Search Documents

By R. W. Smith

The flotation response and electrokinetic behavior of Chlorella vulgaris and Chlamydomonas sp . when using dodecylamine hydrochloride as collector were studied in the absence and presence of Pb(I1). Also, an electron microscopy study was made of the algae after Pb accumulation. The flotation behavior of the organisms after Pb accumulation appears at least partially related to the location of the abstracted Pb onto or into the organisms.

Jan 1, 1991

By Jose Antonio Silva Souza, Emanuel Negrao Macedo, Iara Ferreira Santos, Otacilio Dias, Luis C. A. Venancio

"The Bayer process alumina production generates 0.7 to 2.0 ton of bauxite residue (BR) and an average of 1.0 ton of C02 per ton of alumina produced. The direct reaction of exhaust gases to reduce the alkalinity of the BR may allow a triple gain: improve the storage conditions, open a range of new applications for BR and sequester from 16 to 102 kg of C02 per ton of alumina. This paper shows the measurement of C02 captured during a pilot scale reaction of a suspension of BR in water with flue gas, produced from direct oil burning. Three different types of reactors were used: one spray tower and two packed columns. The inlet and exhaust gases were analyzed using electrochemical and non dispersive infrared sensors. The pH of the suspension was monitored during and after the reaction to evaluate the buffer effect.IntroductionAlumina is mainly produced from refining bauxite using the Bayer process. This process consists of extracting the aluminum oxide through its dissolution in caustic soda and the separation of insoluble residues. The remaining bauxite residue (BR) although not particularly toxic, poses risks to the environment due to the large volume and its reactivity. According to the latest technology it is stacked on sealed deposits specially constructed. To date the vast majority of the bauxite residue that has been produced was stored and were not incorporated into any existing production processes to significant amount with exception to some extent of the sintering process residue at Russia and China [l]. On the other hand, a strategic goal was set in 2010 by the Bauxite and Alumina Committee of the International Aluminium Institute (formed by the majority of the industries) to reuse 20% of BR in 2025 [2]. China has established the same objective for 2015[1].This paper is a sequel of two others presented at TMS 2010/2012 [3-4] and describes:• A pilot plant assembled for testing the carbonation process of BR with gaseous effluents similar of the refining process.• The methodology used to measure the amount of C02 captured during the carbonation process.• The results of the carbonation process, the C02 capture and the buffering effect of the pH.This process is studied in order to consume minimal power and use the effluent as efficiently as possible in different reactors from low cost to ultra high performance. The process is designed for minimizing the power consumption and equipment investment."

Jan 1, 2013

By Thomas F. Hagelien, Stein Tore Johansen, Svend Graadahl

"The dross appearing at the surface of an aluminium melt consists, in addition to aluminium, of mainly oxide films but also other non metallic inclusions. The inclusion particles are often poorly wetted by the metal and accumulate at the metal surface. However, in impeller stirred reactors for melt refining stirring may be exceedingly strong and inclusions accumulated at the metal surface may be entrained into the bulk metal. Concerning inclusion removal, such effects may deteriorate the effectiveness of melt refining.In this paper we investigate the surface entrainment of buoyant particles in a water model of an impeller stirred refining reactor. Under varying operating conditions the particle concentration in the bulk liquid is measured by laser light attenuation. The observed entrainment of particles is explained theoretically and metallurgical consequences of the findings are discussed.IntroductionIn the modern cast house continuos metal refining and casting is well established. One of the most distinct advances in metal refining has been the introduction of degassers which introduce bubbles into the metal. The bubbles act to remove dissolved elements together with nonmetallic inclusions. In the most efficient degassers stirring by different kind of rotors is applied to increase mass transfer and inclusion removal. However, it has been known from experience in the cast house that for violent stirring the concentration of some types of inclusions may increase from inlet to outlet of the reactor. This phenomenon has so far been attributed to production of dross in the reactor due to oxidation of metal. Due to removal of inclusions by bubbles the dross layer inside the reactor must be expected to increase in thickness by time. The removed inclusions and produced oxides due to strong stirring is removed at regular interval by the operators of the degasser. However, a fundamental understanding of the phenomena which take place in and below the dross layer is lacking. In the present work we add buoyant particles to the surface of a refining reactor and investigate how these particles may be entrained and dispersed into the liquid depending on the flow conditions. Similar investigations for gas stirred ladles have been performed by Taniguchi et al [1] who found that below a critical gas flow rate all particles are trapped at the surface. Above the critical gas flowrate the surface velocity are sufficiently high to entrain the particles. In this situation it looks like all particles are ready for the melt and a balance between terminal velocity and turbulent dispersion determine the concentration in the liquid."

