Article

Particle residence time distributions in circulating fluidised beds

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

This paper gives experimental measurements of the particle residence time distribution (RTD) made in the riser of a square cross section, cold model, circulating fluidised bed, using the fast response particle RTD technique developed by Harris et al. (Chem. Eng. J. 89 (2002a) 127). This technique depends upon all particles having phosphorescent properties. A small proportion of the particles become tracers when activated by a flash of light at the riser entry; the concentration of these phosphorescent particles can subsequently be detected by a photomultiplier. The influence of the solids circulation rate and superficial gas velocity on the RTD were investigated. The results presented are novel because (i) the experiments were performed in a system with closed boundaries and hence give the true residence time distribution in the riser and (ii) the measurement of the tracer concentration is exceedingly fast. The majority of previous studies have measured the RTD in risers with open boundaries, giving an erroneous measure of the RTD.Analysis of the results suggests that using pressure measurements in a riser to infer the solids inventory leads to erroneous estimates of the mean residence time. In particular, the results cast doubt on the assumption that friction and acceleration effects can be neglected when inferring the axial solids concentration profile from riser pressure measurements.An assessment of particle RTD models is also given. A stochastic particle RTD model was coupled to a riser hydrodynamic model incorporating the four main hydrodynamic regions observed in a fast-fluidised bed riser namely (i) the entrance region, (ii) a transition region, (iii) a core-annulus region and (iv) an exit region. This model successfully predicts the experimental residence time distributions.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... For experimental measurements, a certain amount of tracers were introduced into the apparatus, and then the solid residence time was determined via detecting tracer signals at the system exits [26,27]. Using this incentive-response technology of the tracer, researchers carried out many studies about solid residence time in the CFB riser [28][29][30][31][32][33][34][35]. The results showed that the solid residence time distribution could be descript by a plug flow with dispersion [34], and the solid residence time distribution curves had a sharp peak with a long tail [31]. ...
... Using this incentive-response technology of the tracer, researchers carried out many studies about solid residence time in the CFB riser [28][29][30][31][32][33][34][35]. The results showed that the solid residence time distribution could be descript by a plug flow with dispersion [34], and the solid residence time distribution curves had a sharp peak with a long tail [31]. Besides, the Positron Emission Particle Tracking (PEPT) of single radio-active tracer particles was also used to study solid residence time under different operating regimes [36]. ...
... In each region, each individual particle has a SRT value, which is defined as the time from this particle entering the region to leaving it. When the system reaches equilibrium state, all particles are tracked for obtaining their SRTs in each region to produce the probability histograms of SRT (i.e., E-curve) [31,61]. The curves are corrected to a zero baseline, and normalized to make ∫ 0 ∞ E(t)dt = 1. ...
Article
The effect of superficial gas velocity (Uf) on solid behaviors in a full-loop circulating fluidized bed is numerically studied using computational fluid dynamics-discrete element method (CFD-DEM). Specifically, the solid mixing and dispersion, solid residence time, solid force and velocity, and particle granular temperature are comprehensively explored. The results show that increasing Uf slows the solid mixing. The solid axial dispersion in the riser is dominated among that in the three regions. Besides, increasing Uf decreases the solid cycle time and leads to a more uniform solid residence time (SRT) distribution in the riser and dipleg. The fluid force is an order of magnitude smaller than the collision force. Increasing Uf shows distinctive influences on the rotational speed and translational velocity in the three regions. Moreover, increasing Uf suppresses the particle granular temperature. The average particle granular temperatures in the riser at four superficial gas velocities (i.e., 5.5 m/s, 6.0 m/s, 6.5 m/s, and 7.0 m/s) are 0.09326 m²/s², 0.08040 m²/s², 0.06531 m²/s², and 0.05047 m²/s², respectively.
... In addition to the pressure signals, the solid back-mixing behavior in the CFB riser is usually characterized by solid residence time distribution (RTD). 20 In general, the macro-scopic flow structure usually exhibits heterogeneous distributions in both axial and radial directions 21 of the CFB riser. Take the fast fluidization (FF) regime for example: the axial void fraction profile commonly shows the "S" shape, while the radial profile usually exhibits the core−annulus structure in the CFB riser; 22,23 moreover, plenty of experimental studies have demonstrated that many clusters exist in the bottom, dense region and the near-wall region. ...
... Traditionally, the impulse stimulus response technique is usually adopted to obtain the solid RTD in the CFB riser. 20,27 On the other hand, a series of empirical models in this area have been developed: one-and two-dimensional diffusional models, the stochastic model, and the core−annulus tracer balance model. 28 However, the majority of these models are based on simplification or depend heavily on lots of semiempirical correlations, thus they are physically unrealistic. ...
... Specifically, for simplifying the geometry of the CFB, according to Grace et al., 68 we choose the L-valve type of loop-seal, and this type is also widely employed in the experimental and numerical work. 14,20 The geometry of the investigated object in the current work is similar to the CFB used in the literature, 14,20 but the dual-side refeed structure (i.e., two sets of recirculation systems arranged symmetrically) is adopted because of its improvements of gas− solid flow uniformity (i.e., homogeneous) in the riser, which is discussed in our other work. The geometry shown in Figure 2a Figure 2c, and a total number of 150 028 hexahedron elements are generated by the commercial software Gambit 2.4.6. ...
... Wei and Du [16] Riser of CFB From the review of the reported experimental methods, several research groups [16], [19]- [25] prefer to employ phosphorescent tracer particles thanks to its simplicity and high accuracy. In these works, the properties of phosphorescent tracer particles were similar to the average properties of the bulk particles. ...
... The excited phosphorescent particles emit light and the afterglow is captured by a detector and collected by a computer data acquisition system as a function of time. This optical method has been perfected by Harris et al. [19]- [21]. In these phosphorescent tracer experiments, phosphorescent particles were used as both bulk and tracer particles. ...
... Therefore, residence time distribution data help in the understanding of fluid dynamics. These data are essential for reactor design, scale-up, plant operation and optimization [19]. ...
Thesis
Full-text available
The aim of the present thesis is to develop a novel experimental technique for determining the residence time distribution (RTD) of solid particles in solid unit operations as well as model development. Initially, a novel optical method was developed to measure the particle RTD. Experiments are carried out with Silicon Carbide (SiC) and the pigment phosphorescent (Lumilux® Green SN-F50 WS) as tracer particle. A preliminary experimental study was conducted in a simple bubbling fluidized bed in order to validate the proposed RTD measurement methodology. In the second step, the developed technique of the concentration measurement was applied to measure the RTD of a deep fluidized bed. The particle RTD curves are determined experimentally in different operating conditions. Finally, a model consisting of the combination of the ideal reactors is proposed to predict the particle residence time distribution in the studied fluidized bed. The predicted output values are then compared with the experimental data to establish a good model fitting data
... Greek letters a mass ratio of coarse particles to fine particles in a binary system, dimensionless In recent 20 years, several relevant reviews have summarized the critical development of solids mixing and RTD issue insightfully from various standpoints (Bi, 2004;Breault, 2006;Gao et al., 2012;Harris et al., 2003c;Nauman, 2008;Werther and Hirschberg, 1997). Due to the wide research scope in a vast variety of fields, however, they might have to restrict their attention to some specific aspects. ...
... Lim et al. (1995) and Berruti et al. (1995) also dedicated some paragraphs to solids mixing in CFB briefly. Furthermore, Harris et al. (2003c) were concerned more about the experimental studies of solids RTD in risers. Summarily, most studies of solids RTD in CFB risers focused on how to improve the accuracy of measurement technology and improve the predictive ability of theoretical models. ...
... For solids RTD in CFB risers, some significant discrepancies arose among numerous publications, and even seemingly contradictory conclusions were drawn by various researchers. First of all for the shape of RTD, some researchers observed the occurrence of a double-peak (Ambler et al., 1990;Bhusarapu et al., 2004a;Kojima et al., 1989;Lin et al., 1999;Patience et al., 1991;Wei et al., 1998;Weinell et al., 1997) or even multi-peak (Bhusarapu, 2005;Wang et al., 1996), whereas more researchers only measured a typical single-peak curve (Andreux et al., 2008;Bader et al., 1988;Harris et al., 2003c;Rhodes et al., 1991;Smolders and Baeyens, 2000;Viitanen, 1993;Zheng et al., 1992). Secondly, as the two widely used parameters to quantitatively measure solids RTD, solids dispersion coefficient, D, and Peclet number, Pe, have a huge discrepancy in magnitude for the tests measured in risers. ...
Article
Solids Residence Time Distribution (RTD) in Circulating Fluidized Bed (CFB) risers has received increasing attention due to its vital role in determining the operating condition, particle property, and reactor geometry in industrial CFB applications. In recent years, various solids RTD experimental techniques and theoretical models have been utilized and proposed to study CFB risers. Some controversial issues, however, also arose in the open publications. By means of exploring the advantages and disadvantages of each available RTD experimental technique and model when they are applied to particles in CFB risers, this study discussed the primary causes leading to the huge discrepancy in magnitude of solids dispersion coefficient and Peclet number, which can achieve 4 orders or span from 1 to 100. On the basis of the massive experiment data collected from the literature, the variations of average residence time, Peclet number and dispersion coefficient of solids with superficial gas velocity, solids mass flux and solids concentration were presented. By applying the transition of flow regime in CFB mode, we provided a helpful way to explain some existing contradictions in the reported effects of operating conditions on solids RTD. The possible reasons were also summarized to clarify why some researchers measured a double- or multi-peak solids RTD curve and the others could not in a similar situation.
... It determines the product directly because the inadequate residence time of the reactant in the reactor can lead to an overor under-reaction not beneficial to the industrial companies. The research on solid RTD has been explored widely in a circulating fluidized bed (CFB) riser or other CFB parts due to the fact that the particles pass the CFB riser and downcomer continuously to form an external loop (Bi, 2004;Breault, 2006;Harris et al., 2003). In contrast, RTD of solids is seldom investigated in a batch-mode dense fluidized bed that is operated in bubbling, slugging or turbulent fluidization regime. ...
... Finally the colored particles were dried totally to be ready for the tests. This coloring method could ensure that the tracer particles have exactly same physical properties as the bulk bed material, which is one of the essential requirements of the RTD technique (Harris et al., 2003). The tracer concentration in each sample was defined here as the mass fraction of the tracer particles. ...
Article
The residence time distribution (RTD) of particles in a gas-solid dense fluidized bed with a continuous solid flow operation and baffles needs much more focus for the purpose of the wide industrial applications. The available models of solid RTD are not adequate in this case due to the additional convective diffusion induced by the cross-flow of the solid feeding and the complex geometry caused by the baffles. To address this problem, this work applied a two-phase Eulerian-Eulerian model combined with the species transport equation to predict solid RTD in a dense fluidized bed. The effects of the continuous solid feeding and baffles were considered. The solid dispersion coefficient D s in the species transport equation was calculated by an analytical solution from kinetic theory of granular flow. D s only needs the value at a molecular level due to the fact that the computational fluid dynamics (CFD) model is able to reproduce the RTD procedure exactly same as the practical experiments. To validate the established CFD model, a series of 3D lab-scale cold flow experiments were conducted in the free and baffled beds for the bed material with various severe non-spherical shapes. The measurement included the solid hydrodynamic characteristics at the outlet of the outflow pipe and solid RTD of the system. The reasonable agreement between the CFD prediction and experimental data demonstrated the good performance of the CFD model. 1D plug flow with dispersion model and the tanks-in-series model were further used to fit the calculated RTD curves. The estimated lateral dispersion coefficient of solids locates in a rational range compared with the data collected extensively from the literature. The results showed that the lateral dispersion coefficient of solids decreases when the baffles are installed.
... There have been many reports of investigations of the RTD of solids in various CFB risers because of its importance in at the exit as a function of time. To improve the measurement of the RTD of the solid, various tracer particles have been used since 1980, including ferromagnetic particles [23,24] , radioactive sand [25] , sodium chloride [26][27][28] , and phosphorescent particles [29][30][31] . These studies mainly focused on the effect of the operating conditions; i.e. , the superficial velocity, solid circulation rate, and the diameter of the riser [23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39] . ...
