Bed Of Pellets Research Articles

Moisture adsorption characteristics of a thin bed of polycrystalline potash pellets: Part II — Modeling and numerical simulation

This paper describes a numerical investigation of heat and moisture transfer in a bed of porous polycrystalline potash pellets. In this study a non-equilibrium mathematical model of temperature and moisture distribution within the bed was developed. Simulations were performed for a bed of potash pellets that is subjected to flows of humid air on the upper boundary and a cold impermeable surface on the lower boundary. The numerical predictions are in good agreement with the experiments and within the experimental uncertainty bounds. The model predicts that beds of porous potash pellets will contain less moisture than other types of potash fertilizer beds that are exposed to the same environmental conditions. The predictions also indicate that most of the accumulated moisture will be located in the top 10% of the bed and that the pore space relative humidity will be much greater than the supply air relative humidity.

Modeling of multiple noncatalytic gas–solid reactions in a moving bed of porous pellets based on finite volume method

A mathematical model is presented to simulate the multiple heterogeneous reactions with complex set of physicochemical and thermal phenomena in a moving bed of porous pellets. This model is based on both heat and mass transfer phenomena of gaseous species in a porous medium including chemical reactions at interfaces whose areas vary during the conversion. This model accounts for both the exothermic and endothermic reactions which can be equimolar or nonequimolar. Furthermore it considers simultaneously the reactions in the nonisothermal transient condition. A powerful technique based upon finite volume fully implicit approach has been implemented to solve the complicated governing equations numerically. The model has been validated by comparing with various experimental and analytical results in two cases: the single pellet scale as well as the counter current moving bed reactor.

Mass‐transfer models for rapid pressure swing adsorption simulation

AbstractIn this study, trends from an adsorption process simulator are compared against experimental results obtained from a Rapid Pressure Swing Adsorption (RPSA) pilot plant for the separation of air over a packed bed of LiLSX pellets. The primary purpose of this study was to examine the impact of two model formulations for intrapellet mass transfer on predicted process performance, the linear driving force model and a rigorous pellet model (the viscous flow plus dusty gas model intrapellet flux equation). Non‐isothermal and non‐isobaric behavior is maintained within the model formulation. Nine PSA data sets at cyclic steady state were measured, with total cycle time varying from 8 to 50 s. At the short cycle time, large discrepancies between the simple lumped model and the rigorous pellet model arose, with the rigorous pellet model tracking qualitative trends in pilot plant data more faithfully than the simple model. The accepted value for the dimensionless cycle time θi below which linear driving force models fail is 0.1. However, our data and models suggest that a more appropriate value would be 1.0. Below θi = 1.0 the simple linear driving force model started to deviate from experimental data and the rigorous pellet model. This suggests that conclusions from studies of single pellets may not apply rigorously to packed beds of pellets in which each pellet is exposed to rapidly varying and coupled boundary conditions in which composition, pressure, and temperature all change with time. © 2006 American Institute of Chemical Engineers AIChE J, 2006

Honeycomb Supports with High Thermal Conductivity for Gas/Solid Chemical Processes

AbstractFor Abstract see ChemInform Abstract in Full Text.

Combined SO 2 and NO x removal at moderate temperature by a dual bed of potassium-containing coal-pellets and calcium-containing pellets

The combined removal of both SO 2 and NO x from a gas mixture that contains NO x , SO 2, O 2, and N 2 was carried out in a dual bed of calcium-containing pellets and potassium-containing coal-pellets. The calcium-containing pellets retain SO 2 and nitrogen oxides are reduced to N 2 by the potassium-containing coal-pellets. The NO x -carbon reaction is catalysed by the potassium species on pellets, which also avoid the massive consumption of carbon by O 2. The optimum temperature for the combined removal of SO 2 and NO x is around 450 °C. This temperature is high enough to ensure the decomposition of Ca(OH) 2, used as CaO precursor [CaO is much more effective for SO 2 retention than Ca(OH) 2], and low enough to avoid an important combustion of carbon by O 2. The ratio between mass Ca-pellets/mass K/carbon-pellets is a key factor in this process in order to avoid the potassium catalyst poisoning by SO 2 chemisorption. Under the experimental conditions of this study, an appropriate mass Ca-pellets/mass K/carbon-pellets ratio of 10/1 was found. With this ratio, the efficiency of the potassium-containing coal-pellets for NO x reduction is similar to that observed in the absence of the calcium-containing pellets for a SO 2-free gas mixture.

