Search Documents

By Abbas H. Mirza

Arsenic and selenium are found in low quantities together with other metals in several nonferrous metallurgical process streams such as scrubber blowdown solution, acid plant wastewater, gas cleaning plant water, etc. Normally, the other metals in these streams are present as cations and can be precipitated by conventional lime treatment and sulfide polishing. However, arsenic and selenium are invariably present as anions, AsO33-and AsO43 in the case of arsenic and SeO32 and SeO42 for selenium. Conventional lime treatment removes most of the arsenic, but it does not remove selenium. The present paper reviews the processes and technologies developed to date on the removal of arsenic and selenium from metallurgical process streams.

Jan 1, 1996

By J. E. Klein

Bench scale tests using a stack of ten whole SAES St909 pellets, 6 grams, to crack methane have been completed with pure methane feed. Test conditions ranged from 600 to 800°C at 101 kPa (760 ton) with a methane feed rate of ten sccm. Methane reformation tests, test where hydrogen was passed through a St909 bed after cracking methane, were performed to determine if and how much methane can be generated under "reformation" conditions. Phase structure determinations of St909, composed of a ZrMnFe alloy with aluminum used as a pellet binder, after methane cracking were complicated by the number of materials that can be formed by the elements in the St909 and the carbon from methane cracking. It was determined that the aluminum not only alloys with the Zr, Mn, and Fe in the St909 alloy, but also forms aluminum transition metal carbides in addition to ZrC as the carbon content and the temperature of the St909 increases. Some A14C3 was also found in some pellets. Passing hydrogen over St909 pellets after methane cracking formed methane at various rates in the temperature range of 600 to 800°C. Temperatures of 800°C gave the best methane cracking performance and efficiencies decreased rapidly with temperature.

Jan 1, 2002

By A. F. Saavedra, N. E. Richards, C. J. McMinn

"The aluminum industry in general and Reynolds Metals Company in particular have continued to support development work for improvement of the Hall cell design and for alternative processes for the reduction of alumina. The evolution of the Hall cell design will continue into the next century as a result of further development of mathematical models, sophisticated control systems and new materials for cell components. This paper discusses the areas in which the new generation cell advances will differ from the present day design. Additionally, Reynolds Metal s has supported work for the development of a carbo thermic process to reduce alumina. Theoretical work has indicated that this process could possibly have economic advantage over the Hall process. Several complicated furnace configurations have been proposed to provide the conditions for successfully reducing alumina. The results of small pilot scale operation and most promising furnace design are discussed. For comparison, projected capital investment costs and operation costs are discussed. IntroductionThe evolution of the Hall-Heroult process will continue into the next century. This evolution will be influenced by economics, more .restrictive environment and worker protection requirements, and technological development. These factors will also cause changes in the competitive position of alternate processes for aluminum production.Several aluminum producers have made major investments in alternate processes for aluminum production in the past three decades and there is continued support in some companies for improvements in the Hall-Heroult process and for alternate processes. The more visible development work in reduction technology, not aimed at significant process changes, is in the areas of higher amperage cells with bus bar designs to reduce harmful magnetic effects, more advanced and automated feed, voltage and electrolyte composition control, improved energy efficiency and sol id waste disposal . Efforts to bring significant change to the reduction process are being made in development of materials that can function as inert anodes and cathodes. These materials could lead to the development of reduction cells with multiple bipolar electrodes operating on oxide feed and fluoride electrolytes. Development work on new processes was reduced in the last decade by several companies because of economic and technological realities. Reynolds Metals supported work on a carbothermic process to reduce alumina and produce commercial pure aluminum. This work progressed through small pilot plant operation, and resulted in development of furnace designs and operation regimes. This paper reviews how this and other technologies could influence the reduction of alumina beyond the year 2000."

