand. Hematite is probably present.

241 K> NEAR BARREN NEAR REMOTE A N> REMOTE BARREN-60 INTERIOR-50 ORE ZONE-30 SEEPAGE ZONE-2Q SEEPAGE ZONE-10 0-

10- 20- r 30- ti 40- 50- ( 60-

ALTERATION CROSS-SECTION

GENERAL ROCK DESCRIPTION ARKOSIC SAND - COMPACTED BUT NOT CEMENTED TYPE GEOCHEMICAL CELL MEDIUM TO COARSE GRAIN FAIR SORTING POWDER RIVER BASIN SUB-ANGULAR 10 SUB-ROUNDED SHAPES WYOMING QUARTZ >80% FELDSPAR <15% RUBIN 1971 CARBON FRAGMENTS <5% PYRITE ±1% BLACK AND GREEN ACCESSORY MINERALS <1%

FIG.6. 7. Roll-front type - hole to hole correlation (Rubin, 1971). Hole number 2 is similiar to number 1 but the ganma-ray curve is within the sand and the anomalies are larger. The SP curve indicates that the peaks are in less oxidized sand than that between them. These indicate a position in the barren or altered zone near the cell. Limonite may be present.

Hole number 3 is at the cell redox interface. All curves indicate a rather laminar sand-clay sequence. The SP is active because of the larger amounts of dissolved metallic compounds, the clayey layers, and the active oxidation taking place. The resistivity curve is low for the same reasons. The ganma-ray curve is reflecting the relative permeability through the presence of the soluble and insoluble uranium mineral. Small amounts of calcite are typical at the lower part of the sand. Carbonaceous trash remnants and tarnished sulfides may be present higher in the sand. Limonite may be present.

Hole number 4 is through the main body of the cell. This is where most of the ore will be found. It is a highly min- erallized, highly reduced zone. The permeability is often rather restricted because of a sharp decrease in grain size at the interface. The gamma-ray curve will show its greatest deflection here in a single anomaly. The SP will indicate a fine-grained sand with little dissolved mineral in the formation water. The resistivity curve will read higher than anywhere in the altered zone because of the low mineral content of the water and the lower porosity of the sand. A neutron curve would show a slightly lower porosity. Calcite may be present, usually low in the sand, but not always. Carbonaceous material and sulfides may be present. Bright sulfides and/or magnetite may be present.

Hole number 5 is similiar to 4 except the gamma-ray curve indicates that the uranium mineralization lies at the bottom of the sand. This is the seepage zone.

6.2.5.2. Vein deposits in granite - Hole to hole correlation, Figure 6.8 Three drill holes are illustrated as intersecting 45° angle faults in a granitic terrain. Mineralization is generally found in silica or in some instances in altered fault zones. In this instance we look for the correlation between holes as a method of determining (in addition to the geological interpretation of the drill core or chips) the dip and continuity of the mineralized zone. Granites generally display a uniform slightly higher background for gamma-ray profiles. In drill hole 1 the gamma- ray signature illustrates the response obtained at a fault near the top which is again repeated near the lower middle of drill hole 2 and probably relates to a still thinner zone on the bottom of drill hole 3. This assumes previous knowledge of the

243 FIG.6.8. Granite deposits - hole to hole correlation.

dip and strike of the faults. Midway in drill hole 1 is a small response obtained of a fault intersection and which can be correlated to a like intersection near the top of drill hole 2. In drill hole 3, near the middle, is a small intersection which is impossible to correlate with any of the others and indicates a separate fault system. Through a simplified diagram, this data typifies some types of occurrences in granite terrain.

6.2.5.3. Stratiform depostis - Hole to hole correlation, Figure 6.9

Typical stratiform mineralization is usually contained within a sandstone member surrounded by other sediments such as siltstones and slates. The sandstone member may also contain some carbonaceous material and sulfides. Where the deposit is in the weathered zone the mineralization is often coffinite and the unweathered zone usually uraninite. Uraninite fills the open spaces and fractures, and is conmonly associated with pyrite. Drill hole 1 illustrates a varying background of ganma- ray and resistivity signatures fairly typical of the sand and

244 FIG.6.9. Stratiform deposits - hole to hole correlation.

245 shaly rock types encountered. The mineralization exhibits a ganma-ray high, with the resistivity exhibiting both extreme highs and lows. This indicates alternating impermeable barriers and highly carbonaceous material with some effect due in part to the pyrite. Drill hole 2 illustrates a continuance of the resistivity signature over what would be the expected uranium zone. Only a slight gamria response is detected. Correlation of the resistivity curve with hole number 1 is good. Still further up- dip, hole 3 indicates a projection of the mineralization illus- trating it must have been at surface and that new mineralization was encountered in the middle of the hole. The gamma-ray profile in the middle of hole 2 indicates sub-economic mineralization, but had a resistivity characteristic similar to hole 1. This suggests a near-miss situation, and indicates the need for additional drill holes at a closer spacing. It also suggests that this type of deposit may have a chance of being "pod like" in occurrence and therefore will require considerable amounts of close drilling to define the ore horizons and pods. Should hole 2 have been placed either up-dip or down-dip from its present position the interpreter could be forgiven for having connected the mineralization between the two holes; which would have given quite an erroneous picture as to the actual aerial extent of the mineralization.

6.2.5.4. Unconformity vein type deposits - Hole to hole correlation, Figure 6.10

This cross-section illustrates a stylized lithology that is found in the Pine Creek Geosyncline. It consists of an overlying sandstone, an unconformity, followed by schists, dolomites and alternating units of graphitic and chloritic rich schists, followed by dolomites and ending in a basement gneiss. A reverse fault shown through the center and the section is further interrupted by a pegpiatite dyke. The gamma-ray profile in drill hole 1 is the background for the sandstone, the overlying schists and the dolomite. The graphitic schists are anomalous and have a distinctive signature. The basement gneiss has a background value about that of most granitic rocks. Hole 2 intersects a fault beneath the sandstone and extends into a dolomite signature followed by a chlorite rich'graphite unit and a mainly graphitic unit. The two graphite rich units have a correlatable gamma-ray signature similar to drill hole 1. In addition to the core that would have been recovered and geologically logged from these holes, there is a directly correlatable signature both in ganma-ray and resistivity which matches between the two holes.

Drill hole 3 intersects an obvious fault. The signature which was found in horizon 1 in drill holes 1 and 2 now can be

246 found near the expected continuation of horizon 2. However, the signature for horizon 2 fails to emerge because it is faulted off, and the bottom part of the hole goes into dolomite and into basement gneiss. The fairly distinct signature between the dolomites and the first graphitic horizon 1 assists in making the correlation that the only mineralization indicated in drill hole 3 is, in fact, that of the uppermost unit found in holes 1 and 2. Note the similarity between the resistivity and ganma- ray signatures in each of these 3 holes in horizon 1. Drill hole 4 has an anomaly at the unconformity and it cannot be suggested that it will automatically have the continuity such as illustrated. However, when the second anomaly occurs, illus- trating the bottom-most horizon signature found in drill holes 1 and 2 with a correct vertical spacing between the horizons, it suggests that the anomaly now found at the unconformity relates to horizon 1 and therefore, to the upward continuation of that unit. In drill hole 4 beneath the dolomite the hole intersects a pegjmatite dyke where the gamma-ray profile rises slightly due to the potassium content of the dyke. Note the change in the resistivity above the contact.

