Search Documents

By I. R. Hanby

A simple solution to the problem of whether to round the calculated number of trucks up or down to the nearest whole number is to adopt as criterion the minimum cost per unit volume/mass moved, assuming all trucks in the fleet are of the same type. When the effects on journey time of varying haul-road slope are calculated for several rolling resistances, it is shown that the optimum angle of 7.5 degrees is moderately independent of both rolling resistance and height climbed; the lower the resistance, the lower the sensitivity to ramp angle, but if haul-road gradients do not exceed 1 in 10 (5.7 degrees) for safety reasons then travelling time will never be more than 7% greater than its theoretical minimum. Further calculations result in a graph of the maximum payload values to incur the minimum cost, for different values of total effective resistance

Jan 1, 1991

Enhancement of Energy Efficiency in Fine Grinding of Copper Sulfide Minerals Using a Pilot-Scale Stirred Media Mill - IsaMill The mineral liberation size of run of mine ores has dramatically reduced over recent years. As a result, the demand for fine grinding which is the most energy-intensive form of grinding has markedly increased in the minerals industry. Furthermore, the fine grinding costs increase exponentially with decreasing product particle size. As a result of the increased energy requirement for fine grinding, it is desirable to optimise fine grinding processes such as stirred media milling (ie IsaMill) to enhance energy efficiency, throughput and the product fineness. In the present study, the influence of slurry rheology, grinding media type and grinding media-to-feed size ratio on energy efficiency in a pilot-scale IsaMill was investigated under certain industrially relevant pulp processing conditions. It was shown that a critical rheological state existed, below which the energy required to achieve an 80 per cent passing (ie P80) 20 ¦m product size was unaffected by shear rheology (ie apparent viscosity and shear yield stress) whilst above which the grinding energy required increased dramatically with increasing shear rheology. Colorado River sand (3.35 mm) showed higher energy efficiency than MT1 ceramic media (2.5 mm) for a target P80 of 20 ¦m in the solids concentration range of 38 to 55 per cent by weight, but the former experienced more than four times the wear rate of the later. The high wear rate of the Colorado River sand may be due to its much lower hardness. For feeds with an 80 per cent passing (ie F80) 70 to 110 ¦m, 2.0 or 2.5 mm MT1 media appeared to be more energy efficient than 3.5 mm MT1 media whilst a media-to-feed size ratio of 41 to 53 is applicable for ~50 ¦m (ie F80) feeds for efficient grinding.

Sep 13, 2010

By Mario Gerards

Maturing populations create an increase in the demand for the type of non-ferrous metals, which are used in the manufacturing of electronic products. The depletion of existing high-grade deposits, make it necessary to boost the efficiency of beneficiation processes and most importantly the yield (recovery rate of the value commodities). Increasing demand now makes it worthwhile to tap into resources, which were previously considered waste, for example secondary deposits such as tailing ponds and electronic waste. In order to achieve increased yields, increasingly finer, or even ultra-fine fractions have to be treated. In most cases, mineral characteristics such as size and density call for non-mechanical, high-capacity technologies to be used. There are various technologies that allow for the separation of fines and ultra-fines. This paper outlines the dynamics and efficiency of pneumatic flotation with emphasis on its application to the recovery of non-ferrous metals, from electronic waste. The conventional agitator flotation technology of an electronic waste processing plant was not able to recover ultra-fine non-ferrous metals satisfactorily, due to losses within the tailings. Because of its adaptability and its inherent suitability for non-ferrous metals, pneumatic flotation has been incorporated to the existing process achieving excellent results. The merit of the process lies in the state of the art technology, which does not utilize compressed air and any rotating parts, which results in a reduction in wear and energy requirement, both features of this innovative technology. In addition, the use of pneumatic flotation resulted in a significant reduction in residence time, when compared to mechanical flotation. Pneumatic technology is able to influence air bubble size in a flotation cell, by changing the speed and pressure of the slurry stream. Higher slurry velocities lead to the formation of smaller air bubbles, which is required for the successful flotation of ultra-fine particles. Conversely, decreasing the slurry velocity leads to bigger air bubbles. The optimum bubble size can be created according to different feed characteristics. This is important in case of feed fluctuations.

