- Email: [email protected]
Gas and solid flow patterns in the loop-seal of a circulating fluidized bed
Gas and solid flow patterns in the loop-seal of a circulating fluidized bed Piero Bareschino, Roberto Solimene, Riccardo Chirone, Piero Salatino PII: DOI: Reference:
S0032-5910(14)00491-4 doi: 10.1016/j.powtec.2014.05.036 PTEC 10291
To appear in:
Powder Technology
Received date: Revised date: Accepted date:
17 February 2014 21 May 2014 26 May 2014
Please cite this article as: Piero Bareschino, Roberto Solimene, Riccardo Chirone, Piero Salatino, Gas and solid flow patterns in the loop-seal of a circulating fluidized bed, Powder Technology (2014), doi: 10.1016/j.powtec.2014.05.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
RI
PT
GAS AND SOLID FLOW PATTERNS IN THE LOOP-SEAL OF A CIRCULATING FLUIDIZED BED
MA
NU
SC
Piero Bareschino Dipartimento di Ingegneria, Università degli Studi del Sannio Piazza Roma 21, 82100 Benevento, Italy Roberto Solimene, Riccardo Chirone Istituto di Ricerche sulla Combustione – CNR, Piazzale Tecchio 80, 80125 Napoli, Italy Piero Salatino Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale Scuola Politecnica e delle Scienze di Base, Università degli Studi di Napoli "Federico II" Piazzale Tecchio 80, 80125 Napoli, Italy
D
KEYWORDS:
AC CE P
TE
Circulating Fluidized Bed; Loop Seal; Non-Mechanical Valve
ABSTRACT
Gas and solids flow patterns in a loop-seal operated as solids re-injection device between the downcomer and the riser of a lab-scale cold CFB apparatus was characterized under a broad range of operating conditions by a combination of methods, including measurement of pressure time series along the CFB loop, solids mass flux along the riser and solids and gas tracing into the loop-seal. Analysis of results from solids and gas tracing tests highlights the role of both solids downflow in the supply chamber and gas cross-flow between the loop-seal chambers on the fluidization patterns. An approximate criterion is given for the onset of gas “leakage” between the two loop-seal sections, relevant to loop-seal applications in dual interconnected fluidized beds (DIFB).
ACCEPTED MANUSCRIPT
1. INTRODUCTION
applications
whenever
PT
The dual fluidized bed technology has recently risen to renewed interest for process distinct
reactive environments (e.g.
oxidative/reducing,
RI
temperature and/or pressure looping) need to be established while sharing a solid
SC
stream which acts as reactant or thermal carrier. Exploitation of the dual fluidized bed concept is currently being explored in fields like: i) CO 2 capture in coal-fired generating
NU
stations and cement kilns (1)); ii) chemical-looping combustion (2-3); iii) biomass
MA
pyrolysis and gasification (4). The dual loop system presents, however, some criticalities that are mostly related to the effective control of solids recirculation between the beds and to the establishment of leak-tight operation of the beds with respect to the
TE
D
gaseous streams. These criticalities involve both design and operational variables of the plant which determine the hydrodynamic behaviour of different plant components
AC CE P
(5-6). Several types of non-mechanical valves (L-valve, J-valve, seal pot and loop-seal) are commonly used in CFBs and DIFBs to control the mass flow rate of solids or to seal gaseous streams. Despite empirical or semi-empirical criteria for correct loop-seal design have been laid (7-15), thorough characterization of loop-seal hydrodynamics, especially under large solid throughput conditions, is still lacking. In the present study the hydrodynamics of a loop-seal is investigated. The seal is operated as solids reinjection device between a 40 mm ID downcomer and a 102 mm ID riser in a lab-scale cold
CFB
apparatus.
The
experimental
procedure
includes
time-resolved
measurements of gas pressure at several locations along the CFB loop and in the loop-seal and of solids mass flux along the riser. The gas flow patterns are characterized by means of gas tracing experiments based on continuous injection of CO2 at the bottom of the loop-seal chambers fluidized with air. Solids flow patterns are
ACCEPTED MANUSCRIPT
characterized by impulsive injection of dye-coloured particles into the supply chamber,
PT
followed by particle tracking by means of a video camera interfaced to a computer. The characterization of the gas and solids flow patterns is carried out under a broad range
RI
of operating conditions by varying: i) the gas superficial velocity in the riser; ii) the
SC
fluidization velocity in the loop seal; iii) the throughput of circulating bed solids in the loop seal. Results are presented and discussed with an emphasis on the establishment
NU
of gas cross-flow between the loop seal chambers and its relevance to effective solids
MA
recirculation and gas “leakage”.
2. EXPERIMENTAL
TE
D
2.1 Apparatus
The lab-scale cold model CFB (Figure 1) consists of a riser, a cyclone, a solids
AC CE P
reservoir, a standpipe and a loop seal. The Plexiglas riser, 0.102 m ID and 5.6 m high, is equipped with a low-pressure drop gas distributor consisting of several stainless steel nets layered one on the other up to 2 mm thickness. A high-efficiency cyclone, 0.09 m and 0.32 m high, is installed at riser outlet. The solids return line consists of the solids reservoir, 180 mm ID 0.40 m high, a 40 mm standpipe and a loop seal. The supply and recycle chambers of the loop seal have a square cross section (50mm side), height of 300 and 250 mm, respectively, and separate wind-boxes. The gap at the bottom of the two chambers is about 50 mm. The CFB loop is equipped with 21 pressure taps. The overall solids mass flux is measured periodically during the test by closing for a pre-set time a butterfly valve located along the standpipe and recording the solids mass accumulated above the valve.
ACCEPTED MANUSCRIPT
2.2 Diagnostics
PT
The pressure along the CFB loop is measured by high-precision piezo-resistive pressure transducers. On-line gas sampling is accomplished at different locations in
RI
the loop seal by means of a probe connected to a CO2 gas analyzer (ABB advance
SC
optima Uras 14). Pointwise and time-resolved measurement of CO2 concentration at the sampling ports serve to characterize the gas flow patterns in the loop seal. All
NU
signals coming from the instruments are logged on a data acquisition unit consisting of
MA
a PC equipped with a data acquisition board. A high-speed, highly light-sensitive mega-pixel CMOS camera (Photron’s FASTCAM ultima APX I2) interfaced to a
TE
in the loop seal chambers.
