Comparison of abundances of chemical elements in mineralized and unmineralized sandstone of the Brushy Basin Member of the Morrison Formation, Smith Lake District, Grants uranium region, New Mexico
Statistical treatment of analytical data from the Mariano Lake and Ruby uranium deposits in the Smith Lake district, New Mexico, indicates that organic…
Public-domain full text preserved in the Mountain Man Mining Library. Original source: pubs.usgs.gov.
UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY Comparison of abundances of chemical elements in mineralized and unmineralized sandstone of the Brushy Basin Member of the Morrison Formation, Smith Lake District, Grants uranium region, New Mexico By C. T. Pierson , C. S. Spirakis , and J. F. Robertson Open-File Report 83-818 This report is preliminary and has not been reviewed for conformity with U.S. Geological Survey editorial standards and strati graphic nomenclature. U.S. Geological Survey, P.O. Box 25046, Federal Center, Denver, CO 80225
CONTENTS Page Abstract 1 Int roduct1 on 1 Analytical Data 1 Data Treatment 4 Observations and Conclusions 5 References Cited 7 ILLUSTRATIONS Figure 1. Chemical characteristics of the Mariano Lake deposit 2 Chemical characteristics of the Ruby deposit 3 TABLES Table 1. Statistical data for the Mariano Lake deposit 9 Statistical data for the Ruby deposit 28
ABSTRACT Statistical treatment of analytical data from the Mariano Lake and Ruby uranium deposits in the Smith Lake district, New Mexico, indicates that organic carbon, arsenic, barium, calcium, cobalt, copper, gallium, iron, lead, manganese, molybdenum, nickel, selenium, strontium, sulfur, vanadium, yttrium, and zirconium are concentrated along with uranium in primary ore. Comparison of the Smith Lake data with information from other primary deposits in the Grants uranium region and elsewhere in the Morrison Formation of the Colorado Plateau suggests that these elements, with the possible exceptions of zirconium and gallium and with the probable addition of aluminum and magnesium, are typically associated with primary, tabular uranium deposits. Chemical differences between the Ruby and Mariano Lake deposits are consistent with the interpretation that the Ruby deposit has been more affected by postmineralization oxidizing solutions than has the Mariano Lake deposit. INTRODUCTION This paper summarizes the chemical characteristics of 299 samples of mineralized and unmineralized sandstone collected from the Mariano Lake and Ruby uranium deposits in the Morrison Formation, Smith Lake district, New Mexico. The samples were collected by J. F. Robertson, N. S. Fishman, and R. L. Reynolds. These data place constraints on the chemistry of the genesis and alteration of primary tabular uranium deposits. The Smith Lake district lies in the western part of the Grants uranium region (Chenoweth and Holen, 1980). The organic-rich character of the Mariano Lake and Ruby deposits indicates they are of primary origin (Granger, 1968); both are found in sandstone in the lower part of the Brushy Basin Member of the Morrison Formation of Late Jurassic age. Additional information on the uranium deposits in the Brushy Basin Member in the Smith Lake district may be obtained from papers by Hoskins (1963), Ristorcelli (1980), Jenkins and Cunningham (1980), Sachdev (1980), Place and others (1980), and Fishman and Reynolds (1982a and b). ANALYTICAL DATA The chemical elements considered in the present study are shown on figures 1 and 2. Most of the analytical results were obtained by 6-step semiquantitative emission spectrography, although about 63 percent of the spectrographic data for the Ruby deposit was obtained by a semi-quantitative, inductively-coupled, argon plasma (ICP) method. Elements determined by other than spectrographic methods are the following: uranium values, obtained by delayed neutron analysis; equivalent uranium, measured by geiger counter; carbonate carbon, organic carbon, and sulfur, obtained by the induction furnace technique; arsenic, determined by a wet chemical method; and selenium, obtained by X-ray fluorescence.
