Search Documents

By Anestis Vlyssides, Elli Maria Barampouti, Sofia Mai, ?postolos Vlyssides

"A new anaerobic reactor variation, the UASB/CO2-stripper reactor was conceived. It consisted of a conventional UASB reactor that recirculated the biogas produced. The sludge characteristics of this system were examined when treating high strength wastewater. From the data collected, it was obvious that the UASB/CO2-stripper system can satisfactorily operate at higher loading rates (25 g COD·L-1·d-1). A pronounced effect on the quality of the granular sludge was observed. The VSS/TSS ratio decreased sharply. A considerable decrease in the mean granule diameter was observed, while the specific gravity of the granules increased. The sludge bed porosity increased significantly leading to a considerable expansion of the sludge bed. These changes in the sludge properties led to a gradual shift of the system towards a fluidized bed process as far as its hydraulic characteristics were concerned.IntroductionIn recent years energy considerations and environmental concerns have increased the interest in direct anaerobic treatment of industrial wastes. The anaerobic method of waste treatment offers, under the present circumstances, a number of significant advantages with trivial or more serious drawbacks over other treatment methods. The major obstacle to the application of the process for this specific purpose is perhaps the difficulty of extending it to a stable and simple operational form. A number of reactor configurations has been developed leading to high biomass concentrations, e.g. the contact process, the filter reactor, the UASB reactor, and the fluidized bed systems, but the process stability problem still remains [1].A prime objective in the operation of anaerobic waste treatment process is the maintenance of a proper pH range. Thus, a possible solution to the process stability problem could be obtained by increasing alkalinity in the anaerobic systems. Therefore, the main objective of this paper is to present and evaluate the results of a system (UASB/ CO2-stripper) with increased alkalinity concentrations."

Jan 1, 2008

By V. A. Luganov

During the treatment of raw materials containing arsenic and antimony, problems in the distribution of these elements in the process products occur. In a number of cases in processing of copper-bearing materials, it is possible (or necessary) to obtain speiss. The purpose of the present research is to study the theoretical aspects and analyze the technological possibilities of producing copper-bearing speiss and the subsequent electrometallurgical processing with copper extraction as well as arsenic and antimony utilization. Copper arsenide (Cu3As) and copper antimonide (Cu3Sb) are the main components of the speiss. A thermodynamic analysis of the electrochemical oxidation of Cu3As and Cu3Sb, carried out using Pourbaix diagrams, showed that in a pH range from 0 to 2, oxidation potential 0.45-0.55V, ions of copper and arsenic and antimony trioxide are in the equilibrium. The kinetic study has shown, that with an increase in the polarization potential, the dissolution process changes from chemical to diffbsion control. An increase in temperature stabilizes the electrolysis process due to dissolution of arsenic trioxide forming on the anode. Electrolysis of speiss using a cathode diaphragm with the solution pH = 1-2, current density 150-200 A/m2 and temperature 60-70°C makes it possible to obtain cathode copper with a current efficiency of 95-96% and to remove the arsenic from the process during the electrolyte cleaning process. The arsenic content in the cathode copper did not surpass 0.002%. As the result of the research, a technological flowsheet for copper-lead speiss processing was developed. At the present time this technology is undergoing pilot testing.

Jan 1, 1999

By N. Solak

CrN, TiN and Ti-B-N coatings were deposited onto H13 hot-working tool steel and alumina substrates by arc-PVD. The coatings were characterized with respect to their mechanical and structural properties. Oxidation behavior of the coatings on alumina substrates was investigated by TG/DTA. Samples coated on H13 substrates were oxidized in a tube furnace under atmospheric conditions at specific temperatures determined from TG/DTA tests of the same coatings deposited on alumina substrates. The oxide scale morphology was characterized by scanning electron microscopy and X-ray diffraction. Oxide scale thickness change and TG data were used to calculate oxidation reaction activation energies. It was determined that boron addition into TiN coating improves not only mechanical properties but also oxidation resistance. Depending on the substrate material different oxidation activation energy values for the TiN and Ti-B-N coatings were determined revealing the importance of film morphology on oxidation resistance.

