- •Recovered Paper and Recycled Fibers
- •Isbn: 3-527-30999-3
- •Introduction
- •Isbn: 3-527-30999-3
- •Isbn: 3-527-30999-3
- •2006, Isbn 3-527-30997-7
- •Volume 1
- •Isbn: 3-527-30999-3
- •4.1 Introduction 109
- •4.2.5.1 Introduction 185
- •4.3.1 Introduction 392
- •5.1 Introduction 511
- •6.1 Introduction 561
- •6.2.1 Introduction 563
- •6.4.1 Introduction 579
- •Volume 2
- •7.3.1 Introduction 628
- •7.4.1 Introduction 734
- •7.5.1 Introduction 777
- •7.6.1 Introduction 849
- •7.10.1 Introduction 887
- •8.1 Introduction 933
- •1 Introduction 1071
- •5 Processing of Mechanical Pulp and Reject Handling: Screening and
- •1 Introduction 1149
- •Isbn: 3-527-30999-3
- •Isbn: 3-527-30999-3
- •Isbn: 3-527-30999-3
- •Isbn: 3-527-30999-3
- •Introduction
- •Introduction
- •Isbn: 3-527-30999-3
- •1 Introduction
- •1 Introduction
- •1 Introduction
- •1 Introduction
- •1 Introduction
- •1 Introduction
- •150.000 Annual Fiber Flow[kt]
- •1 Introduction
- •1 Introduction
- •Introduction
- •Isbn: 3-527-30999-3
- •Void volume
- •Void volume fraction
- •Xylan and Fiber Morphology
- •Initial bulk residual
- •4.2.5.1 Introduction
- •In (Ai) Model concept Reference
- •Initial value
- •Validation and Application of the Kinetic Model
- •Inititial
- •Viscosity
- •Influence on Bleachability
- •Impregnation
- •Impregnation
- •Impregnation
- •Impregnation
- •Impregnation
- •Impregnation
- •Impregnation
- •Impregnation
- •Impregnation
- •Impregnation
- •Introduction
- •International
- •Impregnation
- •Influence of Substituents on the Rate of Hydrolysis
- •140 116 Total so2
- •Xylonic
- •Viscosity Brightness
- •Xyl Man Glu Ara Furf hoAc XyLa
- •Initial NaOh charge [% of total charge]:
- •Introduction
- •Isbn: 3-527-30999-3
- •Introduction
- •Isbn: 3-527-30999-3
- •Introduction
- •Introduction
- •Isbn: 3-527-30999-3
- •In 1950, about 50% of the global paper production was produced. This proportion
- •4.0% Worldwide; 4.2% for the cepi countries; and 4.8% for Germany.
- •1150 1 Introduction
- •1 Introduction
- •1 Introduction
- •Virgin fibers
- •74.4 % Mixed grades
- •Indonesia
- •Virgin fibers
- •Inhomogeneous sample Homogeneous sample
- •Variance of sampling Variance of measurement
- •1.Quartile
- •3.Quartile
- •Insoluble
- •Insoluble
- •Insoluble
- •Integral
- •In Newtonion liquid
- •Velocity
- •Increasing dp
- •2Α filter
- •0 Reaction time
- •Increasing interaction of probe and cellulose
- •Increasing hydrodynamic size
- •Vessel cell of beech
- •Initial elastic range
- •Internal flow
- •Intact structure
- •Viscosity 457
- •Isbn: 3-527-30999-3
- •1292 Index
- •Visbatch® pulp 354
- •Index 1293
- •1294 Index
- •Impregnation 153
- •Viscosity–extinction 433
- •Index 1295
- •1296 Index
- •Index 1297
- •Inhibitor 789
- •1298 Index
- •Index 1299
- •Impregnation liquor 290–293
- •1300 Index
- •Industries
- •Index 1301
- •1302 Index
- •Index 1303
- •Xylose 463
- •1304 Index
- •Index 1305
- •1306 Index
- •Index 1307
- •1308 Index
- •In conventional kraft cooking 232
- •Visbatch® pulp 358
- •Index 1309
- •In prehydrolysis-kraft process 351
- •Visbatch® cook 349–350
- •1310 Index
- •Index 1311
- •1312 Index
- •Viscosity 456
- •Index 1313
- •Viscosity 459
- •Interactions 327
- •1314 Index
- •Index 1315
- •Viscosity 459
- •1316 Index
- •Index 1317
- •Xylose 461
- •Index 1319
- •Visbatch® pulp 355
- •Impregnation 151–158
- •1320 Index
- •Index 1321
- •1322 Index
- •Xylan water prehydrolysis 333
- •Index 1323
- •1324 Index
- •Viscosity 459
- •Index 1325
- •Xylose 940
- •1326 Index
- •Index 1327
- •In selected kinetics model 228–229
- •4OMeGlcA 940
- •1328 Index
- •Index 1329
- •Intermediate molecule 164–165
- •1330 Index
- •Viscosity 456
- •Index 1331
- •1332 Index
- •Impregnation liquor 290–293
- •Index 1333
- •1334 Index
- •Index 1335
- •1336 Index
- •Impregnation 153
- •Index 1337
- •1338 Index
- •Viscose process 7
- •Index 1339
- •Volumetric reject ratio 590
- •1340 Index
- •Index 1341
- •1342 Index
- •Index 1343
- •1344 Index
- •Index 1345
- •Initiator 788
- •Xylose 463
- •1346 Index
- •Index 1347
- •Vessel 385
- •Index 1349
- •1350 Index
- •Xylan 834
- •1352 Index
Introduction
Hydrocyclones are gravity separators of relatively simple mechanical design, and
have no moving parts. In order to function, they require an internal vortex flow as
well as a difference in densities between the liquor and the particles to be separated.
Figure 6.17 shows the streams around a hydrocyclone. The pulp is fed with the
feed stream QF at the concentration cF. The fraction containing the heavier particles
is concentrated in the underflow of the cyclone in the stream QU at the consistency
cU. The lighter solids are discharged with the overflow stream QO. The
underflow is also referred to as “apex flow” and the overflow as “base flow”.