... To improve the measurement of the RTD of the solid, various tracer particles have been used since 1980, including ferromagnetic particles [23,24] , radioactive sand [25] , sodium chloride [26][27][28] , and phosphorescent particles [29][30][31] . These studies mainly focused on the effect of the operating conditions; i.e. , the superficial velocity, solid circulation rate, and the diameter of the riser [23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39] . ...
Article
The authors investigate the effects of the direction of the gas jet on the solid residence time distribution in a CFB riser. Tracer technique was employed to calculate the RTD of solids. A Eulerian–Eulerian model with kinetic theory of granular flow and species transport was used to simulate the motion of tracer particles in a CFB riser. For a comparative analysis of the direction of the gas jet, simulations of vertical, horizontal and hybrid jets were carried out. The direction of the gas jet significantly influenced the axial and radial structure of bed, and hence affected the RTD for solid particles. The mean residence time of solids was changed, and the results showed that 16.3 s, 14.8 s, and 11.4 s with vertical, horizontal and hybrid jets nozzles, respectively.
... As such, a characterisation method that accounts for size should be implemented. Reactor characterisation studies present in the literature propose a number of possible solutions such as assuming that solids of different sizes travel at the same speed, using detectable radioactive or fluorescent tracers or even using laser diffraction (Pareek et al., 2001;Harris, Davidson and Thorpe, 2003;Bhusarapu, Al-Dahhan and Dudukovic, 2004;Sievers et al., 2016). Whilst no reports were found in the literature showing the implementation of laser diffraction to determine particle travel time based on size, this technique could be applied to characterize the ACR. ...
... Whilst soluble tracers with assumed rheological equivalence are used often, radioactive or phosphorescent particles have also been employed. A disadvantage of the latter is the time and effort required to ensure that the selected particles have the same properties as those being studied (Harris, Davidson and Thorpe, 2003;Bhusarapu, Al-Dahhan and Dudukovic, 2004;Gao, Muzzio and Ierapetritou, 2012). One study by Pareek et al. (2001) actually used laser diffraction to determine titania Time (min) powder contents at the outlet of a three phase annular bubble column reactor, but no investigation in the RTDs of different particle sizes was carried out (Pareek et al., 2001). ...
Conference Paper
Valorisation of agricultural wastes, such as Sugar Beet Pulp (SBP), for production of biofuels and value-added chemicals, has garnered increasing interest in recent years. Through physicochemical means, lignocellulosic material can be pretreated to release monosacharides which can then be upgraded by fermentative and biocatalytic routes. Previous UCL-led research has examined many aspects of utilisation of waste streams from sugar refineries. Vinasse, a glycerol-rich waste product of bioethanol production, was used as a nutrient source for enzyme production. Sugars from SBP, such as D-glucose, L-arabinose and D-galacturonic acid and which make up approximately 25% w/w, 21% w/w, and 20% w/w of the total pulp weight, respectively, were solubilised through operations such as steam explosion pretreatment and depolymerisation of the released polysaccharides. These SBP monosaccharides were then employed in bioconversion reactions using thermostable enzymes. This Thesis aims to study SBP as a feedstock for the enzymatic production of value added chemicals. It also aims to translate key reactions in the valorisation process from batch mode into a continuous flow process in a scalable, 100 mL, Agitated Cell Reactor (ACR). Initial Residence Time Distribution characterisation of the ACR showed that it provided excellent plug flow properties, equivalent to 13 stirred reactors in series. The ACR was able to handle SBP slurries over a range of solids loadings (1% w/v – 5% w/v) and residence times (3.8 min – 19.0 min). The SBP suspension was shown to be shear thinning with measured viscosities in the range of 0.0011 Pa.s at 1% w/v and 0.0339 Pa.s at 10% w/v. A set of correlations was developed that enable prediction of the feed viscosity as a function of SBP concentration and shear rate. The SBP particle size distribution ranged from 15.0 μm (D10) to 446 μm (D90) with a median size of 128 μm. Studies on the particle flow through the ACR demonstrated that steady state could be achieved, but that larger particles had longer residence times than smaller particles through the ACR. Dilute acid pretreatment (DAP) of SBP was investigated as an alternative to previous work on steam explosion as it would be more compatible with continuous operation. DAP using sulfuric acid at concentrations up to 75 mM and 80 °C was performed. These conditions showed good release of polymeric L-arabinose, which increased with higher temperatures and acid concentrations (70% w/w at 75 mM and 80 °C). Cellulose, which is more heat- and acid-resistant than SBP pectin, was only slightly hydrolysed into D-glucose, creating the potential for selective sugar fractionation. When compared to steam explosion pretreatment, flow DAP in the ACR obtained similar throughputs (3.5 and 3.1 g(L-arabinose).hr⁻¹, respectively), but productivity (throughput in terms of reactor volume) was an order of magnitude higher (3.5 and 25.6 g(L-arabinose).L⁻¹.hr⁻¹). Monomerisation of the polymeric L-arabinose could be achieved in a continuous flow enzyme-membrane as in previously described work. Finally, valorisation of the L-arabinose monomers by a continuous-flow, two-step enzymatic process in the ACR was demonstrated. L-gluco-heptulose is a rare ketoheptose which has potential cancer and diabetes treatment applications. The one-pot two-step production of L-gluco-heptulose using a thermostable transaminase (TAm) and transketolase (TK) both isolated from Deinococcus geothermalis DSM11300 was also carried out in the ACR. The initial goal was to use immobilized TK and TAm enzymes in order to intensify the bioconversion process. While TK could be successfully immobilized on both Nickel-chelated beads and Epoxymethacrylate resin, the TAm immobilization proved challenging with only low levels of retained activity. Consequently, the flow studies were performed with soluble TK and TAm enzymes. ACR bioconversions compared favourably with well-mixed batch reactions yields using the same reaction time (2 hours). Initial studies demonstrated the conversion of model substrates L-arabinose, L-serine and α-ketoglutaric acid into L-gluco-heptulose. Subsequently it was shown that L-gluco-heptulose could be synthesised equally well using SBP-derived L-arabinose. Concentrations of the intermediate product hydroxypyruvic acid (HPA) and L-gluco-heptulose obtained in continuous mode were 2.62 mM and 0.60 mM, respectively using SBP derived L-arabinose and 1.21 mM and 0.31 mM, respectively, using model solutes. This was equivalent throughputs of 170.5 µM.hr⁻¹ and 39.0 µM.hr⁻¹ for the SBP derived L-arabinose and 81.0 µM.hr⁻¹ and 20.0 µM.hr⁻¹ for the model solutes. Higher final L-gluco-heptulose concentrations could be obtained by increasing starting L-arabinose concentrations. The continuous process demonstrated here has clear potential for use within a SBP biorefinery. Future work needs to focus on alternative methods of TAm immobilization to enable process intensification and scale-up to pilot scale in order to demonstrate robust commercial operation.
... The profile of cross-section solid holdup along the CFB riser in baffles-free case is consistent with experiment measurements and simulation results (Jiradilok et al., 2008;Xu et al., 2018). Large value in the recycling port and small value in the top region are attributed to the entrainment effect and exit restraint effect, respectively (Harris et al., 2003;Kim et al., 2008;Luo et al., 2015). Solid holdup for the ring-type 1 to 3 cases is larger than the baffles-free case in the higher part of riser while smaller than the baffles-free case in the lower part of riser. ...
... Solid residence time (SRT) is a quantitative criterion to assess the solid back-mixing and the degree of chemical reaction in the CFBs Chen et al., 2017;Hua et al., 2014;Shi et al., 2015). In experiments, the incentive-response technology of tracer is usually used to obtain the time-related information (Bhusarapu et al., 2004;Guío-Pérez et al., 2014;Harris et al., 2003;Hua and Wang, 2018). However, the experimental methods suffer from high cost and hard operation, thus the numerical simulation is expected to be an important supplement. ...
Article
Gas-solid distribution is non-uniform in the circulating fluidized bed (CFB), which deteriorates gas and solid mixing, causing a lower system performance. In this work, the effect of ring baffles on system performance of a full-loop CFB is comprehensively studied using the computational fluid dynamics-discrete element method (CFD-DEM). Detailed information (i.e., solid velocity, solid flux, solid circulation rate, solid inventory height, solid residence time, and solid dispersion) are obtained and analyzed. Meaningful results include: (i) ring baffles improve the inter-particle and inter-phase interactions, and solid particles become more uniformly dispersed in the riser; (ii) introduction of ring baffles increases the solid holdup in local regions of the riser while has a slight influence on pressure drop; (iii) increasing the number of ring baffles and the spacing between ring baffles increases the solid circulation rate and solid inventory height, decreases solid cycle time in the system and SRT in the riser, increases SRT in the dipleg, and has a slight impact on SRT in the cyclone; (iv) the average solid dispersion coefficients in the riser along three directions are 0.675 × 10⁻³, 0.522 × 10⁻³, and 8.415 × 10⁻³ m²/s, and the insertion of ring baffles decreases the overall dispersion intensity in the axial direction while has a slight influence in the radial direction.
... In consideration of the great advantages and broad applications of the CFB, plenty of experimental efforts have been made to reveal gas-solid flow dynamics, such as general flow patterns and regime transition (Chew et al., 2012;Monazam and Shadle, 2011;Qiu et al., 2014), particle mixing and circulation (Lim et al., 2013;McMillan et al., 2013;Zi et al., 2016), solid velocity and resident behavior (Bhusarapu et al., 2006;Harris et al., 2003;Pantzali et al., 2013), pressure fluctuation and heat and mass transfer (Kalita et al., 2013;van der Schaaf et al., 2004;van Ommen et al., 2011). However, the experimental methods have obvious disadvantages in the aspect of high cost and hard operation, thus the computational fluid dynamics (CFD) has been used as an alternative to obtaining the intrinsic mechanism of gas-solid motions in the CFB with the development of numerical algorithm and computational capacity. ...
... By means of these numerical tools, gas-solid flow dynamics is mainly obtained in the single component of the CFB. For the gas-solid flow dynamics in the riser, attention is mainly paid to gas-solid distribution Zhang et al., 2008aZhang et al., , 2008bZhou et al., 2014), solid dispersion feature (Jiradilok et al., 2008), solid resident behavior (Harris et al., 2003), and effects of internal devices (Zhao et al., 2015a(Zhao et al., , 2015b. Besides, the optimization of loop seal (Seo et al., 2011;Zhao et al., 2015aZhao et al., , 2015b and the separation efficiency of cyclone (Chu et al., 2011;Elsayed and Lacor, 2013) are also widely investigated. ...
Article
Multiple cyclones are adopted to enlarge the capacity of circulating fluidized bed (CFB). In this work, the effect of cyclone arrangements on the gas-solid flow dynamics in the three-dimensional full-loop CFB is investigated by the computational fluid dynamics coupled with discrete element method (CFD-DEM). Flow patterns, pressure distribution, and non-uniform distribution of the gas-solid flow in the CFB is comprehensively studied. Results show that the CFB with multiple cyclones gives rise to some unique characteristics, including: (i) solid back-mixing behavior mainly occurs close to the front and back walls and in the four corners of the riser; (ii) the spiral directions of the internal and external vortexes of gas flow in the cyclone are same; (iii) the closed-loop pressure is obtained and the pressure drop in the standpipe and L-valve are nearly four times of that in the cyclone; (iv) gas-solid flow shows non-uniform distribution in the multiple cyclones, and the middle cyclones have higher solid holdup and solid velocity than the corner cyclones; (v) the axial symmetry arrangement for the multiple cyclones is better than the central symmetry arrangement in terms of the uniform distribution of solid flux in the cyclone and solid inventory in the standpipe. The results provide meaningful understanding for the design, scale-up, and optimization of CFB apparatuses.
... Therefore, RTD measurement can be an efficient tool for better understanding the hydrodynamic conditions in the reactor. This information can be applied for designing reactor as well as scale-up, operation, and optimization Harris et al., 2003). ...