Jan 1, 1999

By N. Ono, H. Fujiwara, S. Otsuka-Yao-Matsuo, T. Omata

"The electrical conductivity and the oxygen release and uptake of sintered indium tin oxide (ITO) upon heating and cooling under the condition that P(O2)/P*=4.9X10-4 (P* = atmospheric pressure), :when P(O2) was approximately 50Pa, were simultaneously measured by using a closed-system oxygen-gas analyzer. A large increasing in -the electrical conductivity - upon the· oxygen release detected in 1130<T<1273K was observed. Quantitative relationship between the number of oxygen atom released from ITO and the conduction electron density in ITO was studied. The release of one oxygen atom agreed with the generation of two conduction electrons, i.e., the efficiency-carrier-generation for the oxygen release from the ITO was almost one. The oxygen atom released must be originally at the quasi-anion site in the C-type rare-earth lattice, i.e., the 16c site in the space group of Ia3, as an interstitial excess oxygen, Oi"". It was concluded that the mechanism of conduction carrier generation and compensation upon SnO2 doping into In2O3 can be expressed by the defect equation, 2SnO2-2Sn1n +2(1-z)e'+zOi""+3O0 x+(l-z)/2O2, and that the oxygen release bringing the increase in the electrical · conductivity can be expressed by the equation, Oi""-1/2O2+2e'.IntroductionThin films of SnO2-doped In2O 3 (generally designated as ITO, for indium tin oxide), which have high optical transparency for visible light and high electrical conductivity, have been used as transparent electrodes for liquid crystal and electrochromic displays. It is well known that the electrical conductivity of ITO depends on the content of SnOi-dopant and oxygen. However, a reliable relationship between electrical conductivity and the absolute oxygen content in ITO has yet to be reported. The elucidation of such a relationship is one of the important clues for realizing extremely highly conductive ITO electrodes."

Jan 1, 2000

By F. J. Tavera

Recently Jameson cell technology has gained acceptance in both mineral and non-mineral applications, mainly because of the high degree of contact between the involved phases in the downcomer, fact that leads to an efficient collection performance. The key for an acceptable performance of the downcomer operation depends on the stability of the jet zone, the mixing zone, and the plug flow zone. in this work the behaviour of the downcomer of a laboratory Jameson cell is reported, as a function of gas hold-up, superficial gas velocity, superficial liquid (or slurry) velocity, and length of the operating zones. The presence of carbon is evaluated in terms of its effect on the gas hold-up and bubble surface area flux. The influence of the downcomer operating conditions on the gas-liquid disengaging cell is presented. Keywords: Jameson cell, downcomer, gas hold-up, mixing, plug flow, conductivity.

Jan 1, 2003

By R. Rajiv Srivastava

Brine leaching of a TSL processed zinc-plant residue (ZPR), containing 83.3% anglesite, has been investigated for lead recovery. A prior treatment of ZPR in acid solution did not show any significant effect of acid concentration in zinc dissolution. The study reveals that the solubility limits of PbCl2 is as an important factor with respect to the brine concentration and pulp density in lixiviant. The dissociated sulfate ions hinder the formation of PbCl2 on prolong leaching (above 60 min), while exhibiting the reverse solubility of lead as PbSO4. A 3-step leaching process could yield >92% leaching efficiency when the parameters were maintained as: 10% pulp density in a 250 g/L NaCl solution, 80 °C temperature, and 60 min time. This leaves the insoluble ferrite, sulfides and oxy-sulfates of zinc and lead in the final residue as revealed by the XRD characterization.

By G. A. Brooks

Reactive ball milling has been proposed as a new route for producing carbides from oxides through carbothermic reduction. It is claimed that through milling oxide and graphite powders to the nanoscale, amorphous materials with very high reactivity are formed, that allows for easy transformation of the starting material to another in a subsequent thermal process. Thermodynamic analysis of this proposed process suggests that unless particle sizes are reduced to below 10nm, then reduction in size alone cannot explain a change in the stability of oxides. Even with nanometer scale powders, the products formed need to have a lower total surface energy than the reactants before a chemical change will occur. A model developed by Bhattacharya and Arzt for predicting localized temperature changes based on milling conditions suggest that temperature rises less than 400 K will be achieved in laboratory conditions. The present calculations suggest that nanometer scale powders formed through high speed ball milling (> I 0 mls) may have undergone a process analogous to melting and quenching. Further work is required to fully distinguish these features.