Sep 1, 2012

By A. M. Raichur, S. P. Vijayalakshmi

Fine coal dewatering is a major problem in the coal industry because the fines hold significant amounts of water and account for two-thirds of the product surface moisture. Present methods include flocculation using polymeric flocculants followed by filtration. But the kinetics of filtration are not favorable with such flocculants. In this study, a novel method involving flocculation with microorganisms for improving the settling rates and dewatering kinetics has been investigated. Two Gram-positive microorganisms, Mycobacterium phlei and Bacillus subtilis were used as bioflocculants. The effect of time, pH and ionic strength of the electrolyte used on adhesion and flocculation were studied. Adhesion tests showed that maximum adsorption occurred in 20 minutes in the pH range of 2-7 and the total amount of bacteria adsorbed was almost the same with both the flocculants. Nearly 98% of the coal flocculated with both the microorganisms in about 5 minutes. The filtration rate of fine coal almost doubled with addition of bacteria and electrolyte and maximum filtration occurred near the point-of-zero-charge of the bacteria.

Jan 1, 2003

By Timo Niitti, Antti Rinne, Janne Turunen, Sami Grönstrand

"This paper describes novel methods to enhance the flow dynamics of a mechanical flotation cell that can be applied to an existing installation. There are several reasons to alter the flow dynamics. The most common motivation is to improve the metallurgical performance of the cell, for instance, to improve the flotation recovery in particle size ranges that are normally difficult to recover.The flow dynamics enhancements are associated with the geometry of the rotor-stator area and additional features in the tank. The rotor-stator area has been modified to provide more concentrated turbulent kinetic energy or respectively more pumping efficiency with less attrition to the slurry. The tank can be fitted with various geometrical features, aimed to create a desired particle suspension profile. Depending on the targeted particle size range and application, badly selected or designed features can also have a negative effect on the metallurgical results.Besides new flotation equipment, the enhancements can be applied to existing flotation cell installations, thus enabling improvements in the mixing profiles, solids suspension, tank sanding and metallurgical performance of operating flotation circuits. Results of several tests carried out in existing plant operations are presented, concerning both flow variables and metallurgy.The common factor to above-mentioned enhancements is the use of modern computational fluid dynamics tools in the design phase, as well as the validation of the obtained CFD results in laboratory and production scale. Validated variables include mixing profile, power draw and local air hold-up.The paper demonstrates the effects of various flow geometries on the mixing profiles and metallurgical results of existing mechanical flotation cells. It shows how modern computational fluid dynamics analysis can be used to screen and select different approaches before carrying out costly trial-and-error experiments in plant scale."

Jan 1, 2006

By Majid Saghaeian, Rasool Hejazi, Parisa Ghobadi

"Pulp handling in mineral processing circuits is very complicated due to its inherent different specifications compared with other fluids such as water. Because any variation in ore property (e.g. size and abrasion index) and pulp specification (e.g. density, solid content and pressure) causes problem in this part and reduces the plant availability and throughput as a result. The most common issues in the hose system are sumps and pipe-line blockage, wearing of the pumps impeller and liner, and pipe tearing specially at elbows and bends. High pressure grinding roll (HPGR) is employed in a closed circuit with a screen as the first stage of grinding in Zarand Iranian Steel Complex (ZISCO). Wearing of the pump and pipeline transferring the screen undersize stream was the main problem from the plant commissioning. The plant couldn't achieve to the design capacity because of this defect. Although various practical solutions such as improving the pipes and impeller material quality and eliminating the elbows, as much as possible, were done in order to downtime minimizing, but damages didn’t vanish. Since HPGR discharge contains somewhat rough and abrasive particles, the mentioned modifications didn't effective very much. Therefore, converting of the wet screen to dry and using of belt conveyor to the screen undersize handling was employed based on the plant capability evaluation and experimental and pilot trial results. Although this reformation imposed another additional cost (for conveyor system), but it had significant advantages such as increasing the plant availability up to 96%, reducing the operating costs of the wear parts, providing more stable operating condition for the HPGR performance and decreasing abrasion rate of the HPGR rolls liner because of the lower moist feed."

Jan 1, 2014

By W. Mirihange

Considerable mechanical property decline is always observed at the Heat Affected Zone (HAZ) of AA 5083, This occurred as a result of the thermal effect on the hardened material, It is highly beneficial to have a prospective metallurgical procedure to overcome this decline of properties. Therefore, experiments were set to evaluate the effectiveness of a quick post weld heat treatment process with the AA 5083, Optimum results were obtained at 473K and with approximately 600 seconds time, The interplay between the time, temperature and mechanical properties are possible to explain by the Lifshitz-Slyozov-Wagner (LSW) theory and by considering the composition of the material.

