Search Documents

By H. F. Seifert

THE classification and preparation of non-ferrous scrap metals is a subject of interest to every individual and corporation that employs in its processes of manufacture non-ferrous metals and alloys and accumulates in such manufacture scrap from these materials for disposition in the form of turnings, borings, short ends, punchings, clippings, grindings, washings, etc. All too frequently the disposition of such scrap is left to individuals who have no knowledge of the subject either from a metallurgical or a commercial standpoint, or both. Much money is thus lost which with a little more intelligent effort might be saved, and in this day of fierce competition and small profits such an opportunity cannot be ignored. The Mineral Resources of the United States for the year 1925, the latest issue available (Part 1, page 221), gives the secondary non-ferrous TABLE 1.-Some Secondary Non-ferrous Metals Recovered in the United States for the Year 1924-25 Material Short Tons Value Copper, including that in alloys other than brass.... 250,600 $ 71,170,400 Brass, scrap remelted 242,300 56,270,800 Lead as metal 112,4201 39,477,100 Lead as alloys 112,460 Zinc as metal 61,430 11,359,000 Zinc -in alloys other than brass 13,300 Tin as metal 1 7,950 35,165,400 Tin in alloys 23,000 Antimony as metal 1 } 3;794,000 Antimony in alloys 10,839 Aluminum as metal 17,700 24 Aluminum in alloys 26,300 ,816,000 Nickel as metal 19 1 1,518,000 Nickel in non-ferrous alloys 2,109 Total 882,600 $243,570,700

Jan 1, 1928

By N. M. Steven, J. M. Moore, J. L. Kuyper

Mineral exploration and research has led to the identification of four types of tungsten mineralization in the NE-trending, Iargelyensialic, inland branch of the late Proterozoic/early Paleozoic Pan-African Damara Orogen. The mineralization (mainly scheelite) occurs in a terrane (the Central Zone) that is characterized by multiple deformation, greenschist/amphibolite-facies metamorphism and numerous peraluminous S-type granitic and pegrnatitic intrusions. The four major types of scheelite mineralization are (in ascending stratigraphic order of host rock): (1) thin (5 cm to 2 meter) quartz+plagioclase+diopside ± scapolite calc-silicate granofels with fmely disseminated scheelite (scheelite-bearing granofels; SBG) developed as discontinuous intercalations within the calc-silicate rocks of the Khan, Spes Bona, Oberwasser and Kuiseb Formations (Nosib and Swakop Groups); (2) gamet+pyroxene± vesuvianite+scheelite ± fluorite replacement skarns hosted by Okawayo, Karibib and lower Kuiseb Formation marbles in the aureoles of late tectonic, unrnineralized leucogranites; (3) metaturbidite-hosted scheelite- (and possibly wolframite-) bearing tourmalinites, both stratiform and crosscutting, in the Kuiseb Formation located in the 'cleavage fronts' of late tectonic granitoids; (4) scheelite-bearing vesuvianite+quartz+garnet skarnoid rocks (several hundred meters in length) in the calcareous metaturbidites of the upper Kuiseb Formation. These scheelite occurrences were discovered by the follow-up of a large number of tungsten anomalies located during a regional stream sediment sampling program. Detailed mapping, ultra-violet surveys and rock sampling proved to be the most effective surface exploration techniques for locating mineralization; soil sampling and geophysics were of almost no use. The largest, continuous scheelite-bearing skarn of economic interest, Otjua, has a drill-indicated reserve of 250 000 tonnes at 0.53 wt5 WO3 and 8 wt% CaF2•

Jan 1, 1994

By Li Xibing

The In order to mine safely with reasonable cost, it is necessary to research the fracture characters of backfill and its matches with rock mass. Four kinds backfill with cement-tailing ratio of 0.250, 0.125, 0.100 and 0.083 respectively were tested, and basing on stress-strain curves tested, the deformation characters and fracture laws as well as the damage constitutive equations investigated with damage mechanics whose calculating results are coincided with experiment results well were proposed. The results of damage analysis show that there have different damage characters in different backfill, e. g. the lower the ratio of cement to tailings the lower the damage values when stresses reach to peak value; the lower the content of cement the more rapid in damage and the more abrupt in breakage process. In addition, using catastrophe theory, an energy model was proposed for backfill and rock mass acted each other in mechanics, and the unstable condition of the system was obtained. Research results show that the higher is the stiffness of rock mass the more stable the system, and the lower is the ratio of cement to tailings, the more unstable the system will be. Further more, according to the catastrophe model proposed, the reasonable matches between backfill and rock mass were researched, and a calculation formula that designs reasonable strength of backfill during mining was put forward. It has been proved by practice that the model and theory researched in this paper can guide filling design in mining. Keywords: Cemented tailings backfill, constitutive equation, catastrophe theory, match of backfill and rock mass

May 1, 2007

By J. D. Morgan

The US economy grew steadily in 1989, reaching an annual rate of $5.3 trillion in the last quarter (US trillion=1012 ). The value of domestically processed, mineral-based materials rose 5% to $315 billion (US billion=109). The value of metals produced from US ores rose 14% to $11.6 billion, while industrial minerals rose 2% to $20.2 billion. Recycled scrap was valued at $10 billion. Increasing economic activity stimulated metal prices. Yearly lows and highs were $2.31 and $3.56/kg ($1.05 and $1.62 per lb) for copper, $1.58 and $2.57/kg (72 and $1.17 per lb) for aluminum, 77 and 97 cents/kg (35 cents and 44 cents per lb) for lead, and $1.58 and $2.09/kg (72 cents and 95 cents per lb) for zinc. Many mining and metal companies reported reasonable profits. Consumer demand strongly affects mining because the US economy is saturated with long-lived products. Consumer, mortgage, and Federal debt continued to rise. 250 million US citizens already possess 172 million licensed motor vehicles and 100 million dwelling units. New car production fell 4% to 6.8 million units, while truck and bus production fell 1 % to four million units. New housing starts fell 9% to 1.4 million units. Imports supplied a significant proportion of total US consumption of several important mineral materials. Imports of crude and refined petroleum costing $50 billion exceeded domestic crude oil production. President Bush expressed "strong concerns about this country becoming further dependent on foreign oil." And the Department of Energy held a series of hearings around the country to develop ideas for a "national energy strategy."

