• Applied Science and Convergence Technology
• /
• v.26 no.6
• /
• pp.184-188
• /
• 2017

Polymer deposition pattern on the ceramic lid surface is analyzed by numerical modeling. Assumption was made that is affected by gas flow pattern from the horizontal and vertical nozzles, temperature profile from the finger-like branches made of graphite and electrostatic potential effect. Calculated results showed gas flow dynamics is less relevant than two others. Temperature and electrostatic effects are likely determining the polymer deposition pattern based on our numerical simulation results.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2007.04a
• /
• pp.99-100
• /
• 2007

유도결합 플라즈마(ICP)를 이용한 CVD 장치에서 플라즈마를 발생시키기 위한 안테나의 구조는 매우 중요하다. 전자 온도와 전자 밀도에 직접적인 영향을 주게 되며, 뿐만 아니라 증착 물질의 두께 균일도에 결정적인 역할을 하게 된다. 본 연구에서는 플라즈마 특성 균일도 최적화를 위하여 2turn 직렬, 병렬, 혼합의 ICP 안테나의 구조에 대하여 플라즈마 특성 및 $SiO_2$ CVD 증착 특성을 계산하였다.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.08a
• /
• pp.148-148
• /
• 2012

A cutoff probe is the novel diagnostic method to get the absolute plasma density with simple system and less assumption. However, high density of ion flux from plasma on probe tip can make the error of plasma density measurement because the dielectric material of probe tip can be damaged by ion flux. We proposed a shielded cutoff probe using the ceramic tube for protection from ion flux. The ceramic tube on probe tip can intercept the ion flux from plasma. The transmitted spectrum using the shielded cutoff probe is good agreement with E/M wave simulation result (CST Microwave Studio) and previous circuit simulation of cutoff probe [1]. From the analysis of the measured transmitted spectrum base on the circuit modeling, the parallel resonance frequency is same as the unshielded cutoff probe case. The obtained results of electron density is presented and discussed in wide range of experimental conditions, together with comparison result with previous cutoff method.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.20-20
• /
• 2000

Back-propagation neural network (BPNN) is the most prevalently used paradigm in modeling semiconductor manufacturing processes, which as a neuron activation function typically employs a bipolar or unipolar sigmoid function in either hidden and output layers. In this study, applicability of another linear function as a neuron activation function is investigated. The linear function was operated in combination with other sigmoid functions. Comparison revealed that a particular combination, the bipolar sigmoid function in hidden layer and the linear function in output layer, is found to be the best combination that yields the highest prediction accuracy. For BPNN with this combination, predictive performance once again optimized by incrementally adjusting the gradients respective to each function. A total of 121 combinations of gradients were examined and out of them one optimal set was determined. Predictive performance of the corresponding model were compared to non-optimized, revealing that optimized models are more accurate over non-optimized counterparts by an improvement of more than 30%. This demonstrates that the proposed gradient-optimized teaming for BPNN with a linear function in output layer is an effective means to construct plasma models. The plasma modeled is a hemispherical inductively coupled plasma, which was characterized by a 24 full factorial design. To validate models, another eight experiments were conducted. process variables that were varied in the design include source polver, pressure, position of chuck holder and chroline flow rate. Plasma attributes measured using Langmuir probe are electron density, electron temperature, and plasma potential.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2014.02a
• /
• pp.221-221
• /
• 2014

The present work proposes an improved numerical simulator for design and modification of large area capacitively coupled plasma (CCP) processing chamber. CCP, as notoriously well-known, demands the tremendously huge computational cost for carrying out transient analyses in realistic multi-dimensional models, because electron dissociations take place in a much smaller time scale (${\Delta}t{\approx}10-8{\sim}10-10$) than time scale of those happened between neutrals (${\Delta}t{\approx}10-1{\sim}10-3$), due to the rf drive frequencies of external electric field. And also, for spatial discretization of electron flux (Je), exponential scheme such as Scharfetter-Gummel method needs to be used in order to alleviate the numerical stiffness and resolve exponential change of spatial distribution of electron temperature (Te) and electron number density (Ne) in the vicinity of electrodes. Due to such computational intractability, it is prohibited to simulate CCP deposition in a three-dimension within acceptable calculation runtimes (<24 h). Under the situation where process conditions require thickness non-uniformity below 5%, however, detailed flow features of reactive gases induced from three-dimensional geometric effects such as gas distribution through the perforated plates (showerhead) should be considered. Without considering plasma chemistry, we therefore simulated flow, temperature and species fields in three-dimensional geometry first, and then, based on that data, boundary conditions of two-dimensional plasma discharge model are set. In the particular case of SiH4-NH3-N2-He CCP discharge to produce deposition of SiNxHy thin film, a cylindrical showerhead electrode reactor was studied by numerical modeling of mass, momentum and energy transports for charged particles in an axi-symmetric geometry. By solving transport equations of electron and radicals simultaneously, we observed that the way how source gases are consumed in the non-isothermal flow field and such consequences on active species production were outlined as playing the leading parts in the processes. As an example of application of the model for the prediction of the deposited thickness uniformity in a 300 mm wafer plasma processing chamber, the results were compared with the experimentally measured deposition profiles along the radius of the wafer varying inter-electrode gap. The simulation results were in good agreement with experimental data.

