Won Gyu Lee

Formation of Tungsten Oxide Defects during Tungsten CMP

Most tungsten chemical mechanical polishing (CMP) processes utilize a subsequent cleaning process referred to as post-CMP cleaning for removal of contaminated layers. Even after a proper post-CMP cleaning has treated polished waters, several defects such as slurry particles, microscratches, and metallic impurities are still easily observable. In this study, a new kind of defect has been observed at the stage between overpolishing and post-CMP clean in the course of the tungsten CMP. The circular-shape defect was composed of very thin tungsten oxide, which could be removed by several treatments such as HF solution dipping, additional oxide CMP, and the treatment of concentrated ammonium hydroxide solution during poly(vinyl alcohol) brushing. Among them, the most available method without any further process issues was the use of concentrated ammonium hydroxide solution.

Comparison of different applied voltage waveforms on CO2 reforming of CH4 in an atmospheric plasma system

Sinusoidal and pulse waveforms of applied voltage were employed for CO2 reforming of CH4 to syngas in an atmospheric dielectric barrier discharge reactor. The discharge power of a pulse waveform was higher than that of sinusoidal waveform at the same applied voltage. The plasma reaction by a pulse waveform enhanced the conversion of CO2 and CH4 and the selectivity of H2 and CO. It was confirmed that CO2 reforming of CH4 can be improved by the adaption of pulse-type power supply in a dielectric barrier discharge reactor immersed in an electrically insulating oil bath.

Hydrophobic properties of films grown by torch-type atmospheric pressure plasma in Ar ambient containing C6 hydrocarbon precursor

The direct deposition of polymeric films with a torch-type atmospheric pressure plasma using benzene, n-hexane and cyclohexane in Ar was performed on several substrates. The surface morphologies of the films deposited with n-hexane and cyclohexane were uniformly smooth for all deposition thicknesses, and the typical water contact angle on the films indicating the degree of hydrophobicity was about 85o. However, the films deposited using benzene had a micro-coarse surface morphology and showed a superhydrophobic property with a water contact angle exceeding 150°. Some trace of oxygen incorporation was shown in all films due to the plasma deposition process in an air ambient. The small amount of oxygenated species did not lead to a decrease of hydrophobicity of the films.

Process conditions for complete decomposition of CHF3 in a dielectric barrier discharge reactor

Plasma decomposition of CHF3 was investigated using a dielectric barrier discharge immersed in an electrically insulating oil bath in a mixture of CHF3, O2, and N2. CHF3 was well decomposed under a relatively high applied voltage in an atmospheric pressure plasma system. The main by-product was CO2 and its selectivity increased with a decrease in the CHF3 concentration in the feed. Complete decomposition of CHF3 was achieved at a typical process range: an applied voltage ≥7.0 kVp, an initial CHF3 flow rate of 3ml/min, and a total flow rate of 500ml/min. The value of energy efficiency and energy density at the center range for the complete decomposition of CHF3 was 0.01 mmol/kJ and 6.00kWh/Nm3, respectively.

Analytical Approach to estimate and Predict the CO2 Reforming of CH4 in a Dielectric Barrier Discharge

CO2 reforming of CH4 to syngas was investigated in a coaxial dielectric barrier discharge reactor immersed in an oil bath. An analytical model was suggested to estimate and predict the reaction phenomena. The model had input parameters as predictor variables (applied voltage, ratio of CH4/CO2, and total flow rate in the feed), output parameters as observed variables, the molar flow rates of reactants (CH4, CO2, CO, H2, and by-products), and energy efficiencies. More than 70% of the output parameters variance could be explained by the input parameter. The model for the CO2 reforming of CH4 in a dielectric barrier discharge reactor would be useful to optimize the experiments. A comparison between input parameters suggests that the reaction should be performed under high total flow rate or low applied voltage to obtain greater energy efficiency; whereas at high applied voltage and total flow rate, the reaction obtains a greater absolute amount of reactant conversion and products, but lower energy efficiency.

Analysis of Plasma Reaction for Decomposition of CHF3 in a Dielectric Barrier Discharge

The decomposition of CHF3 in a mixture with O2 and Ar was investigated in a coaxial dielectric barrier discharge at atmospheric pressure. CHF3 decomposition increased linearly in regard to specific energy input (SEI), whereas energy efficiency decreased. The main product was CO2, and its selectivity increased with high SEI and the presence of O2 in the feed, but an increase of O2 in the feed led to a decrease in decomposition rate. An increase in total flow rate led to an increase of the absolute amount of CHF3 decomposition and energy efficiency; however, the decomposition of CHF3 decreased. A complete CHF3 decomposition occurred under an SEI of 1.54 kJ/L with the selectivity of CO2 and CO as 89.87% and 7.00%, respectively. Optical emission spectroscopic analysis could explain the available reaction pathways for CHF3 decomposition in the CHF3/O2/Ar atmospheric plasma and show the possibility of F2 and HF formation.

Abnormal silicon oxide growth on cobalt silicide in arsenic-doped N+ active areas

Abstract During the deposition process of silicon oxide, we found an unexpected oxide growth on a CoSi 2 layer in an arsenic-doped N+ active area, showing the degradation of contact resistance in this area. The role of arsenic was investigated in order to explain its effects on the abnormal oxidation. This abnormal oxidation was caused by the out-diffusion of silicon atoms from the substrate and was readily influenced by the concentration of arsenic in the silicon substrate and by heat treatment enhancing the out-diffusion of arsenic.

Novel Formation of Zirconium Titanate from Multi-layered Structure of ZrO2 and TiO2 deposited by PRTCVD

PRTCVD (pulsed rapid thermal chemical vapor deposition) method has been introduced to prepare zirconium titanate (ZrTiO4) films from a multi-layered structure composed of ZrO2 and TiO2 as a primitive oxide, which allows for the preparation of a given film with different element per cycle and precise control over the thickness of the film. The feasibility of PRTCVD was evaluated and it was confirmed that a multi-layered structure composed of different primitive oxides can be easily produced by PRTCVD and transformed into a multi-component single phase oxide such as zirconium titanate by an appropriate thermal annealing process.

Prediction of the infinite-dilution partial molar volumes of organic solutes in supercritical carbon dioxide using the Kirkwood-Buff fluctuation integral with the hard sphere expansion (HSE) theory

Two thermodynamic models were used to predict the infinite dilution partial molar volumes (PMVs) of organic solutes in supercritical carbon dioxide: (1) the Kirkwood Buff fluctuation integral with the hard sphere expansion (HSE) theory incorporated (KB-HSE fluctuation integral method) and (2) the Peng-Robinson equation of state with the classical mixing rule. While an equation of state only for pure supercritical carbon dioxide is needed in the KB-HSE fluctuation integral model, and thus, there is no need to know the critical properties of solutes, two molecular parameters (one size parameter σ12 and one dimensionless parameter α12) in the KB-HSE fluctuation integral model are determined to fit the experimental data published on the infinite dilution PMVs of solutes. The KB-HSE fluctuation integral method produced better results on the infinite dilution PMVs of eight organic solutes tested in this work than the Peng-Robinson equation of state with the classical mixing rule.