Wettability between Porous MgAl 2 O 4 Substrates and Molten Iron

Transcription

1 Materials Transactions, Vol. 50, No. 11 (2009) pp to 2556 #2009 The Japan Institute of Metals Wettability between Porous MgAl 2 O 4 Substrates and Molten Iron Naotaka Fukami*, Ryohei Wakamatsu*, Nobuya Shinozaki and Kyoko Wasai Department of Materials Science and Engineering, Faculty of Engineering, Kyushu Institute of Technology, Kitakyusyu , Japan The wettability between porous MgAl 2 O 4 substrates and molten iron was investigated by the sessile drop method at 1833 K as a basic study in order to elucidate the interactions between molten iron and a refractory material. For the study, MgAl 2 O 4 substrates with 2, 8, 13, 27, and 39% porosity were prepared. The contact angle increased with the substrate porosity. In substrates with 2%, 8%, and 13% porosity, the contact angles were independent of time. In contrast, for substrates with 27% and 39% porosity, the contact angle decreased rapidly during the first hour and then gradually reached a steady value. The decrease in contact angle with time was attributed to the interfacial free energy. The work of adhesion was 1.27 Jm 2, and it was suggested that the interfacial bond consists of not only a physical bond but also a chemical bond. [doi: /matertrans.m ] (Received June 17, 2009; Accepted August 27, 2009; Published October 25, 2009) Keywords: wettability, contact angle, molten iron, MgAl 2 O 4, porosity, interfacial free energy 1. Introduction Steel refining is characterized by many interfacial phenomena related to refractory materials, slag, and molten steel. The properties of these interfaces are very important from the viewpoint of elucidating the separation of slag from molten steel, the floating of nonmetallic inclusion, and the corrosion of a refractory material by slag or molten steel. It is wellknown that molten steel penetrates into the refractory through pores, thereby contributing to the corrosion of refractory. A longer refractory life can be expected if no molten steel penetrates the refractory. This penetration phenomenon is highly correlated to the wettability of the refractory, and many investigations have focused on the wettability between molten iron and solid oxides to clarify the mechanism of refractory corrosion. 1 4) These studies have focused chiefly on investigation of the influence of the composition of molten iron on the wettability between molten iron and solid oxides. Few studies have examined the influence of structure and composition of solid oxides, such as MgAl 2 O 4, on wettability. This study focuses on the effect of the structure of a solid oxide on wettability; in particular, it examines the influence of pores on the wettability between porous MgAl 2 O 4 substrates and molten iron. 2. Experimental 2.1 Sample preparation Cylindrical 1.0-g iron specimens were prepared from commercial electrolytic iron that included mass% oxygen. MgAl 2 O 4 (purity: mass%, average particle size: 0.2 mm) and graphite (purity: 99.9 mass%, average particle size: 20 mm) powders were used as starting materials for substrates. These powders were mixed in a ball mill with ethanol for 24 h, and after drying, were CIP formed at 150 MPa as disk-shaped specimens with diameters of 20 mm and thicknesses of 5 mm. The disk-shaped specimens were sintered at 1823 K for 3 h at a heating rate of 300 K h 1 in air and were then used as substrates. Substrates with 2, 8, 13, 27, *Graduate Student, Kyushu Institute of Technology Gas Outlet Water-cooling cap Thermocouple(outside) Lanthanum-chromite heater Alumina tube Camera Silica glass Fig. 1 Gas Inlet Alumina support Ceramic boats with titanium sponge Substrate Metal droplet Thermocouple(inside) Schematic view of experimental apparatus. and 39% porosity were prepared by controlling the amount of graphite powder used as a precursor. In this study, the porosity value corresponded to the apparent porosity measured by the Archimedes method. By X-ray diffraction, it was confirmed that the prepared substrate consisted entirely of a polycrystalline stabilized spinel. The surfaces of the substrates obtained by the above procedure were polished using No emery paper. 2.2 Experimental procedures The experiments were carried out by the sessile drop method. Figure 1 shows a schematic view of the experimental apparatus. The experimental procedure is as follows. A cylindrical iron specimen was loaded on a ceramic substrate and placed horizontally on an alumina support in an electric resistor furnace. The internal chamber of the furnace was evacuated and then backfilled with argon gas, and the temperature was raised to 1833 K at a heating rate of 300 K h 1 while simultaneously passing argon gas at a flow rate of 2: m 3 s 1. The argon gas was deoxidized

