Both numerical simulation and experimental investigation were performed to develop a prediction method for extreme jet flow frequency.Extreme jet flow frequencies of different orders were in the relation of multiples.Numerical simulation could describe truly the motion of wave system and contact surface in the receiving tube.The simulation results of the extreme jet flow frequency were lower than those of experiment，but the error was less than 5%.Moreover，extreme jet flow frequency of each order，which was influenced more greatly by pressure ratio，increased with increasing outlet pressure and pressure ratio.Compared with the condition of the extreme jet flow frequency of the first order，the isentropic refrigeration efficiency was higher under the condition of the second and the third order.The above conclusions provide guidance for the rotary gas wave refrigerator to run stably and effectively in practice.

Fluid flow and mass transfer enhancement were investigated numerically with the software Fluent in a sinusoidal wavy-walled tube for pulsatile laminar flow，and the comparison between numerical and experimental results was performed.Three parameters were considered here：net flow Reynolds number，oscillatory fraction of flow rate and Strouhal number.Among them，the effect of Strouhal number was especially investigated.The numerical results of flow patterns and mass transfer enhancement below the moderate Reynolds numbers showed that there existed an optimum Strouhal number corresponding to the maximum mass transfer enhancement effect，and a good agreement was found between numerical and experimental results.

In order to make further study on the electrohydrodynamic enhancement mechanisms of boiling heat transfer,the growth process of boiling bubbles in the electric field was visually investigated with a high-speed camera.The bubble dynamic growth process was observed in the electric field and boiling heat transfer action influenced by the change of bubble behaviour in the electric field was analyzed.Experimental results showed that the bubble elongated along the direction of the electric field.The bubble’s aspect ratio and departure frequency increased with increasing electric field intensity, while its departure volume decreased with increasing electric field intensity.The bubble deformation resulted primarily from the effect of electric stress acting on the bubble surface.The bubble became slender and longer along the direction of the electric field because the electric stress compressed the bubble in the equatorial direction and extended the bubble along the axial direction.The boiling heat transfer coefficient and critical heat flux increased due to the change of bubble behavior in the electric field.

Using de-ionized water as working fluid，experiments were conducted to study the forced convective flow and heat transfer characteristics in ultra-light porous copper foam embedded in a copper channel.Through the analysis of experimental parameters，the effects of heat flux，pore density of the metal foam and mass flux on pressure drop，wall temperature and heat transfer coefficient were obtained during the laminar flow range.It was found that the metal foam could enhance the heat transfer and lower the wall temperature dramatically.The heat transfer capability increased with the increase of the Reynolds number.Compared with the common copper channel without the embedment of metal foam，the maximum Nusselt number during convective flow through the copper channel embedded with the metal foam was thirteen times larger，but the pressure drop was obviously increased.The increase of the Reynolds number and the pore density of metal foam also led to lager pressure drop.

An experimental set-up was designed and constructed to investigate the water vapor permeation through membranes.The test section mainly consisted of an airflow channel, a membrane, and a water tank, which formed a sandwich structure.The channel had a 5 mm×50 mm cross-section and a 800 mm length.Flowing in the channel was an air fluid with a low humidity.The membrane was sandwiched between the channel and water tank, forming an air gap with a higher humidity between the water and membrane.Moisture was transferred from the water to the air stream in the channel through the air gap and membrane.Tests were conducted on two membranes that included the PVDF (polyvinylidene fluoride) and PES (poly ether sulfone) membranes.The water vapor permeations through membranes were investigated.The membrane resistance was obtained by subtracting the convective moisture resistance in the channel and the moisture resistance caused by the air gap from the total moisture resistance.Through combination with the sorption curves of membranes, the diffusivities of moisture in membranes were estimated.The results showed that the diffusivities of the PVDF and PES membranes were in the order of 10^-6 kg·m^-1·s^-1, and the former was about 4 times as great as the latter.

