Search Documents

By H. D. S. Miller

"This paper introduces a new approach for the design of rib pillars. The approach involves an empirical method for pillar design and a discussion of the pillar stability used for open stope mining methods.The empirical method predicts the degree of stability of rib pillars. Pillar load is estimated using two and three dimensional numerical modelling. Rock strength and the shape effect of pillars are used to account for pillar strength. The method has been calibrated with case histories from Canadian open stope metal mines and is compared with other methods used [or pillar design.INTRODUCTIONThe role of rib pillars in open stope mining is to provide stability to a mining block by limiting rock mass displacements and restricting the exposure of a rock mass in stope backs and stope walls. They also provide flexibility in ore extraction by dividing a stope block into multiple working faces.The use of backfill in open stope mining methods has reduced somewhat the need for optimum pillar design. In contrast to permanent pillar methods, the dimensions do not need to be minimized because the pillars are usually recovered in a later stage of mining. However, the consequences of poor pillar design are often overlooked and can still be severe. A pillar that does not perform its intended role may result in excessive stope or pillar sloughing, difficult and expensive pillar recovery, loss of permanent access and remedial costs such as development rehabilitation, artificial support or low productivity.Pillar design concepts of the past were not developed for open stope mining conditions and generally have not been confirmed for open stoping. The focus of this paper is to introduce an empirical design relationship specifically for open stope pillars. It provides a stable, yielding or failed pillar assessment for a given design."

Jan 1, 1989

By Murray W. Grabinsky

The in situ performance of Cemented Paste Backfill (CPB) systems can be monitored to improve understanding of the paste?s behaviour and therefore optimize the geomechanical design of the cemented paste backfill system (i.e., minimize binder content, minimize fill barricade complexity, maximize fill rate-of-rise during backfilling, and minimize cure time before resuming production blasting in proximity). However, undertaking such a monitoring program is inherently complex, because the rate-of-rise and accompanying self-weight induced stresses occur at a time rate that is similar to the rate of hydration and associated increase in the paste?s stiffness and strength. In addition, arching within the stope complicates the overall stress distribution. Therefore, to obtain a complete picture of the CPB?s response, it is necessary to measure several variables (3-D total pressures, water pressures and suctions, hydration rates, and blasting induced vibrations) at a given ?point?, and take all such measurements at many points throughout the fill mass as well as at barricade locations. The overall testing program can therefore be logistically difficult to implement and have an apparent high cost. However, if the program is successful, the benefits arising from improved mine design can significantly outweigh the seemingly high initial capital and labour costs of the monitoring program. A case study involving an instrumented paste fill in a tall and narrow Alimak stope is briefly outlined. Plans for a similar experiment in a large long-hole stope are then considered.

May 1, 2008

By C. L. O?Brian

Note: The figures quoted and used herein have been collected and furnished by the Dominion Bureau of Statistics. Without the help and willing co-operation of the officials of the Burreau, this study would not have been possible. Introductory and General Coal has one unique quality that distinguishes it from other minerals, one that is too, often forgotten. This is the quality of ubiquity. No other single mineral supplies such a wide and varied range of human requirements nor, up to the present, is there any indication of a rival in this regard. There is, consequently, reason for individual study of each of the major portions of the overall market for coa1. The provision of heat for our homes is a necessity of life in our Northern climate and the largest single source of this heat is coal. It is intended to present here a factual survey of the use ?of coal in home heating throughout Canada and the effect of these requirements upon the production of coal and the development of this phase of the mineral industry. The coals used for house heating in Canada fall into two more or less distinct classes, the very low volatile fuels such as anthracite and including coke, and the bituminous coals, including the sub-bituminous and lignite types. The bituminous coals are, of course, consumed in very large quantity by industry and the railways and are the most important type both in tonnage and in respect to the area supplied. Canada consumes annually for all purposes some 4,400,000 tons of anthracite, 4,300,000 tons of coke exclusive of Petroleum coke, and some 40,000,000 tons of bituminous coals. This is a fair measure of the relative importance to the overall economy although it will be seen later that the concentration of the household use of anthracite and coke into one area does create a considerable problem.

