Search Documents

By C. Doyle, D. M. Hoffman, C. Díaz, J. Hrepic, W. J. Koshut, R. H. Hanewald

"Recently, lining life of the INMETCO submerged arc furnace has been dramatically extended from an average of eleven months in the 1980s to a current record of 34 months. This paper summarizes test work and operating results describing the development and establishment of slag parameters that reduce corrosion of the magnesite-chrome refractory lining.The desired slag parameters are now maintained by utilizing waste streams as slag additives. This paper also discusses the successful evaluation of a rebonded-fused-grain magnesite- chrome refractory as a lower cost, more thermo-mechanically forgiving alternative to a fused-cast magnesite-chrome refractory that had been the refractory of choice in previous years at INMETCO.IntroductionSince 1978, INMETCO, a wholly owned subsidiary of INCO Limited, has been well known for recycling by-products from the production of stainless steel, namely, EAF bag- house dust, mill scale and grinding swarf. INMETCO operates the only smelter in North America dedicated to the recovery of nickel, chromium, and iron from hazardous and nonhazardous waste. The process was developed in the research laboratories of INCO. In recent years, the process has been successfully modified to accept a wider variety of other specialty steel waste streams including: spent pickling solutions, pickling and waste water treatment filter cakes, acid scale, carbon brick, spent bag-house bags, torch cutter dust, spent dolomitic and chromium-bearing refractories, and rechargeable nickel-cadmium batteries."

Jan 1, 1997

By T. P. Lim, N. K. Davé

"A modified accumulator method was devised to measure surface radon flux from porous media. It consists of a collector hood type of accumulator with a homogenizer pump and a pressure balancer. The latter was used to compensate any pressure changes produced during sample withdrawal and hence eliminate additional radon transport from the media. Errors resulting from radon back diffusion due to decreasing concentration gradient with time across the interface were also eliminated by measuring radon concentrations in the accumulator for two different periods of time and calculating the true radon flux, J(o), from the solution of a time-dependent radon transport equation in the absence of any pressure and temperature gradients.Core samples from the observation site were also withdrawn for laboratory determination of porosity, moisture content, bulk and grain densities, Ra-226 source concentration and the fraction of Rn-222 produced in the void volume. The data were used to calculate the radon diffusion coefficient for the media.Using the present method, the true value of surface radon flux for an abandoned uranium tailings pile was obtained as J(o) = 2020.2 mBq/m2/sec (54.6 pCi/m2/sec) compared to the value J = 1406.0 mBq/m2/sec (38.0 pCi/m2/sec) obtained by the conventional method. The bulk diffusion coefficient of radon for the media was obtained as 1.2 x 10-7 m2/sec.IntroductionRadon-222 (hereafter called radon) released from uranium mill tailings is a continuous source of environment radioactivity in terms of its transport as an airborne gas and its decay products. The particulate daughter products readily attach to aerosols or dust particles and can further be transported by various other environmental pathways. Typical radon release rates from existing tailings piles vary between 740 and 14,800 mBq/m2/sec (20-400 pCilm2/sec), depending on the nature of the tailings material, moisture content, etc. The United States Nuclear Regulatory Commission (NRC) is proposing radon release rates of 74 mBq/m2/sec (2 pCi/m2/sec) above that of background."

Jan 1, 1982

By Charles J. Zablocki, Robert J. Kamilli

"Buried, possibly mineralized granite cupolas at the Silsilah tin deposit in Saudi Arabia (lat. 26°06' N, long. 42°40' E) have been successfully located using a closely spaced electric-field-ratio profiling technique. In this study, electrical fields at 27 and 270 Hz across grounded electrodes spaced 50 m apart were measured along six traverses. This method is interpretable by inspection, unlike the dipole-dipole technique, which requires mathematical modeling for interpretation.In the study area, the technique allowed the authors to identify and distinguish among unroofed granite cupolas, cupolas with their aplite-pegmatite apical contact zones intact, strong (quartz-rich) and weak (relatively quartz-poor) greisens, dikes, faults, and pervasively argillized rocks.This method should be useful in the valuation of other shallow mineral prospects in similar settings in other parts of the world, especially deposits for which the location of vertical contrasts is important, such as veins or mineralized zones associated with the apexes of small igneous cupolas."