If the core information from the hole drilled is combined with the electrical and gamma-ray logs a better picture can be built up of the probable geologic interpretation from very few holes in a geologic section.

247 7. SELECTED REFERENCES

References: Borehole Effects and Mud.

Pirson, Sylvain J. "Handbook of Well Log Analysis." Prentice-Hall, Inc. Englewood Cliffs, New Jersey. 1963.

Dakhnov, V. N. "Geophysical Well Logging." Quarterly of the Colorado School of Mines. Golden, Colorado. Vol. 57, No. 2, April 1962.

Vennard, John K. "Elementary Fluid Mechanics." John Wiley & Sons. 4th Edition, 1961.

Czubek, Jan A. "Quantitative Interpretation of Garmia Ray Logs." Institute of Nuclear Research VI Dep. Krakow, Poland. C 1962.

LeRoy, L. W. and LeRoy, D. 0., "Subsurface Geology." Colorado School of Mines.

Keller, G. V. and Frischknecht, F. C., "Electrical Methods in Geophysical Prospecting." Pergamon Press.

Stokes, W. L. and Vames, D. J., "Glossary of Selected Geologic Terms." Colorado School of Mines.

"Log Interpretation Reference Data Handbook", Gearhart-Owens Industries, Inc. 1972, Fort Worth.

McAdams, William H. "Heat Transmission." McGraw-Hill Book Company, Inc., 1954.

Sears, Francis W. "Electricity and Magnetism." Addison-Wesley Publishing Company, Inc., 1955.

Zemausky, Mark W. "Heat and Thermodynamics." McGraw-Hill Book Company, Inc., 1943.

Formation Evaluation Data Handbook.," Gearhart-Owens Industries, Inc., 1974.

"An Introduction to Single Element Differential Tanperature Logging.", GO International document.

Holmes, Charles S. and Smith, Samuel C. "Calculation of Circulating Mud Temperatures." J. P. T., June 1970.

Zimmerman, D. W. "Dating of Ancient Pottery by Therriiolumines- cence." Research Laboratory for Archeology, University of Oxford, England.

248 References: Radioactivity Methods. Radioactivity and Ganrna Ray Logging.

Cork, James M. "Radioactivity and Nuclear Physics." D. Van Nostrand Company, Inc. 1957.

Lapp, R. E. and Andrews, H. L. "Nuclear Radiation Physics." Prentice Hall, 1949.

"Formation Evaluation Data Handbook." Gearhart-Owen Industries, Inc., 1974.

Smith, Orsino C. "Identification and Qualitative Analysis of Minerals." D. Van Nostrand Co., Inc., 1953.

"National Resource Evaluation, Preliminary Report." Energy Resource and Development Authority, June 1976.

Daniels, Scott and Blackmon, Starkey. "Borehole Geophysical Investigations in the South Texas Uranium District." Journal of Research of the U.S.G.S., Vol. 5, No. 3, May-June 1977, pp 343-357.

Rosholt, Jr., John N. "Natural Radioactive Disequilibrium of the Uranium Series." U.S.G.S. Bullentin 1084-A.

Kowalski, E. "Nuclear Electronics." Springer-Verlag, 1970.

Nicholson, P. W. "Nuclear Electronics." John Wiley and Sons, 1974.

Senftle, F.B., Tanner, A. G., Philbin, P. W., Boynton. G. R., and Schram, C. H. "In-situ Analysis of Coal Using a Cf - Ge(Li) Borehole Sonde." Mining Engineering (AIME) Vol. 30, No. 6, pp 666-674.

Alder and Poessler, 5th Edition. "Introduction to Probability and Statistics." W. H. Freemen and Company, 1972.

Birks, J. B. "Scintillation Counters." McGraw Hill Book Co.

Czubek, J. A. "Quantitative Interpretation of Ganma Ray Logs." Institute of Nuclear Research. Kracow, Poland.

Hallenburg, James K. "Interpretation of Gamma Ray Logs." WGA Earth Science Bulletin, September 1973.

Dodd, P. H. Eschliman, Dennis H., NATO. "Borehole Logging Techniques for Uranium Exploration and Evaluation." Methods of Prospecting for Uranium Minerals, London 10/21/71.

249 Scott, J. H. "Computer Analysis of Ganma Ray Logs." Society of Exploration Geophysicists. Geophysics, Vol. XXVIII, No. 1-6, 1963.

Scott, James H. "The Gamlog Computer Program." AEC RME 143.

Scott, J. H. "The Grade Computer Program for Calculating Uranium Ore Reserves." AEC RME 145.

Moore, Donald C. "Interpretation of Radiometric Logs in Thin and Dipping Beds." BFEC Publications, 1978.

Dodd, P. H. "Quantitative Logging and Interpretation Systems to Evaluate Uranium Deposits." SPWLA Proceedings, 1966 Paper P..

Czubek, Jan A. "Natural Selective Ganrna Logging." Nukleonika, Vol. XIII, No. 1, 1968.

Czubek, J. A. "Recent Development of the Nuclear Geophysics Methods to the Mineral Prospection and Exploitation." ACS Paper 15, Houston, February 22, 1970.

Jonas, J. Thomas. "Digital Data Processing Techniques Applied to the Natural Gamma Ray Log." Thesis, Colorado School of Mines, Golden, Colorado, 1975.

George, Evans, Allen, Key, Ward, Mathews. "A Borehole Gamma Ray Spectrometer for Uranium Exploration." BFEC, DOE GJBX -82 (78) May, 1978.

Killeen, P. G. "A Ganma Ray Spectral Logging system." Geological Survey of Canada, 1977.

Princeton Gamna Tech. "The PGT Probe." Document, 1978.

Darnley, A. G. "Airborne Gamma Ray Survey Techniques." Geological Survey of Canada, 1971.

Edwards, J. M., Ottinger, N. H., and Haskell, R. E. "Nuclear Log Evaluation Potash Deposits." Proceedings SPWLA, 1967, Paper L.

Grasty, R. L. and Holman, P. B. "Optimum Detector Sizes for Airborne Gamma Ray Survey." Resource Geophysics and Geo- chemistry Division, Geological Survey of Canada, Paper 74-1, Part B.

Kellogg, William C. "Aerial Radioactivity Surveying. Techniques of Successful Application." Lockwood, Kessler and Bartlett, Inc., Consulting Engineers, Pasadena, California, February, 1968.

250 Belknap, W. B., Dewan, J. T., Kirkpatrick, C. V., Matt, W. E., Pearson, A. J., and Robson, W. R. "API Calibration Facility for Nuclear Logs." American Petroleum Institute Proceedings, 1959 pp 289-317.

Conaway, J. C. "Problems in Garrma-Ray Logging: The effect of dipping beds on the accuracy of ore grade determinations." Current Research, Part A, Geologic Survey of Canada, Paper 70- 1A, 1979.

Crew, M. _E. and Berkoff, E. "TWOPIT, A Different approach to Calibration of Ganma-Ray Logging Equipment." Log Analyst, II, No. 6, 1970, pp 26-32.

Czubek, J. A. "Differential Interpretation of Gamma-Ray Logs, 11 Case of the Dynamic Gamma-Ray Curve." Warsaw, Poland, Report No. 7931, Nuclear Energy Information Center of the Polish Government, Comissioner for Use of Nuclear Energy, 1972.

Duray, J. R. "A Brief Review of the Basis for, and the Proce- dures Currently Utilized in, Gross Gamma-Ray Log Calibration." U.S. Department of Energy open-file report, GJBX-61(76), 1976 12 p.