Jan 1, 1988

By R. Tunold, Y. J. Zhang

"The cathodic kinetics of aluminum deposition from aluminum chloridealkali chloride melts was studied by use of electro-cnemical transient techniques at 700 °C. Aluminum deposited as an insoluble product on a glassy carbon electrode and as a partly soluble product on a tungsten electrode. The deposition was found to be a simple diffusion controlled process involving the exchange of three electrons. The diffusion coefficient of aluminum chloride was calculated - 4.sx10-s cm2 ·S-I in AlCl3 -NaCl-LiCl melts and 2.6x10-s cm2 ·S-' in AlCl3 -NaCl-KCl melts. For AlCl3 -NaCl-CsCl system the diffusion coefficient exhibited some uncertainty due to nucleation. The determined rate constant of aluminum deposition reaction was on the order of 10-' -10-2 cm'S:' Constant potential electrolysis was carried out. TiB2 was well wetted by aluminum. The species of melts penetrated into graphite and glassy carbon but not into tungsten. The diffusion coefficient of species in glassy carbon was evaluated.IntroductionAluminum chloride electrolysis is attractive for its power saving of 30%. Different process to circumvent the conventional chlorization to obtain aluminum chloride economically, such as directly chlorizating mixture of Al~03 and C in a cell with A-shaped laminated bipolar electrodes by rising chlorine (1), or by use of a composite anode (2), will promote the electrolysis technology to be realized industrially. With the appearance of the Alcoa process periodically there are studies on the kinetics of aluminum deposition. Some discrepancies seem to exist on the mechanism of aluminum deposition. Borne studies (3,4,5) showed that the deposition of aluminum was hindered by diffusion process in the melts. M.Gabco et a1. (6) claimed a preceding dissociation of a complex being the rate controlled step. This paper presents our results on the mechanism of aluminum deposition from the study by use of transient techniques and some results from constant potential electrolysis."

Jan 1, 1988

By V. M. Paretsky

Oxygen-flame converting of a solid ground matte containing heavy non-ferrous metals in a wide range of concentrations is one of the perspective trends meeting the demands of modern non-ferrous metallurgy. As. a result, of semicommercial tests on oxygen-flame smelting of nickel matte containing 13,5-17,5% Ni; 0,45-0,65% Co under continuous autogenous conditions a converter matte containing 68 8-75,0% Ni and 0,48-1-,0% Co and ferrite -calcium slag containing 1,4-8,4 Ni and 0,16-0,56 Co were pro¬duced. Nickel recovery in a converter-matte was 76,2%, copper - 58,'2% and cobalt recovery in. a slag was 90%. Slag cleaning in an electric furnace allowed to after-recover nickel and cobalt in an alloy to produce a dumpy slag containing 0,03% Ni and 0,02% Co. The development of modern' ecologically pure energy-saving autogenous technologies of the treatment of sulphide raw materials containing heavy non-ferrous metals makes certain demands of the perfection of a process of converting rich sulphide-metallic intermediate products obtained at autogenous smelting. These-demands were connected first of all with the fact that converting process stages because of process periodicity are more unfavourable from the point of ecology, power intensity and unit stability. Besides, during rich matte treatment the problems connected with heat shortage when using air blast and heat excess-when using top oxygen blast to treat luquid matte a-rise and when enriching air blast by oxygen up to 26% fast erosion of the brickwork of a tuyere zone takes place. The technology of Outokumpu and Kennecott companies-on solid- copper matte converting in a flash smelting furnace using air blast is not universal as its possibilities are limited by certain composition of mattes. Besides, flash smelting to produce highly basic or combined slags requires larger enrichment-of blast by oxygen or extra fuel feed. Naturally a flowsheet with continuous converting of matte of various composition up to white matte production is simple,

Jan 1, 1993

By K. Mansfield

Spent Pot Lining (SPL) is an unavoidable waste product of the electrolytic process in the smelting of aluminium. SPL is considered to be a hazardous waste in various countries because it contains significant quantities of absorbed fluorides along with traces of cyanide. In 1995, Portland Aluminium undertook the investment of AUD$26,400,000 to construct a Research and Development processing facility at Portland, Victoria, Australia, to treat SPL. Extensive research, development and design were carried out over 5 years prior to the investment decision. This study led Portland Aluminium to install an Ausmelt pyrometallurgical furnace to combust and melt the SPL producing a hydrogen fluoride (HF) laden gas, which was cooled and then converted to aluminium fluoride (AlF3). Both the cooling and AIF3 manufacture were processes developed by Portland Aluminium. Many commissioning, equipment reliability and process problems have been encountered that had not been experienced in the trials at Ausmelt or in the developmental process for the treatment of the furnace offgases to form AlF3. Designs based on anticipated process expectations have had to be re-engineered by the SPL team to ensure plant and operational success. In June 2001, the installation of the final major component - a slag granulation process - was completed. This has produced a vitrified (glassy) slag that is expected to have leachability qualities acceptable to the Victorian Environment Protection Authority (EPA) to allow commercial end uses of the slag material to be developed. Today (September 2001), the smelter has successfully used the fluoride product, namely 65- 70% AIF3, in aluminium production and the SPL plant has significantly increased throughput with expectations to be at annual design rates of 12,000 tonne of SPL in early 2002. "The Alcoa Portland SPL Process" has been technically developed and is a proven process for the treatment and recovery of valuable components from SPL materials.

