A post-riser system in RFCC unit has a significant influence on product yield and distribution, as well as long term operation of the unit. Quick separators are the core equipment of the system. The fundamental research, development and commercialization of quick separators of post-riser system in China were reviewed. The influence of key geometric configuration and size on the two-phase flow field, separation efficiency and pressure drop were also analyzed and discussed. In order to reduce the contact time of catalyst and oil gas and to discharge oil gas as soon as quickly, a rough cut cyclone was coupled with a pre-stripper. Internals were mounted in the system, which were also called FSC and CSC system, to diminish the fluctuating processing vortex core and to reduce the influence of pre-stripping steam. Then a post-riser system called VQS was proposed, which included a vortex quick separator and an isolated shell. The study was conducted to optimize the geometric configuration of the separator, and the one with long and downward spiral arms was found to have excellent performance. Furthermore, the simulation results showed that severe short cut flow occurred in the vicinity of the exit of arms, leading to significant drop of separation efficiency. Then, a new system called SVQS was proposed by adding an annular cover and a tube into the system. As a result, the separation efficiency considerably increased about 30% for 18 μm talc, meanwhile the separation pressure drop raised only 400 Pa. In order to reduce the separation time, a quick separator was proposed and optimized. The separation time was reduced to less than 1 s, and separation efficiency was closed to 75% for 44 μm talc, which was slightly lower than separation efficiency of cyclone.

As a physical technique, small angle X-ray scatter (SAXS) can be used to study the geometric structure of inhomogeneous electron density zone inside the matter at nanometer level. This review focuses on advantages and features of synchrotron radiation X-ray source SAXS, and the newly progress of its application to characterizations of sol-gel process, synthesis process and structure of zeolite, in particular, the existence behavior of silicoaluminum species in sol-gel system, the influence on synthesis of zeolite and the fractal structure, particle size, mesoporous size, morphology and interface features of zeolite. Finally, the problems existing in SAXS application to zeolite-related researches are discussed and the future research direction is proposed.

Zeolite Y is one of the most important solid acids in the field of heterogeneous catalysis in the past, present and even for a long time in the future. The ultra-stable Y zeolites are the most important form for practical applications. The present review briefly summarized the structure changes of Y zeolite framework under steaming environment, including the removal of framework aluminum, the repair of framework vacancies by migrating silicon insertion, the generation of extra-framework aluminum and the formation of mesopores in the volume of zeolite grains. The emphasis of the present review was on the recent experimental and theoretical progresses about steaming dealumination. The necessity to continue conducting in-depth study on the post-synthesis modification of zeolite Y was stressed.

To improve the gas-solid two-phase mixing and eliminate the effect of secondary flow on the back-mixing near the riser wall in the feedstock injection zone, a novel fluidizer technology is proposed. A protective layer is produced by fluidizer to prevent long-duration contact between catalysts and feed near the wall. Three schemes for this technology (reverse-impact, concurrent-flow and crossed schemes) are used to improve the catalyst-feed mixing and back-mixing near wall by using 3-D CFD simulation. The amount of fluidizer of the optimal scheme is further optimized. The results show that the best is the reverse-impact scheme and the worst is the crossed scheme. The optimal reverse-impact scheme(when the mass flux of fluidizer is equal to the 15% mass flux of feed) can improve the catalyst-feed mixing, prevent the expanding of secondary flow and weaken the intensity of back-mixing at the |r/R|>0.9 near the wall.

The flow, heat transfer and reaction behaviors in the fluid catalytic cracking (FCC) riser before and after the injection of terminator were simulated by the computational fluid dynamics (CFD) method. Distributions of velocity, solid catalyst volume fraction, temperature, and species concentration in FCC riser were obtained after the injection of terminator. Besides, the action zone of the terminator and the effect on cracking reactions were analyzed with different mass fraction and teeming height of the terminator. The simulation results showed that with injection of the terminator, the velocity of oil gas was increased but volume fraction and temperature of the catalyst as well as temperature of the oil gas were decreased. The change in flow and the heat transfer caused the reduction in degree of the cracking reactions and the secondary reaction of oil gas in the riser, which resulted in a higher yield of gasoline and lower yields of dry gas, LPG and coke. The action zone and the effect on cracking reaction varied with different mass fraction and teeming height of the terminator, which should be evaluated by actual process conditions.

The phase change emulsion(PCE) containing phase change material(PCM) of n-tetradecane was used to directly absorb heat of the hydration by the solid-to-liquid phase change of n-tetradecane. Once n-tetradecane in PCE was solidified in the hydrator, the PCE would convert to slurry. The impacts of different pressures from 5.0 to 6.28 MPa on the yield of methane hydrate and hydration rate in the slurry at the temperature 277.6 K were researched at a semi-batch hydrator. Considering the influence of methane dissolution in n-tetradecane on the hydration process, a PVT apparatus to measure the solubility of methane in liquid n-tetradecane was designed, and a more precise calculation process was established to describe the methane hydration. The results showed that:at low temperature conditions, the solubility of methane in liquid n-tetradecane distributed linearly versus pressure, and the methane hydrate formation process was intensified significantly compared to indirect heat removal hydration.