Jan 1, 1990

By Kay JR

In 1984 a new Regional Geological Mapping Program (RGMP) was commenced by the Geological Survey of Queensland (GSQ); the new program was in response to the withdrawal of the Bureau of Mineral Resources from regional mapping. Areas for mapping were selected in northern and south-eastern Queensland on the basis of potential for mineral discoveries, level of company interest, inaccuracies in the first edition mapping and interests of other organisations (government and universities) in joint projects.Geological provinces of particular interest in northern Queensland are the Hodgkinson Basin, parts of the Georgetown Inlier not covered by earlier BMR-GSQ mapping, Broken River Embayment, Burdekin Basin, Lolworth-Ravenswood Block, Campwyn Block and Anakie Inlier. The 1:250,000 sheets covering these areas of interest are shown in Figure 1 including those already completed, those in progress and those planned for future investigation.A flow chart showing the various phases for RGMP projects is shown in Figure 2. Data collection has two facets: (i) field recording and (ii) office work including air photo interpretation, study of company reports, literature research, petrography and various laboratory studies. Two main strands of information collected are represented by a data base and compilation sheets. Compilation sheets are generally at 1:25,000 scale and are available for purchase. The current data base is very fragmentary and can be accessed only by direct contact with the relevant geologist or via written reports. A more accessible field data base is in the process of being introduced this year; this field data base (REGMAP) is based on standardised data recording and computer storage.

Jan 1, 1986

By Liang Shen

The Cooper River Bridge, the longest cable-stayed bridge in North America, replaces two existing bridges along U.S. Highway 17 and connects the City of Charleston and the town of Mount Pleasant, South Carolina. Two diamond shaped towers support the 1,546 ft long main span over Cooper River and allows for a 1000 ft wide navigation channel. The Charleston high-level approach is 4351 ft long and provides a 250 ft wide navigation channel for Town Creek, while the Mount Pleasant high-level approach is 2,090 ft long. Due to the high seismicity in the region, potential vessel impact, and hurricanes, large lateral loads are expected. From various types of foundations, 10-foot diameter drilled shafts were selected for their high capacity against vertical and lateral loadings. The shafts are founded on a stiff to very stiff lightly cemented calcareous sandy clay or sandy silt layer, called Cooper Marl, which underlies the coastal sediments of the interbedded clays and sands. A group of 11 shafts supports each main tower, while a rock island protects it from vessel collision. A group of 4 to 8 shafts foundations support the other piers of the main span unit. A pair of shafts supports the High Level Approach piers. The shaft tip elevations vary from 120 ft to 230 ft below water line. An elaborate geotechnical investigation was conducted on the site, including a load test program on full-scale shafts under axial and lateral loads. The results of the tests were used to calibrate the lateral and axial soil resistance. The objective of this paper is to present the different aspects of the bridge foundation design.

Jan 1, 2013

Filtration of pulps of varying densities is an essential operation in the hydro-metallurgical production of zinc. The use of filters naturally applies only to the "wet" plants, and in these varying types and sizes are in commission. At Risdon, the Moore suction filter is used for the heavier pulps, except in one instance, where a modification of the Butters-Cassell filter is installed.Filter presses are in general use, and of these, one is a washing unit; the plates, and frames are of cast iron or mild steel, except in one case which is a centre feed press having wooden frames.FILTER DESIGNThe problem of filtration presented itself very early in the history of Risdon and the first developments occurred in what was termed the 250 -lb. plant. Here experiments were made with many types of filters, and as this was the pilot plant on which the design of the 10 ton plant (10 tons of zinc per day) was based, quick decisions were necessary.An Oliver type filter was chosen for the heavier pulp or residue, and a Shriver wooden frame filter press for the lighter pulps.To this day, but with a number of variations, filter presses continue to function unopposed in their particular field; however, the Oliver filter was soon superseded because it was found impossible to wash the filter cake efficiently without repulping; consequently its use was abandoned and the Moore filter adopted.Both- these filters have been modified and developed in many ways to meet the requirements of a zinc plant, but they retain their original basic features.

Jan 1, 1949

By G. St. J. Perrott, N. A. Tolch

"The measured strength of an explosive depends to a certain extent on which it is determined. An earlier investigation showed the relative propulsive strength of dynamites as determined in one type of apparatus, ballistic pendulum.4 The aim of this paper is to precept strength data on these same dynamites determined in another type of apparatus, the Trauzl lead block, and to compare the results with those of the ballistic pendulum. Trauzl lead block have been in use many years for the determination of thy strength of explosives, under moderate confinement.5 The common practice is to shoot 10 grams of explosive, in tinfoil wrapper,- and then to measure the volume of the cavity produced by the shot. The volume of the original borehole is deducted from the gross expansion and sometimes a correction is made for the expansion produced by the detonator. Dry sand is used for stemming the charge.It has been found that the expansion in Trauzl blocks is not strictly proportional to the charge. Naoum5 believed that for scientific accuracy the charge should undoubtedly be regulated to give the same average expansion for each sample tested, but for expansions of 200 to 400 cubic, centimeters the deviation from proportionality is not sufficient to give results appreciably in error.Neubner7 found that the proportionality curve is tangent to the charge-expansion curve at about 300 cubic centimeters expansion, provided that no deductions are made for the volume of the original borehole or the effect of the detonator; the total or gross expansion is proportional to the charge between expansions of 250 to 350 cubic centimeters without serious error."

