• Korean Journal of Materials Research
• /
• v.15 no.5
• /
• pp.318-322
• /
• 2005

Ti-capped NiSi contacts were formed on $p^+/n$ junctions to improve the leakage problem and then Cu was deposited without removing the Ti-capping layer in an attempt to utilize as a diffusion barrier. The electrical characteristics of these $p^+/n$ diodes with Cu/Ti/NiSi electrodes were measured as a function of drive-in RTA(rapid-thermal annealing) and silicidation temperature and time. When drive-in annealed at $900^{\circ}C$, 10 sec. and silicided at $500^{\circ}C$, 100 sec., the diodes showed the most excellent I-V characteristics. Especially, the leakage current was $10^{-10}A$, much lower than reported data for diodes with NiSi contacts. However, when the $p^+/n$ diodes with Cu/Ti/NiSi contacts were furnace-annealed at $400^{\circ}C$ for 40 min., the leakage current increased by 4 orders. The FESEM and AES analysis revealed that the Ti-capping layer effectively prohibited the Cu diffusion, but was ineffective against the NiSi dissociation and consequent Ni diffusion.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2011.02a
• /
• pp.188-188
• /
• 2011

Recently, scaling down of ULSI (Ultra Large Scale Integration) circuit of CMOS (Complementary Meta Oxide Semiconductor) based electronic devices, the electronic devices, become much faster and smaller size that are promising property of semiconductor market. However, very narrow interconnect line width has some disadvantages. Deposition of conformal and thin barrier is not easy. And metallization process needs deposition of diffusion barrier and glue layer for EP/ELP deposition. Thus, there is not enough space for copper filling process. In order to get over these negative effects, simple process of copper metallization is important. In this study, Cu-V alloy layer was deposited using of DC/RF magnetron sputter deposition system. Cu-V alloy film was deposited on the plane SiO2/Si bi-layer substrate with smooth surface. Cu-V film's thickness was about 50 nm. Cu-V alloy film deposited at $150^{\circ}C$. XRD, AFM, Hall measurement system, and AES were used to analyze this work. For the barrier formation, annealing temperature was 300, 400, $500^{\circ}C$ (1 hour). Barrier thermal stability was tested by I-V(leakage current) and XRD analysis after 300, 500, $700^{\circ}C$ (12 hour) annealing. With this research, over $500^{\circ}C$ annealed barrier has large leakage current. However vanadium-based diffusion barrier annealed at $400^{\circ}C$ has good thermal stability. Therefore thermal stability of vanadium-based diffusion barrier is desirable for copper interconnection.

• Korean Journal of Materials Research
• /
• v.28 no.7
• /
• pp.428-434
• /
• 2018

Hydrogen evolution on a steel surface and subsequent hydrogen diffusion into the steel matrix are evaluated using an electrochemical permeation test with no applied cathodic current on the hydrogen charging side. In particular, cyclic operation in the permeation test is also conducted to clarify the corrosion-induced hydrogen evolution behavior. In contrast to the conventional perception that the cathodic reduction reaction on the steel in neutral aqueous environments is an oxygen reduction reaction, this study demonstrates that atomic hydrogen may be generated on the steel surface by the corrosion reaction, even in a neutral environment. Although a much lower permeation current density and significant slower diffusion kinetics of hydrogen are observed compared to the results measured in acidic environments, they contribute to the increase in the embrittlement index. This study suggests that the research on hydrogen embrittlement in ultra-strong steels should be approached from the viewpoint of corrosion reactions on the steel surface and subsequent hydrogen evolution/diffusion behavior.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.20 no.3
• /
• pp.1050-1060
• /
• 1996

Recently, attention has been paid to the flame diagnostic by noncontact methods which dose not deform the flame shape. One of them is a method which is using the radical luminous intensity. Generally, this diagnostic method using radical luminous has been investigated its reliability by applying to laminar flame. This study, however, investigated each radical luminous signals through stocastical analysis like auto-correlation, cross-correlation, phase and coherence which were acquired from measuring radical luminous intensity of OH, CH, $O_{2}$, radicals in turbulent diffusion flame. To compare radical luminous intensity in flame with temperature, ion current and concentration , radious distribution of each properties was investigated and considered. In radical luminous intensity, correlation in the reaction zone of flame was higher than in correlation in combusted gas zone. And radious distribution of radical luminous intensity was corresponded with radious distribution of temperature, ion current and concentration. The result of the study confirms that a radical luminous flame diagnosis is possible in the turbulent diffusion flame.

