Search Documents

By Chen Xia, G. Deschenes, Wan-Tai Yen

"Ammonium thiosulphate is a potential replacement for the cyanide in gold leaching. However, the high reagent consumption hinders its practical applications. In order to reduce the thiosulphate consumption, chemical additives were used. The gold leaching tests were conducted on a sulphide copper-bearing gold ore with 0.3M thiosulphate, 0.03M copper sulphate and 3M ammonium hydroxide solution at 44% solids and pH 10.2. Sulphate, sulphite, amino acid and chelating agents were initially used as additives. It was found that chelating agents (EDTA and NTA) could reduce the thiosulphate consumption from 30 kg/t to 17.8 kg/t with some increase of gold extraction. The improvement from other additives was not significant. Other approaches were also attempted to reduce thiosulphate consumption with some success. The thiosulphate consumption could be reduced to about 17 kg/t by reducing copper sulphate concentration fiom 0.03M to 0.0015M; adding 0.04M of EDTA; replacing copper sulphate with 0.001M of nickel sulphate. The increase of pulp density fkom 44% to 60% solids could reduce the thiosulphate consumption to 17 kg/t as well. Limiting the dissolved oxygen content by sparging nitrogen reduced the thiosulphate consumption further to 10 kg/t. It could be concluded that the regulations of copper and oxygen supply in the lixiviant were the keys to reduce the reagent consumption.INTRODUCTIONCyanidation, as the predominant gold leaching method, has been used in mining industry for more than a century. The low reagent requirement and the high performance are the reasons for its great success. However, cyanidation cannot deal with ores containing large amount of copper, nickel, arsenic and zinc minerals. This type of ore may consume more reagents and result in a poor gold extraction. On the other hand, the carbonaceous materials may preg-rob the gold from pregnant solution and make cyanidation a problem leaching method. Further more, the high toxicity of cyanide represents another challenge to modern gold mines. In the beginning of the year 2000, an environmental disaster caused by an accidental spill in Yugoslavia was broadcasted world-widely (Gurevitch 2000). Gold mine operators are facing higher social pressures and increasing costs on environmental protections. The usage of cyanide in mining industry has been banned in some countries and locations such as Czech Republic, Japan, Turkey, California, Colorado, Wisconsin and Montana States. Thus, a non-cyanide lixiviant is required."

Jan 1, 2002

By James A. Finch, Robert E. McIvor

"This paper presents the development of a new grinding circuit steady-state modeling system. It is fundamentally based on the “functional performance equation”, which has now been used to improve the performance of numerous grinding circuits. This equation characterizes the two distinct efficiencies that are at work in a ball milling circuit, and how these relate to overall circuit productivity. It does so based on the particle size chosen to distinguish between coarse and fine (product size) material. Now, by applying it to all the material size classes, a simple yet reliable ball mill model was created. Combined with cyclone recoveries, the complete, detailed grinding circuit performance can be exactly reproduced mathematically by the computer program. It can then be used with separate process equipment application engineering methods to systematically engineer grinding circuit improvements. The development, features, uses and limitations of the system are described.INITIAL DEVELOPMENT OF THE CIRCUIT MODELThe Functional Performance Equation for Closed Circuit Ball (or Pebble) MillingIt has been shown that closed grinding circuit performance can be effectively characterized by the “functional performance equation” for closed ball or pebble milling circuits (McIvor, 2006a). One method of its derivation is summarized here for convenient reference.First select a cut-off size to define “coarse” versus “fine” or product size material for the grinding circuit. Any particle mesh size can be used to differentiate between coarse material and fines, but the circuit target P80 is often an effective choice for these purposes.Any energy expended on “fines” is wasted. It also generates “extreme fines” often detrimental to downstream processing (McIvor and Finch, 1991). Only the portion of the mill power draw that is expended on coarse material is effectively used. This portion can be estimated from the mill feed and discharge size distributions (McIvor, 1988a) by taking the average of the percentage of “coarse” material in the mill feed and discharge. We can define the thusly calculated percentage of the mill power that is expended on coarse material in the mill as the ball mill circuit “classification system efficiency” (CSEff). The “effective mill power” (EMP) can be calculated from the total mill power (TMP) as follows."

