Search Documents

By Walter W. Focke, Zola Kwela, Willem van der Merwe, Arao Manhique

"The advantages of the de Wet zirconium process for the manufacture of AZST from zircon sand include: Fewer processing steps; removal of radioactivity by leaching rather than by precipitation; alkali reagent recovery in the form of a sodium silicate solution that is essentially free of radioactivity; fewer waste streams and less waste to discard.Here we report on our efforts at maximizing the alkali recovery in the sodium silicate product stream: Milled zircon, d50 ˜ 9 µm, was fused with caustic soda pearls in open reaction vessels at temperatures between 650°C and 850°C. Fusion times of 1, 2, 4, 24 and 336 hours (with intermittent milling) were investigated.For prolonged fusion times the fusion products approached equilibrium compositions. However, the phases Na2ZrSiO5 and Na4Zr2Si3O12 observed at 850°C did not form at 650°C. Since they do not hydrolyse in water they limit the recovery of alkali. For 2-hour fusions a broad spectrum of non-equilibrium reaction-products, that are rich in sodium silicates, is formed. Optimum process conditions, with respect to NaOH recovery, correspond to a 2:1 stoichiometry and fusing for 2-hour at 850°C. At these conditions the alkali recovery is ca. 78% and the zirconia yield is ca. 57%."

Jan 1, 2003

By P. T. Nguyen

The analysis of particle capture by an air bubble by two separate phenomena of collision and attachment has been difficult due to the uncertainty in determining when and where the two phenomena begin and terminate at the bubble surface, respectively. Therefore, a unified analysis was developed by solving the motion equation for the particle capture as one process. No distinction be-tween collision and attachment was made. Both the liquid film described by the micro-hydrodynamic resistance functions and the surface forces were included in the new model. The model predictions with the DLVO (Derjaguin-Landau-Verwey-Overbeek) surface forces did not agree with the experimental trajectories of latex and silica spheres. The model predictions with the inclusion of the non-DLVO attractive surface force agreed with the experimental data. The non-DLVO force reduced the repulsion between the non-attaching particle and the bubble, and significantly increased the attraction between the attaching particle and the bubble. The unified analysis is capable of predicting particle capture without the use of the induction time. The unified analysis also shows significant influence of the short-range hydrodynamic interactions on the collision which has been previously ignored.

Jan 1, 2014

By Richard K. Komarek, Antonios Zavaliangos, Roman T. Dec

Roller presses are widely utilized in mineral industry for fine milling and compaction. Recently used models are providing only limited information about the material behavior in the roll bite. This paper introduces a 2-dimensional finite element model for analyzing those two basic processes. A pressure-dependent plasticity model with linear elasticity was used as a constitutive model for the particular material. Coulomb friction law was used to describe phenomena on the material/roller surface interface. Two basic roller geometries are considered: smooth (used for milling and compaction) and with cavities, commonly used for briquetting. Comparison of the results obtained from the model with experiment indicates that they are following the general trends of experimental values recorded with laboratory roller presses. It is shown, that the model is suitable to analyze material behavior through all stages of the rolling process: feeding zone, nip region, neutral plane and release section. It can predict pressure, shear stress, slip of the material over the roller surface, material velocity field, failure of the compact during release and the course of other important process variables. This allows the process engineer to gain an improved understanding, leading to a better process and equipment design and control.

Jan 1, 2003

By Caroline Olsen

The objective of this study was to achieve Zn activation control in an industrial Cu flotation circuit treating a complex Cu-Pb-Zn-Au-Ag sulphide ore at the LaRonde Division of Agnico-Eagle Mines Limited (Québec, Canada). In order to provide process relevant results, a laboratory testing methodology was developed to duplicate plant operational parameters. The action mechanisms of ZnSO4, TETA (triethylenetetramine) and NaHSO3used as depressants in order to limit ZnS activation were examined. Laboratory flotation tests with and without Cu2+ions (the ions responsible for ZnSactivation)were performed. All depressants showed equal or better flotation performance relative to baseline tests. For example, in the presence of Cu2+, the Zn recovery reduced from 32.2% to 17.5% in the copper rougher concentrate after adding ZnSO4(with a dosage of 50 g/t). TOF-SIMS analyses confirmed the relative decrease in surface Cu in the samples treated with ZnSO4and TETA. For the case with TETA, this indirectlyindicates the formation of a copper TETA complex in solution, thereby reducing available Cu for transfer. The TOF-SIMS and XPS data indicate that the addition of ZnSO4and NaHSO3resulted in an increase in surface metal hydroxides and SO3 on ZnS grains inhibiting both Cu and collector attachment.ZnSO4 was subsequently tested with success in the LaRonde concentrator. The addition of 40 g/t ZnSO4 at the conditioning (or cleaning) stage led to a 2% reduction in Zn content in the Cu and PM concentrate. Recently, improvements towards the control of copper sulfate addition in the downstream Zn flotation circuit resulted in the reduction of Cu2+ in process water. As this ?cleaner? water is recycled for Cu and PM flotation, the addition of ZnSO4 is now used only as required.