2 Wettability between Porous MgAl 2 O 4 Substrates and Molten Iron 2553 with a heated titanium sponge and metallic magnesium. Images of a drop of molten iron were taken when the chamber reached a temperature of 1833 K. Images were taken initially at intervals of 1 min for 10 min, then at intervals of 5 min until the end of the 3-h holding time. The contact angle was measured from the shape of droplet. The crystal structure of the substrate was analyzed by X-ray diffraction (Rigaku, RINT2000). The microstructure and composition of the substrate were examined using a scanning electron microscope (SEM, Hitachi, S-3000N) equipped with a energy dispersive X-ray spectrometer (EDAX, Genesis 4000). The oxygen content of the iron specimen was measured by an oxygen analyzer (LECO, TC-136). 3. Results and Discussion 3.1 Changes in contact angle with time Figure 2 shows the changes in the contact angle with time and the effect of porosity on the contact angle. The contact angle was high so that the porosity of substrate was high. In substrates with 27% and 39% porosity, the contact angle decreased rapidly during the first hour and then gradually reached a steady value. In contrast, the contact angles of the other substrates were independent of time. The manner in which the changes in the contact angles of substrates with 27% and 39% porosity were taken into consideration is given below. In general, to evaluate the wettability of a solid by a liquid, the contact angle () can be calculated in terms of the solid vapor, solid liquid, and liquid vapor interfacial free energies by using Young s equation: cos ¼ð SV SL Þ= LV ð1þ SV : solid{vapor interfacial free energy SL : solid{liquid interfacial free energy LV : liquid{vapor interfacial free energy In this study, the surface tension of molten iron is equivalent to LV. It is well known that surface-active elements influence the surface tension of molten iron. Many previous studies have focused, in particular, on the influence of oxygen as a surface-active element on the surface tension of % 8% 13% 27% 39% Holding time (h) Fig. 2 Change in contact angle with time and effect of porosity on contact angle. Table 1 Oxygen content of iron after experiment. Porosity (%) Oxygen content (mass%) molten iron. The oxygen content of iron after experiment ranged from mass% to mass%, regardless of the substrate porosity, as shown in Table 1. According to Takiuchi, 5) the surface tension of molten iron can be calculated using the experimental temperature and the oxygen content of iron; based on Takiuchi s calculation method, the surface tension of molten iron was determined to be 1.8 Jm 2. In addition, the surface free energy of ordinary oxides, SV, has been reported 6) to be about 1.0 Jm 2. Therefore, it could be assumed that the surface free energy of MgAl 2 O 4 is equal to 1.0 Jm 2. The contact angle [as calculated by eq. (1)] between molten iron and MgAl 2 O 4 substrate with 2% porosity is shown in Fig. 2 to be 108. Using these values, the interfacial free energy between molten iron and the MgAl 2 O 4 substrate was determined to be 1.56 Jm 2. The following relationship is established: LV > SL > SV. In this experiment, the contact angle was measured when the cylindrical iron specimen melted and formed a droplet on the substrate. After melting, molten iron spread on the substrate by its own weight until a three-phase interface composed of vapor, liquid, and solid that satisfies eq. (1) was formed. Then, according to the relationship SL > SV, molten iron spreads on the substrate while simultaneously penetrating through the pores of the substrate, for which additional surface free energy is required. Shinozaki 7) reported that it was difficult for molten manganese to penetrate through the pores of a substrate, although molten manganese wetted a dense MnAl 2 O 4 substrate. As indicated in Fig. 2, the contact angle of a substrate with 2% porosity is more than 90, and wetting does not occur. It is supposed that the molten iron would spread on the substrate while simultaneously penetrating through the pores. Because LV > SL, molten iron spreads over the pores to a greater degree than it spreads over the dense substrate. Therefore, it was surmised that it took much time until the shape of the droplets was stable because there were so many pores on the surface of substrate. Thus, it is now clear why the contact angle of the substrates with 27% and 39% porosity decreased in the first hour of the experiment. 