• The Korean Journal of Applied Statistics
• /
• v.34 no.1
• /
• pp.99-114
• /
• 2021

With the Fourth Industrial Revolution based on new technology, the semiconductor manufacturing industry researches various analysis methods such as detecting process abnormalities and predicting yield based on equipment sensor data generated in the manufacturing process. The semiconductor manufacturing process consists of hundreds of processes and thousands of measurement processes associated with them, each of which has properties that cannot be defined by chemical or physical equations. In the individual measurement process, the actual measurement ratio does not exceed 0.1% to 5% of the target product, and it cannot be kept constant for each measurement point. For this reason, efforts are being made to determine whether to manage by using equipment sensor data that can indirectly determine the normal state of each step of the process. In this study, the Functional Data Analysis (FDA) was proposed to define a process abnormality detection process based on equipment sensor data and compensate for the disadvantages of the currently applied statistics-based diagnosis method. Anomaly detection accuracy was compared using machine learning on actual field case data, and its effectiveness was verified.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.32 no.2
• /
• pp.142-148
• /
• 2012

This study presents the assessment results of adhesive properties on the interface between a silicon wafer and nano-scale polymer thin film pattern through UAFM images by using the contact resonance frequency of the cantilever. For the experiment, we varied surface treatment processes for the silicon wafer and fabricated a 300nm polymer thin film pattern through lithography. Images from the optical microscope were used to compare the produced test specimens for adhesive condition and the critical load value from the nano scratch test was used to verify the adhesive condition of the nano pattern. Each test specimen resulted in a $1{\mu}m{\times}1{\mu}m$ surface image and subsurface adhesive image. Adhesive condition was evaluated by image contrast differences on the interface according to the changing amplitudes and phases of contact resonance frequency.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.35 no.5
• /
• pp.459-465
• /
• 2011

Excessive heat may be generated during the semiconductor manufacturing process. Therefore, precise control of temperature is required to maintain a constant ambient temperature and wafer temperature in the chamber. Compared to an industrial chiller, a semiconductor chiller's power consumption is high because it is in continuous operation for a year. Because of this high power consumption, it is necessary to develop an energy-efficient chiller by optimizing the operation. The competitiveness of domestic products is low because of the high energy consumption. We experimentally investigated a domestic semiconductor by conducting load change, temperature rise and fall, and control precision experiments. The experimental study showed that the chiller had 2.1-3.9 kW of cooling capacity and 0.56-0.93 of EER. The control precisions were ${\pm}1^{\circ}C$ and ${\pm}0.6^{\circ}C$ when the setting temperatures were $0^{\circ}C$ and $30^{\circ}C$ respectively.

• Korean Journal of Optics and Photonics
• /
• v.28 no.2
• /
• pp.75-82
• /
• 2017

We present a new type of receiver-transmitter optical system that can be adapted to the sensor head of a displacement-measuring interferometer. The interferometer is utilized to control positioning error and repetition accuracy of a wafer, down to the order of 1 nm, in a semiconductor manufacturing process. Currently, according to the tendency of scale-up of wafers, an interferometer is demanded to measure a wider range of displacement. To solve this technical problem, we suggest a new type of receiver-transmitter optical system consisting of a GRIN lens-Collimating lens-Afocal lens system, compared to conventional receiver-transmitter using a single collimating lens. By adapting this new technological optical structure, we can improve coupling efficiency up to about 100 times that of a single conventional collimating lens.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2011.02a
• /
• pp.143-144
• /
• 2011

최근 광전자 분야에서는 미래 에너지 자원에 대한 관심과 함께 GaN 기반 발광다이오드에 대한 연구가 활발히 진행되고 있다. 특히 InGaN/GaN 양자 우물 구조는 푸른색, 녹색 발광다이오드 구현에 있어 우수한 물질적 특성을 가지고 있다고 알려져 있다. 하지만 우수한 물질적 특성에도 불구하고 고인듐 고품위 막질 성장의 어려움으로 인해 높은 효율의 녹색 발광다이오드 구현하는 것은 여전히 어려운 실정이다. 이를 극복하기 위한 대안 중에 하나인 선택 영역 박막성장법(Selective Area Growth)은 마스크 패터닝을 통해 열린 영역에서만 박막을 성장하는 방법으로써 인듐 함량을 향상 시킬 수 있는 방법으로 주목 받고 있다. 선택 영역 박막 성장법을 이용하여 GaN를 성장하기 위해 그림 1의 공정을 통하여 n-GaN층 위에 SiO2 마스크를 포토리소그라피와 Reactive Ion Etching (RIE)를 이용한 건식 식각 공정을 통해 형성한 후 Metal Organic Chemical Vapor Deposition (MOCVD) 장비를 이용하여 선택적으로 에피를 성장하였다. 성장된 마이크로 피라미드 발광다이오드 구조는 n-GaN 피라미드 구조위에 양자우물 및 p-GaN을 성장함으로써 p-GaN/MQW/n-GaN 구조를 갖는다. 이렇게 생성된 피라미드 구조의 에피를 이용하여 발광다이오드를 제작한 후 그에 대한 전기적, 광학적 특성을 측정하였다. 2인치 웨이퍼의 중심을 원점 좌표인 (0,0)으로 설정하였을 때 2인치 웨이퍼에서 좌표에 해당하는 위치에서의 Photoluminescence (PL) 측정한 결과 일반적인 구조의 발광다이오드의 경우 첨두치가 441~451nm인데 반해 피라미드 구조의 발광다이오드의 경우 첨두치가 558nm~563nm 임을 알 수 있었다. 이를 통해 피라미드 구조 발광다이오드의 경우 일반적인 구조의 발광다이오드에 비해 인듐의 함유량을 증가시킬 수 있다는 것을 알 수 있다. 본 논문에서는 선택 영역 박막 성장법을 이용하여 마이크로 피라미드 InGaN/GaN 양자 우물 구조 구현과 광 전기적 특성에 대해 더 자세히 논의 하도록 하겠다.

