Gibbs ensemble Monte Carlo method with TraPPE-UA force field was used to study the vapor-liquid phase equilibrium of systems containing crotonaldehyde. To lay the foundation for the vapor-liquid phase equilibrium simulation, quantum chemistry calculations with B3LYP 6-311G (d, p) and MP2 6-31G (d, p) basis sets were firstly performed to obtain the lacking bond parameters and torsional parameters for crotonaldehyde molecular. Based on these parameters, a new TraPPE-UA force field was developed. Vapor-liquid phase equilibrium data of pure crotonaldehyde and its mixture with ethanal were calculated using the new force field and compared with experiment data. The saturated liquid densities of pure crotonaldehyde predicted by the new force field were in close agreement with experimental results, with deviations from 0.28% to 1.23%. This demonstrated that the new TraPPE-UA force field manifested high simulation precision for crotonaldehyde. For the binary system consisting of crotonaldehyde and ethanal, the deviation between simulations and experiments increased with rising temperature, and the minimum and maximum values were 0.13% and 7.44%, respectively.

Micro-droplet formation in the crossing micro-channel was numerically simulated by the 3D VOF model. Three flow regimes to generate droplet with monodispersity (squeezing, dripping and jetting) and other regimes (turbulent jetting, nodular flow, tubular flow, slip flow) were obtained. The simulation results agreed well with experiment observations, which validated the numerical model. The flow regimes were mainly affected by flow rate, surface tension and viscosity of two phases. With increasing flow rate of the continuous phase, droplet size would decrease and generation frequency would increase, whereas the effect of flow rate of the disperse phase was the opposite. Surface tension and viscosity of the continuous phase played a leading role under the conditions of high and low Ca number of the continuous phase, respectively. When Ca number was low, droplet size decreased in the squeezing and dripping regimes and increased in the jetting regime with decreasing surface tension, while generation frequency shows the reverse changing tendency compared with droplet size. When Ca number of the continuous phase was high, however, droplet size decreased with increasing the viscosity of the continuous phase, while generation frequency changed inversely. In addition, contact angel had little effect on the squeezing regime, but droplet could not be generated in the dripping and jetting regimes when contact angel decreased.

Water-sparged aerocyclone (WSA) is a new gas-liquid mass transfer equipment, which takes the advantage of coupling effect of air cyclone and liquid jet fields. In order to improve its mass transfer performance and establish design principles, the effect of diameter (D_e) and length (S) of the vortex finder in the WSA on its mass transfer performance and air pressure drop in air stripping of ammonia was investigated and optimized. A new concept, mass transfer efficiency per gas pressure drop, η_p, and its calculation equation were put forward. The WSA with smaller D_e and longer S showed higher mass transfer efficiency than that with larger D_e and shorter S in air stripping of ammonia. However, this kind of WSA had higher air pressure drop, which would consume more energy. Calculation of η_p, which was a comprehensive indicator considering energy consumption and mass transfer efficiency, showed that it increased with the increase of D_e/D and S/H,but it deceased a little when S/H increased from 0.71 to 0.93. The optimal D_e/D and S/H should be 0.43 to 0.53 and 0.70, respectively. These results could be used as a guide for the design of WSA with good mass transfer performance.

The heat transfer and pressure drop characteristics of moist air in metal foam during the flow and dehumidification process were investigated experimentally, and the influence of different factors was analyzed. Heat transfer rate and pressure drop increased with increasing relative humidity as a result of increasing condensate water. When relative humidity increased from 50% to 90%, heat transfer and pressure drop respectively increased by 67% and 62%. With the increase of inlet air temperature, heat transfer rate and pressure drop increased. Heat transfer rate and pressure drop decreased with increasing cooling water temperature, pressure drop increased with increasing pore density, while heat transfer at first decreased slightly and then increased because of the influence of condensate water. Heat transfer rate and pressure drop increased with the increase of windward height of copper foam.

The flow characteristics and pressure drop of gas-coal mixtures under high pressure and concentration through a Venturi with different structures, including diameter ratio(0.44, 0.55, 0.7), convergence angle (2.5°, 5°, 9°), diffuser angle (2.5°, 8°, 13°) and throat length (23d, 43d, 80d) were investigated in the dense phase pneumatic conveying of pulverized coal system. A similar trend but a varied degree in non-dimensional pressure distribution for the gas-coal mixture through different Venturis was observed. The most remarkable influence parameter was diameter ratio, which had the greatest influence on pressure drop. The increase in pressure drop of the Venturi and remarkable increase in pressure drop in the diffuser were observed with the decrease of diameter ratio. The major role of other structural parameters was to change pressure distribution inside the Venturi, while having little influences on total pressure drop. The greatest convergence pressure drop for the convergence angle series occurred in the 2.5°. There was little difference of convergence pressure drop between 5° and 9° under high pressure and concentration. The greatest throat pressure drop for the throat length series occurred in the 80d. The least diffuser pressure drop for the diffuser angle series occurred in the 8°. By introducing solids momentum flux, an empirical equation of pressure drop of gas-coal mixture through the Venturi was established. It showed good prediction with deviations from most experimental value within 30%.

In order to explore the mechanism of droplet impact on a spherical concave surface, a numerical model using the coupled level set and volume of fluid method was developed. The mechanism of droplet impact and breakup during spreading was obtained. The impact characteristics were closely related to impact velocity. Droplet impact on a spherical concave surface showed a smaller spread factor, earlier time of central normal jet and larger jet velocity than the impact on a flat surface. The maximum spread factor and relative maximum spread velocity increased exponentially with Reynolds number. The maximum length of jet increased logarithmically with Reynolds number. The maximum jet velocity decreased exponentially with Reynolds number. Through comparison of theoretical analysis models of maximum spread factor, the development direction of analysis model of maximum spreading factor of droplet impact on a concave surface was presented.

