D0I:10.13374/i.isrm1001-053x.1989.06.019 第11老第6期 北京科技大学学报 Vol.11 No.6 1989年11月 Journal of University of Science and Technology Beijing Nov,1989 The Deoxidation Process of GCr15 Steel' Cai Huaide(蔡怀德)Zhang Jian(张鉴)Tong Fusheng(佟福生) Zhang Beingcheng(张炳成)Qu Dianlou(曲殿楼) Shi Liyong(时立永)Zhang Mingy:(张明义)** ABSTRACT:It is shown that the average CO)in the steel can be decreased to 12.86 ppm and there is an optimal %Si]at the end of melting as 0.11%, at which [O)after vacuum carbon deoxidation (VCD)is minimum.A kinetical model has been deduced to simulate the VCD process.The results show that adding Si before Al is superior to the adding Al before Si with respect to the final [O)in the steel.It is suggested that the melting point of covering slag should be lower than 1550C,and (FeO+MnO)should be less than 1%,so as to effectively refine GCr15 steel. KEY WORDS:deoxidation,optimal silicon content,covering slag,refining In secondary refining of bearing steel,there are two kinds of deoxidation: vacuum carbon deoxidation and procipitation deoxidation with silicon and alu- minimum.The former has the merit of leaving no nonmetallic inclusions in steel,yet it can not reduce the oxygen content in the steel to the required level. As regards precipitation deoxidation,there are different opinions about the removal of deoxidation products.The refining process with covering slag develo- ped rapidly in recent years is very valid for removing the oxides in molten steelt1,2,3),so much attention should be paid to it. On account of the above mentioned,a study on the deoxidation process of bearing steel has been carried out in 50kg vaccum induction furnace in order to search after the reasonable parameters for VCD process. Manuscript Received March 27,1989 ··Dept.of Metallurgy ···Xining Steel Works in China 493
第 卷第 期 北 京 科 技 大 学 学 报 年 。 ’ ‘ 了 蔡怀德 。 夕 方 。 张 鉴 。 夕 。 夕 体福 生 石 ‘ 张炳 成 曲殿楼 ￡ 时立 永 夕￡ 张明 义 今 〔 〕 〔 〕 。 ， 五 〔 〕 。 ，· 〔 〕 。 ， “ ， 卜 ， · ， ， ， ， “ ， ， 一 ’ ’ ， · ， · 二 ， … DOI ：10．13374／j ．issn1001－053x．1989．06．019
1 Experimental Method 1.1 Experimental Conditions All experiments have been carried out in a 50kg vaccum induction furnace. The crops of GCr15 steel products are used for the charge,which is melted in an electroremelted magnesia crucible.The temperature of steel before refining is 1678t10C.The flow rate of argon for stirring is INL/min.The pressure in the furnace is less than 0.133kPa. 1.2 Variants of Deoxidation There are two variants of deoxidation as follows (a)VCD-deoxidation with Al-refining with covering slag under atmosphere -adding FeSi at the later stage of refining. (b)VCD-deoxidation with FeSi-deoxidation with Al-refining with cove- ring slag under atmosphere. CaO-Al2O,-CaF2 system is used as covering slag,slag and alloys were dried at400~500°C. 2 Experimental Results 2.1 Change of Oxygen in the Steel with Time during VCD The change of oxygen in the steel with time during VCD is shown in Fig.1. It