Search Documents

By R. C. A. Minnitt, A. Akpabio

This study presents a rigorous comparison between ordinary kriging and commonly used machine learning algorithms, those being, linear regression, support vector regression, decision trees, random forests (RF), and k-nearest neighbours for spatial interpolation of platinum grade estimates in a complex ore body within the Bushveld Igneous Complex. Using only X and Y coordinates as predictors, both ordinary kriging and machine learning models were evaluated at point and block supports under traditional and spatial block cross validation frameworks. While naive validation results suggested superior performance for k-nearest neighbour and random forest (R² = 0.92 and 0.86, respectively), these were revealed to be overly optimistic under spatial dependence. Spatial block cross validation results demonstrated substantial declines in model performance, with R² often falling below zero, particularly for decision trees and k-nearest neighbour, indicating strong overfitting and limited generalisability. Ordinary kriging exhibited more stable, albeit modest, performance under spatial validation, reflecting its strength in geostatistical interpolation when contextual geological variables are unavailable. The study underscores the critical importance of spatially aware validation in resource estimation and highlights that machine learning models constrained to spatial coordinates behave as interpolators rather than true learners of geological variability. Recommendations are provided for future work incorporating geological information to enhance predictive robustness.

Mar 30, 2026

By E. G. King

THIRD-LAW entropy data for zirconium-containing compounds are sparse. Kelley a lists entropies at 298° K. for the oxide, tetrachloride, nitride, and silicate, as well as for the metal. The present paper presents heat capacities between 51° and 298° K. and entropies at 298.15° K. for the zirconates of the alkaline-earth metals-calcium, strontium, and barium. The authors believe magnesium zirconate does not exist as a stable compound at room temperature. Entropies of formation from the constituent oxides also are given. MATERIALS The materials used in this investigation were synthesized and analyzed in the Berkeley Laboratory of the Bureau of Mines. Calcium zirconate was prepared by heating an intimate mixture of reagent-grade calcium carbonate and pure zirconia for 30 hours between 1,200° and 1,300° C. and for 32 hours between 1,400° and 1,500° C. As part of the heating schedule, the sample was ground and mixed seven times. Periodic analyses and minor adjustments of composition were made. Analysis of the final product gave 31.19 percent calcium oxide (theoretically 31.28 percent) and 68.67 percent zirconia (theoretically 68.72 percent). The X-ray diffraction pattern agreed with that in the ASTM catalog.

Jan 1, 1960

By Norman P. Goss

What would appear to be a very simple process on paper has proved to be overrun with a multitude of unseen difficulties. For more than 70 years attempts have been made to cast metals in a continuous manner, but it has been only within the last Io years that some degree of success has been achieved.1 Historical Development Bessemer's Experiments—Method I Bessemer was the first to try to cast molten metal in a continuous manner. His method is fully described in U. S. Patent No. 49053, issued in 1865. The patent drawing is shown in Fig. I. In this method the molten metal was poured between a pair of rolls of large diameter, which were cooled by water sprays. Means were provided to prevent moisture from coming into direct coptact with the molten metal. The width of the strip was controlled by the flanges. On paper this should work to perfection, but it is beset with many difficulties still awaiting solution. Bessemer failed to reduce his invention to commercial practice. About 1890 Bessemer's method was revived by Norton and associates (U. S. Patent No. 441375). They formed the Fluid Metal Rolling Co. at Maywood, Ill., and spent $250,000 in endeavoring to develop the technique for this simple method. Finally they had to abandon it. Naturally, they did carry the technical development far beyond anything Bessemer had done, but even so it was far from being a useful commercial success. In recent years it has undergone further development by C. W. Hazelett (U. S. Patent No. 2058447), who has succeeded in obtaining long lengths of copper, brass, aluminum, and even stainless steel. He introduced many' ingenious devices but the greatest obstacle still remained—the disintegration of the roll surface when the surface is heated and chilled in rapid succession. Difficulty is also experienced in maintaining a uniform supply of molten metal to the rolls. The patents pertaining to this method would indicate these to be the major obstacles. Experimental work by Mellen—Method 2 A departure from the method introduced by Bessemer, was tried by Grenville Mellen, of East Orange, K. J., sometime between 1914 and 1920, who obtained a number of patents.2 His method is illustrated in Fig. 2. This method failed to make an impression upon the art; however, it is of value in pointing out the difficulties the inventor ran into and the way in which he tried to meet them. The split mold sections form two endless chains. When going through the center of the apparatus, these sections engage each other to form a complete mold. As the engaged mold section moves downward at a constant speed, molten metal is teemed into it. As this is going on, another set of cooperating mold sections move into position on top of the one being filled. The molten metal is teemed into the mold through a silica tube. According to the teachings of this patent, the level of molten

