Search Documents

By M. Pelegrin Rodríguez, J. Castellanos Suarez, R. Causse Miyares, A. Mosqueda Martínez, Rubén Aja

"This paper will describe the methods and the results of studies to integrate polymetallic nodule extractive metallurgy methods into existing and operating technologies or plants with the aim of reducing the capital cost of deep-sea mining challenges and to streamline the supply chain of raw materials where such opportunities exist (Moa Bay and others). It will also assess the possibility to carry out pre-industrial processing operations with productivity corresponding to pilot mining operation in the Clarion Clipperton Fracture Zone (5–10% of assumed annual production) with the use of existing facilities. The present study shows the results of high pressure acid leaching of polymetallic nodules (PN) mixed with laterite non-commercial ore. It also presents PN processing in the presence of ammoniacal tailings obtained from the Caron process. The experiments were carried out according to 24 full factorial design. The regression equations were determined for extractions of Ni, Co, Cu, Zn, Fe and Mn. The obtained results show that increasing temperature, the acid/ mineral ratio and the leaching time have a positive and significant effect on extraction of all the studied metals. However, increasing the tailings addition favors the extraction of the Co, Zn and Mn. The extraction of Fe diminishes, while the extraction of Ni is not affected. In the following conditions; 180 min, 260°C, 0.5 kg acid /kg minerals and 2.5 kg PN/kg tailing; the following extractions were achieved; 96% Ni, 90% Co, 97% Cu, 99.9% Zn, 43% Mn and 26% Fe."

Jan 1, 2017

Base metal skarns have provided most of South KoreaÆs Pb-Zn production since the end of the Korean War, with the bulk of production coming from the Yeonwha group of mines. Currently only the Kumho Zn-Pb mine remains in production, while magnetite skarns are mined as iron ore at the old Yomisan and Ulsan base metal deposits. Base metal skarns within Korea fall into two groups; those hosted in Cambro-Ordovician limestone sequences, situated mainly in sub-basins in the northeast of the peninsula and those hosted within carbonate volcaniclastics of the Cretaceous Gyeongsang Basin, in the southeast. Sulphide mineralisation is dominated by sphalerite, lesser galena, arsenopyrite and pyrrhotite, with subordinate chalcopyrite and molybdenite with Ag-Co credits. Skarns are composed of pyroxene-clinopyroxene-garnet-carbonate-quartz and magnetite, grading into silica-chlorite hornfels. Most of the significant deposits display both pipe-like and stratabound morphology. In the Hwanggang district, a zoned distribution of Ordovician carbonate hosted, polymetallic Mo-W and Zn-Pb-Cu-Ag base metal deposits is observed around the margin of a Cretaceous granite intrusion. Similar polymetallic zoning is recognised in the Sangdong district, where deposits include Sangdong (W), Abisan (Au-As) and Yomisan Northwest (Zn-Mo) and Yomisan Southeast (Zn-Cu-Co), hosted in Ordovician carbonates in proximity to Cretaceous quartz porphyry intrusives. The Taebaeksan district hosts the largest deposits at the Yeonwha group of mines. Although production ceased in 1993, significant resources remain, hosted in Ordovician limestone and dolomite that have been intruded by Cretaceous quartz porphyry dykes. Deposits in the Cretaceous Gyeongsang Basin consist of bedding concordant, vertically stacked base metal skarn mineralisation at several stratigraphic horizons, connected by carbonate-base metal veins with a pipe-like geometry. Evidence such as massive sulphide layers and restricted proximal carbonates suggest mineralisation was initially partly syn-depositional, with both base metals and carbonate venting onto a lake floor, being reconstituted to skarns by later magmatic activity. Skarn mineralisation is universally closely associated with the Cretaceous Bulgugsa Series acid intrusive activity. A common intrusive-structural relationship is recognised at some deposits, with bedding plane and imbricate thrust faulting, introducing felsic intrusives as sills. These intrusions probably introduced the base metals, carbonate, boron, heat and silica, which metasomatically reconstituted the carbonate lithologies to form skarn assemblages. The Korean base metal skarn deposits are regarded as Zn-Pb chimney-manto skarns and display similarities to the Mexican manto type deposits.

