• Journal of the Semiconductor & Display Technology
• /
• v.22 no.4
• /
• pp.131-135
• /
• 2023

Neural network-based time series models called time series neural networks (TSNNs) are trained by the error backpropagation algorithm and used to predict process shifts of parameters such as gas flow, RF power, and chamber pressure in reactive ion etching (RIE). The training data consists of process conditions, as well as principal components (PCs) of optical emission spectroscopy (OES) data collected in-situ. Data are generated during the etching of benzocyclobutene (BCB) in a SF₆/O₂ plasma. Combinations of baseline and faulty responses for each process parameter are simulated, and a moving average of TSNN predictions successfully identifies process shifts in the recipe parameters for various degrees of faults.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.32A no.1
• /
• pp.152-158
• /
• 1995

Eteching characteristics of the Indium Tin Oxide (ITO), which is transparent conductor, was investigated with CH4/H2 and Ar as etching gases for the Reactive Ion Etching (RIE). With CH4/H2 for the etching gas, the highly selective etching characteristics for the ITO on GaAs was obtained. It was examined that the dominant etching parameter for the selective etchning of ITO on GaAs structure was the chamber pressure. But, the etching selectivity for ITO on InP was poor eventhough we tried systematic etching. RIE etching conditins using CH4/H2 gas was limited due to the formation of polymer on the substrates. In the case of Ar gas for the reactive gas, the selectivity of ITO on BaTiO3 was above 10. The etch rete of ITO was more sensitive to the etching parameters than that of BaTiO3, which was almost constant with different etching parameters.

• Transactions on Electrical and Electronic Materials
• /
• v.16 no.4
• /
• pp.169-172
• /
• 2015

Although, there are several research studies on the engineering of the graphene layers using different etching techniques, there is not any comprehensive study on the effects of using different etching masks in the reactive ion etching (RIE) method on the quality and uniformity of the etched graphene films. This study investigated the effects of using polystyrene and conventional photolithography resist as a etching mask on the engineering of the number of graphene layers, using RIE. The effects were studied using Raman spectroscopy. This analysis indicated that the photo-resist mask is better than the polystyrene mask because of its lower post processing effects on the graphene surface during the RIE process. A single layer graphene was achieved from a bi-layer graphene after 3 s of the RIE process using oxygen plasma, and the bi-layer graphene was successfully etched after 6 s of the RIE process. The bilayer etching time was significantly smaller than reported values for graphene flakes in previous research.

• Journal of Magnetics
• /
• v.12 no.2
• /
• pp.81-83
• /
• 2007

Etching of NiFe films covered with an organic photo-resist or Ti was successfully performed by an inductively coupled plasma-reactive ion etching (ICP-RIE) system using $CHF_3/O_2/NH_3$ discharges exchanging $CHF_3$ for $CH_4$ gas gradually. Experimental results showed that the organic photo-resist mask can be applied to the NiFe etching. In the case of the Ti metal mask, it was found that the etching-selectivity Ti against NiFe was significantly varied from 7.3 to ${\sim}0$ by changing $CHF_3/CH_4/O_2/NH_3$ to $CH_4/O_2/NH_3$ discharges used in the ICP-RIE system. These results show that the present RIE of NiFe was dominated by a chemical reaction rather than a physical sputtering.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.14 no.6
• /
• pp.236-243
• /
• 2004

The oxide etching process was characterized in a magnetically enhanced reactive ion etching (MERIE) reactor with a $CHF_3CF_4$ gas chemistry. A statistical experimental design plus one center point was used to characterize relationships between process factors and etch response. The etch response modeled are etch rate, etch selectivity to TiN and uniformity. Etching uniformity was improved with increasing $CF_4$ flow ratio, increasing source power, and increasing pressure depending on source power. Characterization of via etching in $CHF_3CF_4$ MERIE using neural networks was successfully executed giving to highly valuable information about etching mechanism and optimum etching condition. It was found that etching uniformity was closely related to surface polymerization, DC bias, TiN and uniformity.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.12
• /
• pp.2221-2225
• /
• 2007

Hollow-type microneedle array can be used for painless, continuous and stable drug delivery through a human skin. The needles must be sharp and have sufficient length in order to penetrate the epidermis. An array of hollow-type silicon microneedles was fabricated by using deep reactive ion etching and HNA wet etching with two oxide masks. Isotropic etching was used to create tapered tips of the needles, and anisotropic etching of Bosch process was used to make the extended length and holes of microneedles. The microneedles were formed by three steps of isotropic, anisotropic, and isotropic etching in order. The holes were made by one anisotropic etching step. The fabricated microneedles have $170{\mu}m$ width, $40{\mu}m$ hole diameter and $230{\mu}m$ length.

