Oscillations were widely reported in continuous fermentation processes. The mechanism studies of different type oscillations in Saccharomyces cerevisiae were summarized. Unique parameters oscillations were observed in continuous ethanol fermentation and similar industrial fermentation processes for other productions. Such a type of oscillation would break the steady state of system running. The mechanism, regulation and possible application of this oscillation were discussed in this review.

A simple estimation method for molar Gibbs free energy and enthalpy of formation for complex oxysalt minerals is proposed. Complex oxysalt minerals are divided into complex oxysalt minerals with water and complex oxysalt minerals without water. Complex oxysalt minerals without water can be considered as the composite of acidic oxides and alkaline oxides, so their molar Gibbs free energy or molar enthalpy of formation can be calculated by the summation of molar Gibbs free energy or molar enthalpy of formation of the acidic oxides and alkaline oxides which are combined into the complex compounds. Molar Gibbs free energy or molar enthalpy of formation of complex oxysalt minerals with water is obtained by the summation of molar Gibbs free energy or molar enthalpy of formation of the part without water and water. The calculation method represented in this paper is very simple and convenient. By comparing the relative errors, the method is better than other methods. Moreover, it can be used to calculate molar Gibbs free energy and molar enthalpy of formation of all complex oxysalt minerals including a series of titanate whiskers that author’s group has studied for many years. The thermodynamic data for the series of titanate whiskers provide necessary fundamentals to their preparation.

Two-stage compression is an effective method to improve the coefficient of performance（COP）of trans-critical refrigeration cycle of CO₂，and COP of a complete inter-cooling mode is higher than that of an incomplete mode. For the trans-critical throttling cycle of CO₂ in a two-stage compression with complete inter-cooling，the throttling cycle can be divided into two types，single-throttling（STDC）cycle and double-throttling（DTDC）cycle. For these two cycles，the effects of outlet temperature of gas-cooler and heat rejection pressure on the COP are investigated，and compared with those in a single-stage compression（STSC）.The results indicate that under given condition，DTDC cycle presents the maximum COP，while the STDC cycle presents the highest optimum high pressure. For the three cycles，increase in gas-cooler outlet temperature decreases the COP，increases the optimum high pressure，and decreases the isentropic compression efficiency. The optimum inter-cooling pressure for STDC cycle and DTDC cycle deviates from the geometric mean of high pressure and low pressure，but this has little effect on the system performance.

Recurrence plots analysis was to the pressure fluctuation signals with chaos and fractal characteristics in the gas-solids recycled fluidized bed. The relationship between characteristic parameters of recurrence plots and the largest Lyapunov exponent was studied and it was concluded that the characteristic parameters of recurrence plots could reflect the extent of chaos. The change of the characteristic parameters in different styles of fluidized bed was investigated. The characteristic parameters increased or decreased swiftly in the styles of bubbling bed and fast fluidized bed, and kept steady in the styles of fixed bed, turbulent bed and pneumatic conveying bed. Recurrence plots could also indicate the style of fluidized bed.

The flow behavior of particles in a two-dimensional riser was modeled with the combination of the discrete element method (DEM) and computational fluid dynamics(CFD).Gas turbulence was modeled by large eddy simulation (LES), and DEM was used to model the collision of particles. A transport energy weighted averaging approach was proposed to coupling the gas phase in the Euler coordinate with the particles in the Lagrange coordinate on the basis of the balance of the transport kinetic energy. Simulated results showed that the clusters grew in size and increased in number along the flow direction. A core-annulus flow structure with a dilute phase in the center and a dense phase near the walls was formed in the riser. Simulated results showed the formation and break-up of clusters and the motion of particles in the riser. Simulated results were in agreement with available experimental data. However, the comparison with different models of drag force remained rather limited. More detailed simulations are required to get a more complete validation.

Turbulent combustion deals with micro-scale coupling of chemical reaction and turbulent flow. The validity of turbulence models used in flow simulation determines the success and accuracy of combustion simulation. In the present work，five variants of k-ε and k-ω family models and Reynolds stress model（RSM）were used to simulate the cool flow field（without combustion）in a bluff-body burner and PIV measurement was conducted to validate the simulation results. Comparison of numerical results with the PIV measurements showed that the simulations based on different turbulence models deviated much from each other for velocity distribution and eddy structure. Numerical results based on the standard k-ε and RNG k-ε models were in good agreement with the PIV data.