Jan 1, 2003

By Reinhard F. Dobner

After a short introduction about the HK&apos;s secondary smelter process and explanations concerning the anodes produced the regulation and buffering of bleed will be described and the principles of technology for processing of bleed will be presented. Modernization aspects, especially installation equipment, and the different types of evaporators used will be described in detail. Besides, experience made with acid-resistant materials - stainless steel, enamel, plastics - will be reported. Environmental aspects such as water and air pollution will be outlined briefly and in conclusion the quality of products obtained as well as quality control will be presented.

Jan 1, 1997

By Gerardo R. F. Alvear, Alistair S. Burrows, Stanko Nikolic, James S. Edwards

"The copper ISASMELT™ process has evolved over more than a quarter of a century and can now be considered mainstream, with single furnaces producing more than 340,000 tpa of copper in matte. The next evolutionary step for the ISASMELT™ technology is to directly challenge the dominance of the Peirce-Smith converter by the implementation of a continuous copper converting process on a commercial scale. The present work describes the status of the continuous converting process – ISACONVERT™.Continuous converting has been performed successfully in pilot scale ISACONVERT™ plants, where operating results have verified the predictions of fundamental research and reinforced the underlying confidence in the design parameters. From a matte feed rate of 250 kg/hr, a calcium ferrite slag and a low-sulfur blister have been produced. Design is now complete for the first commercial application of the ISACONVERT™ process. The present paper describes the status of ISACONVERT™ by summarising results of recent pilot plant tests and describing the first commercial plant design."

Jan 1, 2009

By Carlos G. Levi

"The principles involved in the rapid solidification of supercooled metal droplets and existing numerical models are reviewed. The limitations of these models for studying alloy solidification are discussed in terms of the effects of alloying elements on the relevant heat transfer, thermodynamic and kinetic parameters. Analysis of the enthalpy-temperature diagram reveals how the extent of rapid solidification is hindered by the addition of solute. A solidification model incorporating thermodynamic and kinetic relationships into a Newtonian heat transfer formulation is used to analyze the thermal history and concurrent solute redistribution during recalescence. IntroductionIt is well acknowledged that microstructural evolution during solidification is intimately associated with the thermal history of the liquid-solid interface. Structural parameters like chemical homogeneity, morphology and size scale of the phases can be related to the interfacial temperature and velocity. These in turn change by the interaction of the interface with the thermal and solutal fields in its neighboring environment.The thermal history of powder solidification can be conveniently described on the enthalpy-temperature diagram for a pure metal shown in Figure 1 and discussed in detail elsewhere [1). A system can conceivably follow any path of decreasing enthalpy in the supercooled region of Figure 1 (~ < 1), starting at the nucleation temperature (oN < 0) on the liquid line, and ending at some point on the solid line. A vertical path represents an isothermal transformation, of which a limiting case is solidification without significant barriers to nucleation and growth, with the interface at the equilibrium temperature TM (path one in Figure 1). A horizontal path represents isenthalpic (or adiabatic) solidification whereby the system could form an amount of solid proportional to the initial supercooling of the liquid without further heat loss to the environment. A liquid that can completely transform to solid under adiabatic conditions is said to be ""hypercooled""."

Jan 1, 1986

By Tatsuo Inoue, Marnoru Mizuno, Toru Matsuoka, Kenichiro Abe

"The distribution of particle reinforcements and the residual stresses in metal matrix composite are simulated by considering the motion of particles and the solidification of matrix associated with temperature variation in the course of centrifugal casting process. In the analysis, a heat conduction equation including heat generation due to latent heat and also mechanical work as well as heat input from successively accumulated element on the air/metal boundary is solved by the finite element technique. After the movement of particles subjected to the centrifugal force and drag are analyzed, the residual stresses are elucidated by using a unified constitutive equation which can describe the behavior both in liquid and solid states. Results of the simulation are compared with the experimental data, and the validity and utility of the procedure is confirmed.IntroductionProcess simulation of metal matrix composite (MMC) is required to determine the mechanical properties of the materials and the efficient process and conditions of the fabrication. Properties of bonding matrix with reinforcements in the course of the processing, for example, affects the strength of the composite, while the final costs of the productions depends on the efficiency of fabrication. Therefore, the simulation of processing of MMC is prerequisite in order to clarify the mechanism to fabricate the material which has been expected to maintain high mechanical properties and the cost performance.Some techniques of processing MMC[l] have been developed so far. Among them, the centrifugal casting technique [2-5] has many attractive characteristics as processing of the MMC: The centrifugal casting technique makes the reinforcements such as particles or short fibers to be distributed in the radial direction by the centrifugal force acting on reinforcements due to the difference in density between reinforcements and matrix. Since the distribution of reinforcements affect the mechanical properties of MMC, the composite with functionally gradient mechanical properties can be provided by the centrifugal casting technique."

