• Proceedings of the Korean Vacuum Society Conference
• /
• 1999.07a
• /
• pp.119-119
• /
• 1999

GaN-based 물질들은 blue와 UV 영역의 LED, LD와 같은 광소자가 상용화되었을 뿐만아니라 HBT, FET와 같은 전기소자로도 널리 응용될 시점이지만 아직까지 해결되지 않은 문제점들이 있다. 그 중에 하나가 바로 GaN의 격자상수와 일치하는 기판이 없어 발생하는 dislocation인데, 이를 해결하기 위한 방법으로 새로운 기판이나, buffer, 또는 새로운 성장방법(ELOG) 등을 시도하고 있으나 dislocation density는 아직 높은 (107~1010cm-2) 상태이다. 이에 본 연구에서는 dislocation을 줄이기 위한 방책으로 InxGa1-xN를 새로운 buffer층으로 사용하여 GaN 박막을 MBE 방법으로 성장하였다. InxGa1-xN를 선택한 이유는 GaN와의 격자상수차이가 In0.12Ga0.88N일 경우 거의 일치한다는 보고가 있으며, 특히 InGaN의 melting point는 GaN의 성장온도 보다는 약간 높기 때문에 GaN 박막을 성장할 때와 식힐 때의 InGaN 원자결합은 약하게 작용되며, 결국 이는 열적인 stress를 줄여주게 된다. 이와 같이 성장된 GaN 박막은 그 결정성을 XRD로 분석하였고, 표면과 계면을 SEM으로 관찰하였다. 그리고 그 광학적 특성을 저온 PL로서 조사하였다. 그 결과를 살펴보면 35$^{\circ}$ 근방에서 GaN(0002) peak가 나온 것으로 보아 wurtzite 구조가 성장됨을 XRD로부터 확인하였다. 그리고 저온 (12K) PL에서는 3.470eV의 D$^{\circ}$X peak뿐만 아니라 3.258eV에 해당하는 peak를 얻었는데, 이는 InxGa1-xN buffer layer의 vapour pressure가 높은 (<50$0^{\circ}C$)에 도달하게 됨으로써 dissociation이 일어나면서 초기 성장이 이루어졌고 이는 다시 계면에서의 inter-diffusion을 발생시킨 것으로 보여진다.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.32 no.5
• /
• pp.50-57
• /
• 2004

For the application of structural health monitoring such as AE detection, a stabilized FBG sensor system with wide dynamic range was proposed. A tunable Fabry-Perot filter with narrow free spectral range(FSR) was used to simplify the multiplexing demodulator for FBG vibration sensors. Stabilization controlling system was also developed for the maintenance of maximum sensitivity of the sensors. In order to verify the performance of the proposed FBG vibration sensor system, we measured sensitivity, and the system showed the average sensitivity of 256 $n{\in}_{mas}/{\sqrt{Hz}}$. Finally, multi-points vibration tests using in-line FBG sensors were conducted to validate the multiplexing performance of the FBG system.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1996.10a
• /
• pp.728-731
• /
• 1996

• Journal of the Korean Ceramic Society
• /
• v.25 no.2
• /
• pp.143-153
• /
• 1988

In order to synthesize the partially stabilized $\alpha$-Sialon, A1N and Y2O3 were added to synthesized $\alpha$-Si3N4. The phase composition, mechanical properties, micro structure, etc, of the synthesized $\alpha$-Sialon were investigated. Partially stabilized $\alpha$-Sialon ceramics could be synthesized from the composition which was a little deviated from x=0.4, x=0.6 composition along the Si3N4.0.1Y2O3:0.9AlN tie line at 1750-180$0^{\circ}C$ for 2 hrs in N2 atmosphere. It is assumed that A1N is more closely related than Y2O3 to the formation of $\alpha$-Sialon, and that A1N is more easily dissolved into $\alpha$-structure than into $\beta$-structure. In Ya2O3-rich phase mechanical properties were observed to be poor because of formation of mellilite, grain growth, and thermal decomposition of $\alpha$-Sialon. The maximum values of M.O.R, KIC and hardness are 723 MPa, 4.5MN/㎥/2 and 19.3 GPa, respectively, and they were observed for the $\alpha$-Sialon ceramics sintered at 178$0^{\circ}C$.

• Korean Journal of Materials Research
• /
• v.9 no.7
• /
• pp.688-694
• /
• 1999

Ti-6Al-4V-N films have been grown onto glass substrates by dc reactive magnetron sputtering from a Ti-6Al-4V-N alloy target at different nitrogen partial pressure, input powers and sputtering times. The influence of various sputtering conditions on structural properties of Ti-6Al-4V-N films was investigated by measuring their X-ray diffraction. The quaternary Ti-6Al-4V-N film is crystallizing in a face centered cubic TiN structure, the lattice parameter is smaller than the TiN parameter as titanium atoms of the TiN lattice are replaced by aluminum and vanadium atoms. The films show the (111) preferred orientation and the (111) peak intensity decreases as the nitrogen partial pressure is increased, but the intensity increases as the sputtering time is increased. The deposition rate and the grain size are alto related with the variation of various sputtering conditions.