Jan 1, 2006

By Florian Kongoli

While calcium ferrite slags have been successfully used in continuous copper converting for the last 30 years, ferrous calcium silicate slag was proposed about 8 years ago as an additional alternative for copper smelting. Although both slags are normal extensions of each other their phase relations have not been completely clarified. In this paper the liquidus relations of both slags have been quantified at various areas that cover both of them. The existing experimental data have been first reviewed and discussed in relation to the solidification procedure. An original physical model is then used to predict various diagrams of the liquidus surface of both slags. It is shown that the ternary liquidus diagrams normally used for FeOx-SiO2-CaO system are not suitable for industrial slags. The new model diagrams of Fe/CaO versus SiO2 and Fe/SiO2 versus CaO as well as Fe/CaO versus Temperature and Fe/SiO2 versus temperature are proven to be much more convenient for ferrite and ferrous calcium silicate slags. Based on these diagrams the effect of SiO2 and CaO on the melting temperature is quantified. The effect of Cu2O is also quantified through the model. Some Fe/CaO versus Cu2O and Fe/SiO2versus CaO at various copper contents are predicted and the results show good agreement with the experimental data. It is also shown that in the presence of liquid copper, Cu2O is dissolved in the slag depending on the oxidation degree of the system and might lower the liquidus temperature. Some discussions are also presented taking into account plant observations and important conclusions are drawn for copper smelting and converting

Jan 1, 2006

By Florian Kongoli

While calcium ferrite slags have been successfully used in continuous copper converting for the last 30 years, ferrous calcium silicate slag was proposed about 8 years ago as an additional alternative for copper smelting. Although both slags are normal extensions of each other their phase relations have not been completely clarified. In this paper the liquidus relations of both slags have been quantified at various areas that cover both of them. The existing experimental data have been first reviewed and discussed in relation to the solidification procedure. An original physical model is then used to predict various diagrams of the liquidus surface of both slags. It is shown that the ternary liquidus diagrams normally used for FeOx-SiO2-CaO system are not suitable for industrial slags. The new model diagrams of Fe/CaO versus SiO2 and Fe/SiO2 versus CaO as well as Fe/CaO versus Temperature and Fe/SiO2 versus temperature are proven to be much more convenient for ferrite and ferrous calcium silicate slags. Based on these diagrams the effect of SiO2 and CaO on the melting temperature is quantified. The effect of Cu2O is also quantified through the model. Some Fe/CaO versus Cu2O and Fe/SiO2versus CaO at various copper contents are predicted and the results show good agreement with the experimental data. It is also shown that in the presence of liquid copper, Cu2O is dissolved in the slag depending on the oxidation degree of the system and might lower the liquidus temperature. Some discussions are also presented taking into account plant observations and important conclusions are drawn for copper smelting and converting

Jan 1, 2006

By Florian Kongoli

While calcium ferrite slags have been successfully used in continuous copper converting for the last 30 years, ferrous calcium silicate slag was proposed about 8 years ago as an additional alternative for copper smelting. Although both slags are normal extensions of each other their phase relations have not been completely clarified. In this paper the liquidus relations of both slags have been quantified at various areas that cover both of them. The existing experimental data have been first reviewed and discussed in relation to the solidification procedure. An original physical model is then used to predict various diagrams of the liquidus surface of both slags. It is shown that the ternary liquidus diagrams normally used for FeOx-SiO2-CaO system are not suitable for industrial slags. The new model diagrams of Fe/CaO versus SiO2 and Fe/SiO2 versus CaO as well as Fe/CaO versus Temperature and Fe/SiO2 versus temperature are proven to be much more convenient for ferrite and ferrous calcium silicate slags. Based on these diagrams the effect of SiO2 and CaO on the melting temperature is quantified. The effect of Cu2O is also quantified through the model. Some Fe/CaO versus Cu2O and Fe/SiO2versus CaO at various copper contents are predicted and the results show good agreement with the experimental data. It is also shown that in the presence of liquid copper, Cu2O is dissolved in the slag depending on the oxidation degree of the system and might lower the liquidus temperature. Some discussions are also presented taking into account plant observations and important conclusions are drawn for copper smelting and converting

