Sungmo Kim

Modelling of plasma etching using a generalized regression neural network

Plasma etching was modelled by using a generalized regression neural network (GRNN). The etching process was characterized with a statistical experimental design. Three etch responses were modelled, which include two etch rates of aluminium and silica and etching profile. GRNN prediction ability was optimized as a function of training factor. Three types of models were constructed depending on the type of prepared data. Type I model corresponds to the model constructed with the original, non-classified data. Type II and III models were built for the classified data without and with the control of data interface, respectively. Compared to type I models, type II models for two etch rates demonstrated more than 25% improvement. By the control of data interface, type III models exhibited more than 15% improvement over type II models. Classification-based models in conjunction with data control thus illustrated much improved prediction of GRNN over those for non-classified models.

Modeling SiC surface roughness using neural network and atomic force microscopy

A prediction model for surface roughness was constructed using a neural network and atomic force microscopy. The silicon carbide etch process was characterized by a 25 full factorial experiment. The experimental ranges of process parameters were 600–900W source power, 50–150W bias power, 4–16mTorr pressure, 0–80% O2 percentage, and 6–12cm gap. The model factors were optimized by means of a genetic algorithm. The optimized model had a root mean-squared error of 0.11nm. From the model, various plots were predicted while being supported by actual measurements. The dc bias induced by each process parameter was correlated to the surface roughness. Increasing the bias power increased the surface roughness. In contrast, the surface roughness decreased as the dc bias was larger than about 600V. The surface roughness was strongly correlated to the source power-induced dc bias only at low bias powers. The pressure effect was clear only as the dc bias was maintained at 480V. For the variations in the O2 percentage, ...

Plasma diagnosis by recognizing in situ data using a modular backpropagation network

Abstract Accurate and fast diagnosis of a plasma equipment is crucial to maintain device yield and throughput. A methodology is presented to identify plasma faults. Particular emphasis is placed on the use of in situ diagnostic data in conjunction with a modular backpropagation neural network (BPNN). The experimental data were collected from a real-time impedance match monitor system. Two match positions were used as the signature of an anomaly in equipment plasma. Fault patterns were experimentally generated with a variation in process factors, including radio frequency source power, pressure, O 2 and Ar flow rates. A total of 30 experiments were conducted and subsequently divided into 20 training and 10 test data. Prediction accuracy and fault sensitivity were measured as a function of training factors, hidden neuron and initial weight distribution. The sensitivity was further evaluated from the standpoint of a single and modular network. Modular network consisted of four single networks, specific to variations in process factors. Adjusting initial weight distribution improved fault sensitivity in either network. Compared to single network, modular network greatly improved fault sensitivity irrespective of thresholds.

Proximity-controlled silicon carbide etching in inductively coupled plasma

Abstract By controlling the proximity relative to plasma source, silicon carbide (SiC) has been etched in a C2F6 inductively coupled plasma. This was accomplished by adjusting the gap between the wafer substrate and the plasma source. Gap effects on SiC etching were extensively examined as a function of process factors as well as at particular two conditions, high source power (plasma density) and high bias power (ion bombardment). Decreasing the gap increased the etch rates for all conditions mainly due to the increased plasma density. The DC bias induced with the gap variation played a little role in affecting the etch rate. Gap effect was more enhanced as the wafer electrode was powered with relatively larger DC bias. Factor effects on the etch rates were quite different depending on the level of plasma density or ion bombardment. Decreasing the gap resulted in a microtrench at the base of the sidewall, which was little affected by the DC bias.

Prediction of Plasma Etching Using a Classification-Based Neural Network

Qualitative models of plasma etching are essential for understanding physical etch behaviors as well as plasma control. Artificial neural networks (ANN) have been widely used in constructing predictive etch models. In most applications, the ANN prediction performance has been examined only in terms of the training factors. A technique for building a predictive model is presented here. This is accomplished by applying a neural network to classification data while optimizing the effect of the data boundary. The technique was evaluated with plasma etch data, characterized with a statistical experimental design. The etch response modeled included the etch rates for silica and aluminum (Al) as well as Al selectivity. Compared to the models for nonclassified data, the classification-based model demonstrated improvement in the prediction errors of 36.3 and 15.8% for the Al and silica etch rates, respectively, while exhibiting a poorer prediction for Al selectivity. However, by controlling the data boundary the Al selectivity model was improved significantly by about 43.8%. Thus, the classification-based modeling technique in conjunction with the control of the data boundary is effective in improving the prediction ability of neural network models.

Wavelet-based uniformity of plasma etching surface

Uniformity of plasma etching has been conventionally examined only in terms of etch rate. A uniformity of etching profile surface is increasingly demanded to improve process quality. This is accomplished by applying a discrete wavelet transformation (DWT) to profile images obtained with a scanning electron microscopy. Applicability of DWT-based profile uniformity was evaluated with a tungsten etch experiment, conducted in an SF/sub 6/ helicon plasma. Its suitability was investigated as a function of process parameters and scale levels. Compared to a conventional metric, the wavelet-based one characterized more effectively the uniformity of profile variations. The proposed metric can be applied to any other plasma-processed surfaces.

Wavelet-based neural network prediction of plasma etch profile nonuniformity

Summary form only given, as follows. Profiles of plasma etching have conventionally been characterized by approximating the slope with an angle or anisotropy. This is critically limited in that detailed variations on the profile surface are inevitably neglected. In current high density plasma etching, this becomes more serious since unexpected microfeatures such as bowing or microtrenching are frequently formed along the profile surface.

Use of neural network to characterize temperature effects on refractive property of silicon nitride film deposited by PECVD

Summary form only given, as follows. Deposition of silicon nitride (SiN) film is one of the most critical processes that determine the efficiency of solar cells. Qualities of SiN film deposited by a plasma-enhanced chemical vapor deposition depend on many process parameters. Predicting film properties is very important to their optimization as well as to gain insight into underlying deposition mechanisms. For plasma-driven processes, however, it has been a difficult task to construct prediction models due to complexity within a plasma. In this study, a predictive model for a SiN PECVD process was constructed and used to understand physical deposition mechanisms. The interpretation was mainly focused on the refractive index, particularly with respect to the substrate temperature.