Jan 1, 2007

By D. I. Smith

Constructing tunnels in congested urban areas requires tunnels to be placed closer together and with less cover than was previously necessary. Contract CQ031 of New York?s East Side Access project will construct four six-meter diameter bored tunnels in a three-up one-down ?T? configuration, with less than three meters of soil between the tunnels. Mixed-face rock and glacial till conditions with a high ground water table will be encountered under active railroad yards. The tunnels will also pass beneath the busiest railroad interlocking in North America with cover as shallow as three meters. This paper discusses the challenges of designing segmentally lined tunnels for these conditions.

Jan 1, 2008

Lead isotope measurements indicate that lead from the A.B.H. Consols mine and two transgressive veins in Browne's shaft is "Thackaringa" type and not "Broken Hill" type.

Jan 1, 1970

By M. Botz

Anatolia Minerals Development Limited (AMDL) is constructing the Çöpler Gold Project in eastern Turkey and commenced mining in the fourth quarter of 2009 with gold production scheduled to begin in the fourth quarter of 2010. The project is a 15,500 tonnes/day oxide heap leach with three-stage crushing, agglomeration, carbon adsorption and electrowinning for final metal recovery. Leach solution copper levels are expected to be relatively low during the first two years of production but will increase in later years due to higher copper content in the mined and processed ore. Over the mine life approximately 40 million tones of ore will be placed on the leach pad, which will contain about 35,000 tonnes of copper. Metallurgical testing indicates copper extraction will vary by ore type and, if controls are not implemented, may lead to copper levels exceeding 1,000 mg/L in pregnant solution. High copper levels in leach solution have the deleterious effect of increasing cyanide consumption and potentially overwhelming the ability of the carbon circuit to effectively adsorb the principal economic metals, which in this case are gold and silver. Using data obtained from laboratory leach column testwork, the heap leach facility was modeled using METSIM process simulation software to estimate the build-up of copper in leach solution over the mine life. Laboratory testwork was also performed to evaluate process conditions for treating a portion of pregnant leach solution via sulfide addition to precipitate saleable copper as copper sulfide (Cu2S).

Jan 1, 2011

By Faruk Oksuz, Clay Haynes, Miguel Sanchez, Patrick Finn, Roger J. Maurer

The Milwaukee Metropolitan Sewerage District’s (MMSD)’s challenges of basement flooding and wet weather sewer overflows are being creatively addressed with the Northwest Side Relief Sewer tunnel. The $132 million, 20-foot diameter, and37,740 feet long conveyance and storage tunnel excavation in dolomite at depths reaching 165 feet below ground surface has been successfully completed within13 months in April 2004. This paper includes Owner’s perspective on the overall project, and the design and construction details by the Designer and the Contractor. The project contract documents included a comprehensive Geotechnical Data Report and the Geotechnical Baseline Report.

Jan 1, 2005

By C. Schuffeneger

"The design of SAG milling circuits for mining projects requires the incorporation of the energy consumption considerations and the evaluation between different flexibility-adding options to absorb the time-dependent variation in the ore lithology. In order to expand the focus of the engineering design, a higher level of deepness and quality in the baseline information must be achieved to elucidate all of the constraints on throughput, energy consumption and product quality generating a design optimization consistent with an integrated vision of the process. This paper presents a review of a collection of typical examples of the copper and gold mining project in the Chilean Andes, in which both process and energy efficiency criteria are considered when the focus is to deliver an integral solution that incorporate those areas in the design. The intensive use of energy in the SAG milling, the process continuity and the quality of the product are a function of the proper design and selection of grinding equipment and plant layout, aligned with the mining plan and associated with projected future expansions and production increases.INTRODUCTION Autogenous and Semiautogenous grinding technology, large diameter and short length mills, begins with the Handsel design. During the decade of 1930’s, Handsel-Hardinge studies led to cascade mill development. Nowadays, 80 years later, continuous improvements in design and technology have been conducted to a completely new generation of gearless SAG mills that have led to an increasing capacity at the concentrator plants. Considering the existing alternatives, SAG mills are the most implemented option for the grinding circuit today. This work presents a review of the SAG mill implementation parameters and circuit considerations and its correlations in the most recent projects in Chile with an ore treatment capacity above 50 ktpd since the decade of the 90’s up to now. From this period, the tendencies will be analyzed using project design data. Based on process design information, the recent SAG mill history in relevant projects related to the new generation of commercial SAG mills, technical associated considerations and performance will be discussed. Circuit selection based on energy utilization will be revised to bring up a more holistic perspective on the SAG mill circuit design concepts. As is now well known, the heart of every concentrator plant are the grinding mills. Higher demands on throughput, availability and power consumption optimizations require not only high performance electrical equipment but also a well selected grinding circuit. The energy efficiency issues associated to the circuits design will be reviewed."

