Search Documents

By Reginals W. Burroughs

Diatomaceous or infusorial earth, also called diatomite; fossil meal, or kieselguhr, is derived from the lowest form of vegetable life, the unicellular alga:. It consists of the siliceous remains of extremely minute aquatic plants, which exist in both salt and fresh water. Light i~ a prerequisite to their development. These very small particles secrete silica in much the same way that molluscs secrete lime to form their shells. The diatoms at death sink to the bottom of the body of water in which they lived, and the organic matter gradually decays and is washed away, leaving the skeletons. Thousands of years are required by this means to develop bed5 of sufficient thickness to be of commercial value. The siliceous remains vary in size according to the species, but all are very minute and, as a general rule, they can be identified only under a high-power microscope. Over one thousand types have been distinguished and named. They are all composed of a hydrous or opalescent form of silica, with silica content ranging from seventy to ninety-five per cent. Magnified, they present an attractive sight, a very fragile structure with many interstitial openings known as cells. Many of them consist of two parts, defined as valves, held together by means of a girdle. Certain species, belonging to the Pinnularia family appear as a long comb, in half longitudinal section. The spaces between the teeth are the cells. This particular genus is very common in Nova Scotian deposits.

Jan 1, 1928

By A. Yousefpour, M. Hojjati, G. Chouinard

In recent years, high performance thermoplastic composites such as PEEK and PEKK have been developed for aerospace applications. These materials offer, among the other properties, high toughness and damage tolerances. These are semi-crystalline materials and their processing can lead to different crystallinity that in turn have some impact on the final properties. The effect of cooling rate on degree of crystallinity of PEEK is studied to identify the processing window for manufacturing. Samples were heated at the constant heating rate of 20°C/min up to 375°C (above melting point) and held for 10 minute. The enthalpy of the crystallization was measured at different constant cooling rates of 0.5, 1, 2, 5, 10, 20, 30, and 40°C/min. After cooling, the samples were heated at constant heating rate of 2°C/min using modulated DSC. At temperature between 90°C to 375°C, two endotherm peaks were observed. Results suggest that endotherm peaks are related to a melting and recrystallization phenomenon of one crystal morphology. However, the second endotherm peak could be associated with another crystal structure or morphology.

Jan 1, 2006

By Ajit K. Sinha

"Electromagnetic surveys were conducted over the Yava sandstone-hosted lead deposit in Nova Scotia to evaluate the usefulness of such surveys for detecting and mapping sedimentaryhosted metallic ore deposits. Both transient and multifrequency systems were used for that purpose. The deposit consists of three zones of disseminated galena which also contain minor amounts of pyrite and sphalerite filling the interstices of the sandstone grains. Each zone overlies a basement depression and is separated from the adjacent zone by an intervening basement high. A conductive shale layer underlies the galena-rich sandstone formation at all three zones.Because the concentration of galena (3070 to 4% ) and other sulfides is too low to affect the resistivity of the sandstone formations for detection by ground electromagnetic (EM) methods, the strategy for our survey was to look for areas of basement depressions and the presence of conductive shale in those depressions by EM depth soundings. The presence of the galena-rich sandstone formations could be inferred because of the close association of these two features with the deposits in the Yava area. Both transient and multifrequency EM systems were successful in mapping the basement depressions and detecting conductive shale at the three known mineralized zones. The technique, therefore, seems to offer good prospects for mapping similar stratabound sedimentary-hosted sulfide deposits."

Jan 1, 1993

By Tyson C. Birkett

Initial indicator mineral sampling covering 425,000 square km of northern Quebec and subsequent follow-up of positive results convinced the Ashton ? SOQUEM joint venture to acquire mineral rights to 380,000 hectares prospective for diamond in late summer, 2000. Drilling in 2001 and 2002 identified eight kimberlitic bodies, all significantly diamondiferous. Commercial size stones (retained on a square mesh sieve with openings 0.85 mm) and encouraging grades have been returned by several bodies tested to date: The Renard bodies are formed by steep-sided breccia pipes with kimberlitic matrix surrounded by irregular outboard breccia zones within country rock. These bodies were emplaced in high-grade metamorphic rocks of the Superior Province of Archean age. Orthogneiss with lesser tonalite hosts the breccias, although regionally paragneiss is also present. Diatreme facies portions and nearby country rock are intruded by minor to major (Renard 1) amounts of hypabyssal-facies kimberlitic rock. Renard bodies are macrocrystic, with large olivine crystals set in a groundmass of olivine, phlogopite, monticellite, spinel, carbonate, apatite and serpentine. Proximity of various Renard bodies, with most in a core area of 0.5 square km will allow planning of an efficient mining operation. Geophysics, especially magnetic surveying, has contributed to identifying bodies of kimberlitic rock.

