Search Documents

By W. A. Bruce

Studies of reservoir behavior of 12 Smackover limestone oil and condensate pools are presented. Buckner, Midway, McKamie, Magnolia, Mt. Holly, Schuler (Reynolds) and Village are treated in considerable detail and subjected to analysis with an electrical device for analyzing reservoir behavior. Results and comparisons of these analyses are presented. In addition to these reservoir-behavior studies, an attempt is made to assemble trans-missibility and storage data on the Smackover limestone formation from cores and logs taken in dry holes and producers. This together with the behavior of individual reservoirs leads to a better understanding of the formation as a porous aquifer. The ultimate purpose of this type of study is to improve predictions concerning the future behavior of these pools in such a system. Predictions are presented for several pools. Introduction In 1937 the Phillips Petroleum Co. drilled its J. D. Reynolds No. I to the Smackover limestone formation at Snow Hill, Arkansas. This well found oil in the porous, permeable, upper part of the Smackover, thereafter referred to as the Reynolds lime. Since then, six oil pools and six condensate pools have been discovered in the Reynolds lime, and more than 75 dry holes have explored parts of this formation. By the first of 1943, 44 million barrels of oil and 83 billion cubic feet of gas had been withdrawn from the various reservoirs. The original reserves estimated by the Arkansas Oil and Gas Commission were 342 million barrels of oil and rzoo billion cubic feet of gas. So far all of the production has been from anticlinal structures, most of which are asymmetric and have their axes roughly perpendicular to the dip of the formation top. This is indicated in Fig. I, which shows the pools, dry holes drilled, and contours on top of the Smackover.' These pools are of particular interest for reservoir-behavior studies because the production and drilling has been controlled, and in most cases the records of reservoir data are complete. In addition to the excellent production and pressure information available, more than one half of the wells drilled were cored and all of them were electrically logged. All of the pools show evidence of water drive. While this paper is in a sense a collection of reservoir-behavior studies for several different pools, there has been an attempt to go one step further and to show the importance of the iormation as a porous continuum or a single aquifer in which the withdrawals of fluids from any pool has an important effect on the behavior of all of the pools. An effort will be made to show that the water drives of these pools are dependent upon a much larger area of the Smackover limestone than that occupied by the petroleum accumulations.

Jan 1, 1944

By W. A. Bruce

Studies of reservoir behavior of 12 Smackover limestone oil and condensate pools are presented. Buckner, Midway, McKamie, Magnolia, Mt. Holly, Schuler (Reynolds) and Village are treated in considerable detail and subjected to analysis with an electrical device for analyzing reservoir behavior. Results and comparisons of these analyses are presented. In addition to these reservoir-behavior studies, an attempt is made to assemble trans-missibility and storage data on the Smackover limestone formation from cores and logs taken in dry holes and producers. This together with the behavior of individual reservoirs leads to a better understanding of the formation as a porous aquifer. The ultimate purpose of this type of study is to improve predictions concerning the future behavior of these pools in such a system. Predictions are presented for several pools. Introduction In 1937 the Phillips Petroleum Co. drilled its J. D. Reynolds No. I to the Smackover limestone formation at Snow Hill, Arkansas. This well found oil in the porous, permeable, upper part of the Smackover, thereafter referred to as the Reynolds lime. Since then, six oil pools and six condensate pools have been discovered in the Reynolds lime, and more than 75 dry holes have explored parts of this formation. By the first of 1943, 44 million barrels of oil and 83 billion cubic feet of gas had been withdrawn from the various reservoirs. The original reserves estimated by the Arkansas Oil and Gas Commission were 342 million barrels of oil and rzoo billion cubic feet of gas. So far all of the production has been from anticlinal structures, most of which are asymmetric and have their axes roughly perpendicular to the dip of the formation top. This is indicated in Fig. I, which shows the pools, dry holes drilled, and contours on top of the Smackover.' These pools are of particular interest for reservoir-behavior studies because the production and drilling has been controlled, and in most cases the records of reservoir data are complete. In addition to the excellent production and pressure information available, more than one half of the wells drilled were cored and all of them were electrically logged. All of the pools show evidence of water drive. While this paper is in a sense a collection of reservoir-behavior studies for several different pools, there has been an attempt to go one step further and to show the importance of the iormation as a porous continuum or a single aquifer in which the withdrawals of fluids from any pool has an important effect on the behavior of all of the pools. An effort will be made to show that the water drives of these pools are dependent upon a much larger area of the Smackover limestone than that occupied by the petroleum accumulations.

