Search Documents

By Mario Giavani

"IntroductionBy the 1980s, the continuous casting process had almost replaced the ingot casting of steel in Japan, Europe and North America. While economic benefits were probably initially the prime motivator, the transition rapidly proved that continuous casting in conjunction with ladle metallurgy offered a much greater degree of process control and made possible routine steel specifications not previously available.This new technology demanded improved procedures for the quantitative addition of the required metallurgical reagents, especially as regards the more volatile or toxic elements such as calcium , boron, carbon, sulphur, lead and tellurium. The injection of a thin steel tube containing the additive into the metal bath of a ladle refining station offers the best procedure for this. In the industry this has been referred to as ""Cored Wire "".Operators using this technology find that they are faced with balancing the injection speed against the small cross section of the tube to feed the required amount of reagent within the short time span available to the operation. Large-ladle shops requiring major quantities usually resort to multiple machines and high speeds. Many machines approach a maximum feed rate of 400 m/min. Higher speeds exceed the melting rate of the system, and the ferro static buoyancy of the tube will cause it to coil up the ladle wall and possibly resurface into the slag layer, and even eject itself above the surface. Under these conditions, exact control of element recovery becomes difficult, especially with the commonest addition, calcium silicon alloy.Many experts are persuaded that the final melting position of the carrier tube has major importance in establishing consistent element yield to the bath.For instance, the vapour pressure of calcium at l500 °C is close to I bar. Reaching a melting depth sufficient to accommodate this value is considered to be critical to achieving an acceptable and reproducible yield.Many studies have elaborated semi-empirical formulae to establish the melt zone depth, connecting this value essentially to the diameter of the tube , the injection rate and the superheat of the steel. The author's experience in steelmaking could not confirm such formulae in many cases, and he set about to develop a mathematical model which would be useful in addressing this problem, taking into account the several parameters which will regulate the behaviour of a filled tube immersed in molten steel."

Jan 1, 1991

By N. Depountis, Th. Rondoyanni, N. Sabtakakis, K. Kavoura, G. Tsiambaos

"Some typical landslide phenomena in Western Greece, in terms of the geological composition and structure of the affected materials, induced by heavy rainfalls are thoroughly studied and analyzed. The studied cases involved composite landslides on Flysch and Neogene weak rock materials that constitute the most critical landslide prone geological formations in Greece. The studied cases named Platanos, Platanitis and Karya include representative landslide sites as regards to the geological composition and structure of the displaced formations. Thus: (a) flysch formation includes different lithological units as shales, marls, siltstones, sandstones and conglomerates and is closely related to the Alpine orogenesis, suffering intense past tectonic movements and (b) the Plio-Pleistocene sediments include clayey marls, marlstones and siltstones. The investigated landslides , affecting mainly transportation routes, are closely related to heavy rainfalls induced by extreme meteorological events and are controlled by active fault tectonics. Long term movement monitoring using borehole inclinometers and surface benchmarks showed that landslides control was effective only when the proper stabilization measures were taken.INTRODUCTIONLandslides represent a major threat to human life, property, infrastructure and natural environment in most regions of the world. They are recognized by the scientific and politic authorities as having a major socio-economic impact and they represent a significant hazard for the population and the properties in particular locations. Landslide hazard expressed as the probability of occurrence within a reference time period and is a function of the spatial and temporal probability (Varnes, 1984; Guzzetti et al. 1999; Lee and Jones, 2004). The spatial probability of landslide initiation (susceptibility) is mainly related to static causal factors (slope inclination, material properties, etc.), while temporal probability is mainly related to dynamic causal factors such as rain input, increased groundwater levels and drainage, earthquakes, etc. (Van Westen et al. 2005)."

Jan 1, 2015

By O. A. Ojo, R. G. Ding, M. C. Chaturvedi

In this study, the fusion zone microstructure of tungsten inert gas weld of a directionally solidified NhAl-base alloy was examined by the standard metallographic techniques involving the use of optical, analytical scanning and transmission electron microscopes equipped with energy dispersive spectrometer. Fusion zone cracking was observed to be closely associated with interdendritic secondary solidification microconstituents identified as consisting of y, y' and NiMo (Y) phases. Depression of solidification temperature to the terminal ternary eutectic reaction temperature of the system during cooling is suggested as an important factor contributing to the fusion zone cracking.