The subcooled boiling flow of water in a vertical tube under high pressure and high subcooling condition was numerically simulated using the two-fluid model, in which the bubble diameter model was a key factor to describe the subcooled boiling process. The influence of bubble diameter on the wall heat transfer was revealed by comparing different bubble diameter models, and the model suitable for high pressure was determined. The influence of pressure and wall heat flux on the flow and heat transfer characteristics was investigated. The calculation results show that the bubble departure diameter decreases with increasing pressure,which leads to the increase in the proportion of single-phase convective heat transfer and the decrease in the surface heat transfer coefficient. Moreover,the extremely nonuniform distribution of bubbles strongly depends on the characteristics of high pressure and high subcooling. Bubbles are aggregated near the wall with the increase of wall heat flux, which has significant influence on heat transfer characteristics of subcooled boiling process.

To enhance the heat transfer on shell side of a double-pipe heat exchanger with helical fins on the outer wall of the inner tube, some pin fins were installed along the centerline of the helical channel formed by helical fins. The heat transfer enhanced by these fins was investigated experimentally and the enhancement mechanism was analyzed by applying the field coordination theory. The heat transfer coefficient in the exchanger with helical fins and pin fins was 1.53—2.53 times as much as that with helical fins only at the same pitch and was 3.08—4.87 times that without the fins for Reynolds number ranging from 4000 to 14000. The performance of the exchanger with pin fins and helical fins was much better at the same mass flow rate, pressure drop and pumping power. Reducing the pitch of pin fins by half, the heat transfer was much better. At the same pressure drop and pumping power, the heat transfer in the exchanger with odd-numbered pin fins within one pitch was a little better than that with even-numbered pin fins. The pin fins increased the radical velocity. Accordingly, the resultant effect of temperature field and velocity field improved the heat transfer.

The convective heat transfer with constant heat flux condition inside a circular microchannel was investigated. The velocity and temperature profiles with a slip velocity at the surface were derived，and then the heat transfer coefficient and Nusselt number were obtained. A model was proposed for effective conduction at the super-hydrophobic surface with different structural parameters and the thermal resistance of the surface with trapped air was calculated. The effective heat transfer coefficient of super-hydrophobic surface was related to the heat transfer coefficient and surface thermal resistance. The calculation gives the following results. The slip of fluid on a super-hydrophobic surface makes the temperature profile inside the channel more uniform，so that the heat transfer coefficient or Nusselt number is increased，which may reach 2. 8 times that without slip under constant heat flux condition. The thermal resistance of super-hydrophobic surface increases with trapped air volume. The effective heat transfer coefficient on super-hydrophobic surface declines seriously as the trapped air volume increases，especially as the trapped air area at the surface increases. There exists a critical thickness for the trapped air on the super-hydrophobic surface with given surface structural parameters，under which the effective heat transfer coefficient is not less than that on general surfaces without slip. Therefore，it is necessary to adopt appropriate structure for the surface，such as rib height and distance between ribs，to avoid trapping air，so that the heat transfer on the super-hydrophobic surface is enhanced.

The mechanism for drying paste materials was investigated in a fluidized bed with inert particles and submerged heating tubes. The correlation equations were obtained for the drying time and specific evaporation rate of water, from which the drying performance can be predicted. The paste TiO₂ material was dried in a fluidized bed in pilot plant with inert particles and submerged heating tubes. The paste material was sprayed in droplet to the surface of inert particles in the range of 200—400 μm with air-blast nozzles. The appropriate drying parameters, operation parameters and equipment parameters were explored. The results show that this drying technology enhances the heat transfer and the paste material with high viscosity can be effectively distributed into the fluidized bed, with uniform temperature in the bed, large flexibility of operation, and low energy consumption. The new drying equipment has high drying capacity, and the heat transfer coefficient is more than 300 W·m^-2·K^-1.

Two super-hydrophobic surfaces with micro-nanostructure were prepared with n-octadecyl mercaptan self-assembled monolayers（SAMs）.The wettability and heat transfer characteristics during the condensation of steam or steam-air mixture were investigated experimentally on a vertical tube. For the condensation of pure steam，the condensation heat transfer on the super-hydrophobic surfaces with micro-nanostructure is worse than that on a smooth hydrophobic surface. In the presence of a low-concentration of noncondensable gas（NCG），the condensation effect is almost the same for the super-hydrophobic surface and smooth hydrophobic surface. During the condensation，the micro-nanostructure，an important feature for the super-hydrophobic surfaces，does not adsorb and trap air to exhibit the super-hydrophobicity. The wetting mode of solid-liquid changes from the Cassie state in the air to the wetting state during steam condensation. The contact angle hysteresis increases. The retardarce of micro-nanostructure to the condensate increases the thermal resistance. The results demonstrate that the solid-liquid wettability of micro-nanostructure significantly influences the heat transfer performance.

