A useful model has been proposed to calculate liquid–liquid equilibria (LLE) of non-aqueous ternary systems, which are classified as type I (only one pair among binary systems is partially miscible), based on a modified Wilson equation. The modified Wilson equation adopted contains the interaction energy parameters of constituent binary systems and the multi-component parameter for the ternary system concerned. In this study, a simple procedure has been developed to estimate those parameters from normal boiling points and solubility parameters which can be obtained by a group-contribution method. Therefore, LLE of ternary systems can be calculated by using pure-component properties alone. The procedure has been adopted for various alkane+methanol+aromatics ternary systems and good correlation performances are obtained.

We investigate the hydrodynamic performance of a baffled stirred tank for an air–water system in the turbulent regime. Comparisons among three different axial impellers were conducted: Maxblend impeller, two coaxial Pitched Blade Turbine (down-pumping mode) and two coaxial Intermig impellers. Measurements of the mean velocity field and the power consumption for gassed and ungassed conditions were obtained. The gas phase was introduced in a controlled manner using a capillary bank mounted at the bottom of the tank. The measured distributions of turbulent intensities for each impeller system for both gassed and ungassed configurations showed significantly different distributions. The flow patterns were also drastically modified for gassed conditions for the conventional agitators. The Maxblend impeller was observed to maintain a certain degree of homogeneity in the turbulent distribution for both gassed and ungassed conditions.

We use a front-tracking method to simulate inward solidification in a cylindrical container with volume change. The problem includes temporal evolution of three interfaces that meet at a triple point. The governing Navier–Stokes and energy equations are solved for the whole domain, setting the velocities in the solid phase to zero and with the non-slip condition on the solid–liquid interface. Computational results show that a cavity forms in the center of the cylinder if the density of solid is higher than that of liquid, i.e. ρ_sl>1.0. In contrast, if ρ_sl<1.0, the solidified product has a conical shape in the central region of the cylinder. The triple point also affects the solidification process as an increase in the growth angle results in a decrease in the cone angle at the top in the case of volume expansion.

Alkali carbonates supported on activated carbon (AC) for CO₂ sorption from moist flue gases by a temperature-swing operation have been proposed. The temperature dependence of CO₂ uptake by the sorbent was investigated to understand the CO₂ sorption/release behavior. The CO₂ sorption behaviors of alkali carbonates-on-AC at various temperatures were well explained based on CO₂ gas-sorbent equilibria. However, the reverse reaction of thermal decomposition of alkali hydrogencarbonates, which were the reaction products after CO₂ sorption, did not explain the sorption behaviors. The CO₂ gas-sorbent equilibrium constants were used to determine the effect of CO₂ partial pressure in the feed gas on the CO₂ sorption capacities. The results showed that the pressure-swing operation was appropriate for sorption using alkali carbonates-on-AC. At low CO₂ partial pressures in the feed gas, the CO₂ sorption capacities achieved using alkali carbonates-on-AC were higher than that achieved using K₂CO₃ solution (i.e., the Benfield process).

Hydrophobic Al₂O₃ membranes were prepared by grafting C₆F₁₃C₂H₄Si(OC₂H₅)₃ molecules, and then applied to a vacuum membrane distillation (VMD) system for concentration of sucrose solution as a model solution of fruit juice. The effects of sucrose concentration and solution temperature on the VMD performance were discussed. By exposing the surface side of the tube membrane to an aqueous solution of 10 wt% sucrose at 60°C and applying vacuum at the inner side of the Al₂O₃ tube to a pressure of 0.5 kPa, permeation flux of over 7 kg m⁻² h⁻¹ was attained through the hydrophobic Al₂O₃ membrane with asymmetric structure. A mathematical model based on Knudsen and Knudsen-viscous models revealed that the permeance of the asymmetric membrane depends on that of the substrate layer, due to its thickness of 1.47 mm.