Jan 1, 2006

By M. W. Bartley

"The Steep Rock iron mine is 141 miles west of Port Arthur, Ontario, and 50 air miles north of the Vermilion iron range in Minnesota. The property lies 27'2 miles north of the village of Atikokan on the south line of the Canadian National railway.The orebodies underlying the bed of Steeprock lake, since de-watered, were known to exist 60 years ago. References to the possibility of iron ore at this location were made in early geological papers and reports. In 1937 the locations of the orebodies were determined by dip-needle surveys, and from 1938 until 1943 diamond and churn drilling were carried on to prove their extent and configuration. It was found that the ore, in addition to being associated with definite types of wall-rock, was controlled by structural features, such as faulting and brecciation. From this geological knowledge the presence of other deposits has been inferred, and these are now being explored and developed.Production at the property commenced in October 1944. During that season 17,000 gross tons of ore was shipped. In 1945, 506,000 gross tons and in 1946, 830,000 gross tons was mined and shipped. In 1946, 164,000 gross tons additional was placed on stockpile for future shipment.TYPES OF 0REBODIES AND MINERALIZATIONThe orebodies at Steep Rock iron mine are tabular. They have an average width of 250 feet, lengths up to 5,000 feet, and persist to known depths of 1,200 feet. They are replacement deposits in brecciated carbonate rock at the contact of underlying pseudo-limestones and overlying volcanics. The ore is composed of hematite and hydrous hematite in the form of goethite, turgite, and limonite. These minerals replaced the brecciated host rocks metasomatically as a result of hydrothermal action. The replacement followed a definite structural pattern along the unconformable contact between the sediments or other rock (now replaced by carbonate) and the volcanics, subsequent to disturbances in the form of minor faulting and shearing. The faulting and shearing provided the channels through which the replacing solutions percolated. It is believed that the late basic intrusions outside the ore zones were genetically responsible for the introduction of the ore-bearing solutions."

Jan 1, 1954

By Tong Gao, Jingyu Qin, Xiangfa Liu, Zengqiang Li, Yaoxian Zhang

"The recycling of aluminum materials is a green industry while Al–Si–Fe is a typical scrap aluminum alloy. The key point of recycling Al–Si–Fe alloys is to separate Al and Fe. A method by introducing Al–Si–Fe alloy into Mg melt has been investigated in this work. A separation layer with gathered Fe–rich and Al–poor Fe3Al0.7Si0.3particleswereobtained at the bottom of the cooled ingot, indicating that ß–Al5FeSi phase evolves to Fe3Al0.7Si0.3,and quantities of Al are released to the upper Mg matrix simultaneously. However, the evolution of Al3Fe by adding binary Al–Fe alloy into Mg melt is different. It is supposed that element Si plays an important role in the phase evolution procedure, thus affects the performance of this method.INTRODUCTIONThe rapid development of aeronautic, automotive and aerospace technologies strongly depends on the consumption of materials while sustainable development is the future direction. Aluminum is a recyclable material and Fe is one of the most common impurities. Fe can be introduced into Al alloys by adding low–purity alloying materials and by using unprotected ferrous crucible, tools, or equipment (Belov, Eskin, & Avxentieva, 2005; Lu& Dahle, 2005).For instance, wear–resistant rings are usually inserted into piston head when producing Al–Si pistons and an infiltration procedure by putting the cast iron rings into Al melt before casting is always essential. As a result, both the re–melting of pistons and the reuse of Al melt for infiltrating rings lead to the enrichment of Fe element, which can be even high to5 wt.%(Viala, Peronnet, Barbeau, Bosselet, & Bouix, 2002).It is widely known that coarse ß–Al5FeSi particles will form in high Fe–contained Al–Si–Fe alloys which is harmful for the mechanical properties. In the past decades, several methods to eliminate the harmful effect of ß–Al5FeSi have been investigated which can be widely referred (Munson, 1967; Lee, Kawamura, Inoue, Cho, & Masumoto, 1997; Suárez–Peña& Asensio–Lozano, 2006; Eidhed, 2008).These methods can be divided into two kinds. One is to aggregate and remove the Fe–rich phase through the methods of centrifugation, gravity filtration or magnetic separation. The other way is modifing the morphology of flake–like ß–Al5FeSi by overheating the melt or adding neutralized elements. However, these traditional methods are rarely used due to respective disadvantages, while the commonly used method in industries is using primary aluminum for dilution. However, only less than 30wt.% of scrap Al can be added into primary Al melt for each melting work while it also results in the decrease of product quality simultaneously(Zhang, Liu, & Liu, 2016)."