QF, cF
QU, cU
QO, cO
Fig. 6.17 Streams around a hydrocyclone.
579
6.4.2
Flow Regime
The gravitational forces which drive the separation in a hydrocyclones are generated
as pressure energy is converted into rotational momentum. In the very basic
design (see Fig. 6.18), the feed flow enters the cyclone tangentially at the upper
end of the cylindrical section and induces a vortex around the axis of the cyclone.
As the suspension swirls downward at the perimeter of the cylindrical and conical
sections, heavy-weight material is concentrated near the wall and is eventually
dragged to the underflow at the apex of the cone. The balance of the liquor together
with the light-weight material rotates towards the axis of the cyclone and proceeds
to the overflow at the opposite end of the cyclone. The overflow escapes
through the vortex finder, a piece of pipe extending into the body of the cyclone
which helps to limit the short-circuit flow from the feed inlet to the overflow.
Overflow (Base)
Underflow (Apex)
Feed
Fig. 6.18 Flow pattern in a hydrocyclone.
Hydrocyclones develop an air core when one of the outlets discharges into atmosphere.
Modern cyclones used in cleaning operate at a backpressure and do
not have an air core.
The geometrical form of the hydrocyclone has a major influence on the separation
efficiency. While manufacturers have developed different designs mainly based
on experience, all of them target at maintaining a largely laminar flow regime.
Unlike pressure screens, hydrocyclones cannot profit from turbulence and the
resulting fluidization. They must be operated at low consistencies in order to
minimize particle–particle interactions. Both turbulence and higher consistencies
will considerably jeopardize the cleaning efficiency.
580 6 Pulp Screening, Cleaning, and Fractionation
The tangential velocity profile observed in a cyclone starts with a forced vortex
flow at the axis. It then passes into a free vortex flow via a transition zone before
the wall effect reduces the velocity to zero again (Fig. 6.19). The axial velocity profile
shows the flow towards the apex at the perimeter of the cyclone body and
towards the base around the center. The pressure loss in the cyclone is a function
of the friction, mainly at the cyclone walls and at the inner wall of the vortex finder.
Locus of zero
axial velocity
Free vortex
v T · r = constant
Forced vortex
v T : r = constant
v tangential
r
v axial
r
Fig. 6.19 Tangential (left) and axial (right) velocity profiles in a hydrocyclone [21].
6.4.3
Sedimentation
Let us make some basic considerations about sedimentation to better understand
what is happening in a hydrocyclone. A characteristic parameter describing sedimentation
is the terminal settling velocity. It describes the state where a particle
moves at constant speed under the influence of frictional and gravitational forces.
Figure 6.20 shows the main forces which act on a settling particle; these are the
weight FW, the buoyancy FB, and the drag force FD.
vsettling
Weight force
Buoyancy force
Drag force
Fig. 6.20 Forces acting on a settling particle.
6.4 Centrifugal Cleaning Theory 581
Weight and buoyancy depend on the specific weight of the particle and the displaced
liquor, respectively. The drag force is a function of the particle movement
and the particle shape. It is always directed in the opposite direction of the velocity
vector. In the Earth’s gravitational field, the three forces are defined as follows:
FW _ _S V g _12_
FB _ _L V g _13_
FD _ cDAP _L
v2s
2 _14_
where qS = density of the particle (kg m–3); qL = density of the liquid (kg m–3);
V = volume of the particle (m3); AP = area of the particle as seen in projection
along the direction of motion (m–2); g = acceleration due to gravity (9.81 m s–2);
cD = drag coefficient; and vS = settling velocity (m s–1).
In the steady state, the forces are in equilibrium, which means that
FW _ FB _ FD _ 0 _15_
Combining Eqs. (12–15) and solving for vS yields Newton’s law for the terminal
settling velocity in the Earth’s gravitational field:
vS _ _2_________________
cD
__S _ _L_
_L
V
AP
_ g _16_
In the special case of a spherical particle with the diameter d (m), where
V = d3p/6 and AP = d2p/4, the above expression becomes:
vS _ __4________________
3 cD
__S _ _L_
_L
_ d g _17_
Note that the settling velocity increases with the density difference between the
particle and the liquor and with the particle diameter. The drag coefficient
depends on the size and shape of the particle, on the viscosity and density of the
fluid, and on the settling velocity itself. When the sphere settles in a creeping,
laminar environment, Eq. (17) converts into Stokes’ law:
vS _
__S _ _L_d2
S g
18 l _18_
where l is the dynamic viscosity of the liquid (Pa·s).
582 6 Pulp Screening, Cleaning, and Fractionation
The solids contained in a pulp stream are very different in shape and size.
While sand particles may come close to spherical shape, pulp fibers obviously do
not. Likewise, there are wide ranges of particle densities from plastics to metals.
In addition, the reinforced gravitational field in the cyclone adds complexity to the
matter. Consequently, meaningful theoretical models for the settling of solids in a
pulp suspension during centrifugal cleaning are not available. We will therefore
use the general form of Newton’s law, as per Eq. (16), for the qualitative evaluation
of separation in a hydrocyclone.
vtangential
vradial
vsettling
r
Net gravitational force Drag force
Fig. 6.21 Forces acting on a particle in a hydrocyclone.
So, what is happening to a particle in a hydrocyclone? The tangential feed provokes
a tangential liquor velocity which makes the particle move along a circular
path around the axis of the cyclone (Fig. 6.21). A radial flow vector describes the
transport of the liquor from the feed inlet at the outer perimeter to the centrical
vortex finder. There is also an axial flow vector which is directed towards the apex
at the cyclone perimeter and towards the vortex finder around the axis. The forces
acting on the particle in a plain perpendicular to the axis are a drag force pointing
against the direction of the settling velocity, and gravitational forces as a function
of the different solid and liquid densities.
