Search Documents

This Bureau of Mines report and its companion report (Information Circular 9142) have been prepared to assist in the preparation of prefeasibility type estimates for capital and operating costs of beneficiation of various types of mineral occurrences using current technology. The handbook provides a convenient costing procedure based on the summation of the costs for the unit processes required in any particular mining or mineral processing operation. The costing handbook consists of a series of costing sections, each corresponding to a specific mineral processing unit process. Contained within each section is the methodology to estimate either the capital or operating cost for that unit process. The unit process sections may be used to generate, in January 1984 dollars, costs through the use of either costing curves or formulae representing the prevailing technology. The mineral processing handbook includes individual cost estimation sections for unit operations associated with comminution, beneficiation, solid-liquid separation, hydrometallurgy, and special applications as well as infrastructure and plant general and administrative costs. When using this system for estimating cost data for a mineral processing facility or for checking or verifying processing costs from an existing facility, a minimum amount of background information is necessary.

Jan 1, 1987

By J. A. Kost

The contract objective was to evaluate the effectiveness of available dust-control technology for double-drum shearers. Eight underground surveys were conducted to evaluate drum design, drum spray nozzle locations, ranging-arm sprays, water quantity, wetting agents, and cutting techniques. The final report is made up of eight dust-control technique reports, each discussing, details of the individual mine surveys.

Jan 1, 1985

By William W. Odell

Certain gas-making processes are in use in Europe, particularly in Germany, which differ radically from those in common use in America. A number of them are of particular interest because they are different. The questions naturally arise: Are any of them applicable to the making of gas in this country and, if so, with what particular type of solid fuel? It is a fact that some of them were developed with consideration of a particular set of conditions and for the utilization of solid fuels available in Germany. The latter fuels, in certain characteristics, are unlike the high-rank fuels commonly used for making gas in the United States; therefore, in attempting to evaluate the various foreign- developed processes with regard to usefulness here, one must give attention to the particular conditions that will prevail in any specified case and also the characteristics and requirements of the process and generator under considera- tion. Three of the processes that were developed and put into commercial operation in Germany, and which have some characteristics in common, are con- sidered in particular in this publication; they are: 1. Lurgi pressure gasification process. 2. Winkler process. 3. The Leuna slagging-type producer-gas process. Each of these processes requires steam and oxygen as gas-maiding fluids. For a number of years, oxygen, has been used in a relatively pure state,. along with steam, in making gas in Germany, and many of the difficulties initi-. ally encountered have been overcome. Although it appears that oxygen recovered from air was first used in this manner as a method of utilizing a byproduct in a plant that used the nitrogen complement in the manufacture of ammonia; more recently, the oxygen was made particularly for the purpose. The fuels - brown coal or lignite used in processes 1 and 2 were not suitable for use in the older and commonly used intermittent-type water-gas generators, and it was necessary, therefore, to devise a different process for gasifying them. A gas of higher heating value than producer gas, was desired, and the use of oxygen (not air) along with steam made possible the production of the desired gas without nitrogen dilution, which characterizes producer gas. Operating data presented herein are based upon observations by, members of the Technical Oil Mission at plants in Germany, on records, and on information vocally given to members of the lässion, including the author, while visiting those plants, and they are believed to be reasonably correct; however, certain hasic observations, comparisons, and conclusions are. made by the author.

Nov 1, 1947

By Mark E. Emerson

1967 will be remembered as the year the world realized it was running out of silver. In May of this year a new era in tin the history of the white metal was begun as the power of the U.S. Treasury to regulate world .silver markets was staggered by one of the most intense speculative flurries in recent times. In the past few months the price of silver has risen from $1, 29 to over $1.80 per troy ounce and with it have spiraled the prices of a host of silver-bearing consumer items. Legislation has, been enacted to conserve our coinage and meager silver stocks and today large quantities of non-existent silver are trading hands at a fast pace on the commodity exchanges throughout the world. Now that the price of silver has been set loose, forces are in motion to dull the world's appetite for silver. At the same time there is strong incentive to provide greater quantities of silver to satisfy the world's hunger' for the metal.

Jan 1, 1967

By Daniel G. DiCanio

This report describes a computer-based model for estimating the tangible costs of injuries and deaths from work-related accidents in underground coal mines. The study undertook to identify and quantify the elements of accident costs, including field visits to coal mines, insurance carriers, and public agencies. The mathematical models developed for each tangible element of accident cost are presented in Volume I of the report. A listing of the computer programs which comprise the model and instructions for using the programs are presented in Volume II. Statistical distributions of mining family characteristics, data on 95 miners killed in underground accidents, and other basic data used to develop the model are presented in Volume III. Application of the model to a data base of 9,286 accidents which occured in underground bituminous coal mines during 1974 yielded a total of $56,900,000 in tangible costs to industry, mining families, and public agencies. Loss of income to the families of miners injured or killed in these accidents accounted for 47 percent of the total.