Emilia, D. A., Allen, J. W., Chessmore, R. B., and Wilson, R. B. "The DOE/SIMPLEC Magnetic Susceptibility Logging System." U.S. Department of Energy open-file report GJBX-75(81), 1981, 42 p.

Fink, J. B. "On K-factors and gamma log calculations." Geophysics, 43, 1978, pp 1546-1550.

Jain, M, Evans, J. L., and Close, D. A. "Nondestructive Assay Technology for Uranium Resource Evaluation, Infinite Medium Calculations." U.S. Department of Energy open-file report, GJBX-85(79), Grand Junction, Colorado, 1979, 70 p.

Koizumi, C. J. "Thin, Dipping Ore Zone Logging Models: Log Studies." U.S. Department of Energy open-file Report, GJBX-54 (80), Grand Junction, Colorado, 1980, 45 p.

Mathews, M. A., Kozumi, C. J., and Evans, H. B. "DOE-Grand Junction Logging Model Data Synopsis." U.S. Department of Energy open-file report GJBX-76(68), Grand Junction Colorado, 1978, .52 p.

Scott, J. H., and Dodd, P. H. "Ganma Only Assaying for Dis- equilibrium Corrections." USAEC Report RME-135, Geology and Mineralogy, Grand Junction, Colorado, 1960, 20 p.

251 Scott, J. H., Dodd, P. H., Droullard, R. F., and Mudra, P. J. "Quantitative Interpretation of Gamma-Ray Logs." Geophysics, 26 1961, pp 284-294.

Wilson, R. D., Stromswold, D. C., Evans, M. L., Jain, M., and Close, D. A. "Spectral Gamma-Ray Logging III: Formation and Thin Bed Effects." SFWLA Twentieth Annual Logging Symposium, Vol. II, Tulsa, Oklahoma, 1979, paper FF.

Wilson, R. D. and Stromswold, D. C. ' "Spectral Gamma-Ray Logging Studies." U.S. Department of Energy open-file report GJBX-21 (81), Grand Junction, Colorado, 1981, 187 p.

Wollenberg, H. A. "Nuclear Methods in Mineral Exploration and Production." J. G. Morse, editor; Elsevier Scientific Publishing Co., 1977, pp 5-36.

References: Garrma-Ray Spectroscopy.

Adams, J.A.S. and Lowder, W. M. (eds.), 1964. "The Natural Radiation Environment." University of Chicago Press, Chicago, 1069 p.

Bowie, S. H. U., David, M., and Ostle, D., 1972. "Uranium Prospecting Handbook." Inst. Min. Metall., London, 346 p.

Bristow, Q., 1979. "Garmia ray spectrometric methods in uranium exploration: airborne instrumentation." Geophysics and Geo- chemistry in the Search for Metallic Ores, Geol. Surv. Can., Econ. Geol. Rep. 31, Paper 10A.

Conaway, J. G., 1980. "Uranium concentrations and the system response function in gamma-ray logging." Geol. Surv. Can., Paper 80-la, pp. 77-87.

Conaway, J. G., Bristow, Q., and Killeen, P. G., 1980. "Opti- mization of gamma-ray logging techniques for uranium." Geo- physics, v. 45, 292-311.

Conaway, J. G., and Killeen, P. G., 1978. "Quantitative uranium determinations from gamma-ray logs by application of digital time series analysis." Geophysics, v. 43, no. 6, 1204- 1221.

Conaway, J. G., and Killeen, P. G., 1980. "Ganma-ray spectral logging for uranium." Can. Inst. Min. Metall., Vol. 73, No. 813, p. 115-123.

252 Davisson, C. M. and Evans, R. D., 1952. "Gainna ray absorbtion coefficients." Reviews of Modern Physics, v. 24, No. 2, p. 79-107

Dodd, P. Hi, and Eschliman, D. H., 1972. "Borehole logging techniques for uranium exploration and evaluation." Uranium Prospecting Handbook, S.H.U. Bowie et al. (ed.), Inst. Min. Me tall., London, p. 244-276.

Dodd, P. H., Droullard, R. F., and Lathan, C. P., 1969. "Bore- hole logging methods for exploration and evaluation of uranium deposits." Mining and Groundwater Geophysics, 1967. Geol. Surv. Can., Econ. Geol. Rep. 26, p. 401-415.

George, D. C., Evans, H. B., Allen, J. W., Key, B. N., Ward, D. L. and Mathew, M. A., 1978. "A borehole gamma-ray spectrometer for uranium exploration." U.S. Dep. of Energy, Grand Junction Office, Report GJBX-82(78).

IAEA, 1974. "Recommended instrumentation for uranium and thorium exploration." Technical Report 158, IAEA, Vienna, 93 p.

IAEA, 1976. "Radiometric reporting methods and calibration in uranium exploration." Technical Report 174, IAEA, Vienna, 57 p.

Killeen, P. G., 1975. "Nuclear techniques for borehole logging in mineral exploration." Borehole Geophysics Applied to Metal- lic Mineral Prospecting - a review. A.V. Dyck (ed.), Geol. Surv. Can., Paper 75-31, p. 39-52

Killeen, P. G., and Bristow, Q., 1976. "Uranium exploration by borehole gamma-ray spectrometry using off-the-shelf instrumentation." Exploration for Uranium Ore Deposits, Proc. Series, IAEA, Vienna, p. 393-414.

Killeen, P. G., and Bristow, Q., 1976. "Radioactive disequilibrium determinations, Part 1: Determination of radioactive disequilibrium in uranium ores by alpha-spectrometry." Geol. Surv. Can., Paper 75-38, p. 1-18.

Killeen, P. G., and Conaway, J. G., 1978. "New facilities for calibrating ganma-ray spectrcmetric logging and surface exploration equipment." Can. Inst. Mining Me tall. Bull., v. 71, no. 793, p.84-87

Killeen, P. G., Conaway, J. G., and Bristow, Q., 1978. "A gamma-ray spectral logging system including digital playback, with recommendations for a new generation system." Current Research, Part A, Geol. Surv. Can., Paper 78-1A, p. 235-241.

253 Lock, G. A. and Hoyer, W. A., 1971. "Natural gamma-ray spectral logging." The Log Analyst, v. 12, no. 5, p. 3-9

Lvborg, L., Wollenberg, H., Rose-Hansen, J., and Nielsen, B.L., 1972. "Drill-core scanning for radioelements by ganma-ray spectrometry." Geophysics, v. 37, no. 4, p. 675-693.

Mathews, M. A.', Koizumi, C. J., and Evans, H. B., 1978. "D.O.E. Grand Junction logging model data synopsis." U.S. Dept. of Energy, Grand Junction Office, Report GJBX-76(78).

Moxham, R. M. and Tanner, A. B., 1977. "High resolution garrma- ray spectrometry in uranium exploration." U.S. Geol. Survey, Jour. Resear., v. 5, no. 6, p. 783-795.

Ostrihansky, L., 1976. "Radioactive disequilibrium determinations, Part 2: Radioactive disequilibrium investigations, Elliot Lake area, Ontario." Geol. Surv. Can., Paper 75-38, p. 19-48. Rhodes, D. F. and Mott, W. E., 1966. "Quantitative intrepre- tation of gamma-ray spectral logs." Geophysics, v. 31, no.2, p. 410-418. Rosholt, J. N., Jr., 1959. "Natural radioactive disequilibrium of the uranium series." U.S. Geol. Surv. Bull. 1084-A, p . 30

Scott, J. H., 1963. "Computer analysis of gamma-ray logs." Geophysics, v. 28, no. 3, p. 457-465.