D
computer and controlled with an image processing software is used for particle tracking
AC CE P
2.3 Materials and Operating Conditions Table 1 reports the properties of the bed material, consisting of quartz sieved in two size ranges, namely 100-400 and 250-500m. The fluidizing gas is technical air at room temperature. Gaseous CO2 from cylinders is used as gas tracer. Black dye-coloured glass beads (500 m) are used as solids tracer. The overall bed inventory is 3.5 kg. Gas superficial velocity in the riser Ug,r ranges between 2.1 and 4.4 m/s, always larger than the terminal velocity (Ut) of the granular solid. Air is injected at the bottom of the loop-seal so that the gas superficial velocity in loop-seal chambers (Ug,ls) is 1, 1.5 and 2 times Umf. Solids mass circulation fluxes in the riser Gs ranges between 2 and 11 kg/(s m2). Bed voidage () along the riser is calculated by processing the pressure drops, P, measured at successive pressure taps located at distance z. Gas static head, gas and solids frictional and acceleration terms can be neglected under the operating conditions
ACCEPTED MANUSCRIPT
investigated (16). Accordingly, voidage of the bottom bed can be evaluated according to
PT
the simplified relationship P=sg(1-)z, where g is the acceleration due to gravity.
RI
2.4 Experimental Procedures
SC
2.4.1 Bed Hydrodynamics
Steady state operation of the CFB loop was investigated at a pre-set fluidization velocity in
NU
the riser and in the loop seal. The steady state condition was recognized by continuous
MA
monitoring of the gas pressure along the loop and by step-wise measurement of the mass flow rate. Once the steady state condition was reached, the gas pressure signal was
D
recorded for about 5 minutes. Different operational regimes corresponded to different
TE
combinations of gas superficial velocity in the riser and in the loop seal. 2.4.2 Solids and Gas Tracing in the Loop Seal
AC CE P
Some experiments were carried out at Ug,r=2.85m/s by performing solids and gas tracing in order to investigate the solids and gas flow patterns in the loop seal. Once the steady state operation of the CFB loop was reached at a pre-set value of Ug,ls, solids tracing was started by injecting a batch of solids tracer (50g) in the standpipe above the loop seal. The tracer particles eventually moved across the loop seal and their motion was recorded by the video camera at 125 fps for 30s. Experiments were iterated three times at a given Ug,ls. An ad-hoc procedure for image analysis has been developed to follow the trajectories of the tracer particles. Gas tracing was carried out under the same operating conditions by continuous feeding of fluidizing gas at a preset CO2 concentration (in the range of 10-20%v) either in the supply or in the recycle chamber. The time series of the CO2 concentration measured at the bottom and the top of the beds of the loop seal were analyzed to infer the gas flow distribution in the loop
ACCEPTED MANUSCRIPT
PT
seal.
3.1 Hydrodynamics of the riser and of the loop seal
RI
3. EXPERIMENTAL RESULTS AND DISCUSSION
SC
Figure 2 reports the relative pressures measured at different locations around the CFB loop during runs operated at Ug,r=2.85m/s. The gas superficial velocity in the loop seal Ug,ls
NU
was varied in the range Umf≤Ug,ls≤2Umf. Analysis of the pressure loop indicates that the
MA
pressure gradient at the bottom of the riser (i.e. at levels above the distributor up to 0.1m) increases as Ug,ls is increased. The pressure gradient in the recycle chamber of the loop seal remains constant at a value consistent with the bulk density of the bed material under
TE
D
bubbling fluidization conditions at any Ug,ls. The pressure gradient in the supply chamber increases from a minimum at Ug,ls=Umf to approach values corresponding to the bulk
AC CE P
density of a fully fluidized bubbling bed as Ug,ls is increased. Figure 3A reports the solids mass flux in the riser as a function of Ug,ls/Umf. Figures 3B and C report the pressure drop and the height of the bed of solids, respectively, in the recycle and in the supply chamber of the loop-seal. Data points correspond to the three experimental conditions reported in Figure 2. The solid mass flux remains substantially constant as Ug,ls is increased. In the recycle chamber of the loop seal, the pressure drop slightly decreases and the bed height remains fairly constant as Ug,ls is increased. In the supply chamber of the loop seal, the pressure drop undergoes a pronounced increase and the bed height decreases as Ug,ls increases from Umf to 1.5Umf, both remaining nearly constant at values larger than Umf. The fluidization velocity in the loop seal affects the local hydrodynamics: as Ug,ls is increased from Umf to 1.5Umf, the bed material present in the supply chamber becomes fully
ACCEPTED MANUSCRIPT
fluidized, consistently with the change of pressure drop and gradient reported in Figures 2
PT
and 3. When the supply chamber is fluidized, a fraction of the solids present in this chamber is transferred to the riser, and the loading of solids at the bottom of the riser is
RI
increased. This feature is reflected by the change of bed height in the supply chamber and
SC
of the pressure gradient at the bottom of the riser.
Figure 4 reports, for both the fine (open symbols) and coarse (closed symbols) quartz, the
NU
solids mass flux in the riser and the bulk density () of the bed at the bottom of the riser as
MA
a function of the gas superficial velocity (Ug,r) for different values of the ratio Ug,ls/Umf in the loop seal. Gs increases almost linearly with Ug,r. As expected, Gs decreases with decreasing Ug,ls/Umf, especially when Ug,ls approaches Umf and for large values of Ug,r.
TE
D
The bottom bed density is fairly constant at a value of about 140kg/m3 for Ug,r>3m/s. It increases as the fluidization velocity in the riser decreases below Ug,r<3m/s. For small
AC CE P
values of Ug,r, as the terminal velocity of bed particles is approached, a dense bed is established at the bottom of the riser with values of in the order of 800-1000kg/m3. The bed density is influenced by Ug,ls/Umf, decreasing as Ug,ls approaches Umf as a consequence of poorer transfer of solids from the loop seal. Altogether, results suggest that loop seal operation in the CFB loop becomes fully effective when both the recycle chamber and the supply chamber become fully fluidized. This condition requires the gas superficial velocity in the loop seal to exceed Umf.
3.2 Analysis of solids flow patterns in the loop seal The hydrodynamics of the loop seal has been investigated by inspection of the gas and solids flow patterns in the loop seal chambers as Ug,ls was changed. Tracer solids were followed after injection in the standpipe above the loop seal. Trajectories of tracer solids
ACCEPTED MANUSCRIPT
were reconstructed by digital frame-by-frame analysis of videorecordings taken during their
PT
motion toward and beyond the gap between the loop seal chambers. Figure 5 reports a selected sample of the trajectories recorded in the experiments and of the corresponding
RI
time-resolved components (VX, VY) of the tracer particle velocity. Experiments were carried
SC
out at Ug,ls equal to either Umf or 1.5Umf, the gas superficial velocity in the riser being steadily set at Ug,r=2.85m/s. Images acquired for Ug,ls=2Umf could not be analyzed as the
NU
stochastic motion of the particles typical of a bubbling fluidized bed hampered following a
MA
single particle along the test. Trajectories are mapped in Figure 5 assuming the bottom-left corner of the supply chamber as origin of the reference axes, taking x positive along the direction pointing toward the riser and y positive along the direction opposed to gravity. The
TE
D
selection of trajectories plotted in Figure 5 correspond to tracer particles that were all initially located at the top of the supply chamber but at different lateral position: one at the
AC CE P
center, one close to the external wall and one close to the separation wall (vertical solid line) with the recycle chamber.
