Search

Spout diameter has been measured in shallow spouted beds by means of an optical fiber probe, for different geometric factors of the contactor (base angle and gas inlet diameter) and under different experimental conditions (stagnant bed height, gas velocity, particle density, particle size, and particle shape). The study has been performed with glass beads and low-density solids (high and low-density polyethylene, polypropylene, and extruded and expanded polystyrene). A correlation that is valid to predict the average spout diameter in cylindrical spouted beds for beds composed of solids of different density, size, and shape has been developed that is in agreement with the experimental data. This equation is interesting for the proposal of real models for gas−solid flow.

The transformation of methanol into olefins over SAPO-18 catalyst in the 250−450 °C range has been studied. The process has been analyzed by following the evolution with time on stream of the mass deposited on the catalyst, of the heat flow evolved, and of the product formation, which are analyzed by means of on-line mass spectroscopy. The results are in agreement with the “hydrocarbon pool” mechanism, and show that the process is conditioned by an initiation step in which the reactive intermediates for the production of olefins are generated. These intermediates are trapped in the SAPO-18 cages. As time on stream is increased, the steps of initiation, maximum production of olefins, and deactivation develop, and the duration of these steps depends on reaction conditions. Water content in the feed has an important attenuating effect on the methanol adsorption and delays the generation of the active intermediates, and also attenuates the subsequent degradation of the intermediates to coke. This attenuation of coke formation is of great importance for reducing catalyst deactivation.

The kinetics of the pyrolysis of tire material has been studied in a conical spouted bed reactor (CSBR), in the range between 450 and 600 °C, for two particle size ranges (between 0.1 and 0.8 mm and between 0.8 and 4.0 mm). The kinetic results have been compared with those obtained for the same material with other technologies, namely, thermogravimetry and a fast heating microreactor. This comparison and the one carried out with the literature results for similar materials have permitted the analysis of the effect of the technology and of the experimental conditions of pyrolysis (especially particle size) on the kinetic results. It has been proven that the CSBR is especially suitable for the kinetic study of these materials, as it allows for obtaining an excellent gas−solid contact, which minimizes the limitations of heat and mass transfer and maintains bed isothermallity at high temperatures, within the temperature ranges and heating rates of interest for scaling-up. In the pyrolysis of small particles (between 0.1 and 0.8 mm), the apparent activation energy is 121.3 kJ mol^-1 and the corresponding frequency factor is 1.0 × 10⁶ s^-1. The effect of particle size is significant and shows the impossibility of totally avoiding the limitations of heat and mass transfer in the pyrolysis of these materials.

The effect of the geometric factors of the contactor and of the operating conditions on the geometry of the spout and fountain has been studied by means of an optical fiber probe. In the fountain, the core or solid ascending zone and the periphery or solid descending zone have been delimited. The materials studied are glass beads and different plastics (polyethylene, polypropylene, and polystyrene, both extruded and expanded). A correlation proposed for the calculation of the average spout diameter in cylindrical spouted beds of a conical base has been proven to be valid for these materials of low density. New correlations are proposed for the calculation of the height and width of the fountain.

The combined effect of both the water content in the reaction medium and the reaction temperature (in the 400−500 °C range) on the acidity deterioration of a HZSM-5 zeolite catalyst in the methanol to olefins process has been studied. Accordingly, the catalyst has been used in reaction-regeneration cycles in an isothermal fixed-bed reactor. The nature of the acid sites has been measured by means of Fourier transform infrared of the pyridine adsorbed. Both the total acidity and the acid strength distribution of the sites have been determined by calorimetric measurements of differential scanning of adsorption of ammonia and temperature-programmed desorption of this base. The results are proof of the irreversible loss of acidity at 450 °C for a water content in the feed above 50 wt %. Irreversible deactivation occurs at 500 °C from low values of conversion, even when pure methanol is fed.

