• YAKHAK HOEJI
• /
• v.42 no.6
• /
• pp.627-633
• /
• 1998

Impurity profiling analysis of methamphetamine seized in Korea was investigated for the evidential and intelligent purpose. Samples were extracted with ethylacetate which contai ns internal standard of dioctylsebacate under basic condition and extracts were analyzed by GC-FID. Ephedrine, chloroephedrine & 1,2-dimethyl-3-phenylaziridine were identified impurities in illicit methamphetamine by GC-MS. These impurities revealed that most of abused methamphetamine in Korea were synthesized from ephedrine as a starting material. For the classification of samples. firstly, 24 impurity peaks were selected after inspection of every peak in 50 samples as the specific markers of impurities. Secondly, corresponding peak retention time and area ratio to the internal standard were calculated and database was created with values of 24 peaks by in-house program. Finally, cluster analysis was attempted with the resultant profiles using the STAR plot, which was based on the Euclidian distance for evaluating similarity among samples. A total of 76 samples were divided into 8 different groups within 90% statistical similarity and inter-batch samples showed similar impurity patterns by this procedure. In conclusion, the analysis of impurities is a suitable index for estimation the common or different origin of methamphetamine sample.

• The Korean Journal of Applied Statistics
• /
• v.34 no.5
• /
• pp.711-721
• /
• 2021

In this paper, we propose the method of adjusting thresholds using impurity indices in classification analysis on imbalanced data. Suppose the minority category is Positive and the majority category is Negative for the imbalanced binomial data. When categories are determined based on the commonly used 0.5 basis, the specificity tends to be high in unbalanced data while the sensitivity is relatively low. Increasing sensitivity is important when proper classification of objects in minority categories is relatively important. We explore how to increase sensitivity through adjusting thresholds. Existing studies have adjusted thresholds based on measures such as G-Mean and F1-score, but in this paper, we propose a method to select optimal thresholds using the chi-square statistic of CHAID, the Gini index of CART, and the entropy of C4.5. We also introduce how to get a possible unique value when multiple optimal thresholds are obtained. Empirical analysis shows what improvements have been made compared to the results based on 0.5 through classification performance metrics.

• Communications for Statistical Applications and Methods
• /
• v.8 no.3
• /
• pp.885-894
• /
• 2001

Decision tree methods is the one of data mining techniques. Classification trees are used to predict a class label. When a tree grows, the conventional splitting criteria use the weighted average of the left and the right child nodes for measuring the node impurity. In this paper, new splitting criteria for classification trees are proposed which improve the interpretablity of trees comparing to the conventional methods. The criteria search only for interesting subsets of the data, as opposed to modeling all of the data equally well. As a result, the tree is very unbalanced but extremely interpretable.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1999.05a
• /
• pp.47-49
• /
• 1999

The air and impurity gaps generated by a defect in homogeneous material have a difference index of refraction compare to the homogeneous material. Therefore an electromagnetic wave has a multiple reflection between the two indices of refraction. If the gap has a small thickness, it is difficult to know the gap in time domain. However, the presence of the gaps can be known in frequency domain by total transmission and phase difference. Also, the thickness and index of refraction can be measured by well known Febry-Perot theory.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.17 no.6
• /
• pp.580-585
• /
• 2004

Silicon dioxide (SiO$_2$) is widely used as a gate dielectric material for thin film transistors (TFT) and semiconductor devices. In this paper, SiO$_2$ films were grown by APCVD(Atmospheric Pressure chemical vapor deposition) at the high temperature. Experimental investigations were carried out as a function of $O_2$ gas flow ratios from 0 to 200 1pm. This article presents the SiO$_2$ gate dielectric studies in terms of deposition rate, refrative index, FT-IR, C-V for the gate dielectric layer of thin film transistor applications. We also study defect passivation technique for improvement interface or surface properties in thin films. Our passivation technique is Forming Gas Annealing treatment. FGA acts passivation of interface and surface impurity or defects in SiO$_2$ film. We used RTP system for FGA and gained results that reduced surface fixed charge and trap density of midgap value.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.5 no.4
• /
• pp.318-331
• /
• 1995

The c - axis oriented single crystal of $LiNbO_3$ and $LiNbO_3$ : 5mol%MgO was success-fully grown by Floating zone method using halogen lamp as a heat source. The effects of the sintering condition of the feed rod and the atmosphere gas during the crystal growth on the be havior of the feed rod/melt interface were studied for growing crystal with the high quality, and then, the optimum growth conditions were determined by studying the experimental param eters, such as gas flow rate, pulling rate, rotation speeds of the feed rod and the seed. The grown crystals were analyzed using the chemical etching to observe the tch pattern and the ICP (Inductively Coupled Plasma) to determine the composition uniformity and the impurity content of Fe. The effects of additive (5 mol % MgO) on the transmittance and refractive index was, also, analyzed. In order to compare the nonlinear optical oharacteristics of $LiNbO_3$ with those of the other optical materials, the nonlinear optical refractive index ($n_2$) was calcu l lated using the measured refractive index.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.34 no.3
• /
• pp.167-173
• /
• 2008