May 1, 2003

By H. Soto, J. Jara, F. Nava

A method has been developed to recover cyanide and copper from cyanidation effluents contaminated with thiocyanate and copper-cyanide complexes. Copper is precipitated as copper thiocyanate or as copper cyanide depending on pH and the concentrations of thiocyanate and cyanide present in the effluent. The copper precipitate presents good dewatering characteristics and can be easily separated from the effluent by thickening and filtration. The solution containing the bulk of the cyanide is then oxidized with ozone to transform the remaining thiocyanate into cyanide. This results in a regenerated solution rich in free cyanide that can be recycled to the cyanidation process. Careful control of the pH of the solution is required. First pH of the effluent is lowered to optimize the precipitation and oxidation stages, then the pH is restored to an alkaline value to regenerate the cyanide solution. Recoveries of over 96% copper are obtained. Free cyanide in the regenerated solution corresponds to up to 160% of the total cyanide contained in the original impure effluent. (The balance being accounted for by the cyanide obtained by the oxidation of thiocyanate). The efficiency of the thiocyanate oxidation stage is 85% to 90%. The method has been demonstrated at bench scale with effluents of various cyanidation plants of northern Quebec. A preliminary economic evaluation indicates that the procedure would be profitable for effluents rich in copper and thiocyanate. This is the case for most barren bleeds of Merrill Crowe and heap leaching operations and also certain CIP tailings of gold ores containing copper impurities. Environmentally the procedure is a better alternative than the current detoxification methods based on the oxidation of cyanide that do not allow for copper recovery and that have difficulties in producing a final effluent with acceptable levels of contaminants such as ammonia, thiocyanate and copper.

Jan 1, 1996

By N. Hwang

There has been increasing interest in the use of robotic machines, and several prototype robotic machines have been developed to automate the grit-blasting process for the purpose of steel bridge maintenance. To utilize such a robotic grit-blast machine effectively, the first consideration is automating the determination of the rust surface area to blast, on the basis of standards of practice, in a rapid manner. This study aims to propose a method to rapidly and accurately determine the rust surface area on steel bridges to blast, with consideration for the standards of practice, from images acquired via a blasting machine. The first step is to perform a color space conversion to transform the input image from a red/green/blue (RGB) color space to a hue/saturation/intensity (HSI) color space. The next step is to detect the presence of rust, using pixel-level classification via the C4.5 decision tree algorithm. Then it is necessary to confirm the blasting area by verifying whether the rust detection result satisfies the specified criteria on the basis of standards of practice. The proposed method was validated on 39 test images with various characteristics with respect to the degree of rusting and rust distribution type. The experimental results showed that the average accuracy rate of rust area classification was about 97.63%, and the success rate of the final decision of blasting area determination was 100.00% for 39 test examples. The whole processing time took an average of only 0.86 seconds per image. The preliminary results demonstrated that the proposed method not only determined whether rust was present in an image and the amount of rust but also indicated whether blasting was necessary, and, if necessary, it rapidly specified the rust surface area that should be blasted on the basis of standards of practice. The proposed method could be successfully incorporated into a robotic grit-blast machine.

Aug 1, 2013

By Burrows. A.

Temperature profiles in the lead smelter sinter bed at Xstrata Zinc Mount Isa sinter plant have been measured using a wireless temperature probe. Corresponding sinter samples have been collected for characterization. The sinter peak bed temperatures were found to be in the range of 1000 - 1200 °C. Microstructures and softening temperatures of these sinters have been characterised. Strong correlations have been observed between peak bed temperature and peak hood temperature, and between mean sinter softening temperature and peak bed temperature. It was found that higher peak bed temperature results in higher sinter softening temperature. For the feed material used in the trials, to obtain sinters with mean softening temperatures above 1050 °C, peak bed temperatures above 1130 °C are required. There appears to be a direct correlation between granule size distribution in the feed and peak bed temperature attained. High peak bed temperatures are observed in sinters with narrow granule size distribution.