is seen that the oxygen content in the steel can be reduced to less than 20ppm (mean CO)=12.86ppm)after more than 10 min,but when the treating time exceeds 15 min,the oxygen content in the steel nearly remains unchanged. However,as observed in the experiments,molten steel boils all the time during VCD.This fact shows that there must be sources supplying oxygen to the molten steel at the later stage of VCD,otherwise,it is impossible to keep [O]constant. 2.2 The Relation botween Initial C%Si]and [O]after VCD The relation between initial C%Si)。and〔O〕after VCD is shown in Fig2. Here CO]is the mean value of oxygen at the later stage of VCD.From this Figure,it is shown that there is an optimal initial %Si],at which CO] after VCD is minimum.When〔%Si)。is more or less than〔%Si)g,〔O〕after VCD would increase.By regression analysis of experimental data a relation of second order between O]after VCD and initial %Si].is obtained as follows: 〔0]=12.44-23.03〔%Si0。+106.1〔%Si)3 (1) 494
一 一 。 土 。 “ 。 招 。 一 ， ， ， ， ， ， 一 。 一 ， 。 一 卜 百 一 一 。 〔 〕 ， ， ， ， 。 ， ， 〔 〕 。 一 一一 〔 一〕 。 〔 〕 〔写 〕 。 〔 〕 〔 〕 ， 口 。 位 〔 〕曹 ， 二 。 〔 写 〕 。 】】 〔 写 〕 ， 〔 〕 〔 〕 一 弓 〔 〕 〔 〕 。 〔 〕 。 一 。 〔 〕 。 。 〔写 〕 竺 尸
15 ●Experimental data -Calculated data' 60 层 12 40 o9 g 11 10 20 8L 10 2030 40 50 0.050.100.150.200.250.30 Time,min Initial〔Si),%. Fig.1 Change of [O]with time Fig.2 Relation between initial during VCD C%Si)。and〔o)after VCD Differentiating the above relation with respect to [%Si]and putting dcO)/ dc%Si.=0,and given the optimal %Si].=0.11%,which is in good agree- ment with that given by G.N.Oiks!(%Si]=0.08~0.12%). 2.3 The relation between deoxidation and decarburization The relation between reduced content of oxygen and that of carbon during VCD is shown in Fig.3.It is indicated that a certain amount of carbon should be burned out. 70 1 60 60 ACO)om/4100△(C)g 40 40 wdd g E 20 20 101520 25. 1100 1600 1800 2000 △C),g Melting point,'C Fig.3 Relation between△〔O)and Fig.4 The relation between the △rC)during VCD melting point of covering (-experimental,----calculated) slag and final〔0] On the other hand,theoretical relation between deoxidation and decarburi- zation can be derived as follows from the weight of steel: △〔0)5mm=1/3(100)△〔C)g (2) It is evident that the difference between the practical and theoretical value of deoxidation is very obvious.This also shows that there are some oxygen 495
孟孟么， 代乙 爪 日合 、 ‘几 。 万 ‘ ‘ 、 、 ， 介 尸 厂 ，上︸土口，︹ 五 麦 五犯 犷王几 嗯山 一 公 以 压 ’ 、 眺 猛 尸 二 泛丁二三 一 一 一 卜巴 ’ 且 三护 划卜 ， 〔 〕 卜 飞 。 艾了 厂 〕 ，呱 〔 〕 。 〔 〕 〔 〕 。 〔 〕 〔 〕 。 ， 〔肠 〕 。 。 ， 一 〔 〔肠 〕 。 。 。 一 权 内鑫‘ “甘血 任 尹 △〔 。 ，蛛 ‘ ， ‘ ’ △了〕 一 厂 子 。 产 ， 广 丈 ‘ － 了 厂 声 一 一 户尸洲 ‘ 卜 虽 闷八 乙犯〕 尽 ， ， 。 。 么〔 〕 △〔 〕 一 ， 一 卜 〔 〕 ， 〔 〕备 二 △〔 〕