Jan 1, 2012

By E. Henry, C. Mayer, H. Rott

Hangingwall deformation monitoring at Kiruna mine is a major concern for Luossavaara- Kiirunavaara AB (LKAB) since underground production is deepening toward the town of Kiruna. In the quest for more effective monitoring techniques, LKAB decided to evaluate the ability of satellite-borne synthetic aperture radar (SAR) interferometry to monitor ground subsidence by generating continuous deformation maps from satellite radar images. While adopted by other geoscience-related industries, SAR interferometry is still relatively untested in the mining industry. This paper describes the basic principles of SAR interferometry and describes how subsidence is monitored at Kiruna mine. The encouraging results are discussed as well as the benefits (continuous deformation mapping, low cost, fast, etc.) and limitations (sensitivity to snow and vegetation cover) of the technique.

Jan 1, 2004

By Christopher J. Bise

INTRODUCTION The principal means for transporting men, supplies, and mined material in an underground or surface mine are rail systems, conveyor belt systems, or a combination of the two. Typically, large underground mines use rail haulage from the mining sections to the hoist, with conveyor belts being used within the sections. Small mines use conveyor belts in the mains and sub- mains as well as in the panels. The size and layout of surface mines also contribute to variations in the way these haulage systems are applied. The following sections address the typical problems confronting mining engineers in designing these haul- age systems. For rail, these include rail selection and turnout design, locomotive tractive-effort calculation, and motor sizing. Belt haulage calculations include belt width and speed specifications, tension and take- up requirements, idler spacings, and motor sizing. TRACK LAYOUTS Introduction The adequacy of mine track layouts is dependent upon such factors as rail weight (Table l), suitable ties, a well-drained roadbed, and properly designed curves, grades, and track alignment. As long as the installation conforms to the manufacturer's recommendations, good service can be expected from the rail system. Typically, mining-engineering calculations for rack layouts have dealt almost exclusively with the design of turnouts. As such, the following section provides the necessary information for computing the theoretically correct dimensions of turnouts. Turnout Design Fig. 1 shows a typical turnout with an identification of the component parts. Fig. 2 identifies the variables that are used to determine the turnout dimensions. The formulas that concern the mining engineer are those listed: [ ]

Jan 1, 1986

By A. D. Davis

Limestone-based technology for arsenic removal from water is an innovative and promising method. The technology offers the potential for low-cost disposal of a stable waste product after arsenic removal, either in an ordinary landfill or by encapsulation in concrete. Leaching research described in this paper shows that the limestone waste, after encapsulation in concrete, passes the Toxicity Characteristic Leaching Procedure (TCLP) test. Leaching was less than 0.05 mg/L, which is 1/100 of the U.S. EPA?s standard for disposal of arsenic in a landfill. Disposal of arsenic-enriched waste is critical for commercial viability of arsenic-removal technologies. Low-cost disposal gives the limestone-based method an advantage that could help communities meet the new maximum contaminant level of 10 parts per billion for arsenic in drinking water. The ability to recycle the waste material in concrete could add a significant economic benefit, further reducing overall costs.