The temperature measurement of combustion flame is very useful in studying flame structure, fuel mixing and combustion,and stable operation and control of combustion equipment. In this paper, the high speed optical sectioning method is introduced to reconstruct the flame temperature field. A 3-D irradiance body can be regarded as a combination of a set of 2-D luminous sections and the corresponding image of the 3-D irradiance body can be regarded as the superposition of the focused and the defocused images. Focusing on each section and taking photos using a single CCD camera, the original luminosity distribution of each section can be retrieved by using the image inversion algorithm. Furthermore, based on the relationship between luminosity and temperature, the temperature distribution of flame can be reconstructed. The candlelight experimental results prove the feasibility of the method and reveal the internal radiation structure of flame. It can be used as a new measurement and diagnostic tool for quantitative characterization of a combustion flame and other two-phase flow parameters field.

A gas-solids fast separator was studied for its potential application in the process of coal pyrolysis to acetylene in plasma. The CFD method was employed to simulate the flow behavior in the gas-solids fast separator based on the realizable k-ε turbulence model and the discrete particle model（DPM）.The separation efficiency and residence time of gas phase in the gas-solids fast separator could be calculated. The numerical simulations were validated by the experimental results at a low velocity of the inlet gas（e. g. ，4 m·s^-1）.With the increase of gas velocity at the inlet，the separation efficiency was increased，and the residence time of gas phase was reduced accordingly. The separation efficiency approached 100% when particle diameter was larger than 20 μm. When inlet velocity was 100 m·s^-1，the mean residence time of gas phase was about 35 ms. To be noted，the performance of the gas-solids fast separator could be improved，for example by shortening the length of the separator，with a reduced residence time of gas phase at ～20 ms. It is expected that the gas-solids fast separator can meet the stringent demand of the coal pyrolysis to acetylene process for the milliseconds reaction，quench and separation.

Compared with gas-liquid two-phase flow，oil-gas-water three-phase flow is much more complex. There is immiscible oil-water，whose interaction and dispersion greatly affects the flow characteristics. The slug flow pattern of oil-gas-water three-phase and its flow pattern transition were studied in a 95 m long，51 mm i. d. horizontal pipe. The oil-gas-water three-phase slug flow pattern could be classified into five sub-flow patterns. The slug flow was W/O or O/W one during its transition to roll wave，which was three-layer flow pattern without mixed-phase on the interface. An even larger superficial gas velocity was needed for the transition boundary of slug flow and roll wave flow when the superficial liquid velocity is large. Besides，the region of roll wave flow pattern became smaller. The above-mentioned transition only happened when the water cut of liquid was between 30% and 70%. At the same superficial liquid velocity，there appeared a minimum superficial gas velocity corresponding to the transition of flow pattern when the water cut of liquid was between 40% and 50%.

Gas phase mass transfer in falling film microreactors (FFMRs) with the absorption of CO₂ into aqueous solutions of NaOH was investigated. The overall gas-phase mass transfer coefficient increases with NaOH concentration, but decreases as the concentration of CO₂ increases. There exists an entrance effect, hindering the mass transfer, which is caused by the dead volume for gas-phase flow in the gas chamber in FFMRs. The entrance effect has a larger impact in a shorter FFMR owing to the relatively large dead volume with respect to that of gas chamber. A decrease in the depth of gas chamber facilitates the mass transfer process. Therefore, the gas-phase entrance or geometry of the gas chamber should be designed appropriately to reduce the entrance effect and improve the mass transfer.

Due to its energy saving, the solid desiccation system has more and more applications. However, the regeneration of the system is beset with high operating cost and complicated equipment. In this study a new electro-osmotic regeneration method was proposed, and the feasibility was explored experimentally. An electro-osmotic regeneration system with a sample box, controlling and measuring apparatus was set up, and its performance was evaluated. In the experiment, an obvious electro-osmotic regeneration effect was observed for the solid to desiccant in the humid air at normal temperature, and the maximum regeneration rate reached 0. 21 g·m^-2·s^-1. It was found that the whole electro-osmosis process was divided into three phases because of the Joule heating effect and electrode corrosion, which hindered the electro-osmotic regeneration. The experiment showed that the Joule heating effect could be reduced by controlling the supplied electric field at the initial stage.

