• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.14 no.2
• /
• pp.93-98
• /
• 2001

YMnO$_3$ films are excellent gate dielectric materials of ferroelectric random access memories (FRAMs) with MFSFET (metal -ferroelectric-semiconductor field effect transistor) structure because YMnO$_3$ films can be deposited directly on Si substrate and have a relatively low permittivity. Although the patterning of YMnO$_3$ thin films is the requisite for the fabrication of FRAMs, the etch mechanism of YMnO$_3$ thin films has not been reported. In this study, YMnO$_3$thin films were etched with Cl$_2$/Ar gas chemistries in inductively coupled plasma (ICP). The maximum etch rate of YMnO$_3$ film is 285$\AA$/min under Cl$_2$/(Cl$_2$+Ar) of 1.0, RF power of 600 W, dc-bias voltage of -200V, chamber pressure of 15 mTorr and substrate temperature of $25^{\circ}C$. The selectivities of YMnO$_3$ over CeO$_2$ and $Y_2$O$_3$ are 2.85, 1.72, respectively. The selectivities of YMnO$_3$ over PR and Pt are quite low. Chemical reaction in surface of the etched YMnO$_3$ thin films was investigated with X-ray photoelectron spectroscopy (XPS) surface of the selected YMnO$_3$ thin films was investigated with X-ray photoelectron spectroscopy(XPS) and secondary ion mass spectrometry (SIMS). The etch profile was also investigated by scaning electron microscopy(SEM)

• Applied Chemistry for Engineering
• /
• v.20 no.1
• /
• pp.99-103
• /
• 2009

HBr and $O_2$ in $Cl_2$ gas ambient for the high density plasma gate etching has been used to increase the performance of gate electrode in semiconductor devices. When an un-doped amorphous silicon layer was used for a gate electrode material, the notching profile was observed at the outer sidewall foot of the outermost line. This phenomenon can be explained by the electron shading effect: i.e., electrons are captured at the photoresist sidewall while ions pass through the photoresist sidewall and reach the oxide surface at a narrowly spaced pattern during the over etch step. The potential distribution between gate lines deflects the ions trajectory toward the gate sidewall. In this study, an appropriate mechanism was proposed to explain the occurrence of notching in the gate electrode of un-doped amorphous silicon.

• Journal of Advanced Marine Engineering and Technology
• /
• v.35 no.2
• /
• pp.232-237
• /
• 2011

Recently, continuous research on cryogenic valves is being carried out with the rapid development of the cryogenic valve-related industry, and especially, high performance of cryogenic valves is being promoted due to the breakthrough development and demand of users, etc., of the mechanical, shipbuilding, semiconductor and display industry and the aerospace industry field, but it is the reality that technical development and research on cryogenic application equipment on vacuum insulation are insufficient. The present research focused on interception of heat exchange with the outside by keeping low pressure after installing a jacket pipe outside a stem and also considered heat transfer properties on changes in pressure of a vacuum part and radius of a jacket which can reduce heat exchange for effective heat transmission control by studying it in a three-dimensional numerical analysis method.

• Korean Journal of Materials Research
• /
• v.13 no.10
• /
• pp.635-639
• /
• 2003

We investigated dry etching of GaAs and AiGaAs in a high density planar inductively coupled plasma system with BCl$_3$and BCl$_3$/Ar gas chemistry. A detailed etch process study of GaAs and ALGaAs was peformed as functions of ICP source power, RIE chuck power and mixing ratio of $BCl_3$ and Ar. Chamber process pressure was fixed at 7.5 mTorr in this study. The ICP source power and RIE chuck power were varied from 0 to 500 W and from 0 to 150 W, respectively. GaAs etch rate increased with the increase of ICP source power and RIE chuck power. It was also found that etch rates of GaAs in $15BCi_3$/5Ar plasmas were relatively high with applied RIE chuck power compared to pure 20 sccm $BCl_3$plasmas. The result was the same as AlGaAs. We expect that high ion-assisted effect in $BCl_3$/Ar plasma increased etch rates of both materials. The GaAs and AlGaAs features etched at 20 sccm $BCl_3$and $15BCl_3$/5Ar with 300 W ICP source power, 100 W RIE chuck power and 7.5 mTorr showed very smooth surfaces(RMS roughness < 2 nm) and excellent sidewall. XPS study on the surfaces of processed GaAs also proved extremely clean surfaces of the materials after dry etching.