Experimental and theoretical researches were conducted to get the two-phase flow boiling heat transfer characteristics of carbon dioxide as refrigerant in horizontal mini-channels. Based on infrared thermal imaging tests and experimental study on heat transfer coefficients, heat transfer coefficients of carbon dioxide were analyzed qualitatively and quantitatively under following experimental conditions: heat flux 2—35 kW·m^-2, saturation temperature -10—15℃, mini-channels inner diameter:1 mm, 2 mm or 3 mm. With the increase of mass flow rate, the value of vapor quality decreased gradually when dryout happened. When mass flow rate was below a critical value, heat transfer coefficient almost kept constant with the increase of mass flow rate after dryout. When mass flow rate was above the critical value, heat transfer coefficient gradually increased with the increase of mass flow rate after dryout. With the increase of tube diameter, mass flow rate decreased and the critical value of heat flux increased when dryout happened. With the decrease of tube diameter, coefficient of heat transfer increased.

Density, self-diffusion coefficients and structural properties of N-methylformamide were studied by molecular dynamics simulation over wide ranges of temperature and pressure. The simulation results agreed well with the experiment of diffusion, and the average relative deviation was 7.3%, which suggested that one can use the simulation method as a powerful tool to obtain self-diffusion coefficients over wide ranges of temperature and pressure, under which it was rather difficult for experiments. In N-methylformamide, oxygen and amino hydrogen formed strong hydrogen bond, while oxygen and carbonyl hydrogen formed weak interaction. The temperature effects were more pronounced than the pressure effects on self-diffusion coefficients and radial distribution functions.

Based on percolation theory and network model, an equation describing the relationship between filtration coefficient and pore size distribution (PSD) of filter media was established. Using this equation, PSD parameters of the filter medium could be obtained from experimental data. Moreover, the result was also compared with PSD parameters estimated by other methods and other data in the literature. In order to validate this method, the obtained PSD parameters were used in the simulation of straining-dominant filtration, and the simulation result was consistent with experimental data. The filtration model based on percolation theory for PSD parameter estimation was relatively accurate and effective.

To investigate the characteristics of pressure fluctuations in the T-abrupt exit, the dynamic pressures of gas-solid two-phase flow in the T-abrupt exit of a CFB riser were measured in a circulating fluidized bed device with FCC catalyst by dynamic pressure sensors. The pressure fluctuations of the blind tube in T-abrupt exit were the strongest and the gas-solid two-phase flow in T-abrupt exit formed a new fluctuation source which was the result of the particles cluster turning back in the blind tube. Wavelet analysis revealed that pressure fluctuations in the T-abrupt exit mainly came from the fluctuation source in the blind tube constituting the main frequency, while the energy ratio of secondary frequency increased with increasing solids mass flux. The standard deviation of the dynamic pressures in the T-abrupt exit was linear with the solids mass flux, and in the inlet of the T-abrupt exit the linear relationship was well-defined. So the real-time solids mass flux could be characterized by the standard deviation of the dynamic pressures in the inlet of the T-abrupt exit.

Hydrophobic solution was prepared by adding 2% perfluorinated octyl fluorine silane and micro-nano particles to modified silicone dilute solution. This hydrophobic solution could be solidified to obtain hydrophobic surfaces, and the contact angle of water on these surfaces could be adjusted by changing the content of micro-nano particles in hydrophobic solution. Using the method above, different micro cylinders groups with contact angles of 99.5°, 119.5° and 151.5° (de-ionized water) were prepared, respectively. Using de-ionized water as working fluid, the flow resistance and pressure drops in micro cylinders groups with different contact angles were measured. The value of friction factor of test section with hydrophobic layer was apparently lower than that without the layer at the same Reynolds number, and this was because the surface tension on hydrophobic interaction interface led to a reduction of friction resistance. The larger the contact angle, the smaller the distance between the micro/nano bumps on the double structure of hydrophobic layer at the same Re, meanwhile the ratio of liquid-air contact area to the total flow area became higher, thus flow resistance decreased. The drag reduction rates attributed to the three kinds of hydrophobic coatings on surfaces of micro pin fins decreased continuously with the increase of Re.

Both main and side reaction kinetics of non-catalytic peroxidation of liquid phase isopropyl benzene (IPB) were investigated, including byproduct formation such as α-methyl benzyl alchohol (MBA) and acetophenone (ACP), and decomposition pathway of ROOR (DCP). Based on free radical chain reaction mechanism, a kinetic model was developed, in which IPB hydroperoxide (IPBHP), MBA, ACP and DCP were considered. The prediction by this kinetic model is agreed well with the experimental data, if the rate constants were used which were obtained by fitting of the experimental data at 373—404 K. The activation energies estimated were larger for all side reactions than for the main reaction. DCP decomposition requires the presence of oxygen. These data obtained could be helpful for deep understand of the reaction mechanism of liquid phase of alkyl aromatics and could be used as the fundamental data for design and optimization of industrial liquid phase oxidation of IPB.

The non-isothermal gasification behavior of coal char with CO₂ was comparatively studied in the newly developed micro fluidized bed reaction analyzer (MFBRA) and a commercial thermogravimetric analyzer (TGA). The reaction kinetics data were calculated according to the methods of single heating rate and combination heating rate. The experimental data demonstrated that heating rate had obvious influence on coal char gasification. Increasing heating rate evidently increased reaction initiation temperature (T_x≈0), and reaction temperature corresponding to the maximum reaction rate (T_Rmax), but it gradually decreased char conversion at a given temperature and the activation energy estimated by the method of single heating rate. Comparing the results from TGA and MFBRA at the same heating rate clarified that both T_x≈0 and T_Rmax were relatively lower for MFBRA, and the higher the heating rate, the larger the difference between MFBRA and TGA. The measurement using MFBRA led to higher activation energy for char-CO₂ gasification, no matter the adopted estimation method of single heating rate (≥5℃·min^-1) or combination heating rate. The difference in kinetics data tested by TGA and MFBRA at higher heating rate (≥5℃·min^-1) could be much related to better heat transfer and lower diffusion inhibition.