A new method to prepare chromic oxide green pigment was proposed. According to the investigation of key factors and related mechanism of the hydrogen reduction and sintering step, chromic oxide green pigment with excellent performance was prepared, and NaOH as byproduct was obtained for recycling use. The results showed that lower hydrogen reduction temperature and suitable addition of Al, were beneficial to obtaining the high quality chromic oxide green pigment. Additionally, sintering mechanism of the chromic oxide green pigment was evaporation-condensation. Based on the chromate clean process and integrated technology of chromium compounds which was developed by the Institute of Process Engineering, Chinese Academy of Sciences, the new method efficiently enables the short and clean production route from chromite ore to chromic oxide green pigment, expanding application of chromate clean production.

A new reaction for the halogenation of thiophene in the coking benzene was investigated. The method was designed to occur in a water-benzene two-phase mixtures solution containing both the permanganate and the chloride. A partial of thiophene will react with chlorine to generate the 2-chlorothiophene or 2,5-dichlorothiophene. The total sulfur and the thiophene contents in the benzene were determined by micro coulometer and GC-FPD respectively. The chloro-thiophenes produced were characterized by GCMS analysis. The preliminary results show that the thiophene in the benzene can not be oxidized by alone potassium permanganate solution even at pH 2—3. When the chlorides were used as the co-reactant the thiophene can be converted to chloro-thiophene. The process could be explained as follow: the chloride was oxidized by permanganate to generate chlorine in the suitable acidic solution, then the chlorine produced react immediately with thiophene to form chloro-thiophenes. The thiophene in the coking benzene could be completely converted by using the new method if the suitable reaction conditions are selected such as the pH value of aqueous solution and the molar concentration of both the permanganate and chloride.

This study presents a molecular dynamics simulation of protein separation by reverse phase liquid chromatography（RPLC）at all-atom level，using silica gel surface with C₄ compounds and cytochrome C as representative adsorbent and protein，respectively. The conformational transition of cytochrome C was monitored during the aqueous adsorption onto the adsorbent and elution with methanol. It was shown that during the adsorption，in which protein was attracted to the adsorbent surface，a simultaneous exclusion of water molecules occurred both at the adsorbent surface and the protein. A conformational transition of the protein conformation that led to the exposure of the hydrophobic amino acid residues，as compared to its native conformation in aqueous phase，was also displayed. During the elution by methanol，the replacement of water molecule by methanol occurred on both the surface of adsorbent and the adsorbed protein. As a result，the adsorbed protein was replaced by methanol. Above-mentioned simulation results reproduced the preferential hydration mechanism of the adsorption by RPLC and provided a molecular insight into the interaction among protein，adsorbent and the solvents at all-atom level，which is helpful for the design of new adsorbent and the optimization of RPLC process.

Adsorbent resin JD-1 was used for the in situ product removal (ISPR) in the pristinamycin production with S. pristinaespiralis E-71. A kinetic model for pristinamycin production and the film-pore diffusion model (FPDM) were combined together to describe the ISPR process in a 2. 5 L stirred tank bioreactor. Model parameters including pristinamycin formation rate constant, product inhibition constant, film mass transfer coefficient and pore diffusion coefficient were determined with the nonlinear least squares method. Theoretical calculations indicated that the model was in good agreement with the practical ISPR process. The effect of the amount and size of the resin on the ISPR results was studied. Experimental results showed that up to 90% of the pristinamycin produced by the fermentation could be recovered by the adsorbent resin, and that the performance got better with the decrease of resin size as well as with the increase of resin amount in a specific range. A yield of 1. 01 g·L^-1 was obtained with 95% adsorbed in situ when 70 g·L^-1 resin (R=0.195 mm) was added.

Robust optimization is time-consuming for uncertainty analysis. Aiming to improve computing efficiency, a sequential robust optimization method was proposed, which combined dual-Kriging model with multi-point sampling criterion. Dual-Kriging model, constructing the relationship between variables and standard deviation of the objective, reduced the computing time of uncertainty analysis in optimization to improve computing efficiency. Multi-point sampling criterion realized sequential iteration process, and improved the global optimum. The test results of a mathematical function showed that the proposed robust optimization method was more efficient in comparison with common surrogate-based robust optimization. Lastly, the proposed optimization method was applied to optimizing the wall-thickness and injection process for a box-shape injection molded part by considering the instability of injection process, and the results showed that the optimization method was effective.