Sep 1, 2012

By ?. Kostadinov

Theoretical study of the behavior of grinding medium presented by TBP (tetraballpebs) was conducted by computer modeling of the position of grinding medium at different speeds and different drum fill factor. For the modeling were used known equations of motion derived for the ball load. Algorithm was created by SolidWorks software, allowing determination of center of gravity of the load in the mill and energy for grinding. These results were compared with similar results obtained by ball grinding medium. The capabilities of both grinding media are demonstrated by graphics under the following conditions: 1.Equal dimensions of the experimental model of the mill; 2.Equal mass of the grinding loads; 3.Comparison of grinding bodies with the same number, mass and density. Keywords: simulation, behavior, grinding medium, balls, tetraballpebs

Sep 1, 2012

By V. I. Ryaboy, V. P. Kretov, S. E. Levkovets, I. V. Ryaboy

"Using the flotation of a gold-containing ore from a deposit in Kazakhstan as an example, we have conducted a study designing dialkyldithiophosphate collectors, in addition to xanthates, for the flotation of hard-toenrich gold-containing ores to increase gold recovery.The increased flotational ability of the dialkyldithiophosphates was determined by their surface activity at the liquid-gas interface and their hydrophobic ability, the changing of which allowed us to control the collecting power and selective ability of the dialkyldithiophosphates. The studying of the surface activity of aqueous solutions was done using the stalagmometric method and the evaluation of the hydrophobic ability was done in percentage of the relative value of the zinc ion binding in the solution in standard conditions.Studying the influence of surface activity in our designed dialkyldithiophosphate collectors, with their close hydrophobic ability, showed that even with a small increase in surface activity in their aqueous solutions, gold recovery increased by 2-3% in laboratory conditions. Using this enabled us first to apply the dialkyldithiophosphate collector BTF-161 at a plant in Kazakhstan, and then to introduce the specially designed dialkyldithiophosphate collector BTF-163, which has been in use since. Further study of dialkyldithiophosphate activity based on this concept showed us that for the range of short-chain dialkyldithiophosphate collectors with a chain length of up to ?5, the effectiveness of ore processing became lower the higher their surface activity got, with relatively small differences in their hydrophobic ability. For long-chain dialkyldithiophosphates with a chain length of over ?5, ore processing effectiveness increased with the increase of their hydrophobic ability.Based on these studies, a new effective dialkyldithiophosphate collector, BTF-1517, was designed. Using this reagent in continuous gold recovery tests allowed us to reach an 84.21% recovery with a concentrate quality of 15.66 g/t, while using the applied collector BTF-163 in similar conditions led to the recovery of 82.87% with a concentrate quality of 15.65%. As a result, the Kazakhstan processing plant decided to conduct industrial tests with the BTF-1517 collector."

Jan 1, 2018

With rapid development of Chinese iron & steel industry, it is of significance to intensify the utilization of domestic complex iron ore resources. The tin, zinc-bearing complex iron ores are typical polymetallic and intractable ones because of the complicated intergrowth relations among the multi-elements, and they have not been efficiently utilized, presently. Based on the thermodynamic analysis of iron, tin and zinc oxides reduction by CO, the technological conditions of tin and zinc recovery and pellets preparation from tin, zinc-containing iron concentrates by selective reduction roasting are studied and a new comprehensive utilization process is developed in this investigation. Under the optimum conditions, the tin and zinc volatilization is respectively in excess of 80% and 70%, and the compression strength of roasted pellet is more than 2500 N/P. Moreover, the residual tin and zinc in the finished pellets is below 0.06%. The finished pellets can be used as high quality iron-making burdens for large-scale blast furnace.

