• Journal of the Mineralogical Society of Korea
• /
• v.14 no.1
• /
• pp.12-20
• /
• 2001

충북 영동군 동창광산에서 산출되는 일라이트(illite)광선의 표면화학특성을 전위차 적정 실험과 FITEQL3.2 프로그램을 이용하여 연구하였다. 정량 Xtjs 회절 분석에 의한 일라이트 광석의 광물조석은 석영 46.6% 일라이트 41.6% 카올리나이트(kaolinite) 11.8%이며 $N_2$BET 방법에 의하여 구한 비표면적은 $6.52 m^2$g이다. 전위차 적정 실험결과를 그란(Gran)법을 적용하여 구한 일라이트광석의 영전하점($pH_{pznpc}$ )은 pH 3.9 총표면 자리 밀도는 21.24 sites/$nm^2$이다. 표면 복합체 모델중 일정 용량 모델을 적용해 일라이트 광석의 표변 특성에 알맞는 모델을 찾아보았다. 일라이트 광석의 표면을 사면체 자리와 팔면체 자리로 나누어 설정한 2sites$-3pK_{ a}$s 모델은 변수값이 수렴되지않았으므로 부적절하다고 판단된다. 일라이트 광석의 표면을 하나의 균질한 흡착표면으로 가정해서 설정한 1 site -1 $K_{a}$ 와 1 site -2 $pK_{a}$ s 모델 사이에는 뚜렷한 차이는 없지만, 1 site -1 $pK_{a}$ 모델의 WSOS/DF 값이 17, 1 site - 2 $pK_a{s}$ 모델은 26으로서 앞 모델이 보다 적절하다. 이 결과는 일라이트 광석 표면에서 수소의 해리와 첨가 반응 중 첨가 반응을 무시하여도 표면반응을 설명하는 데 큰 무리가 없음을 시사한다. 가장 적절하다고 판단되는 1 site -1 $pK_{a}$ 모델의 $pK_{a}$ 값은 4.17, specific capacitance는 $6F/m^2$ 표면 자리 농도는 $1.15\Times10^{-3}$ mol/L 이다.

• Journal of the Korean GEO-environmental Society
• /
• v.11 no.1
• /
• pp.19-26
• /
• 2010

The batch experiments and response surface methodology (RSM) have been applied to the investigation of the cadmium (Cd) adsorption by coal fly ash (CFA). CFA having maximum Cd removal mass of 8.51 mg/g were calculated from Langmuir model. Cd removal reaction with different initial pH ranged from 4 to 9. When the initial pH was higher, Cd was removed more by adsorption and precipitation. These results suggest that the lower pH cause an increase of $H^+$ ion concentration which competed with Cd ions for exchange sites in CFA. Also, The Cd adsorption was mathematically described as a function of parameters initial Cd concentration ($X_1$), initial pH ($X_2$), and initial CFA mass ($X_3$) being modeled by use of the Box-Behnken methods. Empirical models were developed to describe relationship between the experimental variables and response. Statistical analysis indicates that tree factors ($X_1$, $X_2$, and $X_3$) on the linear term (main effects), and tree factors ($X_1X_2$, $X_1X_3$, and $X_2X_3$) on the non-linear term (Interaction effect; cross-product) had significant effects, respectively. In this case, the value of the adjusted determination coefficient (adjusted $R^2=0.9280$) was closed to 1, showing a high significance of the model. Statistical results showed the order of Cd removal at experimental factors to be initial initial pH > initial Cd concentration > initial CFA mass.

• The Korean Journal of Applied Statistics
• /
• v.12 no.1
• /
• pp.67-81
• /
• 1999

2차 반응표면분석을 위한 부분합성계획은 실험 비용이 많이 들거나, 실험 자체가 어려워서 시간의 소비자 많은 경우, 실험오차의 독립추정이 가능할 때 효과적이다. 반응표면분석에서는 회전성과 기울기 회전성을 만족하는 것이 중요하다. 따라서 본 논문에서는 회전성과 기울기 회전성의 관점에서 부분합성계획을 살펴보고 또한 인자의 수와 중심점의 수가 변함에 따라서 어떤 $\alpha$(중심에서 축점까지의 거리)의 값이 최적의 실험계획이 되도록 하는지를 연구하였다.

• Journal of the military operations research society of Korea
• /
• v.33 no.1
• /
• pp.59-74
• /
• 2007

Computer simulation models are in general quite complex and time-consuming to run, and therefore, a simpler meta-model is usually constructed for further analysis. In this paper, JANUS, a war-game simulator, is used to describe a certain tank combat situation. Then, second-order response surface and artificial neural network meta-models are developed using the data from eight different experimental designs. Relative performances of the developed meta-models are compared in terms of the mean squared error of prediction. Computational results indicate that, for the given problem, the second-order response surface meta-model generally performs better than the neural network, and the orthogonal array-based Latin hypercube design(LHD) or LHD using maximin distance criterion may be recommended.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.37 no.7
• /
• pp.841-849
• /
• 2013

In this study, a 2-step stamping model with an additional 1st stamping tool is proposed to reduce stamping flaws in the curved parts of a dimple in a nuclear fuel spacer grid. First, the strains of curved part of dimple are characterized via a comparison with strain solutions in pure bending. A reference 2D finite element (FE) model of 1-step stamping is then established, and the corresponding maximum strain is obtained. By varying the values of design variables of the 1st stamping tool in the 2-step stamping model, FE solutions are obtained to express the strain as a function of process variables, which provides the optimum values of process variables. Finally, applying these optimum values to a 3D FE model, we demonstrate the enhanced formability of the proposed 2-step stamping model.