2,4-Dichloro-5-fluoropyrimidine was prepared by reaction of 5-fluorouracil (5-FU) and phosphorus oxychloride in the presence of N,N-dimethylaniline (DMA) as acid acceptor. The effect of molar ratio of 5-FU to phosphorus oxychloride, molar ratio of 5-FU to N,N-dimethylaniline, reaction temperature and reaction time was investigated. Furthermore, reaction mechanism was discussed. 5-FU and POCl₃ ratio 1:10, 5-FU and DMA ratio 1:1.5, reaction temperature 114℃, reaction time 2 h were the optimum process parameters of synthesizing 2,4-dichloro-5-fluoropyrimidine. Under this process condition, maximum yield of 92.2% of 2,4-dichloro-5-fluoropyrimidine could be obtained. The pH of waste water was then adjusted by adding sodium hydroxide, and the effect of pH on recovery of DMA and yield of by-product disodium hydrogen phosphate was studied. When pH was adjusted to 7, maximum recovery of DMA and maximum yield of by-product disodium hydrogen phosphate could be obtained.

The disproportionation reaction between (CH₃)₃SiCl and SiCl₄ catalyzed by AlCl₃ involves three reactions; the reaction between (CH₃)₃SiCl and SiCl₄ to form (CH₃)₂SiCl₂ and CH₃SiCl₃ (R1), the reaction between (CH₃)₂SiCl₂ and SiCl₄ to form CH₃SiCl₃(R2) and the reaction between (CH₃)₃SiCl and CH₃SiCl₃ to form (CH₃)₂SiCl₂(R3). Density functional theory was used to study these reactions. The possible catalytic mechanisms of AlCl₃ were stepwise reaction pathway of forming intermediate CH₃AlCl₂ and co-adsorption pathway of two reactants on AlCl₃. The co-adsorption mechanism was more likely for these three reactions. The energy barrier ranking of these reactions was R2 >R1 >R3. By analyzing the thermodynamics data of these reactions, the effect of temperature on the Gibbs reaction energy of R2 was almost negligible within a temperature range of 543.15 K to 593.15 K. However, the Gibbs reaction energy of R1 and R3 significantly decreased at 593.15 K. These studies indicated that through appropriate temperature regulation, the reactions of R1 and R3 could be the dominant reactions while R2 would be effectively suppressed, influencing product distribution and facilitating formation of (CH₃)₂SiCl₂.

Gold was supported on a functionalized MOF-5 by impregnation method. The catalyst Au/MOF-5 was characterized by powder X-ray diffraction (XRD), N₂ adsorption-desorption, infrared spectra (IR), thermogravimetric analysis and differential thermal analyses (TG-DTA), transmission electro microscopy (TEM), and inductivity coupled plasma (ICP). Its catalytic performance was examined in one-pot synthesis of structurally divergent propargylamines via three component coupling of aldehyde, alkyne, and amine (A³) in 1,4-dioxane. The results showed that the Au/MOF-5 catalyst displayed high activity (high aldehyde consumption rate and turn-over number at 120℃: 18.7 mmol·(g Au)^-1·h^-1 and 14.7, based on total gold content) and 100% selectivity for propargylamines. Various aromatic and aliphatic aldehydes were also able to undergo the A³ coupling on the catalyst Au/MOF-5. It was also found that the conversion of benzaldehyde was higher on the catalyst for aryl aldehydes with electron-withdrawing group bonded on benzene ring than for those with an electron-donating group. There was good conversion of benzaldehyde for secondary amines and N-alkyl aniline such as piperidine, pyrrolidine, morpholine, diethylamine, and N-methylaniline. It was also found that the conversion of benzaldehyde decreased with increasing chain length of straight-chain alkyl substituted phenylacetylene. Au/MOF-5 catalyst could be recycled at least three times, and the activity only decreased slightly.

It has been accepted that separation of low boiling-point gas mixture via hydrate technology has the advantage of operating at around 0℃, and therefore can reduce the refrigerating energy consumption dramatically. It is necessary to study the simulation algorithm for the operation process because this separation technique is still at the stage of theoretical research and has not fulfilled practical industrial application. In this work, a process of hydration and absorption two-tower series operation was simulated by using the non-equilibrium stage approach, which involved hydrate formation kinetic model, hydrate dissociation kinetic model, and mass transfer kinetic model when gas was absorbed in water-in-oil emulsion. The effect of operation parameters of two series towers, such as temperature, pressure on separation efficiency was investigated. The calculation results are of significance for designing separation process for low boiling-point gas mixture via hydrate technology, and can also lay a theoretical foundation for industrial application of this separation technique.

Carbon dioxide absorption into a mixed aqueous solution of functional ionic liquid, tetramethyl ammonium glycinate ([N₁₁₁₁][Gly]) and monoethanolamine (MEA) was performed in a plate falling film reactor with different flow patterns. The effect of countercurrent gas flow on flow pattern transition was investigated. Absorption of carbon dioxide into the mixed aqueous solution of MEA and [N₁₁₁₁][Gly] was studied with different liquid flow rates, liquid temperatures, gas flow rates and inlet carbon dioxide concentrations. Falling liquid film was found to take the patterns of corner rivulet flow, falling film flow with dry patches, and complete falling film flow depending on gas and liquid flow rates. Reduction of gas flow rate could lower critical flow rate of pattern transition. Absorption rate increased with increasing liquid flow rate with flow patterns of corner rivulet flow and falling film flow with dry patches, and then became unchanged with the pattern of complete falling film flow. CO₂ absorption rate reached a maximum with the pattern of complete falling film flow, however, corner rivulet flow had the largest liquid phase mass transfer coefficient.

Highly concentrated KCl and MgCl₂ solutions were treated with direct contact membrane distillation (DCMD). The effects of salt concentration (1.0—4.0 mol·L^-1) and circulation velocity (0.1—0.5 m·s^-1) on permeate flux were investigated. Permeate flux decreased with increasing concentration, which was attributed to vapor pressure reduction caused by decrease of water activity in the solution. Reversed flux occurred in the DCMD process of highly concentrated MgCl₂ solution (>3.0 mol·L^-1) under the test conditions, which was due to the combined effect of lower water activity and higher viscosity of the fluid. Permeate fluxes increased with increasing circulation velocity, which was more effective to reduce the negative effect of temperature and concentration polarization for MgCl₂ solution with higher viscosity. Membrane distillation-crystallization (MDC) of a highly concentrated solution of KCl-MgCl₂-H₂O was investigated for recovery of KCl crystalline product and pure water with batch operation mode. The effect of way of cooling on supersaturation of solution, morphology, and crystal size distribution was studied. KCl crystalline product with regular shape, tidy surface and concentrated particle size distribution was obtained by natural cooling and maximum degree of supercooling generated under natural cooling was lower than under quick cooling, which was beneficial to control nucleation amount and growth rate of KCl.

