Search Documents

By John Wallace, Chib Iheagwara

"Refinements in sensor design have produced low-cost rugged elements to act as loop coupled with stable detection hardware for eddy current measurement in difficult environments. This has allowed direct dimension gauging of nonferrous and ferrous products above the curie point for absolute dimensions and electrical conductivities independently. The software was developed for the analysis of electrical conductivity profiles in CZ silicon crystal growth, then refined for monitoring bar properties and tube properties. Since in most metal working operations, speeds at the minimum of a few hundred feet per minute are encountered the design of the gauge was optimized for a rapid measurement. Independently extracting electrical conductivity provides data in some metal systems for determining the temperature of the monitored product.IntroductionThe primary reason for the use of eddy currents in gauging is that in hostile environments of metal working techniques such as laser gauging and ultrasonics are not always robust enough for in process measurements. Our goal is to use the technique of eddy current measurement as a quantitative control element in various hot and cold processing operations. Inspection in the early stages of metal and semiconductor processing is usually too costly to introduce in a well developed process and these tools are usually only employed in process development. Until recently there has been no efficient general analysis to allow real time measurements in the cylindrical and the probe geometry that could separate dimensional changes from material property changes both which effect the eddy current response. The basis for these measurements is that forward electromagnetic problem for conductors can be solved exactly for the test geometry which allows an inverse analysis to be made by relaxing the dimensions and conductivity to values which reproduce the measured response[!]. This transformation is based on analysis developed in Robert Siegfried' s thesis and was explored both by modeling and by physical measurement. This analysis was extended over the last few years to allow a more compact data set to be used to exact physical properties. The main result was that the technique would work on high ·quality physical data on as little as three test frequencies with the penalty being that the analysis is computational intensive. The main result of this work was that a quantitative inverse analysis of eddy current response would allow a general instrument to be used for a variety of measurements rather than being tailored for a specific measurement on a specific material test geometry. The development of these instruments has progressed over the last ten years for monitoring real time solidification, crystal growth[2] and hot seamless tube forming where the use of standards in traditional eddy current is limiting. The specific instrument design was dictated by the requirements of the algorithm for both frequency selection and levels of systematic noise that could be tolerated. Because the data is need for real time information this dictated that the system be a parallel multifrequency eddy current instrument feeding the same sensor. The sensor have to be simple and robust and usually a single loop or several loop of heat resistant wire or tube. This presentation is about using inversion of multifrequency eddy current data in real time to produce dimensional and property data on cylindrical conductors and planar conductors in real time. The object of the measurements is to produce dimensional outputs on measurements of conducting material without having to establish a detailed standard for each material or size tested. The dimensional measurements using eddy currents is complicated by material variations in conductivity due to temperature and composition. Those variations that are not geometric are often not of critical interest and complicate the measurement. The dimensional changes are often of more importance"

Jan 1, 1997

By Ali S. Al-Watban

"In this paper, the room temperature friction and deformation properties of various polycrystalline diamonds; i.e. aggregates of diamond crystals with a ""binder"" phase of cobalt and a polycrystalline coating produced by a chemical vapour deposition processes (CVD) were studied when softer metallic and ceramic sliders were used. The experiments were conducted· at relatively low sliding speeds (10 mm/min), in a vacuum of 10-5 mbr after having cleaned the specimen surfaces by out-gassing them for about 30 min at 800 °C. As a result of increasing the number of traversals significant wear of the coated diamond by softer metallic sliders (aluminium and mild steel) was observed. With ceramic sliders, it is shown that the multiple traversals result in the formation of wear groove. With diamond silicon carbide, high friction and adhesion lead to significant metal transfer which then pre-empts cumulative deformation of the harder materials. With the diamond-cobalt composite, the friction is surprisingly low, there accumulation of wear debris on the tip of the slider but no permanent transfer of metal and the harder specimen is not damaged even after 5000 traversal.IntroductionPolycrystalline diamond is mainly produced as tool materials for the machining of hard metals and composites. Thus, the sliding friction and wear resistance are amongst the most important properties relevant to those applications.During cutting by PCD the wear mechanism is mostly influenced by the chemical affinity between the tool material and the workpiece. For example, diamond based materials are not recommended for machining ferrous materials because carbon is highly solubie in iron: This leads to the graphitisation of diamond particles and hence a high rate of diffusion wear is observed (Brookes and Lambert, 1982)."