ro NO SIGNIFICANT DIFFERENCE BETWEEN MINERALIZED AND ADJACENT UNMINERAL- Denotes presence of 30% or more qualified values in data -I-/Samples with uranium content equal to or greater than 100 ppm are arbitrarily considered "mineralized", whereas samples with uranium content less than 100 ppm are considered "unmineralized." Figure 1.--Chemical characteristics of the Mariano Lake Deposit
na Be ENRICHED IN MINERALIZED SANDSTONE DEPLETED IN MINERALIZED SANDSTONE NO SIGNIFICANT DIFFERENCE BETWEEN MINERALIZED AND ADJACENT UNMINERAL- Denotes presence of 30% or more qualified values in data -L/ Samples with uranium content equal to or greater than 100 ppm are arbitrarily considered "mineralized", whereas samples with uranium content less than 100 ppm are considered "unmineralized". Figure 2.--Chemical characteristics of the Ruby Deposit
DATA TREATMENT In order to compare the chemical characteristics of the Mariano Lake and Ruby deposits, the samples used in this study were divided into four arbitrary, partly overlapping groups for each deposit (tables 1 and 2). The groups, which range from mineralized to essentially non-mineralized, are defined with respect to uranium content as follows: 1) equal to or greater than 1,000 ppm (G1000); 2) equal to or greater than 100 ppm (G100); 3) less than 100 pom (L100); and 4) equal to or less than 20 ppm (L20). For the Mariano Lake deposit, 98 samples (including those of the G1000 group) contain more than 100 ppm uranium, and 73 samples (including those of the L20 group) contain less than 100 ppm uranium. The numbers of samples containing more than 1,000, or less than 20 ppm, are 54 and 7, respectively. For the Ruby deposit, 81 samples (including those of the G1000 group) contain more than 100 ppm uranium, and 47 (including those of the L20 group) contain less than 100 ppm. Of these, 46 contain more than 1,000, and 16 contain less than 20 ppm uranium. Most of the conclusions concerning apparent enrichment or depletion of the "mineralized" groups (G1000 and G100) as compared to the "unmineralized" groups (L100 and L20) were obtained by comparing the two nonoverlapping groups (G100 and L100). In some cases it was helpful to have the G1000 and L20 groups, where trends of enrichment or depletion generally tend to exhibit higher contrast than between the G100 and L100 groups. With the aid of a computer, the geometric means and deviations for each of the elements in the four groups were computed for each deposit (tables 1 and 2). The computations are straightforward for all elements for which no qualified values are present. Where qualified values are present, special methods described below were used. Qualified values are of the three following types: 1) element not detected (N); 2) element detected, but present in an amount less than the lower limit of determination (L) for that element; or 3) present in an amount greater than the upper limit of determination (G) for that element. For cases where data are either singly censored on the left (data contain only N or L values, or on the right (data contain only G values) of the normality curve, a method devised by Cohen (1959, 1961) and programmed by VanTrump, Jr. (1978) was used to calculate geometric means and deviations. In this procedure, log normality for the data is assumed, and the geometric means and deviations, which should be considered as estimates, are calculated from functions of the following quantities for each element: 1) the geometric mean and deviation of the unqualified values; 2) the numerical value of the limit of determination; and 3) the number of qualified values in the set of data for that element. In Cohen's method, N's are not distinguished from L's, and moreover, as the percentage of qualified values increases, the accuracy of the geometric mean and deviation decreases. Tables 1 and 2 present the numbers and types of qualified values as well as the geometric means and deviations, the number of analyses, the limits of detection, and the minimum and maximum values for each chemical element studied at the Mariano Lake and Ruby deposits. Where qualified values are present, the geometric mean may have a lower value than the minimum (unqualified) value given by the table. This apparent inconsistency arises because the technique of Cohen mathematically reconstructs that part of the frequency distribution curve that lies below the limit of analytical
determination and calculates a mean based upon this part of the curve plus the part determined by the unqualified values. Because the accuracy of a geometric mean estimated by Cohen's method decreases with an increase in the percentage of qualified values, an asterisk is appended whenever the percentage reaches 30 percent or more for a given element. No values for the mean or deviation are reported when the percent of qualified values is greater than 75 percent. The limits 30 percent and 75 percent are chosen arbitrarily because the usefulness of qualified data varies depending upon the precision and limits of detection of the analytical technique used, as well as upon the variability of element concentrations within the group of samples studied. In order to identify statistically significant differences between geometric means of elements among the various mineralized and unmineralized groups, a "t" test, described by Natrella (1963, Chapter 3, p. 26-28), was used. This test, which employs the mean and variance of the logarithms of the chemical data as well as the numbers of samples in each of the two groups being tested against each other, was made at the 95 percent confidence level. Where more than 75 percent qualified values are present, the "t" test was not used, but rather judgment as to relative enrichment or depletion was made on the basis of the percentages of qualified values present in the groups