Jan 1, 1945

By L. K. Witika, J. H. Masinja

"Investigations of flotation characteristics of Lumwana and Kansanshi ores using formulated frothers were carried out. Of the six frothers tested, two came from the alcohol family, two from the glycol family and last two were formulated products. The initial work involved determining the optimum mesh of grind and frother dosage for Lumwana and Kansanshi ores. Further work involved carrying out surface tension, bubble size and froth stability measurements. Finally, water recovery tests were conducted at Lumwana and Kansanshi mine sites. For froth stability tests, six frothers were tested on a two-phase (water–air) system using the Bikerman column flotation cell, while water recovery tests were conducted using a total of seven frothers, including the standards for each plant. The results of the testwork showed that (i) although a stable froth is essential for optimum recovery of valuable minerals, a higher molecular weight frother, like Dowfroth 250, caused poor selectivity when used on fine ores and (ii) a lower molecular weight frother was suited for Lumwana ore, while a mid to higher molecular weight formulated product was suitable for Kansanshi sulphide ore. INTRODUCTION Investigations of flotation characteristics of Lumwana and Kansanshi ores using formulated frothers were carried out. The initial work of this project involved determining the optimum mesh of grind and frother dosage for Lumwana and Kansanshi ores. Further work involved carrying out surface tension, bubble size and froth stability measurements. Finally, water recovery tests were conducted at Lumwana and Kansanshi mine sites. In order to investigate the gangue entrainment levels in the Lumwana and Kansanshi sulphide ores, two well-known methods (froth stability and water recovery) were used. Firstly, froth stability tests were conducted using six frothers, two from the glycol family (Dowfroth 200 and Dowfroth 250), two from alcohol family (MIBC and pine oil) and other two from the formulated family (Betafroth 20 and Betafroth 245), at the University of Cape Town’s Centre for Mineral Research (CMR) laboratories. Froth formation and stability measurements were conducted using the Bikerman column flotation cell."

Jan 1, 2015

By G. Belardi, N. Shehu, L. Passeri

"This paper illustrates the possibilities of applying what is essentially a physical soil-washing technique for treatment of soil contaminated by semi-volatile hydrocarbons in order to reduce the level of pollution to that permitted by Italian regulations.The ground around a factory producing lubricating-oil additives consists of silty-sandy soils which have long been contaminated by heavy hydrocarbons, especially branched alkyl benzenes and naphthalene derivatives. Pollution is particularly marked in some zones (hot-spots) where Total Petroleum Hydrocarbon (TPH) values are as high as 7000 ppm. The hydrogeological characterization showed the presence of a gravely soil in a sandy/silty matrix, with big stones making preferential paths to the water circulation. The permeability is 10-5 m/s and the water-table occurs at about 3 m below ground level.After laboratory and pilot-scale studies on representative samples of in-situ soils with TPH between 2700 and 7000 ppm a flow-sheet is proposed for reducing TPH values below the acceptable limits in Italy, namely 700 ppm for hydrocarbons with more than 12 carbon atoms and less than 250 ppm for those with fewer.On the basis of indications provided by laboratory tests a mainly physical treatment process of soil washing has been developed. This is used to clean-up soil from a site polluted by non-volatile hydrocarbons with a high TPH value. The process involves a high-density attrition stage followed by wet screening at 0.5 mm and then wet-gravity separation of the minus 0.5 mm fraction in the centrifugal field.The gravity separation products consist of a light fraction with particle size finer than 15 micron which contains most of the pollutants, and a heavy coarse fraction with TPH values conforming to the limits set by Italian regulations (less than 700 ppm hydrocarbons whose structure contains more than 12 C atoms, and less than 250 ppm of hydrocarbons with less than 12 C atoms)."