Jan 1, 2006

By M. Urda-Kiel

Stainless steel is increasingly used as a cathode starting sheet. It was shown earlier that even when depositing copper on copper, a substrate-influenced zone appears in the deposit structure, followed by a transition zone and finally by the electrolysis-influenced structure. On copper, the substrate-influenced zone has a thickness of twenty micrometers and on titanium, eventually more than fifty micrometers. Initial nucleation and growth have a considerable influence. In this paper, copper was deposited in a channel cell on four different 304 stainless steel substrates: mechanically polished and anodized, respectively, at 0, +400 and +800 mV/SHE in 1M H2SO4 at 20°C. Copper was deposited under electrowinning conditions after solvent extraction (SX) (40 g/l Cu++, 180 gh/l, H2SO4, 20 or 60°C). The results showed that, although Auger surface analyses of the substrates were only slightly different one from another including the oxide thickness, nucleation and initial stages of growth were very sensitive to the surface treatment of the substrate. Potential step and current step methods gave similar results : progressive nucleation followed by 3D diffusion controlled growth. The highest nucleation frequency was observed on stainless steel anodized at 0 mV/SHE, under various electrolysis conditions. The potential influence of these findings on industrial practice is discussed.

Jan 1, 1999

By C. T. Yavuz

Higher environmental standards have made the removal of arsenic from water an important problem for environmental engineering. Iron oxide is a particularly interesting sorbent to consider for this application. Its magnetic properties allow relatively routine dispersal and recovery of the adsorbent into and from groundwater or industrial processing facilities; in addition, iron oxide has strong and specific interactions with both As(III) and As(V). Finally, this material can be produced with nanoscale dimensions which enhance both its capacity and removal. The objective of this study was to evaluate the potential arsenic adsorption by nanoscale iron oxides, specifically magnetite (Fe3O4) nanoparticles. We focused on the effect of magnetite particle size on the adsorption and desorption behavior of As(III) and As(V). The results show that the nanoparticle size has a dramatic effect on the adsorption and desorption of arsenic. As particle size is decreased from 300 nm to 12 nm the adsorption capacities for both As(III) and As(V) increase nearly 1000 times. Interestingly, this increase is more than expected from simple considerations of surface area and suggest that nanoscale iron oxide materials sorb arsenic through different means than bulk systems. The desorption process however, exhibits some hysteresis with the effect becoming more pronounced with small nanoparticles. This hysteresis most likely results from a higher arsenic affinity for magnetite nanoparticles. This work suggests that magnetite nanocrystals and magnetic separations offer a promising method for arsenic removal.

Jan 1, 2006

By A. A. Bustos

Four Air Liquide Shrouded Injectors (ALSI ) were installed and operated for over one year in a Peirce-Smith converter at the Falconbridge nickel smelter in Falconbridge, Ontario, Canada. A shrouded injector consists of an inner pipe and an annulus, through which high oxygen enriched air and nitrogen flow, respectively The nitrogen cools the injector and protects the refractory in the vicinity of the injector tip, allowing for substantial increases in oxygen flow rates to the converting reactions. Several injector designs were tested to study the effect of changes in the cross-sectional area and changes in the entrance geometry of the annulus. An optimum design would minimize the injection pressure required for a given flow rate to achieve choked flow conditions at the tip of the injector. The operation of the ALSI was simple and did not require any special care or maintenance, either between blows or between converter cycles. In particular, the shrouded injectors did not require punching, as is the case with conventional tuyeres. In terms of refractory performance, the ALSI, operating at 30-40 % oxygen enrichment, were superior or equal to the conventional tuyeres, which operated at 21-23% oxygen. The results of the demonstration program led Falconbridge Limited to implement the ALSI technology on a commercial scale The paper presents the demonstration program results in terms of gas flow rates, accretion formation, punchless operation and refractory wear.

Jan 1, 1999

By L. Z. Villanueva

An investigation and evaluation has been performed to determine the potential utilization of electron beam irradiation technology for sulfur oxides removal from reduced-S02-strength metallurgical gases from copper smelters. These off-gases, though too weak to be converted to usable, concentrated: sulfuric acid by conventional means, are characterized by SO2 content higher than 2,000 ppm and a complex chemical composition, including high oxygen content and its generation occurring in a cyclical, fluctuating pattern. This work was based on information supplied by principal smelter companies in Chile. A laboratory facility with flow rate of 20 Nm3/h and equipped with an electron accelerator of beam energy 8.00 keV, has been used for testing. Influence on SO2 removal efficiency of operational parameters such as gas temperature, water vapor content, ammonia feed ratio and irradiation dose was established. The main conclusion of this work, from experimental laboratory tests as well as engineering studies, is that the Electron Beam Process (EBDS) is a cost-effective, simple and appropriate chemical process means for emission abatement. EBDS can, on a site-specific basis, adequately augment SO2 removal to upgrade overall abatement as required and in an optimal manner.

