La
Metallurgia Italiana
International Journal of the Italian Association for Metallurgy
n. 7/8 Luglio Agosto 2018 Organo ufficiale dell’Associazione Italiana di Metallurgia. Rivista fondata nel 1909
trentasettesimo convegno nazionale
AIM
Bologna // 12-13-14 settembre 2018 Organizzato da
con il patrocinio di
www.aimnet.it/37aim.htm Sponsorizzato da
Hardness Testers
HEAT TREATMENT
A c c r e d i t ed TM
GROUP
TAVENGINEERING
Heat Treating
La Metallurgia Italiana
La
Metallurgia Italiana
International Journal of the Italian Association for Metallurgy
n. 7/8 Luglio Agosto 2018 Organo ufficiale dell’Associazione Italiana di Metallurgia. Rivista fondata nel 1909
International Journal of the Italian Association for Metallurgy Organo ufficiale dell’Associazione Italiana di Metallurgia. House organ of AIM Italian Association for Metallurgy. Rivista fondata nel 1909
Direttore responsabile/Chief editor: Mario Cusolito Direttore vicario/Deputy director: Gianangelo Camona Comitato scientifico/Editorial panel: Livio Battezzati, Christian Bernhard, Massimiliano Bestetti, Wolfgang Bleck, Franco Bonollo, Bruno Buchmayr, Enrique Mariano Castrodeza, Emanuela Cerri, Lorella Ceschini, Mario Conserva, Vladislav Deev, Augusto Di Gianfrancesco, Bernd Kleimt, Carlo Mapelli, Jean Denis Mithieux, Marco Ormellese, Massimo Pellizzari, Giorgio Poli, Pedro Dolabella Portella, Barbara Previtali, Evgeny S. Prusov, Emilio Ramous, Roberto Roberti, Dieter Senk, Du Sichen, Karl-Hermann Tacke, Stefano Trasatti Segreteria di redazione/Editorial secretary: Valeria Scarano Comitato di redazione/Editorial committee: Federica Bassani, Gianangelo Camona, Mario Cusolito, Carlo Mapelli, Federico Mazzolari, Valeria Scarano Direzione e redazione/Editorial and executive office: AIM - Via F. Turati 8 - 20121 Milano tel. 02 76 02 11 32 - fax 02 76 02 05 51 met@aimnet.it - www.aimnet.it
siderweb LA COMMUNITY DELL’ACCIAIO
Gestione editoriale e pubblicità Publisher and marketing office: Siderweb spa Via Don Milani, 5 - 25020 Flero (BS) tel. 030 25 400 06 - fax 030 25 400 41 commerciale@siderweb.com - www.siderweb.com La riproduzione degli articoli e delle illustrazioni è permessa solo citando la fonte e previa autorizzazione della Direzione della rivista. Reproduction in whole or in part of articles and images is permitted only upon receipt of required permission and provided that the source is cited. Reg. Trib. Milano n. 499 del 18/9/1948. Sped. in abb. Post. - D.L.353/2003 (conv. L. 27/02/2004 n. 46) art. 1, comma 1, DCB UD Siderweb spa è iscritta al Roc con il num. 26116 Stampa/Printed by: Poligrafiche San Marco sas - Cormòns (GO)
n. 7/8 Luglio Agosto 2018 Anno 110 - ISSN 0026-0843
Science and technology in steelmaking / Scienza e tecnologia nella produzione siderurgica Reaction kinetics of molten iron oxides reduction using hydrogen M. Naseri Seftejani, J. Schenk 5 Hot ductility behavior of a high alloy steel J.-H. Min, S.-H. Kwon, S.-D. Lee, S.-W. Moon, D-.K. Kim, J.-S. Lee, Y.-U. Heo, C.-H. Yim 15 The interfacial convection in fluxes in the continuous casting process P. R. Scheller, Y. Lin, Q. Shu 23 Attualità industriale / Industry news Manifestazioni AIM
30
Bypassing Problems Related to Water Cooling: Case Study for Applying ILTEC in a 100-ton EAF edited by: M. B. Hanel – Mettop GmbH, Leoben Austria A. Filzwieser – Mettop GmbH/PolyMet Solutions GmbH Leoben, Austria R. Degel – SMS group GmbH, Düsseldorf, Germany 31 Latest results in EAF optimization of scrap-based melting process: Q-MELT installation in Kroman Celik edited by: M. Ansoldi, D. Patrizio – Danieli & C. Officine Meccaniche, Buttrio, Italy M. Piazza – Danieli Automation S.P.A., Buttrio, Italy O. Kuran – Kroman Celik San. A.Ş., Darıca/Kocaeli, Türkiye 41 Water leak detection in EAf based on Tenova’s off-gas technology: recent developments and results in lucchini RS, Lovere, Italy edited by: M. Luccini, V. Scipolo, D. Zuliani - Tenova Goodfellow Inc. L. Poli - Steelmaking and Casting Line Manager D. Masoero - Tenova S.p.a. 58 Scenari / Experts' Corner Strong Potential of Commercialized High Mn Steel Products and Process for Various Applications edited by: J. Choi – POSCO technical research laboratories, Republic of Korea 69 Atti e notizie / Aim news Calendario degli eventi internazionali / International events calendar 78 Rubrica dai Centri
79
l’editoriale La Metallurgia Italiana
Prof. Carlo Mapelli Politecnico di Milano
4
Ogni edizione dell’International Congress on Science and Technology of Steelmaking è un’interessante occasione per stabilire quali possano essere gli sviluppi dell’industria siderurgica nel prossimo futuro a livello globale. L’evento appena conclusosi a Venezia e organizzato da AIM non ha fatto eccezione, anche in forza della significativa partecipazione di ricercatori ed operatori siderurgici sia europei che asiatici. E’ possibile delineare almeno quattro linee di sviluppo che sono state suggerite dalla partecipazione e dalle discussioni svoltesi durante le sessioni tecniche. 1. Il crescente coinvolgimento della siderurgia asiatica verso la filiera da forno elettrico, in cui la siderurgia europea, ed in particolare quella italiana, rappresentano un punto di riferimento a livello globale. Questo segnala la progressiva metamorfosi della siderurgia asiatica da una configurazione tipica dei paesi in via di sviluppo, che puntano ad incrementare la massa netta di acciaio in circolazione, verso quella dei paesi sviluppati che puntano maggiormente al riciclo del deposito di rottame presente sui propri territori. 2. L’automazione considerata come elemento determinante per accrescere la competitività e la sostenibilità del sistema siderurgico. La sostenibilità non va però intesa solo come crescita di efficienza del sistema in termini di risparmio energetico e diminuzione degli sfridi, ma come strumento per migliorare le condizioni di lavoro degli operatori, soprattutto in contesti che vedono crescere l’età media degli stessi. Dato che le risorse economiche a disposizione non sono infinte e l’industria siderurgica si caratterizza per un utilizzo intensivo dei capitali, è necessario compiere scelte strategiche su come ripartire gli investimenti in automazione, ossia scegliere se privilegiare sistemi che collaborino con il personale oppure se privilegiare sistemi basati sullo sviluppo della cosiddetta intelligenza artificiale. 3. L’applicazione di quest’ultima ai sistemi si produzione siderurgica non appare un aspetto scontato, poiché risulta essere ancora piuttosto carente la capacità di modellizzare adeguatamente ed in linea alcuni aspetti fondamentali concernenti l’interazione tra scoria ed acciaio, la formazione e il movimento delle inclusioni non metalliche, i processi di solidificazione e le relative segregazioni. 4. Il crescente interesse per acciai austenitici caratterizzati dall’alligazione con elevate concentrazioni di manganese, dove la coreana POSCO ha evidenziato il proprio vantaggio competitivo acquisito rispetto a tutti i concorrenti in un settore dalle grandi prospettive e in forza della combinazione di tenacità, resistenza e duttilità che caratterizzano tali tipologie di acciaio. Oltre a riservare i dovuti complimenti all’impeccabile organizzazione della struttura operativa di AIM, gli atti del congresso non rappresentano solo un buon compendio dello stato dell’arte ma anche una raccolta di spunti di riflessioni e di sfide che i mondi della ricerca e della produzione siderurgica stanno vivendo o dovranno presto affrontare.
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica
Kinetics of molten iron oxides reduction using hydrogen M. Naseri Seftejani, J. Schenk
In conventional iron- and steel- making routs, hydrogen is used in molecular state. However, in hydrogen plasma smelting reduction (HPSR), hydrogen is used in plasma state to reduce iron oxide. In this study, the kinetics of molten iron oxide reduction using hydrogen in molecular and plasma state are discussed. HPSR has been established as an alternative to existing iron- and steel-making processes to reduce the emission of CO2. In HPSR, a H2-Ar mixture is injected into the plasma arc zone in a plasma reactor. Hydrogen particles become partially atomized and ionized. Thermodynamics calculations show that hydrogen particles in plasma state are strong reducing agents for iron oxides. This study provides an overview for generating hydrogen plasma in an arc plasma reactor and compares the reduction rate of iron oxide using various reducing agents. The results show that hydrogen in the plasma state is the strongest reducing agent. In addition, to increase the reduction rate of iron oxide, the temperature of the hydrogen particles should be increased.
KEYWORDS: HYDROGEN PLASMA – SMELTING REDUCTION– HPSR – IRON OXIDE– PLASMA ARC
INTRODUCTION The use of coal, coke, and natural gas in iron- and steel-making processes has led to the generation of the greenhouse gas CO2. The quantity of CO2 emitted depends on the iron- and steel-making route. The integrated HyL3-electric arc furnace process route, which produces 1125 kg/ton of hot metal, is the best steel-making route in terms of greenhouse gas emissions [1]. The iron and steel industries produce 7% of the total anthropogenic CO2 emissions [2]. Hydrogen plasma smelting reduction (HPSR) uses hydrogen to reduce iron oxides to directly produce crude steel products, thus eliminating the use of carbon. Therefore, HPSR is considered a next-generation steel-making process. Hence, understanding the kinetics of this method is important for controlling the process. HPSR provides good reducing conditions by the application of hydrogen and high plasma temperature to intensify the reduction processes; thus, it not only facilitates the production of iron in a onestage process but also prevents the introduction of carbon in the product, which in turn makes the metallurgical equipment compact [3]. It has been reported that reducing iron ores by hydrogen is 20% cheaper than conventional steel-making processes; moreover, steel produced in this manner is of higher quality and has greater flexibility [4]. In HPSR, plasma is generated by creating an electric arc between a hollow graphite electrode as the cathode and molten iron oxide bath as the anode, with continual input of a mixed gas containing argon and hydrogen. The basic flowsheet of the HPSR process has been presented in our previous study [5]. In the process, argon or nitrogen is used to conduct the current
La Metallurgia Italiana - n. 7/8 2018
in the plasma arc; argon is preferred due to its low ionization energy and high conductivity. Hydrogen operates as the reducing agent; hence, a mixture of hydrogen and argon is injected into the arc zone in the reactor through the hollow graphite electrode. Collision of electrons with hydrogen molecules at high temperatures leads to the activation of the hydrogen molecules. The injection of gases through the electrode directly to the arc zone guarantees optimal conditions for atomization and ionization. Excited hydrogen molecules provide a potentially very useful
Masab Naseri Seftejani, Johannes Schenk Chair of Ferrous Metallurgy, Montanuniversitaet Leoben, Leoben, Austria
5
Science and technology in steelmaking way for the reduction of stable metal oxides. Because monoatomic hydrogen is unstable, the production of pure hydrogen with an industrially usable lifetime is almost impossible. Nevertheless, in plasma arc discharge reactors, mixtures of H and H2 and hydrogen ions can be produced. Produced metastable mixtures have a suitable lifetime to reduce iron ores [6]. PHYSICAL AND THERMODYNAMIC PROPERTIES OF HYDROGEN THERMAL PLASMA HPSR consists of a plasma reactor in which the plasma arc formed at the tip of the hollow graphite electrode leads to the atomization and ionization of hydrogen molecules. Plasma arc is formed by the light electrons (negative particles) and heavy hydrogen ions (positive particles), which are two different gases. Among the various ionization processes of atoms, the
dominant ionization process in the electric arc (HPSR process) is collisional ionization by energetic electrons [7]. In HPSR, the arc temperature is between 20000 °C and 25000 °C. Thus, the atomization and ionization of hydrogen molecules take place in the arc zone. In contrast, with increasing distance from the arc, the recombination process becomes more dominant. The rates of ionization and recombination should be balanced by the reciprocal process at thermodynamic equilibrium [7–9]. Fig. 1 shows the dissociation and ionization of a mixture of 50 mol.% molecular hydrogen and 50 mol.% molecular argon with respect to plasma temperature, as calculated by FactSageTM 7.1 at equilibrium. It is obvious that above 15000°C, the amount of ionized hydrogen (H+) exceeds the amount of atomized hydrogen (H ).
Fig. 1 – Composition of Ar−H2 mixed plasma as a function of temperature (FactSageTM 7.1, Database FactPS)
Various phenomena may occur due to the collision of two particles in plasma. Particles can change their energy or momentum, which in turn can lead to neutral particles getting ionized and ionized particles getting neutral. The dominant collision mechanism for electrons colliding with atoms is elastic scattering. In this mechanism, at first, the electron momentum is changed and then excitation and ionization occur. Similarly, the collision 6
between ions and atoms is also governed by elastic scattering [9, 10]. During HPSR, almost all types of collisions occur in the arc zone. The amount of energy transferred during collision between electrons and atoms is negligible. In contrast, the kinetic energy significantly changes during atom-molecule collisions [11].
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica REACTION RATES OF HYDROGEN IONIZATION IN PLASMA STATE When electrons collide with atoms, the atoms get ionized. The
following equation is used to determine the degree of hydrogen ionization at temperature T:
(1) where x = nr+1⠄nr is the degree of ionization, nr and nr+1 are the number densities of atoms in the r and r + 1 state, respectively, T is the temperature, ℎ is Planck’s constant, K is the Boltzmann constant, and K is the number of hydrogen atoms [7]. Obviously, the ionization fraction sharply increases with increasing temperature. According to the Maxwell-Boltzmann distribution function, the mean velocity of particles depends on the square root of the temperature [7, 10–12]. Dembovsky [13] reported that the density of the active particles in an electrically insulated surface
and in the absence of an externally applied electric field depends on the particle temperature and the density of the particles. Therefore, there are two ways to increase the number of active hydrogen particles in HPSR: (1) by increasing the injection flow rate of hydrogen gas and (2) by increasing the temperature of the plasma arc. To calculate the production rate of new electrons per unit volume, the ionization collision frequency is multiplied by the electron density ne. This source rate Se of electrons is given by
(2) where ne and nn are the electron density and neutral atom density, respectively, đ?œŽion is the collision crosssection for electronimpact ionization, and ve is the electron velocity. In contrast, the rate of recombination is determined by the sink rate. The
value is the inverse of Se. The following formula shows a good approximation of the ionization rate (đ?œŽionve) of hydrogen atoms by the Maxwellian distribution as a function of electron temperature Te [7, 14, 14]:
(3) The equation shows that at partial ionization of hydrogen, temperature is the main influencing parameter. In this process, electrons collide with hydrogen atoms.
To obtain the rate of hydrogen ionization, the following reaction is considered:
(4) If the relative velocity of the hydrogen particle H and electron e is VHE and VHe is the corresponding collision cross-section for
these two particles, the probability of the collision of an electron with a hydrogen atom is given by
(5)
La Metallurgia Italiana - n. 7/8 2018
7
Science and technology in steelmaking where nH is the density of hydrogen atoms. In addition, the probability of total collisions is
(6) where ne is the density of electrons and KHe is the collision rate or reaction rate [7, 11]. KINETICS OF MOLTEN IRON OXIDE REDUCTION USING HYDROGEN In HPSR, the plasma arc zone is limited to a cone shape between the tip of the graphite electrode and the surface of the liquid iron oxide. Therefore, only this zone is at the plasma temperature, and it is the preferred area for atomization and ionization. The temperature of the surrounding area decreases with increasing distance from the arc zone. The gas temperature in the freeboard region of the reactor is about 1700 °K. Therefore, it is important that the reduction reactions in the arc zone take place in a short period. Hence, the kinetics of the reduction using hydrogen in the plasma state plays an essential role in this process. Many studies have evaluated the reduction rate of solid iron oxides in conventional steel-making processes involving hydrogen and other reducing agents. However, research on the kinetics of liquid iron oxide reduction is scarce. A few researchers have evaluated the reduction of iron oxides using hydrogen in plasma state. HPSR is one of the main important topics in terms of reducing liquid iron oxide in a plasma reactor using hydrogen and was proposed by the Chair of Ferrous Metallurgy of Montanuniversitaet Leoben in 1992. A lab-scale plasma facility, equipped with a mass spectrometer to analyze the off gas, was installed at the Chair of Ferrous Metallurgy. The reduction
rates of different iron ores and iron oxides by hydrogen in molecular and plasma states were evaluated previously [15–20]. The results showed that the reduction rate was improved with increasing hydrogen flow rate. In contrast, the hydrogen utilization degree was lower at high flow rates. Katayama et al. [21] investigated the reduction rate of liquid iron ore by a mixed H2−H2O gas to simulate blast furnace slag. Nagasaka et al. [22] reported that the reduction rate of pure liquid iron oxide by hydrogen was very high and gas-phase mass transfer was predominant. The reduction of hematite by microwave-assisted non-thermal hydrogen plasma was studied by Rajput et al. [23]. They indicated that hematite was reduced by excited hydrogen molecules even at temperatures below 573 °K. However, molecular hydrogen could not reduce hematite under the same conditions. The reason was that plasma dissociated and excited hydrogen molecules, which in turn decreased the activation energy from 46 to 5.36 KJ/mol [24]. Mechanism of iron oxide reduction by hydrogen Plasma smelting technology for the reduction of iron ores has been investigated since the late 1970s [15, 19, 20, 25, 26]. The stoichiometric equation of hematite reduction by hydrogen is given by
(7)
For a reaction to take place, the reactants must first collide to cope with the activation barrier. The increase in the temperature of the system increases the number of molecules with sufficient energy to react. The reduction rate is defined by the slowest step of the process, which restricts the overall reaction rate. Kamiya et al. [27] presented a mechanism for iron oxide reduction using hydrogen plasma through the following steps: 1- Mass transfer of hydrogen in the gas phase to the reaction zone 2- Mass transfer of oxygen through a liquid film from molten iron oxide to the reaction interface 3- Adsorption of hydrogen particles at the reaction interface
8
4- Adsorption with dissociation of iron oxide at the reaction interface 5- Reduction at the reaction interface and formation of water vapor 6- Desorption of water vapor from the reaction interface 7- Mass transfer of water vapor from the reaction interface Hayashi et al. [28] proposed the mechanism of iron oxide reduction using hydrogen in a gas-conveyed system (Fig. 2). In this model, the mass transfer step through the gas flow and interfacial reduction reactions were considered.
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica
Fig. 2 – Reduction reaction model for a single iron oxide particle [28] Nagasaka et al. [22] suggested the following limiting steps for the reduction of iron oxide using hydrogen: 1. Mass transfer in gas flow 2. Interfacial reduction reaction 3. Mass transfer of oxygen in the liquid phase to the reaction surface They represented that the rate of reduction in a gas-liquid system was mainly defined by the chemical reaction at the inter-
face when sufficient flow rate of the reducing agent was used. IRON OXIDE REDUCTION RATE WITH HYDROGEN The reduction rate of pure liquid Fet0 depends on the partial pressure of H2 in the H2−Ar mixture. As shown in Fig. 3, the reduction rate is determined by the interfacial chemical reaction rate with respect to hydrogen partial pressure.
Fig. 3 – Effect of H2 partial pressure in gas mixture on the apparent chemical reaction rate in the reduction of pure liquid iron oxide using hydrogen at 1673 K [29]
La Metallurgia Italiana - n. 7/8 2018
9
Science and technology in steelmaking The reduction rate is given by
(8) where r is the specific reduction rate (Kg−oxygen/m2s), Ka[H2] is the apparent rate constant for hydrogen reduction (Kg−oxygen/ m2s Pa), PH2 and PH20 (Pa) are the H2 and H2O partial pressures, respectively, and K'H is the gas ratio of (PH2 ⁄ PH20) at equilibrium with liquid Fet0 and pure solid iron. For this reaction, the specific reduction rate or apparent rate constant Ka[H2] using
hydrogen was 1.6 × 10-6(Kg−oxygen/m2s Pa) at 1673 K. As a result, (PH2 −PH20⁄ K'H) was the driving force of the reaction. The following equation shows the reduction rate of liquid iron oxide by CO with consideration for the gas composition and partial pressure of CO:
(9) where Ka[CO] is the apparent rate constant for CO reduction (Kg−oxygen/m2s Pa), Pco and Pco2 (Pa) are the CO and CO2 partial pressures, respectively, and K'CO is the gas ratio of (Pco / Pco2). The apparent rate constant for hydrogen reduction is approximately two orders of magnitude greater than that for CO reduction. Therefore, the reduction rate of liquid Fet0 by hydrogen was faster than that by CO. Hayashi and Iguchi [28] studied the reduction of pure liquid Fet0 by hydrogen at 1773 °K. In their experiments, very fine Fet0 particles were reacted with H2−N2 mixture in a particle-gas conveyed system. The flow rate and hydrogen concentration were in the range of 0.6–3 Nl/min and 5–30 vol%, respectively. They showed that the chemical reaction rate was controlled by PH2 in the gas phase, and the apparent rate constant was reported to be 1.58 × 10-6 (Kg−oxygen/m2s Pa), which was in good agreement with the results of Nagasaka et al. [29]. Moreover, the reduction rate above the wustite melting point was 13.2 times higher than that in the solid phase.
In HPSR, fine iron oxide is injected into the plasma zone via a hollow graphite electrode that is in contact with partially ionized hydrogen at high temperatures. Therefore, the inflight reduction should be taken into account. Moreover, the most important advantages of HPSR are the use of ionized hydrogen and the high temperatures of the iron ore particles. Kamiya et al. [27] have constructed a lab-scale plasma furnace for investigating the reduction rate of iron ore, which is shown in Fig. 4. A thoriated tungsten electrode was used as the cathode with consideration of the non-transferred arc attachment. A mixture of hydrogen and argon was injected into the plasma zone via a plasma torch cooled by water. Iron oxide was melted in a 60-mm-diameter water-cooled copper crucible. After complete melting of iron oxide in the crucible, the plasma gas (H2−Ar mixture) was then injected into the furnace. The gas flow rate was between 10 and 30 Nl/min for up to 75 g of iron oxide and the input DC power was 8.3 kW.