Jan 1, 1944

By R. D. Pehlke, J. V. Gluck

A study was made of the effect of small additions of metallic solutes on the thermodynamic activity of zinc in dilute solution with molten bismuth in the range 450o to 650°C. The experimental measurements were made in a multielectrode galvanic cell using. a fused LiCl-KCl electrolyte. The activity of zinc in binary solution exhibited a strong positive deviation from Raoult's law and obeyed Henry 's law to at least 0.050 mole fraction. The ternary addition of lead, gallium, or indium increased the activity of zinc. Cadmium or tin additions slightly decreased the zinc activity. Copper, mercury, silver, or antimony caused moderate decreases, while gold strongly decreased the zinc activity. The results are expressed in terms of Wagner's interaction parameter ejZn and a second-order cross-interaction parameter. The interactions for copper, gold, silver, and antimony showed a linear dependence with reciprocal absolute temperature, while the other interactions were essentially independent of temperature over the experimental range. Since tin and antimony decreased the zinc activity, the hypothesis of a purely periodic variation with atomic number was rejected. A semiquantitative explanation was possible with a free-electron model when relative electronegativities were used to express changes in the electron/atom ratio caused by the solute addition. Random solution and quasi-chemical models correctly predicted the majority of the interactions for which binary data were available to apply them. A knowledge of the thermodynamic interactions in liquid (and solid) metallic solutions is of considerable theoretical and experimental interest in the formulation of quantitative thermodynamic relations for metallurgical processes. The approaches to defining the interactions may be either phenomenological (i.e., "What mathematical function adequately describes observed behavior and permits extrapolation to new situations?") or theoretical, such as a specific model

Jan 1, 1968

By E. C. Franz

IN an effort to understand high cycle fatigue, as well as to study the mechanism of fracture in general, a number of researches have been undertaken whereby the fracture properties of a metal have been determined under conditions of repeated cyclic prestsain employing high strains with relatively few cycles. Liu, Lynch, Ripling, and Sachs' showed the effect of a definite strain history consisting of small increments of tensile or compressive prestrain upon the flow and fracture characteristics of metals after a limited number of repeated loadings. Liu and SachsW emonstrated by cyclic straining of the aluminum alloy 24s-T4 that the least reduction in ductility was obtained when the compressive stress was nearly equal to the tensile stress. Attempts have been made to correlate the reduction in ductility under such prestraining with conventional fatigue data, as was done by Liu and co-workers.' The following research was conducted to determine if such reduction as has been noted could be correlated with conventional fatigue data or whether the effect resulted from causes extraneous to the basic mechanism of fatigue failure such as strain aging. It might be pointed out in this connection that the reduction in ductility has occurred in alloys that might be susceptible to strain aging or similar behavior, such as alloy 24s-T4. In the present investigation, the test specimens were prepared from the aluminum alloy 75s-T6 and annealed electrolytic copper. Alloy 75s-T6 is a complex precipitation hardenable alloy, whereas electrolytic copper would not be expected to exhibit strain aging effects, particularly at room temperature. The experimental procedure consisted of prestraining test specimens repeatedly by increasing tension in small increments to various total prestrain values, followed by testing to failure in a tension test. The prestraining as well as the final testing was performed on a tensile machine using special fixtures9 0 maintain concentricity of the specimens. After prestraining, the specimens were unloaded as rapidly as possible and immediately reloaded so that the total time lapse between strains was held to a few seconds. Logarithmic or true natural strain was used in all calculations. The results are presented in Figs. 1 and 2. In Fig. 1, the retained ductility, area after prestrain area at fracture is plotted against the tensile prestrain, initial area area after prestrain as abscissa with the number of stress reversals executed to attain the required prestrain plotted as parameter. Several specimens that fractured during

