Search Documents

By Chonghun Han, Tae-Hwa Choi, Young-Hwan Chu

"A statistical optimization method is proposed to maintain stable operation in a blast furnace with stave coolers. The objective of this study is to find the operating condition which maximizes productivity as well as minimizes the heat load on the stave coolers. In this method, we first analyzed historical data and calculated the heat flux of stave coolers and the burden distribution by mathematical models. In this approach, we determined the variables to optimize and manipulated variables to adjust these response variables. Next, we collected the necessary data for the data variables after preprocessing such as removal of noise and outlier. Finally, we constructed empirical models describing the patterns of response variables in terms of manipulated variables by partial least squares regression method, and then these models were used as objective functions for entire optimization problem. By using appropriate optimization algorithm, we determined the compromised optimum operating condition considering all response variables from the past operating conditions.IntroductionBlast furnace operation which is accompanied by complex physical and chemical phenomena is an essential process for steel industry. In this process, iron ore which is charged into the blast furnace from the top goes downward. It is reduced to iron by CO gas and is melted by the heat of combustion of coke and pulverized coal. The pig iron is tapped at the bottom in liquid phase after removing the slag, which floats on the pig iron.The cooling equipment of the blast furnace wall is changed from the plate type to the stave cooler recently. In order to cool the shell uniformly and eliminate shell spray cooling, Pohang No.2 blast furnace of POSCO (Pohang Iron & Steel Company) was equipped with 11 rows of stave coolers. These staves are only 350mm-thick including the 170 mm-deep refractory on the hot face."

Jan 1, 2001

By M. Potesser

In the primary and secondary copper industry the anode furnace is the most important process step in the refining procedure to reach an enhancement in value and quality of the copper by a selective oxidation and reduction. By combining the advantages of the conventional tilting hearth anode furnace and the benefits of the drum type anode furnace a new aggregate was developed and patented. The cross-section of the furnace is elliptical and the whole unit is tiltable. During the melting period, the contact surface between the burners and the bath or the scarp is enlarged - an increased energy input is the consequence. For the refining period the furnace is rotated, whereby a high bath level is guaranteed. Due to the longer residence time of the gas in the copper melt, not only the gas consumption is reduced but also the efficiency/quality of the refining process is improved. This investigation shows a metallurgical comparison between the new anode furnace concept with the elliptical profile/cross section and the conventional, tiltable hearth anode furnace as well as the drum type anode furnace. The CFD modeling of the currents in these different furnaces delivered helpful data to compare the geometry of the different furnaces. The investigation covered the refining of blister copper from primary metallurgy, from secondary metallurgy as well as from a mixture of both copper melts. The paper underlines the possibilities for a reduction of the tap-to-tap time and for an increase in production capacity, which can be achieved by applying the new furnace concept in comparison to the conventional furnaces.

Jan 1, 2006

By Kim Verbeken, Evelien Van de Steene, Marc Verhaege, Jeriffa De Clercq

"AbstractThe industry is confronted with stricter discharge levels and an increasing awareness to avoid toxic waste (such as heavy metal containing sludge obtained in physicochemical water treatment) and recover valuable metals. This can be realized by ion exchange or by electrodeposition. Both techniques however show an optimum concentration range to efficiently remove and/or recover metals: electrolysis can recover the metal in its metallic form but suffers from low efficiency when applied to diluted solutions, whereas ion exchange can only be applied economically in the low ppm range.A combination of ion exchange followed by electrodeposition was applied to an industrial wastewater. The strong acidic ion exchange resin was stripped with H2SO4 giving a solution of 9g/L nickel. From this solution, nickel was electrodeposited in an ammonia buffer on a Ti-mesh cathode with high current efficiency. It is shown that organic compounds (surfactants, brighteners, …) present in the industrial wastewater do not influence the recovery process.IntroductionAlmost all rinse waters emerging from metal plating plants are generated during the washing operation of plated materials. These solutions with low metal content must either be recycled within the process or be discarded to the surface water after decreasing their metal content to acceptable levels determined by the local environmental regulations, which become more and more stringent. Today, the most commonly used practice is a physico-chemical treatment (precipitation) and a resin post treatment to meet the standards. A more sustainable approach should use a technology to recover efficiently the metal or metal salt without the production of sludge and the effluent should be consistent with environmental regulations."

