• Proceedings of the KIPE Conference
• /
• 2006.06a
• /
• pp.387-389
• /
• 2006

The LCD TV has many variable functions for consumer. Low temperature driving below minus 20 degree is also one of key functions. Since LCD is not self-luminance device, it is needed backlight system. Recently EEFL(External Electrode Fluorescent Lamp) widely substitutes for CCFL(Cold Cathode Fluorescent Lamp). It is more cheaper, higher efficiency, and easy to drive parallel connection compared to CCFL. In this paper, several conditions for low temperature driving of EEFL are investigated and a new boost-half bridge inverter for low temperature driving is proposed. Mode analysis is described and experimental results is represented to verify validity of the proposed inverter.

• Animal cells and systems
• /
• v.4 no.3
• /
• pp.239-243
• /
• 2000

Most space-related life science programs are expensive and time-consuming, requiring international cooperation and resources with trans-disciplinary expertise. A comprehensive future program in "life sciences in space" needs, therefore, well-defined research goals and strategies as well as a sound ground-based program. The first half of this review will describe four key aspects such as the environment in space, previous accomplishments in space (primarily focusing on amphibian embryogenesis), available resources, and recent advances in bioinformatics and biotechnology, whose clear understanding is imperative for defining future directions. The second half of this review will focus on a broad range of interdisciplinary research opportunities currently supported by the National Aeronautics and Space Administration (NASA), National Institute of Health (NIH), and National Science Foundation (NSF). By listing numerous research topics such as alterations in a diffusion-limited metabolic process, bone loss and skeletal) muscle weakness of astronauts, behavioral and cognitive ability in space, life in extreme environment, etc., we will attempt to suggest future opportunities.

• Asian Journal of Turfgrass Science
• /
• v.10 no.4
• /
• pp.275-283
• /
• 1996

An investigation was performed to reveal the relation between the storage and decomposition of the titter DNA of a Zoysia japonica grassland on Mt. Kwanak. The loss constant k of litters was 0.167. The times required for the decomposition of half, 95% and 99% of accumulated DNA on the grassland floor were 3.8, 16.6 and 27.6, respectively. The amount of DNA which is turned to living organism in the ecosystem is higher than that of crude protein. In the case of crude protein, the decay constant k was 0.181. The times needed for the decomposition of half, 95% and 99% of accumulated crude protein on the Z. japonica grassland floor were 3.8, 16.6 and 27.6 years, respectively. Key words: Zoysia japonica, Mt. Kwanak, Litter DNA, Crude protein, Decomposition, Accumulation.

• Korean Journal of Plant Taxonomy
• /
• v.40 no.1
• /
• pp.59-64
• /
• 2010

Herein is described a new species of Impatiens violascens B.U. Oh & Y.Y. Kim (Balsaminaceae) discovered on Bohyeon Mountain in Yeongcheon-si, Gyeongsangbuk-do, Korea. The new species is distinct from a closely-related species, I. textorii, by having a flower half the size that of I. textorii, sharply narrowed lower sepal, one half- to one time-coiled spur, and distal lobes of elliptical type laterally united petals. The key to these two species is provided.

• Tunnel and Underground Space
• /
• v.6 no.3
• /
• pp.260-266
• /
• 1996

Safety concern is one of the most important parameters in the design of explosive building demolition. Laboratory experiments were performed to investigate the failure behaviour of concrete columns and the effects of protection materials. Fourteen reinforced columns with two sizes were constructed and the effects of protection materials were tested for two kinds of materials: non woven fabrics and wire net. The results showed that control of gas effects is a key to the control of flying chips. It was recommended to use both wire net and non woven fabrics as primary and secondary protection materials. Such protection method was successfully applied to the explosive demolition of 16 and 17-strory apartment buildings.and the results of a simulation on a model tunneling workings using diesel equipments are introduced. In case of typical model of tunneling face, the gas concentration of human height is about one third of roof concentration and right side half of the tunnel shows better environment than left half. NOx concentration of workings can be estimated about 0.45ppm which is much lower than permissible level(5 ppm).