Jan 1, 1999

By Allan J. MacRae

A copper alloy to be used for coils in cast copper cooling blocks has been identified and samples prepared for metallurgical examination and destructive testing to compare with cast cooling blocks manufactured from copper and nickel-copper alloy pipes. The copper alloy allows for the formation of a metallurgical bond between the pipe and the cast copper during the casting process, which does not normally occur with pure copper or nickel-copper alloy pipes. The alloy's coefficient of thermal expansion is also much closer to that of pure copper compared with nickel-copper alloys, so there is less strain induced at the pipe-cast copper interface due to differential thermal expansion. It is expected that this new pipe coil alloy and associated casting techniques will allow for the production of cast copper cooling blocks that have greatly enhanced service life in high heat load applications compared to existing technologies.

Jan 1, 2002

By Florian Kongoli

While calcium ferrite slags have been successfully used in continuous copper converting for the last 30 years, ferrous calcium silicate slag was proposed about 8 years ago as an additional alternative for copper smelting. Although both slags are normal extensions of each other their phase relations have not been completely clarified. In this paper the liquidus relations of both slags have been quantified at various areas that cover both of them. The existing experimental data have been first reviewed and discussed in relation to the solidification procedure. An original physical model is then used to predict various diagrams of the liquidus surface of both slags. It is shown that the ternary liquidus diagrams normally used for FeOx-SiO2-CaO system are not suitable for industrial slags. The new model diagrams of Fe/CaO versus SiO2 and Fe/SiO2 versus CaO as well as Fe/CaO versus Temperature and Fe/SiO2 versus temperature are proven to be much more convenient for ferrite and ferrous calcium silicate slags. Based on these diagrams the effect of SiO2 and CaO on the melting temperature is quantified. The effect of Cu2O is also quantified through the model. Some Fe/CaO versus Cu2O and Fe/SiO2 versus CaO at various copper contents are predicted and the results show good agreement with the experimental data. It is also shown that in the presence of liquid copper, Cu2O is dissolved in the slag depending on the oxidation degree of the system and might lower the liquidus temperature. Some discussions are also presented taking into account plant observations and important conclusions are drawn for copper smelting and converting

Jan 1, 2006

By Florian Kongoli

While calcium ferrite slags have been successfully used in continuous copper converting for the last 30 years, ferrous calcium silicate slag was proposed about 8 years ago as an additional alternative for copper smelting. Although both slags are normal extensions of each other their phase relations have not been completely clarified. In this paper the liquidus relations of both slags have been quantified at various areas that cover both of them. The existing experimental data have been first reviewed and discussed in relation to the solidification procedure. An original physical model is then used to predict various diagrams of the liquidus surface of both slags. It is shown that the ternary liquidus diagrams normally used for FeOx-SiO2-CaO system are not suitable for industrial slags. The new model diagrams of Fe/CaO versus SiO2 and Fe/SiO2 versus CaO as well as Fe/CaO versus Temperature and Fe/SiO2 versus temperature are proven to be much more convenient for ferrite and ferrous calcium silicate slags. Based on these diagrams the effect of SiO2 and CaO on the melting temperature is quantified. The effect of Cu2O is also quantified through the model. Some Fe/CaO versus Cu2O and Fe/SiO2versus CaO at various copper contents are predicted and the results show good agreement with the experimental data. It is also shown that in the presence of liquid copper, Cu2O is dissolved in the slag depending on the oxidation degree of the system and might lower the liquidus temperature. Some discussions are also presented taking into account plant observations and important conclusions are drawn for copper smelting and converting