• Journal of the Korean Vacuum Society
• /
• v.1 no.1
• /
• pp.126-133
• /
• 1992

This study deals with the crystal growth and the optical characteristics of CdS thin films activatedby silver. CdS:Ag thin films were deposited by using an electron beam evaporation(EBE) technique in vacuumof 1.5X 10-'torr, voltage of 4 kV, current of 2.5 mA and substrate temperature of 250$^{\circ}$C CdS:Ag photoconductivefilms prepared by EBE method show high photoconductivity after annealing at about 550"c for 0.5 h in air andAr gas.The grain size of CdS:Ag thin films annealed in Ar atmosphere (1 atm) was grown over 1 ym and the thicknessof the films is 4-5 pm. The analysis of X-ray diffraction patterns shows that the crystal structures are hexagonal.The diffraction line by (00.2) plane can only be observed, indicating that c-axis of hexagonal grows preferentiallyperpendicular to the substrate. The profiles of photoluminescence spectra of CdS:Ag films show Gaussian typecurves at room temperature, the maximum peak spectral sensitivity of CdS:Ag is located at the wavelength of520 nm.We annealed CdS:Ag thin films in air and Ar vapor in order to make the CdS photoconductors having theintensive photocurrent, the broad distribution of the photocurrent spectrum and the large value of the ratioof the photocurrent (pc) to the dark current(dc). We found that CdS:Ag thin films annealed in air atmospherewas the best one.air atmosphere was the best one.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2011.02a
• /
• pp.159-159
• /
• 2011

Among the semiconductor ternary compounds in the I-III-$VI_2$ series, $CulnS_2$ ($CulnSe_2$) are one of the promising materials for photovoltaic applications because of the suitability of their electrical and optical properties. The $CuInS_2$ thin film is one of I-III-$VI_2$ type semiconductors, which crystallizes in the chalcopyrite structure. Its direct band gap of 1.5 eV, high absorption coefficient and environmental viewpoint that $CuInS_2$ does not contain any toxic constituents make it suitable for terrestrial photovoltaic applications. A variety of techniques have been applied to deposit $CuInS_2$ thin films, such as single/double source evaporation, coevaporation, rf sputtering, chemical vapor deposition and chemical spray pyrolysis. This is the first report that $CuInS_2$ thin films have been prepared by Aerosol Jet Deposition (AJD) technique which is a novel and attractive method because thin films with high deposition rate can be grown at very low cost. In this study, $CuInS_2$ thin films have been prepared by Aerosol Jet Deposition (AJD) method which employs a nozzle expansion. The mixed fluid is expanded through the nozzle into the chamber evacuated in a lower pressure to deposit $CuInS_2$ films on Mo coated glass substrate. In this AJD system, the characteristics of $CuInS_2$ films are dependent on various deposition parameters, such as compositional ratio of precursor solution, flow rate of carrier gas, stagnation pressure, substrate temperature, nozzle shape, nozzle size and chamber pressure, etc. In this report, $CuInS_2$ thin films are deposited using the deposition parameters such as the compositional ratio of the precursor solution and the substrate temperature. The deposited $CuInS_2$ thin films will be analyzed in terms of deposition rate, crystal structure, and optical properties.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.02a
• /
• pp.334-334
• /
• 2012