sources during VCD process. 2.4 The Relation between the Melting Point of Covering Slag and Final [O] The relation between the melting point of covering slag and final cO]is shown in Fig.4.It is indicated that the final co)increases rapidly with the increasing the melting point of the covering slags above 1550C.So good results can only be obtained by the rational prescription for the covering slag of CaO- Al2Oa-CaF2 system so as to keep its melting point under 1550C. 2.5 The Influence of Deoxidation Variants on the Final [o] The influence of deoxidation variants on the final cO]is shown in Fig.5. It is evident that variant b is better than a with respect to deoxidation.So,it is suggested that variant b should be used for refining ball bearing steel. 3 Discussion 3.1 The Deoxidizing Power of Carbon in Molten Steel 〔C)+〔O)=C0(e),logK=1160/T+2.003, K1873=pco/C%Cf。〔%0]f。=419.1 (3) For GCr15 steel,fe and fo are 1.33 and 0.282 respectively,so C%0]=pco/(K1873C%C)fcfo)=6.36×10-3pco (4) The equilibrium CO)under different pressure Pco are shown in Table. Pco,kPa 101.3210.131.010.1330.0660.033 COJ,ppm 63.626.360.640.0840.0420.021 In the experiment,the mean [O]after VCD is 12.86ppm,which is much higher than the value in Table at Pco<0.133 kPa. 3.2 The Main Factors Affecting VCD (1)The Vacuum degree requircd for reducing Al2O3 by [C] A120,+3〔C)=2〔A1)+3C0(g,logK=-61388/T+26.862 (5) K1873=a1p8o/c8=1.188510-6 pc0=〔%C)fc×10-2×(1,1885/C%A12.f:)13 (6) when [%A1]=0.04,far=1.31,Pco=10.13kPa When the furnace pressure is dropped below 10.13kPa,it means that Al2O3 may be reduced below by〔C)atl60o°C. (2)The vacuum degree required for reducing Mgo by [C) 496
一 〔 〕 〔 〕 〔 〕 。 一 。 〔 〕 〔 〕 ， ， · 一 。 ， 。 卜 〔 〕 〔 〕 ， ， 。 。 〔 〕 。 · 〔 〕 。 ， 。 。 ， 声 〔 〕 。 。 尤 ， 。 · 〔 〕 · 。 一 。 一 。 。 〔 〕 。 。 ， 。 。 。 。 。 〔 〕 ， 。 。 。 。 。 ， 〔 〕 ， ， 夕 · ， 。 卜 〔 〕 口、、‘、 目， 匕八 内 矛 。 〔 〕 〔 〕 ， 一 二 灵 ， · 急 。 忍 。 一 。 。 二 〔 百 〕 · 。 一 。 〔 〕 · 灵 ’ 尹 〔 〕 。 ， 。 ， 。 二 。 ， 〔 〕 〔 〕
MgO()+〔C)=Mg0(g)+C0(g),logK=-31684/T+13.09 K1843=1.479×10-4 (7) Pco=1.42kPa,ie.Mgo may be reduced by [C)at 1600C when pressure (PMs+Pco)is lower than 2.84 kPa. (3)The vacuum degree required for decomposition of MgO Mg0c)+=Mg(g)+〔O],logK=-32580/T+10.88, K1873=a,pMs=3.09×10-7 (8) For pure iron,f。=1,C%O〕=3.09×10-7/pMg (9) For GCr15 steel,.f。=0.282,〔%0〕=1.096×10-6/pwe (10) The relationship between PMs needed for decomposition of MgO and (O]is shown in Fig.6.It is obvious that Mgo may decompose when Pye is lower than 0.0667 kPa and [O]is less than 20 ppm. From thermodynamic calculation mentioned above,it is evident that there are indeed some oxygen sources supplying oxygen to the molten steel during VCD,that is why the experimental co]is always higher than the theoretical (with exception of ferrostatic pressure). 70 1--Pure'iron 2--GCr15 stee 40 50t 30 40 [BUIA 20 10L 50 55606 Cao in covering slag,% 00.03330.06670.10000.1333 乃Mg,kPa Fig,5 The influence of deoxidation Fig.6 The relaticn between Pys variants on the final O] needed for deccmpcsiticn cf (a-deoxidation with Al before Si: Mg0andC0]at1600°C b-deoxidation with Si before Al) 3.3 The kinetic mocel cf VCD Generally,the rate of ICD is expressed by (402）,=-0o1co (11) If decomposition of Mgo is regarded as zero order reactiont),then at the rate,oxygen supplied by Mgo to the molten steel can be given as 497