Jan 1, 1976

By Dirk van Zyl, S. Chander

Minerals have been critical to human society from the earliest and their use has been intertwined with the development of civilization. In the Western World, the mining, processing and extractive metallurgy associated with ores was first described in detail by Agricola in the 16th century. He considered these collective activities to be the most necessary and the most profitable of professions, and wrote, "without doubt, none of the arts is older than agriculture, but that of the metals is not less ancient; in fact they are at least equal and coeval, for no mortal man ever tilled a field without implements. In truth, in all works of agriculture, as in the other arts, implements are used which are made from metals, or which could not be made without the use of metals; for this reason the metals are of the greatest necessity to man." Ours is still a materials-dependent society that relies heavily on minerals for raw materials. Today our dependence on minerals is exceptionally great. To meet the need for fuel and non-fuel minerals in a representative year, in the United States alone, about one billion tons of coal and 2.7 billion tons of ore are mined. These mineral products are essential for fertilizers, for construction materials, metals and alloys, for manufacturing household and industrial items, for transportation, and for the generation of electric power. It is instructive to consider the diversity of uses found in our society for a single mineral product - silver, for example. Experts believe that in the past 5000 years we have mined about a million tons, or 31.9 billion troy ounces of silver. In 1988, 133 million troy ounces of silver were used in the Unites States. Most of the silver used was for techno- logical, medicinal and industrial purposes, Table 1. X-ray films, for example, carry comparatively large amounts of silver to lessen patient's radiation exposure. Silver does not prevent tooth decay, but each year about two million troy ounces of silver is used for dental work in the United States. In small quantities it is used in medicine. No metal - not even copper conducts heat and electricity so efficiently as silver. Silver wires lace solar cells, and silver oxide batteries power hearing aids, calculators, submarines and satellites. Hardened with tungsten or molybdenum, miniature discs of silver pass current from wire to wire in cars, telephones and computers. A dish- washer timer alone may have 50 such electrical contacts, which open and close without excessive heat or friction. Silver is an important component of solar reflectors, since no other material reflects light so well or uniformly. In our personal lives, silver remains in high esteem to adorn our tables, provide precious objects and jewelry. In many parts of the world, silver in the form of coins, jewelry and bullion is used for financial independence and security against the vagaries of paper currency. In India, silver is used in its purest form as thin foils to decorate sweets and other edibles. It is obvious that this mineral product, silver, is of great value and utility in our society. Minerals and Environment The benefits of minerals do not come without problems. Much of the silver, for example, lies locked up in copper, lead-zinc or precious metal ores; to extract it creates vast quantities of waste material. On the average, in order to recover one ounce of silver about a quarter ton of ore must be mined. To meet the world's annual demand for silver, approximately 100 million tons of waste is generated every Year. The problem with silver is repeated with all minerals. Conversion of mineral ores into usable materials necessitates the generation of large quantities of waste products that create tremendous disposal problems. In addition to waste material discarded during mining, wastes may be generated during processing as gases, liquids or solids that must be subsequently converted into forms more acceptable for disposal. Associated with the handling, treatment and management of waste products are a number of environmental problems. Environmental problems are not new to the minerals industry. In his Geographia, Strabo (63 B.C. - 20 A.D.), gives an account of the conditions of labor under which men

Jan 1, 1992

By R. Adsero

"A high altitude review of the state of the mining industry provided by a study of multifactor productivity shows a continuation of lower output year on year for the capital and labor inputs applied. The story is sadly similar for both U.S. mining and the Australian mining sector. In fact, according to PwC analysts, the world’s mining equipment productivity has declined 20 percent during the past seven years (Lumley, 2014). Although the reasons for the continuing decline in mining productivity are varied, the need for improvement is clear.The world’s mining industry fleet of more than 400 large draglines moves billions of tons of overburden and ore (phosphate, limestone, etc.) annually. This discussion will review a few of the tools available for improved productivity in dragline operations.Benchmarking equipment performance exposes the possibilitiesIn the September 2013 Mining Engineering cover story Graham Lumley and R. Adsero suggested that analyzing best practice equipment performance from an industrywide database for specific equipment performance improvement goals would be beneficial (Adsero, 2013). Benchmarking equipment performance by normalizing differences and then comparing specific measures of performance will provide a road map to improvement for those following the leaders. For those already performing at best practice levels, a benchmark confirms and bolsters what has already been achieved. As described in the SME February 2014 paper by this author, best practice is the average of the top 10 percent in annual output. Filling performance gaps identified by a comparison to best practice by investigating the details of the why and taking action to change, makes all the difference. This is true for all major mining equipment. Given the large gap between average performance and best practice performance most mining operations can benefit. Best practice performance provides a view of what is possible to the 90 percent majority of mining operations that are performing below the top 10 percent.For dragline mining operations, identifying specific characteristics and practices of leading dragline performance will provide a reference for comparison."