Oct 1, 1930

By P. T. Bissert, T. J. Lutz, J. A. Yonkey, G. T. Homce

"The U.S. National Institute for Occupational Safety and Health (NIOSH) has been researching refuge alternatives (RAs) since 2007. RAs typically have built-in pressure relief valves (PRVs) to prevent the unit from reaching unsafe pressures. The U.S. Mine Safety and Health Administration requires that these valves vent the chamber at a maximum pressure of 1.25 kPa (0.18 psi, 5.0 in. H2O), or as specified by the manufacturer, above mine atmospheric pressure in the RA. To facilitate PRV testing, an instrumented benchtop test fixture was developed using an off-the-shelf centrifugal blower and ductwork. Relief pressures and flow characteristics were measured for three units: (1) a modified polyvinyl chloride check valve, (2) an off-the-shelf brass/cast-iron butterfly check valve and (3) a commercially available valve that was designed specifically for one manufacturer’s steel prefabricated RAs and had been adapted for use in one mine operator’s built-in-place RA. PRVs used in tent-style RAs were not investigated. The units were tested with different modifications and configurations in order to check compliance with Title 30 Code of Federal Regulations, or 30 CFR, regulations. The commercially available relief valve did not meet the 30 CFR relief pressure specification but may meet the manufacturer’s specification. Alternative valve designs were modified to meet the 30 CFR relief pressure specification, but all valve designs will need further design research to examine survivability in the event of a 103 kPa (15.0 psi) impulse overpressure during a disaster. BackgroundThe Mine Improvement and New Emergency Response Act of 2006, known as the Miner Act (U.S. Department of Labor, 2006), was enacted in the wake of three mine accidents involving explosions or fire that claimed 19 lives that year. Intended to help improve underground coal mine accident preparedness, the Miner Act includes provisions that target mine safety issues in areas such as emergency response planning; adoption of new technology; training and education; and mine safety standards enforcement. Section 13 of the Miner Act specifically directed the U.S. National Institute for Occupational Safety and Health (NIOSH) to provide for research into the effectiveness and viability of refuge alternatives (RAs) for underground coal mines, and the U.S. Department of Labor to act on the results of such research, as appropriate. These mandates culminated in the 2009 adoption of changes to the Title 30 Code of Federal Regulations, or 30 CFR, mining health and safety regulations (U.S. Department of Labor, 2008), requiring underground coal mines to supply mine emergency RAs and associated components so as to provide a life-sustaining environment for persons trapped underground. Such RAs can be either selfcontained mobile units or built-in-place facilities. The regulatory changes also include provisions establishing requirements for the approval of RAs and their components by the U.S. Mine Safety and Health Administration (MSHA), among which are numerous criteria for providing a safe breathable atmosphere under positive pressure within the RAs. One criterion for maintaining a safe RA atmosphere is the inclusion of an air pressure relief valve that will activate at a maximum of 1.25 kPa (specified as 0.18 psi, or approximately 5.0 in. H2O), or as specified by the RA manufacturer, above mine atmospheric pressure in the RA (U.S. Department of Labor, 2008)."

Oct 1, 2016

By D. G. Westlake

Resistance was measured as a function of temperature, 77° to 350°K,for niobium with hydrogen concentrations up to 3.76 at. pct. Cooling curves exhibited a discontinuity in slope that was attributed to the initiation of hydride formation. A thermal hysteresis was explained in terms of diffusion and the effects of mechanical constraint on nucleation and growth. The heat of solution of hydrogen in saturated Nb-H solid solution was 2. 737 * 0.045 kcal per g-atom, including an unknown stress contribution. The critical ternperatures for hydride formation during cooling and for embrittlement are correlated but not coincident. This, too, has been explained in terms of mechanical constraint and stress-itlduced tratzsfomtation. Nb-H alloys undergo a ductile-brittle transition with decreasing temperature.'-3 Similar behavior has been noted for v-H.1,5 For the latter system, there is evi-dence6 suggesting coincidence of the temperature of minimum ductility and the critical temperature T, for initiation of metal hydride precipitation during cooling. A decrease in ductility due to the presence of brittle hydride particles is readily explainable. The fact that ductility begins to decrease at some temperature higher than T,4,5 has led us6 to suggest that precipitation may be strain-induced locally during tensile testing. Wood and Daniels' have reported that a plot of temperatures for 10 pct reduction in ductility vs hydrogen concentration in niobium is not coincident with, but parallels, the solvus.7,8 They believe this indicates a correlation of precipitation and embrittlement. Unfortunately, the two reported solvuses 7,8 to which Wood and Daniels refer are not in agreement with each other. Furthermore, neither solvus was determined experimentally: each was a plot of calculated solubilities. Thus. neither is highly dependable for determining whether embrittlement coincides with initiation of hydride precipitation. Longson 9 studied phase relations in the Nb-H system between 77° and 473°K by dilatometry. Comparison with metallographic observations convinced him that discontinuities in the slopes of his dilatometric plots were coincident with the initiation of hydride precipitation. Plotting the logarithm of the hydrogen concentration vs the reciprocal of the temperature at which precipitation was initiated during cooling, he obtained a straight line shown in Fig. 1. Walter and chandler,'' using X-ray techniques to detect the for mation of a second phase? have obtained three experimental points in the temperature range 295° to 338°K which fall within Longson's experimental error for this plot. A similar plotg of temperatures at which dissolution was completed during heating was a parallel line shifted toward higher temperatures. A straight-line plot was obtained, also, for the temperatures of the ductile-brittle transitions,9 but it was shifted toward still higher temperatures. Longson calculated an activation energy of 2.980 * 0.300 kcal per mole for the hydrogen embrittlement mechanism, compared with 3.180 * 0.400 kcal per mole for hydride formation. Despite this agreement, he concluded, "It seems unlikely that embrittlement results from hydride formation..." because hydrogen embrittlement was observed at temperatures above the solvus. It is our contention that embrittlement is correlated with precipitation of the hydride.6 Because only Longson has investigated Nb-H phase relations at sub-ambient temperatures, we deemed it advisable to re-determine the temperature dependence of the solubility using the resistometric technique already proven successful for the V-H system.6 Coincidence of our solvus with temperatures for embrittlement would