Clearly, the settling velocity must be larger than the radial velocity in the cyclone
for a particle to be separated to the underflow. Nevertheless, the tangential velocity
represents the most important flow vector in the hydrocyclone because it controls
the gravity forces acting on the particle.
The acceleration term is determined by the tangential velocity vT (m s–1) and the
distance between the particle and the center of rotation, r (m). When substituting
the acceleration due to the Earth’s gravity g by the centripetal acceleration vT
2/r,
Eq. (16) can be rewritten to give:
vS _ vT _2_________________
cD
__S _ _L_
_L
V
AP
1
r _ _19_
6.4 Centrifugal Cleaning Theory 583
Apparently, higher tangential flow velocities vT and smaller distances r increase
the settling velocity. This means that a cyclone of a smaller diameter is more efficient
for the removal of small particles than a large-diameter cyclone. Likewise,
higher tangential flow velocities improve the efficiency. Both the cyclone diameter
and the tangential velocity are physically limited by the necessity to maintain the
typical laminar flow pattern.
The density difference between the liquid and some particles (e.g., plastics or
light-weight wood components) may be very low. This means that high velocities
and small radii are needed for cleaning to be efficient. In a typical cleaner, the
centrifugal force is so much larger than the Earth’s gravity that it does not make
any difference whether the cleaner is installed vertically or horizontally.
For the cleaning of pulp, the relevant solids density qS is the apparent fiber density
– that is, the density of the swollen fiber consisting of the liquor-saturated
fiber wall and liquor-filled lumen. It has been suggested that for chemical pulp,
the influence of the fiber shape on the drag force and consequently on cD is not
significant [22].
The derivations described above are valid for particles which have a larger density
than the fluid. When a particle is lighter than the fluid, its weight becomes
smaller than the buoyancy, and the vector for the settling velocity shown in
Fig. 6.20 is directed upwards. This is when the particle begins to float to the surface
rather than settle to the bottom. Consequently, the drag force points downwards.
When then the terminal settling velocity is calculated in analogy to
Eq. (19), the solid and liquid densities in the numerator of the density term
change place:
vS _ vT _2_________________
cD
__L _ _S_
_L
V
AP
1
r _ _20_
So, the separation of light-weight particles to the overflow is controlled by the
same factors as the separation of heavy-weight particles to the underflow, the difference
being that there is no need to overcome the radial velocity for separation
to occur. In theory, this circumstance facilitates the separation of light-weight particles
compared to heavy-weight particles. However, in practice the density difference
between light-weight material and liquor is often very small, and any support
for obtaining a reasonable separation efficiency is welcome.
6.4.4
Underflow Thickening
In all cleaning operations, pulp fibers are heavier than liquor. Consequently, fibers
become concentrated in the underflow of the hydrocyclone. In an analogy to
screening, the thickening factor T is defined by:
T _
cU
cF _21_
584 6 Pulp Screening, Cleaning, and Fractionation
6.4 Centrifugal Cleaning Theory 585
The underflow thickening depends on the specific design of the cyclone applied,
with typical thickening factors ranging between 1.5 and 3.0.
Thickening leads on the one hand to poorer separation when particle–particle
interaction hinders the free movement of material to be separated. On the other
hand, thickening may cause plugging of the cone which takes the cyclone out of
operation. If underflow thickening is of major concern for a certain application,
special cyclones with a dilution near the apex can be employed.
6.4.5
Selective Separation
As in screening, the selective separation of different types of solids contained in
the feed stream is of major importance for all contaminant removal and fractionation
applications. As described earlier, separation in a hydrocyclone depends
mainly on differences in the particles’ densities and specific surfaces. Hence, the
selectivity of separation in a contaminant removal application improves with the
density difference between debris and pulp.
Likewise, the density difference between individual pulp fibers determines how
selectively they can be separated in a cyclone. The apparent density of a pulp fiber
results from its diameter and wall thickness. Thick-walled, smaller-diameter fibers
(as found in softwood latewood) have a higher apparent density than thin-walled,
larger-diameter fibers (as found in softwood earlywood). When pulp is subjected
to fractionation in a cyclone, the thick-walled fibers will proceed preferably to the
underflow, forming the coarse fraction. Despite their lower apparent density, thinwalled
fibers are still heavier than the liquor. They report not only to the overflow (fine
fraction) but also to the underflow. It is therefore easier to obtain a relatively pure fine
fraction in the overflow than to obtain a pure coarse fraction in the underflow [23].
0
5
10
15
20
25
30
0 2 4 6 8 10
Proportion in each class, %
Fiber wall thickness, μm
Fine fraction
Feed
Coarse fraction
Fig. 6.22 Example of fiber wall thickness distributions of feed and fractions
after several stages of centrifugal cleaning;bleached softwood kraft pulp [23].
Separation in a hydrocyclone is influenced by a variety of factors such as the
complex fiber morphology, particle–particle interactions and short-circuit flows
within the cyclone. In practice, separation in a single cyclone is far from ideal,
and several stages of cleaning are needed to obtain fractions of significantly different
character, such as those illustrated in Fig. 6.22.
6.5
Centrifugal Cleaning Parameters
In this subsection, we will review those parameters that affect the operation and
determine the performance of a cleaning system, and their qualitative influences
on the cleaning efficiency.
These parameters include operating conditions, such as flow rate and pressure
drop, feed consistency and temperature. They also include equipment-specific parameters,
mainly the cyclone diameter. In addition, we need to observe the furnish
characteristics of both the pulp fibers and the contaminants.
Some of the above parameters can be adjusted, but some are intrinsic to a special
process step or piece of equipment. The chosen combination of adjustable
cleaning parameters depends on the individual requirements of the application,
and is usually a compromise within performance limits and operating constraints,
because the optimization of single parameters often leads in opposite directions.
Due to the complexity of the involved mechanisms and the fact that system
design is usually based on rules of thumb with supportive testing, the discussion
of parameters below is of qualitative nature only.
6.5.1
Cyclone Parameters
Since there are no moving parts, the performance of a hydrocyclone is determined
by its geometry. Design details vary between cyclone manufacturers and target,
for instance, at the minimization of the short-circuit flow from the feed to the
overflow, at lower or higher reject thickening, or at the prevention of cone plugging.