Jan 1, 2008

By Hong Ro Lee, Seang Suk Cho, Byung Sun Chun, Chang Whan Won

"The preparation of Al203-SiC-TiC composite powders was studied by SHS process using silica (source of Si02) powder instead of Si which is expensive. And, aluminum powder was used as reducing agent of silica and charcoal activated was used as carbon source. Effects of the molar ratio in raw material mixtures and compaction pressure and initial temperature of reactant on the products and combustion process were studied. The most important variable affecting the synthesis of Al203-SiC-TiC was the molar ratio of carbon. The combustion reaction generated in powder state without preheating. But the unreactants exist in the product on all molar ratio. When the preheating temperature is over 400 °C, the combustion reaction was able to be generated in pellet state. Without preheating, the combustion reaction was possible by Ti02 addition. The optimum condition in this system was the molar ratio of Si02:Ti02:Al:C=2.5:0.5:4:3.1. INTRODUCTIONCeramics have demonstrated a capability as a cutting tool, especially in competition with WC-Co cermets (ceramic-metal composites) as inserts for metal turning and milling operation. The advantage of ceramics compared with carbides is retention of high hardness, strength, and chemical inertness to temperatures in excess of 1000 ·c . This allows use of the ceramics at much higher machining speeds than can be tolerated by carbides. However, the ceramics have lower toughness than the carbide materials, and have only been used successfully in the limited operations of turning and milling. There are several classes of cutting tools. The addition of TiC or PSZ, or both, and reinforcement with SiC overcame the poor fracture toughness and thermal-shock resistance that limited Al203 use in the past [1]."

Jan 1, 1998

By Pengcheng Li, Wei-Tong Du, Yu Wang, Gang Wen

Carbon dioxide could be utilized as a weak oxidant and a kind of coolant to oxidize elements, meanwhile, helping control the temperature during the converter vanadium extraction process. However, the optimum content of CO2 and the cooling effect of CO2 at low content have not been reported. In this study, experimental research based on the influence of different CO2 contents from 0% to 25% injected to the vanadium-containing hot metal was carried out, as well as contrast experiments of O2-N2 mixed blowing. The results indicated that the optimum content of CO2 was 15%. Under the optimum condition, the oxidation of [C] was the lowest and the oxidation rate of [V] was 96.9%, while the temperature was also lower than the O2-N2 mixed blowing. This paper provide a potential property for utilizing CO2 during the converter vanadium extraction process.

Jan 1, 2016

By P. F. Stratton, M Huggahalli, N Saxena

"Several studies have been carried out to determine how the optimum atmosphere conditions in the heat treatment of metals can be established for a given product by using different gas injectors and varying atmosphere flow rate and composition. The most common approach taken to gain insight into furnace atmospheres has been to use commercial computational fluid dynamics codes and examples of their use are shown. However, the complexity of these codes, the difficulty in establishing boundary conditions, their cost and computational demands limit their usability in solving day-to-day production problems.A simplified, web-based model of a furnace atmosphere is described where the conservation equations are solved to determine the composition profiles of the atmosphere in the furnace. Although the model is restricted in scope to certain types of furnace, it is shown to give better results than CFD when boundary conditions are uncertain. The application of the model to the understanding of existing furnace conditions and to determining ways of optimising the use of gases are discussed. IntroductionThe important role of protective atmospheres during heat treatment of metals, furnace brazing of assemblies and soldering of printed circuit boards is well established. The atmosphere in the furnace prevents air ingress and oxidation of the parts, helps flush away any flux or binder from braze and solder pastes and provides an important mode of heat transfer [1]. In a batch furnace having typically one or two gas injectors, the gas utilisation and process can be optimised by minimising air leaks and mantaining atmosphere homogeneity with fans or other atmosphere circulation techniques. In a continuous furnace, however, atmosphere optimisation is more difficult as furnace inlets and exits can also let in air. The atmosphere requirements of different stages of heat treatment, brazing or soldering vary [2]. Achieving these atmospheres in different sections of the open-ended furnace is a relatively complex task. Methods such as atmosphere zoning and the use of physical curtains and gas injectors have evolved with time, though it is hard to predict how effective these techniques will be for a given furnace and process. Experts in the field tend to rely on experience and rules-of-thumb rather than scientific method to achieve the required results."

