Search Documents

By John E. Conley, John A. Ruppert

W. B. Mather—A minor recommendation that may be offered to improve the paper is the inclusion of a map of Oregon showing the general location of the various deposits. This is of especial importance to persons like myself who are not familiar with the various areas and counties mentioned. N. S. Wagner and R. S. Mason (authors' reply)—We agree that an index map of Oregon showing the location of the various nonmetallic deposits would add to the paper (fig. 1).

Jan 1, 1951

By John A. Ruppert, John E. Conley

W. B. Mather—A minor recommendation that may be offered to improve the paper is the inclusion of a map of Oregon showing the general location of the various deposits. This is of especial importance to persons like myself who are not familiar with the various areas and counties mentioned. N. S. Wagner and R. S. Mason (authors' reply)—We agree that an index map of Oregon showing the location of the various nonmetallic deposits would add to the paper (fig. 1).

Jan 1, 1951

By Baojie Zhang

The main goal of this study was not only to investigate the effectiveness of the FGX Dry Separator for removing pure rock from run-of-mine coal producing a better feed stock for conventional coal preparation plants but also to optimize the FGX Dry Separator performance for producing a salable clean coal product without using the conventional wet coal preparation processes. A Model FGX-1 Dry Separator with feed throughput capacity of 10 tph was extensively tested at the Illinois Coal Development Park using two coal samples having distinctly different cleaning characteristics. Statistically designed experimental programs were conducted to indentify critical process variables and optimize FGX Dry Separator performance by systematic adjustments of critical process variable parameters. The coal cleaning performance of the FGX Dry Separator was evaluated for the particle size range of 63.5 x 4.76 mm in most cases, although FGX Dry Separator feed consisted of nominal -63.5 mm run-of-mine coals. The best cleaning performance obtained from the FGX Dry Separator is described by specific gravity of separation (SG50) and probable error (Ep) values of 1.98 and 0.17, respectively. The SG50 and Ep values for individual size fractions were as follow: 1.90 and 0.12 for the +50.8 mm size fraction, 1.95 and 0.18 for the 50.8 x 25.4 mm size fraction, 2.01 and 0.19 for the 25.4 x 12.7 mm size fraction, and 2.03 and 0.23 for the 12.7 x 4.76 mm size fraction. These process efficiency measures produced a clean coal with ash content of 13.38% from a feed coal with ash content of 34.45%. Ash content of tailings and middlings streams were 85.09% and 39.57%, respectively. Total sulfur contents of corresponding streams were 3.87%, 4.68%, 6.87%, and 4.62%. For a relatively easy to clean Illinois No 6 seam coal (Cleaning Index: 0.72), only about 0.42% of the clean coal (i.e., 1.6 float fraction) present in the feed was lost to the tailings stream. For a relatively difficult to clean coal (Cleaning Index: 0.53), about 0.98% of the clean coal present in the feed was lost to the tailings stream.

Jan 1, 2012

By Oliver C. Ralston

This paper reports the results of the use of salt in some research work carried on during the past 3 years at the Salt Lake City Station of the Bureau of Mines, which is quartered in the University of Utah, whose Metallurgical Research Department cooperates with the Bureau. Only the work involving the use of salt is described. The authors wish equal credit given to J. F. Cullen, C. L. Larson, C. Y. Pfoutz, C. E. Sims, H. C. Neeld, R. W. Johnson, F. G. Moses, N. C. Christensen, and G. H. Wigton, who have all in some way contributed toward the success of this work, while connected with the station in one way or another. Lead Metallurgy Popularly Supposed to be Without Problems In lead metallurgy it is difficult at first to find metallurgical problems, as the prevailing conception has been that lead is so easily concentrated in its various ores and so easily reduced to its metallic form that the only progress possible is that involved in cheapening methods of mining and of reduction. Most of the problems usually considered are of minor nature and do not involve radical metallurgical changes. During the three years of work of the Salt Lake Station, the following classification of problems in the metallurgy of lead has been made: Oxidized Ores. 1. Ores containing only lead. 2. Ores containing lead and silver or gold. 3. Ores containing lead and zinc, with or without silver. Sulphide Ores. 1. Ores of pyrite carrying some lead. 2. Ores of complex zinc-lead-iron sulphides. 3. Complex sulphides containing precious metals.