Jan 1, 1994

By Andre Thess, Jurij Kolesnikov, Christian Karcher

"During the motion of molten metal under external non-homogeneous magnetic field the induced electric currents produce Lorentz body forces in the opposite direction to the flow, causing deformation of velocity profile, vortical structures, and adding a pressure drop across the magnetic field. The same Lorentz force acts on the magnet system. We analyse these phenomena in metallurgy, namely, at molten metal pumping, stirring, continuous casting, and in contactless Lorentz force flowmeter (LFF). Experiments with liquid metal alloy, including direct measurements of Lorentz forces, velocities, and flow structure visualization were conducted. Based on this study the industrial prototypes of LFF systems were designed, built, and tested in a secondary aluminium production. A molten metal flow at high temperature (700°C-850°C) can be effectively measured. The LFF technique method allows measuring precisely the volumetric and mass flow rates linked with Lorentz force, and the mass accumulated within production process.IntroductionThe present work demonstrates the development of a non-contact electromagnetic system for measurement of a liquid metal flow. Our approach embodies the Lorentz force velocimetry [1-6] technique which is physically based on a measurement of the force acting on an external magnet system the field of which interacts with the flow. This force is exactly equal to the braking Lorentz force induced in a flowing conducting fluid. In what follows a measurement system based on the Lorentz force velocimetry principle. For monitoring and controlling the production process it is desirable to measure continuously the mean velocity of the liquid metal in order to deduce the mass-flux and volume-flux from it. The form and dimensions of the magnet system are in agreement with geometry of the operating channel. We consider here a version of magnet systems in a composition of permanent magnets.The LFF possesses a number of advantages which make it attractive for the measurement of flows in hot and aggressive melts. The relation between the measured force and the desired volumetric flow rate is often linear (for closed channel or at a constant liquid metal level in open channel). For the converting of the measured force into volumetric and mass flow rate we should know the calibration factor. The calibration factor is independent of the viscosity and density of the liquid metal and depends only on the magnetic field magnitude, the electrical conductivity of the liquid metal and the geometry of the channel in which the liquid metal flows. In aluminium production, for instance, there exists a number of standard alloy compositions having welldefined conductivities for which the calibration factors can thus be tabulated. We consider below consistently effects in liquid metals arising under influencing magnetic field, construction principle of LFF and measurements of metal flows at industrial conditions."

Jan 1, 2011

By P. Mazumder

Numerical analysis is carried out to examine the effects of thermosolutal convection on the formation of complex patterns in directionally solidified binary alloys. A finite-difference analysis is used for dynamic modeling of a two-dimensional prototype of Vertical Bridgman system that takes into account heat transfer in the melt, crystal, and the ampoule; melt flow and solute transport. Actual temperature data from experimental measurements are used for accurately describing the thermal boundary conditions. A range of complex dynamical behavior is predicted in the melt flow due to flow transitions and this is found to be directly related to the spatial patterns observed experimentally in the solidified alloys. The model is applied to single phase solidification in the Al-Cu and Pb-Sn systems to characterize the effect of convection on the macroscopic shape of the interface. The application of the model to hyper-peritectic alloys in the Sn-Cd system shows that the presence of oscillating flow can give rise to a novel convection induced microstructure in which a tree-like primary phase in the center of the sample is embedded in the surrounding peritectic matrix.