Jan 1, 2006

By Henk Dijkman

Existing metallurgical operations can be optimized by implementing engineered bioreactor systems in the process configuration. For instance, metals can be recovered safely and economically from process and waste streams using biogenic sulfide. This paper focuses upon possible applications of high rate biotechnology in metallurgy and mining. Metal recovery from liquid streams (process bleed streams, leach water, waste streams), integrated removal of (fugitive) SO2, treatment of acid blowdown and other options will be described. Practical examples are given. A large-scale application of biotechnology at Pasminco's Budel Zink refinery is described. Here, acid plant blowdown is treated together with electrolyte magnesium bleed. Concentrated zinc sulfide (10 tpd) is produced biologically, eliminating the production of gypsum and closing the refinery sulfur cycle. Recovery of metals from solution at Kovehute Pribram in the Czech Republic is the second practical example. In this case, H2S is produced on-site for metal removal from an alkaline slag dump leachate. At the Caribou mine in New Brunswick, Canada, acid mine drainage is treated with biogenic sulfide. This plant is situated upstream of a lime neutralization plant. As a result, reusable ZnS is produced, the volume of the sludge (gypsum and hydroxides) from the lime plant is reduced by 50%, the sludge is free of heavy metals and the final effluent concentration of heavy metals is in the ppb range. In the second phase of this project, metals will be recovered from stored tailings, resulting in significant revenues.

Jan 1, 2002

By Pekka Hanniala

The falling real price level of copper metal and changes in the treatment and refining charges has been characteristic to the copper industry during the last few decades. Outokumpu's continuous development of the Flash Smelting Technology and increased productivity of the smelters based on these improvements may be one of the basic reasons that have influenced to these trends. From the early beginning of the Flash Smelting Technology, it has been characterized by low operation costs and low emissions, and the vision has been to develop the technology towards Direct Blister Flash Smelting. In Direct Blister Flash Smelting, both smelting and converting are carried out in one furnace making it possible to reduce the number of unit operation steps and thus lower both operation and investment costs. The merger of large copper production companies taken place during the last few years is a sign of a new age in copper production business. Now, the aim is to increase the productivity of copper production not just by utilizing the economy of scale but also by increasing integration and optimization of the whole production chain from mine to refinery. This also means increasing win- win based cooperation between companies. A Production Network concept based on the Flash Converting process enables existing smelters to decrease the operating costs and expand the smelting capacity with minimum investment and at the same time to decrease the emissions of the smelters. The new structure in ownership and partnership between copper production companies makes it easier to adopt the Production Network concept to both new and existing plants, where there exists a lot of built-in capacity. The paper presents Outokumpu's prediction on how the new options for the smelters will contribute to further changes in the structure of copper production. These visions are illustrated by giving examples on how the combination of matte production, Flash Converting, and Direct-to-Blister Flash Smelting could be adapted in existing and new smelters.

Jan 1, 2002

By Junji Shibata

As one of the effective usages of various wastes discharged in non-ferrous metal industry, the synthesis of hydrotalcite was carried out using aluminum dross as a raw material. Hydrotalcite is one of the inorganic anion exchangers and has layered structure of complex hydroxide. The removal of toxic metal ions such as As(III), As(V), Cr(VI) and Se(IV) with the hydrotalcite was investigated, and the removal properties were compared from the viewpoint of the anion exchange amount and exchange rate. Hydrotalcite can be obtained by a co-precipitation method from the aluminum dross and the waste MgCl2 solution discharged in an aluminum regeneration process. Slight amounts of SiO2, Al2O3, Fe2O3 and so on remain in the reaction product as an impurity component. Interlayer distance of the obtained hydrotalcite is about 0.3nm. Hydrotalcite changes to Mg-Al oxide by the calcination at 773K for 3h, and then hydrotalcite is formed again by a rehydration operation after calcination. The toxic metal ions are removed from aqueous solution by the anion exchange reaction in the pH region where toxic metal ions exist as the anionic species like AsO2-, HAsO42-,CrO42-, and SeO32-. The exchange amount is considerably concerned with the valence of anionic species and the size of anionic species. The removal of toxic metal ions were investigated from the engineering aspects such as the low concentration limit to be removed, the amount of hydrotalcite to be used for the perfect removal of toxic metal ions and so on.