Jan 1, 2000

By M. A. Llavona

In Spain, as in many countries, the gold content of jewellery must be officially hallmarked. The official methods of assay are gravimetnc cupellation (a method which is highly accurate but both time and sample consuming), potentiometry, gravimetry and volumetry. In this paper, a procedure for measuring gold by flame atomic absorption in jewellery samples is described. The method is based on the acid digestion of a 40 mg sample. The quantity of aqua regia necessary for dissolving the gold, the ideal weight of a sample in order that the error may be minimal, the effect of the acids employed, the type of material used for its storage, and the necessity of employing a buffer to avoid the ionization of the solution are described. The method proposed is simple, highly accurate and causes little damage to the jewellery.

Jan 1, 1999

By Peter R. Whiteley

"This paper discusses gaseous, solid and liquid injection technologies employed in the aluminium industry. The reasons for injection systems are presented, together with a description of the processes used. This includes a review of injection in crucibles, furnaces, launders, and into special purpose metal treatment boxes. The paper covers both the primary and secondary aluminium industries. IntroductionMany injection technologies are used in the aluminium industry and are in a rapid state of development. This paper provides a review of these systems and presents some thoughts on the possible future application of injection to further improve productivity and reduce losses associated with current applications.The Needs for InjectionThere are many issues facing the process metallurgist in the aluminium industry which call for the application of injection technologies now and in the future. In order to categorise these issues the industry may be broken into primary and secondary aluminium, although some of the concerns occur in both areas.The forementioned issues are variously concerned with improving metal quality, improving productivity, and reducing process problems.The use of injection techniques will be discussed under the heading of solid, gaseous and liquid injections, and the objectives and benefits of each covered briefly."

Jan 1, 1996

By R. Colás, A. Salas, M. P. Guerrero Mata, R. Garza

"The use of the finite element method (FEM) for studying the plastic behavior of metals has become a common practice over the past few years, because it has been proved to be a good means of evaluating processes parameters, and predicting and improving processes, avoiding the traditional trail and error technique. FEM was used via a commercial package (Abaqus) to create two dimensional models to simulate uniaxial tensile tests for a galvanized low carbon steel. The aim of the work was to analyse the plastic behaviour of a distribution of grains on the zinc cover of a set of galvanized low carbon steel samples. A mesh with lines 2 mm apart was drawn on one of the faces of the samples that were deformed at room temperature, at constant crosshead speed, stopping the deformation at different times. Before and during the stops the deformation measurements were recorded for further calculations in an image analyser in order to compare these with computational predictions, as well as validate the numerical model. The results of this study can be extended to other testing conditions, besides the model can be used for obtaining constitutive equations of the materials.INTRODUCTIONThe use of the finite element method has been widely spread during the last two decades in most engineering areas, it is a good means of obtaining approach values at a first try [1-4]. Thanks to the software and hardware developments it is now possible to run huge problems within hours or minutes, making the technique more accessible and useful for the industry.The aim of this work is to create a finite element model for the uniaxial simple tension test, and compare the results with some experimental results, and asses the validity of the model.The plastic behaviour of the materials with FEM has been studied for a long time [5] and there is always something new to learn about this behaviour of the metals in particular, as well as to characterized new materials and alloys [6,7]. The tension test has been used since XIX century to get knowledge of the materials [5,8], it is the test that has been used the most and the only that is fully standardized, although it has the limiting of low strains, it is still the most widely used, especially in the industry.Two different finite element models in two dimensions (2D) have been developed, one called the mesh model and the other one the grain model, attention was focused on the geometry and strain prediction. The model is validated with experimental test, by using the data and directly comparing it with the numerical predictions. All the tests were carried out at the same constant crosshead speed at room temperature. An analysis of the grid distortion using FEM and analytical methods has been carried out."