Jan 1, 1970

By M. J. Pool, with Appendix by Larry Kaufman, R. W. Krenzer

Thermodynmic properties of the Cu-Mn system have been deterrrzined in the temperature range 973" to 1273°K by measuring the vapor pressure of manganese in equilibrium with alloys of compositions varying from 0.13 to 0.91 rnole fraction manganese using the Knudsen effusion technique. Activities of manganese calculated from the measured vapor pressures show large positive deviations. Calculated heats and entropies of mixing are positive with a maximum occurring at approximately 0. 7 mole fraction manganese. ThE investigation of thermodynamic properties of binary systems involving transition metals is of considerable interest at the present time. The incomplete subvalency electron levels of transition metals lead to unusual behavior in some alloys, and considerable information can be obtained by determining the thermodynamic characteristics of these systems and relating the thermodynamic behavior to observed physical properties. The Cu-Mn system, shown in Fig. 1,' provides an example of an alloy system in which the determination of the equilibrium phase diagram and various physical properties has received considerable attention, but very little work has been done on measurement of the thermodynamic properties. Elemental copper and manganese have comparable electronegatives, atomic sizes, and cohesive energies, but differ in crystal structure. They form a continuous series of solid solutions only at high temperatures and the structure varies in a continuous manner from fcc at the copper-rich end to fct at the manganese-rich end.' The purpose of this investigation was to measure the vapor pressure of manganese over a series of Cu-Mn alloys and from these values determine the thermodynamic properties of the binary system. Vapor pressure measurements were made in the temperature range 973" to 1273"K, remaining within the solid-solution region for each alloy studied. EXPERIMENTAL PROCEDURE The relative thermodynamic properties of solid Cu-Mn alloys were determined from vapor pressure measurements by means of the Knudsen effusion tech- nique. Several discussions of the theory can be found in the literature. In the case of copper and manganese, it is assumed that manganese is the only component subliming, so the vapor pressure of manganese over the various alloys can be determined directly by measuring the weight loss of the effusion cell. This is a valid assumption since the vapor pressure of manganese is about four orders of magnitude higher than the vapor pressure of copper in the temperature region of interest. The vapor pressure of manganese over the alloys at any given temperature is then related to the weight loss of the effusion cell per unit area per unit time. An upper limit of 1.5 x 10"s has been measured for the ratio of polyatomic to monoatomic molecules for manganese vapor in equilibrium with solid manganese at 1180°K, so it was assumed that manganese vapor is monoatomic. The effusion cells, fabricated from high-purity alumina, are 26 mm high, 18 mm OD, and have a wall thickness of 1 mm. Three different sized holes of 2.03, 2.54, and 3.18 mm diam were drilled in the lids from the inside face to minimize chipping, and the lids were then slip-cast onto the cells to provide a gas-tight seal. Alumina was selected for this particular application both because of its resistance to thermal shock and reaction with manganese vapor. In the ideal case, the orifice of the effusion cell should have a knife edge so that effusion through the orifice

Jan 1, 1970

By F. J. Fayers, R. L. Perrine

For many years the problem of increasing ultimate recovery of oil from a reservoir has been a subject of interest to the oil industry. At present, a standard secondary recovery technique is to flood the reservoir with water once natural forces of the system are depleted. The permeability characteristics of the rock, together with interfacial forces which exist between the two fluids and the porous material. allow only part of the oil in the rock formation to be recovered by the water-flooding process. Laboratory experiments have shown that when detergents are added to the flooding water, the flow properties of oil in a porous rock arc improved. This improvement results in greater oil recovery. Even though higher recoveries are obtained using detergents, the extra cost associated with the method may make it economically unsatisfactory. To establish economic feasibility requires experimental laboratory measurement of fluid and rock properties together with a calculational procedure to predict field results. Pilot tests in the field may also be required. This paper presents a mathematical procedure for predicting detergent flood behavior. A number of reasonable simplifying assumptions are made to obtain a tractable system of equations. The resulting method may be regarded as a generalization of Buckley-Lev-crett' theory to solve simultaneously for water saturation and detergent concentration profiles. Consequently this type of calculation can be used in much the same manner as Buckley-Leverett theory is used to evaluate a water flood. The mathematical method employed is the so-called method of characteristics for solving hyperbolic partial differential equations. This application of the method is instructive because several other problems in reservoir analysis may be analyzed in a similar manner. FLUID FLOW EQUATIONS The flow behavior of oil and water without detergent through porous reservoir rock is usually expressed by Darcy's law and a continuity equation for each phase. If gravitational forces are neglected, vrg = kr (s)µr v p rg.........(1) vrg = kr (s)µr v p rg.........(1) ? pa va = F ?/?t (sd) where v is the volumetric flux, k represents permeability, S is saturation, and P the pressure. Fluid viscosity is denoted by 11, p is the density, 4 is porosity, and I the time. The subscripts w and o refer to water and oil. The capillary forces in the system lead to a difference in pressure between phases. This may be expressed by the relation: Pm -P0 =Pc(S).........(5) The capillary pressure? P,, is determined experimentally as a function of saturation. DETERGENT BALANCE EQUATION Addition of detergent to the water phase affects the flow properties of the system. Capillary pressure and relative permeability become functions of detergent concentration, c, as well as fluid saturation. That is: P--P (S,c) and k,,,.- - -- k,,,. At low oil saturation values, the presence of detergent reduces capillary pressure and reduces the relative permeability ratio. Detergents are surface-active and adsorb on rock surfaces. The quantity adsorbed, a, will depend on the concentration c of detergent solution and may also depend on the saturation; thus, a = a(S,c). In this analysis we have assumed that the detergents used are water soluble but not oil soluble, and their movement is therefore restricted to the water phase. The dispersion of detergent in water is neglected by comparison with transfer by flow within the system. Detergent movement is determined by the water flux and adsorption: ? cv m = -F ?/?t (cs + a) We will assume that the quantity of detergent in solution is too small to significantly change water density or viscosity. When the permeability and capillary pres-

