Liquid diffusion coefficients are very important basic data for reasearch and many applications in chemical engineering.Because of the complex structure of liquids and the complexity of diffusion process, the research is far from maturity, there has not been a rigorous theoretical equation for the prediction of liquid diffusion coefficients.It is necessary to obtain data by experimental method and theoretical model. A metallic diaphragm cell was prepared and demarcated by measuring the diffusion coefficient of aqueous solutions of KCl at 25 ℃. The reliability of the experimental method was verified by determining the system of sucrosewater at 25 ℃ with the known diffusion coefficient.The diffusion coefficients of aqueous glucose solutions from 298.15 K to 328.15 K were determined by this metallic diaphragm cell.Based on the integral diffusion coefficients, the relationship between differential diffusion coefficients and concentration were correlated.The influence of temperature on diffusion coefficient was also discussed. Based on all these data, a semi-empirical model containing temperature for correlating the diffusion coefficients was proposed. The fitting result of this model was satisfactory.

The equilibrium composition of thiophene hydropyrolysis during steam stimulation was calculated by using the method of multiplier and the concept of minimum Gibbs free energy.The reaction mechanism was analyzed by using laboratory data and equilibrium calculation.The results indicated that hydrolyzation, pyrolysis,hydrodesulfurization, hydrogenation and water gas shift were the main reactions of thiophene hydropyrolysis.When the molar ratio of water to thiophene was less than 4,a higher molar ratio favored both desulfurization and higher gas yield.Only higher gas yield was achieved if the reaction temperature was higher and the pressure was lower.Complete conversion of thiophene could be achieved when the molar ratio of water to thiophene was greater than 4.Temperature,pressure and feed ratio had little effect on the equilibrium composition under this condition.

The resolution of ibuprofen enantiomers was conducted on a chiral stationary phase of Kromasil CHI-TBB column (250 mm×4.6 mm I.D.,5 μm,Eka Chemicals AB, Bohus, Sweden).The thermodynamic properties of the enantiomers in supercritical/subcritical fluid chromatography were investigated in the ranges of 293—323 K and 12.0—21.0 MPa,and compared with those in high performance liquid chromatography in the range of 288—328 K.The mobile phases were isopropanol/carbon dioxide (2∶98, by mass), 690.0 ml•min^-1 (at ambient temperature and pressure), and isopropanol/hexane (3∶97, by volume),2.0 ml•min^-1,respectively.The plots of lnk′ and lnαversus 1/T were straight lines in HPLC.The plots of lnαversus 1/T were still straight lines but the linear relationship of lnk′versus 1/T could not be observed in SFC. The differences of enthalpy changes and entropy changes for the transfer of enantiomers from the mobile phase to the stationary phase were calculated.In both cases,｜Δ_R,SΔH⁰｜＞｜TΔ_R,SΔS⁰｜, therefore,the enthalpic contribution to the overall enantiomer transfer energy is more important than the entropic contribution in the temperature range examined, and furthermore, lower temperature is better for chiral resolution of ibuprofen.

Three-dimensional flow field in the vortex quick separation system of FCC disengager is simulated based on the Differential Stress Model（DSM）provided by CFX5,is studied firstly to improve the separation efficiency.Predictions with the Differential Stress Model are in reasonable agreement with experimental results measured by intelligent five-hole probe, attesting to the suitability of Differential Stress Model for simulating flow field in the vortex quick separator. Both the numerical calculation and the experimental results show that turbulent flow field in the vortex quick separator is symmetric.There is a short-circuit flow near the outlet of the vortex quick separator, which has negative effect on the particle separation and is a key factor to decrease the separation efficiency.The configuration dimensions (S) and the characteristic angle (α) of the vortex head have influences on the tangential velocities in the vortex quick separation system, so they should be optimized. The stripping gas has a little effect on the gas-solid separation efficiency and can be beneficial for discharging the gas.

Two pieces of typical slags specimens were sampled from 0.25 MW test furnace wall to represent fouling and slagging, respectively.Each
mineral element dynamic distribution was analyzed along the slag sample’s thickness direction by using electron probe microanalyzer.Enrichment of iron,sodium and sulphur was the substantial reason leading to forming slag’s initial layer, while enrichment of sodium,silicon,aluminum and sulphur was the most outstanding trait of the fouling sample’s initial layer.Iron and calcium played much more important role in the course of slagging than fouling.Silicon and aluminum showed identical distribution regularities in two individual samples.

