Recently，ceramic membranes have been widely used in process engineering and enjoyed rapid development on account of its inherently superior physical integrity，chemical resistance，and separation performance. Based on the theory and method of application-oriented ceramic membranes design，the recent developments in the preparation of ceramic membranes with quantitative control，the selection and design of ceramic membranes used for different application processes，and the fouling control of ceramic membranes during industrial processes were reviewed. The solutions for the problems and the preferential projects for future study were proposed.

Mixed refrigerant cycles can improve refrigeration efficiency, and are used widely in natural gas liquefaction systems. The effect of mixed refrigerant cycle (MRC) stages on refrigeration performance is significant. Thermodynamics models of the main pieces of equipments in the MRC were built. Natural gas liquefaction processes with MRC were simulated. Most of the process parameters, including shaft power of compressors, refrigeration efficiency and exergy efficiency of different numbers of stages of MRC were obtained. With increasing number of stages of MRC, the shaft power of compressors decreased, while the refrigerant efficiency and exergy efficiency increased. But the rate of decrease and increase declined. Multi-stage cycles made the process configuration complex and operation difficult. Different sizes of refrigeration systems had different optimum number of stages. The larger the size of refrigeration system, the greater the number of stages. A balance must be made between energy efficiency and complexity.

A comparative experiment prototype of an air-source heat pump water heater（ASHPWH）system was established for the comparative study of instantaneous heating and circulation heating modes. At ambient temperature of (19±0.5)℃，176 kg of water was heated from 20℃ to 55℃ with the two modes respectively. The average COP of the instantaneous heating mode is 24% higher than that of the circulation heating mode，and the average condensation heating power is also increased by 20%. The instantaneous heating mode not only has a higher COP，but also has a higher condensation heating power，resulting in a lower power consumption and a shorter heating time. At the same time，the condensation pressure，the compression ratio，the maximum outlet temperature of the compressor and other important parameters with the instantaneous heating mode are superior to those with the circulation heating mode. The present study shows that ASHPWH system has superior thermal performance in the instantaneous heating mode than in the circulation heating mode.

Decoupling chamber is an important component of a pulse combustor which is not only for noise elimination, but also to ensure the acoustics boundary conditions at the exit of tailpipe as to maintain the performance of the whole combustor. Thus, a decoupling chamber pressure control system was designed and installed at the end of the tailpipe of a valveless self-excited pulse combustor. The relative pressure in the decoupling chamber was controlled from about -10 kPa to 10kPa through a blower, and the influence of pressure variation on the heat transfer in tailpipe was investigated. Experimental results showed that when the pressure in decoupling chamber is higher or lower than atmospheric pressure, the heat transfer coefficient in tailpipe would be increased, the heat transfer coefficient in tailpipe is relatively higher when pressure in decoupling chamber is lower than atmospheric pressure,and both the magnitude of pressure amplitude and the velocity ratio in the combustor can reflect the heat transfer coefficient in tailpipe.

Due to the design of fan-shaped reentrant cavities，the microchannel with periodically changeable cross-sections is divided into constant diameter region and arcuate region. The structural parameters，including the lengths and widths of the constant diameter region and the arcuate region have an important influence on the behavior of fluid flow and heat transfer in the microchannel with periodically changeable cross-sections. The influence of structural parameters with three-dimensional numerical analysis was investigated in this paper. The steady，laminar flow and heat transfer equations were solved by using a finite-volume method. The analysis showed that the influence of different structural parameters on fluid flow and heat transfer was very significant. When u =4 m·s^-1，the optimized structural parameters were 0.15<L₁/L₂<2 and 1.4<W₂/W₁<2.1.

The design calculation and CAD system for fin-and-tube heat exchanger was developed. This software, suitable for the industrial application of domestic heat exchanger manufacturers, provides an all-in-one service for the designers integrated functions of thermal performance design, sizing and strength design calculation, costing calculation and a CAD system. In the performance design module, some measures such as pre-screening and optimization of the search strategy were selected to reduce the calculation times, and all the solutions meeting the thermal indices and discrete design parameters can be obtained at the same time. The sizing calculation and strength design were slightly coupled, which makes the results meet different strength criteria flexibly. The structure parameters passed to the CAD system were divided into four parts, through which the strength of the components would not be influenced with the adjustment of parameters in the CAD system.