A study has been made of the effect of both the geometric factors of the contactor (base angle and gas inlet diameter) and the experimental conditions (stagnant bed height, particle diameter, and gas velocity) on the trajectories of the particles in the three zones (spout, annulus, and interface) of spouted beds. The trajectories have been determined from the experimental results of the vertical component of the particle velocity measured by means of an optical fiber probe. Cycle time distributions have been calculated from the trajectories, and they have been compared with the experimental results obtained by monitoring colored particles using video-imaging treatment. Original correlations have been proposed for calculating the average cycle time from the geometric factors and experimental conditions.

The catalytic transformation over a HZSM-5 zeolite of key components of the liquid product obtained by the flash pyrolysis of biomass, namely, acetaldehyde, ketones (acetone and butanone), and acetic acid, has been studied, and great differences in reactivity and degradation to coke have been found. Acetaldehyde has a low reactivity to hydrocarbons, with a noticeable deactivation caused by coke deposition, which can be attributed to its capacity for oligomerization with the trimer trimethyltrioxane obtained as an intermediate product. The transformation of ketones [less reactive than the alcohols studied in part I of this work (see the preceding paper in this issue)] and of acetic acid (which gives rise to acetone as the primary product) mainly occurs through decarboxylation and, to a lesser degree, dehydration. Above 400 °C, this transformation gives olefins and aromatics according to a reaction scheme similar to that better known for the reaction of alcohols. The generation of coke (attenuated by the presence of water in the reaction medium) is more significant than in the corresponding process for alcohols, and it limits the formation of aromatics and increases the formation of olefins (intermediate products of the reaction scheme).

The effects of temperature and space time on the transformation over a HZSM-5 zeolite catalyst of several model components of the liquid product obtained by the flash pyrolysis of vegetable biomass (1-propanol, 2-propanol, 1-butanol, 2-butanol, phenol, and 2-methoxyphenol) have been studied. The transformation of alcohols follows a route similar to that of methanol and ethanol toward the formation of hydrocarbon constituents of the lumps of gasoline and light olefins. Phenol and 2-methoxyphenol have a low reactivities to hydrocarbons, and the deposition of coke of thermal origin caused by the condensation of 2-methoxyphenol is noticeable. The generation of catalytic coke and the deactivation by this cause attenuate as the space time and water content in the feed are increased. To avoid the irreversible deactivation of the HZSM-5 zeolite, operations must be carried out at a temperature below 400 °C. Above this temperature, the increase in product aromaticity is also significant.

The effect of the operating conditions (base angle, air inlet diameter, stagnant bed height, particle diameter, and air velocity) on the fountain geometry (height and amplitude) has been studied. Likewise, particle ascending and descending zones in the fountain have been delimited. These studies have been carried out by means of an optical fiber and by an image treatment system. An equation has been proposed for the calculation of the fountain height from the vertical component of the particle velocity in the axis at the base of the fountain. The contactor base angle has a major incidence on the fountain geometry. The external diameter of the fountain and that of the core are conditioned by the diameter of the spout at the base of the fountain. The diameter of the core (or ascending zone) decreases with the fountain level for all experimental conditions.

A study has been carried on the applicability of correlations previously proposed for the design of conical spouted beds, for the treatment of materials of density similar to that of glass, and for the handling of plastic materials (polyethylene, polypropylene, and extruded and expanded polystyrene). The correlations proposed allow for calculation of the geometric factors (angle and design of the base) of conical spouted beds for stable operation, the minimum velocity for spouting and dilute spouting, the operating pressure drop in both regimes, the peak pressure, the bed voidage, and the evolution of this voidage as the gas velocity is increased between the regimes of a spouted bed and a dilute spouted bed.

The conversions and yields of the fractions of gases (C1, C2), LPG (C3, C4), gasoline (C5−C12), and coke and their compositions have been studied in the catalytic cracking of blends made up of polystyrene and polystyrene−butadiene with a LCO (light cycle oil, which is a secondary refinery product) used as the solvent. The catalysts were prepared from HZSM-5 zeolites of different total acidities. The reaction was carried out in the 450−550 °C range and under the standard conditions of FCC units. The important effects both of the total acidity of the zeolite and of the temperature on the conversion and yield of styrene are demonstrated. Total recovery of the monomer styrene is attained at 550 °C. Likewise, the LCO is valorized, as high yields of both the C5−C12 aromatic fraction and LPG (olefinic) are obtained with low yields of gases (C1, C2) and coke.