Jan 1, 2018

By Michael L. Wayman

Native copper is copper which exists in nature in elemental, metallic form, rather than being chemically combined as is the case with the more usual oxides, sulphides, carbonates, etc. Several other metals, including gold, silver and platinum also occur in native form as do, rarely, iron, lead and mercury; a few naturally occurring native metal alloys, notably electrum (gold-silver) and tumbaga {gold-copper or gold-silver-copper) are also known. An analysis of the relative availabilities of native metals as nuggets in a useful size range111 shows native copper to be by far the most abundant of the native metals, exceeding both gold and silver by at least two orders of magnitude. While such an estimate is crude, it serves to illustrate why copper was one of the first metals to have been used by ancient peoples. Native copper is known to occur on all inhabited continents, although its exact occurrence and abundance in ancient times cannot be unambiguously determined because many surface deposits may have been totally exploited and are therefore no longer identifiable. It is clear, however, that native copper does or did occur in all areas inhabited by the early metal-using cultures, for example in the Andes, in Central America and in the Near East. It is remarkable that most of the world's native copper occurs in the deposits of the Lake Superior region, primarily on the Keweenaw Peninsula of northern Michigan and on Isle Royale, an island in Lake Superior some 50 miles north of the Keweenaw deposits. Canadian deposits of native copper are found in Newfoundland and Labrador, Nova Scotia, Quebec, Ontario, British Columbia, the Yukon and the Northwest Territories (on Victoria Island, Bathurst Inlet and the Coppermine River area). It was to locate the latter deposits that the explorer Samuel Hearne was sent overland from Hudson's Bay in 1770; they were also surveyed by Sir John Franklin's expeditions in the 1820s.

Jan 1, 1999

By David J. Forrester

Light metals, most commonly aluminum, magnesium and titanium, and alloys containing them, are used in many industrial applications where lightness, hardness, ductility and resistance to corrosion are needed. However, these metals and their alloys can be potential ignition sources in underground coal mines. There have been documented cases in the past where coal mine explosions have been attributed to firedamp/coal dust ignitions involving light metals or their alloys. When light metals and their alloys are brought into contact with oxygen bearing material, such as iron oxide (rust) in the presence of heat, then a "thermite" reaction can occur. This source of heat can be generated from frictional contact or from merely a glancing blow. This paper outlines the hazards associated with light alloys, how to address the hazard and regulatory aspects.

May 1, 2001

By L. A. Wolff

"Second only to profitability (but not always), productivity is the most sought after measurement for any operation. Whether it is in mining, manufacturing, construction, etc. managers always strive to know how productive various plants are. Even the Federal Government publishes, on a monthly basis, the nation's factory productivity.Just what is productivity and how is it measured? This is an important topic that needs definition. There could be as many definitions as there are people who use it. In the mining industry, the problems of defining productivity would be analogous to the definition of equipment availability. Each mine uses its own definition to suit its needs. Webster's Dictionary primary definition for productivity is ""the quality or state of being productive,"" with productive defined as ""yielding or furnishing results"". This is a definition open to a broad spectrum of interpretation. A discussion of how productivity is currently used in the mining industry and an introduction to a potentially different approach to its measurement and meaning is outlined in this paper.Defining ProductivityTypically, mining companies track several productivity measurements. These are partial measurements and can include the following: tons per specific equipment class per hour, tons per man-hour, and tons per shift. None of these, however, track the over -all performance of an operation, that is, a unit of output per unit of input.For a mining operation, output can be defined as the ore for mill processing. Input includes the required components of production to get to the first step of processing.The Wesbter's Dictionary definition exposes a major fault with the popular conception of productivity. The critical issue of accountability is not addressed. Anyone can produce, but how effectively and efficiently are resources used? Quantifying an inclusive measurement is difficult. Statistics referring to an increase in productivity are often published. For example, the statement, ""gold productivity has increased by 10%"" has many potentially dangerous interpretations. One must ask, what does this statistic really mean? Did gold companies produce 10%” more gold? Were less manpower or equipment required to produce an equivalent amount of gold? Did ore grades increase resulting in more ounces of gold produced, etc. Obviously, production and productivity can have similar implied applications and meanings, but also two different, distinct meanings.To mitigate a problem like this, a standard needs to be developed within the mining industry. Because of varying ore grade, changing waste to ore ratios, different mining methods, etc. all of which have a direct impact on mining, development of a standard is difficult. Kendrick(1) discusses the use of Total Factor Productivity as a productivity measurement. Summarized, the total factor productivity measure is an indication of the change in efficiency in the use of basic labour and capital factors associated with an operation.This measurement has little meaning when calculated at a given point in time. It should be an on-going calculation, tracking an operation over a period of time, to evaluate trends. These trends can be used to determine if an operation's productivity is what it should be, and alert management to potential problems before they become major and adversely affect an operation."