Jan 1, 2003

By Elger. G. W.

The Bureau of Mines investigated preheating of continuously fed, fragmented ferrous scrap charges by furnace offgases and utilizing substitutes for imported fluorspar to condition electric steelmaking slags. Three types of continuous scrap charging procedures were investigated to determine electrical energy consumption in a I-st (short ton) electric arc furnace. Cold and preheated scrap charges were continuously fed at rates averaging 37.5 and 43.7lb/min, respectively. The feed rate varied appreciably from test to test owing to hangup of the scrap in the charge bin. Approximately 7 pct less electrical energy was consumed in melting scrap preheated to 8400 to 1,1100 F by furnace off-gases than in melting cold scrap. Overall energy consumptions were 888 kW'h/st for cold scrap, 829 kW'h/st for preheated scrap" and 637 kW'h/st for conventional backcharged scrap. Stack gases from scrap preheating averaged 1200 F and a flow rate of 1,615 scfm compared with 2200 F and 1,302 sefm for cold-charged scrap. Slag conditioners tested included synthetic fluorspar, boric acid, hydroboracite, used aluminum potlining wastes and aluminum anode tailing wastes, and Sorelflux B. The boron-containing conditioners fluidized the slags the best, followed in order by the conditioners containing fluorine and titanium.

Jan 1, 1987

By Walter Kawa

The hydrogenation of phenyl ether, benzyl ether, and phenyl sulfide in the presence of iron catalysts was studied by the Bureau of Mines to gain a better understanding of the role of iron catalysts in the hydrogenation of coal. The carbon-oxygen and carbon-sulfur bonds in these compounds may be similar to some of the carbon-oxygen and carbon-sulfur bonds in coal which react during the early stages of liquefaction by hydrogenation. Experiments were made in batch autoclaves for 4 hours; hydrogen was at a pressure of about 540 atmospheres. In the hydrogenation of phenyl ether at 400° C, metallic iron catalyzed both hydrogenolysis (splitting-hydrogenation) and nuclear hydrogenation (addition of hydrogen to aromatic rings) reactions, Products were principally benzene and cyclohexane. The activity of metallic iron was decreased by adding carbon monoxide or a reactive sulfur-containing organic compound. Ferrous sulfide catalyzed the conversion of phenyl ether to benzene, phenol, and small amounts of cyclohexane. Hydrogenolysis of benzyl ether was catalyzed by metallic iron and to a lesser extent, by ferrous sulfide at 250° C. No nuclear hydrogenation was observed with either catalyst. Conversion of phenyl sulfide to benzene was nearly quantitative in experiments made at 250° C with equimolar quantities of phenyl sulfide and metallic iron or iron as ferric oxide. Conversions of metallic iron and ferric oxide to ferrous sulfide were nearly stoichiometric. No reaction of phenyl sulfide occurred in the presence of ferric oxide and water, and very little occurred in the presence of metallic iron and water. Ferrous sulfide also catalyzed the hydrogenolysis of phenyl sulfide to benzene, but conversion was lower than in experiments with metallic iron or ferric oxide. The activity of ferrous sulfide was increased by adding water and decreased by adding hydrogen sulfide. The results of this study suggest that ferrous sulfide is an active form of iron when iron is used as a catalyst for the hydrogenation of coal.

Jan 1, 1963

By William J. Davis, R. Blair Hrabi, Sally Pehrsson, Amelia V. Logan, Ross Sherlock