Monoethanolamine (MEA) aqueous solution has remarkable absorption capability for acid gas such as CO₂ and has a potential application prospect in the field of gas separation and purification. Absorption temperature, gas flow rate and MEA concentration were taken into account as influence factors on reaction rate by monitoring the variation of pH in MEA aqueous solution. A conception called specific activity describing the absorption performance of the solution was introduced. The descending rate of pH decreased, saturation time was extended and the pH of final saturation state increased with the increase of solution concentration. With increasing temperature, the descending rate of pH increased and saturation time was shortened, but both of them would maintain a relatively stable value when temperature was high enough. When gas flow rate of CO₂ increased, the pH of solution was gradually reducing at a faster rate, and finally, remained stable. The pH of saturated solution reached the lowest level at the flow rate of 70 ml·min^-1. The monitoring method was easy to operate and the result was highly accurate so that it could provide a reference for development of CO₂ absorption solutions.

A novel amino acid functionalized polyamine carboxymethyl chitosan (CTGCS) was designed to improve adsorption ability of chitosan (CS) for heavy metals. Chitosan resin (GCS) was firstly obtained via cross-link reaction of chitosan and glutaraldehyde. After following modification by tetraethylenepentamine and chloroacetic acid, CTGCS was successfully synthesized. The samples at different reaction steps were characterized by using FTIR and XPS to understand reaction mechanism. Moreover, the adsorption ability of as-synthesized CTGCS for Ni²⁺ from aqueous solutions was investigated. The adsorption process conformed with the Freundlich isotherm model and could be fitted by the pseudo-second order kinetics model. The maximum adsorption capacity of CTGCS for Ni²⁺ was significantly improved up to 533.8 mg·g^-1, about 1.5 times more than that of TGCS, indicating that CTGCS was a promising adsorbent for heavy metals.

The hydraulic characteristics and mass transfer efficiency of a new kind of oriented grid packing extractor were investigated by experimental study. The throughput of the new extractor increased greatly than the traditional packed extractor. When velocity ratio L_R (dispersed phase:continuous phase) equaled 1:5, flooding throughout reached 150 m³·m^-2·h^-1. The new packed extractor also exhibited good extraction efficiency, and thus could have excellent application prospects in many areas as solvent deasphalting in petroleum refining process, phenol removal from wastewater in coal chemical industry, etc.

In this paper, a novel fault detection method based on sparse non-negative matrix factorization (SNMF) is proposed. NMF (non-negative matrix factorization) is a new dimension reduction technique that can find a low-rank matrix approximation from the original data. In contrast to the conventional multivariate statistical process monitoring methods, for example PCA, NMF has no assumption about the nature of latent variables, except for non-negativity. Combining linear sparse coding and NMF, SNMF can learn much sparser representation via imposing sparseness constraints. During factorization, low-rank matrix is orthogonalized to remove redundant information and concentrate information on fewer directions of projection. Then, SNMF is used to extract the latent variables that drive a process and new statistical metrics are defined for fault detection. Kernel density estimation (KDE) is adopted to calculate the confidence limits of defined statistical metrics. Afterwards, the proposed method is applied to the Tennessee Eastman process to evaluate the monitoring performance, comparing with conventional NMF and PCA. The results from the experiment show the feasibility of the new method.

In order to achieve automatic diagnosis of burning-plate fault on the cathode bar and real-time monitoring of current distribution in the process of copper electrolysis, based on the principle of infrared imaging, an infrared camera was used to shoot the image of copper electrolytic tank, and the gray value on the cathode bar in the processed image was extracted. A mathematical model between gray value and cathode bar temperature was developed on the basis of field tests, and the surface temperature of the cathode bar could be calculated. A partial least squares (PLS) method was used to develop a mathematical model between surface temperature and current value of the cathode bar. The square of current value after integration was found to be in quasi-linear relationship with surface temperature of the cathode bar, the coordinates of cathode bar and ambient temperature. Contrast between the current on the cathode bar derived from the model and the current measurement on the field showed that the current on the cathode bar could be accurately determined with this method, which realized real-time monitoring of current value in the copper electrolysis process. In addition, the cathode bar with burning-plate fault could be accurately and automatically diagnosed, and this fault could be predicted by monitoring the current value on the cathode bar, achieving automatic diagnosis of the fault in the copper electrolysis process, thereby reducing burning-plate fault on the cathode bar and resulting in good economic benefits.

In the refrigeration and air conditioning system, sensors are independent component for physical data measuring and operating state monitoring. Sensor faults, especially sensor biases output will lead to incorrect measurement, inappropriate controlling strategy and further energy consumption rise. Based on the pattern recognition theory, the fault detection task could be considered as a one-class classification problem. Therefore, a powerful pattern recognition-based one-class classification algorithm, Support Vector Data Description (SVDD) was used to detect the sensor biases occurring in a chiller system. The practical fault-free data were used as training dataset to develop the SVDD model so as to detect the sensor faults. The method and its fault detection efficiency were validated by test dataset with different artificially introduced levels of sensor biases. The SVDD-based fault detection method worked well with chiller practical operating measurements, but the fault detection efficiencies of different sensors with different level faults were inconsistent.

Flow maldistribution in the inlet of the plate-fin heat exchanger along the longitudinal direction affects heat transfer among different layers. Nowadays experimental methods are used to determine the parameters of the leading flow structure, however, optimization of these parameters has not been reported yet. A new multi-objective optimizing method was proposed based on traditional leading flow structure. Considering the factors influencing fluid distribution in the inlet of the plate-fin heat exchanger, including diameter, space and number of hole on the leading flow structure, fluid velocity, location of the leading flow structure in the inlet of the exchanger and inlet diameter of the exchanger, the multi-objective optimizing models of pressure drop and heat transfer per unit time were established. Orthogonal experiment was conducted to create the simulation model and deduce non-linear mapping equations of the leading flow structure. The leading flow structure was optimized using back propagation network (BP) combined with genetic algorithm and the final parameters of the structur were determined. Fluid flow and distribution in the inlet of the plate-fin heat exchanger before and after optimization were simulated using the commercial software, Fluent, and were compared with those without the leading flow structure. The leading flow structure after optimization could improve fluid distribution along the longitudinal direction obviously, and meanwhile, improve heat transfer and enhance performance of the exchanger.

