Solid State Phenomena Vol. 346

Paper Title Page

Abstract: In semiconductor device manufacturing, it is often necessary to process devices that incorporate both Si3N4 and SiO2 with etching solutions. Recently, we found that a cellulose-based polymer (DC01) selectively adsorbs onto silicon nitride and acts as a protective film against etching with buffered fluoric acid. However, the mechanism behind this selective adsorption remains unclear, and a protective film persists after etching. QCM measurements and calculations of the surface free energy indicate that the affinity between DC01 and Si3N4 was significantly stronger than that of SiO2. Notably, the same behavior was not observed with cellulose or cellulose acetate, indicating that this phenomenon is unique to our original cellulose. Additionally, we successfully prepared films using a facile dipping method. Furthermore, we discovered that the strongly adhesive DC01 film can be completely removed by subjecting it to oxidative conditions, such as SC-1 or SC-2.
131
Abstract: This study investigated the etching kinetics of Si3N4 in various concentration of H3PO4 solution and the effect of Si3N4 etching enhancers on the etching process, particularly for 3D NAND trench structures. 30 wt% H3PO4 was used to etch Si3N4, which can produce higher Si3N4/SiO2 etching selectivity and similar Si3N4 etching rate compared to a conventional 85 wt% H3PO4. 30 wt% H3PO4 showed significantly improved etching performance for the Si3N4/SiO2 3D NAND structure as compared to 85 wt% H3PO4. In particular, the transportation ability of H3PO4 into 3D NAND trench structures can be improved by reducing viscosity of etchant, which can be obtained by reducing the concentration of H3PO4. In addition, Si3N4 etching enhancers were introduced to accelerate the Si3N4 etching kinetics in 30 wt% H3PO4. Addition of such additives improved the Si3N4 etching rate and Si3N4/SiO2 etching selectivity while suppressing oxide regrowth. The results provide valuable insights for optimizing selective Si3N4 etching process in 3D NAND structures.
137
Abstract: In the process of manufacturing three-dimensional Not AND (3D NAND) flash memory devices, oxide regrowth occurs during the selective Si3N4 etching process on the Si3N4/SiO2 multi-stack structures. The oxide regrowth issue must be suppressed. Thus, effects of two factors, the mass transfer ability of etching byproduct and the stack number of 3D NAND structure, on oxide regrowth were investigated in this study using finite element method (FEM) simulation. Using FEM simulation, we predicted that increase in the stack number of the 3D NAND structure and decrease in the diffusivity of Si3N4 etching byproducts highly aggravated the oxide regrowth. Therefore, it was suggested that an etchant capable of promoting H2SiO3 diffusion behavior would inhibit the oxide regrowth during the selective Si3N4 etching process of the 3D NAND structure.
143
Abstract: Abstract. Advanced semiconductor technology features complicated three-dimensional nanostructures and nanoconfined spaces such as nanosheets, supervias, deep contact holes and nanocavities. Uniform wet etching of such nanoconfined spaces across different feature sizes or critical dimensions (CD) is extremely challenging. Typically, etch rate decreases with decrease in CD size. In this paper we report methods to achieve uniform wet etch rate (ER) of SiO2 across different CD sizes by mixing organic solvents in the etching solution. We also report a reversal of etch rate trend where SiO2 structure of smaller CD etches faster than a larger CD, by tuning the ratio of organic to water solvents in the etching solution. We also investigate the impact of parameters such as solvent type, wall material, surface tension and ionic strength on ER. Our data suggests, while surface tension and ionic strength show no impact, the type of wall material, surface potential and organic solvent amount in the etching solution show a strong influence on SiO2 ER. Also, zeta potential could explain most of our results but not all, suggesting that surface potential is not the only factor impacting CD dependent ER in a nanoconfined spaces.
149
Abstract: A method for continuous measurement of particle and particle precursor material in high purity liquids using aerosolization and condensation particle courting (A+CPC) is presented. A process for correlating wafter defects with the concentration values measured using A+CPC using Sequential Spin Coating + Surface Enhanced Particle Sizing (SSC + SEPS) is described. The method was applied to monitor the rinsate of an ion exchange resin following the SEMI C93 guide. The results showed correlation between the methods and the concentration values were used to calculate a deposition factor (liquid concentration/surface concentration) on the order of 1E5 (#/ml)/(#/cm2).
157