Drying of the thin film coated on a particle surface is a basic step for spray-fluidized-bed drying. In this work, the relationship between the vapor pressure and the moisture content of the wet materials is analyzed and a physical model for the drying of the thin film coated on a particle surface is proposed to study the mechanism of spray-fluidized-bed drying.The numerical simulation is in good agreement with the experimental data.In addition, the effects of some particle parameters，such as density，specific heat，diameter，thermal conductivity，on the drying are discussed.The results can be used for the scale-up and optimal design of spray-fluidized-bed drier.

During the thermogravimetric measurement of reaction rates involving mass change in a solid, such as coal char gasification reactivity, the sample is contained in a cup-like crucible, and CO₂ has to access through the top surface of the particle bed.CO₂ mass transfer limitations control their reaction in such situations.Different types of crucibles and coal char initial mass loadings were employed to study the effect of gas mass transfer on coal char gasification rate.The reaction temperature was 1140 ℃,which was in the chemical controlled reaction regime for the coal char gasification.The conclusion indicated that the coal char gasification rate strongly dependent on the crucible type and char initial mass loading.However, the gas mass transfer around and inside a crucible could be simulated by the computational fluid dynamics PHOENICS package.By simulating the space above the bed, it was confirmed that diffusion could supply most of the CO₂ to the bed surface and molecular diffusion governed the mass transfer.Also, a mono-dimensional model was developed.Two parameters were used to indicate whether mass transfer was constraining the reaction either inside or to the bed according to the model.And, the CO₂ concentration calculated by PHOENICS software was in good agreement with what was predicted by the model developed in this work so that the mono-dimensional model could play an important role on studying the gas-solid reaction kinetics when a cup-like crucible was employed during the thermogravimetric measurement.

Chemical reaction takes place when two chemical liquids flow parallelly and diffuse into each other.A solution for chemical reaction kinetics under the convective coordinate was presented to solve convection-diffusion chemical reaction equations. Under this coordinate, these equations became reaction-diffusion equations without convection terms, viz. only reaction and diffusion terms was calculated in each time step. The flux of the liquids was shown in the increase of calculation region as time step increased;and the time to react and diffuse of each particle was equal to the time after they flow into the calculation region, so the time to react and diffuse for every particle in the liquids differed from each other. This method avoided the spurious numerical oscillations in calculation caused by the flow discontinuity in front of the materials, which showed serious unstable influence on solution.Numerical examples demonstrated that this algorithm was effective.

With the help of the description methods in graph theory and the adaptive adjustment method, a steady sate simulation model for fin-and-tube heat exchanger with complex circuit arrangement was developed.This model could provide better compatibility and accuracy for the design of heat exchanger with complex circuit arrangement.Using this model, six typical fin-and-tube heat exchangers with different circuit arrangements were analyzed.The counter flow arrangement could not always achieve the best performance.With appropriate compound arrangement, the performance of heat exchanger could be improved by 2%—4%.

The process of removing dilute CO₂ from air by using the mixtures of K₂CO₃ and piperazine (PZ) was conducted in a random packed tower at 25 ℃.The results showed that PZ increased the absorption rate of CO₂ into aqueous K₂CO₃ much more effectively than MEA or DEA. The volumetric overall mass transfer coefficient (K_Ga) of dilute CO₂ absorption into K₂CO₃/PZ was measured.The K_Ga value was evaluated over the ranges of main operating variables: the concentration of CO₂ in inlet gas, gas flow rate, liquid loading, CO₂ loading in liquid phase,and the concentrations of K₂CO₃ and PZ. The test showed that K_Ga could be remarkably improved by increasing liquid loading and the concentration of PZ, and decreasing the concentration of CO₂ in inlet gas, as well as the gas flow rate and CO₂ loading in liquid phase.

A full-size numerical simulation of an airlift loop reactor with three types of different gas distributors was carried out.The standard κ-ε model and Euler multiphase flow model were used for the simulation of gas-liquid two-phase turbulent flow.All details of gas holdup and flow velocity distribution in the reactor were obtained.It was found that different gas distributor affected greatly global gas holdup and two-phase velocity in the inner tube,and consequently affected greatly gas holdup distribution and gas-liquid contact,which influenced the progress of reaction.The results showed that the contact of gas and liquid was poor with gas distributor of one loop.Gas distributors with multi-loops were recommended in big-diameter loop reactors.The computing results were compared with experimental data on global gas holdup and the agreement was satisfactory.