Jan 1, 2001

By Michel G. Drouet

Dross, a major by-product of all processes involving molten aluminum, forms at the surface of the molten metal as the latter reacts with the furnace atmosphere. It generally represents 1 to 5 wt% of the melt, depending on the process, and contains on average about 50% free aluminum dispersed in an oxide layer. Since aluminum production is highly energy-intensive, dross recycling is very attractive from both the energy and the economic standpoints. The conventional recycling process using salt rotary furnaces is thermally inefficient and environmentally non-acceptable because of the production of salt slags. Hydro-Quebec has developed and patented a new salt-free technology using a rotary furnace heated by an electric arc between two graphite electrodes, called DROSCAR®. A 600-kW pilot plant in operation at LTEE is in use to demonstrate the process. This process provides aluminum recovery rates over 90%, using a highly energy efficient, environmentally sound production method. In 1994, 400 tonnes of aluminum dross were treated in this facility and several tests on various types of dross have also been conducted in early 1995. A report on the results will be presented.

Jan 1, 1995

By Carlos M. Fontes Vieira, Ruben J. Sanchez Rodriguez, Carlos E. Gomes Ribeiro

"In this study we attempted to model a response surface for the apparent dry density of ternary blends of calcitic marble particles. The crushed marble was divided into three granulometric ranges. Simplex network methodology was used to determine the mathematical relation between granulometric ranges and apparent dry density (ADD) of particle mixtures. A highly significant regression was obtained and the simplex network methodology proved to be able to predict high packing particle compositions.IntroductionProblems with packing and expected porosity of particles are common and widely recognized to be important in terms of both fundamental physics and engineering applications. For example, densification of a powder mass is important in the shaping of solids in ceramics, powder metallurgy or composite synthesis; chemical processes such as solid-liquid separation (sedimentation, thickening and filtration), raw material handling (stockpiling, blending, drying, storage and transportation) and pyrometallurgical processes (agglomeration, coke making, ore sintering, blast furnace iron making and new iron making processes) [1,2].Many factors can affect particle packing, but in practice, there are three main factors: (dimensionless) particle size distribution, particle shape and absolute particle size [2].Geometrical and practical methods were developed to predict properties in mixtures. Zou, Gan and Yu [2] present a paper treating a prediction of the porosity of multi-component mixtures of cohesive (fine) and non-cohesive (coarse) particles. Bahiense et al. [3,4] used a simplex network to determine a mathematical model for obtain the optimal compositions between aggregates to a greater packing of a ternary particles mixture. Despite these techniques for predicting particle packing, some studies in particulate composites production don't present clear criteria for choice of particle distribution [5,6]. As the best particle packaging depends on the size, distribution and shape of these particles, individual models can be constructed for specific combinations of these factors."

Jan 1, 2014

By Osarnu Inoue

In Mitsubishi Continuous Copper Smelting and Converting Process, smelting, slag-matte separation and matte converting are continuously carried out in series in three furnaces connected by launders. As a result of this operational configuration of the furnaces, influence of the change in operating condition in upper stream, such as variation of feeds in the smelting furnace, propagates to down stream with some delay. In order to achieve stabler operation, process control algorithm based on the dynamic characteristics of each furnace would be very effective. In this study, the dynamic process simulation model was developed for the smelting furnace. The model consists of two mathematical modules, i.e. the thermodynamic module and the mixing module. The calculation results using operating data obtained at Naoshima Smelter showed good agreement with the actually observed change in temperature, composition and flow rate of matte and slag at the outlet of the furnace. In addition, operation parameters such as reaction efficiency of oxygen and coal, dead volume of melts were estimated.

Jan 1, 1999

By J. Jakumelt

The coupled simulation of mold filling and solidification for an investment casting process presents difficult. This is due to the complexity of the carious physical process and the geometry, To achieve this task in a precise way, the CFD program FLUENT and the casting simulation tool CASTS were closely coupled. Any physical quantity, such as velocities or temperatures, can be exchanged via files after each time step of the simulation. The communication and operating sequence of the programs is controlled by PERL-scripts. This enables the expertise of FLU ENT for flow simulation and the strength of CASTS, for casting simulation to be combined. The coupled simulation was successful applied to an aerospace investment casting. Several physical models were tested for correct treatment of the casting filters, the porosity of the mold and the viscosity of the liquid metal. The results demonstrate that even complex investment casting geometries can be investigated using CASTS-FLUENT.