Fig. 4 – Schematic of experimental apparatus [27] 10
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica Kamiya et al. [27] reported that the reduction rate of liquid FeO was limited by the chemical reaction between FeO and atomic hydrogen in the plasma arc zone. The reduction reaction took place in the cavity formed on the molten pool surface due to the momentum of the plasma jet gas. Moreover, when sufficient hydrogen was used, the reduction rate was independent of the flow rate. Their results were in agreement with previous results [25, 27], except for the effect of hydrogen concentration on hydrogen utilization. They showed that the hydrogen utilization degree was constant at about 44% for hydrogen concentrations between 10 and 30 vol% in the H2−Ar mixture and was about 60% at hydrogen concentration lower than 10 vol% [27]. However, Badr et al. [16] reported that with increasing hydrogen concentration, the degree of utilization decreased and the maximum utilization was achieved at a hydrogen concentration of 26 vol%. A higher degree of hydrogen utilization at a low concentration of H2 was in a good agreement with the results of Nakamura et al. [30]. Nakamura et al. [30] mentioned that the reduction rate might be improved by the separation of slag from the reduced iron. This behavior was also confirmed by Lemperle [25], who showed that a higher utilization degree was obtained at lower H2 concentrations due to the lower rate of recombination of hydrogen particles in plasma state.
Effect of basicity on reduction rate In HPSR, the temperature and basicity of the slag are the parameters influencing the reduction rate. Partial solidification of the slag leads to a decrease in the reduction rate because of limited oxygen transport to the reaction surface. Badr [16] experimentally assessed the effect of slag basicity on the reduction rate; however, no linear correlation was found. Further, at a high CaO content, the hydrogen utilization degree decreased. Kamiya et al. [27] also did not observe any obvious changes in the reduction rate on varying the basicity from 0 to 2. Nagasaka et al. [24] studied the effect of additives such as CaO, Al2O3, and TiO2 on the reduction rate of liquid Fet0 by hydrogen at different flow rates of H2−Ar mixture. They observed that an increase in the gas flow rate improved the reduction rate of iron oxide, i.e., the flow rate played an important role in these binary slags. As a result, the mass transfer process was the rate-limiting step in the gas phase. Comparison of iron oxide reduction rates with hydrogen by various researchers Fig. 5 shows the reduction rate of FeO by hydrogen in solid and liquid states and at the melting point of FeO reported by different studies.
Fig. 5 – Specific reduction rates of molten FeO and wustite by pure H2 and CO at various temperatures [16]
As can be seen, the reduction potential of hydrogen increased significantly with respect to that of the solid phase just above the FeO melting point. However, above the melting point, the
La Metallurgia Italiana - n. 7/8 2018
logarithms of the reduction rate increased linearly with increasing the temperature up to the plasma temperature.
11
Science and technology in steelmaking Comparison of reduction rates of molten iron oxide by hydrogen, solid carbon, Fe-C melt, and CO In terms of the reduction of molten iron oxide by CO, Nagasaka et al. [22] represented most of the related studies [31–33] in a plot for comparison. The results were in good agreement with each other, except for those of Kato et al. [16], which were later corrected by Soma [31]. Nagasaka et al. [22] showed that the reduction rate of FeO by CO increased by almost one order of magnitude just above the melting point. Many researchers have studied the reduction of molten FeO by solid carbon [32, 34–36]. One of the main results was that the reduction rate of pure liquid FeO by solid carbon was signifi-
cantly affected by temperature. Moreover, the reduction rate of liquid iron oxide by Fe−C was significantly higher than that by solid carbon up to 1893 K. A comparison of the reduction rates of liquid iron oxide by hydrogen, solid carbon, Fe−C melt, and CO is represented in Fig. 6. The reductions by H2 and CO were referred to the interfacial chemical reaction. In contrast, for the reductions by solid carbon and Fe−C melt, the overall rates were considered. Consequently, the reduction rate by hydrogen was one or two orders of magnitude higher than those by the other reductants [22].
Fig. 6 – Comparison of reduction rates of pure liquid iron oxide by solid carbon, Fe−C melt, CO, and H2 [22] Summary and conclusion Thermodynamic calculations at equilibrium show that hydrogen and argon are ionized at plasma arc temperatures. Moreover, the hydrogen species in plasma state is the strongest reducing agent in terms of Gibbs free energy. The reduction abilities of different hydrogen species can be expressed in the following order: H+>H+2>H+3>H>H2. To evaluate the reduction of iron oxide by hydrogen, the reduction rate should be taken into account and the ionization degree of the hydrogen particles should be assessed. To cover those, the main parameters 12
influencing the ionization degree are analyzed. It is found that the ionization degree increases with increasing temperature and number density of the hydrogen particles. Finally, the reduction rates of iron oxides by various reducing agents are studied and the reduction abilities are graphically compared. Consequently, the iron oxide reduction rate by hydrogen is one or two orders of magnitude higher than those by other reductants.
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica REFERENCES [1]
Lisienko VG, Lapteva AV, Chesnokov YN, Lugovkin VV. Greenhouse-gas (CO2) emissions in the steel industry. Steel Transl. 2015; 45(9):623–6.
[2]
Kim Y, Worrell E. International comparison of CO2 emission trends in the iron and steel industry. Energy Policy 2002; 30(10):827– 38.
[3]
Fridman AA. Plasma chemistry. Cambridge, New York: Cambridge University Press; 2008.
[4]
Hiebler H, Plaul JF. Hydrogen plasma- smelting reduction- an option for steel making in the future [METABK] 2004 [cited 2016 Jan 20]; 43(0543-5846):155–62.
[5]
Naseri Seftejani M, Schenk J, editors. Fundamentals of hydrogen plasma smelting reduction (HPSR) of iron oxides, a new generation of steelmaking processes. Conference proceedings; 2018. (Iron and steel Society).
6]
Robino CV. Representation of mixed reactive gases on free energy (Ellingharn-Richardson) diagrams. Metall and Materi Trans B 1996; 27(1):65–9.
7]
Keidar M, Beilis I. Plasma Engineering: Applications from aerospace to bio and nanotechnology. San Diego: Elsevier Science; 2013.
8]
Piel A. Plasma physics: An introduction to laboratory, space, and fusion plasmas / Alexander Piel. Heidelberg: Springer; 2010.
9]
Janev RK, Reiter D, Samm U. Collision processes in low-temperature hydrogen plasmas: 1Institut f¨ur Plasmaphysik, Forschungszentrum J¨ulich GmbH, EURATOM Association, Trilateral.
10]
Lichtenberg AJ, Lieberman MA. Principles of plasma discharges and materials processing. 2nd ed. Hoboken, N.J., Chichester: Wiley-Interscience; 2005.
11]
Feinman J. Plasma technology in metallurgical processing. USA: Iron and steel Society; 1987.
12]
Freemire B, Hanlet PM, Torun Y. Determining electron temperature and density in a hydrogen microwave plasma. Journal of Thermophysics and Heat Transfer 1996; 10(3):426–35.
13]
Dembovsky V. How the polarity of a surface reacting with a low temperature plasma affects the thermodynamic variables in metallurgical reactions. Achta Physic Slovenia 1984; 34(1):11–8.
14]
Goldston RJ, Rutherford PH. Introduction to plasma physics. Bristol: Institute of Physics Pub; 1995 (2000 [printing]).
15]
Plaul JF. Schmelzreduktion von hämatitischen Feinerzen im Wasserstoff-Argon-Plasma [Doktoral]. Leoben, Austria; 2005.
16]
Badr K. Smelting of iron oxides using hydrogen based plasmas [Doctoral Dissertation]. Leoben, Austria.
17]
Badr K, Bäck E, Krieger W. Reduction of iron ore by a mixture of Ar-H2 with CO and CO2 under plasma application. 18th International Symposium on Plasma Chemistry 2007.
18]
Kirschen M, Badr K, Pfeifer H. Influence of direct reduced iron on the energy balance of the electric arc furnace in steel industry. Energy 2011; 36(10):6146–55.
19]
Bäck E. Schmelzreduktion von Eisenoxiden mit Argon-Wasserstoff- Plasma [Doctoral Dissertation]. Leoben, Austria; 1998.
20]
Sormann A. Untersuchungen zur Schmelzreduktion von Eisenoxiden mit Wasserstoff als Reduktionsmittel [Doktoral Disserta-
La Metallurgia Italiana - n. 7/8 2018
13
Science and technology in steelmaking tion]. Leoben, Austria; 1992. 21]
Katayama, H., S. Taguchi, and N. Tsuchiya H, Taguchi s, Tsuchiya N. Reduction of iron oxide in molten slag with H2 gas. ISIJ Int. 1982 [cited 2016 Jan 20]; 68:2279–86.
22]
Nagasaka T, Hino M, Ban-ya S. Interfacial kinetics of hydrogen with liquid slag containing iron oxide. Metall and Materi Trans B 2000; 31(5):945–55.
23]
Rajput P, Bhoi B, Sahoo S, Paramguru RK, Mishra BK. Preliminary investigation into direct reduction of iron in low temperature hydrogen plasma. Ironmaking & Steelmaking 2013; 40(1):61–8.
24]
Sabat KC, Rajput P, Paramguru RK, Bhoi B, Mishra BK. Reduction of oxide minerals by hydrogen plasma: An Overview. Plasma Chem Plasma Process 2014; 34(1):1–23.
25]
Weigel A, Lemperle M, Lyhs W, Wilhelmi H. Experiments on the reduction of iron ores with an argon hydrogen plasma. ISPC-7 Eindhoven 1985:1214–9.
26]
Gilles HL, Clump CW. Reduction of iron ore with hydrogen in a direct current plasma jet. Ind. Eng. Chem. Proc. Des. Dev. 1970; 9(2):194–207.
27]
Kamiya K, Kitahara N, Morinaka I, Sakuraya K, Ozawa M, Tanaka M. Reduction of molten iron oxide and FeO bearing slags by H2-Ar plasma. ISIJ Int. 1984; 24(1):7–16.
28]
Hayashi S, (None), Iguchi Y. Hydrogen reduction of liquid iron oxide fines in gas-conveyed systems. ISIJ International 1994; 34(7):555–61.
29]
Ban-ya S, Ighuchi Y, Nagasaka T. Rate of reduction of liquid wustite with hydrogen. tetsu-to-hagane; 70:1689–96.
30]
Nakamura Y, Ito M, Ishikawa H. Reduction and dephosphorization of molten iron oxide with hydrogen-argon plasma. Plasma Chem Plasma Process 1981; 1(2):149–60.
31]
Soma T. Smelting reduction of iron ore. Bulletin of the Japan Institute of Metals 1982; 21(8):620–5.
32]
Tsukihashi F, Amatatsu M, Soma T. Reduction of molten iron ore with carbon. ISIJ Int. 1982 [cited 2016 Jan 21]; 22:688–95.
33]
Nagasaka T, Iguchi Y, Ban-ya S. Effect of additives on the rate of reduction of liquid iron oxide with CO. tetsu-to-hagane 1989; 75(1):74–81. Available from: URL: http://ci.nii.ac.jp/lognavi?name=nels&lang=jp&type=pdf&id=ART0001818203.
34]
Takahashi K, Amatatsu M, Soma T. Reduction of Molten Iron Ore with Solid Carbon. tetsu-to-hagane 1975 [cited 2016 Jan 21]; 61:2525–30.
35] Sasaki Y, Okamoto Y, Soma T. Kinetics of Reaction between Iron Oxide Slags and Solid Carbon. ISIJ International 1978; 64(3):367–75. Available from: URL: http://ci.nii.ac.jp/lognavi?name=nels&lang=jp&type=pdf&id=ART0001787448. 36]
Sato A, Aragane G, Kamihira K, Yoshimatsu S. Reduction Rate of Molten Iron Oxide by the Solid Carbon or the Carbon in Molten Iron. tetsu-to-hagane 1987 [cited 2016 Jan 21]; 73:812–9.
14
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica
Hot ductility behavior of a high alloy steel J.-H. Min, S.-H. Kwon, S.-D. Lee, S.-W. Moon, D-.K. Kim, J.-S. Lee, Y.-U. Heo, C.-H. Yim
Hot ductility behaviour is studied in a cast steel at the intermediate temperature range. A low carbon cast steel is heated to 1,673K (1,400°C) and cooled to 673-873K at the heating and cooling rate of 10K/s. Tensile test is performed at the target temperature. Ductility decreases abruptly, shows the lowest at 823K (550°C), and then recovers gradually as the tensile testing temperature decreases. Fracture surface shows the intergranular fracture with small craters at 823K. Further study on the fracture behaviour reveals that the upper bainite structure existing the cementite at the prior austenite grain boundary is crucial for the embrittlement. Segregated phosphorus triggers the initial micro-crack generation at the grain boundary cementite/matrix interface. Detailed embrittlement mechanism and the fabrication way for sound cast iron are discussed.
KEYWORDS: SLAB DEFECT - INTERGRANULLAR FRACTURE - PHOSPHORUS SEGREGATION REDUCTION IN AREA - INTERMEDIATE TEMPERATURE
INTRODUCTION Slab imperfection in the continuous casting process is one of the major factors causing economical loss. Most of slab imperfection is related with the defects on the slab surface. Finding the origin of those defects is important in views of understanding the detailed embrittlement mechanism in the academic field as well as saving production cost in the industrial field. Surface cracks occur mainly in the midface or corner region of cast steel, and typically include transverse cracks and corner cracks. One of the most important factors that affects transverse cracks is steel composition. Steel containing aluminum, niobium, vanadium and over 1 pct manganese is particularly susceptible to cracking. Hater et al. [1-7] have reported that the cracks occur at the austenite grain boundaries (GBs) which are enriched in aluminum, probably in the form of AlN. Similar to the effects of Al, steel containing Nb and V also precipitates MC carbide at GBs in between 973 K (700°C) and 1173 K (900°C), and shows intergranular cracking. As the application condition of the constructive steel moves to the harsh and severe environments, the needs for the production of high strength steels are increasing. To achieve high strength, low carbon steel includes a large amount of alloying elements such as Mn, Ni and Cu. However, the tendency of crack generation at the surface or sub-surface increases as the contents of those elements increase. In the fabrication of thick plate, quality control at the continuously-casted slabs is very important. To achieve the sound slab, the secondary cooling condition is typically set as a hard cooling condition, and so the slab surface temperature at the
La Metallurgia Italiana - n. 7/8 2018
unbending zone is controlled below 873K (600°C) [8,9]. Generally, it is reported that the ductility decreases significantly in the brittle zone between 873K and 1173K (900°C), and gradually recovered in the intermediate temperature range (below 600°C) [10]. However, the unexpected increase of crack generation in the slab containing large amount of alloying elements is observed at the intermediate temperature range. The detailed reason and embrittlement mechanism are not clarified yet.
J.-H. Min, S.-H. Kwon, S.-D. Lee, J.-S. Lee, Y.-U. Heo, C.-H. Yim Graduate Institute for Ferrous Technology, Pohang University of Science and Technology, 77Cheongam-Ro, Nam-Gu, Pohang 37673, Republic of Korea
S.-W. Moon Technical Research Laboratory, POSCO, Pohang 790-785, Republic of Korea
D-.K. Kim Dong-A University, 225 Gudeok-ro, Seo-gu, Busan, Republic of Korea
15
Science and technology in steelmaking It is the aim of this research that elucidates the embrittlement mechanism of the continuously-casted slab at intermediate temperature. Tensile properties are firstly obtained at the temperature range of 673K (400°C) ~873K (600°C). The extensive analyses from fracture surface to grain boundary precipitates are applied to clarify the detailed embrittlement mechanism. On the bases of the experimental results, the fracture mechanism and the governing factor of brittle fracture illuminate.
is fabricated by continuous casting process. The slab was machined to the tensile specimen as shown in Fig. 1(b). Tensile test was performed using a Caster&Thermomechanical simulator(40334, Fuji electronic industrial, Saitama). All the specimens were heated to 1673 K (1400°C) at a heating rate of 10K/s for the dissolution of precipitates, held for 300Sec. at the temperature, and then cooled to the target temperature(673, 723, 773, 823, and 873 K) at a cooling rate of 10 K/s (Fig. 1(a)). Tensile test was performed at the target temperature after holding for 60sec.
Experiment A slab specimen which has the chemical composition in Tab. 1
Fig. 1 – (a) Schematic drawing of heat treatment for tensile test. (b) Specimen size Stress-strain curves were obtained during tensile test. Displacement of the specimen converted to strain by dividing with total gage length. To confirm tensile behaviour, the test was repeated to maximum 3 times at the same target temperature. The reduction in area(RA) at the fractured specimen was measured [11]. Fracture surfaces were observed using a field-emission scanning electron microscope (FE-SEM, JSM-7100F, Japan Electron Optics Ltd. (JEOL), Tokyo). The micro-cracks below the fracture surface was also investigated using a FE-SEM and an FE electron probe micro analyzer (EPMA, JXA-8530F, JEOL, Tokyo). To observe the micro-crack, the fracture part of tensile-tested specimen was cut in half and the cross-sectional specimen was
prepared by micro-polishing and etching in a 3% Nital etchant. Segregation behavior of the impurity elements was investigated using a nano secondary ion mass spectroscope (NANOSIMS, NANO-SIMS 50, CAMECA, Gennevilliers Cedex).Cs+ gun maintained with current of 0.4 pA and impact energy of 16 keV was used for the detection of 12C-, 31P, 32S, 55Mn16O-, 63Cu. Cementite formation at the prior austenite GB and in a grain was also observed in an FE- transmission electron microscope (FE-TEM, JEM-2100F, JEOL, Tokyo) equipped with Oxford energy dispersive spectroscope (EDS, AZtec, Oxford instrument, Abingdon). Accelerating voltage for the observation was 200 keV and the beam size for EDS analysis was about 1 nm.
Tab. 1 – Chemical Composition (wt%)
16
C
Mn
P
S
Cu
Nb
Ni
Fe
0.05~0.07
1.6~2.0
0.0054
0.0010
0.1~0.3
0.02~0.04
0.30~0.50
Bal.
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica Results Fracture behavior of the cast steel at intermediate temperature Stress-strain curves obtained from the tensile test are shown in Fig. 2. Total elongation decreases, showsminimum at 823
K, and then increases again as testing temperature decreases. Comparing tensile curves at 873 and 823 K, yield stress is 1.5 times larger at 823 K than that at 873 K. There is the possibility of to bainite transformation at 823 K.
Fig. 2 – (a) Stress-strain curves and (b) RA values RA value showed abrupt decrease as testing temperature decreases from 873 to 823 K. Ductility recovers again at the lower temperature. Comparing the case [10] of low alloy steels which
show the recovery of ductility below 873 K, this is an unexpected result. To understand the reason of ductility variation, fracture surfaces were observed in FE-SEM.
Fig. 3 – FE-SEM images of fracture surfaces: (a) 873 K, (b) 823 K, (c) 773 K, (d) 723 K, (e) 673 K, and (f) high magnification image of fracture surface (823 K)
La Metallurgia Italiana - n. 7/8 2018
17
Science and technology in steelmaking Ductile fracture surfaces are shown at 673, 723, and 873 K where RA values are relatively high (Fig.s 3(a), (d), and (e)). However, intergranular fractures are observed at 773 and 823 K where RA values are low (Fig.s 3(b) and (c)). High magnification image reveals large number of small craters on the intergranular fracture surface at 823 K (Fig. 3(f)). Interestingly, small particles which are probability GB precipitates are observed in the core of craters. Role of GB cementite on the micro-crack formation Micro-cracks are observed at the tensile-fractured specimen at 823 K. As shown in Fig. 4(a), crack forms along the prior austenite GB (PAGB) at 823K. It was observed that GB cementite particles of 1Îźm or less in size distributed near the crack.
Martensite-austenite (M-A)constituent was present in the packet or block boundary. Bright field (BF)-TEM image of this specimen shows cementite formation at PAGB or lath boundary where the microstructure is close to upper bainite structure (Fig. 4(b)). On the other hand, cementite is mainly distributed inside of a grain at 673 K where ductile fracture occurs (Fig.s 4(c) and (d)). GB cementite is hardly observed at the GB. This is a typical low bainite structure. EDS spectra of cementite and matrix are compared in Fig. 4(f). Cementite shows slightly higher Mn content than that of matrix. It is considered that the GB cementite is related to GB embrittlement at 823 K. The M-A phase which is mainly distributed in the inside of the grain has probably no relation with intergranular fracture.
Fig. 4 – FE-SEM and BF-TEM images, and EDS spectra in the tensile-fractured specimen:(a) FE-SEM and (b) BF-TEM image of tensilefractured specimen at 823 K, (c) FE-SEM and (d) BF-TEM image of tensile-fractured specimen at 673 K, and (e) EDS spectra of GB precipitate and matrix. 18
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica The initiation of micro-crack at 823 K was investigated using EPMA analysis. Secondary electron (SE) and back scattered electron (BSE) images in Fig.s 5(a) and (b) show the formation of multiple micro-cracks at PAGB. GB cementite position was identified at the EPMA carbon map in Fig. 5(c). The correlation between GB cementite and micro-crack positions is studied by
super imposing GB cementite positions on SE and BSE images. Micro-cracks were formed at the GB cementite/matrix interfaces. EPMA sulfur map in Fig. 5(d) shows sulfur segregation at the each micro-crack region. This segregation of sulfur will further weaken the neighboring GB.
Fig. 5 – (a) SE and (b) BSE Image of micro-crack formed at 823 K. (c) and (d) are EPMA carbon and sulfur maps, respectively.
To clarify the origin of initial decohesion at the GB cementite/matrix interface, nano-SIMS analysis was conducted. Fig.6. shows the nano-SIMS maps at the crack-free GB in the tensilefractured specimen at 823 K. GB cementite is clearly identified in the carbon and manganese map (Fig.s 6(a) and (d)). Phosphorus is enriched around the GB cementite (Fig. 6(b)), and the weak sulfur segregation around the GB cementite is also observed(Fig. 6(c)). Copper also segregates around the GB cementite. Detailed segregation behaviors at the GB cementite/
La Metallurgia Italiana - n. 7/8 2018
matrix interface and the cementite-free GB are shown in Fig.s 6(f) and (g), respectively. Phosphorus segregation is dominant at the GB cementite/matrix interface. However, the cementite-free GB shows relatively low level of segregation. Phosphorus segregation at the GB cementite/matrix interface weakens cohesive strength at the interface. Therefore, the formation of micro-crack initiates at this weakest interface under tensile stress as shown in Fig. 5.
19
Science and technology in steelmaking
Fig. 6 – (a)~(e) Nano SIMS maps at 823 K, and line scan results of (f) GB cementite/matrix interface and (g) cementitefree GB.