Jan 1, 2006

By Abhishek Lahiri

Titanium dioxide is extracted from titaniferous ores by chloride and sulfate processes. These two processes produce wastes in the forms of iron chloride, iron sulfate and rare earth oxides gangue, each of which has to go through a further treatment before it can be disposed safely, which may be expensive. The shortage of high-grade ores is also making the two conventional processes less efficient. In view of the above mentioned problems, an alternate beneficiation process of roasting titaniferous ores with soda ash is discussed. The advantage of this process is that, there is no production of wastes. The impurities present in the ore can be selectively separated after the roasting process. The roasting of titaniferous ores with soda ash was carried out between temperatures of 873K and1173K for several hours. The roasting process resulted in the formation of complex alkali compounds and various sodium titanate compounds, depending on the Na2CO3to ore ratio. The phases formed after roasting were characterized using the X-ray powder diffraction (XRD), X-ray florescence (XRF) and scanning electron microscopy (SEM) techniques. The phase equilibria in the Na-Ti-O system are evaluated and compared with the phases obtained after the roasting process. The effects of temperature and the proportion of sodium carbonate on titaniferous ores are also discussed. From the experimental data, the overall reaction kinetics and phases formed were analyzed to ascertain the mechanism of overall reaction and the rate-limiting steps.

Jan 1, 2006

By Liping Li

A comprehensive three-dimensional two-phase mathematical model was developed to represent a channel type reactor with intense bottom gas agitation. Tracer dispersion and particle tracing methods were used to conduct the analysis based on the solution of the Navier-Stokes equations. Different operational parameters, such is gas volume flow rate. bubbler separation aspect ratios and liquid Input flow rates, were investigated for the purpose of reactor optimization. The numerical calculations were compared with experimental results obtained at the University of Missouri and a good general agreement was observed. Relatively low longitudinal mixing was observed characterizing the system as a "plug-flow" type reactor, in contrast to the intense vertical mixing, under the conditions investigated. Some general mixing characteristics are also discussed In connection to the behavior of other metallurgical reactors.

Jan 1, 1996

By F. Mucciardi

With the increasing trend to continuous smelting and converting, granulation of copper matte is gaining prominence as a means of decoupling the smelting and converting operations. Moreover, granulation is also being seen as a possible avenue to using higher grade oxygen in converters. By feeding cold solid matte, the oxygen concentration can be increased to compensate for the lower enthalpy feed. This leads to a higher grade of SO2 in the off gas and a lower total flow rate. Current granulation technology relies on the cooling of a falling matte film by water jets. This technology while effective has several drawbacks including those related to safety. Noranda and McGill University have developed a new process (patent pending) for solidifying copper matte (or slag). It overcomes many of the negatives of current technology. The system is based on an air cooled twin roll caster. This: new system uses no water and instead incorporates heat pipe technology to transfer the heat from the solidifying matte to the cooling air to produce a thin strip of 1 to 2 mm in thickness. Testing of a prototype unit demonstrated its viability. This paper describes this newly developed, twin roll casting system that is completely waterless. The advantages of this process as well as the findings from several tests are also detailed.

Jan 1, 1999

By Donato Firrao, Giorgio Scavino, Paolo Matteis

"The fatigue behavior of two high-strength sheet steels used to produce car bodies, a dual-phase steel and a TWIP one, both with ultimate tensile strength close to 1 GPa, is investigated.Unnotched tensile specimens are fatigue tested at room temperature with constant load amplitude and zero load ratio, in order to determine their 1 million cycles fatigue limit, both in the asfabricated sheet condition and after the resistance spot welding of one small homologous sheet square onto the calibrated length of the tensile specimen. Moreover, pre-cracked compact tension specimens are tested with load ratio 0.1 to determine the Paris fatigue crack growth curve.The fatigue results are completed with crystallographic, microstructural, tensile, and fractographic examinations, and the influence of the microstructure and of the welding processes is discussed.IntroductionCar bodies are increasingly made with high-strength weldable steels, in order to reduce vehicles' weight (to save fuel and reduce air pollution) and to improve the vehicle occupant safety in case of car crash. The main mechanical requirements of these steels in service are fracture strength and energy absorption, relevant for crashes, as well as fatigue endurance, relevant for ordinary car usage. [1]Steel sheets for car-body applications are usually made by continuous casting, hot rolling, cold rolling, continuous annealing (or other continuous heat treatment), and protective coating (in some cases). The sheets are then cold formed (deep drawn) to produce car body parts. Finally, the car bodies are built by assembling several deep-drawn parts, often consisting of different steels, usually through the Resistance Spot Welding (RSW) process, whereby two or three overlapping sheets are welded by local Joule-effect heating [2].Dual-Phase (DP) steels are widely used, low alloy steels with a microstructure consisting of ferrite and martensite, obtained from a final heat treatment which consists of heating at an intercritical temperature and then quenching. [3,4]"

Jan 1, 2012