Scott, J. H., and Dodd, P. H., 1960. "Gamma-only assaying for disequilibrium corrections." U.S. Atomic Energy Comm. RME-135, p. 1-20

Scott, J. H., Dodd, P. H., Droullard, R. F., and Mudra, P. J., 1961. "Quantitative interpretation of ganma-ray logs." Geo- physics, v. 26, no. 2, p. 182-191

Senftle, F. E., Moxham, R. M., Tanner, A. B., Boynton, G. R., Philbin, P. W., and Baicker, J. A., 1976. "Intrinsic germanium detector used in borehole sonde for uranium exploration." Nuclear Instruments and Methods, v. 138, p. 371-380.

S iegbahn, K., 1968. 1 'Alpha-, be ta-, and gamna-ray spec tros copy.'1 North Holland Pub- Co. Amsterdam, p. 1742

Toens, P. D., van As, D., and Vleggaar, C. M., 1973. "A facility at the national nuclear research centre, Penlindaba for the calibration of gamma-survey meters used in uranium prospecting operations." J.S. Afr. Inst. Min. Melall., v. 73, p. 428.

254 References: Spontaneous Potentials.

Sheriff, R. E. "Encyclopedic Dictionary of Exploration Geophysics." Society of Exploration Geophysicists. Tulsa, 1973.

Lynch, Edward J. "Formation Evaluation." Harper and Row. New York, 1962.

Dakhnov, V. N., translated by Keller, G. V. "Geophysical Well Logging." Colorado School of Mines, Volume 57, Number 2, April 1962.

Ives, David J. G., and Janz, G. J. (editors). "Reference Electrodes, Theory and Practice." Academic Press. New York, 1969.

Doll, H. G. "The S.P. Log: Theoretical Analysis and Principles of Interpretation." AIME Technical Publication 2463, 1948.

Telford, Gedart, Sheriff, and Keys. "Applied Geophysics." Cambridge University Press. Cambridge, 1976.

Wyllie, M. R. J. "A Quantitative Analysis of the Electro- chemical Component of S. P. Curve." AIME publication.

Gondouin, M., Tixier, M. P., and Seinard, G. L. "An Experi- mental Study of the Influence of the Chemical Composition of Electrolytes on the S. P. Curve." Journal of Petroleum Technology, February 1957.

Garrels and Christ. "Solutions, Minerals, and Disequilibria." Freemen Cooper & Company. San Francisco, 1965.

Evers, John F., and Iyer, Babu G. "A Statistical Study of the SP Log in Fresh Water Formations of Northern Wyoming." SPWLA 16th Annual Logging Symposium. June 1975.

Doll, H. G. "The S. P. Log in Shaley Sands." T. P. 2912 AIME, 1949.

Guyod, Hubert. "Interpretation of Electric and Gamma Ray Logs in Water Wells." Log Analyst, Volume VI Number 5. January - March 1966.

Hill, Shirley, and Klein. "Bound Water in Shaley Sands." Log Analyst, Volume XX, Number 3. May - June 1979.

255 Evseeva, L. S., Ivanov, K. E., and Kochetkov, V. I. "Some Laws of the Formation of Epigenetic Uranium Ores in Sandstones, Derived from Experimental and Radiochemical Data." Atomnaya Energiya, Volume 14, Number 5, pp 474-481. May 1963 (U. S. source unknown).

Sato, Motoaki and Mooney, Harold M. "The Electrochemical Mechanism of Sulfide Self-Potentials." Geophysics, Volume XXV, Number 1. February 1960.

Duncan, D. W. and Bruynesteyn, A. "Microbiological Leaching of Uranium." 1970 Uranium Symposium, Socorro, New Mexico.

Hallenburg, J. K. "Use of the Spontaneous Potential Curve in a Mineral Mapping Technique." SPWLA 19th Annual Logging Symposium, 1978.

Gondouin, M. and Scala, C. "Streaming Potential and the SP Log." SPE paper Number 864-G.

Wyllie, M. R. J. "An Investigation of the Electrokinetic Component of the Self Potential Curve." Petroleum Transactions, AIME TP2940 Volume 192, 1951.

Doll, H. G. "Selective SP Logging." Journal of Petroleum Technology, T.P. 2850, Volume 189, 1950.

References: Mechanical Methods. Calipers.

Selby, Samual M. "Standard Mathematics Tables," 17th Edition. The Chemical Rubber Company, 1969.

References: Magnetic Susceptibility Methods.

Sears, F. W., "Electricity and Magnetism." Addison-Wesley Publishing Company, Inc. Cambridge, MA 1946.

Zablocki, C. J. "Magnetite Assays from Magnetic Susceptibility Measurements in Taconite Production Blast Holes,." Geophysics, Volume 39, Number 2, April 1974, pp 174-189.

Anderson, Wallace L. "Theory of Magnetic Susceptibility Measurements with Coil Pairs." Geophysics, Volume 33, Number 6, December 1968, pp 962-971.

256 Telford, W. M., Geldart, L. P., Sheriff, R. E., and Keys, D. A. "Applied Geophysics." Cambridge University Press, 1976.

Keller, G. V., and Frischkneckt, Frank C. "Electrical Methods in Geophysical Prospecting." Pergamon Press, 1966.

Breiner, S. "Applications Manual for Portable Magnetometers." Geometries publication, 1973.

Ellis, John R., Austin, S. Ralph, Droullard, Robert F. "Mag- netic Susceptibility and Geochemical Relationships as Uranium Prospecting Guides." U.S.A.E.C. (U.W.D.O.E.) Publication AEC-RID-4, Grand junction, Colorado.

References: Radioactivity Methods. Density Logging and Neutron Logging.

Tittman, J. "Radiation Logging." Petroleum Engineering Conference, University of Kansas, April 2-3, 1966. Reprinted in SPWLA Reprint Volume: "Gamma Ray, Density, and Neutron lagging," 1978.

Owen, J. D. "A Review of Fundamental Nuclear Physics Applied to Ganma Ray Spectral Logs." The Log Analyst, September 1966. SPWLA Reprint Volume: "Ganma Ray, Density, and Neutron Logging!1 1978.

Sherman, H. and Lock, S. "Depth of Investigation of Neutron and Density Sondes for 35 Percent Porosity Sand." SPWLA Symposium Transactions, 1975. SPWLA Reprint Volume: "Ganma Ray, Density, and Neutron Logging," 1978.

Felder, B. And Boyeldieu. "The Lithodensity Log." 6th European SPWLA Symposium Transactions, 1979, paper 0.

Woolson, W. A. and Gritzner, M. L. "Borehole Model Calculations for Direct Uranium Measurements with Neutrons." U.S. Department of Energy, Grand Junction, Colorado Contract, Science Applica- tions, Inc., January 1977.

Woolson, W. A. and Gritzner, M. L. "Evaluation Models of Active Neutron Logging Tools for Direct Uranium Measurement." U. S. Department of Energy, Grand Junction, Colorado contract, Science Applications, Inc., September 1978

Lapp and Andrews. "Nuclear Radiation Physics." Prentice-Hall, Inc.