being compared. For example, beryllium in the Mariano Lake deposit has percentages of unqualified values of 67 percent, 44 percent, 0 percent, and 0 percent, respectively, in the G1000, G100, L100, and L20 groups. In this case no statistical tests could be made between mineralized and unmineralized groups because no calculations of mean and variance could be made for the L100 or L20 groups. Beryllium, however, was judged to be enriched in the mineralized sandstone as compared to the unmineralized sandstone because the former group contains a much greater percentage of samples in which beryllium is found in amounts above the detection limit of the analytical method used. OBSERVATIONS AND CONCLUSIONS Interpretations concerning relative enrichment, depletion, or lack of difference in elemental composition between mineralized (G1000 and G100 groups) and unmineralized (L100 and L20 groups) sandstone in the Mariano Lake and Ruby deposits are shown graphically in figures 1 and 2. Results of statistical comparisons of geometric means are listed below. 1) The following elements are relatively enriched in the G1000 and G100 groups at both deposits compared to unmineralized or weakly mineralized rock (L20 and L100 groups): arsenic, calcium, organic carbon, cobalt, copper, gallium, iron, lead, manganese, molybdenum, nickel, selenium, strontium, sulfur, uranium, vanadium, yttrium, ytterbium and zirconium; 2) beryllium, potassium, and sodium are enriched in mineralized rock at Mariano Lake, but not at the Ruby deposit; 3) aluminum, and magnesium are enriched at the Ruby deposit, but not at Mariano Lake; 4) titanium is depleted at the Ruby deposit, but not at Mariano Lake; 5) chromium is depleted in the mineralized samples at Mariano Lake, but data are inadequate for comparisons of the Ruby deposit; 6) lanthanum shows essentially no change among the groups at the Ruby deposit, but data are inadequate to make a determination for the Mariano Lake deposit; and 7) scandium is probably enriched at Mariano Lake, but data are inadequate for a determination at the Ruby deposit.
Some of the differences in the chemical characteristics of the Mariano Lake and Ruby deposits suggest that the latter has been more extensively affected by oxidizing solutions than the former. Arsenic, iron, sulfur, and selenium are included in pyrite. Thus the lesser amounts of these elements at the Ruby deposit indicate a lesser pyrite content for the Ruby ore. Removal of pyrite by oxidation is a means of accounting for the lower pyrite content. The higher barium content of the Ruby deposit may also be explained by the oxidation process postulated above. Such, oxidation would produce sulphate. Because barium is extremely insoluble in the presence of sulphate (Plummer, 1971), any barium in the oxidizing solution would be precipitated as barite. Although barium is more concentrated around the Ruby deposit (L100) than within the deposit, the barium contents of the Mariano Lake and Ruby deposits and rocks immediately around these deposits are much higher than the barium contents of Morrison sandstones in the southern San Juan basin remote from mineralization (Spirakis, Pierson, and Granger, 1981) and higher than in average sandstones (Turekian and Wedepohl, 1961). Thus, barium is enriched in these deposits. As noted by Fishman and Reynolds (1982b), authigenic calcite is more abundant at the Ruby than at the Mariano Lake deposit. The substitution of strontium for either barium in barite or calcium in calcite may be the reason that the Ruby deposit has a higher strontium content than the Mariano Lake deposit. The higher manganese content of the Ruby deposit probably also reflects the higher calcite content. The relative lack of molybdenum and vanadium in the Ruby deposit compared to the Mariano Lake deposit may be a reflection of the leaching of these elements by the oxidizing solution. All of these differences between the deposits are consistent with the interpretation that the Ruby deposit is more oxidized than the Mariano Lake deposit. In order to determine which elements are characteristically enriched in primary tabular uranium deposits, the group of elements enriched in the Smith Lake district was compared to the group of elements enriched in the primary deposits of the Ambrosia Lake area (Spirakis and others, 1981) as well as to the group of elements enriched in primary tabular deposits in the Morrison Formation elsewhere on the Colorado Plateau (Shoemaker and others, 1959). These comparisons suggest that the elements typically enriched in primary tabular deposits of the Morrison Formation include aluminum, arsenic, barium, organic carbon, cobalt, copper, iron, lead, magnesium, manganese, molybdenum, nickel, selenium, sulphur, strontium, uranium, vanadium, and yttrium. The precipitation of molybdenum, selenium, uranium, and vanadium might be due to reduction or adsorption by organic carbon. Arsenic, cobalt, copper, iron, nickel, some selenium, and sulfur probably occur mostly as sulfides in the ores. The enrichment in yttrium may be the result of the substitution of yttrium for uranium in uranium minerals (P. Hansley, oral commun., 1982). The formation of chlorite and clay minerals in the ores may account for the enrichment of aluminum and magnesium. Some of the barium, manganese, and strontium enriched in the ores might occur in clay minerals, but in at least some of the deposits, the enrichment is related to the formation of post-ore sulphates and carbonates and is not necessarily produced by primary oreforming processes. The enrichment of lead is probably due to the radioactive decay of uranium, thus it also is not related to the primary ore-forming process. Details of element concentration and residence for the Mariano Lake deposit are given by Fishman and Reynolds (1982a, p. 13-15).
