• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.41 no.8
• /
• pp.869-874
• /
• 1992

The physical and electrical properties of TiN/TiSiS12T bilayer were studied. The TiN/TiSiS12T bilayer was formed by rapid thermal anneal in NHS13T ambient after the Ti film was deposited on silicon substrate. The Ti film reacts with NHS13T gas to make a TiN layer at the surface and reacts with silicon to make a TiSiS12T layer at the interface respectively. It was found that the formation of TiN/TiSiS12T bilayer depends on RTA temperature. In this experiment, competitive reaction for TiN/TiSiS12T bilayer occured above $600^{\circ}C$. Ti-rich TiNS1xT layer and Ti-rich TiSiS1xT layer and Ti-rich TiSiS1xT layer were formed at $600^{\circ}C$. stable structure TiN layer TiSiS12T layer which has CS149T phase and CS154T phase were formed at $700^{\circ}C$. Both stable TiN layer and CS154T phase TiSiS12T layer were formed at 80$0^{\circ}C$. The thickness of TiN/TiSiS12T bilayer was increased as the thickness of deposited Ti film increased.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.2
• /
• pp.224-233
• /
• 1991

The multilayer expanson of single layer NN (Neural Network) was needed to solve the linear seperability problem as shown by the classic example using the XOR function. The EBP (Error Back Propagation ) learning rule is often used in multilayer Neural Networks, but it is not without its faults: 1)D.Rimmelhart expanded the Delta Rule but there is a problem in obtaining Ca from the linear combination of the Weight matrix N between the hidden layer and the output layer and H, wich is the result of another linear combination between the input pattern and the Weight matrix M between the input layer and the hidden layer. 2) Even if using the difference between Ca and Da to adjust the values of the Weight matrix N between the hidden layer and the output layer may be valid is correct, but using the same value to adjust the Weight matrixd M between the input layer and the hidden layer is wrong. Recognitron III was proposed to solve these faults. According to simulation results, since Recognitron III does not learn the three layer NN itself, but divides it into several single layer NNs and learns these with learning patterns, the learning time is 32.5 to 72.2 time faster than EBP NN one. The number of patterns learned in a EBP NN with n input and output cells and n+1 hidden cells are 2**n, but n in Recognitron III of the same size. [5] In the case of pattern generalization, however, EBP NN is less than Recognitron III.

• Journal of Advanced Marine Engineering and Technology
• /
• v.32 no.3
• /
• pp.461-467
• /
• 2008

It was studied the effect of a pre-deposited ultrathin Al layer as part of a buffer layer for the growth of GaN. AlN buffer layers were deposited on a Si(111) substrate using an RF sputtering technique, followed by GaN using hydride vapor phase epitaxy (HVPE). Several atomic layers of Al were deposited prior to AlN sputtering and the samples were compared with the others grown without pre-deposition of Al. And it was also studied the influence of AlN buffer layer thickness on the growth of GaN. The peak wavelength of the photoluminescence (PL) was varied with increasing the thickness of the GaN and AlN layers. The optimum thickness of AlN on a Si(111) substrate with an ultrathin Al layer was about $260{\AA}$. Scanning electron microscope (SEM) images showed coalescent surface morphology and X-ray diffraction (XRD) showed a strongly oriented GaN(0002) peak.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.15 no.7
• /
• pp.564-569
• /
• 2002

Heteroepitaxial GaN films were grown on sapphire substrates in order to study the effects of the buffer layer's surface roughness and thermal treatment on the epitaxial layer's quality. For this, GaN buffer layers were grown at $550^{\circ}C$ with various TMGa flow rates and durations of growth, and annealed at $1010^{\circ}C$ for 3 min after the temperature was raised by 23 ~ $92^{\circ}C/min$, and then GaN epitaxial layers were grown at $1000^{\circ}C$. It has been found that the buffer layer's surface roughness and the thermal treatment condition are critical factors on the quality of the epitaxial layer. When a buffer layer was frown with a TMGa flow rate of $24\mu mole/min$ for 30 sec, the surface roughness of the buffer lather was minimum and when the thermal ramping rate was $30.6^{\circ}C/min$ on this layer, the successively grown epitaxial layer's crystalline and optical qualities were optimized with a specular morphology. The minimum full width at half maximum(FWHM) of GaN(0002) x-ray diffraction peak and that of near-band-edge(NBE) peak from a room temperature photoluminescence (PL) were 5 arcmin and 9 nm, respectively.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.08a
• /
• pp.219-219
• /
• 2013

The effect of an electron blocking layer (EBL) on V-I curves in GaN/InGaN multiple quantum well is investigated. For the first time, we found that curves were intersected at 3.012 V and analyzed the reason for intersection. The forward voltage in LEDs with an p-AlGaN EBL is larger than without p-AlGaN EBL at low injection current because the Mg doping efficiency for p-GaN layer was higher than that of p-AlGaN layer. However, the forward voltage in LEDs with an p-AlGaN EBL is smaller than without p-AlGaN EBL at high injection current because the carriers overflow from the active layer when injection current increases in LEDs without p-AlGaN EBL and in case of LED with p-AlGaN EBL, the carriers are blocked by EBL.

