Epstein N., Grace J.R. Spouted and Spout-Fluid Beds: Fundamentals and Applications
Подождите немного. Документ загружается.
Gasification, pyrolysis, and combustion 267
37. A. C. L. Lisboa and A. P. Watkinson. Pyrolysis with partial combustion of oil shale fines in a
spouted bed. Can. J. Chem. Eng., 70 (1992), 983–990.
38. K. C. Teo and A. P. Watkinson. Rapid pyrolysis of Canadian coals in a miniature spouted bed
reactor. Fuel, 65 (1986), 949–959.
39. A. Jarallah and A. P. Watkinson. Pyrolysis of Western Canadian coals in a spouted bed. Can.
J. Chem. Eng., 63 (1985), 227–236.
40. T. Tam. Pyrolysis of oil shale in a spouted bed pyrolyser. M.A.Sc. thesis, University of British
Columbia (1987).
41. G. Rovero and A. P. Watkinson. A two-stage spouted bed process for auto-thermal pyrolysis
or retorting. Fuel Proc. Technol., 26 (1990), 221–238.
42. R. Aguado, M. Olazar, M. San Jos
´
e, G. Aguirre, and J. Bilbao. Pyrolysis of sawdust in
a conical spouted bed reactor. Yields and product composition, Ind. Eng. Chem. Res., 39
(2000), 1925–1933.
43. A. K. Janarthanan and L. D. Clements. Gasification of wood in a pilot scale spouted bed
gasifier. In Dev. Thermochem. Biomass Conversion,vol.2,ed.A.V.BridgwaterandD.G.B.
Boocock (Amsterdam: Kluwer Academic, 1997), pp. 945–959.
44. M. Chen, J. Wang, X. Wang, X. Zhang, S. Zhang, Z. Ren, and Y. Yan. Fast pyrolysis of biomass
in a spout-fluidized bed reactor–analysis of composition and combustion characteristics of
liquid product from biomass. Chin. J. Proc. Eng., 6 (2006), 192–96.
45. J. Jurewicz and A. Lemoine. Solid oxide fuel cell combustible by plasma spouted bed gasi-
fication of granulated biomass. Prep. Pap. Amer. Chem. Soc. Div. Fuel Chem., 49 (2004),
916–917.
46. C. J. Lim, S. K. Barua, N. Epstein, J. R. Grace, and A. P. Watkinson. Spouted bed and
spout-fluid bed combustion of solid fuels. In Fluidised Combustion: Is It Achieving Its
Promise? Third International Fluidised Bed Combustion Conference, Institute of Energy,
London (1984).
47. J. C. Zhao, C. J. Lim, and J. R. Grace. Coal burnout times in spouted and spouted-fluid beds.
Chem. Eng. Res. Dev., 65 (1987), 426–429.
48. J. C. Zhao, C. J. Lim, and J. R. Grace. Flow regimes and combustion behavior in coal-burning
spouted and spout-fluid beds. Chem. Eng. Sci., 42 (1987), 2865–2875.
49. C. J. Lim, A. P. Watkinson, G. K. Khoe, S. Low, N. Epstein, and J. R. Grace. Spouted, fluidized
and spout-fluid bed combustion of bituminous coals. Fuel, 67 (1988), 1211–1217.
50. F. W. Shirley and R. D. Litt. Advanced atmospheric fluidized-bed combustion design-spouted
bed. Final Report to U.S. Department of Energy from Battelle Columbus Laboratories, Febru-
ary 1988.
51. H. B. Vuthaluru, T. M. Linjewille, D. Zhang, and A. R. Manzoori. Investigations into the
control of agglomeration and defluidisation during fluidised-bed combustion of low-rank
coals. Fuel, 78 (1999), 419–425.
52. H. B. Vuthaluru and D. Zhang. Control methods for remediation of ash-related problems in
fluidised-bed combustors. Fuel Proc. Technol., 60 (1999), 145–156.
53. H. B. Vuthaluru, D. Zhang, and T. M. Linjewile. Behaviour of inorganic constituents and ash
characteristics during fluidised-bed combustion of several Australian low-rank coals. Fuel
Proc. Technol., 67 (2000), 165–176.