To enhance heat exchange under high temperature condition，a nanofluid with alumina nano-particles suspended in propylene glycol and water（base liquid）was prepared.The boiling point，thermal conductivity，specific heat and viscosity were measured respectively by ebulliometer，transient hot-wire method，comparison method and rotation viscometer.The heat exchange coefficient and flow resistance of 1%—5%（vol）nanofluids were measured in the plate-pin oil cooler test.The experimental results showed that nanoparticles could significantly enhance the heat transfer capability of base liquid.Particle volume fraction and fluid temperature had important effect on thermal properties of nanofluids.The boiling point of nanofluids was over 120℃，a little lower than that of base liquid.Adding particle volume fraction would decrease specific heat but increase thermal conductivity，viscosity and heat exchange coefficient.In Test 2，the mean heat exchange coefficient of 5%（vol） nanofluid in the plate-pin oil cooler was 124.56％ higher than that of base liquid，while the flow resistance rose moderately.

Classical orthokinetic mechanism assumes that the fine particles near the core particle move straightly until hitting the core particle.Actually，airflow detours the core particles and the trajectories of fine particles will be deflected from straight lines by the airflow.The collision efficiency was introduced to denote the proportion of fine particles that collide with core particles.By modeling the collision efficiency of acoustic agglomeration，the effect on acoustic agglomeration was studied.The results showed that collision efficiency was very low for PM_2.5 especially for PM₁，which meant that the fine particles were hardly collected by orthokinetic agglomeration.Collision efficiency decreased as acoustic frequency increased，but it enlarged the range of optimum frequency of agglomeration.Higher sound pressure level（SPL）favored both collision efficiency and agglomeration kernel，so acoustic agglomeration increased in the higher-intensity acoustic field.Acoustic agglomeration was ineffective for SPL less than 140 dB.Temperature produced little influence on orthokinetic agglomeration at a low temperature，but a high temperature reduced both collision efficiency and agglomeration rate.

Ship’s ballast water is the main way of spreading harmful organisms between the geographically isolated water bodies, but it still can not be prevented with an environment friendly method at present.It is regarded as the most feasible online approach that the large number invasive microbes in ballast water are killed by advanced oxidant.However, it is a challenge how the reactive free radicals contact the large number microbes in large-volume and high-flow rate ballast water.In the paper, the Eulerian-Eulerian two-equation model was used to simulate the mixing process in which gas-liquid jet including reactive free radicals impinges into the ballast water in crossflow.The results showed that scarf vortices were formed in the upstream,leading to counter-rotating vortices in the downstream in confined crossflow, which enhanced the mixing process.However, the gas was not easy to diffuse in the gas-liquid jet on the downstream wall surface.

A theoretical model of macro kinetics for methanol oxidation in the porous anode of DMFC was developed.First, an intrinsic kinetics expression including the coverage ratios of CO and OH was derived from the dual site mechanism of methanol hydroxylation on Pt-Ru catalyst surface.Second, the balances for both material and charge in a differential volume of the porous anode were calculated to the model equations coupled for describing concentration and potential distributions in the anode.Then the generalized macro kinetics model was obtained with dimensionless variables and parameters.The physical characteristics of the porous anode, including the thickness of catalyst layer, specific area, effective diffusion and conductive coefficients, as well as the catalytic characteristics related to CO and OH coverage ratios as functions of the variable of thickness in particular, were included in this model.Furthermore, the expressions of effectiveness and polarization curve for the porous electrode were presented.The model was a boundary problem of a set of nonlinear second order differential equations, from which two equivalent differential equations can be obtained separately by decoupling.The model equation was solved by using Newman’s BAND (J) program while inserting a subroutine to calculate the coverage ratios at each note.Through comparing the model predicted values with experimental polarization data reported，it was found that the predictions agreed with the data satisfactorily under lower current densities, and were regularly lower than the data under higher current densities as CO₂ bubbling became considerable.Detailed analysis of macro kinetics showed that crucial issues for improving performance of the porous anode should be to increase the activity of Pt set of the catalyst for methanol electro decomposition, and to optimize both micro- and macro-structures of the anode for mitigating the influence of two phase flow.This work may also provide an analysis tool for the performance of anodes in DEFC as well as DBFC.