Jan 1, 1999

By W. Dresler

Since 1900, smelters have produced in Canada over 10 million tonnes of iron silicate waste slag containing significant quantities of nickel, cobalt and copper. The slag is reduced with aluminum and calcium oxide additions to produce a copper, nickel, and cobalt enriched alloy and an obsidian slag. It was found that the obsidian slag contained very low copper, nickel, and cobalt contents, indeed below 0.06% of each the base metals. These metals have been collected in a type of ferroalloy with up to 1.6% Cu, 2.1% Ni, and 0.7% Co. These alloys can be recycled according to standard procedures in base metal smelters. Surprisingly, the obsidian slag was found to display good cementitious and pozzolanic properties, which is a requirement for cement production. The strength producing properties of the obsidian slag are considered to be exceptional at the 28 day standard test age. In most cases, the obsidian concrete is characterized by significantly greater strengths than control concrete tests and concrete containing fly ash or blast furnace slag. Compressive strength data at 365 days for the obsidian concrete is impressive.

Jan 1, 2003

The heap leaching of base metal ores can be applied within a number of different scenarios, each requiring its own set of parameters to be determined during the metallurgical testwork programme. The occurrence of acid-gangue reactions and its concomitant precipitation reactions is a factor common to all cases, and requires that the optimal irrigation acid strength be found that optimises the combination of extraction kinetics, acid consumption and possible premature blinding of the heap due to precipitation reactions. For the heap bioleaching of base metal sulphide ores the oxygen requirement needs to be determined for the design of the aeration system, and the operating conditions employed need to also facilitate bacterial growth over and above the acid strength optimisation mentioned above. For the heap bioleaching of chalcopyrite, the need to preserve heat in the heap becomes yet a further parameter of critical importance. For the heap leaching of uranium ore a choice is required between either chemical or bacterial oxidation, based on economic and processing factors. Frequently the operating parameters need to be determined for a circuit in which heap leaching is combined with tank leaching and/or pressure leaching. The processing implications and metallurgical testwork approach for each case is discussed.

Jan 1, 2007

By Steven R. McCutcheon

ABSTRACT - The Bathurst-Newcastle district, a subcircular area approximately 50 km in diameter, contains 37 massive-sulfide Zn-Pb-Cu deposits that have defined tonnages, and has 58 additional significant occurrences. The largest deposit is Brunswick No . 12, with total reserves (including past production) of 148 million tonnes grading about 12.50% combined Pb-Zn. The deposits/ occurrences are spatially associated with a felsic volcanic pile that contains a significant sedimentary component. In the vicinity of the Brunswick deposits the pile is divisible into lower and upper sequences. The lower is characterized by megacrystic rhyolite (QFAS) and coarse- to fine-grained quartz ( ± feldspar) wackes and/or tuffites (QAS); the upper sequence is characterized by aphyric to sparsely porphyritic rhyolites (HWR), micro-breccias, and acid hyaloclastites with interbedded dark gray sedimentary rocks. A greenish gray to dark gray siltstone (GGS) separates the two sequences and hosts the sulfide deposits. Lithogeochemistry of !east altered samples shows that the QFAS and HWR have an igneous character, whereas the QAS and GGS have a sedimentary sig­ nature. The two types can be distinguished on the basis of total alkalies (Na2O + K2O) versus SiO2/Al2O3, and Ti02 versus Fe/(Fe+ Mg). Some trace elements, such as Sr and Y, also help differentiate the two types. The QFAS differs from HWR by having higher La/Yb ratios but lower Zr and Th. The Bathurst deposits do not show a close spatial association with rhyolites but rather are stratiform bodies (lacking clastic textures) within fine-grained sedimentary rocks. The scarcity of barite and other sulfates indicates that there was little mixing between the metal­ bearing fluids and cold seawater. High Fe/(Fe+ Mg) ratios in chlorites also indicate the fluids did not contain a significant seawater component. These features and the large size of many deposits suggest that the deposits formed from metal-rich pore waters that were released catastrophically from an acquifer in the footwall stratigraphy.