Jan 1, 2001

By Christian Wögerer

For material or process development as well as for quality checking of materials, thermal diffusivity measurement may be the tool of choice in many applications. Especially if diffusivity is measured locally resolved, a lot of information can be read out of such a ?diffusivity image?. This paper explains the method to measure coplanar specimen like plates and lists some interpretations of diffusivity images, like inhomogeneity of sintered materials, structural voids or inclusions. Diffusivity is a material property, describing transportation of heat in solids in spatial and temporal terms. With thermo cameras it is possible, to measure diffusivity of larger areas simultaneously. The result is an area of diffusivity values that may be plotted or displayed like an image. Locations with lower diffusivity values could be caused by the presents of structural differences in the material, by decomposition of the material components or by inclusions of other material like air.

Jan 1, 2009

By Dong Liang, Honggang Zhong, Zhenxing Yin, Qijie Zhai

"2507 super duplex stainless steel (SDSS) has been widely applied in a marine environment, attributing to its excellent pitting corrosion resistance and mechanical properties which are particularly determined by the nitrogen precipitation behaviors and volume fraction of ferrite. In this paper, various cooling rates were obtained by applying four different moulds with the common gating system. By which the influence of cooling rates on nitrogen precipitation behaviors and the ferrite fraction in cast 2507 SDSS was comparatively investigated. The results showed that nitrogen porosity could easily precipitate in cast 2507 SDSS under standard atmospheric pressure. The number of nitrogen holes increased with decreasing the cooling rates, and the holes mainly distributed in three zones, i.e. the interface between the horizontal and vertical columnar grains, top vertical columnar grains and the final solidification region. Additionally, higher ferrite fraction was obtained with lower cooling rates.IntroductionDuplex stainless steels (DSS) have a dual-phase microstructure consisting of ferrite (d) and aus-tenite (?). Due to the combined advantageous properties of d and ? phase, their application rapid-ly increased in industry [1, 2]. Super duplex Stainless Steel (SDSS) is the third generation of du-plex stainless steels, of which the alloy elements Cr, Ni and N content are higher than the stand-ard 2205 DSS. Because it has high contents of alloy elements, the SDSS has better mechanical properties, and its pitting resistance equivalent number (PREN) reached a point 43. Hence SDSS has been used in petroleum chemical industry, the marine equipment and harsh corrosion environments [3]. SDSS contains ferrite stabilizing elements, such as Cr, Mo and Si, and austenite stabilizing elements like Ni, Mn and N [4]. Many researchers have studied the effects of these elements on the stabilization of d and ? phase and the ratio of the two phases. High nitrogen con-tent is another prominent feature of SDSS. Nitrogen addition to SDSS is very effective for improving its mechanical properties [5] and corrosion resistance [6], such as pitting resistance in NOx environment. To increase the nitrogen content in steels, a lot of work had been done. N. Nakamura and S. Takaki [7] examined the behavior of nitrogen absorption in Fe-23Cr alloy powder by sintering in nitrogen gas atmosphere, they found nitrogen content easily up to 1.0 wt. %. Using devised NH3 gas injection system, W. Park et al. [8] successfully produced austenitic Fe–23.5Cr alloys containing over 3 wt. % of nitrogen. Further-more, W.P. Tong et al. [9] stated that nitrogen percent was extraordinary high (about 10 wt. %) in pure iron plate. However, the above methods are difficult to be applied to large-scale industrial production."

Jan 1, 2012

By S. Z. El-Tawil, K. A. El-Barawy, M. B. Morsi, M. M. Nasr, Z. I. Zaki

"Tue sulphatization of zinc concentrate produced from heated zinc concentrate of Um Ghieg area, eastern desert by ammonium sulphate as sulphatizing agent is carried out at different conditions to achieve the appreciable formation of soluble sulphates. The maximum zinc recovery as soluble sulphate equfils to 95% upon using 1.5 stoichiometry ammonium sulphate at 600 °C for one hour. Tue obtained leached zinc sulphate solution is subjected· to purification process to prepare high pure zinc metal and zmc compounds such as pure zinc sulphate, zinc caibonate and zinc oxide. The products obtained are analyzed chemically and identified by X-ray analysisTue mechanism and kinetics of the sulphatization process of zinc concentrate are considered where the activation energy is calculated and equals to 33.3 K.J.mol""1 A process flowsheet for the overall sulphatization process of zinc concentrate by ammonium sulphate is proposed.IntroductionConsiderable reserves of zinc ores are present in Um Ghieg area eastern desert, Egypt. The possibility of extraction zinc from these ores and preparing its valuable compounds by choosing a more feasible technology become of vital importance. An attention is focused on the sulphatizing roasting processes (1-9) to achieve the appreciable formation of soluble sulphates. Previous publications are reported in our division by El. Tawil and others (10-16) using ammonium sulphate as sulphatizing agent for the successful treatment of nonferrous bearing materials.Therefore, the present work aims to study the different factors affecting the sulphatizing roasting process of local zinc coocentrate by anunonium sulphate to extract completely zinc and preparing its main compounds. Tue mechanism and kinetics of the different chemical reactloos that 1ake place in the process are considered."