Liquid-phase oxidation of p-xylene is one of the most significant reactions in the petrochemical industry.The aerobic oxidation of p-xylene catalyzed by μ-oxo bis［tetra-para-chlorophenylporphinato］iron and cobalt acetate ［(T_p-ClPPFe)₂O/Co(OAc)₂］ without solvent and bromides was reported, and the co-catalysis between (T_p-ClPPFe)₂O and Co(OAc)₂ was proposed.The influence of reaction time, reaction temperature, pressure, air flow rate, the amount of porphyrins and the mass ratio of Co(OAc)₂ /(T_p-ClPPFe)₂O on the p-xylene aerobic oxidation was studied.The optimized conditions for the p-xylene oxidation with air were obtained by using orthogonal design and confirmed by critical tests.Under the optimized conditions of the amount of porphyrins 17.5 mg, the ratio of Co(OAc)₂/(T_p-ClPPFe)₂O 60, pressure 0.8 MPa, temperature 180℃, reaction time 4.5 h and air flow rate 0.20 m³·h^-1, the conversion of p-xylene was 65.0%, the selectivity of p-toluic acid and terephthalic acid were 81.0% and 13.0% respectively.

Tungsten substituted mesoporous molecular sieves W-SBA-15 catalysts with different Si/W ratios were directly synthesized under hydrothermal conditions by using Na₂WO₄ as tungsten precursor and tetraethyl orthosilicate(TEOS) as silica precursor.The prepared W-SBA-15 catalysts were characterized by means of XRD, HRTEM, UV-Vis DRS, NH₃-TPD and H₂-TPR.The characterization data indicated that amorphous tetrahedral and octahedral tungsten species were mainly incorporated into the framework of SBA-15 (Si/W mole ratio of 44), and further increase of the amounts of tungsten species led to the appearance of WO₃ in the channels of W-SBA-15 (Si/W mole ratio of 30 and 20).The metathesis catalytic performance of W-SBA-15 catalysts was better than that of the conventional WO₃/SiO₂ catalysts prepared by impregnation.Under the conditions of 300℃, 3.0 MPa and 6.4 h^-1, butene conversion was more than 50% with the propylene selectivity exceeding 96% during the 24 h continuous operation, using W-SBA-15 as the catalyst.The W-SBA-15 with the Si/W mole ratio of 44 exhibited the best dispersion of tungsten species inside the SBA-15 and the best catalytic performance of olefin metathesis among the W-SBA-15 catalysts.

Lactic acid (LA) is a commercial fine chemical, used in food and medicine, which is readily available by biomass fermentation.As a chemical feedstock, LA is widely used due to the renewable biomass resources.In this paper, BP neural network was applied to simulating the dehydration reaction of lactic acid.The optimal reaction conditions were determined through orthogonal design that could be used to investigate the interrelated effects of pH, feed liquid flow rate, carrier gas flow rate and temperature on the yield of acrylic acid(AA).With the training samples from experimental data of orthogonal design, a new model that could simulate the process of the dehydrated reaction was established based on BP neural network theory.Then the network system trained was used to anticipate the effects of different ingredients and their interactions on AA yield.The three-dimensional graphs produced by the network could effectively express the relationships between reaction conditions and catalytic activity.Finally, with the help of the network, the highest AA yield of 27.45% was obtained, which was very close to the experimental result with the relative error of -0.4391%.

Liquid-phase oxidation of toluene by air is an environmentally benign process to produce benzoic acid and benaldehyde.Based on the compositional analysis of the oxidation products and the experimental observations, a reasonable mechanism and reaction network were proposed, and a kinetic model was derived.The model well described the kinetics data which were obtained in a bubble column reactor under conditions close to commercial operations.Using this kinetic model, a commercial process of liquid-phase oxidation of toluene was simulated.The results showed that the reaction in the industrial reactor was limited by the feed amount of oxygen.Measures like increasing air flow rate, increasing both gas and liquid feed rates, or using oxygen-enriched air were proposed to enhance the oxidation reaction.Simulations revealed that such measures could significantly enhance the production.Using an enriched air with an oxygen mole ratio of 31.1%, the toluene conversion reached 22.24%.Increasing air flow by 1.59 times, the toluene conversion increased to 18.61% and the productivity increased to 1.31 times.If increasing both gas and liquid feed rates simultaneously, the productivity increased to 1.71 times.