"Abstract - The Meadowbank gold deposits are hosted by a series of polydeformed and metamorphosed auriferous iron formations located in the Neoarchean Woodburn Lake group, western Churchill province, Nunavut. The supracrustal rocks in the Meadowbank area consist of a thick sequence of intermediate-composition volcanic rocks with intercalated iron formation, ultramafic volcanic rocks, and quartzite to quartz arenite. U-Pb geochronology gives an interpreted age of 2711 ± 3 Ma for the host strata at Meadowbank.The intermediate volcanic rocks are predominately volcaniclastic and have a geochemical composition that is consistent with an active continental margin setting. Iron formation deposition was coeval with the intermediate volcanism, as indicated by incorporation of volcanic detritus in the chemical precipitate. The geochemistry of the iron formation indicates that it precipitated from a fluid that at one time was at high temperature (>250°C), reducing, and likely acidic. Intercalated ultramafic rocks rarely show spinifex textures and have trace element ratios that are consistent with mantle plume-related undepleted mantle asthenosphere, likely recording episodic mantle upwelling associated with rifting in an active continental margin. The contact between the ultramafic volcanic rocks and the overlying quartz arenite is a disconformity marked by a quartz pebble conglomerate. However, similar geology on the structural hanging wall and footwall of the contact suggests that the conglomerate does not represent a significant hiatus in deposition, and is more likely a prograding terriginous siliciclastic unit.Three metamorphic events are recognized. The first, M1, is a cryptic greenschist facies event that is pre-D2. The second event, M2, is a mid-greenschist to amphibolite facies, syn-D2 event. M3 is the last event recognized, and is a post-tectonic greenschist facies event that is regional but inhomogeneous in extent, possibly reflecting thermal aureoles around post-tectonic 1.7 to 1.8 Ga Nueltin granites. The structural geology of the area is complex with four phases of deformation recognized, two of which had a significant effect on the geometry of the deposit. All of these regional events are interpreted as Paleoproterozoic in age. Relationships between deformation fabrics and mineralization, as well as the overall geometry of the mineralized envelopes, suggest that the deposit formed during syn- to late-D2. Superimposed on the mineralization are D4 structural elements that postdate gold mineralization.The main control on gold mineralization is replacement of magnetite by pyrrhotite and pyrite in high-strain corridors. The composition of amphiboles and chlorites associated with gold mineralization is remarkably consistent and shows no spatial or temporal variation, suggesting that it was buffered by the iron-rich nature of the host rocks. The bulk metasomatic effect on the intermediate volcanic rocks is characterized by the addition of K2O and the loss of CaO, Na2O, and MgO, with little variation in the total iron content. The elemental losses are likely the result of destruction of feldspars, stilpnomelane, and ferroactinolite, which are relatively common in the unmineralized rocks, and the formation of sericite, chlorite, and grunerite."

Jan 1, 2004

By G N. Phillips

Global annual gold production has been remarkably constant over the last decade at ~2500 t/a3. The major producers today are China (annual production over 350 t and rising), Australia (250 t), USA (240 t), Russia (200 t) and South Africa (190 t and falling); these figures are representative of recent years but approximate. Demand from jewellery, investment, industrial and dental uses has been in excess of mine production for the last decade, by approximately 1000 t/a.Ongoing exploration is the life-blood for gold mining, and more so than for many other metals. Explorers consider sovereign risk, security, labour costs, commercial terms, land availability and inherent geologic prospectivity to determine where to focus their activities. Exploration success over the last decade has been shared by established major producers like Australia, Canada and USA, several countries in West Africa, and many parts of the Pacific margin. Although gold-only deposits have dominated world gold production historically, the gold-plus types, especially copper-gold, are becoming more important.Australia experienced a major period of exploration success from 1979 that has led to annual gold production rising from 17 t in 1979 to 250 - 300 t/a recently. Today, Australia has economic demonstrated resources of over 9000 t Au, which has risen progressively from 220 t in 1979. The majority of production has come from gold-only deposits in the Yilgarn craton in the southern half of Western Australia, but there are a few large Proterozoic deposits in central Australia including Olympic Dam, Telfer and Starra, which are mostly gold-plus types with copper and gold. Large goldfields in the younger host rocks of eastern Australia include the gold-only deposits of Victoria and North Queensland, and gold-plus deposits of New South Wales.CITATION:Phillips, G N, 2013. Australian and global setting for gold in 2013, in Proceedings World Gold 2013 , pp 15-22 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 26, 2013

