Characteristics of inclusions in high-Al steel during electroslag remelting process

International Jourmnal of Minerals, Metallurgy and MaterialsVolume 19, Number 4, Apr 2012, Page 295DOl: 10.1007/s12613-012-0554-xCharacteristics of inclusions in high-Al steel during electroslag remeltingprocessCheng bin Shil2), Xi chun Chen?), and Han-jie Guo'2)1)State Key Laboratory of Advanced Metallugy, Universty of Science and Tchnology Bejig, Beijing 10083., Caina2) School ofMallurgical and Ecological Engineering, University of Science and Technology Bejing, Beijing 00083, China3) Research Istiute of High Temperature Materials, Ceatral lron and Steel Research Intitue, Bejng 00081, China(Received: 23 March 2011; revise: 20 May 2011; accepted: 23 May 2011)Abstract: The characteristics of inclusions in high-Al steel refined by electroslag rermelting (ESR) were investigated by image analysis,scanning electron microscopy (SEM), and energy dispersive spectrometry (EDS). The results show that the size of almost all the inclusionsobserved in ESR ingots is less than 5 um. Inclusions smaller than 3 pum take nearly 75% of the total inclusions observed in ceach ingot. Inclu-sions observed in ESR ingots are pure AIN as dominating precipitates and some fine spherical Al2O3 inclusions with a size of1 μm or less. Itis also found that protective gas operation and slag deoxidation treatment during ESR process have significant effects on the number of in-clusions smaller than 2 pm but lttle effects on that of inclusions larger than 2 um. Thermodynamic calculations show that AIN inclusions areunable to precipitate in the liquid metal pool under the present experimental conditions, while the precipitation of AIN inclusions could takeplace at the solidifying front due to the microsegregation of Al and N in liquid steel during soidifcation.Keywords: electroslag remelting; aluminum nitride; inclusions; thermodynamics[This work was financially supported by the Intemational Science and Technology Cooperation and Exchange of Special Projects (No.2010DFR50590).]1. Introductionprecipitated during ESR have been reported [6-11], there isno report on the characteristics of inclusions in high-Al steelAIN inclusions are typical precipitates in high-A1 steel.during ESR process.The precipitated AIN inclusions could induce slab trans-In the present work, the characteristics of inclusions inverse cracking during rolling process [1]. Although electro-high-Al steel produced by ESR under different remeltingslag remelting (ESR) has many advantages [2 -3], reducingconditions were studied. The formation and behavior of in-the nitrogen content in liquid steel during ESR process isclusions during ESR process were also discussed based onrelatively dificult. Removal of nitride inclusions is the mainthermodynamics. In addition, the effects of remelting at-way to lower the nitrogen content during ESR process [4].mosphere and slag deoxidation treatment on the charac-Croft et al. [5] reported that the precipitation of AIN in-teristics of inclusions were analyzed.clusions in Al-killed steel is due to the enrichment of solutein liquid phase during soldification. Yin [1] reported that2. ExperimentalAIN precipitation in high-Al steel is much easier than that in2.1. Experimental procedureregular Alkilled steel at the austenite temperature range.Although various studies on characteristics of inclusionsA consumable electrode with the diameter of 80 mm andCoresponding author: Han-jie Guo E mail: guohanjic@ustb.edu.cn中国煤化工包Springer。