A new method was proposed in a cold riser experimental apparatus with height of about 18 m and inner diameter of 100 mm according to the result measured by PV-6D optical fiber to calculate the local particle flux and velocity based on all the sampling time. The new method was compared with the method used in the referring paper. The results showed that the values of local particle flux and velocity calculated by the two methods had a great discrepancy. The maximum, minimum and mean relative errors between the cross-sectional mean particle flux and the measured value calculated by this paper and the referring paper were 606.9%, 241.3%; 221.4%, 89.5% and 388.9%, 145.6%, respectively. Thus, the value of particle flux measured by this paper was relatively low. The cross-sectional mean particle velocity calculated by the referring paper was higher than the operating gas velocity, and the gas-solid slip velocity and slip coefficient respectively varied from -1.6 to -4.7 m·s^-1 and 0.56 to 0.90, respectively, and thus there was a great difference to the actual gas-solid flow in riser. The cross-sectional mean particle velocity calculated by this paper was lower than the operating gas velocity. The gas-solid slip velocity and slip coefficient varied from 0.6 to 9.6 m·s^-1 and 1.11 to 2.14, respectively. There were some problems when using the reflecting optical fiber to measure the particle concentration and it was the main reason that resulted in the higher particle flux, slip velocity and slip coefficient measured by this paper. Furthermore, two fitting functions were come up with to calculate the particle circulation in the riser according to the measuring result by the optical fiber, substituting for the recent volumetric method.

A CFD-PBM coupling model was developed to simulate turbulent gas-solid flow in FCC riser reactors with consideration of kinetics of granular flow as well as agglomeration and core breakage of particles. Three representative methods of numerical moments, namely, the quadrature method of moments (QMOM), the direct quadrature method of moments (DQMOM), and the fixed pivot quadrature method of moments (FPQMOM), were used to solve PBEs in the model and the simulation results were evaluated to compare the effect of each numerical method. All three MOMs could predict reasonably both radial and axial distributions of time-averaged volume fraction and velocity of particles in FCC riser reactors. Compared to DQMOM and FPQMOM, QMOM yielded calculations most agreeable to the experimental data with regards to time-averaged solid density distribution, solid mass flux distribution, and axial pressure drop. The study suggested that QMOM could be applied to solve PBEs in CFD-PBM coupling model for such simple structured reactors.

The behavior of particle flow in a new type of gas-solid air loop reactor (GSALR), which acted as a particle mixer, was numerically simulated by multi-scale computational fluid dynamics (CFD) with the structure-dependent EMMS drag model. The suitability of the drag model was verified by agreement of the simulated results and the experimental data in time averages of solid holdup and particle velocity. The particle upward velocity increased but the mean solid holdup on beds decreased when the superficial gas velocity increased in the draft tube. Several mixing regions of particle crossflow and mixed flow in GSALR improved the efficiency of radial particle mixing. In the groove region, the distributions of solid holdup and particle velocity were more uniform, while a particle concentrating area was formed at circular overlap due to particle flow from the groove. In the region of 0≤L≤0.058 m and 0< r/R< 0.3, the solid holdup was increased and particle flow was evidently enhanced along the radial direction.

Numerical simulation was adopted to study the chemical stripping process in an annular resid fluid catalytic cracking (RFCC) stripper coupling Eulerian model as well as desorption, thermal cracking and catalytic cracking models. Solid volume fraction distributions and species contents in the gas oil along the stripper were investigated. Furtherly, the effect of adding high temperature activated catalysts in the catalyst inlet on stripping was analyzed. The results showed that the simulation results were in acceptable agreements with the experimental data. The simulated solid volume fraction at the bottom of the stripper was about 0.3-0.45, while in the upper section it showed a sharp decrease to 0.2. From bottom up, the contents of gas and gasoline components increased significantly to about 20% at the top, those of heavy components decreased to nearly 30% and of diesel components about 30% do not change much. With adding high temperature activated catalysts in the catalyst inlet, the contents of gas and gasoline components at the gas outlet increased, while those of diesel and heavy components decreased. When adding 60% of activated catalysts, the content of heavy components decreased to 18%.

The behavior of non-sphere particles in a gas-solid dense phase fluidized bed allocated with three vertical internals was simulated by the Eulerian two-fluid model. The effects of particle shapes were taken into account by an inter-phase drag model, and more effort was devoted to the effects of the boundary conditions of particles at walls. Meanwhile, experiments of a three dimensional lab-scale fluidized bed were conducted to provide available validation data. The quantitative analyses included axial profile of the pressure, radial profile of solids concentration, and power spectral density of particle mass flow rate at the bed outlet. It showed that specular reflection coefficient only affects the macroscopic behavior of the bed slightly, but the local particle motion near walls significantly for the bed without internals. While for the bed with internals featuring a remarkable increase in wall area, specular reflection coefficient can significantly influence dynamics of gas and solids. Therefore, the value of specular reflection coefficient should be selected in an appropriate range for beds with internals, to predict the system reasonably.

The gas-solid flow in gas-solid bubbling fluidized bed was simulated by a combined approach of computational fluid dynamics (CFD) and the two fluid model (TFM), based on the modified EMMS drag model. A new image processing method derived from image calibration was applied to study bubble characteristics with focus on overall and radial distributions of bubbles as well as bubble profiles in terms of averaged equivalent diameter, rising velocity, degree of sphericity, and bubble lifetime at various superficial gas velocity. The results showed that smaller bubbles most located at the bottom of beds and wall region whereas larger bubbles most located at the central region of beds. With increase of superficial gas velocity, the bed height were gradually increased and bubble profiles of the averaged equivalent diameter, the frequency of occurrence, the rising velocity and the lifetime were all increased except that the degree of sphericity was decreased. Nevertheless, once the superficial gas velocity increased to a certain level, its increase would no longer have any significant impact on the rising velocity.