• Journal of Microbiology and Biotechnology
• /
• v.25 no.1
• /
• pp.33-43
• /
• 2015

GAM-1 and GAM-2, two themostable glucoamylases from Aspergillus niger B-30, possess different molecular masses, glycosylation, and thermal stability. In the present study, the effects of additives on the thermal inactivation of GAM-1 and GAM-2 were investigated. The half-lives of GAM-1 and GAM-2 at 70℃ were 45 and 216 min, respectively. Data obtained from fluorescence spectroscopy, circular dichroism spectroscopy, UV absorption spectroscopy, and dynamic light scattering demonstrated that during the thermal inactivation progress, combined with the loss of the helical structure and a majority of the tertiary structure, tryptophan residues were partially exposed and further led to glucoamylases aggregating. The thermal stability of GAM-1 and GAM-2 was largely improved in the presence of sorbitol and trehalose. Results from spectroscopy and Native-PAGE confirmed that sorbitol and trehalose maintained the native state of glucoamylases and prevented their thermal aggregation. The loss of hydrophobic bonding and helical structure was responsible for the decrease of glucoamylase activity. Additionally, sorbitol and trehalose significantly increased the substrate affinity and catalytic efficiency of the two glucoamylases. Our results display an insight into the thermal inactivation of glucoamylases and provide an important base for industrial applications of the thermally stable glucoamylases.

• Molecules and Cells
• /
• v.24 no.2
• /
• pp.210-214
• /
• 2007

The 90-kDa heat shock protein (HSP90) normally functions as a molecular chaperone participating in folding and stabilizing newly synthesized proteins, and refolding denatured proteins. The HSP90 inhibitor geldanamycin (GA) occupies the ATP/ADP binding pocket of HSP90 so inhibits its chaperone activity and causes subsequent degradation of HSP90 client proteins by proteasomes. Here we show that GA reduces the level of endogenous c-Jun in human embryonic kidney 293 (HEK293) cells in a time and dose dependent manner, and that this decrease can be reversed by transfection of HSP90 plasmids. Transfection of HSP90 plasmids in the absence of GA increases the level of endogenous c-Jun protein, but has no obvious affect on c-Jun mRNA levels. We also showed that HSP90 prolongs the half-life of c-Jun by stabilizing the protein; the proteasome inhibitor N-benzoyloxycarbonyl (Z)-Leu-Leu-leucinal (MG132) blocks the degradation of c-Jun promoted by GA. Transfection of HSP90 plasmids did not obviously alter phosphorylation of c-Jun, and a Jun-2 luciferase activity assay indicated that over-expression of HSP90 elevated the total protein activity of c-Jun in HEK293 cells. All our evidence indicates that HSP90 stabilizes c-Jun protein, and so increases the total activity of c-Jun in HEK293 cells.

• The Bulletin of The Korean Astronomical Society
• /
• v.42 no.2
• /
• pp.37.2-37.2
• /
• 2017

Since the discovery of quasars about a half century ago, it has been an open question what triggers supermassive black holes at the center of galaxies. I will talk about AGNs and their host galaxies in the context of galaxy assembly and evolution. In particular, I will focus on the fundamental, but unanswered question whether black holes play a key role in galaxy evolution from my personal perspective.

• Proceedings of the Korean Magnestics Society Conference
• /
• 2013.12a
• /
• pp.90-90
• /
• 2013

Since discovery, NdFeB permanent magnet has replaced application of the conventional magnets rapidly because of its superior physical and mechanical properties. With increasing consumption of power combined with energy resource depletion, energy efficiency is becoming more and more inportant. According to recent reports, almost almost half of the electric power is consumed by motor, and NdFeB magnets which are the core component of the motor play a key role on improving energy efficiency of the devices. In parallel with finding alternatives energy resources, research works improving energy efficiecy have been conducted world wide. Althogh NdFeB magnets usage have been expanded to various applications, key materials such as Nd and Dy, resouces lean heavily on specific area, China. Magnetic industry revently experienced skyrocketing price fluctuatioin of rare earth at around 2008. Chineses government's regulations worsened the situation and arose a necessity to develop methods to minimize rare earth use. In this presentation, POSCO's recent research works on rare earth reduction is presented including novel powder alloying method using nitrate precursors. Also, future R&D plans for rare earth free magnets is briefly introduced as well.

• Journal of Semiconductor Engineering
• /
• v.1 no.1
• /
• pp.31-37
• /
• 2020

We present an approximate DRAM architecture for energy-efficient deep learning. Our key premise is that by bounding memory errors to non-critical information, we can significantly reduce DRAM refresh energy without compromising recognition accuracy of deep neural networks. To validate the key premise, we make extensive Monte-Carlo simulations for several well-known convolutional neural networks such as LeNet, ConvNet and AlexNet with the input of MINIST, CIFAR-10, and ImageNet, respectively. We assume that the highest-order 8-bits (in single precision) and 4-bits (in half precision) are protected from retention errors under the proposed architecture and then, randomly inject bit-errors to unprotected bits with various bit-error-rates. Here, recognition accuracies of the above convolutional neural networks are successfully maintained up to the 10^-5-order bit-error-rate. We simulate DRAM energy during inference of the above convolutional neural networks, where the proposed architecture shows the possibility of considerable energy saving up to 10 ~ 37.5% of total DRAM energy.