磁场中直拉单晶工艺技术进展

附加磁场的熔体直拉法(MCZ法)是近年开发的一种颇有前途的生长优质硅单晶的新工艺技术。由于磁场抑制了硅熔体中的热对流,减少了熔体的温度波动,因而能减少晶体的生长条纹,降低或控制晶体中的氧含量,提高晶体结构完整性,显著降低单晶硅片的翘曲度,并能拉制高阻硅单晶。从而为超大规模集成电路提供了优质硅材料。本文系统论述了MCZ法的发展历史、方法与原理、工艺与设备、近年的新成果和发展前景。     

CUSP磁场对直拉硅单晶氧浓度分布影响的数值模拟

利用有限元分析软件对φ300mm直拉硅单晶生长过程进行模拟,分析了保持CUSP磁场对称面与熔体坩埚界面交点处的径向分量不变的情况下,硅单晶中氧浓度分别随CUSP磁场通电线圈距离、通电线圈半径的变化规律.随着通电线圈距离和半径的增大,晶体熔体固液界面氧浓度均逐渐降低.随着通电线圈距离和半径的增大,硅熔体径向磁场强度逐渐增大,对坩埚底部熔体向晶体熔体固液界面处对流的抑制作用加强,固液界面下方熔体轴向流速减小,使得从坩埚底部运输上来的富氧熔体减少,继而固液界面处的氧浓度降低.随着线圈距离和半径的增大,为保持所需磁场强度,施加电流也逐渐增大,从而能耗增大,与增大通电线圈距离相比,增大通电线圈半径所需的电流较大.通过实验,将CUSP磁场对单晶中氧浓度分布影响的数值模拟结果与实际晶体生长进行了对比,实验结果验证了数值模拟的结果.     

切克劳斯基硅熔体中液流图式研究

单晶硅目前是电子工业中最重要的材料。尽管已经知道用直拉法(SZ法)生长的硅单晶的某些特性——例如旋涡缺陷——与熔体流动情况有密切关系,因而迫切要求了解它的熔体流动图,但由于硅熔体是强腐蚀剂,关于它的流体行为的物理测量很困难,精确地获得硅熔体流动图目前尚不可能。近年来,国外一些学者引用流体动力学对 CZ 法硅熔体做了很多理论估计。在这个     

MCZ硅单晶

MCZ硅单晶即磁场直拉硅单晶,于1980年由日本索尼公司首先拉制成功,至今已五年多,在这段时间里,无论对磁场直拉工艺的研究,还是对该材料在应用领域的研究都取得了很大进展。当器件集成度进入兆位时代后,为提高器件性能和合格率,对硅单晶的质量要求就更加严格,由于MCZ硅的优越性非常突出,     

直拉法硅表面层熔体流动所遵循的线性规律

本文定量地研究了硅自由表面层熔体的流动规律,用大量实验数据确立了自由表面层熔体径向流动速度与熔体纵横比(D/H)值的线性对应关系。根据这个关系可以成功地解释和处理硅单晶生长过程中的许多相关问题。     

直拉硅单晶制备中来自于石英坩埚的受主杂质污染定量分析

半导体硅单晶制备中一般要求掺入一定量的杂质以控制其电阻率。但是,由于坩埚受熔体侵蚀引入杂质污染,影响硅棒的实际电阻率,降低产品合格率。特别对于连续拉晶过程中,长时间的熔体侵蚀造成硅棒尾部杂质含量急剧升高。讨论了由于石英坩埚中受主杂质污染对硅单晶棒电学性能的影响,并对杂质渗入量进行了定量分析。结果表明,随着晶体生长时间的延长,石英坩埚的Ba涂层被破坏,杂质的渗入速率快速增加。试验中杂质总渗入量超过700μg。     

CZ硅中金属杂质的起源

ＣＺ硅中金属杂质的起源为促进晶体净化技术的发展，日本三菱材料公司的科研人员研究了直拉单晶硅中金属杂质的来源。实验计算了从不同杂质来源混入熔体中的金属杂质的质量，并计算了熔硅与硅单晶中杂质的浓度；为了证实其计算结果，实验分析了单晶生长后剩余熔体样品。比...     