University of Science and Technology Beijing and Springer-Verlag Berlin Heidelberg 2012MHCNMH G296Im. J. Miner. Metall. Mater, VoL.19, No.4, Apr 2012the length of 600 mm was remelted under different remelt-Pre melted slag composed of 60wt% CaFz, 20wt% CaO,ing conditions in a 50-kg scale ESR fumace. The chemicaland 20wt% Al2O3 was used in each experiment. Thecomposition of the consumable electrode used in each heatpre-melted slag was calcined at 500°C for at least5h to reis given in Table I. .move moisture before ESR experiments.Table 1. Chemical composition of consumable electrode used2.2. Chemical analysis and microscopic observationin each heatwt%The total oxygen and nitrogen contents in each ESR ingotCMn_SiPSCrNiCuMoSol.Alwere determined by the inert gas fusion-infrared absorpti-0.07_ 1.70 0.14 0.014 0.002 0.16 3.04 1.09 0.34 0.84ometry and inert gas fusion-thermal conductivity method,respectively. The contents of soluble aluminum in steelFig.1 shows a schematic diagram of the experimentalsamples were determined by the inductively coupledapparatus. The inner diameter of the water- cooled copperplasma-mass spectroscopy (ICP -MS). The analyzed resultsmold is 170 mm. Three heats were conducted to investigateare listed in Table 2.the effects of remelting atmosphere and slag deoxidationTable 2. Oxygen, nitrogen, and soluble aluminum contents intreatment on the characteristics of inclusions during ESRESR ingotsprocess. The fixed remelting conditions in each beat areRemelting condition Sample No.__ 0 N Sol. Alsummarized as follows: AC curent of about 1500 A, volt-Case AE0.0032 0.00320.84age of 49 V, remelting rate of approximately 40 kg/h, andthe outlet temperature of cooling water of 25°C. The dfferCase B0.0012 0.00330.81Case C0.0010 0.00321.05ent remelting conditions adopted in ESR experiments arelisted as follows: case A (the first heat) - - remelting underatmosphere, without slag deoxidation treatment; case B (theIn order to analyze inclusions in ESR ingots, steel sam-second heat)- - remelting under Ar gas atmosphere (Ar gasples were also prepared for image analysis, scanning elec-flow rate of 50 NL/min), without slag deoxidation treatment;tron microscopy (SEM), and energy-dispersive spectrometrycase C (the third heat) - - remelting under Ar gas atmos-(EDS).phere (Ar gas flow rate of 50 NL/min) combining with slagFifty view fields were observed for each steel sample todeoxidation treatment, and adding deoxidant (30w%analyze the size, distribution, and mumber of inchusions us-Al+15wt% Al2O3+15wt% CaF2+40wt% Fe powder) intoing an image analyzer (Leica MEF4A) by 200 magnifica-slag.tions. The total area of 50 view fields is 10.19 mm2. The ef-fective minimum size of inclusions that can be detected bythe image analyzer is 0.587 μm. It should be emphasizedStub-that the size of particles derived by the image analyzer is the+ - Gas protective capequivalent diameter, which is defined as the diameter of acircle with the same area as the particle.Consumable electrodeSixty inclusions in each sample were randomly selected(uppryfor SEM-EDS measurement, and the morphology and➢Water ouletchemical composition of each inclusion were analyzed bySlag poolSEM-EDS (JEOL JSM-6480LV).Liquid metal poolSolidified ingot3. Results and discussionWater inletWater cooledmold3.1. Characteristics of inclusions in high-Al ESR ingotsCooling←校+一WaterFig. 