Jan 1, 1993

By Eirik Sundvor, Hany Doss, Peter Vogt, Kathleen Crane

"Abstract - In 1989-1990, the SeaMARC II side-looking sonar (11 to 12 kHz) and swath bathymetric system imaged more than 80 000 km2 of the seafloor in the Norwegian-Greenland Sea and southern Arctic Ocean. One of our main goals was to investigate the highly oblique (115°) junction between the Mohns and Knipovich ridges, and to determine, using new morphotectonic data, whether or not this junction is stable or in a state of readjustment. These findings are presented in this paper.Along the northern Mohns Ridge, we observed a highly segmented rift valley undisturbed by transform faults. In contrast to the elevated Reykjanes Ridge, the Mohns Ridge is dominated by deep en echelon fault-bounded troughs, which are bordered by high rift valley volcanoes. These segment boundaries are interpreted to be sites of recent to present volcanic activity, based on their high acoustic reflectivity. The northwestern flank of the Mohns Ridge is marked by shallow quasi-linear ridges parallel to the present orientation of the Mohns Ridge. The orientation and geographic position of these ridges suggest that they were formed at the Mohns Ridge. Only in the northernmost cases are some of these off-axial ridges non-linear. The degree of the non-linearities increases slightly with proximity to the Mohns-Knipovich bend. In addition, because no transform fault was detected between the Mohns and Knipovich ridges, we conclude that the Knipovich Ridge must be obliquely spreading to the NW-SE."

Jan 1, 1999

By C. K. Baxter, A. Odeshi, M. Mohammadi, E. D. Cyr

"Metal additive manufacturing has been a revolutionary step in designing new complex shapes with faster and cleaner building capacity in comparison to subtractive manufacturing processes. The ability to print lightweight alloys such as aluminum has made metal 3D printing attractive to different industrial sector from aerospace to energy and bioengineering. Of particular interest in this study, are the mechanical and microstructural properties of AlSi10Mg-200°C alloys. In this project, samples were 3D-printed in horizontal direction using an EOS M290 machine by Direct Metal Laser Sintering (DMLS) technique. Uniaxial tensile tests and high strain-rate impact e tests were carried out to investigate the mechanical properties and to study the repeatability of the process. A constitutive model is developed to predict the results of the mechanical tests at varying strain rates to better understand the high strain-rate behaviour of these alloys. Using optical microscopy, features such as scan tracks and melt pools within the AlSi10Mg-200°C samples were observed, and the microstructure was observed to be homogenous. The previously listed tests were compared to the results obtained for similar cast alloy counterpart, i.e. A360.0.INTRODUCTION A relatively new way of manufacturing parts is on the increase as rapid prototyping becomes increasingly popular throughout the world. Additive manufacturing is a process which, given a 3D model, uses metallic material in powder form to build a part layer by layer until completion (Gu, Meiners, Wissenbach, & Poprawe, 2013). This way of rapid prototyping is very attractive in comparison to other possibilities because of its ability to accommodate complex design and geometries. Conventional manufacturing methods have limitations when it comes to complex geometries, but additive manufacturing allows for increased design flexibilities while maintaining a light and stable final product (Murr et al., 2012). It also allows for a reasonable cost, especially for small batch sizes that would otherwise be extremely expensive."

Jan 1, 2018

By Herbert D. Drechsler, James B. Stephenson

"AbstractThe paper identifies the major components of price change which affect the determination of the internal rate of return of a mining project. It illustrates several price indexes which may be used to measure general price increases, identifying techniques which might incorporate them into mine valuation procedures. The ""hurdle rate"" for new investment is discussed.IntroductionDECISIONS regarding the value of an orebody and whether or when to mine an orebody are based on an estimate of the cash flows which would be associated with its exploitation. These pro-forma cash flows represent cash in or out of a possible mineable deposit and include exploration and construction costs, sales revenues, operating expenses, taxes, debt borrowings, interest and debt repayment. It has long been recognized that when combining the cash flows occurring over the life of the mine, dollars received or paid out early have greater value than those received or paid out later. This time value of money has been incorporated into mine evaluation calculations through the use of present-value or discounting techniques. Another factor affecting mine valuation, inflation, is not accounted for by valuation techniques now in use. In recent years, price changes have become more volatile. In the past, price change could easily be ignored when evaluating potential mines. Today, the combination of larger investments and greater price volatility leads to great concern about the effects of price change on mine valuation."