Jan 1, 1999

By J. M. Figueiredo, A. R. Silva, M. C. Coelho, R. M. Guedes

"With the continuous depletion of conventional sources of non ferrous metals, low grade complex sulphide deposits have been regarded as an alternative source of supply of zinc, copper, lead and also silver and gold. In spite of very important efforts in research in this domain it is still a long way to go in order to achieve reliable technological solutions both in ore dressing and metallurgical treatments which enable these complex materials to be regarded as commodities. The RLE process to treat bulk concentrates such as those obtained from portuguese ores, under development at LNETI and QUIMIGAL premises, combines experienced unit operations and innovative configurations and is an effective contribution for the future use of these materials. The results of three years of experimental work and main conclusions on the overall feasibility of the process are presented.IntroductionIn spite of the very well-known problems associated to the development of complex sulphide deposits, the economical importance of the enormous reserves of these materials, widespread all over the world, is so evident that its future use as source of supply of basic and precious metals and other minor elements is almost unavoidable.Critical steps in that direction are mainly concerned with favourable economical conditions and sound metallurgical practice for the treatment of nonferrous metals concentrates, which are nowadays still considered as non-conventional raw materials.Since Portugal is one of the countries where these orebodies occur more extensively, Portuguese authorities have dedicated since the beginning of the seventies, a big effort on finding economical and reliable processes for the recovery of copper, lead, zinc and also gold and silver, and for this purpose, several alternatives on ore dressing and metallurgical processing were studied, however without full sucess.Five years ago LNETI, a research laboratory, and QUIMIGAL, a chemical company, both state owned, decided to launch a research programme for the development of a metallurgical process for the treatment of NFM concentrates, which received financial support from NATO under its programme ""Science for Stability""."

Jan 1, 1988

By Micheal McCarter, Biswadeep Saha, Meenakshisundaram Ramanathan, Chai Ren, Sivaraman Guruswamy

"Three-point bend test was used to study the effect of electrochemically charged hydrogen and magnetic field on the embrittlement behavior of AISI 4340 steel in this work. Acoustic emission signals were collected during the test and analyzed to gain an understanding of the crack growth process. Cathodic hydrogen charging in AISI 4340 steel resulted in drastic reduction of ductility and strength. Hydrogen embrittlement was characterized by a change in fracture surface from dimpled ductile surface to quasi cleavage type fracture. No significant effect was observed in the presence of magnetic field. The acoustic emission signals collected during the test provided an indication of the extent of energy released during the crack growth process.*Support of this work by NSF through the award DMR-0854166 is gratefully acknowledgedIntroductionGrowing interest in hydrogen as a clean fuel and use of steel and other iron based alloys for storage and transport of hydrogen has led to extensive study of the influence of hydrogen on the mechanical properties of these alloys. Hydrogen interaction with materials may result in degradation of mechanical properties. Embrittlement, either ductile or brittle, can be described as a condition which decreases the toughness of the material or increases the ease of crack growth. Hydrogen present even in few parts per million is sufficient to embrittle steels at room temperature [1]. In most cases, hydrogen exposure is through an electrochemical process in aqueous corroding environments. In such cases, they are prone to be exposed simultaneously to electrically induced magnetic fields, which are also believed to affect the hydrogen embrittlement susceptibility of these alloys. So it is also important to study the effect of magnetic field on the hydrogen embrittlement susceptibility. Hydrogen solubility by electrochemical process in aqueous solution is proportional to square root of current density for iron. [2,3]"

Jan 1, 2012

By Keith Rozenburg, Ivar Reimanis, Jesus Chapa

"Experiments in conjunction with finite element analysis (FEA) are used to better predict crack propagation in ductile/brittle graded composites. Cu/W graded composites were fabricated by stacking discrete layers of powders and hot pressing. Crack propagation was studied by loading a notched beam in four point bending and examining the subsequent crack deflection. The notch was placed within a particular composition of material, oriented in a direction perpendicular to the gradient direction. Elastic and plastic mismatch between different constituents in the graded composite induce shifts in the phase angle of loading, thereby causing the crack to deflect. FEA indicates that thermal residual stress has a negligible ""influence on the crack path. The combined FEA and experimental results suggest that elastic mismatch is the single most important material property-related factor in controlling crack deflection.IntroductionMany 'applications would benefit or may even require materials whose performance or functionality must vary spatially. While the concept of functionally graded materials has existed for nearly one hundred years, it is only within the last fifteen years, that research efforts have focused on generic issues in processing, properties and performance of functionally graded materials. This generic research effort has brought together scientists and engineers from traditionally different disciplines to address common issues with graded composites. The current work focuses on issues in structural joining. A wide variety of applications require some type of joining between dissimilar materials. Examples include thermal barrier coatings for turbine engines, wear resistant coatings for tools, ceramic engine parts, and microeiectronic devices, as well as many traditional applications involving metallic welding. The joints may either be discrete, involving a single interface, or they may· be spatially graded in composition. Extensive theoretical and experimental information exists on discrete bimaterial and trimaterial joints, but relatively little exists for graded joints or composites"