The micro-mixing time in a continuous system is studied with the incorporation model, in which the direct and indirect solutions are compared and the assumptions and parameters are modified to suit for the continuous system. Results showed that the direct solution is more close to the real mixing state, which is strongly affected by the change of the concentrations in the surroundings caused by the ratio of the continuous flows. While, the real concentrations in the aggregate have significant effect on both the solutions, even more than ten times, especially c_H⁺ in the aggregate. By fitting, the relationships of X_s and t_m resulted from both solutions are similar,lgX_s is linear to lgt_m when t_m <0.01 s, and is quadratic when t_m is larger.

The process of bubble growth on heating wall in subcooled boiling includes the micro-layer evaporation on heating wall and the bubble top coagulation when the bubbles grow to a certain size and emerge into the subcooled mainstream fluid. Based on this consideration, a model for the single bubble growth of subcooled flow boiling in vertical narrow rectangular channel was proposed.Compared with experimental results, the error of the simulation results using the proposed model is less than ±25%. The simulation results indicated that as the wall superheat increases, the bubble growth gets faster, with the subcooled degree of mainstream increases, the bubble growth in later stage would be slowed, with the contact angle increases, the contact radius of the bubble bottom and the wall tension would be strengthened, resulting in faster bubble growth to make the bubble to be flat and more easily exposed to the mainstream. The velocity of mainstream has no significant effects on bubble growth rate.

The interface motion in the injection molding process is simulated with the Level Set two-phase flow method，which can avoid both to deal with the complex boundary conditions and to extend the melt velocity to out of the melt boundary by the Ghost method at each time step. The physical governing equation systems are solved by the finite volume method on a non-staggered grid. The Level Set and its reinitialization equation are solved by the finite difference method. Two injection processes，that is the injection process at the horizontal central plane and the vertical central plane of the cavity，are simulated. The influences of different injection velocities，different quantities of the inlets and different Reynolds numbers on the interface motion are studied. The positions of the interface at different time and the pressure distributions when the injection process is over are presented. The features at different stages of the melt motion and how such stages generated are analyzed. The numerical results，which coincide with the experiments，show that when the difference between the width of the injection inlet and that of the mold is not large，if low or middle injection velocities are used，the melt will not break and can be transported steadily to the whole mold cavity. The higher the injection velocity，the shorter the time to reach stable extension flow.

In order to ignite the open flow gas of gas well containing sulfur immediately and reliably, the original open flow device is redesigned according to the conditions of gas wells in SINOPEC Southern Company. A mathematical model for the improved open flow device is proposed based on computational fluid dynamics, and the opening flow and concentration field situations are simulated. Combined with experimental results in situ, the simulation results are optimized, and then the range of ignition position for open flow gas is determined, which provides a theoretical and technical supports for the installation of ignition device.

An 8-lump kinetic model is proposed for Daqing naphtha pyrolysis based on the analysis of experimental data obtained in a self-designed cracking experiment device. A computational program was written by Marquardt⁺⁺ method and Matlab language and kinetic constants were calculated. The error analysis showed that most of the apparent activation energies are between 112 and 317 kJ&#8226;mol^-1,the raw material lump relative error is approximately about 10%, while the relative error for the main target products is basically within 7%, in which the average relative error for ethylene yield is 1.62%. The proposed model can describe the reaction process of Daqing naphtha pyrolysis and predict the distribution of products.

The hydrogenation kinetics of Shenhua coal liquefaction residue was studied in a microreactor. The experiment was carried out at reaction temperatures from 425℃ to 485℃ and reaction time from 0 min to 30 min with 6 MPa hydrogen pressure. The product was divided into three types, oil and gas (Oil+G), preasphaltene and asphaltene(PAA) as well as tetrahydrofuran-insolubles (THFIS). A model to describe the hydrogenation of Shenhua coal liquefaction residue was proposed. Results indicated that the proposed model can predict well the conversions to the three types of products. The activation energies of THFIS to PAA, PAA to Oil+G and PAA to THFIS were calculated as 147.41, 34.81 and 173.48 kJ·mol^-1, respectively.

A series of CuZnAlSi bifunctional catalysts with different SiO₂ contents were prepared by a complete liquid phase method, and the catalysts were characterized by using in-situ XPS, XRD, BET and NH₃-TPD. The catalytic performance was evaluated in a slurry reactor. Based on these results, it was deduced that SiO₂ in the CuZnAlSi catalyst could interact with AlOOH resulting in the change of some physicochemical properties, such as the pore structure and surface acidity, thus the catalytic activity for DME synthesis has got improved. It was found from the in-situ XPS characterization that the surface active species for the methanol synthesis in the process of DME production is composed by the reduced Cu⁰ species and ZnO. However, the stability of the catalyst decreased owing to the introduction of SiO₂, which may due to a lower interaction among the components of Cu, Zn and Al.