Jan 1, 1977

ABSTRACT Understanding process histories and worker exposures in nickel production operations is critical in determining the role nickel compounds play in respiratory cancer incidence. As part of a series on Vale Canada’s (previously known as Inco) nickel processing operations, this paper includes descriptions of Ni electrolytic operations at refineries in Port Colborne, Ontario (1922–1984), and Thompson, Manitoba (1961–present). Industrial hygiene measurements, by area and personal sampling, of workplace aerosols at these operations are reported.

Jan 1, 2014

By Y. Guglielmi, C. Gout, P. Henry, P. Dick, C. Nussbaum

For a decade, observations and experiments conducted in underground environments worldwide (mines, underground research laboratories - URL) made it possible to bridge the scale gap between laboratory scale and large scale faults. Indeed, underground rock laboratories allow developing decameter scale relatively well constrained experiments at the smallest distance from the hypocenters. URL’s experiments allowed developing new monitoring technologies coupling pressure-strain-seismic measurements and reproducing complete earthquakes sequences analogous to natural ones. They also allowed exploring past movements of faults and sometimes also paleofluids in three dimensions and characterizing different fault architectures and hydromechanical properties in unaltered conditions. This is particularly important in the study of faults affecting clay formations which could not be observed at the Earth’s surface because of the high alteration. Here we show how examples from the Mont Terri (Switzerland) and Tournemire (France) The URLs may help in (i) defining the concepts of fault architecture in shales, (ii) estimating fault hydromechanical properties from field decameter scale experiments. Some key results are that faults which are critically stressed or at the rupture limit can reactivate for stresses variations within the range of those produced by shallow underground excavations, and that slight micrometer-to-millimeter scale reactivations may lead to large factor-of-10-to-1000 permeability increases. Such promising results show that URL’s experiments may bring new insights on some key processes that control the relations between fault stability and permeability evolutions in very permeable layers.

Jan 1, 2015

By A. B. C. Stothart

"IntroductionIn the past four years Ontario Mines have made great progress in the development of reliable mine conveyance safety dogs. History shows that, prior to 1945, a rope failure on a descending conveyance was usually followed by unsatisfactory safety dog operation.It is proposed in this paper to discuss briefly;The history of this development. Progressive .development of one type (E. Long).Free fall testing of safety dogs. Anti-trailing rope device.HistoryOn February 2nd., 1945, a hoisting rope broke at the Paymaster mine. The safety dogs proved ineffective and permitted the cage to crash through to the bottom of the shaft, with disastrous results.Following this accident, the Ontario Government appointed a Technical Committee to investigate The Safety of Hoisting Equipment and Hoisting Practice in Ontario Mines. This Committee consisted of Dr. C. R. Young (Chairman) and Professors T. R. Loudon, E. A. Allcut, V. G. Smith, and Lloyd M. Pidgeon. The Ontario Mining Association (at the request of the Ontario Government) also appointed a Committee consisting of Messrs. R. E. Dye, R. D. Parker, and R. L. Healy."