At Ug,ls=Umf the streamlines of the tracer particles in the supply chamber converge toward the top of the gap between the chambers. A significant part of the supply chamber (lying below the dashed line) appears stagnant whereas a limited part of the gap between the two loop seal chambers is characterized by flow of bed solids. The velocity component along the x axis (VX) is nearly the same for all the tracer particles and constant with time (about 1.5 mm/s), but it increases significantly as they approach the gap between the two chambers. The velocity component along the y axis (VY) is nearly the same for all the tracer particles and decreases as they approach the gap between the two chambers. The tracer particle that initially was located close to the supply chamber axis presents the largest value of VY.
ACCEPTED MANUSCRIPT
As Ug,ls is increased (Ug,ls =1.5Umf), it can be recognized that the convergence of the
PT
trajectories followed by different tracer particles in the supply chamber takes place at a later stage, well in the upper part of the gap between the two chambers. The stagnant
RI
region of the bed in the supply chamber is less extended than that observed at Ug,ls=Umf.
SC
Correspondingly, a larger fractional area of the gap between the two chambers experiences solids flow. VX is initially negligible for all the tracing particles except when
NU
they approach the gap. VY is similar for all the tracer particles and constant with time (about
MA
12.5 mm/s) in the supply chamber; it becomes positive and increases with time in the recycle chamber. The tracer particle that was initially closer to the external wall is trapped by the stagnant region of the bed and stops there. In the range Umf
TE
D
solids motion in the supply chamber is pretty much streamlined, being the random effect due to bubble motion negligible. On the contrary, at Ug,ls=2Umf, the stagnant region
AC CE P
disappears and bubbling fluidization establishes throughout the bed.
3.3 Analysis of gas flow patterns in the loop seal Figure 6 shows an outline of the gas flow pattern in the loop seal during gas tracing experiments. Two separate gas streams of equal flow rates are fed to the bottom of the supply (QSCin) and recycle (QRCin) chambers of the loop seal, characterized by different CO2 concentrations xSCin and xRCin, respectively. Experiments have been carried out by feeding CO2-enriched air at a pre-set inlet concentration either to the supply or to the recycle chamber under the same operating conditions established in the solids tracing tests (Ug,r =2.85m/s, Ug,ls=Umf, 1.5Umf, 2Umf). The gas cross-flow between the two chambers has been represented by two streams: one (QRCbypass) directed from the recycle chamber toward the supply chamber, the other (QSCbypass) in the opposite direction. It has been assumed that the CO2 concentration in the by-pass stream is that
ACCEPTED MANUSCRIPT
of the fluidizing gas fed to the corresponding chamber. From the known values of QSCin,
PT
QRCin, xSCin, xRCin and the measured values of the CO2 concentration (xSCout, xRCout) at the outlet, it is possible to calculate all the other gas flow rates (QRCbypass, QSCbypass,
RI
QSCout and QRCout) on the basis of global material balance and on a balance on CO2.
SC
Figure 7 shows xSCin, xRCin, xSCout and xRCout as a function of Ug,ls during gas tracing experiments in both loop seal chambers. It can be recognized that xSCout is nearly equal
NU
to xSCin both in the supply chamber gas tracing mode (i.e. when the supply chamber is
MA
fluidized with CO2-enriched gas) and in the recycle chamber gas tracing mode (i.e. when only air is used to fluidize the supply chamber). On the contrary, xRCout shows significant departures from xRCin: it is smaller than the inlet pre-set CO2 concentration
TE
D
when gas tracing occurs in the recycle chamber gas tracing mode, and it is larger than the CO2 concentration in air when gas tracing is performed in the supply chamber
AC CE P
tracing mode. Working out the data reported by means of simple material balance equations, the calculated QRCbypass turns out to be negligible (in order of 0.01m3/h), whereas QSCbypass is of about 0.2m3/h, irrespective of Ug,ls. Taking into account that the gas flow rate at incipient fluidization is about 0.45m3/h for a single chamber, QSCbypass represents a significant fraction of the effective gas superficial velocity in the supply chamber. The large values of QSCbypass found in the experiments are consistent with the extensive defluidization observed in the supply chamber at Ug,ls≈Umf (Fig. 5). The criterion for fluidization of bed solids in the supply chamber can be approximately based on the following relationship:
U geff,ls U mf
Gs Ar mf Als s (1 mf )
[1]
where Ueffg,ls, s, mf, Ar, Als are the effective gas superficial velocity, bed solids density,
ACCEPTED MANUSCRIPT
bed voidage at incipient fluidization, cross sections of the riser and of the supply
PT
chambers, respectively. The RHS of Equation (1) expresses the net gas superficial velocity established in the emulsion phase of the bed in the supply chamber when the
RI
bed is fluidized and the downflow of bed solids is taken into account. This value could
SC
be negative for low values of Umf or large values of Gs. It is worth noting that a negative value of the net gas superficial velocity in the emulsion phase of the bed would not be
NU
per se undesirable from the standpoint of the effective recirculation of solids, but it
MA
might be unacceptable when “leakage” of gas from the supply chamber to the recycle chamber must be prevented due to process constraints. This is the case, for instance, when the loop seal is used as solids non-mechanical valve in dual interconnected
TE
D
fluidized bed systems with distinct reactive environments. The relationship (1) is satisfied in all the experimental conditions tested, except when Ug,ls=Umf. In tests
AC CE P
carried out at Ug,ls approaching Umf, the supply chamber of the loop seal was largely defluidized, consistently with the pressure drop data and with the analysis of bed solids tracing, and extensive “leakage” of gas took place from the supply chamber to the recycle chamber.