3.2 Influence of pores on contact angle It is well known that the wettability of a solid oxide by a liquid is affected by the surface properties of the solid oxide. Figures 3 and 4 show the X-ray diffraction pattern of the substrate after experiment and SEM images of the substrate surface before and after the experiment. As indicated in Fig. 3, each substrate consisted of MgAl 2 O 4, and it was found that a mineral was not formed on the surface of the substrate by reaction with molten iron. As shown in Fig. 4, pores with sizes of several micrometers were scattered over the surface of the substrates with 2% and 8% porosity prior to the experiment. The surface of the substrate with 13% porosity exhibited larger pores, in the range mm, than

3 2554 N. Fukami, R. Wakamatsu, N. Shinozaki and K. Wasai % 27% 13% 8% 2% 20 Fig θ / Equation (2) 80 0 after experiment 2% 60 X-ray diffraction pattern of MgAl2 O4 substrate after experiment. before experiment Intensity (a.u.) Spinel [MgAl2O4] 8% 13% 27% 39% Fig. 4 SE images of each MgAl2 O4 substrate surface before and after experiment. Fig Porosity (%) 40 Relationship between contact angle and porosity. substrates with 2% and 8% porosity. In addition, in substrates with more than 27% porosity, assemblies of many pores greater than 20 mm in size were observed. The distribution and total amount of pores before and after the experiment are similar. The wettability of the surface of a porous material has been defined by Cassie and Baxter8) as follows: ð2þ cos C ¼ A1 cos 1 A2 A1 : surface area ratio of solid A2 : surface area ratio of pore C : apparent contact angle 1 : actual contact angle Equation (2) suggests that the apparent contact angle is sufficiently large so that pores with numerous different surface area ratios are present. Figure 5 shows the relationship between the contact angle and the porosity. In Fig. 5, the contact angle refers to the mean value of the contact angle measured in Fig. 2. In substrates with 27% and 39% porosity, the mean value of the contact angle after one hour was adopted. The apparent contact angle increases with the porosity, as shown in Fig. 5 Moreover, the apparent contact angle is close to that predicted by eq. (2).The pore surface area ratio was measured from SEM images (Fig. 4) by an image processing method using the strength of contrast. Figure 6 shows an example of the image used for the measurement. The gray part of the image corresponds to MgAl2 O4, and the black and white parts correspond to pores. The pore surface area ratio is the mean value of measurements obtained post experiment from ten positions in the contact zone between solidified iron and the substrate. Figure 7 shows the relationship between the contact angle and the pore surface area ratio. The apparent contact angle increases with the surface area ratio, as shown in Fig. 5 Moreover, the apparent contact angle is very close to that predicted by eq. (2). It is concluded that the pore surface area ratio influences the apparent contact angle of the MgAl2 O4 substrate with the molten iron. From Figs. 5, 7, it is clear that the relation between contact angle and the pore surface area ratio is closer to eq. (2) than the relation between contact angle and the porosity. Therefore, it is thought that the influence of the pore on the contact angle of the MgAl2 O4 substrate with the molten iron can be determined more quantitatively by adopting the pore surface

4 Wettability between Porous MgAl 2 O 4 Substrates and Molten Iron 2555 MgAl 2 O 4 Pore Fig. 6 SE image of substrate surface and its image by treatment with contrast. 160 Equation (2) Pore surface area ratio (%) Mg O Al Fe Fig. 7 Relationship between contact angle and pore surface area ratio. area ratio for eq. (2). Shinozaki 7) reported that the contact angle of ZrO 2 and MnAl 2 O 4 with molten manganese increased with the porosity of the substrate. The results of the present study, along with those of Shinozaki, support the theory of Cassie and Baxter as an explanation for the increase of apparent contact angle of porous substrates with the molten metal. 