Jan 1, 2008

By Samuel Nunoo

Gibraltar Mine is an open pit copper-molybdenum mine in central British Columbia. The stability and potential failure modes of pit walls in the Granite Pit are controlled by the orientations of joints and faults. The structural geology also has an impact on as-mined bench geometries. Geotechnical mapping and surveying were conducted using stereo photographs and photogrammetry software. Detailed digital terrain models were constructed to determine the pit wall geometry and the locations and orientations of larger structural features. The pit wall geometry is compared to conventional survey results and the mapped structures are compared to results from a nearby geotechnical borehole.

Aug 1, 2013

By T. G. Joseph

ABSTRACT: The Canadian oil sands provide some of the harshest mining conditions in the world. Asoperations demand ever bigger tools, the toll on large mobile mining equipment is high, with structural life often half of that expected. Soft ground conditions predominantly lead to fatigue failure scenarios, poor stability, and inconsistent payload evaluations. These challenges are not solely a concern for Canadian oil sands mines; there are far-reaching implications for surface mining equipment in any poor ground-operating environment. This paper focuses on the basics of payload evaluation, motion response, and a simple estimate of structural life for haulers.

Jan 1, 2011

"HPGRs operate in a wide array of applications in mining and minerals. These range from relatively coarse crushing of copper and gold ores at one end of the spectrum to the processing of fine iron ore pellet feed and fine industrial mineral products at the other. Traditionally these circuits involve the treatment of relatively moist materials in pre-grinding circuits; using single stage HPGRs in open circuit or as a closed circuit operation with screening or partial product recycle. More recently, HPGR is increasingly deployed in dry grinding in combination with air classification. The high throughput demonstrated by larger HPGRs has stretched the application towards coarser size reduction duties for a number of new projects. At the same time, the basic principle of inter-particle crushing for generating a product with a high fines proportion gives ground to investigate applying the technology further towards finished product duties. The advantages that HPGR brings are clear. Multi Stage grinding enables energy efficient size reduction across more of the comminution circuit. Primary HPGR serves for a coarse grinding duty, whereas a secondary HPGR is geared towards generating a fine product for downstream processing or even a finished product. This publication summarizes some of the features of staged HPGR circuits and discusses some examples of operating and projected applications.INTRODUCTION In most HPGR applications, the grinding units are used as a single process stage, either in open circuit, such as for pebble crushing [Rose 2002, Westermeyer 2000, Van der Meer 2015] or in closed circuit with classification [Van der Meer 2009, 2012] or product recycling [Van der Meer 2010]. Where a higher plant capacity is required, the HPGRs are commonly installed in multiple lines in a parallel arrangement [Parker 2005, Koski 201, Vanderbeek 2006]. HPGRs can take a wide range of feed size distributions, and grind this down to a product with a high fines content. In general view, there is a limit on the size reduction range that even an HPGR can provide. Grinding down from a very coarse ore (e.g. 70-100 mm) to a very fine product (e.g. 100 microns range) is stretching the ability of a single stage HPGR circuit. Fine grinding generally requires a classification stage that returns the classifier oversize to the head of the HPGR circuit. From one point of view, it can be demonstrated that when recycling a large proportion of fines and middlings back to the HPGR feed results in an effective feed with an overrepresentation of middlings and fines, which will eventually reduce the grinding efficiency and consequently result in an increased recirculation mass flow [Van der Meer 2014]. From another point of view the large recycling mass puts a strain or even a limit on the capacity of present-day conveyor capacity."