Jan 1, 1992

By C L Humphrey

Biological monitoring programs for environmental protection require techniques that will provide both early detection of possible adverse effects, and assessment of the ecological significance of these effects. Monitoring techniques must include responses sensitive to the impact, that can be subjected to rigorous statistical analysis and for which statistical power is high. Such issues in baseline research of 'what and how to measure?' and 'for how long?' have been the focus of a program being developed to monitor and assess effects of mining operations on the effectively pristine, freshwater ecosystems of the Alligator Rivers Region (ARR) in tropical northern Australia. Application of the BACIP design, utilising a form of temporal replication, to univariate data (single species) and multivariate (community) data is described. For new ventures in Australia (such as mining), monitoring programs should be designed to establish with high probability that an impact on the ecosystem no greater than a prescribed amount has gone undetected - as opposed to simply allowing development to continue as long as 'proof' of impact is not available. Such a precautionary approach represents a shifting of the 'burden of proof' and would require with the BACIP design, a minimum number of baseline years in order to achieve some desired level of statistical power in a test for impact. Power analyses conducted on two to five years (depending upon the technique) of baseline data from streams of the ARR indicate that the BACIP design would not impose developmental constraints on mining and that a policy of reversal of the burden of proof would provide the cornerstone required for managing such development in an ecologically sustainable manner.

Jan 1, 1994

By Stefan Stein, Jorge Davo, Adrian Villanueva, Abhay Kulkarni, Shane Hoff

Pellet sizing plays a key role in the productivity of the entire pelletization and subsequent induration processes, whereas uneven pellet sizing can increase the circulating load of the balling circuit inducing operational instability leading to production loses. The present work focuses in comparing the agglomeration kinetics, expressed as the number of pellets produced in a determined target size range, of different organic binder-bentonite binder systems. The batch experimental results show that the binder system has a great influence in the resulting pellet sizing, which could lead to advantages in terms of balling circuit efficiency and eventually operational de-bottlenecking.

Oct 1, 2019

By Alexsandro Berger

Many thickener operations continue to be constrained by their ability to process consolidated underflows with high yield stress and associated rheological characteristics. To maximise separation efficiencies within a thickener operation the use of both engineering and reagents are always considered. BASF is developing a Novel Flocculant Technology (NFT), which combines both of these aspects to enhance the performance of the thickener operation, improving the areas of pumping efficiency, underflow density and throughput. The development progression through stages of laboratory scale batch test work to pilot scale dynamic testing has demonstrated the ability of BASF NFT to maintain effective flocculation performance whilst increasing the rate of underflow consolidation, and significantly reducing the associated underflow yield stress of the treated material. This paper will present application data through full scale dynamic testing conducted on a granite washing circuit to demonstrate the performance benefits that NFT offers to a thickener operation. Compared to the conventional flocculant treatment, within the operation parameters and conditions employed during the full scale trials, the NFT produced a reduction in underflow rheology (yield stress) of around 45 % relative to a given solids and improved the underflow consolidation by providing at least 20% relative higher underflow solids loading than the upper limit of the given typical thickener operation. Keywords: flocculation, thickener operation, rheology, yield stress, underflow density, throughput, slump test

Sep 1, 2012

By Adrian Villanueva

In this study, we compare the influence of two water soluble polymer-based organic binders added to bentonite, to form a hybrid-binder system for Iron Ore Pelletization. The kinetics of the agglomeration process and resulting properties of green pellets were analyzed, and examined in detail through Scanning Electron Microscopy (SEM) and Confocal Laser Scanning Microscopy (CLSM) analytical techniques. It was encountered that the differences in the mechanical properties of the pellets produced by each binder system could potentially be explained by the ability of the polymer to interact with bentonite at a macromolecular level.

Oct 30, 2017

By Douglas Glenn Stalheim

Production of structural hot strip coils in yield strengths up to 700 MPa and higher is increasing worldwide. In addition to requiring higher strengths, structural designs require good formability, toughness and weldability. Toughness and formability come from optimization of ductility properties while weldability comes from lower carbon equivalent alloy designs. The key to the production of optimized cost effective hot strip produced structural coils is to create the correct balance of a proper alloy/processing design to create as fine and uniformly distributed cross sectional grain size as possible. In addition to the creation of the fine/uniformly distributed cross sectional grain size, the correct microstructure needs to be created. In the higher yield strength grades, not only does the grain size/distribution play a role but a proper understanding and control of microalloying precipitation kinetics is required. This paper will present various examples of hot strip alloy designs to produce cost effective hot strip structural coil in yield strengths up to 700 MPa with good ductility and weldability. In addition, steelmaking/casting parameters will be discussed to assure successful production. Optimized reheating, rolling and cooling parameters of these alloy designs will be introduced in an accompanying paper.

