|
abstract 摘要
The complex refractive index of SiGe alloys at 632.8 nm has been measured as a function of the Ge concentration using in situ ellipsometry while the material is slowly removed from a silicon substrate using reactive-ion etching (RIE).
利用反应离子刻蚀(RIE)缓慢刻蚀材料表面的同步椭偏仪测量了作为Ge浓度函数的在632.8 nm波长下SiGe合金的复数折射率。
Homogeneous, strained epitaxial SiGe films on silicon substrates were used.
在硅基板上使用了均匀、应变的外延SiGe薄膜。
The Ge concentration was obtained by Rutherford backscattering.
Ge浓度由卢瑟福背散射得到。
If an unknown SiGe structure is etched with RIE, in situ ellipsometry yields combinations of the ellipsometric anglesΨ and Δ with time.
若使用RIE刻蚀未知SiGe结构,同步椭偏仪结合椭偏角度Ψ和Δ与时间得出结果。
Starting at the Si substrate, these points are, on a point-to-point basis, converted into combinations of complex refractive index and depth in a numerical procedure.
从硅衬底开始,这些点基于点对点,采用数字过程转换成复数折射率和深度的组合。
For this inversion of the ellipsometry equations, the known relation between the real and the imaginary part of the refractive index of SiGe is used.
通过椭偏方程转换,使用已知SiGe折射率的实部和虚部关系。
Finally the refractive indices are converted into Ge concentrations.
最后转换折射复数为Ge的浓度。
Thus the depth profile of the Ge concentration in an unknown epitaxial SiGe structure can be inferred from an in situ ellipsometric measurement during RIE of the unknown structure.
由此未知SiGe 外延结构的Ge浓度纵向分布可以由对未知结构的反应离子刻蚀同步椭偏测量推断出。
The obtained resolutions in depth and Ge concentration are 0.3 nm and 0.3%, respectively.
所得到的深度和Ge浓度分辨率为0.3 nm和0.3%
Il. EXPERIMENT
实验
The experiments were done in a reactor equipped with a 12 in. water-cooled, quartz-covered rf-powered (13.56 MHz) electrode.
本实验是在一个12寸水冷石英外罩的RF(13.56 MHz)功率电极的反应腔中进行的。
The CF4, gas was fed through a MKS mass flow controller.
CF4气体通过MKS的质量流量控制器流入。
The flow rate was kept at 100 sccm at all times.
气体流量总保持在100 sccm。
A throttle valve in the pumping line to a 500 L/s Leybold turbomolecular pump kept the pressure at 25 mTorr in all experiments.
在所有实验中在排气管路上的蝶阀和500 L/s Leybold 涡轮分子泵保持保持工艺压力在25 mTorr。
The wafer was placed on the powered electrode.
硅片被放置于功率电极上。
The power used for the reported experiments varied between 10 and 400 W.
本实验的功率适用范围在10到400 W。
The reactor is capable of working in the RIE (reactive ion etching) mode and also in the plasma etching mode.
反应腔拥有在RIE (反应离子刻蚀)模式和等离子刻蚀模式下工作的能力。
In the latter case the pressure is raised and the wafer is mounted on the grounded electrode.
后者的压力升高硅片放置在接地的电极上。
For the experiments reported in this article we have exclusively used the RIE mode, but we kept the power at a comparatively low level.
在本文实验报告中我们单独使用RIE模式,但我们将功率保持在一个相对低的水平。
When the power is raised, the flux and energy of the ion bombardment go up.
当功率上升,离子轰击能量流增加。
This in turn causes a more profound modification of the etched surface, e.g., due to roughening or amorphization.
这将导致对于刻蚀表面的严重修正,譬如粗糙度和非晶化。
In the plasma-etching mode the passivation layer that is formed on the surface is likely to be thicker than in the case of RIE.
在等离子刻蚀模式下,表面形成了比使用RIE情况下更厚的钝化层。
In order to keep the impact of the plasma on the surface properties of the etched sample as low as possible, the low-power RIE conditions are preferred.
为了尽可能降低灯离子体对表面性质的影响,首选低功率RIE。
Under these conditions the etch rate can be very low, e.g., about 0.05 nm/s..
在此条件下,表面刻蚀速率可以非常慢,如大约0.05 nm/s.
Since the measuring speed of the ellipsometer is limited (in our case about one measurement per second), a low etch rate implies a high depth resolution (if the intrinsic precision of the ellipsometer is better than the thickness of the material etched away between two data points).