Nov 1, 2016

By H. Copur

Experimental studies for quantification of force and brittleness indices relevant to rock cutting performance are presented. A simple interpretation method is developed, which quantifies the force-penetration response obtained from indentation tests. Basic parameters of the quantification are force increment and decrementrates, force increment and decrement periods, and number of increment and decrement data points. Blocks of eleven different rock samples were collected from different operating mines in Turkey and subjected to indentation tests to calculate the indices. The samples included harzburgite, serpentinite, trona, limestone, claystone, two different sandstone samples, siltstone and three different chromite ore samples. The concept was validated by a set of full-scale linear cutting tests using a conical cutter for measuring the cutting performance in relieved and unrelieved conditions, i.e. cutter forces, specific energy, size distribution of rock cuttings. The rock cuttings were sieved to determine coarseness index, indicates coarse chip generation characteristics, and size distribution. The force index is moderately correlated with the cutting performance and mechanical properties of the rocks, and poorly correlated with the coarseness index and size distribution of cuttings. A general and moderate trend is observed between the brittleness index and the coarseness index, cutting performance for unrelieved cutting conditions and mechanical properties of the rocks. However, additional data is required to fill the data gaps and to potentially establish relationship between the brittleness index and all other parameters.

Jan 1, 2003

By John A. Mathews

It is a great honor to be asked by the Board of Directors of this Institute to deliver the Henry Marion Howe lecture. The invitation carries with it a great responsibility, which I accept with considerable hesitation and with a feeling of unworthiness to perform the duties expected of me in this hour devoted to the memory of the life, character, and work of the late Professor Howe. Many of you recall the scholarly address delivered here a year ago by Prof. Albert Sauveur, and the beautiful tribute he paid to his friend and colleague of many years in the great work of changing the art of iron making into the science of iron. My association with Professor Howe was of brief duration—only one year—and the relation was that of pupil to teacher. My feeling toward him in later years was always that of a humble disciple at the feet of a master. He attained his third academic degree in the year that I was born. I recall the pride I felt when, after entering the steel industry, I received from him, from time to time, letters asking for information or, perchance, asking my opinion, and he was always most punctilious in acknowledging the source of such bits of information in his writings. It was also my good fortune to sit at the feet of another great teacher— Sir William Roberts-Austen, at the Royal School of Mines. I mention him because his name was used by Osmond in coining the word "aus-tenite," which is the subject of this address, and Osmond's suggestion was heartily approved by Howe. These three international metallurgists were the best of friends and delighted to honor the achievements of one another. There are other reasons for mentioning Roberts-Austen at this time. It was an introduction from Howe that opened to me the door of his laboratory as well as of his heart. It was while at Roberts-Austen's laboratory that I worked side by side with William Campbell

Jan 1, 1925

By W. G. Freeman

It is unlikely, as In the past, improvement in truck efficiency and capacity can fully offset the effects of inflation, fuel costs/shortages and ever increasing haul distances. There-fore, investigations of haulage systems less sensitive to these problems are warranted. Thus, this U.S.B.M. contract research study on alternative continuous haulage systems was prepared. This study identifies the physical, economic and technological parameters in the application of continuous haulage systems with special attention to under pit slope conveyors and skip hoisting systems. The evaluation data was gathered from extensive literature searches. European/North American mine visits, and industry comments. The study concludes that continuous underground haulage is a viable method within the proper geometric configuration.