2 shows the size distribution of inchusions observedWatercooled base platein ESR ingots remelted under different conditions. It can beobserved that the size of all inclusions detected in the threeFig 1. Schematic diagram of the experimental apparatus.samples is less than 20 um. The statistical results of inchu-The produced ESR ingots in case A, case B, and case Csions in ESR i中国煤化工zer are summa-were recorded as samples EI, E2, and E3, respectively.rized in TableTYHCNMHG.C.B. Shi et al, Characteristics of inclusions in high-Al steel during electroslag remelting process2977As shown in Table 3, the maximum size of inclusions日- EI60- 0甲)-E2detected in samples E1, E2, and E3 is 18.28, 18.90, andA E317.59 μm, respectively. Fig. 3 shows the relationship be-50tween the size of inclusions and the ratio of the number ofdifferent size inclusions to the total number of inclusions. It宣40OAis clear from Figs. 2 and 3 that the inclusions of 1-2 μm ini307Dsize take u the largest proportion of the total inclusions andsecondly the inclusions of 0-1 μm in size in each sample. Itg 20can also be observed that inclusions in the range of 0-5 μmaccount for more than 90% of the total inclusions in each” I0ingot, and the inclusions smaller than 3 μm take nearly 75%.0Inclusions larger than 10 μm take up a very small proportion0-1 4-5 8-9 12-13 16-17in each sample. The number of inclusions larger than 3 um2-6-7 10-11 14-15 18-19in each sample decreases with increasing size as shown inDiameter of inclusions 1 jμumFigs. 2 and 3. The nunber of inclusions with the size largerFig 2. Size distribution of inclusions observed in each ESRthan 5 μm per mm2 in three different ESR ingots is almostingot.the same with the increase of size.Table 3. Statistical results of inclusions in ESR ingotsSample Number of observed Maximum equivalent Minimum equivalent Total mumber of inclusions in Ratio of total area of inctusions to to-sNo.inclusions/ mm2 circle diameter/ um circle diameter/ um the observed 50 view fields tal area of observed view fields / %E120718.280.58721050.29217018.90).58717340.22311117.5511270.200.35in sample E2 is less than that in sample E1.口-E1日E2.0.30- 命As shown in Table 3, the total pumber of inclusions in-AE30.25.dthe observed 50 view fields in samples E1, E2, and E3 isobviously decreasing in order. The ratio of the total crOSSsectional area of inclusions to the total area of the observed中view fields in sample E1 is larger than that in E2 or E3. In0.15sample E3 remelted in case C, the ratio of the total cross0.10sectional area of inclusions to the total area of observedview field is the smallest. In samples E1, E2, and E3, the ra-tio of the total cross sectional area of inclusions to the totalarea of the observed view fields is 0.29%, 0.22%, and0.009的0.20%, respectively.2-36-710-11 14-15 18-19The SEM images and EDS analysis results of typical in-Diameter of inclusions/ μnclusions observed in samples E1, E2, and E3 are shown inFig 3. Relationship between the size of inclusions and the ra-Fig. 4, respectively. Figs. 4(a) 4(e) show the morphologytio of the number of the different size inclusions to the totaland composition of typical inclusions observed in sample E1.number of inclusions.It can be observed that the morphology of typical AIN in-As shown in Figs. 2 and 3, the number of inclusionsclusions is clusters, near-spherical shape, or quadrangle withsmaller than 5 um per mm2 in sample E3 is much less thanclear angularities in the ESR ingot remelted in case A. Thethat in samples E1 and E2, especially for inclusions smallersize of most AIN inchusions is about 2 μm. Meanwhile,than 2 um. It can be scen by comparing different remeltingsome spherical中国煤化工m or less in sizeconditions that the number of inclusions smaller than 2 μm can also be fouYHCNMHG.C.B. Shi et al, Characteristics of inclusions in high-Al steel during electroslag remelting process299lgK.