For industrial processes, there are not only correlations among variables, but also autocorrelation in temporal structure of these variables, therfore, a new sensor fault detection and diagnosis method based on min/max autocorrelation factors (MAF) was proposed in this work. Firstly, MAF analysis of historical normal data was made. Then, strong autocorrelation factors and weak autocorrelation factors were obtained. Based on these factors, the statistics for fault detection were constructed and corresponding contribution plots were derived. The proposed method was applied to the continuous stirred tank reactor (CSTR) and compared with the principal component analysis method. Simulation results demonstrated that the proposed method could detect sensor faults with slow variation more quickly with less false-alarm. The contribution plots based on MAF can explain complicated sensor fault more reasonably than principal component analysis (PCA), which is a classical multivariate statistical method for process monitoring.

For the multi-contaminants batch water network optimization problem, a mixer-integer nonlinear programming model with the objective of minimum fresh water usage was established. Meanwhile, a solution strategy combined with matrix encoding genetic algorithm and trust region sequential quadratic programming(TRSQP)was proposed. In this paper, the matrix encoding method is easy to establish scheduling module, so the water reused relationship between the water units is more clearly defined.The integer variables was optimized by matrix encoding genetic algorithm and the continuous variables was optimized by TRSQP, the optimal strategy integrated the two methods to achieve the solution of MINLP model with the MATLAB software. The proposed method was used to solve two typical examples and the results obtained were superior to those in literature, which demonstrated that the approach introduced by this paper was effective.

The corrosion and friction behavior of four kinds of ceramic grinding balls (zirconia, alumina, agate, silicon nitride) was investigated in a milling-leaching combined process for the acidolysis of Panzhihua ilmenite in sulfuric acid. The order of corrosion rates of the balls were: silicon nitride>alumina>zirconia>agate balls. Corrosion of zirconia mainly conformed to the mechanism of friction, while collision and surface cutting were responsible for corrosion of agate balls. Corrosion of alumina ball was due to abrasive wear, and presence of ilmenite significantly increased weight loss. Silicon nitride balls experienced obvious surface spalling abrasive wear. The weight loss rates of the four ceramic grinding balls were over 80% caused by physical abrasion, and the aluminum and silicon nitride balls suffered an obvious combined effect of corrosion and friction. The weight loss rates of alumina and silicon nitride ball were 1.76 and 14.52 mg·cm^-2·d^-1, respectively, which were much higher than those of zirconia and agate balls. The results provided good reference to chasing the grinding ball materials in the milling-leaching process for acidolysis of ilmenite.

A natural organic oil spill adsorption material was developed by steam explosion and esterification modification of rice straw. The oil-absorption capacity and water-absorption capacity of the raw rice straw were 5.66 g·g^-1 and 7.59g·g^-1, respectively. The conditions of esterification modification of rice straw were optimized. The optimal reaction temperature, catalyst concentration and reaction time were 80℃, 0.75%, 6 h, respectively. The oil-absorption capacity and water-absorption capacity of the modified rice straw were 8.54 g·g^-1 and 0.10 g·g^-1, respectively. Furthermore, the rice straw before and after modification were characterized by BET, XRD, FT-IR. The modified rice straw had larger pore diameter after steam explosion, appearance of non-crystallizing state and acetyl group of characteristic peak after esterification, which proved the feasibility of esterification modification of rice straw.

Five different oil shale kerogens were analyzed with advanced testing techniques (XRD, ¹³C NMR and FT-IR) to obtain important structural information. The long-range disordered state of aggregation was also confirmed by XRD. Using ¹³C NMR spectroscopy, 18 characteristic parameters of carbon skeleton structure were obtained. Detailed information about the chemical structure of kerogen was obtained through verifying the results of ¹³C NMR experiments with the qualitative analysis of kerogen groups made by FT-IR technique. The aliphatic structure of oil shale kerogen was up to 60%—90%, with the methylene carbon occupying the dominant position of 77%—90%, and most of the aliphatic structure were straight-chain while less alicyclic or branched chain. In addition to Huadian oil shale with a monocyclic aromatic carbon structure, the other four oil shale kerogens were co-existing monocyclic and polycyclic rings, with the latter mainly peri-condensed. The oxygen structure was C-O (alcohol, phenol, ether) and C=O, and carboxyl class was more than carbonyl class C=O. More C=O in kerogen was due to a low degree of metamorphism, while a high degree of metamorphism resulted in more C-O.

In order to analyze the growth mechanism of product layer in SO₂ and NO_x removal by calcium-based sorbent, the single-crystal Ca(OH)₂(0001) surface was used to simulate the pore wall and external surface of the sorbent. The growth of product layer was analyzed under different reaction environments. The growth mechanism of product layer in SO₂ and NO_x simultaneous removal by calcium-based sorbent was proposed, combining crystal growth theory and discrepancy between SO₂ and NO_x single removal. SO₂ removal reaction firstly occurred in the local area of the single-crystal surface to result in expansion of surface and formation of convex structure. Subsequently, the reaction and expansion area gradually extended, which led to formation of the continuous dense product layer. Due to solubility and diffusion of NO₂ removal product in the water film, NO₂ reaction had an obvious etching effect on the sorbent surface. When SO₂ removal reaction coexisted with NO₂ removal reaction, the etching action of NO₂ removal reaction could inhibit formation of the continuous dense SO₂ removal product layer. Meanwhile, the presence of NO₂ removal product could change the growth habit of the SO₂ removal product. Therefore, the sorbent surface structure after simultaneously reacting with SO₂ and NO_x was loose.

