低气压双频容性耦合电负性等离子体物理特性的研究

发布时间：2018-05-11 02:35

  本文选题：双频容性耦合等离子体 + 电负性气体　； 参考：《大连理工大学》2015年博士论文

【摘要】：双频容性耦合等离子体(Dual-Frequency Capacitively Coupled Plasmas, DF-CCPs)能够独立控制轰击到基片表面的离子通量和能量,从而可以缓解刻蚀过程中刻蚀速率与器件损伤的矛盾和减弱薄膜沉积过程中的充电效应,因而在国际上被广泛地应用到半导体芯片的刻蚀和沉积工艺中。电子密度及离子能量作为等离子体最基本的参数,对于理解等离子体的特性及优化工艺过程具有重要的意义。到目前为止,有关DF-CCPs的实验研究主要以Ar放电为主,而对于在实际刻蚀工艺中被大量使用的电负性气体,像O2、 CF4, c-C4F8等,实验研究工作较少。本文从实验上系统地研究了在O2、Ar/O2、Ar/CF4、 Ar/O2/CF4放电中,控制变量(高频功率、低频频率和功率、气压等)对电子密度和离子能量分布等的影响。同时,采用了流体模型及PIC/MC (Particle-In-Cell and Monte Calro)模型与实验结果进行对比验证。本论文研究对象大多是具有氧化性或腐蚀性的气体,传统的诊断手段(如,Langmuir探针)无法有效地使用,然而微波发卡探针和四极杆质谱仪基本上不受腐蚀性气体影响,可以较准确地测量电子密度和离子能量分布。在第1章,简单地介绍了低温等离子体在集成电路工业中的应用、低温射频等离子体源、DF-CCP的研究进展及热点问题,给出了本论文的研究内容安排。在第2章,简单地介绍了双频容性耦合放电装置的关键参数,详细地介绍了实验中所使用的诊断手段(微波发卡探针、光探针、四极杆质谱仪、电压-电流探测器)的原理、结构和使用方法。其中,微波发卡探针可以测量电子密度,光探针可以测量特定谱线的发光强度,四极杆质谱仪能够分辨出不同质荷比的离子并给出相应的离子能量分布,电压-电流探测器可以测量电压、电流、功率等电学参数。最后,简单介绍了本论文采用的数值模型,包括流体模型及PIC/MC模型。在第3章,基于上述的实验设备和诊断手段,系统地研究了在O2和Ar/O2放电中高频功率、低频功率、气压对电子密度的影响。研究表明：在O2放电中,电子密度主要由高频功率决定,随着高频功率的增加电子密度线性增大；在较高高频功率下,随着低频功率的增加电子密度减小,而当高频功率较低时电子密度会随着低频功率增加而增大；随着气压的升高,电子密度先快速增大而后缓慢减小；Ar的添加导致电子密度的增大,但不影响电子密度随控制变量的变化趋势。同时,采用了PIC/MC模拟对实验结果进行验证,二者取得较好的一致性。在第4章,系统地研究了在Ar/O2放电中低频频率、低频功率、气压等对离子能量分布的影响。研究表明：低频频率和低频功率是影响离子能量分布的主要参数,并且二者的作用相反,即随着低频频率增大,能宽逐渐变窄,高能峰向着低能区移动,而随着低频功率的增大,能宽逐渐变宽,高能峰向着高能区移动；气压的升高导致激烈的共振电荷交换碰撞,产生更多的低能电子。同时,采用了PIC/MC模拟对部分实验结果进行验证,二者给出相同的变化趋势,但在数值方面二者存在着差异,针对差异分析了可能的原因。在第5章,系统地研究了在Ar/CF4和Ar/O2/CF4放电中高频功率、低频功率、气压对电子密度和离子能量分布的影响。研究表明：在Ar/CF4放电中,电子密度主要由高频功率决定,低频功率影响很小,O2的添加会导致电子密度的下降但不会影响电子密度随控制变量的变化趋势；离子能量主要由低频源(频率和功率)决定,高频功率对其影响较小,气压的升高会导致能量分布中低能离子的增多。在第6章,研究了在O2放电中驱动频率对电子密度的影响。在相同的输入功率下,不同驱动频率的等离子体吸收功率不同,很大一部分输入功率耗散在匹配网络,在100MHz下,损失的功率高达50%以上。在固定等离子体吸收功率的条件下,电子密度随着频率从13.56 MHz升高到40.68 MHz而增大,当频率进一步从60 MHz升高到100 MHz时,电子密度在2.6 Pa和13.3 Pa下呈现不同的变化趋势。在固定极板电压的条件下,电子密度随驱动频率的增大先增大后减小,在40.68 MHz时达到最大值。在第7章中,给出了本论文的主要结论、创新点及未来的工作计划。
[Abstract]:The dual frequency capacitive coupled plasma (Dual-Frequency Capacitively Coupled Plasmas, DF-CCPs) can independently control the ion flux and energy of the bombardment on the substrate surface, thus alleviating the contradiction between etching rate and device damage and reducing the charge effect in the process of thin film deposition, so it is widely used in the world. In the etching and deposition process of semiconductor chips, the electron density and ion energy are the most basic parameters of the plasma. It is of great significance to understand the characteristics of the plasma and optimize the process process. So far, the experimental research on DF-CCPs is mainly based on the Ar discharge, and it has been widely used in the actual etching process. Electronegative gases, such as O2, CF4, c-C4F8 and so on, have less experimental work. In this paper, the effects of control variables (high frequency power, low frequency frequency and power, pressure and so on) on the electron density and ion energy distribution in O2, Ar/O2, Ar/CF4, Ar/O2/CF4 discharge are systematically studied. The fluid model and PIC/MC (Particle-) are used. The In-Cell and Monte Calro) model is compared with the experimental results. Most of the objects in this paper are oxidizing or corrosive gases, and the traditional methods of diagnosis (such as Langmuir probes) can not be used effectively. However, the microwave hairpin probe and quadrupole mass spectrometer are not affected by corrosive gases, and can be more accurately measured. In the first chapter, the application of low temperature plasma in the integrated circuit industry, the research progress and hot issues of DF-CCP are briefly introduced. The research content of this paper is given. In the second chapter, the key parameters of the dual frequency capacitive coupling discharge device are briefly introduced. The principle, structure and using method of the diagnostic means (microwave hairpin probe, optical probe, quadrupole mass spectrometer, voltage current detector) used in the experiment are introduced in detail. In this method, the microwave hairpin probe can measure the electron density. The light probe can measure the luminous intensity of