4. CONCLUSIONS The hydrodynamics of a loop-seal operated as solids re-injection device in a lab-scale cold CFB apparatus was characterized by a combination of methods, including solids and gas tracing. For a given riser gas superficial velocity, the loop seal fluidization velocity exerts a decisive influence on the hydrodynamics of the loop seal, as it significantly affects the gas and solids flow patterns in the supply and recycle chambers of the loop-seal. As the loop seal is kept at gas superficial velocities just beyond incipient fluidization, the supply
ACCEPTED MANUSCRIPT
chamber is only partly fluidized and an extended stagnant zone is observed at its bottom.
PT
As fluidization becomes more vigorous, stagnant zone extension is reduced and more solids are transferred to the riser. Analysis of solids and gas tracing tests highlights the role
RI
of solids downflow in the supply chamber and of gas cross-flow from the supply to the
SC
recycle chamber on the fluidization patterns. An approximate criterion is given for the onset of gas “leakage” between the two loop seal sections, relevant to loop seal applications in
MA
NU
dual interconnected fluidized beds.
5. ACKNOWLEDGMENTS
The present study has been carried out in the framework of the project: “Ricerca di Sistema
D
- Accordo MiSE-CNR Gruppo Tematico Carbone Pulito - Fondo per il Finanziamento
AC CE P
TE
Attività di Ricerca e Sviluppo di Interesse Generale per il Sistema Elettrico Nazionale”.
LITERATURE CITED
1. Hughes, R.W., Lu, D.Y., Anthony, E. J. and Macchi, A., Design, process simulation and construction of an atmospheric dual fluidized bed combustion system for in situ CO2 capture using high-temperature sorbents, Fuel Processing Technology, 86, 1523-1531, (2005). 2. Kronberger, B., Lyngfelt, A., Löffler, G., and Hofbauer, H., Design and Fluid Dynamic Analysis of a Bench-Scale Combustion System with CO2 Separation-ChemicalLooping Combustion, Industrial Engineering Chemistry & Research, 44, 546-556, (2005). 3. Hossain, M.M. and de Lasa, H. I., Chemical-looping combustion (CLC) for inherent CO2 separation-a review, Chemical Engineering Science, 63, 4433-4451, (2008). 4. Pfeifer, C., Proll, T., Punchner, B. and Hofbauer, H., H2-rich syngas from renewable source bu dual fluidized bed steam gasification of solid biomass. In Bi, X., Berruti, F. & Pugsley, T., (Eds.), Fluidization XII (pp. 889-896) Engineering Foundation. New York, (2007). 5. Kaiser, S., Löffler, G., Bosch, K. and Hofbauer, H. Hydrodynamics of a dual fluidized bed gasifier. Part II: simulation of solid circulation rate, pressure loop and stability, Chemical Engineering Science, 58, 4215-4223, (2003). 6. Bai, D., Issangya, A.S., Zhu, J.-X. and Grace, J.R., Analysis of the Overall Pressure Balance around a High-Density Circulating Fluidized Bed, Industrial Engineering Chemistry & Research, 36, 3898-3903, (1997).
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
7. Cheng, L., Basu, P. and Cen, K., Solid circulation rate prediction in a pressurized loop seal, Transactions of the Institution of Chemical Engineers, 76, 761-763, (1998). 8. Cheng, L. and Basu, P., Effect of pressure on loop seal operation for a pressurized circulating fluidized bed, Powder Technology, 103, 203-211, (2000). 9. Basu, P. and Cheng, L., An analysis of loop seal operations in a circulating fluidized bed, Transactions of the Institution of Chemical Engineers, 78, 991-998, (2000). 10. Botsio, E. and Basu, P., Experimental Investigation into the Hydrodynamics of Flow of Solids through a Loop Seal Recycle Chamber, Canadian Journal of Chemical Engineers, 83, 554-558, (2005). 11. Kim, S.W., Namkung, W. and Kim, S.D., Solids Flow Characteristics in Loop-seal of a Circulating Fluidized Bed, Korean Journal of Chemical Engineers, 16, 82-88, (1999). 12. Kim, S.W., Namkung, W. and Kim, S.D., Solid recycle characteristics of loop-seals in a circulating fluidized bed, Chemical Engineering Technology, 24, 843-849, (2001). 13. Kim, S.W., Kim, S.D. and Lee, D.H., Pressure balance for circulating fluidized beds with a loop-seal, Industrial Engineering Chemistry & Research, 41, 4949-4956, (2002). 14. Kim, S.W. and Kim, S.D., Effects of particle properties on solids recycle in loop-seal of a circulating fluidized bed, Powder Technology, 124, 76-81, (2002). 15. Basu, P. and Buttler, J., Studies on the operation of loop-seal in circulating fluidized bed boilers, Applied Energy, 86, 1723-1731, (2009). 16. Geldart D., Gas Fluidization Technology, Wiley, New York, (1986).
ACCEPTED MANUSCRIPT
TABLE CAPTIONS Properties of the granular solids investigated
PT
Table 1
FIGURE CAPTIONS
RI
SC
NU
Figure 6 Figure 7
MA
Figure 5
D
Figure 4
TE
Figure 2 Figure 3
Experimental apparatus: (A) cyclone; (B) solids reservoir; (C) loop seal; (D) loop seal magnification. Pressure loops for different Ug,ls/Umf ratio;Ug,r = 2.85 m/s. Solids mass flux (A), pressure drops (B) and bed height (C) in the loop-seal chambers as a function of Ug,ls/Umf ratio;Ug,r = 2.85 m/s. Solids mass flux (A) and bottom bulk density (B) as a function of Ug,r for different Ug,ls/Umf ratio. Solids flow patterns in the loop seal: trajectories and time-resolved velocity of selected tracer particles as a function of Ug,ls. Outline of the gas flow pattern in the loop seal. Inlet and outlet CO2 concentrations in the loop seal chambers as a function of Ug,ls during gas tracing experiments.
AC CE P
Figure 1
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 1
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 2
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 3
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 4
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 5
TE
AC CE P
Figure 6
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
Figure 7
ACCEPTED MANUSCRIPT
Table 1. Properties of the granular solids.
TE
D
MA
PT
Quartz 368 250-500 2300 0.09 2.67
RI
NU
calculated
AC CE P
(1)
Quartz 253 100-400 2300 0.047 1.77
SC
Material Sauter mean diameter (dp), m Size range, m Particle density (s), kg/m3 Incipient fluidization velocity (Umf)(1), m/s Terminal velocity (Ut)(1), m/s
AC CE P
Graphical Abstract
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights The hydrodynamics of a loop seal operated as solids reinjection device in a cold
PT
CFB was investigated.
RI
Gas and solids flow patterns in the loop seal was characterized by means of solids and gas tracing.
Solids downflow from standpipe and gas cross-flow between loop seal chambers
NU
play a key role on fluidization patterns.