Thesis
In water treatment, flocculation creates large and weight flocs enough to be removed by the downstream processes of sedimentation and filtration. Among the various existing technologies, the jet clarifier is considered as an effective and compact system as it couples flocculation and clarification in a single unit. For the design of jet mixing, much experimental work has been done and many correlations have been proposed. However, these correlations are case specifics, and, to date, there is no comprehensive view for the flocculation aspect.In order to evaluate the performance of the jet clarifier for turbidity removal and understand hydrodynamics to propose the optimal operating conditions and design criteria, three different configurations of the continuous jet clarifiers are figured out. The first one is a prototype of a 3D jet clarifier studied at two scales and implemented at Samsen Water Treatment Plant, Thailand; these two reactors were designed to investigate the performance and mean Residence Time Distribution (RTD) for various injected flow rates. The results indicated no effect of reactor sizes, and a reduction of the initial turbidity (50 NTU) was achieved with an efficiency of approximately 80% under optimal conditions.Moreover, in order to better understand local phenomena controlling the efficiency of the jet clarifier, a Quasi-biDimensional (Q2D) jet clarifier was designed at the TBI-INSA-Toulouse, France allowing the application of optical metrological methods. Hence, measurements of instantaneous velocity field were performed by means of Particle Image Velocimetry (PIV). The processing of experimental PIV data highlighted a strong circulation induced by the jet in the flocculation zone. At this location, the range of velocity gradient (G) is 2 to 15 s-1 whereas the residence time decreases from 4 to 1 hour. Based on the hydrodynamic analysis, the Camp number (Gt) in the flocculation zone is shown to be constant at around 30,000 for different jet flow rates (from 11L/hr to 49L/hr). The efficiency of such the jet clarifier can thus be foreseen. Plus, measurements of the number of flocs and their size distributions were performed by means of shadowgraphy and image analysis. Thanks to a coupling between the different experimental results obtained in the Q2D jet clarifier, it was possible to relate the evolution of the number of flocs along the jet to the recirculation loop present in the flocculation zone. The relative independence of the floc size distributions on the flow rate is discussed in light of the Camp number, which can explain the efficiency of the jet clarifier in terms of flocculation.Finally, due to reactor design, the simulations using CFD code showing encouraging results were presented at the end of the manuscript. Here as well, characteristic time scales and velocity gradient were used to perform the first comparisons.
... RTD decreases with increase in screw speed and material feed rate [17]. RTD can be determined by computational particle fluid dynamic approach [18]. As it is difficult to characterize the boundary conditions for solid flow, open and closed boundary conditions are hence imposed [19][20][21]. ...
... In the literature, there exist empirical correlations for the one-dimensional (1D) analysis of the slip factor and drag force 50 and also the utilization of the slip velocity for determining the solid residence time. 51 Figure 10 presents the radial (a) and axial (b) distributions of the drag force and slip velocity of the solid phase, which are calculated as the sectional-averaged particle-scale information of the solid phase. Nonuniform distribution of the slip velocity appears along the radial direction (Figure 10a), with the appearance of large slip velocity in the central core region of the riser and small ones near the sidewalls of the bed. ...
Article
In this work, numerical simulation is carried out in a three-dimensional full-loop pilot-scale circulating fluidized bed to explore the shape effect of the riser cross section on the typical flow characteristics of the bed via the multiphase particle-in-cell (MP-PIC) method. The gas and solid phases are modeled with the large eddy simulation and Newton’s law of motion in the Eulerian and Lagrangian frameworks, respectively. The proposed model has been well validated with experimental data, followed by evaluating the typical core–annulus structure and the nonuniformity of the solid phase distributed along the radial and axial directions of the riser. Then, the particle-scale information of the solid phase distributed in different parts of the system is explored. The results demonstrate that (i) the square riser gives rise to a higher solid inventory in the standpipe owing to the stronger circulation intensity; (ii) the thickness of the solid back-mixing layer reduces along the riser height; the solid back-mixing tends to concentrate in the four corners, while it is not obvious near the sidewalls of the square riser; and (iii) nonuniform distribution of the particle-scale information of the solid phase (e.g., mass, flux, drag force, and slip velocity) can be observed. The square riser gives rise to comparatively more uniform axial mass distribution, a larger rising solid flux, larger horizontal transportation velocity between the core and annulus regions, and a larger horizontal dispersion coefficient in the riser, as compared with the corresponding ones in the circular riser.
... In the RRZ, gas back mixing decreases with the increase of gas velocity. Besides, the particle residence time also decreases, which brings high gas-catalysts contact efficiency [21][22][23] . The flow pattern in RRZ is fast fluidization where the solid concentration axial distribution exhibits from exponential shape to C shape depending on the reactor exit configuration [24][25][26] . ...
Article
To eliminate the feeding device coking deposit in commercial pyridine synthesis reactor, a coupled fluidized bed reactor is proposed in this paper. The coupled reactor is composed of a feeding zone (FZ) with novel feeding scheme, a riser reaction zone (RRZ) to increase the selectivity of pyridine over 3-picoline and a fluidized bed reaction zone (FRZ) for enough gas-solid contact time. The flow patterns are fast fluidization, fast fluidization and turbulent fluidization in FZ, RRZ and FRZ respectively. By using pilot scale experimental apparatus, the reaction performance of this coupled reactor is investigated. During 15 days of continuous operation, there is no coke deposit in the feeding device of this coupled reactor. Experimental results show that pyridine and 3-picoline product yield reaches as high as 75 %, which indicates this coupled reactor can provide the same gas-solid contact time as that in commercial reactors. The selectivity of pyridine over 3-picoline is around 2.5 and is higher than that in commercial reactors, which is around 2.2. Core-annulus model and Dispersion model are proposed to model different zones of this coupled reactor, and the predicted average deviation from experimental data is 12 %. The prediction results show that the main reactions take place at FZ and FRZ. RRZ contributes the least conversion, because of the limit by the mass transfer between the core and annulus.
... In the past few decades, the circulating fluidized bed (CFB) technique has been one of the most promising techniques in physical and chemical engineering fields because of its higher thermal efficiency over a wide range of operating conditions. 1 For a CFB boiler, heat produced from a vigorous burning rate is carried out via convective, conductive, and radiative heat transfers. Thus, the system maintains at a lower temperature, leading to a lower emission of SO 2 and NOx. 2 Although plenty of experiments have been conducted about voidage field, 3 pressure signal, 4 solid residence distribution, 5 and solid dispersion property 6 in CFBs for decades, the gas−solid hydrodynamics and heat transfer are not yet comprehensively understood due to the complex multiphase and multiphysics processes. Generally, superficial gas velocity dominates solid dispersion, bubble coalescence, collision dynamics, and heat transfer process. ...
Article
Full-text available
Comprehensive understanding of heat transfer and erosion characteristics is of significance for circulating fluidized bed (CFB) optimization. In this work, a high-fidelity simulation of the full-loop CFB is conducted using the CFD-DEM method. After qualitative and quantitative model validations, the impact of superficial gas velocity on flow dynamics, particle temperature evolution, and cyclone erosion is comprehensively explored. The results show that increasing superficial gas velocity decreases solid holdup and improves flow uniformity. Probability density of particle temperature in different CFB components shows distinctive distributions. The average particle temperature augments as the superficial gas velocity augments in fast fluidization regime, while it shows the opposite tendency in dilute phase transport regime. Meanwhile, the cyclone erosion region gradually concentrates in the inter-section of cylinder and cone parts. The results benefit for in-depth understanding of heat transfer and erosion characteristics in CFB apparatuses.
... However, in the downer, the RTD curve is narrow in width and sharp in height, implying that the back-mixing degree of particles is rather low. The flow pattern of the particles is closer to plug flow in the downer than in the riser, which is consistent with the commonly reported conclusions from simulations [38,39] and experimental results [40,41]. The near-plug-flow characteristics of the downer is beneficial for achieving high reaction selectivity and yield of desirable products. ...
Article
The interactions of gas and solids are important in determining the flow structures in the Circulating Fluidized Bed (CFB) downer. Proper description of the interphase drag model is thus crucial for simulating the downer. Considering that cluster with several numbers of particles can be present in the downer, an interphase drag model was developed by considering the dynamic equilibrium of the amount of particles in and out of the cluster in the downer, which was then incorporated into the Computational Particle Fluid Dynamics (CPFD) model. CPFD simulation results show that the predicted particle flow behaviors are in good agreements with experimental data under different operating conditions. Solids circulation rate and gas velocity can significantly affect particle flow behaviors especially in the acceleration region near the inlet. More amounts of particles flow faster in the center while a few part of particles flow slower near the wall, resulting that the particle residence time is shorter and uniform in the center and longer near the wall. The residence time distribution (RTD) curve of particles is narrower with a higher peak in the downer than in the riser. Simulation results indicate that less back-mixing of particles is present in the downer, and the flow structure in the downer approximates the ideal plug flow. The present method considering the cluster effect is reasonable in predicting the flow behavior and RTD of particles in the downer.
... In terms of operation, increasing the superficial gas velocity or decreasing the solid circulation rate, leads to better mixing in bubbling beds [21,27]. In the fast fluidisation regime, an increase in the superficial gas velocity or a decrease in the solid circulation rate decreases the mean residence time of solids, decreases the variance and increases the coefficient of variation [40]. A change in the geometry of the bed, e.g. by the addition of baffles, by modification of the aspect ratio, by adding stages, or by modifying the exit has been found to alter the mixing behaviour [21,23,41]. ...
Article
Full-text available
A model for chemical looping combustion has been developed to allow the effect of different residence time distributions of oxygen carrier particles in the air and fuel reactors to be investigated. The model envisages two, coupled fluidised bed reactors with steady circulation of particles between them. The results show that the process is sensitive to the residence time distributions, particularly when the mean residence time of particles in the reactors is similar to the time required for them to react completely. Under certain operating conditions, decreasing the variance of the residence time distribution, leads to a greater mean conversion of the particles by the time they leave the reactors and higher mean rates of reaction in the beds. In this way the required inventory and circulation rate of solids could be reduced, which would lower the capital and operating costs of a CLC process. Since the residence time distribution of solids is important, it should be taken into account when modelling or designing a chemical looping combustion process, e.g. by using a tanks-in-series model. This work indicates that if the number of tanks, N ≤ 5, knowing N to the nearest integer is generally sufficient, unless a high degree of accuracy is needed. As N increases, the sensitivity of the coupled system decreases, so for N > 5, knowing the value to the nearest 5 or 10 tanks is sufficient. This is valid whether N is the same or different in the two reactors. Chemical looping combustion is one example of a reactor-regenerator system, so the results are also relevant for other processes of this type, such as fluidised catalytic cracking.
... Besides, the slip velocity between gas phase and catalysts 36,37 increases the catalysts residence time. Three types of models are proposed to predict the catalysts residence time distribution in circulating fluidized bed reactors, 38 which are the dispersion model, core-annulus/incremental tracer balance model, and stochastic model. In this paper, the model proposed by Wei 39,40 is employed to predict the catalysts particle residence time distribution. ...
Article
Pyridine has been generally synthesized by aldehydes and ammonia in a turbulent fluidized bed reactor. In this paper, a fast fluidized bed reactor was proposed for pyridine synthesis. Experiment result shows that the yields of pyridine and 3-picoline decrease while the selectivity of pyridine over 3-picoline is increased. A model was proposed to predict the performance of the fast fluidized bed reactor, the average prediction deviation is 6%. The influence of mass transfer, heat transfer and backmixing of gas phase is represented by a modification factor, and the mean value of this modification factor is 0.75 within the experiment operating conditions. By model prediction, the reaction should be terminated when the critical point of R2 is reached to avoid over reaction. To optimize the pyridine and 3-picoline product yield and minimize coke product yield, the reaction temperature should be kept around 723 K.
... Such models have typically been used to describe non-ideal fluid flow [15]. RTD models have been applied successfully to powder processing, such as twin screw co-rotating mixers, circulating fluidized beds, and continuous horizontal blenders [6,16,17]. ...