Non-homogeneous gas-solid flow in a circulating fluidized bed (CFB) riser was numerically simulated with CFD models. According to the data and information obtained from the time-averaged CFD flow fields, a novel non-ideal reactor model named equivalent reactor network (ERN) model was developed for the CFB riser. Six parameters for characterizing the structure of a reactor network and the criteria for equivalence checking of the established reactor network were proposed. Systematic analysis on each model parameter and the correlations between them were made elaborately. Values of these parameters were determined subsequently. It was found that the developed ERN model, which was based on the time-averaged CFD flow fields, was capable of describing reasonably non-ideal gas-solid flow behaviors in the CFB riser.

Particle velocity distributions were investigated in draft tube and annulus region of a φ600 mm CSVQS cold-state setup under conditions of gas velocities at 0.2 and 0.3 m·s^-1 in the draft tube, at 0.03 and 0.07 m·s^-1 in the annulus region and 0 and 0.13 m·s^-1 in the stripper. Mean residence time distribution of particles in the pre-tripper section was also calculated by the method proposed here. Experimental results demonstrated that gas flowed as center-airlifting circulation in the pre-stripper section and most of the gas went into the draft tube under the above-stated operating conditions. Under constant gas velocity in the draft tube and increased gas velocity in the annulus region, the radial velocity distribution of particles changed from steep to smooth without stripper wind whereas that distribution changed slightly with stripper wind. Under the same gas velocity in the annulus region and increased velocity in the draft tube, the radial velocity distribution of particles changed from smooth to steep without stripper wind whereas distribution remained quite steep with stripper wind under the two different velocities in the draft tube. Compared to that in the draft tube, the radial velocity distribution of particles in annulus region kept almost no change with varied operating conditions. When the gas velocity was increased in the draft tube or was decreased in the annulus section, the average residence time distribution of particles was narrowed and the mass fraction of particles was decreased. When the circulation strength of particles in the pre-stripper section was increased, the average residence time distribution of particles was narrowed but the mass fraction of particles was changed irregularly.

Three different cyclone separators with diameter of 300 mm, PV-1, PV-2 and PV-3, were designed by changing either the direction of rotation or the diameter of vortex finder. PV-1 differed from PV-2 in the direction of rotation, while PV-3 differed from PV-1 in the diameter of vortex finder. These cyclone separators were assembled center-symmetrically in three paralleled arrangements as assembly of same cyclones (Parallel-Ⅰ), assembly of various rotation (Parallel-Ⅱ) and assembly of various vortex finder (Parallel-Ⅲ). Parallel-Ⅰ was consisted of four PV-1 cell cyclones; Parallel-Ⅱ was consisted of two PV-1 cyclones and two PV-2 cyclones; Parallel-Ⅲ was consisted of two PV-1 cyclones and two PV-3 cyclones. Separation performances of single cyclone and cyclone assemblies were studied in a cold state experimental setup under a condition of solid loading at 5 g·m^-3 and inlet velocities ranging from 14-26 m·s^-1. The flow fields in these paralleled cyclones were simulated with FLUENT software. The results show that paralleled cyclones had higher efficiency than single one with no hump in the curve of efficiency versus inlet velocity. Compared to Parallel-Ⅰ, Parallel-Ⅱ was lower in both total pressure drop and efficiency due to weaker swirl flow. Gas throughput was evenly distributed among each cell cyclone and no cross flow was observed in the common dust bin of either Parallel-I or Parallel-Ⅱ. However, Parallel-Ⅲ had higher total pressure drop than Parallel-Ⅰ. Because Parallel-Ⅲ was composed of cell cyclones with different vortex finder, gas throughput in Parallel-Ⅲ was no longer evenly distributed and an average deviation of about 6.0% was observed between the inlet and outlet of each cell cyclone. A cross flow in the common dust bin was also happened from PV-3 to PV-1, which forced some collected particles back to the inner flow, weakened the stability of vortex flow and decreased separation efficiency. Therefore, the same type of cyclone separators should be assembled center-symmetrically in parallel in order to ensure high separation efficiency of paralleled cyclones.

In order to explore mechanism of concentration and type of inorganic salts on polarization and deformation of water droplet during electric dehydration, water droplets of various salt concentrations and types were studied microscopically in pulsed electric field of both high frequency and high voltage. Water droplets of all tested sodium chloride (NaCl) concentrations showed a trend that the deformation degree increased first but decreased later when the electric frequency and duty ratio were increased. The deformation degree of water droplet increased with increase of NaCl concentration, as a result of increased electric conductivity and ion collision. The electric force on multivalent ions, which is several times stronger than that on monovalent ions, enhanced velocity and momentum of their collision, so the deformation degree of water droplet increased upon increase of ionic valence. Furthermore, the ion activity and hydrolysis affected the deformation degree. In the same high-voltage pulsed electric field, the deformation degree of sodium-containing water droplets changed in the order from high to low as Na₃PO₄> Na₂CO₃> Na₂SO₄> NaCl> NaNO₃ whereas that of chlorine-containing water droplets changed in the order as MgCl₂> CaCl₂> KCl> NaCl> NH₄ClCl. The experimental findings shall facilitate investigating mechanism of electrostaticde-emulsification at high voltage and high frequency.