水平超导磁场直拉法硅单晶熔体过热现象

基于水平超导磁场及特定热场结构，对直拉法硅单晶生长过程中的熔体过热现象进行了数值模拟和实验研究。在数值模拟中发现了熔体对流和温度变化的特殊趋势，随后在实验中发现了相应位置晶体直径突然收缩现象，依此推测拉晶过程中生长界面附近熔体温度及温度梯度的变化是引起晶体直径收缩的直接原因。此外发现改变熔体温度梯度将改变熔体过热现象发生的位置和幅度，模拟中对两种极端的功率条件进行了计算，并在实验中观察到了预期的差别。水平超导磁场对于熔体对流的抑制作用、特定的热场结构和工艺条件，都是引起熔体过热和晶体直径收缩现象发生的原因。     

硅单晶微缺陷研究进展

目前很多无位错硅单晶经过Sirtl腐蚀剂或改良的Sirtl腐蚀剂腐蚀后,在横断面上可看见呈漩涡状分布的浅底蚀坑。所谓硅单晶中微缺陷,目前主要就是指的这种缺陷。在硅器件制备热处理中,这种缺陷上将产生金属微沉淀、绝缘微沉淀及层错,损害硅器件性能。例如,使P-N结反向漏电流增大,产生局部微等离子击穿,使电荷耦合器件的寄存时间缩短,对集成电路影响更大。因此研究这种缺陷的机理,探索消除它们的方法,对于发展硅材料、提高硅器件性能具有重要意     

Improving Percentage of Finished Product of FZ NTD Silicon

本文根据中子嬗变掺杂的原理、工艺,着重分析了影响NTD硅成品率的因素:命中率及断面均匀性.通过采用换位法、旋转辐照工艺,较大幅度地提高了NTD硅单晶的命中率、断面均匀性及成品率.     

锰硅合金沉淀脱磷工艺制度的优化

选用钙系脱磷剂、CaO—CaF2渣,对锰硅铁合金进行沉淀脱磷,着重研究脱磷剂加入量和锰硅铁合金中硅含量增量对脱磷效果的影响。结果表明：当脱磷剂加入量由2．5％增加到7．5％时,脱磷效果增加5．56％,而合金中硅含量由19％增加到25％时,脱磷效果将至少增加23．22％,说明锰硅合金中硅含量增加对脱磷效果的作用远大于单一增加脱磷剂的作用。合金熔体中硅含量的增加降低了氧势,同时提高了磷的活度,这将有助于脱磷剂中的钙以Ca2P3而非CaO形式进入渣中;此外,采用沉淀脱磷工艺的锰硅合金熔体中碳含量也有明显下降,有利于中、低碳锰铁生产中碳含量的控制。     

Review of Microstructure Evolution in Hypereutectic Al–Si Alloys and its Effect on Wear Properties

Al–Si alloys with silicon content more than 13 % are termed as hypereutectic alloys. In recent years, these alloys have drawn the attention of researchers due to their ability to replace cast iron parts in the transportation industry. The properties of the hypereutectic alloy are greatly dependent on the morphology, size and distribution of primary silicon crystals in the alloy. Mechanical properties of the hypereutectic Al–Si alloy can be improved by the simultaneous refinement and modification of the primary and eutectic silicon and by controlling the solidification parameters. In this paper, the effect of solidification rate and melt treatment on the evolution of microstructure in hypereutectic Al–Si alloys are reviewed. Different types of primary silicon morphology and the conditions for its nucleation and growth are explained. The paper discusses the effect of refinement/modification treatments on the microstructure and properties of the hypereutectic Al-Si alloy. The importance and effect of processing variables and phosphorus refinement on the silicon morphology and wear properties of the alloy is highlighted.     

固相辅助电极型传感器的物理模型

用研制的Mg₂SiO₄(s)+MgO(s)硅传感器,对高碳冶金熔液中的硅进行了测量,发现辅助电极的物相结构和熔体中的氧含量对测量结果有明显的影响.首先提出了辅助电极型传感器的物理模型,导出了辅助电极的局域平衡氧势与熔体氧势和硅传感器测量结果的关系式,给出了覆盖常数和结构常数的物理意义.用实验数据拟合得到了该传感器的结构常数为6.66;用该模型对小于1的不同覆盖常数的硅传感器测得同一熔体的氧势进行了校正,得到了一致的局域平衡氧势值,证明了模型与实际吻合得相当好.     