Article
Full-text available
Residence-time-distribution (RTD)-based models are key to understanding the mixing dynamics of continuous manufacturing systems. Such models can allow for material traceability throughout the process and can provide the ability for removal of non-conforming material from the finished product. These models have been implemented in continuous pharmaceutical manufacturing mainly for monitoring purposes, not as an integral part of the control strategy and in-process specifications. This paper discusses the steps taken to develop an RTD model design space and how the model was statistically incorporated into the product’s control strategy. To develop the model, experiments were conducted at a range of blender impeller speeds and total system mass flow rates. RTD parameters were optimized for each condition tested using a tank-in-series-type model with a delay. Using the experimental RTD parameters, an equation was derived relating the mean residence time to the operating conditions (i.e., blender impeller speed and mass flow rate). The RTD parameters were used in combination with real-time upstream process data to predict downstream API concentration, where these predictions allowed validation across the entire operating range of the process by comparison to measured tablet assay. The standard in-process control limits for the product were statistically tightened using the validation acceptance criteria. Ultimately, this model and strategy were accepted by regulatory authorities.
Article
In this paper, a three-dimensional compressible mathematical model of circulating fluidized bed (CFB) combustion was established, by using the Euler-Lagrangian method, based on the Computational Particle Fluid Dynamics (CPFD). The model considers heat and mass transfer, the phase chemical reaction. The distribution of temperature, gas composition, particle residence time and velocity of the co-combustion of oil shale retorting solid waste with cornstalk particles in the CFB were obtained, by using simulation calculation under a variety of operating parameters. The results of the model are established and analyzed. The optimal operating conditions of co-combustion of oil shale retorting solid waste and cornstalk particles in the CFB were obtained, which provide meaningful values for the experiment, design and working condition of the CFB.
Article
A pneumatic transport reactor can be used for continuous carbon capture processes using a dry sorbent because it can handle large quantities of flue gas. To design efficient reactors, it is necessary to understand the internal characteristics of a reactor with a complicated gas-solid flow. Computational fluid dynamics using an Eulerian-Eulerian approach was adopted to simulate gas-solid two-phase flow to better understand the gas-solid behaviors and heat transfer characteristics in a pneumatic transport reactor. Numerical simulations were used to analyze the pressure difference, solid mass flux, and heat transfer coefficient. The results showed that the gas-solid behavior was unstable and that localized particle flow affects the heat transfer characteristics. The degree of particle mixing near the solid return inlet was lower than that at greater heights within the reactor; in the inlet region, the heat transfer coefficient is not uniform in accordance with the non-uniformity of solid particle behavior.
Article
Cross-flow bubbling fluidized beds (BFBs) have been widely used in dual fluidized bed systems such as chemical and heat looping. Understanding the residence characteristics of solids in such system is important for better design and optimization of reactors. Previously we experimentally measured the residence time distributions (RTDs) of sands in a cross-flow rectangular BFB by using coal particles as tracer. A computational investigation of solids RTD using multi-fluid Eulerian method combined with the species transport equation showed that the RTD of sands could be correctly represented by that of coal particles, and the simulation results well agreed with experimental data. Parametric studies demonstrated that, under the considered operation conditions, the influence of tracer injection time period on the predicted solid residence times was nearly ignorable. Simulation results revealed that in the investigated cross-flow BFB the solids RTD is closely related to solids inventory and solids flux. Through proper data processing, it was found that the descending part of solids RTD profile can be uniquely fitted by an empirical exponential function. A semi-empirical approach was thus developed and further validated, for the first time in the literature, to predict the entire profile of solids RTD, in which the ascending part of solids RTD profile is obtained through CFD simulation whereas the descending part is given by the fitted empirical exponential function.
Article
For non-catalytic gas–solid reaction, it is desirable to match the mean residence time (MRT) of particles and complete conversion time (tc) in a fluidized bed. In this study, the MRT differences (MRT ratios) between the coarse particles and the fine particles were investigated in a continuous fluidized bed with a side exit by varying the superficial gas velocity, feed composition and particle size ratio. The results show that the MRT ratio increases firstly and then decreases with increasing the gas velocity. By controlling the gas velocity and the feed composition of coarse particles, the MRT ratio can be modulated from 1.8 to 10.5 at the gas velocity of 1.0 m/s for the binary mixture with the size ratio of 2.2. The MRT ratio can reach to ~ 12 at the gas velocity of 1.2 m/s for the particles size ratio of 3.3. The present study has endeavored to obtain fundamental data for an effective plant operation to meet the need of synchronously complete conversion of particles with different sizes during the film diffusion controlling reaction.
Conference Paper
Efficiency of a photoreactor depends on the irradiation dose. Fluid residence time distribution (RTD) reflects hydrodynamic behavior of the flow. A computational model was built on a base and fitting a previous radiotracer experiment. Results of three simulations for three different configurations and height flow rate are presented and discussed below. This paper shows usefulness of CFD modeling as an imaging tool, which can be used to retrieve detailed, local information about the flow.
Article
The rotary kiln is an essential device in chemical and metallurgical industries, with applications in a wide range of solids manufacturing processes. In particular, in the preparation of industrial chemical catalysts, the kiln has become a popular reactor for continuous calcination of catalysts ranging from millimeter-sized extrudates to micron-sized powders. As granular and powder flow behaviors are difficult to characterize, the design and scale-up of rotary calcination processes are often performed empirically. The goal of this research is to improve the fundamental understanding of powder flow in rotary kilns to aid in optimization of the continuous calcination process. For successful calcination to occur, the residence time of the particles must exceed the time required for heating and subsequent treatment. For uniform treatment of the feed, the particles must also exhibit low axial dispersion. In this work, the mean residence time and axial dispersion coefficient for a cohesive fluid catalytic cracking powder were determined in a pilot plant kiln by measuring the residence time distribution. This study utilized a pulse test developed by Danckwerts. Results were fit to the Taylor solution of the axial dispersion model and compared to the Sullivan prediction for mean residence time. It was found that the mean residence time decreases as the feed rate, kiln incline, and rotation rate increase. It was also found that the axial dispersion coefficient increases with speed of rotation and angle of incline. However, the axial dispersion coefficient decreases as the feed rate is increased.
Article
Full-text available
For continuous manufacturing of pharmaceuticals, understanding the dynamics of how a material flows through the process is critical with respect to the development of a control strategy for product quality assurance. Such understanding of the process dynamics can be obtained by characterization of the residence time distribution (RTD). The RTD for a process is not fixed and can vary due to changes in operating conditions or physiochemical properties of the blend. As such the RTD needs to be evaluated over the range of operating condition that can impact process dynamics (e.g. throughput, impeller rotation rate etc.). In this paper, we demonstrate that the dimensionless RTD (normalized with respect to the mean residence time) is invariant with throughput and impeller rotation rates under certain conditions for the two continuous direct compression processes. We present a case study to illustrate the utility of this relationship for predicting the process dynamics at different operating conditions (i.e., throughputs) and evaluating the impact of variations in the process dynamics on the control strategy for a continuous direct compression process.
Article
In this computational study, we model the mixing of biomass pyrolysis vapor with solid catalyst in circulating riser reactors with a focus on the determination of solid catalyst residence time distributions (RTDs). A comprehensive set of 2D and 3D simulations were conducted for a pilot-scale riser using the Eulerian-Eulerian two-fluid modeling framework with and without sub-grid-scale models for the gas-solids interaction. A validation test case was also simulated and compared to experiments, showing agreement in the pressure gradient and RTD mean and spread. For simulation cases, it was found that for accurate RTD prediction, the Johnson and Jackson partial slip solids boundary condition was required for all models and a sub-grid model is useful so that ultra high resolutions grids which are very computationally intensive are not required.. We discovered a 2/3 scaling relation for the RTD mean and spread when comparing resolved 2D simulations to validated unresolved 3D sub-grid-scale model simulations.
Chapter
It has been realized in recent decades that a proper investigation of gasification reactor requires the detailed information over the entire flow field, as well as time, at multiple scales. Such detailed information needs the use of sophisticated measuring techniques with capability to provide the required information over the entire flow field, as well as time, at multiple scales. Aside from the mean velocities and volume fractions, information about the flow fluctuations or dynamics (quantified in terms of cross-correlations and auto-correlations) is also desirable. In addition, it is preferable if such techniques are amenable to automation to reduce extensive human involvement in the data collection process. While such data are “stand-alone” sets of information, which can be used for design and scale-up strategies, it also provides information that is crucial to establish the validity of conventional models like phenomenological flow models describing residence time distribution (RTD), as well as more recent and sophisticated models like those based on computational fluid dynamics (CFD). In fact, it almost seems imprudent to validate CFD predictions on overall holdup and flow rates, because these spatial integrals of point properties are simply averages of a complete flow field that a CFD code is designed to and claims to compute. Thus, fair validation must involve validation at multiple scales, for which one needs experimental information also at multiple scales (and not just spatial and temporal averages). Several experimental techniques have been reported in past to quantify the flow field in gas–solid gasification reactors, with each technique having its own advantages and disadvantages. In this chapter, details of pressure, solid velocity, solid fraction, and RTD measurement techniques will be presented. Techniques will be divided majorly in two types, invasive and non-invasive. The postprocessing methods for each technique, advantages, and limitations will be discussed. Finally, some of the recent findings on gas–solids circulating fluidized bed using radioactive particle tracking (RPT) technique will be discussed in detail to explain the use of the experimental techniques for design and scale-up of these reactors.
Article
Suspended Heating Surfaces (SHSs) in large-scale circulating fluidized bed (CFB) boilers have a significant impact on internal (Gs,i) and external (Gs,o) circulating solid flow rates that is important for heat transfer. In this paper, the distribution of Gs,i and Gs,o in a 650 × 680 × 3500 mm³ CFB apparatus was simulated and verified by experimentations. On this foundation, the effects of SHSs structural layouts on Gs,i and Gs,o were investigated in a 350 MW supercritical CFB boiler. Results showed that Gs,i increased by 65.8% as SHSs number increased from 0 to 20, whereas Gs,o decreased by 17.8%, so Gs,o/Gs,i decreased correspondingly. Meanwhile, increasing SHSs height had a greater improvement on Gs,i than width, which would be more conducive to the heat transfer near SHSs. Finally, empirical formulas of Gs,o, Gs,i and layout parameters of SHSs were developed, and the research results can provide meaningful reference for the operation optimization of large-scale CFB boilers.
Article
Bubbling fluidized bed is extensively used in petrochemical and other process industries. The performance of the bubbling fluidized bed can be improved through the placement of various internals, among which structured packing is proposed to be a promising type of internal. To address the advantage of such packing over the bubbling fluidized bed without any internals, three-dimensional Computational Fluid Dynamics (3D CFD) simulations were performed to compare the gas-solids flow hydrodynamics and bubble behaviors in the bubbling fluidized beds with and without structured packing. A method of capturing and characterizing the bubble was proposed based on the simulation results. Experimental data from the pilot-scale bubbling fluidized bed was used to examine the accuracy of the CFD model. Simulation results show that the distributions of solids hold-up and velocity were more uniform in the bubbling fluidized bed with structured packing than that without internals. The bubble diameter and the bubble number in the packed bubbling fluidized bed were less than half and about 10 times, respectively, of those in the empty bubbling fluidized bed, indicating that much more amounts of smaller bubbles were present in the packed bed compared to the empty bed. It was found that the improved gas content in the fluidized bed at higher gas velocities was mainly due to the increased bubble diameter in empty bubbling bed while it was mainly due to the increased bubble number in packed bubbling bed. The degree of gas back-mixing was reduced slightly while the degree of solids back-mixing was reduced by around 50% in the packed bubbling fluidized bed compared to those in the empty bubbling fluidized bed. Simulation results indicate that the bubbling fluidized bed with structured packing can provide larger interfacial area between gas and solids and can significantly reduce solids back-mixing, which is beneficial for improving the performance of the bubbling fluidized bed.