Jan 1, 1955

By K. K. Kohli, S. S. Peng, R. E. Thill

Introduction Since the adoption of the Surface Mining Reclamation and Control Act of 1977, which mandates that surface subsidence be an intergral part of the underground coal mine design, there has been a sudden rise of demand on guidelines for minimizing surface subsidence and its associated structural damages. Unfortunately there is no sufficient amount of systematic subsidence data to fill such a gap, although considerable research has been devoted to the subsidence induced damages of surface structures in the past. However, it was said that many surface subsidence surveys have been performed and monitored by coal companies during the past twenty years, and that it will be a great source of information for subsidence engineering design if they can be collected and assembled together. An effort was therefore initiated in the Department of Mining Engineering, West Virginia University, under contract with the U. S. Bureau of Mines in December 1978 to contact each individual coal company in the Northern Appalachian Coal Field for any surface subsidence survey data. After 10 months' survey, the data collected include: 17 longwall panels and 2 room and pillar sections. The data in each panel and section cover the complete set of subsidence survey raw data, underground mine layout maps, surface topo maps, geological loggings showing the stratigraphic sequences and thickness of the overburden, and if available, locations and types of surface structures. On-site visits were followed to study the surface terraines and structures. Interviews were conducted with the owners of the surface structures in most cases about, if any, the history and characteristics of damages. Most subsidence surveys started with the first panel and ended with the second or sometimes, but rarely, the third panel. The first panel may be located in a virgin area but frequently it was adjacent to (on the tailentry side) a mined-out room and pillar section. It must be noted that all of the subsidence data is restricted only to the amount of vertical subsidence and that the horizontal displacement measurement was completely ignored. Primarily due to the fact that vertical subsidence is the most striking feature in a subsidence trough, which is mistakenly considered to be the only critical contributor to surface structural damages, and that a horizontal displacement survey is much more complicated and time-consuming.

Jan 1, 1981

By E. P. Abrahamson II

The effect of transition element binary solid-solution additions upon the recrystallization temperature of vanadium has been investigated. Cr, Fe, Co, Ni, Mo, Ru, Rh, Os, and IY lower the recrystallization temperature, while Ti, Mn, Zr, Cb, Pd, Hf, Ta, W, Re, and Pt raise it. The majority of elements have their greatest effect in less than 0.15 at. pet. Both the rate of change of recrystallization temperature with atomic percent solute and the limit of initial linearity are found to Correlate with the free atom grozdnd state outer electron configuration of the sollte element. THE work of Abrahamson and Grant1 and Abraham-son and Blakeney2 has shown that the solute's elec- tron configuration appears to be a prime determinant of both brittle-to-ductile transition and recrystallization temperature in dilute binary alloys. It has also been shown by Abrahamson3 that the limit of linearity, i.e., the limit of initial linear change of recrystallization temperature with atomic percent solute, is also a function of the electron configuration of the solute in iron base alloys. Because of the systematic nature of the solute electron configuration correlation, it is to be expected that some systematic variation might be observed upon changing the solvent. Thus, this study of recrystallization in vanadium base alloys is the first of a series aimed at defining the role played by the electron configuration of solute in determining the recrystallization characteristics of dilute transitional alloys. PROCEDURE All alloys were made using 99.8 pct V containing 0.011 C, 0.024 N, 0.08 0, 0.006 H, 0.035 Ta, and