Jan 1, 1950

By P. J. Le Roux

"Mining worldwide and definitely in South Africa, is constantly under pressure to reduce its cost structure so as to sustain profitability. In underground gold mines where an open stope mining method is employed, dilution often has a significant effect on the viability of sustaining profits. Target Mine practices Open Stope mining Method and it was found that in some open stopes dilution was in excess of 10%, which has a significant impact on the sustainability of the mine.Dilution in excess of 10% can result in the reduction of the recovered grade from 5,5 to 4,5 grams per ton (g/t). The reduction of 1 g/t in recovered grade results in a potential loss of about ZAR21 Million per month bases on a gold price of ZAR240 000 per kilogram. Based on Life of Mine projections the potential loss of income could be as much as ZAR3,3 Billion. A reduction in dilution would have the opposite effect. There are a number of factors which affect dilution of which falls of ground in open stopes is a major contributor. The falls of ground can be attributed to a number of factors such as beam failure as a result of a larger than normal expected roof area (hydraulic radius too large), poor ground conditions, and poor blasting. The cost of damage to or loss of trackless equipment as a direct result of the falls of ground in open stopes, is very significant. The review of financial figures has indicated that this could be as high as ZAR491 million. This combined with the added cost of transport, hoisting, secondary blasting, milling and plant treatment costs of ZAR293 results in an estimated opportunity loss of ZAR784 million.Currently there is a significant amount of data available in the mining industry which could be effectively used to develop suitable back analysis techniques but to date this has not been used effectively. If dilution can have such an impact on current and future mining ventures then the optimization of back analyses for the prediction of dilution in open stoping could assist significantly in the reduction of dilution in massive open stopes. Rockmass classifications, geotechnical information, blast techniques, blast design, the stress strain environment and hydraulic radius all have an effect on, or play a role in the evaluation of dilution. Each of these factors will be taken into consideration to ultimately determine a measurable or calculated percentage of dilution in massive open stopes.The amount of overbreak in an open stope can be determined by subtracting the planned stope volume in m3 from the actual measured final stope volume in m3, which is obtained from the CMS (cavity monitoring system). This is in turn divided by the planned stope volume in m3 to determine the percentage overbreak. The CMS wireframe is imported into the geological model and its grade re-evaluated. From this the actual percentage dilution for this open stope can be determined. The dilution obtained can result in a major reduction of recovered grade for the open stope, but can also increase the recovered grade. The latter is normally the exception rather than the rule.When analysing data from Target mine the following was achieved:• Using 11 years of data a method of measuring and predicting the percentage dilution in open stoping was developed. This took into account rock mass quality, stress-strain state, and the hangingwall hydraulic radius (size of stope hangingwall exposed).• Implementation of this prediction method resulted in a reduction in falls of ground in open stopes. The benefit of this was a reduction in the damage to mechanised equipment resulting from less falls of ground which had a positive effect on the profit margins of the mine. As a direct result the recovered grade from the open stopes increased due to the reduction in the amount of dilution.• A design criterion, Dilution Stress-Strain Index (DSSI), was developed which allows the user to calculate, with certainty, the stability of the open stope and determine if m"

Jan 1, 2015

By G. B. Harris, C. W. White

A study to investigate the recovery of base metals, notably copper, nickel, cobalt, zinc and manganese from iron-free concentrated chloride solutions using ion exchange was undertaken. In such solutions, apart from nickel and, to a lesser extent manganese, these metals are present predominantly as complex anions, and therefore conventional separation techniques based on cationic resins are not applicable. The results demonstrated that zinc, copper and cobalt can all be separately recovered leaving nickel and manganese behind. Specifically, zinc is a major nuisance element in nickel and cobalt processing, and the testwork showed that it can be preferentially removed prior to copper and cobalt. Selective resins for both copper and cobalt were identified, and it was demonstrated that the strength of the chloride background matrix has a significant effect on the efficiency of separation. An additional major advantage in such systems is that stripping can mostly be effected with water. Finally, some observations are made on the hurdles that must be overcome when attempting to introduce and commercialise new technologies.

Jan 1, 2015

By F Kelly

"A central composite experimental design was used to determine empirical relationships between the operating variables of a ball mill and the characteristics of the ground products. The independent variables selected for the study were the work index and size distribution of the feed material, the feed rate and the pulp density. The measured responses included product size distribution, product surface area, viscosity of pulp discharge, weight of dry solids in mill, mill time constant and emitted sonic vibrations. The ball mill was operated continuously in open circuit during each experimental run.Multivariable regression techniques were used to derive linear empirical models from the experimental results. Of the four factors studied, the grindability of the feed material significantly affected every measured response. The size distribution of the ground product was affected significantly by the feed rate and the size distribution of the feed, but not by the pulp• density. Feed throughput capacity and pulp density were significant to the viscosity of the pulp discharge, the weight of solids in the mill and the mill time constant. Significant first-and second-order models are given for the above relationships. The relation-ships between the characteristics of the sonic vibrations of the operating mill and the various dependent variables have not as yet been established."

Jan 1, 1970

By David Scott, Denis V. Lachance, Daniel Marcotie

"LocationThe Heath Steele mine is located 50 km northwest of Miramichi, New Brunswick.Brief DescriptionHeath Steele is a 2700 t/d, copper-lead-zinc operation which employs conventional crushing, grinding, and differential flotation.HistoryThe Heath Steele mine is located in northeastern New Brunswick approximately 50 km northwest of the city of Miramichi and 65 km southwest of Bathurst. The massive sulphides containing ore values in zinc, lead, copper and silver were discovered in 1953 as a result of a joint venture between the American Metal Company and the International Nickel Company of Canada. The discovery was named after Mr. Heath Steele, manager of exploration for Amco at the time of the discovery.Mill production began in February 1957; however, the mine was forced to shutdown in 1958 due to depressed metal prices and metallurgical problems. It was re-opened in 1962 and milling operations resumed at 725 t/d until 1968 when the milling capacity was boosted to 1450 t/d and then to 2700 t/d in 1969.OreThe main ore zones are massive sulphides with very fine mineral grain. The chalcopyrite, galena and marmatitesphalerite occurrences are finely disseminated.Liberation is accomplished by grinding to 68% to 72% <400 mesh. Average mill feed assays are 0.8% to 1.2% copper, 1.8% to 2.5% lead, and 5.5% to 6.5% zinc, with 60 to 75 g/t silver."