257 Givens, W. W., Mills, W. R., Dennis, C. L., and Caldwell, R. L. "Uranium Assay Logging using a Pulsed 14-Mev Neutron Source and Detection of Delayed Fission Neutron Source and Detection of Delayed Fission Neutrons." Geophysics Volume 41, Number 3, June 1976, pp 468-490.

Bivens, Smith, Jensen, Jacobs, and Rice. "Pulsed Neutron Uranium Borehole Logging with Epithermal Dieaway." Sandia Laboratories, Albuquerque,New Mexico 1976.

Steinman, D. K., John, Joseph. "252Cf-Based Logging System for In-Situ Assay of Uranium Ore." IRT Corporation report Intel-RT 7019-005, circa 1977.

References: Dipmeters. Electric Logging Systems.

Holt, Olin R. "Relating Diplogs to Practical Geology." Dresser- Atlas document.

LeRoy, D. 0. and LeRoy, L. W. "Subsurface Geology." Colorado School of Mines, 1977.

Rodriques, A. R. and Pirson, S. J. "The Continuous Dipmeter as a Tool for Studies in Directional Sedimentation and Directional Tectonics."

Cow, J. W. "The High Resolution Dipmeter Reveals Dip-Related Borehole and Formation Characteristics." SPWLA Transactions, 1970 - D.

Okitsu, Fiunio. "Quick Interpretation of the High Resolution Dipmeter (HDT)." SPWLA Transactions, 1976 - I.

Fundamentals of Dipmeter Interpretation." Schlumberger Well Services document, 1970.

258 APPENDIX A

Calibration Procedure for Gross-Count Ganma-Ray Logging

The determination of the calibration factor, k by the tail- factor method is outlined for a log made in the model N3, located at the U.S. Department of Energy facility in Grand Junction, Colorado. A cut-away view of N3 is shown in figure 1A. The borehole is uncased in the barren and mineralized zones. When the model N3 is not in use, it is filled with water in order to keep the zones saturated. The model is completely sealed (except for the borehole) in a steel container and approximately 60% is above ground level.

A calibration is made under a reproducible set of standard conditions. The premise of this calibration procedure is that logs taken in the field, under conditions unlike the standard can be adjusted (by appropriate correction factors) to the standard condition. These standard conditions are:

Condition Standard Borehole diameter 4.5 inches (11.4 cm) Borehole fluid air Borehole casing none Ore zone moisture 12.3% (by weight) Ore zone bulk density 2.09 g/cm3 (wet)

The assigned radiometric grade for the calibration model used here is 0.240% eU308 and represents an average of dry-weight determinations. Each determination is made by a closed-can radiometric assay system, calibrated by certified U.S. New Brunswick Laboratory uranium counting standards (NBL #103). The thickness of the mineralized zone is 4.19 feet, determined by measurement at the time of construction.

The following procedures are recottmended:

1. The chart recorder drive (for analog systans) or the reporting intervals (for digital and point-by-point systems) should be indexed to the depth of the probe in the borehole model. 2. The recorder response for all types of systems should be capable of being read to a precision of + 1% or better of the nominal response in the mineralized zone. The probe should be positioned in the calibration borehole model so it will remain in contact with the borehole wall during the calibration. 3. For all dynamically calibrated systems, the borehole model is logged from the bottom to the surface.

259 N-3 TEST PIT

• 5.0' f I.I

- 4" 1.0. pipe

— 4 1/2

Barren Zone

"t />buiii tdr»|» 1.83 gm/cm5 G« 0.240 % e U30e < 3 />Bulk(wet)= 2.09flm/cm GT- 1.006 %e U,0e-feet 12.3% moisture (by weight)

Barren Zone

— 4 I D. pipe I 1

FIG.1A. Cut-away view of model N3 at Grand Junction, Colorado.

4. For analog systems, the chart paper grid should be sub- divided so a grid line occurs at successive 10 centimetre intervals. It is recommended that the chart depth scale be 50 - t (0.5 metre equals one centimetre of chart paper). For static calibration, the system should be positioned, then allowed to reach equilibrium by waiting for the ratemeter to stabilize before beginning the next measurement. 5. For digital and point-by-point systems, the depth based reporting interval should be uniform and not be greater than 10 centimetres.

260 DEFLECTION (INCHES)

FIG.2A. Log of model N3.

The probe electronics should be operating properly. It is important that the lower level electronic threshold be properly set. A simple method to determine the best setting of the lower level (energy) threshold is the gain-plateau method. The purpose of this method is to adjust the threshold so that it is above the electronic noise, below the signals of interest, and at the same time, least sensitive to the drift in electronic gain. The method consists of measuring and then plotting count rate (or counts for a fixed time interval) versus the setting of the lower level threshold adjustment. The settings (x~axis) should range from the electronic noise (high count rate) to where the counts (y-axis) of interest begin to decrease. The operating threshold setting is chosen in the middle of the flat portion (plateau) of the curve connecting the point-wise readings. A useful criteria to apply to the operating threshold setting is that small changes in gain will not change the output response by more than 1%. Use a source to obtain a count-rate appreciably higher than background.

The tail factor grade determination method results in an error when used for thin zones, especially high-grade zones. The error is caused by the thinness of the zone and the selection of the point Ei prior to deadtime corrections. The total-count method is recommended for all thin zones (see example 2 in the text).

261 DEFLECTION (INCHES)

FIG.3A. Step 2 of calibration factor determination.

DEFLECTION (INCHES)

FIG.4A. Measurement of E2 deflection (step 3). DEFLECTION (INCHES)

FIG.5A. Step 4, measuring the further deflections.

The steps required to obtain a calibration (k) factor from a log (figure 2A) by the tail-factor method are:

1. Determine the half-amplitude point ei (non-deadtime corrected Ei) at the upper boundary and record the deflection. A deflection is the distance from the reference baseline to a particular point on the curve. 2. Beginning at ei, mark off regular intervals (in this example, every 1/2 foot) and continued to,or beyond the lower boundary (figure 3A). 3. The point e2 is the next regular (1/2 foot) interval at or beyond the lower boundary half-amplitude point. Measure and record the e2 deflection (figure 4A). 4. Measure and record the deflections n^ (i = 1, 2, ... ) beginning with the first interval just after ei, that were marked in step 2 (figure 5A), but not including ez. 5. From the log or the log heading, find the deflection scale factor and convert the deflections ei, e2, and the n£ to count rate. 6. Using the measured deadtime (see appendix B), correct theei, e2, and n£ for deadtime using N = n/(l-nx) (figure 6A).

263 CALIBRATION WORK SHEET

n N CORRECTED* DEFLECTION SCALE FACTOR COUNT RATE COUNT RATE (inches! (counts/sec/inch) (counts/sec) (counts/sec)

Ei 1.98 X 5.000 9.900 9.993

E z 0.86 X 5.000 4,260 4,267

m 3.55 X 5.000 17.750 18.052

n2 3.92 X 5,000 19.600 19,969

n3 3.98 X 5;000 19.900 20.280

"« 3.98 X 5,000 19,900 20,280

3.97 X 5.000 19.850 20,228

no 3.95 X 5.000 19.750 20,124

3.72 X 5.000 18.600 18,932

"" 2.65 X 5.000 13.250 13,417

'Dead time correction, using

T= 0.942 • 10 * aeconde

in equation 2B

N = n

1 - nT

FIG.6 A. Dead time corrections.