The mesoscopic dynamics (MesoDyn) method was used to investigate the charge effects and charge matching interactions between surfactant and silicate species on the mesophase structure of SBA-15. The simulation results showed that a uniform hexagonal mesophase of SBA-15 was formed when no charge or one unit positive charge was distributed to the silicate species while 0.093 unit positive charges were distributed to each EO unit of the P123 surfactant. Moreover, the charge effects of surfactant and silicate species both had a significant impact on the mesophase structure of SBA-15. On the one hand, the silicate species with positive charges could promote the formation of surfactant liquid crystal phase. On the other hand, the surfactants with some weak charges could not only put the silicate species in order, but also promote the concentration of silicate species on the interface. The simulation results validated and interpreted the “surfactant/silicate species cooperative formation mechanism” of SBA-15. The MesoDyn method was developed to investigate the formation mechanism of ordered mesoporous molecular sieves. This work could make up for the lack of experimental and characterization technologies.

As its activity falls in high salinity water, the usual surfactant cannot be used for washing because of much salt, and increased surface tension and corrosion. It was reported in literature that some special surfactant can resist salinity up to 110000 mg·L^-1 （alkaline earth metals up to 2500 mg·L^-1）.A compound α-olefine polyether carboxylate whose range of resisting salinity up to 150000 mg·L^-1 （alkaline earth metals up to 20000 mg·L^-1）was introduced. Its washing and anti-corrosion effect in sea water was excellent. In the production of high-salinity heavy oil, it can reduce the viscosity of oil with formation water salinity as high as 150000 mg·L^-1 （alkaline earth metals up to 20000 mg·L^-1）from 100000 mPa·s (50℃) to 50 mPa·s (50℃).

The industrialization of electrochemical synthesis of 3,6-dichloropicolinic acid(3,6-DCP) from 3,4,5,6-tetrachloropicolinic acid(3,4,5,6-TCP) in a tank-type undivided electrolytic cell with silver cathode was investigated. The experimental results showed that the electrolytic performance greatly depended on the operation parameters, such as flow rate of electrolyte, temperature, and concentration of NaOH. The reduction of 3,4,5,6-TCP to 3,6-DCP was favored by increasing temperature and flow rate of electrolyte. Industrialization tests of electrochemical synthesis of 3,6-DCP were conducted under the following conditions: silver cathode, temperature 313 K, flow rate 6 cm·s^-1, concentration of NaOH 8%(mass), concentration of 3,4,5,6-TCP 5%（mass）, cathode current density 600 A·m^-2. The current efficiency and conversion ratio of 3,4,5,6-TCP were more than 76% and 99%, respectively. The yield of 3,6-DCP was up to 90% and DC power consumption was 2. 939 kW·h·(kg 3,6-DCP)^-1. The purity of the product was higher than 95%, and its melting point was 151—153℃.

La_0.8K_0.2Mn_0.95Cu_0.05O₃ perovskite-type oxide was synthesized by the complex method of citric acid, then was supported on γ-Al₂O₃ to obtain a series of loadings by using the impregnation method, complexing method and mixing method. Temperature programmed oxidation (TPO) was used to investigate their catalytic behavior of model soot oxidation in a feed gas containing various concentrations of SO₂. The catalysts were characterized with XRD, BET, and FT-IR. XRD and BET results showed that the catalysts exhibited good structure of perovskite-type before being supported, and the specific surface area increased after being supported. TPO tests revealed that the activity of La_0.8K_0.2Mn_0.95Cu_0.05O₃/γ-Al₂O₃ prepared by the impregnation method with 20%(mass) loading was better. La_0.8K_0.2Mn_0.95Cu_0.05O₃ catalyst presented higher catalytic activity than those supported ones, i. e. La_0.8K_0.2Mn_0.95Cu_0.05O₃/γ-Al₂O_3, however, sulfur-resistance of catalysts was improved when supported on γ-Al₂O₃. FT-IR spectrum confirmed that with increasing concentration of SO₂, the SO^2-₄ species formed and accumulated on the catalyst surface would inhibit the catalytic activity.