20
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica Discussion Intergranular fracture mechanism of a cast steel at intermediate temperature Up to now, the embrittlement of steels at the intermediate temperature was studied actively in the high strength low alloy (HSLA) steel [12,13]. When the HSLA steel is isothermally held in between673 and 873 K, the ductilebrittle transition temperature (DBTT) increases. The intergranular brittle fracture occurs below the room temperature. This behavior is known as temper embrittlement. There were many attempts to interpret the origin of temper embrittlement. Impurity segregation such as phosphorus, zinc, tin, and so on, was explained as the possible origin of GB embrittlement. M. P. Seah [12] reported that the nose of embrittlement was mainly formed in between 773 and 873 K, and phosphorus segregation at the GB was maximized at the nose. This temperature dependent segregation behavior is related to the segregation kinetics. Even though the equilibrium GB segregation concentration of P increases as the holding temperature decreases, actual segregation concentration decreases at the low temperature due to its low diffusivity. Matrix phase which defines the diffusivity of impurity element is also an important factor for GB segregation. Diffusivity of phosphorus in is one order of magnitude smaller than that in [14]. GB phosphorus segre-
gation will be much smaller in -matrix than that in -matrix. Therefore, phosphorus segregation could be maximized at the upper bainite temperature region. In this study, the embrittlement behavior at intermediate temperature is related to the distribution of cementite at PAGB as shown in Figs. 3 to 6. Strong phosphorus segregation at the GB cementite/matrix interfaces has triggered the decohesion of those interfaces. Preferred segregation of impurities (P, S) at the GB precipitate/ matrix interface was reported in the several literatures [15-18]. This is understood as the higher interface energy at the phase boundary. On the base of experimental results, the intergranular fracture mechanism of a cast steel at intermediate temperature is established in Fig. 9. Cementite is formed at the PAGB. Active segregation of impurity (P) occurs at the GB cementite/matrix interface (Fig. 6). Formation of micro-cracks initiates at the weak GB cementite/matrix interface (Fig. 5). Micro-crack makes a free surface for the segregation. Sulfur segregates actively at the free surface which is formed by the decohesion of the GB cementite/matrix interface (Fig. 5)[19]. Sulfur segregation around the micro-cracks decreases the cohesive strength of the neighboring un-cracked GB. Micro-crack grows and coalescences with the neighboring micro-crack. Finally, intergranular fracture occurs by the linking of all the micro-cracks.
Fig. 8 – Intergranular fracture mechanism of a cast steel at intermediate temperature
La Metallurgia Italiana - n. 7/8 2018
21
Science and technology in steelmaking On the intergranular fracture surface, the small craters are remainders shown in Fig. 3(f). The crater is formed during accommodating deformation after the initial decohesion of GB cementite/matrix interface for a considerable time Conclusion Embrittlement mechanism of a cast steel which contains relatively large amount of alloying elements (Mn, Cu, Nb, Ni) was studied at the temperature range of 673~873 K. The characteristic embrittlement behavior was observed at the upper bainite temperature range. RA values increased again at the low bainite temperature range. Through the detailed studies on the embrittlement mechanism at the intermediate temperature, following conclusions have been reached: 1. Embrittlement of cast steels at the intermediate temperature range is predominantly influenced by P segregation.
2. Active segregation of P at the GB cementite/matrix interface is confirmed by nano-SIMS analysis. 3. Upper bainite structure where cementite precipitates at PAGB accelerates the embrittlement by the easy decohesion of GB cementite/matrix interface. 4. S segregation at the micro-crack area is confirmed by EPMA analysis. S segregation at the micro-cracks promotes the growth and coalescence of micro-cracks. 5. It is recommended that slab temperature is controlled to avoid the upper bainite temperature range at the unbending zone. Acknowledgement This work was supported by the project (Project. No. 4.0014031) from POSCO.
REFERENCES 1]
Hater M, Klages R, Redenz B: OpenHearthProc; 1973. Vol. 56. 202-217p.
2]
Sengupta J, Thomas BG, Wells MA: Metallurgical and Materials Transactions A; 2005. Vol.36. 187–204p
3]
Brimacombe JK, Sorimachi K: Metallurgical Transactions B; 1977. Vol.8. 489–505p
4]
Minz B, Abushosha R: Ironmaking and steelmaking; 1993. Vol.20. 445-452p
5]
Maehara Y, Ohmori Y: Materials Science and Engineering; 1984. Vol.62. 109-119p
6]
Crowther DN, Mohamed Z: Metallurgical transaction A; 1987. Vol.18. 1929-1939p
7]
QIAN G, Cheng G: ISIJ International; 2014. Vol.54.1611-1620p
8]
Ray AK, Bhor PK: Ironmaking & Steelmaking;2000; Vol.27. 189-193p.
9]
ITO Y, MURAI T: ISIJ International; 2011. Vol. 51. 1454–1460p.
10]
Suzuki HG, Nishimura S: Tetsu-to-Hagane; 1979. Vol. 65. 2038-2046p.
11]
Yamanaka K, Terasaki F: Tetsu-to-Hagane; 1979. Vol. 65. 1410-1417p
12]
Seah MP: Acta Metall; 1977. vol. 25. 345-357p.
13]
Jaffe LD, Buffum DC: JOM; 1957. 8p.
14]
Brandes EA, Smithells metals reference book (B). 6th ed. Lomdon: Butterworths; 1983.
15]
Hong CW, Heo YU: Mater. Charact; 2017. vol. 124. 192-205p.
16]
Yamaguchi M, Shiga M: Science; 2005. vol 307. 393~397p
17]
Heuer JK, Okamoto PR: Journal of Nuclear Materials; 2002. vol 301. 129-141p
18]
Rice JR: Materials Science and Engineering; 1989. vol 107. 23-40p
19]
Lejček P et. Al: Surface Science; 1992. vol. 269/270. 1147-1151p.
22
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica
The interfacial convection in fluxes in the continuous casting process P.R. Scheller, Y. Lin, Q. Shu
In this paper the generation of interfacial convection is investigated on the basis of the samples taken in the mould directly in the continuous casting process and quenched in water. Besides of the mass transfer as the result of the chemical reaction, the solidification structure near to the interface with liquid steel was investigated. The results show near to the interface the layered structure containing crystalline and partially glassy layers with the thickness of 50 to 150 μm. They occur as well under the conditions when thermal convection in the flux layer exists as under the conditions when thermal convection does not exist. In each kind of layer a different frequency of crystallisation nuclei were recorded. The reduced iron and chromium oxides produces small metal droplets with <1 μm to 2 μm in diameter which act as nuclei for the crystallisation of slag. These oxides are products of the chemical reactions at the flux/steel interface and are transported by the convection into the bulk of flux layer. Their attendance and distribution in partially glassy and crystalline layers contribute to the different solidification structures besides of the thermal layering. The sources for differences of temperature are rapid local heating at the flux/steel interface and cooling at the place of sintered layer fusion in the casting process. In the quenched samples it is a strong cooling at the slag/metal interface. All of these results indicate strongly the existence of the interfacial convection.
KEYWORDS: INTERFACIAL CONVECTION - MASS TRANSPORT - THIN LAYER - CASTING FLUX - FLUX SOLIDIFICATION
INTRODUCTION In the metallurgical process technique the understanding of transport phenomena is a key knowledge besides of physical chemistry of reactions. As the reactions at interfaces are usually rapid the transport from the interface to the bulk of the reacting phases determines the total reaction rate. Looking to the metal-slag reactions the transport of reactants in the given slag phase limits the kinetics. In e.g. the continuous casting the thin liquid flux layer at the meniscus reacts with liquid steel before it infiltrate the gap between the steel shell and the mould wall. Transport phenomena in the slag phase as thermal and interfacial convection and diffusion determine the reaction rate between both phases. The experimental results presented in this paper focus on the detection of the occurrence of interfacial convection. Furthermore its contribution to the total mass transfer is analysed for the industrial continuous casting process. Convection flow in the fluid phases have a decisive effect on mass and heat transport. In the case of reactions between the metal and the slag, the flows and the mixing in the slag phase determine the kinetics of the reactions because of higher viscosity (1). Here a distinction can be drawn between large area, thermal convective flows taking place under specific conditions, and convective flows in the vicinity of the interface occurring, for example, finally as a result of chemical reactions at the phase boundary (2). Gradients of interfacial energy along the interface produce interfacial convection which increases the mass transport within La Metallurgia Italiana - n. 7/8 2018
and through the boundary layers. These gradients can be generated by the gradients of temperature, concentration, electrical potential which in turn are usually caused by local differences of species transfer through the interface and their diffusion. In slag-metal systems such convection reaches high intensity (3-5) because of high values of interfacial energies. In the most of systems the interfacial energy decreases with increasing mass flow rate through the interface of diffusing components. The instability of the interface between two undisturbed and still fluids is therefore depen-
Piotr R. Scheller Beijing University of Science and Technology, China Technical University Bergakademie Freiberg, Germany
Yong Lin, Qifeng Shu Beijing University of Science and Technology, China
23
Science and technology in steelmaking dent on the quote of the diffusion coefficients in both phases of the passing through component and on the quote of the kinematical viscosities of both phases. Therefore the fluid dynamic controlled instability occurs if e.g. is D1/D2 =1 and ν1/ν2 > 1 (6). These conditions are still fulfilled for liquid slag/metal systems (7). The induced flow generates in this case microwaves at the interface which is termed the Kelvin-Helmholtz instability and was reported in few papers (2, 8-9). ANALYSIS OF THE PROBLEM AND ITS BASIC DESCRIPTION Convection Flows For the problem of interest here relates to how local disruption of the interfacial tension affects the flows in a slag layer on top
of the liquid metal. The details of this analysis are described in the previous papers (2, 7, 10). The movement of a briefly accelerated volume element or fluid layer is determined by a given interfacial tension force Fσ (corresponding to Δσ) acting as a result of the disturbance in a local area on the interface, the inertia force Fρ and the viscous force Fη. As the velocity in metallurgical systems cannot normally be measured in experiments at such short distances (movement at the boundary layer), only the variables describing the acting forces and the characteristic length can be considered for a description of the problem (variables Lc, σ, ρ, η). An analysis of the problem with the aid of the similarity theory gives the following expression to describe fluid movement in the boundary layer area:
(1) where fjfjfjffkf is denoted the Steinmetz number Ste as derived in the previous papers (2,11). An identical expression is obtained by division of the known numbers Re and We, Re²/We. By linking the acting forces in the form of the above-mentioned value (equation 1), the overall relationship for convection flows is maintained without flow velocity being a factor. For an identical disruption of the interfacial energy (τmax = Δσ/ Δx = constant) increasing layer thickness will give a constantly strong impulse at an even greater distance from the disruption. When viewed perpendicularly to the phase boundary, layers further away from the interface are also moved due to friction and the impulse decreases with the distance from the interface. As stated previously, this gradient is the greater, the thinner the layer. For the respective fluid, however, grad τ decreases with increasing Ste (because Ste ~ Lc or Δy) and in parallel greater fluid volume near the interface is moved. It is therefore logical to use the Steinmetz number Ste to characterize the flow conditions in thin fluid layers caused by gradients of
interfacial tension. Mass transport Three dimension free numbers as Sh, St and Bo, Sherwood, Stanton and Bodenstein respectively, are usually used to describe the mass transfer. With respect to the analysis of the mass transport description discussed in the introduction, the Bo-number is suitable for the description of the problem. For improved evaluation of the measured values the modified Bodenstein number Bo* is new defined, as the mass flow caused by diffusion cannot be separated from mass flow caused by convection in experimental investigations. It defines the quote of the total (measured) mass flow rate density of analysed species which results from the transferred mass through the interface in the reaction time to the mass flow rate density caused by diffusion
(2) If the upper fluid (e.g. slag) is assumed as completely still without any convection so the species i will be transported in this fluid only by the diffusion. If the mass transport is controlled only by diffusion, than is Bo*=1. If the interfacial reaction controls the total rate, than is the Bo* ≤ 1. Otherwise, if the reaction products or transferred species 24
are transported away by convection (e.g. thermal or interfacial convection), than is the Bo* > 1. If interfacial convection occurs they are squeezed from the diffusion layer to the bulk followed by “fresh” liquid. The mass transfer through the interface causes the local change in the interfacial energy which itself generate the movement La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica at or near the interface. Following, the relationship between the interfacial convection and mass transport have been pro-
posed in the previous paper (2):
(3) Experimental The industrial casts were performed at ThyssenKrupp Nirosta in Bochum works. The cast steels were Cr-Ni stainless steels AISI 304, 321 and 316Ti with Ti contents of 0.01, 0.3 and 0.3 mass % respectively. The temperature of liquid steel in the mould was approx. 1460°C. Different mould fluxes were used for casting. The composition of the most frequently used flux is listed in Table 1. During industrial continuous casting samples
were taken from the mould as described previously (12). At this sampling method the whole vertical square section including liquid steel, liquid slag, sintered and powder layer is taken in the cc mould in the steel container with 50 mm diameter and quenched immediately in water. After cutting, polishing and preparation of the sample the square section was investigated using LM, SEM and EDS.
Tab. 1 – Chemical composition of casting powder before use. Carbon and moisture losses are already taken into account. In delivery state: C (free) 2.2-2.9%, CO2 5.5-6.8%. FeO
MnO
P
SiO2
Al2O3
TiO2
CaO
MgO
F
Na2O
K2O
Li2O
1.27
0.04
0.04
33.5
7.05
0.14
38.6
0.81
8.40
8.23
0.50
1.08
Results and discussion The main components of the investigated fluxes are CaO and SiO2, which make up together approx. 60% of the mass. The relevant physical properties of fluxes for continuous casting such as viscosity and surface tension are adjusted by adding CaF2, Na2O, K2O, Li2O and FeO, MnO and MgO. Even small additions and changes in composition can have a considerable effect on these properties. The composition of liquid flux changes during casting because of the reactions with steel at the interface. The change in the chemical composition of the slag from 13 samples taken during casting of AISI 304 is plotted as a bar chart in Figure 1. Greatest changes show SiO2 (decreasing) and TiO2 (increasing). Oxygen is supplied to the liquid metal mainly from (FeO)
and (SiO2) in the flux, with [Ti], [Mn] and [Cr] in the steel melt being oxidised. Designation ( ) and [ ] indicates the slag-phase or metal-phase respectively. The greatest exchange takes place between (SiO2) and [Ti]. Thermal convective flow, as well as phase boundary convection in the slag layer decisively influences the mass transport and therefore the distribution coefficient e.g. Δ(TiO2)/[Ti] during the reaction time (2, 12, 13). The change of the chemical composition of the mould sample indicated in Figure 1 is representative for slag layer height greater than 6 mm where both kind of convective flows are in action (2, 12, 13). The results described above show that intense mass exchange between slag and liquid metal takes place.
Fig. 1 – Change in the chemical composition of slag during casting of AISI 304 compared to virgin casting powder (C and moisture losses are taken into account). La Metallurgia Italiana - n. 7/8 2018
25
Science and technology in steelmaking In Figure 2 the square section over the whole quenched liquid layer, from liquid metal to sintered layer, is presented. Different solidification structures show clearly the pattern caused by thermal flow. It is generated because the interface to liquid steel is approx. 300 K hotter compared to the upper part with
contact to the sintered layer as reported previously (2). The patterns are visible because of the different solidification structure which is partially glassy and partially crystalline. The length scale of these patterns lies in the range of millimetres.
Fig. 2 – Square section over the slag sample taken from the CC mould and quenched in water. Since [Ti] was present in steel and the change in the (TiO2) concentration was the highest of all components relative to the content in steel, the effect of the convective flows on the kinetics of the metal-slag reaction was examined on the basis of [Ti] oxidation and absorption into the slag. The effect of interfacial convective flows on the kinetics of the mass transfer with liquid metal was investigated via the ratio Δ(TiO2) / [Ti] (where Δ represents the difference in the chemical composition of sampled slag and the original powder) as a function of the slag layer thickness (2, 10). The ratio increases continuously with slag layer thickness. The relationship between the Bo* and the convection conditions in the slag, as defined by the Steinmetz number Ste containing slag layer thickness as Lc, are examined below on the example of casting steel AISI 304. The experimental results eva-
luated in this way are shown in Figure 3. It can be clearly seen that Bo* increases with increasing Ste. Even with the smallest measured slag layer thickness, Bo* is greater than 1 (i.e. the total mass transport is greater than the diffusive mass transport). It can be reliably assumed that under the experimental conditions, the interfacial flows always occur as soon as liquid slag is formed. They make a continuous contribution to the total mass transport up to approx. 20 times the mass flux by diffusion (see values of Bo* in Fig. 3). For values of Ste higher than ca. 5·103 (marked with thick stroke), thermal convection occurs and the relation between Bo* and convective flows has to be described using another function as reported previously (2). In this case the flow conditions are described by the product Ste•Ra, which takes into account as well the interfacial convection as the thermal one.
Fig. 3 – Modified Bodenstein-number Bo* for TiO2-absoption into the casting flux as a function of Steinmetznumber Ste (valid in the range of Ste up to 5 .103) for cast steel grade a) AISI 304.
26
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica Further investigations of the slag solidification structure show specific structure near to the interface with metal. The overview is shown in Figure 4. Alternating crystalline and partially glassy layers arranged parallel to the interface with metal are clearly visible. Glassy structure in this sense means not real glass, as can be detected by XRD, but structure with
much less crystals than the other one. The thickness of the layers is 50 to 80 μm, in other samples up to approx. 150 μm. The slag layer thickness on the meniscus of liquid steel was in this case 5 mm. At these conditions no any thermal convection can occur (2).
Fig. 4 – Solidification structure near to the metal interface (on the right side). Detailed inside view to the both kinds of layer are presented in Figure 5 a) and b). In the crystalline structure a big number of metallic droplets exists a) and in the partially glassy structure much less b). During casting Fe, Cr, Mn and Ti in steel are oxidised at the interface and introduced into slag, compare Figure 1. During cooling down of samples (similar situation exists in the gap between strand shell and the mould wall) iron oxide and chromium oxide are reduced by residual carbon (11) producing metal droplets with <1 μm to 2 μm in diameter. They are composed of iron with 2 to 7 mass % chromium. As the reduction process runs before the slag solidifies they act as nuclei for the slag crystallisation. In nearly each crystal a
metallic droplet can be detected in the centre. The quenching of the samples was performed in this way that the steel container with slag and liquid steel was immersed into water with the bottom side. As the steel has higher thermal conductivity compared to slag, the heat was mainly removed downwards through the liquid and then solidified steel. Following, the first slag layer adhering to steel cooled with higher rate than the layer above it. This could be the reason for partially glassy solidification. Layer in bigger distance from the interface have higher temperature and during cooling more crystals can be produced, as shown in CCT diagrams
Fig. 5 – Solidification structure in the a) crystalline and b) partially glassy layer; magnification from Fig. 4. White points are metallic droplets. La Metallurgia Italiana - n. 7/8 2018
27
Science and technology in steelmaking e.g. in previous investigations (14, 15). The occurrence of layers and their order is difficult to explain if we assume the static state in liquid slag. In Figure 2 was shown that thermal convection in any case exists (even during cooling of the sample) if the slag layer is thicker than 6 mm (2). In the sample shown in Figure 4 with 5mm slag layer thickness thermal convection do not exists. Therefore it is evident that the movement of liquid slag producing small scale vorticities and layering must be caused by interfacial convection. Such phenomena were previously investigated in physical models (16). The different solidification structure is finally the result of local thermal history and probably small differences in the content of at the interface oxidised iron and chromium. Interfacial convection occurs when gradients of interfacial tension along the interface exist. On the basis of oxygen transfer through the interface investigated in similar samples it was previously concluded that the interfacial tension should be close to zero (17). Other investigations of interfacial tension between liquid slag or flux and steels using the drop weight method show values between 0.6 to 1 Nm-1 (18). However, the values measured with drop weight method correspond to nearly equilibrium state which cannot be assumed in the real continuous casting process. In the last case high mass transfer rates exist because of the specific flux composition, its low viscosity and short contact time between phases. Therefore much lower interfacial tension can be expected. On the other hand the precipitated metal droplets in the slag and clear interface between slag and metal, even if sometimes perturbed, indicate that interfacial tension should have in average a certain value higher than zero. It is not homogeneously distributed at the
interface because of intense mass exchange and because of convection flows with connected renewal of the interface. All these factors lead to local gradients of interfacial tension producing interfacial turbulences.
LIST OF SYMBOLS Bo Bodenstein-number Bo* modified Bodenstein-number D diffusion Fη viscous force Fρ inertia force Fσ surface tension force Lc Characteristic length mass flux density Ra Reyleigh-number Re Reynolds-number Sh Sherwood-number Ste Steinmetz-number We Weber-number
η dynamic viscosity ν kinematic viscosity ρ density σ interfacial tension τ shear stress [] metal phase () slag phase
28
Conclusions Samples were taken in the mould during the industrial continuous casting process and quenched in water. The sampling technique enables the investigation of the whole slag layer between the contact to liquid steel and the sintered layer. The investigation of slag samples was focussed to the area near to the interface with liquid steel. The results can be summarised as follows: 1. Near to the interface with metal a layered structure in the solidified slag was found. In alternating order the layer consist of crystalline and partially glassy structure. The thickness of these layers lies between 50 and 150 μm. 2. In the centre of nearly all crystals metal droplets with <1 μm to 2 μm in diameter exist. They are produced by reduction of iron and chromium oxide during cooling of samples and act as nuclei for slag crystallisation. 3. The order of layers and their length scale indicate strongly the existence of interfacial convection in the liquid state. ACKNOWLEDGEMENT The ThyssenKrupp Nirosta are gratefully acknowledged for performing the trials and sampling and the University of Science and Technology Beijing for providing of Research Grants.