Jan 1, 1962

By T. J. Smolik, Harman, D. W. Fuerstenau

By means of electrokinetic measurements, the surface properties of the aluminosilicate polymorphs (sillimanite, andalusite, and kyanite) and also mullite have been found to depend on the ratio of A10 bonds to SiO bonds in the surface. The Al-toSi ratio in the surface is dependent upon the crystal lographic properties of the mineral, which in turn determine coordination numbers of A1 and of Si, cleavage characteristics, and density of the various solids. Mullite with a greater bulk Al-toSi ratio has the highest zpc of the aluminosilicates. By treatment in dilute acids or by cycling between acidic and basic solutions, alumina is selectively dissolved from the surface of these orthosilicates. This shifts the apparent zpc to lower pH values as the fraction of SiO groups in the surface increases. Thus, surface properties can be altered by procedures that alter the A l-to-Si ratio in the surface. Flotation of these aluminosilicates results from electrostatic interaction of the collector ions with the mineral surfaces. As a result, flotation response shifts in accord with any shift in apparent zpc by pretreatment. Separation of minerals by flotation is based on differences in the physicochemical properties of their surfaces.'-3 For many minerals, adsorption of flotation collectors results from their functioning as counter ions in the electrical double layer at solid-water interface. In such systems, separation depends on differences in the zero point of charge (zpc) of the various minerals. The zpc is defined as the concentration of potential determining ions in solution at which the surface is uncharged. The potential determining ions for inorganic oxide-water interfaces are hydrogen and hydroxyl.4,5 This work is concerned with the effects of crystal structure, composition, and surface preparation on the surface properties and flotation behavior of aluminosilicates. Differences resulting from crystal structure were studied by measuring electrokinetic potentials and determining the flotation behavior of three different aluminosilicate materials having the same chemical composition (Al2O3 SiO2), namely andalusite, kyanite, and sillimanite. Furthermore, the effect of the Al2O3-to-SiO2 ratio on flotation behavior and surface properties was investigated by including synthetic mullite, 3A12O3 .2SiO2. Gaudin and co-workers 6,7 have clearly shown how crystal structure determines the native floatability of certain minerals. Those solids in which the surface is comprised mainly of broken residual bonds possess naturally hydrophobic surfaces. It was found by Bakakin8 that for a number of oxidized lead minerals, differences in floatability were due to the distance of Pb atom from the surface. Only limited work has been done on relating electrical properties of surfaces to crystal structure and composition. Healy et. a1.9 found that the zero points of charge of various manganese dioxides ranged from about pH 1.5 to pH 7.5, and these differences in the zpc's were interpreted in terms of the atomic packing in the various crystals. The research presented here clearly shows that crystal structure and chemical composition do affect the surface properties of aluminosilicates and, in addition, that these effects can be altered by changing conditions under which the surfaces are prepared. CRYSTAL STRUCTURE OF THE ALUMINOSILICATES Sillimanite, andalusite, and kyanite are composed of regionally metamorphosed rocks differing mainly in degree of metamorphism. All three minerals have the same chemical composition, Al2O3 - SiO2, of which sillimanite is the high temperature polymorph and kyanite and andalusite the lower temperature polymorphs. Andalusite and sillimanite crystallize in the orthorhombic system while kyanite crystallizes in the triclinic system. The structure of these three minerals is essentially chains of aluminum-oxygen octahedra linked laterally by silicon and aluminum. The arrangement of these chains in the three minerals is shown in Fig. 1.l4

Jan 1, 1967

By Claudia Gasparrini

Gold, silver and platinum-group elements occur normally in very low concentrations in their ores. Because their market value is in many situations high enough to justify their full recoveries, mineralogical studies before any routine recovery is attempted are particularly important for the description of their manner of occurrence and decision on the appropriate recovery techniques. Gold and silver, although sometimes treated by similar metallurgical processes, may require different approaches and treatments because of the different manner in which they occur in their ores: gold has a relatively simple mineralogy limited to native gold, electrum and less often tellurides. Silver has a very complex mineralogy including over 200 different minerals. On the other hand, gold is more often than silver recovered from ores where its fine grain size requires the breaking of the host minerals by chemical methods. Mineral, grain sizes and host minerals all affect the metallurgy of the platinum-group elements..

Jan 1, 1984

By H. S. Kalish, F. J. Dunkerley

INTRODUCTION AND PREVIOUS WORK THE determination of the low temperature tensile properties of tin and tin-lead alloys was initiated as part of an extensive research program on the phasial equilibria of tin binary systems below 13.2°C and the property changes associated therewith. A survey of the literature revealed that, although some room temperature tensile data were available for tin and a few tin-lead alloys, there existed no survey report of the low temperature tensile properties of these alloys. Accordingly, the tensile data for tin and a series of pure tin alloys containing from 0.01 to 50 pct lead were measured at eleven temperatures from -196 to +20°C and are presented here in survey form. In addition to the importance of these tensile data as a preliminary to certain fundamental research, this information promises also to be of commercial importance because of the 'extensive use now being made of tin-lead solders in refrigeration and gas liquefaction equipment. Occasional failures of the solders in low temperature service have been attributed by some metallurgists to low temperature mbrittlement, while others hold that failure results from the transformation of the tin-rich phase of the solder from the solid white (beta) to the powdered gray (alpha) form. From the data previously available in the literature, neither of these explanations can be completely corroborated. Cohen and van Lieshout1 have established the beta to alpha tin transformation at 13.2°C and have shown that strain accelerates the transformation. Further, they showed that although I pct lead retarded the transformation, it was not eliminated. It will be pertinent, then, to note whether the beta to alpha transformation of the tin-rich phase will occur in a short-time tensile test. Polanyi and Schmid2 studied the tensile . properties of tin single crystals at -185°C and report that tin is brittle at-this temperature. In 1946, Kostenets3 reported that pure tin was brittle at -196 and -253°C, while lead remained ductile at both these temperatures. He also studied the tensile properties of tin-lead alloys containing from 10 to 75 pct lead at -253, -196 and +17°C. The ultimate tensile strength increased more than two times in going from room temperature to -196°C and about three times that value on further cooling to -253°C. All the alloys were brittle at -6°C except those containing more than 75 pct lead. Even at -253°C, the 75 pct lead alloy retained 36 pct elonga¬tion in a 30 mm-ga. length. Creep is also an important phenomenon