Jan 1, 2003

By Kenichi Moriyama

The Sumitomo Toyo Copper Smelter started operation in 1971, as an epoch-making smelter at which 9996 of the charged sulfur could be captured. In the past ten years, the amount of treated copper concentrate has-been increased from 1,300 l/ day to 1;900 t/ day by the extremely effective application of the oxygen enrichment for a flash smelting furnace and converter furnace. The sulfuric acid production capacity at the acid plant has also been increased up to 1,900 t/ day corresponding to the increase of copper concentrate smelting rate without installing an additional acid plant. Furthermore the sulfur fixed ratio is kept at a high level, more than 99.996. despite the SO. strength fluctuation of the wide range of 596 to 1396 which is attributed to the combination of flash smelting furnace and, P.S. converter. These -operational results at the? acid plant were achieved by several modifications. and two examples are briefly described as follows; (1) Converter modifications in order to cope with the high SO, strength -gas (2) Introduction of the anodically protection shell and tube type acid coolers In addition. the quality of sulfuric acid has been positively increased. For example, the concentration of iron and arsenic, which were the main Impurities in ?sulfuric acid, was successfully .reduced.

Jan 1, 1995

By Donghai Li, Cheng Pan, Xuewei Lv, Mei Liu, Enguang Guo

"Phosphorite is the main raw material for the production of phosphorus in chemical industry. More attentions have been drawn to the phosphorite concentrate beneficiated from low-grade phosphorite due to the decrease of high-grade phosphorite. For another, the agglomeration process is necessary for the phosphorite concentrate after beneficiation process. The mineralogy and sintering process of the phosphorite from Leshan, China were studied in this work. The result shows that the main phases in the phosphorite are fluorapatite (Ca3(P04)3F), hydroxylapatite (Cas(P04)3(0H)), quartz (Si02) and dolomite (CaMg(C03)2). The yield of the sinter increases with increasing the carbon dosage to 7%, and then decreases when it continuously increases. Similar regularities were found for the strength and the content of P20s of the sinter. The best strength of the sinter as well as its P205 content were discovered when the carbon dosage is 7%. Fluorapatite ((Ca5(P04)3)F) is the main phase in the sinter while fuggerite (Ca2MgSiz01) is the main binder phase.IntroductionPhosphorus is an important foundation industrial raw material of national defense industry, food industry and chemical industry, where it is used to produce red phosphorus, phosphonium hydroxide, phosphorus pentasulfide and phosphoric anhydride[l-2]."

Jan 1, 2013

By George E. Cook, Peter C. Levick, Francis M. Wells

"Relationships are established between the peak current of a pulsed, rectangular current waveform and the pulse current duty cycle under conditions of constant arc power. By appropriate choice of these interrelated parameters, it is shown that the arc pressure may be varied over a wide range even though the arc power is held constant. The methodology is suggested as a means of countering the effect of gravity in 5-G welding, while maintaining constant heat input to the weld. Combined with appropriate penetration sensors, the methodology is additionally suggested as a means of controlling penetration. IntroductionBoth low frequency pulsation (typically 0.1 - 20 Hz) and high frequency pulsation (> 100 Hz) are often used with GTA welding. Benefits derived from pulsing include improved weld quality and better process control.Advantages cited for low frequency pulsation include claims that the total heat input to the weld is reduced, the weld bead size is reduced, weld penetration is increased, grain size is reduced, porosity is reduced, weld distortion is reduced, thinner materials can be welded successfully, and the tolerance of the welding process to material and production variations is improved (1-16). With respect to the latter, pulsing aids control of weld bead penetration by increasing the tolerance to component dimensions and joint fit-up. It also reduces the sensitivity to disparities in heat sink, and it helps to reduce distortion in thin section material or distortion attributable to poor clamping. In certain materials it can reduce the sensitivity to surface oxides and to cast to cast compositional variations (16). Low frequency pulsation is also useful in welding thick materials mated by a prepared groove. In these applications the low frequency pulsation is used to obtain proper sidewall fusion as the tungsten electrode is weaved from one side of the groove to the other. In this case the weld current pulsation is synchronized with the cross seam weave pattern, such that the high current pulses occur at each sidewall. This type of low frequency pulsation has proven particularly useful in welding with narrow width groove designs (17-21). Finally, Ecer (22) has shown that the pulsed current arc is more stable against magnetic disturbances than an equivalent constant current arc."