、 、 。 〔 〕 ， 二 一 嫂 。 只 一 。 。 ， · · 扭 〔 〕 “ 、 一 。 · 。 〔 〕 ， 一 。 ， 。 一 ， 。 ， 〔 〕 一 ， 。 ， 〔 〕 一 “ 。 〔 〕 · 〔 〕 。 、 ， 少 厂 ， ， 从 乃 一 〔 〕 丁 ﹄ 匕丹曰“ 尸 任‘ 、 、 ， ， 一 、 、 、 、 一勺 ‘ 抽 匕 匕一 ，肠、 － 任 ︸︸ 山勺 “ ，土 ︹︺ ‘ 、 又 、 、 ‘ 卜、 、 、 一尸尸尸户尸口 一 八目 今︸，土 八曰 洲一陌口 ， 失 、 、 呜峨 。 卜 ‘ 卜 〔 〕 一 一 二 · 七 ， 〔 勺 “ 。 卜 ， 手 一 ， 灭 一 手 〔 〕 、 ， · 一 夕 ， 一 下方 “ “ ’ ， ，
(9),=最 (12) Hence,the total rate of VCD may be expressed as follows: dco]/df=(dco]/dt):+(dco]/dt)2 =-(F1/V)]+(F2/Wm)k! (13) Putting1=(F1/Vm)kto1,2=(F2/Wm)k'gives d〔O)/dt=-k1CO)+k2 (14) Integrating equation (14),gives 〔0]=1/k1〔(k,〔O)o-k2)e-k1t+k2 (15) From Fig.1,it is shown that when t approaches infinite,the mean value of (O)will approach 12.86ppm as a limit i.e. limC0〕=12.86，thus from(15),k2=12.86k1 In this experiment,F1=1600cm2,taking k[o=0.02cm/s,from references [10,11],then k1=F:kto]/Vm=Fikto]Pa/Wm=5.6x10-3 k2=12.86k1=7.2×10-2 As the mean initial COJo is 51.24ppm,substituting this into (15),the later becomes C0]=12.86+38.4exp(-5.6×10-3t). The calculated results by the equation are shown on the solid line in Fig.1. Comparing the solid curve with the experimental data,a conclusion can be drawn that the simulation of the kinetic model is satisfactory. 3.4 On the Optimal Initial [%Si In case C%S%S,there will be a lot of manganese in steel to be oxidized as shown in Fig.7.The product of MnO will be formed a quantity of 2MnO.SiO2,MnO.Al2O3 and so on with SiO,and Al2O,in the melting process,which are difficult to reduce or decompose.Furthermore,a decrease in interfacial tension between molten slag and steel with the increase of Mno will hinder the separation of slag from metal,as a result of this,there will be an increase of O]. On the other hand,in case C%S J>%S:,first of all,there will be not enough oxygen to react with carbon to perform VCD of necessary intensity. Secondly,a quantity of SiOz particles will be formed in the process of melting, these SiO2 particles have high melting point and are wettable,therefore,it is difficult to get them together and float off the molten steel,as a result of 498
〔 〕 产 牙 。 、产 ， 〔 〕 〔 〕 ， 〔 〕 一 犷二 。 〔 〕 甲 口 ‘ 凡 。 ， 仔 ’ 〔 〕 一 〔 〕 ， 〔 〕 〔 庵 〔 〕 。 一 一 ， ， 。 一 〕 ， 〔 〕 二 。 ， ， 二 ， ， 秃 二 。 。 。 ， 掩， 吞 二 壳 。 二 甲 。 一 。 存 一 ， 〔 〕 。 ， ， 洲尸 〔 〕 一 。 一 。 丁 一 。 ， 。 一 〔 〕曹 〔写 ‘ 〕 。 〔 ， 〕 ， · 宜 ， · 口 ， 住 · ， 几 ， ， 〔 〕 ， 〔 〕 。 〔 〕言 ， ， ， ， ， ， 口 声 尸
these two points,there will be an increase in oxygen too. 3.5 Effectiveness of Refining with Covering Slag Molten steel is often exposed to atmosphere in induction furnace due to: (1)The covering slag is in a semi-molten state on account of low temperature; (2)The convex meniscus of molten steel repeal the covering slag to the peri- phery of metal bath. 0.30 The exposed metal bath would be 0.27 oxidized by the reactions below: 0:24 O(UN) 0.21 〔Fe]+CO]=(FeO),log(as。