A series of TiO₂-SiO₂ supported cobalt catalysts were prepared by using the incipient wetness impregnation method, and their activity and selectivity were evaluated for the selective hydrogenation of cinnamaldehyde to cinnamyl alcohol.The effects of Co loading, calcination temperature, reduction temperature and rare earth promoters on the hydrogenation performance of the catalysts were tested.The results showed that the catalytic performance of cobalt catalyst depended on the crystal size of Co on the catalyst surface.The catalyst with bigger Co crystals exhibited higher hydrogenation activity and selectivity of cinnamyl alcohol.The optimal cobalt loading, calcination temperature and reduction temperature of the catalyst for cinnamaldehyde hydrogenation were found to be 15%, 823 K and 823 K, respectively.The addition of rare earth La or Ce enhanced the dispersion of Co on the catalyst surface, and increased the catalytic activity.

In order to utilize energy fully a mathematical model of parallel-feed multi-effect evaporation system for conventional design was established.The model included two sequential system—multi-evaporation system and multi-level preheating system.In this model, the energy-saving measures, such as condensate flash, using extra steam and thick liquor of each effect to preheat raw liquid were adopted.Solid separation was also considered in the process of evaporation.Material balance and energy balance equations of complex parallel-feed multi-effect evaporation system were expressed in matrix equation.By holding or omitting correlated block-matrix, the matrix equation could be used to describe the difference of parallel-feed multi-effect evaporation system with and without solid separation, with and without condensate flash, with and without preheating raw liquid.The model could also be used in the calculation of different effects to realize the universality of the model.The solid-liquid equilibrium data in the multi-effect evaporation system with solid separation was fitted by the intelligent fitting method, which helped to solve this problem by computer.The model was solved by the iterative method and matrix method.The result showed that the model had universality.Using extra steam to preheat raw liquid and condensate flash were effective methods to save energy.The steam consumption of four-effect parallel-feed evaporation with former energy saving methods was decreased by about 30% for evaporation of NaOH solution containing NaCl.The algorithm increased the convergence speed and stability.

The steam power system (SPS) should be operated under the optimal scheme to meet the varying demand for steam and power and to reduce cost and save energy for petrochemical industry.A mixed integer linear program (MILP) model was presented for multi-period operational optimization of utility system.The objective function of the model was to find the minimum annual operational cost of utility systems during the multiple periods, by considering fuel cost, feed-water cost, maintenance cost and startup/shutdown charge.An improved particle swarm optimization(PSO) was employed to obtain the optimum solution of the MILP model.This improved PSO utilized two linearly changing inertia weights to make particles have the tendency to reach global optimum.The application of the proposed optimization approach was illustrated through one case study.

A novel robust PID design method is presented in this paper.This method gets the reasonable PID controller parameters based on min-max principle，which can guarantee the control quality when the model varies in a certain range，decrease the sensitivity to the model uncertainty and increase the robustness of the closed-loop system.Since the min-max problem is not computationally tractable，the authors address this problem by first deriving an upper bound on the robust performance objective and then translating the min-max problem into the minimization problem with the upper bound on the worst-case objective.The paper proves the robust stability of the PID design method.The simulation result of CSTR problem is given to demonstrate the effectiveness of the robust PID design method.

Process monitoring and fault diagnosis is an important problem in chemical processes.Aiming at the real chemical process with its complicated mechanism, nonlinear characteristics and the numerous predictor variables within serious multicollinearity, a novel fault diagnosis model was constructed by combination of isometric feature mapping (ISOMAP) with linear discriminant analysis (LDA).The resulting discriminate model based on this data-driven approach ISOMAP-LDA was divided into two steps. In the first step that performed the ISOMAP manifold learning algorithm was in the original high space for nonlinear dimensionality reduction and feature extraction, and in the second step that built the LDA discriminate model was built in making use of the extracted feature variables.Finally, the proposed ISOMAP-LDA approach was applied to the multiple faults diagnosis of Tennessee Eastman chemical process.The results showed that this method had great power in nonlinear dimensionality reduction and strong generalization ability.At the same time, the ISOMAP-LDA fault diagnosis model was more concise and could be used to observe the structure of the set of samples in the embedding space.