The surface of quartz glass was modified by the pulsed laser deposition of carbon clusters. A carbon target was irradiated by a pulsed laser transmitted through a quartz glass substrate to excite the carbon clusters deposited on the surface of the quartz. After continuous laser irradiation for several minutes, the resulting film on the irradiated spot (2 mm in diameter) was found to be optically transparent, but to exhibit electrical conductivity. The Raman spectrum of the irradiated region contained clear G and 2D band peaks, a likely sign of thin layers of graphite structures, but it also contained a D band peak originating from defects. The surface-modified regions were analyzed in detail using transmission electron microscopy and X-ray photoelectron spectroscopy. The analyses suggested that silicon carbide (SiC) generated by the laser-induced reaction between the carbon clusters and quartz substrate may have played an important role in the formation of the transparent conductive region on the surface of the quartz glass.

In wastewater treatment plants (WWTPs), some variables such as BOD₅ and COD that are related to effluent quality, are difficult to measure directly online due to technical or economic limitations. To deal with this problem, a soft sensor that is based on a deep neural network with a named stacked autoencoder (SAE) is developed for WWTPs. Neural networks with deep structure are superior to shallow ones when facing complex problems in modern applications, which makes them suitable for wastewater treatment processes. However, deep structures are difficult to train when using traditional learning algorithms, and there are no general guidelines for identifying the proper network structure for a specific application. In the present work, a deep learning technique called the greedy layer-wise training algorithm is employed to train a deep neural network, and a genetic-algorithm strategy is developed for identifying the number of neurons in each hidden layer. In order to demonstrate its usefulness, the proposed soft sensor is tested through the test-bed Benchmark Simulation Model No. 1 (BSM1) with different weather conditions. The results indicate that the proposed soft sensor based on a deep-structure neural network can achieve better prediction and generalization performance in comparison with commonly used methodologies.

A reaction kinetics study was performed for all cis-polyaniline nanotube (ACPN) film modified electrode for fast ascorbic acid detecting. An ACPN film modified electrode exhibited excellent electrocatalytic activity, a low detection limit and a high sensitivity to ascorbic acid (AA). It is found that AA can be electro-catalytically oxidized on an ACPN film even at a concentration less than 1 mM. A mechanism of AA electrooxidation on the ACPN film modified electrode was proposed. Furthermore, the detection sensitivity was regulated by tuning the film thickness, and the optimized fabrication conditions for the ACPN film modified electrode was obtained. Based on the reaction kinetics study, the catalytic rate constant for the best electrode was found to be 6.3×10⁶ cm³·mol⁻¹·s⁻¹. It is expected that such a sensor based on the ACPN film can be widely applied for AA detection.

Life Cycle Assessment (LCA) has become more widely and internationally accepted around the world, including in Thailand. The objective of this study is to calculate and identify the amount of greenhouse gas emission in order to provide an environmental profile for the broiler industry. In this study, LCA methodology was applied to chicken feed production for 12 different formulas from 3 participating factories in Thailand. The results of the study reveal that producing one kilogram of chicken feed requires 0.12–0.36 MJ of energy consumption, both in electricity and fuel. In terms of greenhouse gases, the results show the emission of 374–473 gCO₂eq for broiler feed and 408–454 gCO₂eq for parent feed. Energy-rich and protein-rich ingredients are the main contributors to the greenhouse gas effect, accounting for 87–96%. Those used in a factory, such as electricity, water and fuel, are less significantly impactful, or 3–11% of total greenhouse gases on average for all formulas. Grain transportation is also not a prominent contributor to the impact, accounting for less than 3% of total greenhouse gases. This is because the grains are produced domestically in Thailand. Feasible options to reduce greenhouse gas emissions include the replacement of soybeans—a conventional, protein-rich ingredient and the replacement of corn—a typically energy-rich ingredient. The results show that GHGs reduces by 1.67–6.96% when 20% of the soybeans are replaced by cassava leaves and by 12.80–22.77% when replacing 50% of the corn with cassava roots.