Jan 1, 1951

By James F. Kelly

"Research and development on the production of liquid fuels from coal is rapidly expanding worldwide. AImost every industrialized country is engaged in coal liquefaction research and process development. In Canada, the federal government, through CANMET, is actively supporting coal liquefaction research and development in a number of areas.A review of the basic principles and some of the processes in the production of liquid fuels from coal is presented with reference to the characteristics of the Canadian coal resource base. An overview of the CANMET coal liquefaction contract program is presented and specific results from the batch autoclave studies carried out with Saskatchewan lignite and Nova Scotia high-volatile bituminous coals are discussed. The development of a laboratory-scale continuous-flow coal liquefaction unit at the Energy Research Laboratories of CANMET to investigate the co-processing of coal and bitumen/heavy oil is also described.IntroductionThe CANMET coal liquefaction program is made up of two interrelated parts. The first part consists of the research and development work being carried out at the Energy Research Laboratories near Ottawa, Ontario. The second focuses on contracted research and development as part of the Coal Conversion Program of Energy, Mines and Resources Canada. The long-term objective of the liquefaction program is to contribute to Canadian liquid fuels production through the definition, research, development and demonstration of different coal liquefaction technologies for the Canadian marketplace.In coal liquefaction, the main objective is to produce liquid hydrocarbons with a higher hydrogen-to-carbon (HIC) ratio than the coal feedstock while reducing heteroatom (S,O,N) content, removing mineral matter and minimizing gas production. Typical atomic HIC ratios for a variety of Canadian fossil fuels are shown in Figure I. Canadian coals have about 500/0 of the HIC ratio present in tar sand bitumens and heavy oils and only about 20% of the ratio available in methane. In general, the cost of synthetic fuels production is related to the cost of hydrogen and, therefore, to the increase of HIC ratio between feedstock and products. On this basis, the production of liquid fuels from coal would be expected to be more costly than liquid fuels produced from bitumens and heavy oils. However, for mined tar sands and some heavy oils this difference tends to be offset by the large sand content (>90 mass %) of tar sand bitumens and the increased production costs of viscous heavy oils. For example, it can easily be shown that for equal weights of distillate <525°C) yield, current coal liquefaction processes required between two and three times less mined coal by weight than the mined tar sand s required for the most technically advanced processes proposed for bitumen upgrading."

Jan 1, 1981

By P. Erskine

"Hyperaccumulator plants can be utilized as ‘metal crops’ in agromining operations. This emerging technology produces ‘bio-ore’ by growing ‘metal crops’ on sub-economic ore materials, such as natural ultramafic soils. In a typical application plant material is periodically harvested, followed by air-drying and incineration to produce the bio-ore intermediate. The bio-ore is of very high grade (>15 wt. % nickel), and of high purity, and may be processed into a number of different products (including nickel metal, nickelbased catalysts and pure nickel salts). Current research efforts from our team focus on Mediterranean climate field trials (Albania, Spain) and tropical climate field trials (Malaysia, Indonesia). Field-scale demonstrations are required to provide evidence of real-life performance and of economic viability. If the trials are successful, agromining may in the near future support local livelihoods with income opportunities as an alternative type of agriculture: to farm nickel.INTRODUCTION AND BACKGROUNDHyperaccumulators, plants with the exceptional ability to uptake and concentrate trace elements in their shoots, can be utilized as ‘metal crops’ in agromining operations (Van der Ent et al., 2013a; Chaney et al., 1998; 2007; Baker et al., 2010). Agromining produces ‘bio-ore’ from harvested ‘metal crops’ grown on unconventional resources such as natural ultramafic soils (Van der Ent et al., 2015a). The bio-ore is a high purity and very high-grade (>15 wt. % nickel) product intermediate that may be turned into a range of different nickel products (nickel metal, nickel-based catalysts and pure nickel salts). The potential for agromining is greatest in tropical regions (Asia-Pacific Region: Indonesia, Malaysia, Philippines, Papua New Guinea and New Caledonia) where very large expanses of ultramafic soils exist (Van der Ent et al., 2013b). Agromine operations may replace existing marginal agriculture on poor ultramafic soils, for example oil palm plantations that require high fertilizer inputs to be profitable, or take place under a cocropping regime. The ability of ‘metal crops’ to achieve high nickel concentrations in biomass (and hence in harvested bio-ore) from soils with just <0.5 wt. % nickel, makes it possible to access non-conventional resources, not accessible by traditional mining techniques. For example, agromining could be implemented both on the large ‘halo’ around the mine leases, and as part of the progressive rehabilitation process after conventional resource extraction. As with all methods for resource extraction, agromining will be finite due to the diminishing concentrations of nickel in the zone accessible by plant roots (Chaney et al., 2014). Nevertheless, considering soil materials with 0.2 wt. % nickel and ‘metal crops’ with a yield of 100 kg nickel ha-1 the agromining venture may be sustainable for decades (Van der Ent et al., 2015a)."