The pyrolysis study of the ionic liquid (IL)-H₂O mixtures pretreated bamboo (Phyllostachys edulis) was investigated. 1-Butyl-3-methylimidazolium chloride ([Bmim]Cl)-H₂O, 1-butyl-3-methylimidazolium tetrafluoroborate ([Bmim]BF₄)-H₂O and 1-butyl-3-methyl pyrrolidinium chloride ([Bmpy]Cl)-H₂O with or without solid acid catalyst (Nafion^®NR50) were used in the pretreatment process. The composition analysis result of untreated bamboo and pretreated bamboo were compared. XRD and SEM result were analyzed to explain the effect of ILs mixture on cellulose crystalline structures and surface structures. According to the result of TG-DTA and GC-TCD, the sample pretreated with [Bmim]Cl-H₂O mixture was used in the pyrolysis process to produce more gaseous products and low char. About 0.12 g gaseous products and 0.19 g char were produced from 1.0 g sample pretreated with [Bmim]Cl-H₂O mixture and added solid acid catalyst, different from the result of untreated sample, 0.04 g gaseous products and 0.24 g char. The potential and problems of IL-H₂O mixture pretreatment used in thermal chemical conversion methods were discussed.

Energy saving and emission reduction technologies for vehicles have been paid more attention in recent years, and it is necessary to investigate how to efficiently recover and utilize the waste heat from vehicle engines to drive automotive air conditioning system. Based on that, a full air-cooled bubble absorption refrigeration test system with 3 kW cooling capacity driven by waste heat, using R124-DMAC as working fluids was designed and built. In the system, engine exhaust gas was simulated by heated air. Operating parameters were tested through changing heat source temperature, chilled water temperature and pump flow rate. The maximum COP of the system could reach 0.54 under the condition of generator temperature 100℃ and evaporating temperature -4℃. Heat source and chilled water temperatures had an important effect on cooling capacity and COP, and this system had good stability. However when evaporating temperature was below 5℃, a rectification unit should be considered in order to improve cooling effects. Additionally, this experiment proved that the air-cooled bubble absorber worked well, providing theoretical support for the refrigeration system driven by waste heat in vehicle air conditioning.

Two kinds of important equipment i.e. selective catalytic reduction system (SCR) and flue gas cooler are applied in coal-fired power plant to achieve energy conservation and to reduce emission. However, few studies have been published on the influence of ammonia escaped from SCR system on flue gas cooler. In this study, the condensates deposited on the wall of the cooler at the downstream of SCR were analyzed by XRD, XRF, SEM and EDS. The results indicated that the white crystalline material in the condensates is mainly ammonium fluoroborate and some intermediates such as ammonium fluoride, boric acid and ammonium fluorosilicate. The formation of crystalline layer took place by coupling reaction between the ammonia escaped from SCR system and both fluorine element and boron element volatilized in combustion of coal. Although the deposition layer has only a limited thickness, it could be hamper directly for operation of flue gas cooler.

The influence of hydraulic retention time (HRT, 24, 18, 15 and 12 h) on the performance of double circle (DC) anaerobic reactor for treating traditional Chinese medicine wastewater was investigated. Additionally, the characteristics of granular sludge were studied, including particle size distribution (PSD), extracellular polymeric substances (EPS), and distribution of microbial communities. When HRT was reduced from 24 h to 12 h, the DC anaerobic reactor demonstrated good COD removal of 90%. Although concentration of acetate increased to 339.69 mg·L^-1, the process was operating favorably without acidification. But humic acid-like and fulvic acid-like absorption peaks were stronger and coenzyme F₄₂₀ absorption was weaker in the EEM spectra of soluble microbial products (SMP) at HRT 12 h. It was not recommended to further reduce HRT. The contents of EPS, protein, and polysaccharide decreased in the sludge at HRT 12 h, and tyrosine played an important role in maintaining stability of the sludge. The phospholipid fatty acids analysis showed that reducing HRT affected significantly distribution of microbial community in the second reaction area of the DC anaerobic reactor. Gram-positive bacteria decreased from 44.24% to 32.69%, while gram-negative bacteria increased from 32.69% to 38.66%.

In order to evaporate and concentrate the thermo-sensitive materials with properties of high viscosity, easy scaling and strong corrosion, a jacketed MVR (Mechanical Vapor Recompression) heat pump evaporation concentration system was designed. Water was used as the test medium for a series of experimental studies. Such parameters as temperature difference, compression ratio, evaporation capacity, coefficient of heat transfer, COP (Coefficient of Performance), SMER (Specific Moisture Extraction Rate), adiabatic efficiency and volumetric efficiency were influenced by operational and structural parameters like evaporating pressure, frequency of compressor and heat transfer area. The performance of the system when the compressor was running at afrequency of 50 Hz was much better than that under other conditions at a lower frequency. The most appropriate range of evaporating pressure was from 65 kPa to 85 kPa when a single evaporating kettle was running, and it was unnecessary to supplement external steam. Two evaporating kettles could be simultaneously used for industrial evaporation concentration by supplementing only a small quantity of external steam. Energy saving of both cases mentioned above was significant.

Zero emission coal (ZEC) process based on calcium looping in which CO₂ is captured through a cyclic carbonation/calcination process is a promising technology for hydrogen production. In this process, H₂S should be removed before raw gas flows into the downstream carbonator/reformer. In this work, the cycled CaO derived from limestone after multiple carbonation/calcination cycles for CO₂ capture was used to remove H₂S from raw gas. Cyclic carbonation/calcination of CaO was performed in a dual fixed-bed reactor and then the cycled CaO was sent to a sulfidation reactor for H₂S removal. The effects of carbonation/calcination cycle number, sulfidation temperature, H₂S concentration and CaO microstructure on sulfidation performance of the cycled CaO from CO₂ capture cycles were investigated. The cycled CaO after the long-term carbonation/calcination cycles for CO₂ capture still had high H₂S removal capacity. Sulfidation conversion of CaO derived from limestone decreased rapidly with increasing number of carbonation/calcination cycles in the first 20 cycles, and then decreased slowly with further increasing cycle number. After sulfidation for 120 min, sulfidation conversion of the CaO after 0 cycle was nearly 100%. And sulfidation conversions of the cycled CaO after 1, 20 and 100 cycles were 94%, 81% and 74%, respectively. H₂S concentration showed a great effect on H₂S removal of the cycled CaO. Sulfidation temperature in the range of 800—1000℃ had a little effect on H₂S removal of the cycled CaO, and the optimum sulfidation temperature was 900℃. As the cycle number increased, specific surface area of the cycled CaO dropped due to sintering. The pores 20—150 nm in diameter of the cycled CaO which were closely related with H₂S adsorption also decreased. Thus, H₂S removal capacity of the cycled CaO decayed with the number of cycles.