The major parameter affecting cleaning efficiency is the cyclone diameter. At a
given pressure drop, cyclones of smaller diameter generate higher centrifugal
forces, but they also process lower flow rates. Hence, the cyclone size is subject to
an economical restriction given by the number of units to be installed for handling
a particular production capacity. Smaller units are also more sensitive to plugging
due to the small diameter of the underflow opening.
586 6 Pulp Screening, Cleaning, and Fractionation
6.5.2
Operating Parameters
6.5.2.1 Flow Rate and Pressure Drop
A higher pressure drop, which is a synonym for an increased flow rate and a higher
tangential flow velocity, improves the separation efficiency. Care must be taken
not to increase the tangential velocity beyond a point where turbulence occurs.
Since turbulence destroys the controlled flow pattern in the cyclone, it is highly
unwelcome in cleaning and must be avoided. In addition, the pressure drop influences
the operating costs of centrifugal separation, which are mainly determined
by the pumping energy required to overcome the pressure drop.
6.5.2.2 Feed Consistency
In order to limit flocculation, hydrocyclones are normally operated below about
0.6% feed consistency. An increase in feed consistency above this level leads to
reduced cleaning efficiency.
6.5.2.3 Temperature
Higher temperatures can have a positive effect on the cleaning efficiency due to
the reduced liquid viscosity. The maximum operating temperature of a pressurized
hydrocyclone is limited to 70–80 °C.
6.5.3
Furnish Parameters
6.5.3.1 Pulp Fibers
With respect to cleaning, pulp fibers are characterized mainly by their density, surface
texture, size, freeness and disruptive shear stress of the fiber network. Together
with the consistency, these properties determine the performance of the furnish
in a hydrocyclone.
The main parameter affecting separation is the apparent density of the fiber.
Depending on the nature of the furnish, this can mean that fibers are separated
according to wall thickness or coarseness. At the same fiber diameter, fibers with
thicker walls tend to be rejected to the underflow. At the same coarseness, fibers
with smaller diameter tend to be rejected. With regard to size, larger fibers and
fiber bundles go to the underflow. The influence of length alone is inferior to the
influences of other fiber properties [24].
Compared to a nonfibrillated fiber, a fibrillated fiber exposes a larger specific
surface area which offers more resistance to the relative flow in the gravity field of
the cyclone. The larger the resulting drag force, the more likely the fibrillated fiber
reports to the overflow.
6.5 Centrifugal Cleaning Parameters 587
6.5.3.2 Contaminants
The nature of a contaminant decides the preferred technical solution for its
removal. The most important contaminant parameters for cleaning are the contaminant
density and the contaminant shape.
Contaminants with densities that deviate far from the apparent fiber density are
easier to remove. Irregularly shaped contaminants can pose a challenge to cleaning
due to their inherently higher drag forces. Large contaminants may plug the
underflow of the cyclone.
A categorization of contaminants and selective ways for their removal are discussed
in Section 6.7.
6.6
Separation Efficiency
A variety of parameters are being used to describe the separation efficiency of
screening and cleaning operations. While overall parameters are usually sufficient
for characterizing the separation of impurities, a more refined approach becomes
appropriate especially for the purposes of fractionation.
6.6.1
Screening and Cleaning Efficiency
The very basic definition of the separation efficiency E is
E _
amout of debris in reject
amount of debris in feed _22_
Traditionally, this equation is employed generally for screens and more or less
exclusively for cleaners. There are some limitations to Eq. (22), however. E turns
unity when all debris is rejected, irrespective of the reject ratio. Likewise, the
operation of merely splitting a feed flow by a plain pipe tee yields a separation
efficiency larger than zero. In total, E disregards the good fiber loss with debris in
the reject stream.
The efficiency of a screen is usually plotted against the reject ratio due to its
overwhelming influence on the efficiency. Nelson has introduced a screen performance
parameter, the screening quotient Q, which can be easily determined by
just two analyses [25]:
Q _ 1 _
cd_A
cd_R _23_
where cd,R = mass concentration of debris in oven-dry reject (kg kg–1); and
cd,A = mass concentration of debris in oven-dry accept (kg kg–1).
588 6 Pulp Screening, Cleaning, and Fractionation
The screening quotient becomes zero for the pipe tee, and unity for ideal separation.
When applied to measurements from a given screen, Q was found to vary
only insignificantly over the range of industrially practiced reject ratios. Under
consideration of the mass balance over the screen, the screening efficiency is
obtained by:
E _
Rm
1 _ Q _1 _ Rm_ _24_
where Rm is the mass reject ratio – that is, the oven-dry reject mass divided by the
oven-dry feed mass. Figure 6.23 shows the screening efficiencies calculated for
different values of Q over the mass reject ratio. Since the performance of a given
screen is characterized by a particular Q, the screen’s operating point will, in theory,
move along a curve of constant Q. Typical values of Q for shives are 0.9 and larger.
0%
25%
50%
75%
100%
0.0 0.2 0.4 0.6 0.8 1.0
Efficiency, E
Mass reject ratio, Rm
0.0
0.5
0.7
0.9
Q = 1.0
Fig. 6.23 Screening efficiency as a function of the mass reject
ratio and screening quotient Q.
Using their plug-flow model, Gooding and Kerekes [1] have derived the screening
efficiency by combining Eqs. (5) and (22):
E _ RPc
V _25_
where Rv and Pc are the volumetric reject ratio and passage ratio of the contaminants,
respectively. Figure 6.24 illustrates screening efficiencies calculated for different values
of Pc over the volumetric reject ratio. Again, the performance of a given screen is
characterized by a particular Pc, and the screen’s operating point will move, in theory,
along a curve of constant Pc. Typical values of Pc for shives are 0.1 and smaller.