The reduction of NO with CO over Cu catalysts in the presence of excess oxygen was investigated.The samples were characterized by thermo-gravimetric analysis (TGA), X-ray diffraction (XRD), and temperature-programmed-reduction of H₂(H₂-TPR) and BET.Cu was supported on modified sepiolite by impregnation.The condition of sepiolite modification was chosen by orthogonal experiments.The acid-treatment can improve the special surface area of Cu/sepiolite greatly.The optimized conditions of sepiolite modification were as follows:the concentration of hydrochloric acid was 1.2 mol•L^-1; the solid/liquid volume ratio was 1∶20;the soaking time was 20 h;and the soaking temperature was 333 K.The catalytic performance was studied under oxidizing conditions.The maximum conversion of NO was observed on the Cu/sepiolite loaded with 5%(mass)Cu.It shows a lowest light-off temperature of 518 K and reaches the maximum NO conversion of 90.3% at 653 K.Calcination above 873 K would cause irreversible folding in the structure of sepiolite.When the calcination temperature is about 673 K, the Cu/sepiolite shows a highest activity.The addition of rare-earth elements Ce and Sm can improve the activity of Cu/sepiolite greatly.And this improvement is not only related to the kind of additive but also to the ratio of them.The effect of the oxygen content and the GHSV (gas hourly space velocity) on the catalytic activities was also studied.The suitable working condition of Cu/sepiolite requires that 2.3% oxygen exists in the gas mixture and the space velocity does not exceed 16000 h^-1.Under the same experimental conditions, the catalytic activity of Cu/sepiolite is better than Cu/ZSM-5.

α-Alkyl benzoic alcohol is hydrogenated over Pd/C in the H₂-HAc-HClO₄ system into the corresponding branched-alkyl benzene.The influence of the reaction temperature, hydrogen gas pressure and the amount of catalyst is investigated.Perchloric acid as co-catalyst plays a key role in relation to the catalyst activity.Room temperature (15—30℃), H₂ pressure (400—800 kPa), and 5%—10% Pd/C catalyst by mass are the optimum conditions for this hydrogenation reaction.Perchloric acid as co-catalyst can lead to formation of PdO_xCl_y cluster on the surface of Pd during the hydrogenation reaction.PdO_xCl_y cluster may be a new active center, which enhances the activity of catalyst.

Detailed reaction kinetic studies involving the hydrogenation of NBR were carried out in the presence of hydrogen storage alloys under mild reaction conditions(30—60℃,THF solvent, air atmosphere).Apparent activation energy(48.74 kJ•mol^-1) of NBR hydrogenation reaction was obtained by changing reaction temperature.The lower apparent activation energy suggests that the substrate is activated easily and the reaction is relatively temperature-insensitive during hydrogenation of NBR solution catalyzed by hydrogen storage alloy.Hydrogenation selectivity of double bonds in NBR catalyzed by hydrogen storage alloy was investigated by characterization using IR and ¹HNMR.The results showed that the 1,4-BD units were hydrogenated prior to 1,2-BD units. The kinetic results for the hydrogenation of trans-1,4-BD units and 1,2-BD units in NBR were also introduced.According to the kinetic and other experimental results, a reaction mechanism was presented which could explain most of expetimental results.

Supported Pd-Sn-Ce catalysts for direct synthesis of diphenyl carbonate (DPC) were prepared with three different Ce loading methods that were sol-gel，co-calcination and precipitation methods.The characterization of the catalysts prepared with different Ce loading methods were investigated by XRD and TPR respectively,and the results showed that the main crystal phases in the different catalysts were all La_0.62 Pb_0.38MnO₃.But CeO₂ and Ce_XSn_1-xO₂ were also detected in the catalyst prepared by sol-gel method;CeO₂ and SnO₂ were identified in the catalyst prepared by co-calcination method and CeO₂ was discovered in the catalyst prepared by precipitation method, too.Specific surface areas of catalysts prepared with different Ce loading methods were determined by SEM and specific surface area instrument, the results showed the SBET of catalyst prepared by sol-gel method was larger than that of the others.The synthesis of DPC was carried out and it was found that the activity and life of catalyst prepared by sol-gel was better than that of the others.Furthermore,optimization of synthesis conditions for different catalysts was also carried out respectively.