Pressure Drop in a Packed Bed under Nonadsorbing and Adsorbing Conditions

Simulation of cyclic adsorption processes for gas separations relies on accurate models of the bed pressure drop during dynamic pressurization, depressurization, and breakthrough steps. This is especially true for rapid pressure swing adsorption (RPSA), where large gas flows can cause significant changes in the axial pressure gradient through the bed, influencing process performance. Under these conditions, there is some question as to the validity of conventional steady-state models used to represent the dynamic, rapidly changing pressures. In this study, the ability of the steady-state Ergun equation to represent pressure drop under dynamic adsorbing and nonadsorbing conditions was tested. The Ergun equation accurately reproduced dynamic depressurization and breakthrough pressure profiles using values of κviscous and κkinetic (the two Ergun parameters) determined experimentally from the steady-state flow of a variety of gases through a packed bed of LiLSX pellets. The full momentum equation was also tested, and errors of <0.1% were observed between the Ergun equation and the full momentum balance under dynamic conditions for a nonadsorbing packed bed. These observations and simulations suggest the Ergun equation can be reliably used to reproduce experimental profiles under dynamic adsorbing conditions where high gas flows result.

Macropore diffusion dusty-gas coefficient for pelletised zeolites from breakthrough experiments in the [formula omitted] system

Simulation of air separation in zeolites requires an accurate mass transfer model particularly under rapid cycling conditions. An earlier study [Todd et al., 2005. Adsorption 11, 427–432] examined the application of the dusty gas model to the simulation of the adsorption of O 2 / N 2 system in lithium-exchanged low silica-to-alumina ratio x-type zeolite (LiLSX) pellets. In that study we characterised Knudsen diffusion ( C K ) and viscous flow ( C v ) structural parameters, two of three parameters that constitute the viscous flow plus dusty gas model ( VF + DGM ) intrapellet flux equation for a macroporous pellet under adsorbing conditions. This paper quantifies the third structural parameter ( C m ), related to molecular diffusion that arises when a multicomponent gas mixture is introduced into the pore network of a sorbent pellet. With these three parameters defined, the mass transfer behaviour for the O 2 / N 2 system in our adsorbent is completely characterised (under the assumption of no surface diffusion). To experimentally characterise C m , a series of breakthrough experiments were performed with a packed bed of LiLSX pellets. Six independent breakthrough runs over a range of conditions were performed with one of these runs used to determine C m . The remaining five runs were used for validation. Our adsorption simulator previously described [Todd et al., 2003. Computers and Chemical Engineering 27, 883–899] was enhanced with the addition of a rigorous wall model and was used to perform all breakthrough simulations. A value of C m = 0.122 was obtained giving a molecular diffusion tortuosity of 2.5. This is typical of tortuosities for porous adsorbent pellets. Experiments and simulations revealed operating conditions were very close to adiabatic during each experimental breakthrough run while a sensitivity analysis on the intrapellet structural coefficients revealed the sensitivity coefficients of C m and C K are similar in magnitude suggesting both mechanisms of diffusion are important. Viscous flow was several orders of magnitude smaller and hence is of minor importance in breakthrough simulations. Simulations also suggest the level of radial discretisation within a sorbent pellet is an important aspect of a discretised pellet model to consider when simulating a non-isothermal and non-isobaric breakthrough experiment. Relatively high pellet discretisations are needed to obtain physically meaningful structural coefficients.

Hydrodynamics in two-phase flow within porous media

Ultra-fast magnetic resonance imaging techniques are used to image liquid distribution in two and three dimensions during air-water co-current down flow through a fixed bed of cylindrical porous pellets of length and diameter 3 mm, packed within a 43 mm internal diameter column in both the trickle- and pulsing-flow regimes. The data acquisition times used were 20 and 280 ms, giving 2-D and 3-D spatial resolutions of 1.4 mm x 2.8 mm and 3.75 mm x 3.75 mm x 1.87 mm, respectively. This work reports images of local pulsing events within the bed occurring during the trickle-to-pulse flow transition. The evolution of the local instabilities is studied as a function of increasing liquid velocity at constant gas velocity.