The matching between oil and catalyst in a downward pointed feed injection scheme was investigated in a large scale model cold riser. Two new parameters were introduced as relative concentration of feed jets and that of catalyst particles, which an oil/catalyst matching index was calculated from the concentration distributions of feed jets and catalyst particles. Experimental results showed that the downward pointed feed injection quickly covered the whole cross section of the riser and significantly shortened the mixing height of oil and catalysts such that the feed oil and the catalyst particles could contact rapidly and uniformly. According to axial distributions of the oil/catalyst matching index, the feed segment under the proposed new scheme of feed injection were divided into three regions, i.e. the initial contact region, the gas-solid diffusion region and the recovery region. A better matching between oil and catalyst could be obtained through a lower nozzle setting angle α, a higher prelift gas velocity U_r and a moderate jet gas velocity U_j. The optimal operating condition was U_r=4.1 m·s^-1, U_j=64.2 m·s^-1 and α=30°. Furthermore, the simulation of axial distribution of the average oil/catalysts matching index were performed by combination with experimental data, which could certainly provide some guidance for the design of downward pointed feed injection scheme.

It does exist that some of abnormal phenomena, such as cavities and particle flow variations from plug flow, are inevitable in cross-flow moving bed operations. In terms of the issues related to the abnormal operation phenomena, the experiment was conducted through a large cold model experimental facility consisting of a φ600 mm×1300 mm semi-conical and semi-cylindrical radial flow moving bed. It showed that, the critical velocity of forming cavity can be effectively increased through introducing coarse additive particles, hence the low elasticity of operation can be improved in cross-flow moving bed to some extent. Both theoretical analysis and experimental results showed that, the inhomogeneity of particle flow can be effectively improved by introducing coarse additive particles.

Gas-solid flow in CFB riser was simulated by a combined approach of computational fluid dynamics (CFD) and discrete element method (DEM). A new cluster analysis method based on image calibration and processing technique was developed to obtain overall, axial and radial distributions of clusters in fluidized beds as well as cluster characteristics in terms of inclination angle, degree of sphericity, and aspect ratio of long over short axis. The clusters showed a wide distribution with mainly small clusters and large clusters formed in the regions near the wall. Along the height direction of the riser, the amount of clusters first increased and then gradually decreased. In addition, clusters were most likely to exist in the form of non-spherical aggregates with large inclination angle and aspect ratio between 2 and 4.

Particle properties have significant effect on gas/solids flow characteristics, which results in significantly different force characteristics exerted in the internals immersed in fluidized beds of different particles. In this study, the dynamic force in a tested slat immersed in fluidized bed was measured by adhering strain gauges on its surface. The force characteristics of the slat were systematically compared in two beds of FCC catalyst particles (Geldart A) and silica sand particles (Geldart B). The experimental results showed that the RMS load density acting on the slat in the bed of Geldart B particles was about 2-3 times higher than in the bed of Geldart A particles. Except installed near the bottom distributor, the measured load density on the slat increased with increasing superficial gas velocity in both beds. However, the effect of the installation height of the slat indicated great difference in the two beds. The measured load density on the slat increased with increasing installation height in the bed with Geldart B particles, while it decreased firstly and then increased with increasing installation height in the Geldart A particles bed. At θ=75°-90°, the measured load densities decreased sharply with increasing inclination angle in both beds. However, at θ=0°-75°, the measured load density decreased slightly with increasing inclination angle in the bed of Geldart B particles, while there was no obvious change in the bed of Geldart A particles.

The microscopic liquid flow characteristics of the structured corrugation foam packing (SCFP) sheets with different combinations of the pore size and the extrusion ratio are observed experimentally in this paper. The tracks of liquid water under different flow rates were recorded by a high-speed camera with a micro lens. A combination of liquid film and quasi-wall flow is verified as the unique liquid flow pattern in SCFP sheets. Besides, the study on the liquid transportation of combined SCFP sheets is carried out. Furthermore, the liquid flow characteristics of SCFP sheets are compared with two classical packing sheets widely used in distillation. The comparison results demonstrate that the unique three-dimensional network structure of SiC skeleton makes SCFP sheets realize larger liquid dispersion degree for single corrugated sheet and better liquid transportation capability for combined corrugated sheets. The study suggests that SiC-foam material has a broad application prospect in the field of structured packing.

To optimize the cylinder height of the closing section, the gas phase flow field and the natural vortex length in a φ600 mm×4150 mm super vortex quick separation (SVQS) system were simulated by using the RNG k-ε turbulent model. The simulation results show that the cross-section location of the vortex end was agreeable to that of 88% attenuation of the maximum tangential velocity in the entrance section of flow partition column, which hence defined the natural vortex length. From the tangential velocity distribution of gas phase in SVQS, the natural vortex length of the quick separation was found to increase with the increase of spout velocity and the decrease of stripping velocity. Based on the numerical simulation and structural characteristics of SVQS, an equation for calculating the natural vortex length in SVQS was developed.

ZSM-5 zeolite, one of the important shape selective catalysts with high molar ratio of silica to alumina, possesses strong acidity and excellent thermal and hydrothermal stability. The acidity and pore structure of ZSM-5 zeolite were adjusted by combinatorially modifying ZSM-5 molecular sieves with phosphorus and aluminum. Results showed lamellar-structured phosphorus and aluminum deposits on the surface of ZSM-5 zeolite at greater than 10% (mass) load of phosphorus and aluminum, which increased its acidity and formed new mesopores about 10 nm in diameter. Such highly acidic porous structures significantly increased the conversion of n-octane and the propylene yield by more than 20%. When used in catalytic cracking of heavy oil, the yield of oil residual and coke was decreased but the yield of propylene was increased by 1%.