By D. R. Hay, E. Scala

Electrotransport has been superimposed on the rate-limiling- step in zone refining which is the impurity diffusion through the liquid at the solid/liquid interface. The efficiency of zone refining is increased and relative resistivity ratios (P300°k/P4 2°K) he increased by 50 pel in tungsten for a given number of posses using- the electrotransport field-aided method. The effiect of zotze refining in the presence of an electric field on the relative resistivity ratio is examined as a function of its intensity and polarity. THE effect of electrodiffusion on the effective distribution coefficient of zone refining has been discussed by Pfann and Wagner.' Control of the distribution coefficient in zone refining is of twofold interest. A distribution coefficient of unity can be increased or decreased, making possible the transfer of some previously inseparable solutes. The general efficiency of zone refining can also be increased. In the conventional zone-refining process impurities are removed by their retention in the molten zone and subsequent deposition at the end of the rod after a complete passage of the zone. In higher-melting-point systems a second mechanism may also operate, the volatilization of impurities from the molten zone. It has been suggested by schadler2 that, in refractory metals, vacuum distillation plays the predominant role in purification. In Koo's results,3 the resistivity ratio decreased beyond a zone-refined length of 6 in. from the starting end, indicating that some purification does in fact occur by true zone-refining action. Hay and scala4 showed that the degree of purity as measured by the relative resistivity ratio method depends on the square root of time in the molten zone (number of passes) and thus appears to be limited by diffusion in the boundary layer at the solidifying interface assuming good mixing in the body of the molten zone where a linear rate would prevail. The final concentration of impurities in electron-beam floating-zone purification depends very much on the operating conditions and results obtained by various investigators2,3,5 are difficult to correlate. The diameter of the starting material, operating vacuum, and rate of traverse are among the factors to be considered. There are, nevertheless, some general observations to be obtained from the literature. Lawley5 reviewed the literature up to 1962 and concluded that the major contaminants generally remaining in tungsten after electron-beam zone refining are molybdenum and carbon. Molybdenum due to its high melting point and distribution coefficient of nearly unity would be expected to be difficult to remove by either volatilization or zone refining. Carbon is not only a common contaminant in vacuum systems, but is also very difficult to remove by conventional vacuum melting. even with vacuums of 10-7 mm Hg. For each of those impurities removed by zone refining the equation of Pfann and Wagner may be applied: It relates the effective distribution coefficient k to a superimposed driving force which accelerates the solute away from the interface with a terminal velocity L" and to the more conventional zone-refining parameters; v = rate of traverse, t = thickness of the diffusion layer at the solidifying interface, D = diffusivity of the solute in the molten zone, and ko = equilibrium distribution coefficient of the impurity.

Jan 1, 1965

By J. Kucera, J. Vreštál

DIFFERENT authors1 ' have been engaged in meas-uring the vapor pressure of pure cobalt. The results of their measurements satisfy the expected temperature dependence of vapor pressure and are in mutual agreement, with the exception of the results by De Dyk Man and ~esmejanov.~ The latter authors, by using the effusion method, have achieved considerably higher values than other workers. By using the isotope exchange method they obtained much lower values than other authors. The determination of thermodynamic functions of the binary Ni-Co system was carried out, as far as we know, only by De Dyk Man and Nesmejanov.4 This determination was based on extremely low values of vapor pressure over pure cobalt obtained by the isotope exchange method. Therefore we considered it useful to repeat these measurements. Furthermore, the reliable values of thermodynamic functions should complement the migration characteristics, the inner magnetic field data obtained by Miissbauer effect measurements, specific heat data, and the K effect and positron annihilation characteristics that have been achieved for the Ni-Co system in the Institute of Metallurgy of the Czechoslovak Academy of Sciences in Brno.5-9 The measurements in this paper were performed by the isotope exchange method. In this method, as has been shown by analysis,10-'3 the vaporization rate, w, can be calculated from the relation w = k/1oSw [1] i0Sw) [1] where L, is the activity per unit mass of the source, S is the area of evaporation, k is the tangent at the point t = 0 to the curve giving the radioactivity de-pendence of the target vs annealing time, t, and wis the probability for the tracer atom condensation on the target surface. The value for cc. can be obtained from the equation w = a + 2P(l-a) [2] where P is Clausing's coefficient and a is the condensation coefficient. In the present work the cylinder shaped containers for annealing the samples were used, characterized by the ratio 1:d = 1.06. Here 1 is the length and d the diameter of the container. For this ratio P = 0.514 and w =0.5, and w is independent of the condensation coefficient. The vaporization rate is related to the vapor pressure p by the relation where R is the gas constant, T is the absolute temperature, and M is the molecular weight of the component investigated. Measurements of the vapor pressure of the i component over the pure metal, p!, and the alloy, pi, at the given temperature and composition enable one to calculate the thermodynamic activity of this component from the equation ai=pi/pi [4] is and the activity coefficient ri=ai/xi [5] xi where xi is the molar fraction of the investigated alloy component. EXPERIMENTAL Evaporization rate measurements were carried out with pure cobalt and eight different Ni-Co alloys, which were kindly delivered by the Centre National de Recherches Me'tallurgiques Liege, Belgium. The composition of the alloys is given in Table I. The cobalt used in these alloys contained the following impurities: 0.016 wt pct Fe, 0.014 wt pct Cu, 0.01 wt pct Mn. Other elements were present in much lower concentrations. From the melts, samples in the form of small cylinders of 12 mm in diam were ma-chined. The thickness of the targets and sources was 2.5 and 1.2 mm, respectively. The lower thickness of the sources was chosen to make their over-all ac-tivity as low as possible. Sources were activated in