• Resources Recycling
• /
• v.12 no.3
• /
• pp.25-30
• /
• 2003

The synthesis of SiC used for the parts of the gas turbine and the heat exchanger, was carried out. In this study, wire cutting slurry of silicon wafer and the graphite rod of spent zinc-carbon battery were applied to the starting materials for the synthesis. The powders of Si or Si+SiC were obtained from the waste material by filtration, gravity separation and magnetic separation. Graphite powder was produced by dismantling, grinding and gravity separation from spent zinc-carbon battery. The synthesis of SiC could be completed from the mixture powders of Si and C or Si+SiC and C at the condition of equivalent ratio of Si and C, atmosphere of Ar or vacuum, temperature of above 1$600^{\circ}C$ and 2 hours reactions. The purity of synthesized Si-C was above 99%.

• Proceedings of the KIEE Conference
• /
• 1999.07d
• /
• pp.1827-1829
• /
• 1999

반도체 소자가 고집적화 되고 미세화 될수록 좁은 면적에 여러 기능을 가진 우물을 형성시켜야하나 기존의 우물로는 고온 장시간 열처리로 인하여 측면 확산이 깊게 되고, 불순물 농도 분포는 표면으로부터 농도가 점차 낮아진다. 따라서 기존 우물의 불순물 분포로는 기생 트랜지스터에 의한 렛치-엎과 알파 입자에 의한 SER의 감소를 위하여 필요한 벌크에서의 고농도 분포를 유지하기가 곤란하다. 이러한 문제는 차세대 반도체 개발을 위해서는 반드시 해결해야 할 것이며 이것을 해결할 수 있는 공정으로는 고 에너지 이온 주입과 저온, 단시간 열처리이다. 고 에너지 이온 주입 시의 불순물 분포를 어떻게 제어할 것인가에 대한 것과 여기서 부수적으로 나타나는 격자 손상과 그 회복 및 잔류결함의 성질을 어떻게 알고 이를 게터링 등에 이용할 것이냐에 대한 것이다. 실리콘 기판 내로 가속된 이온은 실리콘 격자와 충돌하면서 많은 1차 결함이 생기고. 이들은 후속 열처리 과정에서 활성화되면서 대부분은 실리콘 격자의 위치에 들어가 활성화되고. 그 나머지는 실리콘내의 격자간 산소, 격자간 실리콘. 격자 빈자리와 상호 작용을 하여 2차 결함을 형성한다. 에피택셜 웨이퍼와 p-type웨이퍼에 비소 이온을 고에너지로 주입후 2단계 열처리에 의한 농도분포변화와 핵생성과 결함성장에 관해 실험하였고, 핵생성온도는 $600^{\circ}C$이하이고, 성장에 필요한 온도는 $700^{\circ}C$이상이다.

• Journal of Applied Reliability
• /
• v.14 no.1
• /
• pp.11-19
• /
• 2014

A colloidal silica slurry has been manufactured by mixing nano silica powders having different grain size to improve the reliability of Sapphire wafer CMP process. The main reliability problem of CMP process such as the breaking of wafer can be prevented by reducing the size of particles in a slurry. While existing commercial colloidal silica slurries are usually made of single grain size silica powder of about 120nm, in the present study 40nm and 100nm silica powders are mixed to achieve a similar removal rate. The new colloidal silica slurry showed wafer removal rate of $3.04{\mu}m/120min$ while that of a commercial colloidal silica slurry was $3.03{\mu}m/120min$. The roughness was less than $4{\AA}$ and scratch was 0. It is also expected that the reduction of the size of nano silica particles can improve the dispersion stability and prolong the useful life of the slurry.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.104-107
• /
• 2005

Recently, according to the development of semiconductor industry, needed to high-integration and high-functionality. These changes are required for silicon wafer of large scale diameter and precision of TTV (Total Thickness variation). So, in this research, suggest that the method of monitoring system is using motor current. This method is needed for observation of silicon wafer grinding process. Motor current sensor is consisted of hall sensor. Hall sensor is known to catching of change of current. Received original signal is converted to the diginal, then, it is calculated RMS values, and then, it is analysed in computer. Generally, the change of force is relative to the change of current, So this reason, in this research tried to monitoring of motor current change, and then, it will be applied to analysis for silicon wafer grinding process. using motor current sensor.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.11 no.4
• /
• pp.12-18
• /
• 2002

Silicon on insulator(SOI) wafer is used in a variety of microsensor applications in which thermal deformations and other mechanical effects may dominate device Performance. One of major Problems associated with the manufacturing Processes of the microaccelerometer based on the tunneling current concept is thermal deformations and thermal stresses. This paper deals with finite element analysis(FEA) of residual thermal deformations causing popping up, which are induced in micrormaching processes of a microaccelerometer. The reason for this Popping up phenomenon in manufacturing processes of microaccelerometer may be the bending of the whole wafer or it may come from the way the underetching occurs. We want to seek after the real cause of this popping up phenomenon and diminish this by changing manufacturing processes of mic개accelerometer. In microaccelerometer manufacturing process, this paper intend to find thermal deformation change of the temperature distribution by tunnel gap and additional beams. The thermal behaviors analysis intend to use ANSYS V5.5.3.