Transparent oxide semiconductors have recently attracted much attention as channel layer materials due to advantageous electrical and optical characteristics such as high mobility, high stability, and good transparency. In addition, transparent oxide semiconductor can be fabricated at low temperature with a low production cost and it permits highly uniform devices such as large area displays. A variety of thin film transistors (TFTs) have been studied including ZnO, InZnO, and InGaZnO as the channel layer. Recently, there are many studies for substitution of Ga in InGaZnO TFTs due to their problem, such as stability of devices. In this work, new quaternary compound materials, tantalum-indium-tin oxide (TaInSnO) thin films were fabricated by using co-sputtering and used for the active channel layer in thin film transistors (TFTs). We deposited TaInSnO films in a mixed gas (O2+Ar) atmosphere by co-sputtering from Ta and ITO targets, respectively. The electric characteristics of TaInSnO TFTs and thin films were investigated according to the RF power applied to the $Ta_2O_5$ target. The addition of Ta elements could suppress the formation of oxygen vacancies because of the stronger oxidation tendency of Ta relative to that of In or Sn. Therefore the free carrier density decreased with increasing RF power of $Ta_2O_5$ in TaInSnO thin film. The optimized characteristics of TaInSnO TFT showed an on/off current ratio of $1.4{\times}108$, a threshold voltage of 2.91 V, a field-effect mobility of 2.37 cm2/Vs, and a subthreshold swing of 0.48 V/dec.

• Korean Journal of Materials Research
• /
• v.14 no.6
• /
• pp.402-406
• /
• 2004

$WSi_2$, $TiSi_2$, $CoSi_2$, and $TaSi_2$ are general silicides used today in semiconductor devices. $WSi_2$ thin films have been proposed, studied and used recently in CMOS technology extensively to reduce sheet resistance of polysilicon and $n^{+}$ region. However, there are several serious problems encountered because $WSi_2$ is oxidized and forms a native oxide layer at the interface between $WSi_2$ and $Si_3$$N_4$. In this study, we have introduced 20 $slm-N_2$ gas from top to bottom of the furnace in order to control native oxide films between $WSi_2$ and $Si_3$$N_4$ film. In resulting SEM photographs, we have observed that the native oxide films at the surface of $WSi_2$ film are removed using the long injector system.

• Fire Science and Engineering
• /
• v.13 no.3
• /
• pp.19-26
• /
• 1999

$CO_2$ extinguishing system is the most $\phi$pular among the gas extinguishing system. $CO_2$ is usually stored with liquified state. But, it gasifies at the tip of nozzle when $CO_2$ was released through the pipe and head. A ro$\alpha$n temperature is very low when $CO_2$ was released. So electrical instrument, magnetic storage equipment and memory semiconductor are electrically or physically injured by cooling effect in a few minutes. So, we intend to find out temperature profile and electrical damage in compartment area, and supply basic d data for research and making standards and code through the full scale experiment. As result of experiment on the damage due to cooling effect from $CO_2$ extinguishing system, i instantaneous discharging temperature. was $-82.5^{\circ}C$ in average. An average temp. in the compartment after discharging $CO_2$ was $-40^{\circ}C$.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.3
• /
• pp.215-222
• /
• 2001

Neural network based models of semiconductor manufacturing processes have been shown to offer advantages in both accuracy and generalization over traditional methods. However, model development is often complicated by the fact that back-propagation neural networks contain several adjustable parameters whose optimal values unknown during training. These include learning rate, momentum, training tolerance, and the number of hidden layer neurOnS. This paper presents an investigation of the use of genetic algorithms (GAs) to determine the optimal neural network parameters for the modeling of plasma-enhanced chemical vapor deposition (PECVD) of silicon dioxide films. To find an optimal parameter set for the neural network PECVD models, a performance index was defined and used in the GA objective function. This index was designed to account for network prediction error as well as training error, with a higher emphasis on reducing prediction error. The results of the genetic search were compared with the results of a similar search using the simplex algorithm.