Jan 1, 1978

By Robert E. Hershey

1. This investigation shows that the most important beds in Claiborne County, from the standpoint of present production and reserves, are the Coal Creek and Jellico. Together, they contain 69 percent of the known recoverable reserves in the county. Reserves in five other beds are tabulated in this report; in order of quantity of reserves, they are: Blue Gem, Lower Pioneer, Pewee, Big Mary, and Rex. 2. Known measured and indicated reserves in all beds, based on a minimum thickness of 14 inches and on 1,800 short tons per acre-foot of coal in place, are estimated to be 60 million tons, as of January 1, 1953. Of this total, 55 million tons is in beds 28 inches and more thick. Areas in each bed were omitted from the estimate because available data relative to the bed characteristics are too meager to permit making an estimate that conforms with the definitions of measured and indicated coal adopted for this study. Should future drilling or development prove reserves in these areas, such reserves should be added to the total estimated reserves.

Jan 1, 1956

By Nathan P. Schmidt

Climate and hydrology data are required during all phases of mining projects, including impact assessment, mine planning and design, operational water management and mine closure and reclamation design. In the Canadian north, data sources are spatially sparse and often short-term in nature, and design parameters are often estimated based on data collected at sites far distant. Confidence in such data can be increased by verification or modification using data collected during properly implemented baseline and operational monitoring programs. A data set that accurately reflects conditions at the project is more readily accepted by regulatory agencies during the permitting process, and allows development of more effective and economic water management, mitigation measures and compensation. Data requirements and collection methods for developing an adequate data set for mine development are reviewed. These are illustrated using examples from oil sands, diamond and metal mines in the Canadian north.

May 1, 2004

By Richard B. Stein

Section 25 has been assembled to give the geological, mining, or metallurgical engineer an overview of some of the important downstream operations that typically follow mining activities. This subject is obviously too broad and complex to be covered in detail within this Handbook. However, the engineer will be interested to learn about the basic fundamentals, and especially how they might affect the economics of the total project from prospecting and mining through to metals recovery and marketing. In this way, it is hoped that the engineer will have a better overall appreciation for the technical and commercial prospects of a mining and metal-producing project.

Jan 1, 2008

By C. Dirscherl

"Declining ore grade, leading to increased material transport and plant throughput requirements, is one of the major challenges in the current mining environment. This is accompanied by rising energy and labor costs, decreasing plant productivity and the requirement for highest plant availability. Siemens’ answer is the latest gearless drive technology for grinding and conveying. Gearless drive solutions are surpassing the mechanical limits of conventional drive systems serving to utilize the principle of economies of scale. By elimination of various components of the drive train the maintenance activities and spare parts inventories are reduced. The advantages are culminating in the incomparable availability of the Siemens’ gearless drive technology of 99.46 to 99.71% - as confirmed by real operational data from the Antapaccay mine in Peru. It was the first plant worldwide to use gearless drives for both applications. KEYWORDS: Gearless Mill Drives, Gearless Conveyor Drives, Availability, SAG Mill, Ball Mill, Overland Conveyor, Antapaccay INTRODUCTION – THE GEARLESS PRINCIPLE The principle idea behind a gearless drive system is to reduce the number of components and diversity of parts. Especially those components that are at the edge of current manufacturing or power limits are eliminated. The used equipment provides the possibility to scale up and increase the power transmitted to the driven machine. This finally results in a possible increase of throughput respectively transported or processed material. In order to apply this principle of simplification we will take a look at the components of a conventional drive system consisting of: - Driven equipment (mill, conveyor, slurry pump…) - Components altering the rated speed (gearboxes, girth gear / pinion) - The electric motor - Starting devices for the motor (VSD, CCV, LRS, clutch…), which are necessary because a direct across-the-line start of high power motors with load would result in a significant drop in the line supply voltage - Alternatively a variable speed drive (VSD) or cycloconverter (CCV) to utilize the possibility of speed variation during operation - The electrical supply system (cables, switchgear, transformers, protection equipment) Intermediate gearboxes are not required when a motor is used with a rated speed that directly matches the speed requirement of the driven equipment. Furthermore the VSD can be used to smoothly start the motor and to vary speed during operation, which means that it satisfies both requirements."

Jan 1, 2015

By Keith Heasley

This paper is a retrospective on the development, application, and enhancement of the LaModel program from a personal perspective. The narrative starts with an explanation of the objectives behind the original development of the program and then introduces some of the unique and very practical features of the program. Next, a demonstration of how LaModel was advantageously used to model stresses and displacements in a multiple-seam mining situation with highly variable topography is presented. This is followed by the details of a number of the major enhancements and expansions of the program over the years. A back-analysis of the Crandall Canyon Mine collapse is then briefly presented, along with the associated development of a recommended calibration procedure for the critical input parameters to LaModel. Finally, some of the present research areas are briefly described, and future enhancements are introduced.