Jan 1, 2018

By J. P. Harrison, K. Gao

It is well known that in situ stress measurements at a given site will generally display significant variability. However, it appears that a robust analysis of such measurements – i.e., one that is faithful to the tensorial nature of the data – is not available. Despite the fact that stress is a tensor, stress magnitude and orientation have long been analyzed separately using classical statistics. In earlier work we demonstrated the applicability of tensor statistics to the analysis of stress variability. Here, we investigate further the difference between tensor and classical statistics with respect to the stress tensor correlation matrix. Our analysis commences with the selection of a series of valid correlation matrices. Then, assuming that standardized tensor components follow the multivariate standard normal distribution, these correlation matrices are used to generate marginal probability distributions of the eigen-parameters (i.e., principal stress magnitude and rotation angle). From these, random stress tensors are generated, their eigenparameters determined, and the probability density distributions of these obtained using both classical and tensor statistics. Our results indicate that, for all correlation matrices, both methods give practically identical distributions for principal stress magnitude. However, the discrepancy between the two methods with regard to probability density distribution of rotation angle is highly dependent on stress component correlation. This discrepancy is usefully quantified in terms of the integral, with respect to orientation, of the difference between the two probability density distributions. This parameter, which we term Area Difference, is closely linked to the magnitude of the correlation matrix determinant, which itself is related to the normal stress correlation coefficient (i.e., ?x ,?y ??). Assuming that Area Difference should not exceed 0.1, we identify the critical determinant for any particular value of normal stress component correlation coefficient. Thus, using published in situ stress data and the obtained critical determinants, we show that for a specific normal stress component coefficient, when the correlation matrix determinant is less than its critical determinant, the more precise tensor statistics should be used to calculate the distribution of rotation angle. These results demonstrate that the customary analysis of stress measurement data may lead to erroneous results, and that tensor statistics appears to be preferred approach.

Jan 1, 2015

By Robert K. Jonas

Mining companies can improve production and reduce costs by using the best available technology to control their operations. Many mines are challenged to use multivariable predictive control due to limited experience or staffing, as well as being concerned over the robustness and sustainability of predictive controls. As a result many mines are not realizing optimal performance from their assets, and thusly are not realizing the production increases or cost reductions that this technology can deliver. Over the past few years, the application of multivariable model-based predictive control has gained acceptance in many mining processes and has proven to be robust and sustainable for some of the most challenging mining processes. New automated modeling and performance monitoring software have greatly simplified the effort to implement and maintain these solutions. In addition, many progressive companies have chosen to co-source the implementation and service of these solutions to assure project success and continued benefits stream. This technology has been successfully applied to a wide variety of processes; grinding mills, alumina digestion, CCD washer and thickeners, kilns, calciners, roasters, flash smelters, heat exchangers, evaporators, and most recently in the demanding process of flotation. This paper discusses the state-of-the art in advance control technologies in the mining industry, including the impact of these technologies on maximizing benefits, and implementing best practices for long term sustainability. Benefits from recent multivariable predictive control applications in copper mining will be reviewed, and the use of the latest predictive control technologies and practices in mining will be discussed.

Oct 1, 2009

By G. A. Sutton

This paper, describes a case example from San Manuel mine, where in situ copper leaching of the supergene zone was done over a 14 year period (1986–1999) in conjunction with underground and open-pit mining operations of a copper porphyry deposit. In 1993–1994, testing of well fields was done to determine the amount of copper that would be recovered from in situ leaching of the supergene zone, which was composed primarily of chrysocolla and copper-clay minerals. It was determined that the leach fluid would reach 70% (‘sweep efficiency’) of the copper minerals and that 75% of them would be leached for a recovery of 52% of the acid soluble copper. However, the final results from Zone 6, which had the longest leach history, showed that approximately 42% of the acid soluble copper was recovered, indicating that the `sweep efficiency’ was closer to 55%. Comparison of drill-core assay data from two holes, one drilled before (ARM158) and the other after (SMO9607) leaching, also indicate that approximately 42% of the acid soluble copper was recovered. The discrepancy between the predicted (52%) and actual (42%) recovery is probably related to a combination of factors, including difficulty in maintaining fluid saturation, channeling effects and gypsum precipitation. INTRODUCTION In situ copper leaching at the San Manuel mine, Arizona, was initiated in 1986. It was recognized that some of the copper oxide resource was not going to be economically recoverable by conventional open pit and heap leach methods using solvent extraction and electrowinning (SXEW). A high stripping ratio would be required to access part of the resource. At the time, in situ leaching was becoming popular in the uranium industry but the approach was virtually unknown in the copper industry, especially not at a large scale.