Transition to pulsing flow in trickle‐bed reactors studied using MRI

AbstractUltrafast magnetic resonance imaging (MRI) is used to provide two‐dimensional (2‐D) images of gas‐liquid distribution within trickle‐bed reactors with data acquisition times of 20 and 40 ms. Gas‐water, cocurrent downflow through a fixed bed of cylindrical porous pellets of length and dia. 3 mm, packed within a 43 mm internal dia. column, was studied in both the trickle‐ and pulsing‐flow regimes. Superficial gas velocities in the range 50–345 mm s−1 (0.06–0.42 kg m−2 s−1), and superficial liquid velocities in the range 0.4–13.3 mm s−1 (0.4–13.3 kg m−2 s−1) were used. MRI is used to investigate the stability of the gas‐liquid distribution in the trickle‐ and pulsing‐flow regimes. At the onset of the transition to pulsing flow, local pulsing, at the length‐scale of the size of the packing elements is observed within the bed. Increasing liquid velocity causes an increase in the number of these local pulses until a velocity is reached at which the system transforms to the rapidly‐changing gas‐liquid distribution typical of pulsing flow. © 2005 American Institute of Chemical Engineers AIChE J, 51: 615–621, 2005

Application of ported kiln technology to the Grate-Kiln Process at Minntac

In 2001, the Minntac Division of US-Steel retrofitted one of their grate-kiln iron ore pelletizing lines with a ported-kiln air-injection system (PKAIS), the first of its kind in the world for oxide pellets. The ported kiln system injects air under the bed of pellets in the rotary kiln, causing rapid oxidation of the magnetite pellets. This rapid oxidation allows for lower kiln temperatures, which reduces fuel demand; allows for less heat load on the annular cooler; and produces higher pellet quality. Both the process and the equipment are patented developments by Metso Minerals. This paper discusses the process, the key equipment and the benefits to both the pellet plant and the downstream blast furnace operations.

New insights to trickle and pulse flow hydrodynamics in trickle-bed reactors using MRI

Ultra-fast Magnetic Resonance Imaging (MRI) is used to image liquid distribution during air–water, co-current downflow through a fixed bed of cylindrical porous pellets of length and diameter 3 mm, packed within a 43 mm internal diameter column in both the trickle- and pulsing-flow regimes. Two different data acquisition times are used, 20 ms and 6.4 s, giving two-dimensional images of in-plane spatial resolution 1.4 mm × 2.8 mm and 351 μ m × 351 μ m , respectively. Superficial gas velocities in the range 27 – 275 mm s - 1 ( 0.033 – 0.338 kg m - 2 s - 1 ) and superficial liquid velocities in the range 0.4 – 13.3 mm s - 1 ( 0.4 – 13.3 kg m - 2 s - 1 ) were used. This work reports images of local pulsing events within the bed occurring during the trickle-to-pulse flow transition. The evolution of the local instabilities is studied as a function of increasing and decreasing liquid velocity, at constant gas velocity. Temporal fluctuations in local liquid holdup were seen even in the trickle flow regime; these fluctuations had much shorter time periods than the pore-scale liquid pulses, and are perhaps associated with fluctuations in local surface wetting of the packing elements. A preliminary assessment of the effect of periodic operation on liquid distribution within the bed is also reported.

Application of Dynamic Mechanical Analysis (DMA) to the determination of the mechanical properties of coated pellets

Pellets containing a model drug, paracetamol, and microcrystalline cellulose (MCC) were designed to vary their mechanical properties by the incorporation of lactose, glyceryl monostearate (GMS), ethanol, or glycerol, and were produced by the process of extrusion and spheronization. The pellets were coated with an aqueous dispersion of ethyl cellulose (Surelease ®) to different levels of weight gain (5, 10, and 20%). The tensile strength, deformability, linear strain, elastic modulus, and shear strength of the coated and uncoated pellets were determined by conventional techniques, which are obtained from diametral compression test of individual pellets and compaction of a bed of pellets. Dynamic Mechanical Analysis (DMA) was performed on single pellets to determine the storage modulus and phase angle of the coated pellets. This work demonstrated that the coating film affected the mechanical properties of the pellets differently depending on the properties of the core pellets. Analysis of variance established a significant increase in the strength of the soft GMS- or glycerol-containing pellets with coating, while the effect of the coating material was not significant with respect to the elastic modulus, storage modulus, and phase angle of such pellets. The effects of the coating material on the elastic modulus, deformability, storage modulus, and phase angle of the rigid lactose-containing pellets were significant. The sinusoidal stress–relaxation cycle of the DMA illustrated the increase in the viscoelasticity of all the pellets after coating. Finally, the work demonstrate the advantages of DMA in determining the reversible or dissipated energy by means of storage modulus or phase angle when compared with the irreversible structural destruction of the pellets by conventional techniques.