Aug 17, 2017

By J. W. Craig

THERE are, broadly speaking, three classes of refractories: acid, basic, and neutral. Fireclay, diaspore, silica, and the sillimanite group of minerals are included in the acid class, which for the most part are not very resistant to basic slags. The neutra! class includes chrome, graphite, and silicon carbide, while most important in the basic class are magnesite and dolomite. Although chrome is a neutral refractory, it has no sharp distinction from magnesitic refractoriness. The two are mineralogically (1) and chemically related, and there is available on the market a complete range of brick from straight chrome to straight magnesite. In many indus trial applications, one. type may be replaced by another. The present paper is confined entirely to such refractories. The following types of basic and neutra! brick, each sold under sorne special trade name, are available for use in Canada: (1) Burned chrome from crude ores (2) Burned chrome from treated ores (3) Chemically bonded (unburned) chrome-magnesite ( 4) Burned chrome-magnesite (5) Chemically bonded ( unburned) magnesite-chrome (6) Chemically bonded (unburned) high-iron magnesite (7) Burned high-iron magnesite (Regular) (8) Burned low-iron magnesite (9) Burned olivine-magnesite (10) Burned magnesitic dolomite-chrome (Magnecon) (11) Chemically bonded magnesitic dolomite-chrome (Magmecom) (12) Dolomitic brick

Jan 1, 1936

By D. A. Dahlstrom

Since the initiation of continuous vacuum filtration over a half a century ago, control of the performance of these units has been largely in the hands of the plant operators, Obviously there has been the introduction of variable speed drives, adjustable speed feed slurry agitation, vacuum control, etc. However, control of , filtration rate, cake moisture content, dissolved solids recovery by cake washing and dryness of solids where thermal drying is practiced is primarily up to the operators judgment and the settings he imposes on the controls he regulates, - filter cycle time, submergence, vacuum level, filter medium condition, wash water- rate, and other such factors. While this may seem to indicate that good operators are a paramount necessity, and certainly always desirable, a considerably more vital factor is a properly designed filter and filter station for the particular application. With improper design, highly skilled operator control will still leave much to be desired in the way of process performance. On the other hand, with correct design and application, the operator can easily and quickly control normal plant surges and quality variations within the flow sheet. Due to the growing complexity and inter dependency between the various processing steps in the modern flow sheet as well as the rising costs of capital investment, labor, maintenance, and losses involved with off quality product or insufficient recovery, automatic or inherent performance control of many of the unit operations within the metallurgical mill must be achieved. Continuous filtration should and can be susceptible to such control. However, as every instrument engineer knows, before control and automation can be achieved, a basic understanding of the operation is mandatory. This is primarily due to three factors. First, as in so many control applications, it may not be possible or feasible to measure the factor that it is desired to control. Thus, secondary variables that influence the major factor are substituted for control purposes. Secondly, it is always desirable to take advantage of any ?self-regulation? within the operation being performed on the equipment. In so doing, not only will automatic control be facilitated, but instrumentation is minimized while experiencing most efficient operation. Finally, automatic control is not a cure-all for improper process and equipment design and application. Hanging instruments on such an operation can be compared to constructing a sizable building on sand, Results will be disappointing.