6.6 Separation Efficiency 589
0%
25%
50%
75%
100%
0.0 0.2 0.4 0.6 0.8 1.0
Efficiency, E
Volumetric reject ratio, Rv
1.0
0.5
0.3
0.1
Pc = 0.0
Fig. 6.24 Screening efficiency as a function of the volumetric reject
ratio and debris passage ratio Pc.
When comparing Fig. 6.23 with Fig. 6.24, the constant-Q curves expose a steeper
inclination at low reject ratios than the constant-Pc curves. This hold true even
after correction between mass reject ratio and volumetric reject ratio. The superiority
of the plug-flow model over the mixed flow model suggests that Eq. (25) is
more appropriate to describe a screen’s performance than Eq. (24) [10].
It must be remembered that all efficiencies calculated from Eqs. (22), (24) and
(25) above are actually contaminant-removal efficiencies. Each of these becomes
100% when the reject ratio is unity – a case which is of no industrial relevance.
Clearly, the economy demands that the amount of good fibers lost with the reject
from a separator is kept at a minimum. Therefore, any contaminant removal efficiency
calculated as per these equations must always be evaluated in conjunction
with the loss of good fibers.
6.6.2
Fractionation Efficiency
6.6.2.1 Removal Efficiency
In the basic case, the screening yield can be adopted for the purposes of fractionation.
The fiber removal function e(l) is defined as the mass of fibers with length in
the interval [l, l+dl] in the reject stream divided by the mass of fibers with the
same length in the feed [26]:
e_l_ _
QR cR_l_
QF cF_l_ _26_
590 6 Pulp Screening, Cleaning, and Fractionation
where cR(l) and cF(l) are the concentrations of the fibers with length in the interval
[l, l+dl] in the reject and feed streams, respectively. Assuming a plug-flow model
and constant passage ratio, this expression can be rewritten using Eq. (5):
e_l_ _ RP_l_ V _27_
While the fractionation objective above is determined by fiber length, other
pulp parameters, such as wall thickness, freeness or coarseness, may be assessed
similarly. In a more general form, the yield of a fiber fraction can be defined for
either the accept or the reject stream, with the selection depending on which
stream is of interest [11]. Then, the fractionation yield, Y, for any property of interest
is defined by:
Y _
QStream of interest cStream of interest_Property of interest_
QF cF_Property of interst_ _28_
6.6.2.2 Fractionation Index
In most fractionation applications it is important to remove as high a portion of
the one fraction while removing as little a portion as possible from the other fraction.
Therefore, the quality of the fractionation is characterized by the removal
functions of both fractions.
This can be quantified by the introduction of a fiber fractionation index, U. In
case of length-based fractionation, U is defined as the average e(l) for long fibers,
EL, minus the average e(l) for short fibers, ES [9]:
U _ EL _ ES _29_
where EL is basically the long fiber removal and ES is the short fiber loss. Unlike
removal efficiency, the fractionation index is penalized by removal of the fraction
which ought to be accepted, in the above case by the fraction of short fibers. U = 1
applies when the reject stream is composed only of long fibers and the accept
stream is composed only of short fibers – that is, perfect separation. In addition,
U = 0 means that the fiber length distribution remains unchanged – that is, no
separation.
The fractionation index increases as the hole size is reduced below the targeted
marginal fiber length, but deteriorates again as the hole size becomes smaller
than about half the marginal fiber length [9]. At similar reject thickening, the fractionation
index is almost twice as high for holed screen plates as for slotted ones [8].
The plug-flow model delivers a fractionation parameter a which is defined in
terms of the passage ratios of long fibers, PL, and short fibers, PS [10]:
a _ 1 _
PL
PS _30_
6.6 Separation Efficiency 591
Since the passage ratios are independent of the reject ratio, a reflects the performance
of a specific screen and can be used to anticipate the effect of changes in
reject ratio. Applying Eq. (25) to long and short fibers and eliminating Rv yields
EL _ E1_a
S _31_
Both the fractionation index and fractionation parameter are plotted within the
field of long fiber removal versus short fiber loss in Fig. 6.25. The solid lines calculated
for different values of a represent the curves on which a screen’s operating
point will move. For a given screen, the optimum point for fractionation lies
where the constant-a curve is tangent to a line of constant fractionation index.
Typical values of a are in the range of 0.4 to 0.7 [10].
0%
20%
40%
60%
80%
100%
0% 20% 40% 60% 80% 100%
Long fiber removal, EL
Short fiber loss, ES
0.0
0.4
0.7
Φ = 0.8 0.6 0.4 0.2 0.0
α = 0.9
Fig. 6.25 Screen operating curves (solid lines of constant a)
and fractionation index (dashed lines of constant U) plotted
in a field of long fiber removal versus short fiber loss [10].
6.7
Screening and Cleaning Applications
6.7.1
Selective Contaminant Removal
The selective removal of solid pulp impurities is by far the predominant application
of screening and cleaning in the production of chemical pulp. An overview
over the most common contaminants and their removal is provided below.
592 6 Pulp Screening, Cleaning, and Fractionation
6.7.1.1 Knots
Typically, knots represent the largest fraction of impurities in the pulp coming from
the digester. Knots originate from the dense sections of branches and heartwood, as
well as from oversized chips which have not been cooked down to their center. Knots
are rather large in size and of dark color. They can cause the failure of downstream
equipment in the pulp mill if they are not efficiently removed from the pulp.
Thus, knot removal (knotting) is normally carried out before washing. Knot separation
from the main stream of pulp is performed in a pressure screen. The separated
knots are then subjected to removal of good fibers in a secondary, atmospheric
screen. Both operations are governed by a barrier screening mechanism.
6.7.1.2 Shives
Shives are smaller impurities consisting of fiber bundles from incompletely
cooked wood. Their removal during screening is more difficult than that of knots.
Shives cause operational problems on the paper machine. In contrast to knots,
shives are mostly bleachable, but they consume higher amounts of bleaching chemicals
and may still remain of darker color than the bulk of the pulp after bleaching.
Shives should be removed before bleaching. Shive separation is carried out in a
system consisting of a number of pressure screens. Whether shive removal follows
barrier or probability screening depends on the aperture size of the screens.