Jan 1, 2002

By F. Robert Hubbard

IMCO Recycling Inc. has developed a new, patented technology for the production of fertilizer, K-SOIL?, as a by-product of its aluminum recycling operations. This new technology can substitute for conventional technology which now results in a saltcake residue that is placed in 'landfills. Four years of research and development work by IMCO culminated in a U. S. patent issued April 5, 1994. As part of the technology development effort, IMCO engaged the services of a nationally renowned research organization and a prominent university. These two organizations have conducted in-depth agronomy studies In the laboratory, greenhouse, and the field. Results have shown K-SOIL? to be performance equivalent, on a nutrient content basis, to conventional potash fertilizer. The first field applications on commercial farms were made in the spring of 1994. K-SOIL? is a potassium-bearIng, granular solid which may find use in fertilizer blends, direct application, and other forms of fertilizer products. The overall benefit of this new technology is to create a useful by-product rather than generate a useless residue and, thereby, reduce or eliminate the amount of salt cake now placed in landfills. As such, It is a significant, step toward IMCO's ultimate goal of complete recycle.

Jan 1, 1995

By Michael J. Nicol

Growing environmental and occupational safety concerns about the use of cyanide in gold processing has increased the interest in more acceptable alternatives. The most likely of these is thiosulfate. However, the thiosulfate process lacks a proven in-pulp method of recovering the dissolved gold, because activated carbon is not effective for the absorption of the gold thiosulfate complex .. Anion exchange resins do offer a possible route for in-pulp recovery. This paper describes work aimed at the evaluation of the effectiveness of commercially available anion exchange resins for the recovery of gold from thiosulfate leach liquors? and pulps. It has been found that strong-base resins are superior to weak-base resins in that they have a larger capacity, which assists in reducing the competition from other anions within a leach solution. Further studies with a selected commercial strong-base resin have shown that competing polythionateanions (particularly tri-and tetrathionate). Lower the maximum possible gold loading but that gold is selectively recovered over other base metal anions in solution. Kinetic experiments found that while? competilig anibns did not affect the initial rate? of gold loading it was important to minimise the contact time of the resin in solution so to prevent displacement of loaded gold by competing anions over longer times. A small-scale resin-in-pulp plant investigation showed gold can be recovered from a leach pulp by a strong-base resin to yield gold loadings of 3000 mg/L, barren pulp containing less than 0.05 mg/L of dissolved gold and. copper loadings of 500 mg/L. Loaded copper was displaced gold and at a loading of 2000 mg/L gold the copper loading was minimal «500 mg/L). The resin was found to recover gold adsorbed on the solids that would, otherwise have. reported to tails if a solid/liquid separation method were used. A patented two-step elution process in which copper and then gold are separately eluted from the resin is also discussed.

Jan 1, 1998

By A. Iván, Patiño C. Francisco, Flores B. Javier, Peréz L. Miguel, Hernandez A. Juan, Trapála P. Norma, Salinas R. Eleazar, Rivera L. Isauro, G. Flores, U. Mizraim

"Tills work is related with the production of tiles using wastes from mining and metallurgical industry of Hidalgo State, as raw material. These wastes present a majority particle size (60 %), minor than 53 m (mesh 270, in Tyler Series) with the following composition; 70.01 % wt. of Si02, 12.82 % wt. of Ah03, 3.80 % wt. of Fe, 0.70 % wt. of Mn, 3.98 % wt. ofK20, 3.34 % wt. of CaO, 2.50 % wt. ofNa20, 0.04 % wt. of Zn, 0.026 % wt. of Pb, 56 grams per ton of Ag and 0.6 grams per ton of Au. In this work, the partial results and experimental conditions for the preparation of some composites to produce tiles are shown. The composite that offers the best results in productions of tiles was that with 66.67 % wt. of mining waste, 26.67 % wt. of red mud and 6.67 % wt. of black mud.IntroductionTo now, mining activities carried out during more than 457 years in Hidalgo State, Mexico; has lead a great volume of wastes; which can be classified into different types, having among the more common; Acid mining water, solid transport water, elements in radical form, abandoned minerals without processing and processing minerals wastes known as burrows, being these the most abundant in the region counting with an approximated volume of 110 million of tons, presenting a fine particle size that can affect population health during windy epochs [1 -4].Form metallurgical point of view, is well known that a fundamental characterization study it is necessary to know the viability of reusing or recycling of these wastes. According this, these residues can also be considered as raw materials in other economical activities, due its Si, Fe, AI, Ca, Na and P oxides contents, its mineralogy and the distribution of chemical elements in the primary and secondary phases [1,2,5,6]."