Jan 1, 1970

By Paul A. Beck, John Towers, W. D. Manly

Recent work on grain growth in high purity aluminum and in a solid solution type alloy of aluminum and magnesium1 showed that the isothermal increase of the average grain diameter D with time follows the equation D = K(tg + A)2 Eq 1 Here tg is that portion of the total annealing time 1, which is available for grain growth, that is the annealing time less the time necessary for complete recrystallization. The parameters A and n depend only on the temperature. A has the dimension of time, and is usually of the same order of magnitude as the time for recrystallization: R. In some instances R may be substituted for A in Eq I. This is the case for example at long periods of annealing or at high temperatures, where both A and R become small in comparison with t. In such instances Eq I is simplified to D = K, tn Eq 2 For high purity aluminum the exponent n increases in the ratio of about I to 5 in. the temperature range of 350' to 6w°C. 70-30 brass has been long known to exhibit grain growth of the continuous type, such as occurs in pure metals. For this reason, it was of interest to determine whether grain growth in brass also follows Eq I and 2 as it does in aluminum. Isotherma1 grain growth data for 70-30 brass were published in recent years by R. S. French2 and by H. L. Walker.' Fig I shows a log D vs. log t plot of French's data, together with the 500, 600 and 700°C data by Walker for 42.8 pct reduction by rolling. In such a log D vs. log t plot Eq 2 corresponds to a straight line with a slope of n. As seen in Fig I, the data of French and the 600°C data of Walker can be represented by straight lines. However, the 500 and 7c0°C data of Walker give curved lines and the average slope of his 500°C line is very different from the slope of the line representing French's data for the same temperature. Furthermore the curvature of the 700°C line suggests a tendency for grain growth to stop after 8 hr. If confirmed, this would constitute a definite deviation from Eq 2, not explainable on the basis of the specimen thickness effect found in high purity aluminum.' Walker's specimens were 0.125 in., or 3.18 mm thick, while the deviation occurs at a grain size of approximately 0.4 mm. A further point of divergence arose recently, when J. E. Burke suggested4 on the basis of Walker's data, that, instead of Eq 2, grain growth in 70-30 brass follows the equation D - D, = K . tgn Eq 3 Here D, is the grain size as recrystallized at the temperature in question. Although it was possible to show5 that, for pure aluminum, for which isothermal grain growth has been carefully investigated, Eq 3 is incompatible with the experimental facts, the situation remained somewhat obscure with respect to 70-30 brass. In view of the discrepancies between the published data on brass from different

Jan 1, 1949

By G. P. Marino

The "conductivity ", Kr, of a spherical void in a linear temperature gradient due to radiation is calculated by integration of the general expression for diffuse radiation between two gray surfaces of arbitrary orientation. The spherical geometry allows an exact evaluation of the phenomenon without resorting to up-proximate expressions for the radiant heat flux. The resulting expression for Kr is three quarters of the approximate value obtained by Loeb. IN the determination of the thermal conductivity of porous materials such as sintered UO2 and ThO2 fuel pellets, the magnitude of the thermal conductivity in the pores due to radiative heat transfer across the pore, Kr, may be important in assessing the total conductivity of the pellets. Loeb' has calculated Kr for various pore geometries by approximating the net flux of heat between two surfaces of temperature difference AT as 4eT~AT and obtained: K,=4yoet3odp [1] where a is the Stefan-Boltzmann constant, E is the emissivity of the radiating surface, To is the average absolute temperature of the region across which radiation occurs, dp is the largest dimension of the gap in the direction of heat flow, and y is a geometric factor. That is: y = 1 for laminar pores and cylindrical pores with axes parallel to the heat flow direction, y = 2/3 for spherical pores, y = /4 for cylindrical pores with axes perpendicular to heat flow direction. It was the purpose of this work to obtain an exact expression for K, in spherical pores by integration of the general expression for the net heat transfer between two incremental areas on the surface of a spherical pore. This procedure yields a value of Kr 3/4 of that obtained by Loeb. I) CALCULATION OF Kr Fig. 1 illustrates the general geometry which defines the problem. The total emissive power for a gray body is given by: E = EoT4 [2] where E is the radiated energy per unit area of the radiation surface per unit time, T is the absolute temperature, o is the Stefan-Boltzmann constant (o = 5.67 X lo- '' watt per sq cm-"K1), and E is the emissivity of the surface. Although Eq. [2] gives the exact expression for the total emissive power of any gray body, all practical problems involve the net exchange of thermal radiation between a body and surrounding bodies. The two major considerations in the solution of such problems are a) shape and relative position of the surfaces and b) the emissivities of the surfaces. In this work it shall be assumed that all incremental areas on a pore have the same emissivity and, further, that little or no reflection occurs (i.e., the radiation hitting a surface is either absorbed or transmitted). Thus, the solution of the stated problem requires only that consideration a be satisfied. Consider two incremental surface areas dA 1 and dA2 of the same material in a general orientation to one another. Let dAl be at temperature TI and dAz be at T2. Fig. 2 illustrates the geometrical parameters involved. That is, 71, is the normal of dA, n2 the normal of dA2, r the line connecting the centers of each area, 81 the angle between nl and r, and Q2 the angle between 7,2 and r. Using Eq. [2] and the Lambert cosine law of diffuse radiation it is easily shown that:293 WdAf- dA, = ecr T*------^fpi------" dAidA2 [ 3a 1 and