Salinity gradient power is a kind of chemical potential energy which exists between sea water and fresh water or water with different salt concentrations. Reverse electrodialysis (RED) is a process that transfers the salinity gradient to energy. Compared with other process, RED shows higher energy density. However, the studies on RED are limited to small membrane stacks. Experiments with a larger setup show that power density decreases significantly. Moreover, there is little research on the energy efficiency analysis of the RED process. In this paper, the energy transfer process of RED was simulated mathematically and the structure of the membrane stack was optimized. The power density and energy efficiency of RED at different structural parameters were also analyzed. Decreasing the thickness of the compartment could enhance both power density and energy efficiency greatly. Increasing flow length could result in higher energy efficiency, but average power density would decrease.

Combining anaerobic-anoxic-oxic reactor with an external nitrification biological aerated filter, the AAO-BAF process was used to treat domestic wastewater. Adjusting the influent COD concentration from 211 to 675 mg·L^-1, the effect of carbon-nitrogen ratio (C/N) on the performance of the AAO-BAF process was investigated. A low or high influent organic concentration had significant effect on the nutrients removal efficiency due to the deficient phosphorus release or nitrate competition in the anoxic zones of the AAO reactor between DPAOs and ordinary heterotrophic organisms. When C/N was between 4 and 7,the average removal efficiencies of COD, TN and PO₄^3- could reach more than 86%, 78% and 90% respectively. Large quantity of residual volatile fatty acid (VFA)would be transferred into the anoxic zones when C/N was very high (such as 9.5), which would cause rapid consumption of NO₃^- by ordinary heterotrophic organisms rather than denitrifying phosphorus accumulating organisms (DPAOs).

Micro injection molding experiments were conducted to investigate the influence of shear history on morphology, microstructure and mechanical properties of the micro-scale parts by changing injection speed based on the iPP parts with thickness of 0.2, 0.5 and 1 mm. Shear rate was analyzed with software Moldflow. Skin-core structure appeared in the thickness direction in both 0.5 mm and 1 mm parts, and imperfect shish-kebab structure appeared in the transition layer between skin layer and core layer of 0.5 mm and 1 mm parts, however, column crystal was shown in the transition layer of 0.2 mm parts. The whole shear level in shear history increased with increasing injection speed for all the parts with three thicknesses. The ratio of skin layer of 0.5 mm and 1 mm parts decreased with increasing injection speed, and yield stress, elongation at break, breaking strength and modulus decreased as the ratio of skin layer decreased. For 0.2 mm parts, the ratio of skin layer increased with increasing injection speed, and yield stress, breaking strength, modulus increased with increasing ratio of skin layer, while elongation at break decreased.

A novel Fe-Sr₂Bi₂O₅ photocatalyst was synthesized with the co-precipitation method and added into a low density polyethylene (LDPE) film. The solid-phase photocatalytic degradation behavior of LDPE composite film under UV and visible light irradiation at room temperature was investigated and compared with the pure LDPE film, CaCO₃/PE,Sr₂Bi₂O₅/PE and TiO₂/PE composite films,and the obtained sample were characterized with contact angle, Fourier transform infrared spectroscopy (FT-IR) and field emission scanning electron microscopy (FESEM), including the analyses for the mechanical properties of reserved elongation at break and reserved tensile strength. Fe-Sr₂Bi₂O₅ had a high photocatalytic activity in degrading LDPE under UV light and visible light irradiation. After UV light irradiation for 10 d, reserved elongation at break and reserved tensile strength were lower than 5% and 40%, respectively; while contact angle was reduced by 22.52°. At the same time, carbonyl groups were detected in LDPE composite films and there were holes on the surface of LDPE around Fe-Sr₂Bi₂O₅. Fe-Sr₂Bi₂O₅ made photodegradation of LDPE more effective compared with Sr₂Bi₂O₅ and TiO₂ under UV or visible light degradation. In conclusion, LDPE could be degraded obviously by Fe-Sr₂Bi₂O₅ under UV and visible light irradiation at room temperature, and Fe-Sr₂Bi₂O₅ could be a new photocatalyst for photocatalytic degradation of polyethylene film in the future.

Comprehensive process mineralogy analysis of waste residue from incomplete reaction of titanium concentrate and sulfuric acid in sulfate process of titanium dioxide industry was made with Mastersizer 2000, XRD, XRF, ICP-AES, SEM-EDS, specific gravity method and screening, to provide guidance to recycling ilmenite. The residue had wet surface with about 45%(mass) soluble substance, including 4.06%(mass)Ti ion which could be hydrolyzed easily. The insoluble particles with various appearances contained 17.14%(mass) TiO₂, 10.78% (mass) Fe₂O₃, and 46.94%(mass) SiO₂. Its particle size distribution was between 1 mm and 100 mm. Its density was 3.21 g·cm^-3. From screening, there were different particle sizes and densities between ilmenite and other impurities. Particles above 120 mm with average density about 3.4 g·cm^-3 were mainly anhydrite. Particles between 18 mm and 75 μm with average density greater than 3.5 g·cm^-3 were mainly ilmenite. Particles below 18 mm with average density about 2.9 g·cm^-3 mainly contained silicon sludge, and became viscous and difficult to separate in water.

The molecular structure of a monohydroxy modified PBO and its AB monomer were proposed, and 2-hydroxy-4-(5-amino-6-hydroxybenzoxazole-2-yl) benzoic acid(HABA)was synthesized by the condensation-cyclization reaction of 4-amino-6-nitroresorcinol hydrochloride(ANR·HCl)and hydroxy terephthalic acid(HTA), then 2-hydroxy-4-(5-nitro-6-hydroxybenzoxazole-2-yl) benzoic acid(HNBA)was reduced to obtain the AB monomer (HABA). The precursor HNBA was one-pot prepared from ANR·HCl and HTA with a yield of 69.25% and purity of 96.81%, and the AB monomer HABA with a purity >98% was obtained by reducing HNBA, with a total yield of 58.38% from ANR·HCl. The novel synthesis technology had the characteristics of convenient source of raw materials,short reaction steps,mild conditions,excellent stability and equi-molar polycondensing group. HABA and HNBA were identified with FT-IR, MS and NMR.