o/〔%O) 09 0.18 =6320/T-2.734 (16) 0.15 cMn〕+〔O]=(MnO),log(MnO)/C%O) 0.12 =12760/T-6.230 (17) 0.09 0 0.060.12.0.180.20.30 From above equation,it is shown Initial〔sio% that the higher the temperature of steel Fig.7 The relation between initial and the higher the Feo and Mno in 〔%Si]。and〔%Mn]. slag are,the more theCOJis.Therefore, it is suggested that during refining with covering slag,(FcO+MnO)must be less than 1%and the melting point of covering slag should be lower than 1550C. Under these conditions,steel bath would be covered better,so [O]in the steel be lower. 4 Conclusions (1)By the use of VCD,CO]in molten GCr15 steel can be reduced to a value lower than 20ppm,generally,a mean value of 12.86ppm is obtained pro- vided that the treating time is longer than 10min,and the pressure in the furnace is lower than 0.133kPa. (2)The optimal silicon content [%Si]at the end of melting is 0.11%. In case %SiJo lower or higher %Si),the CO]after VCD would be higher. Hance,it is reasonable to control the initial silicon content within the range of 0.08~0.15%。 (3)Oxides,such as MgO,Al2Os etc,are the major sources supplying oxygen to the metal bath at higher vacuum degree.Hence,the time of VCD should be controlled within 10~15min. (4)In order to refine GCr15 steel with covering slag effectively,the melting point of CaO-Al2Os-CaF2 slag system should be lower than 1550C,and (FeO +MnO)lower than 1%. 499
， 、 。 一 卜 。 。 卜 月 山今召“ … 曰爪卜︸ 、 。 。 曰 冰。俄 后︸署︸︺。 。 。 ， 。 口 琦尹 洲 下一匹口 才 。 于 团 产 丫 叮 〔 互。 〔 〕 。 〔 〕 。 咨 〔 〕 〔 〕 ， 。 。 〔 〕 一 〔 〕 〔 〕 ， 〔 〕 一 。 ， ， 〔 〕 。 ， 、 、 ， ， ， 〔 〕 。 、 、 、 ， ‘ 。 卜
REFERENCES 1 Hammar O,ct al.Tool and Alloy Steel,1985;(1):4 2 Zhang Jian.Dalian Special Steels,1986:(2):1 3 Shigern Inoue,ct al.Tetsu-to-Hagane,1985;(12):s1081 4 Toshio Onoye,et al.Tefsu-to Hagane,1984;(12):s883 5 Cai Huaide.Master Thsis,BUIST,1988 6 Oiks G N,et al.Steel USSR,1960;(4):308 7 Yoshimoto Wanibc,ct al.Trans ISIJ,1983;(7):608 8 Shigenori Kawai.Tetsu-fo-Hagane,1977;1975 9 Qu Ying.Principles of Steelmaking,Metall.Press,1980:240 XX材材材烘州州以以X州然州州州州好州材湖州以州烘州州州州州以 New Silicon Nitride Ceramic Inserts ST-Si,N.ceramic ieserts are made with new type additives and advanced P/M techniques.It is suitable for machining those materials which are very dif- ficult to be machined with hard metals,such as wear-resistant alloyed cast irons, chilled irons,quenched steels,Ni alloys hard-facing Ni coatings etc.,and for high-rate machining of conventional irons and steels.In machining hard Ni cast irons (HRC 55-59),Cr 27 (HRC 59-62),and Cr15Mo3 (HRC 60-67),ST-Si,N. shows superior performances to other ceramic inserts. 以州州州州州州州州州州以以州州州州州州州烘州州州州汾材州州州州州州州州州以 500
产 ， 月 夕 ， ￡ ， ， 一 一 夕 ， 了 ， 一 “ 一 ” ， ， ， ， ， ， ， 。 ， ， “ 一 一 夕 ， 一 ， ， 尸 告讨 伏绪杖咭怡心咭 倪 备材 含尝心长褚备况 李心弓二心污二弋摇李李弓掀谧尝令弓浮心魂李亡考只毛片弓片毛李二绍只谧李今笔二心浦卜拭 片毛只魂 魂孚心魂 弓欠绳矜魂 毛片弓李之摇李右弓李牛弓李念弓 ， ， ， 尹叫 一 ， 一 ， 一 ， ， ， 一 ， 一 一 ， 一 ， 一 ， 一 名绳只弓李汉 李右摇 寸魂欠毛李尝弓吕名李心代李 谧 弓二伙 浮拭 二铭 李片 李李名吕魂 谧矛侣 李挑 兴弓寻伙 李尝毛二心名二心绍 魂李佣 李魂片弓二伙 李今弓二希魂欠魂李令摇二心弓李沈 拭绳矜弓