Pearlescent pigments have been widely used in cosmetic applications. Up to date; the most widely used pearl effect pigment is the mica-based pigment, which uses natural mica as the substrate that is in turn coated with metal of oxide interference layer. However, when natural mica is employed as a base material the final product often has a yellowish color, mainly due to the fact that natural mica contains low levels of iron as an impurity[1,2]. This study was focused on developing a pearl pigment which might have a similar sparkling effect as snow. This effect was found to be due to its structure and purity, and this concept was also applied to development of our pearl pigments. More specifically, this invented pearl effect pigments are the mixture of glass-flake and glass-flake coated metal oxides and present the optical properties of snow matrix such as refractive index and particle size, unlike only the glass-flake or glass-flake coated metal oxides to be applied in. Using base material having similar physical properties (refractive index and particle size) as snow matrix as platelet for pearl effect pigments, these invented pigments present a three-dimensional glittering effect of the snow matrix. With this invented figments an applied; we achieved the beauty of snow crystal from makeup products containing these pigments.

• Korean Journal of Materials Research
• /
• v.26 no.8
• /
• pp.430-437
• /
• 2016

Aluminum oxide ($Al_2O_3$) thin films were grown by atomic layer deposition (ALD) using a new Al metalorganic precursor, dimethyl aluminum sec-butoxide ($C_{12}H_{30}Al_2O_2$), and water vapor ($H_2O$) as the reactant at deposition temperatures ranging from 150 to $300^{\circ}C$. The ALD process showed typical self-limited film growth with precursor and reactant pulsing time at $250^{\circ}C$; the growth rate was 0.095 nm/cycle, with no incubation cycle. This is relatively lower and more controllable than the growth rate in the typical $ALD-Al_2O_3$ process, which uses trimethyl aluminum (TMA) and shows a growth rate of 0.11 nm/cycle. The as-deposited $ALD-Al_2O_3$ film was amorphous; X-ray diffraction and transmission electron microscopy confirmed that its amorphous state was maintained even after annealing at $1000^{\circ}C$. The refractive index of the $ALD-Al_2O_3$ films ranged from 1.45 to 1.67; these values were dependent on the deposition temperature. X-ray photoelectron spectroscopy showed that the $ALD-Al_2O_3$ films deposited at $250^{\circ}C$ were stoichiometric, with no carbon impurity. The step coverage of the $ALD-Al_2O_3$ film was perfect, at approximately 100%, at the dual trench structure, with an aspect ratio of approximately 6.3 (top opening size of 40 nm). With capacitance-voltage measurements of the $Al/ALD-Al_2O_3/p-Si$ structure, the dielectric constant of the $ALD-Al_2O_3$ films deposited at $250^{\circ}C$ was determined to be ~8.1, with a leakage current density on the order of $10^{-8}A/cm^2$ at 1 V.

• Nuclear Engineering and Technology
• /
• v.5 no.2
• /
• pp.137-149
• /
• 1973

A method of activation analysis, based on the irradiation and counting of an iron wire which contains manganese impurity as the single comparator. has been elaborated by critical evaluation of nuclear data involved in activation and activity measurement. The variation of effective cross section is investigated as a function of the spectral index and other parameters such as a measure of the proportion of epithermal neutrons in the reactor spectrum. The errors induced by shifts in the neutron spectrum in the irradiation positions are discussed. The known amount of each element is irradiated simultaneously together with the single comparator, and the obtained values are compared with the known amount of each element. The results show that en general the random errors are not greater than those obtained by using the conventional relative method, but the systematic errors were up to about 20%. This method is applied to the determinations of fourteen rare earth elements in monazite as well as other seven elements in the standard kale powder. The satisfactory reproducibility of the present method makes possible the determination of the elements with an accuracy attainable with the conventional relative method.

• Journal of the Korean Ceramic Society
• /
• v.37 no.8
• /
• pp.781-786
• /
• 2000

An Er-doped sulfide fiber was drawn, and its spectroscopic properties were analyzed. Compositions of a 1000 ppmwt Er3+-doped core and an undoped clad were Ge30-Ga1-Asg-S61 and Ge30-As8-S62, in at.%, respectively. Refractive index of the core composition was approximately 0.01 high than that of the clad. In order to enhance the mechanical stability as well as to prevent infiltration of impurity ions such as OH-, an UV-curable polymer was used for the coating. The optical loss of a fiber formed directly from a polymer coated core rod without cladding was ∼15 dB/m at 1.06$\mu\textrm{m}$. In the case of a fiber with core/clad structure, the optical loss was so high that the stimulated emission of erbium fluorescence was not evident. It is believed that presence of inhomogeneous core/clad interface and crystalline aggregates precipitated in the clad region were responsible for the high optical loss. On the other hand, fluorescence characteristics of Er3+ embedded in the core region were more or loss deteriorate compared to fiber preform, which is attributed to the redistribution of the Er ions along with the partial crystallization of the core glass during the fiberization process.