Jan 1, 1960

By E. Daniel Baker

"The term ""oxide disseminated gold deposits"" refers to near surface, gold-bearing veins, faults, and shear zones that typically contain appreciable amounts of oxidized ore or gossanous material, resulting from oxidation of sulphides.The concept of oxide gold deposits underwent a dramatic change after the discovery of the Carlin gold mine in north-central Nevada in the early 1960's. Now the term implies large tonnage, low-grade bulk minable deposits that are processed by milling and/or heap leach methods.Oxide gold deposits may now be classified into two types: (1) primacy ores derived directly from the effects of hydrothermal alteration on host rocks, and (2) secondary oxide ores derived from the supergene oxidizing effects of circulating post-mineral meteoric waters on gold-bearing sulphides. Each type of deposit has unique diagnostic mineralogic and metallurgical characteristics that must be addressed to achieve successful exploration strategies and viable metallurgical exploitation.The author would like to thank the management of Newmont Metallurgical Services and Newmont Mining Company for allowing presentation of this paper. Thanks are due Sevkett Acar, Donald M. Hausen, Jerry R. Odekirk, and D. Erik Spiller for their critical review of the manuscript. Also, thanks are extended to all those cited and uncited whose hard work and dedication have led to the discovery and exploitation of near-surface oxide gold deposits. Our knowledge would be greatly limited without the dedication and research of many others."

Jan 1, 1997

By R. E. J. Putman

The, principle purpose of applying control to any process is to improve -its profitability. Even when considering the impact of a plant on its surrounding ecology this is still, true, since the sanctions felt most immediately from violations of the regulations are the economic penalties or fines which result. 'Economic performance can be improved within a plant in several ways. Steps can be taken to cut the cost or quantity of resources (power, water, steam, reagents) consumed per ton of throughput. Alternatively the operation of the plant can be changed so as to increase product yield per ton of throughput. Or the cost of labor and plant overhead per ton of product can be reduced by raising the total 'throughput, assuming-that the increased amount of product can be sold for the same unit price. In designing a program to improve plant performance, all three routes should be borne in mind. To maximize the return on investment over a longer period, it is also desirable to set up a comprehensive plan of improvement, equipment being specified at each step in the implementation so as to ensure that it fits in with later requirements later implementation being either by means of modular additions or by having the necessary interfaces built into the original equipment. It is much cheaper in the long run to buy features which have been factory-installed, than to have to make provision for their later addition in the field. This paper examines some ways of improving plant performance by the application of various process control techniques using both analog and digital con¬trol equipment. A general philosophy of application is suggested and specific examples described.

Jan 1, 1972

By J. D. Sullivan, Chester R. Austin, J. L. Nunes

IT is characteristic of most shales and surface clays that a bloated or vesicular structure is produced by burning to a sufficiently high temperature, usually about 150° to 200°F. above the normal maturing temperature used when making common brick of the same material. Early investigators recognized this phenomenon and attributed it to the release or expansion of gases in the vitrified piece. A number of theories have been advanced on the source of the gas and the manner in which it is evolved, but none has been proved conclusively. T. E. Jackson's1 early theory that bloating was due to oxygen evolved by ferric oxide in forming ferrous oxide or ferrous silicate was never substantiated by experimental work. Orton2 thought this explanation unlikely, as all clays should bloat at the same temperature, if this were true. He found that clays bloated in the range from 1100° to 1706°C. Also, many clays containing iron did not bloat at all, but fused. Orton also stated that oxidation of different types of carbon was a source of gas during burning of clay and should be expelled before vitrification was reached. Seger3 said that he had dissolved ferric oxide in glass with no apparent evolution of gas. Matson4 came to the conclusion that ferruginous days burned to vitrification in an oxidizing atmosphere did not have the iron in a ferrous form. According to Wilson,5 the gases responsible for bloating might be entrapped air, water, sulphur dioxide, carbon monoxide, oxygen, and hydrocarbons. He spoke of two causes of bloating: (I) incomplete oxidation below the vitrification temperature, and (2) decomposition of compounds and evolution of gases above the vitrification temperature. Decomposition of calcium sulphate was given as the principal example of the latter. Bleininger and Montgomery6 stated that if any time after vitrification began the evolution of gaseous matter was rapid enough so that pressure was produced within the day, bloating was certain to take place, and a vesicular structure would be formed. They recognized that initial structure, rate of heating, kiln atmosphere, and many other variables affected bloating. Work done by the Emergency Fleet Corporation7 indicated that clays suitable for bloating should be high in compounds of metallic oxides, carbon, sulphur, sodium, potassium, or other equivalent compounds, which either act as a flux or give off gas. Orton2 studied the chemical status of the carbon, iron, and sulphur in clays during various stages of burning, using a Huron shale that bloated under oxidizing or reducing conditions. He found that only a small amount of sulphur remained in the clay and that it could be detected only by dissolving the clay in hydrofluoric acid; i.e., the sulphur was in combination with the silica. Orton said that this sulphur was probably in the form of iron sulphide dissolved in the glass phase. The iron sulphide was not broken down by heating, but was dissolved in the basic slag or glass,