• Journal of the Mineralogical Society of Korea
• /
• v.17 no.2
• /
• pp.157-168
• /
• 2004

The surface chemical properties of aqueous kaolinite and halloysite were studied using a potentiometric titration experiment and a computer program FITEQL3.2. Among the surface complexation models a constant capacitance model was selected for this study. The 2 sites - 3 p $K_{a}$ s model, in which the surfaces were assumed to have tetrahedral and octahedral sites, was reasonable for the description of the experimental data. The surface charges of both minerals were negative above pH of 4. The higher the pH, the lower the proton surface charge densities of both minerals. The ≡ $SiO^{[-10]}$ site played an important role in cation adsorption in acid and neutral pH range; whereas the ≡ Al $O^{[-10]}$ site was in an alkaline pH range. The optimized intrinsic constants of kaolinite, p $K_{a2(Si)}$$^{int}$, p $K_{al(Al)}$$^{int}$ and p $K_{a2(Al)}$$^{int}$ were 4.436, 4.564, and 8.461 respectively, and those of halloysite were 7.852, 3.885, and 7.084, respectively. The total Si and Al surface sites concentrations of kaolinite were 0.215 and 0.148 mM, and those of halloysite were 0.357 and 0.246 mM. The ratio of Si and Al surface site densities ([≡SiOH]:[≡AlOH]) of both minerals was 1 : 0.69. The total surface site density of kaolinite, 3.774 sites/n $m^2$, was 1.6 times larger than that of halloysite, 2.292 sites/n $m^2$./TEX>.

• Journal of Korean Society of Environmental Engineers
• /
• v.39 no.10
• /
• pp.568-574
• /
• 2017

The purpose of this study was to optimize experimental conditions (time ($X_1$) (ranging of 26.36 - 93.64 min), concentration of sulfuric acid ($X_2$) (ranging of 0-2.5%) and temperature ($X_3$) (ranging of $136.4-203.6^{\circ}C$) for an organosolv pretreatment process to extract lignin from waste wood. The resulting quadratic model equation using RSM (response surface methodology) represented y (lignin yield) = $-79.89+0.91X_1+9.8X_2-2.54{\times}10^{-3}X_1{^2}-2.11X_2{^2}$. The $R^2$ (coefficient of determination) value of 0.8531 for a model indicates this model has statistically significant predictors at the 10% levels. The predictive results optimized by quadratic model produced a lignin yield of 12.46 g/100 g of dry wood under conditions of $178.2^{\circ}C$ and 2.32% $H_2SO_4$. The lignin yield was more affected by the acid catalyst concentrations than the reaction temperature, but the reaction time was not an influential factor for improving lignin extraction from waste wood in this organosolv pretreatment. According to ANOVA (analysis of variance), the significance probability (p-value) of model was smaller than 0.001 and simulation of obtained model equations showed a good reproducibility based on actual organosolv tests under optimal conditions.

• Clean Technology
• /
• v.23 no.4
• /
• pp.413-420
• /
• 2017

The batch experiments and response surface methodology (RSM) have been applied to the investigation of the Cs adsorption with zeolite X synthesized using coal fly ash generated from the thermal power plant. Regression equation formulated for Cs adsorption was represented as a function of response variables. The model was highly relevant because the decision coefficient ($r^2$) was 0.9630. It was confirmed from the statistical results that the removal efficiency of Cs was affected by the order of experimental factors as pH > Cs concentration > temperature. The adsorption kinetics were more accurately represented by a pseudo second-order model. The maximum adsorption capacity calculated from the Langmuir isotherm model was $151.52mg\;g^{-1}$ at 293 K. Also, according to the thermodynamic parameters calculated from Vant Hoff equation, it could be confirmed that the adsorption reaction was an endothermic reaction and a spontaneous process.

• Applied Chemistry for Engineering
• /
• v.10 no.1
• /
• pp.85-92
• /
• 1999

The complete catalytic oxidation of vinyl chloride was investigated over chromium oxide supported ${\gamma}$-alumina using a fixed bed micro-reactor at temperature between 240 and $300^{\circ}C$ and concentration between 600 and 3500 ppm. The oxidation of vinyl chloride was nonlinear in the concentration of vinyl chloride and zeroth order in the concentration of oxygen. The addition of HCl and $H_2O$ as products to the feed stream didn't influence the conversion of vinyl chloride. Several kinetic rate model were tested to describe the data over the range of condition investigated, and developed a model which provide the best correlation of experimental data. The resulting model of kinetic rate was derived by assuming that the reacting occurred via adsorption and subsequent decomposition of the vinyl chloride onto the oxygen covered chromium oxide surface, with the reaction being inhibited by the adsorption of vinyl chloride. The percent standard deviation between the predicted and experimental was about 5.2%, and the activation energy was 18.9 kcal/mol.

• Journal of Korea Soil Environment Society
• /
• v.4 no.1
• /
• pp.97-108
• /
• 1999

Reduction 2, 4, 6-trinitrotoluene by zero valent iron was studied in a batch reactor under anoxic conditions. Results showed that the removal of trinitrotoluene (TNT)followed a pseudo-first order reaction and the rate was linearly dependent on the available reactive surfau area of the zero valent iron surface area, resulting a rate constant of 0.0981min$^{1}m$$^{-2}m$. High concentrations of the final product, presumably triaminotoluene which needs to be treated by other means, accumulated in the solution. However , little amount of TAT was extracted from the metal surface by using acetonitrile or phosphate buffered water (pH 7.0). Other common major intermediate in biological TNT degradation, a group of aminodinitrotoluenes, was not detected in the solution. Therefore, it is postulated that the reduction of nitro group by $Fe^0$ occurs simultaneously in all three positions and a TNT reduction model by zero valent iron was suggested.