By Ann M. Hagni

Flash reduction of zinc oxides is an alternative, environmentally safe method to leaching and electrowinning for zinc recovery. Zinc and lead are volatilized in a flash reactor, a non-hazardous residue is generated, and the fuel source is inexpensive coal in the flash reduction process. Results from experimental laboratory tests yielded low zinc recovery for this process. Mineralogical and textural characterization of the residues from the flash reduction tests for different feed materials and temperatures were conducted to determine the cause for the low zinc recovery. Reflected light microscopy, cathodoluminescence microscopy, scanning electron microscopy-energy dispersive spectroscopy, and X-ray diffraction were utilized in this applied mineralogical research. The original zinc calcine is composed primarily of fine-grained, iron-bearing zincite particles, 25-40 um in diameter, in a porous texture. Pellets processed at 1150°C were found to produce reaction rims of franklinite-magnetite-jacobsite solid solution that are 125-200 um thick and surround cores of unreacted iron-bearing zincite in a Zn-Fe-Si-Ca glass matrix. The reaction rims have subsequently oxidized to form an outer 10 ~m thick rim of hematite. Slags from calcines processed at 1275°C are primarily triangular particles with porous centers, cenospheres, and spheres of iron-bearing zincite, in addition to rectangular particles composed of non-porous franklinite-magnetite solid solution cores with iron-bearing zincite rims. Slag particles from calcine processed at 1200°C are predominantly 100-250 ~m triangular-shaped, spheres, and cenospheres containing fine-grained franklinite and franklinite-magnetite solid solution in an iron-bearing zincite matrix. Slag from calcine processed at 1300°C are also predominantly 1 00-250 um triangular-shaped, spheres, and cenospheres but contain coarser-grained iron-bearing zincite. The zincite in the penets and the slags typically contain 10-20% iron. Based upon these mineralogical observations, it is concluded that the significant amounts of iron in the zincite prevent volatilization of the zinc in the flash reduction process.

Jan 1, 1995

By G. Caetani

I. INTRODUCTION. WE owe to the courtesy of Mr. R. M. Raymond, Manager of the El Oro Mining & Railway Co., Ltd., the permission of publishing in this paper the results of a series of experiments and tests made in view of determining the economical limit to which the fine grinding of the ore by tube-mills could be carried on. For the sake of clearness, the following definitions are given of certain terms used in this paper, as well as of the units and screens in the experiments. Slime : In a general way " slime" is that mill-product, of which from 90 to 95 per cent, will pass a 200-mesh screen (0.067 mm. aperture), and which is treated as one class of material in the El Oro cyanide-plants. As a screen-product, slime constitutes all material that *ill pass a 200-mesh, screen. In sizing, the following products are distinguished:- Coarse Sand : That material remaining on a 100-mesh (0.111 mm. aperture) screen. Fine Sand : That passing through. 100-mesh and remaining on a 200-mesh (0.067 mm. aperture) screen. Slime: That passing through 200-mesh. Fines, which pass through 200-mesh and settle in 30 seconds, and silt, which does not settle in 1 minute. Sand-index : The number proportional to its total fineness considered from the economical point of view, as explained on page 93. Absolute extraction : The percentage of values actually brought into solution. Actual extraction: The percentage of values actually brought to the ziuc-boxes. Units : Dollars, U. S. Currency; pounds; and short tons (2,000 lb.). Gold is valued at $20.67, and silver at $0.50 per ounce. Screens : In the sizing-tests the screens used were:¬ Mesh (per 1 in.), . 24 30 40 60 80 100 150 200 250 Aperture (mm.), . 0.740 0.515 0.424 0.250 0.159 0.111 0.090 0.067 0.058 Strengths of solution : Strengths are expressed in the percent- . age of KCN contained, though the sodium cyanide, which is

Jan 1, 1906

By William A. Fuchs

A study of carbonaceous ore at the Santa Fe mine (Corona Gold Inc.) shows that there is no quantitative relationship between organic carbon content and preg rob, although much of this material preg robs. However, there is a relation between K20 and preg rob. Preg rob generally occurs only in those samples with greater than 2% K20. X-ray diffraction and chemical . studies show that the critical factor is the illite content of mixed (possibly interstratified) illite-smectite. Apparently illite, at a certain threshold, "geo-activates" the carbon. The potential for easily monitoring preg rob using a gamma ray spectrometer is discussed.