Formation of silica and silicates during the solidification of Fe−Si−O alloys

Small (35 g) ingots of Fe?Si?O alloys have been solidified under controlled conditions and the form, distribution, and compositions of the oxides in the cast structures are reported. The composition of a melt was established by charging a preselected amount of silicon, 0.002 < pct Si < 0.7, and equilibrating the melt with the silica crucible at 1550° or 1650°C. The concentration of FeO and the morphology of the oxide particles in the cast structures indicate that during solidification the process of segregation causes the oxygen potential in the interdendritic liquid to increase over that in equilibrium with the silicon content of the liquid. It appears that nuclei for the formation of the oxide phases, which probably are in the liquid prior to the onset of solidification, become surrounded by solid iron and are made ineffective. It is possible at lower silicon levels that homogeneous nucleation may result by the melt becoming sufficiently supersaturated in oxygen to form FeO or an iron-silicate liquid phase.     

Oxygen solubility in silicon-containing Fe-Co melts

Thermodynamic analysis of the oxygen solutions in silicon-containing Fe-Co melts is performed. The equilibrium constant of silicon deoxidation of iron-cobalt melts, the activity coefficients for infinite dilution, and the interaction parameters for melts differing in composition are determined. The dependences of the oxygen solubility in the melts under study are calculated for different cobalt and silicon contents. The deoxidizing capacity of silicon increases substantially as the cobalt content in a melt increases. The curves of oxygen solubility in Fe-Co melts have a minimum; the minimum oxygen solubility shifts to a low silicon content as the cobalt content in the melts increases. The silicon contents for the minima in the curves of oxygen solubility and the minimum oxygen concentrations corresponding to the silicon contents are determined.     

Sintering with Kovdor magnetite concentrates in the batch

Electron-microscope data show that the magnetite crystals in iron ore have different microstructure, depending on the temperature and time of ore formation. Thus, in sedimentary–metamorphic iron quartzites and magmatic skarns, the structure of the magnetite crystals is homogeneous and the composition is close to stoichiometric. In Kovdor ore, the magnetite crystals are heterogeneous. The matrix contains isomorphic Al, Mg, Ti, and other impurities as individual spinel microphases. The reduction of magnetite crystals in conditions that resemble sintering indicates that heterogeneous crystals disintegrate on sintering, with the formation of two ore phases: solid solutions of magnetite and wustite that are not involved in liquid-phase strengthening of the sinter. In the final stage of fluxed-sinter production, calcium–silicon silicate binders of melilite composition are formed in the product in place of the melt; these binders are not strong. On the basis of the research findings, it is important, in assessing iron-ore fields, to pay attention not only to the content of iron and silicon oxides in the ore but also to the structure of the magnetite crystals, since the iron in the magnetite determines the direction of melt formation in processing.     

VOD炉冶炼20Cr13不锈钢硅脱氧工艺

 以实际工艺流程30 t EAF-VOD-LF-模铸冶炼20Cr13不锈钢为背景,通过热力学计算,对VOD精炼硅脱氧处理后,重新造渣过程中炉渣碱度、合金含量以及钢液温度对钢液中平衡氧含量的影响进行了研究。研究结果表明：钢液中的氧含量随炉渣碱度的增加呈现明显的减小趋势,实际生产中炉渣碱度一般应确保在2以上;为了保证添加合金进行脱氧的冶炼效果,应保证钢液中的硅的质量分数在0.3%以上;降低冶炼温度有利于降低钢液平衡氧含量;炉渣碱度对平衡溶解氧含量和实际生产全氧含量的影响规律基本相同,拟合出平衡溶解氧含量与实际全氧含量之间存在线性关系。     

生产工艺及成分体系对含磷无取向硅钢磁性能的影响

结合工业化生产的中低牌号无取向硅钢,探讨了生产工艺及成分体系对含磷无取向硅钢磁性能的影响。结果表明,磷元素具有晶粒细化作用,能够抑制热轧组织再结晶,以及减少成品晶粒尺寸,因而会对磁性能产生显著影响。无铝钢中,随着磷含量增加,消除应力退火前、后,磁感变化不大,而铁损逐渐上升;含铝钢中,无论硅含量高低,随着磷含量增加,磁感均单调、快速递减,但随着硅含量升高,磷含量对铁损的影响程度逐渐变弱。采用中间退火处理后,少量的磷含量便能有效改善磁感,而对铁损影响不大。