The equations that describe the chemical reaction system are strongly nonlinear.Normally there are many solutions of the nonlinear problem.In this paper an extended homotopy-continuation algorithm was proposed, based on which the steady state multiplicity characteristics of an ideal reacting system was studied.In order to deeply understand the essential characteristics of the reacting system, an investigation of how the steady state solutions changed when one single parameter varied continuously was performed.Finally a spatial image of steady state solutions in the parameter plane was given to show the multiplicity characteristics of the reacting system.

In order to effectively reduce fresh water consumption and parallel processing flow rates of regeneration in a batch water system with multi-contaminant, a superstructure and its optimization design method for a batch water system with multi-contaminant were proposed, which featured parallel processing for continuous regeneration of wastewater with different qualities.Buffer tanks and regeneration units were installed to accomplish separate wastewater processing depending on wastewater quality.A mathematical programming model was established to reduce fresh water consumption and regeneration processing flow rates.An example problem was solved by commercial software GAMS.The computational results showed that the proposed method could deal with the batch water system with wastewater parallel processing in accordance to water quality effectively.Minimal fresh water consumption and regeneration flow rates can be achieved simultaneously.

Static characteristics of internally heat-integrated distillation column (HIDiC) separating binary mixtures of benzene-toluene and of propylene-propane under different compression ratios, feed thermal conditions and distribution modes of heat duty were investigated through simulation studies, and comparative studies against conventional distillation columns (CDIC) and heat pump assisted distillation column (VRC) were also performed to evaluate the energy-saving effect of HIDiC.The simulation results demonstrated that the ideal HIDiC held its superiority in energy efficiency only for the separation of binary close-boiling mixtures.The results obtained showed that the energy-saving effect of VRC was the best for benzene-toluene, with the value of 40% compared with the CDIC, and in the case of propylene-propane system，the HIDiC provided the best energy-saving effect of 60%—80%.The paper proposes a more reasonable heat duty distribution mode called temperature and heat duty matching (THM) for HIDiC.Compared with the mode of averaged heat duty (AH), the heat transfer area in HIDiC can be reduced significantly at a relative small compression ratio using THM.

Ethylene glycol process is a long process, which includes more than ten fractionating columns and fixed-bed catalytic reactor.In the long process, there are many uncertain factors, such as catalyst deactivation and so on.Therefore the simulation of the whole ethylene glycol process is very difficult.In this paper, ASPEN PLUS software was used as a platform.Support vector machine, optimization method, mechanism modeling methods are integrated to develop the model of ethylene glycol process.With actual process data, the process model was intelligently rectified.The results of model simulation agreed with actual data fairly and the whole process model can satisfy the needs of process simulation.

The variances of macro- and micro- mixture fractions were used to represent the performance of macro- and micro-mixing.Multi-scale simulation and study of turbulent mixing in jet reactor were made and the characteristic times of macro- and micro-mixing were obtained under different operation conditions.The results showed that in the case of constant entrained flow velocity, the mixing time to reach complete mixing decreased when the nozzle velocity increased.In the case of constant nozzle velocity, the mixing time to reach complete mixing increased when the entrained flow velocity increased.In the case of constant velocity ratio of nozzle and entrained flow, the mixing time to reach complete mixing decreased when both velocities increased.In all the cases studied, the controlling step of the turbulent mixing in jet reactor was micro-mixing process.

In propane dehydrogenation to propene, the catalyst is continuously deactivated owing to coke formation, which needs to be burnt off periodically to regenerate the catalyst.In this paper, an effectiveness factor modified coke burning-off model was used and a dynamic heterogeneous model for the reactor was established, which took into consideration the pore and film diffusion effects.Simulation study showed that there was a large temperature difference between gas and solid.At the beginning of coke burning-off, because the accumulation of reaction heat, the temperature of solid increased rapidly, which was more obvious near the inlet of the reactor.The temperature quickly decreased to the same temperature with gas when the carbon burning was close to completion.Coke burning-off gradually advanced along the bed, and the hot spot of the bed also gradually advanced from inlet to outlet.The hot spot temperature had little change in different bed position.