In order to improve the reaction rate of pyrotechnical reagent, transition metal oxides (TMO), Fe₂O₃ and CuO, were added into potassium perchlorate and potassium nitrate respectively by mixing and crystallization methods. Accelerating rate calorimeter (ARC) was used to study the catalytic behavior of TMO. Results illustrated that Fe₂O₃ can catalyze pyrotechnical reagent with potassium perchlorate, and the addition is better with crystallization method. The maximal reaction rate of pyrotechnical reagent with potassium perchlorate added with Fe₂O₃ by crystallization method is 8.10 min^-1, which is 4.09 times of that without catalyst. The time to maximal rate is 8.52 min, which is lower than that without catalyst by 92.8%. It was also found that CuO can catalyze pyrotechnical reagent with potassium nitrate, and the addition is better with mixing method. The maximal reaction rate of pyrotechnical reagent with potassium nitrate added with CuO by mixing method is 7.38 min^-1, which is 1.30 times of that without catalyst. The time to maximal rate is 53.55 min, which is lower than that without catalyst by 66.3%.

The dielectric microwave transmission window is one of the key components used in power couplers for microwave chemical reactors. It should keep low microwave power reflection to avoid burning the power source, and at the same, should airproof the cavity of reactor to separate the reactant from the microwave equipments. Based on multi-physics field model together with finite-element method (FEM), the mechanical and thermal stress characteristic of the dielectric microwave transmission windows during the processing are calculated. Valuable results for applications in industry are obtained. Finally, a novel gasket made by PTFE is proposed to add on the microwave coupled window design, which can reduce the stress and improve reliability.

The flow rate of gas-water mixture in the membrane reactor could be measured with laser Doppler velocimetry（LDV）. The impact of deflector，aeration intensity and flow rate of effluent on the shear forces of the membrane surface was investigated. The results showed that the deflector significantly increased the tangential velocity and turbulence of the mixture on the membrane surface. As a result，the shear force on the membrane surface was enhanced and in turn the formation of cake layer was delayed，thus alleviating concentration polarization. The vertical and horizontal velocities increased with increasing effluent flow rate. Also，aeration intensity affected the flow rate of the mixture remarkably. It was found that mean and pulse velocities were positively proportional to aeration intensity in logarithmic relationship，respectively，indicating significant effects of aeration intensity on turbulence and shear forces.

To enhance CO₂ absorption in aqueous solution of N-methyldiethanolamine (MDEA), four kinds of ionic liquid—tetramethylammonium glycin(［N₁₁₁₁］［Gly］), tetraethylammonium glycin(［N₂₂₂₂］［Gly］), tetramethylammonium lysine(［N₁₁₁₁］［Lys］) and tetraethylammonium lysine(［N₂₂₂₂］［Lys］)were synthesized and added in MDEA aqueous solution as activator. The performance of CO₂ absorption of the mixed solutions at total mass fraction 30% were tested by the constant volume method, and the proton transfer mechanism in promoting CO₂ absorption was also analyzed. The results indicated that these ionic liquids could greatly enhance the absorption, and the absorption rate increased with increasing concentration of ionic liquid. The mixed absorbents with lysine were of high absorption loads, and the initial absorption rate of ［N₁₁₁₁］［Gly］-MDEA aqueous solution was higher than the others used in this work. Moreover, the regeneration efficiency of the mixed solution with added ［N₁₁₁₁］［Gly］was 98%, higher than any other absorbents.

Crystallization is one of the primary processes for producing p-xylene. For the existing p-xylene crystallization kinetics，parameters were all acquired by regressing the profile data of temperature and concentration changes in the crystallization process with non-linear optimization method. Due to the lack of p-xylene crystal particle size distribution information，the accuracy of these crystallization kinetics parameters can hardly be guaranteed. In this paper，the p-xylene crystal size distribution data in the crystallization process are collected with an OPUS on-line particle size measure instrument based on ultrasonic spectroscopy analysis principle. The p-xylene crystallization kinetics is investigated by the method of moments through seeded batch cooling crystallization experiments with a binary mixture of p-xylene and m-xylene，in which the original p-xylene concentration is 82%(mass). The kinetic equations of p-xylene are obtained by regressing the experimental data with the multiple linear least square method. The results suggest that in the melt suspension crystallization process of p-xylene，the relative supersaturation has a larger effect on the crystal nucleation rate than the crystal growth rate. The agitation rate also has prominent effect on the nucleation rate，while the magma density does not.