REFERENCES CITED Chenoweth, W. L., and Holen, H. K., 1980, Exploration in the Grants uranium region since 1963, Rautman, C. A., compiler, Geology and mineral technology of the Grants uranium region, 1979: New Mexico Bureau Mines and Mineral Resources, Memoir 38, p. 17-21. Cohen, A. C. Jr., 1959, Simplified estimates for the normal distribution when samples are singly censored or truncated: Technometrics, v. 1, no. 3, p. 217-237. , 1961, Tables for maximum likelihood estimates; singly truncated and singly censored samples: Technometrics, v. 3, no. 4, p. 535-541. Fishman, N. S., and Reynolds, R. L. , 1982a, Origin of the Mariano Lake uranium deposit, McKinley County, New Mexico: U.S. Geological Survey Open-File Report 82-888, 52 p. , 1982b, Diagenesis of the Morrison Formation, Smith Lake uranium district, McKinley County, New Mexico: U.S. Geological Survey Workshop on Diagenesis, Golden, Colorado, March 16-18, 1982, p. 18. Granger, H. C., 1968, Localization and control of uranium deposits in the southern San Juan Basin mineral belt, New Mexico--An hypothesis, in Geological Survey research 1968: U.S. Geol. Survey Prof. Paper 6UtT-B, p. B60-B70. Hilpert, L. S., 1969, Uranium resources of northwestern New Mexico: U.S. Geological Survey Professional Paper 603, 166 p. Hoskins, W. G. , 1963, Geology of the Black Jack No. 2 mine, Smith Lake area, j_n_Kelley, V. C., chairman, Geology and technology of the Grants uranium region: New Mexico Bureau Mines and Mineral Resources Memoir 15, p. 49-52. Jenkins, J. T. , and Cunningham, S. B., 1980, Depositional environment of Brushy Basin Member, Morrison Formation, in Gulf Mariano Lake mine, McKinley County, Rautman, C. A., compiler, Geology and mineral technology of the Grants uranium region, 1979: New Mexico Bureau Mines and Mineral Resources, Memoir 38, p. 153-161. Natrella, M. G., 1963, Experimental statistics: U.S. National Bureau of Standards Handbook 91. Place, J., DellaValle, R. S., and Brookins, D. G., 1980, Mineralogy and geochemistry of Mariano Lake uranium deposit, Smith Lake district, in Rautman, C. A., compiler, Geology and mineral technology of the Grants uranium region, 1979: New Mexico Bureau Mines and Mineral Resources, Memoir 38, p. 172-184. Plummer, L. N., 1971, Barite deposition in central Kentucky: Economic Geology, v. 66, p. 252-258.