Jan 1, 1948

By James Lorimer

THE topography at the Stan Trg mine facilitated early exploration by adits; in consequence adit levels were developed at horizons 865, 795, and 760 meters above sea level, and the levels in the mine 'are named according to their elevation. Exploitation started from the 760-m. level . upwards while simultaneously first steps were taken with regard to mining in depth. A deep dewatering adit 2.5 km. long now gives access to the deposit at the 610-m. horizon which also is connected with the 760-m. adit level by a three-compartment main hoisting shaft. Main levels have been developed internally from

Jan 1, 1936

By Linn Bradley

F. H. WILLCOX, Pittsburgh, Pa. (communication to the Secretary*). -We must keep in mind, in balancing the savings-to be anticipated by the most efficient combustion of gas, the best heat absorption by stoves or boilers, and by heat conservation of the thermal equivalent carried in. blast-furnace gas-against the cost in operation, maintenance, depreciation and interest charges of the equipment required to realize the savings, that in boilers and hot-blast stoves we are basing our economies upon two different, combustion phenomena. In a boiler, evaporating water at a heat efficiency of 70 per cent., approximately 47,550 B.t.u. is consumed per boiler horsepower. Taking the blast, furnace gas cited in the above paper, we are using about 38 ¾ lb. of cold-washed gas per boiler horsepower. Using the hot-cleaned gas, the same equivalent heat is furnished by 34 1b. of the gas , This leaves of the 38 ¾ lb. required in the case of cold-washed gas, 4 ¾ lb. or 6,650 B.t.u. available for raising more steam when dry-hot gas is used, which means that an equivalent amount of coal firing can be dispensed with. Per ton of iron, this figure 2/3 X 12,000/38.75 X 4.75 = 980 lb. gas or 1,372,000 B.t.u. or 100 lb. coal. At $1.50 per ton of coal, the net saving per ton of pig iron produced by the use of dry-hot over wet-cold cleaning will, therefore, be about $0.06 2/3. At a stove heating 8,500 lb. of air per toil iron from 150° F., one-quarter of the dry-hot cleaned gas, or 3,000 lb., would heat the air to 1,450° F. 3,000 X 1,400 X 0.70 = (8,500 X 0.255) = 1,300. The cold-wet gas would heat the same amount of air to 1,280° F. 3,000 X 1,225 X 0.70 = (8,500 X.0.255) = 1,130.

Jan 4, 1917

By B. A. Rogers

ACCORDING to petrographic investigations, 1-4 cooled steel furnace slags contain a number of substances that have been shown to be ferro-magnetic5,6 and hence capable of undergoing appreciable changes of susceptibility with temperature. Since any alteration of the suscepti-bility of a constituent affects the average susceptibility of the sample, information concerning the nature of the substance should be deducible from its magnetic behavior. The possibility of obtaining information concerning slags in this way led to the construction of the apparatus to be described. FUNDAMENTAL PRINCIPLES OF APPARATUS The most convenient method of observing changes in mass suscepti-bility? with temperature is the one employed by Curie7 in his original investigation of the changes in magnetic properties of substances at elevated temperatures. As the method has been described frequently in the literature and in textbooks, only a brief review of the fundamental principles need be given here, but the particular arrangements used in the present work require a more detailed description.

Jan 1, 1939

By Hans Lundberg

IN last year's report on the progress of geophysics, the airborne magnetometer was the featured new development. At that time only a relatively small number of surveys had been made. During 1947, the airborne magnetometer was used in making many surveys and is again one of the high lights of the year in geophysics. This report is mainly concerned with geophysics as applied to the search for minerals exclusive of petroleum. In 194,5 and 1946, the use of the magnetic method reached a peak, partly because of an active period of prospecting brought on by an active mining market, particularly in Canada.