Jan 1, 2005

By G. Mahieu, L. Perez-Barnuevo, P. Barbeau-Verreault, O. Gravel

Raglan mine is a copper-nickel mining operation, part of Glencore, and located at the northern border of the province of Quebec, in Canada. The concentrator is fed by a mix of four different underground mines, thus it must accommodate periodic variations in run-of-mine mineralisation and composition. In order to concentrate nickel and copper sulphides, the flotation circuit also includes a regrind step at the cleaner stage. This leads to challenges in maintaining optimal recoveries and in diagnosing potential disturbances. This paper advocates for the combination of automated mineralogical analyses, reconciliated mass balances from diagnostic sampling campaigns and innovative multidimensional visualisation to assist in tackling these challenges. Through this approach, it has been possible to quantify the regrind step influence on the liberation of valuable mineral particles in the final flotation concentrate. The methodology also allows for virtual sieving of the products in order to track unit loss in the flotation process. The proposed analysis scheme presents an opportunity to easily diagnose flotation circuit performance based on mineralogical information. Keywords: Flotation diagnostics, multi-dimensional visualisation, mineral liberation analysis

Jan 1, 2020

By N. Mulaudzi, S. Archer

"Direct cobalt electrowinning was developed to produce unrefined cobalt metal as an alternative to producing a mixed hydroxide product (MHP), which is the common cobalt intermediate product in African Copperbelt operations. MHP is preferred due to the high cost of producing high-grade cobalt cathode. Direct cobalt electrowinning test work results have shown that the purity of the unrefined cobalt metal produced from a low-grade copper solvent-extraction raffinate stream was ~97% (w/w), which is significantly higher than the MHP purity (up to 45% (w/w) Co). The current study investigates whether direct cobalt electrowinning would increase the profitability of African Copperbelt copper–cobalt oxide operations, when compared with the MHP flowsheet. An order-of-magnitude economic evaluation was conducted using dynamic differential discounted cash flow models on a “bare-bones” basis, using a 10% discount rate. Three variables were evaluated, namely; reagent utilisation extents, precipitation extents and filter-press wash efficiency. Multivariable risk analysis was also conducted using the Monte Carlo simulation technique to understand the impact of key project drivers. The financial study results indicated that a direct cobalt electrowinning flowsheet could potentially result in a higher mean net present value (NPV) compared with the MHP flowsheet. Cobalt recovery was found to have the most significant impact on the NPV. In addition to increased profitability of African Copperbelt operations, direct cobalt electrowinning has the potential to reduce the environmental footprint of operations as well as addressing the Democratic Republic of Congo government’s drive towards export of higher value metal products instead of low-grade intermediates."

Jan 1, 2015

By Erik S. Bernard

"INTRODUCTION Reinforced Concrete (RC) members such as beams and columns are usually required to include stirrups to maintain post-crack shear capacity, and some may include stirrups that act as confining steel to maintain structural integrity during a seismic event (eg. NZS3101, ACI 318). To achieve seismic resistance, the spacing between stirrups is usually required to be much closer than is ordinarily prescribed for non-seismically resistant structures. Each stirrup must also fully enclose the longitudinal reinforcement with long end-hooks to ensure that the contained concrete is sufficiently restrained during reverse-cycle loading and the reinforcing bars are confined to prevent buckling. The close spacing between stirrups can lead to congestion of reinforcement within members, and can also be expensive to construct due to the labor-intensive nature of steel fixing. A possible alternative to closely-spaced stirrups is the use of fibers to provide restraint and confinement during a seismic event, although this is not yet recognized in codes. Previous experimental research (Jiuru et al, 1992; Filiatrault et al, 1995; Bayasi and Gebman, 2002; Billington and Yoon, 2004) has demonstrated that steel fibers are relatively effective in this application, but suffer the problem that their stiffness can make them difficult to consolidate within an already tight reinforcement cage, especially if they are relatively long (>40 mm). Moreover, steel fiber reinforced concrete (SFRC) is subject to embrittlement with the passage of time, leading to a steady loss of ductility beyond crack widths of 1-2 mm (Banthia and Trottier, 1994; Bernard and Hanke, 2002; Bernard, 2014; Bjontegaard et al, 2014; Bernard, 2015). Since cracks during seismic events are generally much larger than this, embrittlement may present a challenge to the long-term efficacy of SFRC during seismic events. An alternative to steel fibers for enhancement of seismic resistance is the use of macro-synthetic fibers. This type of fiber usually offers the advantage of being flexible so that consolidation is easier between reinforcing bars. They also offer superior durability compared to steel fibers and can reduce flexural crack widths (Bernard, 2016). Several recent small-scale investigations (Ma et al, 2012; Carnovale, 2012; Osorio et al, 2014) have indicated that macro-synthetic fibers can produce meaningful improvements in seismic resistance if used in combination with conventional steel reinforcing bars. The present investigation has extended research in this field by examining whether embossed macro-synthetic fibers can be used to increase the spacing between stirrups while maintaining seismic resistance."