• Proceedings of the KIEE Conference
• /
• 2000.07c
• /
• pp.1491-1493
• /
• 2000

Continuing improvement of microprocessor performance involves in the device size. This allow greater device speed, an increase in device packing density, and an increase in the number of functions that can reside on a single chip. However this has led to propagation delay, crosstalk noise, and power dissipation due to resistance-capacitance(RC) coupling become significant due to increased wiring capacitance, especially interline capacitance between the metal lines on the same metal level. So, MSSQ which has the permittivity between 2.5-3.2 is used to prevent from these problems. For pattering MSSQ(Methylsilsequioxane), we use RIE(Reactive Ion Etching) and MERIE(Magnetically enhanced Reactive Ion Etching) which could provide good anisotropic etching. In this study, we optimized the flow rate of $CF_{4}/O_2$ gas, RF power to obtain the best etching rate and roughness and also analyzed the etching result using $\alpha$-step profilemeter, SEM, infrared spectrum and AFM.

• Journal of Sensor Science and Technology
• /
• v.24 no.1
• /
• pp.10-14
• /
• 2015

This paper reports the results of a study on the optimization of the etching profile, which is an important factor in deep-reactive-ion etching (DRIE), i.e., dry etching. Dry etching is the key processing step necessary for the development of the Internet of Things (IoT) and various microelectromechanical sensors (MEMS). Large-area etching (open area > 20%) under a high-frequency (HF) condition with nonoptimized processing parameters results in damage to the etched sidewall. Therefore, in this study, optimization was performed under a low-frequency (LF) condition. The HF method, which is typically used for through-silicon via (TSV) technology, applies a high etch rate and cannot be easily adapted to processes sensitive to sidewall damage. The optimal etching profile was determined by controlling various parameters for the DRIE of a large Si wafer area (open area > 20%). The optimal processing condition was derived after establishing the correlations of etch rate, uniformity, and sidewall damage on a 6-in Si wafer to the parameters of coil power, run pressure, platen power for passivation etching, and $SF_6$ gas flow rate. The processing-parameter-dependent results of the experiments performed for optimization of the etching profile in terms of etch rate, uniformity, and sidewall damage in the case of large Si area etching can be summarized as follows. When LF is applied, the platen power, coil power, and $SF_6$ should be low, whereas the run pressure has little effect on the etching performance. Under the optimal LF condition of 380 Hz, the platen power, coil power, and $SF_6$ were set at 115W, 3500W, and 700 sccm, respectively. In addition, the aforementioned standard recipe was applied as follows: run pressure of 4 Pa, $C_4F_8$ content of 400 sccm, and a gas exchange interval of $SF_6/C_4F_8=2s/3s$.

• Journal of the Korean Vacuum Society
• /
• v.1 no.1
• /
• pp.145-152
• /
• 1992

Si surfaces exposed to CHF3/C2F6 gas plasmas ih reactive ion etching (RIE) have been characterized by X-ray photoelectron spectroscopy (XPS). CHF3/C2F6 gas plasma exposure of Si surface leads to the deposition of residual film containing carbon and fluorine. The narrow scan spectra of C 1s show various bonding states of carbon as C-Si, C-F/H, C-CFx(x $\leq$ 3), C-F, C-F2, and C-F3. The chemical bonding states of fluorine are described with F-Si, F-C and F-O. And the oxygen and silicon are also detected. The effects of parameters for reactive ion etching as CHF3/C2F6 gas ratio, RF power, and pressure are investigated.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.389-392
• /
• 2005

차세대 반도체 공정을 위한 많은 노력 중 Reactive Ion Etching(RIE)에 대한 연구의 중요성은 계속되고 있으며, 현재 제조공정 라인에서는 공정상의 오류를 줄이는 노력에 주목 하고 있다. 본 논문에서는 이러한 점을 고려하여 반도체 제조 장비에서 발생하는 실시간 데이터에 대해 신경망을 이용하여 각 장비파라미터의 허용범위를 검출하고, 제안된 방법의 성능평가를 위하여, 생산라인에서 수립된 데이터를 활용하였다. 기존의 통계적 공정제어(SPC) 제서 지시되는 방법이 아닌 신경망 모델을 통하여 random variability를 고려한 control limit을 제시한다.