Jan 1, 1948

By James M. Henderson, John B. Pfeiffer

The recently completed liquid sulfur dioxide plant at American Smelting and Refining Co.'s Tacoma copper smelter in Washington last April is one step by the firm to meet ambient air quality standards in the vicinities of its three US copper smelters. At a capital cost of $18.5 million, the liquid SO2 plant will work together with an existing sulfuric acid plant to treat all gases generated by the copper converters. Designed to handle up to 45,000 cfm of converter gas utilizing the Dimethylaniline (DMA) process, this plant is expected to recover about 51% of the total smelter input.

Jan 11, 1974

By John V. Beall

Many Americans will judge the energy crisis by the picture shrink on their TV screen. And they are right to make this assessment considering the large areas of the country with marginal generating capacity, (See Fig, 1). The Federal Power Commission (FPC) recommends 20-25% reserve generating capacity above peak summer loads and 15-20% as minimal. The New York metropolitan area is the most concentrated power service area in the United States, and a spot check with Consolidated Edison, serving this area, revealed that on Friday, August 18, the company was operating at 5% reserve.

Jan 10, 1972

By M. A. Bryant, J. C. Goodwin, T. B. Faddies, L. D. Kornze

The Mercur Gold Mine is located 56 km (35 air miles) southwest of Salt Lake City in north central Utah; In July 1981 Getty Mining Company made the decision to put Mercur into production. The first gold bar was poured in April 1983. This paper provides details on: 1) the history of the camp, 2) past production, 3) geology of the deposit, 4) the exploration and development program, 5) ore reserve estimates, 6) detailed geologic pit mapping, 7) production blasthole sampling techniques and 8) the use of geostatistics (kriging) in handling assays and gold extractions to produce the required mill feed.

Jan 1, 1985

By Roderick K. Davey

Introduction in any business, it is essential that we select those alternatives which are not only technically feasible, but will be the most profitable to the business m terms of corporate objectives Some of the studies and activities requiring support of financial analysis to select the alternatives which most closely satisfy the management objectives and technical considerations are as follows 1 ) Cutoff grade determination. 2) Ore reserve estimation. 3) Production forecasts (long- and short-term) for an operating mine, development of a new mine, or expansion of an existing mine. 4) Evaluation of haulage alternatives. 5) Evaluation of equipment.

Jan 1, 1979

By K. U. Snowden, J. N. Greenwood

EARLY studies by Gough and sopwith' have shown that the fatigue resistance of certain metals increased when the test was carried out in a partial vacuum; lead showed this effect to a marked degree. This communication describes surface deformation differences between high-purity lead fatigued in air and in a vacuum of approximately 5 X l0-3 mm Hg. Both air and vacuum tests were carried out by the same alternating flexure machine at a frequency of 500 cpm. Flat cantilever-bend specimens with tapered sides were cut from strip, annealed at 100°C, and chemically polished.

Jan 1, 1959

By T. Y. Yan, W. F. Espewheid

The various ion exchange reactions occurring during restoration of formations contaminated with ammonia during in situ uranium leaching are reviewed. Data are presented on the restoration of a previously leached laboratory packed column wing lime saturated brine. Although this restoration fluid selectively and effectively removed the ammonia, 24.5 pore volumes are required to reduce the ammonia concentration from 500-3 mg/L (29- 0.18 gr per gal). The essence of NaCl in the restoration fluid at levels in formation water is beneficial in improving the selectivity of the system for NH4+ removal and reducing Ca + + ion consumption

Jan 1, 1984

By R. Chanchani

It was reported earlier that francolite and dolomite from south Florida deposits cannot be selectively floated using conventional conditioning with sodium oleate as the collector. However, dolomite could be separated from francolite if the minerals were first conditioned at pH 10 followed by reconditioning at a lower pH. An optimum separation was achieved after reconditioning at a pH below 4.5. Reconditioning pH and collector concentration were determined to exert a major influence on the selectivity. Flotation, adsorption, and transmission infrared spectroscopic results are presented to elucidate the mechanisms underlying the observed selectivity.

Jan 1, 1986