3.3 Work of adhesion The work of adhesion, W ad, is defined as the energy needed to separate the liquid from the solid surface. It is given by eq. (3) as follows: W ad ¼ SV þ LV SL ¼ LV ð1 þ cos Þ ð3þ From the above discussion, LV was determined to be 1.8 Jm 2 and the contact angle was determined to be 107 ; these results were obtained from the calculation based on Cassie and Baxter s contact angle theory with 2% porosity. Using these values, the W ad is determined to be 1.27 Jm 2 from eq. (3). In addition, the work of adhesion per mole of MgAl 2 O 4 atoms existing in the interface is calculated from eq. (4) based on W ad. 9) W ad mol ¼ðM=Þ 2=3 N 1=3 fw ad ð4þ Fig. 8 EDS spectra obtained from SE image of substrate with 2% porosity after experiment in Fig. 4. where M is the molecular weight of the substrate, is the density of the substrate, N is Avogadro s constant, and f is the filling coefficient. A report in the literature indicates that f ¼ 1 for sintering solid oxide. 10) W ad mol is determined to be kjmol 1 according to eq. (4). W ad mol is equivalent to the bond energy at the solid liquid interface. The physical bond energy is less than 40 kjmol 1, and chemical bond energy is less than 400 kjmol 1. 11,12) It is concluded that both physical bond energy and chemical bond energy were brought to bear at the interface between molten iron and substrate to create a strong bonding state. Figure 8 shows EDS spectra obtained from SEM images of substrate with 2% porosity after the experiment illustrated in Fig. 4. As indicated, a small amount of Fe is the only substance other than MgAl 2 O 4 that is detected on the surface of the substrate. Because minerals other than MgAl 2 O 4 are not found to be present on the surface of the substrate, it is concluded that Fe and oxygen in the molten iron have been incorporated into the MgAl 2 O 4 matrix as FeO. 13) Tanaka 3) reported a relationship between the work of adhesion per

5 2556 N. Fukami, R. Wakamatsu, N. Shinozaki and K. Wasai mole of Al 2 O 3 and oxygen content in molten iron. From Tanaka s report, as well as on the basis of the results of the current study, it was concluded that in the range of oxygen content from mass% to mass%, the bond between molten iron and Al 2 O 3 is a physical bond because W ad mol is 40 kjmol 1 or less. On the other hand, at an oxygen content of 0.05 mass% or more, W ad mol is kjmol 1 or more. In the range of oxygen content from mass% to mass%, it is observed that the bonding state between molten iron and MgAl 2 O 4 is stronger than that between molten iron and Al 2 O 3. The reason seems to be because FeO can be transferred into MgAl 2 O 4 from molten iron, even at a low oxygen content. 4. Conclusion The wettability between molten iron and MgAl 2 O 4 substrates with porosities ranging from 2% to 39% was investigated in order to elucidate the wettability phenomenon that occurs between molten iron and a refractory material. The conclusions reached are summarized as follows: (1) The contact angle increased with the porosity of substrate. In substrates with porosities of 2%, 8%, and 13%, the contact angles were independent of time. In contrast, in substrates with 27% and 39% porosity, the contact angle decreased rapidly in the first hour and then gradually reached a steady value. (2) The pore surface area ratio was measured from SEM images that were enhanced by image processing using strength of contrast. The contact angle increased from 108 to 130 when the pore surface area ratio increased from 2.4% to 47%. (3) It is suggested that the adhesion between molten iron and the MgAl 2 O 4 substrate is driven by an interfacial bond that comprises not only a physical bond but also a chemical bond. REFERENCES 1) K. Ogino, K. Nogi and O. Yamase: Testu-to-Hagane 66 (1980) ) K. Nogi and K. Ogino: Tech. Assoc. Refract. Japan 41 (1989) ) Y. Tanaka, N. Shinozaki and K. Mukai: Tech. Assoc. Refract. Japan 44 (1992) ) T. Suzuki and T. Koseki: Testu-to-Hagane 92 (2006) ) N. Takiuchi, T. Taniguchi, N. Shinozaki and K. Mukai: J. Jpn. Inst. Met. 55 (1991) ) D. T. Livey and P. Murray: J. Am. Ceram. Soc. 39 (1956) ) N. Shinozaki, T. Noboritate and K. Wasai: Tech. Assoc. Refract. Japan 56 (2004) ) A. B. D. Cassie and S. Baxter: Trans. Farady Soc. 40 (1944) ) N. Shinozaki, J. Morita and K. Wasai: J. JILM 55 (2005) ) F. S. Galasso: STRACTURE AND PROPERTIES OF INORGANIC SOLIDS, (AGNE Gijutsu Center, 1984) p ) W. J. Moore: PHYSICAL CHEMISTRY, 4th Edition, (Tokyo Kagaku Dozin Co., Ltd, 1965) p ) W. J. Moore: PHYSICAL CHEMISTRY, 4th Edition, (Tokyo Kagaku Dozin Co., Ltd, 1965) p ) N. Fukami, S. Kawagishi, N. Shinozaki and K. Wasai: Tech. Assoc. Refract. Japan 60 (2008)