Jan 1, 2016

By G. Kolosay, J. Rajala, G. Deschênes, A. Romanov

ABSTRACT: The Kupol project, located in Far East Russia, treats a high-grade, free-milling ore containing 25.5 g/t Au, 298.3 g/t Ag, and 0.5% pyrite,with a P70of 53 microns. Most of the gold occurs as elec-trum and the silver is predominantly present in the forms of acanthite, stephanite, pyrargyrite, andproustite. The mill reached the design throughput of 3,000 tpd six months after start-up. AKnelson con-centratorrecovers10.4% gold and 0.5% silver. Leaching (14.5 hours pre-treatment and 116 hours cyanidation)of gravity tailings is performed with the CANMET Enhanced Leach Process (CELP),using only 490 ppm NaCN (cyanide consumption of 1.1 kg/t). The precious metals are recovered by counter-current decantation and the Merrill-Crowe process. Calcium hypochlorite destroyscyanide and thiocyanate. Overall, gold and silver recoveries for 2008 were 95.4% and 85.6%,respectively (2.8% and7.5% higher than targets). In the first eight months of 2009, gold and silver recoveries were 95.0% and82.6%, respectively. By August 2009, the mill processed 1.42 million tons of ore with production of 1.1 million ounces of gold and 11.3 million ounces of silver. CELP performance was not affected by changes in feed grade and ore mineralogy. Additional modifications to the leaching strategy allowed further reduction in cyanide consumption and effluent treatment costs.

Jan 1, 2011

By J. P. Mogan, E. D. Dainty

Polynuclear aromatic hydrocarbon (PAH) levels were measured in underground mines in the vicinity of diesel vehicles equipped with several types of emission control hardware. In general, the emission control devices appeared to reduce PAH levels, but the companion mutagen study suggested that the reduction might not always be beneficial. All underground PAH concentrations were below the threshold at which a significant impact on health might be expected, but recent animal exposure studies have indicated that PAHs may not be the primary hazard from chronic diesel fume exposure.

Jan 1, 1994

By M. G. Mena, M. S. Mena-Junio

Silver (Ag) filler is the most attractive choice among all the conductive fillers of die attach materials for the attachment of silicon devices because of its balanced properties and relatively low cost. However, a major problem of silver is that it electrochemically migrates in the presence of moisture and applied bias. In microelectronic devices, silver migration can occur between adjacent conductors/electrodes, which leads to the formation of dendrites and eventually results in short-circuit failure. In order to assess the reliability of these conductive fillers, an investigation for two types of Agfilled die attach materials was done using the water drop test at room temperature and using a 1000 hour temperature-humidity-bias (THB) test under 85 ºC and 85% RH conditions. Four different bias/voltages and six different electrode distance spacing were used. Both dendrite and cloud/lace-like growths were observed as a result of the electrochemical migration of silver, Water drop test results revealed that increasing the applied bias/voltage and decreasing electrode spacing causes migration to occur faster, thus resulting into shorter time to fail (TTF). It was also observed that the growth rate of the dendrites from the cathode increased linearly as the dendrite approached the anode, probably due to a combination of increased ion availability and electric field. A minimum applied bias of 3 volts was required for migration to occur in Epoxy A as compared to 2 volts for Epoxy C. The higher migration voltage requirement for Epoxy A was attributed to the higher volume resistivity of the said epoxy. In addition to the minimum voltage requirement, it was observed that a minimum electric field of about 1 volt/mm is required for migration to occur. The 1000 hour THB testing did not result into any failures. It is surmised that the presence of condensed moisture is another requirement for migration. Such condition is not present during the THB testing.