7. Calculate Ft (Ei + E2)"using the deadtimercorrected values and add the results to the sum of the N-^. Ft is the tail factor. It will be different for each different interval used. If 0.5 foot intervals are used, Ft is 1.38. If 10 centimetres intervals are used, Ft =0.91. 8. Multiply the sum, the result of step 7, by the interval between each of the N£. In this example, a 1/2 foot interval is used (figure 7A).

The value of Ft is approximately

where i is the depth interval used (i.e. 0.5 foot, 10 cm, etc.).

264 AREA BY THE TAIL FACTOR METHOD

1.38 (E, + Ez) 19.679 Ni 18.052 N: 19.969 N3 20,280 N« 20.280 Ns 20,228 Ne 20,124 Ny 18.932 No 13.417 SUM 170,961 counts/second

AREA = SUM x INTERVAL A = 170.961 x 0.5 A = 86,481 counts - feet sec

FIG. 7A. Step 8, area determination.

9. Using the assigned value of GT for the standard calibration model, calculate k according to k = GyT/A.

In this exapmle, GT = 1.006% eU308 feet (from figure 1A), then

_ 1.006 _ 5 1,177 x 10" % eU3O0/count per second

A total-area method (see figure 3.6 in the text) may be easier to use than the tail-factor method for the determination of the calibration factor. However, its utility here is not straightforward to apply. In general, the method used for calibration should be the same one used to reduce field logs. To use the total-area method merely change steps 2, 3, 7, and 8 to read:

2. Beginning 0.5 metre (1.5 feet) before ei mark off regular intervals and continue to 0.5 metre (1.5 foot) past the lower boundary. 3. Eliminate 7. Eliminate 8. Multiply the sum by the interval between each point of Ni.

265 APPENDIX A

Deadtime Determination for Gross-Count Gamma-Ray Logging

It takes time, T for a system to process one event,from detection through to the recording of the event. During this interval, the system will not process another event because it is busy or "dead." Hence the common reference to the term "deadtime."

If, in one second, N gamma rays arrive randomly at the detector and the system only counts n of them, on the average, the system spends a fraction of one second, nx processing the n events so that an average of N(1 - nx) counts were counted during the one second. Thus,

N(1 - m) = n (IB) or, rearranging, N = V2— (2B) 1-nx yiere n is the detected count rate, N is the probable true count rate and the deadtime is x.

The deadtime must be measured for each gross-count gamma- ray system. One deadtime determination procedure is called the TWOPIT method (Crew et al., 1970) which utlizes pairs of calibration models. Another method, called the "two-source method", can be used to measure the deadtime. This latter method requires only two modest radioactive sources employed in a geometry that is easily reproduced.

Two-Source Method

When using the two-source method, one source is brought near the probe and the count rate ni is recorded (this corresponds to the actual count rate Ni at the detector). A second source is brought near the probe and the count rate ni2 recorded (corresponding to Ni2). The first source is removed , the count rate n2 is recorded (corresponding to N2). Finally the second source is removed and the background count rate n^ (Nb) is recorded. Then

(Ni - Nb) + (N2 - Nb) = (Ni 2 - Nb) (3B)

By substituting (2B)

ni , n2 ni2 1 1-nix l-n2x l-ni2x l^b - = 0 (4B)

266 This relation reduces to an equation, in deadtime:

(5B)

(6B)

b = -2(ni + n2 - ni2 nb) (7B)

c (ni + n2) - (ni2 + nb) (SB)

The general solution of (5B) is

-b ± (b2 - 4ac)% T (9B) 2a (the positive root is an unphysical result).

While there is sane latitude in choosing the sources for this method, sources of roughly equal strength are recommended. Always pay careful attention to the geometry; it should be such as to be readily reproduced. Also, avoid situations where the source gamna rays are scattered into the detector.

If the deadtime of the logging system is found to vary with count rate, a series of two-source determinations must be made over a wide range of count rates. Only one pair of approximately equal strength sources is required. Different count rates can be achieved by varying the source-to-detector distance.

WOP IT Method (Crew et al., 1970)

The TWOPIT method uses one or more pairs of calibration models. The calculational technique presented here is an approximate one. Consult the reference cited for the exact calculational technique. The basic TWOPIT procedure is to log each model of the pair; one model must have a relatively high uranium grade and the other a relatively low grade. The ratio

low uranium grade R high uranium grade (10B) should be less than about 0.2 for this method to work satisfactorily. The measurements should be taken under identical conditions. If the models have thick, homogeneous, mineralized zones, the static measurements in the center of either zone (or preferably, an average over the peak log anomaly)

267 FIG. IB. TWOPIT method: low-grade model, G = 0.4436% eU3Os.

can be used. For each pair of models, the deadtime T can be determined from the relation

T = Wt> (11B) where L is the average count rate in the low-grade model and H is the average count rate in the high-grade model.

An example using the TWOPIT procedure is outlined in the sequence that follows:

1. Measure the average deflection in the middle of the thick zone for each pair of high/low-grade models. The average deflection is the distance from the baseline to the flat part of the log. See figure IB for the log of the low- grade model, and figure 2B for the log of the high-grade model. 2. Convert each average deflection to count rate by multiplying the measured deflection by the appropriate scale factor. Thus, from figure IB,

L = 7.20 inches x 5000 counts/second/inch = 36,000 counts/second. From figure 2B,

H = 7.42 inches x 20,000 counts/second/inch = 148,000 counts/second.

268 FIG.2B. TWOPIT method: low-grade model, G = 2.05% eU308.

3. Using the appropriate ratio R, calculate the deadtime using equation 11B for each high/low model pair. In the example shown, R =0.444/2.05 =0.216. Substitute this value for

R, along with L and H, into equation 11B-6 . The deadtime is t = 0.000 000 942 seconds = 0.942 x 10 seconds. If more model pairs are utilized, calculate the average deadtime for all pairs and use the average.

269 APPENDIX A

Correction Factors for Gross-Count Gamma-Ray Logging

Certain conditions of the borehole and the formation require that a correction be made to the area value^determined under the log response,because the conditions differ from the standard. There are three categories of correction factors: those that apply to the logging system, the borehole conditions, and the formation conditions. The correction for the deadtime of the electronics has already been discussed in appendix B.

Borehole correction factors

The borehole conditions that require a correction factor are: borehole size/water (or drilling mud) factor Fw and the casing factor, Fc. There is no significant effect on the gross-count gartma-ray log as a function of borehole size when the borehole is air filled. The presence of water (or mud) in the borehole will, however, significantly affect the log response. The water (or mud) attenuates the ganma rays coming from the formation. The determination of the borehole size/water factor requires a sufficiently thick, mineralized model or formation with different size boreholes. The procedure for the determination of any borehole size correction factor is to position the detector (probe) in the middle of each mineralized zone with and without water (or mud) in the borehole. Record the average count rate each time. It is important to keep the probe along the wall of the borehole, since the probe will likely ride along the borehole periphery during field logging. The ratio of the dry borehole average count rate to the wet borehole average count rate is the correction factor. Figure ICis an example of this correction factor for different probes in water. Because the fluid in the borehole attenuates the gamma-ray flux, the correction factor is always greater than one. The correction curves in figure 1C should extrapolate to a correction factor of 1.00 for a borehole diameter equal to the diameter of the probe.