Indices D diffusion Σ total i substance ir
La Metallurgia Italiana - n. 7/8 2018
Scienza e tecnologia nella produzione siderurgica REFERENCES 1]
E. Steinmetz: Arch. EisenhĂźttenwes. 43 (1972), 151
2]
P.R. Scheller: Ironmaking & Steelmaking 29 (2002) 2,154
3]
T. Takasu; J.M.Toguri: Philosophical Trans. Royal Soc. A: Mathematical, Physical and Engineering Sciences 356 (1998) No 1739, 967
4]
M.A. Rhamdhani; K.S. Coley; G.A. Brooks: Proc. 43rd Annual Conf. of Metallurgists of CIM 2004, Hamilton, CA, 203
5]
M.A. Rhamdhani; K.S. Coley; G.A. Brooks: Met. Mat. Trans. B 36B (2005) 5, 591
6]
C.V. Sterling; L.E. Scriven: AICHE J. 5 (1959), 514
7]
P.R. Scheller: Proc. SANO Symposium, University of Tokyo, pp 84-95, Tokyo Oct. 1-3, 2008
8]
P.R. Scheller: Habilitation Thesis: Flow conditions and mass exchange in thin liquid layer with special view to continuous casting fluxes, Aachen University RWTH, May 1998
9]
Y. Chung; A. Cramb: Met. Mat. Trans. B 31B (2000) 10, 957
10]
P.R. Scheller: steel research int., 76 (2005), 581
11]
P.R. Scheller: High Temp. Materials and Processes 22 (2003) 5-6, 387
12]
Scheller, P. R. Proc. 3rd Europ. Conf. Cont. Casting, Madrid, Spain, Oct. 20-23. 1998. Vol.2. pp. 797-806
13]
P.R. Scheller, steel research 72 (2001) 3, pp 76-80
14]
J.L. Klug; R. Hagemann; N.C. Heck; A.C.F. Vilela; H.P. Heller and P.R. Scheller: steel res. int., 84 (2013) 4, pp. 344-351
15]
Q. Shu; Z. Wang; J.L. Klug; K. Chou and P.R. Scheller: steel res. int., 84 (2013) 11, pp. 1138-1145
16]
P.R. Scheller: Sano Symposium Proc., The University of Tokyo, Japan, Oct. 2-3, 2008, pp. 84-95
17]
P.R. Scheller: steel res. int., 81 (2010) 10, pp. 886-890
18]
R. Hagemenn; R. Schwarze; H.P. Heller and P.R. Scheller: Met.Mat.Trans. B, 44B, 2, (2013), pp. 80-90
La Metallurgia Italiana - n. 7/8 2018
29
Le manifestazioni AIM AIM meetings and events 2018
37° CONVEGNO NAZIONALE AIM Convegno - SEGR Bologna, 12-14 settembre EOSC 2018 - 8TH EUROPEAN OXYGEN STEELMAKING CONFERENCE Convegno Internazionale Taranto, 10-12 ottobre TRATTAMENTI TERMICI DEGLI ACCIAI PER STAMPI A CALDO E A FREDDO PER IL SETTORE AUTOMOTIVE GdS - Centro TTM Ivrea c/o Confindustria Canavese, 11 ottobre GLI ACCIAI INOSSIDABILI - 10a EDIZIONE Corso - SEGR Milano, 17-18-24-25 ottobre/7-8-14-15 novembre RISCHIO AZIENDALE E CONTRATTUALE GdS - Centri FOR e A
OTTIMIZZAZIONE DEI TRATTAMENTI TERMOCHIMICI E DEI PROCESSI MECCANICI NELL’INDUSTRIA MECCANICA GdS - Centro TTM Provaglio d’Iseo c/o GEFRAN, 8 novembre LA PRODUZIONE DI GETTI PER APPLICAZIONI STRUTTURALI. ASPETTI METALLURGICI E DI PROCESSO GdS - Centro P Travagliato (BS) c/o IDRA, 9 novembre RIVESTIMENTI - 1° modulo Rivestimenti PVD e CVD Corso modulare - Centro R Roma, 14-15 novembre FAILURE ANALYSIS Corso - Centro CCP 20-21-28-29 novembre UTENSILI DIAMANTATI GdS - Centro MP Vicenza, 22 novembre
24 ottobre LA SORVEGLIANZA SANITARIA ED EPIDEMIOLOGICA NEL SETTORE METALLURGICO TRA TUTELA DEL LAVORATORE E DEL DATORE DI LAVORO GdS - Centro AS Brescia, 25 ottobre
CLEAN TECH - 4TH EUROPEAN CONFERENCE ON CLEAN TEHNOLOGIES IN THE STEEL INDUSTRY Convegno Internazionale Bergamo, 28-29 novembre CREEP Corso - Centro ME Milano, 11-12 dicembre
Per ulteriori informazioni rivolgersi alla Segreteria AIM e-mail: info@aimnet.it oppure visitare il sito internet www.aimnet.it
30
La Metallurgia Italiana - n. 7/8 2018
Attualità industriale Bypassing Problems Related to Water Cooling: Case Study for Applying ILTEC in a 100-ton EAF edited by: Martina B. Hanel, Andreas Filzwieser, Rolf Degel With the ILTEC Technology, Mettop´s patented water-free cooling solutions, it is possible to overcome the disadvantages of water by using an alternative cooling medium, namely the ionic liquid IL-B2001. The main characteristics, which makes IL-B2001 so favourable, are the neglectable vapour pressure, the wide liquidus range and the not flammable, non-explosive and atoxic behaviour. Since it is well known that there are multiple reasons of creating damage caused by leaking of water cooled panels in EAF´s, it is of superior interest to eliminate water as the cooling medium in certain areas: within the upper shell there might happen hydration of the refractory material caused by small leakages in the upper side wall cooling panels leading to molten steel perforation or in the off-gas parts were there can occur cooler perforation due to corrosive off-gases. Another problem are prewear hot spots within the sidewall of the refractory lining leading to additional down time causing production losses which can be alleviated. Not to forget the priceless benefit of a safe workplace. Using the example of a 100 t EAF, the different possibilities of implementing ILTEC are pointed out, from the bottom and the upper shell to the roof and off-gas parts of the furnace. In order to show the saving potentials, the economic benefits of increased furnace availability by implementing improved cooling solutions is provided. In addition, the benefits of increasing safety, illustrated according to different damage scenarios will be highlighted and the impact of a change to ILTEC discussed. It is the well-engineered technology and the sophisticated process control that in the first place prevents damage based on immediate leak detection combined with the outstanding properties of IL-B2001 in case of failure or leakage that makes ILTEC creating new pathways towards safe and effective cooling. KEYWORDS: WATER-FREE - COOLING - IMPROVED SAFETY - INCREASED FURNACE AVAILABILITY NOVEL COOLING OPTIONS - ILTEC - EAF
Martina B. Hanel Mettop GmbH, Leoben Austria
Andreas Filzwieser Mettop GmbH/PolyMet Solutions GmbH Leoben, Austria
Rolf Degel SMS group GmbH, Düsseldorf, Germany
INTRODUCTION All different metallurgical furnaces and aggregates that deal with high temperatures have in common, that they need a powerful cooling system, since temperatures from 800 up to 2000 °C are required for the production of metals. Worldwide, industrial scale cooling systems predominantly operate with water. This is mainly due to the high thermal conducti-
31
vity and the convenient availability of water, making it broadly used and favourable. However, water also has several disadvantageous characteristics [1-3]. Due to its restriction to a maximum operation temperature of 60 °C and the risk of explosion - both due to volume expansion and possible hydrogen explosion - the cooling medium water can even cause severe dangers.
La Metallurgia Italiana - n. 7/8 2018
Industry news On taking a closer look at the safety issues it has to be noted, that due to water damages - especially in the field of highly stressed areas – several fatal accidents occur every year worldwide [4,5]. This was the main motivation to rethink the existing cooling systems in metallurgy and to finally develop the innovative “cooling with ionic liquids” concept. This water free cooling concept has now been approved in industry for several years and hence the possible application and resulting advantages with an EAF are discussed here. Since it is well known that there are multiple reasons for damage caused by leaking of water within the different parts of the EAF, it is of superior interest to eliminate water as the cooling medium: within the upper shell, hydration of the refractory material, caused by small leakages in the side wall cooling panels that lead to molten steel perforation, can occur. In off-gas parts, cooler perforation due to corrosive off-gases can take place. Hot spots within the sidewall of the refractory lining in the bottom vessel, leading to additional down-time causing production losses, are a problem that can be solved. Not to forget the priceless benefit of a safe workplace. Water-free cooling - characteristics of the iltec technology With the novel ionic liquid cooling technology, virtually all negative effects of a water cooled system are eliminated and even additional benefits can be provided. By definition, ionic liquids (IL) are salts with a liquidus temperature below 100 °C. They have no noticeable vapour pressure below their thermal decomposition point and – depending on the actual composition – there is just a minor or absolutely no reaction with liquid melt or slag. Furthermore, the temperature range for cooling a system is much wider than water. An additional benefit resulting from this relatively high cooling medium temperature is the fact that hydration and corrosion problems are avoided.
The ILTEC Technology is characterised by the following: • Instead of water the ionic liquid IL-B2001 is being used as cooling medium • IL-B2001 is liquid at room temperature and can be used at a maximum operating temperature of up to 250 °C (on a short term basis, 200 °C on long term) • In case of a leak in the cooling system, the IL-B2001 will disintegrate into its components without a sudden increase in volume and without the formation of hydrogen. There will be no explosion when getting in contact with liquid metal, and so work safety can be guaranteed • No cooler corrosion problems will occur, as the IL-B2001 can be used at higher temperatures (above the dew point of the exhaust gases) • Due to the higher temperature (up to 250 °C), the dissipated heat can be recovered. This advantage will play a particularly important role in the future, not only in case of local legal requirements • Novel approaches to cooling beneath bath levels or at highly stressed areas because of the lack of explosive reactions and damage in case of leakage Overall it is of utmost importance for industrial applications, that the requirements regarding critical heat flux, thermal limits and impacts of health, safety and environment are being fulfilled. Research activities, and even more commercial operations, have proven that the ionic liquid IL-B2001 and the ILTEC Technology can lead to an industrial change regarding safety standards by becoming the new best available technology (BAT) [2,6,7]: When substituting water by the ionic liquid damages will decrease dramatically. Due to the lack of explosions the economic damage, the environmental damage and finally the personal damage will be substantially decreased and the area of acceptable risk (compare the two graphs in Figure 1) will move towards lowering the acceptability of consequences.
Fig. 1 - 2-dimensional risk analysis of a state of the art water cooled area (left) and using the ionic liquid as cooling medium (right) [8] La Metallurgia Italiana - n. 7/8 2018
32
Attualità industriale IL-B2001 – the ionic liquid at a glace Basically, ionic liquids are salts, meaning that they consist solely of anions and cations. Caused by their poorly coordinated ions, many of them are liquid even at room tempera-
ture [9-11]. After years of intensive research, the most optimised ionic liquid for the use as a coolant was developed. The main properties of this special and unique ionic liquid, with the trade name IL-B2001, is given in the table below.
Tab. 1 - Characteristic properties of IL-B2001] Symbol
Value
Unit
Range
Operation temperature
50-200
[°C]
ΔT = 150 °C
Short term stability
250
[°C]
Decomposition temperature
450
[°C]
Melting range
< 15
[°C] [kg/dm3]
Density
ρ
1.25 – 1.14
Specific heat capacity
cp
1.38 – 1.70
[J/gK]
50 – 200 °C
Dynamic viscosity
η
20 – 5
[mPa·s]
50 – 200 °C
Electrical conductivity
κ
30 – 130
[mS/cm]
The positive properties, that make IL-B2001 perfectly suitable as a cooling agent, can be emphasised as follows: • Broad operation temperature range • Non-explosive, non-flammable • High electrical conductivity • Relatively low viscosity • Non-corrosive due to chlorine free production procedure • Non-toxic, not harmful • Sufficient heat removal • No altering, non-consumable Iltec technology - hardware and features: A fundamental characteristic of ILTEC is the application of a closed circuit loop for the ionic liquid – the primary cooling circuit. The prevention of any contact between the ionic liquid and water and/or air makes IL-B2001 a not consumable good. Experiences over the last years have proven, that IL-B2001 does not change its original properties. The basic design and the required equipment of the ILTEC cooling technology remain more or less the same for all application. Specific details however, especially the supply capacity of the cooling medium and the dimensioning of the component parts, are tailor-made to meet individual customer-specific demands.
33
50 – 200 °C
An exemplary design of an ILTEC facility is given in Figure 2 and the main components can be summarised: • Storage tank to hold the entire amount of IL-B2001, the freeboard volume above the liquid level is purged with nitrogen in order to prevent absorption of water through moisture in the air • Two identical pumps (one for redundancy) guarantee the flow of the IL through the entire pipe system • Two heat exchangers to remove the heat to the secondary cooling circuit; one in operation, one for redundancy • Numerous measuring devices (digital as well as analogue) throughout the entire system to measure temperature, flow, pressure, differential pressure and level of liquid in the tank • Variety of valves, adjusting wheels and shut-off devices for all different operation modes • Depending on the application, a distribution unit close to the cooling application might be part of the system By using this basic hardware, water free cooling of all different areas within the electric arc furnace can be realised. However, the following numbers for the different layouts and dimensioning of the ILTEC facilities are only estimates based on different customers´ data. For each specific application, a detailed feasibility study will be conducted to determine the most optimised solution.
La Metallurgia Italiana - n. 7/8 2018
Industry news
Fig. 2 - Basic design of an ILTEC system (left) and photo of the optical appearance of the ionic liquid IL-B2001 Application within a 100T EAF - upper shell cooling to increase safety When thinking of cooling the shell of an EAF, the main design the side walls are welded steel pipes, only few furnaces have casted copper cooled parts. However, in both cases, the cooling medium water can be substituted by the ionic liquid IL-B2001 and the multiple benefits can be highlighted: • Increased safety because of a water-free furnace area
• • • •
Decreased damage in case of leakage due to the l ack of volume expansion and explosion Fast leak detection in case of individual supply and supervision of each panel Avoidance of hydration of refractory of the botom vessel in case of leakage Easy, fast and safe repair work in case of breakage
Fig. 3 - Different approaches for ILTEC Technology in the upper shell of an EAF: cooling of all panels with one main supply line and a ring distributor (Option 1 - left side) and individual supply of each single cooling panel (Option 2 - right)
La Metallurgia Italiana - n. 7/8 2018
34
Attualità industriale Considering a 100 t EAF, there are different approaches to realise ILTEC cooling for the entire upper shell. In Figure 3 the flow sheets of two different possible solutions are given and discussed. One option is (left flow sheet) a simple substitution of the cooling medium within the existing piping system can be conducted, in this case one single supply line pumps the IL-B2001 from the storage tank to the ring distribution unit at the furnace. In this way the supply of the individual cooling panels is realised. Having passed the panels, the cooling medium is recollected and transported to the heat exchanger unit in one single line. The other option, a technically more sophisticated approach of supplying each cooling panel (or unit which might be a
serial alignment of panels) with an individual line, is shown in the right flow sheet. The transportation of the cooling medium near the furnace is again implemented by a large supply line, but in this case the distribution unit divides the total flow in individual supply lines with smaller diameter. Each single supply line is equipped with flow, temperature and pressure measuring devices to supervise the operating condition. For both examples, the cooling performance of the entire facility (for average values) will be calculated based on the precondition as given by the customer. For one specific case the dimensions and the capacity of these two different options can be summarised as given in Table 2.
Tab. 2 - Basic numbers of different layouts for cooling the upper shell of the EAF Option 1
Option 2
Unit
Supply lines cooling panels
1
18
---
Total Flow rate IL-B2001
300
300
m3/h
Temp. difference average (peak)
15 (30)
15 (30)
°C
Cooling load average (peak)
2500 (5000)
2500 (5000)
kW
Inlet temperature IL-B2001
50
50
°C
Outlet temperature IL-B2001
65 (80)
65 (80)
°C
Amount of IL-B2001
6500
8700
kg
The mayor difference of these possible options is the more sophisticated supervision possibilities of individual supply lines vs. complexity of the system and amount of IL-B2001. Therefore, an optimised mixture (for example 5 supply lines for a quarterly merging of the furnace circumference) can be regarded, always depending on the customer’s needs. Application within a 100T EAF - bottom vessel cooling for an increased lifetime With the possibility to provide a cooling solution with more than sufficient heat removal AND a perfectly safe operation mode, a totally new approach of cooling the bottom vessel can be realised. Since there is no danger in case of leakage, cooling the refractory beneath bath level can contribute to a tremendous increase in lifetime and lead to decreased production costs and increase of furnace availability. The benefits of cooling beneath bath level can be summarised: • Intensified cooling – steep temperature gradient, less infiltration zone • Less wear – increased lifetime of refractory 35
• Decreased down time – more productivity • Decreased repair work/gunning – lower production costs • Increased safety for employees and systems • Increased inner furnace diameter by decreasing refractory thickness The installation of cooling beneath the bath level can technically be realised in different ways. There is a simple implementation of a cooled copper plate (comparable to staves) within the brick lining, for example by the substitution of the insulation layer or a decrease in thickness of the brick lining, as given on the left side of Figure 4. As a result, for the best possible cooling performance, the coolers themselves should in any case be part of the concept. Beside a conventional copper cooling plate in-between the steel shell and the brick lining, a solution with high intensity coolers can further improve the refractory resistance and increase the lifetime of the vessel. In Figure 4 two different installation scenarios are given. The so-called CFM cooling elements (Composite Furnace Modules, which consist of a combination of a casted copper cooler and the refractory La Metallurgia Italiana - n. 7/8 2018
Industry news mass casted upon it) are installed to more or less replace the brick lining. Due to the intensified cooling performance, the formed accretion layer produced by the intense heat removal, leads to a non-wear steady-state condition [12-14]. For the subsequent estimation with regards to the capacity and dimensions of the ILTEC Technology for cooling the bottom vessel, two different cases are shown, Option 1 for copper plates and Option 2 for CFM cooling elements. On considering copper plate coolers, the heat removal might be
lower, caused by the isolating brick lining in front of the cooler. As peak values always the double portion of the average value is considered. However, the CFM cooling elements are capable of up to 400 kW/m2 cooling performance as peak level, because of the design and improved contact between copper cooling fingers and the refractory [12,15,16]. Therefore the facility´s layout is designed for 200 kW/m2 heat removal in average.
Fig. 4 - Different possibilities for installation within the lower part of the EAF, left: installation of a plate cooling element between steel shell and brick lining, right: example of a high intensity cooler (CFM) combined with a copper plate cooler at the lower part
Compared to the case of the upper shell, always two panels are connected (serial array). Based on the assumption of average heat removal values, the capacity of the facility is given as summarised in Table 3. There are differences in capacity for the entire pumping and
piping system (50 and 100 m3/h IL-B2001) but also regarding the maximum cooling performance of the entire lower furnace part at full load (at simultaneous use of both heat exchangers) of 1600 and 6400 kW.
Fig. 5 - ILTEC Technology for cooling the bottom vessel with 5 independent supply lines, Option 1 (left) with copper cooling plates and Option 2 (right) with high intensity CFM coolers
La Metallurgia Italiana - n. 7/8 2018
36
Attualità industriale Due to the higher capacity, the piping dimensions for the high intensity CFM coolers have to be larger, also due to the larger amount of IL-B2001 needed. Still this is only a theo-
retical approach for this comparison and has to be adapted for each specific case after a detailed study and analysis of the actual state-of-the-art situation on site.
Tab. 3 - Basic numbers for cooling the bottom vessel of the EAF
Option 1 (Plate cooler)
Option 2 (CFM cooler)
Unit
Supply lines cooling panels
5
5
---
Average (peak) heat removal
50 (100)
200 (400)
kW/m2
Total Flow rate IL-B2001
50
100
m3/h
Temp. difference average (peak)
30 (60)
60 (120)
°C
Cooling load average (peak)
800 (1600)
2300 (6400)
kW
Inlet temperature IL-B2001
50
50
°C
Outlet temperature IL-B2001
80 (110)
120 (170)
°C
Amount of IL-B2001
2800
4000
kg
With the novel approach of cooling beneath bath level, a notable increase in lifetime of the furnace is achieved, that leads to a reduction of production costs at different stages: less down time for repair and exchange of refractory and decreased costs for refractory material due to less gunning and less refractory wear. Intensified cooling at highly stressed areas (for example near the burner/injector or near an eccentric tap hole) can lead to a debottlenecking of those areas. Furthermore, with a better cooling performance a decrease of the overall refractory thickness can be realised leading to an increase in hearth diameter for even more production Application within a 100T EAF - off-gas duct for heat recovery One immense benefit of using a cooling medium at a temperature of up to 200 °C, is the possibility to increase the operation temperature within the refractory lining to some extent. This can eliminate problems arising from hydration or corrosion due to surface temperatures below the dew point (e.g. sulfuric acid formation at presence of sulphurcontaining gases). Even more, when considering the huge amount of energy and heat that is lost within the off-gas, the higher temperatures of 200 °C can be directly used for
37
heat recovery. Benefits of exchanging water in the off-gas duct: • Heat recovery due to outlet temperature of up to 200 °C • Bypassing corrosion problems in cold spots • Increased safety for employees and systems (no water in the furnace in case of leakage) In Figure 6 the flow sheet and in Table 4 the dimensions of the facility are given, based on a heat load peak value of 200 kW/m2. For this system, as shown in the figure, a simple replacement of water as the cooling agent has been taken into account. It can be calculated that at 100 kW/m2 the overall heat load is 2000 kW. To optimise the heat recovery a frequency controlled pump can be used and the flow rate and therefore the cooling effect can be controlled by the outlet temperature of the ionic liquid. This value can be set at 200 °C to guarantee the most efficient heat recovery. In addition, a combined cooling of off-gas duct and roof part can be taken into account.
La Metallurgia Italiana - n. 7/8 2018
Industry news Unit
Supply line
1
---
Total Flow rate IL-B2001
Variable
m3/h
Temp. difference average (peak)
50 (100)
째C
Cooling load average (peak)
2000 (4000)
Kw
Inlet temperature IL-B2001
150 (100)
째C
Outlet temperature IL-B2001
200
째C
Amount of IL-B2001
2800
kg
Fig. 6 - Dimensions and basic numbers of the ILTEC facility for cooling the off-gas duct Summary and outlook The current paper describes the opportunity to increase furnace safety and operation efficiency by implementing an industrially-proven alternative cooling method. In addition, it demonstrates the substantial economic improvement by highlighting the increased lifetime of refractory lining and the potential of heat recovery. The given applications for the EAF demonstrate the many possibilities to employ the new cooling technology as an alternative to water in existing cooling circuits for different reasons. The replacement of water in the upper shell is more or less initiated by safety considerations. The approach of implementing cooling within in the side walls beneath bath level, combined with a new design of coolers, is driven by
the increase in refractory lifetime and an increase in furnace availability. However, this technology is not only limited to the given examples, but can also be used in multiple other applications where water cooling is not possible due to safety reasons, making it best available technology (BAT) in the future. An important future trend in terms of a cost-effective and environmentally friendly operating mode inevitably leads to a better utilization of waste heat. The higher outlet temperature and the higher temperature difference Total Flow rate between inlet and outlet make IL-B2001 the perfect cooling medium for effective heat recovery. The temperature of up to 200 째C of the cooling medium can either be used for conversion into electricity or as a direct heating agent.