The effects of a sweeping treatment on the activity recovery, coke content, H/C ratio, and kinetics of coke combustion have been studied in the coke deposited on an HZSM-5 zeolite. The reaction test studied is the transformation of acetone into hydrocarbons at 400 °C, and the gases used in the sweeping are He, H2, and steam. Aside from the partial elimination of coke, which occurs with all these gases, each one has a different effect. Thus, sweeping with H2 causes catalyst rejuvenation. Sweeping with He promotes coke aging and a reduction in the H/C ratio. Steam is efficient for coke elimination but causes dealumination. The combustion of the remaining coke is disfavored (the activation energy is increased, and the kinetic constant is decreased) by the sweeping treatment that decreases the H/C ratio of the coke.

A study has been made of the valorization of polyolefins, PE (polyethylene) and PP (polypropylene), dissolved in LCO [light cycle oil obtained as a product stream in a fluid catalytic cracking (FCC) unit] by their cracking on three catalysts prepared based on HY zeolites of different acidity and different porous structure. The reaction has been studied in the 450−550 °C range and under conditions similar to those of the FCC. The determining factor in the performance of the catalyst is the accessibility of the acid sites by the dissolved macromolecular chains, where a moderate acidity of the zeolite is sufficient. Whereas a feed of dissolved PE contributes to a decrease of the yield of coke with respect to that corresponding to the cracking of LCO, a feed of PP increases the yield of coke. The optimum temperature for maximizing the yield of gasoline, and minimizing that of coke, is 450 °C. The gasoline fraction (C5−C12) is mainly aromatic with C8+ components. The stream of gaseous products is largely made up of propylene and butenes.

Kinetic results of polyolefin (polyethylene and polypropylene) pyrolysis obtained in a conical spouted bed reactor are compared with those obtained by thermogravimetry and in a high-heating-rate microreactor. The results are evidence of the good performance of this original reactor for kinetic studies of flash pyrolysis of plastics at high temperatures, which is due to bed isothermicity and to the fast and uniform coating of sand particles with melted plastic and, consequently, to lower limitations of heat and mass transfer. Agglomeration problems are minimum in this reactor, which is due to the good gas−solid contact in the spouted bed and to the vigorous particle movement attained with the conical geometry of the reactor.

Basic aspects (coke deposition, acidity deterioration) of the deactivation by coke of a catalyst (prepared from a HZSM-5 zeolite) used in the transformation of aqueous ethanol into hydrocarbons in the 350−450 °C range have been studied. The experiments were carried out in an isothermal fixed-bed reactor with feeds of ethene, mixtures of ethanol and water at different mass ratios, and diethyl ether. The importance of temperature and water content in the reaction medium on the mechanism of coke evolution is demonstrated. The effect of these variables on deactivation is explained by a kinetic model in which the role of coke precursors (ethene, light olefins, and gasoline) is considered. This role depends on the operating conditions, particularly the temperature.

A simulation model has been proposed and used to predict the behavior of a laboratory pseudoadiabatic reactor for the transformation of methanol into hydrocarbons. The model incorporates recent advances in the kinetic modeling of the main reaction steps and of the deactivation by coke of the catalyst in the 350−450 °C range and is efficient for analyzing the effect of process conditions (space time, water content and inert gas in the feed, and inlet temperature) on the evolution with time on stream of the temperature profile along the reactor and on the yields of the product lumps.

A kinetic model for the transformation of aqueous ethanol into hydrocarbons on a HZSM-5 zeolite has been proposed for the 623−723 K range on the basis of experiments carried out in an isothermal fixed-bed reactor with feeds of ethene, ethanol−water mixtures of different mass ratios, and diethyl ether. The kinetic model is of interest for directly upgrading the liquid product obtained by the fermentation of carbohydrate wastes. It is based on the latest progress made in the catalytic transformation of methanol and quantifies the attenuating effect of water in the reaction medium in all the steps of the kinetic scheme.

