• Journal of the Korean institute of surface engineering
• /
• v.43 no.3
• /
• pp.159-164
• /
• 2010

Surface roughness of deposited or etched film strongly depends on ion bombardment. Relationships between ion bombardment variables and surface roughness are too complicated to model analytically. To overcome this, an empirical neural network model was constructed and applied to a deposition process of silicon nitride (SiN) films. The films were deposited by using a pulsed plasma enhanced chemical vapor deposition system in $SiH_4$-$NH_4$ plasma. Radio frequency source power and duty ratio were varied in the range of 200-800 W and 40-100%. A total of 20 experiments were conducted. A non-invasive ion energy analyzer was used to collect ion energy distribution. The diagnostic variables examined include high (or) low ion energy and high (or low) ion energy flux. Mean surface roughness was measured by using atomic force microscopy. A neural network model relating the diagnostic variables to the surface roughness was constructed and its prediction performance was optimized by using a genetic algorithm. The optimized model yielded an improved performance of about 58% over statistical regression model. The model revealed very interesting features useful for optimization of surface roughness. This includes a reduction in surface roughness either by an increase in ion energy flux at lower ion energy or by an increase in higher ion energy at lower ion energy flux.

• Journal of the Semiconductor & Display Technology
• /
• v.23 no.2
• /
• pp.129-134
• /
• 2024

This study aims to talk about the necessity of solving the PFC gas emission problem raised by the recent development of the semiconductor industry and the remote plasma source method monitoring system used in the semiconductor industry. The 'monitoring system' means that the researchers applied machine learning to the existing monitoring technology and modeled it. In the process of this study, Residual Gas Analyzer monitoring technology and linear regression model were used. Through this model, the researchers identified emissions of at least 12700mg CO2 to 75800mg CO2 with values ranging from ion current 0.6A to 1.7A, and expect that the 'monitoring system' will contribute to the effective calculation of greenhouse gas emissions in the semiconductor industry in the future.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.9
• /
• pp.875-880
• /
• 2015

Traditional semiconductor process control has been performed through statistical process control techniques in a constant process-recipe conditions. However, the complexity of the interior of the etching apparatus plasma physics, quantitative modeling of process conditions due to the many difficult features constraints apply simple SISO control scheme. The introduction of the Advanced Process Control (APC) as a way to overcome the limits has been using the APC process control methodology run-to-run, wafer-to-wafer, or the yield of the semiconductor manufacturing process to the real-time process control, performance, it is possible to improve production. In addition, it is possible to establish a hierarchical structure of the process control made by the process control unit and associated algorithms and etching apparatus, the process unit, the overall process. In this study, the research focused on the methodology and monitoring improvements in performance needed to consider the process management of future developments in the semiconductor manufacturing process in accordance with the age of the APC analysis in real applications of the semiconductor manufacturing process and process fault diagnosis and control techniques in progress.

• YAKHAK HOEJI
• /
• v.47 no.2
• /
• pp.110-119
• /
• 2003

The purpose of the present study was to kinetically investigate the carrier-mediated uptake in the hepatic transport of organic anions, and to simulate the ″in vivo counter-transport″ phenomena, using kinetic model which was developed in this study. The condition that the mobility of carrier-ligand complex is greater than that of free carrier is not essential for the occurrence of ″counter-transport″ phenomenon. To examine the inhibitory effects on the initial uptake of a ligand by the liver, it is necessary to judge whether the true counter-transport mechanism (trans-stimulation) is working or not. The initial plasma disappearance curves of a organic anion were then kinetically analyzed based on a flow model, in which the ligand is eliminated only from the peripheral compartment (liver compartment). Moreover, ″in vive counter-transport″ phenomena were simulated based on the perfusion model which incorporated the carrier-mediated transport and the saturable intracellular binding. The ″in vivo counter-transport″ phenomena in the hepatic transport of a organic anion were well demonstrated by incorporating the carrier-mediated process. However, the ″in vivo counter-transport″ phenomena may be also explained by the enhancement of back diffusion due to the displacement of intracellular binding. In conclusion, one should be more cautious in interpreting data obtained from so-called ″in vivo counter-transport″ experiments.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.11 no.3
• /
• pp.135-145
• /
• 2011