the specific line. The quadrupole mass spectrometer can distinguish the different mass charge. The corresponding ion energy distribution is given and the voltage current detector can measure the electrical parameters such as voltage, current and power. Finally, the numerical model used in this paper, including the fluid model and the PIC/MC model, is briefly introduced. In the third chapter, the O2 and Ar/O2 discharge based on the above experimental setup and diagnosis are systematically studied. The effect of high frequency power, low frequency power and pressure on electron density. The study shows that in O2 discharge, the electron density is mainly determined by high frequency power, and the electron density increases linearly with the increase of high frequency power; at high frequency power, the electron density decreases with the increase of low frequency power, while the electron density will be low when the high frequency power is low. With the increase of low frequency power, the electron density increases quickly and then decreases with the increase of the pressure. The addition of Ar leads to the increase of the electron density, but does not affect the change trend of the electron density with the control variable. At the same time, the PIC/MC simulation is used to verify the experimental results, and the two are in good agreement. In the fourth chapter, The influence of low frequency frequency, low frequency power and pressure on the ion energy distribution in Ar/O2 discharge is systematically studied. The study shows that low frequency and low frequency power are the main parameters affecting the ion energy distribution, and the effect of the two is opposite, that is, as the frequency increases, the energy is gradually narrowed, and the high energy peak moves toward the low energy region. With the increase of low frequency power, the energy width gradually widened and the high energy peak moved toward the high energy region; the increase of the air pressure led to the intense resonance charge exchange collision and more low-energy electrons. At the same time, the PIC/MC simulation was used to verify the results of some experiments. The two people gave the same trend of change, but there were differences between the two in numerical aspects. The possible reasons for the difference are analyzed. In the fifth chapter, the influence of high frequency power, low frequency power and pressure on the electron density and ion energy distribution in Ar/CF4 and Ar/O2/CF4 discharge is systematically studied. The study shows that in Ar/CF4 discharge, the electron density is mainly determined by high frequency power, low frequency power is small, and the addition of O2 leads to the density of electron density. The decrease of the degree of degree does not affect the change trend of the electron density with the control variable; the ion energy is mainly determined by the low frequency source (frequency and power), the high frequency power has little influence on it, the increase of the pressure will lead to the increase of the low energy ion in the energy distribution. In the sixth chapter, the influence of the driving frequency on the electron density in the O2 discharge is studied. Under the input power, the plasma absorption power of different driving frequencies is different. A large part of the input power is dissipated in the matched network. Under 100MHz, the loss power is up to 50%. Under the condition of fixed plasma absorption power, the electron density increases with the frequency from 13.56 MHz to 40.68 MHz, when the frequency is further from 60 M. When the Hz is raised to 100 MHz, the electron density varies under 2.6 Pa and 13.3 Pa. Under the condition of the fixed plate voltage, the electron density increases first and then decreases with the increase of the driving frequency, and reaches the maximum at 40.68 MHz. In the seventh chapter, the main conclusions, innovation and future work plan of this paper are given.

【学位授予单位】：大连理工大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：O53
，

本文编号：1872050