Jan 1, 1918

By D. R. BUDINSKI

Clinton Creek Mine in west-central Yukon produced 0.94 million tonnes of fibre from 15 million tonnes of asbestos ore between 1968 and 1978. The mine closed in 1978. The deposit lies along the margin of a medium sized ultramafic pluton that intrudes schistose and gneissic rocks of the Yukon Metamorphic Complex. Country rocks are metamorphic equivalents of a mixed sedimentary and volcanic assemblage deposited during Precambrian or early Paleozaic time. Movement along the north west-trending Tintina Fault System probably caused the fairly high meromorphic grade that altered the original rocks to schists and gneisses. The ultramafic pluton hosting Clinton Creek Mine is one of several bodies that were emplaced along a deep seated imbricate fault system parallel to the Tintina Trench. Most of rhe intrusions were hydrothermally altered to serpentinite, bur apparently only the one at Clinton Creek contains commercial quantities of asbestos. The Clinton Creek pluton was probably injecred slowly at relatively low temperatures into the crustal rocks as a "mush" that cooled gradually, allowing sufficient time for serpentinization and chrysotile fibre crystallizarion.

Jan 1, 1984

By S. T. Murray

The European industrial minerals industry and its markets are largely self-sufficient. The companies who are members of the Industrial Minerals Association (IMA-Europe) number around 150 contributing 5 billion Euros to the EU?s GDP and range in size from global publicly quoted corporations to small family owned private companies. Their means and capability to implement sustainable development vary considerably. However the leading companies provide an excellent set of best practices, which IMA as a trade association with recognised core competencies in the field of sustainable development is ideally placed to promulgate throughout its membership. As an active participant in the EU Voluntary Scheme for SDI reporting IMA is not only working to increase the level of response to the SDI questionnaire but is also developing with its members a new sustainable development charter which its members will voluntarily adopt as a guiding principle for their implementation of sustainable development This work will be a consultative process involving all IMA?s section members as well as other trade associations, NGOs and MEPs. Public awareness of the benefits of minerals and their contribution to sustainable development has been raised by IMA?s awareness campaign now being followed by a more targeted programme aimed at MEPs. Further work in the future will consider product stewardship and life-cycle analysis. IMA is already positioning itself with the help of its members to assimilate the minerals industry of the accession territories in the sustainable development process.

Jan 1, 2003

By White A, Price G

The main assets of a mining company are its orebodies; the second are the people who find and use them.The graphs in Figure 1 were derived from a paper by Rob Macdonald of Rothschild Australia presented to The AusIMM Centenary Conference in Adelaide earlier this year. They show that the value of the industry on the Australian Stock Exchange halved over the past decade, despite the re-investment of capital equivalent to its full market capitalisation. Why is this so? The average dividend yields for the ASX Mining Index was never above five per cent during that period and barely averaged 2.5 per cent. Why then is $Ai million capital re-investment per employee since 1960 not enough to generate a rate of return to investors better than lousy bank interest? Do the investing public no longer regard our industry's assets to be world class?

Jan 1, 1993

Thank you and good morning. I am delighted to be here and to see so many of my colleagues and competitors. The general topic of this week's conference - World's Best Practice in Mining and Processing - suggests a focus on mining and processing. I am somewhat of a `jack of all trades and a master of none'. But I'm not even a jack when it comes to mining and processing technology and practices, and so I will leave those topics to speakers better 'qualified than I. Instead, what I would like to do is give you my thoughts on the best management practices of mining. You see, the times are changing for the mining business. There is great uncertainty in our industry. Anyone who received their education in an English-speaking school probably remembers the famous poem `Invictus' by William Ernest Henley. The two most famous lines from that poem are, if you remember, `I am the master of my fate; I am the captain of my soul'. The fact of the matter is that the metals industry, including the copper industry, is not the master of its fate. To a great extent we're at the mercy of forces over which we have little control, such as the world economy or political situations in individual countries. The copper industry is like a surfer at Manly - we can't influence where the wave is going; all we can do is stay in front of the curl and on our board. Historically, discovering, developing and processing ore reserves were the primary facets of our capital intensive business. To stay competitive today, we must invest equally in intellectual capital - our employees - in order to succeed in the face of economic, regulatory and political challenges and global competition (Schonberger, 1986). Everyone seems to have a different theory about being competitive. A lot of people use sports analogies when talking about competition. Others, like American businessman and former US presidential wanna-be Ross Perot, take a more interesting approach. In Mr Perot's hypothetical corporation, competition is a lot like having a snake in a factory. If someone sees a snake, the first thing they do is create a committee on snakes. Then, the committee on snakes meets and discusses the problem for a couple of years. The most likely course of action is nothing. Everyone figures that the snake hasn't bitten anyone yet, so they just let him crawl around the factory floor. Perhaps a vice president of serpentine affairs is appointed to continue watching the snake - making sure it doesn't grow. Ross Perot sees it a little differently. In his corporate world, business needs to build an environment where the first guy who sees the snake kills it. Of course in our industry, we better make sure the snake isn't on an Endangered Species List before we go after it! I am not going to talk about competition and competitive forces in terms of sports or snakes, but I would like to share some thoughts with you about how the mining industry can better utilise competitive thinking to get ahead and stay there. You may be wondering how an executive from a 160-year-old company could know much about competitive spirit, but the founding ancestors of Phelps Dodge were wizards at identifying