Accurate process characterization and optimization are the first step for a successful advanced process control (APC), and they should be followed by continuous monitoring and control in order to run manufacturing processes most efficiently. In this paper, process characterization and recipe optimization methods with multiple outputs are presented in high density plasma-chemical vapor deposition (HDP-CVD) silicon dioxide deposition process. Five controllable process variables of Top $SiH_4$, Bottom $SiH_4$, $O_2$, Top RF Power, and Bottom RF Power, and two responses of interest, such as deposition rate and uniformity, are simultaneously considered employing both statistical response surface methodology (RSM) and neural networks (NNs) based genetic algorithm (GA). Statistically, two phases of experimental design was performed, and the established statistical models were optimized using performance index (PI). Artificial intelligently, NN process model with two outputs were established, and recipe synthesis was performed employing GA. Statistical RSM offers minimum numbers of experiment to build regression models and response surface models, but the analysis of the data need to satisfy underlying assumption and statistical data analysis capability. NN based-GA does not require any underlying assumption for data modeling; however, the selection of the input data for the model establishment is important for accurate model construction. Both statistical and artificial intelligent methods suggest competitive characterization and optimization results in HDP-CVD $SiO_2$ deposition process, and the NN based-GA method showed 26% uniformity improvement with 36% less $SiH_4$ gas usage yielding 20.8 ${\AA}/sec$ deposition rate.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.27-29
• /
• 2005

Using a generalized regression neural network, etch microtrenching was modeled. All neurons in the pattern layer were equipped with multi-factored spreads and their complex effects on the prediction performance were optimized by means of a genetic algorithm. For comparison, GRNN model was constructed in a conventional way. Comparison result revealed that GA-GRNN model was more accurate than GRNN model by about 30%. The microtrenching data were collected during the etching of silicon oxynitride film and the etch process was characterized by a statistical experimental design.

• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.49 no.8
• /
• pp.487-494
• /
• 2000

This paper describes the design, fabrication and experiments of surface-micromachined aluminum micromirror array with hidden pin-joints. Instead of the conventional elastic spring components as connection between mirror plate and supporting structure, we used pin-joint composed of pin and staples to support the mirror plate. The placement of pin-joint under the mirror plate makes large active surface area possible. These flexureless micromirrors are driven by electrostatic force. As the mirror plate has discrete deflection angles, the device can be ap;lied to adaptive optics and digitally-operating optical applications. Four-level metal structural layers and semi-cured photoresist sacrificial layers were used in the fabrication process and sacrificial layers were removed by oxygen plasma ashing. Static characteristics of fabricated samples were measured and compared with modeling results.

• Proceedings of the KWS Conference
• /
• 2005.06a
• /
• pp.97-99
• /
• 2005

Laser arc hybrid process is actively researched as a new welding method since it has several advantages by the combination of laser beam and electric arc. By the coupling of two different heat sources, laser and arc mutually assist and influence. High power laser can make the deep keyhole and arc plasma can form the large bead shape. In this paper the effect of two different heat sources to weld bead are investigated and as a result of analysis, it is shown that the lower part of keyhole is heated by laser and the upper part of weld pool is dominantly heated by arc.

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.177-178
• /
• 2007

플라즈마 공정은 집적회로 제작을 위한 미세 박막의 증착과 패턴닝에 핵심적으로 이용되고 있다. 본 연구에서는 플라즈마공정감시와 제어에 응용될 수 있는 모델을 제안한다. 본 모델은 광방사분광기 (Optical emission spectroscopy-OES)정보와 역전파 신경망을 이용해서 개발하였다. 제안된 기법은 Oxide 식각공정에서 수집한 데이터에 적용하였으며, 체계적인 모델링을 위해 공정데이터는 통계적 실험계획법을 적용하여 수집되었다. Raw OES 정보대신, Actinometric OES 정보를 이용하였으며, 신경망의 예측성능은 유전자 알고리즘을 이용해서 증진시켰다. OES의 차수를 줄이기 위해 주인자 분석 (Principal Component Analysis-PCA)을 세 종류의 분산(100, 99, 98%)에 대해서 적용하였다. 최적화한 모델의 예측에러는 323 $\AA/min$이었다. 이전에 PCA를 적용하고 은닉층 뉴런의 함수로 최적화한 모델의 예측에러는 570 $\AA/min$이었으며, 개발된 모델은 이에 비해 43% 증진된 예측 성능을 보이고 있다.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2582-2584
• /
• 2003

역전파 신경망 (BPNN)은 반도체 공정 모델링에 효과적으로 응용되고 있다. 뉴런의 활성화 함수는 동일한 값을 가지며, 이로 인해 예측정확도를 증진하는 데에는 한계가 있었다. 본 연구에서는 난수발생기(Random generator-RG)를 이용하여 뉴런 경사들이 다중값을 가지도록 최적화하였다. 본 기법은 은닉충의 뉴런수의 함수로 고찰하였으며, 종래의 고정된 경사를 갖는 모델과 그 성능을 비교 평가하였다. 평가에 이용된 데이터는 플라즈마 식각 공정데이터이며, 모델에 이용된 응답은 식각률과 프로파일 각이다. 비교결과 종래의 모델에 비해 예측정확도가, 식각률의 경우 19%-43%, 프로파일의 경우 10%-56% 정도 향상하였으며, 이는 제안된 기법이 모델개발에 매우 효과적으로 적용될 수 있음을 보여준다.