The casing factor, Fc is obtained by a procedure similar to that for the borehole size/water factor correction. A section of casing (having a certain wall composition and thickness) about six times as long as the garrma-ray detector is slipped over the probe and centered at the ganma-ray detector to duplicate the sidewall condition encountered in the field. The average count rate in the mineralized zone is measured. The casing factor is the ratio of the average count rate in an air-filled borehole without casing to the average count rate in the air-filled borehole with the probe surrounded by the casing segment. An example of a casing factor for a type of iron casing is shown in

270 FIG AC. Water factor correction. Gross-count gamma-ray probe.

figure 2C. The casing correction factor for no casing is unity. For any finite thickness of casing wall, the correction factor is greater than one.

Formation correction factors

There are three formation parameters that affect the quantitative interpretation of the gross-count ganma-ray log: the average electron density, disequilibrium, and an abnormally large average atomic number (Z). These correction factors are based on changes relative to the calibration conditions.

The change in the average electron density accounts for a variation in probe response. This is due to Compton scattering of gamma rays by formation electrons. The area of the gross- count energy spectrum decreases as the bulk density increases. However, the shape of the gross-count energy spectrum does not change. (The condition of large average atomic number, hence anomalously large electron density, is excluded here but dealt with later on.)

271 DETECTOR TYPE ' Nol DETECTOR SIZE < 3.2 cm x 4.4 em (I !A» inch x I Vi inch) BOREHOLE FLUID ' AIR

UNFILTERED PROBE (E P 1.4

<(0 1.2 o I.I 1.0 1/8 inch= 3.2 mm

l/B 1/4 3/8 1/2 CASING WALL THICKNESS (inches)

FIG.2C. Casing correction factor. Typical iron casing. Gross-count gamma-ray probe.

In sedimentary environments it has been found easier to measure the in-situ moisture than the in-situ density. A simple algorithm based on moisture (relative to the calibration model moisture) has been employed to approximately compensate for probe response variations due to bulk density. The correction factor algorithm is : dry weight fraction of the standard model (1C) dry weight fraction of the formation or 100% ~ % H20 of the standard model (2C) 100% - % H20 ot the formation This correction factor is applied to the deadtime corrected area under the log.

The quantitative assessment of uranium by the measurement of natural gamma-ray activity from the daughters of uranium is based on the equilibrium distribution of the sources of the gamma-ray activity. About 2% of the gartma rays come from the uranium group ^^U through 2 3 01h) and 98% of the ganrna rays

272 come from the radium group (Z26Ra through 206Pb) in the uranium series. If the relative equilibrium concentration (secular equilibrium) is disturbed, the measured gamma-ray activity is no longer indicative of the uranium mineralization present in the formation. The degree to which this is so is defined as the disequilibrium factor, F, and will satisfactorily approximate the real disequilibrium amount:

"d - -<4D)

where %U308 represents the chemically determined assay grade and %eU30s represents the radictnetrically determined equivalent grade. The disequilibrium factor is best determined in the laboratory using samples obtained in the field. This correction is applied directly to the area determined from the log. Grade inppm maybe substituted for U308 .

The radiometric equivalent grade used in the determination of the disequilibrium factor is determined by the closed-can technique. Ore samples are pulverized (-20 mesh), dried and fully packed in tarred metal cans, sealed and then weighed. The time and net weight of each sample are recorded directly on the can. (Typical can dimensions are 10.2 cm in diameter by 5.4 cm high and 0.356 mm thick.) The reason for sealing the pulverized sample is to allow radon (222Rn) to build up to its equilibrium value with the radium (226Ra) in the sample. The half-life of radon is 3.8 days, so that after about six half-lives (22 days) the radon is nearly in equilibrium. Then the activity is measured and the radiometric equivalent grade is determined (by comparison with standard samples). The act of obtaining samples and the subsequent laboratory processing of the samples are additional causes of disequilibrium, thus the reason for the closed-can procedure. There is an alternative method which is described elsewhere (Scott et at., 1960) that permits the radiometric assay to be determined within a period of about two days.

A Z-effect correction is required when the average atomic number of the formation is appreciably different from that of the standard calibration condition. As the average atomic number of the formation increases, an absorption process (photoelectric effect) disproportionately decreases the low energy region (<500 keV) of the gross-count ganma-ray spectrum relative to the higher energy region of the spectrum (>500 keV). Large Z elements in sufficient concentrations are typical causes^ for example, barium (Z=56), an additive to some drilling muds, and uranium (Z=92).

A method that avoids grade underestimation due to the Z- effect is to use a filtered probe. The graded mechanical filter consists of concentric wraps of lead (3.5 mm), cadmium (1.6mm),

273 Average Unfiltered Grade (KellsOa) 0.2 0.4 0.6 0.8 1.0 1.2

20 40 60 80 100 (x1000) Average Count Rate (cps)

FIG.3C. Effect and correction.

and brass or copper (0.9 urn); the lead wrap is outermost. (The thicknesses indicated are approximate.)

To determine a correction factor curve for the Z-effect models (or formations) having known grades that span the grade range of interest are needed. The grades must be known relative to the standard calibration conditions. The models are logged under standard conditions with a mechanical filter probe. The filter can be wrapped around the outside of the probe casing for convenience. Care must be taken to ensure that the filter is longer than the ganma-ray detector so that it does an adequate job of shielding the ends of the detector from surface and near surface borehole activity. The models are also logged under standard conditions without a filter. Each set of data is corrected for deadtime losses. The correction factor is the ratio of unfiltered to filtered count rates.

In practice, the Z-effect correction can be applied to the gross-count unfiltered data at either of two stages: to the deadtime corrected count rates (point-by-point) or to the average grade determined from the GT analysis. A plot typical of the correction versus deadtime corrected, unfiltered grade, and

274 also versus deadtime corrected unfiltered count rate is shown in figure 3C. The curve through the points has been estimated.

In summary, the correction factors covered herein are: deadtime, borehole size/water, casing, density (moisture), disequilibrium, and Z-effect.

Except for the deadtime, each is independent from each other and each is applied to the measured log area by multiplication.

275 APPENDIX A

Glossary

SI base units are the metre (m), kilogram (kg), second (s), ampere (A), kelvin (K), candela (cd) and mole (mol).

Multiply Radiation units becguerel 1 Bq (= 2.7027x10"11 Ci) disintegrations per second (1 dis/s) 1.00 xl0° Bq u 10 curie (=3.7 x 10 dis/s) 1 Ci = 3.70 xlO Bq roentgen 1 R = 2.58 xlO-" C/kg gray 1 Gy 1.00 xlO0 J/kg) (= 2 rad 1 rad = 1.00 xlO- Gy 1.00 xlO"2 J/kg) rem (in radiation protection only) dimensions of

Mass

unified atomic mass unit 12 (1/12 of the mass of C) 1 u 1.661 xlO"27 kg pound mass (avoirdupois) 1 lbm 4.536 xlO-1 kg = 1 ounce mass (avoirdupois) 1 ozm = 2.835 xlO g ton (long)(=2240 lbm) 1 ton 1.016 xlO3 = 2 kg ton (short)(=2000 lbm) 1 short ton = 9.072 xlO kg 3 tonne (=metric ton) 1 t = 1.00 xlO kg Length

statute mile 1 mile = 1.609 xlO0 km yard 1 yd = 9.144 xlO-1 rn foot 1 ft = 3.048 xlO-1 m inch 1 in = 2.54 xlO-2 m mil (= lO"3 in) 1 mil = 2.54 xl0~2 irm