MA
An approximate criterion for the onset of gas “leakage” between the loop seal
TE
D
sections was given.
AC CE P
SC
S0032-5910(14)00491-4 doi: 10.1016/j.powtec.2014.05.036 PTEC 10291
To appear in:
Powder Technology
Received date: Revised date: Accepted date:
17 February 2014 21 May 2014 26 May 2014
Please cite this article as: Piero Bareschino, Roberto Solimene, Riccardo Chirone, Piero Salatino, Gas and solid flow patterns in the loop-seal of a circulating fluidized bed, Powder Technology (2014), doi: 10.1016/j.powtec.2014.05.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
RI
PT
GAS AND SOLID FLOW PATTERNS IN THE LOOP-SEAL OF A CIRCULATING FLUIDIZED BED
MA
NU
SC
Piero Bareschino Dipartimento di Ingegneria, Università degli Studi del Sannio Piazza Roma 21, 82100 Benevento, Italy Roberto Solimene, Riccardo Chirone Istituto di Ricerche sulla Combustione – CNR, Piazzale Tecchio 80, 80125 Napoli, Italy Piero Salatino Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale Scuola Politecnica e delle Scienze di Base, Università degli Studi di Napoli "Federico II" Piazzale Tecchio 80, 80125 Napoli, Italy
D
KEYWORDS:
AC CE P
TE
Circulating Fluidized Bed; Loop Seal; Non-Mechanical Valve
ABSTRACT
Gas and solids flow patterns in a loop-seal operated as solids re-injection device between the downcomer and the riser of a lab-scale cold CFB apparatus was characterized under a broad range of operating conditions by a combination of methods, including measurement of pressure time series along the CFB loop, solids mass flux along the riser and solids and gas tracing into the loop-seal. Analysis of results from solids and gas tracing tests highlights the role of both solids downflow in the supply chamber and gas cross-flow between the loop-seal chambers on the fluidization patterns. An approximate criterion is given for the onset of gas “leakage” between the two loop-seal sections, relevant to loop-seal applications in dual interconnected fluidized beds (DIFB).
ACCEPTED MANUSCRIPT
1. INTRODUCTION
applications
whenever
PT
The dual fluidized bed technology has recently risen to renewed interest for process distinct
reactive environments (e.g.
oxidative/reducing,
RI
temperature and/or pressure looping) need to be established while sharing a solid
SC
stream which acts as reactant or thermal carrier. Exploitation of the dual fluidized bed concept is currently being explored in fields like: i) CO 2 capture in coal-fired generating
NU
stations and cement kilns (1)); ii) chemical-looping combustion (2-3); iii) biomass
MA
pyrolysis and gasification (4). The dual loop system presents, however, some criticalities that are mostly related to the effective control of solids recirculation between the beds and to the establishment of leak-tight operation of the beds with respect to the
TE
D
gaseous streams. These criticalities involve both design and operational variables of the plant which determine the hydrodynamic behaviour of different plant components
AC CE P
(5-6). Several types of non-mechanical valves (L-valve, J-valve, seal pot and loop-seal) are commonly used in CFBs and DIFBs to control the mass flow rate of solids or to seal gaseous streams. Despite empirical or semi-empirical criteria for correct loop-seal design have been laid (7-15), thorough characterization of loop-seal hydrodynamics, especially under large solid throughput conditions, is still lacking. In the present study the hydrodynamics of a loop-seal is investigated. The seal is operated as solids reinjection device between a 40 mm ID downcomer and a 102 mm ID riser in a lab-scale cold
CFB
apparatus.
The
experimental
procedure
includes
time-resolved
measurements of gas pressure at several locations along the CFB loop and in the loop-seal and of solids mass flux along the riser. The gas flow patterns are characterized by means of gas tracing experiments based on continuous injection of CO2 at the bottom of the loop-seal chambers fluidized with air. Solids flow patterns are
ACCEPTED MANUSCRIPT
characterized by impulsive injection of dye-coloured particles into the supply chamber,
PT
followed by particle tracking by means of a video camera interfaced to a computer. The characterization of the gas and solids flow patterns is carried out under a broad range
RI
of operating conditions by varying: i) the gas superficial velocity in the riser; ii) the
SC
fluidization velocity in the loop seal; iii) the throughput of circulating bed solids in the loop seal. Results are presented and discussed with an emphasis on the establishment
NU
of gas cross-flow between the loop seal chambers and its relevance to effective solids
MA
recirculation and gas “leakage”.
2. EXPERIMENTAL
TE
D
2.1 Apparatus
The lab-scale cold model CFB (Figure 1) consists of a riser, a cyclone, a solids
AC CE P
reservoir, a standpipe and a loop seal. The Plexiglas riser, 0.102 m ID and 5.6 m high, is equipped with a low-pressure drop gas distributor consisting of several stainless steel nets layered one on the other up to 2 mm thickness. A high-efficiency cyclone, 0.09 m and 0.32 m high, is installed at riser outlet. The solids return line consists of the solids reservoir, 180 mm ID 0.40 m high, a 40 mm standpipe and a loop seal. The supply and recycle chambers of the loop seal have a square cross section (50mm side), height of 300 and 250 mm, respectively, and separate wind-boxes. The gap at the bottom of the two chambers is about 50 mm. The CFB loop is equipped with 21 pressure taps. The overall solids mass flux is measured periodically during the test by closing for a pre-set time a butterfly valve located along the standpipe and recording the solids mass accumulated above the valve.
ACCEPTED MANUSCRIPT
2.2 Diagnostics
PT
The pressure along the CFB loop is measured by high-precision piezo-resistive pressure transducers. On-line gas sampling is accomplished at different locations in
RI
the loop seal by means of a probe connected to a CO2 gas analyzer (ABB advance
SC
optima Uras 14). Pointwise and time-resolved measurement of CO2 concentration at the sampling ports serve to characterize the gas flow patterns in the loop seal. All
NU
signals coming from the instruments are logged on a data acquisition unit consisting of
MA
a PC equipped with a data acquisition board. A high-speed, highly light-sensitive mega-pixel CMOS camera (Photron’s FASTCAM ultima APX I2) interfaced to a
TE
in the loop seal chambers.