54. H. B. Vuthaluru and D. Zhang. Effect of coal blending on particle agglomeration and deflu-
idisation during spouted-bed combustion of low-rank coals.
Fuel Proc. Technol., 70 (2001),
41–51.
268 A. Paul Watkinson and Antonio C. L. Lisboa
55. M. J. San Jos
´
e, M. J. Aguado, S. Alvarez, and J. Bilbao. Combustion of sawdust and forestry
waste in conical spouted beds. Rev. de Inform. Tecnol. 13 (2002), 133–137.
56. M. Khoshnoodi and R. Z. Abidin. Utilization of Malaysian peat in spouted bed burner. In
Proceedings of the 10th Malaysian Chemical Engineering Symposium (Penang: Institute of
Chemical Engineers, Malaysia, 1994).
57. E. R. Altwicker, R. K. N. Konduri, and M. S. Milligan. Spouted bed combustor for the study
of heterogeneous hazardous waste incineration. AIChE Meeting, Philadelphia, Paper 82c
(1989).
58. R. S. Baker and R. C. Wilkinson. The commercial development of spouted/fluidised bed
combustion in Australia. In Fluidised Combustion: Systems and Applications. Institute of
Energy Symposium Series, Vol. 1, No. 4 (1980), pp. 1–11.
59. D. O. Albina. Combustion of rice husk in a multiple-spouted fluidized bed. Energy Sources,
25 (2003), 893–904.
60. D. O. Albina, Emissions from multiple-spouted and spout-fluid beds using rice husks as fuel.
Renew. Energy, 31 (2006), 2152–2163.
61. M. Barz. Sewage sludge combustion in a spouted bed cascade system. China Particuology, 1
(2003), 223–228.
62. M. G. Rasul. Spouted bed combustion of wood charcoal: performance comparison of three
different designs. Fuel, 80 (2001), 2189–2191.
63. H. A. Arbib and A. Levy. Combustion of low heating value fuels and wastes in the spouted
bed. Can. J. Chem. Eng., 60 (1982), 528–531.
64. J. Bilbao, M. Olazar, and M. J. San Jos
´
e. Dispositivo para la combustion en continuo de
residuos solidos, Oficina Espanola de Patentes y Marcas, 2,148,026 (2001), 9 pages.
16 Spouted bed electrochemical
reactors
J. W. Evans and Vladim
´
ır Ji
ˇ
ri
ˇ
cn
´
y
16.1 Introduction
Electrochemical reactions are used in the chemical, metallurgical, pharmaceutical, and
other industries, as such reactions typically allow precise control of reactions by adjust-
ment of voltage. Such reactions usually carry a high capital and energy cost, however,
and there has been much investigation to improve space-time yield, current efficiency,
and electric power consumption.
Figure 16.1 shows schematically the geometric arrangement of several electrochemi-
cal reactors. Broadly, these designs have two electrodes separated by a liquid electrolyte,
and in some instances a separator capable of ion transport. Electrochemical reactions
are inherently heterogeneous, occurring at the interface between an electronic and an
ionic conductor; mass transport of ions to and from such interfaces therefore plays an
important role. Unlike other heterogeneous reactions, such as reaction of a gas on a solid
catalytic surface, transport on the other side of the interface (of charge, i.e., electrical
current) is also significant. The need for this latter conduction motivates many aspects
of design in Figure 16.1, as discussed in this chapter. An impor tant parameter of the
designs in Figure 16.1 is the electrode surface area per unit of reactor volume; the higher
this value, the higher the space-time yield, other things being equal. Moving along the
top row and down to the bottom row should provide designs in which space-time yields
are high and perhaps designs in which high mass transfer rates can be exploited as well.
Another design consideration arises from the nature of the reaction products: solid,
gas, or solution. If no additional phases are formed, such as in an organic synthesis in
which reaction products at both electrodes could remain in solution, then no issue arises.
If a solid is formed, as in the electrodeposition of metals from solution, then designs
such as the narrow gap cell or the Swiss-roll cell are unsuitable, because their flow paths
are rapidly blocked by deposits. This is true of the extreme form of the narrow gap cell,
the electrochemical microreactor in which the interelectrode gap is typically of the order
of 100 μm. In the case of gas evolution at an electrode, the possible interference of gas
bubbles with either electrolyte flow or conduction in the electrolyte must be considered.