The behavior of capturing sulphur by calcined calcium-based industrial wastes such as red mud and carbide slag was evaluated at fluidized bed boiler temperature and compared with that of limestone. The change of phase and microstructure for these materials during reaction was characterized. The results show that with increase of reaction time the calcium conversion is higher in red mud than in carbide slag, and that in limestone is the smallest under the same conditions. The optimum temperature for sulphur capture is 850－900℃ for red mud and 950－1000℃ for carbide slag. With increase of SO₂ concentration the calcium conversion and the rate of sulfuration reaction for red mud increases for the same reaction time, and the effect of particle size on sulfur capture of red mud is not significant. Ca₂SiO₄ and Ca(OH₂ are main calcium compounds in red mud and carbide slag, respectively. The range of pore size for calcined red mud and carbide slag is 5－20 nm that is the most favorable for sulfur capture, while that for calcined limestone is 45－420 nm. These calcium-based industrial wastes are of better performance of capturing sulfur at the fluidized bed boiler temperature because of their suitable pore structure.

The carbonation characteristics and the decline of reactivity for recycle Ca-based CO₂ absorbent were investigated. The microstructure, pore radius and specific surface area of the absorbent changed with the number of cycles, leading to the decline of the absorbent reactivity. The results show that, the lamellar structure in the absorbent disappears, which changes to bigger cubic crystals, and most apertures were blocked. It reduces the specific surface area and total cumulative volume and changes the distribution of pore radius, and then the reactivity of Ca-based CO₂ absorbent and the absorption capability decline as the number of cycles increases.

The mesoscopic dynamics model was built to study the flocculation between colloidal particles and adsorbing linear polymer chains. The model was investigated especially for the flocculation of sludge and polyacrylamide and the dewatering of sludge. The flocculation model of colloidal particles with the negative electricity and positively charged polymer chains was built by establishing a coarse-grained model and using the force field of empirical potential function. The appearance of the aggregate structures was obtained by simulation and bridging of colloidal particles and neutralizing the charge on the surface of particles by polymer chains, leading to flocculation was proved by the model. The variation of the potential energy during flocculation as a function of normalized time was also studied. The morphology of aggregates was observed directly by the use of transmission electron microscope. At last, the influence on flocculation created by the variations of the exponent A and B of polymer chains showed that higher ionic degree facilitated flocculation. But the floc was more complex at the same time, which was unfavorable for sludge dewatering.

High effectiveness is one of the most attractive advantages of the anaerobic ammonium oxidation (anammox) process for nitrogen removal from ammonium-rich wastewater. Granulation is considered to play a vital role in development of high-load bioreactors. The hydraulic retention time (HRT) of the 1.1 L anammox EGSB (expanded-granular sludge bed) reactor was progressively shortened from 6.9 h to 0.30 h; while the influent nitrite concentration was fixed at 360 mg·L^-1 and the effluent recycling was maintained at 0.5 constantly. A super high nitrogen removal rate, 50.75 kg·m^-3·d^-1 was recorded, which was 2 times of the highest value reported before. Granulation of the anammox microorganisms occurred in the high-load anammox EGSB reactor. The average diameter of the high-load anammox granules was （2.51±0.91）mm and the specific anammox activity reached 1.899 kg·(g VSS）^-1·d^-1. The extracellular polymers (ECP) content was 143.00 mg·（g VSS）^-1, which was amazingly high when compared to heterotrophic microbial granules. The extracellular proteins increased at a higher rate along with the increase of hydrodynamic shear stress, resulting in over-production of extracellular proteins. Thus the stability of the anammox granules decreased and sludge washout became inevitable.

The partial nitrification (PN) process was investigated at 30℃ in an air-lift circulation (ALC) reactor fed with simulated wastewater. The results proved that it was a highly-efficient process for ammonia removal. The turnover of the ALC reactor was 12—28 times per day with hydraulic retention time between 0.86 h and 2.00 h. When influent concentrations of ammonia were increased from 358.5 to 942.3 mg·L^-1，the average volumetric ammonia removal rate reached as high as 5.5 kg N·m^-3·d^-1，which was at the top level reported in the literature. During the operation，the relative standard deviation of effluent substrate concentration，ammonia removal efficiency and nitrite accumulation was 4.3%—26.5%，3.1%—16.8% and 0.4%—5.3%，respectively，in spite of fluctuation of influent concentration，flow rate or pH. The stability of process performance could be attributed to both the efficient retention of sludge (4.0—5.2 g VSS·L^-1) and the high specific activity of biomass (2.71 g N·（gVSS）^-1·d^-1 determined from the kinetic study).