REFERENCES 1]
K. Verscheure, A.K. Kyllo, A. Filzwieser, B. Blanpain, P. Wollants, Furnace cooling Technology in Pyrometallurgical Processes, Sohn International Symposium: Advanced Processing of Metals and Materials, Proceedings Non-Ferrous Materials Extraction and Processing, Volume 4, 2006
2]
S. Wallner, A. Filzwieser and J. Kleicker, Some aspects for the use of water cooled furnace walls -Water the best refractory? Proceedings of the International Copper Conference, 2003
3]
N. Voermann et al., Furnace Cooling Design for Modern, High-Intensity Pyrometallurgical Processes, Proceedings of the 4th Int. Conf. Copper 99, Vol. V: Smelting Operations and Advances, 1999, p. 573
La Metallurgia Italiana - n. 7/8 2018
38
Attualità industriale 4]
H.J. Oterdoom, Furnace Explosions with a Focus on Water, COM 2014 – Conference of Metallurgists, Vancouver, Canada, 2014
5]
A. Filzwieser, S. Konetschnik, I. Filzwieser, S. Wallner and R. Preiss: Mettop´s new Cooling Technology is the safest way to cool a Furnace, Conference of Metallurgists, Vancouver, Canada (2014)
6]
A. Filzwieser, A. Siegmund, I. Filzwieser and M.B. Hanel, Using Mettop´s Ionic Liquid Cooling Technology – A Breakthrough Technology, Proceedings of the AISTech 2016, p. 2827
7]
M. B. Hanel, A. Filzwieser, I. Filzwieser, S. Wallner and S. Ruhs, ILTEC – Mettop´s Revolutionary AND Safe Cooling Solution, Proceedings of the Copper Conference, 2016
8]
R. Preiss, Methoden der Risikoanalyse in der Technik, TÜV Austria Akademie GmbH, 2009, ISBN-10 3-901942-09-2
9]
H. Joglekar, I. Rahman and B. Kulkarni, The Path ahead for ionic liquids, Chem. Eng. Technol. No. 7, 2007
10]
P. Wasserscheid and T. Welton, Ionic liquid in synthesis, Volume 1, Wiley-VCH Verlag GmbH & Co. KGaA., Weinheim, Germany, 2008
11]
IoLiTec: Wärmeträgermedien, Thermofluide, http://www.iolitec.de/ Warmespeicherung-Transport/waermetraegermedien-thermofluide.html, access on 20/03/2013
12]
Kyllo, A.K. and N.B. Gray: Composite Furnace Module Cooling Systems in the Electric Slag Cleaning Furnace, Proceedings of the EMC, 2005, p.1027
13]
A. Fallah-Mehrjardi, P.C. Hayes, S. Vervynckt, E. Jak, Investigation of Freeze-Linings in a Nonferrous Industrial Slag, Metallurgical and Materials Transactions B, Volume 45B, 2014, p. 850
14]
A. Fallah-Mehrjardi, P.C. Hayes, E. Jak, Understanding Slag Freeze Linings, JOM, Vol. 66, No. 9, 2014, p. 1654
15]
A.K. Kyllo, N.B. Gray, D. Papazoglou, and B.J. Elliot: Developing Composite Furnace Module Cooling Systems, JOM, (2000) p. 66
16]
L.R. Nelson et al: Application of a high-intensity cooling system to a DC-ARC Furnace production of Ferrocobalt at Chambishi, Proceedings: Tenth International Ferroalloys Congress, Cape Town, South Africa, 2004, p. 508
39
La Metallurgia Italiana - n. 7/8 2018
in evidenza
Corso
Failure analysis X Edizione
21-22-28-29 novembre 2018 Milano . Monza . Soncino Organizzato da
In collaborazione con SRL Centro Ricerca - Prove Materiali - Tarature
Il termine inglese, “failure analysis” indica, in generale, lo studio delle cause all’origine di uno scopo non raggiunto. In ambito tecnico il termine è strettamente legato al concetto di avaria, in campo meccanico al concetto di rottura. Trovare le cause che hanno portato ad una “failure” consente di individuare le responsabilità, siano esse collocabili in fase di progetto, di fabbricazione o di esercizio e di predisporre le adeguate misure correttive; la “failure analysis costituisce dunque, se correttamente e regolarmente utilizzata, un fattore essenziale nello sviluppo tecnologico. Come ormai sua tradizione il Corso, intende fornire ai partecipanti un quadro completo dei presupposti e degli strumenti su cui si basa l’intera disciplina della Failure analysis. Verranno così affrontati e descritti i vari possibili meccanismi di danno e la loro dipendenza dalle condizioni d’esercizio che poi sfociano in difetti che propagandosi portano a rottura, le tecniche d’indagine oggi disponibili, la strumentazione d’indagine usualmente impiagata. La presentazione degli argomenti sarà affidata a docenti con comprovata competenza ed esperienza specifica sugli argomenti trattati, esperienza derivante da anni di attività in ambito industriale e accademico. Coordinatore del Corso: Carlo Fossati
IL PROGRAMMA COMPLETO E TUTTE LE INFORMAZIONI SONO DISPONIBILI SUL SITO www.aimnet.it
#corso #formazione #failure #analysis #indagine #rottura #difetti #materiali
Attualità industriale Latest results in EAF optimization of scrap-based melting process: Q-MELT installation in Kroman Celik edited by: Marco Ansoldi, Damiano Patrizio, Manuele Piazza, Orhan Kuran Competitiveness in Electric Arc Furnace production is pushing steelmakers to increase the throughput at the lowest expenditure. The technological challenge is therefore to operate the existing facilities at their top capabilities and to exploit materials and energy sources at the upmost yield and efficiency. Moreover, seizing the opportunity offered from the price variability of metallic feedstock calls for flexible practices, versatile equipment and tools to rapidly tune the melting profiles without hindering steel quality and equipment lifetime. The present work analyzes the results achieved in Kroman Celik after the installation of Q-MELT Automatic EAF system and the revamping of the furnace chemical package. Q-MELT is conceived with the double target to maximize the productivity minimizing the operational cost and it is designed as a centralized control interacting with multiple technological packages. The main features of the system are described such as electrode regulation and foamy slag control, charging optimization, off-gas analysis and closed loop injectors control. Moreover, these modules provide the process supervisor application with important information regarding the process status, thus allowing the adoption of a unified control strategy. The supervisor implements a robust statistical approach to identify process deviations in real time. Process data are collected, clustered and filtered, extracting the average and deviation trends of the key process variables. Comparing the real-time to the expected trend, the system performs an adaptive process control and acts on specific actuators. The chemical package upgrade includes the M-ONETM sidewall injectors for gaseous fuel, efficient injection of oxygen and coal, modern post-combustion units dedicated to supply soft oxygen to optimize CO combustion and LimeJetTM units committed to dose slag formers (lime and dolomite). The installation of the LindarcTM real-time gas analyzer and of the new Q-REG electrode regulator complete the Q-MELT system and represent the core of the adaptive process control.
KEYWORDS: EAF - GAS ANALYSIS - POST-COMBUSTION - PROCESS CONTROL - YIELD - SIDEWALL INJECTION
Marco Ansoldi, Damiano Patrizio Danieli & C. Officine Meccaniche, Buttrio, Italy
Manuele Piazza Danieli Automation S.P.A., Buttrio, Italy
Orhan Kuran Kroman Celik San. A.Ş., Darıca/Kocaeli, Türkiye
La Metallurgia Italiana - n. 7/8 2018
41
Industry news INTRODUCTION In todayâ&#x20AC;&#x2122;s highly competitive market situation, it is of outmost importance for EAF steel makers to optimize their processes in order to reduce operating costs and to improve safety and reliability of the equipment. A common trend observed during the last five years is a generalized reduction in oxygen and fuel utilization, mostly driven by a worse quality charge mix and a lower productivity level requirement from saturated markets. The comprehensive study on the chemical energy efficiency in EAF practice reported in their paper by Patrizio and Pesamosca [1] is clarifying the theoretical background explaining the observed tendency. Whenever the charge mix is poor in carbon and other oxidizing elements (silicon, manganese, chromium) the overall energy efficiency of oxygen drops down and turns into expensive yield losses. Coal energy input is lower, compared to gaseous fuel burners. Intensive oxygen practices lead therefore to overall higher heat losses to the fumes treatment plant, incomplete CO combustion, iron oxidation and poor slag quality. The current market situation is mostly driven by price competition. Volumes have kept shrinking in the recent period,
especially in the steel commodities and in countries with mature economies. Operating expenditure is the most important performance steelmakers are taking care of. Existing steel production over capacity is not pushing for higher production levels, while high quality commercial bar feedstock is widely available at competitive prices, mainly imported from Far East. Scrap prices are volatile and highly variable on short period. Low bulk density feedstock and local market are cheaper and preferred to expensive imported or shredded feedstock. The question is therefore how to rapidly adapt to this highly dynamic market environment still holding competitive performances to keep proper profit margins. The solution calls for a highly automated EAF, equipped with efficient specialized equipment, advanced monitoring and sensing systems and adaptive integrated process control. In July 2016 Kroman Celik, located in Gebze (Turkey), a producer of reinforcing bars and wire rods, choose DANIELI QMelt process control and MORE chemical package to upgrade their 150 ton EAF (Fig. 1), targeting to improve the EAF productivity and to reduce operative expenditures.
Fig. 1 - Kroman Celik EAF
42
La Metallurgia Italiana - n. 7/8 2018
Attualità industriale KROMAN CELIK EAF REVAMPING The main electric arc furnace design features are reported in table 1. Tab. 1 - Kroman EAF design features Start up (year)
2010
Power supply type
AC
EAF Supplier
DANIELI
EAF Type
EBT
Tapped steel weight
150
t
Hot heel weight
25
t
Productivity
175
t/h
Annual Production
1.300.000
tpy
Inner panel diameter
7200
mm
Electrode diameter
7200
mm
Pitch circle diameter
1350
mm
Transformer nominal rating
140
MVA
Max secondary voltage
1350
V
Since its start-up in 2010 the furnace performances have been outstanding, with a very fast learning curve [2]. More recently, the progressive deterioration of the scrap bulk density and quality called for a reduction in oxygen use. Specific electric energy consumption increased and productivity dropped, mostly affected by the longer power-off time required by loading as many as 4 buckets of scrap. Due to the above items, it was decided to equip the meltshop with the Q-Melt Dynamic Heat Suite package, which collects all the melting information to perform automatic and dynamic adjustments during melting and, specifically, to perform the metallurgical control of the electric arc furnace process. The system integrates Q-Reg, the latest release of the dynamic electrode regulation with foamy slag control, managing coal and lime injection to keep the arcs shielded and balanced at the highest power input, although not entailing excessive radiation losses. Real-time process optimization and control is based on LindarcTM, a innovative fast gas analyzer performing insitu laser spectrometry and allowing the closed loop control of postcombustion [3]. The chemical package was revamped with the most recent
La Metallurgia Italiana - n. 7/8 2018
MORE’s technologies for technological gases and solids’ injection management [4]: M-ONE sidewall injectors to improve oxygen and coal efficiency and Limejets to integrate also lime and dolomite injection, to optimize the slag foaming practice. ADAPTIVE PROCESS CONTROL: Q-MELT The well-known inherent high variability and low measurab lity of the EAF process call for a set of responsive yet very robust control strategies. Q-MELT process supervisor (Melt Model) was designed and developed to address these aspects by means of a data-driven approach to electric arc furnace control. It implements a statistical approach to identify process deviations in real time. Process data are automatically collected, clustered and filtered, and the average and standard deviation trends of the key process variables are extracted. This set of statistical figures is called fingerprint and gives a picture of the expected and normal process behavior. By means of the comparison between the real-time and the expected trend, the system performs an adaptive process control and acts on specific actuators (Fig. 2).
43
Industry news
Fig. 2 - Melt Model implementation of a trend-based adaptive process control
Melt Model applies this very general approach to control the decarburization process. At the start of the heat, the fingerprint of the off-gas composition (%CO, %CO2, %H2O) and of other controlled variables (injectors O2 flow, dispensers C rate and others) is retrieved from the historical data base. The extraction is done considering a proper set of filtering criteria (practice, charge materials and others). The resulting fingerprint represents the expected behavior of the heat.
Comparing the fingerprint and the real-time trends, the application detects whether the decarburization process is proceeding with the expected rate or the profile requires some adjustments. The control strategy dedicated to this is named soft landing; it dynamically adjusts the oxygen injection to hit the final carbon and temperature without over-oxidizing the heat (figure 3).
Fig. 3 - Soft landing control in action optimizing the decarburization profile in real time
44
La Metallurgia Italiana - n. 7/8 2018
AttualitĂ industriale From the first valid cartridge sample measurement on, the application also tracks the bath carbon / temperature / dissolved oxygen thanks to its integrated process models (Fig. 4). These
models are fitted on historical data stored by the system, so their output is tuned on the specific EAF considered and it adapts to slow process changes over time.
Fig. 4 - Melt Model bath decarburization tracking
The cartridge measurements results, as well as the steel analysis, are also considered by the application to control the decarburization profile: when available, this information is used to further increase the accuracy of the final bath composition and temperature. Q-REG ELECTRODE REGULATOR Q-REG is DANIELIâ&#x20AC;&#x2122;s latest release of the electrode regulator system for electric arc furnaces. The controller is based on the innovative PAC (Programmable Automation Controller) highperformance platform which has plenty computational power to manage the complex electrode regulation control with extremely low response times. Dynamic electric energy control Controlling the position of each electrode column, the system dynamically adjusts the electrical set-points to adapt to the furnace and network conditions and to achieve the highest possible active power input. In this field, Q-REG main features are: fast response hydraulic counter pressure control and touchdown function (lower electrode breakage risk), boring-down dynamic control (auto-regulation of the electrical
La Metallurgia Italiana - n. 7/8 2018
working point, to increase the power as soon as possible), automatic supply voltage compensation (uniform operation and power inputs without operator intervention), transformer over-current and thermal protection with secondary insulation control (safer EAF operation). A faster hydraulic system response and stiffer mechanical design further improve the electrode positioning performances. The traditional current-voltage-impedance control modes were also extended including an advanced MIMO (Multiple Input, Multiple Output) model of the furnace electrical system. The model is used to predict in real time the electrical effects of one phase on the other two. With this approach, the system predicts the effects of phase unbalancing, rather than just record them, and it thus reacts faster to the changing conditions inside the EAF. Electrical energy management is further optimized according to additional process variables. By means of the foamy slag status evaluation, the system detects whether the slag is properly shielding the arcs or not, and adapts the electrical setpoints accordingly to minimize energy losses and power-on time.
45
Industry news
Fig. 5 - Q-REG electrode regulator overview
Q-RAY arc irradiance supervisor An innovative real-time arc irradiance supervisor was added to further optimize the electrical profile. This module evaluates the total irradiative heat flux along the furnace walls to monitor the thermal load on the watercooled panels (Fig. 6). The system thus modifies the electrical set-points to maximize the active power until the end of the heat, without severe stress to the panels and the refractory
lining in case of operation with uncovered arc. The main characteristics of Q-RAY control strategy released in Kroman Celik are: > Possibility to dynamically unbalance the electrode current if the temperature of the corresponding panel reaches the alarm threshold. > Possibility to reduce the tap position if the predicted panelsâ&#x20AC;&#x2122; temperature moves quickly towards the trip threshold.
Fig. 6 - Q-RAY arc irradiance supervisor overview
These features were enabled to improve the panelsâ&#x20AC;&#x2122; protection; it also reduced the current on each electrode by 5%. Arc radiation evaluation also allowed the implementation of a tap-holding control strategy to keep the maximum tap adopted in the static profile as long as possible during bucket melting. This feature further reduced power-on time without negative consequences on the furnace panels.
46
Dynamic foaming slag control During the refining phase, arc coverage by the foamy slag plays a key role. The dynamic foamy slag control continuously monitors the slag condition, evaluating the Arc Coverage Index (ACI), a proprietary function based on arc voltage and current real time fast signal processing.
La Metallurgia Italiana - n. 7/8 2018
AttualitĂ industriale
Fig. 7 - Dynamic foamy slag control
The status and tendency of the ACI are considered by a controller acting on the coal injection flow to increase the electrical energy transfer, while keeping the arcs shielded by the foaming slag (Fig. 7). As arc coverage is evaluated for each electrode, coal injection is regulated acting on each single carbon dispenser, to focus the action on the injectors closer to the unstable phase. Towards the end of the heat, dynamic regulation is applied also to lime-dolomite injection, to recover proper slag basicity and viscosity and thus decrease the bath thermal losses.
LINDARTM OFF-GAS ANALYSIS AND POST-COMBUSTION CONTROL Q-MELT technological packages suite for process optimization comprises a very important feedback to characterize the furnace freeboard conditions in real time. This is done by means of LINDARC real-time off-gas analyzer. The technology is based on the Tunable Diode Laser Absorption Spectroscopic (TDLAS) technique [3, 10] and is engineered to carry out the measurement directly in the EAF off-gas stream. The system continuously measures composition and temperature of the gases crossing the laser path (Fig. 8).
Fig. 8 - LINDARC installation and laser beams The measurement ranges for the species monitored are reported in Table 2. A complete analysis is performed in a cycle
La Metallurgia Italiana - n. 7/8 2018
time of 2 seconds.
47
Industry news Tab. 2 - LindarcTM system features Molecule
Range
Temperature
O2
0 ÷ 25%
CO
0 ÷ 100%
CO2
0 ÷ 100%
H2O
0 ÷ 50%
0 ÷ 1600 °C 400 ÷ 1600 °C 400 ÷ 1600 °C 600 ÷ 1600 °C 400 ÷ 1600 °C 2s
Temperature Response Time Conventional extractive systems require gas sample suction, filtering, cooling and conditioning before sending it to multiple gas analyzers. As a consequence they are inherently slow and lag 20-30 seconds behind the process. Gases sample extraction and manipulation introduce intensive maintenance: extractive systems are prone to clogging, water condensation and air infiltration. This can impair the analysis information sent to the process control and limits the feedback control actions. An in-situ laser-based system addresses effectively all these aspects. LindarcTM installation For practical reasons, the off-gas analysis system is installed
downstream to the furnace. Between the roof elbow and the movable sleeve, a gap is intentionally left open, entraining enough air to supply combustion oxygen and to cool down the fumes below 850°C at the settling chamber. As air infiltrates and mixes, it reacts with the gas stream coming from the furnace. The measurement point is chosen to ensure that the laser beams interact only with the EAF offgas, avoiding any dilution effect with the secondary air from the gap. The correct location of the beams is right at the core of the EAF gas stream and it is selected with the support of specific CFD simulations (Fig. 9) considering melting and refining operating conditions.
Fig. 9 - Gasdynamics to select LindarcTM place
Two water-cooled lances protrude inside the fixed duct through the duct walls, just downstream the movable sleeve. The fingers shield the laser beam to reach the uncontaminated core of the hot off-gas stream. The transmitter units are installed on a side of the duct. They beam the laser light through pipe channels flushed by ni48
trogen, a non-absorbing inert gas. At the lance exit, the laser beams cross the gas stream that selectively absorbs the spectra lines corresponding to specific molecules resonance frequency. The residual laser light reaches the receiver unit at the opposite side of the duct.
La Metallurgia Italiana - n. 7/8 2018
Attualità industriale
Fig. 10 - LindarcTM receiver and control cabinet The laser transmitter and receiver units are aligned with the water cooled fingers and are protected by heavy duty housings (Fig. 10) to prevent any damage, due to the very harsh environment typical of this location. The cooling system and purging gases are managed by a dedicated cabinet, located in a safe and easily accessible location away from the EAF. Dynamic post-combustion control Before the advent of off-gas measuring technologies, the only way to perform post-combustion was to add some extra oxygen to the stoichiometric value via the burners, based on pure speculation. Even if this practice was, and still is, widely
used among steelmakers, it negatively affects electrode, carbon, oxygen consumptions and also metallic yield. Static input of extra oxygen cannot be efficient because CO and hydrocarbon evolution in the EAF freeboard are all but steady values [3]. They are indeed highly variable and hardly predictable from the melting program and the charge recipe. Real-time off-gas analysis provides an effective answer to this need. The fumes composition measurement is considered by a Closed Loop Control (CLC) to manage the dedicated postcombustor units (Fig. 11), as well as the burners’ oxygen/ natural gas ratio if necessary.
Fig. 11 - Postcombustor unit
A key factor of this control is the very fast response time ensured by the LINDARC™ technology. The control is based on the evaluation of the Post Combustion Degree (PCD), defined as follows: (2) La Metallurgia Italiana - n. 7/8 2018
The formula assesses the oxidation level at the furnace freeboard. The lower the PCD, the higher extra-oxygen will be input by post-combustor units. The higher the PCD, the lower the burners oxygen-natural gas ratio will be, with the lowest limit defined by the oxygen-natural gas ratio set in the burner profile. Each individual unit (post- combustor, burner and si49
Industry news dewall injector) is independently configured, introducing specific limits for the flow rates and timing within the dedicated HMI application. Area performance indicator and process data analysis platform Given the high level of process automation achieved with the
Q-MELT suite, a proper user interface is needed to give to the operator a clear picture of the process status and of the operation of each dynamic control. The EAF Area Performance indicator was thus redesigned to accommodate all these information (Fig. 12).
Fig. 12 - EAF Area Performance Indicator showing main process KPIs, planned and applied profiles, and process models outputs Along the process, the application summarizes the process status with a panel comprising the most important KPIs (heat grade, charged weigth and specific consumptions, among others). The main plots are designed to show the differences between the static process profiles and the ones adapted by the dynamic controllers, both in terms of electrical and chemical set-points. During the refining step, as slag and bath conditions are more relevant, the plots show the arc coverage trend and the estimated bath temperature and Carbon %. The operations of all the dynamic controllers, namely the electrode regulation, the dynamic foamy slag management,
the post-combustion control and the end-point soft-landing management, are summarized by dedicated indicators on the footer, to have a clear overview of which controller is currently adapting the corresponding set-point. The huge amount of data acquired automatically by Q-Melt is a significant additional value of the system, as their analysis helps technologists to monitor the process performaces over time and optimize the applied profiles. At Kroman Celik, the complete Q3Intelligence suite for EAF was installed, which allows multiple data analysis levels (Fig. 13).
Fig. 13 - Q3Intelligence EAF dashboard and process trends analysis
50
La Metallurgia Italiana - n. 7/8 2018
AttualitĂ industriale The web-based dashboard, available and released also on the mobile platform, is a lightweight and fully customizable panel to monitor the desired KPIs over the selected time period. For more advanced statistical analysis, a Excel-based, fully customizable analysis tool allows to process huge amount of data (full process trends of thousands heats) and to extract average and deviation trends in a matter of seconds. Process technologists may select among all relevant variables made available also by the offgas probe to correlate them with different practices, charge recipes, heat energy, duration classes and raw materials used.
NEW CHEMICAL PACKAGE The new EAF layout is reported in Fig. 14. It is composed of: > 4 M-ONE combined oxygen and carbon injectors (to replace supersonic oxygen injectors and carbon lances); > 2 Post-combustors. > 1 Oxygen jet at the sump. > 2 LIMEJET injectors for pneumatic conveying of lime and dolomite. > 3 MOLI lime dispensers to dose and inject lime and dolomite into the EAF.
Fig. 14 - New EAF layout The rationale behind the new layout concept is to concentrate the heat input in the cold spots of the furnace and to locate the lime and dolomite injection at the arc hot spots. Post-combustors are installed at the cold spots, too. They are intended to cooperate with the sidewall injectors, in particular with M-ONE units, which can be considered as carbon monoxide generators. Post-combustors are designed to blow soft oxygen in order to accomplish the carbon monoxide combustion inside the furnace close to the bath and inside the scrap during melting. The gas utilities were rerouted from the
La Metallurgia Italiana - n. 7/8 2018
existing valve stands in order to limit the capital investment. M-ONE injectors M-ONE is integrating three functions in a single unit: > Mixed swirled flame burner; > High efficiency supersonic coherent oxygen lancing; > High momentum coal injection. The all-in-one design (Fig. 15) was designed to reduce the number of devices installed in the furnace shell, in order to ease the assembly work, to reduce the capital investment and to improve the reliability of equipment.