Article
Circulating fluidized bed (CFB) has been commonly operated with polydisperse particles with a wide particle size distribution (PSD), which significantly affects solid transportation and the resulting thermochemical performance. Dense gas-solid flows in a pilot-scale CFB riser are simulated via a multiphase particle-in-cell (MP-PIC) coupled with a heterogeneous drag model. Gas-solid hydrodynamics together with the particle-scale information with a wide PSD is discussed. The results show that in the riser, the non-uniformly spatial distribution appears with the presence of a core-annulus structure and a dilute-upper dense-bottom profile. The wide PSD gives rise to the axial segregation phenomenon with the accumulation of large particles in the bottom region. Enlarging the superficial gas velocity and PSD width respectively decreases and increases the segregation intensity of solid phase. As compared with gas velocity, the PSD width performs limited impacts on solid concentration, particle-scale information along the axial direction, and solid residence time. The horizontal solid dispersion coefficient is at the scale of 10⁻³ m²/s, while the vertical dispersion is at the scale of 1 m²/s. For a wide PSD, particles with larger sizes have larger slip velocities but smaller dispersion coefficients in the riser.
Article
The hydrodynamic and thermochemical characteristics in the coal-direct chemical looping combustion (CLC) process is studied by a self-developed computational fluid dynamics - discrete element method (CFD-DEM) approach featuring particle-scale simulations of collisions, heat and mass transfer, drying process, coal pyrolysis, gasification, and heterogeneous reactions between gas species and oxygen carriers. A polydisperse drag model is adopted to accurately calculate gas-solid interactions. After comprehensive model validations, the flow pattern, pressure drop, gas products composition, particle temperature, combustion efficiency, and solid residence time (SRT) are qualitatively and quantitatively analyzed. The results show that increasing the coal feeding rate slightly improves the temperature of coal particles and accelerates the release of moisture in coal particles. Finer oxygen carriers promote the conversion of intermediate gas products into CO2 and H2O, therefore improving the combustion efficiency in the CLC process. The SRT distribution with an early peak and a long tail is profoundly revealed under different operating parameters. A dual-side coal feeding arrangement remarkably improves the uniformity in the CLC system.
Article
Full-text available
Solids velocimetry based on cross-correlation algorithm has been widely used in multiphase systems. Recent studies have revealed that its measurement qualities depend critically on fluidization regimes, and the underlying mechanism was supposed to be the competition between solids convection and mixing. In this study, a method was proposed to analyze the local solids residence time distribution (RTD) from the results of Eulerian-Lagrangian simulations. Systematic simulations were then conducted to acquire the local solids RTDs in various gas-solids fluidization regimes. Afterward, the local Péclet numbers were quantified. It was found that the measurement qualities are intimately related to the local Péclet numbers and the solids convection-mixing competing mechanism. Present study revealed the underlying mechanism of the success and failure of cross-correlation based solids velocimetry, and proved that the cross-correlation algorithm is ideal for solids convection-dominated systems or regions in a system, but caution is needed when solids mixing is important.
Article
A 1 MWth pilot-scale chemical looping combustion (CLC) system is numerically simulated in the Eulerian-Lagrangian framework. Multiple chemical reactions including coal pyrolysis, char gasification, volatiles combustion, and oxidation of oxygen carriers are considered in the model. The influence of solid fuel feeding rate on the performance of the CLC system is analyzed and the parcel-scale information is obtained. The numerical results are synthetically compared with the experimental measurements. A non-uniform distribution of the gas species is observed in two reactors. Meanwhile, the relatively uniform parabolic profiles of solid holdup are displayed at different radial directions while the profiles of CO2 concentration at different radial directions are quite asymmetrical and non-uniform. Due to the strong oxidation reaction, the temperature profile in the AR is higher than that in the FR. The lognormal probability plots of solid residence time present an early peak with a long tail. Increasing coal feeding rate will promote the formation of product gas yields and solid temperature but suppresses the transport of solid particles in the CLC system.
Article
In this work, the cluster dynamics in the riser of a pilot-scale circulating fluidized bed with the presence of wide particle size distribution (PSD) are numerically simulated via the multiphase particle-in-cell (MP-PIC) approach. Specifically, the gas and solid motions are tracked in the Eulerian and Lagrangian frameworks, respectively. The gas-solid numerical model is firstly validated with the experimental data. Then, the spatial distribution of particle clusters together with the cluster dynamics (velocity, length, aspect ratio) in the bed operating with different parameters are explored. The results demonstrate the ability of the MP-PIC model on predicting the important features of the cluster in the riser. Clusters mainly distribute in the lower part of the riser. The closer of the cluster to the riser wall, the larger vertical length, and larger aspect ratio appear. Moreover, the presence of riser outlet and solid inlet slightly affects the spatial distribution, the size and aspect ratio of clusters in the riser. The superficial gas velocity has an obvious impact on the cluster volume and the cluster vertical length, while the PSD width significantly affects the horizontal length and the aspect ratio. The results obtained in this work provide meaningful insights regarding the cluster dynamics in the riser, which will be beneficial to the deep understanding of this specific structure in the practical operation.
Article
Solid mixing process plays a crucial role in evaluating the reaction performance of multiphase reactor. However, there still lacks the systematic research on solid mixing for the multistage circulating fluidized bed with enlarged sections. Solid mixing mechanism in a multistage riser is investigated numerically in this study. Some individual characteristics of solid mixing are found. It is shown by analyzing residence time distributions that solid mean residence time and the degree of solid dispersion increase with the increasing solid flux or the decreasing superficial gas velocity. Nevertheless, particle diffusion coefficient increases as solid flux decreases or superficial gas velocity increases. Moreover, the diffusion coefficient in enlarged sections decreases with the increasing height, which is contrary to the variations in traditional risers. Compared with the traditional riser, particle diffusion is weak, but overall solid mixing is enhanced in the multistage riser. Furthermore, two mixing patterns that gas-solid flow develops towards the plug-type flow in the center region of riser and the mixed-type flow in the enlarged section are identified. In addition, the dominant recirculation time of solids in the enlarged section is predicted as 1.5 s under examined cases.
Article
Olive oil production urges the energy companies to exploit the potential of the residues as biomass fuels for clean energy production. Circulating fluidized bed (CFB) gasifier gains global interest within these several years due to its promising solution for converting biomass material to renewable energy. However, the multi-physics processes and multiscale structures inside the gasifier impede the experimental measurements. As an alternative, a reactive multiphase particle-in-cell approach is developed in this work to simulate the gasification process of olive oil waste in a pilot-scale full-loop CFB gasifier. After the model validation, the heat transfer properties of solid phase and the effect of size-induced segregation on solid thermochemical properties are explored. The results show that the low-temperature biomass particles injected highly influence the spatial distribution of solid temperature and solid heat transfer coefficient (HTC). A larger particle size gives rise to a higher HTC. Moreover, the temperature and HTC of biomass are larger than that of sand. A wide and narrow residence time distribution of sand in the riser and cyclone can be observed. The scale of the biomass dispersion is respectively larger and the same as that of sand in horizontal and vertical directions. Particles in the dense region have a smaller HTC, Reynolds number, and temperature. Superficial gas velocity improves the uniformity of solid distribution. Elevating the gas velocity and initial bed temperature enhances solid dispersion intensity.
Article
This work presents a novel technique with fast response for Residence Time Distribution (RTD) measurements in gas-solid unit operations (e.g., fluidized bed reactors). This technique is based on an optical method which eliminates the requirement of knowing the velocity and concentration profiles at the exit section of the system. Experiments were carried out with SiC particles and a phosphorescent pigment used as a tracer. A concentration measurement system was developed to measure the tracer concentration in SiC/pigment mixtures. The corresponding pigment concentrations were evaluated at the bottom of this system using a photomultiplier. The pigment concentration was derived from the integral of the signal intensity received by the photomultiplier. Then, a calibration curve was established which provided the empirical relationship between the integral and pigment concentration. In order to validate this RTD measurement technique, a series of experiments was performed in a bubbling fluidized bed and the effect of the bed height was studied. It was shown that the experimental RTD curves were in good agreement with the theoretical RTD of bubbling fluidized beds. This solids RTD measurement technique can be used to provide a better understanding of the hydrodynamics of complex solids unit operations.
Article
Traditional methods for measuring the residence time distribution (RTD) of particles in a fluidized bed are complex and time-consuming. To this regard, the present work proposes a new measurement method with remarkable efficiency based on digital image analysis. The dyed tracers are recognized in the images of the samples due to the difference of colors from bed materials. The HSV and the well-known RGB color space were employed to distinguish the tracers. By enhancing the Saturation and the Value in HSV and adjusting the gray range of images, the recognition error is effectively reduced. Then the pixels representing the tracers are distinguished, based on which the concentration of the tracers and RTD are measured. The efficiency, accuracy and repeatability of the method were validated by RTD measurements experiments. The method is also fit for distinguishing the target particles from multi-component systems consisting of particles of different colors.
Article
Biomass gasification process in a three-dimensional pilot-scale circulating fluidized bed (CFB) gasifier is numerically simulated via the multiphase particle-in-cell approach. The physical and thermochemical properties of gas and solid phases (i.e., temperature, density, and viscosity, pressure, and species) in the gasifier under high temperature are comprehensively explored. The results show that the size- and density-induced segregation causes the accumulation of biomass particles in the standpipe. Elutriation of biomass particles leads to the presence of low-temperature endothermic zones and high-temperature exothermic regions in the riser. The large contents of combustible gas species exist in the left part of the riser. Continuous decrease of O2 is due to gas species oxidation, while the reduction of N2 results from the release of gas species. The dense region has a large content of N2 but a small content of combustible gas species. Large turbulent gas viscosity appears in the riser. Enlarging the gas velocity decreases the solid concentration but increases the gas temperature. Gasification temperature has an insignificant influence on the solid concentration and the contents of combustible gas species. Based on the relationship between the solid transportation and gas–solid thermochemical properties, the current work provides a guideline for the optimization of CFB gasifiers, for example, the loop seal is advised to be optimized to enhance the gasification performance of this kind of apparatus.
Article
This work studies the particle exchange rates in horizontal fluidized beds equipped with different weir designs between compartments. These particle exchange rates provide information on the axial dispersion of the solid material within the process. For this purpose discrete particle modelling (DPM) was used to determine the particle exchange on microscopic level. This method uses a coupled CFD-DEM approach to observe particle dynamics in a fluid field. The model was validated against exchange rates in a lab-scale setup as determined by Particle Tracking Velocimetry (PTV) with very good quantitative agreement, showing the suitability of the method for the evaluation of weir designs. Simulations were performed for different weir designs and under variation of the hold-up mass, the feed rate and gas velocity to predict their transport behavior in a pilot-scale 3D horizontal fluidized bed. The results indicate that the solids transport behavior is strongly dependent on the used weir design and the main driving force for the particle transport that can be influenced by the process conditions. The installation of weirs between two compartments induces a transport resistance, while the base type without the installation of a weir between the two chambers represents the fastest possibility for mixing the particles of a two-compartment system. It has been observed that the general trend shows higher particle recirculation rates for the overflow weir and base configuration (no weir), whereas the underflow and sideflow weir applications improve the solids transport through the horizontal fluidized bed.
Article
Powder flow and heat transfer in rotary calciners are difficult to predict, which can result in reduced product quality. This work attempts to improve our understanding of powder flow in a rotary calciner. We compare experimental results to existing models. In this study, a moderately cohesive fluid catalytic cracking (FCC) powder was passed through two pilot plants rotary calciners with and without dams. Residence time distributions were measured, and the resulting mean residence times and axial dispersion coefficients were compared to predictive models. It observed that increasing the calciner incline and speed of rotation resulted in decreased mean residence time (MRT) and feed rate only had a small effect on MRT and axial dispersion. Increasing the height of dams increases the mean residence time and lowers the dispersion coefficient. Bed depth profiles at various operating conditions and geometries, with and without dams, were also determined and compared to available models.
