Although pulse-modulated plasma has overcome various problems encountered during the development of the high aspect ratio contact hole etching process, there is still a lack of understanding in terms of precisely how the pulse-modulated plasma solves the issues. In this research, to gain insight into previously observed phenomena, SiO₂ etching characteristics were investigated under various pulsed plasma conditions and analyzed through plasma diagnostics. Specifically, the disappearance of micro-trenching from the use of pulse-modulated plasma is analyzed via self-bias, and the phenomenon that as power off-time increases, the sidewall angle increases is interpreted via radical species density and self-bias. Further, the change from etching to deposition with decreased peak power during processing is understood via self-bias and electron density. It is expected that this research will provide an informative window for the optimization of SiO₂ etching and for basic processing databases including plasma diagnosis for advanced plasma processing simulators.

The cutoff probe (CP) is a precise diagnostic technique that measures electron density using two cylindrical probe tips and holders inserted in a low-pressure plasma. However, a recent study has reported that the insertion of the probe tips and holders can perturb the plasma condition and act as a source of error in the measurement of electron density. This study proposes a planar cutoff probe (PCP), which does not have invasive probe tips and holders, to minimize the plasma perturbation during plasma measurement. We embed the PCP on the existing substrate chuck of the plasma chamber instead of inserting an additional probe holder into the chamber. The adjusted conditions for the radii of the probe’s antennas and the gap distance between the two probe antennas are obtained via an electromagnetic simulation. The comparative experiment results show that the inferred electron densities of the optimized PCP are in good agreement with those of the previous CP and within reasonable discrepancies stemming from the differences in measurement positions. We believe that the PCP, which can be embedded in the inner surface of the chamber and measure the electron density during the plasma (wafer) process, would be an alternative technical tool to realize advanced process control of plasma based on non-invasive diagnostic methods.

In Ho Seong , Jang Jae Lee , Chul Hee Cho , Yeong Seok Lee, Si Jun Kim , and Shin Jae You. Applied Science and Convergence Technology 2021;30:176-82. https://doi.org/10.5757/ASCT.2021.30.6.176

Recently, the uniformity in the wafer edge area that is normally abandoned in the fabrication process has become important for improving the process yield. The wafer edge structure normally has a difference of height between wafer and electrode, which can result in a sheath bend, distorting important parameters of the etch, such as ionic properties, resulting in nonuniform etching. This problem nowadays is resolved by introducing the supplemented structure called a focus ring on the periphery of the wafer. However, the focus ring is known to be easily eroded by the bombardment of high-energy ions, resulting in etch nonuniformity again, so that the focus ring is a consumable part and must be replaced periodically. Because of this issue, there are many simulation studies being conducted on the correlation between the sheath structural characteristics and materials of focus rings to find the replacement period, but the experimental data and an analysis based on this are not sufficient yet. In this study, in order to experimentally investigate the etching characteristics of the wafer edge area according to the sheath structure of the wafer edge, the etching was performed by increasing the wafer height (thickness) in the wafer edge area. The result shows that the degree of tilt in the etch profile at the wafer edge and the area where the tilt is observed severely are increased with the height difference between the wafer and electrode. This study is expected to provide a database for the characteristics of the etching at the wafer edge and useful information regarding the tolerance of the height difference for untilted etch profile and the replacement period of the etch ring.

The previously proposed circuit model has proven to be a powerful tool to analyze and improve the characteristics of the cutoff probe. This model revealed the mechanism of formation of the transmission microwave frequency spectrum of the cutoff probe, and allows for a more precise measurement of electron density by taking into account electron-neutral collision frequency and sheath width. However, a recent study has found that there is a discrepancy between the circuit model and 3D full E/M wave (EMW) simulation, in that the circuit model is applicable to the measurement of low-density plasma but not high-density plasma; this indicates that the basis of the circuit model for the cutoff probe seems to be incomplete (Kim et al 2015 2015 KPS Fall Meeting ). To resolve this problem, the current paper proposes a new model of the cutoff probe, named the transmission line (TL) model, based on TL theory. Results of the TL model are compared with EMW simulation, with the TL model found to reproduce the results of EMW simulation in both high-density as well as low-density plasma.

Although the recently developed cutoff probe is a promising tool to precisely infer plasma electron density by measuring the cutoff frequency ($f_{\mathrm{cutoff}}$) in the S${}_{21}$ spectrum, it is currently only applicable to low-pressure plasma diagnostics below several torr. To improve the cutoff probe, this paper proposes a novel method to measure the crossing frequency ($f_{\mathrm{cross}}$), which is applicable to high-pressure plasma diagnostics where the conventional $f_{\mathrm{cutoff}}$ method does not operate. Here, $f_{\mathrm{cross}}$ is the frequency where the S${}_{21}$ spectra in vacuum and plasma conditions cross each other. This paper demonstrates the $f_{\mathrm{cross}}$ method through three-dimensional electromagnetic wave simulation as well as experiments in a capacitively coupled plasma source. Results demonstrate that the method operates well at high pressure (several tens of torr) as well as low pressure. In addition, through circuit model analysis, a method to estimate electron density from $f_{\mathrm{cross}}$ is discussed. It is believed that the proposed method expands the operating range of the cutoff probe and thus contributes to its further development.