Jan 1, 2016

By K. Esmeli, A. Ozkan

The enhancement of shear flocculation of a galena suspension by ultrasound in the presence of sodium isopropyl xanthate and potassium ethyl xanthate was investigated by studying the effects of surfactant concentration on flocculation efficiency, contact angle, and zeta potential. Strong flocculation was obtained under alkaline conditions (pH > 9) and a high ultrasound power level (150 W). Although the ultrasonic treatment increased the negativity of the zeta potential of the galena particles, this did not decrease the flocculation efficiency, thereby showing that the hydrophobic interactions between the particle were stronger than the electrical double-layer repulsion resulting from xanthate adsorption. The findings indicate that the ultrasonic treatment promoted the adsorption of surfactants onto the mineral surfaces.

Jul 5, 2023

Ilmenite (FeTiO3) is the primary global source of titaniumdioxide which is commonly used as a whitening agent in papermanufacture and a pigment in paint manufacture. In WesternAustralia ilmenite is processed by the Becher process (Becher,1963, Becher et al, 1965) where ilmenite particles are feddirectly into the furnace with coal and heated to form acomposite of elemental iron and TiO2. The mechanism of thecarbothermic reduction of ilmenite has been extensively studied(Grey, Jones and Reid, 1973; Jones, 1973; Grey and Reid, 1974;El-Tawil, Morse and Francis, 1993) and is reasonably wellaccepted.

Jan 1, 1997

By Rahul Singh, Sunil Kumar, Erich Dohm, Maoming Fan

The rapid increase of P2O5 use for food, biofuels, and LiFePO4 batteries etc. accelerates the depletion rate of phosphate reserves. In the phosphate industry, a significant amount of ultrafine phosphate has been discarded as slime tailing due to processing difficulties. The major problems associated with ultrafine particle flotation are low probability of bubble-particle collision and adhesion, and high probability of unwanted gangue particles entrainment. This paper investigates the enhanced ultrafine phosphate particle flotation probability by Eriez’ CavTube generated fine/ ultrafine bubbles and the improved flotation selectivity by wash water etc. Ultrafine phosphates with various gangue minerals of quartz/muscovite, calcite/dolomite, and iron/ silica were studied. Microscopic observations show that the entrained ultrafine gangue minerals can be minimized by using wash water in column flotation. A three-factor threelevel central composite experimental design was conducted for ultrafine phosphate flotation. For comparison, both benchtop mechanical cell and laboratory column flotation tests were performed at their optimal respective flotation conditions. Using Eriez CavTube column flotation technology for Industries Chimiques du Senegal (ICS) ultrafine slime tailings, a concentrate grade of 36.0% P2O5, was achieved at 89.4% P2O5 recovery. Improved flotation performance was achieved using column flotation as compared to benchtop mechanical cell flotation. The benefits of Eriez’ column flotation technology realized from this study include the high P2O5 grade and recovery.