D
computer and controlled with an image processing software is used for particle tracking
AC CE P
2.3 Materials and Operating Conditions Table 1 reports the properties of the bed material, consisting of quartz sieved in two size ranges, namely 100-400 and 250-500m. The fluidizing gas is technical air at room temperature. Gaseous CO2 from cylinders is used as gas tracer. Black dye-coloured glass beads (500 m) are used as solids tracer. The overall bed inventory is 3.5 kg. Gas superficial velocity in the riser Ug,r ranges between 2.1 and 4.4 m/s, always larger than the terminal velocity (Ut) of the granular solid. Air is injected at the bottom of the loop-seal so that the gas superficial velocity in loop-seal chambers (Ug,ls) is 1, 1.5 and 2 times Umf. Solids mass circulation fluxes in the riser Gs ranges between 2 and 11 kg/(s m2). Bed voidage () along the riser is calculated by processing the pressure drops, P, measured at successive pressure taps located at distance z. Gas static head, gas and solids frictional and acceleration terms can be neglected under the operating conditions
ACCEPTED MANUSCRIPT
investigated (16). Accordingly, voidage of the bottom bed can be evaluated according to
PT
the simplified relationship P=sg(1-)z, where g is the acceleration due to gravity.
RI
2.4 Experimental Procedures
SC
2.4.1 Bed Hydrodynamics
Steady state operation of the CFB loop was investigated at a pre-set fluidization velocity in
NU
the riser and in the loop seal. The steady state condition was recognized by continuous
MA
monitoring of the gas pressure along the loop and by step-wise measurement of the mass flow rate. Once the steady state condition was reached, the gas pressure signal was
D
recorded for about 5 minutes. Different operational regimes corresponded to different
TE
combinations of gas superficial velocity in the riser and in the loop seal. 2.4.2 Solids and Gas Tracing in the Loop Seal
AC CE P
Some experiments were carried out at Ug,r=2.85m/s by performing solids and gas tracing in order to investigate the solids and gas flow patterns in the loop seal. Once the steady state operation of the CFB loop was reached at a pre-set value of Ug,ls, solids tracing was started by injecting a batch of solids tracer (50g) in the standpipe above the loop seal. The tracer particles eventually moved across the loop seal and their motion was recorded by the video camera at 125 fps for 30s. Experiments were iterated three times at a given Ug,ls. An ad-hoc procedure for image analysis has been developed to follow the trajectories of the tracer particles. Gas tracing was carried out under the same operating conditions by continuous feeding of fluidizing gas at a preset CO2 concentration (in the range of 10-20%v) either in the supply or in the recycle chamber. The time series of the CO2 concentration measured at the bottom and the top of the beds of the loop seal were analyzed to infer the gas flow distribution in the loop
ACCEPTED MANUSCRIPT
PT
seal.
3.1 Hydrodynamics of the riser and of the loop seal
RI
3. EXPERIMENTAL RESULTS AND DISCUSSION
SC
Figure 2 reports the relative pressures measured at different locations around the CFB loop during runs operated at Ug,r=2.85m/s. The gas superficial velocity in the loop seal Ug,ls
NU
was varied in the range Umf≤Ug,ls≤2Umf. Analysis of the pressure loop indicates that the
MA
pressure gradient at the bottom of the riser (i.e. at levels above the distributor up to 0.1m) increases as Ug,ls is increased. The pressure gradient in the recycle chamber of the loop seal remains constant at a value consistent with the bulk density of the bed material under
TE
D
bubbling fluidization conditions at any Ug,ls. The pressure gradient in the supply chamber increases from a minimum at Ug,ls=Umf to approach values corresponding to the bulk
AC CE P
density of a fully fluidized bubbling bed as Ug,ls is increased. Figure 3A reports the solids mass flux in the riser as a function of Ug,ls/Umf. Figures 3B and C report the pressure drop and the height of the bed of solids, respectively, in the recycle and in the supply chamber of the loop-seal. Data points correspond to the three experimental conditions reported in Figure 2. The solid mass flux remains substantially constant as Ug,ls is increased. In the recycle chamber of the loop seal, the pressure drop slightly decreases and the bed height remains fairly constant as Ug,ls is increased. In the supply chamber of the loop seal, the pressure drop undergoes a pronounced increase and the bed height decreases as Ug,ls increases from Umf to 1.5Umf, both remaining nearly constant at values larger than Umf. The fluidization velocity in the loop seal affects the local hydrodynamics: as Ug,ls is increased from Umf to 1.5Umf, the bed material present in the supply chamber becomes fully
ACCEPTED MANUSCRIPT
fluidized, consistently with the change of pressure drop and gradient reported in Figures 2
PT
and 3. When the supply chamber is fluidized, a fraction of the solids present in this chamber is transferred to the riser, and the loading of solids at the bottom of the riser is
RI
increased. This feature is reflected by the change of bed height in the supply chamber and
SC
of the pressure gradient at the bottom of the riser.
Figure 4 reports, for both the fine (open symbols) and coarse (closed symbols) quartz, the
NU
solids mass flux in the riser and the bulk density () of the bed at the bottom of the riser as
MA
a function of the gas superficial velocity (Ug,r) for different values of the ratio Ug,ls/Umf in the loop seal. Gs increases almost linearly with Ug,r. As expected, Gs decreases with decreasing Ug,ls/Umf, especially when Ug,ls approaches Umf and for large values of Ug,r.
TE
D
The bottom bed density is fairly constant at a value of about 140kg/m3 for Ug,r>3m/s. It increases as the fluidization velocity in the riser decreases below Ug,r<3m/s. For small
AC CE P
values of Ug,r, as the terminal velocity of bed particles is approached, a dense bed is established at the bottom of the riser with values of in the order of 800-1000kg/m3. The bed density is influenced by Ug,ls/Umf, decreasing as Ug,ls approaches Umf as a consequence of poorer transfer of solids from the loop seal. Altogether, results suggest that loop seal operation in the CFB loop becomes fully effective when both the recycle chamber and the supply chamber become fully fluidized. This condition requires the gas superficial velocity in the loop seal to exceed Umf.
3.2 Analysis of solids flow patterns in the loop seal The hydrodynamics of the loop seal has been investigated by inspection of the gas and solids flow patterns in the loop seal chambers as Ug,ls was changed. Tracer solids were followed after injection in the standpipe above the loop seal. Trajectories of tracer solids
ACCEPTED MANUSCRIPT
were reconstructed by digital frame-by-frame analysis of videorecordings taken during their
PT
motion toward and beyond the gap between the loop seal chambers. Figure 5 reports a selected sample of the trajectories recorded in the experiments and of the corresponding
RI
time-resolved components (VX, VY) of the tracer particle velocity. Experiments were carried
SC
out at Ug,ls equal to either Umf or 1.5Umf, the gas superficial velocity in the riser being steadily set at Ug,r=2.85m/s. Images acquired for Ug,ls=2Umf could not be analyzed as the
NU
stochastic motion of the particles typical of a bubbling fluidized bed hampered following a
MA
single particle along the test. Trajectories are mapped in Figure 5 assuming the bottom-left corner of the supply chamber as origin of the reference axes, taking x positive along the direction pointing toward the riser and y positive along the direction opposed to gravity. The
TE
D
selection of trajectories plotted in Figure 5 correspond to tracer particles that were all initially located at the top of the supply chamber but at different lateral position: one at the
AC CE P
center, one close to the external wall and one close to the separation wall (vertical solid line) with the recycle chamber.