Jan 1, 2014

By M. Vasudevan

In sulfide mineral flotation, the efficiency of separation is severely affected by the presence of non-sulfide gangue minerals, such as altered silicates and serpentines. Although this is an acknowledged fact, most studies in mineral processing have focused on value minerals, and studies on gangue minerals affecting flotation performance are quite rare, and even those that do are limited in scope. This paper is an attempt to highlight the use of grade-recovery curves in studying the effects of gangue minerals. Specifically, we present the flotation response of problematic Ni and Cu ores which show highly uncharacteristic but similar grade-recovery characteristics despite the very different mineralogy and try to understand this behavior and the underlying mechanism. Finally, we propose chemical modifiers as solutions to overcome the detrimental effects of these gangue minerals and improve efficiency of flotation performance. Keywords: Non-sulfide gangue, grade-recovery, sulfide minerals, serpentines, talc, shale, slimes, spiking, modifiers

Sep 1, 2012

By T. Onnela

Metso Mining and Construction Technology is one of the world leading suppliers of equipment and knowhow for aggregate and mining industry. The processing capacity, ease of use, and reliability are featuring Metso?s state-of-the-art cone crushers. Crushing is a complicated process affected by numerous parameters. The optimized crusher kinematics is one of the key factors for high performance. To fine tune the crusher kinematics and crushing chambers one needs to know how different parameters affect. The main cone crusher parameters are head angle and diameter, gyration point, shaft speed, eccentricity, stroke length, and inclination and design of crushing cavity. Selecting the correct combination of these parameters is a multi task optimization challenge. In practical life there are more factors to achieve satisfactory performance, like crushing chamber wear and profile management keeping the performance constant over the whole wear life, control of crusher frame loads, ability to handle disturbances in feed material etc. The extensive research work has lead to a solution that enables the highest possible material density in the crushing chamber. This design provides maximum reduction of processed material and thus increased share of 0/10 mm product. High density in the cavity offers further benefits, such as high throughput capacity, less material contact with wear parts and even load on crusher frame and bearings. The challenge that an equipment supplier involved with high density crushing is how to design a robust cone crusher, which is able to cope with the high crushing forces due to high resistance from the crushed material against the gyration movement. Metso has launched a new series of cone crushers HP3, HP4 and HP5 that are all able to work in high density crushing applications The test results and evidence from over 300 installations clearly prove the effectiveness and importance of high density crushing. Keywords: cone crusher, crushing chamber, crusher kinematics

Sep 1, 2012

By Anatoly Usov

The detailed researches of a broad circle of problems connected to development of the electric pulse method destruction materials (EPD) of a various nature and area of use are executed. The electrical and power characteristics of process, physical fundamentals of selectivity of a breakdown and destruction are investigated, the device of electric pulse (EP) disaggregation and ?omminution of materials are developed, the technological efficiency of disintegration of industrial minerals estimated. Due to its technological advantages, the EPD could have a great potential in minerals mining and processing technologies. Low specific power and weight-and-size characteristics of EP installations based on the existing out-of-date electric equipment are at present the main reasons hindering the effective application of the method in mining and processing technologies. The main studies carried out in Russia in the field of electro technical equipment development for EP technology are presented. The pulse capacitors service life affects greatly the economic efficiency of the process because the capacitors cost accounts for a considerable, and in some cases, the most part of the operational costs. Determined and analyzed in the study are the ways which could significantly improve the specific characteristics of high voltage pulse generators (VPG), such as to use of high-frequency voltage transformation to develop a charger; to use pulse transformers (PT) with iron magnetic circuits made of materials characterized by high magnetic permeability, in pulse generating circuits; as well as to use special ways which could produce voltage pulses with a short front. The experiments have proved the possibility to adapt these technological solutions for EP devices, and a further strategy of the EP technologies development and application has been suggested on this base. Keywords: electric pulse disaggregation, electric pulse installation, charger battery, high voltage pulse generator