Pt/TiO₂/ZSM-5 catalyst with dehydrogenation-cracking bifunctionality catalytic activity was prepared by modifying ZSM-5 zeolite with titanium dioxide (TiO₂) via sol-gel method and then loading Pt to the titanium modified ZSM-5 zeolite by incipient impregnation method. The as-prepared catalyst was characterized by means of XRD, N₂ adsorption-desorption, TEM, XPS and NH₃-TPD to analyze crystal structure, pore properties, morphology, valence states of active metal and acid properties of the catalyst, and the catalytic performance for the cracking of n-butane into light olefins was investigated. The results showed that the introduction of titanium dioxide provided additional acid sites to ZSM-5 zeolite, especially increasing the strong acid centers and enhancing the activation of n-butane. In addition, after reduction in hydrogen atmosphere, the partially reduced Ti³⁺ species was generated which was catalyzed by platinum due to the strong metal-support interaction (SMSI) between Pt and TiO₂. The formation of appropriate amount of Ti³⁺ species enhanced the electron density around Pt, and thus weakened the adsorption of ethene and propene on Pt atoms. After reduction by hydrogen at 450℃,n-butane conversion of 76.1% and yield of light olefins (C₂⁼-C₃⁼) of 50.9% were achieved at the reaction temperature of 625℃ over Pt/10TiO₂/ZSM-5 catalyst, which was 16.7% and 12.6% higher than those of Pt/ZSM-5 catalyst, respectively.

ZSM-5 zeolites with similar crystal size and Si/Al ratio were synthesized in the presence of n-butylamine (NBA), tetrapropyl ammonium (TPA⁺) and no template (NT), respectively. The samples were characterized by XRD, TEM, NH₃-TPD, N₂ sorption analysis and applied to n-hexane cracking to investigate the difference of their acid characters, cracking selectivity and lifetime. The results showed that the samples had different acid properties and acid distribution in spite of similar Si/Al. Under the same initial conversion, the samples exhibited different product distribution. ZSM-5 using NBA and TPA⁺ showed higher selectivity of ethylene and propylene than ZSM-5 without template. However, the latter had much longer stable period than the former two and its conversion could stay above 96% for 1200 min.

Mass loss behavior in air and solid-state reaction mechanism of ammonium molybdate with LaHY were investigated by TG-DSC technique, and the physical structure, specific surface area, and surface acid sites by XRD, BET and NH₃-TPD techniques. It showed that surface speicies of Mo arising from decomposition of (NH₄)₆Mo₇O₂₄·4H₂O, over solid-state reaction of (NH₄)₆Mo₇O₂₄·4H₂O with LaHY, allocates in LaHY molecular sieve cage as molybdenum-oxygen clusters to form a single-phase complex nMoO_x·LaHY. The single-phase complex contributes to the lattice shrinkage, the reduced lattice parameter a₀, and the decrease in specific surface area. The as-prepared nMoO_x·LaHY possesses a slightly variation of weak acid center sites, an increase in medium strong acid center sites, a slightly decrease in strong acid center sites, and an increase in total acid sites, in comaprision with LaHY. The hydrodesulfulrization (HDS) performance of the single-phase complex nMoO_x·LaHY was evaluated by using 0.6% (mass) dibenzothiophene(DBT)/decane as the representative reactants. The conversion of DBT in HDS process at 290℃ and 310℃ over nMoO_x·LaHY catalyst with 5.0% (mass) of Mo loading, reached 56.38% and 88.79% under conditions of 4.0 MPa, 20 h^-1 of space velocity and 500:1 of H₂/oil volumetric ratio, which increased 12 and 28 percentage than that of MoO₃/Al₂O₃ with 20% (mass) of Mo loading, respectively. The nMoO_x·LaHY catalyst exhibited higher activity for DBT HDS reaction.

The cracking performance of light hydrocarbon model compounds such as n-paraffins, i-paraffins, naphthalenes, aromatics and n-olefins on a mesoporous catalyst based on ZSM-5 molecular sieve was investigated in a small scale fixed bed reactor. The experimental results showed that the cracking performance in the order from good to poor was:n-olefins, n-paraffins, naphthalenes, i-paraffins, and aromatics. For n-paraffins, i-paraffins and naphthalenes, the yields of total light olefins were largely different whereas the selectivities of total light olefins were all around 56.57%. The main liquid components were pentene, benzene, toluene and xylene besides unreacted feed. In case of straight-run naphtha, the yield of ethylene plus propylene and the yield of total light olefins increased with the increase in reaction temperature and the decrease in weight hourly space velocity. At a condition of reaction temperature of 680℃, a weight hourly space velocity of 4.32 h^-1 and a steam-to-oil weight ratio of 0.35, naphtha catalytic pyrolysis yielded 41.94% (mass) of total light olefins and 35.87% (mass) of ethylene plus propylene, which the ratio of propylene to ethylene at above 1.2 was much higher than that of naphtha steam cracking. A large amount of olefins and paraffins in the liquid sample of naphtha catalytic pyrolysis indicated that the liquid sample still had good cracking performance. A new characterization parameter (KF), a function of H/C atomic ratio, relative density and molecular weight of feeds, was proposed to be more suitable than KCP for characterizing catalytic pyrolysis of light hydrocarbons.