Jan 1, 1990

By M. Tanaka, H. Narita, T. Fuchikami, K. Morisaku, K. Nagao, T. Yoshida, K. Kuroda

The effect of various N-substituents on the performance of amide-containing tertiary amine compounds in the extraction of Rh(III) from HCl solutions was investigated to optimize the structure of these compounds for industrial use. It was found that N-2-ethylhexyl-bis(N-di-2-ethylhexyl-ethylamide)amine and tris(N-di-2-ethylhexyl-ethylamide)amine, which have long, branched alkyl chains, were suitable for Rh extraction, because they were soluble in a practical diluent (80 vol% n-dodecane?20 vol% 2-ethylhexanol) and stable against highly concentrated HCl/Rh solutions, and showed high selectivity for Rh(III) over Pd(II) and Pt(IV).

Jan 1, 2014

By E. Rezaei

Glue is an animal protein that is used with other organic agent such as thiourea, in copper electrorefining. The concentration of glue has to be controlled carefully, because the electro-deposition process is often very sensitive to small fluctuation in the operating variable. This additive (glue) prevents rough and nodular copper cathode deposits. The nodular deposits may grow causing short circuits, which results in losses in production. Addition agents are effective in controlling metal growth; they modify the normal nucleation and lattice growth steps and refine the grain structure. In this research glue concentration was determined in an industrial copper electrolysis plant. A non-linear model with high capability was constructed and after optimizing its parameters, the accuracy and precision of the model were evaluated. For determination of glue concentration, electrochemical voltammograms in optimum condition (Differential pulse voltammetry technique, Hpulse=50 my, scan rate=10 mv/s) were plotted. Some information from voltammograms was transferred to the model as inputs and the calibrated model analyzed the information and the glue concentration was determined with high precision and accuracy. In this way there is no need to separate or add agents to remove interferences. This method has been successfully applied at Sarcheshmeh copper complex.

Jan 1, 2007

By H. E. T. Haultain

Research is a much over-overworked word, sometimes a much abused word; the result of this has been that many who really need the word avoid it, shy away from it just as we do from that most useful word camouflage. We have not been altogether happy in our substitutes. Some years ago I spoilt a very good proposition by using the word investigation. There were those who thought that men and methods were to be investigated instead of simply processes. They expected disruption instead of co-ordination. The bad psychology up-set the scheme. The word experiment is even worse in its psychology. To the mind of the "practical" business man experiment denotes ignorance, waste of time and money and presupposes failure. It is not so long ago that the recognized way of improving the methods about a mine was to get a new manager. Until comparatively recently the average life of a boss in a mine in the West was under, rather than over, two years. Certainly was this the case in the smaller mines. Directors did not recognize that a cock-sure new boss was a greater experiment than a certain amount of feeling-his-way on the part of the old one. We always say that no two mines are the same but we are only now realizing to what extent the fine points of the game can and must be learnt on the job itself. One of the main differences between a small property and a large one lies in this study of the fine points of the game. In the early stages operation is the main point-get her going and keep her going.

Jan 1, 1924

A series of small scale tests have been conducted to evaluate the effect of micro-fracture development, imposed by blasting, on the material properties of the rock as well as its resistance to crushing and grinding. The uniaxial compressive strength, tensile strength, Poisson?s ratio, P-wave, Swave velocity, point load test index, crushing and grinding work indices were evaluated for samples without any blasting load and for samples which had received a variety of blast loads ranging from a single impact with low strength detonating cord, to combination of impacts at different times and significant impacts with powder factors close to those used in practical blasts. It was shown that the rock mechanics parameters change even at low impacts, suggesting easier crushing of the rock. Damage values appear to be affected by timing as stress wave superposition affects the damage process. Fragmentation is achieved at higher impact levels while work index modification of the fragments is also observed when high powder factors are used, suggesting improvements in grinding.

