Gon-Ho Kim

In low-temperature plasma diagnostics, optical emission spectroscopy (OES) is widely employed as a non-invasive technique for analyzing electron energy distribution functions (EEDFs). Conventional OES methods typically assume a pre-selected EEDF model, chosen in advance based on domain knowledge of the electron heating physics of the target plasma. However, such assumptions limit diagnostic accuracy and may introduce errors when the actual electron heating physics deviates from the assumed model. This study proposes a physics-informed model-selective diagnostic method that identifies the most appropriate non-Maxwellian EEDF by quantitatively comparing emission line ratios predicted by a collisional-radiative model with experimental OES spectra. Three analytic EEDF models—generalized Maxwellian, bi-Maxwellian, and beam-shifted Maxwellian—are considered as candidates. The method was applied and validated in inductively coupled plasma and VHF-CCP systems exhibiting distinct electron heating characteristics. The validity was confirmed through comparison with Langmuir probe measured EEDF and a physics-based analytic model within 10%. The proposed method significantly improves diagnostic accuracy by reducing EEDF estimation error by more than a factor of five and lowering reaction rate uncertainty by over an order of magnitude.

C2C (chamber-to-chamber) matching of the plasma processing reactors, called tool-to-tool matching, is one of the most important procedures to perform mass production in the organic light emitting diode display or semiconductor manufacturing facilities. Plasma processes, which adopt the RF power, are governed by the generation rate of reactive radicals and ions and their transportation to the reacting surfaces. Therefore, the characteristics of plasma heating, which determine the energy gain and relaxation of the electrons, are important information for performing C2C matching in the electron impact collisional inelastic processes dominated plasmas. In this study, we introduced the plasma heating characterizing PI (plasma information) index by using the limited sensor data accessible in the mass-producing fab with the help of a simple machine learning methodology to extract the information about the plasma heating properties. This index is applied to the C2C matching of the large-area capacitively coupled RF discharge-based PECVD (plasma-enhanced chemical vapor deposition) reactors successfully. In addition to the PI parameterization of the plasma heating characteristics in the large-area inductively coupled discharge applied plasma etcher introduced in the previous study, we could introduce the plasma heating characterized PI index for the PECVD reactors. From this, we could validate its utility to the artificial intelligence modeling, such as the plasma information-based virtual metrology and plasma information-based advanced process control, to optimize the productivity of the mass-producing factory.

This study developed Plasma Information-based Virtual Metrology (PI-VM) to predict etching process results and analyze process phenomena. Using a dual-frequency capacitively coupled plasma (CCP) etcher with C4F8/Ar/O2 plasma, we etched low aspect ratio (AR) trench patterns in amorphous carbon layer (ACL) hard masks and SiO2 molds, and developed the PI-VM statistically by integrating plasma information (PI) variables that reflect domain knowledge. The passivation effect of fluorocarbon plasma was analyzed by varying the gas ratios and the effect of ion energy was analyzed by changing the low frequency (LF) power. In the PI-VM results, the density ratios of the passivation precursor CF2 to the etchant F and O were selected as key factors for predicting the process. The selection of radical density ratios as features confirmed the dominance of plasma chemistry in low AR etching. Demonstrating high predictive accuracy with minimal data, PI-VM offers significant potential to enhance the development of semiconductor process recipes.

Meter-scale of the large area inductively coupled plasma etchers with the capacitive power coupling are widely applied for the mass production of OLED (organic light emitting diode) display panels. Because of the large area-to-volume ratio of the etcher, the balance between the power loss and absorption is easily located in the capacitive coupling mode rather than the ideal inductively coupled mode. Therefore, the process results are sensitively governed by the power absorption and plasma heating properties of the reactors. We have introduced a new PI (plasma information) parameter, the ratio of the stochastic heating to Ohmic heating of the plasmas, which is monitorable by using the optical emission spectroscopy data of the processing etchers. With the help of this plasma heating characteristic index, we could optimize the process recipes with the detailed control of the etched hole sidewall passivation and related species generation rate in the plasmas; thus, chamber-to-chamber matching in the huge mass production fab with the higher efficiency was possible. It was demonstrated that the introduced PI index with plasma heating mechanism characterization could be applicable to the VM (virtual metrology) modeling as one of the good information supplying core variables. This PI index has shown a very high correlation with the plasma sheath and ion flux governing phenomena for a large number of mass-produced OLED display glasses. From these results, the introduced plasma heating mechanism-based PI index is expected to be utilized as a good reference index for their performance analysis or PI-VM modelings.

The effect of collisions on the motion of magnetized ions in sheath and presheath plasma regions was investigated through the measurement of ion incident angle of a hydrogen ion at a graphite surface. The experiment was conducted in hydrogen and deuterium plasmas where the ion mean free path is 5–10 times larger than the ion gyro radius and with varying magnetic field angle ψ from 0° to 90° normal to the target surface. The hydrogen ions actively reacted with carbon, leading to the formation of conical tips with axes directed along the incident ion flow direction. The ion incident angle was measured from the etched graphite images taken by scanning electron microscopy. The measured angles were compared to those calculated using Ahedo's fluid magnetic sheath model. In addition, we adopted the nominal Bohm criterion at the electrostatic sheath edge due to the larger ion gyro radius than the sheath. The results show that the ion incident angle was inclined to the normal direction with respect to the magnetic field angle because of the effect of ion collisions on ion motion in the presheath. The collisional effect on the ion motion is drastic for an oblique magnetic field angle ψ > 85°. This study demonstrates that the collisional property of the ions is crucial to guide the ion motion in magnetic (pre)sheath and to determine the ion incidence angle at the surface, even in collisionless and weakly magnetized plasmas.

Data-driven science and technology offer transformative tools and methods to science. This review article highlights the latest development and progress in the interdisciplinary field of data-driven plasma science (DDPS), i.e., plasma science whose progress is driven strongly by data and data analyses. Plasma is considered to be the most ubiquitous form of observable matter in the universe. Data associated with plasmas can, therefore, cover extremely large spatial and temporal scales, and often provide essential information for other scientific disciplines. Thanks to the latest technological developments, plasma experiments, observations, and computation now produce a large amount of data that can no longer be analyzed or interpreted manually. This trend now necessitates a highly sophisticated use of high-performance computers for data analyses, making artificial intelligence and machine learning vital components of DDPS. This article contains seven primary sections, in addition to the introduction and summary. Following an overview of fundamental data-driven science, five other sections cover widely studied topics of plasma science and technologies, i.e., basic plasma physics and laboratory experiments, magnetic confinement fusion, inertial confinement fusion and high-energy-density physics, space and astronomical plasmas, and plasma technologies for industrial and other applications. The final section before the summary discusses plasma-related databases that could significantly contribute to DDPS. Each primary section starts with a brief introduction to the topic, discusses the state-of-the-art developments in the use of data and/or data-scientific approaches, and presents the summary and outlook. Despite the recent impressive signs of progress, the DDPS is still in its infancy. This article attempts to offer a broad perspective on the development of this field and identify where further innovations are required.