Feb 1, 2024

By E. C. Grunsky, B. W. Smee

"The development of low-cost, rapid multielement analytical techniques has generated large geochemical databases in many exploration programs. When a sampling program consists of several thousand samples, the resulting data matrix is enormous and effective interpretation using all of the elements individually becomes burdensome. However, the application of multivariate statistical techniques can extract geochemical patterns related to the underlying geology, weathering, alteration and mineralization. Imaging the results over topography enhances the interpretation of these patterns. Examples of this approach are shown from mineral exploration programs in Canada and Mexico. IntroductionThere has been a dramatic change in the effectiveness of using geochemical survey data in the past ten years. This is mostly due to:• lower cost of geochemical analysis;• lower detection limits;• new methods of multi-element analysis;• a marked decrease in the turnaround time;and• the intelligent use of computers in presenting geochemical data.Geochemical surveys commonly have two objectives: locating abnormal concentrations of ore-forming or pathfinder elements and characterizing the underlying host lithologies. Geochemical surveys often employ a wide range of sample media that may typically include whole rock lithogeochemistry, various soil horizons, till and basal till sampling, stream sediments, lake sediments, and various forms of weathered regolith. Each of these sample media will exhibit a unique element distribution that must be factored in the interpretation phase of a geochemical survey study. This can become a cumbersome process if univariate data handling methods are employed."

Jan 1, 2003

By Carl D. Ealy

Work is being done to web enable the FHWA Deep foundation Load Test Database. Originally designed to function in the client server environment via modem, access to the database both for data maintenance and user queries has been upgraded to take advantage of the Internet and current and future information technology innovations. The DFLTD will be part of an Automated Geotechnical Information and Design Aid System, the object of which is to assist those involved with managing and maintaining our nation's highway infrastructure. The system will also meet the needs of university and other research institutions for a statistically meaningful source of high quality data for the development and verification of improved analysis methods, in-situ testing methods, etc.

Jan 1, 2002

By R. Hardy

LaModel is a boundary-element program capable of calculating the displacements and stresses in thin bedded deposits such as coal, salt, or potash. Throughout its history, it has continually been upgraded in order to expand analysis detail and stay current with the latest advances in computer technology. The most recent enhancements include: an increased grid size capability, a significant increase in the speed of the automatic grid generation using an AutoCAD visual interface, implementation of subsidence-based horizontal strain calculations for multiple-seam situations, and pillar safety factor calculations. To demonstrate these new features, a multiple-seam case study of a longwall panel undermining an active set of main entries is presented.

Jan 1, 2006

By D. W. Duncan, A. Bruynesteyn

Acidic mine waters containing solubilized uranium have been observed in many uranium and some copper mines. In every case, the sulphide and iron oxidizing bacterium, Thiobacillus ferrooxidans, was also present. This bacterium provides the ferric iron required to oxidize u+4, both by oxidizing available ferrous iron and by oxidizing pyrite or other available sulphide minerals. This latter oxidation also provides sulphuric acid and may increase extraction by exposing more uranium. The enhancement of bacterial activity and the results of some field tests are discussed.

Jan 1, 1971

By N. F. Faria, F. Monteiro, B. O. Taiwo, A. Palangio, D. Fonseca, J. Manuel

This paper explores the interpretation and application of Particle Size Distribution (PSD) curves in optimizing blast design and enhancing downstream operations. The study looks at important factors like D50, D80, fines content, and the uniformity index to show how PSD curves affect the efficiency of different mining processes. The first section of this paper discusses the impact of blast design parameters such as burden, spacing, and explosive energy on the resulting PSD, highlighting their significance for downstream operations like crushing, and material handling. The section concluded that proper alignment of PSD with downstream size classes, like primary crusher inlet sizes, among others, can result in substantial cost reductions and productivity gains. The second part presents a case study from dstgroup aggregate quarry in northern Portugal, utilizing fragmentation analysis to evaluate blast performance and promote continuous improvement. The section specifically examines the effect of stemming material size on blast fragmentation, comparing the impact of using aggregate with a D80 of 12.32 mm versus a D80 of 21.94 mm. The D80 value represents the particle size at which 80% of the stemming material is smaller, making it a critical parameter in evaluating the stemming's effectiveness in confining explosive energy. The drilling and blasting designs were meticulously planned using O-PitDev and O-PitSurface software to ensure accuracy in blast parameters and layout. To assess the fragmentation resulting from these different stemming material sizes, the study utilized WipFrag 4 software. The software allowed for precise measurement of particle size distribution after the blast, as well as a detailed comparison between the two stemming sizes. Results from the analysis indicate that using bigger-sized stemming material led to improved blast confinement, which in turn resulted in better fragmentation compared to the smaller-sized material. The reviewed blast videos demonstrate that the larger aggregate effectively contained the explosive energy within the borehole, thereby reducing premature venting and directing more energy towards breaking the surrounding rock mass. This led to finer fragmentation with D80 = 1240.1 mm, D50 = 685.9, and Dmax = 2580 mm, which is advantageous in downstream operations like crushing and milling. On the other hand, the smaller stemming material (D80 = 21.94 mm) did not confine the explosive energy as efficiently, resulting in coarser fragmentation (D80 = 1327.37 mm, D50 = 750.76, and Dmax = 2370 mm) and reduced overall blast efficiency. These findings highlight the significant influence of stemming material size on blast performance and fragmentation outcomes. 