At Ug,ls=Umf the streamlines of the tracer particles in the supply chamber converge toward the top of the gap between the chambers. A significant part of the supply chamber (lying below the dashed line) appears stagnant whereas a limited part of the gap between the two loop seal chambers is characterized by flow of bed solids. The velocity component along the x axis (VX) is nearly the same for all the tracer particles and constant with time (about 1.5 mm/s), but it increases significantly as they approach the gap between the two chambers. The velocity component along the y axis (VY) is nearly the same for all the tracer particles and decreases as they approach the gap between the two chambers. The tracer particle that initially was located close to the supply chamber axis presents the largest value of VY.
ACCEPTED MANUSCRIPT
As Ug,ls is increased (Ug,ls =1.5Umf), it can be recognized that the convergence of the
PT
trajectories followed by different tracer particles in the supply chamber takes place at a later stage, well in the upper part of the gap between the two chambers. The stagnant
RI
region of the bed in the supply chamber is less extended than that observed at Ug,ls=Umf.
SC
Correspondingly, a larger fractional area of the gap between the two chambers experiences solids flow. VX is initially negligible for all the tracing particles except when
NU
they approach the gap. VY is similar for all the tracer particles and constant with time (about
MA
12.5 mm/s) in the supply chamber; it becomes positive and increases with time in the recycle chamber. The tracer particle that was initially closer to the external wall is trapped by the stagnant region of the bed and stops there. In the range Umf
TE
D
solids motion in the supply chamber is pretty much streamlined, being the random effect due to bubble motion negligible. On the contrary, at Ug,ls=2Umf, the stagnant region
AC CE P
disappears and bubbling fluidization establishes throughout the bed.
3.3 Analysis of gas flow patterns in the loop seal Figure 6 shows an outline of the gas flow pattern in the loop seal during gas tracing experiments. Two separate gas streams of equal flow rates are fed to the bottom of the supply (QSCin) and recycle (QRCin) chambers of the loop seal, characterized by different CO2 concentrations xSCin and xRCin, respectively. Experiments have been carried out by feeding CO2-enriched air at a pre-set inlet concentration either to the supply or to the recycle chamber under the same operating conditions established in the solids tracing tests (Ug,r =2.85m/s, Ug,ls=Umf, 1.5Umf, 2Umf). The gas cross-flow between the two chambers has been represented by two streams: one (QRCbypass) directed from the recycle chamber toward the supply chamber, the other (QSCbypass) in the opposite direction. It has been assumed that the CO2 concentration in the by-pass stream is that
ACCEPTED MANUSCRIPT
of the fluidizing gas fed to the corresponding chamber. From the known values of QSCin,
PT
QRCin, xSCin, xRCin and the measured values of the CO2 concentration (xSCout, xRCout) at the outlet, it is possible to calculate all the other gas flow rates (QRCbypass, QSCbypass,
RI
QSCout and QRCout) on the basis of global material balance and on a balance on CO2.
SC
Figure 7 shows xSCin, xRCin, xSCout and xRCout as a function of Ug,ls during gas tracing experiments in both loop seal chambers. It can be recognized that xSCout is nearly equal
NU
to xSCin both in the supply chamber gas tracing mode (i.e. when the supply chamber is
MA
fluidized with CO2-enriched gas) and in the recycle chamber gas tracing mode (i.e. when only air is used to fluidize the supply chamber). On the contrary, xRCout shows significant departures from xRCin: it is smaller than the inlet pre-set CO2 concentration
TE
D
when gas tracing occurs in the recycle chamber gas tracing mode, and it is larger than the CO2 concentration in air when gas tracing is performed in the supply chamber
AC CE P
tracing mode. Working out the data reported by means of simple material balance equations, the calculated QRCbypass turns out to be negligible (in order of 0.01m3/h), whereas QSCbypass is of about 0.2m3/h, irrespective of Ug,ls. Taking into account that the gas flow rate at incipient fluidization is about 0.45m3/h for a single chamber, QSCbypass represents a significant fraction of the effective gas superficial velocity in the supply chamber. The large values of QSCbypass found in the experiments are consistent with the extensive defluidization observed in the supply chamber at Ug,ls≈Umf (Fig. 5). The criterion for fluidization of bed solids in the supply chamber can be approximately based on the following relationship:
U geff,ls U mf
Gs Ar mf Als s (1 mf )
[1]
where Ueffg,ls, s, mf, Ar, Als are the effective gas superficial velocity, bed solids density,
ACCEPTED MANUSCRIPT
bed voidage at incipient fluidization, cross sections of the riser and of the supply
PT
chambers, respectively. The RHS of Equation (1) expresses the net gas superficial velocity established in the emulsion phase of the bed in the supply chamber when the
RI
bed is fluidized and the downflow of bed solids is taken into account. This value could
SC
be negative for low values of Umf or large values of Gs. It is worth noting that a negative value of the net gas superficial velocity in the emulsion phase of the bed would not be
NU
per se undesirable from the standpoint of the effective recirculation of solids, but it
MA
might be unacceptable when “leakage” of gas from the supply chamber to the recycle chamber must be prevented due to process constraints. This is the case, for instance, when the loop seal is used as solids non-mechanical valve in dual interconnected
TE
D
fluidized bed systems with distinct reactive environments. The relationship (1) is satisfied in all the experimental conditions tested, except when Ug,ls=Umf. In tests
AC CE P
carried out at Ug,ls approaching Umf, the supply chamber of the loop seal was largely defluidized, consistently with the pressure drop data and with the analysis of bed solids tracing, and extensive “leakage” of gas took place from the supply chamber to the recycle chamber.
4. CONCLUSIONS The hydrodynamics of a loop-seal operated as solids re-injection device in a lab-scale cold CFB apparatus was characterized by a combination of methods, including solids and gas tracing. For a given riser gas superficial velocity, the loop seal fluidization velocity exerts a decisive influence on the hydrodynamics of the loop seal, as it significantly affects the gas and solids flow patterns in the supply and recycle chambers of the loop-seal. As the loop seal is kept at gas superficial velocities just beyond incipient fluidization, the supply
ACCEPTED MANUSCRIPT
chamber is only partly fluidized and an extended stagnant zone is observed at its bottom.