The ceramic membrane process has been encountered with a problem of high energy consuming owing to that high crossflow velocities are necessary for the concentration polarization and membrane fouling control. On the other hand, the external loop airlift reactor is widely used in the field of chemical engineering. Thus, the application of air sparging in dextran ultrafiltration with ceramic membrane process is studied at different gas flow rates (Q_G). 87% increase in permeation flux was achieved while filtering 2 g·L^-1 dextran (2×10⁶) solution with the airlift crossflow operation of Q_G=400 L·h^-1. Hydrodynamics of the gas-liquid two-phase flow was characterized and the flux enhancement mechanism in the case of slug flow in tubular ceramic membrane was discussed.

Aqueous two-phase extraction of 2,3-butanediol from fermented broths was investigated by using ethanol/potassium carbonate system. The bioconversion of saccharified hydrolyzate of Dioscorea zingiberensis into 2,3-butanediol was carried out by Klebsiella pneumoniae. The partitioning experiment indicated that the optimal conditions for aqueous two-phase extraction of 2,3-butanediol from the fermentation broth clarified by hollow fiber membrane were 22% (mass) of ethanol and 26% (mass) of potassium carbonate. Under this optimal condition, aqueous two-phase extraction of 2,3-butanediol was carried out directly from the fermentation broth without filtration. The recoveries of 2,3-butanediol and acetoin in the top phase as well as reducing sugar in the bottom phase were 97%, 97% and 87%, respectively. The removal rates of cells and proteins were 99% and 94%,respectively. Pyruvic acid, citric acid, malic acid, fumaric acid and butane diacid were partitioned into the bottom phase. Aqueous two-phase system of ethanol/potassium carbonate exhibited a great potential in the industrial separation of 2,3-butanediol.

Effects of ultrasound on the adsorption and desorption of chromium (Ⅵ) on activated carbon were studied. Results showed that within the range of experiment concerned, the adsorption ratio of chromium(Ⅵ) decreases with pH increasing regardless ultrasound application. However, with ultrasound, the equilibrium of the system moves to the direction in the decrease of adsorption, which becomes severe with the increase in pH. With ultrasound, as the initial Cr(Ⅵ) concentration increases from 20 mg·L^-1 to 110 mg·L^-1, the removal efficiency of Cr(Ⅵ) drops from 99.9% to 79.8%, while the equilibrium adsorption capacity increases from 3. 3 mg·g^-1 to 15. 0 mg·g^-1, which is similar to that without ultrasound. The adsorption ratio continuously increases until to an equilibrium without ultrasound,while, it appears a slight decrease after being quickly to the equilibrium, and then gradually increases to the equilibrium again in case with ultrasound. The desorption ratio of Cr(Ⅵ) is very small in distilled water regardless ultrasound application. While, the desorption can be greatly improved by the addition of NaOH, and this promotion effect is remarkable for the ultrasonic system.

Chlorophylls were extracted by using ultrasonic from Spirulina platensis. Single factor examination and response surface analysis experiments were adopted to investigate the effects of extraction time, extraction solvent, solvent concentration, ratio of liquid to solid and extraction grade. The results showed that the optimal process parameters for this method were: extraction time of 56.5 min, ethanol concentration of 48.3% (vol) of ethanol/acetone solvent, and ratio of liquid to solid of 7. 9 ml·g^-1. The optimized chlorophylls extraction yield was 1.28%. The comparison experimental results indicated that the yield of chlorophylls by ultrasonic extraction was higher than that obtained from conventional solvent extraction.

A multi-objective optimization based dynamic fuzzy recurrent neural network（FRNN）modeling method was designed to control the pH neutralization process by generalized predictive controller（GPC）. To improve the FRNN fitting accuracy and simplify the network structure，a multi-objective optimization algorithm was proposed to optimize fuzzy rule numbers，center points and widths of Gaussian membership functions. The dynamic FRNN model was then obtained to precisely fit the pH neutralization process. Based on the dynamic model，the local linear model could be obtained at every control period. The complex non-linear optimization problem of GPC was changed into a simple quadratic linear programming problem. Simulation results showed the feasibility of the proposed method.