The effect of operating conditions (base angle, gas inlet diameter, stagnant bed height, particle diameter, and gas velocity) on longitudinal and radial profiles of bed voidage has been studied in the spout, annulus, and both the fountain core and periphery of spouted beds, by means of an optical fiber probe. Correlations are proposed for calculation of the local bed voidage in spout and annular zones, along the spout axis and in proximity to the contactor wall.

The effect of the operating conditions (base angle, gas inlet diameter, stagnant bed height, particle diameter, and gas velocity) on spout geometry has been studied in shallow spouted beds by means of an optical fiber probe. The experimental results have been fitted to an empirical equation for calculation of the average spout diameter in shallow spouted beds. The proposed equation takes into account the effect of base angle and gas inlet diameter, which are parameters that have not been previously taken into account in the literature (for deep beds) and have a great importance in the generation of the spout in shallow spouted beds.

The kinetic behavior has been studied for cracking a commercial feedstock on a hybrid catalyst (with 5 wt % of HZSM-5 zeolite) made up of 80 wt % of a catalyst equilibrated in a fluid catalytic cracking (FCC) unit and of 20 wt % of a catalyst prepared based on a HZSM-5 zeolite. A laboratory-scale reactor which allows for operation under conditions similar to those of commercial FCC units has been used. From the analysis of the gases and gasoline obtained in the 500−550 °C range, with contact times between 2 and 10 s and for a catalyst/feedstock = 6, the influence of HZSM-5 zeolite presence in the catalyst on gas and gasoline composition and on gasoline research octane number has been analyzed.

The performance of original equipment provided with a conical spouted bed reactor has been studied in flash pyrolysis of sawdust with an inert gas (N2) in the 350−700 °C range and with 50 ms of average gas residence time. The effect of pyrolysis temperature on the yields of gas, liquid, and char on gas and liquid composition and on char properties has been studied. The maximum yield of liquid (corresponding to 70 wt %) is obtained at 450 °C and its composition determined by GC/MS is similar to that reported in the literature for bubbling fluidized beds. Although temperatures above 600 °C are required for the development of the char porous structure, the yield of CO2 obtained under these conditions is unacceptable.

A kinetic model for the methanol to olefins (MTO) process on a catalyst based on a silicoaluminophosphate SAPO-34 has been proposed. The model takes into account four individual steps for the production of ethene, propene, butenes, and remaining hydrocarbons (pentenes + paraffins). The kinetic parameters have been obtained by experimentation in an isothermal fixed-bed reactor, in the 623−748 K range. In virtue of the results, it has been proven that the water introduced in the feed (required for attenuation of coke deactivation) inhibits olefin formation. This inhibition is taken into account in the kinetic model.

Combustion of the coke deposited on a commercial catalytic cracking catalyst is studied by differential scanning calorimetry. The conditions of the industrial operation in a fluidized catalytic cracking unit causing deactivation (feed, contact time, and temperature, C/O ratio) and those used for regeneration (catalyst stripping and temperature) have been reproduced. Combustion heat of coke is strongly dependent on combustion temperature and on deactivation severity (reaction conditions). Kinetic equations for the formation of gaseous combustion products, CO, CO2, and H2O, have been determined, which are useful in the design of industrial regenerators.

The simulation of the MTG process has been studied in a fluidized bed with circulation of the catalyst (prepared based on a HZSM-5 zeolite). The simulation has been carried out by taking into account the activity distribution of the catalyst particles in the bed and by using experimentally determined kinetic models for the reaction at zero time on stream and for the catalyst deactivation. The results of the simulation have been proven in an experimental laboratory unit by operating in the range between 380 and 420 °C, with different values of space time and of average residence time of the catalyst.

The vertical components of particle velocities in the spout and annular zones of conical spouted beds have been experimentally determined for different values both of contactor geometric factors (angle and air inlet diameter) and of operating conditions (particle diameter, stagnant bed height, and air velocity) by means of an optical fiber probe. Based on the effect of the operating variables on the longitudinal and radial profiles of the particle velocities, a correlation has been determined for their calculation at any position in the spout zone. The correlation proposed by Németh and Pallai for the calculation of the average particle velocity at the wall in cylindrical spouted beds has been proven to be valid in conical spouted beds.