Sep 1, 2012

By E. Lee

With the growing environmental concern, as well as the increasing concern for sustainability of natural resources, the need of improving the performance and efficiency of today?s crushers is be-coming increasingly important. However, the development of existing crushers requires an extended basic knowledge about the crushing process itself. The aim of the present study is therefore to find the theoretically optimal way of crushing compressively under given circumstances. In the present paper, optimizations have been performed with the purposes of maximizing product yield and minimizing energy consumption. Performed on an ideal crushing stage free from physical limitations, the optimization results have been obtained in terms of number of compressions and their corresponding compression ratios. Complementing earlier results on interparticle breakage (Hulthén et al., 2006), presented here are thus the results from corresponding optimizations with single particle breakage and interparticle breakage. It is concluded that modern crushers are not working optimally in any of the studied aspects for gneiss. In fact, optimizations indicate that a lowering of the compression ratio, as well as a shortening of the crushing sequence, are theoretically favourable. The outcome of this study is therefore highly promising as to giving new ideas and suggestions of improvements in existing crushers, and can even result in redesign of compression crushers.

Jan 1, 2014

By Vladimir Grujic, Milorad Grujic, Ivan Djurkovic, William Riggs

"In this paper, we have studied efficiency of different chemical structures of frothers on copper ore Majdanpek. The goal was to find out the best suitable chemical structure, which allows good bubble formation, size and interaction, forces on mineral surface. The particle size distribution was between 10 and 100 µm. Our research was oriented on three groups of frothers: alcohols, poly glycol ethers and poly glycols.We also used different collectors based on similar or less similar alcohol group in chemical structure as in frothers, expecting higher stability of air bubble in contact with copper minerals. Minimal potential energy was determinate in the hydrodynamic conditions for different particle sizes. It was founded that it is proportional to air bubble volume on 0.88° and contra proportional to the kinematics viscosity of pulp.Analyzing the results, we find out that poly glycols decreased surface tension the most. Also, it increased amount of coarser particles to the top of pulp. This group is less sensitive to pH value compared to the others too.From the class of frothers with higher efficiency in flotation process, we pointed OrePrep X-75 and X-133 as frothers with expressive frothing characteristic in the alkaline and low alkaline circuits. Special suitability of those frothers for floating copper minerals makes their hetero polar molecules with more than one hydrocarbon radicals connected to one polar group. However, those froters have no nether high or low solubility and are non-ionization. Those froters indicate some small collecting characteristics too. In dispersion systems, those two frothers make two types of bubbles: smaller di < 1.2 mm; V = 0.0008 cm3 and middle di < 16 mm; V = 2.2 cm3 (di – Bubble diameter, V – Bubble volume). Transportation speed of air bubbles it is possible to calculate based on equation: ?s = 27.5 * di 1.4 cm/s. Distribution of free surfaces for chalcopyrite is average 80% in the class –74µm +10µm and 20% in the class -10µm. If we get all class sizes, we find that recovery (R) depends from the time by equation:"

Jan 1, 2003

Compressive crushing is applied in crushing plants used in mineral, mining, and aggregate in-dustry for fragmentation of rock materials. From a global perspective, the total volume of crushed rock is massive and corresponds to several billion tons. All societies have to fulfil a daily need of rock material products and its derivatives such as metals and minerals. The business for metals and minerals is global while the aggregates production is mainly a local industry. This paper presents a broader overview of optimisation with respect to product yield, use of natural resources and energy consumption. High level plant optimisation of compressive crushing plants is covered as well as real-time optimisation and optimisation of single production units. Each of three optimisation areas are presented and a discussion around how these three areas will be able to communicate with each other is given. Several models, e.g. process performance behav-iour with respect to wear, as well as fitness functions can be common for all three areas. Well-defined and representative fitness functions are of the utmost importance in order to assure that the optimisation results reflect the production targets. It is concluded that great benefits can be achieved by linking together different areas of optimisa-tion. The future for compressing crushing lies, without a doubt, in tailored processes both with re-spect to machines optimised for each customer, plant parameters and operation routines as well as an on-line adjustment, which takes into account current information and control possibilities.

