JOURNAL OF CHEMICAL ENGINEERING OF JAPAN Vol. 50(2017) No. 9

When rotation of an impeller is started, the torque is larger than that at a steady state. This torque is important for the design of the impeller. However, the relationship between the starting torque and the rotational speed and the shape of the impeller has not yet been sufficiently investigated. The present study investigates the influence of the rotational speed, the number of blades, and the blade width on the starting torque of a vertical paddle impeller. Furthermore, experiments and a CDF study were conducted to examine the mechanism generating the starting torque. The starting torque was separated into the 1st and 2nd starting torque. The 1st starting torque was generated while the impeller was accelerating and CFD results showed that the flow was only around the impeller blades. It was found that the 1st starting torque correlated with the angular acceleration of the impeller and the fluid density. The 2nd starting torque was generated after the impeller reached a steady rotational speed. In this period, the negative pressure region at the back of the blades increased. It was also found that the 2nd starting torque originated from the profile drag of the blades and was proportional to the square of the impeller rotational speed and the blade width. Also, it was proportional to the 0.6th power of the number of blades. In the present study, the relationship between the 2nd starting torque and the torque under completely baffled conditions was also investigated, and it was confirmed that they were almost equal. From this result, it was found that the measurement of starting torque was a possible method for obtaining the torque under completely baffled conditions.

This paper discussed the feasibility of four imidazolium-type ionic liquids (IL) as membrane materials for the separation of C₆ and C₇ aromatic/aliphatic or alicyclic hydrocarbon mixtures. Novel nanoparticle-supported membranes composed of these ILs were fabricated successfully. Liquid membranes composed of IL and aluminum oxide nanoparticles mixtures, supported by a hydrophobic porous membrane, were prepared and their efficiencies in the separation of aromatic compounds through vapor permeation were evaluated in terms of their aromatic selectivity, total permeate flux, and permeability. For an n-hexane/benzene mixture, the IL [emim][HSO₄] gave the highest separation factor of 85 and the IL [emim][B(CN)₄] showed the highest permeate flux of 0.18 kg/(m²·h) (emim: 1-ethyl-3-methylimidazolium). Additionally, a vapor sorption test for the hydrocarbons was conducted in the ILs and sulfolane, which is a typical solvent for aromatic compounds. All the ILs showed absorption tendencies similar to that of sulfolane. The permeabilities of the ILs were 32–977 Barrer for the aromatic compounds, 0.61–186 Barrer for the alicyclic ones and 0.31–197 Barrer for the aliphatic ones. The permeabilities obtained were independent of the operating partial pressure.

Preparation of anion exchangers using melamine sponge (MS) and adsorption of nitrate ion was investigated. Carbonized melamine sponge was prepared at 500°C with ZnCl₂ impregnation (Z6) to develop the pore structure. Z6 was further treated by CH₃I to introduce quaternary nitrogen onto the carbon surface and referred to as Z6-Q. Based on the result of X-ray photoelectron spectroscopy (XPS), Z6-Q contained more quaternary nitrogen than MS and Z6. Adsorption of nitrate ion on Z6-Q reached an equilibrium state the most quickly among the prepared samples. Adsorption kinetic curves indicated that the rate-limiting process was diffusion inside the pore for the adsorption of nitrate ion. In the adsorption of nitrate ion ranging 0.1–0.7 mmol/g, Z6-Q released equimolar chloride ion, suggesting that Z6-Q could have anion exchange sites to adsorb nitrate ion.

The mechanism by which ice slurry crystallizes under a vacuum was investigated in this work. The crystallization kinetic models for two different aqueous solutions (glycerol; a combination of glycerol and nano-SiO₂) were developed based on the classic Avrami theory. The values of the Avrami exponent n under different experimental conditions were between 0.58 and 0.84, indicating the one-dimensional flake thickening growth of ice crystals. Glycerol inhibited the growth of ice crystals due to the hydroxyl-containing structure, and nano-SiO₂, which served as a nucleating agent, could enhance the crystallization process effectively. The growth morphology of ice crystals, including the size and shape, were also observed. Results showed that ice crystals produced by the vacuum method were generally oval in shape. Adding nano-SiO₂ to the glycerol solution reduced the size of the ice crystals, suggesting that nano-SiO₂ is a promising additive for ice slurry systems. The study of crystallization mechanism of ice slurry can be useful for future design of vacuum ice production systems.

Effect of temperature on nitric oxide (NO) and sulfur dioxide (SO₂) removal in a dielectric barrier discharge (DBD) reactor at 25, 50, 100, 150, 200, and 300°C was investigated. The reduced electric field (E/N) changed from 40 to 113 Td at 25°C, and it increased from 77 to 217 Td at 300°C, promoting the formation of active radicals (N and O). In the NO/SO₂/N₂/O₂ system, the increased temperature enhanced the decomposition of O₃ and the formation of NO, thereby reducing the NO removal efficiency. However, increasing temperature enhanced the O formation and promoted the rate of SO₂ removal reactions, both of which contribute to SO₂ removal. In the NO/SO₂/N₂/O₂/NH₃ system, when the temperature was increased, the generation of OH radicals and the rate of related reactions were promoted, enhancing the NO and SO₂ efficiency slightly more than that in the NO/SO₂/N₂/O₂ system. In the NO/SO₂/N₂/O₂/C₂H₄ system, the addition of C₂H₄ promoted the NO removal, but it did not affect the SO₂ removal. When the temperature was increased, the generation of HO₂ and the rate of NO removal were promoted markedly, increasing the NO removal.