Jan 1, 2009

By O. M. McRae

The Courtland-Gleeson area is in Cochise County about 15 miles east of Tombstone in southeastern Arizona. Rocks exposed in the area range in age from Pre-cambrian to Quaternary. The Precambrian is repre-sented by the Pinal schist. Paleozoic rocks include the Cambrian Bolsa quartzite and Abrigo limestone, Mississippian Escabrosa limestone, and the Permo-Carboniferous Naco Group limestones. Rocks believed to be of Triassic and/or Jurassic age include the Copper Belle monzonite porphyry, Turquoise granite, and Gleeson quartz monzonite. The Cretaceous is represented by andesitic volcanics and various sedimentary rocks. Tertiary rocks include the intrusive and extrusive Sugarloaf quartz latite porphyry and numerous dikes of various rock types. Quaternary conglomerates cover the northern and eastern part of the area. The geologic structure is extremely complex. All rocks as old as early Tertiary have been involved in either tilting, normal faulting, high angle reverse faulting, imbricate thrusting or folding and overfold-ing. Major deformation was initiated during the first half of the Tertiary. Northeast-southwest compres-sional forces caused thrust plates containing lower Paleozoic rocks to override Cretaceous rocks. Subsequently, two large slide blocks of lower Paleozoic rocks, acting under the influence of gravity, moved eastward. The imbricate structure and the gravity slide blocks were then broken by late Tertiary normal faults. INTRODUCTION The Courtland-Gleeson area, or Turquoise Mining District, has long been recognized as an area of extreme structural complexity. Gilluly 2 referred to the entire district as a "gigantic thrust breccia". In 1962, Bear Creek Mining Company completed a two year program of detailed geologic mapping supported by exploration drilling. The purpose of this paper is to present some of the results of this investigation and to provide a structural interpretation based on these results. The geologic map that accompanies this report is generalized and shows only the major and significant features. The area was initially mapped on aerial photographs on a scale of 1 in. equals approximately 500 ft. These data were transferred to a planimetric base using a vertical sketch master. LOCATION The Courtland-Gleeson area is in central Cochise County, about fifteen miles east of Tombstone, and about twenty-two miles northeast of Bisbee in southeastern Arizona (Fig. 1). The area is readily accessible over graded gravel roads from both Tombstone and Elfrida. GENERAL GEOLOGY Rocks exposed in the area range in age from Pre-cambrian to Quaternary. wilson3 and Gilluly2 have published very detailed descriptions of the principal rock units exposed in the area. Only a summary description is provided in Table I. A great number of previously unmapped narrow Tertiary dikes and small irregular masses of hornblende andesite porphyry, lamprophyre, and aplite intrude the Gleeson quartz monzonite in the western part of the map area. The scale of the map prohibits showing them. The aplite dikes are related to, and bear the same structural relationships as, the Sugarloaf quartz latite dikes found in this area. For that reason, and because both the hornblende andesite porphyry and the lamprophyre are post-mineral and post-thrusting, their omission from the map is not considered significant. Because so many of the contacts in the district are fault contacts, because bedding faults appear to be prevalent but are difficult to recognize and map, and because igneous intrusions are so common throughout the sedimentary section, it is virtually impossible to measure an exact total thickness of any of the sedimentary rock units. Nevertheless, neither the field mapping program nor the diamond drilling provided any basis for suspecting that the thicknesses of the sedimentary units are significantly greater or less than what might be inferred from surrounding districts.

Jan 1, 1967

By Dean E. Harris

The realignment of U.S. Highway 95 in Sandpoint, Idaho, is being designed on a site underlain by glaciolacustrine soil that includes very thick deposits of soft, sensitive clay to estimated depths of greater than 200 m. The highway realignment will include construction of numerous embankments, mechanically stabilized earth (MSE) retaining walls, and pile-supported bridges. The area of the proposed new highway corridor also includes existing historic and commercial buildings, a railroad, sensitive environmental areas, and economically vital tourism/recreational assets. The evaluation of pile axial capacity included the use of a variety of methods, including total stress, effective stress, and empirical methods using the results of piezocone soundings and conventional borehole and laboratory tests. These methods were used along with the findings from an instrumented static loading test to design the piled foundations for the bridges. A critical factor throughout the design of the pile foundations was controlling settlements to levels that could be tolerated by the bridges. Also, because of the proximity of the pile foundations to the existing structures and the proposed new embankments, the project has required careful consideration of the potential interaction between these elements. The analysis of piles also considered the potential for settlement of embankments, drag loads on the piles, and downdrag.