Jan 1, 1995

By James Douglas

As the operation of light railroads is important to the mining industry, the following statistics of the Arizona Southeastern Railroad may be of interest. When the traffic of the Bisbee copper-mines reached 20,000 tons per year the necessity of better transportation than that afforded by eighteen- and twenty-horse and mule teams became imperative on the score of both cheapness and convenience. Before deciding on the construction of a railroad, the Copper Queen Company made a thorough experiment in transportation

Jan 1, 1899

By John R. Marks

Introduction An engineer is a builder and a problem solver. He uses knowledge of the properties of matter and energy to furnish the needs of society. Engineers form the bridge between scientific researchers and the general public, most of whom are technologically illiterate. Mining, metallurgical, and petroleum engineers extract minerals from the ground and process them into usable forms. Then, other engineers further process them into society's needs. Minerals engineers deal with more uncertainty than other engineers. Each ore deposit has its own personality and each ore is metallurgically unique. Minerals engineers, therefore, must be flexible, often applying larger safety factors and contingencies than other engineers, and frequently functioning as a jack of all trades. A minerals engineer can work in coal, oil, oil shale, base metals, ferrous metals, precious metals, industrial minerals, strategic minerals, or in exploration. Other mining jobs include banking and financing, education and research, marketing and sales, and government. As a minerals engineer, you may be among the first to work on the ocean floor or the moon. College Preparation A difference between college and industry is the type of problems to be solved. A typical college problem has an exact answer derived from given information. An industry problem is more apt to have a solution chosen from a number of alternatives. Information required for the solution is not given. It must be measured, calculated, looked-up, estimated, or otherwise derived. Solutions involve trade-offs, economics, and human factors, as well as application of physical laws and theory. When a college problem is answered, you move on to the next one. In industry, however, the headaches may just be starting with the implementation of the solution. Students often complain that they see little use for the theory they are learning. This is valid. A good professor will link theory and practice. Another difference between college and industry is the price of failure. In college, you can make a half-hearted effort in an assignment and take your C or D. On the job, however, poor engineering results in potentially effective projects being rejected and unsound projects being undertaken. Poor performance results in cost overruns, schedule holdups, oversights, disgruntled employees, irate bosses, unsafe structures, and unreliable systems. The costs are time, money, and possibly lives. The student pays the university for his work - not the other way around. So the rewards for good schoolwork often seem nebulous. But education must be viewed as an investment. Working for good grades will help you think better, learn more, receive better job offers, and perform more effectively on the job. There is no substitute for hard work. Avoid taking the easiest electives. If you desire a more liberal education or dislike rigorous physical sciences, mathematics, or engineering courses, change majors. Remember, you are training to be an engineer, not a computer programmer or market analyst. At least one summer of industry employment is necessary for minerals engineering students. The sooner you gain actual experience, the sooner you will know whether to change majors. Most large companies hire students during summer. It is best to send out applications around Christmas break. First and second year students usually work as laborers. Upper-classmen may work as engineering technicians. Students must sometimes choose between returning to a company at a higher position or trying a different company at a lower position, for more varied experience. Both alternatives have advantages. Do your best on these summer jobs. The company is watching. If you do well, the chances are good you will get a job offer when you graduate. Good references are important when you begin job hunting. Interviewing occurs near the end of your school career. The standard procedure is to send your resume with a cover letter to each company you are interested in. The resume for a graduating engineer with little experience need not exceed one page. The cover letter must be written for the individual company.