51
Industry news
Fig. 15 - M-ONE injector and box installation The powerful mixed swirled flame, spreading on a larger area improves the burner efficiency, involving larger portions of scrap and having a reliable self-cleaning effect against tip
clogging. It does not require any low oxy-fuel flame to prevent slag sticking and skull formations (Fig. 16).
Fig. 16 - M-ONE injector: burner mode flame Oxygen lancing is therefore more effective: the advanced De Laval nozzle design allows a rapid scrap meltdown and an early access to the melt. Combined coal and oxygen injection promotes immediate control of the liquid slag interaction with refractories at the
52
slag line. In the past, the distance between the oxygen and the coal injection points required time to complete the reduction reactions and FeO infiltration in the refractory wall quickly dislodged the refractory bricks from the carbonaceous binder at the slag line.
La Metallurgia Italiana - n. 7/8 2018
Attualità industriale
Fig. 17 - M-ONE injector: CFD analysis of coal particle acceleration The design exploits the proximity between oxygen and coal jets to further accelerate the solid particles with the faster gas stream. Extensive CFD parametric modeling was performed in order to maximize the solid particles acceleration by the jet’s entrainment effect, while avoiding negative effects on the supersonic oxygen free-jet structure (Fig. 17). Lime and dololime injection Lime injection consists of a fully automatic system using a
dry day bin for grain sized materials storage. The accurately weighted pressure vessels pneumatically convey the solids to dedicated sidewall injectors. Limejet injectors are devices specifically designed for grain-sized slag formers injection at a velocity of 60 to 80 m/s, sufficiently high to reliably penetrate the thick slag layer and disperse even the finer particles of flux at a distance in excess of 1.5 m (Fig. 18).
Fig. 18 - Lime particles’ jet
Sidewall injection is done close to the slag: the short travelling path limits fines’ loss to the off-gases. Consumptions savings and less dust load to the baghouse are direct benefits of the high material efficiency.
La Metallurgia Italiana - n. 7/8 2018
System response time and slag reactions kinetics are very fast, making lime and dolomite injection suitable for dynamic control and integration with the electrodes regulator.
53
Industry news
Fig. 19 - Limejet Sidewall Injector
The Limejet injector integrates a powerful burner function in the same tip (Fig. 19). The flame efficiently heats and melts the scrap in front of it, making a hole in the charge and pre-
venting the formation of skulls at the wall, commonly reported with conventional simple pipe installations, which are often prone to severe clogging.
Fig. 20 - Slag Former injection system in Kroman Celik
The best feed-rate set-point delivers the required amount in the suited process time, without accumulation of unreacted lime at the wall. The system therefore effectively controls the slag V-ratio and local slag temperature. The solution installed at Kroman Celik (Fig. 20) allows to load also magnesia binders, to aim at slag double saturation.
54
A MgO-FeO saturated slag promotes refractory savings (longer campaigns) and slag apparent viscosity, improving foamy slag [5]. These functions are especially valuable during superheating, when slag foam tends to decline as a direct consequence of FeO concentration and the viscosity drop that is following the temperature raise [6, 7, 8, 9].
La Metallurgia Italiana - n. 7/8 2018
AttualitĂ industriale RESULTS AND CONCLUSION The EAF retrofit in Kroman Celik was completed in 9 days only and the first heat was performed on March 9, 2017.
Fig. 21 - Control room with Q-Melt in operation The learning curve has been very fast with a rapid ramp up of the productivity.
The first results achieved after few weeks of operation are quite interesting, as reported in Table 3.
Tab. 2 - Achieved results
Power-on-on Tap-to-Tap Electric Energy Oxygen Natural gas Coal Lime+Dolomite
min min kWh/t Nm3/t Nm3/t kg/t kg/t
Process fine-tuning continued along the following period and more consolidated results were achieved. The differences between the average EAF performances of the 3 months after
-1 -2.4 -25 +1.5 +0.3 -3.8 -0.7
startup (1602 heats) and the 6 months before startup (2867 heats) are reported in Table 4.
Tab. 4 - Achieved results after 3 months from startup
Power-on-on Tap-to-Tap Electric Energy Oxygen Natural Gas Coal Pig iron Lime Dolomite Electrode
La Metallurgia Italiana - n. 7/8 2018
min min kWh/t Nm3/t Nm3/t Kg/t % kg/t kg/t kg/t
-2,9 -2.7 -20 +2,6 +0.7 -1.0 -5 -3.6 -1.8 -0.30
55
Industry news Coal consumption was decreased by 1 kg/tls, but if we consider carbon input from pig iron, the total carbon benefit could be extended up to 3.2 kg/tls; as additional benefit, there was the possibility to cut the pig iron cost from the charge. Despite the oxygen input increase and the carbon input decrease, the steel oxidation was not affected due to the following reasons: > The extra oxygen was given mainly in the post combustion phase during the bucket melting. > The carbon reduction was reduced thanks to the more effective side wall injectors and to the injection reduction in arc covered conditions.
> The oxygen dynamic control during flat bath operation allowed to control the level of oxidation within the established target. The main result achieved was the electrical energy reduction by 20 kWh/t. However, if we consider the changes in the chemical input, the total energetic saving was equal to 42 kWh/t. This increased the energetic efficiency, intended as ratio between steel enthalpy and total energy input, by 3%. The comparison between the slag composition in the 2 months before startup and the 2 months after startup shows a higher level of process repeatability, leading to a lower standard deviation of the values reported in Table 5.
Tab. 5 - Standard Deviation reduction
-on Fe2O3 V-ratio MgO
A further benefit achieved, partially explained by the above table, was the shell campaign duration, which was extended by 17% with respect to the previous campaigns. The remaining thickness of the bricks was sufficiently good to suggest a further increase of the shell refractory life in the next campaigns. The results achieved are only the first step in the continuous improvement plan agreed between Kroman, DANIELI and MORE. The equipment described in this paper is a complete suite of tools to manage and control the furnace with a fully automa-
56
% % %
STDEV reduction -16.2 -12.6 -51.7
tic approach. The Q-Melt architecture is designed to collect process operating knowledge along time and to recognize variations by tracking the key performance indicators. It thus detects changing conditions and self-adapts accordingly by selecting the best practices, to keep the production performances at the optimal level. Acknowledgment The authors would like to thank Kroman Celik A.S. personnel and management for the support given during the project execution and commissioning.
La Metallurgia Italiana - n. 7/8 2018
Attualità industriale REFERENCES 1]
D. Patrizio, A. Pesamosca, “Latest trends in EAF optimization of scrap-based melting process: balancing chemical and electrical energy input for competitive and sustainable steelmaking”, 11th EUROPEAN ELECTRIC STEELMAKING CONFERENCE AND EXPO, 25-27 May 2016, Venice, Italy.
2]
MORE highlights n. 13, April 2011
http://www.more-oxy.com/repo/archive/hl13.pdf 3]
D.Tolazzi, M. Piazza, O. Milocco, “Installation and Operational Results of LINDARCTM Real Time Laser Off-Gas Analysis System at Acciaierie Bertoli Safau – ABS (Italy) Electric Arc Furnace”, Metec & 2nd Estad 2015, 15-19 June 2015, Dusseldorf, Germany
4]
D. Tolazzi, C. Candusso, S. Marcuzzi, “New Developments and operational results in the use of fixed side-wall injectors in the electric arc furnaces”, 11th EUROPEAN ELECTRIC STEELMAKING CONFERENCE AND EXPO, 25-27 May 2016, Venice, Italy.
5]
Pretorius, Carlisle, “Foamy slag fundamentals and their practical application to electric furnace steelmaking”, Proceedings of the 16th Process Technology Conference; 1998 Nov 15-18; New Orleans, LA, USA.
6]
Mc Gill, M. Iacuzzi, A. Beasley, “Applications and Operating Results of Pneumatic Lime/ Dololime Injection Technology in Severstal (SDI) Columbus EAFs”, AISTech 2015 Steel Conference, Cleveland, USA
7]
L. Wolfe, J. P. Massin, T. Hunturk, W. Ripamonti, “Lime Injection Technology – A Viable Tool For The Electric Arc Furnace”, technical paper: http://www.carmeusena.com/sites/default/files/brochures/steel/tp-lime-inj-eaf-2008.pdf
8]
Wolfe, J. Korn, “Overview Of Lime Injection In The Electric Arc Furnace”, technical paper: http://www.carmeusena.com/sites/ default/files/brochures/steel/tp-lime-inj-eaf-2007.pdf
9]
R. McClanahan, L. Kibler, L. Wolfe, A. C. Dyar, J. M. Compton, “Comparative Analysis of Dolomitic Lime and Chinese Magnesite Practices in Electric Arc Furnace Steelmaking Slags”, technical paper: http://www.carmeusena.com/sites/default/files/brochures/steel/tp-nsbhp-20paper-aise-202003.pdf
10]
M. Piazza, F. Bianco, D. Patrizio, M. Ometto “EAF process optimization through a modular automation system and an adaptive control strategy”, 11th EUROPEAN ELECTRIC STEELMAKING CONFERENCE AND EXPO, 25-27 May 2016, Venice, Italy.
La Metallurgia Italiana - n. 7/8 2018
57
Industry news Water Leak Detection in EAF based on Tenova’s off-gas technology: recent developments and results in Lucchini RS, Lovere, Italy edited by: M. Luccini, V. Scipolo, D. Zuliani, L. Poli, D. Masoero Tenova’s EFSOP® Water Detection System (WDS) samples and measures the Electric Arc Furnace (EAF) off-gas complete chemistry (H2, CO, CO2, O2, H2O) continuously and uses those measurements to alert operators in real-time of statistically abnormal high humidity or abnormal high hydrocarbons in the EAF, which might pose a serious risk during the scrap melting and refining phases. Specifically the EFSOP® WDS alerts operators when the system metrics deviate significantly from the established reference normal practice due to such events as (but not limited to) high hydrocarbons or water contained in the scrap or furnace panel or other water leaks. The proposed paper reviews the critical factors needed for the effective development of Tenova’s WDS and highlights the results obtained in the 60 tons EAF at Lucchini RS, Lovere, Italy after completing the WDS tuning in August 2016, with particular attention to the preventive alarms generated by the WDS in case of panel water leaks.
KEYWORDS: WATER LEAK DETECTION - EAF
M. Luccini, V. Scipolo, D. Zuliani Tenova Goodfellow Inc
L. Poli Lucchini RS
D. Masoero Tenova S.p.A.
INTRODUCTION LUCCHINI RS, LOVERE, ITALY Lucchini RS is a steel manufacturing group which offers a diversified range of high-tech products and services and operates globally. Lucchini RS produces in its Lovere Plant in Italy a wide range of steel castings and forgings used in all industrial sectors such as power generation, petrochemical industry, off shore, plant engineering, steel industry and cement works. Thanks to the latest melting, refining technologies and auxiliary equipment, the steel-making plant is capable of producing all types of steel from carbon, alloy and stainless qualities with a high degree of purity. Lucchini RS's quest for quality in all its casting and forging products begins with the production of liquid steel from carefully selected scrap. Melting takes place in a 60 ton capacity electric arc furnace powered by a 38 MVA transformer. The results presented in this paper were obtained during the tuning and commissioning of 58
Tenova Goodfellow WDS for the 60 tons EAF in Lovere, Italy, between April and August 2016. After successful completion of system tuning tests and the real-time detection of several all actual water leaks with only ~ a 1% false alarm rate, the WDS technology was formally accepted by Lucchini RS management and has been actively monitoring the furnace since September 2016. Success factors for eaf off-gas water detection Three basic off-gas analysis technologies are now available for use on an EAF: - Insitu laser systems use a tunable diode laser to transmit a beam in the near IR range through the off-gas for subsequent pick-up by an optical detector. The transmitted laser’s wavelength is modulated around the particular spectroscopic line of the gaseous species of interest. The amount of absorption in the detected beam is subsequently used to calculate La Metallurgia Italiana - n. 7/8 2018
Attualità industriale the concentration of that particular species in the off-gas. EAF insitu laser systems use up to 3 separate lasers, one laser for CO2 and H2O vapor, one for CO and one for O2. While insitu laser systems can analyze several gases, lasers cannot analyze many mononuclear diatomic gases including H2 [1]. In total about 10 insitu laser EAF off-gas analysis systems have been installed since first developed in about the year 2000 by Linde and Air Liquide [2]. - Extractive systems use a water cooled probe, heated line and analyzer to continuously extract and analyze a sample of EAF off-gas from the fume duct. Various analytical methods are employed to continuously analyze a complete spectrum of off-gas chemistry in real-time including CO, CO2, O2 and H2. While there are various extractive systems available on the market, with well over 80 EAF steel plant installations worldwide since its commercial introduction in 1998, EFSOP® extractive technology is the established industry standard for off-gas analysis and EAF process optimization [3,4,5,6,7]. As discussed in this paper, Tenova Goodfellow recently developed a complete EFSOP® analyzer that includes H2O vapor to the spectrum of gases that are analyzed. - NextGen® hybrid technology combines the best features of insitu laser and extractive to achieve highly reliable full spectrum off-gas analysis (CO, CO2, O2, H2 and H2O), fast response times, reduced maintanence and lower installation costs.[10,11,12] Tenova’s 2015 commercial launch of NextGen® technology focused on North America where 9 systems have now been installed or are underway. NextGen® is now available internationally with several systems pending in Europe, South America and Asia. Effective water detection under all EAF operating conditions requires that off-gas analysis systems must demonstrate; a. Exceptional reliability; b. Fast response times; c. The capability to analyze both H2 and H2O vapor to detect leaks in both reducing & oxidizing freeboard conditions; d. The capability to provide “Operator Alerts” with minimal false alarms; (a) System Reliability It is most important for a water detection system to be active and working reliably from “start to end” of the heat and especially whenever the power is on. Any gaps in water detection measurements during the course of the heat can represent a potentially serious situation especially if a water leak were to occur when the detection system is inoperative for whatever reason. Also to be effective for process control and
La Metallurgia Italiana - n. 7/8 2018
water detection, the off-gas analysis technology must analyze “true furnace gas” before dilution with combustion air introduced at the 4th hole. This requires careful positioning of the off-gas chemistry insitu laser transmitter/detector or the extraction probe. Insitu laser systems can be categorized as “passive” technology – they rely on passive transmission of a laser beam through the off-gas fume from the emitter to the detector. Any attenuation or interference of the laser signal that prevents its sufficient detection will result in interrupted off-gas analysis. To help mitigate laser beam attenuation problems, today’s insitu optical off-gas systems use continuous N2 purged, water cooled horizontal steel pipes to shorten the laser transmission path length [8]. As with the extractive probe, care must be taken to position the open gap between the horizontal steel pipes directly within the cone of off-gas exiting the EAF to ensure that the laser beams pass through only undiluted EAF process gas. While the shortened path length, horizontal or vertical probe configuration has considerably reduced beam attenuation problems compared to original full path length insitu designs, because of the passive nature of laser transmission there still remains a real risk that 1 or more of the insitu laser beams may suffer from periodic and unpredictable interruptions in signal transmission especially when dust loading is particularly high during melting. Any lost laser signals during melting would limit effectiveness of the water detection system during critical periods when hung-up scrap can fall into bath and create a metal slosh event that can trigger a water leak related explosion. On the other hand, extractive off-gas systems such as EFSOP® and the NextGen® hybrid system are “active” technologies - whenever power is switched on, these two systems automatically switch to an active, positive high suction to rapidly extract a sample of off-gas from the fume duct at high flow rate for chemical analysis. Extractive & NextGen® systems are designed to automatically switch to an active N2 or instrument air purge during “power-off” periods to clean the probe and filters before the start of every heat and whenever power is off during the heat. Historically, such technology has demonstrated exceptional reliability – when properly maintained, EFSOP® and NextGen® technology has consistently demonstrated better than 99% reliability to provide continuous off-gas chemistry from start-to-end of the heat. As shown in Figure 1, the probe is designed to operate directly inside the cone of off-gas exiting the EAF at the 4th hole thereby ensuring the system is sampling true process
59
Industry news gas before dilution with combustion air. Analyzing undiluted EAF process gas is critical for both EAF process control and effective water detection. The probe is positioned vertically
to reduce debris build-up and to use gravity assist for more effective particulate removal during the N2 purge cleaning cycle.
Fig. 1 - Off-gas system 4th Hole configurations; Insitu laser (left), EFSOP® & NextGen® extractive probe (right)
(b) Fast Response Times When liquid H2O enters the EAF it quickly produces H2O vapor a portion of which will further react to produce H2. These gaseous products of a water leak are immediately present in the freeboard off-gas inside the furnace. Because of the fume system’s high suction rate, the freeboard gases including H2 and H2O vapor will be extracted from the EAF through the 4th hole in a matter of a few seconds. As a result, properly designed off-gas based water detection systems have an inherent quick response time. Insitu optical systems have been reported to provide analytical results at about 2 second intervals. By comparison, traditional extractive systems have a probe tip to analysis response time of about 20 to 30 seconds. Historically, such response times have proven very effective for dynamic closed loop burner, injector and fume suction control and process optimization because they avoid problems associated with “chasing your tail” control logic. For more effective water detection response, Tenova Goodfellow’s WDS used in both EFSOP® and NextGen® systems was redesigned to provide a probe tip-to-analysis response time of 8 seconds or less, proven by the results of many Controlled Water Trials conducted in different installations. In summary, it can be concluded that when combined with the high fume system suction rate, the insitu optical system and the latest EFSOP® & NextGen® systems have exceptio60
nally fast response times as is required for effective water detection. (c) Capable of Analyzing Both H2 and H2O Vapor The relative quantities of H2O vapor and H2 in the EAF freeboard are chemically linked to each other. Under normal practice, both H2O vapor and H2 gas are naturally present in the EAF from electrodes water sprays, from moisture in the scrap charge and as products of combustion from the burners, post combustors, injectors and residual oils on the scrap. All of these sources can be considered as contributing to “normal” levels of H2O vapor and H2 in the EAF freeboard. Once present in the EAF, H2O vapor and H2 are continuously reacting with Fe, FeO, CO2, CO and C, the dynamics affecting the actual “H2O vapor to H2 ratio” in the freeboard off-gas are quite complex and variable. Adding to this complex situation, at any time during the course of the heat, the EAF can effectively switch between an oxidizing and reducing freeboard off-gas chemistry depending on the level of burner firing, oxygen lancing, post combustion and fume system suction. When liquid water unexpectedly leaks into the EAF it will immediately begin to boil to form H2O vapor. Depending on the location of the liquid water leak and whether the EAF is operating in an overly oxidizing or reducing condition, a proportion of the resulting H2O vapor will further La Metallurgia Italiana - n. 7/8 2018
Attualità industriale react and dissociate to H2 gas thereby leading to an “abnormal” amount of H2O vapor and H2 in the EAF freeboard. The equilibrium off-gas chemistry will shift towards more H2 if the furnace is operating in a more reducing condition. However, the thermodynamics will favor a shift towards more H2O vapor if the furnace is more oxidizing which for example, can happen if fume system suction is too high. Because of the complex interdependence of these chemical reactions and their dependence on the level of combustion within the EAF and the amount of fume system suction, it is simply not possible to predict the relative proportion of H2 and H2O vapor in the freeboard off-gas in advance of a water leak. Hence, for water detection to remain effective under all operating situations, it is absolutely necessary that the offgas analysis technology be capable of analyzing both H2O vapor and H2 gas. Importantly, unlike insitu laser which cannot analyze H2, EFSOP® and NextGen® technologies analyze both H2 & H2O vapor and hence are capable of providing coverage for water leaks in both reducing & oxidizing freeboard conditions. (d) Capable of Providing Effective “Operator Alerts” with Minimal False Alarms As discussed above, there is always a “normal” background level of H2O vapor & H2 in the freeboard off-gas inside the EAF. However, it is important to note that the absolute level of “normal” H2O vapor and H2 in the off-gas and the ratio of H2O vapor to H2 in the off-gas will be varying from start to end of the heat and from heat-to-heat depending on how wet and/or oily the scrap charge is, on the amount of burner firing and post combustion at any point during the heat, on the level of the electrode sprays and on the level of fume system suction. To be effective, a water detection system must be able to quickly and correctly distinguish between “abnormal” H2O vapor & H2 levels due to a water leak into the EAF and “nor-
mal” background levels of H2O vapor & H2 due to operating practice. Tenova Goodfellow has developed a software package that can dynamically and quickly analyze the actual H2 and H2O vapor off-gas chemistry in real-time, dynamically adjust the baseline “normal” water levels upwards/downwards in response to changing scrap & furnace conditions so as to correctly distinguish between “normal” and “abnormal” conditions, and then provide the EAF operators with an HMI showing well defined “alerts” that clearly indicate when H2O vapor levels exceed normal levels. WDS MODEL DESCRIPTION As introduced above, Tenova Goodfellow’s Water Detection System (WDS) relies on a proprietary set of algorithms to differentiate between statistically “normal” & “abnormal” H2O vapor levels in the EAF. In order to best adapt to different process conditions in the EAF and to the heat-to-heat process variability, the current WDS technology employs a proprietary control algorithm that automatically and dynamically adjusts the “expected H2O% in normal conditions” in response to the EAF process evolution, coupled with a second algorithm which directly monitors the measured H2O%. Figure 2 schematically shows the structure of the WDS control algorithm and its two “cores algorithms”. The first model is responsible for the estimation of the expected H2O% in the given EAF conditions and based on the full range of off-gas chemistry (including H2); a second model monitors the behaviour of the measured H2O% and its variation in time. The two independent models are combined to provide a reliable indication whether the real time H2O% concentration in the EAF is to be considered “normal” for the current process conditions or if it is the indication of excess humidity which needs to be monitored.
Fig. 2 - WDS algorithm block scheme architecture
La Metallurgia Italiana - n. 7/8 2018
61
Industry news Figure 3 graphically illustrates the methodology for triggering the “Operator Alerts”. While the system is not a failsafe method, it does provide operators with valuable real-time alerts indicating the statistical probability of abnormally high amounts of H2O vapor in the EAF, based on direct H2O vapour & H2 concentrations measured by EFSOP® or NextGen® in relation to the overall off-gas chemistry composition. The blue curve is the direct measured H2O vapour concentration. The green dotted curve represents the expected H2O% vapour concentration in normal conditions, based on the EAF process characteristics and the off-gas real-time chemistry status measured by the EFSOP® or NextGen® system. The yellow dot-
ted curve represents the threshold limit calculated to detect a potentially hazardous situation and/or an out-of-ordinary H2O vapour concentration. The threshold limit calculation is the result of a tuning campaign, which typically involves both the process modelling for the specific EAF (chemical package study, operations) and controlled water injection trials to simulate the effect of a water leak. Finally, a specific algorithm is introduced to evaluate the behaviour of the measured H2O% vapour concentration above the warning threshold, in order to maximize the system effectiveness in detecting unusual conditions and to minimize the possibility of false alarms.