Jan 1, 2000

By Tiffany Brown

In the thiosulfate gold leaching process, thiosulfate is used to complex gold following electrocatalytic oxidation of the native metal by Cu(II)-ammines. Unfortunately, Cu(II) complexes also catalyse the undesirable conversion of thiosulfate to polythionates. We have been seeking alternatives to ammonia, which fulfil the role of ammonia but minimize thiosulfate decomposition. One aspect of these studies has involved the determination of the rate of thiosulfate oxidation by the Cu(II) complexes from UV-Visible spectral changes. Various ligand classes, including pyridyls, polyamines, aminocarboxylates and picolinates, have been tested and the majority found to inhibit thiosulfate decomposition. Interesting variations in reactivity with the nature and denticity of the ligand and L/M ratio have been observed. In the case of open chain polyamine ligands with 2 to 5 nitrogen donors, the rate of thiosulfate oxidation was generally found to decrease with the denticity and the concentration of each ligand. Saturation of the Cu(II) coordination sphere appears to inhibit the oxidation of thiosulfate when excess thiosulfate is present. The Cu(II) complex of the branched polyamine ligand, tris(2-aminoethyl)amine (tren) exhibits this behaviour in excess thiosulfate. Interestingly, when the [Cu(tren)(OH2)]2+ complex is in excess oxidation of thiosulfate was observed. This unexpected finding prompted a spectrophotometric study of this reaction which has revealed that the oxidation of thiosulfate involves several processes, two of which involve the formation and decay of a tetrathionate radical.

Jan 1, 2003

By Dumitru Mitrica, Vasile Soare, Elvira Alexandrescu, Mihai Ghita, Marian Burada, Petru Moldovan, Raluca Maria Florea

"Aluminum metal matrix composite Al-Si /SiCp was prepared by reactive gas injection and subsequent remelting. The influence of the remelting process on the microstructure and physicalmechanical properties of the resulted material was investigated. Samples resulted from the synthesis process contained silicon carbide agglomerations in the vicinity of primary silicon and large pore concentrations. In-situ formed silicon carbide presented mostly a polymorphic structure with particle sizes varying from 1 to lOµm. Physical and mechanical properties were largely influenced by the porosity distribution. The remelting process provided a decrease in the composite material porosity and pore size, with significant improvements in the particle distribution and physical-mechanical properties of the composite material. Additions of up to lwt.%Mg in the starting alloy improved the in-situ carbide formation and particles distribution in the solidified material. However, stable and large particle agglomerations were detected in the material in absence of conventional stirring or controlled solidification techniques.IntroductionThe remarkable properties of composite materials (yield strength, stiffuess, hardness and resilience at a reduced specific weight) make them attractive for applications in automotive, aerospace and military industries. Particle reinforced aluminum metal matrix composites (PAMMC) are known for their high strength to weight ratio. The recent development of P AMMC created the possibility to obtain large particles concentrations, exceeding 10% by volume ofreinforcement (SiC, AIN, Al203, etc.). The improvement ofhigh temperature strength and wear resistance made them appropriate for the use in automotive applications, such as the pistons, rotors, disc brakes, and propeller shafts [ 1]. The present main disadvantage of the P AMMCs is their high costs for obtaining and processing related to the final product performance. In order to benefit from the advantages that offer composite materials properties a reasonable production cost - performance ratio must be obtained [2]. Recent innovations in the manufacturing of aluminum-based metal matrix composites provided notable performance improvements but no feasible solutions were found for lowering their present high prices."