Jan 1, 2012

By Zhong Zhang, Xijia Wu, Prakash C. Patnaik

"The integrated vehicle health management (IVHM) aims to provide life cycle management decisions for military platforms and their propulsion systems with reduced life cycle cost and increased operational readiness and safety. Among all the issues of IVHM, understanding the physics of failure and predicting lifetimes of the critical components are at the core: the decision makers have to know when and how components may fail in order to plan ahead. The difficulties to tackle failure problems lie in determining the material response to cyclic and dwell loading in a hot and/or corrosive environment, which all combined to induce multiple deformation and damage mechanisms, resulting in mix-mode failures. Decisions on component replacement (material substitution) and mission planning must be based on quantitative knowledge of the failure physics to mitigating the risk. This paper addresses the critical failure issues including creep, low cycle fatigue (LCF), thermomechanical fatigue (TMF); and presents a mechanism-based approach to life prediction. The mechanism-based approach, by its physics, is widely applicable over the entire operational spectrum of both structures and engines, and therefore is suitable for IVHM purposes."

Jan 1, 2015

By Michael S. Moats, Addin Pranowo, Gerardo R. F. Alvear F., Nigel Aslin

The toxicity of arsenic is well known and documented. However, the presence of arsenic in copper electrorefining anodes and electrolyte is critically necessary to produce high quality cathode. Arsenic as well as antimony and bismuth concentrations vary depending on their relative concentration in the anode copper received from the smelter. This paper will discuss the behaviour and benefits of arsenic in copper electrorefining and the detrimental effects of antimony and bismuth on cathode quality and tankhouse performance. This will include the minimization and mitigation of problems associated with arsenic, antimony and bismuth including cathode contamination, floating slimes, scaling and anode passivation.

Jan 1, 2014

By Y. Okita

The Goro Nickel Process, developed over a ten-year period, uses a number of novel processing steps and treats two ore types, limonite and saprolite, together. Nickel and cobalt are solubilized using a 270°C pressure acid leach in sulphuric acid, which rejects the bulk of the Fe into the leach residue as hematite particles. The leach liquor, after CCD, is aerated to oxidize any ferrous iron to the ferric form and is neutralized with limestone to precipitate iron as hydroxides together with gypsum from the neutralization of the free acid. Further neutralization with lime, together with steps to remove suspended solids, reduces the iron to trace concentrations. Nickel and cobalt are directly recovered from the low-iron leach solution by primary solvent extraction using Cyanex 301® (SXl) in a high-strength hydrochloric acid solution. Nickel is subsequently separated from cobalt in a secondary solvent extraction circuit using tri-isooctylamine. A granular nickel oxide is made by pyrohydrolysis, and this allows the recycling of hydrochloric acid to SXl. The process is currently being commercialized to produce approximately 60,000 t/y of nickel as nickel oxide granules and approximately 5,400 t/y of cobalt as cobalt carbonate. This paper describes the process development to control the Fe concentration in the SX1 feed solution.

Jan 1, 2006

By R. Therrien, L. Gosselin, J. Raymond

ABSTRACT: The ground-source heat pump is a low-carbon technology that uses the earth’s renewable geothermal energy to heat and cool buildings, with an ef?ciency 3 to 4.5 times that of conventional systems. These high performances are achieved by exchanging heat with the earth using a ?uid that is groundwater, surface water or closed loop water. Expensive installations, such as water wells or buriedcoils, are however necessary to transport the ?uid in and out of the earth, making ground-source heat pumps more costly than conventional systems. Installation costs can be reduced in mining environments due to the abundance and the accessibility of resources to be used with heat pumps. Examples of system design calculations using data from the Mouska and Doyon mines in Abitibi show signi?cant installation cost savings using dewatering mine water or ground heat exchangers installed below oxidizing waste rocks. Ground-source heat pump technology is reviewed and scoping calculations are presented to illustrate how such systems can be installed in mining environments.