Jan 2, 1985

By R. J. Hickson

The Grasberg copper-gold deposit was discovered in Spring 1988, just one year prior to the planned closure of Freeport Indonesia's original Ertsberg open pit copper mine. In spite of Grasberg's remote location, 600m above the Ertsberg, atop a hitherto inaccessible mountain, Freeport-McMoRan brought their latest open pit mine into production in the last week of 1989, less than two years after discovery. This paper describes the start-up of the operation, highlighting synergies drawn from both the original Ertsberg experiences and the current Freeport Indonesia underground mines.

Jan 1, 1991

By R. L. Crosby

A part of the liquidus of the magnesium-zirconium system was determined by chemical analysis of suction samples taken from the melt in the approximate temperature range of 850° to 1,450° C. The melt was prepared in a furnace pressurized with an inert gas. The average zirconium value in weight-percent of the liquidus composition at each sampling temperature was 0.96 at 867° C, 1.40 at 1,062° C, 2.05 at 1,262° C, and 2.96 at 1,402° C. The equilibrium solid phase was zirconium.

Jan 1, 1965

By R. S. Rickard, H. D. Peterson, J. D. Miller, M. C. Fuerstenan

Pure chrysocolla is floated with chelating agents that form insoluble complexes with copper at ambient temperature. Complete flotation is obtained with potassium octyl hydroxamate as collector at pH 6. Flotation response is enhanced with increased temperature when low additions of hydroxamate are involved. A natural ore was floated with 0.4 1b per ton octyl hydroxamate at pH 6.5 and 58°C, and a flotation recovery of 76% was obtained with a concentrate grade of 31.6% copper. The flotation characteristics of the oxide copper minerals, malachite, azurite, and cuprite, have not presented the difficulty for concentration as have those of the copper silicate, chrysocolla. The copper carbonates and oxides respond reasonably well to flotation with conventional collectors, whereas chrysocolla will not respond to flotation with fatty acids or xanthates under normal flotation conditions. In this view then, other reagents will have to be devised to function as collectors for chrysocolla. The most obvious general class of reagents for this purpose would seem to be the organic copper chelating compounds. The utility of chelating agents as collectors in flotation systems has already been demonstrated. For example Vivian1 has floated cassiterite using ammonium nitrosophenylhydroxylamine. Holman2 studied the flotation of nickel oxide ores with dimethylglyoxime and also suggested the use of taurine on oxidized lead ores. A rather detailed study on the application of certain chelating agents to some flotation systems was presented by Gutzeit.3 This work indicated that the formation of surface insoluble chelates is probably responsible for flotation in many cases. The role that soluble chelating agents assume was also presented, that is with effective removal of polyvalent cations by complex formation, effective depression of quartz can be obtained. DeWitt and Batchelder4,5 have shown that oximes function well as collectors for chalcocite, malachite, azurite, and cuprite. Ludt and DeWitt6 have presented the possibility of using dyes, such as octyl malachite green, as a collector for chrysocolla. These works presented by DeWitt et al suggest that organic copper chelating compounds offer considerable promise for concentrating chrysocolla, but that the most suitable chelating reagent from the standpoint of cost and flotation response has yet to be determined. The present investigation was undertaken to study the flotation response of chrysocolla to selected copper chelating compounds, that is ethylenediamine, hexamethylenetetramine, potassium octyl hydroxamate, dimethylglyoxime, and benzoin a-oxime. EXPERIMENTAL MATERIALS Chrysocolla: The chrysocolla used in the microflo-tation studies was from New Mexico and analyzed 20.8% copper. X-ray diffraction showed this material to be amorphous in character, but some faint lines of quartz were noted. A natural oxide copper ore from Utah was also used in the investigation. This ore contains malachite principally but also contains chrysocolla and tenorite together with some sulfide copper. The sulfide copper comprises from 15 to 20% of the total copper content of the ore. Water: Conductivity water was used in the microflo-tation experiments, while Golden tap water was used in the experiments with the natural ore. Reagents: Reagent grade n-amyl alcohol was used as frother in the experiments with pure mineral, while MIBC was used in the work with the ore. Ethylenediamine, hexamethylenetetramine, dimethylglyoxime and benzoin a-oxime were reagent grade in quality. Pure potassium octyl hydroxamate was prepared as follows: 1) 1.0 mole of KOH in 140 cc of methanol was combined with 0.6 mole of hydroxyl-amine hydrochloride in 240 cc of methanol at 40°C (KC1 precipitates under these conditions and to effect complete removal of KCl, the system was cooled to 10°C and filtered), 2) a light methanol wash was then given the KC1 cake, 3) the filtrate was agitated at room temperature and 0.33 mole of the methyl ester of the organic acid (i.e. methyl octanoate) was

Jan 1, 1965

By S. R. Gilbert, B. W. Burdett, L. B. Hales

A brief review of grinding control systems is given with a review of the most popularly reported control loops. The trend toward more sophisticated control systems in terms of control strategies and control equipment is discussed, and the reasons for the popularity of distributed control equipment are outlined. The evolution of the interface between the operator and the process and its importance in plant control is described. Finally, the authors' view is given as to the future of grinding control.