Ristorcelli, S. J. , 1980, Geology of eastern Smith Lake ore trend, Grants mineral belt, Rautman, C. A., compiler, Geology and mineral technology of the Grants uranium region, 1979: New Mexico Bureau Mines and Mineral Resources, Memoir 38, p. 145-152. Sachdev, S. C., 1980, Mineralogical variations across Mariano Lake roll-type uranium deposit, McKinley County, jin_ Rautman, C. A., compiler, Geology and mineral technology of the Grants uranium region, 1979: New Mexico Bureau Mines and Mineral Resources, Memoir 38, p. 162-171. Shoemaker, E. M., Miesch, A. T., Newman, W. L., and Riley, L. B., 1959, Elemental composition of the sandstone-type deposits, Garrels, R. M., and Larsen, E. S., 3d, compilers, Geochemistry and mineralogy of the Colorado Plateau uranium ores: U.S. Geological Survey Professional Paper 320, p. 25-54. Spirakis, C. S., Pierson, C. T., and Granger, H. C., 1981, Comparison of the chemical composition of mineralized and unmineralized (barren) samples of the Morrison Formation in the Ambrosia Lake area, New Mexico: U.S. Geological Survey Open-File Report 81-508, 43 p. Turekian, K. K. , and Wedepohl, K. H., 1961, Distribution of .elements in some major units in the earth's crust: Geological Society America Bulletin, v. 72, p. 175-192. VanTrump, Jr., George, 1978, The STATPAC system user's guide for D0010 Fisher K-statisties, U.S. Geological Survey Administrative Report, 6 p.
Table 1. Statistical data for the Mariano Lake deposit EXPLANATION Types of analyses: S -- emission spectrograph.y Chem -- wet chemical method Ind induction furnace XRF -- X-ray fluorescence DNA -- delayed neutron analysis Qualified values: N -- element not detected L -- element present in an amount less than the lower determination limit G -- element present in an amount greater than the upper determination 1 imit Geometric mean and deviation: 1) Where no qualified values are present, the geometric mean and deviation are respectively the antilogs of the log mean and log standard deviation of the chemical data. 2) Where qualified data are present, the geometric mean and deviation are estimated by the technique of Cohen (1959, 1961). Inasmuch as the reliability of a geometric mean decreases as the amount of qualified values increases, an asterisk is appended when 30 percent or more qualified values are present. The geometric mean and deviation are not reported (two asterisks) when 75 percent or more qualified values are present. Three asterisks indicate that no data are available. 3) In some cases where qualified values are present, the geometric mean estimated by Cohen's technique may be less than the minimum value in the data. This will depend upon the number of qualified values, the geometric mean and deviation of the unqualified values, and the limit of determination of the analytical method used. Note: The limit of determination given is that of the lower limit unless G values are present in the data. In these cases, both upper and lower limits are given.
A1%-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower determination limit--. G1000 G100 Lower As ppm-Chem determination limit--! G1000 G100 05% ppm
Ba-ppm-S Lower determination limit--! ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 10,000 10,000 5,000 2,000 Comments The difference in the abundances of barium between the mineralized rock and the adjacent unmineralized (U <100 ppm) rock is not statistically significant at a 95 percent confidence level. The geometric means of barium in and around this deposit, however, are much higher than the background value (560 ppm) for the Morrison Formation in the southern San Juan Basin (Spirakis, Pierson, and Granger, 1981) and higher than the barium concentration of average sandstones (Turekian and Wedepohl, 1961). Consequently, barium is considered as enriched in this deposit.
Be ppm-S Lower determination limit 1 ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 1.67* 1.27*
Comments The high number of N and L values in the data make the geometric means somewhat uncertain but the absence of detectable beryllium in the samples with less than 100 ppm U in contrast to the detection of beryllium in about half of the samples with more than 100 ppm U suggests that an enrichment of beryllium occurs in the mineralized rocks.
Org C%-Ind Local determination limit--. 01% Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 Carbonate G100
C%-Ind Lower determination limit--. 01% Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 .05* 8.33* G100
Ca%-S Lower determination limit--.05% Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Comments The low concentration of calcium relative to the Ruby deposit as well as to the Ambrosia Lake deposits (Spirakis and others, 1981) is consistant with the field observation that the calcite content of the Mariano Lake deposit is low.
Co ppm-S Lower determination limit--5 ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 1.27* 1.50* .49*
Comments The high numbers of N and L values in these data and the large geometric deviations indicate that the geometric means for cobalt are only approximate. The detection of cobalt in a higher proportion of samples with uranium greater than 100 ppm than in samples with uranium of less than 100 ppm suggests a possible enrichment of cobalt in uranium ore.