The effect on combustion in air of the nature of the coke deposited in HZSM5 zeolites used in the MTG process has been studied. This coke is highly hydrogenated and unstable, and its H/C ratio decreases during combustion or when a previous thermal treatment is carried out. Coke H/C ratio greatly affects its reactivity during combustion; consequently, a severe thermal equilibration treatment is recommended for reproducibility of results. Combustion kinetics of equilibrated coke, when it is released from the catalyst, has been proven to be similar to that of the coke deposited on other catalysts for several processes. Lower coke reactivity for aging and combustion, on being deposited within the HZSM5 zeolite, must be attributed to air−coke contact restrictions due to the location of the coke, which partially impedes the flow of air into the crystals.

The influence of reaction conditions in the transformation of methanol into gasoline (temperature, time on stream, and contact time) on the deposition and nature of coke (composition, H/C ratio) and on its location in the porous structure of a H-ZSM5 zeolite-based catalyst has been studied in an isothermal fixed-bed integral reactor. The distribution of the coke within the porous structure of the catalyst is similar to that proposed for other reactions on H-ZSM5 zeolites, and the highly hydrogenated character of coke and its instability is noteworthy. Coke deposition has been related to catalyst acidic site deterioration and to a kinetic model for catalyst deactivation in an integral reactor.

Filming equipment combined with an image treatment program has been used to study the positions of the particles in the jet spouted bed (or dilute spouted bed in conical contactor). The properties determined were the density functions for distribution of particle trajectories and of their maximum heights in the bed and local bed voidages. The spout geometry has been determined for this regime. By extending the experimental study to systems with different values of contactor inlet diameter, stagnant bed height, and relative air velocity with respect to the minimum needed for the jet spouted bed, the effect of these variables on the properties that describe the solid flow has been analyzed.

A study has been made of the activity recovery of H-ZSM5 zeolite based catalysts, which have been used in reaction−regeneration cycles in the transformation of methanol into gasoline. Catalyst stability is sensitive to catalyst calcination conditions (temperature and time) and to zeolite Si/Al ratio. From experiments carried out on automated reaction−regeneration equipment provided with an isothermal fixed-bed reactor, optimum catalyst equilibration conditions have been determined, in order to minimize the irreversible deactivation in the regeneration step. The effect of this step on the catalyst acid structure has been studied.

Local bed voidages have been experimentally determined by means of an optical fiber probe in conical spouted beds of different geometry (angle, inlet diameter) and working under different operating conditions (particle diameter, stagnant bed height, and air velocity). Based on the experimental results, original correlations have been proposed and proven for calculation of bed voidage along the axis of the spout and near the contactor wall. A correlation has been proposed for calculation of bed voidage at any bed position as a function both of bed voidage along the spout axis and at the wall and of the spout radius.

Both simulation and optimization of the transformation of methanol into gasoline and light olefins (intermediate lump in the kinetic scheme) carried out in an isothermal fixed-bed reactor under reaction−regeneration cycles have been studied by using the kinetics of the following:  main reaction, catalyst deactivation, and catalyst reactivation. The catalyst is based on a HZSM-5 zeolite. Optimization of lumps of light olefins and gasoline has been carried out by calculating the optimum values of the length of time for reaction and combustion steps. The effects of other operating conditions (temperature and space time) on the optima have been studied. Suitability of the simulation routine has been proven by carrying out cyclic experiments of reaction−regeneration under the optimum conditions calculated and by comparing the experimental results with those predicted by simulation.

The nature of the active sites and the distribution of acidity strength of a SAPO-34 silicoaluminophosphate has been studied by FTIR spectrophotometry and by combined calorimetry-gas chromatography of ammonia desorption. On the basis of the in situ FTIR analysis of the surface species generated in the transformation of methanol into olefins and of the evidence of the existence of oxonium ions as intermediates, a mechanism has been proposed for the reaction, in which the oxonium ions are successively methylized by means of propagation reactions in the solid phase and ethyloxonium, propyloxonium, and so forth, are formed. This mechanism is mainly conditioned by shape selectivity.