From the considerations of the last paragraph, it is clear that reactors in which the
electrodes are particulate, and there is relative motion between particles and electrolyte,
Spouted and Spout-Fluid Beds: Fundamentals and Applications, eds. Norman Epstein and John R. Grace.
Published by Cambridge University Press.
C
Cambridge University Press 2011.
270 J. W. Evans and Vladim
´
ır Ji
ˇ
ri
ˇ
cn
´
y
Planar
electrodes
Narrow gap cell ''Swiss-roll'' cell
Fluidized bed of Rotating cylinder
conductive or inert
particles
electrode
External draft tube
Spouted beds
electrodes
separators
Internal draft tube
Figure 16.1. Electrochemical cell principles.
might be especially beneficial. For reactions in which neither gases nor solids form, a
packed bed of particles could be expected to function well.
1
(Electrodes that are porous
are similar.) If solids or gases form, an electrode in which the particles are in motion
would seem preferable – that is, a fluidized or spouted bed, or a bed in which particles are
mechanically agitated, as in rotating cylinder electrodes. In all these cases the electrodes
become, in a sense, three-dimensional; when operating properly, electrochemistry occurs
throughout the whole volume of the bed of particles.
16.2 Spouted bed electrodes
The sketches of particulate electrodes in Figure 16.1 are incomplete in that they do not
show how electrical connection is made to the particles, nor do they show the counter-
electrode that completes the cell design. More detailed designs appear in Figure 16.2.
Simplest in concept is the spouted bed electrode (SBE) of A, in which a central jet of
electrolyte in a cylindrical-conical bed of particles sweeps a flow of particles upward.
At the top of the bed the jet (spout) expands, and the upswept particles fall back onto the
top of the bed. The particles then move slowly downward in an annular region, dense
with particles, in a manner characteristic of spouted beds, as described throughout this
volume. Just visible in the sketch is a “current feeder” (or “current collector,” depending
on whether the bed acts as an anode or cathode) in the form of a conducting material
placed outside the bed. This is electrically connected to one side of the DC power supply.
Above the bed is a counterelectrode, such as a disk of conducting material, connected to
the other side of the power supply. From the previous section, it will be clear that only
particles that are in electrical connection with the power supply will function electro-
chemically. Therefore particles in the annular region will be active, but those in the spout
Spouted bed electrochemical reactors 271
Anode
Metal particle
Current collector
Fountain
Annulus
Spout
Nozzle
A
DE
BC
Anolyte
in
V
N
- Nozzle electrolyte flowrate
Catholyte in
Anodic
connection
Cathodic
connection
Electrolyte
outlet
Particle
distributor
Anode
draft
tube
Cathode
Electrolyte
inlet
DSA
anode
Catholyte
out
Cathodic
current
collector
Anolyte
out
Weir
Annulus
Draft tube
Falling region
Annulus
Nozzle
Electrolyte drain
Bead drain
Cell
Catholyte
outlet
Circulating
particulate
bed
Flow
distributor
Catholyte
inlet
Anolyte
inlet
Anolyte
outlet
Anode
θ
Figure 16.2. Geometrical arrangements of SBEs: (A) conventional conical-cylindrical spouted bed
without draft tube or separator
1
; (B) conical-cylindrical spouted bed with draft tube and
separator
2
; (C) rectangular spouted bed with draft tube and separator
11, 12
; (D) conical-cylindrical
SBE with draft tube and without separator
3,4,34
; (E) circulating rectangular SBE with separator
and without draft tube.
35
will not. (Less facile electrical contact of the particles is one disadvantage of fluidized
bed electrodes compared with SBEs.)