Article
This paper reports on experiments conducted with a cold flow model utilized for the investigation of the particle residence time distribution and mixing characteristics in a bubbling fluidized bed with continuous solids exchange. The investigated system is of a rectangular cross section (0.4 × 0.2 m) with a bed height of 0.17 m. A measurement device based on an alternating current bridge circuitry coupled with lock-in amplifier technology was built in the scope of quantifying the solids residence time distribution, whereby a pulse-injected ferromagnetic tracer creates the input signal. The implementation of a profound mathematical routine ensures the reproducible calculation of the particles mean residence time and characteristic values describing particle mixing phenomena. Therefore, the E-curve was modeled by mathematical convolution of the exit age distributions available for an ideally mixed continuous stirred tank reactor and a plug flow reactor with axial dispersion. It is shown that the in-bed mixing is highly dependent on the fluidization rate as well as on the solids circulation rate. Albeit the lowest superficial gas velocity equals a fluidization number of 4.7, the formation of dead spaces and short-circuit flows was observed under these conditions. Axial dispersion coefficients in the range of 5·10-3 to 7·10-1 m² s⁻¹ were obtained.
Article
The understanding of gas–solid flow is of great importance for various high-density circulating fluidized bed (HDCFB) practices. In this work, the effects of exit geometry on the gas–solid flow behavior in an HDCFB riser were numerically studied by using the computational particle fluid dynamics (CPFD) method. The simulated results were compared against experimental data on bed pressure and solid volume fraction distributions along the riser and a satisfactory agreement was found between them, which indicated the capability of the CPFD for the HDCFB riser simulation. Thereafter, simulations were conducted for the HDCFB risers with six different exit geometries and comparisons of hydrodynamics characteristics between each other were made. It was found that exit geometry significantly affected the axial solid volume fraction distribution through the whole riser. The abrupt exits gave rise to higher SRT than the smooth ones due to the solids back-mixing and packing effects. The angle between the exit duct and the horizontal plane also has a great influence on the gas–solid hydrodynamics and the abrupt exit with 15° was found to be appropriate for HDCFB riser due to its promotion on solid dispersion and reduction in solids back-mixing and packing.
Article
Fluidised bed technology is a very common industrial process and has a wide range of applications in process engineering. Though indirectly, a common way to infer the end product uniformity is through investigating the solid residence time distribution (RTD) in the system. A good understanding of the particle RTD is crucial for the proper design, optimisation, and scale-up of a fluidisation process. Flexible ribbon particles have a very different flow behaviour in a fluidised bed compared to that of spheres which in turn influences the RTD of the particles. However, this issue is still not very well addressed to date. In this paper, a double chain model is adopted to model cut tobacco strips in a fluidised bed riser. The spring constants of this chain model are calibrated, and the model is validated against experimental datum. Finally, the particle RTD as a function of different process parameters is investigated.
Article
The full-loop gas-solid hydrodynamics in a pilot-scale circulating fluidized bed (CFB) is numerically investigated by using multi-phase particle-in-cell (MP-PIC) method. In this method, fluid motion is resolved under Eulerian framework while particle motion is tackled under Lagrangian framework. The numerical results agree well with experimental data, demonstrating the reasonability of the MP-PIC method in simulating dense fluid-particle systems. The main features of gas-solid flows in the full-loop CFB are qualitatively predicted, with a typical core-annulus flow structure in the riser. A large pressure gradient of gas phase in the bottom region and top region of riser reflects the turbulent and chaotic gas-solid flows in these two regions. Enlarging superficial gas velocity leads to a more non-symmetrical pattern of axial gas velocity, indicating the enhanced influence of recycle structure on gas-sold uniformity under the higher superficial gas velocity. The solid residence time (SRT) distribution in the riser exhibits a feature of early peak and a long tail, which can be well fitted by a lognormal probability distribution function. Enlarging superficial gas velocity gives rise to a decreased mean value and narrower distribution of the solid residence time in the riser, reflecting the depressed solid back-mixing intensity in this condition.
Chapter
This chapter deals with the practical aspects of designing, fabricating, operating, and understanding experimental fluidized beds. It discusses some of key components to be considered in designing and building a fluidized bed system. The main component of an experimental fluidized bed system is the fluidization column. Common materials used to fabricate these columns are metals (e.g. mild steel or stainless steel) and acrylic (e.g. Plexiglas), depending on the purpose of the experiments. Fluidization behaviour can be characterized qualitatively in terms of “flowability” and “fluidizability.” Both parameters amalgamate various properties and conditions that influence how bulk solids behave. The chapter presents some practical measurements to illustrate how these properties can be used to describe changes in particle properties. It summarizes various invasive and non‐invasive measurement systems. Each of these types of measurement requires careful calibration prior to conducting experiments. Finally, the chapter describes the operation of fluidized beds.
Article
Accurate prediction of transport phenomena is critical for VPU reactor design, optimization, and scale-up. The current study focused on the validation and application of a multiphase CFD model within an open-source code MFiX for hydrodynamics, temperature field, and residence time distribution (RTD) simulation in a non-reacting circulating fluidized bed riser for biomass pyrolysis vapor phase upgrading (VPU). First, an Eulerian-Eulerian approach three-dimensional CFD model was employed to simulate the pilot-scale VPU riser on the supercomputer Joule. Excellent quantitative agreement between experimental and simulated results was achieved for pressure drops and temperature field in a range of operating conditions. Then the validated multiphase CFD model was applied to predict gas and solid residence time distributions (RTDs) since prediction and analysis of RTD is an important tool to study the complex multiphase flow behavior and mixing inside chemical reactors. The predictions show that solid mean residence time is 3.5 times the gas residence time; the solid RTD is more sensitive to the process gas flow rate than the solids circulation rate.
Article
We report a joint experimental, numerical and theoretical study of particle residence times in a novel vortex-based vessel for thermal processing of suspended particles. The tracer pulse-response method, in which the particle phase itself is employed as the tracer, is used to measure the particle residence time distribution (RTD) within a laboratory-scale model of a class of Solar Expanding Vortex Receiver-Reactor (SEVR). The operating parameters of particle size, gas volumetric flow rate and inlet velocity were systematically varied to assess their influence on the particle RTD and to determine the mechanisms controlling the behaviour of the two-phase flow in the SEVR. The particle RTD behaviour is also described by a compartment model consisting of a small plug flow reactor followed by a series of two interconnected continuously-stirred tank reactors (CSTRs).
Article
Computational Particle Fluid Dynamics (CPFD) simulation was performed to study coal pyrolysis in a Circulating Fluidized Bed (CFB) downer containing binary particles (coal particles and heat carrier particles). A suitable drag model, which considers the cluster effect by assuming dynamic equilibrium for particles in and out cluster, was incorporated into the CPFD model, which shows acceptable accuracy in describing the hydrodynamics of binary particles in the downer. The predicted mixing index of temperature was also in good agreement with experiment, indicating the model can also be used to study heat transfer between binary particles in the downer. The coal pyrolysis kinetics were further incorporated into the model. Simulation results demonstrated that the rapid mixing of coal and heat carrier near the downer inlet was beneficial for the rapid temperature increase of coal particle. The predicted coal pyrolysis behaviors further demonstrated that the gas-solids can flow uniformly with short residence time and low back-mixing in the downer, which facilitates the rapid pyrolysis of coal. It was concluded that the CPFD modeling can provide reasonable simulation results with respect to hydrodynamics, heat transfer and chemical reactions in the downer for coal pyrolysis.
Article
A series of ideally mixed fluidized beds is used to obtain a narrow product size distribution in industrial-scale applications. Thereby, the fluidized bed compartments or stages are divided from each other by weirs with defined openings (slots) for particle exchange. The knowledge of particle exchange streams and residence times in different compartments, respectively, is crucial for the process design and optimization of such processes. This particle exchange behavior between different compartments in 3D-fluidized bed systems cannot be easily evaluated by experiments.
Article
This book provides an introduction to mixing phenomena in continuous flow systems and particularly to those mixing effects that are important in the design of chemical reactors. The book is intended for practicing engineers and scientists who must design and analyze complex flow and reaction systems with limited data. For the relatively few situations where the system is well understood, the methods presented provide a unifying, conceptual framework. Far more often, they allow the design to proceed or experiments to be interpreted when total systems knowledge is incomplete. The book is written in the mathematical language of science, but the applications are very real with a dozen examples to be found in any large chemical plant or oil refinery. Nearly all the mathematics is accessible to a person with an undergraduate degree in chemical engineering, and most of the material can be read and understood by anyone with a good technical background. Although primarily conceived as a self instructional guide for the practicing engineer, the material presented has also been used for a variety of formal and informal courses, both in academic and industrial settings. The problem sets at the end of each chapter reflect examples, homework assignments, and examinations that have been used in such courses. As a college text, the book is best suited for a graduate level, special topics course on mixing or reactor design. It may also be used as a supplementary text for courses in chemical reaction engineering when the instructor chooses to emphasize industrial design methods rather than kinematics and mechanisms. (from preface)
Article
A unique phosphor tracer technique was used to allow experimental measurement of residence time distribution (RTD) of solids in the circulating fluidized bed. The instantaneous input of light impulse and the detection of the light emission by the excited phosphor particles permit on-line continuous determination of solids tracer concentration with time at different radial and axial positions within the riser and downer. The measured RTD curves in the downer show that less axial solids mixing takes place in the downer than that in the riser. The two-peak distribution of RTD curves in the riser is clearly seen in the experiment.
Article
This paper summarizes several years of testwork in laboratory and bench scale circulating fluidized bed reactors (CFB) using the thermal decomposition of NaHCO//3 as model reaction. Selected results are presented concerning the gas and solids phase fluid dynamic behaviour and the influence of the chemical reaction on CFB operating conditions.
Article
Tests were conducted in a 30.5-cm-diameter riser with 76 μm FCC catalyst to determine the degree of gas/solid contacting in the “fast-fluid bed” regime. Data were obtained at superficial gas velocities of 3.7 to 6.1 m/s and solid fluxes of 98 to 195 kg/s-m². The conveying gas was air at ambient temperature and 103 kPa. Data were obtained on axial pressure profiles, particle velocity and mass flux as a function of radial and axial position, radial and axial gas dispersion, and solids residence time distributions. Density and voidage profiles were calculated from these data.
Article
Residence time distributions have become a standard tool of chemical reactor analysis. One reason for this popularity is the usefulness of residence time data for prediction of reaction yields. A second reason is the apparent ease with which residence times can be measured using inert tracers. In open systems, however, such transient response experiments do not give the residence time distribution but give other distribution functions which are not only directly useful for predicting reaction yields. In the general case, residence times cannot be determined solely through the use of inert tracers. Instead, they must be deduced by deconvolution of reaction yield data or by mathematical manipulation of a detailed system model. Even the reaction yield approach is not completely unambiguous since the presence of a reaction can sometimes affect the residence time distribution for reactant molecules. The axial dispersion model is used to illustrate these principles.
Chapter
Experimental studies of risers have shown that these systems are often characterized by complex flow phenomena such as non-uniform spatial distribution of particles, large slip velocities between the phases, and the existence of several possible pressure gradients and solids holdups for specified values of gas and solid flow rates (see Chapter 2 — Hydrodynamics). The particle concentration profile influences the distribution of residence times of particles and may lead to recirculation of particles against the direction of their net motion. These effects are critically important in predicting the conversion in systems in which the particles react chemically with the gas or catalyse a gaseous reaction. Hence, an accurate understanding of the mechanism responsible for the cross sectional distribution of solids is necessary to predict the performance of these systems. Empirical correlations have generally proven unsuccessful as they are typically limited to the database used to develop them and ignore the effect of radial non-uniformity of the basic variables. However, fundamentally based models have made some progress in this direction. These models can be used to predict how various parameters vary as system conditions change. This is very important, especially to determine the effects of scale-up, design and optimization. A review of these models is the subject of this chapter.
Chapter
Circulating fluidized beds (CFB) exhibit very complex hydrodynamics, caused by interactions between the gas and solid phase. The motions of gas and solids are driven by many mechanisms that are difficult to identify and to describe.