The effect of both the Cr/Zn atomic ratio of the Cr2O3−ZnO catalyst and the Si/Al ratio (different surface acidity) of the ZSM-5 catalyst on the activity and selectivity to gasoline is studied, when a physical mixture of both catalysts is used for transformation of syngas into gasoline in a single step. Furthermore, the composition of each catalyst required for a good compromise between activity and selectivity and for a high aromatic content in the gasoline has been determined. The kinetic behavior of these catalysts has been compared with that of others which have been prepared by combining conventional metallic catalysts in methanol synthesis with a ZSM-5 catalyst.

The thermal and catalytic cracking of several plastic wastes (polypropylene, polystyrene, polystyrene−polybutadiene) dissolved in refinery LCO has been studied in a riser simulator, under operating conditions (catalyst, temperature, contact time, C/O ratio) that are similar to commercial FCC units. The results of the product distribution show the interest of this strategy, which may be integrated into refinery operation for large-scale tertiary exploitation of plastic wastes.

The skeletal isomerization of n-butenes over a chlorinated alumina catalyst has been studied, by operating under reaction−regeneration cycles in an isothermal fixed-bed integral reactor. The limiting values of the reaction and regeneration temperatures for avoiding irreversible deactivation by loss of chlorine have been determined. Reversible deactivation by deposition of coke, which is completely removable by combustion, is unavoidable. The effect of the operating conditions upon the deposition of coke and the deterioration of the porous structure as a consequence of the deposition have been analyzed.

A simulation model for a catalytic polymerization reactor in a dilute spouted bed regime is described. The model, which is of dispersed plug flow, is based on solving the mass conservation equation for the monomer. The input parameters in the simulation program are the geometric factors of the contactor, the kinetic equations of polymerization and of catalyst deactivation, reaction temperature, monomer concentration in the feed, mean residence time of the catalyst in the reactor, and bed voidage. The following are calculated:  (a) parameters that describe the state of the catalyst in the reactor (particle size distribution of the catalyst coated with polymer, catalyst amount, activity distribution); (b) hydrodynamic conditions of the bed (feed flow, dispersion coefficient, radial and longitudinal distribution of gas velocities, and alcohol concentration profiles); (c) catalyst flow required in the feed; (d) polymer yield. The simulation model has been verified with results from a pilot plant using, as test reaction, the polymerization of benzyl alcohol on a silica/alumina catalyst in order to obtain thermally stable polybenzyls.

A kinetic model based on the postulates of Langmuir−Hinshelwood has been proposed for the catalyst deactivation by coke in heterogeneous polymerizations. The originality of the model lies in the fact that deactivation is also taken into account during the initiation period by defining the activity as the useful fraction of active sites throughout the polymerization. The validity of the proposed deactivation kinetic model has been shown for the polymerization of gaseous benzyl alcohol in an ample range of operating conditions:  catalysts (silica/aluminas and a HY zeolite) with different acidity and different porous structure, temperature, and monomer concentration. The results are compared with those of the previously defined model, where activity referred to the maximum polymerization rate.

From experimental results obtained in a wide range of operating conditions (temperature and contact time) in an isothermal fixed-bed integral reactor, the usefulness of the model of Schipper and Krambeck (Chem. Eng. Sci. 1986, 41, 1013−1019) for simulating the operation has been proven in the 300−375 °C range. The validity of a deactivation kinetic model dependent on the composition of the three lumps for the MTG process (oxygenates, light olefins, and the rest of the products) has been proven, for a wide interval of compositions of the lumps. These compositions correspond to different states of catalyst deactivation. The use of this model in the simulation of the integral reactor avoids noticeable errors inherent to the nonconsideration of the effect of composition on the deactivation.

From the results obtained for different geometric factors of the contactor (angle and inlet diameter) and for different operating conditions (particle size, stagnant bed height, and air velocity), the validity of a linear correlation based on the ratio between drag and gravitational forces is proven for the calculation of bed voidage in conical spouted beds when they operate in the transition between the regimes of a spouted bed and a jet-spouted bed. The parameters of this equation are calculated from the correlations previously proposed for the calculation of the bed voidage corresponding to these extreme regimes. The dependency on these parameters of the geometric factors and operating conditions has been determined.