Jan 1, 2014

By Hiroshi Hashimoto

For the application of sodium carbonate slag treatment to product high purity copper on an industrial scale, the counter-flow batch process that reduces the amount of utilized slag is proposed from the viewpoint of the post-treatment of slag. Based on the distribution ratios of several impurities between the slag and molten copper determined previously, the effectiveness of this process is discussed using a mathematical model and confirmed experimentally. According to the model, the slag consumption to remove Sb to a required level in case of the counter-flow batch process with 3 reactors is calculated to be around 30 % of that in the case of an ordinary batch process, where pure sodium carbonate is used as a slag at every stage.

Jan 1, 2003

By Jafar Safarian, Merete Tangstad, Leiv Kolbeinsen

"Silicon and its binary alloys have several applications in metallurgical, chemical, photovoltaic and semiconductor industries. Silicon goes to its liquid state within its production, refining and when it is alloyed with other elements. Therefore, the characterization of silicon melts thermodynamics is important and useful. The liquidus and activities in several silicon binary systems have been recently calculated by the application of two liquidus constants, which were established using the experimentally determined liquidus data. In the present study, the thermodynamic properties of six silicon binary melts, which are simple eutectics, are calculated by analytical approaches. In this case, the changes in the partial and integral Gibbs free energies and enthalpies of mixing are calculated and compared with the experimental data in literature. The enthalpies are calculated by two methods and the calculation results are compared to the data from the literature. It is shown that thermodynamic properties of silicon binary melts can be accurately predicted without the use of complicated thermodynamic software.IntroductionSilicon and its alloys have several industrial applications. Silicon is used for making aluminium-silicon alloys to produce cast parts, mainly for automotive industry. Silicon in the form of ferro-silicon is used to deoxidize and alloy steel and cast iron. The photovoltaic (PV) industry has had a rapid growth in the recent years and silicon is the dominant material used for the fabrication of solar cells. Silicon is primarily produced through carbothermic reduction of quartz in electric arc furnaces and the produced metallurgical grade silicon (MG-Si) contains around 1 wt% impurities. It is further refined to produce solar grade silicon (SoG-Si) with above 99.9999 wt% Si to fulfil the chemical requirement for PV application. Since molten silicon is processed all along the metallurgical route, obtaining fundamental thermodynamics knowledge about silicon melts is valuable for the related industries."

Jan 1, 2012

By Jorge A. S. Tenorio, Chuanwei Zhuo, Joner O. Alves, Yiannis A. Levendis

"The ever increasing number of automobiles and the consequential increase in consumption of tires have generated a pressing environmental issue, concerning the disposal of unserviceable tires. The worldwide disposed of waste tires is approximately 1 billion units per year and this amount is expected to increase by 2% each year. This work presents a microstructural characterization of nanomaterials synthesized in a catalyst system from the effluent of burning waste tire chips in a horizontal two-stage laminar-flow furnace. Controlled combustion of waste tire chips took place at temperatures of 900 or 1000 °C, and stainless steel meshes were used to synthesize the nanomaterials at 1000°C. Produced materials were analyzed by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The catalyst meshes were covered by materials with diameters of 20 to 200 nm and lengths of about 40 µm. They possessed structures similar to those of carbon nanotubes.IntroductionThe motor vehicles on the road have been increasing annually. World motor vehicle production in 2009 approached 62 million units. Consequently, the consumption of tires has also been on the rise; about 1.4 billion units were produced around the world in 2007 [l]. The disposal of scrap tires is a serious environmental problem. The worldwide disposed of waste tires is almost 1 billion units per year with an expected increase of around 2% each year. However, less than 7% of produced waste tires are recycled (excluding reuse, retreading or incineration). The waste rubber tires are a thermoset polymer, which cannot be reprocessed like thermoplastics polymers, therefore the recycling of this material requires special techniques that demand big investments [2, 3].A solution to the disposal of waste tires is the sequential pyrolysis/combustion process. Past research showed that waste tires are an attractive potential fuel due the high heating value (29-3 7 MJ/kg), which is comparable, or even higher to those of typical bituminous coals [ 4, 5]. Pyrolysis is a process wherein the material is thermally decomposed at elevated temperatures at the absence of an oxidizing gas. Combustion is characterized by the addition of the oxidizing gases during the process. Several studies indicate that the pyrolysis and combustion processes for the tires treatment have a great energy conversion potential [4, 5, 6]."