Jan 1, 1970

By John Towers, Paul A. Beck, W. D. Manly

Recent work on grain growth in high purity aluminum and in a solid solution type alloy of aluminum and magnesium1 showed that the isothermal increase of the average grain diameter D with time follows the equation D = K(tg + A)2 Eq 1 Here tg is that portion of the total annealing time 1, which is available for grain growth, that is the annealing time less the time necessary for complete recrystallization. The parameters A and n depend only on the temperature. A has the dimension of time, and is usually of the same order of magnitude as the time for recrystallization: R. In some instances R may be substituted for A in Eq I. This is the case for example at long periods of annealing or at high temperatures, where both A and R become small in comparison with t. In such instances Eq I is simplified to D = K, tn Eq 2 For high purity aluminum the exponent n increases in the ratio of about I to 5 in. the temperature range of 350' to 6w°C. 70-30 brass has been long known to exhibit grain growth of the continuous type, such as occurs in pure metals. For this reason, it was of interest to determine whether grain growth in brass also follows Eq I and 2 as it does in aluminum. Isotherma1 grain growth data for 70-30 brass were published in recent years by R. S. French2 and by H. L. Walker.' Fig I shows a log D vs. log t plot of French's data, together with the 500, 600 and 700°C data by Walker for 42.8 pct reduction by rolling. In such a log D vs. log t plot Eq 2 corresponds to a straight line with a slope of n. As seen in Fig I, the data of French and the 600°C data of Walker can be represented by straight lines. However, the 500 and 7c0°C data of Walker give curved lines and the average slope of his 500°C line is very different from the slope of the line representing French's data for the same temperature. Furthermore the curvature of the 700°C line suggests a tendency for grain growth to stop after 8 hr. If confirmed, this would constitute a definite deviation from Eq 2, not explainable on the basis of the specimen thickness effect found in high purity aluminum.' Walker's specimens were 0.125 in., or 3.18 mm thick, while the deviation occurs at a grain size of approximately 0.4 mm. A further point of divergence arose recently, when J. E. Burke suggested4 on the basis of Walker's data, that, instead of Eq 2, grain growth in 70-30 brass follows the equation D - D, = K . tgn Eq 3 Here D, is the grain size as recrystallized at the temperature in question. Although it was possible to show5 that, for pure aluminum, for which isothermal grain growth has been carefully investigated, Eq 3 is incompatible with the experimental facts, the situation remained somewhat obscure with respect to 70-30 brass. In view of the discrepancies between the published data on brass from different

Jan 1, 1949

By G. E. Karantzavelos

Introduction Pure alunite (KAl3 (SO4 )3 (OH)6 ) contains 37% A1203. In the alunite ore found on Milos Island, Greece, opal is the principal gangue while some aluminum, iron, and titanium oxides are also present. Processing of alunite ore leads to the recovery of potassium sulfate, alusil (alumina-silica mixture), and sulfur dioxide. An interesting process has been developed in Greece (Svoronos, 1977) consisting mainly of roasting the alunite ore mixed with KCl to disrupt the mineral lattice in preparation for separating its components. The thermal decomposition of alunite at approximately 600°C occurs by the reaction: KAl3 (SO4)3 (OH)6 + nSiO2 + 6KC1 - 4 K2SO4 + 3 A1203 . nSiO2 + 6 HCl + 3H20 According to the process pure alunite mixed with 543 kg KCl, 840 kg K2SO4, 370 kg alusil, 264 kg HCl, and 65 kg H2O are produced. The roasted calcine is leached in hot water. The potassium sulfate leach slurry is then thickened and washed to obtain a solid residue (alusil) plus a solution rich in potassium sulfate. Assuming the alunite ore feed consists of 50% to 60% alunite, the solid residue might consist of approximately 22% to 27% A1203, silica, iron, and titanium oxides. This leach residue could be digested in a hot sodium hydroxide solution to dissolve alumina. An amount of the alumina (approximately 10%) would not dissolve. In addition, part of the silica would dissolve requiring removal from the solution in a subsequent desilication operation. The same leach residue could be attacked by mineral acids such as a diluted hydrochloric acid to extract aluminum chloride (Sawyer et al., 1983). Alumina and aluminum are presently produced from bauxite of which Greece has important reserves. If this aluminous material (alusil) could be beneficiated by any physical process, then it would be considered as a raw material for the production of fire-clay refractories instead of an alternate source of alumina. Experimental procedure Extensive laboratory work (mineralogical and chemical analyses) indicates that alusil is a mixture of liberated (mostly) alumina and silica particles, not a natural silicate mineral. Table 1 shows the chemical assays of two alusil samples that were obtained from a roasted alunite ore leaching pilot plant. The samples used for the experiments were a very fine alusil powder (K80 = 400 mesh, 38 µm). These were pulped in tap water. Desliming of the pulp before each flotation test was ineffective. Flotation tests were performed in a 2000 ml Denver cell. A standard 500-g sample was first conditioned at a pulp density 50% solids for 5 min with hydrofluoric (HF) acid solution (40%). The temperature of the conditioned pulp was measured and kept nearly constant at 50°C. The pH of the conditioned pulp was also regulated to 2.5-3 with HF solution. Then ARMAC C solution (2%) as well as two drops of fuel oil and one drop of AEROFROTH 65 were added and the pulp was reconditioned for 2 min. After conditioning, the viscous pulp was transfered to the flotation cell. It was then diluted with warm water to about 15% solids and floated to separate silica from alumina particles. A rougher froth product that was cleaned and recleaned was obtained from the flotation tests.