=lgAN - = lgaAan -lgax-lgay=lgf;= Z(e'[%j]+r{[%j})(3)aAI"aNwhere e| and r| are the first-order and second-order in--lgfA[%AI]-lgJN[%N](2)teraction parameters, respectively. The first order interactionwhere fAs and fi are the activity coefficients of dissolvedparameters used in the present study are listed in Table 4.aluminum and nitrogen in liquid steel, respectively. TheyThe available second-order interaction parameters are sum-marized as = -0.004, 州一-0.0006, and呤=can be calculated using the following equation:-0.0011 + 0.17/T[15].Table 4. First-order interaction parameters used in the present studye{CSMnNiaMoAlA0.0910.0560.035[14]0.0330.0350.03 [14]0.011+63/T[15]N0.130.048-0.020.0590.0070.00-0.0460.009-0.0110.01-0.421-0.066-0.0210.07-0.1330.006-0.055-0.0130.0051.9-5750/T[16]Note: data without notation are from Ref. [17].The liquidus temperature Tkq and solidus temperature0.04口aTon of the studied high-Al steel can be calculated using theQbfollowing equations [18-19]:c0.03 上1772Ko(°C)= 1536 -{00.3[%C]- 22.4%C] -0.16+13.55[%Si]-0.64[%Si} + 5.82[%Mn]+ .03[%Mn]} + 4.2[%Cu]+ 4.18[%Ni]+1748 K0.0I%Nij2 +1.59%C]-0.007%Cr} (4a)T01(°C)= 1536-{415.5%C]+ 12.3[%Si]+006.8[%Mn]+ 124.5[%P]+ 183.9[%S]+4.3[%Ni]+ 1.4[%Cr]+ 4.1[%AI}(4b)0.0.80.91.01.1[%Al]The calculated liquidus temperature Tig and solidus tem-Fig. 5. Stability diagram of AIN precipitation in high-Al steelperature Tol by Eqs. (4a) and (4b) are 1772 and 1748 K,re (a- -sample E1, b- -sample E2, and c- sample E3).spectively. The thermodynamic condition for AIN particlesprecipitation in liquid phase can be expressed as follows:When the Al content in liquid steel is 0.84%, the lowestN content for precipitating AIN inclusions at the liquiduslgaxaN = lgfAI[%AI]{N[%N]= 5.620-12849.457(5temperature of 1772 K is 0.025% calculated using Eq. (5). ItTcan be seen that the calculated lowest N content for AINUsing the relevant parametrs, the sablity diagam of formation in the liquid metal pool is much greater than thatAIN precipitation can be obtained as shown in Fig. 5. Thein the studied steel. Therefore, it is impossible to precipitatedashed line and solid line shown in Fig. 5 were calculatedAIN inclusions in liquid steel at the liquidus temperature orby combining Eqs. (3) and (5) as well as thermodynamicabove. When the Al and N contents are 0.84% and 0.0032%lata given in Table 4, respectively. The analyzed Al and Nin liquid steel, the calculated critical temperature for AINcontents for samples EI, E2, and E3 are also shown asprecipitation is 1577 K, which is much lower than thepoints a, b, andc in Fig. 5. It is clear from Fig. 5 that AINsolidus temperature of 1748 K. This result indicates thatinclusions are unable to precipitate in liquid steel with the AlAIN inclusions are unable to precipitate in the liquid metaland N contents of points a, b, and c above the solidus tem-pool, or at the'中国煤化工ling through theperature and liquidus temperature.slag pool, orDlets at the elec-:YHCNM H G'.300In. J. Miner. Metall Mater, VoL19, No.4,Apr 2012trode tip in ESR process based on thermodynamic equilib-smaller than 2 μm in sample E1 is much more than that inrium analysis. Meanwhile, thermodynamic calculations in-sample E2 or E3. Meanwhile, considering that the size of alldicate that even if there are AIN inclusions in the consum-the Al2O3 inclusions observed by SEM-EDS in ESR ingotsable electrode, these AIN inclusions still can decompose atis 1 μm or less as well as that the oxygen content in samplethe above-mentioned three stages. Fu et al. [6] reported thatE1 is much higher than that in samples E2 and E3, it can bemost of the original inclusions in the consumable electrodeinferred that Al2O3 inclusions take up a large proportion incould be removed at the stage of metal droplets fallingthe extra amount of inclusions in sample E1. As a result ofthrough the slag pool and formation of droplets at the elecprotective gas operation or slag deoxidation treatment, thetrode tip during ESR process. Inclusions in the final ESRoxygen contents in