As the use of very narrow slotted screens becomes common, shives tend to be
removed increasingly by the barrier principle.
6.7.1.3 Bark
Bark originating from incomplete debarking of the wood represents one of the
most challenging impurities. Bark is of dark color, has a similar density as wood,
and disintegrates easily.
There is normally no dedicated process for the removal of bark from pulp, but
the primary removal of bark should take place in the woodyard before chipping.
The remaining bark is removed from the pulp, together with other contaminants
during the course of screening and cleaning.
6.7.1.4 Sand and Stones
Sand and stones mainly come along with the wood chips, but may originate also
from tiling or concrete tanks. Rocky material can cause equipment failure and is
responsible for the wearing of equipment. The removal of stones and sand is
therefore best carried out as soon as possible in the fiberline.
Larger stones can be separated from the pulp by screening. Cleaning takes care
of any type of rocky material including sand. When narrow slotted screens are
used in a screening application, sand is rejected on a barrier principle and carried
through the subsequent screening stages. Special precautions must be taken in
such a case to minimize wear in the system caused by sand accumulation.
6.7 Screening and Cleaning Applications 593
6.7.1.5 Metals and Plastics
Metals and plastic can enter the fiberline with the wood, they may break away
from equipment, or they may enter the process accidentally. Like stones, metals
can cause the breakdown of equipment and must be removed to protect sensitive
machinery. Plastic contaminants adversely affect the quality of the final product
by causing operational problems in paper-making.
Because of the large density difference, metals can be easily separated from
pulp by centrifugal cleaning. Plastics are generally more difficult to remove, but
as certain types of plastic are less dense than pulp they can sometimes be separated
by reverse centrifugal cleaning.
6.7.2
Fractionation
Fiber fractionation generally follows the probability mechanism of separation.
Despite the limitations placed on fractionation efficiency by currently available
screening and cleaning equipment, fractionation applications are gaining increasing
attention and the prospects of value-added, tailor-made fibers have stimulated
the imagination of product developers.
With regard to paper-making properties, pulps containing long and thick-walled
fibers generally produce a higher tear index. Pulps with thin-walled fibers, and
those containing fines, have better optical properties, higher tensile strength,
internal bond strength, elongation and density [23,27].
The different fractions can be separately refined or treated otherwise, and may
then be recombined, or not. A market pulp producer with two dewatering
machines may fractionate his pulp to increase the long fiber content of the furnish
sent to one machine in order to produce a high-value reinforced pulp. A
paper producer with a multi-layer headbox may direct the shorter fibers to the surface
layers to improve sheet smoothness and optical properties, while placing the
longer fibers in the core to provide strength [10]. Besides strength, fiber fractionation
can also substantially improve the porosity of a pulp by removing the short
fibers and fines that reduce porosity [28].
In total, the fractionation of pulp creates a multitude of new opportunities for
the alternative utilization of the fiber raw material. Nevertheless, fractionation is
practicable only in mills which can make use of all the obtained fractions.
6.8
Systems for Contaminant Removal and Fractionation
6.8.1
Basic System Design Principles
We have seen above that there are various purposes for operating a screening or
cleaning system. While fractionation is of increasing interest, most applications
594 6 Pulp Screening, Cleaning, and Fractionation
still target the removal of large, heavy-weight, or light-weight contaminants. There
is a fundamental difference between contaminant removal and fractionation with
respect to the amount of material to be separated. After fractionation, the smaller
pulp fraction is seldom less than 20% of the pulp in the feed stream. In contrast,
the contaminants to be removed during screening or cleaning are typically no
more than 3% of the feed stream pulp.
Both contaminant removal and fractionation are subject to the condition that
the rejected portion contains only a minimum of the acceptable portion. Modern
screening and cleaning equipment removes unwanted matter quite efficiently
from the feed stream and produces an accept stream of high purity. In order to
achieve this, the reject stream must contain a relatively large amount of acceptable
material in addition to the matter to be rejected.
In a contaminant-removal system, economic reasons call for the minimization
of good fibers lost with the removed contaminants. Such systems usually consist
of a number of separators which can be operated in different arrangements. On
the one hand, contaminant removal is usually most efficient in a cascade feedback
arrangement. On the other hand, generally accepted rules for designing fractionation
systems are yet to be developed. In fact, it is uncertain if such rules will ever
exist, as fractionation tasks are custom-designed for a particular application.
In many cases, the design of separation systems is based on experience and
rules of thumb, because the interrelation of equipment, operating and pulp furnish
parameters is not yet fully understood. The resulting systems are often safe
to operate, but do not necessarily represent the best process solution and economy.
Screening and cleaning systems tend generally to be complex because of the
large number of design and operating parameters. Their function is challenged by
the circumstance that the optimum performance of the system is typically
achieved with equipment working near its point of failure (i.e., plugging). As
mechanistic models are further developed, the basic understanding of effects on
screen capacity, reject thickening and screening efficiency will improve. Computer
simulation provides valuable support in this respect [19,29].
In the following sections we will examine some common systems for contaminant
removal, as well as a few potential fractionation systems. However, before
doing this it may be appropriate to highlight some general aspects regarding the
design of separation systems.
Slotted screen baskets are quite susceptible to damage by junk material such as
metal bolts or rocks. A damaged screen basket leads to inferior screening efficiency
and requires costly replacement. Therefore, it has proven advantageous to
protect slotted screens from junk by the installation of an upstream perforated
screen. A protective screen is also highly recommended for cleaning systems to
avoid damage or blocking of hydrocyclone cones. When the amount of junk material
is low, protective screens can be operated with intermittent reject discharge.
As a result of reject thickening, industrial separation techniques involve dilution
at various points, both in the form of internal dilution to the equipment and
in the form of dilution between stages. The objective of dilution is first, to avoid
6.8 Systems for Contaminant Removal and Fractionation 595
plugging at the reject outlet and second, to adjust the feed consistency between
stages. It should be noted that most of the illustrations in this chapter lack such
dilution streams in order to avoid unnecessary complexity.