Jan 1, 1984

By C. B. Ndimande, C. Ntsele, M. Lisso, L. Arlow

"In the processing of UG2 platinum ore, stirred milling technologies are dominating regrind applications that require grinding the particles to P80s’ in the region of 45 µm. For this application stirred mills are 30–40% more energy efficient and require a smaller foot print compared to traditional tumbling mills. In this work a study was carried out using a batch pin mill and three scales of Stirred Media Detritors. A laboratory scale pin mill, a pilot scale Stirred Media Detritor with an installed power of 18.5 KW and two industrial scale Stirred Media Detritors in continuous operation, with an installed power of 355 KW and 1,100 KW were used. The industrial scale SMDs were in the secondary grinding section of a mill floatmill float (MF2) UG2 processing plants. The tests were focused on assessing the effect of feed rate and percent solids on power consumption and product fineness at lab scale, floatation response of the SMD product at pilot scale and on an industrial scale, the effect of feed rate on energy consumption and product fineness. The measurements obtained were feed flow rates, mill power and from the samples cut, particle size distribution of both feed and product. Specific energy was observed to increase with increasing percent solids over the 30wt% to 80wt% studied. Floatation response tests of the pilot scale SMD products show improvement in grade. Specific energy and reduction rations are observed to decrease with an increase in feed rate.INTRODUCTIONThe purpose of comminution in most mineral processing applications is to liberate the valuable minerals from gangue so that they can be recovered in subsequent separation stages. Most precious and base metal plants treat low grade fine-grained ores which require a higher degree of size reduction to yield the necessary degree of liberation (Partyka & Yan, 2007). Many authors have shown that the relationship between product size and grinding energy consumption is exponential, the finer the product from the comminution device, the higher the energy consumption (Gao & Weller 1993; Napier-Munn, Morell, Morrison & Kojovic, 1996; Jankovic, Valery & La Rosa, 2003). The suggested reason for this kind of relationship is that smaller particles have fewer flaws and higher strength than larger particles. Particles with fewer flaws and higher strength require higher energy for breakage to occur. One of the challenges most plants face is that of efficient energy utilisation in the liberation of the mineral from these complex ores. Conventional tumbling mills are not energy efficient in fine and ultra-fine grinding applications because some of the energy is spent on moving the charge and does not go directly into breakage. Stirred mills have been found to be about 30% to 40% more efficient and require a smaller foot print compared to traditional tumbling mills for regrind applications that require product size with P80s’ in the sub 45 µm regions (Radziszewski & Allen, 2013). Stirred mills are able to produce fine product because they can achieve a high number of stress events per unit time and volume compared to tumbling mills (Kwade, 1999a). To minimise energy usage without compromising product quality, the mining industry has adopted stirred milling technologies for fine and ultra-fine grinding (Radziszewski & Allen 2014; Kwade, 1999b). The south Africa platinum group metal (PGM) industry alone had over 40 stirred media mills installed by 2011 (Rule, 2011). Of that 40, 13 were in the ultra-fine grinding (UFG) application and the rest in main stream inert grinding (MIG)."

Jan 1, 2016

By R. D. Leitch

The work of the United States Bureau of Mines on stream pollution by coal-mine drainage has included a study of seasonal variations in acidity and volume of several streams in Pennsylvania. The acidity of these streams has been determined during the two or three years prior to 1930 and the results may be considered to represent "normal" low water conditions. The summer of 1330, however, was unusually dry and represents what may be termed "extreme low water conditions on these streams. This report is a comparison of the results obtained for extreme low water of 1930 and for other years. The streams investigated were Buffalo Creek in Somerset County, Indian Creek (lower 5 miles) in Westmoreland County, Little Toby Creek in Elk and Jefferson Counties, Daguscohonda Run in Elk County and West Creek in Elk and Cameron Counties. These will be treated separately in the order given. Most of the descriptive matter and physical data pertinent to the streams and drainage basins are taken from published reports of the Pennsylvania State Water Supply Commission, 1917.

