The 193nm photoresist (ArF resist) degradation mechanism in dielectric etching was investigated by using an ultra-high-frequency electron-cyclotron-resonance plasma. This investigation focused on via-hole etching. It was found that the bottom-antireflection coating (BARC) etching condition is a critical factor in the reduction of striation and pitting after via-hole etching. X-ray photoelectron spectroscopy and scanning electron spectroscopy studies revealed that argon-less and low-incident-ion-energy conditions in BARC etching can keep the resist surface smooth and maintain a carbon-rich micromask-less state because decomposition of the C–H or OC–O bonds is suppressed. As a result, resist damage after via-hole etching is reduced remarkably. Furthermore, in the via-hole etching, it was also found that the characteristics of the fluorocarbon polymer, i.e., deposition rate and flourine-to-carbon ratio of the fluorocarbon polymer, stacked on the resist surface during etching strongly affect the ArF resist degradation. Low-sticking-coefficient radicals such as CF2 and a low amount of deposition thickness are suitable for damage-less etching. In regard to the formation of striations at the pattern corner, the sputtering effect was taken into consideration. As a result, in the case of via-hole etching, line-edge-roughness in the trench pattern was improved by about 50%, and a striation-less and pitting-less hole etched profile was obtained by using either an argon-and-xenon (20%) mixture as a dilution gas or a fluorocarbon gas at low flow rate under low gas pressure.

The T(re) was consistently greater than T(au) when T(core) was measured in hyperthermic individuals before, during, and postexercise. As T(core) increased, T(au) appeared to underestimate T(core) as determined by T(re). Clinicians should be aware of this critical difference in temperature magnitude between these measurement techniques when assessing T(core) in hyperthermic individuals during or postexercise.

The bowing mechanism in high-aspect-ratio contact hole (HARC) etching was investigated by taking into account reactive sticking on the sidewall of the hole. Sticking coefficients of radicals on the sidewall have been estimated by comparing the observed deposition profile with the calculated one. It was found that the coefficients of C rich radicals and CF x radicals were 0.5 and 0.004, respectively, and that F radical reaction probability to the fluorocarbon polymer is 0.07. These coefficient values were deduced that the excessive flux of O and F onto the sidewall of a hole causes bowing during HARC etching. It was also indicated that the bowing can be suppressed by reducing of the flux of oxygen. These findings were confirmed by the results of experiments using an ultra-high frequency-electron cyclotron resonance (UHF-ECR) plasma.

In this study, atomic layer etching (ALE) of silicon nitride (SiN) using a cyclic process with monofluoromethane chemistry was investigated. Results show that an appropriate desorption time must be chosen at a specific adsorption time to achieve SiN ALE. The results also show that the infinite selectivity of SiN over Si can be achieved using the cycle process. To further understand this behavior, the adsorption and desorption effects were also studied. The results revealed a mechanism to obtain a high Si selectivity and a dominant factor that causes the Si loss. To further understand the ALE capability, we studied and compared the etched profiles and resulting surface roughness obtained by both a conventional process and an ALE process. The results show that the ALE process can achieve a high Si selectivity and a non-detectable level of Si surface damage, compared with a conventional continuous etching process.

No abstract