Jan 1, 2012

By K. Osseo-Asare

"The molecular structural and solution chemical characteristics that underlie the surface activity of solvent extraction reagents are highlighted, with particular emphasis on the tendency towards reverse micelle and microemulsion formation. The physicochemical factors that lead to microemulsion-enhanced solvent extraction with chelating extractants are elucidated. Membrane-based contactor systems are described which are uniquely suited for enhanced solvent extraction with microemulsions. The possibility of combining solvent extraction with microemulsion-mediated materials synthesis is discussed. IntroductionThe discovery of new selective extractants has resulted in major breakthroughs which have helped to establish solvent extraction as a versatile separation technology (1-3). A complementary strategy towards new process development is to seek new ways of using the available reagents. This paper explores this second approach. From a molecular structural standpoint, an extractant may be viewed in terms of two components: a polar functional group and a hydrocarbon radical. In the conventional view of solvent extraction, this molecular structure serves two main purposes: The polar functional group provides the binding sites for complexation of metal ions, while the organic group is needed to render the resulting metal-extractant complex organic-soluble. But are these the only properties of the organic and the polar groups that are relevant to the solvent extraction process?In this paper, attention is directed to the fact that solvent extraction reagents possess the molecular structures of amphiphiIes; that is, they are characterized by the presence of spatially separated hydrophilic and hydrophobic groups on the same molecule. A case is made for the view that this amphiphilicity offers interesting opportunities to develop new separation processes which collectively may be described as enhanced solvent extraction (4-6)."

Jan 1, 1996

By Zhao Guangwen, Wu Tan, Qiuzhan Tian, Duan Shuzhen

"A thorough study of lost currents in bipolar cells has been made for the development of the great potential of bipolar cells in energy saving. A proper equivalent circuit for bipolar cells has been found, and mathematic methods for computing lost currents at all kinds of polarization have been developed.At linear polarization, the lost currents can be computed by solving either a tridiagonal matrix or an ordinary differential equation of second order. This kind of values can be used to evaluate the minimal lost currents at given conditions. At other kinds of polarization, the lost currents can be computed by the same methods as above if the necessary experiments of measuring the cell voltages of every electrolytic compartment have been made. This kind of values can be used to design bipolar cells at practical conditions.Extensive experiments have been made to evaluate the effects of electrode polarization, emf of galvanic cell for the formation of electrolyzed substance, gaseous products, geometric dimensions of electrolytic cells, special resistance of electrolyte used, and the total and ordinal number of electrolytic compartments on lost currents. It is the first time to make such a systematic research by experiments in the world. Experiments showed that theoretic models were in good agreement with measured data. Experiments also showed that D.C. energy consumption of 0.45 kwh/kg of Pb for lead electrolysis from molten chlorides could be achieved in a cell with 7 bipolar electrodes.IntroductionIt is very promising to apply bipolar electrolytic cells to electrolysis of aluminum, magnesium, lead, zinc and perchlorinates for their great potential of energy saving for example, bipolar cells had been used in the production of magnesium and saved more than 30% of energy, being one of the three main techniques of energy saving in the production of titanium sponge (1)."

Jan 1, 1988

By K. Osseo-Asare

The molecular structural and solution chemical characteristics that underlie the surface activity of solvent extraction reagents are highlighted, with particular emphasis on the tendency towards reverse micelle and microemulsion formation. The physicochemical factors that lead to microemulsion-enhanced solvent extraction with chelating extractants are elucidated Membrane-based contactor systems are described which are uniquely suited for enhanced solvent extraction with microemulsions. The possibility of combining solvent extraction with microemulsion-mediated materials synthesis is discussed.