Article
The velocity and the axial dispersion of solids within a bed (50 mm in inner diameter, 3600 mm in height) with high gas velocity were analyzed by impulse response of tracer particles. FCC-particles (“group A” powder according to Geldart classification) were employed for the present investigation.Local particle velocity at the axis of the bed increased with increased gas velocity, while the fraction of solids decreased. The axial dispersion coefficient within the bed ranged from 1 to 900 cm2/s. This result suggested unsteady and random behaviour of solids within the fast fluidized bed. Local slip velocity at the axis of the bed was mostly larger than the terminal velocity of a single particle.
Article
A set of stochastic mathematical models have been developed to simulate the residence time distribution of solids in the riser of a circulating fluidised bed. The models simulate the motion of single particles moving up and down the riser using a Markov chain. Two models are presented in detail: (i)a core-annulus solids interchange model, and(ii)a four zone model that follows from the fast fluidised bed hydrodynamic profile. Both discrete and continuous time versions are presented. Each model incorporates different sections to account for the different flow regimes that exist within a riser. Simulations are linked to actual experimental conditions using local particle transfer rates between each model section. Simulations are able to predict the influence of changes in the solids flux, riser height and riser exit geometry. The influence of core-annulus solids transfer is also investigated. Comparison with a range of experimental data is presented.
Article
The residence time distribution (RTD) of solids is essential for the design of CFB reactors where conversion proceeds with time. The residence time distribution for the solids is measured at different working conditions. The resulting average residence times are correlated as function of gas velocity and solids circulation rate and are compared with literature data. In order to predict the RTD of the solids, the solids/gas flow is firstly described by a plug flow with dispersion. A fitting procedure gave experimental Péclet numbers, which were correlated as function of gas velocity and solids circulation rate. A more fundamental approach based on the core/annulus flow structure is thereafter used to predict the residence time. The riser is divided in a dilute core with particles flowing upward and a denser film moving slowly downward along the wall. Particle interchange between the two regions is described by a convective interchange flow. The model is used to predict experimental RTD-curves. The interchange flux between core and annulus corresponds well with radial fluxes reported in literature. Although the RTD is qualitatively well described, experimental curves show a higher dispersion than the calculated ones. To improve the core/annulus approach, further research is necessary.
Article
A comprehensive, predictive model is presented that describes the flow and distribution of solids within a circulating fluidised bed (CFB). Dense particle sheets or streamers are assumed to fall at the walls, surrounding a rapid dilute suspension upflow. Mechanisms and relationships for interchange of solid particles between the streamers and dilute suspension are given. The fraction of solids reflected internally at the reactor exit and the suspension density at the solids return location constitute the model boundary conditions. Variations with reactor height in axial suspension density, fractional wall coverage by streamers, and renewal rate of solids at the wall are predicted. Although developed primarily for prediction of circulating fluidised-bed combustor operation, the model may be used for other CFB applications. Model results may be incorporated in detailed modelling of bed-to-wall heat transfer. Model relationships utilise data obtained from the University of British Columbia 115 kWth pilot-scale unit. The model's predictive capability was tested against data provided by Studsvik Energy of Sweden obtained from their 2.5 MWth combustor. Good prediction of the Studsvik density profiles was obtained.
Article
Mixing of gas, solids and heat in a commercial FCC regenerator, 5.76 m in diameter, was investigated. This regenerator was operated at a high superficial gas velocity (>1.2 m/s) and high temperature (650–710°C). It is classified as a highly efficient regenerator. The profiles of coke content in the catalyst, gas composition, temperature and pressure were obtained by sampling and instrumental measurement. It was confirmed that a proposed axial dispersion model could be used for a satisfactory prediction of these experimental profiles. The values of gas and solids dispersion coefficients are also presented.
Article
This paper describes an experimental investigation of single-particle behaviour in a cold pilot-scale model of a circulating fluidized bed combustor (CFBC). In the system, sand is recirculated by means of air. Pressure measurements along the riser are used to determine the suspension density. A radioactive tracking facility, which detects single radioactive particles, is developed and applied to determine the dynamic picture of the particle trajectories in the simulated boiler. The tracer particles are observed to move between the zone above and below the secondary air inlet with a mean frequency of about 1 Hz under the present operating conditions. This relatively high frequency is due to the fact that most of the particle trajectories take place just around the secondary air inlet. It is found that the upward particle velocity in the upper dilute transport zone decreases with the particle size or density, which results in a decreased number of particle observations for the larger particles with the riser height. The experiments show that the mean particle residence times in the zones above and below the secondary air inlet are almost independent of the particle characteristics. The overall mean particle residence time in the riser is proportional to the magnitude of the internal particle recirculation, which increases with the particle size.
Article
The residence time distribution of limestone sorbent particles has been studied in order to increase the understanding of the conditions for sulphur capture in fluidised bed boilers. Two methods were used. The ‘steady state method’ involves the study of residence time for various particle size fractions. The ‘transient method’ is based on the transient increase in the amount of sorbent carryover with the fly ash, following initial limestone addition to a fresh bed (i.e. a bed with little or no sorbent). For the boiler investigated both methods gave similar results, showing that the major fraction of the sorbent, 80–85%, had a residence time of one hour or more.
Article
This paper presents the experimental results with radioactive tracer particles in an 80MWth circulating fluidized-bed boiler. Batches of γ-ray emitting tracer particles were injected into the standpipe. The response curves of the impulse injection were measured by a set of successive scintillation detectors located at different parts of the boiler. The results show that there are significant differences of the response signals when boiler loading changes. A model has been developed to obtain information from the experimental data. The particle recirculating rate is obtained by fitting the model to the experimental data. In the same way, the average particle velocities in the furnace are estimated. The impact of loading on the particle recirculating rate and on the hydrodynamics of the boiler is discussed.
Studies on the axial solid mixing mechanisms in gas—solids cocurrent upflow and downflow circulating fluidized bed systems have revealed that, among the many influencing factors, flow direction has the most profound influence on the axial solids mixing. When the flow is in the direction of gravity (downflow in the downer), axial solids dispersion is very small and the flow pattern approaches plug flow; when the flow is against gravity (upflow in the riser), axial solids dispersion is significantly larger and the flow pattern deviates significantly from plug flow.Solids mixing is found to be mainly due to the dispersion of dispersed particles in the downer while two solids mixing mechanisms co-exist in the riser: the dispersion of dispersed particles and the dispersion of particle clusters. Dispersion due to dispersed particles is very small in both the riser and the downer, indicating that dispersed particles pass through the system in a near plug flow pattern. Dispersion due to particle clusters in the riser, on the other hand, is very significant, contributing to the large axial solids backmixing and the bimodal solids residence time distribution in the riser.
Article
The lateral and axial mixing of dispersed particles in a circulating fluidized bed were measured by using the phosphor tracer technique. The measured residence time distribution was satisfactorily described by a proposed two dimensional plug flow dispersion pow. Compared with that in the cocurrent downflow circulating fluidized bed (CDCPB), the dispersion Peclet numbers for lateral and axial mixing of dispersed particles is about the same value with that in CDCFB (downer). Correlations of the Peclet numbers under the operating conditions are presented.
Article
A catalyst velocity of 11 fps in a pilot unit transfer line was determined by injection of catalyst tagged with zirconium--niobium-95. Similar injections have illustrated mixing patterns in the reactor proper. The examples presented of the types of data which may be obtained demonstrate the unique ability of the tracer technique to provide vital information concerning the effects of operating conditions and structural designs on solids-mixing patterns in fluidized systems. (auth)
Article
The dynamics of fluidized beds have been a subject of intensive study over the past decade as the importance and use of fluidization techniques have increased. A major question is defining the regime of fluidization, which depends on many parameters and greatly affects the dynamic behavior of the particles. In this paper the distribution of different phases in a riser reactor at a fluidized catalyst cracking (FCC) plant, which is used in the oil refinery industry, has been studied. The tracers used were [sup 41]Ar and [sup 140]La. Axial and radial dispersion coefficients have been determined for modeling purposes.
Article
The development of tracer pulse techniques for measuring mean film and droplet velocities (film and droplet residence times) in vertical two phase annular flow is described. Results of experiments are discussed, and compared to a computer simulation of a tracer pulse in annular flow as described by the interchange model. The experiments show that the mean film velocity can be determined accurately, but not the droplet velocity. The computer simulation supports the experimental results. The effect of interchange is also shown to be sufficiently large in much of the annular flow region to give average film and droplet residence times which are not greatly different. On décrit la mise au point de méthodes, basées sur les impulsions d̂un indicateur, pour mesurer les vitesses moyennes de pellicules et gouttelettes (temps de séjour de pellicules et gouttelettes) dans un écoulement annulaire à deux phases. On discute les résultats expérimentanx et les compare aux données obtenues par reproduction, sur un ordinateur, d̂un impulsion d̂un indicateur dans un écoulement annulaire, tel que décrit par le modèle d̂échange. Les expériences indiquent qu'on peut déterminer avec exactitude la vitesse moyenne des pellicules mais non pas celle des gouttelettes. La reproduction sur ordinateur vient à l̂appui des résultats expérimentaux. On voit aussi que l̂effet de l̂échange est suffisamment prononcé dans une grande partie de la région de l̂écoulement annulaire pour indiquer des temps de séjour moyens des pellicules et des gouttelettes qui diffèrent peu.
Article
Axial and lateral mixing of fine particles in a binary-solid riser have been investigated using a phosphor tracer method. The measured bimodal residence time distribution (RTD) demonstrated two types of axial dispersions of the fines: the dispersion of discrete particles and that of clusters. A proposed one-dimensional, bimodal dispersion model describes the bimodal RTDs very well. The axial Peclet number of the fines is not sensitive to the fraction of coarse particles, gas velocity and solids circulation rate. Lateral solids dispersion was determined by measuring the solids RTD at different radii using a point source tracer injection. A two-dimensional dispersion model describes the measured RTDs satisfactorily. Lateral solids mixing decreased as coarse particles were added into the riser. Correlations of the axial and lateral Peclet numbers obtained fit the experimental data well. On a étudié le mélange axial et latéeral de particules fines dans une colonne montante binaires-solides à l'aide d'une méthode de traçage au phospore. La distribution du temps de séjour (DTS) bimodale mesurée illustre deux types de dispersions axiales des particules fines: la dispersion de particules discrètes et la dispersion de grappes. On propose un modèle de dispersion unidimensionnel bimodal qui décrit très bien la DTS bimodale. Le nombre de Péclet axial des fins n'est pas sensible à la fraction des particules grossières, à la vitesse du gaz et a la vitesse de circulation des solides. On a déterminé la dispersion latérale des solides en mesurant la DTS des solides à différents rayons à l'aide d'une injection de traceur ponctuelle. Un modèle de dispersion bidimensionnel décrit les DTS de manière satisfaisante. Le mélange latéral des solides diminue avec l'ajout de particules grossières dans la colonne. Des correlations obtenues pour les nombres de Péclet axial et latéral montrent un bon accord avec les données expérimentales.
Article
Longitudinal solids mixing was studied experimentally in circulating fluidized bed risers of internal diameter 0.152 m and 0.305 m. Superficial gas velocity and mean solids flux used were 2.8–5.0 m/s and 5.0–80 kg/m2·s, respectively, and the bed solids had a surface volume mean diameter of 71 μm and a particle density of 2,456 kg/m3. A sodium chloride tracer was used in impulse injection experiments. A simple, one-dimensional dispersion model describes measured solids mixing satisfactorily. Peclet numbers (UoL/Dz) found, in the range 1.0–9.0, were correlated with the riser diameter and mean solids flux. The modeling approach described here permits residence time distribution curves to be calculated directly from the knowledge of superficial gas velocity, mean solids flux, and riser diameter. Longitudinal solids mixing in the riser decreased with increasing riser diameter. The results are consistent with recent hydrodynamic studies.