• International Journal of Precision Engineering and Manufacturing-Green Technology
• /
• v.5 no.5
• /
• pp.613-621
• /
• 2018

The electrically assisted brazing of a ferritic stainless steel with nickel-based filler metal is experimentally investigated. During electrically assisted brazing of a lap joint, the temperature of the joint is first rapidly increased to a brazing temperature and held nearly constant for a specific period using a pulsed electric current. Microstructural analysis results strongly suggest that the electric current during electrically assisted brazing enhances diffusion between the filler metal and the ferritic stainless steel, thus inducing significantly thicker diffusion zones compared with induction brazing. The mechanical test results show that the strength of the electrically assisted brazing joint is comparable to or even superior to those of the joint fabricated by induction brazing, while the process time of the electrically assisted brazing is significantly shorter than that of induction brazing.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.33 no.4
• /
• pp.288-297
• /
• 2009

A serpentine channel geometry often used in a fuel cell has a strong pressure gradient between adjacent channels in specific regions. The pressure gradient helps some amount of reactant gas penetrate through a gas diffusion layer(GDL). As a result, the overall serpentine flow structure is slightly different from the intention of a designer. The purpose of this paper is to examine the effect of serpentine flow structure on current density distribution. By using a commercial code, STAR-CD, a numerical simulation is performed to analyze the fuel cell with high aspect ratio of active area. To increase the accuracy of the numerical simulation, GDL permeabilities are measured with various compressive forces. Three-dimensional flow field and current density distribution are calculated. For the verification of the numerical simulation results, water condensation process in the cathode channel is observed through a transparent bipolar plate. The result of this study shows that the region of relatively low current density corresponds that of dropwise condensation in cathode channels.

• Ocean Systems Engineering
• /
• v.13 no.3
• /
• pp.313-324
• /
• 2023

The vertical distribution of suspended sediments in the mangrove-mud coast is complicated due to the characterization of cohesive sediment properties, and the influence of hydrodynamic factors. In this study, the time-evolution of suspended sediment concentration (SSC) in water depth is simulated by a one-dimensional model. The model applies in-situ data measured in October 2014 at the outer station in Cu Lao Dung coastal areas, Soc Trang, Vietnam. In the model, parameters which have influence on vertical distribution of SSC include the settling velocity Ws and the diffusion coefficient Kz. The settling velocity depends on the cohesive sediment properties, and the diffusion coefficient depends on the wave-current dynamics. The settling velocity is determined by the settling column experiment in the laboratory, which is a constant of 1.8 × 10^-4 ms^-1. Two hydrodynamic conditions are simulated including a strong current condition and a strong wave condition. Both simulations show that the SSC near the bottom is much higher than ones at the surface due to higher turbulence at the bottom. At the bottom layer, the SSC is strongly influenced by the current.

• ETRI Journal
• /
• v.36 no.3
• /
• pp.490-497
• /
• 2014

The market performance of WiBro in Korea has not been as expected, and its rapid diffusion in the near future is unlikely owing to the existence of competing services. There has been little research on the factors affecting this low market diffusion. This study is based on an analytical framework in which a lack of service capability and the insufficiency of service-related activities have resulted in the current poor market performance. An expert survey was conducted on WiBro specialists and verified using the analytical hierarchy process method. The result of this analysis is as follows: underinvestment in network deployment and marketing, insufficient promotional policies, and a shortage of service capabilities are to be analyzed as the main causes of WiBro's low market diffusion.

• Journal of Powder Materials
• /
• v.7 no.3
• /
• pp.137-142
• /
• 2000

In order to analyze the densification behaviour of stainless steel powder compacts during hot isostatic pressing (HIP) at elevated temperatures, a power-law creep constitutive model based on the plastic deformation theory for porous materials was applied to the densification. Various densification mechanisms including interparticle boundary diffusion, grain boundary diffusion and lattice diffusion mechanisms were incorporated in the constitutive model, as well. The power-law creep model in conjunction with various diffusion models was applied to the HIP process of 316L stainless steel powder compacts under 50 and 100 MPa at $1125^{\circ}C$. The results of the calculations were verified using literature data. It could be found that the contribution of the diffusional mechanisms is not significant under the current process conditions.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.16 no.2
• /
• pp.203-209
• /
• 2018

To correctly predict the neutron behavior based on diffusion calculations, it is necessary to adopt well-specified boundary conditions using suitable diffusion approximations to transport boundary conditions. Boundary conditions such as the zero net-current, the Marshak, the Mark, the zero scalar flux, and the Albedo condition have been used extensively in diffusion theory to approximate the reflective and vacuum conditions in transport theory. In this paper, we derive and analyze these conditions to prove their mathematical validity and to understand their physical implications, as well as their relationships with one another. To show the validity of these diffusion boundary conditions, we solve a sample problem. The results show that solutions of the diffusion equation with these well-formulated boundary conditions are very close to the solution of the transport equation with transport boundary conditions.