• Korean Journal of Materials Research
• /
• v.11 no.7
• /
• pp.575-579
• /
• 2001

GaN buffer layer and epilayer have been grown on sapphire (0001) by metal organic chemical vapor deposition (MOCVD). GaN buffer layer ranging from 26 nm to 130 nm in thickness was grown at 55$0^{\circ}C$ prior to the 4 $\mu\textrm{m}$ thick GaN epitaxial deposition at 110$0^{\circ}C$. After GaN buffer layer growth, buffer layer surface was examined by atomic force microscopy (AFM). As the thickness of GaN buffer layer was increased, surface morphology of GaN epilayer was investigated by scanning electron microscopy (SEM). Double crystal X-ray diffraction (DCXRD) and Raman spectroscopy were employed to study crystallinity of GaN epilayers. Optical properties of GaN epilayers were measured by photoluminescence (PL). The epilayer grown with a thin buffer layer had rough surface, and the epilayer grown with a thick buffer layer had mirror-like surface of epilayer. Although the stress on the latter was larger than on the former, its crystallinity was much better. These results imply that the internal free energy is decreased in case of the thick buffer layer. Decrease in internal free energy promotes the lateral growth of the GaN film, which results in the smoother surface and better crystallinity.

• Korean Journal of Materials Research
• /
• v.20 no.4
• /
• pp.187-193
• /
• 2010

Chromium nitride (CrN) samples with two different layer structures (multilayer and single layer) were coated on bipolar plates of polymer electrolyte membrane fuel cells (PEMFC) using the reactive sputtering method. The effects with respect to layer structure on corrosion resistance and overall cell performance were investigated. A continuous and thin chromium nitride layer ($Cr_{0.48}\;N_{0.52}$) was formed on the surface of the SUS 316L when the nitrogen flow rate was 10 sccm. The electrochemical stability of the coated layers was examined using the potentiodynamic and potentiostatic methods in the simulated corrosive circumstances of the PEMFC under $80^{\circ}C$. Interfacial contact resistance (ICR) between the CrN coated sample and the gas diffusion layer was measured by using Wang's method. A single cell performance test was also conducted. The test results showed that CrN coated SUS316L with multilayer structure had excellent corrosion resistance compared to single layer structures and single cell performance results with $25\;cm^2$ in effective area also showed the same tendency. The difference of the electrochemical properties between the single and multilayer samples was attributed to the Cr interlayer layer, which improved the corrosion resistance. Because the coating layer was damaged by pinholes, the Cr layer prevented the penetration of corrosive media into the substrate. Therefore, the CrN with a multilayer structure is an effective coating method to increase the corrosion resistance and to decrease the ICR for metallic bipolar plates in PEMFC.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.12
• /
• pp.2258-2263
• /
• 2006

In this parer, hump characteristics of short-channel nMOSFETs induced by moistures of the ILD(inter-layer dielectric) layer in the shallow trench isolation (STI) process are investigated and the method for hump suppression is proposed Using nMOSFETs with various types of the gate and a measurement of TDS-APIMS (Thermal Desorption System-Atmospheric Pressure ionization Mass Spectrometry), hump characteristics were systematically analyzed and the systemic analysis based hump model was presented; the ILD layer over poly-Si gate of nMOSFET generates moistures, but they can't diffuse out of the SiN layer due to the upper SiN layer. Consequently, they diffuses into the edge between the gate and STI and induces short-channel hump. In order to eliminate moisture in the ILD layer by out-gassing method, the annealing process prior to the deposition of the SiN layer was carried out. As the result, short-channel humps of the nMOSFETs were successfully suppressed.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2010.02a
• /
• pp.124-124
• /
• 2010

Much interest has been focused on InGaN-based materials and their quantum structures due to their optoelectronics applications such as light emitting diode (LED) and photovoltaic devices, because of its high thermal conductivity, high optical efficiency, and direct wide band gap, in spite of their high density of threading dislocations. Build-in internal field-induced quantum-confined Stark effect in InGaN/GaN quantum well LED structures results in a spatial separation of electrons and holes, which leads to a reduction of radiative recombination rate. Therefore, many growth techniques have been developed by utilizing lateral over-growth mode or by inserting additional layers such as patterned layer and superlattices for reducing threading dislocations and internal fields. In this work, we investigated various characteristics of InGaN multiple quantum wells (MQWs) LED structures grown on selectively wet-etched porous (SWEP) GaN template layer and compared with those grown on non-porous GaN template layer over c-plane sapphire substrates. From the surface morphology measured by atomic force microscope, high resolution X-ray diffraction analysis, low temperature photoluminescence (PL) and PL excitation measurements, good structural and optical properties were observed on both LED structures. However, InGaN MQWs LED structures grown on SWEP GaN template layer show relatively low In composition, thin well width, and blue shift of PL spectra on MQW emission. These results were explained by rough surface of template layer, reduction of residual compressive stress, and less piezoelectric field on MQWs by utilizing SWEP GaN template layer. Better electrical properties were also observed for InGaN MQWs on SWEP GaN template layer, specially at reverse operating condition for I-V measurements.

• Progress in Superconductivity and Cryogenics
• /
• v.18 no.2
• /
• pp.5-7
• /
• 2016

We present an experimental investigation of the superconducting transition temperatures, $T_c$, of superconductor/ferromagnet bilayers with varying the thickness of ferromagnetic layer. FeN was used for the ferromagnetic (F) layer, and NbN and Nb were used for the superconducting (S) layer. The results were obtained using three different-thickness series of the S layer of the S/F bilayers: NbN/FeN with NbN thickness, $d_{NbN}{\approx}9.3nm$ and $d_{NbN}{\approx}10nm$, and Nb/FeN with Nb thickness $d_{Nb}{\approx}15nm$. $T_c$ drops sharply with increasing thickness of the ferromagnetic layer, $d_{FeN}$, before maximal suppression of superconductivity at $d_{FeN}{\approx}6.3nm$ for $d_{NbN}{\approx}10nm$ and at $d_{FeN}{\approx}2.5nm$ for $d_{Nb}{\approx}15nm$, respectively. After shallow minimum of $T_c$, a weak $T_c$ oscillation was observed in NbN/FeN bilayers, but it was hardly observable in Nb/FeN bilayers.