Jan 1, 2003

By D. B. Felsher

"Lakeshore Gold’s Bell Creek Mill is located in Timmins, Ontario. The mill has been in operation since 2009 and has undergone several small expansions since then. In 2012 - 2013 the first major mill expansion was commissioned raising the mill capacity from 2,000tpd to 3,150tpd. This was done through the installation of a new ROM ore truck dumping facility, jaw crusher, ore storage dome, SAG mill, leach tank, CIL tank, and electrowinning cell. This paper will cover the equipment installed to raise the tonnage as well as the commissioning of the new equipment including problems encountered and how these problems were solved.IntroductionLake Shore Gold acquired the Bell Creek Mill, along with the Bell Creek Mine, from the Porcupine Gold Mines Joint Venture, a joint venture between Goldcorp Canada Ltd. and Kinross Gold Corporation, in December 2007.The Bell Creek mill, which processed ore from the Bell Creek Mine has consistently achieved recoveries of over 95%, is a conventional gold mill circuit, involving crushing and two stage ball mill grinding, gravity and leaching, followed by Carbon-In-Pulp (CIP) process for gold recovery. The facility had been placed on care and maintenance in 2002. At the beginning of 2009, the mill had been refurbished with an initial capacity of 800 tonnes per day and was subsequently refurbished to a capacity of 1,500 tonnes per day near the end of the third quarter of that year. The Mill was then expanded in 2010 to a capacity of 2,000 tonnes per day.In late 2011, the Company commenced a 50% expansion of its milling facility, from a processing capacity of 2,000 tonnes per day to 3,150 tonnes per day. Phase 1 of the expansion, which mainly involved enhancements to the solution circuit (“back end”) and increased the mill’s capacity to 2,500 tonnes per day, was completed at the end of 2012. Commissioning of Phase 2 of the expansion, to a capacity of 3,150 tonnes per day, commenced at the end of July 2013."

Jan 1, 2015

The development of research activities at Risdon and the implementing of a policy of instrumentation in plant processes have, in recent years, entailed a considerable increase in research staff and created an urgent need for more space and modern laboratory facilities. The existing laboratories have been in use for 25 years and are now inadequate. New buildings have been authorized.The new buildings will comprise, firstly, a laboratory building of three floors giving a totai .floor space of approximately 11,000 sq.ft., and a pilot plant building as an annexe to the laboratory building, of approximately 4,900 sq.ft. The laboratory building will accommodate from 20 to 22 research workers, and additional plant operatives will be used for the control of processes in the pilot plant building. The new buildings are designed to provide facilities for three principal functions; the investigation and improvementof existing processes, the development of new processes up to the pilot plant stage, and the control and maintenance of instruments.LOCALITY AND ORIENTATION OF BUILDINGSThe site of the buildings is elevated ground on the eastern side of the main plant, near the main entrance and general office building. There will be no obstruction to sunshine at any time of the day, and the laboratory building will command views of wide radius from wooded hills across the River Derwentto the open sea on the south-eastern quarter.The buildings are located to provide space for future extensions. The main laboratory building faces approx...

Jan 1, 1949

Three computerised sets of minerals information are being developed by New Zealand Geological Survey in co-operation with the Resource Management and Mining group of the Ministry of Energy: 1. Geological Resource Map: a file of information on the location, geology, geochemistry, production, and potential use of individual occurrences of aggregate, mineral, geothermal, coal, oil, and gas resources. The computer program can be used for search and retrieval of information and the production of maps and tables. The data is being published as a series of 1:250 000 maps with accompanying text and tables. 2. Minerals Bibliography: an index and retrieval system for mineral exploration related; open-file company reports, N.Z. Geological Survey files, scientific research publications and university theses. 3. Minerals WhoÆs Who: an annotated directory of geologists involved in mineral exploration, mining and research.