Jan 1, 1996

By Kotaro Ogura

"Ethylene has been produced in aqueous solution from CO2 by the electrochemical reduction driven by a natural energy. This process is useful for storing a large amount of the natural energy. In the closed system, the conversion efficiency of CO2 is almost 100%, and the maximum selectivity for the formation of ethylene is more than 70%. On the other hand, the current efficiency for the competitive reduction of water is less than 10%. The electrolysis is practicable under such special conditions as three-phase interface consisting of gas, solution and metal, concentrated solution of potassium halide, low pH and copper or copper halide-confined metal electrode. These requirements are thoroughly examined, and the grounds to reply upon are revealed. A series of chemical apparatuses including an electrolytic cell in a large scale are designed for the ethylene production, which allow us continuously to supply raw CO2 and to extract the product.IntroductionIn connection with the depletion of the petroleum reserve, much attention has been paid to the utilization of natural energies such as sunlight, wind force and waves. However, these energies are always changeable, and some means of storing them are necessary for the common use. In general, natural energies are transformed to the electric energy at the time of utilization, but the electricity is not very storable. This is one of the reasons why the application of natural energy is not so expanded. On the other hand, the storing of natural energy by plants is very skilful. In the photosynthesis, solar energy is transformed to chemical energy by fixing carbon dioxide."

Jan 1, 2009

By Joerg Hammerschmidt, Marcus Runkel, Andreas Wirtz, Kent Pope

"Fluidized bed reactor systems in the minerals industry have been progressively developed and widely implemented over a period of more than sixty years for a multitude of process applications. Their versatility is manifested in calcining, roasting and combustion treatment of minerals including solid fuels. Experiences from metallurgical processes have expanded to new applications in the oil shale processing, biomass combustion and renewable energy sector. This paper focuses on the most important aspects of fluidized bed technology for a variety of Outotec processes. Energy recovery is a major topic for all applications and will be addressed, as well as environmental aspects, particularly regarding the off-gas treatment. Outotec is one of the pioneers of roasting and the fluidized bed process including gas cleaning and sulphuric acid production where applicable. Process development is strongly supported by significant investment in R&D, thus enabling process adaptation and scale-up to meet specific customer/market requirements.Introduction to Fluidized Bed TechnologyThe early days of fluidized bed technology were characterized by the use of bubbling fluidized bed reactors (BFB). This technology was popular for roasting zinc and copper ores. The introduction of circulating fluidized bed technology led to increased plant capacities and improved process control for the traditional processes. Furthermore, the technology has been applied in many other industries, including the renewable energy industry, which will be part of this design and process comparison. The paper starts out with the fundamentals on particle fluidization as the main common design parameter, leading to the importance of heat recovery and offgas treatment.In fluidized reactor systems, the particle concentrations can reach several hundred kilograms per cubic meter of fluidization gas. This high solid concentration leads to large surface areas and enables direct contact of gas molecules with particle surface molecules. Excellent conditions for enhanced heat and mass transfer thus provide nearly ""ideal"" heat exchange conditions and better reaction kinetics. The bubbling bed exhibits a better heat transfer coefficient within the bed in comparison to the circulating fluidized bed (CFB). The different conditions of gas velocities and solid concentrations allowing the classification of fluidization reaction regimes are shown in the Reh diagram in Figure 1."

Jan 1, 2014

By Christiana Mystrioti

Greece is the only EU country with extensive nickel laterite deposits. Since the 1960s, the Greek laterites are treated for ferronickel production via a pyrometallurgical route, which involves pre-reduction of the ore in rotary kilns and reduction smelting in electric arc furnaces. Due to the rising cost of energy and the decreasing grade of mined laterites, the pyrometallurgical treatment is economically marginal. For this reason, alternative treatment processes are currently under investigation. Potential applicability of hydrometallurgical treatment, using either H2SO4 or HCl, under atmospheric pressure conditions, for Ni and Co extraction from two low grade laterite ores, was examined in the present work. The two samples were selected to represent the different mineralogical composition of the saprolitic and limonitic ore deposits of the Greek laterites. The investigated parameters were temperature (65–90 °C), acid concentration (1–4 N) and solid to liquid ratio (10–30% w/v). The results indicated that HCl was more efficient than H2SO4 for the treatment of the saprolitic ore; Ni and Co extractions were very high for both elements, up to 98% and 96% respectively, while H2SO4 could leach efficiently Ni, up to 100%, but extracted only 34% of Co. Hydrochloric acid was also more efficient for the treatment of the limonitic ore, with Ni and Co extractions up to 98.3% and 87.6% respectively. With H2SO4 the maximum extractions were 71.3% for Ni and 52.1% for Co.