This paper described the flow,heat and mass transfer, and heterogeneous catalytic reaction in three-way catalytic converter, analyzed the correction of the radial heat conductivity, and compared the diffusion coefficients when the exhaust gas was considered as a binary and multicomponent mixture respectively.A two-dimensional symmetrical transient model applicable to cold-start emission applications was integrated, which was solved by the finite volume method.The model can be used to simulate and explain effects of the flow and temperature distribution on the performance of the converter.Computational results and their analyses were described in the paper Ⅱ.

The effects of flow distributions on the light-off，warm-up performance and the temporal variation of the solid temperature were studied through numerical simulation,and an experiment of the flow distribution was carried out.The predicted velocity distribution agreed with experimental results satisfactorily.The following conclusions were obtained from the predicted light-off curves and the steady conversion efficiency,that the light-off time increases and the total outlet conversion efficiency reduces as the flow uniformity index decreases,and the conversion efficiency decreases as the radius reduces due to the reduction of residual time caused by the increase of the velocity.The analyses on the distribution of temperature in the monolith show that the solid temperatures in area near front face rise as the radius increases, and the radial gradient of solid temperature becomes greater as the flow uniformity index decreases.

To solve dynamic optimization problems of chemical processes with numerical methods, a new algorithm,iterative genetic algorithm (IGA) was developed, of which the main idea was to iteratively call genetic operations and gradually approximate the optimal control profile.The first step of IGA was to discretize time interval and control region to make the continuous dynamic optimization problem a discrete problem.Genetic operations were then used to seek the best control profile of the discrete dynamic system.Lastly,region-reduction strategy and iteration were used to increases numerical accuracy and enhance robustness.IGA was shown by a case study to be convenient, feasible, and efficient.When applied to optimizing feed-rate of Lee-Ramirez bioreactor, IGA displayed advantage over general methods.The result by IGA was better than that in the reference.IGA approach could be regarded as a reliable and useful optimization tool when the gradients are not available.

The multi-objective and layered optimization method is developed from the application background of optimization and control for complex industrial processes.The meaning of multi-objective and layered optimization is explained in detail and the steady-state optimization based on quadratic programming is presented to solve the multi-objective and layered optimization in the context of two-stage model predictive control formulation.Simulation results demonstrate that the steady-state optimization algorithm presented possesses good control performance and practical significance.

The pressure difference fluctuation signals of the air-water two-phase flow in a horizontal pipe were analyzed by statistical and fractal theory.Seven parameters, i.e. the averaged value of the pressure difference, standard deviation, skewness，energy ratio, box dimension，correlation dimension and Hurst index were obtained and be used as the characteristic vector of an improved BP neural network with self-adapted learning ratio. Learning form training samples, the network could accomplish the objective identification of the unknown flow patterns.The simulated results showed that the flow pattern characteristic vector which was obtained by statistical and fractal parameters could reflect the difference between various flow patterns and the recognition possibility of the network could reached up to about 93 percent.Moreover, the improved BP neural network had the merits of fast convergence for simulation and avoidance of local minimum.

The flow patterns of oil/water two-phase flow in vertical upward pipes were characterized by the analysis of symbolic time series based on the conductance fluctuating signals.The study showed that the symbolic sequence temporal irreversibility T_fb and chi-square χ²_fb statistics had little change with oil-in-water flow pattern variations for water cut (K_w) ranging from 61% to 91% but showed irregular sudden changes with transitional flow pattern variations for water cut 51%.When distinguishing the transitional flow pattern from oil-in-water flow pattern, the symbolic time series analysis method presented more unique characteristics and was a useful assistant diagnostic tool for the identification of oil/water two- phase flow patterns.

Brass tube in simulated circlating cooling system was studied by means of electrochemical impedance spectrum(EIS) and electrochemical noise(EN).EIS shows that the inhibitor can react with metal surface or corrosive product and inhibit the corrosion process by forming protective film.With the flow rate increasing, the diffuse speed of oxygen increases and the process of film forming accelerates in HAl77-2 tube，resulting in a decreasing corrosion speed.EN shows that the flow rate has little effect on the susceptibility of pitting corrosion of HSn70-1 tube.But the susceptibility of pitting corrosion of HAl77-2 tube increases with an increasing flow rate.The result is in full agreement with actual conditions.It shows that the electrochemical noise technique is good for corrosion monitoring.The flow rate has a significant effect on the sensor placed countercurrently and has little effect on the sensor placed concurrently.So the sensor should be placed concurrently in application to decrease the effect of flow rate.The result of experiment shows that the sensor can successfully monitor the corrosion in flowing seawater.