Jan 1, 1986

By Dale Dean, Bill Hancock, Raul Mondragon, Carlos Aguilar, Chris Martin

"In 1942, commercial mining operations began in what is now the Lucky Friday Ag/Pb/Zn mine in Idaho’s Silver Valley. Wholly owned since 1962 by Hecla Mining Company, the mine has produced more than 145 million ounces of silver and has been a mainstay producer in Idaho’s Silver Valley for over 70 years.However, by the early 21st century, the mill had fallen, technologically, seriously behind the rest of the industry. Accordingly, starting in 2005, the current mill management team set to revitalizing the milling operation. This paper will describe how the mill was systematically evaluated and the aging infrastructure replaced with more current technology. Driven by internal mill review studies supported by QEMSCAN mineralogical balancing, the mill has in recent years been through waves of upgrading – from the installation of new mechanical and column flotation equipment to the introduction of on-stream analysis, a technology well-accepted around most of the industry but alien to this technologically conservative mining district.The experience at Lucky Friday is a testament to the need for ongoing structured mill optimisation and technology upgrading in established mining operations. This paper will describe engineering experiences associated with upgrading equipment in an operating mill, contained in a confined building. Experiences in teaching the team new ways of operating a mill, and finally the metallurgical rewards that have been reaped will be described.INTRODUCTIONMining in Idaho’s Silver ValleyIdaho’s Silver Valley is a 64-km long valley flanked by the US Northern Rocky mountains, seldom more the 3-km wide, and stretching across much of Idaho’s northern panhandle, roughly 150 km south of the Canadian border. Mining has been continuously active in the area for more than 140 years, during which time, the Silver Valley has become one of the richest silver producing districts in the world, and as of 1985 had out-produced the 500 years of production from the famed Potosi district in Bolivia. Since 1885, almost 100 deposits have been discovered and worked in the district (Chapman, 2000)."

Jan 1, 2011

By L. A. Ruth

Vision 21 is a government/industry/academia cost-shared partnership to develop the technology basis for integrated energy plants that will, early in the 21stcentury, result in the deployment of ultra-clean plants that produce electricity and “opportunity” products. Vision 21 plants will use fossil fuel feed-stocks in combination with other domestic resources, e.g., biomass, municipal waste, and petroleum coke. Opportunity products could include transportation fuels, steam, high-value chemicals, synthesis gas, and hydrogen. Performance targets for Vision 21 plants, example plant configurations, and the key technologies that will be needed are described.

Jan 1, 2000

By Zhouhua Jinag

The attempt has been made to manufacture high nitrogen austenitic stainless steels by pressurized electro-slag remelting furnace with maximum nitrogen pressure of 7MPa. To obtain high nitrogen content in steels, the compound electrodes with different nitrogen sources (FeCrN, Si3N4) were prepared. Using the Si3N4 as the nitrogen alloying source, the silicon contents in ingots were prone to be out of the specification range, and the electric current fluctuated greatly and the surface qualities of the ingots were poor. The surface qualities of the ingots were improved using FeCrN as nitrogen alloying source. The sound and compact macrostructure ingot with the maximum nitrogen above 1.0wt% could be obtained. A series of high nitrogen austenitic stainless steels have been successfully developed by this method. The mechanical and pitting corrosion resistance properties of high nitrogen austenitic stainless steels were investigated. The results show that the steels exhibit excellent mechanical and pitting corrosion resistance properties.

Jan 1, 2009

By R. Dahlke, J. A. Finch

A campaign was undertaken during the greater part of 2001 at Noranda’s Brunswick mine concentrator to investigate the role of gas (air) superficial rate (Jg) distribution to the cells in the final Zn cleaning stage (bank of seven cells), i.e. the Jg profile. Three profiles were tested: “balanced” (each cell with same Jg), “increasing” (from cell 1 to 7 proportionally), and “decreasing,” along with an “as found” profile in a parallel control bank. The experimental design allowed for two confounding factors, changes in ore type and circulating load. The increasing profile consistently gave the best metallurgy (down-the-bank grade / recovery). The reason for improved performance was the response of the first cells in the bank where the increasing profile gave the highest selectivity of sphalerite against nonsulphide gangue (NSG). The other profiles gave lower selectivity and higher zinc recovery in the first cells, which combined to reduce overall bank performance. The higher selectivity with the Increasing profile is attributed to the low gas rate in the first cells reducing water recovery and hence limiting entrainment of NSG. The increasing profile was subsequently configured for all four stages in the zinc cleaner circuit.