Three kinds of polypropylene (PP) with different viscosities were blended with poly(vinyl alcohol) (EVOH) to prepare 90/10 PP/EVOH blends, and the blends were then molded into hollow parts via water-assisted injection molding (WAIM). Samples were cut from locations near the gate (marked as 1^#) and the end of flow channel (marked as 2^#) of the WAIM blend parts, and the phase morphologies at three regions across the residual wall of the samples were observed with scanning electron microscope. The toluene permeabilities of the samples cut from the WAIM blend parts were measured. The phase morphologies at the three regions in the samples were interpreted with the aid of flow fields within the mold cavity under high-pressure water penetration during WAIM. Coarse fibers appeared in both outer and inner layers and droplets in the core layer for the sample taken from location 2^# of the WAIM blend with high viscosity ratio, which resulted in modestly improved barrier performance of the sample compared with the WAIM corresponding PP sample (about 2.4 times). Slim fibers existed in the outer and inner layers and coarse and long fibers in the core layer for the samples taken from location 2^# of the WAIM blends with low viscosity ratios, which resulted in greatly improved barrier performance of the samples compared with the WAIM correponding PP sample (about 9.8 times for the blend with the lowest viscosity ratio). Average diameters of dispersed fibers for the samples taken from location 1^# of three WAIM parts were larger than those taken from location 2^#, which resulted in lower barrier performance for the former samples.

In order to improve production efficiency, simplify procedure, and produce polyacrylonitrile (PAN) spinning dope, which is stable and has high viscosity-average molecular weight (M_η), a new PAN dissolution method with twin-screw extruder was developed. A mode of T-combination was used to frame the second order twin-screw and a filter was installed in the outlet port. A mixed solvent suitable for screw extruder based on PAN dissolution theory and dissolution ability of solvents was selected. The dissolved PAN was used in dry jet wet spinning and high strength fiber with low linear density was made. The twin screw method of dissolving PAN satisfied high performance fiber production requirement, and the solution quality was stable and more productive.

A novel reactive flame retardant named hexakis (4-diethyl phosphate hydroxymethyl phenoxy) cyclotriphosphazene (HPHPCP) was synthesized from hexachlorocyclotriphosphazene (HCCP), p-hydroxy benzaldehyde and diethyl phosphite. Chemical structures of intermediate and HPHPCP were characterized by Fourier-transform infrared spectroscopy (FTIR), proton and 31P nuclear magnetic resonance (NMR). Thermal stability of the HPHPCP was determined via thermogravimetric analysis (TGA). Halogen-free flame-retardant rigid polyurethane foams (RPU) were prepared by using HPHPCP as a reactive flame retardant. The effect of HPHPCP content on the thermal and flame-retardant properties of RPU was investigated with TGA, and limiting oxygen index (LOI). The temperatures of 5%(mass) loss (T_5%) of HPHPCP in nitrogen and air were 162.7℃ and 153.0℃, and the residues at 800℃ of HPHPCP in nitrogen and air were 42.1% and 12.0%(mass), respectively. HPHPCP could enhance thermal stability of RPU. The T_5% of RPU-20%HPHPCP was 193.9℃, 20℃ higher than pure RPU. Moreover, limiting oxygen index (LOI) of RPU increased with increasing HPHPCP, and LOI of RPU-30%HPHPCP was 27%.

The effect of soluble phosphorus on CaSO₄·2H₂O (DH) dehydration to α-calcium sulfate hemihydrate gypsum (α-HH) in hydrothermal CaCl₂ solutions and the crystal shape regulation were investigated. The action mechanism of soluble phosphorus was studied by means of growth of hemihydrate crystal. The system pH was the main factor to determine whether DH could be dehydrated to α-HH in 24% CaCl₂ solution at 98℃. Acicular a-HH crystal could be obtained when pH was less than 2.0. High-strength α-HH crystal with aspect ratios from 0.5 to 3.0 could be observed when polycarboxylic crystal habit modifier NS was added. The corresponding compressive strength was more than 20 MPa. The presence of soluble phosphorus was favorable for α-HH formation by lowering system pH, increasing supersaturation degree and reducing interfacial energy. On the other hand, acid radical ions affected α-HH crystallization behavior and adsorption results of medium crystal agent NS.

In order to enhance electrical conductivity and rate capability of Li₄Ti₅O₁₂, the Li₄Ti₅O₁₂/C powder was prepared by the sol-gel method, using PEG as carbon source. The performance of Li₄Ti₅O₁₂/C coated by three different molecular weight carbon sources: PEG (400, 600, and 1000), was investigated with TG, XRD, SEM, TEM, constant current charge-discharge, rate discharge and EIS. The material obtained by using PEG1000 as carbon source had the best electrochemical performance. Initial discharge capacity was 143.5 mA·h·g^-1 at 0.1C rate, and discharge capacity remained 105 mA·h·g^-1 at 2C rate, capacity retention ratio was 73.17%, in addition, it had the smallest resistance.

Monoammonium phosphate (MAP) was mixed with ammonium nitrate (AN) by the mechanical mixing method (1) and the solution mixing method (2). The UN gap test and Koenen test of UN-Recommendations on the transport of dangerous goods, were used to determine the propagation of detonation and the effect of heating under confinement, and DSC was used to study the thermal decomposition characteristic. The results of UN gap test showed that samples with different mixing methods had the same result. When the MAP content reached 25%, the result was negative, i.e., the modified AN did not propagate detonation. In the Koenen test, when the MAP content in the modified AN-1 and modified AN-2 reached 30% and 25% respectively, the samples were not sensitive to intense heat under test confinement. The results of DSC showed that the T_onset of modified AN-1 was lower than AN, but the T_onset of modified AN-2 was higher than AN. The mechanical mixing method could improve the passivation effect of MAP on AN.