Cr ppm-S Lower determination limit 1 ppm Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100
Cu ppm-S Lower determination limit 2 Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 Fe%-S Lower determination limit--. Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 ppm 05%
Ga ppm-S Lower determination limitSample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Enrichment of gall of the relatively high G1000 ium in the geometric G100 Comments G1000 and G100 mean of the L20 K%-S -5 ppm groups is group. uncertain because Lower determination limit--0.1% Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100
Mg%-S Lower determination 1 Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 imit--approximately .03% G100 Comments Statistical tests enriched in the Mariano give no reason Lake deposit. Mn to believe that magnesium ppm-S Lower determination limi Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 1,500 G100 1,500 t--5 ppm has been
Mo ppm-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower determi G1000 3,000 nation limit G100 3,000 Na%-S Lower determination limit . Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 ppm 5.51* 1%
Ni ppm-S Lower determination limit--2 ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 1.82* 1.48*
Comments Formal statistical tests between the means of mineralized and unmineralized groups were not made because the latter has 75 percent or more qualified values; however, because the mineralized groups contain higher percentages of values above the limit of determination than do the unmineralized groups, it is concluded that Ni is relatively enriched in the mineralized groups.
Pb ppm-S Lower determination limit--10 ppm Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 Total S%-Ind Lower determination limit--. Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 01%
Sc ppm-S Lower determination limit--5 ppm Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 3.54* G100
Comments Scandium is probably enriched in the ore because higher percentages of values above the detection limit are found in the G1000 and G100 than in the L100 and L20 groups. Se ppm-XRF Lower determination Sample set G1000 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's limit approximately .3 ppm G100
Sr ppm-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's The data indicate Lower determination limit--2 G1000 G100 that Comments strontium is enriched i m-s Lower determination limit--, Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 ppm n the ore. ,01%
U ppm-DNA Lower determination limit--.] Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 1,049 63,600 4,402 G100 63,600 1,280 V ppm-S Lower determination limit--! Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 7,000 1,661 G100 7,000 L ppm ppm 1,000
Y ppm-S Lower determination limit 10 ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 8.71*
Comments Comparisons of the geometric means, the maximum values, and the proportion of samples with detectable amounts of yttrium between the groups of samples with more than 100 ppm uranium to the group of samples with less than 100 ppm uranium all indicate that yttrium is enriched in ore.
Yb ppm-S Lower determination limit--! Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 Zr ppm-S Lower determination limit--10 Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 1,500 ppm ppm
A1%-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower determi G1000 nation limi G100 t .05% As ppm-Chem Lower Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's determination G1000 1 imit--approximately 1 ppm G100
Table 2.--Statistical data for the Ruby deposit EXPLANATION Types of analyses: S emission spectrography Chem -- wet chemical method Ind -- induction furnace XRF -- X-ray fluorescence DNA -- delayed neutron analysis Rad -- radiometric counting Qualified values: N element not detected L -- element present in an amount less than the lower determination limit G -- element present in an amount greater than the upper determination limit Geometric mean and deviation: 1) Where no qualified values are present, the geometric mean and deviation are respectively the antilogs of the log mean and log standard deviation of the chemical data. 2) Where qualified data are present, the geometric mean and deviation are estimated by the technique of Cohen (1959, 1961). Inasmuch as the reliability of a geometric mean decreases as the amount of qualified values increases, an asterisk is appended when 30 percent or more qualified values are present. The geometric mean and deviation are not reported (two asterisks) when 75 percent or more qualified values are present. Three asterisks indicate that no data are available. 3) In some cases where qualified values are present, the geometric mean estimated by Cohen's technique may be less than the minimum value in the data. This will depend upon the number of qualified values, the geometric mean and deviation of the unqualified values, and the limit of determination of the analytical method used. Note: The limit of determination given is that of the lower limit unless G values are present in the data. In these cases, both upper and lower limits are given.
Ba ppm-S Determination limit--! ppm (lower); 5000 ppm (upper) Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 50,000 50,000 5,000 1,052.1 3,100 1,082.7 Comments The difference between the barium content of mineralized rock ppm) and unmineralized rock ppm) is not statistically significant. Barium, however, is depleted in highly mineralized rock ppm) compared to adjacent unmineralized rock ppm or U L20 ppm). These data suggest that barium is more concentrated around the ore than in the ore. Compared to sandstones remote from mineralization (Turekian and Wedepohl, 1961; or Spirakis, Pierson, and Granger 19£1) barium is enriched both in ore and in rocks near ore.