To solve the problem of optimal real-time control and large memory expense of partial differential equation（PDE）model with constraints，an approach is presented for optimal control with input/state constraints based on reduced-order model，which was derived from dynamic PDE in the form of a discrete-time low-order state-space model with high accuracy by using proper orthogonal decomposition（POD），singular value decomposition（SVD）and Galerkin projection. Then，by combining the reduced-order model,the modified optimal control method,the single-step rolling optimal control with certain input/state constraints based on quadratic programming which was demonstrated with the extremum validating optimal control based on linear quadratic regulator each other,was constructed. The controlling simulation results in the convection-diffusion-reaction process showed the efficacy and accuracy of proposed algorithm.

The molecular weight distribution (MWD) of polymer is one of the most important performance indexes, which is a kind of typical binary modeling. A method based on hybrid discrete orthogonal polynomial neural network (DOPNN) was proposed to model the MWD of polymers. First, the space and time variables were decomposed by the hybrid neural network. The DOPNN was used to obtain the space model of MWD, and the relationship between MWD and input variables (namely the time variables) was converted into the function between weight vector of space model and input variables. Second, the recurrent neural network was used to obtain the time model. Last, the modeling destination was reached by combining the two NN models mentioned above. The mathematical expression of the model was similar with the traditional discrete state-space expression. Based on the model, an easy way to design the control strategy could be achieved. In space modeling, the weight vector of NN was equivalent to the moment of MWD. That is to say, the weight vector of neural network was of practical significance, so that the grey box model could be obtained. A solution to forecast the number of hidden nodes of neural network was provided. The experimental system investigated was the styrene polymerization in CSTR. The results of the experiment indicated that the NN model was able to capture the MWD as well as to provide accurate moment of MWD through the weight vector of NN model.

Batch processes are very important in most industries and are used to produce high-value-added products, which causes their monitoring and control to emerge as essential techniques. Several multivariate statistical analyses, including multi-way principal component analysis (MPCA), have been developed for the monitoring and fault detection of batch processes. In this paper, an improved statistical batch monitoring and fault diagnosing approach based on variable-wise unfolding was proposed to overcome the drawbacks of traditional MPCA and the AT method proposed by Aguado. The proposed method did not require prediction of the future values while the dynamic relations of data were preserved by using time-varying score covariance, and principal-component-related variable residual statistics was introduced to replace SPE-statistics, thus avoiding the conservation of SPE statistical test and providing more explicit information about the process conditions. As a result, the root cause that violated the Hotelling T² test but still satisfied the SPE test could be unambiguously identified, which was impossible in the MPCA. In addition, time-varying contribution charts were proposed to diagnose anomalous batch process. The proposed method was applied to detecting and identifying faults in the simulation benchmark of fed-batch penicillin production. The simulation results clearly demonstrated the power and advantages of the proposed method in comparison to the MPCA and AT method.

The Borstar technology contains a loop reactor and a fluidized bed reactor connecting in series. Aiming at minimizing the grade transition time and the quantity of off-specification product, a grade transition model involving catalyst changeover was presented for this process. Meanwhile, the models of instantaneous and cumulative polyethylene properties, including melt index and polymer density in each reactor, were presented for the tanks-in-series process involving catalyst changeover. The results showed that if grade transition involved catalyst changeover, the recommended optimal strategy was to feed new catalyst step by step. Further more, it is necessary to consider the respective demands for manipulated variable profiles by old and new catalysts and the effect of old catalyst mass fraction on resin properties in each reactor. In this way, the transition time was remarkably reduced, the sharp fluctuation of manipulated variables was restrained and the profiles of polymer properties were smoothed up.

Based on the molding operator’s thought during the molding trial-runs, a hybrid intelligent model employing case-based reasoning, surrogate model and fuzzy inference was constructed. First, the idea of case-based reasoning was adopted for the initial process parameters setting, which simulated the molding operator’s behavior that one often recalls previous cases and sets the initial process parameters of the current one by referring to the previous ones. If the case-based reasoning failed to set the initial process parameters, a surrogate model was applied to determine the initial process parameters, which imitated a skilled operator’s “know-how” and intuitive sense acquired through long-term experience. Then, the molding trial would be run on the molding machine. Finally, a fuzzy inference based on expert knowledge was developed for correcting defects and optimizing process parameters during the molding trial run until the part quality was found satisfactory. A corresponding intelligent system was developed to be integrated with injection machine by communicating with the controller, and experimental studies showed that the intelligent system could be used in a shop-floor production environment.