samples E2 and E3 are very low, whichingot are the newly precipitated inclusions during thefurther results in low Al2O3 inclusions content in ESR ingots.solidification of liquid steel in the water-cooled mold.It is diffcult to detect inclusions smaller than 1 μm due toSimilar results have been reported by Li et al. [8] and Kay etthe limitation of the lower measuring limit of SEM.al. [1] based on experimental studies. Few inclusionsWith the formation of metal droplets at the electrode tip,observed in samples E1, E2, and E3 are expected to beliquid steel begins to solidify at the bottom of the liquidoriginal inclusions in the consumable electrode.metal pool in the water-cooled mold during ESR process.The chemical reaction for formation of Al2O3 inclusionsDuring the cooling of liquid steel in the water-cooled moldin liquid steel can be written asfrom the liquid metal pool temperature to the liquidus2[A1] + 3[0]= (Al203)(),temperature, and then to the solidus temperature, the soluteis rejected into interdendritic liquid phase, which results inSG8 =-1205115+ 386.714T (/mo) [20](6the enrichment of Al and N in residual liquid steel betweensolid steel dendritic arms at the solidifying front because ofThe activity of Al2O3 is unity (relative to pure solid Al2O3the difference in solubility of solutes between liquid andas the standard state). The equilibrium constant K6 for reac-solid phases [22]. When the product of [%Al]x[%N] in in-tion (6) can be expressed asterdendritic liquid steel exceeds the equilibrium value forAIN precipitation at certain temperature, chemical reactionlgK6 =Ig-aALO3-=lgaA12O, -2gax -3lgao=(1) for AlIN formation may occur. Because complete diffu-sion of solutes in solid phases is impossible, the following-2lg fA[%AI]- 3lg fo[%O](7equation can be used to calculate the solute content in rwhere fo is the activity cofficient of dissolved oxygen insidual liquid phase during solidifcation by assuming nodiffusion in solid phase [22-23]:liquid steel, and can be calculated using Eq. (3), combiningwith the first-order interaction parameters given in Table 4C =C(l-fs)4-)and second-order interaction parameters as ro"= 0.0033 -25.0/T[16], r = 0.0001 [21], and r°*= 0.0003 [21].where C is the concentration of the solute in residual liquidThe measured oxygen content in ESR ingot is the sum ofphase during solidification, Co is the initial concentration ofthe dissolved oxygen and the oxygen combined as oxide in-the solute in liquid steel,Js is the solid fraction, and k; is thectusions. By combining Eqs. (6) and (7), the dissolved oxy-equilibrium partition ratio of solute i between liquid andgen content can be calculated at the liquidus temperaturesolid phases. During the cooling of liquid steel, the Al and Ncontents in liquid phase can be calculated as follows:1772 K as 0.00015%, which is much less than the measuredtotal oxygen content in ESR ingots, especially in sample E1.Therefore, it can be concluded that nearly all oxygen in[%AI]=[%Al](1-5)#u-)(9a)high-Al ESR ingots is expected to exist as Al2O3 inclusions.[%N]=[%N](-fs)4-1)(9b)As showm in Figs. 2 and 3, there are few inclusions withlarge size (greater than 5 μm) in samples EI, E2, and E3where kA and kw are equal to 0.6 and 0.27, respectively [1,Inclusions smaller than 2 μm take nearly 60% of the total24]. The relationship between the product of [%A]x{%N]inclusions in each ingot. The number of inclusions largerand the solid firaction fs is shown in Fig. 6. The calculatedthan 2 μm is almost the same in different ingots. It can be :critical value中国煤化工N] for AIN pre-observed from Figs. 2 and 3 that the number of inclusionscipitation atMYHCNMHCB Eq. (5) is also.CB. Shi et al, Characteristics of inclusions in high-AI steel during electroslag remelting processshown in Fig. 6. It can be observed that the product offraction during the solidifcation. When the product of[%AI]x[%N] increases with increasing solid fractionfs. Al[%AI]x[%N] exceeds the calculated critical value for AINand N in the residual liquid phase between dendritic armsprecipitation at the liquidus temperature, AIN would pre-would enrich at solidifying front with the increase of solidcipitate in the residual liquid phase at the solidifying front.0.10(a)。