Design A suffers from the spreading of the upflowing jet of electrolyte and particles
higher in the bed, so that above one to a few bed diameters the cross-section is mostly
fluidized, with poor electrical contact of the particles. Design B, an SBE with a central
draft tube, is intended to allow vertical scaleup, while maintaining the slowly moving
272 J. W. Evans and Vladim
´
ır Ji
ˇ
ri
ˇ
cn
´
y
downward bed in the annular region. In this design the draft tube also functions as a
current collector, and the peripher y of the bed is contained within a separator, outside
of which sits the cylindrical counterelectrode. The remaining designs in Figure 16.2
are similar in concept to the second, but intended to facilitate aspects of cell operation,
such as the addition and removal of particles. In design C the geometry is no longer
cylindrical; the flat current feeder sits in front of the plane of the figure, whereas the flat
counterelectrode sits behind that plane, with a separator between it and the bed so there
is no direct electrical contact between counterelectrode and bed. If the counterelectrode
is a mesh, then the separator is conveniently supported by the counterelectrode. The
bed in C has a weir, which is intended to limit vertical growth of the bed in the case of
particle growth by deposition (not shown is a discharge of particles from the volume to
the right of the weir). Much of the rest of this chapter is devoted to SBEs of a design
similar to C, particularly their application to the electrodeposition of metals.
SBEs with draft tubes have been extensively studied for use in metal electrodeposition
from aqueous solutions in which the bed is connected to the negative side of the power
supply (via the current collector) and the particles grow as metal deposits on them
because of cathodic reaction. Typically the reaction at the counterelectrode (an anode)
evolves oxygen.
In metal electrodeposition, particle movement must be sufficient that particles are not
joined by electrodeposited metal, lest the bed stop moving and become a rigid mass.
Consequently, care must be taken to avoid “dead regions” in the SBE in which particles
are stationary, or at least these regions must be electrochemically inactive. The jet of
electrolyte entering through a nozzle at the bottom of the bed typically moves mostly
upward through the draft tube and annular region (although at high particle circulation
rates the motion of electrolyte in the annular region can be partly or wholly downward).
Particles from the bed fall though the gap between the bottom of the draft tube and
the (inclined) sidewall of the bed and are then swept up the draft tube. At the top of
the draft tube the particles f all out of the spout of electrolyte and land on top of the
bed to again undergo a slow descent, during which they are in electrical contact with
the current collector (mostly via their fellow particles) and undergo electrochemical
reaction.
Particle circulation, electrolyte flow, and electrochemical reactions are affected by
features of the cell design, notably:
r
Shape of the reactor body
r
Draft tube(s) position and shape
r
Annular regions – shape, dimensions, wall at bottom
r
Electrolyte nozzle
r
Electric current and liquid flow arrangements
r
Separator(s)
16.2.1 Shape of the reactor body
Early versions of SBEs were mostly cylindrical.
1–5
This geometry is particularly conve-
nient if design A of Figure 16.2 is used, in which the macroscopic flow of electrolyte and
Spouted bed electrochemical reactors 273
electric current are both vertical; a separator is then unnecessary. However, the design i s
not efficient in its use of either (floor) space or energy. There are limitations on scaleup
in the direction of current flow, because the bed of particles is fairly conductive and,
therefore, the path of least resistance within the bed itself is particle to particle, rather
than through the electrolyte. It is therefore the particles closest to the counterelectrode
that are most electrochemically active, whereas those close to the bottom of the bed
are essentially inert, particularly for tall beds. Consequently, an SBE of this design
usually leads to a reactor of large horizontal dimensions if significant productivity is
to be achieved. Energy consumption per unit of production is largely a matter of cell
voltage, and design A is likely to have a high voltage owing to the need to conduct
electricity through a few centimeters of electrolyte between the top of the bed and the
counterelectrode. This might also be true for design D.
A more extensively studied reactor shape is that of design C. This design has been
used for both laboratory-scale studies,
2,6–10
and pilot-scale investigations
11,12
of electro-
deposition of metals, including zinc (from both acid and alkaline electrolytes) and copper.
It has also been used to study scaleup, from the viewpoint of ensuring proper particle
motion, including beds with multiple draft tubes.
13
The design has the disadvantage of
needing a separator, but the proximity of the counterelectrode and bed, separated by
only a few millimeters of separator, yields reasonable cell voltages, even at high current
density (current per unit of separator area).
16.2.2 Draft tube
The draft tube of design C in Figure 16.2 is internal to the bed of particles; this is the com-
mon arrangement. However, external draft tubes have also been used (see Figure 16.1)
and might then better be called hydraulic lifts.