Jan 1, 2019

By E. Larsen

Marine tailings disposal may in some cases be the best solution to one of mining’s major environmental problems. The public standing of this solution is poor, partly because of past bad practices. Among the tailings problems spreading of super-fines and toxic substances stand out. The two are often interrelated as heavy metals, insoluble to sea water, tend to adsorb on fines and spread piggyback on them. Only one paper seen, has addressed the mineralogy of fine particles in suspension. The topic was a secondary tailings deposit. Studies on the sedimentation of fines in large river estuaries show that different minerals sediment sequentially as the salinity of the river increase. Mineralogy in the particle size range where particle separation takes place, has so far, naturally been in focus for the mineral processors. Next to nothing is known about the fines which may cause environmental problems. Tailings disposal has, due to its complexity, to a large extent been taken over by specialists in other fields, biologists, engineering geologists, chemists. The basics of the subject ought also to be given more input by the mineral processing community. A drawn out battle about tailings discharge into some fjords has taken place in Norway. The detractors focus at fine particles (could be dangerous nanoparticles) spreading and emphasize the effects. Calculation is simple: 5 million tonnes tailings annually discharged, containing 1 wt. % < 1µm determined by laser diffraction, results in 50 000 tonnes of nanoparticles let loose in the environment. The industry`s difficulty to answer, is lack of basic research on this topic. Test work to develop a methodology to address the problem is described. One thing the mineral processors know is that final tailings may be comprised of tailings from separate separation processes, with different properties, some of which may require special treatment. One tailings stream from a Norwegian mine is investigated in such a context.

Jan 1, 2016

By Simon Weixiao Sun

Conventionally, carbon steel has been the primary material for salt mine headframe construction, protected by specialty paint layers to prevent corrosion. These layers require meticulous QA/QC during fabrication and erection, and the top UV protection layer typically loses effectiveness after an average of 10 years, necessitating ongoing maintenance. As headframes age, significant structural repairs are often required. Alternatively, stainless steel can eliminate corrosion concerns, though its high initial cost has limited its use in headframe design. However, when factoring in long-term maintenance costs, a 'maintenance-free' stainless steel headframe can be an attractive option for stakeholders. This paper compares the structural performance and economic considerations of painted carbon steel and stainless steel headframes, including initial capital and long-term maintenance costs. Additionally, it explores a hybrid approach, strategically using carbon steel in low-contamination areas to reduce initial costs and significantly minimize future maintenance efforts.

May 4, 2025

By Emmanuel Henry

The Kiruna Mine, operated by Luossavaara-Kiirunavaara AB (LKAB), is a large-scale sublevel caving operation located in Arctic Sweden. Iron ore is extracted at a rate of 60 000 tpd from a four-kilometre long, 80-100 m thick and 60-degree dipping orebody. Actual production depth is 800 m under the top of the footwall and about 550 m under the average regional ground level. Because no support is left between the footwall and the hanging wall and despite the relatively low mining depth, extraction results in the formation of a huge stress shield modifying the original stress field and concentrating it just beneath the mining level. Effects of stress concentration on the rock mass are monitored through field observations and a seismic system. Severe stress-induced deterioration of openings is visible to a certain radial distance from the mining level: pillar crushing and fault slips are more and more common and extended. The seismic system, installed in collaboration with CANMET, confirms the field observations and completes them in the rock mass: fault slips and major seismic events are recorded. The system proves its effectiveness to outline the extension of the dynamic response of the rock mass, which actually seems much larger than field observations let think.

May 1, 2001

By Michael K. Watkeys, Jacobus E. Potgieter, E. Guy Charlesworth, Carl R. Anhaeusser, Roger L. Gibson, Alexander F. M. Kisters

Intrusive features of cupriferous noritoids from the deeply eroded Proterozoic granite-gneiss terrane of the Okiep Copper District in South Africa illustrate the emplacement of basic magmas into a mid- to lower-erustal environment. The small, easterly trending intermediate to basic intrusions of the so-called Koperberg Suite display a wide variety of geometries, occurrence and emplacement modes, including dike-, sill- and plug-like bodies. Coeval granulite-facies metamorphism, partial melting and ductile deformation of country rock gneisses during intrusion of the basic magmas led to emplacement and dilation processes of the intrusions which differ significantly from those described for dikes and sills at shallow crustal levels. Regional tectonic stresses are largely negligible due to very low differential stress [( )] during deformation and intrusion, which permits emplacement at high angles to the principal compressive stress and which follows favorably inclined structural anisotropies. Dilation of the basic intrusions is largely determined by compositional differences of host rocks and occurs by thermal erosion, i.e., assimilation of wall rock gneisses. Low length-to-width ratios of basic to ultrabasic dikes and predominantly diapir-like outlines of mesocratic bodies are indicative of low viscosity contrasts between the mafic magmas and high-grade metamorphic host-rock gneisses. Buoyancy-controlled ascent of the basic to intermediate intrusions is expressed by a broad compositional zoning of basic bodies with respect to the granite-gneiss stratigraphy. Various intrusion modes within the stratigraphic column of the granite-gneiss sequence reflect a lithological and structural stratification of the seemingly homogeneous granite-gneiss terrane.