(b)(-Calculated value fromfs..0.080.08 --- Calculated critical value from Tgq系0.00.060.04 tAIN precipitation0.04写0.040.02No AIN precipitation0.00后-0.840后s=0.841-Js- 0.800-;0.0 0.2 0.4 0.6 0.8 1.o0.0 0.2 0.4 0.6 0.8 1.00.0 0.2 0.40.6 0.8 L.oSolid fraction,fSolid fraction,JSolid fractionfsFig. 6. Relationship between AIN precipitation and solid fraction: (a) case A; (b) case B; (C) case C.The critical value of solid fraction fs for AIN precipitation4. Conclusionswas calculated as 0.840 using Eqs. (5) and (9) for the steelwith 0.84% Al and 0.0032% N as shown in Fig. 6(a). For(1) Almost all inclusions in high-Al ESR ingots are lesscase B and case C, the critical values ofs for AIN precipita-than 5 μm in size. Inclusions smaller than 3 μm take nearlytion were calculated as 0.841 and 0.800 as shown in Figs.75% of the total inclusions in each ingot. Inclusions ob-6(b) and 6(c), respectively. When the values of solid fractionserved in ESR ingots are pure AIN as dominating precipi-fs exceed these critical values during solidification, the en-tates and some fine spberical Al2O3 inclusions of 1 μm orrichment of Al and N at the solidifying front can result inless in size.AIN precipitation.(2) Under the condition of protective gas remelting com-Directional solidification proceeds from the bottom of thebining with slag deoxidation treatment, the oxygen contentshallow liquid metal pool during ESR process. The growthin the ESR ingot can be reduced to 10x10 , which is muchdirections of dendrites are mostly perpendicular to the bot-lower than that in the ingot produced under atmosphere ortom of the liquid metal pool. With the growth of dendrites,Ar gas atmosphere.dendrites will be interconnected with each other, and then(3) Thermodynamic calculations show that AIN inclu-the residual liquid steel among interconnected dendrites willsions are unable to precipitate in the liquid metal pool duringbe closed up. Under this condition, the diffusion of solutesESR process under the present experimental conditions.in interdendritic liquid steel was hindered, which would in-However, AIN inclusions can precipitate at the solidifyingduce the enrichment of Al and N at the solidifying front.front where the solid fraction is greater than the calculatedThereafter, AIN inclusions would precipitate at the solidify-critical values of 0.840, 0.841, and 0.800 for case A, case B,ing front caused by the enrichment of Al and N. AIN inclu-and case C, respectively.sions formed in interdendritic liquid steel are nearly unableto float out, which are closed up by interconnected dendrites.ReferencesThereafter, some AIN clusters form among dendrites be-cause fine precipitates collide with each other, which has[1] H.B. Yin, Inclusion characterization and thermodynamics forhigh-Al advanced high-strength steels, lron Steel Techmol,been confirmed by SEM analysis as shown in Fig. 4. Elimi-25(2006), p.64.nation of inclusions formed at the solidifying front during2] V. Weber, A. Jardy, B. Dussoubs, D. Ablitzer, S. Ryberon, V.soldification by floating up is almost impossible [11, 22],Schmit, S. Hans, and H. Poisson, A comprehensive model ofespecially during ESR process [1, 25], due to its relativelythe electroslag remelting process: description and validation,rapid cooling rate. Under this condition, the precipitatedMetall. M/000 NI~2 .1AIN inclusions at the solidifying front will be left in the f-3] S.K Mai中国煤化工二nd R Kawalla, De-nal ESR ingot.velopmen:YHCNMHGseelthrougheloo.302Int J. Miner. MetalLl Mater, VoL19, No.4, Apr 2012troslag refining process, ISU Int, 49(2009), No.6, p.902.No3, p.337.