Y zeolite, the main component responsible for the catalytic activity, selectivity, and stability of catalysts in fluid catalytic cracking (FCC), has long been used extensively in oil refining. The acidity and pore structure of Y zeolite determine its catalytic performance. To tune acidic property, post-synthesis treatment of NaY was carried out with two different modifying media, i.e., the calcination atmosphere of CeY and the ammonium ion exchange with different pH value. The structures and physical properties of the modified Y zeolites were characterized by IR spectroscopy, X-ray photoelectron spectroscopy (XPS) and low temperature N₂ physical adsorption; while their catalytic performance was evaluated on VGO oil in a fluidized bed reaction ACE unit. When Ce ion was chosen as the modifying medium, the distribution of various types of acids (Brønsted/Lewis acid) and the catalytic performance of USY zeolites could be controlled by changing calcination atmosphere. When cationic ammonium was used, the amount of Brønsted acid could be regulated and pore volume could be increased in the ultra stable Y zeolites by precisely adjusting pH value of the first ammonium ion exchange medium.

The Cr-doped ZSM-5 zeolites were synthesized by in-situ hydrothermal method and were used in studying Cr effect on catalytic cracking of n-butane to produce light olefins. Zeolite structure, chemical state of Cr species and acidity of zeolites were characterized by several techniques, including X-ray diffraction, nitrogen adsorption and desorption, ultraviolet visible diffuse reflectance spectroscopy, Raman spectroscopy, ammonia temperature-programmed desorption and pyrrolidine (Py)-FTIR. Cr species were found to partially incorporate in the ZSM-5 framework and present in the form of monodispersion. No significant difference in the weak acidity of zeolites at low Cr content. However, a small amount of Cr species in zeolites could significantly enhance conversion of n-butane and improve yield of ethylene and propylene. When the molar ratio of Cr/Al was 0.04, the conversion of n-butane was 99.6% and the yield of ethylene and propylene was 53.8% at reaction temperature of 650℃, which was 15.5% and 5.0% higher than those of unmodified ZSM-5, respectively. The superior catalytic performance could be attributed to synergistic effects of dehydrogenation of active Cr species and cracking function of the acidic zeolite, which promoted highly efficient conversion of n-butane.

A series of La-modified ZSM-5 were prepared by hydrothermal treatment and incipient wetness impregnation. Compared with those of pristine ZSM-5, an increase in lattice constants and the amount of strong, weak and the total acid cites of La-modified ZSM-5 by the hydrothermal treatment occurs. The amount of acid cites of by La-modified ZSM-5 by incipient wetness impregnation increases, however, the lattice constants does not change. High yields of ethylene and propylene were obtained on both kinds of modified samples tested in the catalytic cracking of hexane. The yields of ethylene and propylene were achieved on the samples modified by hydrothermal treatment at WHSV 6 h^-1 (23.39% of ethylene, 25.17% of propylene), which are 2% and 4% larger than those of the pristine ZSM-5 samples (21.19% of ethylene, 21.04% of propylene), and slightly better than those of the samples modified by incipient wetness impregnation. In the long-term experiment of hexane cracking with at WHSV 4 h^-1, the samples modified by hydrothermal treatment (2000 min) and incipient wetness impregnation (1600 min), exhibited much longer durability than the pristine ZSM-5 sample (800 min). It suggested that the hydrothermal treatment is better modification method in the preparation of La-modified ZSM-5 catalyst to improve the performance of hexane catalytic cracking.

β zeolites synthesized by various methods were used to study the FCC (fluid catalytic cracking) reactions and to explore the impact of silicon/aluminum ratio in β zeolites on FCC product distributions. Compared to high silicon-containing ZSM-5 zeolites, β zeolites in FCC increased both octane number and yield of gasoline. Also, β zeolite showed a stronger functionality of isomerization with enhanced yields of iso-butane and olefin isomers (or tert-olefins)in light gasoline, which enriched raw material resources for alkylation and etherification processes.

The cracking performance of ethylcyclohexane on a mesoporous catalyst based on ZSM-5 molecular sieve was investigated in an apparatus with a fixed-bed reactor. The results showed that ethylcyclohexane had good cracking performance, the conversion was above 80%, the yield of ethylene and propylene reached 41% (mass), the yield of total light olefins was near 50% (mass), and the main components of liquid products were benzene, toluene and xylene. As reaction temperature increased, the yield of ethylene increased monotonously, and those of propylene and butene showed maxima. With the consideration of both thermal and catalytic reactions, a kinetic model involving 14 reactions was established for the catalytic pyrolysis of ethylcyclohexane. The experimental data were obtained at reaction temperatures of 600, 620, 640 and 660℃, and the reaction rate constants were estimated by genetic algorithm regression analysis. Apparent activation energies and pre-exponential factors were subsequently calculated according to the Arrhenius equation. All of the apparent activation energies obtained were below 90 kJ·mol^-1, indicating that the mesoporous ZSM-5 catalyst provides high catalytic activity for ethylcyclohexane catalytic pyrolysis. The model predicted values were in good agreement with the experimental data, and the average relative errors of the main products were below 10%.