Jan 1, 2011

By JOHN G. RAYNER

An examination was made of the filtration characteristics of a haematite slurry using a range of polyacrylamide based flocculants in solutions containing up to 3 per cent sodium chloride. The polymers used were "high" and "Iow" molecular weight and had from 0 to 40 per cent charged groups.Increasing charge density tends to reduce the dose required to achieve optimum performance measured by filtration rate and by cake moisture. In general increased salt addition tends to increase the dosage required to achieve these optima. Both these observations result from the change in polymer extension and accord well with the bridging theory of flocculation.The dose required for optimal cake moisture is up to three times greater than that required for optimal filtration rate. This effect is reasonably consistent for all polymers studied and appears to be non-linearly dependent on salt concentration, with the difference being small at 0 and 3 per cent but quite significant at 1 per cent.

Jan 1, 1987

By Lucien F. Trueb

The microstructure of explosion-bonded pairs of similar and dissimilar metals has been investigated by electron microscopy. A review of the specific problems encountered and the methods used for obtaining surface replicas and thin-film transmission specimens of the bond interface is given. The bond area is mainly characterized by continuous and practically diffusionless metallurgical bonding. The very large shear stresses induced along the collision front of the plates being joined causes extreme grain elongation and a symmetrical pattern of subgrains in the bonding direction. The bond zone is also characterized by a very high density of dislocations and pressure-induced twins. Localized heating occurring during the cladding process can result in partial re-crystallization or the formation of thin layers of molten material. The force of precisely controlled explosions causing a high-velocity impact between metal plates has been used for several years to achieve metallurgical bonding between an extremely wide variety of metals. This method essentially consists of accelerating a plate to high velocity toward a stationary plate by a detonating explosive. Since the restrictions to bonding are not those encountered with conventional nethods, it becomes possible to bond pairs of metals having widely different mechanical properties that are immiscible or form brittle intermetallic compounds. Many applications of such composite metals are found in the field of corrosion protection as well as numerous other fields; for example, explosion bonding is being applied for fabricating the materials used by the United States Mint in the new sandwich-type coins. The primary condition for establishing a metallurgical bond is that absolutely clean metal surfaces be brought together. Any metal exposed to the atmosphere is covered with oxides, adsorbed gases, and other contaminants; even a very forceful impact of two such surfaces is not sufficient for bonding. Cowan and Holtz-man,"' who reviewed the dynamics of colliding plates in detail, showed that in order to achieve a good bond the explosion conditions must be chosen in such a way that the plate collision velocity is less than the sonic velocity, in which case no oblique shock waves are attached to the collision front. A pressure wave is then generated ahead of the collision line, and the material forming the colliding surface of each of the plates flows forward and is ejected in the form of a spray, the so-called jet. The dynamic elastic limit of the metals must be exceeded so that there is sufficient plastic deformation. At the point where the jet formed by the junction of the inner surface layers of both plates separates from the combined plates, the material experi- ences a very high shearing strain and the pressure can reach several hundred kilobars. This process strongly influences the microstructure of the bond zone as will be seen later. Behind the collision front, uncontami-nated layers of internal material are brought together under high pressure and are thus metallurgically bonded. I) STRUCTURE OF EXPLOSION BONDS The different types of explosion bonds that can be obtained depend on the explosion conditions, and have been investigated by Cowan and Holtzman,1'2 Holtzman,3 Klein; Bahrani and crossland,' and Buck and Horn-bogen. The preferred kind for practical applications is the so-called wavy bond, typical examples of which are given in Fig. 1 showing light micrographs of various metal-to-metal interfaces. In forming this type-- of bond the collision energy is mainly expended in jetting, the formation of waves, and localized melting. Beyond the crest of the waves, eddy-shaped areas are observed in which the two metals are mixed in a complex pattern of streaks. Cowan and Holtzmanl first proposed that this wavy pattern is analogous to periodic eddy shedding in the flow of a viscous fluid around an obstacle (Von Karman's eddy street). The mass of metal ahead of the stagnation point, which is associated with the jet and has forward momentum, plays the role of an obstacle and the eddies created in the flow of solid metal around the stagnation point are preserved in the final clad specimen. This idea has been reviewed more recently by Klein4 and the variables involved in the wave formation have been discussed in some detail by Bahrani and crosslands and Buck and Hornbogen.6 Several studies of the structure of explosion bonds by light metallography have already been published.1-6 Aside from the waviness and the eddies which were mentioned above, the most striking characteristic of the area in the vicinity of the bond interface is a very considerable longitudinal grain deformation which appears to be strongest at the metal-to-metal boundary and dies out as one moves away from it. Large twins are often observed within the deformed grains, and molten areas are found in the center of the eddy-shaped structures situated beyond the crest of the waves. The large hydrostatic pressures and shear stresses occurring at the interface modify the mechanical and chemical properties of the bond zone. Increases in hardness in this area have been reported by various authors.396 The defects along the interface can also cause a local increase of the chemical reactivity and thus might be expected to boost the etching rate. However, the effects of this preferential etching cannot be observed by light microscopy due to its inherently limited resolution power. The same limitation precludes the observation of morphological features directly along the bond interface as well as the interface itself. Furthermore, no information can be gained by light-