Jan 1, 1942

By J. D. Sullivan, Chester R. Austin, J. L. Nunes

IT is characteristic of most shales and surface clays that a bloated or vesicular structure is produced by burning to a sufficiently high temperature, usually about 150° to 200°F. above the normal maturing temperature used when making common brick of the same material. Early investigators recognized this phenomenon and attributed it to the release or expansion of gases in the vitrified piece. A number of theories have been advanced on the source of the gas and the manner in which it is evolved, but none has been proved conclusively. T. E. Jackson's1 early theory that bloating was due to oxygen evolved by ferric oxide in forming ferrous oxide or ferrous silicate was never substantiated by experimental work. Orton2 thought this explanation unlikely, as all clays should bloat at the same temperature, if this were true. He found that clays bloated in the range from 1100° to I700°C. Also, many clays containing iron did not bloat at all, but fused. Orton also stated that oxidation of different types of carbon was a source of gas during burning of clay and should be expelled before vitrification was reached. Seger3 said that he had dissolved ferric oxide in glass with no apparent evolution of gas. Matson4 came to the conclusion that ferruginous clays burned to vitrification in an oxidizing atmosphere did not have the iron in a ferrous form. According to Wilson,5 the gases responsible for bloating might be entrapped air, water, sulphur dioxide, carbon monoxide, oxygen, and hydrocarbons. He spoke of two causes of bloating: (I) incomplete oxidation below the vitrification temperature, and (2) decomposition of compounds and evolution of gases above the vitrification temperature. Decomposition of calcium sulphate was given as the principal example of the latter. Bleininger and Montgomery6 stated that if any time after vitrification began the evolution of gaseous matter was rapid enough so that pressure was produced within the clay, bloating was certain to take place, and a vesicular structure would be formed. They recognized that initial structure, rate of heating, kiln atmosphere, and many other variables affected bloating. Work done by the Emergency Fleet Corporation7 indicated that clays suitable for bloating should be high in compounds of metallic oxides, carbon, sulphur, sodium, potassium, or other equivalent compounds, which either act as a flux or give off gas. Orton2 studied the chemical status of the carbon, iron, and sulphur in clays during various stages of burning, using a Huron shale that bloated under oxidizing or reducing conditions. He found that only a small amount of sulphur remained in the clay and that it could be detected only by dissolving the clay in hydrofluoric acid,i.e., the sulphur was in combination with the silica. Orton said that this sulphur was probably in the form of iron sulphide dissolved in the glass phase. The iron sulphide was not broken down by heating, but was dissolved in the basic slag or glass,

Jan 1, 1942

By R. G. Robins

The effective removal of arsenic(V) from process streams and waste waters by precipitation with excess iron(II1) is sometimes attributed to the formation of basic ferric arsenates. The authors have investigated this system extensively using a variety of techniques and conclude that the solid phase produced is iron(II1) oxyhydroxide (ferrihydrite) with arsenic(V) adsorptively bound to the 2-3 nm particles.

Jan 1, 1992

By R. P. King

The essential features of an image analysis system for the measurement of mineral liberation are described. Good procedures for image acquisition and image processing prior to image analysis are defined. The basic stereological methods for the measurement of volume fraction, surface area per unit volume, particle size distribution and liberation spectra are given. The appropriate methods for fractal correction when measuring perimeters of irregular features and for frame correction when measuring linear intercepts are developed. Practically useful and accurate procedures for the stereological correction of particle size and liberation distribution methods are given.