May 1, 2003

By S. Vaidya, R. C. McKellar, N. K. Sinha

Nominally [001]-oriented nickel-base single crystal (SX) superalloy CMSX-10 was investigated at 900°C under constant load, equivalent to an initial stress of 700 MPa, for creep strains from 7% to fracture strain (18.2%). The cross-sectional dimensions of initially uniform cylindrical specimens were measured along their entire length after the tests. The cross-section changed from circular to near elliptical shape. The anisotropic deformation, measured in terms of eccentricity, increased with the increasing strain level, but for a given strain level, it remained constant along the entire length of the initially uniform gauge section. The results are discussed in terms of diameter, eccentricity, area, local stress and microstructural observations. The most striking observation was the fact that all the specimens displayed a yoke-like shape, or the reduction of cross sectional area, at two locations within the originally uniform gauge section. Such observations have not been reported in the literature for single crystal superalloys.

Jan 1, 2006

By F. Hashiuchi

In 1996, the Saganoseki custom smelter & refinery successfully shifted to a single flash smelting furnace operation and increased the copper production capacity to 450,000mtpy in 1998. As the next step, Saganoseki planned to further increase the productivity, especially in the area of P.S. converters and acid plant. As the results of modifications and operational improvements, the gas washing & cooling sections of the Saganoseki acid plant were consolidated from 3 to 2 lines, and increased the gas treatment capacity. The S02 converters in the acid plant were also modified to increase the acid production. By increasing the matte grade from 65 to 68%, six PS converters have been successfully integrated into four, while maintaining the production capacity and reducing the environmental burden in 2005. This paper discusses recent operational conditions and the results of implementation of technological improvements.

Jan 1, 2007

By A. L. Williamson

Bio-mining is a cost-effective, energy efficient and potentially environmentally protective method of exploiting a variety of ore deposits, from high-grade base metal deposits of nickel to lower-grade uranium and rare earth deposits such as are found in the uraniferous quartz-pebble conglomerate beds of the Quirke Syncline, Elliot Lake, Ontario. The effectiveness of bio-mining in Elliot Lake depends on the success of bacterially-influenced dissolution of metals of economic importance. The determination of both the dissolution efficiency and rate controls for uranium and rare earth elements from bio-extraction processes is critical to assess the potential for success of bio-mining along the Quirke Syncline. This poster provides an overview of proton-mediated biogeochemical dissolution of more than 50 elements, including uranium and rare earth elements from samples of the low grade uraniferous ore collected as drill core from the Elliot Lake area. The elemental data is quantified using advanced spectroscopic tools (ICP-MS and ICP-AES) to analyze solutions obtained from a series of environmentally controlled biologically-mediated uranium ore extraction experiments. The findings are being to guide design of larger scale bio-leaching experiments to further assess the potential for success of bio-mining along the Quirke Syncline.

Aug 1, 2010

By K. Sun, S. A. Argyropoulos

The assimilation and the exothermic behavior of Zn-Zr metallic powder compacts into molten aluminum was studied by measuring simultaneously the temperatures inside at the centerline of the compact, at its surface, and inside the molten aluminum bath, as well as the net downward force acting on the compact. In addition, the exothermic behavior of these compacts was also investigated in an argon protected atmosphere. The temperature rise inside the compact increases significantly with the mass ratio of zirconium in the compact and to a lesser degree with the aluminum bath temperature. Experimental results on the recovery of these compacts in molten aluminum will also be presented.

Jan 1, 2008

By P. Fellner, A. Sýkorová, J. Thonstad, J. Jurišová, V. Khandl

Adsorption of S02 on alumina used in the aluminium industry, so-called smelter grade alumina, was studied in the temperature range 15 C-120 C. The specific surface area of the smelter grade alumina was 88 m2/g. It was found that at temperatures lower than 40 C, sulphur dioxide is bonded to alumina reversibly by physical forces, and the adsorption can be described satisfactorily by the Langmuir adsorption isotherm. The heat of adsorption was estimated to be -33 kJ.mor1. At temperatures in the range 80 C- 120 C, which prevail in dry scrubbers in the aluminium industry, the heat of adsorption was determined to be -56 kJ.mor1. When S02 was adsorbed at temperatures higher than 80 C, about 30 % of the S02 could not be desorbed when heating the samples up to 250 C. In the presence of S02 and oxygen, the formation of sulphate was determined at temperatures above 90 C.