Jan 1, 2000

By H. D. B. Wilson

"IntroductionRECENT ADVANCES in geo-chemistry and in •our knowledge •of crystal structure provide a key which enables us to create a systematic picture of the genesis of many ore deposits. This genetic scheme may be used with confidence in assessing certain types of orebodies and may be of some aid in actual exploration, since it affords a rational explanation of the dependence of the composition •of the ore minerals on that of the rock type in which they occur with which they -are •associated. For purposes of the present paper, .the elements chosen are •chromium, vanadium, iron, and titanium. The oxides of all these metals occur in deposits of similar or related type, and it is proposed to show that, for each, the composition of the •ore mineral is related to •the composition of the enclosing :rock, and that one of the principal factors controlling the composition of the ore is the ionic size of its constituent elements.Geochemical PrinciplesThe crystals of the rock-forming silicates and of the magmatic oxide minerals are essentially ionic types. This means that they are composed •of oppositely charged ions such as Mg²+ and 0²-, which are held together by electrostatic attraction. The ions of each element are spherical, with characteristic size known as the ionic radius. Crystals are built up by the packing of these spheres in the most stable arrangement compatible with the size and the size and valence of the ions• involved.Many of the principles governing the formation of ionic crystals are •simple ,and logical."

Jan 1, 1953

By G. E. Wittur

"The twenty-one members of the Organization for Economic Cooperation and Development, comprising eighteen western European nations plus Canada, the United States and Japan. account for two-thirds of world steel production. The Special Committee for Iron and Steel was formed by the OECD to study trends and problems in the steel industries of member countries. The paper begins with a summary on the Organization, including an historical sketch, its structure, member-ship and objectives. The steel industry of the entire OECD is briefly described before proceeding to a more detailed discussion of various aspects of the steel industries of European OECD members, emphasizing practices and conditions that differ from those in North America. Benefits to Canada of representation at meetings of the OECD Special Committee for Iron and Steel are described."

Jan 1, 1967

By C. W. Langenberg, D. E. Macdonald, R. S. Strobl

"The successful marketing of Alberta coals will increasingly depend, to a large degree, on an extensive coal quality information base. This information, at present, is generally based on a deliverable, cleaned coaland is confined to existing mine sites. Coal quality information is not as well understood for raw in situ coal outside of mine sites and even less is known about the geological factors that control it.If new uses and markets for Alberta coal are to be developed, such as gasification, coprocessing, and exploiting high rank coals for combustion, it will be imperative that the coal quality parameters important for these utilizations be known. These parameters need to be known for existing deliverable coals and, just as important, for as yet unexploited in situ coal. Factors such as the Cl, No, K, Co, P and trace element contents should be addressed to meet the requirements of these new utilizations.The geological characterization and explanation of what factors control these coal quality parameters is part of research being done at the Alberta Geological Survey. In-seam vertical profiling has been undertaken within the plains and foothills/mountains coals to determine exactly where coal quality variations occur and why. In the Ardley coal zone at the Highvale Mine it has been found, for example, that the ash content of the mined coal is made up of discrete partings and of microsized particles associated with inertinite macerals. The in-seam characterization of coal seams will lead to a better understanding of in situ coal quality variations, leading to new uses and markets for Alberta coal."