Jan 1, 1995

Refractory gold concentrates often contain submicroscopic gold that is encapsulated within the crystal matrix of iron sulfide minerals, such as pyrite, pyrrhotite and arsenopyrite. To recover the gold, the host mineral must generally be broken down chemically by oxidative processes, such as roasting, pressure oxidation or bacterial leaching, which expose the gold for subsequent recovery by leaching in cyanide solution. The focus of attention in these pretreatment processes is usually the oxidation of the sulfides to elemental sulfur, sulfur dioxide gas or sulfate ions. Less attention is paid to the deportment of iron and the changes in its oxidation state, although this can have a profound effect on gold and silver liberation, as well as downstream operating costs. Iron sulfide minerals break down completely during pressure oxidation, and dissolve in the sulfuric acid solution that is gener-ated from oxidation of the sulfides. This dissolution liberates the tiny gold particles that were originally trapped in the sulfide crystals, and gold recovery during subsequent cyanidation is usually very high (>95%). Iron goes into solution in the oxidation process, initially as ferrous sulfate, but this compound is rapidly oxidized to ferric sulfate, which then hydrolyzes and reprecipitates. The form of the precipitate varies depending on the operating conditions in the autoclave and the presence of certain metal cations. When the acidity in the autoclave is quite low (<20 g/L H2SO4) and the temperature is high (>200°C), the formation of hematite is favored. When the acidity is high (>20 g/L H2SO4) and the temperature is relatively low (160 to 200°C), the formation of basic iron sulfate is favored. If the ore or the leach solution contains significant levels of certain cations (such as Na+, K+, NH4+, Ag+ or Pb2+) and the acidity is high (>20 g/L H2SO4), jarosite compounds are favored. Hematite is the desired iron product in the autoclave discharge, for both metallurgical and environmental reasons, but it is difficult to operate an autoclave under the conditions required for effective liberation of gold without converting some of the iron to basic iron sulfate and/or jarosite. These compounds fall into a category of iron compounds known generically as iron hydroxy sulfates, all of which can cause significant processing and environmental problems in the downstream gold process. This paper deals specifically with basic iron sulfate: the conditions under which it is formed in an autoclave, the problems that are caused by its presence in the feed to a cyanidation plant and possible remedial strategies that can be adopted, both in the autoclave and downstream.

Jan 1, 2010

Refractory gold concentrates often contain sub-microscopic gold that is encapsulated within the crystal matrix of iron sulphide minerals such as pyrite, pyrrhotite and arsenopyrite. To recover the gold, the host mineral must generally be broken down chemically, by oxidative processes such as roasting, pressure oxidation or bacterial leaching, which expose the gold for subsequent recovery by leaching in cyanide solution. The focus of attention in these pre-treatment processes is usually the oxidation of the sulphides to elemental sulphur, sulphur dioxide gas or sulphate ions. Less attention is paid to the deportment of iron and the changes in its oxidation state, although this can have a profound effect on gold and silver liberation, as well as down stream operating costs. Iron sulphide minerals break down completely during pressure oxidation, and dissolve in the sulphuric acid solution that is generated from oxidation of the sulphides. This liberates the tiny gold particles that were originally trapped in the sulphide crystals, and gold recovery during subsequent cyanidation is usually very high (>95%). Iron goes into solution in the oxidation process, initially as ferrous sulphate, but this is rapidly oxidized to ferric sulphate, which then hydrolyzes and re-precipitates. The form of the precipitate varies depending on the operating conditions in the autoclave and the presence of certain metal cations. When the acidity in the autoclave is quite low (<20 g/L H2SO4) and the temperature is high (>200OC) the formation of hematite is favoured. When the acidity is high (>20 g/L H2SO4) and the temperature is relatively low (160 to 200OC), the formation of basic iron sulphate is favoured. If the ore or the leach solution contain significant levels of certain cations (such as Na+, K+, NH4+, Ag+ or Pb2+) and the acidity is high (>20 g/L H2SO4), jarosite compounds are favoured. Hematite is the desired iron product in the autoclave discharge, for both metallurgical and environmental reasons, but it is difficult to operate an autoclave under the conditions required for effective liberation of gold without converting some of the iron to basic iron sulphate and/or jarosite. These compounds fall into a category of iron compounds known generically as iron hydroxy sulphates, all of which can cause significant processing and environmental problems in the downstream gold process. This paper deals specifically with basic iron sulphate; the conditions under which it is formed in an autoclave, the problems that are caused by its presence in the feed to a cyanidation plant, and possible remedial strategies that can be adopted, both in the autoclave and downstream.

Jan 1, 2010