Application of dynamic mechanical analysis (DMA) to determine the mechanical properties of pellets.

Pellets of a wide range of mechanical properties were produced by the process of extrusion and spheronisation using various formulation factors. A range of mechanical properties from a simple fracture load to detailed load/displacement curves obtained when pellets were subjected to diametral compression test and a bed of pellets was compacted, were used to provide measure of tensile strength, deformability, linear strain, elastic modulus, yield and shear strength. Such conventional techniques resulted in irreversible damage to the structure of the pellets and were unable to establish the viscoelastic properties of the pellets. The application of the dynamic mechanical analysis (DMA), however, allowed the determination of (1) an accurate Young's modulus of elasticity, which was found to be between 8.4 and 24-fold higher than that determined from the diametral compression test, (2) the presence of a reversible elastic deformation even after the yield point in terms of storage modulus and (3) a change in the values of the phase angle, which illustrates the increase in viscoelasticity of the pellets formed with ethanol, glyceryl monostearate (GMS) or glycerol, while a decrease in viscoelasticity with the incorporation of lactose into the microcrystalline cellulose (MCC) pellets. This work further demonstrated that the only feasible technique for determining the elastic and plastic deformability of the pellets is the one which subjects the specimen to stress/relaxation cycles and can determine the dissipated energy in terms of loss modulus or phase angle, and that is DMA.

Properties of ceramic foam catalyst supports: mass and heat transfer

Mass and heat transport properties have been determined for 30 PPI α-Al 2O 3 ceramic foam containing 6 wt.% γ-Al 2O 3 washcoat. The foam was loaded with 5 wt.% platinum and the rate of carbon monoxide oxidation measured for a 0.3 cm cylindrical segment of the foam operating with mass transfer controlling at 550 °C. This gave a mass transfer factor versus Reynolds number correlation that was equivalent to a packed bed of particles. A correlation for the radial heat transfer coefficient in a bed of ceramic foam was determined by measuring outlet temperatures achieved when air at varying flow rates and inlet temperatures was passed through a bed of foam pellets. Correlation parameters of a 1D model were fitted from 700 to 1000 °C using a Simplex optimization routine. Radial heat transfer coefficients were two to five times higher than those predicted from packed bed correlations.

Effect of Drag‐Reducing Polymer on the Rate of Cementation of Copper Ion on Zinc Pellets

AbstractThe effect of Polyox WSR‐301 drag‐reducing polymer on the rate of cementation of copper from copper sulfate solution over packed beds of zinc pellets was investigated. The mass transfer coefficient was found to increase with increasing superficial liquid velocity for solution free of polymer. The mass transfer data were correlated in the absence of drag‐reducing polymer, using the following equation Jd = 12.55 Re–0.5 for the conditions 10 &lt; Re &lt; 1970 and 1265 &lt; Sc &lt; 1393. In the presence of polymer, starting from a Reynolds number (Re) 550, the rate of mass transfer was found to decrease by an amount ranging from 7.5 to 51 % depending on Re and polymer concentrations. The percentage decrease in the rate of mass transfer increased with increasing Re, passed through a maximum at Re = 1400 and then decreased rapidly with further increase in Re.

‘High Conductivity’ Monolith Catalysts for Gas/Solid Exothermic Reactions

The replacement of conventional packed beds of catalyst pellets with novel 'high conductivity' honeycomb catalysts in industrial externally cooled multitubular fixed-bed reactors for exothermic gas/solid processes offers potential advantages besides reduced pressure drops. Near-isothermal operation could result in safer reaction operation, better catalyst thermal stability, improved selectivities, higher throughputs and more economical reactors with larger tubes. We report an experimental and theoretical investigation of this concept.