Jan 1, 2014

By Peter Rubie, John Russell

"For many years, optimizing the relining of hard rock grinding mills has been complicated by the large number of variables, the low frequency of relines (resulting in a slow relining experience assessment and review cycle), and the potential economic risk associated with some changes to relining equipment or work practices. A technology that has its roots in the automotive manufacturing industry is being utilized to construct a statistically accurate, configurable, and dynamic simulation of the mill relining process. The detailed effects of various mechanization and process changes can be rapidly modeled and compared. Application of this discrete event analysis technology in the mineral processing industry delivers a range of benefits to many industry stakeholders. For example, mill maintenance managers and contract relining companies can embrace the opportunity to benchmark current performance against world’s best practice. Concentrator plant designers and mill manufacturers, informed by this technology’s scientifically defensible data set, can accurately plan for maximum plant efficiency when designing an infrastructure and equipment combination that truly maximizes annual plant availability. Time-motion studies of grinding mill reline data at mills of varying dimensions, liner configurations, mill relining equipment, operator and tool variables, have been integral to the development of this dynamic simulation tool and its associated data set. Access to mill relining simulation technology will enable mineral processing industry professionals to take advantage of an increased capacity to develop mineral concentrator performance optimisation initiatives within the given financial, physical and technological constraints of the individual concentrator plant."

Jan 1, 2014

The present work aimed to verify the possibility of using froth flotation to separate the constitu-ent minerals of an alaskite rock waste having as the main minerals quartz and feldspar. Three stages of tests have been carried through. In the first stage was used an amphoteric collector for flotation, diaminedioleate, at pH 2 and, after analyzing the final products, a very poor selectivity was ob-served. During the second stage of tests, different collectors that are normally used for feldspar flo-tation as amines, sulfonates and soybean oil were employed. Once more, it was not possible to reach reasonably selective results. In the third stage, a salt of etheramine was used for collecting the feldspar and hydrofluoric acid was utilized for controlling the pH. A simulation of three flotation steps (rougher, cleaner and recleaner) permitted to attain a good selectivity in the flotation process. A concentrate with a silica content of 96.6% was obtained. The low concentrate content of alkalis totalizing 0.52% and a feldspar recovery of 96.5% to the tailings, represented by the high content of Na, K, and Al2O3 in the tailings, demonstrated that the concentrate was made up mostly of quartz.

Jan 1, 2014

By Rodrigo Ladrón, Gonzalo Rund, Gonzalo Montes-Atenas

Modelling ore comminution processes have been a highly fruitful scientific topic. The models have been proved to be largely dependent on the selected comminution technology. Among them, semiautogenous (SAG) mill technology has been widespread implemented in plants but no complete understanding of the mechanisms involved in the process is available. Tracing back in history, SAG models considered only some granulometric cut sizes such as the t10 and attempted to relate these parameters with the energy involved in the impact between particles. Other models evolved towards incorporating physical properties of the material being ground and the mill operating conditions. These material properties have assisted to obtain a more comprehensive description of the breakage mechanisms; nevertheless, the gathering of all known parameters still have not delivered a model able to describe the process. This may be a result of emergent properties or phenomena originated from interactions inside the process which are hard to predict turning the size reduction operation into a complex system. Cellular automata provide one answer to describe complex systems by following simple rules. The aim of this research work is to establish the preliminary rules of thumbs enabling the implementation of cellular automata able to predict the particle size distribution at the exit of a SAG mill. The methodology involved developing a simple 2D model of a ROM sample. Plant data from two mine sites were correctly described by a set of three rules: two rules describing the particle reduction and one rule associated with material transport across the mill. Further analysis related with expanding this model to 3D and its impact on the liberation of valuable material is also discussed.

Jan 1, 2014

By A. P. Das

Biomining technology focuses on achieving effective recovery of precious metals from their ores and concentrates using microbial techniques. Biomineral extraction of manganese from low-grade manganese ores using bioleaching was investigated in this paper. Five fungal strains were selected for screening of bioleaching of low-grade Mn ore collected from Sanindpur iron and manganese mines, Koira Sector, Keonjhar, Odisha, India. A fungal strain, Aspergillus niger was selected as it showed high efficiency of manganese leaching. Growth of the fungus was determined in terms of sbiomass dry weight and the final pH was monitored. The selected fungal strain was used for bioleaching of low-grade manganese ore of -150µm size particle size at 2 % pulp density with 2g/100ml sucrose concentration for duration of 20 days. About 60 % of Mn bioleaching was obtained with A. niger. The fungal species proved to be useful for manganese biomining.