Work at Berkeley
13
determined that the particle circulation rate increases as the
position of the bottom of an internal draft tube is raised. If the draft tube is too
high, however, it becomes difficult to start (restart) particle motion as there is ini-
tially a tall packed bed immediately above the nozzle, restricting its flow. The work at
Berkeley has also determined the effect of such design parameters as draft tube width
and length and the slope of the inclined surfaces bounding the bottom of the bed. The
last were designed to eliminate “dead zones” in the corners, in which particles were
stationary.
SBEs with external draft tubes have worked well, as particle motion is easy to start.
The draft tube is longer than an internal one, however, and particle residence time in the
draft tube might be greater. This can be an issue i f undesired side reactions occur in the
draft tube – for instance, in the chemical dissolution of zinc in acidic electrolytes when
the intention is to electrodeposit zinc in the annular region.
16.2.3 Annular regions
The regions on either side of the draft tube are referred to as “annular,” even though
they are not truly annular in shape in some of the designs of Figure 16.2. These are the
274 J. W. Evans and Vladim
´
ır Ji
ˇ
ri
ˇ
cn
´
y
regions in which electrochemistry takes place and in which, in the case of deposition of
solids, sustained gentle downward motion of the particles is necessary. Such motion is
encouraged by smooth walls and by eliminating horizontal surfaces or corners in which
particles are likely to rest. For example, at a transition from a vertical to an inclined
surface, it is best to have a curve with a radius much larger than that of the largest
particles.
An alternative to inclined walls at the bottom of the annular region is to allow corners,
such as those in design C, but to arrange the current collector and counterelectrode so
that they are not near the regions of stationary particles.
16.2.4 Nozzle
Some investigation of nozzle design was reported by Verma et al.
13
Robinson et al.
11,12
developed a nozzle relying on a Venturi effect for sweeping the particles into a draft
tube.
16.2.5 Electric current and liquid flow arrangements
Several aspects of current and liquid flow have been discussed already. An additional
point can be seen in design C (Figure 16.2), in which a flow diverter in the shape of an
inverted V is placed above the draft tube to encourage disengagement of particles from
the spout.
11,12
In this way, car ryover of particles via the electrolyte exit is minimized,
even for a small headspace above the bed.
16.2.6 Separators
The separator has the purpose of preventing electrical contact between the particles in
the annular regions and the counterelectrode in designs B, C, and E of Figure 16.2. It
is of a material that is not an electronic conductor, but allows the passage of at least
some of the ions in solution (conduction of current is otherwise impossible). Porous
plastics
7,10,14
or plastic meshes have been the most common separators, although ion
exchange membranes are another possibility. It is essential that the particles not stick
to the separator. A particle stuck in this position, when the bed is electrochemically
active, forms a growing deposit that is likely to extend through the separator and short
the cell. Smoothness is therefore an important characteristic of a suitable membrane, as
are toughness, to resist the abrasion by particles, and low resistance to ionic conduction.
If the separator has limited permeability to electrolyte flow, it becomes possible to
operate the cell with somewhat different electrolytes on the two sides of the cell. For
example, in electrowinning zinc, one might use a highly acidic (highly conducting)
electrolyte on the anode side (to minimize cell voltage) with a less acidic electrolyte
on the cathode side (which has the effect of increasing current efficiency). The authors
of this chapter believe that identifying a suitable separator material is usually the most
difficult task in developing the SBE for applications entailing solid deposition, such as
metal electrowinning.
Spouted bed electrochemical reactors 275
16.3 Applications of SBEs
As with all electrodes, the SBE obeys Faraday’s law so that the productivity (say, in
kg/s) is proportional to the product of current and current efficiency. There is, therefore,
incentive to increase the current. Ultimately there is a limit, imposed by mass transfer
from electrolyte to electrode surface, on the current achievable in a cell, but it is usual
to operate cells well below this limiting current. Increase in current typically results
in an increase in cell voltage and sometimes in a decrease in current efficiency. The
consequence is that the electrical energy expenditure per unit of product is proportional
to cell voltage over current efficiency and increases with current. A compromise between
lower capital cost per unit of product and higher energy cost per unit of product is
therefore a feature of all cell development, including in application of SBEs. The SBE
usually has an advantage over more conventional designs in that it can be operated at
greater current density (current per unit area of separator or, equivalently, space-time
yield) for the same voltage.