Jan 1, 2010

By P. A. Riveros

The literature on the arsenic disposal practices of the metallurgical industry and the long term stability of the disposed arsenic compounds were reviewed from a Canadian perspective. The review was complemented with visits to selected operating metallurgical sites to obtain information on their specific arsenic disposal practices and the behaviour of their impounded residues. The arsenic disposal procedure currently favoured by the industry involves the formation of an insoluble ferric arsenate compound which is allowed to sediment at the bottom of tailings or residue ponds. It has been shown recently that this poorly formed crystalline compound is similar to ?arsenical ferrihydrite? which is ferrihydrite containing strongly adsorbed arsenate anions. Despite concerns about its long term stability on thermodynamic grounds, arsenical ferrihydrite appears to be stable for many years in the proper environment which includes a slightly acidic pH and oxidizing conditions. A high Fe/As ratio and the presence of heavy metals appear to increase the stability of arsenical ferrihydrite. High temperature operations, such as those encountered in the autoclave treatment of refractory gold ores, are conducive to the formation of scorodite, FeAsO42H2O, and/or a series of ferric arseno-hydroxy-sulphate compounds depending on the solution composition. Scorodite has several advantages over arsenical ferrihydrite as a disposal compound including a lower iron demand, a higher density and a greater thermodynamic stability. New procedures have been developed to generate scorodite at ambient pressure. These procedures would offer significant reductions in capital cost because of the elimination of the need for autoclaves. Lime precipitation which was widely used in the past is being abandoned as a result of strong evidence showing that calcium arsenate compounds decompose slowly in contact with atmospheric CO2 to form calcium carbonate and soluble arsenic acid. In all cases, the long term stability of the disposed arsenic compounds depends on a number of factors including disposal site characteristics, particle crystallinity and size distribution, the presence of complexing agents and the effect of bacterial activity.

May 1, 2002

By Jennifer L. Durocher

The speciation of Uranium in silica-rich depositional layers in the proximity to uranium ore deposits was investigated using micro X-ray fluorescence (VESPERS beamline), X-ray Photoelectron Spectroscopy, Laser-Ablation ICP-MS, Scanning Electron Microscopy, X-ray diffraction and micro-Raman Spectroscopy. The heterogeneous Si-rich laminations contain micrometer-size inclusions of secondary uranyl-minerals. The U concentration in these layers range from several ppm to lower weight percent and the U 4f spectra indicate variable U6+ : U4+ ratios. Correlation between the concentrations of U and other metals (M) suggests the presence of M-U-bearing nanoparticles or species within the amorphous silica matrix.

Aug 1, 2010

By S. Nafisi

Permanent mould casting may be regarded as a near net shape manufacturing process where simple to very complex shaped engineering components are fabricated with high degree of precision in a metallic mould, In this group, High Pressure Die Casting (HPDC), have attracted more attention because of its production rate, where both liquid and semisolid metals (SSM) can be cast with high degree of precision and versatility. The SEED process, Swirled Enthalpy Equilibration Device, the new process for semi solid metals slurry production route, consists of two steps: The initial step involves extracting specified amount of heat from the molten metal to generate liquid-solid slurry, i.e, mushy structure, followed by drainage of remaining liquid to produce a compact free standing feedstock for rheocasting operation. In this paper, microstructural evolution during the process and some initial casting results of automotive components an also reported.

Jan 1, 2005

By David Weiss

There is significant interest in magnesium alloy castings because of their low density and high strength-to-weight ratio. Lightweight, structural materials are in demand for aerospace components, hand tools, electronics enclosures and have found increased use in automobiles. Increased use of magnesium will depend on the resolution of a number of materials issues and development of more robust and economical casting processes. Most magnesium components are produced via high-pressure die-casting. High levels of porosity generally prevent die-castings from being used for structural applications. Structural castings are usually produced using sand casting, a relatively expensive casting process. Some work has been done on development of lower cost structural casting processes, such as lost foam casting, permanent mold casting or low-pressure casting. New programs are also emphasizing improvements to the die casting process. Research and cost reduction needs are discussed with respect to current programs and future components.

Jan 1, 2006

By E. A. Haack, L. A. Warren

Microbial biofilms are essentially biologically driven, geochemically reactive interfaces between solid surfaces and aqueous solution. Field investigation of acid rock drainage (ARD) associated biofilms at the Onaping nickel mine (Onaping, ON, Falconbridge Ltd.) characterizing: (1) micro-geochemical gradients (pH and 02) (2) metal fractionation amongst biofilm solid components, and (3) microbial communities; over both seasonal and diel timescales, indicates biofilms are highly metal reactive. Results show that biofilms accumulate metals seasonally within two microbially controlled solid fractions: amorphous hydrous manganese oxide (HMO) biominerals and the organic biofilm matrix. The biofilm geochemical microenvironment, driven largely by microbial activity, dynamically impacts biofilm metal behaviour over diel timescales. These results indicate that microbes can play significant roles in bulk system reactive metal transport. On-going investigation seeks to characterize the microbial players and the microbial-geochemical linkages involved in Mn oxyhydroxide formation and subsequent metal sequestration, with the goal of engineering this microbial activity as a bioremediative approach. This approach will have utility beyond specific application to the Falconbridge Onaping site to ARD mine environments in general, as the likely microbes involved in these processes are typically common across ARD systems.