Quartz optical fibers are of great significance in many fields. For example,they can be used to fabricate strain sensors and temperature sensors.If the surface of optical fibers is metal-coated by the electroless plating method, the measuring precision and stability of the optical fiber sensors can be improved remarkably, additionally the life-span of sensors can be prolonged.In this work, an electroless nickel-phosphor plating process for optical fibers was presented.The influences of the content of NaH₂PO₂•H₂O, bath pH value and temperature and the plating deposition rate on the quality of the plating coatings were studied with orthogonal testing.The results indicated that the content of NaH₂PO₂•H₂O was the key factor, which influenced the adhesion force of the plating coatings.The nickel-phosphor plating bath presented in this work possessed excellent stability and the plating deposition rate could be accelerated.Using the process described in this work, nickel-phosphor coatings with great brightness and high adhesion force can be prepared successfully.Optimum conditions were: NiSO4 30 g•L^-1, NaH₂PO₂•H₂O 20 g•L^-1, C₃H₆O₃ 4 ml•L^-1, Na₂CO₃ 3 g•L^-1, NH₄Cl ₄ g•L^-1, saccharin 1 mg•L-1, temperature 84 ℃, pH 4.6.

The effect of ionic strength and initial protein concentration on the diffusion of bovine serum albumin (BSA) in an anion exchanger, Q Sepharose FF，was studied by batch adsorption experiments.The pore diffusion model was used to analyze the diffusive mass transfer of BSA in matrix pores.The pore diffusivity was determined by fitting the uptake curves of dynamic adsorption for BSA at different ionic strengths and initial protein concentrations,and the pore diffusion coefficient was obtained.The results showed that the pore diffusivity decreased with increasing initial protein concentration.The pore diffusivity increased with increasing NaCl concentration till 0.10 mol•L^-1.However, it became lower at a salt concentration of 0.15 mol•L^-1.This indicated that there existed an ionic strength where the pore diffusion coefficient exhibited its maximum value.

A novel method for protein purification was presented, which was the integration of organic antisolvent and compressed gas isoelectric precipitation.In this work the study on isoelectric precipitation of BSA in the compressed carbon dioxide-aqueous system was carried out using ethanol as a precipitation assistant and CO₂ as a volatile acid.The precipitation was obvious and it proved that ethanol played two roles in the system.Firstly，ethanol facilitated BSA precipitation at the isoelectric point.Secondly,it reduced the buffer capacity of BSA.This resulted in a decrease of the pressure that was required to reach the isoelectric point so that precipitation could be attained with an increase of maximum BSA concentration that could be acidified to the isoelectric point.Furthermore，the BSA stability in the above processes was investigated, it was shown that there were no irreversible changes in the conformation of BSA, and BSA existed stably in all cases involved.All the experiments indicated that the improvement provided the novel system with an upper hand over the system reported before.The results would be helpful for the development of protein separation and purification techniques.

The gas antisolvent (GAS) process makes use of the ability of gases to dissolve in organic liquids and to lower the solvent power of the liquid for the solutes in solution, thus causing the solids to precipitate.Ethyl-cellulose(EC) is the ramification of cellulose.It is the first time to use the gas antisolvent process to produce EC microparticles,which would be mainly used as a hydrophobic coating material or matrix former for modifying the release of drugs from oral dosage forms.In the experimental conditions, different size particles of EC, with mean diameters ranging from 2 μm to 15 μm, were obtained.Several parameters, such as pressure, temperature and different organic solvents,were investigated.The results showed that the mean particle size increased as the temperature of the particle formation vessel increased;while the particle size and its distribution decreased as the pressure of the particle formation vessel increased.The acetone solution of EC could make smaller particles than the ethanol solution.This work is the first step to make sustained releasing, target, adhesive and drug containing EC micropariticles.

The characteristic of the o-chlorophenol degradation by pulsed high voltage discharge was investigated in present work.The concentration of o-chlorophenol declined faster by a shortening distance, increasing peak voltage or frequency.A degradation efficiency of 98.7% for o-CP is achieved after degradation for 30 min under the conditions of 2 cm of the electrode distance, 30 kV of the peak voltage and 150 Hz of the frequency.It was found that the discharge form was determined not only by the peak voltage and the electrode distance, but also by the salt content in the solution. The increase in salinity could cause a transition from spark discharge to stream discharge and then to corona discharge. The concentration of o-chlorophenol declined faster in the spark and stream discharge than in the corona discharge. The degradation of o-chlorophenol is attacked by hydroxyl radical to form 2-chloro- hydroguinone.The later is oxidized to 2-chloro-benzoquinone.Therefore, the ring is opened to form simple organic acids such as metacetonic acid, oxalic acid, acetic acid and formic acid.