[4] Z.B. Li, Electroslag Metallungy Theory and Practice, Metal-[15] H. Ohta and H. Suito, Actities in CaO-SiO2-Al2O3 slags andlurgical Industry Press, Bejing, 2010, p.63.deoxidation equilibria of Si and Al, Metall. Mater. Trans. B,5] NH Croft, A.R. Entwisle, and GJ. Davies, Origins of den-27(1996), No.6, p.943.dritic AIN precipitates in aluininmkilld stel castings, Met.[16] H. Itoh, M. Hino, and S. Ban-ya, Asessment of Al deoxida-Techmol, 10(1983), p.125.tion equilibrium in liquid iron, Tesu-1o -Hagane, 83(1997),6] J. Fu and J. Zhu, Change of oxide inclusions during electro-No.12, p773.slag remelting process, Acta Metall. Sin, 7(1964), No.3, p.250.[1刀] The 19th Comite in Steelmaking of the Japan Society for[7] D.G. Zhou, X.C. Chen, J. Fu, P. Wang, J. Li, and M.D. Xu,the Promotion of Science, Steelmaking Data Sourcebook,Inclusions in electroslag remelting and continuous castingGordon and Breach Science Publishers, New York, 1988,bearing steels, J Univ. Sci. Technol. Bejing (in Chinese),p.280.22(2000), No.1, p.26.[18] M. Suouki, R. Yamaguchi, K Murakami, and M. Nakada,[8] Z.B. Li, W.H. Zhou, and Y.D. Li, Mechanism of removal ofInclusion particle growth during solidifcation of stainlessnon-metallic inclusions in the ESR proces, lron Steel,steel, ISU In., 41(2001), No.3, p.247.15(1980), No.1, p.20.[19] T. Binran, Handbook of Iron and Steel, VolL1, 3rd, Edited by[9] Z.B. Li, J.W. Zhang, and X.Q. Che, Control of content andIron and Steel Instiute of Japan, Marnuzen, Tokyo, 1980composition of Do-meallic inclusion in ESR steel, J. lronSteel Res., 9(1997), No.2, p.7.[20] M.A.T. Andersson, P.G. Jnsson, and M.M. Nzotta, Applica-[10] s. Ahmadi, H. Arabi, A. Shokuhfar, and A. Rezaei, Evalua-tion of the sulphide capacity concept on high-basicity ladletion of the electroslag remelting process in medical grade ofslags used in bearing steel production, ISU Int, 391999),316LC stainless steel, J. Mater. Sci Technol, 25(2009), No.5,No.11, p.1140.p.592.[21] H Ohta and H. Suito, Calcium and magnesium deoxidation[1] D.A.R. Kay and RJ. Pomfret, Removal of oxide inclusionsin Fe-Ni and Fe-Cr alloys equilibrated with CaO-Al2O3 ancduring ac electroslag remelting, J. lron Steel Inst, 209(1971),CaO-Al2O-MgO slags, ISI Int, 43(2003), No.9, p.1293.p.962.[22] S.K Choudhary and A. Ghosh, Mathematical model for pre-[12] J.C. Labbe and A. Laimeche, Study of the behaviour of alu-diction of composition of inclusions formed during solidifica-minium nitride in the iron and steel industry, J. Eur. Ceram.tion of liquid stee, ISIUJ Int, 49(2009), No.12, p.1819.Soc, 16(1996), No.8, p.893.[23] W. Kurz and DJ. Fisher, Fundamentals of Solidification, 3rd[13] H. Wada and R.D. Pehilke, Nitrogen solubility and aluminumEd, Trans Tech Pubilcations, Switzerland, 1992, p.280.nitride precipitation in liquid iron, Fe-Cr, Fe-Cr-Ni, and[24] ET. Turkdogan, Fundamentals of Steelmaking, The InstituteFeCr-Ni-Mo alys, Metall. Trans. B, 9(1978), No.4, p.441.of Materials, London, 1996, p.298.[14] J.H. Park, S.B. Le, D.S. Kim, and JJ. Pak, Thermodynamics[25] A. Mitchell, Oxide inclusion behavior during consumableof titanium oxide in CaO-SiO2~Al2Os-MgOms-CaF2 slagelectrode remelting, lronmaking Steelmaking, 1(1974), No.3,equilibrated with Fe-11mass%Cr melt, ISU Int, 49(2009),p.172.中国煤化工MHCNM HG.