The interaction mechanism of benzo［a］pyrene（BaP）-cadmium（Cd）combined pollution in water body with Arthrobacter oxydans was investigated by using HPLC and FTIR analytical techniques. The experimental results showed that BaP，Cd and BaP-Cd combined pollution all posed adverse effects on the normal growth of the strain with the sequence of influence BaP-Cd＞Cd＞BaP. The concentrations of BaP and Cd in aqueous media were reduced to 0.39 mg·L^-1 and 0.65 mg·L^-1，respectively，from the initial concentration of 1 mg·L^-1 after treatment for 5 days at 30℃，130 r·min^-1.Adding H₂O₂ at a specific concentration and keeping the treatment system under alkaline condition were favorable to disposing of BaP-Cd combined pollution. Infrared scanning analysis showed that the structure of BaP changed with the action of microbe. Hydroxy，carbon-carbon double bond，aminoacyl and alkyl were confirmed to be the key functional groups for the strain to biodegrade BaP and adsorb Cd.

Successive concentration and purification of simulated nickel-electroplating rinsing wastewater was carried out by integrating membrane process with electrodeionization. The concentrate compartments were filled with ion exchange resins to enhance the separation. The concentrate stream of the primary EDI procedure was operated in closed circuit circulation. The influence of the volumetric ratio of resins in concentrate compartments on the separation was examined. It was found that the best performance could be achieved when anion to cation resin ratio of 6∶4 was adopted. With feed Ni²⁺ concentration of 50 mg·L^-1 and pH of 4.25，the Ni²⁺ concentration of effluent dilute stream could reach 2.78 mg·L^-1 while that of the effluent concentrate stream was as high as 11171 mg·L^-1，which gave a concentration ratio of higher than 220. The effluent dilute stream of the primary EDI was then sent to the second EDI stack for deep desalting. Dilute product with resistivity of 1.6—2.0 MΩ·cm was then obtained，which could be recovered as pure water for electroplating. The membrane process integrated with EDI could find its potent role for zero emission and resource reuse of heavy metal wastewater.

The decomposition of H₂O₂ at the Ti-based oxide electrode coated with IrO₂, RuO₂ and TiO₂ (Ti/ IrO₂/RuO₂/TiO₂) prepared by thermal decomposition, was investigated in the electrolysis system of constant potential and the non-electrolysis system respectively. Additionally, the influence of the decomposition of H₂O₂ caused by Ti-based oxide electrode on the oxidation of aniline was also investigated. The results showed that both higher loading of oxides and higher pH were able to accelerate the decomposition of H₂O₂ in the non-electrolysis system and in this case the decay of H₂O₂ was mainly caused by the catalytic action of the oxides coating. In the electrolysis system with Ti-based oxides electrode as anode, the decay rate of H₂O₂ increased with increasing anodic potential. In this case, the decay of H₂O₂ involved two mechanisms: catalytic decomposition and electrochemical oxidation. It was also found that the catalytic decomposition of H₂O₂ at the oxides electrode was useless to the oxidation of aniline while the electrochemical oxidation of H₂O₂ was only slightly helpful to the oxidation of aniline. This work suggested that using the appropriate anodes of less H₂O₂ decomposition as well as reasonable potential in the electro-Fenton process could achieve high chemical efficiency of H₂O₂.

To solve the problem of stacking complexity and different shrinkage perpendicular to the membrane during constrained sintering process, an improved pore evolution model with key parameters was improved. Considering the aggregation,initial porosity as the critical parameter of model was adjusted. Then the shrinkage and shrinkage rate in the perpendicular direction to the membrane could be calculated with the theory of constrained sintering, and its effect on the pore size of supported membrane was investigated. On the basis of the improved model, the change of mean pore size, porosity and thickness of zirconium supported membrane with sintering temperature could be predicted quantitatively in the temperature range from 800℃ to 1200℃. The prediction calculated from the improved model fitted the experimental data better than that from the previous model. It provides an effective tool in predicting and controlling the pore size of ceramic membranes during the sintering process. Furthermore, substituting the pore size, porosity and membrane thickness calculated from the model into Hagen-Poiseuille law predicted the pure water flux accurately.