A reaction and diffusion model is developed for the process of solid-state polymerization(SSP) to increase the molecular mass of poly(L-lactic acid)(PLLA).Three reactions are considered: forward and backward polycondensation reactions and thermal degradation reaction.The rate constants of these reactions and diffusivity of water are obtained by fitting the model to experimental data,on the basis of which the activation energies were estimated.Degradation is found to be insignificant at low reaction temperature, while at high reaction temperature it becomes important and cannot be neglected.The diffusivity of water has an important effect on molecular mass growth.Small particle is favorable to obtain a high molecular mass product.

The rare earth ions (RE³⁺:Nd³⁺,Sm³⁺)were loaded into the HMS by the in situ incorporation technique in which RE(NO₃)₃ was used as RE³⁺ precursor.The EPR spectra of synthesised samples were beeline.The distribution of zero region split parameter arose from substituting silicon atomic with rare earth.The EPR spectra of RE-HMS(RE:Nd,Sm) calcined nanoscale samples showed one isotropic line centered at g_iso=2.187,2.163，with peak to peak line width of about 0.03 T and 0.03 T,respectively.The mesoporous solid size of RE-HMS (RE:Nd,Sm) was 31 nm in Nd-HMS and 21 nm in Sm-HMS.The pore size of mesoporous materials RE-HMS (RE:Nd,Sm) from adsorption isotherms was 2.27 nm,2.18 nm,respectively.

The castor oil polyurethane-acrylate hybrid emulsion (PUA) was prepared by the emulsion co-polymerization of castor oil polyurethane dispersion and methyl methacrylate (MMA). The effects of castor oil polyurethane dispersion, MMA content, the initiator and polymerization temperature on the properties of PUA emulsion and its film were investigated. The structure of PUA was characterized by Fourier transform infrared (FTIR). It was found that the PUA emulsion with excellent properties could be obtained by using PU dispersion with opalescent or translucent appearance, and the oil-soluble initiator (AIBN) was better than water-soluble initiator (K₂S₂O₈) in the polymerization system.The mechaniscal properties and water resistance of PUA coating film were improved, while the average particle size, viscosity of PUA emulsions and PUA film gloss decreased as with increasing MMA content.The optimal MMA content was determined as 20%—30% of the total solid content of PUA emulsion.The mechanisms of particle formation and particle size growth of PUA emulsion polymerization were proposed.

China has the richest resource of sodium sulfate in the world, mostly located in its undeveloped western area.Sodium sulfate is also a byproduct of many industrial processes,mostly containing pollutants.The market capacity of sodium sulfate is limited and its economics is poor. Therefore, it is important to develop a new process to utilize sodium sulfate economically.A novel process for manufacturing soda ash and sulfuric acid from sodium sulfate is presented in this paper. It uses calcium oxide and aqueous sodium dichromate solution as a working medium and consists of the following reactions.Reaction of sodium sulfate and calcium oxide in aqueous sodium dichromate solution to get calcium sulfate precipitate and aqueous sodium chromate solution; reduction and decomposition of calcium sulfate to regenerate calcium oxide and get sulfur dioxide; transformation of sulfur dioxide to produce sulfuric acid; carbonation of aqueous sodium chromate solution to regenerate sodium dichromate and get sodium bicarbonate precipitate; removal of chromate from sodium bicarbonate and production of sodium carbonate. The process has the features of low cost, little corrosion and pollution. Hopefully it would become a new way to utilize sodium sulfate.

Polythene film used as confinement plays an important role in flammable cloud explosion.Experiments of C₂H₂/air cloud explosion were performed at 7.8% ethyne concentration.Polythene film with different thicknesses(0.02, 0.04 mm and 0.06 mm) and different layers (1—4) were adopted in the experiments.The results showed that, the maximum overpressure was 2.410 kPa in the explosion of hemisphere cloud with radius 0.5 m, while, as the thickness of film was 0.06 mm,the maximum overpressure reached 3.642 kPa.When 4 layers of film were adopted, the maximum overpressure reached 13.125 kPa.A theoretical model, considering the influences of intensity and layers of polythene film on explosion, was established and solved by SIMPLE algorithm.Compared with the test data, the maximum and average relative deviations of calculation results were 14.8% and 4.73% respectively.The results are of great significance on forecasting gas cloud explosion with weak confinements.