Fig. 3 - Graphical methodology for triggering a water detection alert. The measured H2O (blue) is compared to the expected H2O (green) and the warning threshold (orange). If the measured H2O crosses upward the warning threshold, a specific algorithm monitors the behaviour of H2O and triggers an alarm for water leak within a specific monitoring period. When the measured H2O% concentration is below the calculated warning threshold, the WDS software will display a “Green Condition” indicating that the statistical probability of excessive amounts of water in the EAF is low. When the measured H2O% exceeds the calculated warning threshold, the WDS will enter in a warning state, or “Amber Alert”, meaning that the situation will be monitored closely for a time period defined together with the plant personnel. In Amber Alert situations, operators are strongly advised monitor the information of the WDS, which is implementing a specific algorithm to identify if the EAF conditions are persistent in time or simply transitory. According to the implemented algorithm, when the metrics exceed their limits for a prescribed period of time, a “Red Alert” is issued indicating that the off-gas
62
chemistry is significantly out of the statistically normal range and that there is a high probability of excess water in the EAF. Red Alerts require immediate protective action by EAF operating staff. The above alarming scheme is designed and tuned to maximize the probability of real leak detection for leaks above a minimum flow rate (L/min), which depends on the single EAF and process operations. A false alarm is defined when a “Red Alert” is triggered but no real leak is identified. That situation can happen when an excess of H2O vapour is produced in the furnace and cannot be explained by the information available to the WDS. When properly tuned, the WDS is designed to target a false alarm rate around 1% of the analyzed heats ( 1 “Red Alert” / 100 normal heats ).
La Metallurgia Italiana - n. 7/8 2018
Attualità industriale RESULTS OBTAINED AT LUCCHINI RS, LOVERE, ITALY (a) Accuracy of H2O estimation in normal conditions As described in the previous paragraph, the WDS model is structured in two “core algorithms”. The first one of them is responsible for the estimation of the H2O% in normal conditions and provides the reference against which the actual measurements are compared. The main challenge of water leak detection systems based on off-gas measurement is the ability to differentiate between regular H2O% variations (due to process and scrap quality) and a typical variations due to extra amount of water entering the furnace. Moreover, leaks happening at different times / furnace conditions might result in different H2O% measurement responses and characteristics. For these reasons, Tenova Goodfellow defined a specific approach to the problem based on the estimation of H2O% in normal process conditions. The estimated H2O% in normal conditions is the result of a specific modelling which takes into account the full chemistry composition of off-gases measured by EFSOP® (or by NextGen®) (CO2, CO, O2, H2) and some process parameters which are related to H2O vapour concentration. The WDS estimation model is the results of a significant data analysis and process modelling specifically tailored to each EAF the system is applied to. For the WDS effective application in the water detection problem, it is of primary importance that the estimated H2O% is accurate enough to describe the behavior of measured H2O% in normal process conditions (when no water leaks are present). As an example, Fig. 4 and Fig. 5 show the WDS estimation in normal conditions during two different heats in the
60 tons EAF at Lucchini RS, Lovere, Italy. Both the heats are two-charges and refining, clearly visible in the graphs. The WDS model is able to accurately estimate the normal H2O% variations during the process in both cases, even though the humidity behavior is clearly distinct in the two situations, especially in the refining phases. The estimation accuracy is the fundamental first step to achieve the detection of unusual behaviors of measured H2O%, possibly connected to a water leak in the furnace. It is evident that the WDS is capable of estimating different H2O% trends related to different processes and EAF conditions. As a result, the calculated WDS warning threshold (orange line) is not constant for all the heats but is adapted to the conditions of each single heat, with great benefits in terms of probability of detection of “outof-ordinary” events and reduced probability of false alarms. (b) Detection of Actual Water Leaks occurred Tenova Goodfellow acknowledges that the WDS response to real water leaks is the most convincing result for the proposed WDS technology. The following figures show the WDS response to two real water leaks in 60 tons EAF at Lucchini RS, Lovere, Italy, occurred during the tuning and commissioning of the system (April - August 2016). Fig. 6 shows the WDS recorded data for the three buckets heat 161274 (April 16, 2016). During the melting of the first charge, the measured H2O% ( blue line ) is close to the calculated expected H2O% for the normal process (green line) and well below the WDS calculated warning threshold ( orange line ). Toward the end of the second charge melting it is evident a sudden increase in measured H2O% above the warning threshold, resulting from a water leak in the burner housing.
Fig. 4 - Example of WDS active during a two charges heat. The blue line represents the measured H2O%, the green line the expected H2O% in normal conditions estimated by the WDS model. It can be appreciated the ability of the WDS model to describe the variations of the measured H2O% purely relying on the off-gas composition and process information. La Metallurgia Italiana - n. 7/8 2018
63
Industry news
Fig. 5 - Example of WDS active during a two-charges heat. It is clear by comparing Fig. 4 and Fig. 5 the different behaviour of the measured H2O, which is still well resolved by the WDS model (blue vs. green line). When the measured H2O% concentration is below the calculated warning threshold, the WDS software will display a “Green Condition” indicating that the statistical probability of excessive amounts of water in the EAF is low. When the measured H2O% exceeds the calculated warning threshold, the WDS will enter in a warning state, or “Amber Alert”, meaning that the situation will be monitored closely for a time period defined together with the plant personnel. In Amber Alert situations, operators are strongly advised monitor the information of the WDS, which is implementing a specific algorithm to identify if the EAF conditions are persistent in time or simply transitory. According to the implemented algorithm, when the metrics exceed their limits for a prescribed period of time, a “Red Alert” is issued indicating that the off-gas chemistry is significantly out of the statistically normal range and that there is a high probability of excess water in the EAF. Red Alerts require immediate protective action by EAF operating staff. The above alarming scheme is designed and tuned to maximize the probability of real leak detection for leaks above a
64
minimum flow rate (L/min), which depends on the single EAF and process operations. A false alarm is defined when a “Red Alert” is triggered but no real leak is identified. That situation can happen when an excess of H2O vapour is produced in the furnace and cannot be explained by the information available to the WDS. When properly tuned, the WDS is designed to target a false alarm rate around 1% of the analyzed heats ( 1 “Red Alert” / 100 normal heats ). (c) Results during normal operations – 1 month data Significant effort has been dedicated by Tenova Goodfellow to achieve high detection capabilities and low false alarm rate, meaning high WDS effectiveness and reliability. The WDS model architecture has been improved by exploiting the use of two independent modules and combining the estimation of H2O% in normal conditions with the monitoring of measured H2O% behavior. The synergy between the two algorithms advanced the WDS capabilities in detecting out-of-ordinary situations against the intrinsic variability of H2O vapour among different heats, related to process and scrap conditions.
La Metallurgia Italiana - n. 7/8 2018
Attualità industriale
Fig. 6 - Example of WDS response to a real leak from a burner entrance housing. On the left the WDS response, on the right is a picture of the holes found at the end of the heat. From the WDS response it is clearly visible when the leak started (end of second bucket melting) and it is evident an unusual excess of measured humidity, clearly above the WDS warning threshold for most of the remaining power-on time.
Figure 8 shows the average measured H2O% (blue line) for each single heat during one month of operation after the commissioning of the system. As clear from the graph, the average measured H2O% exhibits an important inter-heat variation, which needs to be properly accounted in the WDS model. To verify the accuracy of the WDS estimation, the difference ( delta ) between the average measured H2O% and the average estimated H2O% by the WDS is shown in green. The green line exhibits a controlled and stable behavior compared to the blue line, proving that the estimated H2O% follows the variations caused by the EAF process and the furnace conditions with good accuracy. The orange line and red line respectively represent the “Amber Alerts” and “Red Alerts” generated by the WDS alarming algorithm during the month of observation: for a total of 407 heats, 6 warning conditions are reported, only 1 turned into a “Red Alert” while the others returned to normal within the “Amber Alert” monitoring time period. In this case, the 1 “Red Alert” did not correspond to a confirmed water leak and has to be considered a false alarm. It is important to note that the WDS does not rely on simple H2O% measurements, which by itself would produce a high number of false alarms, due to the clear variation of the blue line in Figure 7. In addition, in order to maximize La Metallurgia Italiana - n. 7/8 2018
the probability of detection, the WDS needs to be highly sensitive and consequently more exposed to false alarm when measurements vary in an unpredicted way. In this scenario, the challenge is to maximize the detection capabilities (being able to alarm for small leaks) maintaining a minimum false alarm rate. This objective is accomplished through the estimation and detection of normal and “out-of-ordinary” H2O% measurements, resulting in a low (but not null) probability of false alarms. It is also important to note that a minimum false alarm rate is showing that the system working-point is at the right threshold to guarantee the maximum detection: it would be easy to be conservative ( no false alarms ) by limiting the detection capabilities and increasing the risk of no detection. Tenova’s aim is to achieve the minimum false alarm rate while guaranteeing the maximum detection capabilities of the system (the ability to correctly alarm for leaks estimated > 40 L/min = 10.5 gal/min). Table 1 shows the statistics for the considered month of operations. Important numbers are the active-time of the WDS model during one heat, which covers on average 93% of the EAF power on in melting and refining. The resulting False Alarm rate is significantly smaller than the target 1%.
65
Industry news
Fig. 7 - WDS response to a real leak, first on-line preventive alarm after the WDS commissioning. The upper part of the graph shows the time evolution of the heat when the leak happened and the re-start in normal conditions after the leak was fixed. The Red Alert generated by the WDS system is show in the lower graph (zoom-in). During the melting of the first bucket, the measured H2O% (blue line) was significantly higher than the calculated warning threshold (orange line). The alarming sequence took place correctly, the EAF was stopped soon after the Red Alert was triggered and the leak was confirmed. CONCLUSIONS When liquid water enters the EAF it will immediately begin to boil producing steam (H2O vapor) which in turn reacts with Fe, FeO, CO2, CO and C to produce varying amounts of H2 & H2O vapor. A serious safety situation will result if liquid water becomes entrapped below molten steel due to a slosh of the bath. Such a situation can often result in two explosions, the first explosion related to a sudden evolution of steam trapped subsurface which ejects molten metal and slag, and a possi66
ble second much more severe explosion due to rapid ignition of combustible H2 & CO gas present inside the EAF. This paper outlines the four critical success factors for effective off-gas based EAF water detection; a) System reliability; b) System response time; c) System capability to analyze both H2 and H2O vapor; d) System capability to provide â&#x20AC;&#x153;Operator Alertsâ&#x20AC;? with minimal false alarms; La Metallurgia Italiana - n. 7/8 2018
Attualità industriale
Fig. 8 - WDS results of 1 month of normal operations, each point is the average measured H2O% over one heat (blue line). The green line is the Delta = Average measured H2O% - Average estimated H2O%. Over 407 heats, 6 Amber alerts were produced, 1 of them turned into a Red Alert (false alarm). Tab. 1 - WDS active time and alarms during 1 month of normal operations.
Number of heats completed (1 month)on Number of heats with WDS active WDS active time % during heat Number of False Alarms reported Off-gas technology can be divided into two categories; Insitu laser systems which transmit a beam in the near IR range through the off-gas for subsequent pick-up by an optical detector. For EAF applications, insitu laser systems require up to 3 separate lasers, one laser for CO2 and H2O vapor, one laser for CO and one laser for O2. Insitu laser technology is not technically able to analyze H2. By comparison, Extractive (EFSOP®) & the newly launched hybrid NextGen® systems use a water cooled probe to continuously extract and analyze a sample of EAF off-gas from the fume duct. Various analytical methods are employed to continuously analyze a complete spectrum of off-gas chemistry in real-time including CO, CO2, O2, H2 and H2O vapor. The products of an EAF water leak are H2O vapor and H2. Since H2O vapor - H2 interactions inside the EAF are quite complex and dynamic, it is simply not possible to predict the relative proportion of H2 and H2O vapor in advance of a leak. As exLa Metallurgia Italiana - n. 7/8 2018
407 395 (97%) Average 93% ( min 85% - max 96%) 1 (<< 1% target) plained in this paper, the H2O vapor to H2 ratio in the freeboard off-gas inside the EAF is continuously changing with H2 being favored under reducing EAF conditions and H2O vapor being favored whenever the EAF freeboard is oxidizing. Hence, effective off-gas based water detection requires real-time, continuous and reliable analysis of both H2O vapor and H2. Tenova Goodfellow EFSOP® extractive & NextGen® hybrid extractive/laser technologies both provide continuous reliable extraction and fast analytical response times of not only CO, CO2 and O2 but most importantly, of both H2 and H2O vapor. As such, the WDS technology here described is now effective in both reducing and oxidizing EAF operations. Significant technological advancements have been achieved for the development of Tenova’s WDS. The development of a model for the EAF process H2O estimation and measured H2O% monitoring, together with a specific alarming algorithm, have improved the capabilities of the WDS system in 67
Industry news detecting out-of-normal conditions in the EAF. The system achieved the necessary reliability and accuracy to be attractive to EAF customers and to effectively being implemented in typical steelmaking operations. As a result, the WDS system was successfully installed in the 60 tons EAF at Lucchini RS, Lovere, Italy, demonstrating significant results since the commissioning in August 2016, both in terms of reliable information and in the real-time detection and alarming of several actual water leaks, thereby avoiding potentially dangerous situations with less than a 1% false alarm rate. This paper presented and discussed the results obtained in the EAF of Lucchini RS, Lovere, Italy both during normal ope-
rations and when real water leaks were detected. Concerning a real water leak situation, the WDS demonstrated to be an effective technology to stop the EAF process soon after the leak started, thereby minimizing the risk of continuing the operations in hazardous conditions. ACKNOWLEDGMENTS Tenova Goodfellow would like to acknowledge Lucchini RS operators and personnel for their great support during the tuning and commissioning phases of the WDS, both regarding the WDS tuning activities and the regular and extraordinary maintenance of the EFSOP system.
REFERENCES 1]
S. Schilt, F.K. Tittel and K.P. Petrov, “Diode Laser Spectroscopic Monitoring of Trace Gases”, Encyclopedia of Analytical Chemistry, pages 1-29, 2011.
2]
S.C. Jepson, US Patent 6748004.
3]
H.D. Goodfellow, “Dynamic Process Control and Optimization for EAF Steelmakers”, MPT International, Nov. 2006.
4]
D.J. Zuliani, H.D. Goodfellow and M. Bianchi Ferri, “EFSOP Holistic Optimization of Electric Arc Furnaces – Past, Present and Future”, 9th European Electric Steelmaking Conference, May 19-21 2008, Krakow, Poland
5]
D. Vensel and M. Khan, “EAF Performance Improvement at Nucor Steel Auburn using Goodfellow EFSOP”, AISTech Conference Proceedings, 2006.
6]
M. Missio, N. Boin, and M. Khan, “Optimization Results at Ferriere Nord using EFSOP Technology, AISTech Conference Proceedings, 2010.
7]
M. Khan. S. Mistry, V. Scipolo and S. Waterfall, “Next Generation EAF Optimization at ArcelorMittal Dofasco Inc.”, AISTech Conference Proceedings, 2013.
8.
W.A. Von Drasek, K.A Mulderink and O. Marin, US Patent 6943886.
9]
D.J. Zuliani, V. Scipolo, M. Khan, O. Negru and W. Bilski, “Real-time Water detection in EAF Steelmaking”, Iron & Steel Technology, January, 2014, pages 84-95.
11]
D.J. Zuliani, “NextGen® Hybrid Multipoint Off-Gas Analysis Technology for Harsh Industrial Applications” Steel Times International, March 2017.
12]
A. Spencer, A. Pal, D. J. Zuliani, I. Todorovic and V. Scipolo, “NextGen® Multipoint Off-Gas Analysis at Steel Dynamics Inc., Butler, IN”, AISTech Conference Proceedings, 2017.
13]
H. Iyer, B. Babaei and V. Scipolo, “EAF Optimization using real-time Heat and Mass Balances at Nucor Steel Seattle”, AISTech Proceedings, 2017
68
La Metallurgia Italiana - n. 7/8 2018
Scenari Strong Potential of Commercialized High Mn Steel Products and Process for Various Applications edited by: Joo Choi POSCO technical research laboratories, Republic of Korea The future of global industries is likely to require much more energy, and the risk of climate change seems to be higher. This leads to the development of new steel with higher strength and toughness with enhanced ductility to cope with the greenhouse gas emissions in the manufacturing industry. High Mn austenite steels have been strong candidates to meet the requirement from manufacturing industries. In terms of economy and metallurgy, Mn is one of the most attractive and cheap elements to obtain austenite phase with a high enthalpy energy. Using physical and magnetic properties of austenite, high Mn steels can be widely used, which are high workhardening, low temperature toughness, low magnetic permeability and high damping capacity. These mentioned properties enable high Mn steels to be commercialized for the automotive application, wear-resistance and cryogenic service for energy industries, and anti-vibration equipment. POSCO has commercialized various kinds of high manganese steel products, followed by TWIP steels for automotive applications that are the world first commercialization for auto parts. In this paper, several applications of high manganese plates and sheet steels are illustrated. It is emphasized that there will be the strong potential of high manganese steels for various applications in the future because the process technology has been successfully developed.
KEYWORDS: HIGH MN STEEL – TWIP STEEL – POSLM PROCESS – POCAST TECHNOLOGY
INTRODUCTION Solving the problem of energy exhaustion and environmental pollution is the most important value that humankind desires. This leads to the development of new steel with higher strength and
toughness with enhanced ductility. In addition, as more and more activities are being carried out in marginal areas such as polar regions for new energy discovery, the development of new materials that can be used in extreme areas
is indispensable to solve the welfare of mankind. In response to this demand, POSCO has developed TWIP (Twinning-Induced Plasticity) steels, which
Fig. 1 - Schematic diagram of High Mn steel making process in POSCO La Metallurgia Italiana - n. 7/8 2018
69
Expertsâ&#x20AC;&#x2122; corner are widely used as automotive steel, wear-resistant steel for pipe to move oil sands, non-magnetic steel for heavy electric equipment and Cryogenic steel for LNG tanks and pipes. Figure 1 shows high Mn steels developed at POSCO according to C and Mn content (1). In addition, we developed high-efficiency, high-function high-manganese steelmaking technology and are working on the development of new products necessary for future industrial transformation. In this paper, several applications of high manganese plates and sheet steels are illustrated. It is emphasized that there will be the strong potential of high manganese steels for various applications in the future because the process technology has been successfully developed. HIGH MANGANESE STEEL MAKING PROCESS PosLM process In the past, high Mn steel was produced by melting of solid state manganese alloy in steelmaking process. However, it was difficult to control the temperature and the composition in the above pro-
cess because a large amount of manganese alloy must be added. In addition, the yield rate of manganese alloy was low and productivity was decreased. In order to solve these problems, POSCO developed PosLM (POSCO Liquid Manganese) process, a low cost mass production process. Fig. 2 and 3 show schematic diagram of high Mn steel making process. First, FeMn (Ferromanganese) with low impurity contents is prepared and stored in a furnace holding FeMn with molten state. Then, the liquid FeMn is mixed with molten steel to control the manganese content while manufacturing high Mn steel. In the POSLM process, the high purity FeMn is produced by the following process. The first, a raw material is put into the submerged arc furnace, and it is become the high carbon FeMn molten metal through the smelting and reduction process by the electric resistance heat. The second, in the FeMn dephosphorization stirring furnace, flux is put into the high carbon FeMn molten metal, and it is mixed by impeller rotation. And then the phosphorus component is removed
by the reaction between flux and molten metal. The produced low phosphorus FeMn is transferred to the decarburizing refining furnace. Finally, in the FeMn decarburizing refining furnace, high purity FeMn is produced after the removal of carbon using a bottom stirring type refining technology. The produced high purity FeMn is supplied to the holding furnace in a molten state. The FeMn holding furnace is built with self-developed refractory for FeMn and it is possible to control the temperature of the molten metal by induction heating. And, it is possible to add other ferro-alloys and control nitrogen of the molten metal in the FeMn holding furnace. Therefore, it is easy to store the FeMn with molten state and holding furnace serves as a buffer station in secondary refining. As a result, the productivity of high Mn steel was improved through the PosLM process. The continuous casting rate was increased and the manufacturing cost was reduced due to an increase in the yield rate of alloy and molten steel.
Fig. 2 - Schematic diagram of High Mn steel making process in POSCO
70
La Metallurgia Italiana - n. 7/8 2018
Scenari
Fig. 3 - Schematic diagram of PosLM process Pilot caster operation for development of high manganese steel In order to develop advanced technologies and commercialize new products, many trials are necessary to develop new facility and find operating conditions. Because troubles such as stand still and breakout are necessarily accompanied during development of advanced technologies, POSCO pilot caster has been operated since 2003. The accumulated casting number becomes more
than 768 in 2017, and about 50 casts are tried on average each year. During TWIP steel casting, the risk of accidents such as breakout is high due to changes in the physical properties of the mold slag, and it is difficult to produce high quality slab. By using pilot caster, we were able to minimize damage, so we could develop POCAST (POSCO Advanced CASting Technology) technology which can produce high quality slabs. In addition, optimization of casting condi-
tion of high Mn steels was carried out using pilot caster. It has been found to be very effective to use the pilot caster to find the optimal operating conditions for high manganese steel. Because operation and maintenance practices of pilot caster are nearly same as commercial caster, the advanced technologies tested at this caster can be easily applied to actual caster.
Fig. 4 - Pilot caster in POSCO La Metallurgia Italiana - n. 7/8 2018
71
Expertsâ&#x20AC;&#x2122; corner Tab. 1 - Main specification pilot caster Item
Specification
Cast type
VB type, Casting radius 5 m, ML : 19.5 m
Slab thickness
100 / 140 mm
Slab width
600~1,000 mm
Cast speed
0.8~2.5, 5~12 m/min
Ladle capacity
10 / 23 tons
Tundish capacity
5 tons (Stopper control)
Mould type
Parallel / Funnel
Oscillator
Hydraulic type (Non-sinusoidal)
Segment
Hydraulic roll gap control,Air-mist cooling, Dynamic soft reduction
POCAST Technology POSCO is developing a lightweight advanced high strength steel which contains high aluminum contents in composition. In particular, TWIP steel has drawn a lot of attention. However, it has been turned out that high aluminum in steel would bring a difficulty in continuous casting procedures. It is caused by the increase of Al2O3 in the mold slag during casting and this invokes changes in the
physical properties of the slag followed by a bad castability. We have developed the POCAST (POSCO Advanced CASting Technology) process in order to reduce the reactivity between aluminum containing steel and the mold slag. Fig. 5 illustrates the basic concept of the POCAST. The molten mold flux is fed into the molten steel for a little drop in temperature of molten steel. Moreover, a slag bear made by POCAST is much smaller than
that of conventional powder casting, which leads to a wide channel between a mold wall and a strand. This mild cooling feature at the initial solidification of shell is helpful to prevent from longitudinal facial crack. High Al steel is well known to react strongly with a lime-silica based mold flux as follows (2).