Polarization data of BH^-₄ oxidation on Cu anode in alkaline solution were measured at steady state by a self-made experimental cell. The experimental polarization curve showed three regions：the region at lower over-potential η（below 0.4 V approximately），the region at higher η（above 0.6 V approximately）and the transitional region at medium η. It was found that the limiting current density is caused by the limiting elemental step rather than the external mass transport. The apparent reaction order with respect to BH^-₄ changes from 0 to 1. The active energy is about 40 kJ·mol^-1 in both regions with lower and higher η，and is higher than 40 kJ·mol^-1 in the transitional region. With the experimental temperatures and NaOH concentration，the hydrogen production rates released by the electrode reaction were measured at ［BH^-₄］/［OH^-］ratios of 1∶40，1∶20，1∶6. 7 and 1∶4，which is strongly dependent on the over-potential. The apparent number of released electron n by the reaction was calculated by using the measured amounts of hydrogen and coulomb. Higher value of n was obtained at lower ［BH^-₄］/［OH^-］ratio，higher temperature as well as higher over-potential. Under the experimental conditions，the value of n varies from 0 to 7，while in the transitional region it is just in the range of 3 to 5. The experimental results indicate that the limiting current density or n value will increase by virtue of proper experimental conditions. However，to considerably increase the current density at lower over-potential requires a modified anode with higher activity.

Guided by postulations in the relationship between structure and functionality, a saturation mutagenesis was performed on the D168 position of a P450 BM-3 (A74G/F87V/L188Q/E435T) mutant which originally transformed indole largely into indigo. Two novel mutants have been obtained based on the differences in color and absorption spectrum of the indole-hydroxylation products followed by product composition analysis using HPLC. One mutant, P450 BM-3 (A74G/F87V/L188Q/E435T/D168W) catalyzes indole into indirubin in 90% with a decrease of k_cat/K_m in 80%, mainly owing to its K_m is increased by 4.8 times as compared with the parent. The other one, P450 BM-3 (A74G/F87V/L188Q/E435T/D168R) transforms indole into indigo in about 87%, which is higher than the parent and its k_cat/K_m has an increase in over 1.37 times as that of the parent. Such results indicate that the amino acid residue substitution on the D168 position combined with the mutation of E435T can influence the hydroxylation regioselectivity of P450 BM-3 as well as catalytic activity, which determines indole-hydroxylation product as indigo or indirubin. Such influence could not be caused by the mutation on a single amino acid residue site, neither E435 nor D168, because of the lack in some synergistic effect. In conclusion, the directed evolution guided by rational hypothesis for engineering the catalytic properties of P450 BM-3 has succeeded in getting two mutants exhibiting novel characteristics for indole-hydroxylation.

The recondensation system of boil-off gas (BOG) in LNG (liquefied natural gas) receiving terminal is simulated with software PRO/Ⅱ8.0 aiming to reduce its high energy consumption, and the key process parameters such as the change of the output of LNG, the mass ratio and the BOG temperature are analysed. An optimized process, named the precooling of BOG, is proposed, in which the sub-cooled high pressure LNG is used to cool the BOG from the compressor before it enters into the recondenser aiming at reducing the compression ratio of BOG compressor. The optimized process can save the work of BOG compressor in 32.5% compared with the existing process and can improve the operation flexibility when the downstream system of gas transportation appears fluctuation in gas load between peak and valley. Therefore, the proposed process has better performance for peak-valley adjustment.

For researching the energy saving mechanism of 168 kA new type cathode aluminum reduction cell, the thermal field, electromagnetic field, flow field of liquid metal were simulated and the surface oscillation of liquid metal in the new type cathode aluminum reduction cell was measured by a dynamic detecting system on the surface oscillation of liquid metal which was designed by the authors. The results showed that the cathode convex in the new type cathode aluminum reduction cell could reduce the velocity of liquid metal and could weaken the surface oscillation of liquid metal. The X velocity and Y velocity of liquid metal in the new type cathode aluminum reduction cell were lower than that in the traditional cathode aluminum reduction cell. The polar distance of the new type cathode aluminum reduction cell was 3.8 cm. It was 1cm lower than that of the traditional cathode aluminum reduction cell. The voltage drop of electrolyte in the new type cathode aluminum reduction cell was 380 mV lower than the voltage drop of electrolyte in the traditional cathode aluminum reduction cell. The energy utilization of the new type cathode aluminum reduction cell was 5% greater than the energy utilization of the traditional cathode aluminum reduction cell.