Sep 1, 2012

By D. M. R. Sekhar

Certain types of rock phosphates of sedimentary origin are applied to acidic soils as fertilizer directly. Recent research in India showed that high grade rock phosphate in fine size when used as a mix with organic manure works as efficiently as di-ammonium phosphate (DAP) in alkaline soils. The rock phosphate from Jhamarkotra, India (sedimentary origin), Egypt (sedimentary origin) and South Africa (volcanic origin) showed good agronomic efficiency when used as Phosphate Rich Organic Manure (PROM). Further PROM shows equal residual effect. The results of efficiency of PROM made using two grades (+ 24% P2O5 and 34% P2O5) of rock phosphate mineral from Jordan show that low grade rock phosphate slimes are as effective as high grade phosphate concentrate when used in PROM. Also PROM is very effective in the saline soil of Eshidiya desert where DAP is not at all effective.

Jan 1, 2014

By G. K. Nimal Subasinghe

The performance of Knelson Concentrators (KC) used in gravity gold circuits often rely on manufacturers recommendations for selecting operating parameters such as, fluidising water flow rate and cycle time. These parameters are not only dependent on the machine parameters but also on the feed characteristics such as, size distribution of the gold and gangue particles and the composition of its constituent minerals. Particles entering a KC experience a centrifugal acceleration due to the high bowl rotational speeds. The required centripetal force to maintain a circular orbit is provided by the fluid drag force created by inwardly moving fluidising water and the Bagnold?s force arising from the particle/particle collisions. When the fluidising water flow rate is low particles move towards the outer rim of the bowl forming a packed bed within the grooves. On the other hand, if the fluidizing water flowrate is high the bed expands and some particles would move towards the centre of the bowl and gets carried to the tailings stream, thereby reducing the amount of material retained within the grooves of the bowl. Based on a force balance on a particle captured within the KC, a criterion which can be used to predict the amount of material recovered in the bowl has been derived and validated by experimental data. It has been shown that this criterion may be used to demarcate the transition of the bed from a packed-bed state to a fluidised state depending on the fluidizing water flowrate. By identifying the conditions for the optimal expansion of the bed comprising different constituent mineral particles, it is possible to estimate the optimal fluidising water flow rate that would maximise gold recovery while minimising gangue recovery. It was also observed that the accumulation of valuable mineral in the bowl takes place by the displacement of the gangue particles in the bed by the incoming heavy mineral particles. The kinetics of this segregation process was also studied using synthetic mineral mixtures. The experimental data suggests that this process follows first order kinetics and that the build up of heavy mineral reaches a maximum which could be used to identify the optimal cycle time. In this work, the suitability of the above model in optimizing the performance of a KC have been reviewed. Also the necessity and merits of testing an unaltered feed samples in a larger KC as op-posed to the commonly adopted GRG test has been discussed.

Jan 1, 2014

By Shobhana Dey

A low volatile medium coking coal from Jharia coal field, India was used for this investigation. The proximate analysis of the sample shows that it contains about 25% ash, 21.4% volatile matter and 52.4% fixed carbon. The sizewise ash analysis of -0.5 mm coal indicates that -0.5+0.15 mm fraction contains lower ash (21.9%) than -0.15 mm fraction (29.7%). The flotation characteristics of the -0.5 mm coal were determined by release analysis. The study reveals that recovery of combustible at 15% ash is about 27% and 43% at 17% ash level. The low recovery is due to the presence of high ash in finer fraction. In order to improve the recovery, the -0.5 mm feed was classified into -0.5+0.15 mm and -0.15 mm fractions, and subjected to flotation separately. The recoverable combustible obtained from the release analysis of -0.5+0.15 mm at 15% ash level is 46%. The effect of collector dosage, frother dosage and aeration rate on flotation was studied. The responses of these variables in collectorless flotation of the -0.5+0.15 mm fraction containing low ash were also studied. A lower aeration and frother dosage favored the generation of clean coal with low ash. The ash content in the tailing stream from -0.5+0.15 mm flotation circuit is close to that of the original -0.15 mm fraction. The former is ground to -0.15 mm and the two fractions are processed together. As -0.15 mm fraction contains relatively higher ash, collector aided flotation using sodium silicate was performed to recover the clean coal with 17% ash. The products of the overall flotation circuit having different ash levels were recommended for use in different applications. The reject from -0.15 mm flotation circuit contained 56.4% ash which can be used for fluidized bed combustion (FBC). This eventually leads to complete utilization of coal.

Sep 1, 2012