A novel support vector data description（SVDD）statistical process monitoring and fault reconstruction & isolation approach is proposed to address non-Gaussian multivariate process systems and overcome the deficiencies of traditional PCA and PLS methods.A robust iterative approach is involved to reconstruct the fault signals and minimize the SVDD monitoring statistics.By comparing the reconstructed statistics with different fault candidates，a fault diagnosis strategy is proposed to isolate the actual process fault.A study of the application in CSTR simulation demonstrates the efficiency of proposed method.

Fault diagnosis is an important technique to ensure chemical processes stability and safety.A fault diagnosis methodology is proposed in this paper for batch chemical processes based on artificial immune system (AIS) and dynamic time warping (DTW) algorithm.Its application to a simulated penicillin fermentation process demonstrated that the proposed AIS could meet the requirement of online dynamic fault diagnosis of batch processes and diagnose new faults through self-learning.

Multi-objective process synthesis can be described as a multi-objective mixed-integer nonlinear programming (MOMINLP) which is difficult to solve.The process model is often based on simplified equations in the past studies, so the calculation results can not be directly used to design a complex process.A three-layer solution strategy for multi-objective synthesis based on multi-objective genetic algorithm (MOGA) in modular simulator Aspen Plus is proposed.The strategy includes simulation layer, synthesis layer and decision layer.Interface is designed to transfer data and information among three layers.The simulation mistakes are treated to avoid influencing the optimization results.A case study showed high suitability of the strategy.The results could reflect the nature of the process and could be used to design the process that meets sustainable development.

In complicated industrial systems, without the first component model and just by use of routine data, factor analysis can model the process operation and take full consideration of general sense of model error, therefore it has a good application prospect.Aimed at the dynamic characteristics in the actual process, the paper proposes a novel method based on dynamic factor analysis (DFA) to model the process data.In DFA, the sample matrix is extended based on the auto-regressive (AR) model, so it can well extract the information of auto-correlation and cross-correlation among process variables.In order to evaluate the operational condition of the process, the paper constructs several statistics as the monitoring indices to measure the features and errors, respectively.The application in the Tennessee-Eastman (TE) process shows the superiority of the DFA-based method.

A new application of catalytic distillation (CD) for the production of ethylene glycol monobutyl ether (EGMBE) from ethylene oxide (EO) and n-butanol was developed.A pilot-scale CD column was built up, which was operated in a semi-batch mode for safety consideration.A dynamic model was developed for the description of the semi-batch CD process.Simulations were performed to investigate the effect of feed flow rate of EO on conversion of reactants and selectivity toward EGMBE, so as to obtain the optimal and safe operating conditions.The simulation provided the typically dynamic profiles of the liquid and vapor flow rates, compositions and temperature, which were compared with the experimental data.The results obtained indicated the model developed was reasonable, and also the potential advantage of using CD for the production of EGMBE.

Advanced process control（APC）technology has been increasingly applied in petro-chemical plants to ensure product quality，increase production yield and save energy.Xylene distillation unit，as one of the most important units in aromatics production，has difficulties in control and has tremendous energy consumption in the reboiler furnace.Especially there is large noise and even steady error sometimes in the measurement of the outlet vaporization fraction of the reboiler furnace，for which the general PID control for the vaporization fraction cannot work.In order to solve the problems in xylene distillation unit，the state space models of xylene column and its reboiler furnace were made by analyzing its mechanism and using practical data.Based on the state feedback predictive control，the advanced process control system was developed and applied in the distillation unit.Using a model from the fuel to the bottom temperature of the column directly without state feedback of practical vaporization fraction，the measurements of vaporization fraction with large noise were ignored from the temperature control strategy.So the operating stability and the separating performance can be improved，and then the energy consumed can be reduced by the optimization of reflux and furnace load.And also，the optimization of the fuel，which means the increase of fuel gas and the decrease of fuel oil，can get more economical benefit.Two application cases show the advantages of the control system.