Article
Any vessel in which solids are transported upward by a gas stream and then recycled to the bottom may be classified as a Circulating Fluidized Bed (CFB). We describe possible CFB operating regimes in the context of this broad classification and highlight commercial processes that employ CFB technology and potential applications. Process design and development require a fundamental understanding of gas and solids hydrodynamics - solids hold-up, mixing and velocity distribution. We discuss techniques used to measure solids mass flux, which is a critical parameter for both design and control. In the last decade, significant research efforts have been devoted to new experimental techniques to measure both gas and solids spatial and temporal distribution. We list these techniques and detail the different modelling approaches that have emerged based on the new data. Characterization of the data is still incomplete and the available models require further refinement to reliably predict the effect of scale, operating conditions and particle characteristics on hydrodynamics.
Article
A simple hydrodynamic model for Circulating Fluidized Beds has been developed. The mathematical model, based on the core-annulus flow structure, is shown to be able to predict the two-phase flow characteristics and it requires only two measurable steady-state parameters, namely the experimental average voidage profile along the riser, or equivalently the pressure distribution, and the net solids circulation rate. The model has been successfully tested using recently obtained literature data covering a variety of reactor configurations and operating conditions.Un modèle hydrodynamique simple a été mis au point pour des lits fluidisés circulants. On montre que le modèle mathématique, basé sur la structure d'écoulement coeur-espace annulaire, est capable de prédire les caractéristiques d'écoulement biphasique; en outre, il ne nécessite que deux paramètres à l'état permanent mesurables, soit le profil de désaturation moyen expérimental le long de la colonne, ou de manière équivalente la distribution de pression, soit la vitesse de circulation des solides nette. Ce modèle a été testé avec succès à l'aide de données publiées récemment couvrant une variété de configurations de réacteurs et de conditions de fonctionnement.
Article
This paper reports measurements of the influence of riser exit geometry upon the particle residence time distribution in the riser of a square cross section, cold model, circulating fluidised bed. The bed is operated within the fast fluidisation regime. The fast response particle RTD technique developed by Harris et al. (Chem. Eng. J. 89 (2002) 127–142) was used to measure the residence time distribution.The geometry of the riser exit is shown to have a modest but consistent influence upon the particle RTD; the influence of operating conditions, i.e. superficial gas velocity and solids flux is more significant.Increasing the refluxing effect of the riser exit increases the mean, variance and breakthrough time and decreases the coefficient of variation of the residence time distribution. Changes in reflux do not have a systematic effect upon the skewness of the RTD.
Article
Steady-state tracer-gas measurements were carried out in a circulating fluidised-bed (CFB) boiler and in a cold experimental (CFB) unit. The main focus was on the radial spread of gas in the core region of the transport zone under different operating conditions. Helium (He) was used as tracer gas, and the local He-concentration was determined by a mass spectrometer. The fluidisation velocity ranged from 1.2 to 4.3 m/s in the boiler and from 1 to 6 m/s in the cold unit. The bed material was silica sand with an average particle diameter of 0.32 mm in both units. Measurements were also carried out in the bottom bed and splash zone of the boiler for comparison with those of the transport zone. In the cold experimental unit, a greater range of operating conditions could be covered than what was possible in the boiler, including a case without solids. In the core of the transport region of the boiler, there is no backmixing of gas and the horizontal mixing of injected tracer gas is characterised by an axi-symmetric spread of tracer gas around the centre line of injection. The horizontal gas-mixing rate is quantified by a dispersion coefficient, which is above 0.01 m2/s in the boiler and around 0.005 m2/s in the cold unit. The difference in dispersion coefficients of the boiler and the cold unit is due to the difference in size. The present results were compared with literature data of dispersion obtained with and without solids in CFB risers.
Article
The inventory function is the quantity of tracer remaining in a continuous-flow system at elapsed time t when steady flow of the tracer is replaced by untraced flow at t = 0. The relations between residence-time distributions, moments and changes in inventory when a tracer is flushed from a system are established. It is shown that inventory measurements could be an attractive way of measuring moments. In particular, the mean residence time is given by the intercept on the baseline of the initial tangent to the inventory curve, and the variance by the area between the inventory curve, the initial tangent and the baseline. It is proposed that dispersion be defined in terms of the variance of the residence-time distribution. This would allow experimentalists to record their results independently of models or theories in addition to comparing their results with the predictions of theories. Methods based on inventory measurements are potentially more accurate than the traditional step- and pulse-response methods. Ways in which inventory measurements might be made are suggested. It is timely that the theory should be presented now because tomographic methods that could be used to measure inventory are starting to appear.
Article
This work compares time, frequency and state-space analyses of pressure measurements from fluidized beds. The experiments were carried out in a circulating fluidized bed, operated under ambient conditions and under different fluidization regimes. Interpretation of results in time domain, such as standard deviation of the pressure fluctuations, may lead to erroneous conclusions about the flow regime. The results from the frequency domain (power spectra) and state-space analyses (correlation dimension, DML, and Kolmogorov entropy, KML, together with a non-linearity test) of the pressure fluctuations are generally in agreement and can be used complementary to each other. The power spectra can be divided into three regions, a region corresponding to the macro-structure (due to the bubble flow) and, at higher frequencies, two regions representing finer structures that are not predominantly governed by the macro structure of the flow. In all fluidization regimes, the measured pressure fluctuations exhibited an intermittent structure, which is not revealed by power spectral analysis of the original signals. Fluctuations with pronounced peaks in the power spectrum and in the auto-correlation function, corresponding to passage of single bubbles through the bed, are non-linear with a low dimension (DML<5.5). For DML<5.5, the Kolmogorov entropy is proportional to the amount of energy in the spectral range of the intermittent structures observed, whereas for DML>5.5 both KML (bits/cycle) and DML are insensitive to changes in the distribution of energy in power spectra. Thus, the state-space analysis reflects that non-linearity is mostly found in the macro-structure of the flow. Fluidized bed time series treated in this work are available at http://www.entek.chalmers.se/∼fijo
Article
A mathematical model, based on the core-annulus flow structure, is proposed to characterize the two-phase flow and the residence time distribution of solids within the riser of circulating fluidized beds. The model is tested against experimental data obtained in a laboratory unit utilizing a tracer technique. While the accuracy of the experimental technique is shown not to be sufficient to allow conclusive results, the model is capable of a good prediction of the observed trends of the characteristic bimodal residence time distributions with an excellent match of the times of arrival of the two peaks. Simulation runs have been performed to study the effects of key operating variables on the residence time distribution.
Article
A novel, non-intrusive method is described for measuring the particle residence time distribution (RTD) in a system with a short mean residence time. The method uses phosphorescent tracer particles activated by a high intensity pulse of light at the inlet. Tracer is detected using a light sensitive photomultiplier unit. Appropriate boundary conditions are maintained by using an annular feeder fluidised bed at the entrance boundary and an inline jet mixer installed at the exit boundary. This well defined arrangement of experimental boundary conditions represents a significant advance in the measurement of unbiased particle RTDs in these systems. The method was developed for measuring the particle RTD in a circulating fluidised bed (CFB) riser, but is applicable to other particle–fluid systems where a fast response measurement is required.
Book
Until recent years knowledge of chemical processing was descriptive and qualitative. In 1810 modern chemical theory was born and process description became quantitative. Then about 1900 the quantitative engineering approach was developed, first for physical changes, called the Unit Operations, and somewhat later for chemical operations. This we call the American approach. In 1957 European chemical engineers brought together the design of chemical and their related physical operations under the name of Chemical Reaction Engineering, or CRE. This approach and name received practically universal acceptance. Today the methods of CRE are widely used in the processing of biochemical and all sorts of other systems, This talk wanders through this development.
Residence time distributions in circulating uidised beds-dimensional analysis and correlation
  • A T Harris
  • R B Thorpe
  • J F Davidson
Harris, A. T., Thorpe, R. B., & Davidson, J. F. (2001). Residence time distributions in circulating uidised beds-dimensional analysis and correlation. In Proceedings of the Tenth International Conference on Fluidisation, Beijing, China, May 2001 (pp. 221-228). New York: Engineering Foundation.
Modelling the mixing of solids in circulating fluidized beds
  • B J Milne
  • F Berruti
Milne, B. J., & Berruti, F. (1990). Modelling the mixing of solids in circulating uidized beds. In P. Basu, M. Horio, M. Hasatani (Eds.), Circulating uidized bed technology III (pp. 575 -580). Oxford: Pergamon Press.
Residence time distribution of gas and solids in a circulating bed New York: Engineering Foundation Hydrodynamics of circulating bed risers: A review
  • Oxford Bai
  • D Nakagawa
  • N Shibuya
  • E Kinoshita
  • H Kato
Oxford: Pergamon Press. Bai, D., Nakagawa, N., Shibuya, E., Kinoshita, H., & Kato, K. (1992). Residence time distribution of gas and solids in a circulating bed. In O. E. Potter, D. J. Nicklin (Eds.), Fluidisation VII (pp. 195 –202). New York: Engineering Foundation. Berruti, F., Chaouki, J., Godfroy, L., Pugsley, R. S., & Patience, G. S. (1995). Hydrodynamics of circulating bed risers: A review. Canadian Journal of Chemical Engineering, 73, 579–602.
Hydrodynamic modelling
  • J Sinclair
Sinclair, J. (1997). Hydrodynamic modelling. In J. R. Grace, A. Avidan, T. Knowlton (Eds.), Circulating uidised beds (pp. 149 -180). London: Chapman and Hall.
Measurements of diiusional phenomena and velocity proÿles in a vertical riser
  • Van
  • D Meer
van der Meer,. (1997). Ph.D. dissertation, University of Cambridge. Van Zoonen, D. (1963). Measurements of diiusional phenomena and velocity proÿles in a vertical riser. In Proceedings of the symposium on the interaction between and particles (pp. 64 –71). London: Institute of Chemical Engineers.
Bed expansion and solid mixing in high velocity beds
  • A Avidan
Avidan, A. (1980). Bed expansion and solid mixing in high velocity beds. Ph.D. thesis, City University of New York.
Characteristics of fast fluidised beds with internals
  • C Zheng
  • Y Tung
  • Li
  • M Kwauk
Zheng, C., Tung, Y., Li, & Kwauk, M. (1992). Characteristics of fast uidised beds with internals. Proceedings of the 7th engineering foundation conference on uidisation. Australia: Brisbane (pp. 275 -283).
Gas/solid ow patterns in a 30.5-cm-diameter circulating uidised bed
  • R Bader
  • J Findlay
  • T M Knowlton
Bader, R., Findlay, J., & Knowlton, T. M. (1988). Gas/solid ow patterns in a 30.5-cm-diameter circulating uidised bed. In P. Basu, & J. F. Large (Eds.), Circulating uidised bed technology II (pp. 161-165). Oxford: Pergamon Press.
Mixing in continuous systems Solids residence time distribution in CFB reactors
  • E B Nauman
  • B A Buuham
  • Wiley
  • G S Patience
  • J Chaouki
  • G Kennedy
Nauman, E. B., & Buuham, B. A. (1983). Mixing in continuous systems. New York: Wiley. Patience, G. S., Chaouki, J., & Kennedy, G. (1990). Solids residence time distribution in CFB reactors. In P. Basu, M. Horio, & M. Hasatani (Eds.), Circulating bed technology III (pp. 599 – 604).
Residence time in circulating fluidised bed
  • Q Wang
  • J Zhou
  • J Tu
  • Z Luo
  • X Li
  • M Fang
  • L Cheng
  • M Ni
  • K Cen
Wang, Q., Zhou, J., Tu, J., Luo, Z., Li, X., Fang, M., Cheng, L., Ni, M., & Cen, K. (1996). Residence time in circulating uidised bed. In M. Kwauk, & L. Jinghai (Eds.), Circulating uidised bed technology V (pp. 128-133). Beijing: Science Press.
Large radioactive particle motion in a gas/solid riser
  • L Godfroy
  • F Jarachi
  • J Chaouki
Godfroy, L., Jarachi, F., & Chaouki, J. (1999). Large radioactive particle motion in a gas/solid riser. In Werther, J. (Ed.), Proceedings of the Sixth International Conference on Circulating Fluidised Beds (pp. 199 -205). Germany: Dechema e.V.