Jan 1, 1994

By E. L. Jones

THE concentration of tin in the tailing from Sullivan ore has been de-scribed by H. R. Banks, superintendent of the Sullivan concentrator, in a paper presented at the annual meeting of the B.C. Division of the Institute in October, 1941(1). The reduction of the concentrate EARLY EXPERIMENTAL WORK Experimental work on the smelting of Sullivan tin concentrate was done in Trail. Preliminary tests in the laboratory indicated that the concentrate could be successfully treated in an electric furnace. A small electro-thermic pilot plant was built and, at the same time, reduction tests were made in an oil-fired reverberatory furnace. Both types of treatment gave satisfactory results. It developed that no fluxes, other than those already in the ore and in the ash of the reducing agent (either coke or coal), were needed. By using an excess of iron, a complete recovery of the tin in the slag was obtained, but with 20 to 30 per cent iron in the tin. A counter-current treatment was thus indicated, viz., the primary slag was first treated with hard-head (iron-tin alloy) from the refining and final slag smelting, producing tin with 2 to 4 per cent Fe. This secondary slag was subsequently treated with scrap iron to produce tin-iron alloy, and a final slag low enough in tin to waste. By following this line of endeavour, a tentative flow-sheet was drawn up and a successful operation was established. Without resorting to electrolytic refining, a metal comparable in purity to standard tin was produced. Metallurgical results and operating costs were found to be very similar with an electric furnace or a fuel-fired furnace. &apos;However, electric furnace smelting was chosen because the furnace required was smaller, and because some doubt existed as to the cleaning of the final slag in the reverberatory furnace reduction.

Jan 1, 1944

By Shanjun Mao

This paper summarizes current achievements of Digital Mining in China, puts forward the concept, framework and key technology of Smarter Mining, and finally points out that there is still a long way to go in order to acheive Smarter mining.

Aug 1, 2013

By Dale A. Sketchley

"Abstract - Large differences in gold content between rock fragments or small volumes of unbroken rock, commonly referred to as the nugget effect, are related to sample and subsample weights, crushing and pulverizing sizes, and density, shape and size distribution of gold grains. Gy’s sampling theory, which relates gold grain characteristics to sample weight, fragment size, and error, provides a method of estimating a sampling constant that quantifies deposit characteristics that contribute to the nugget effect.Sampling control tests provide methods for determining the sampling constant. Sampling control charts, comprising nomographs that relate sample weight, error and particle size, use the sampling constant to optimize sample weights and comminution sizes, and to minimize the variability introduced by incorrect or inadequate sampling and sub-sampling procedures. The sampling constant is useful in classifying ores into arbitrary categories according to expected sampling difficulties. Spatial variations in the sampling constant can be used to aid in defining ore types for modeling geological domains, which should have unique sampling protocols. Examples that illustrate the relationship between the sampling constant and expected sampling problems, and the spatial variation of the sampling constant, are given for various deposits. Case histories of sampling programs are presented.Quality control procedures consist of monitoring sample contamination amd accuracy and precision of analytical results. Accuracy, precision and contamination are monitored by including standards, duplicates and blanks into batches of samples being analyzed. Analytical results of standards and blanks are plotted on a batch-and-time basis with control lines at plus or minus two/three standard deviations as confidence limits of the accepted round-robin value. Examples are given that illustrate a laboratory system in statistical control and the various types of problems encountered when interpreting quality control data. These include sample transcription errors, degradation, contamination, sampling errors, detection limit problems, instrumental drift, lack of statistical control, laboratory bias, procedural problems and data tampering.Where bias is negligible, precision, as a function of concentration, can be obtained from duplicates using the Thompson-Howarth precision algorithm, which also allows calculation of the practical detection limit. Examples are given that illustrate variations in precision, with grain size, for different types of duplicates, as well as mineralogical changes and sampling errors."

Jan 1, 1998