6.8.2
Systems for Contaminant Removal
6.8.2.1 Arrangement
As mentioned above, the contaminant level in chemical pulp is far below the
mass reject ratio of industrial separation equipment. Consequently, a large portion
of acceptable fibers can be found in the reject of a single separator, together
with the contaminants. Economical constraints of pulping, however, require that
undesirable contaminants taken from the screening system carry along as few
good fibers as possible.
Hence, it is common to use a combination of separators, where, for instance, a
second screen is used to reduce the amount of good fibers in the reject of the first
screen, and a third screen to remove good fibers in the reject of the second one.
Such a simple cascade arrangement is shown in Fig. 6.26.
A
A
R
R
R
A
F
Fig. 6.26 Three-stage screening in cascade feedback arrangement.
F = Feed;A = Accept;R = Reject.
In a cascade system, the reject from one screen passes on to the feed of the
screen in the next stage. In a cascade feedback arrangement, only the accept of the
first stage proceeds to the downstream step in the pulp production process, while
the accepts of the other stages are in each case brought back to the feed of the
preceding stage (Fig. 6.26).
It should be noted that, in a cascade feedback screening system, sand accumulation
can lead to substantial wear and to the need for frequent exchange of screen
baskets. As screen slots become narrower, an increasing portion of the sand com-
596 6 Pulp Screening, Cleaning, and Fractionation
ing with the feed to the first screening stage is rejected. If the following stages
have screens of similar aperture size, the repeated rejection of sand effects a relative
increase in the sand concentration in the reject of each stage. If one of the
following stages has a screen of larger aperture size, sand may be accepted by this
screen and flow back to the preceding stage, where it is rejected again. Both of
these phenomena are inherent to screening systems operating with narrow slots.
Depending on the sand contamination of the pulp furnish, the installation of special
sand cleaners in between stages may be required to reduce the accumulation
of sand in the system.
Similar to pressure screening systems, hydrocyclones are normally arranged in
feedback cascades (Fig. 6.27). At four to five stages, cleaning systems often have
more stages than screening systems with two to four. This is stimulated by a lower
quantity of contaminants in the feed of cleaning systems and more difficult separation
tasks.
J
R
D
D
D
A
F
Fig. 6.27 Four-stage cleaning in cascade feedback arrangement
preceded by protecting pressure screen. F = System feed;J = Junk;
A = System accept;R = System reject;D = Dilution.
In a cascade feed-forward scenario, accepts from other stages are mixed with the
primary accept. Figure 6.28 illustrates a simple two-stage feed-forward system,
which is common for the barrier screening application of knot removal. The secondary
screen of the knot removal system is usually a piece of equipment which
combines several unit operations, including screening, washing and dewatering.
6.8 Systems for Contaminant Removal and Fractionation 597
A
A
R
R
F
Fig. 6.28 Two-stage screening in cascade feed forward
arrangement. F = Feed;A = Accept;R = Reject.
In a cascade feed-forward system for shive removal, the reject from the secondary
screen could be treated in a refiner, after which the accepts of the tertiary
screen could be combined with the accepts of the primary screen, while the rejects
of the tertiary screen go back to the refiner. However, the quality requirements of
chemical pulps do not, in most cases, allow feed-forward operation of shive
screening and, in some cases, not even reject refining.
6.8.2.2 Fiber Loss versus Efficiency
For an exemplary shive screening application where an incoming pulp contains
1% of shives, Fig. 6.29 shows the mass balance for pulp over a single screen,
assuming a 20% mass reject rate and a 90% shive removal efficiency. The reject
stream contains a huge amount of good fibers (in fact 95% of the rejected pulp)
and almost one-fifth of the good fibers from the feed pulp are lost to the reject.
Total pulp
Shives
500 t/d
5.0 t/d
Total pulp
Shives
400 t/d
0.5 t/d
Total pulp
Shives
100 t/d
4.5 t/d
Feed Accept
Reject
Fig. 6.29 Mass balance for single screen;20% mass reject
rate, 90% shive removal efficiency.
Keeping the same assumptions (i.e., 20% mass reject rate and 90% shive
removal efficiency in the primary screen), we can consider a three-stage screening
system operated in cascade feedback mode (Fig. 6.30). Due to the repeated screening
action, the amount of good fibers in the system reject is reduced to 1% of the
feed pulp. In general, the good fiber loss can be reduced by adding another
screening stage or by decreasing the reject ratio. However, the flow regime in the
pressure screen places a physical limit on both the reject ratio and the number of
stages in a multi-stage screening system. That is why there is a minimum loss of
598 6 Pulp Screening, Cleaning, and Fractionation
good fibers with the system reject from the last stage of a pressure screening cascade.
When the economic feasibility of equipment and operating costs versus the
loss of good fibers is taken into consideration, the number of stages in a screening
system for shive removal is typically three or four. As an indication, the related loss of
good fibers in everyday operation seldomfalls below the amount of rejected shives.
Total pulp
Shives
500 t/d
5.0 t/d
System feed
Total pulp
Shives
491 t/d
0.55 t/d
Primary accept
Total pulp
Shives
613 t/d
5.55 t/d
Primary feed
Total pulp
Shives
151 t/d
5.49 t/d
Secondary feed
Total pulp
Shives
123 t/d
5.0 t/d
Primary reject
Total pulp
Shives
28 t/d
0.49 t/d
Tertiary accept
Total pulp
Shives
38 t/d
4.94 t/d
Secondary reject
Total pulp
Shives
113 t/d
0.55 t/d
Secondary accept
Total pulp
Shives
9.4 t/d
4.45 t/d
Tertiary reject
PRIMARY SCREEN
SECONDARY
SCREEN
TERTIARY
SCREEN
Fig. 6.30 Mass balance for three-stage feedback cascade;20% primary
mass reject rate, 25% secondary and tertiary mass reject rates, 90% shive
removal efficiency in each screen.