Jan 1, 1990

By Kimio Itagaki

The distribution ratios of Ag, Pb, As, Sb and Bi for the Cu2S-FeS or Ni3S2-FeS matte / FeOx SiO2, FeOx-SiO2-CaO or FeOx-CaO slag equilibration in a MgO crucible were determined experimentally at 1523~1573K under controlled partial pressures of S2 and O2 at given PSO2 of 10.1, 50.7 and 101.3 kPa (0.1, 0.5 and 1 atm) and be observed dependency of distribution ratio on the matte grade, SO2 pressure and slag basicity were analyzed on the basis of thermodynamics,

Jan 1, 2003

By Kimio Ilagaki

The distribution ratios of Ag, Pb, As, Sb and Bi for the Cu2S-FeS or Ni3S2-FeS matte I FeOx-SiOl, FeOx-Si02-CaO or FeOx-CaO slag equilibration in a MgO crucible were determined experimentally at 1523 ~1573K under controlled partial pressures of S2 and 02 at given PS02 of 10.1,50.7 and 101.3 kPa(O.I, 0.5 and I atm), and the observed dependency of distribution ratio on the matte grade, S02 pressure and slag basicity were analyzed on the basis of thermodynamics.

Jan 1, 2003

By W. Dresler

Design details of a 100 kVA DC electric-arc reduction furnace are described. The furnace is equipped with a 610 mm (24 inch) diameter graphite crucible hearth. A 76 mm (3 inch) diameter top graphite electrode is used as a cathode. The current passes through the bottom of the crucible and is carried to a bottom plate (the anode) by conductive magnesia carbon bricks. The furnace can be operated up to a maximum of 120 V, 1800 A, and presently 100 kW. The start-up operating experiences and metallurgical results of one reduction test with a Montana chromite containing 38% Cr2O3 are discussed.

Jan 1, 1994

By H. C. Ko

Heats of solution of calcium carbon-ate (calcite) and calcium oxide in hydrochloric acid were determined by solution calorimetry. Combination of these two heats yields a value of 42.643±0.024 kcal/mole for the enthalpy of calcination of calcite at 298.15 K. Literature data on ?H°298.15 of calcination of calcite were reviewed and compared. Thermodynamic properties for the calcination process were calculated from 298.15 to 1,500 K using Gibbs energy functions. This research was undertaken by the Bureau of Mines as an effort to provide thermodynamic data for the advancement of minerals technology, energy conservation, and environmental preservation.

Jan 1, 1982

By D. F. Malan, T. E. Oates

A recent experimental pillar extraction project at a UG2 bord-and-pillar mine presented a unique opportunity to compile a new pillar database. Currently, the South African hard rock bordand- pillar mines are designed using the Hedley and Grant formula with a modified K-value. This empirically derived formula was developed for uranium mines in the Elliot Lake district of Canada. The use of this formula for the design of pillars in South Africa is questionable. Very few pillar failures have nevertheless been observed and its current calibrations for the various reef types are possibly too conservative. A new UG2 pillar database of 66 pillars, of which seven are classified as failed, was compiled by the authors. This enabled a revised ‘first-order’ calibration of the K-value for the Hedley and Grant formula. The new estimated value for the UG2 is K = 75 MPa. This gives a pillar strength that is more conservative than the PlatMine formula. This work should nevertheless be considered as only a preliminary calibration as the database was small. Further work is also required to determine whether the exponents in the formula for the width and height parameters are appropriate for UG2 pillars.