Jan 1, 2004

By Garth A. Maki, Douglas M. Smith

"The largest Canadian Lamella thickener installation to date, with 13,600 square feet of effective clarification area, is being successfully operated at the Rocanville Division of the Potash Corporation of Saskatchewan. This paper describes the operational principles and design features of the Lamella thickener, the testwork leading to its specification f or the Rocanville phase two expansion project and a review of the performance of the full- size unit which was ultimately installed.IntroductionThickening equipment operating in the Saskatchewan potash industry is generally situated within the main refinery building or a separate insulated an d he a ted structure due to the harsh winter climate and the need to closely control plant brine temperatures to en sure good flotation circuit performance Because these thickeners process large volumes of slurries, conventional basin-type designs tend to be large, ranging up to 46 metres in diameter or more. As part of the general effort to minimize capital costs, the potash industry has begun investigating the use of alternate high capacity, compact thickener designs. In this regard, PCS Rocanville undertook a program in 1978 to evaluate a LamellaTM thickener and one other type of ‘High Rate&apos; thickener as alternatives to a conventional design for the phase two expansion project.This paper describes the operational principles and design features o f the Lamella thickener (manufactured by Axel Johnson lnc.), the testwork leading to the specification of a Lamella for the Rocanville expansion project and are view of the performance of the full size unit (1486 m2 of the effective settling area) which was ultimately installed and is now successfully operating in the Rocanville plant."

Jan 1, 1983

By Michael Hopp

The beginning of the electroforming technology dates itself on the beginning of the industrial revolution 1800 to 1820 approximately. Basis was the existence of electricity in order to be able to drive the electro-form-process. With the development of the electroforming, big names of the science-history, like Michael Faraday (1791? 1867), Sir Humphry Davy (1778?1829), Max Schlöter (1878?1946), Werner von Siemens (1816?1892), and many others. Already about 1840, the first monograph originated to electroforming (38). While in the first years the electro-forming most extensive only to the coating of art - or other usefulness-objects was used. At the beginning of the 20th century the formation of objects from pure metals became more and more, later also from alloys to application. One of the first bigger electroforming factory, the ?Württembergische Metallwarenfabrik? was working 1894 in Geislingen / Steige, Germany. In the industrial application, the electroforming-technology is to be imagined away for miscellaneous coatingtasks and the form-body-preparation no longer today (18). A little bit confusing is the german name ?galvano-technology?. It based on the Italian physiologist and naturalist Luigi Galvani (1737 ? 1798). This technically seen is however wrong, right, the name is electroforming. The physical bases were founded mainly by Michael Faraday and Sir Humphry Davy.

Oct 1, 2003

By K. D. Peaslee, J. M. Soto, IéDé Smith, A. D. Saikia

Spinel-forming refractories reduce penetration of metal and slags through a tightening of the lining resulting from the volume expansion associated with spinel formation. Sintered spinel is often added to reduce the overall volume increase that results. Tests were carried out to evaluate the difference between performance in spinel¬forming and spinel-containing refractory mixes. Eight refractory linings, one (1) pure magnesia, three (3) magnesia-base spinel-forming and four (4) alumina-base spinel- containing dry vibratable mixes, were tested for four hours in contact with a low-carbon, low-manganese steel in a coreless induction furnace at temperatures in excess of 1600°C. Afterwards, sample were cut, and wear and penetration were measured. For magnesia-based mixes wear was minimized using 20-30wt. spinel formation although further experiments are required to determine the optimal value. For alumina-based mixes, there seemed to be an optimal ratio of spinel added to spinel formed that minimized penetration and wear at the slag line. Sidewall accretion occurred and the accretion was the highest for the mix with the ratio of spinel formed to spinel added near the optimum value. Therefore to design an optimum refractory mix the benefits of higher ratios of spinel added to reduce penetration and wear at the slag line will need to be balanced with the tendency to form accretions on the sidewalls of the furnace.

Jan 1, 2004