Electrochemical reduction of 4-chlorophenol (4-CP) in basic solution was investigated by using cyclic voltammetry and in situ FTIR.The results of cyclic voltammograms showed that Ag electrode exhibited good performance for the electro-reduction of 4-CP and the reduction reaction could take place at a more positive potential.The study on possible mechanism of 4-CP on Ag electrode by in situ FTIR spectroscopy suggested that 4-CP was electro-reduced to its radical ion initially, then its radical ion was transformed to phenolate at a more negative potential and electrochemical dechlorination of 4-CP was realized.

In this paper, calcium alginate gel beads were prepared through a homemade microcapsule preparation system with multi-nozzles under forced high-voltage electrostatic field.The effect of operation parameters on the large-scale preparation of gel beads was analyzed.The results showed that the air pressure of feed was the decisive factor affecting the size and yield of gel beads, and the size distribution was dependent on several factors including the distribution of electric field, pulse potential, air pressure of feed, pulse frequency, pulse width and threshold potential.The gel beads with good sphericity, smooth surface and low dispersivity (CV<20%) can be prepared by adjusting and optimizing above parameters.It is more important that the yield of gel beads by the microcapsule preparation system is 20—40 folds higher than the traditional electrostatic droplet generator, which makes it possible to realize the large-scale preparation of gel beads.

A six-parameter model of the open-circuit voltage (OCV) for all-vanadium redox flow battery (VRB) was proposed.This model indicated that OCV was associated with the electromotive force of the cell, VO₂⁺/VO²⁺ electrode potential, and the transport behavior of vanadium ions in electrolyte across the membrane.Two plateaus and a turning point were obtained in OCV curve by this model, indicating that the calculated results were in good agreement with the experimental data.The dependence of concentrations of four kinds of vanadium ions on time was also investigated.The result showed that the electrolyte imbalance was caused by the differences of Donnan balance and diffusion coefficients of different vanadium ions across the membrane.This model can provide theoretical fundamentals for operation process and engineering guidance for electrolyte management of VRB.

Phenolic resin type waste PCB (wPCB) is one of the most important kinds of waste electrical and electronic equipment (WEEE) in China.It is usually hard to recycle for its high content of brominated flame retardants, so the key of recycling the wPCB is to debrominate, i.e., to convert the organic bromides into inorganic ones, e.g., hydrogen bromide (HBr).Hence, pyrolytic debromination experiments were conducted to reveal the effect of blending calcium carbonate (CaCO₃) with the wPCB in pyrolysis.The results of thermogravimetric analysis indicated that CaCO₃ could absorb HBr to restrain them from reacting with organic molecules when the wPCB was blended with CaCO₃ powders in pyrolysis.Further, a pyrolysis experiment blending the wPCB with CaCO₃ in a quartz glass tube reactor was performed by means of silver nitrate titration to determine the content of inorganic bromides.The results showed that 93.4% inorganic bromides were trapped when the wPCB and CaCO₃ blended in the same quantity, and two times of inorganic bromides than without blending CaCO₃ were produced.Moreover, when the wPCB was pyrolyzed with CaCO₃, the contents of organic bromides in the pyrolysates, measured by GC/MS, were 25% less than those when just wPCB was pyrolyzed.It proved that the CaCO₃ helped to debrominate in the pyrolysis of the wPCB.

Quick and effective modeling tools and searching algorithms are the critical issues in realizing on-line combustion optimization in coal-fired utility boilers.In order to explore the applicability of novel modeling tools and optimization algorithms and to overcome the drawbacks of the existing methods, in the present study support vector regression (SYR) model was proposed to capture the functional relation between the NO_x emissions and operational parameters of a utility boiler.A large number of thermal field test samples, which were recorded by DCS in the actual power plants, were employed to establish the models.It was found that the predicted NO_x emissions by SVR showed better agreement with the measured than those by neural networks.SVR model was more suitable to non-linear problem with a large number of samples.Subsequently, a Gaussian probability density distribution (GPDD) based optimization algorithm was described and then applied to searching the optimal inputs of SVR model for NO_x reduction.The results showed that GPDD outperformed the existing GA.Less than one minute of optimization time period required for GPDD was suitable for on-line application.The current work will lay a foundation for the further extension of GPDD’s application to actual power plants.