Jan 1, 1931

By Kenneth M. Earl

In its program of investigation of deposits of strategic minerals, the Bureau of Mines did some development drilling at the Douglas Deposit in an attempt to find an additional source of supply of copper. The deposit, which was opened in 1878, was worked until 1883 and is reported to have produced between 2 and 3 million pounds of copper. Heap-roasting and smelting and later gravity concentration were employed in ore reduction. Old slag dumps and tailing piles from these operations still remain. In 1917, the American Smelting & Refining Co., purchased the mine, and in June 1918 production was started in a 125-ton mill, using both tables and flotation. The mill was closed December 31, 1918, and soon after the milling plant Was dismantled and removed. The mine has remained idle ever since. The deposit occurs in a mineralized zone that follows the schistosity. The principal metallic minerals are pyrite) chalcopyrite, pyrrhotite) arsenopyrite, sphalerite, magnetite) slight chalcocitel and cuprite. To search for additional ore shoots and to prove the known ore shoot at depth, the Bureau of Mines drilled seven holes, which totaled 2,117.4 feet. No additional ore was developed by this work.

Jan 1, 1950

Historically, mine ventilation practices as it was referred to then and which was enforced by legislation of those days, focused primarily on the conditions in the working place and little attention given to actual exposures of workers to contaminants in the working environments. With the advent of the Leon Commission of Inquiry into Safety and Health in the South African Mining industry in 1994, findings in the report indicated that more attention should be given to occupational health of workers. The upshot of this report resulted in the promulgation of the new Mine Health and Safety Act (MHSA) with certain sections of it devoted to all the aspects of occupational medicine and occupational hygiene. This paper describes the present situation with regards to statistics on occupational hygiene performance and compensation paid out to persons who have contracted occupational diseases, as well as the way forward in terms of legislation implemented in September 2002 on the chronic and acute issues affecting the mining industry in SA.

Jan 1, 2003

By Elmars Ence, Harold Margolin

The titanium-aluminum system has been investigated in the composition region 0 to 34 pct Al in the temperature range 800" to 1450°C. The phases encountered in this region were: a,ß, TiAl3, The reactions observed are as follows: 13 ; L - 6T. A1, 0 (15 pct Al) + 6 (18 pct Al) - yTi Al (16.5 pct Al) at -1170°C andP (7 pctpct Al) + y (9.2 pet Al) - ah(9 pct Al) at - 1060°C. A eutectoid reaction: y — a +d belm 700°C is postulated. THE Ti-A1 system has been the sub.ect of a number of investigations. Several of these1-3 have indicated extensive solubility of A1 in both a and ß titanium. Ence and Margolin, as a result of work on the Ti-A1-0 system:'5 have shown that the a solubility ofalumi-num is considerably restricted, and found that a compound, Ti2A1 existed in the region formerly designated as Q a This structure was identified as hexagonal with c/a = 0.803. Pietrokowsky confirmed the existence of this phase7 and pointed out that the structure was isomorphous with Ti3Sn.8 Although the existence of Ti,A1 has been confirmed by others,9-12 there has been considerable disagreement regarding the composition and mode of formation. A detailed reinvestigation of the Ti-A1 system has been published by Sagel et al.13 The tentative diagram indicated two phases, a, and e, occurring in the former all a region. The a, phase has the same structure as Ti2A1 and was indicated as having a broad region of solubility. Epsilon, a tetragonal phase, was shown to form within the a, solubility region. AS a result of unpublished work,14 Ence and Mugy3 olin disagreed with the interpretation of Sage1 et al., although some common features existed. The present study was conducted in an attempt to clarify prevailing uncertainties. EXPERIMENTAL PROCEDURE Alloy Preparation—Alloys were prepared from iodide titanium (99.9 pct Ti with less than 0.01 pct Zr) or Bureau of Mines electrolytic titanium (BHN-73) and high purity aluminum (99.99 pct). Alloys were arc-melted in argon or helium atmosphere as 15-g buttons in the composition range up to 34 pct Al. Depending on need, alloys were made in increments varying from 0.25 to 2 pct Al. Up to five 15-g buttons of some compositions were made. Weight losses during melting did not exceed about 1 pct of the charge and chemical analysis revealed that both titanium and aluminum were evaporated. In general, nominal and analyzed compositions agreed within 0.5 pct and therefore nominal compositions were used in plotting the data. Unavoidably, some overlap of composition occurred, particularly where small increments of A1 were used. The small increments were primarily used to obtain alloys which would fall into the narrow a -, r region and this purpose was achieved. In general, the data obtained from Bureau of Mines base titanium and iodide titanium alloys were in agreement. Forging—Both as-cast and forged samples were employed. Samples were heated for hand forging by an oxygen-hydrogen torch to the region 1200" to 1300°C. A titanium anvil and cover plate were used to prevent any iron pick-up and consequent formation of a low melting point compound. Frequent reheating kept the temperature in the desired range. Under these conditions, alloys containing up to 15.5 pct could be successfully forged. To check on the depth of contamination, forged alloys were heated slightlybelow the ß transformation temperature. Metal was removed to a depth below the contaminated surface so revealed, and the subsequent results obtained from these samples agreed closely with those secured from as-cast material. Heat Treatment—Prior to heat treatment, all alloys were vacuum annealed to remove hydrogen. Samples