The treatment of real leachate from municipal landfill with high ammonia nitrogen(NH⁺₄-N) content was studied by using a combined process of lab-scale anoxic/anaerobic up-flow anaerobic sludge bed (UASB)-A/O. The method for achieving and stabilizing partial nitrification in the A/O reactor was also investigated. Experimental results clearly showed that COD and nitrogen can be simultaneously deeply removed with this combined biological system, average COD and NH⁺₄-N have been reduced from 6537 mg·L^-1 and 2021 mg·L^-1 in the raw leachate to 300 mg·L^-1 and 15.6 mg·L^-1, respectively. For the anoxic UASB reactor, the average organic loading rate reaches to 6.5 kg·m^-3·d^-1, and its average organic removal rate is 5.3 kg·m^-3·d^-1. The denitrification and methanogenesis are simultaneously conduced in the UASB reactor, and almost 100% denitrification has been achieved. A stable and long term partial nitrification have been realized by the cooperative effect of FA and FAN, the ammonia nitrogen has been removed through the highly effective partial nitrification with above 99% nitrite accumulation rate.

The dynamic characteristics of enzyme activities and bacteria species groups in the substrate of stream banks have been studied through the construction of a special riverine ecosystem, in which the ecological embankments are made up of prefab spherical bricks with porous concrete. The microbial biomass, three kinds of bio-enzymes and five types of bacteria species groups in the substrate of ecological embankments were scouted. Results showed that there are significantly distribution differences (p<0. 05) in the microbial biomass and bio-enzyme activities in the ecological embankments. The activities of dehydrogenase, urease and cellulase are significantly higher in June and September than those in March and December, and the activities of these bio-enzymes in aquatic plants district of the ecological embankments are also higher than those of other places. Microbial biomass appears a peak in June, and its spatial characteristic is similar with the enzyme activities, which all have an accumulated ecological effect on the hydrophyte rhizosphere. The performance indexes of microorganism in the water level fluctuation zone are all superior to those in other places. Aerobic bacteria (cellulose decomposing bacteria, ammonifying bacteria, nitrosation bacteria and nitrobacteria) and anaerobic bacteria (denitrifying bacteria) have the same distribution in the ecological embankments, the quantity of bacterium also appears a peak in June and September, and the dynamic characteristics of each type of bacteria in the special riverine ecosystem are consistent with the biomass and bio-enzyme activities.

In order to study the mechanism for bacteria leaching of pitchblende, several leaching experiments were conducted, including three powder leaching experiments with bacteria and iron, with bacteria and without iron, without bacteria and iron, and two specimen leaching experiments with bacteria and without iron, without bacteria and iron. Variations in the concentrations of bacteria, ferrous ions, total iron ions and uranium ions, pH value and redox potential of the leaching solutions were monitored in the powder leaching systems. The contents of O, Mg, K, P, S, Fe and U in the tailings from powder leaching experiments were analyzed and the surface morphology of the samples after leaching were observed. Results of the powder leaching experiments show that Acidithiobacillus ferrooxidans can effectively oxidize ferrous ions, sulfur in reduction state and elemental sulfur, resulted in the increase in redox potential and the decrease in pH value. In the leaching system with bacteria and iron, the leaching of uranium from the ore can be accelerated and the leaching rate is increased with the higher concentrations of bacteria and total iron ions, the lower pH value and the higher redox potential. Although the concentration of the total iron ions is relatively low in the leaching system with bacteria and without iron, the bacteria can also accelerate the leaching of uranium from the ore and increase the leaching rate. Results of the specimen leaching experiments indicate that the pitchblende ore has been directly oxidized by the bacteria, since corrosion pits can be observed on the specimen surface in the leaching system with bacteria and without iron. Corrosion pits have not been observed on the specimen surface in the leaching system without bacteria and iron.

In order to explore the activator migration through porous media in MEOR process，the flooding experiments of different major activators in filled sand cores and crude oil saturated cores were performed. In addition，the migration of activator with different injection rate and concentration in the cores were also investigated. Results showed that clogging of the cores is the lowest with glucose compared with corn slurry and starch solution. For the low price and good activation effect，corn slurry was studied further，the suitable injection conditions of which were found to be 10 g·L^-1 of concentration and 1.2 ml·min^-1 of injection rate. The activator migration in crude oil saturated cores is more smooth than that in filled sand cores.