Jan 1, 2006

By J. Reinhardt, R. Uken, S. S. Chinnasamy

Au mineralization in the Kilpwal Gold Mine is confined to laminated quartz-carbonate lodes and associated alteration along the Klipwal Shear Zone (KSZ) in the upper Mozaan Group of the Mesoarchaean Pongola Supergroup. The brittle-ductile shear zone is characterized by inner and proximal hydrothermal alteration domains that are comprised of chlorite, epidote, sericite, pyrite, arsenopyrite and gold. Possible sources of gold are: (i) paleo-placer U-Au rich conglomerates from the lower Mozaan Group; (ii) the underlying mafic volcanics of the Nsuze Group; (iii) greenstone belts in the basement to the Pongola basin. Post-Pongola granites probably provided the heat to mobilize auriferous fluids into the KSZ.

Jan 1, 2011

By F. H. Samuel, J. E. Gruzleski, S. Valtierra, H. W. Doty, F. J. Tavitas-Medrano

319-type cast Al-Si-Cu-Mg alloys are commercially important due to their popular use in automotive applications. Artificial aging heat treatments are commonly performed in these alloys to achieve precipitation hardening and, therefore, improve their mechanical properties. The present work was carried out to investigate the effect of conventional (at a constant temperature, as specified by the standard T6 heat treatment regime), and non-conventional multi-temperature and interrupted aging cycles on the hardness, tensile and impact properties of two 319-type alloys (with and without Sr+ Mg additions). The results show that certain multi-temperature aging cycles have a more favourable effect in improving some mechanical properties than conventional cycles.

Jan 1, 2006

By Adriana Ramirez, Bradford C. Wesstrom, Blanca Olave

The Freeport McMoRan El Paso Copper Refinery optimizes electrolyte quality using a number of systems. The facility controls copper, arsenic, antimony and bismuth using electro-winning cells within the Tank House and Nickel Carbonate Plant Liberator Circuit (NCP). The NCP manages sulfuric acid content utilizing recycled electrolyte in the slimes autoclave as well as operation of the Acid Purification Unit, APU. The El Paso Refinery removes common sulfates using metal hydrolysis with sodium carbonate. Finally, a Mechanical Vapor Recompression, (MVR), at the water recovery plant removes sodium sulfate generated by the NCP. This paper will provide a summary of these operations. INTRODUCTION In the mid 1920’s Nichols Copper Company, in Laurel Hill, New York, recognized the need for a modern copper refining facility in the southwest. Because of the many railroad heads, affordable power and abundant labor supply, and the close proximity to regional copper smelters in Mexico, Arizona, and New Mexico, El Paso became the ideal location. In December 1928, ground was broken for the joint venture refinery between the Nichols Copper Company, Phelps Dodge, and the Calumet and Arizona Mining Company. In September 1930, Nichols Copper Company became a part of Phelps Dodge Corporation after a stock exchange. In July 1938, Nichols Division officially became Phelps Dodge Refining Corporation. In March 2007, Freeport McMoRan Copper & Gold acquired Phelps Dodge. Water is a precious commodity in the desert southwest, making conservation of this natural resource an important goal for the El Paso Operations. In the 1980’s, work on the treatment of slimes was improved through the use of pressure leaching of the refinery slimes using an autoclave. Due to the high tellurium content in the leached slimes solution, it was crucial to implement a method for its removal before the solution returns to the tank house electrolyte. A copper telluride cementation operation was installed for tellurium recovery and the ability for the recycle of the leachate and wash water to the refinery electrolyte. In October of 1997, ground was broken for the construction of the water recovery plant using an MVR to reuse process water, reducing water consumption by 25%. In 2002, electrolyte purification was further improved with the installation of an Acid Purification Unit, (APU) (Sheedy, Pajunen, & Wesstrom, 2007). The APU recycles 90% of the acids (both sulfuric and arsenic acids) back to the tank house electrolyte. The solution that passes through the APU is treated with soda ash to precipitate metals out of solution as carbonates and generates a sodium sulfate solution. This sodium sulfate solution is easily processed in the new water recovery plant, where the majority of the water recycles back into the plant as pure condensate. In 2012, the power plant at the refinery upgraded its Reverse Osmosis System to provide RO water to the tankhouse. This insured that the refinery would not have to bleed electrolyte for metals contained in well water such as: calcium, magnesium, iron and sodium chloride.

Jan 1, 2019