3(SiO2) + 4[Al] = 3[Si] + 2(Al2O3) As shown in Fig. 6, Al2O3 pickup in slag was over 30 % in weight when a steel containing 1.5 % aluminum was cast with a conventional mold powder. On
the other hand, by molten mold flux casting with POCAST, Al2O3 pickup in slag was below 5 % at the first and gradually increased. It implies little change in the
physical properties of the mold slag than those of conventional powder casting and enables to cast high Al steel.
Fig. 5 - Concept of molten mold flux feeding effect of POCAST process for Al2O3 pick-up 72
La Metallurgia Italiana - n. 7/8 2018
Scenari
Fig. 6 - Al2O3 contents of mold powder and POSCAST according to casting length PosSCAN Technology In order to produce high quality slabs, we have developed the technology, named PosSCAN (POSCO Simultaneous Controller & Analyzer for Non-detective mold flow), for reducing the defects caused by the mold powder entrapment. The mold powder entrapment occurs by
several reasons. The main reason is the vortex flow due to the serious bias flow induced by the alumina-clogging in the submerged entry nozzle. PosSCAN is the technology based on the real time visualization system for molten steel flow at meniscus in the mold as shown in Fig.7. This system could detect the occurren-
ces of biased flows. When biased flow is detected, the system controls the mold flow using the electro-magnetic stirrer and orders the surface scarfing process. We have reduced over 30% of the inclusion defects using this real time visualization system.
Fig. 7 - Real time visualization of mold flow. La Metallurgia Italiana - n. 7/8 2018
73
Expertsâ&#x20AC;&#x2122; corner HIGH MANGANESE STEEL PRODUCTS TWIP steel for automotive applications POSCO is developing various automotive steel products with advanced performance and the recent developing trends including Ultra-AHSS (U-AHSS) and eXtra-AHSS (X-AHSS) is summarized as shown in Fig. 8 (3) TWIP steel of the new alloy system with reduced Mn content has been developed in POSCO (4-7). The microstructure to obtain the excellent mechanical properties was also designed to ensure 1)
austenite single phase with no martensite at room temperature, 2) appropriate microstructural stability, i.e., that SFE (Stacking Fault Energy) is controlled properly for twinning, 3) no carbides after continuous annealing, and 4) no delayed fracture. The strength and elongation balance of TWIP steel shows the largest value among other steels as shown in Fig. 9. In addition, the HDF (Hydrogen Delayed Fracture) was measured by using cupping specimens with drawing ratio of 2.0 and followed by keeping and checking the time to form cracks at
room temperature. The results showed that the resistance to hydrogen delayed fracture was excellent. TWIP steel shows also excellent LDH (Limit Dome Height) of about 43 mm, which is superior to that of the EDDQ (Extra Deep Drawing Quality) grade low carbon steel. The Vbending result showed no crack at the zero radius of die. HER (Hole Expanding Ratio) was measured as 45~50%. TWIP steel has been well applied to the important parts of a car to protect passengers.
Fig. 8 - Relationship between tensile strength and elongation in various steel grades Wear-resistant steel for pipe to move oil sands New wear-resistant steel containing high Mn has been developed to meet the customer demand, which would be able to be applied for harsh environment such as energy fields, where it is very troublesome and dangerous to exchange the pipes. High Mn steel shows extraordinary abrasion resistance characteristics due to its high strength and high work hardening ability. The work hardening ability is the property that the material becomes harder when deformation is applied, and can be maximized by controlling the carbon-manganese content in steels. That is, the hardness becomes
74
very high after deformation although the hardness of the original material is not quite high. Due to this property, a hardened layer is formed on the surface of the material when it is contacted with the actual wear particles. On the other hand, high Mn wear-resistant steel has not only abrasion resistance but also high toughness because the inside of the steel retains soft. Therefore, it is a material that can be applied to a severe wear environment requiring both properties simultaneously. The wear resistance of high Mn steel is about six times higher than that of carbon steel. In addition, it shows about twice anti-corrosion performance than
that of Hadfield steel in erosion and corrosion environment which is similar to actual slurry transport environment. In particular, slurry pipe application requires not only wear resistance but also piping performance and weldability, and POSCO succeeded in developing and commercializing these technologies. The field test results of the customer for slurry pipe applications showed that the erosion resistance was superior to that of API X70 steel pipes, and it was confirmed that it is possible to secure equal or better performance in comparison with special steel pipe like CCO (Cr Carbide Overlay). These pipes are manufactured in a special way to ensure high erosion
La Metallurgia Italiana - n. 7/8 2018
Scenari and corrosion resistance. Hence, this type of pipe is very expensive although it has excellent wear resistance. In this regard, the economic value of high Mn wear-resistant steel is even greater. POSCO has been conducting various joint
evaluations with several customers in order to apply the superior performance of high Mn wear-resistant steel to various industries. In addition to the above slurry pipes for oil sand transportation, high Mn wear-resistant steel is expected
to be widely applied in fields such as various industrial machinery and structural material parts requiring extreme wear resistance.
Fig. 9 - High Mn steel for wear-resistant applications and comparison of wear resistance (6) Non-magnetic steel for heavy electric equipment POSCO has expanded the application of high Mn steel by developing high Mn non-magnetic steel. Materials for nonmagnetic applications should have a permeability of 1.05 or less and STS304 is typically used. High Mn steel also can be used as excellent non-magnetic ma-
terial because it forms a stable austenite phase with excellent non-magnetic property at room temperature according to the contents of C and Mn. Since high Mn non-magnetic steel has higher strength than STS304, it is possible to maintain the strength even if the usage of material is reduced during the design of the structure. As a result, the overall cost is
reduced and thus the potential is higher in terms of economy. While STS304 increases the permeability by phase transformation after deformation, high Mn non-magnetic steel maintains a low permeability because the austenite phase is stable during deformation.
Fig. 10 - High Mn steel for non-magnetic applications
La Metallurgia Italiana - n. 7/8 2018
75
Expertsâ&#x20AC;&#x2122; corner Cryogenic steel for LNG tanks and pipes With increasing interest in the environment, the demand for LNG fuels has increased significantly. Meanwhile, safe mining, transport, and storage of LNG are challenges for the industry. The required properties of cryogenic materials used for LNG tanks are excellent yield strength at room temperature and high impact toughness at cryogenic temperature (-196°C). Typical ferritic steels have properties that suddenly change into brittle materials as the temperature drops. This temperature is called as DBTT (Ductile to Brittle Transition Temperature). However, high Mn steel maintains a stable austenite phase even at cryoge-
nic temperatures, which shows excellent impact toughness at cryogenic temperatures due to the absence of DBTT despite the drop in temperature. The yield strength is the stress at which the material exhibits a certain permanent deformation, and high Mn steel shows high yield strength at room temperature by adding a specific alloy. Cryogenic materials are currently being commercialized such as 9Ni steel, STS304, Al alloy, INVAR alloy, and etc. POSCO has developed a high Mn steel for cryogenic service with low manufacturing cost and excellent toughness at extremely low temperatures. In particular, 9Ni steel that is a typical cryogenic material contains a large amount of Ni
to ensure toughness. On the other hand, high Mn steel has excellent austenite stability by using Mn which is lower price than Ni. In addition, we have developed welding technologies suitable for high Mn steel to provide total solutions for products and welding. In Feb 2018, the new cryogenic highmanganese steel developed by POSCO has gone into service for the first time in a 50,000 DWT LNG bunker tank on a bulk carrier. This ship transport limestone used by POSCO from Gangwon-do in Korea to Gwangyang Steel Works (8). It is also equipped with an eco-friendly and highly efficient dual-fuel engine that can use bunker coil and LNG.
Fig. 11 - High Mn steel for cryogenic applications
CONCLUSIONS The future global industries are likely to need much more energy even though energy crisis will be expected, and the risk of climate change seems to be handled more importantly. Many governments with large markets are strengthening the environmental regulation and emphasizing energy efficiency. Steel producing companies have developed and suggested new materials to 76
cope with the tremendous need of global industries. Since Mn is most effective element in terms of usefulness and cost, high Mn steel can show variety of superior material properties. POSCO has developed the innovative process technologies: PosLM for steel making process to manufacture competitive high Mn steel products and POCAST to produce TWIP steel. Furthermore, POSCO has commercialized
high manganese steel sheets and plates in many fields: automotive applications, wear-resistance service using high work-hardening rate, cryogenic service with excellent toughness, non-magnetic service with low magnetic permeability, and anti-vibration equipment with excellent damping capacity.
La Metallurgia Italiana - n. 7/8 2018
Scenari REFERENCES 1]
J. Choi: Commercialization of New Advanced High Mn Steels and their Manufacturing Technologies in Asia Steel 2018, Bhubhaneshwar, India (2018).
2]
Cho, J. W., Yoo, S., Park, M. S., Park, J. K., & Moon, K. H. (2017). Improvement of Castability and Surface Quality of Continuously Cast TWIP Slabs by Molten Mold Flux Feeding Technology. Metallurgical and Materials Transactions B, 48(1), 187-196.
3]
Ohjoon Kwon, K. Y. Lee, Gyosung Kim and Kwang-Geun Chin; Mat. Sci. Forum, 638 (2010), 136.
4]
Kayoung Choi, Chang-Hyo Seo, H. Lee, Sungkyu Kim, Jai Hyun Kwak, Kwang-Geun Chin, Kyung-Tae Park and Nack J. Kim; Scripta Materialia, 63 (2010), 1028.
5]
S. K. Kim, Jung-wook Cho, Woo-jin Kwak, Gyosung Kim and Ohjoon Kwon; in METEC2007 edited by Dieter Ameling, DĂźsseldorf, Germany (2007).
6]
S. K. Kim, Gyosung Kim and Kwang-geun Chin; in International Conference on New Developments in Advanced High-Strength Sheet Steels edited by J. G. Speer, Orlando, USA (2008).
7]
D. Fairchild; 2016 POSCO Global EVI Forum, Incheon, Korea, Oct. 31 (2016).
8]
http://www.lngworldshipping.com/news/view,posco-highmanganese-steel-debuts-in-lng-bunker-tank_50885.htm
La Metallurgia Italiana - n. 7/8 2018
77
Aim news Calendario degli eventi internazionali International events calendar
QUOTE SOCIALI AIM 2018 (ANNO SOLARE) Benemeriti (quota minima) 1.750,00 €
August 13-15, Austin, USA Annual International Solid Freeform Fabrication Symposium - An Additive Manufacturing Conference August 16-18, Qingdao, China 7th International Conference on Modelling and Simulation of Metallurgical Processes in Steelmaking (SteelSim) Semptember 5-8, Las Vegas, USA, MEI2018 (Mining Expo International)
Sostenitori (quota minima)
750,00 €
Ordinari (solo persona)
70,00 €
Seniores
25,00 €
Juniores
15,00 €
La quota dà diritto di ricevere la rivista dell’Associazione, La Metallurgia Italiana (distribuita
September 9-13, Oxford, United Kingdom, Eurosuperalloys 2018
in formato digitale). Ai Soci viene
September 11-14, Graz, Austria 15th International Forgemaster Meeting (IFM)
partecipazione alle manifestazioni
riservato un prezzo speciale per la AIM e per l’acquisto delle
September 12-14, Xi’ An, China, 25TH IFHTSE
pubblicazioni edite da AIM.
September 9-13, Oxford, United Kingdom, Metallurgie im Wandel 4.0
Per ulteriori informazioni, iscrizioni, rinnovi:
September 26-28, Darmstadt, Germany, Materials Science and Engineering October 14-18, Seattle, USA, Furnace Tapping 2018 Conference October 16-19, Stockholm, Sweden, 3rd Ingot Casting, Rolling and Forging Conference, ICRF 2018 October 8-12, Lillafüred, Hungary, The 5th International Conference on Competitive Materials and Technology Processes
AIM, Via F. Turati 8 20121 Milano Tel.: 02 76021132/76397770, fax: 02 76020551 e-mail: amm.aim@aimnet.it www.aimnet.it
October 10-12, Taranto, Italy, 8th European Oxygen Steelmaking Conference (EOSC 2018) November 28-29, Bergamo, Italy, 4th European Conference on Clean Tech Technologies in the Steel Indutry (CLEAN TECH 4)
78
La Metallurgia Italiana - n. 7/8 2018
Le Rubriche - Centri di studio Attività dei Comitati Tecnici CT METALLI LEGGERI (ML) (riunione del 18 maggio 2018) Manifestazioni in corso di organizzazione - Il convegno “Materiali metallici e processi produttivi innovativi per l’aerospazio”, organizzato insieme ad altri Centri di Studio AIM, si terrà a Napoli nei giorni 19 e 20 luglio. Il coordinatore Monetta rivede e commenta il programma finale per quanto riguarda gli aspetti tecnici, produttivi e commerciali. - Una sessione sulle leghe leggere si svolgerà durante il 37° Convegno Nazionale AIM di Bologna nella giornata del 13 settembre. Iniziative future - Il presidente Grillo suggerisce di riprogrammare alcune delle manifestazioni che, in passato, hanno riscosso positivo interesse da parte del mondo industriale quali la GdS “Igiene delle leghe di alluminio”, già effettuata nel 2015, e la GdS “Leghe di alluminio”, già effettuata nel 2017. - Il CT propone anche organizzare una manifestazione sul tema della saldatura delle leghe leggere e una sulla laminazione. CT MATERIALI PER L’ENERGIA (ME) (riunione dell’8 giugno 2018) Iniziative future - Corso sul Creep: Bassani richiede che il corso sia tenuto entro fine 2018 per rispettare la cadenza triennale e il calendario fornito a Fondimpresa, sulla base del quale le aziende possono ottenere finanziamenti per la partecipazione alle iniziative di formazione AIM. Viene esaminata la locandina del corso precedente e si decide di riproporre la medesima formula, con 2 giorni di corso di base e 1 giorno di corso avanzato, durante il quale si tratterà anche l’additive manufacturing e le interazioni del creep con la fatica. Si ipotizzano le date del 23-24 e 30 ottobre 2018, salvo conferma della disponibilità dei docenti. Stato dell’arte e notizie - Il presidente Gavelli informa che al 37° Convegno Nazionale AIM di Bologna ci sarà una sessione interamente dedicata ai materiali per l’energia, con memorie su argomenti tradizionali (invecchiamento dei materiali, rafforzamento acciai) e più innovativi (materiali per batterie). - Arianna Gotti, che è stata nominata coordinatrice del Gruppo di Lavoro Creep e ECCC, segnala che il gruppo sta svolgendo le sue attività in modo molto proficuo, pur lavorando su base volontaria. Si stanno effettuando prove su un giunto saldato e ci si concentrerà poi su nuovi tipi di acciaio, compresi gli inox martensitici e austenitici. - La GdS “Leghe di Nickel e Superleghe” La Metallurgia Italiana - n. 7/8 2018
viene rimandata all’inizio del 2019 e comprenderà anche le leghe di cobalto. - Resta viva ma non ancora definita l’idea di una GdS su “Materiali per l’eolico” Stato dell’arte e notizie -Il CT “Metallurgia delle Polveri” ha modificato il proprio nome in “Metallurgia delle Polveri e Tecnologie Additive” in quanto queste ultime sono a tutti gli effetti classificabili come tecnologie di polveri, in rapida evoluzione, e non esiste ancora un CT in AIM che se ne interessi specificamente. -Due nuovi membri, presenti alla riunione come ospiti, sono stati accettati ed entreranno a far parte del CT. CT METALLI E TECNOLOGIE APPLICATIVE (MTA) (riunione del 12 giugno 2018) Iniziative future - Si discute della possibile riedizione di una GdS sul titanio: il presidente Debernardi consiglia una parte introduttiva sul titanio e sulle sue leghe, sulle principali caratteristiche meccaniche e microstrutture ed eventualmente sulla tribologia e sull’additive manufacturing; inoltre ritiene utile lasciare ampio spazio ai produttori e ai distributori perché potrebbero interessare molto gli utilizzatori. - Il CT Leghe Leggere ha proposto la co-organizzazione di una giornata sulla saldatura delle leghe leggere, cui il CT MTA parteciperà volentieri. Stato dell’arte e notizie - Le cariche all’interno del CT MTA, in assenza di nuova candidature, vengono rinnovate per un altro anno: presidente Debernardi, vicepresidente Varalda, segretario Gerosa. - Vista la scarsa partecipazione alle riunioni, i membri del CT si stanno attivando per trovare nuovi membri interessati alle attività del Centro. CT CONTROLLO E CARATTERIZZAZIONE PRODOTTI (CCP) (riunione del 29 giugno 2018) Manifestazioni in corso di organizzazione - Nel maggio 2019 si organizzerà la undicesima edizione del corso di Prove Meccaniche: la durata tornerà ad essere di 4 giorni (2+2 giorni in settimane diverse) perché i 3 giorni della precedente edizione non offrono abbastanza tempo per discussioni e approfondimenti. La sede sarà a Milano, salvo una possibile giornata presso un laboratorio di prove. Il coordinatore Trentini chiede di contattare ancora gli sponsor, già presenti
nelle precedenti edizioni, perché danno un importante contributo alla conoscenza. - Nell’autunno del 2019 si prevede di organizzare la quarta edizione del corso “Prove non Distruttive”; i coordinatori Trentini e M. Cusolito hanno l’incarico di definire i programmi, la sede e gli eventuali sponsor. Vista la insoddisfacente partecipazione alla precedente edizione del corso, Trentini preparerà una lettera circolare per gli altri CT per raccogliere idee e sollecitare eventuali partecipazioni. La lunga discussione ha poi permesso di evidenziare ulteriori aspetti delle PnD che potrebbero essere inclusi nel programma del corso. - IL CT CCP offrirà la propria collaborazione al CT Metallurgia Fisica e Scienza dei Materiali per organizzare una nuova edizione del corso di microscopia elettronica, già tenuto nel 2017 con notevole interesse. - Si è discusso di una possibile giornata su “Interlaboratori ed accreditamento”, possibilmente in collaborazione con ALPI: Trentini preparerà un elenco di prove che potrebbero essere di interesse. Nella prossima riunione si raccoglieranno le idee pervenute e si nominerà un coordinatore se si pensa di poter sviluppare l’argomento. Stato dell’arte e notizie - Il presidente Trentini porge il benvenuto ai due nuovi membri del CT.
- Il vicepresidente Toldo ha partecipato alla riunione dei Presidenti di Centro: la presidenza AIM sollecita una maggiore partecipazione alle riunioni e l’organizzazione di più attività. Per questo motivo la data delle riunioni sarà fissata tramite un doodle per aumentare il numero dei presenti, e parimenti si svilupperà la possibilità di partecipare alle riunioni via Skype; inoltre si cercheranno azioni condivise con altri CT per incrementare il numero degli eventi organizzati. CT AIM/ASSOFOND – FONDERIA (F) (riunione del 05 luglio 2018) Manifestazioni in corso di organizzazione - La riunione è stata interamente dedicata alle sessioni tecniche del XXXIV Congresso di Fonderia che si svolgeranno nei giorni 15 e 16 novembre 2018 a S. Eufemia – Brescia presso il Museo Mille Miglia. Il presidente Caironi sottopone all’analisi dei presenti le candidature ricevute per presentare una memoria durante le sessioni tecniche. Una buona parte delle candidature vengono valutate positivamente, alcune invece necessitano di chiarimenti per essere ammesse. Sono comunque attese ulteriori memorie per completare le sessioni previste. La sessione plenaria raccoglierà le presentazioni con tematiche trasversali di maggiore interesse, mentre le memorie specifiche sono state allocate alle sessioni più tecniche (Metallurgia non ferrosi, Metallurgia ferrosi, Processo metalli non ferrosi, Processo metalli ferrosi). 79
in evidenza
Corso
Gli acciai inossidabili X Edizione
17-18-24-25 ottobre . 7-8-14-15 novembre 2018 Milano Centro Congressi Fast Organizzato da
L’Associazione Italiana di Metallurgia propone una nuova edizione del Corso avanzato sugli acciai inossidabili, dedicato a tecnici, ricercatori, professionisti e in generale agli operatori italiani del settore. Questa iniziativa offre ai partecipanti l’opportunità di approfondire e aggiornare le proprie conoscenze ed in particolare consente alle aziende del settore di programmare un’adeguata azione di formazione e aggiornamento del proprio personale. La decima edizione del Corso si sviluppa su un unico modulo distribuito in 8 giornate. Il Corso tratta in modo esaustivo la metallurgia ed i trattamenti termici degli acciai inossidabili, le diverse famiglie e le relative proprietà, i processi produttivi, le lavorazioni ed i prodotti, la saldatura e la sinterizzazione. Ampio spazio sarò inoltre dedicato alle norme di riferimento e alle specifiche di acquisto, oltre che ai criteri di progettazione e alla scelta del materiale in funzione degli impieghi. L’ultima giornata vedrà protagonista il mercato e le applicazioni degli acciai inossidabili. Le lezioni, di carattere monografico, sono connesse tra loro in modo logicamente consequenziale, così da facilitare ai partecipanti l’apprendimento e l’approfondimento panoramico degli argomenti trattati. I docenti di estrazione industriale e in minor parte accademica sono in grado di fornire ai partecipanti, nelle lezioni e nei dibattiti che le seguono, la diretta testimonianza delle proprie esperienze professionali. Per favorire inoltre il contatto tra i partecipanti e le realtà del mercato degli inossidabili, è organizzata durante le prime due giornate del Corso (17 e 18 ottobre 2018) la presentazione di “tavoli informatori” approntati a cura di diverse aziende sponsor. Coordinatori del Corso: Mario Cusolito, Sandro Fraccia
#corso #formazione #acciai #inossidabili #inox #metallurgia #processi #prodotti
IL PROGRAMMA COMPLETO E TUTTE LE INFORMAZIONI SONO DISPONIBILI SUL SITO www.aimnet.it
8th European Oxygen Steelmaking Conference Taranto . Italy 10-12 October 2018 Organised by
2018
With the support of
Exhibition & Sponsorship opportunities As an integral element of the event, EOSC 2018 will feature a table top Exhibition that will enable excellent exposure for company products, technologies, innovative solutions or services. At this opportunity the Organizers will set an area strategically located. This area will be a focal point of the Conference, so that enough time will be available to guarantee a perfectly targeted potential customerâ&#x20AC;&#x2122;s environment.
Companies will be able to reinforce their participation and enhance their corporate identification by taking advantage of the benefits offered to them as Sponsor of the Conference. Companies interested in exhibiting and/or sponsoring the event may contact the Organising Secretariat: aim@aimnet.it
CONTACTS
AIM - Associazione Italiana di Metallurgia Via Filippo Turati 8, 20121 Milan - Italy Tel. +39 02 76021132 Fax +39 02 76020551 E-mail: aim@aimnet.it
www.aimnet.it/eosc2018