The driving force for electrochemical reactions is the overpotential at the electrode
in question. This is the difference between the electrode potential at which it is in
equilibrium with the adjacent electrolyte and the actual electrode potential. The r ate of the
electrochemical reaction (say in moles/m
2
s) is a complicated function of overpotential,
mass transfer, electrolyte composition and temperature. The reader is referred to standard
texts on electrochemistr y for details.
15
In the case of SBEs it is usual for the overpotential
to vary across the electrode. Potential gradients are necessary, both in the electrolyte and
from particle to particle, t o drive current; it is uncommon for these two gradients to match,
as particle-to-par ticle conduction in the SBE is better than conduction in the electrolyte.
Hence the overpotential, reaction rate, and local space-time yield are typically highest
at positions closest to the counterelectrode and small deeper into the bed. Consequently
the overall space time yield of the SBE increases only marginally beyond a certain bed
thickness (direction of macroscopic current flow), and most applications have employed
beds only a centimeter or two in thickness. Scaleup is then in the two dimensions parallel
to the separator or counterelectrode, coupled with stacking of cells next to each other.
Figure 16.3 illustrates (left side) laboratory scale cells (one disassembled) used at UC
Berkeley and (right side) the scaleup of a cell to a final pilot-scale cell with dual draft
tubes. Models based on the porous-electrode theory of Newman and Tobias
16
have been
advanced to guide in the selection of a suitable bed thickness or other dimensions.
4,17
Models aiding in prediction of particle and electrolyte hydrodynamics have also been
published.
17
16.3.1 Metal electrowinning
Most of the world’s zinc and much of its copper are produced by technology that entails
the electrodeposition of metals from aqueous solutions generated by leaching ores or
processed concentrates. Similar technology is seen in the production of lead, nickel,
cobalt, and other metals. This production is almost entirely in cells with designs that
date back well over a century. The cells contain numerous parallel slablike electrodes
276 J. W. Evans and Vladim
´
ır Ji
ˇ
ri
ˇ
cn
´
y
Figure 16.3. Scaleup of SBE developed at UC Berkeley. The two devices in the left-hand
photograph (one shown disassembled) have beds approximately 100 mm in width. One of these
also appears in the right photograph (far right), in which the largest SBE is 270 mm wide.
Table 16.1. Comparison of conventional and spouted bed electrowinning.
Technology
Current density
(A/m
2
)
Voltage
(V)
DC electrical energy
consumption
(kWh/kg metal)
Copper: conventional 200−300 2 2
Copper: SBE 1200−4500 1.6−2.5 1.3−3
Copper: SBE – matte leaching 1100−3300 0.9−2.2 0.8−1.9
Zinc: conventional 300 3.5 3.3
Zinc: SBE – acid electrolyte 1000−5000 2.7−3.7 2.8−3.4
Zinc: SBE – alkaline electrolyte 2500−4400 3.0−3.2 2.5
(about a meter in size) that are alternating cathodes and anodes. In these conventional
cells, the cathodes grow from thin “starter sheets” of the same metal or are deposits
forming on “cathode blanks” of some inert metal. The anodes are “inert” in a sense that
they are i ntended not to pass into solution, but rather to be the site of another reaction,
such as the generation of oxygen. Electrodes are spaced a few centimeters apart. An
important parameter indicative of the space-time yield is the current density (current
per unit area of slab surface), whereas cell voltage is indicative of energy consumption
per kilogram of metal. Both these economic parameters are adversely affected by lower
current efficiency. Table 16.1 gives representative values for commercial electrowinning
of copper and zinc,
18,19
as well as typical results from laboratory and pilot-scale SBEs.
Electrowinning of metals in SBEs is well illustrated by the case of copper and zinc.
The standard electrode potential for copper is greater than that of hydrogen, which means
that, even in acidic copper salt electrolytes, copper, instead of hydrogen, deposits onto
a cathode (by reduction of hydrogen ions). Hence copper electrowinning from copper
sulfate electrolytes typically proceeds at high current efficiency in either conventional