β-Zeolite, an important acidic catalyst, its structure and acidity can significantly influence the catalytic activity. Through two combinational treatments of acid-calcination and acid-steaming, the structure and acidity of β-zeolite were tuned. After treatment, the modified β-zeolites was characterized by N2 adsorption and desorption, X-ray photoelectron spectroscopy (XPS), and infrared spectra (IR). The activity for catalytic cracking of n-octane at 500℃ was evaluated. It showed that BET specific area and Lewis acid sites (L₁₄₄₅ and L₁₄₅₅) were tuned by different treatments. The catalyst by acid-calcination treatment exhibited better activity for catalytic cracking of n-octane than acid-steaming treatment. The higher BET surface area and the Lewis acid sites account for the higher catalytic cracking activity of the catalyst by acid-calcination treatment. The concentrations of Brønsted acid and two types of Lewis acid as well as textural properties were tuned, which lead to the difference in catalytic activity between the original and final β-zeolites.

ZSM-5/EU-1 co-crystalline zeolites were designed by two-stage synthesis technique with surfactants of diquaternary ammonium and hexamethonium bromide, that both surfactants displayed similar structure-directing effect. By adjusting pre-crystallization process, crystal coexistence of two zeolites was achieved and various ZSM-5/EU-1 co-crystalline zeolites were prepared with controllable phase proportions. Study on gel composition at each stage showed that gel composition at the first stage, especially the template concentration, had pronounced effect on phase proportions for the resulting ZSM-5/EU-1 co-crystalline zeolites. Compared to the mechanical mixture and pure-phase counterpart, ZSM-5/EU-1 co-crystalline zeolites had the largest BET surface area and the highest pore volume, especially for external surface area and mesopore volume. In catalytic cracking n-hexane reaction, ZSM-5/EU-1 co-crystalline zeolite with 39% EU-1 phase content exhibited the most excellent stability of reaction activity and the highest selectivity to light olefins, in particular propylene. At reaction time of 120 min, ZSM-5/EU-1 of 39% EU-1 had the selectivity of light olefins 10.7 percentage points higher than the mechanical mixture of 39% EU-1 and pure ZSM-5 zeolite, the propylene selectivity 8.5 and 9.1 percentage points as well as the P/E ratio 0.42 and 0.46 higher than the mechanical mixture of 39% EU-1 and pure ZSM-5 zeolite, respectively.

Low-carbon olefin separation is important in petrochemical industry. Separation of methane/ethylene mixture was studied by hydrate formation in water-in-oil (w/o) emulsion during isochoric and isothermal processes in an agitated reactor. With pressure-volume-temperature (PVT) method, the ethylene hydrate formation kinetics and separation efficiency were evaluated at various operating conditions of temperature and emulsion consumption. Ethylene concentration in gas mixture decreased from 34% to 9.97% and its recovery achieved to 90.36% with the use of 100mL emulsion at temperature of 273.35 K. Lower temperature and more emulsion were favorable for the recovery of ethylene. A kinetics-controlled model using fugacity difference as the driving force was proposed, which fitted the experiment data well and could predict experimental output accurately from operating parameters.

The performance of reactor filter in the adsorptive desulphurization unit of catalytic gasoline (S-Zorb) will pose direct impacts on the separation and recycling of adsorbent. By utilizing high frequency sensors and experimental set-up of filtration performance during the process of pulse cleaning, the dynamic pressure characteristics varied with time in different measurement points along filter candles were investigated. Also, the effects of the length of filter candle, cleaning pressure and pulse width on dynamic pressure characteristics were analyzed. As shown by the results, the maximum pressure in the filter candle remained stable after gradually increasing from the open end to the blind end. Meanwhile, lower maximum pressure in different measurement points along the whole filter candle was found under the same operating conditions when the longer filter candle was utilized. In these experiments, the dynamic pressure near the open end of 1750-2000 mm filter candles reduced to negative pressure after reaching the maximum pressure. Due to the combination of airflow and pulse pressure wave, the sustained oscillations in the dynamic pressure profile can be discovered. Furthermore, the cleaning efficiency of long filter candle can be improved through the method of properly increasing cleaning pressure and pulse width.

Structured packing has been widely used in distillation processes. An economic feasibility analysis of a new structured packing, the structured corrugation SiC-foam packing (SCFP-SiC), was compared to traditional structured corrugation metal packing (SCMP) in the whole life cycle from raw materials, production to end use by life cycle cost (LCC) theory. Results showed that the production cost of SCFP-SiC was higher than that of SCMP but was less affected by price fluctuation of raw materials. In end use, SCFP-SiC could have overall cost saving advantages by whether decreasing the tower height to reduce the fixed cost or decreasing the reflux ratio to reduce the variable cost. Assuming that SCFP-SiC could increase 1.5 times more theoretical plate number, the total cost would be lower for SCFP-SiC than SCMP in only 6.8 months' separation operation of xylene mixture. Considered its remarkable corrosion resistance, SCFP-SiC could significantly benefit end users from economic viewpoint as a SCMP replacement in separation processes.

The characteristics of slip velocity between gas and solid in a short-contact cyclone FCC reactor was simulated using the Eulerian-Eulerian two fluid model, especially about radial distribution of the tangential slip velocity between gas and solid as well as its dependence on operational and physical parameters in the separation chamber. Simulation results showed that the tangential slip velocity between gas and solid had a two-humped distribution in the radial direction. With increase of the inlet gas velocity or decrease of the ratio of catalyst to gas, the tangential slip velocity between gas and solid weakened but the tangential velocity of solid particles enhanced, which would strengthen the centrifugal force on particles and improve the gas-solid separation efficiency. The particle density exhibited little influence on radial distribution of the tangential slip velocity between gas and solid. However, when the particle diameters were increased, the dust outlet would easily be clogged which was not conducive to running the reactor for a long cycle of time. Finally, a model of the average cross-section tangential slip velocity between gas and solid was established with no more than ±7.0% difference among model simulations and calculated results.