Jan 1, 1987

By J. Buchwald, R. W. Heckel

THE DeHoff method' of analysis of ellipsoids of constant shape provides a quantitative metallo-graphic technique whereby the actual size distribution of the ellipsoids may be determined from polished and etched random planar sections. This method provides a useful metallographic tool since it appears to be the only method available for determining the size distribution of nonequiaxed second-phase particles. This method, however, assumes that the type of particle being studied is known to be either oblate or prolate of a fixed minor to major axis ratio, q. This, in general, means that one assumes either oblate or prolate ellipsoids of a given q and checks the total volume and surface area of the second phase as determined from the particle size distribution to values of these quantities determined by other techniques. Suitable check methods may be point counting, lineal fraction, and areal fraction methods for obtaining the total volume and the Smith and Guttman2 method of intercepts for obtaining the total surface area. If the initial guess at particle shape and q were incorrect, i.c., the volume and surface information obtained from the particle size distribution data does not agree with those values obtained by other techniques, another assumption must be made and the particle size dist distribution must be recalculated. A current study in this laboratory has involved a large number of these calculations. To facilitate the handling of the calculations, the DeHoff method was programmed in the FORTRAN language for an IBM 1620 computer. The purpose of this note is to briefly describe this program. The complete program along with a list of variable names, a sample input, and a sample output may be obtained from the senior author. Basically, the program provides for the calculation of particle size distributions for a series of q values for both oblate and prolate ellipsoids for each set of input data. The total volume of second-phase particles per unit volume are calculated for each calculated particle size distribution. The output of the program is arranged to allow a rapid comparison of the calculated volume and surface values to data obtained by independent check measurements. A condensed flow chart of the program in the notation of DeHoff1 is given in Fig. 1 The calculation of the number of particles per unit volume in the jth size class, Ni, according to the DeHoff method' is: where k(q) is the shape-correction factor given by DeHoff as a function of q for both oblate and prolate ellipsoids, A is the size-class increment, k is the number of size classes, ni is the number of particles in the ith class on the planar measurement surface, and B(j,i) are the saltykov3 coefficients which correct for measurement surfaces which do not pass through the major and minor axes of the particles. The values of /3(j,i) are included as a two-dimensional array in the program. Particular values of k(q) are included in the program in increments of q of 0.05. The first calculation is carried out for q = 1 (spheres), the second for q = 0.95 oblate, the third for q = 0.95 prolate, and so forth. The number of q values chosen for the calculation is determined by a control card in the input for each set of data.

Jan 1, 1965

By M. H. Pilkuhn, H. Rupprecht

The junction luminescence of GaAsxP1-x diodes containing up to 47 pct Gap was studied. Diodes were prepared by diffusing zinc into n-type material which was eithw boat- or vapor-gvown. Observations concerning the energy and the line shape of the spontaneous emission are presented and discussed; in particular the influence of inhomogeneous phosphorus distribution is considered. The role of reabsorption of the junction radiation is pointed out. With increasing current, the peak emission shifts to higher enwgy. This is alm true in cases where slow line narrowing occurs simultaneously. The length dependence of the threshold current density for lasing at 77 °K was examined for lasers made from the same material. It was found that longer lasers have lower threshold curvent densities, Estimates of the characteristic laser losses are given. Data concwning the range of Gap concentration in which stimulated emission was observed are presented. FORWARD-BIASED p-n junctions in mixed alloys of GaAsxPl-, emit light whose wave length depenfs on the composition of the alloy ranging from 8400A (x = 1) to 5350A (x = 0) at 77°K. Laser action in vapor-grown material was first reported by Holonyak and evacqua,' and in boat-grown material by Ainslie, Pilkuhn, and Rupprecht.' In both cases this was observed for low Gap concentration where GaAs,Pl-, is still a "direct" semiconductor with the lowest conductance-band minimum at k = 0 and a subsidiary minimum in the [loo] direction. As the Gap concentration is increased, the subsidiary [loo] minimum in the conduction band is lowered and finally becomes the lowest minimum for Gap concentrations above 50 pt,3 making the alloy an ''indirect" semiconductor. It is of interest to determine up to which Gap concentration stimulated emission can be observed. EXPERIMENTAL Boat-grown material was prepared by N. Ainslie of IBM by horizontal Bridgman technique. The material was n type and silicon-doped up to a concentration of 10'' cm-'. Vapor-grown material was made by E. Hull of IBM or received from Merck & Co. Diodes were fabricated in a similar manner as in the case of GaAs by zinc diffusion. Details about the diffusion can be found in a later paper.4 All diodes were of Fabry-Perot structure with two parallel sides cleaved and the other two sides sawed. Their width ranged from 50 to 150 p and their length from 50 to 700 p. RESULTS AND DISCUSSION Spectral Distributions. A typical display of the junction luminescence of a GaAsxPl-, diode is shown in Fig. 1. The liquid-nitrogen and room-temperature curves which were taken at a very low current (25 ma dc) show a relatively sharp high-energy emission which is band-edge radiation. They also show a broad low-energy emission which is due to transitions to deep impurity levels. This latter emission line may have different shapes and positions depending on the nature of the material. It can sometimes be quite near the band gap giving rise to "bumps" in the high-energy line. The results presented in this paper will be concerned only with the high-energy emission which becomes more intense relative to the low-energy emission when the current is increased and the temperature is lowered.

Jan 1, 1964