• Journal of the Korean Vacuum Society
• /
• v.19 no.3
• /
• pp.190-198
• /
• 2010

Asymmetric model for plasma uniformity by Ar and $CF_4$ was modeled by the antenna structure, the diameter of chamber, and the distance between source and substrate for the development of plasma equipment for 450 mm wafer. The aspect ratio of chamber was divided by diameter, distance from substrate, and pumping port area. And we found the condition with the optimized plasma uniformity by changing the antenna structure. The drift diffusion and quasi-neutrality for simplification were used, and the ion energy function was activated for the surface recombination and etching reaction. The uniformity of plasma density on substrate surface was improved by being far of the distance between substrate wall and chamber wall, and substrate and plasma source. And when the antenna of only 2 turns was used, the plasma uniformity can improve from 20~30% to 4.7%.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.08a
• /
• pp.270-270
• /
• 2012

Low temperature atmospheric pressure plasmas (APPs) have been known to be effective for living cell inactivation in the water [1]. Many earlier research found that pH level of the solution was changed from neutral to acidic after plasma treatment. The importance of the effect of acidity of the solution for cell treatments has already been reported by many experiments. In addition, several studies have demonstrated that the addition of a small amount of oxygen to pure helium results in higher sterilization efficiency of APPs [2]. However, it is not clear yet which species are key factors for the cell treatment. To find key factors, we used GMoo simulation. We elucidate the processes through which pH level in the solution is changed from neutral to acidic after plasma exposure and key components with pH and air variation with using GMoo simulation. First, pH level in a liquid solution is changed by He+ and He(21S) radicals. Second, O3 density decreases as pH level in the solution decreases and air concentration decreases. It can be a method of removing O3 that cause chest pain and damage lung tissue when the density is very high. H2O2, HO2 and NO radicals are found to be key factors for cell inactivation in the solution with pH and air variation.

• Journal of the Korean Vacuum Society
• /
• v.18 no.3
• /
• pp.164-175
• /
• 2009

CFD is used to analyze gas flow characteristics, power absorption, electron temperature, electron density and chemical species profile of an internal antenna type inductively coupled plasma system. An optimized grid generation technology is used for a complex real-scale models for industry. A bare metal antenna shows concentrated power absorption around rf a feeding line. Skin depth of power absorption for a system is modeled to 50 mm, which is reported 53 mm by experiments. For an application of bipolar plates for hydrogen fuel cells, multi-sheet loading ICP nitriding system is proposed using an internal ICP antenna. It shows higher atomic nitrogen density than reported simple pulsed dc nitriding systems. Minimum gap between sheets for uniform nitriding is modeled to be 39 mm.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.5
• /
• pp.35-41
• /
• 2004

In computational study on RF(Radio Frequency) plasmas, a 1D fluid models with an advantage of a short computational time are often adopted. However, in order to obtain realistic calculation results under a typical chamber geometry with unequal-sized electrodes, modeling of the plasma space is an issue to be investigated. In this paper, it is focused on that how much a 1D model can approximate a 2D model. 1D fluid models with unequal-sized electrodes, which have spherical and frustum geometry systems, were developed and their results were compared with those of 2D model with Gaseous Electronic Conference cell structure. Behavior of $N_2$ RF plasmas has been simulated using 1D and 2D fluid models and a technique to take account of unequal-sized electrodes in a 1D fluid models has been examined. Features of the plasma density and the electric potential were discussed as characteristic quantities representing the asymmetry of the chamber geometry.

• Korean Chemical Engineering Research
• /
• v.44 no.5
• /
• pp.520-527
• /
• 2006

A realistic surface model is presented for prediction of various surface phenomena such as polymer deposition, suppression and sputtering as a function of incidence ion energy in high density fluorocarbon plasmas. This model followed ion enhanced etching model using the "well-mixed" or continuous stirred tank reactor (CSTR) assumption to the surface reaction zone. In this work, we suggested ion enhanced polymer formation and decomposition mechanisms that can capture $SiO_2$ etching through a steady-state polymer film on $SiO_2$ under the suppression regime. These mechanisms were derived based on experimental data and molecular dynamic simulation results from literatures. The model coefficients are obtained from fits to available beam and plasma experimental data. In order to show validity of our model, we compared the model results to high density fluorocarbon plasma etching data.