Recovery of ethylene, an important raw chemical material, from refinery gas can effectively reduce production cost and improve profit of refineries. A new hybrid hydration and absorption process for recycling ethylene from refinery gas was designed and optimized by using Aspen Custom Modeler. Conditions at various efficiencies of ethylene recovery were optimized through operational parameters, such as oil flow rate, theoretical plate numbers, water content and temperature.

By screening the auxiliary components and preparation methods, a Fe₃O₄ composite oxide catalyst was prepared for production of hydrogen from methane. Based on the properties of neural network, an improved back-propagation network model was developed to simulate the relationship between components of the catalyst and catalytic performances of the catalyst life and the formation rate of hydrogen. The model network structure and the computer-aided design procedures were established after investigation of the structural organization, the training method, the activation function, and the generalization ability of artificial neural network. Upon used the Levenberg-Marquardt training method, the network convergence was improved significantly and a formulation model of artificial neural network was achieved with strong generalization capability. To further enhance efficiency of catalyst design, a hybrid genetic algorithm was employed for global optimization of design. After six cycles of design optimization, a series formulation of Fe₃O₄ composite oxide catalysts for production of hydrogen from methane was developed. When one of the optimized catalysts was applied in hydrogen production, the catalyst life and formation rate of hydrogen were 4.46 h and 1.16 mmol·min^-1·(g Fe)^-1, respectively, which were better than those of previously reported catalysts.

Considered economic requirement for maximizing propylene yield of fluidized catalytic pyrolysis process in two-stage riser, as well as complex characteristics of the FCC process such as strong nonlinearity, coupling multivariable and unavailability in online measurement of product yields, a self-optimizing control strategy on a basis of multi-objective optimization was proposed. First, a multi-objective optimization framework for maximizing propylene yield while minimizing dry gas output was created from steady state model and operational constraints of the process, and solved for optimal operation condition with a complete and uniform Pareto distribution by standardized normal constraint method. Secondly, a self-optimizing scheme of cascade controls was generated from relationships between the optimal operation condition and the active constraints. Product yield that could not be measured online were estimated by an unscented Kalman filter transformation. Compared to the tracking control on temperature setpoints at the riser outlet, the self-optimizing control method could spontaneously adjust operating condition under circumstances of interference and reduce the disadvantageous impact of noise factors to optimizing process operation.

The two-stage-riser catalytic pyrolysis technology of heavy oil for maximizing propylene, features the possibility of feedstock diversification, high-yield of propylene and high-value gasoline, which enables a bright future for the related industrial applications. Modeling, nonlinear analysis, and control are important aspects for achieving high mass/energy conversion and therefore improving the competitiveness of the TMP technology. This work concentrates on the dynamic modeling for process control purpose. First, mass and energy balance models for the first riser, the second riser, the stripper, the combustor, and the dense region of the regenerator are established and connected by the spent and regenerated catalyst flow among them, forming an integrated dynamic model of the TMP process. It showed that numerical simulation of the main process variables to perturbations in potential manipulated variables not only validates the effectiveness of the set model but also presents strong interactions among two risers and the regenerator system. In total, the built model is well-suited to be adopted as a reference model for subsequent nonlinear analysis and model-based advanced control investigation.

Macroporous vesicle material with more than 100 nm diameter was synthesized via hydrothermal method using poly(ethylene glycol)-block-poly(propylene glycol)-block-poly(ethylene glycol) (P123) as template and tetramethyl orthosilicate (TMOS) as silica source. Moreover, the radical initiators with silane coupling groups such as (2-bromo-2-methylpropionyl)oxy-butyl triethoxysilane (BPE) and (2-bromo-2-methylpropionyl) oxy-butyl chlorodimethylsilane (BPCM) were synthesized by a two-step method. Then BPE and BPCM were grafted onto the inner surface of vesicle via coupling reaction of silane group and Si-OH. The elemental analysis confirmed that BPCM gave higher grafting efficiency. After that the atom transfer radical polymerizations (ATRP) of methyl methacrylate (MMA) and butyl methacrylate (BMA) were initiated in the pores of vesicle, respectively, and the monomers gradually reacted on surface-initiators to form polymer brushes. Transmission electron microscopy (TEM) results suggested that the vesicles retained after grafting of polymers into the inner surface of material successfully. The grafted polymers were cleaved by chemical process and analyzed by gel permeation chromatography (GPC). The number average molecular weights (M_n) of PMMA and PBMA were 7600 and 11400, respectively, the polydispersity indexes (PDI) of PMMA and PBMA were both 1.25.

The alumina sol with high content of Al₂O₃ was prepared by acidolysis-peptizing with aluminum as an additive. Its physical and chemical properties were characterized by pH meter, rheometer, digital viscometer and zeta potential analyzer. Catalytic performance of catalyst that was prepared with alumina sol was evaluated through a fixed-bed. The results showed that the alumina content can reach 20% (mass) of the alumina sol sample with aluminum powder pH of 2.3. The alumina sol had good stability and chlorine was not higher than 15% (mass), which can meet the FCC cracking catalyst requirements.