For the injection-molded plastic part with more than one injection gate and the part with an insert，there would be a welding stage of the melt fronts in the whole filling process. How the melt fronts welded along the thickness direction is one of the key factors affecting the quality of the part，especially the mechanical properties and fiber orientation. In this paper the velocity and pressure of the melt plastic in the welding stage were simulated with the finite element method（FEM）of the T6P3 triangular element（quadratic velocity and linear pressure）.Two examples were used for the comparison of the velocity and pressure of the melt plastic in the thickness direction in the welding stage. One had an even thickness while the other not. For the given examples，the simulation results illustrated that the plastic welding process would be influenced by the thickness uniformity of the region where weld lines existed. The results of this paper could be the input data for other simulation of plastic injection molding，such as mechanical properties and fiber orientation，etc.

Polyurethane dispersions (PUDs) with solid content above 50% were synthesized by using polyether diol containing sulfonate (SPPG) as hydrophilic monomer in the soft segment, dimethylol propionic acid (DMPA) as hydrophilic monomer in the hard segment, poly(hexamethylene neopentylene adipate) (PHNA), isophorone diisocyanate (IPDI), hexamethylene diisocyanate (HDI) and piperazine by the acetone method. The results showed that the zeta potentials of PUDs were between -52 mV and -72 mV and the PUDs displayed excellent stabilization. The average particle sizes decreased, the particle size distributions narrowed and the viscosity increased when the hydrophilic group content and the mole ratio of HDI to IPDI increased. The viscosity evidently decreased when the mole ratio of —SO₃Na to —COOH decreased as a volume fraction of small particles between 21% and 28%.TEM analysis showed polydispersity and different sizes of the spherical particles. Dynamic mechanical analysis showed compatibility of soft and hard segments of the PUDs and a low glass transition temperature of the PUD film.

To improve the performance of glycidyl azide polymer（GAP），by using trimethylolpropane（TMP）as a crosslinker，isophorone diisocyanate（IPDI）as the curing agent，glycidyl azide polymer/ethylene oxide-tetrahydrofuran copolyether（PET）dual-soft segments energetic polyurethane elastomer was synthesized. The elastomer was characterized with FTIR，DSC，XRD and other means. The experimental results showed that some differences in the reaction rates of PET/IPDI and GAP/IPDI were observed at the content of catalyst was 0.1%，but PET and GAP could participate in the formation of polyurethane elastomer together. When PET was introduced in the elastomer，the tensile strength of elastomer increased by 0.9 MPa，elongation increased by 156%. The T_g of two soft segments were observed and its values changed with the content of two soft segments. GAP/PET/TMP/IPDI elastomer was an amorphous polymer and began to decompose at 183℃.

To obtain strong hydrophobic property of silica，on the basis of traditional modification process with silane coupling agent stearic acid was introduced as complex modifier to modify the nano-silica powder. The modified silica particles were characterized with Fourier transform infrared analysis（FTIR），X-ray photoelectron spectroscopy（XPS），contact angle tests and sedimentation tests. The complex modification mechanism was also discussed. The results showed that amino group（—NH₂）was grafted successfully onto the surface of nano-silica particles by modification with silane coupling agent. After the second modification with stearic acid，amide bond（—CONH—）was formed by amino group（—NH₂）and carboxyl group（—COOH）of stearic acid. As a result，the —（CH₂）₃COHN—（CH₂）₁₆CH₃ group was obtained on the surface of nano-silica particles，which had excellent hydrophobic property. Because of the complex modification，the stearic acid molecules were grafted on the surface of nano-silica particles by chemical bonds，and the contact angle could reach 140°.

The non-metallic powder recycled from waste paper-based printed circuit boards (PCBs) was analyzed with scanning electron microscopy (SEC), X-ray fluorescence (XRF), Fourier transform infrared spectrum (FTIR), and the mechanical properties, heat resistance of the non-metallic powder filled polypropylene (PP) composites were tested. The properties of non-metallic powder recycled from waste paper-based PCBs and its effect on the performance of the composites were studied. The results showed that the waste paper-based PCBs was resin polymer with a little fiberglass, and its polarity was due to the presence of hydroxide groups, carbonyl groups, acetal groups and silanol groups on the particle surface. Modified non-metallic powder and PP powder in the composite exhibited excellent compatibility, and modified non-metallic powder enhanced the bending strength, Vicat softening temperature (VST) and flame retardance capability of the composite. The bending strength of the composite increased with increasing filler proportion of the nonmetal powder, bending strength increased by 9.1 MPa and VST rose by 2℃ when the 0.08 cm（180 mesh） non-metallic powder filler proportion reached 30%. The non-metallic powder could be used as filler in the PP composite to improve the properties of the composite.