Pressure buildup in gas‐liquid flow through packed beds due to deposition of fine particles

AbstractIn order to understand the increase in pressure drop in hydrotreating reactors due to deposition of fine solids, experiments were conducted with a model suspension of kaolin clay in kerosene. The suspension was circulated through packed beds of catalyst pellets in the trickle‐flow and pulse‐flow regimes, and the increase in pressure drop measured as a function of particle concentration in the bed. The increase in pressure drop was linear with particle concentrations over the range 0–60 kg.m−3. A consistent approach to modeling the pressure drop behavior was to determine an effective porosity of the packed bed as a function of the concentration of fine particles, then use this porosity in the Ergun equation as a basis for calculating the two‐phase pressure drop.

Adsorption and pressure swing desorption of NOx in Na-Y zeolite: experiments and modeling.

Pressure swing NOx adsorption-desorption cycles were performed in the temperature range 200-350 degrees C using a fixed adsorbent bed of compressed Na-Y pellets and using a honeycomb coated with Na-Y powder. The experiments were performed using a synthetic gas mixture mimicking exhaust from a lean burn internal combustion engine. Na-Y zeolite coadsorbs NO and NO2 as N2O3, which in the regeneration were displaced by competitively adsorbed water molecules from a hydrated air stream. The performance of the fixed bed in these NOx adsorption and displacement desorption processes were modeled with a one-dimensional model. The kinetic and thermodynamic parameters from the fixed bed model were implemented in a model for the operation of the monolith. The experimental adsorption and desorption NOx concentration profiles in the monolith were reasonably well reproduced by the model. The water content of the flushing stream and the stripping gas flow rate are key process parameters. Technically, both parameters can be optimized in a valveless system with rotating honeycomb adsorbent comprising a NOx adsorption, a water injection and a NOx evacuation section.

Characteristics of fixed beds packed with anisotropic particles—Use of image analysis

Fixed beds of anisotropic particles are commonly involved in chemical engineering processes: beds of cylindrical catalyst pellets or beds of wood shavings in paper industry. The characterization of the spatial variations of their structural properties is of great importance for flow modelling purposes. In this work, we intend to produce new data concerning a bed of long cylinder ( h/d=5.29) and a bed of flat plates ( e/a=0.209). This study focuses on the characterization of top and bottom effects, as well as wall effects. The experimental procedure is based on the study of sections of packed beds consolidated with a resin. The captured images of successive transverse and longitudinal sections of the beds are treated and analysed, thanks to an image analysis software. The local variation of the voidage is determined in both the transverse and the longitudinal directions of the beds. Furthermore, mean intercept lengths and mean numbers of particles are measured and discussed. Concerning flat plates, an anisotropic ratio calculated from intercept measurements is compared to that previously obtained from pressure drop measurements. The main results of this work concern the enlightening of wall effects and end effects importance for the two studied beds compared to beds of spheres studied in the literature.

Simulation of structured catalytic reactors with enhanced thermal conductivity for selective oxidation reactions

The replacement of conventional-packed beds of pellets with “high conductivity” honeycomb catalysts in industrial externally cooled multitubular fixed-bed reactors is investigated by modeling and simulation for the oxidation of methanol to formaldehyde and for the epoxidation of ethylene to ethylene oxide, which involve a consecutive and a parallel reaction scheme, respectively. Results suggest that near-isothermal operation of the fixed-bed reactors can be achieved using monolithic catalyst supports based on relatively large volume fractions of highly conductive materials. Pressure drops are reduced to less than 1%. The selectivity is favored by the excellent control of the intraporous diffusional resistances resulting from the thin catalytic washcoats. Reactor designs based on larger tubes are feasible at the expense of greater volume fractions of catalyst support. A critical aspect is represented by the restrictions on the specific load of catalyst per reactor volume resulting from the poor adhesion of very thick catalyst layers onto metallic surfaces. Such a difficulty can be circumvented by maximizing the geometric surface area of the monolith (e.g. minimizing the honeycomb pitch), enhancing the catalytic activity (e.g. increasing the load of active components), and increasing the coolant temperature (if the selectivity is not adversely affected).