Jan 1, 2004

By L. D. Jaffe, R. K. Pitler

IT would clearly be advantageous to produce molten titanium, suitable for alloying and casting, directly from the relatively inexpensive tetra-chloride, without using a metallic reducing agent. Accordingly, a preliminary investigation has been made of the production of titanium by hydrogen reduction of the tetrachloride in an electric arc furnace. For the reaction: TiCl4(g) + 2H2(g) ? Ti(s) + 4HCl(g), Lockhart and his colleagues' have calculated equilibrium constants at 800" to 1500°K, based on tables of Brewer.' Extrapolating these data on the assumption that the heat of reaction varies linearly with temperature, and using Brewer's value for the heat of fusion, there is obtained for the reaction: TiCl4(g) + 2H2(g) ? Ti(l) + 4HCl(g) an equilibrium constant at 2100°K (100" above the melting point) Kp = (PHCl)4/pTlcl4 (Ph2)2 = 0.167 atm. If this is correct, starting with hydrogen saturated with TiC14 at 25°C, where the vapor pressure of the latter3 is 12.6 mm, and maintaining the overall pressure at 1 atm, 99 1/4 pct of the TiC14 would be reduced to Ti at equilibrium. An arc furnace designed for the melting of titanium and lined with water-cooled copper was used for the preliminary experimental check. Its inside diameter was 2 7/16 in. and height 6 in. The water-cooled electrode was tipped with 3/4 in. diam tung-sten. In the one run carried out, the furnace was charged with a starting batch of 293.35 g of scrap titanium,' analyzing 99.24 pct Ti. After the system was twice pumped out and flushed with argon to remove air, the arc was struck in argon and kept at 400 amp dc, electrode negative, for 1 min to melt the starting batch. Commercial tank hydrogen, dried by passing over silica gel and bubbled through technical grade TiC14 at room temperature, was then admitted to the furnace. The inlet tube directed this gas downward toward the arc and melt; the gas outlet was at the top of the furnace. The quantity of gas passed through was not measured accurately but is estimated to be about 50 liters per min at a few mm of mercury above atmospheric pressure. The arc was held in the H2-TiC14 mixture at 400 amp with little difficulty. After 10 min the arc burned through the copper lining above the melt, admitting water and ending the run. (This had happened frequently in straight melting runs under argon, and is not thought due to the atmosphere used.) In the furnace were found an ingot, several small pieces, and a little metallic powder, evidently formed when water entered the hot furnace. Some titanium also clung to the electrode. The portions were weighed and separately analyzed for titanium. Their total weight was 321.76 g of 94.82 pct overall purity. Much of the impurity consisted of tungsten broken off when the titanium was removed from the electrode. No detailed impurity analysis was made because the starting materials were impure and flooding the furnace introduced further contamination. Thus, the metal recovered contained 305.08 g Ti, as compared to 291.12 g in the starting batch. Titanium could not have been carried over from previous melting runs in the furnace, as the interior was sandblasted and the electrode ground clean before the run. Evidently at least 13.96 g of Ti were produced from TiCl4. Since this would require complete reduction of 430 liters of TiCl, gas at 12.6 mm of Hg, the efficiency of the process must have been rather high. Experimental work is continuing. No reason is seen why the process could not be scaled up to the size of a 30-ton steel-melting arc furnace. In a large unit there would be a large molten pool from which liquid metal could be withdrawn for alloying and casting. The problems of electrodes and power supply are similar to those arising in arc furnaces for melting titanium. Unreacted TiC14 and the HC1 in the exhaust gases could be recovered; unreacted H2 could be dried and re-used. The melt would be saturated with hydrogen, whose solubility in liquid titanium is unknown. Small amounts of hydrogen seem to have little effect on the properties of titanium.5 If the amount introduced was so large as to impair properties or cause casting difficulties, it could be reduced by holding the melt under an inert gas or by pumping on it, prior to casting. Acknowledgment The authors would like to thank I. Preble and M. Goldman for their valuable aid in the experimental work and the thermodynamic calculations, respectively. References 1R. J. Lockhart, J. J. Ward, M. J. Hussey, and J. W. Clegg: In unpublished report, "U. S. Air Force Project Rand. Titanium and Titanium Base Alloys." Battelle Memorial Institute (1949) 37-44. 2 L. Brewer et al: Chemistry and Metallurgy of Miscellaneous Materials—Thermodynamics. Ed. by L. L. Quill. pp 13-39, 60-192. (1950). McGraw-Hill Book Co. 3 K. Arii: Scientific Reports. Tohoku Imp. Univ. (1933) 22, 182. 4 R. S. Dean, J. R. Long, F. S. Wartman, and E. L. Anderson: Transactions AIME (1946) 166, 369-381; Metals Technology (Feb. 1946) TP 1961E. * R. I. Jaffee and I. E. Campbell: Transactions AIME (1949) 185, 646-654; Journal of Metals (Sept. 1949) TP 26813.

Jan 1, 1951