因为椭偏仪的测量速度是有限的(在本文中每秒测量一次),低刻蚀速率意味高纵向分辨率。(如果椭偏仪本身的精确度比两个数据点之间刻蚀掉的材料膜厚要小)
The ellipsometer used in these experiments is an automated rotating compensator ellipsometer in the polarizer sample- compensator-analyzer (PSCA) configuration.
在这些实验使用的椭偏仪是偏光样品补偿分析(PSCA)构造的自动旋转补偿椭偏仪。
The polarizer is a Rochon prism, the analyzer a Glan- Thomson prism.
偏光镜为Rochon棱镜,分析器为Glan- Thomson棱镜。
A He-Ne laser is used as a light source, and a photomultiplier tube as a detector.
使用He-Ne激光作为光源,光电倍增管作为探测器。
In order to minimalize the influence of the detection system on the measurement, the light passes a depolarizing diffusor located immediately behind the analyzer.
为了减少测量中检测系统的影响,光通过在分析仪直后的去偏光散布仪。
An optical fiber transports the light to the photomultiplier.
光纤传输光线到光电倍增管。
The signal of the photomultiplier is sampled 256 times each rotation of the compensator.
光电倍增管信号在每次补偿器旋转时采样256次。
The required trigger signals as well as an index pulse are supplied by an optical encoder, mounted on the shaft of the rotating compensator.
需要的触发信号和指数脉冲由光学编码器提供,放置在旋转的补偿器的轴上/
The compensator is unsupported mica.
补偿器是无支持的云母。
The data are taken by a 16 bit analog-to digital converter (ADC), located in an Analog Device RTI850 interface card in an IBM PC/AT.
数据由嵌在IBM PC/AT的模拟装置RTI850接口卡16位模数转换器(ADC)采集。
The computer also performs the Fourier transform of the data, which is necessary to extract the ellipsometric angles Ψ and Δ.
计算机也可以实施对数据的傅立叶变换,这对于提取椭偏仪角度Ψ和Δ非常有用,
The minimum time between two measurements is 0.5 s.
两次测量间最小时间为0.5 s。
The instrument has a precision of about 0.01” in Ψ and in Δ.
仪器的Ψ 和Δ的精确度为0.01”。
The principle of a rotating compensator, instead of a rotating analyzer, offers, among others, the advantage of a nonambiguous determination of Δ, with approximately the same precision for all Δ values.
旋转补偿器原理代替其他提供的旋转分析的优点在于,明确的决定Δ,对所有Δ值有合适的精确度。
The angle of incidence was chosen between 74.0” and 75.0”.
入射角度选择在74.0”到75.0”之间。
This range was chosen because it offers a good compromise between surface sensitivity and stability (which will be discussed more extensively later).
选择这一范围的原因是它提供了在表面敏感度和稳定性(这将在之后进行更广泛的讨论)之间很好的折衷。
The SiGe layers have been grown on 125-mmdiam Si( 100) wafers by the UHV/CVD (ultra-highvacuum/ chemical-vapor-deposition) technique.
SiGe层由UHV/CVD(高真空化学气相淀积)技术生长在Si( 100)的硅片上。
The system is equipped with two hot-wall horizontal quartz reactors, with load-lock and transfer chambers.
系统装备有两个热壁水平石英反应器,装载和传送腔。
The base pressure in the reactors is 2.5 X 10E-9 Torr at 550 “C.
反应器在550度下的基准压力是2.5 X 10E-9 Torr。
Prior to loading the cassette in the system, the wafers were cleaned with a modified RCA procedure, then dipped in a 10% HF solution for 10 s.
在将片盒装载到系统之前,硅片由RCA过程清洗,然后在10% HF溶液中浸泡10 s。
A gas mixture of pure silane and of germane (9.4%/He) was used for the deposition.
用纯硅烷和锗烷(9.4%/He)的混合气体来淀积。
The silane flow rate was kept constant at 14.2 sccm, corresponding to a partial pressure of 1.3 mTorr.
硅烷流量保持固定在14.2sccm,对应的分压为1.3 mTorr。
The growth temperature was 518 “C, as probed by a profile thermocouple located inside the reactor.
生长温度在518 C,由位于反应腔下的热电耦检测。
These growth conditions, similar to the ones used by Meyerson, Uram, and Le- Goues, provide excellent epitaxial quality.
这些生长条件,与Meyerson, Uram, 和Le- Goues,使用的相同,提供了极好的外延质量。 |
|