Abstract: This paper provides a comprehensive review of current nanoparticle analysis methods specifically designed for sub-20 nm particles. The techniques are categorized into three main groups: ensemble (I), single particle counting (II), and separating & counting (III) methods. The study further presents a comparative analysis of test results obtained using these three categories of methods on polydisperse polymer nanoparticles. For this investigation, four different tools capable of sub-10 nm particle analysis were utilized. The techniques employed in this study include dynamic light scattering (I), atomic force microscope (II), scanning mobility liquid particle sizer (III), and cryo-electron microscopy (I). The findings indicate that, for particles larger than 8 nm, the three tools—excluding dynamic light diffraction—yield similar results. Conversely, in the small size range (<7 nm), cryo-electron microscopy consistently demonstrates reliable outcomes.
164
Abstract: Metal contamination is a critical issue in semiconductor manufacturing processes, where controlling the concentration of metals in the deionized water (DIW) used for wafer rinsing and chemical dilution is necessary. In our previous study, we reported an ultra-trace level measurement method using solid-phase extraction (SPE) with a monolithic resin and inductively coupled plasma mass spectrometry (ICP-MS). However, this method had limitations regarding rapid metal concentration confirmation and contamination risks due to manual processing and sample delivery. This study proposes an on-line metal measurement system that uses an automated SPE technique coupled with ICP-MS to address the shortcomings of the conventional method. The system provides continuous on-site monitoring of ultra-trace metal concentrations in DIW, reducing preconcentration volume by up to 60% compared to the conventional method owing to minimizing contamination by automating all processes. Moreover, the system achieved reliable results similar to the conventional method, high analytical stability, with the relative standard deviations (RSDs) within 5% for a concentration of around 0.05 ng/L (n=4) using a test solution. We also evaluated the metal removal filter as an application of the system mentioned above, simultaneously analyzing multiple elements at the single-pg/L level and confirming the performance of the filter.
170
Abstract: In this study, emissions of photoresist resin deposited on wafters in a FOUP are compared using impinger/GCMS at-line measurements and on-line Vocus AMC measurements based on chemical ionization mass spectrometry. While GCMS measurements are routinely used for VOC measurements, the Vocus AMC monitor allows simultaneous, real-time detection of acids, bases, and condensable organics, providing unique insights into the cleanroom and FOUP microenvironment Photoresist resin outgassing measurements show comparable absolute concentrations between the two methods (GCMS and CI-TOF) for targeted compounds in real world sampling scenarios. The rate of outgassing was tracked in real time with the Vocus and also allowed identification of unknown compounds emitted from the resin. Continuous monitoring of the cleanroom air around the FOUP highlights the importance of the cleanroom environment which can directly affect the FOUP microenvironment, and additionally we show that Entegris Barrier Material EBM can guarantee a good wafer protection.
176
Abstract: A new measurement system has been introduced to perform real-time measurements of molecular air contaminants at different location in a clean room down to a concentration of approximately 0.1 ppb. The system consists of an air sampling unit coupled to a proton transfer mass spectrometer and a Cavity Ring-Down Spectrometer. The measurements show that the time average concentrations of more than 10 different volatile impurity molecules exceed 1 ppb.
183
Abstract: Isopropyl alcohol (IPA) was used in the drying process to remove the contamination after the wet cleaning process. Regardless of the types of chemical solutions, the IPA is more strictly controlled because it is used. However, when the IPA was exposed to low acid concentration, the IPA polymers were generated to the di-isopropyl ether, isopropyl acetate, and di-isopropoxy methane. And the pH of IPA rapidly changed by about 1.99 in ppm level concentration. Generated IPA polymers were adsorbed on the wafer surface. These polymers can occur defects, and the probability increases as the concentration increases. Scanning mobility particle sizer (SMPS) measurement system can be measured the impurity in IPA solution. The total impurity concentration was increased by the concentration of H2SO4 with the same results in both gas chromatography mass spectrometry (GCMS) and time of flight secondary ion mass spectrometry (ToF-SIMS). SMPS results can be correlated with the concentration of impurities. In this paper, IPA polymers were defined by acid contamination, and these polymers have affected the Si wafer surface. In addition, the SMPS measurement system was introduced for the detection of impurities in IPA solutions.
189

Showing 21 to 30 of 56 Paper Titles