Jan 1, 1969

By Chris Fleming, John R. Goode, James Brown

"Data will be produced from two uranium feasibility studies that were conducted in 2008, both involving uranium recovery from leached slurry by resin-in-pulp (RIP) with strong base resins. Excellent metallurgical results were produced in RIP pilot plants in both studies, and the loaded resins were eluted with strong sulphuric acid and recycled to RIP in an integrated process. Resin losses by either abrasion or mechanical breakage were immeasurably low.Metallurgical results from these two campaigns will be presented, along with an economic analysis of RIP versus the alternative process of CCD or filtration and solvent extraction.INTRODUCTIONThe recovery of uranium directly from pulp with strong base resins in RIP contactors was practiced commercially in the USA in the 1960s, but the process never established itself as a viable alternative to solid/liquid separation and solvent extraction. This was due mainly to the necessity to do a sand-slime separation partially resulting from the unavailability of coarse resin beads, inadequacies in the design of the RIP contactors and screens to separate resin from pulp, as well as problems with the strength and durability of resins in agitated pulps.There have been significant advances in the last 30 years in each of these areas. All the commercial resin manufacturers are now producing very tough, durable, strong base resins, and are manufacturing them in a large bead size that is better suited to RIP applications. In addition, with the emergence and wide spread implementation of the carbon in pulp process in the gold industry in the 1980s, many innovative techniques for preparing leached slurries and separating carbon from pulp have become firmly established. These techniques have proven to be equally applicable to resin separation from pulps in RIP contactors."

Jan 1, 2010

By Gerald G. Johnson

The use of the Scanning Electron Microscope to characterize Coal Mine Dust has been possible because of two circumstances. First, the imaging process of the SEM allows multiple signals (secondary electrons, back scattered electrons, x-ray emission, etc.) to be viewed. Second, the use of a computer to analyze the signals and perform statistical tests on the results. The parameters described are size, shape, chemical composition, mass distributions, and particle thickness. Although the system is intended for particular systems, those of agglomerates (whether tile same or not) are handled by special aspects of the computer programs. The work on fibers is not completed at the time of this report. This work discusses the hardware (developed especially for this process) and software (both on an IBM 360/370 and a PDP 11/20). The latter system allows direct interaction between the operator of the SEM and computer for maximum information at the least time.

Jan 1, 1979

By A. López-Valdivieso, M. A. Corona-Arroyo, M. Llamas-Bueno

Magnetite (Fe3O4) ore is concentrated by low-magnetic-field drums to recover a magnetite concentrate that is low in silica (SiO2). The presence of SiO2 in the magnetite concentrate for steel production increases the steel processing costs, so a major challenge in magnetic concentration is to lower the SiO2 grade in the Fe concentrate. This work presents studies that were carried out on the removal of SiO2 from the magnetite concentration at the plant scale. Studies were performed with a three 36 × 96″ drum unit processing 46 ton/h of rougher magnetite concentrate. These studies showed that silica appears in the magnetite concentrate by three mechanisms, namely SiO2 entrainment in Fe3O4 chains, heterocoagulation between Fe3O4 and SiO2 particles, and mineral locking of SiO2 to Fe3O4. The SiO2 entrainment mechanism had the highest contribution (75%) to the SiO2 recovered in the Fe concentrate. Electrokinetic studies on Fe3O4 and SiO2 showed that heterocoagulation occurs because of the low negative zeta potential of the minerals at the pH of the plant slurry. This study also showed that a high percentage of ultrafine size SiO2 particles trapped in Fe3O4 agglomerates was not removed by the dilution water for the slurry fed to the drum. To lower the SiO2 recovery in the drum magnetic concentration process, work should be directed towards the removal of trapped SiO2 in Fe3O4 agglomerates.