Jan 1, 1986

By Giuseppe Bonifazi

Minerographic and petrochemical research was carried out on crude ore samples from the complex polymetallic Zn-Cu-Pb sulfide ores of , Sotiel in Huelva province west of Seville, Spain. A computerized image analysis system was used to define the structural and textural characteristics of the ore to obtain information on the degree of liberation of the different minerals (i.e. pyrite, sphalerite, galena, chalcopyrite). Techniques of digital multispectral image processing of reflected light microscopy images were used. Particular structures were identified by means of mineral associations and numerically quantified to obtain useful information for designing beneficiation processes.

Jan 1, 1991

By T. T. Chen

Canadian Electrolytic Zinc has developed the "cement-stabilization of jarosite residue" Garofix) process to stabilize zinc plant jarosite residues for disposal. The cement-jarosite mixtures cure quickly, have good mechanical strength and low hydraulic permeability; they possess good neutralizing capability and chemical stability. In the jarofix process, Portland cement partly decomposes jarosite to form (SO4MOH)12.nH2O, ferric hydroxide and an alkali sulphate. Approximately 40% of the jarosite reacts in mixtures containing 30% cement. The jarofix product also contains various Ca-Al-Fesilicate-sulphate-hydrate phases and gypsum, which bond the individual particles together imparting mechanical strength and reducing the permeability. Most of the reactions take place in the first few hours of curing. Ammonium jarosite is slightly more reactive than sodium jarosite but releases NH3 gas; water-soluble Zn in the jarosite residue is stabilized in the cement reaction products.

Jan 1, 1996

By H. H. Kellogg

The thermochemical properties of molten solutions in the binaries Cu-Bi, Cu-Sb and Bi-Sb, as well as the ternary Cu-Bi-Sb, have been correlated and described in mathematical form by use of the associated species concept combined with the three-suffix Margules equations to account for the non-idea 1 behavior of the solution species. Agreement of the correlations with experimental data is excellent (relative deviation of less than 2.5% in the values of the activity coefficients) in most cases. The correlations are applicable over the entire composition range of each system, and in the temperature range 700°-1250°C.

Jan 1, 1985

By Ing. R. Geutskens

Zinc-bearing dust from blast furnace gas-scrubbing operation is a well known environmental health problem in the iron and steel industry. Hoogovens IJmuiden (The Netherlands) developed processes to solve this problem. 90% of the zinc is concentrated in 25% of the dust (the fine particle fraction) by means of slurry-hydrocyclonage. The resulting fine particle fraction is treated in a pressure leaching process. Hydrochloric waste pickle liquors have been used as a leaching liquor. By creating optimal leaching conditions more than 98% of the zinc is brought into an acid' solution, whereas nearly all the iron remains in the residue. The zinc can be separated selectively from the solution. The residue can be reused in the blast furnace (via the sintering plant). In the article information is given about the research work and -results of the pressure leaching process as well as the possibilities of the process for solving comparible problems.

Jan 1, 1990

By I. Gaballah

The kinetics of oxychlorination of pure CI-2O3 with C12+02 gas mixtures is studied by ThermoGravimetric Analysis (TGA) in non and isothermal conditions up to 1000 °C. The reaction of Cr2O3 with C12+O2 chlorination gas mixtures starts at about 550 °C and the reaction product is CrO2Cl2. The oxychlorination rate of Cr2O3 is almost proportional to the chlorine partial pressure and to the fourth root of the oxygen partial pressure in the chlorinating gas mixtures. At 800 °C, the maximum chlorination rate is obtained using a gas mixture having a C12/O2 ratio equal to 4. The apparent activation energy of the oxychlorination reaction of Cr2O3 is 87 and 46 kJ/mol at temperatures lower and higher than 650 °C, respectively. By varying the chlorine and the oxygen partial pressures of the oxychlorinating gas mixtures at temperatures higher than 550 °C, it is possible to extract selectively the chromium chlorinated compounds from the chromium bearing materials. a. Laboratoire Environnemem et Minéralurgie, BP 40, 54501 Vandceuvre, France, b. École Nationale Supérieure de Géologie, Rue du Doyen Marcel Rouhault, 54501 Vandoeuvre Cedex, France, c. Centre National de la Recherche Scientifique, 3 rue Michel - Ange, 75794 Paris Cedex, France, d. Insritut National Polytechnigne do Lorraine, Rue de la Font de Haye. 54501 Vandceuvre Cedex, France.