On a laboratory scale opposed multi-burner gasifier (OMBG), the fly ashes at different sampling mouths are collected and analyzed by SEM, EDS, XRF and Malvern mastersizer. Most fly ash particles produced in the gasification are irregular, aggregate or spherical. As for the composition of the particles, carbon is the main content, while S, Fe and Na get enriched. At the same time, the concentration of Al and Si in the fly ash particles is lower than that in the original slag. From the nozzle plane to the exit of gasification chamber, the carbon content of particles decreases along the axes of gasifier. The carbon content of particles decreases rapidly from the nozzle plane to No. 7 sampling mouth and declines slowly from No. 7 sampling mouth to the chamber exit. The size of particles generated in the gasification appears a triple-humped-distribution with peaks at 0.1—0.2 μm, 2 μm and 14 μm. The particle size distribution in different sampling places is different. Above the impact plane, more ultra-fine particles are found and coarse particles are larger in location near the impact plane. In symmetrical up and down locations of the impact plane, the particle size distributions are similar, but there are more coarse particles below the impact plane. The coarse particle size decreases and the proportion of fine particles increases below the impact plane, while the proportion of coarse particles increases at the chamber exit.

SrB₆O_10^·5H₂O flower-like and jujube-like superstructures consisting of nanoplates were prepared by a facile solution route，in which NH₄HB₄O_7^·3H₂O and SrCl_2^·6H₂O were used as raw materials and the liquid precipitation method was adopted. The superstructures were characterized by SEM，TEM and XRD. The results of XRD and chemical analysis indicated that the products were single phase SrB₆O_10^·5H₂O and the purity was 99.5%. Experiments were carried out to investigate the growth mechanism of the flower-like structures with the assistance of the TEM and SEM analysis. The mechanism of spherical aggregation of the SrB₆O_10^·5H₂O nuclei followed by the strong anisotropic 2D crystal growth was proposed. The effects of reaction temperature and reactants concentrations were also studied. It was found that the sizes of the aggregates and nanoplates increased with the elevating of temperature. The structure of samples transformed from disordered aggregations into flower-like and jujube-like superstructures，and the temperature of 70℃ was a crucial point for the change of substructure nanoplates from irregular shape to rectangular shape，and from disordered aggregation to ordered aggregation. With a lower concentration of the strontium reactant，the degree of orderliness decreased and the aggregates consisting of disorderly arranged nanoplates were observed.

Atom transfer radical polymerization（ATRP），reversible addition-fragmentation chain transfer（RAFT）and conventional free radical polymerization（FRP）of polyethylene glycol methyl ether methacrylate（PEGMEMA）and polyethylene glycol dimethacrylates（PEGDMA）were conducted respectively，and their reaction kinetics and gelation behavior were investigated with respect to the polymerization rate，vinyl conversion，gel point，and evolution of network. All the three systems experienced autoacceleration of the reaction. The rate autoacceleration in the ATRP was caused by diffusion-controlled radical deactivation，while that in the RAFT was attributed to diffusion-controlled addition of propagating radicals to RAFT-capped chains. In the ATRP and RAFT systems，the slow growth of primary chains allowed adequate chain relaxation and diffusion of reacting species，which reduced the probability of intramolecular cyclizations and limited the formation of microgels，leading to earlier gelation，as compared with the FRP system. The crosslinking networks in the ATRP and RAFT developed mainly through gel radical propagation with monomers，along with incorporation of branched polymers，resulting in a gradual increase in gel fraction and network density.

3,3′-Disulfonated-4,4′-difluorophenyl sulfone (SDFDPS) was synthesized from 4,4′-difluorophenyl sulfone (DFDPS) with industrial grade. The starting material DFDPS was purified by sublimation before use. The reaction condition was optimized by varying the molar ratio of SO₃ to DFDPS, reaction temperature and reaction time. The purity and yield of the product was determined by using HPLC, UV, ¹H NMR and FTIR. The optimized reaction condition was found as that the molar ratio of SO₃ to DFDPS is 3.0∶1 with a reaction temperature of 110℃ for 20 h, under which neither monosulfonated product nor DFDPS residue was detected. The product SDFDPS was obtained with a total yield of 75% upon recrystallization twice from ethanol/water. Poly(arylene ether sulfone) copolymer with 60% sulfonation degree prepared from the synthesized SDFDPS shows a high relative viscosity, which also indicates the high purity of the synthesized SDFDPS.