Area

hectare 1 ha = 1.00 xlO" m2 (statute mile)2 1 mile2 = 2.590 xlO0 km2 acre 1 acre = 4.047 xlO3 m2 yard2 1 yd2 = 8.361 xlO-1 m2 foot2 1 ft2 = 9.290 xlO-2 m2 inch2 1 in2 = 6.452 xlO2 inn2 barn 1 b = 1.00 xlO-28 m2

Multiply by to obtain

Volume

yard3 1 yd3 = 7.646 x 10"1 m3 foot3 1 ft3 = 2.832 x 10"2 m3 inch3 1 in3 = 1.639 x 10" nm' gallon (Brit, or Imp.) 1 gal (Brit) = 4.546 x 10-3 m3 3 gallon (US liquid) 1 gal (US) = 3.785 x 10-3 m litre 1 1 = 1.00 x 10-3 m3

276 Force dyne 1 dyn 1.00 x io-5 N kilogram force 1 kgf 9.807 X 10° N poundal 1 pdl 1.383 X 10"1 N pound force (avoirdupois) 1 lbf 4.448 X 10° N Density

3 3 -5 3 pound mass/inch3 1 lbm/in3 = 2.768 x 101 kg/m3 pound mass/foot 1 lbm/ft = 1.602 x 10 kg/m

Energy

3 British thermal unit 1 Btu = 1.054 X 10 J calorie 1 cal = 4.184 X 10°1 9 J electronvolt 1 eV * 1.602 X 10-7 J erg 1 erg = 1.00 X 10- J foot-pound force 1 ft-lbf = 1.356 X 10°6 J kilowatt-hour 1 kW-h = 3.60 X 10 J

Temperature, energy/area time

Fahrenheit, degrees-32 °F-32 ] 1 ( C

Rankine 2 R J 9 I K 2 1 Btu/ft2 s = 1.135x10"2 ' W/m2 1 cal/cm min = 6.973 x 10 W/m

International System of Units (SI)

A ampere \ ft ohm 2 * A Sngstrom Pa pascal(=N/m ) * a are rad rad(100 rad = 1 Gy) * atm atmosphere rad radian * bar bar R rontgen(l R = 258 uCi/kg) * b barn s second(of angle)

Bq becquerel S second(time) -1 cd candela S s iemens(= ohm ) C coulomb sr steradian Ci curie(1 Ci = 37 GBq) T tesla * d day t tonne(1000 kg) 0 degree(of angle) u unified atomic mass unit °C degree Celsius v" volt eV electronvolt W watt F farad Wb weber g gram Gy gray(l Gy = 1 J/kg) * ha hectare Prefixes H henry. 1

* h hour d (deci) 10"2

J joule c (centi) 10"3 K Kelvin m (milli) 10"

* approved for use with SI for the time being.

277 6 Hz hertz(= cycles per second) y (micro) 10"-9 kilogram n (nano) 10 12

1, ltr litre(one may now use ltr p (pico) 10-15 throughout if the ell and the one f (femto) 10" on the typeface are identical or a (atto) 10"lf similar, e.g. not 1 1 but 1 ltr) da (deka) 101 lm lumen h (hecto) 10' 3 lx lux k (kilo) 106 metre M (mega) 10 minute (of angle) G (giga) 109 min minute(time) T (tera) 1012 mol mole P (peta) 1015 N newton E (exa) 1018

278 LIST OF CONSULTANTS

J.K. Hallenburg Geophysical Consultant, 336 East 29th Street, Tulsa, Oklahoma 74114, USA

John Duray Subsurface Systems Department, Bendix Field Engineering Corp., Box 1569, Grand Junction, Colorado 81501, USA

V.L.R. Furlong Pancontinental Mining Limited, AMP Centre, 50 Bridge Street, Sydney, NSW, 2000 Australia

P.G. Killeen Radiation Methods Section, Geological Survey of Canada, Resource Geographical and Geochemical Division, Department of Mines and Resources, 601 Booth Street, Ottawa, Ontario, K1A 0E8 Canada

Scientific Secretary

P. Barretto Division of Nuclear Fuel Cycle, IAEA

279 HOW TO ORDER IAEA PUBLICATIONS

An exclusive sales agent for IAEA publications, to whom all orders and inquiries should be addressed, has been appointed in the following country:

UNITED STATES OF AMERICA UNIPUB, 345 Park Avenue South, New York, NY 10010

In the following countries IAEA publications may be purchased from the sales agents or booksellers listed or through your major local booksellers. Payment can be made in local currency or with UNESCO coupons.

ARGENTINA Comision Nacional de Energi'a Atomica, Avenida del Libertador 8250, RA-1429 Buenos Aires AUSTRALIA Hunter Publications, 58 A Gipps Street, Collingwood, Victoria 3066 BELGIUM Service Courrier UNESCO, 202, Avenue du Roi, B-1060 Brussels CZECHOSLOVAKIA S.N.T.L., Spalena 51, CS-113 02 Prague 1 Alfa, Publishers, Hurbanovo namestie 6, CS-893 31 Bratislava FRANCE Office International de Documentation et Librairie, 48, rue Gay-Lussac, F-75240 Paris Cedex 05 HUNGARY Kultura, Hungarian Foreign Trading Company P.O. Box 149, H -1389 Budapest 62 INDIA Oxford Book and Stationery Co., 17, Park Street, Calcutta-700 016 Oxford Book and Stationery Co., Scindia House, New Delhi-110 001 ISRAEL Heiliger and Co., Ltd., Scientific and Medical Books, 3, Nathan Strauss Street, Jerusalem 94227 ITALY Libreria Scientifica, Dott. Lucio de Biasio "aeiou". Via Meravigli 16, 1-20123 Milan JAPAN Maruzen Company, Ltd., P.O. Box 5050, 100-31 Tokyo International NETHERLANDS Martinus Nijhoff B.V., Booksellers, Lange Voorhout 9-11. P.O. Box 269, NL-2501 The Hague PAKISTAN Mirza Book Agency, 65, Shahrah Quaid-e-Azam, P.O. Box 729, Lahore 3 POLAND Ars Polona-Ruch, Centrala Handlu Zagranicznego, Krakowskie Przedmiescie 7, PL-00-068 Warsaw ROMANIA Ilexim, P.O. Box 136-137, Bucarest SOUTH AFRICA Van Schaik's Bookstore (Pty) Ltd., Libri Building, Church Street, P.O. Box 724, Pretoria 0001 SPAIN Diaz de Santos, Lagasca 95, Madrid-6 Diaz de Santos, Balmes 417, Barcelona-6 SWEDEN AB C.E. Fritzes Kungl. Hovbokhandel, Fredsgatan 2, P.O. Box 16356, S-103 27 Stockholm UNITED KINGDOM Her Majesty's Stationery Office, Agency Section PDIB, P.O.Box 569, London SE1 9NH

U.S.S.R. Mezhdunarodnaya Kniga, Smolenskaya-Sennaya 32-34, Moscow G-200 YUGOSLAVIA Jugoslovenska Knjiga, Terazije 27, P.O. Box 36, YU-11001 Belgrade

Orders from countries where sales agents have not yet been appointed and requests for information should be addressed directly to: ^ Division of Publications ^ International Atomic Energy Agency £ Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria

INTERNATIONAL SUBJECT GROUP: IV ATOMIC ENERGY AGENCY Chemistry, Geology and Raw Materials/Geology VIENNA, 1982 PRICE: Austrian Schillings 460,-