Pyrolysis characteristics of typical organic components of medical waste were studied by using a fixed bed tubular reactor.The composition and volume fraction of pyrolysis gas were measured by on-line gas analyzers.The effect of pyrolysis temperature was also analyzed.The results indicated that the material characteristics and the final pyrolysis temperature had great effects on the release and volume fraction of pyrolysis gas.The results indicated that the absorbent cotton and bamboo had a lower initial temperature than polypropylene when pyrolyzed.The maximum of gas generation rate and the gas yield increased with the pyrolysis temperature, as well as the proportion of the combustible gas in the total gas products.

The structural evolvement of polyacrylonitrile (PAN) oxidized fiber during the carbonization process and the microstructure of carbon fiber were investigated by the combination of DSC/TG, FT-IR, XRD, elemental analysis, SEM, TEM and HRTEM.The results showed that remnant —C≡N groups continued cyclization reaction in the pre-carbonization process.N hexahydric rings gradually dehydrogenized and denitrogenized.Then they formed C hexahydric rings during the carbonization process.T-700 carbon fiber had high tensile strength due to its smaller d₀₀₂, thicker L_c and higher crystallinity.The microstructure of carbon fiber was composed of 4 parts from skin to core.The high density of sheets packing in the outer skin and the loose structure in the core could be noted.The compact skin of carbon fiber containing lots of nano-crystallites consisted of stacked carbon sheets.The skin-core structure of carbon fiber was evolved from PAN precursor and oxidized fiber.There were strip microstructure and globular microstructure in carbon fiber.The strip microstructure of carbon layers resembled that of PAN molecular chains in precursor fiber.The globular microstructure of carbon layers encircled the center, which was similar to that of the oxidized fiber.It indicated that the microstructure of molecular chains had significant structural interrelationship.

Methyldihydroxyethylbenzyl ammonium chloride（MDBAC）was synthesised with MDEA and benzylchloride as raw materials under the irradiation of constant power microwave.The qualitative analysis of the product was carried out with HLC.The structure of the product was characterized with IR,¹H NMR and ¹³C NMR.Through response surface methodology (RSM), the mathematical relational model between conversion and affecting factors was established, and the microwave synthesis technique of MDBAC was optimized.Under the optimized conditions, the reaction rate was 60 times of the conventional method.The critical micelle concentration (CMC), surface tension, Krafft point and melting point of the product were determined.

A novel apparatus for in situ optical measurement of glass transition temperature （T_g）was developed.The apparatus consisted of a high-pressure cell equipped with quartz windows, a data acquisition system, a heating and temperature-controlling system, a pressure-controlling system, and a CO₂ feeding system.The volume changes of poly-(ethylene-terephthalate) (PET) when it was swollen by CO₂ were measured in real-time by the apparatus, at the temperatures of 50, 60, 70, 80, 90, 100 and 120℃, and under the pressures of 0.1, 5, 10, 15, 20 MPa of CO₂.Consequently, the glass transition temperatures in SC-CO₂ were obtained under various conditions as indicated above.To examine the accuracy and reliability of this method, comparison of the results obtained by the apparatus under atmospheric pressure with those obtained by the standard differential scanning calorimetry (DSC) were made, and these two methods produced results within acceptable precision (1.6%).Thus the method developed in this work can be reliably used to measure the glass transition temperature of polymers in SC-CO₂.

It is of scientific and practical importance，to do experimental research on the process of premixed gas explosion in linked vessels.In this paper，different firing positions，initial pressures and initial densities were investigated to obtain their impacts on gas explosion pressure in linked vessels.The results showed that if fuel was burned in a big vessel，a higher explosion pressure would result in the linked smaller one.The rate of pressure rise was also rapid.The influence of initial density on the linked vessels was basically consistent with that on a single vessel.If the initial pressure increased，explosion pressure in linked vessels increased as well.Furthermore，it increased more quickly in the small vessel than in the large one.Consequently，during the industrial process，the most effective method is the partition of explosion，that is，to install a separation device on the joint between the vessel and the connecting pipe to prevent explosion from spreading through pipes.