May 3, 2023

By Fred C. Bond and

This paper is a continuation of two earlier papers,12 and presents new data on the grindability of various ores and other materials—the results of several years of intermittent research work on the nature of grinding and interpretation of test results. It is divided into five parts: 1. The tabulated results of several hundred grindability tests on a wide variety of ores. 2. A method of correcting screen analyses and calculating from them the surface and other properties. 3. A description of a new method of measuring the "surface energy" of an ore by pendulum impact tests. 4. Test results. 5. A general discussion of the test results and their application to grinding problems. 1. GRindability of Various Ores This part is a continuation of the paper listed as reference No. 1. The method of testing is described in detail in that paper. The sample to be tested is crushed to all through 6 mesh in rolls set at 1/16-in. opening in closed circuit with a screen. Of this sample 700 C.C. is ground dry in a laboratory ball mill under constant conditions of mill speed and ball charge in closed circuit with a screen of the desired size at 250 per cent circulating load. The mill is cylindrical, 12 by 12 in., and rotates at 70 r.p.m. It contains a charge of 285 iron balls weighing 20,125 grams and ranging from 1½ to ¾ in. in diameter. Each test is continued for a sufficient number of periods to establish equilibrium at 250 per cent circulating load, or until the composition of the circulating load is constant for successive periods. Tests are conducted at the mesh size to which the ore is to be ground in practice, and the results are expressed as the net production in grams of screen undersize per revolution of the mill. This is designated as the

Jan 1, 1939

By Fred C. Bond and

This paper is a continuation of two earlier papers,12 and presents new data on the grindability of various ores and other materials—the results of several years of intermittent research work on the nature of grinding and interpretation of test results. It is divided into five parts: 1. The tabulated results of several hundred grindability tests on a wide variety of ores. 2. A method of correcting screen analyses and calculating from them the surface and other properties. 3. A description of a new method of measuring the "surface energy" of an ore by pendulum impact tests. 4. Test results. 5. A general discussion of the test results and their application to grinding problems. 1. GRindability of Various Ores This part is a continuation of the paper listed as reference No. 1. The method of testing is described in detail in that paper. The sample to be tested is crushed to all through 6 mesh in rolls set at 1/16-in. opening in closed circuit with a screen. Of this sample 700 C.C. is ground dry in a laboratory ball mill under constant conditions of mill speed and ball charge in closed circuit with a screen of the desired size at 250 per cent circulating load. The mill is cylindrical, 12 by 12 in., and rotates at 70 r.p.m. It contains a charge of 285 iron balls weighing 20,125 grams and ranging from 1½ to ¾ in. in diameter. Each test is continued for a sufficient number of periods to establish equilibrium at 250 per cent circulating load, or until the composition of the circulating load is constant for successive periods. Tests are conducted at the mesh size to which the ore is to be ground in practice, and the results are expressed as the net production in grams of screen undersize per revolution of the mill. This is designated as the

Jan 1, 1939

By G. L. F. Powell

Bulk samples of silver have beat undercooled lo a large degree in contact with crystalline siliceous and ceramic oxide materials, indicating lhat stable oxides and oxide compounds are not the heterogeneous nucleants which normally catalyze cryslallization within a feu degrees of the equilibrium freezing point. Since 400-g samples were undercooled 250°C in contact with cristobalite, it is concluded that the undercooling required for homogeneous nucleation of sil11er is greater than 250°C. Undercooling was Jacilitated by precleaning the crucibles with a high-fires -sure blast of gas. IN numerous experiments on the undercooling of metal melts it has been assumed that a solid glassy surface, being non crystal line, will have no effect in catalyzing crystal nucleation from the melt. Conversely, it is usually accepted that any crystalline surface, in contact with a metal melt, will catalyze nucleation of metal crystals from the melt at degrees of undercooling less than that at which homogeneous nucleation will occur. Both these assumptions were implicit in the claim that homogeneous nucleation was observed in experiments1'2 on the undercooling of small metal droplets (10 to 100 mp diam). The droplets were thought to be freed of suspended crystalline particles which would otherwise catalyze heterogeneous nucleation. They were supported on flakes of silica glass which were assumed to be amorphous, and therefore incapable of acting as substrates for heterogeneous nucleation. In later experiments, large samples of molten metal were undercooled to temperatures similar to those achieved by the droplet technique by melting in a fused-silica crucible3 or a boro-silicate glass slag.4 Since these samples were usually fully enclosed in glass at the moment of freezing, it seemed possible that nucleation was homogeneous when maximum degrees of undercooling were achieved, despite the large size of the samples (up to 500 g). In experiments on the undercooling of silver5 the metal was contained in vitreous silica crucibles, the

Jan 1, 1968