Jan 1, 2000

By Saskia Duyvesteyn

Small ion-exchange resin particles provide faster mass transport than conventional-sized beads. Unfortunately, high pressures are needed to pump solutions through columns containing small resin particles. It is preferable to disperse fine resins in the solution being treated. Screens have been used to separate the resin from the solution, but screens do not retain fines created by attrition of the resin particles. We have developed a separation process based on flotation, which relies on the surface characteristics of the resin, instead of size. The resin is conditioned with a surfactant before dispersing it in the solution to be treated; this increases the hydrophobicity of the resin sufficiently for froth flotation, without introducing surfactant directly into the solution. The surfactant adsorbed on the resin surface does not appear to impair the kinetics of metal ion removal. A flowsheet for an effluent treatment process is proposed; other applications are clearly possible.

Jan 1, 1998

By Jinrong Zhang

Fresh cyanide solution was treated by near UV light with the assistance of photocatalysts. Cyanide was oxidized mainly to cyanate. Photochemical behavior of the anatase and rutile forms of TiO2 was compared. It is found that anatase has a much higher activity than that of rutile. Bubbling of oxygen in the reactor showed no appreciable increase in cyanide oxidation, implying that the reaction is surface controlling. A very interesting finding in this study is that a natural mineral, ilmenite, can replace chemical titanium dioxide as the photocatalyst. Ilmenite without any treatment did not catalyze the oxidation of cyanide. But, after being roasted at high temperature with oxygen, the oxidation was enhanced dramatically. The mechanisms of photooxidation of cyanide and the roasting of ilmenite are proposed.

Jan 1, 1991

By K. C. Liddell, B. Dou

"In producing multilayer films for magnetic data storage applications, the interfacial roughness must be controlled in order to obtain high quality materials with reproducible properties. The present understanding of the effect of roughness on multilayer properties such as magnetoresistance is sketchy, particularly for electrodeposited films. In this study, quantitative atomic force microscopy was used to characterize surface morphology before and after Ni films of varying thickness were deposited on Cu cathodes. A multinuclear multilayer growth mechanism is supported by both the deposition conditions and the AFM images, and a multinuclear multilayer model gave a satisfactory fit to measurements of the density of peaks on ultrathin Ni films between 0 and 8 nm thick.Alloys consisting of alternating ferro-and paramagnetic layers have begun to replace Permalloy in the sensing elements of magnetic hard drives. The operating principle of the new read heads is giant magnetoresistance (GMR), and they are smaller and more sensitive than previous inductive heads (1). Experimental studies on sputtered multilayers have emphasized the role of layer thickness and interlayer roughness in determining the presence and size of the GMR effect (e.g., 2,3,4,5,6). However, roughness in electrodeposited multilayers has rarely been measured, even though read heads are likely to continue to be made by this process for some time to come. Cu/Ni can be considered a model system for understanding processing-microstructure-property relationships in electrodeposited multilayers and is the most widely investigated of all such systems (e.g., 7,8,9,10,11,12). Nevertheless, the roughness of the interfaces in Cu/Ni electrodeposits has not been systematically characterized by quantitative techniques, and in all studies reported so far, the total deposit thicknesses were considerably greater than the 10 nm needed for most sensor applications (1). The thinnest such films were about 50 nm (7)."

Jan 1, 1998

By M. H. Lin, S. Kou

"Czochralski pulling has been a very widely used process for growing single crystals for electronic and optical applications. Dopant segregation can significantly degrade the quality and yield of single crystals. We have used the floating-crucible technique for segregation control in Czochralski pulling, our floating crucibles being unique in that the melt passageway is very long. As such, dopant back diffusion is suppressed, even when the cross-section of the melt passageway has to be relatively large to help the melt go through or when the pulling rate is very low. Single crystals of model materials K-doped NaN03 and Pb-doped Sn have been grown with very uniform dopant concentrations. IntroductionSingle crystals of electronic and optical materials are often doped with a small amount of impurities to develop required physical properties, e.g., P and B in Si for n-type and p-type semiconductors, respectively, and Nd in YAG lasers. Dopant segregation can result in nonuniform physical properties, thus degrading the quality and yield of the resultant crystals.Figure 1 is an illustration of the conventional Czochralski process for crystal growth. Several techniques have been developed to control dopant segregation in Czochralski pulling, including the push-pull technique (1,2), the drop-pull technique (3-7) and the floating-crucible technique (8-13). The first two techniques require precise sensing devices and feedback control systems to keep the feeding rate equal to the pulling rate. The last technique is simple since the feeding rate is automatically equal to the pulling rate."

Jan 1, 1994