Jan 1, 1993

By R. K. Carnochan

MANY improvements have been made in recent years in the milling of industrial minerals and it is the purpose of this paper to review some of the more important innovations that are now being used in the preparation of these products. Some of the observations which will be made have been prompted by what was seen on a recent trip to industrial mineral milling plants in the north and central sections of the eastern United States. The flotation plants visited were those of the Valley Forge Cement Company at West Conshohocken, Pa., the International Agricultural Corporation at Mount Pleasant, Tenn., and the Aluminum Ore Company at Rosiclare, Ill. Besides these, sand plants in Pennsylvania, West Virginia, and Ohio, coal-cleaning plants in Pennsylvania, pyrophyllite, mica, feldspar, clay, and kyanite plants in North Carolina, and a zinc mill and the Tennessee Valley Authority ceramic laboratory in Tennessee were visited. The milling of industrial minerals can be considered under three broad headings, namely, crushing, grinding, and purification. Usually there is both a separation and grinding problem in treating an industrial mineral, but the problems are distinct, although sometimes a separation is made while grinding, as in the case of a Raymond throw-out mill, which grinds and at the same time throws out most of any hard material present in its feed.

Jan 1, 1939

By Reis. E., P. Pino Véliz

"Along the process of defining the fundamentals of a ferronickel project, several questions require proper answers from the very beginning such as the size and characteristics of the laterite deposit (oxide or silicate), the mineralogical variability, and the selection of the most valuable process route. In general, these elements will be defined along the progress of the exploration and feasibility studies, and definitive engineering. Also, from the very preliminary estimates, three questions are very common: what should be the plant capacity (annual tonnes of contained nickel), what will be the probable capital required (CAPEX), and what will be the estimated timeline (schedule) between the initial stage, say the feasibility study and the plant start-up and ramp-up to designed capacity. The accuracy of the capital cost estimates also progresses along the different stages of development from ±50% at the Preliminary Economic Assessment (PEA) stage down to ±5% for the Detailed Engineering and Construction stage. These accuracies of estimates have been, in some ways, accepted by the mining community and investors. The industry accepted benchmark for ferronickel projects in terms of CAPEX are then in the order of US$50,000 per tonne of nickel contained in the FeNi product. In this paper, the authors analyze in some details three ferronickel projects for lateritic ores that have selected the Rotary Kiln - Electric Furnace (RKEF) as the suitable technological route and have started production in the period 2006–2016. Some key aspects of the ferronickel projects are examined: 1) size and type of deposit; 2) flowsheet selection; 3) evolution of capital cost as a function of capacity from the feasibility stage to construction; 4) comparison between planned production scale and actual ramp up; 5) estimated capital expenditure (CAPEX) compared with actual figures; 6) nickel price evolution and timing of the approval for the project to proceed to construction and production start-up. The objective of this paper is to identify key elements that should be addressed in future projects and that are proposed as benchmark figures for similar projects."

Jan 1, 2017

By E. Ghali

For engineering applications, magnesium is usually alloyed with one or more elements which include aluminium, manganese, rare earth metals, lithium, zinc, and zirconium. Under normal environmental conditions, the corrosion resistance of magnesium alloys is comparable or better than that of mild steel. The oxide film thickness on pure magnesium after exposure for only ?10 seconds at ambient temperature is 2.2 ± 0.3 nm. Exposure to humid air or to water leads to the formation of a thick hydrated amorphous film that has an oxidation rate less than 0.01 pm/ y. E-pH diagram and high temperature corrosion are discussed. The pH values between 8.5 and 11.5 correspond to a relatively protective oxide or hydroxide film that is more stable at higher pH. Structural applications of magnesium alloys are limited by creep strength rather than by oxidation. Active and active-passive behaviours of magnesium alloys are discussed. The pilling Bedworth Ratio values of oxide and hydroxide are 0.79 and 1.77 respectively and so the formation of Mg(OH)2 inside the oxide film could lead to the development of cracks. Microbiologically influenced corrosion and rational biodegradation of magnesium alloys are discussed because of their interest for medical care.

Jan 1, 2006

By N. Melnyk, H W. Habgood

The conventional low temperature process for extracting' helium from natural gas involves liquefying the whole gas stream in order to remove the relatively small fraction of helium. A process based on the selective diffusion of helium through a semi-permeable barrier is, in principle, more economical and flexible, particularly for the treatment of small re-serves and low concentrations. The well-known permeability of glass to helium offers the possibility of such a diffusion process and McAfee of the Bell telephone Laboratories showed in 1958 that practical separations could be achieved by passing natural gas over bundles of fine capillary glass tubing. Developmental studies have been carried out involving larger units and more extended field tests than those of McAfee. Fine capillary tubing of Pyrex glass 0.0035 in. diameter was prepared by drawing ordinary 6 mm. laboratory tubing in a furnace. Extraction units were pre-pared containing about 10,000 such tubes, 4 ft. long and cemented inside

Jan 1, 1961