Jan 1, 1997

Halide media are considered as suitable alternatives to cyanide for extracting gold and have received much attention lately. The extraction of gold by iodide-iodine mixture has been patented and briefly studied without much attention given to the fundamental aspects of this process. This paper reports recent studies on the kinetics of gold extraction in iodide media, using hypochlorite, iodine and hydrogen peroxide oxidants. The kinetics of dissolution was found to be dependent on the solution pH, reagent concentration and most importantly the iodide- oxidant molar ratio. Without proper control, the gold surface can be passivated due to the formation of solid products. Thermodynamic modelling was carried out to determine the equilibria governing the dissolution process. The formation of AuI solid was also identified as a possible product during the dissolution.

Jan 1, 1992

By Robert Mensah-Biney

The loading characteristics of gold-tetrabromide species onto a weak-base resin, PAZ-4, were evaluated using a completely mixed batch reaction system. The time. to attain equilibrium loading was substantial, on the order of several days. However, abnormally high loading capacity values of 15 to 20 meq Au/g dry resin (1000 to 1300 kgAu/tonne) were attainable in 24 hours. The loading of gold-bromo species by P AZ-4 was visualized as consisting of two processes: (i) the rapid adsorption of [AuBr4r species by simple ion exchange mechanisms and (ii) reduction of [AuBr4r to metallic gold by an electron exchange mechanism. Elution of gold from loaded resin was feasible provided the resin had not been loaded to high capacity values. If loading is carried to high levels, complete elution becomes impractical and the gold may be recovered by ashing the loaded resin. For the high loading capacities observed this option would appear to be cost effective.

Jan 1, 1991

By S. Goto, S. Yamashita, K. Hata

"The average consumption of zinc dust for purification of zinc leaching solution is more than 25 kg per tonne of electrolytic zinc. Authors proposed to remove impurities in zinc leaching solution by electrolysis instead of cementation by zinc dust. The effects of copper and arsenic on electrodeposition of cobalt in zinc sulfate solution were examined fundamentally using a potentiostat. It is observed that cobalt is easily deposited on a cathode at potentials under -480mV with arsenic. Mesh cathodes of stainless steel and lead anodes were used in a small electrolytic cell and the electrolyte was circulated through 'the electrodes during the electrolysis. The effect of electrolyte circulating rate , pH, cell temperature, current density, and concentration of copper and arsenious ions on removal of cobalt were studied. Removal of Cd and Ni are also examined.IntroductionCobalt is difficult to remove from a zinc leaching solution by a simple cementation reaction using zinc dust. However, it is easily precipitated by zinc dust with co-addition of arsenious or stibinous oxide and copper ion in practical operations(I), (2), (3). The consumption of zinc dust for purification is 16-152 kg per tonne of electrolytic zinc and 65% of this zinc dust consumption is used for cobalt purification. The average retention time for purification of cobalt is 4.5 hours (1). A more economical process for purification is desired instead of zinc dust. The mechanism of this acceleration effect of copper and arsenious oxide on removal of cobalt are explained by the authorsas follows; a galvanic cell is formed between zinc dust and copper which is precipitated firstly, and cobalt is reduced on copper as Co-As alloy by this galvanic current(2). This assumption was confirmed in experiments using a galvanic cell with a membrane( 4). The following galvanic reactions occur between zinc dust and copper precipitate; on copper cathode,2H+ + 2e -= Hz(g)2Coz+ + 2HAs02 + 6H ++ 10e -= 2CoAs + 4HzOand on zinc anode,Zn = Zn = 2+ 2e –This galvanic current becomes almost zero in a zinc sulfate solution which includes only cobalt and no arsenious oxide, and cobalt is weakly deposited on a copper cathode. Cobalt deposition requiers a high degree of cathodic polarization. However, the galvanic current flows almost constantly in an electrolyte, which includes arsenious oxide, and the cathode potential is kept at about -500mV. Cobalt is precipitated as Co-As on the cathode.The authors assumed that cobalt may be removed by electrolysis at a very low current density instead of zinc dust and the following experiments were carried out. In the first part of this paper, the fundamental experiments using a potentiostat on the behaviour of cobalt in zinc sulfate solution are described. In the second part, the results of the electrolysis are reported. Cobalt is removed by a new type of an electrolytic cell, which was desigoed to <Iccelerate the rate of deposition of cobalt. The removal of cadmium and nickel in a solution which includes cobalt and arsenic are also examined."

Jan 1, 1997