May 4, 2025

By O. Cermak, F. Degonld

Although 3D fragmentation analysis is a relatively new technology, it is rapidly gaining popularity among mid-sized mining operations and quarries. The results provided by this technology are adequate, but like other picture analytic methods, it mainly focuses on the surface of the muck-pile. These results are incomplete or incorrect as falling rocks and boulders from the stemming zone are constantly influencing the muck-pile surface. This obstacle can be overcome by collecting a large amount of data and analyzing it over time. However, this is not always an ideal solution, as companies prefer to receive results immediately after the blast to make resource allocation decisions, production planning, and plan for the next blast. Over the past year and a half, we have conducted a study on various muck-pile fragmentation analysis results from 3D modeling in Central Europe, comparing the results between the surface and inside of the muck-pile. By analyzing multiple shot results, we were able to determine the percentage difference between the top and inside fragment sizes and calculate the range of correction coefficients for different rock types. Applying these coefficients to a single muck-pile surface fragmentation analysis allows us to determine the actual fragment sizes inside the muck-pile. This enables us to make informed decisions regarding loading, hauling, material mixing, and proper crushing or milling. By overcoming the limitation of surface-based analysis, we can now accurately evaluate blast performance, which leads to better processes optimization and increase efficiency. As 3D fragmentation analysis becomes more widespread, we expect to see even more benefits for the mining and quarrying industry

Jan 1, 2024

By G. Van Weert, O. Wang

"As part of a general review of processing double refractory ores through Barrick Gold’s autoclaves / CIL facility at Goldstrike, blanking the preg robbing native carbon in the process solids with hard (m.p. 55 - 60 º C) paraffin wax was studied, specifically in relation to pressure oxidation. The improvement in CIL gold extraction was found to depend on the wax addition point, with hot, dry contacting of the ore in the presence of wax - prior to POX - yielding the largest, most consistent improvement in gold extraction.INTRODUCTIONDouble refractory gold ores present two barriers to gold extraction. The first is generally a physical barrier: the gold is encapsulated or dispersed in a matrix which is impervious to the gold lixiviant. In most cases, including Carlin Trend, NV, deposits, this involves a pyrite or arsenopyrite matrix, although minor gold may also be dispersed in competent silica (quartz) or iron oxides. Gold can be rendered accessible by oxidative destruction of such sulphide matrices. At Goldstrike this is practised by roasting and autoclave oxidation (Thomas, 1994). It is the latter route to which this paper refers.The second barrier refers to the actual gold recovery. Gold may be extracted from the pressure oxidized solids with cyanide, forming a gold cyanide complex. One characteristic of this complex is that it adsorbs on graphitic surfaces. This is taken advantage of on a large scale by gold millers, who use manmade activated, i.e., graphitized carbon particles to remove the gold from solution during leaching: the Carbon-In-Leach, or CIL process. The gold cyanide bearing carbon particles are larger than the ore solids and can be separately recovered by screening. Goldstrike uses the CIL technology to recover gold. The second barrier becomes problematic when the ore naturally contains graphitic orcarbonaceous material, hereafter referred to as native carbon (versus the man-made activated carbon). This material may also adsorb gold cyanide in competition with the activated carbon. Since the native carbon is ground with the ore it is much smaller than the activated carbon. Any competition between the two carbons may produce a loss of gold from the pregnant leach liquor to the process tailings. This phenomenon is called “preg robbing”. Goldstrike ores come from the Carlin trend, which is known for the occurrence of native carbon (Adams & Burger, 1998; Schmitz, Duyvesteyn, Johnson, Enloe & McMullen, 2001; Tretbar, 2004)."

Jan 1, 2012