Be ppm-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower determination limit 1 G1000 G100 Lower Org C%-Ind determination limit--. G1000 G100
ppm 01%
Carbonate C%-Ind Lower determination limit-- Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 -.01%
Comments Although the percentages of qualified values are less than 75 percent for both the G1000 and G100 groups, the geometric means are not reported because of the extremely high geometric deviations. Ca%-S Determi Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's nation limi G1000 t .05% (lower); G100 12% (upper)
Co ppm-S Lower determination limit approximately 1 ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 3.92* 3.38* 1.29* 1.06* Comments The N and L values in the data make the geometric means somewhat uncertain but the differences in the means suggest an enrichment of cobalt in the ore. Cu ppm-S Sample set Lower determination limit--! ppm G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's
Fe%-S Lower determination limit--. 05% Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100
Sample set Ga ppm-S Lower determination limit 5 ppm G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 3.94*
Comments There are too many L values for gallium in unmineralized or weakly mineralized sets for the means to be accurately determined. The high numbers of L's in the unmineralized sets do, however, suggest an enrichment of gallium in the mineralized rocks. K%-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower determi G1000 nation limit--0. G100
La ppm-S Lower determination limit--20 ppm Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 26.4* determination G1000 G100 21.6* Mg%-S 19.6* 22.1* limit approximately .03% G100 .113*
Mn ppm-S 1 Lower determination limit--! ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 3,000 7,000 5,000 Comments 1 Data are from 6-step semi-quantitative spectrographic analyses. ICP semi-quantitative spectrographic analyses are not included because the determination limit for manganese was 200 ppm for the Ruby samples, which is much greater than the 1 ppm determination limit for the 6-step method. Moreover, there were 43 percent qualified values in the ICP data, but none in the 6-step results.
Mo ppm-S Lower determination limit--2 ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 1,200 4.38* 1,200 1.67*
Comments Because of the high percentages of qualified values in the L100 and L20 groups, no statistical tests can be made between them and the G1000 and G100 groups. However, molybdenum is judged to be enriched in the ore, albeit sporadically, on the basis of the very high maximum values in the ore as well as upon the lower percentages of qualified values present in the G1000 and G100 groups.
Na%-S Lower determination limit--0 Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 Ni ppm-S Lower determination limit--2 Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 .1% ppm
Pb ppm-S Lower determination limit--10 ppm Sample set G1000 G100 No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's 20.80* 5.90*
Comments The geometric mean of lead in the L100 set is based on too many L values to be accurate. However, the relatively low value of the estimated geometric mean and the higher proportions of L values in the unmineralized sets indicate an enrichment of lead in the ore.
Total S%-Ind Lower determination limit-- Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 determination G1000 3,641 G100 Se ppm-XRF .01% .011* .009* limit approximately 0.3 ppm G100 3,641
Comments A few very high values of selenium in the mineralized rock suggest a sporadic enrichment of selenium in the mineralized rock.
Sr ppm-S Lower determination limit Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 m-s Lower determination limit--. Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 G100 Ppm 01% Comments The data indicate a possible depletion of titanium in the mineralized rock ppm) compared to adjacent nonmineralized rocks (IK100 ppm). This depletion is statistically significant at a 95 percent confidence level but not at a 99 percent confidence level.
U ppm-DNA Lower determination limit--0.1 ppm Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's G1000 1,010 20,100 3,093 eU determination 1 G1000 4,400 2,599 G100 20,100 1,068 ppm-Rad imit--approximately 1 ppm G100 4,400 1,505
V ppm-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower determination limit--! G1000 G100 4,100 Lower 4,100 Y ppm-S determination limit--10 G1000 G100 ppm ,000 ppm 10.8* 9.0* Comments Yttrium may substitute for uranium in uraninite (P. Hansley, oral commun., 1982) and this may explain the enrichment of yttrium in the ore.
Yb ppm-S Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Sample set No. of analyses for this element Minimum value Maximum value Geometric mean Geometric deviation No. of N's No. of L's No. of G's Lower determination limit--! G1000 G100 Lower .95* 1.20* Zr ppm-S determination limit--10 G1000 G100 ppm .89* ppm 28.3*
2.50*