Jan 1, 2011

By Takaaki Ishikawa

The distribution equilibria of lead, zinc and copper between CaO-Si02-Ah03 melts and liquid copper were measured at 1623K under a controlled H2-C02 atmosphere. The distribution ratios were plotted against the oxygen partial pressure, and reasonable oxide forms dissolved in the melts were estimated from the slopes of the plots. The activity coefficients of zinc oxide (ZnO), lead oxide (PbO) and cuprous oxide (CuO0.5) increased with increasing slag basicity, defined by XCao/Xsio2 The temperature dependence of the activity coefficients was also measured.

Jan 1, 2003

By W. G. Davenport

Chapter 10 showed that any given smelting task, e.g. production of 65% Cu matte, 1500 K, from CuFeS2, can be done with many different combinations of: blast%02 hydrocarbon fuel blast temperature. This chapter repeats the Chapter 10 calculations but reports the results in terms of the mass industrial oxygen mass hydrocarbon fuel blast preheat energy combinations that will produce 65% Cu matte, 1500 K, from CuFeS2 concentrate. It also determines the energy inputs equivalent to these combinations.

Jan 1, 2001

By Hiroshi Hashimoto

For the application of sodium carbonate slag treatment to product high purity copper on an industrial scale, the counter-flow batch process that reduces the amount of utilized slag is proposed from the viewpoint of the post-treatment of slag. Based on the distribution ratios of several impurities between the slag and molten copper determined previously, the effectiveness of this process is discussed using a mathematical model and confirmed experimentally. According to the model, the slag consumption to remove Sb to a required level in case of the counter-flow batch process with 3 reactors is calculated to be around 30 % of that in the case of an ordinary batch process, where pure sodium carbonate is used as a slag at every stage.

Jan 1, 2003

By M. Z. Quadir, M. A. Islam, P. J. Hazell, G. J. Appleby-Thomas, J. P. Escobedo

The mechanical response of closed-cell aluminium metallic foams subjected to low and high strain-rate loading has been investigated. A set of quasi-static and dynamic (shock) compressive tests have been conducted on closed-cell aluminium foams (CYMAT) with densities of 0.50 and 0.30 g/cc. Post-mortem characterization via optical microscopy and electron backscatter diffraction (EBSD) was performed on pristine and deformed specimens to elucidate the dominant deformation mechanisms in these materials. The combination of these techniques allowed for the assessment of critical deformation parameters such as changes in cell geometry and morphology, as well as microstructural evolution and deformation of the aluminium cellular network. These findings aim to aid in the design and development of optimized material structures for impact and blast protection.

Jan 1, 2015

By Mokka Rao, Asdrubal Serrano

"A mathematical model is developed which describes the movement of the minerals present where the vertical and angular velocities are related with the different operative parameters involved in the process of separation of silica from iron ore, which permit maximise the capacity of the spirals to obtain major productivity. The process is realised in two steps, of the rough dressing and other cleaning. Through the results obtained for chemical analysis of the final products of each spiral and the equation of movement of particles, it is proved the applicatibility and efficiency of mathematical model to operating parameters of the spiral. Moreover, it is determined though the recuperation of the final products of quality in the use spiral HC 1870 and HG 8/7. INTRODUCTIONThere is prospetive of increasing the production the iron ore in agreement with the demand of siderurgical proyects united with the external demand of quality 40 million tons per year. The Ferrominera de Orinoco Company is experimenting two types of modem spirals, Humprey (HC 1870) and Reichart (HG 8/7) which have limited capacity. For this reason this investigation designed a mathematical model with which larger units can be designed for increasing the capacity of the spiral."

Jan 1, 1999

By Hui-Gai Li

The fine Ti oxide inclusions precipitated during solidification are favorable for intragranular ferrite nucleation. The oxygen content before solidification affects the composition and size of inclusions. The composition of precipitated inclusions in Mn-Ti alloyed steel was studied by means of thermodynamic calculation and experiment in this paper. The micro-segregation of Mn, Ti and O during solidification was considered in calculation and unidirectional solidification apparatus was employed in experiment to ensure that alloy elements diffused along solidification direction. The results showed that: inclusions are mainly Ti oxide when oxygen content is 10ppm; they are Ti oxide and Mn-Ti oxide when it is 20ppm; Mn-Ti oxide dominates when it is 30ppm. With the increase of oxygen content, the size of inclusions increases. The maximum oxides size is about 3µm from statistical results of experiment.

Jan 1, 2008