By Guojun Qi

This work relates to wafer bumping technologies for flip chip packaging applications in the electronics industry. Nickel and its alloys are alternative under bump metallization (UBM) materials because of their slower reaction rates with Sn-based solder as compared to Cu-based UBMs. In this study, we compared the morphologies of the intermetallic compounds (IMC) formed between Sn-bearing solders (Sn3.5Ag and eutectic Sn-Pb) and two types of nickel- based UBMs, electroless Ni-P (EN) and sputtered nickel. Both chunky-type and needle-type of intermetallics were observed between the EN UBM and the solders. In the case of sputtered Ni UBM, there is only a layer of scallop-type intermetallics formed. The morphology change and growth kinetics of the IMC in thermal aging process were analysed. Kirkendall voids were noticed from samples of Sn3.5Ag solder vs Ni-P UBM that have undergone prolonged thermal aging at high temperature.

Jan 1, 2003

By Akira Yazawa

In the field of pyrometallurgy, the equilibrium evaluation has recently become easy by use of thermodynamic software, but it is strongly desired to estimate the amounts of non-equilibrium minor components to simulate practical processes. However, simple physico-chemical technique can not be applied to evaluate practical non-equilibrium components for many reasons. As an engineering solution for postulated equilibrium system, the activity coefficient is modified and introduced as the simulation parameter referred to as the "adjustment coefficient." This modified thermodynamic calculations were adopted to various sulfide roasting processes where many solid phases are participating and the system is not necessarily in equilibrium. Successful simulation was realized by assigning suitable adjustment coefficients for minor components which form a pseudo solid solution in the calcine. Some examples from sulfation roasting and dearsenizing roasting were discussed. In the smelting system, the loss of copper in slag is caused by equilibrium and non- equilibrium reasons, that is, chemical dissolution and entrainment of copper in slag. The total copper content of slag was simulated by use of the adjustment coefficient for entrained copper.

Jan 1, 1998

By W. E. Straszheim

Computerized microscopic techniques were used to determine the size and identity of the ash-forming mineral components, as well as their association with the organic coal matrix, in order to aid in the development of physical methods of beneficiation which require such characterization for the most effective cleaning. Examples of particle size measurements, mineral phase identification, and especially coal-mineral association results are given for selected samples of raw and physically cleaned coals. These measurements can reveal certain characteristics of coals which can be be of significant value in predicting coal recovery and mineral matter rejection for coals undergoing beneficiation.

Jan 1, 1989

By Xiangwen Wang

Electrochemical measurements associated with the electrowinning of lead metal from concentrated brine (NaCI) - lead chloride (PbCI,) electrolytes at temperatures in the range 328 to 353°K (55 - 80°C) were conducted. The electrodeposition of lead metal onto a lead rotating-disk electrode may be described by the electrodic reaction: {PbCl 4 2 + 2e- = Pb° + 4Cl-} The deposition-overpotential requirements are low (approximately 41mV per decade of current density - Am+), with a corresponding high exchange current (approximately 0.018 A.cm2). The diffusion coefficient of the electroactive PbC142- species in a 3.5 NaCl electrolyte at 353°K was insensitive to the total PbCI, concentration (0.01 to 0.1M), and its value was found to be (1.21*0.1)~10-5 cm2Is with a temperature-sensitivity coefficient (E,/R) of approximately 1.98x103K-1 (328OK(T<353OK).

Jan 1, 1993

By Guojun Qi

This work relates to wafer bumping technologies for flip chip packaging applications in the electronics industry. Nickel and its alloys are alternative under bump metallization (UBM) materials because of their slower reaction rates with Sn-based solder as compared to Cu-based UBMs. In this study, we compared the morphologies of the intermetallic compounds (IMC) formed between Sn-bearing solders (Sn3.5Ag and eutectie Sn-Pb) and two types of nickel- based UBMs, electro less Ni-P (EN) and sputtered nickel. Both chunky-type and needle-type of intermetallics were observed between the EN UBM and the solders. In the case &apos;of sputtered Ni UBM, there is only a layer of scallop-type intermetallics formed, The morphology change and growth kinetics of the IMC in thermal aging process were analysed. Kirkendall voids were noticed from samples ofSn3.5Ag solder vs Ni-P UBM that have undergone prolonged thermal aging at high temperature.

Jan 1, 2003