A photoresponsive tracer that changed color under ultraviolet (UV) irradiation was prepared by immobilizing a photoacid generator and methyl orange on a poly(vinyl chloride) (PVC) carrier resin to evaluate the performance of dry-type powder photoreactors. The color transition of the tracer during exposure to UV light was recorded using a video camera. By analyzing the recorded images, the color change could be continuously detected as a change in the green component of the tracer’s RGB color value. The tracer changed color as a function of the received cumulative radiant energy. Two consecutive reactions occurred on the tracer surface. The first reaction involved the release of a proton upon exposure of the photoacid generator to UV irradiation followed by the reaction of the released proton with methyl orange, which caused the color change. The rate constant of each reaction could be estimated by analyzing the rate of change of the green RGB value under UV irradiation. This photoresponsive tracer enables the visualization of the photoreactions in dry-type powder photoreactors, and allows both the radiant energy received by materials inside the reactor and their reaction kinetics to be determined.

In this paper, a new model simplification method, thermodynamic equilibrium analysis solver, and solvent extraction process simulator are developed for the modeling and simulation of the solvent extraction process with saponified PC88A for purifying rare earth metals. Firstly, the number of nonlinear equilibrium equations to be simultaneously solved is decreased by the proposed model simplification method such that the initial guess problems and poor convergence or divergence problems associated with solving the nonlinear equations can be overcome. Secondly, a new thermodynamic equilibrium analysis solver combined with the proposed model simplification method is proposed to estimate the equilibrium concentrations of the rigorous first principle equilibrium model without any numerical problems. Thirdly, a solvent extraction process simulator combined with the proposed thermodynamic equilibrium analysis solver is developed to estimate all the concentrations of all stages of the solvent extraction process. Simulation study has confirmed that the proposed method can simulate the solvent extraction process without numerical problems such as the initial guess problems and poor convergence or divergence problems. A trial version of the rare earth solvent extraction process simulation software is freely available on the web at http://pse.knu.ac.kr. The proposed modeling and simulation method can contribute immensely to the efficient design of the solvent extraction process with saponified PC88A.

The pyrolysis of residual crude palm oil from empty fruit bunch was investigated to determine the optimal conditions using a two-level experimental design performed in a 3,000 cm³ tube SS316 stainless steel reactor. The influence of the reaction temperature, residual CPO feed rate, and N₂ flow rate over MgO on the product distribution yield were determined using a distillation gas chromatograph (DGC) and physicochemical analyses. A two-level factorial design was used to investigate the parameters that influence biofuel yield and product distribution. Additionally, the hydrocarbon molecular structure under optimal conditions was identified using a gas chromatograph equipped with a mass spectrometer. Finally, the optimal conditions of the catalytic cracking of residual CPO were determined as follows; a temperature of 500°C, a 3 cm³ min⁻¹ CPO feed rate, a 129.45 cm³ min⁻¹ N₂ flow rate and 42.78 v/v% of MgO. This process produced 68.712 wt% of diesel-like fraction content with an intricate composition of aliphatic hydrocarbons consisting of 48.34 wt% alkane and 33.66 wt% alkene that was also diesel-like.

The present paper proposes a monitoring framework for a nonlinear multi-mode process based on the kernel sparse representation based classifier (KSRC). In KSRC, samples from multiple modes are projected onto a high dimensional feature space using the kernel trick. In the kernel space, a test sample is regressed with the training samples using an l₁ constraint sparse regression. The l₁ constraint ensures that most of the regression coefficients corresponding to the training samples which are not from the same mode as the test sample shrink to zero. Mode assignment is then achieved by investigating the regression errors obtained from different modes. Comparing to other methods, KSRC considers the interrelationship between samples and can better capture the data generating mechanism. In order to reduce the scale of the l₁ constraint regression problem, the present paper suggests to use kernel principal component analysis (KPCA) for dimension reduction. For the purpose of monitoring, two Bayesian monitoring statistics are constructed by integrating the monitoring results in different modes using the posterior probability of a test sample falling into each mode, which is calculated based on the regression error. The confidence limits of the monitoring statistics are obtained through kernel density estimation (KDE). Application studies to a simulation example and an ironmaking blast furnace show the advantages of the proposed monitoring strategy.

Regioselective enzymatic glycosylation of D-glucose and various structural alcohols for synthesizing β-D-glucopyranosides was successfully conducted. Among the β-glycosidases from seven potential plant seeds, the low-cost β-glycosidase from Prunus persica seed displayed excellent activity for salidroside preparation. Several crucial parameters, such as the solvent, buffer pH, buffer content, substrate molar ratio, enzyme dosage, and temperature were examined for yield optimization. Under the optimal conditions, the initial reaction rate and yield of salidroside were as high as 3.04 mM/h and 18.90%, respectively. Moreover, the alcohol acceptor recognition of the enzyme in synthesizing β-D-glucopyranosides was also investigated. The experimental data indicated that β-glycosidase activity and yield varied widely among all the tested alcohols due to the specific spatial structures of various acceptors.