Jan 1, 1962

By C. W. Covington, H. C. Cook, J. F. Libsch

Sputtered aluminum thin films were prepared in each of two conventional bell-jar vacuum systems. One system utilized an inner "getter sputtering" enclosure; the second system was a standard diode sputterirlg arrangement. Consistent and repeatable film properties were obtained in both systems provided sufficient cleanup and presputter time was allowed. The various problems associated with aluminum sputtering are discussed, with pavticular attention to the vzinimization of film contamination by outgassing and oxidation. The growth and structure of the aluminum thin films were studied with the aid of electron wzicroscopy and related to deposition rate, substrate temperature, and film thickness. The resistivity of the films was correlated to film structure and surface roughness. Resistiuities as low as 3.5 microhm-cm (2.3 x bulk resistiuity) were measured for relatively thick films (20,000). THIS investigation was undertaken to explore the problems and film characteristics associated with the sputtering of aluminum. Earlier work reported on aluminum sputtering has stressed the contamination of both the cathode and film as well as abnormally low deposition rates.' Data on sputtering yields,3 however, indicate that aluminum sputtering could be practical. Recently, it was shown by Theuerer and Hauser3 that aluminum sputtering was feasible in a conventional oil-diffusion pumped vacuum system provided an inner "getter sputtering" enclosure was employed to minimize gaseous contamination. This technique employs the getter action of aluminum to produce extremely low reactive gas partial pressures in the deposition zone. Aluminum films have already found use in solid-state electronic devices such as transistors and integrated microcircuits. To the present time, most aluminum films have been prepared by evaporation due to the ease with which this metal can be evaporated. Controlled sputtering of aluminum offers a number of advantages over evaporation, however, such as 1) manufacturing compatibility with other metals which must be sputtered, 2) better control of process variables resulting in more repeatable and uniform films, 3) generally better adhesion resulting from higher-energy metal atoms, 4) built-in material supply for continuous operation, and 5) the ability to deposit alloys. The data presented in this paper are the result of two separate studies. One study utilized a small "getter sputtering" enclosure (System A) and was concerned with the structure and properties of thin aluminum films up to about 1600A thick. In this thickness range, study of the film structure was possible with transmission electron microscopy. The second study was conducted with a conventional diode sputtering apparatus (System B) and covered $ much wider range of film thickness up to about 34,000A. Emphasis was placed on developing techniques which would extend aluminum sputtering to an existing in-line continuous deposition process.4 I) EQUIPMENT AND PROCEDURE System A. Fig. 1 shows a schematic of the "getter sputtering" inner chamber used and Fig. 2 is a photograph of this chamber mounted on an 18-in. access ring of a standard liquid-nitrogen trapped bell-jar vacuum system. This chamber was designed after a similar system used by Theuerer and auser.3 The configuration allowed the glow to emanate in all directions from the cathode. Argon (99.99') was allowed to enter only at the top of the chamber such that most of the reactive gases entering with the argon were reacted with the aluminum vapor and deposited on the chilled walls of the chamber before reaching the deposition zone. Provision was also made as shown, Fig. 1, for heating and cooling the substrate from about 20" to 400°C. Two substrate materials were used: Corning 7059 glass (12 by 12 by 0.048 in.) and carbon films about

Jan 1, 1967