• Polymer(Korea)
• /
• v.33 no.2
• /
• pp.144-152
• /
• 2009

The thermal and optical properties of poly {1-(cholesteryloxycarbonylalkanoyloxy) ethylene}s (PCALEn, n=2$\sim$8,10, the number of methylene units in the spacer) were investigated. All of the homologues formed monotropic cholesteric phases with left-handed helical structures. PCALEn with n=2 or 10, in constrast with PCALEn with $3{\leq}n{\leq}8$, did not display reflection colors over the full cholesteric range, suggesting that the helical twisting power of the cholesteryl group highly depends on the length of the spacer connecting the cholesteryl group to the polyethylene chain. The glass transition temperatures decreased with increasing n. The isotropic-cholesteric phase transition temperatures decreased with increasing n up to 7 and showed an odd-even effect. However it became almost constant when n is more than 7. This behavior is rationalized in terms of the change in the average shape of the side chain on varing the parity of the spacer. This rationalization also accounts for the observed variation of the entropy gain for the clearing transition. The thermal stability and degree of order in the mesophase and the temperature dependence of the optical pitch observed for PCALEn were significantly different from those reported for cellulose tri(cholesteryloxycarbonyl)alkanoates. The results were discussed in terms of the differences in the chemical structure and flexibility of main chain and the number of the mesogenic units per repeating unit.

• Journal of IKEEE
• /
• v.6 no.2 s.11
• /
• pp.136-145
• /
• 2002

The simulator for the analysis of the lattice temperature under the steady-state condition is developed. The heat flow equation using the Slotboom variables is discretized and the integration method of the thermal conductivity without using the numerical analysis method is presented. The simulations are executed on the $N^+P$ junction diode and BJT to verify the proposed method. The average relative error of the lattice temperature of $N^+P$ diode compared with DAVINCI is 2% when 1.4[V] forward bias is applied and the average relative error of the lattice temperature of BJT compared with MEDICI is 3% when 5.0[V] is applied to the collector contact and 0.5[V] is applied to the base contact. BANDIS using the proposed method of integration of thermal conductivity needs 3.45 times of matrix solution to solve one bias step and DAVINCI needs 5.1 times of matrix solution MEDICI needs 4.3 times of matrix solution.

• Proceedings of the KSME Conference
• /
• 2004.11a
• /
• pp.811-816
• /
• 2004

The tactile force sensor was studied using Fiber Bragg Grating (FBG). The FBG is able to multiplex easily and is immune to electromagnetic environment. A sensor frame was designed to a cantilever beam type. Strain of a beam is related with the peak shift of a bragg wavelength. Finite Element Method (FEM) was used for getting an appropriate thickness from 0.2 mm to 0.3 mm thick. FEM results showed that 0.3 mm thick was suitable for the force range 10 N. The force resolutions of 0.039 N and 0.113 N were obtained with optical spectrum analyser and tunable Fabry-Perot filter, respectively.

• Journal of KIISE:Software and Applications
• /
• v.26 no.5
• /
• pp.604-612
• /
• 1999

데이터웨어하우스는 데이터분석을 위한 특수 목적의 데이터베이스로, 주로 데이터분석을 위한 질의(이하 OLAP 질의)가 던져진다. 그런데 OLAP질의는 수백만개의 레코드를 가지 테이블에 대해 하나 이상의 aggregation 함수와 group-by 연산자가 포함되므로, 질의 처리 시간은 수 분에서 수 시간이 걸린다. 이를 개선하기위하여 데이터 큐브를 구현함에 있어서의 문제는 디스크 공간이 한정되어 있기 때문에 평균 질의 처리 시간이 훨씬 짧아진다하더라도 모든 셀들을 실체화할수 없다. 따라서 한정된 디스크 공간을 최대한 활용하면서 가능한 빠른 평균 질의 처리 시간을 얻을 수 있도록 데이터 큐브의 일부만을 실체화 시켜야한다. 본 논문의 주제와 관련된 연구로는 Harinarayan[4] 이 제안한 greedy 알고리즘이 있다. 이 알고리즘은 1) 데이터 큐브를 격자구조로 표현한후, 2) 격자의 위에서부터 아래로 차례로 뷰들을 방문하면서 방문한 뷰가 실체화되 경우 데이터 큐브에 주는 이익을 계산한다. 3) 그 중 가장 이익이 큰 것을 선택한다. 3) 그중 가장 이익이 큰 것을 선택한다. 2)와 3)의 과정은 k개의 뷰를 선택할 경우, k번 반복된다. 이 알고리즘의 운영 시간은 데이터 큐브를 구성하는 뷰의 개수가 n개이고 그 중에서 k개를 실체화할 경우에 O(kn2)이다. 본 논문에서는 운영 시간을 향상시킨 수정된 greedy 알고리즘을 제안한다. 알고리즘 내부에서 실체화할 뷰를 선택할 때 격자를 단순화시킨 트리를 사용함으로써, 알고리즘 운영시간을 O(kn2)에서 O(kn)으로 향상시켰다.

• Korean Journal of Optics and Photonics
• /
• v.19 no.1
• /
• pp.25-30
• /
• 2008

In this paper, a novel architecture for device miniaturization of multimode interference-based (MMI) couplers is proposed. The coupling length of $N{\times}N$ MMI couplers scales as the square of the MMI region width so that the use of these structures with large-N ports can require large chip size. To solve the design problem, the butterfly grating-assisted MMI couplers, that have smaller device dimensions than conventional MMI couplers, are discussed and evaluated. Numerical simulations and novel design rules for such structures derived through theoretical analysis are presented.