PT
As fluidization becomes more vigorous, stagnant zone extension is reduced and more solids are transferred to the riser. Analysis of solids and gas tracing tests highlights the role
RI
of solids downflow in the supply chamber and of gas cross-flow from the supply to the
SC
recycle chamber on the fluidization patterns. An approximate criterion is given for the onset of gas “leakage” between the two loop seal sections, relevant to loop seal applications in
MA
NU
dual interconnected fluidized beds.
5. ACKNOWLEDGMENTS
The present study has been carried out in the framework of the project: “Ricerca di Sistema
D
- Accordo MiSE-CNR Gruppo Tematico Carbone Pulito - Fondo per il Finanziamento
AC CE P
TE
Attività di Ricerca e Sviluppo di Interesse Generale per il Sistema Elettrico Nazionale”.
LITERATURE CITED
1. Hughes, R.W., Lu, D.Y., Anthony, E. J. and Macchi, A., Design, process simulation and construction of an atmospheric dual fluidized bed combustion system for in situ CO2 capture using high-temperature sorbents, Fuel Processing Technology, 86, 1523-1531, (2005). 2. Kronberger, B., Lyngfelt, A., Löffler, G., and Hofbauer, H., Design and Fluid Dynamic Analysis of a Bench-Scale Combustion System with CO2 Separation-ChemicalLooping Combustion, Industrial Engineering Chemistry & Research, 44, 546-556, (2005). 3. Hossain, M.M. and de Lasa, H. I., Chemical-looping combustion (CLC) for inherent CO2 separation-a review, Chemical Engineering Science, 63, 4433-4451, (2008). 4. Pfeifer, C., Proll, T., Punchner, B. and Hofbauer, H., H2-rich syngas from renewable source bu dual fluidized bed steam gasification of solid biomass. In Bi, X., Berruti, F. & Pugsley, T., (Eds.), Fluidization XII (pp. 889-896) Engineering Foundation. New York, (2007). 5. Kaiser, S., Löffler, G., Bosch, K. and Hofbauer, H. Hydrodynamics of a dual fluidized bed gasifier. Part II: simulation of solid circulation rate, pressure loop and stability, Chemical Engineering Science, 58, 4215-4223, (2003). 6. Bai, D., Issangya, A.S., Zhu, J.-X. and Grace, J.R., Analysis of the Overall Pressure Balance around a High-Density Circulating Fluidized Bed, Industrial Engineering Chemistry & Research, 36, 3898-3903, (1997).
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
7. Cheng, L., Basu, P. and Cen, K., Solid circulation rate prediction in a pressurized loop seal, Transactions of the Institution of Chemical Engineers, 76, 761-763, (1998). 8. Cheng, L. and Basu, P., Effect of pressure on loop seal operation for a pressurized circulating fluidized bed, Powder Technology, 103, 203-211, (2000). 9. Basu, P. and Cheng, L., An analysis of loop seal operations in a circulating fluidized bed, Transactions of the Institution of Chemical Engineers, 78, 991-998, (2000). 10. Botsio, E. and Basu, P., Experimental Investigation into the Hydrodynamics of Flow of Solids through a Loop Seal Recycle Chamber, Canadian Journal of Chemical Engineers, 83, 554-558, (2005). 11. Kim, S.W., Namkung, W. and Kim, S.D., Solids Flow Characteristics in Loop-seal of a Circulating Fluidized Bed, Korean Journal of Chemical Engineers, 16, 82-88, (1999). 12. Kim, S.W., Namkung, W. and Kim, S.D., Solid recycle characteristics of loop-seals in a circulating fluidized bed, Chemical Engineering Technology, 24, 843-849, (2001). 13. Kim, S.W., Kim, S.D. and Lee, D.H., Pressure balance for circulating fluidized beds with a loop-seal, Industrial Engineering Chemistry & Research, 41, 4949-4956, (2002). 14. Kim, S.W. and Kim, S.D., Effects of particle properties on solids recycle in loop-seal of a circulating fluidized bed, Powder Technology, 124, 76-81, (2002). 15. Basu, P. and Buttler, J., Studies on the operation of loop-seal in circulating fluidized bed boilers, Applied Energy, 86, 1723-1731, (2009). 16. Geldart D., Gas Fluidization Technology, Wiley, New York, (1986).
ACCEPTED MANUSCRIPT
TABLE CAPTIONS Properties of the granular solids investigated
PT
Table 1
FIGURE CAPTIONS
RI
SC
NU
Figure 6 Figure 7
MA
Figure 5
D
Figure 4
TE
Figure 2 Figure 3
Experimental apparatus: (A) cyclone; (B) solids reservoir; (C) loop seal; (D) loop seal magnification. Pressure loops for different Ug,ls/Umf ratio;Ug,r = 2.85 m/s. Solids mass flux (A), pressure drops (B) and bed height (C) in the loop-seal chambers as a function of Ug,ls/Umf ratio;Ug,r = 2.85 m/s. Solids mass flux (A) and bottom bulk density (B) as a function of Ug,r for different Ug,ls/Umf ratio. Solids flow patterns in the loop seal: trajectories and time-resolved velocity of selected tracer particles as a function of Ug,ls. Outline of the gas flow pattern in the loop seal. Inlet and outlet CO2 concentrations in the loop seal chambers as a function of Ug,ls during gas tracing experiments.
AC CE P
Figure 1
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 1
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 2
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 3
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 4
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Figure 5
TE
AC CE P
Figure 6
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
Figure 7
ACCEPTED MANUSCRIPT
Table 1. Properties of the granular solids.
TE
D
MA
PT
Quartz 368 250-500 2300 0.09 2.67
RI
NU
calculated
AC CE P
(1)
Quartz 253 100-400 2300 0.047 1.77
SC
Material Sauter mean diameter (dp), m Size range, m Particle density (s), kg/m3 Incipient fluidization velocity (Umf)(1), m/s Terminal velocity (Ut)(1), m/s
AC CE P
Graphical Abstract
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights The hydrodynamics of a loop seal operated as solids reinjection device in a cold
PT
CFB was investigated.
RI
Gas and solids flow patterns in the loop seal was characterized by means of solids and gas tracing.
Solids downflow from standpipe and gas cross-flow between loop seal chambers
NU
play a key role on fluidization patterns.
MA
An approximate criterion for the onset of gas “leakage” between the loop seal
TE
D
sections was given.
AC CE P
SC