When comparing the single-stage and three-stage screening balances depicted
in Figs. 6.29 and 6.30, another observation relates to the screening efficiency. Due
to the internal circulation within the three-stage system, the accepted pulp contains
10% more shives than the accept from the single-screen case. It should be
noted that multi-stage screening helps to minimize the loss of good fibers but at
the same time reduces the screening efficiency.
6.8.3
Systems for Fractionation
The wide range of tasks achievable by fractionation has been repeatedly indicated
above. Because of the specialty of each case, general design principles for the fractionation
of chemical pulps are not yet established. Thus, the information in this
subsection is restricted to some general comments.
6.8 Systems for Contaminant Removal and Fractionation 599
The flow rates of the different fractions are defined by the particular application,
and a low reject ratio is not necessarily part of the fractionation requirements. It
may be advantageous to perform fractionation in a multi-stage system. In contrast
to contaminant removal, the efficiency of fractionation can be improved by multistage
systems. Two simple fractionation systems using screens are illustrated in
Fig. 6.31.
A
A
R
R
F A
R R
F A
(a) (b)
Fig. 6.31 Two-stage fractionation systems with feedback (a) cascade and (b) series.
Remember that holed screens fractionate better than slotted ones. While feedback
is clearly important for obtaining a higher fractionation efficiency, it has
been shown that both cascade and series arrangements may yield similar fractionation
results at a given mass reject rate [19]. According to the example shown in
Fig. 6.32, the best achievable fractionation occurs between about 30% and 60%
mass reject ratio. Note that the location of the fractionation index peak shifts
along the mass reject ratio axis dependent on of the relative amount of fractions
of interest in the feed pulp.
0.0
0.2
0.4
0.6
0.8
0.0 0.2 0.4 0.6 0.8 1.0
Fractionation index, Φ
Mass reject ratio, Rm
Single screen
Two-stage
cascade
feedback
Two-stage
series
feedback
Fig. 6.32 Fractionation index as a function of the mass reject
ratio for single-stage and two-stage fractionation with holed
screens;length-ba sed fractionation, simulation results [19].
600 6 Pulp Screening, Cleaning, and Fractionation
6.9
Screening and Cleaning Equipment
There is an abundance of different types of commercial separation equipment,
most of which are available as several variants. Some examples of more recent
design are detailed in the following section.
6.9.1
Pressure Screens
In the Impco HI-Q Fine Screen, the pulp feed enters the unit tangentially in the
upper section. Heavy contaminants separate centrifugally into the junk trap.
Then, as the pulp suspension enters the screening zone, prerotation vanes
increase its tangential velocity to improve screening efficiency. The accept passes
through the screen apertures, which are kept clean by pulsation provoked by the
special bump elements attached to the closed rotor. The reject proceeds to the bottom
of the screen where it is diluted and discharged through the reject nozzle [30].
Fig. 6.33 The GL&V Impco HI-Q Fine Screen [30].
In Metso’s DeltaScreen, the pulp suspension is fed tangentially into the bottom
part of the unit, where the heavy debris is trapped and can be removed through
the junk nozzle. The pulp proceeds upwards through the rotor to the screening
zone. The screening process takes place as the pulp moves downwards between
the foil rotor and the fixed screen basket. As with some other pressure screens,
the DeltaScreen can be equipped with a cyclone-type separator on top of the
machine, which offers the possibility for removal of light-weight contaminants
[31].
6.9 Screening and Cleaning Equipment 601
Fig. 6.34 The Metso DeltaScreen [31].
In the Impco HI-Q Knotter, the pulp feed enters the unit tangentially in the
upper section. Heavy contaminants are separated centrifugally into the junk trap.
The accept passes from the outside of the screen through the screen apertures to
the inside and proceeds to the accept nozzle at the bottom of the screen. Hydrofoils
rotating at the accept side provide the pressure pulses which keep the screen
apertures open. At the same time, the rotor action does not break up the knots.
On the reject side outside the screen basket, the knots are diluted and washed as
they proceed downwards to the reject nozzle [32].
Fig. 6.35 The GL&V Impco HI-Q knotter [32].
602 6 Pulp Screening, Cleaning, and Fractionation
Metso’s DeltaCombi is a combined knotter and fine screen that operates in similar
fashion to the DeltaScreen. In the lower section, the screen is equipped with
an additional rotating screen basket with holes for knotting. The feed pulp must
first pass this coarse screen basket from the outside to the inside. The pulse-generating
stationary foils of the knotting section are located on the accept side of the
screen basket. The coarse reject is taken out from the bottom part. The accept
which has passed the coarse screen basket is led upwards through the rotor of the
fine screen, and then enters the fine screening section between the foil rotor and
stationary fine screen basket [33].
Fig. 6.36 The Metso DeltaCombi [33].
The Noss Radiscreen features a different design, with the pulp feed entering
the unit in an axial direction. Accepted fibers pass through the two conical screens
plates fixed in the housing, while the reject proceeds to the reject nozzle at the
housing perimeter. As the rotor vanes pass along the screen plates, their peripheral
velocity increases by the radius towards the reject end of the screen plates, and
this leads to increased turbulence in the critical zone of higher consistency.
Radiscreens are available with perforated screens for both knotting and screening
applications. Their design does not require internal dilution, and features a comparatively
small pressure drop and low power consumption [34].
The abundance of screen designs makes it impossible to present all variations
offered by screen suppliers in this book. Tailor-made solutions are available for
special applications, with recent developments including, for example, screen baskets
with intermediate dilution [35] or intermediate deflocculation [36] half-way
down the screen basket to reduce the effects of reject thickening, or different surface
profiling along the length of the screen basket providing increased turbulence
towards the reject end of the screening zone [37].
6.9 Screening and Cleaning Equipment 603
Fig. 6.37 The Noss Radiscreen [34].
6.9.2
Atmospheric Screens
6.9.2.1 Secondary Knot Screens
Secondary knot screening is a barrier screening application targeted at the recovery
of good fibers from the knot stream coming from the primary knotter. Modern
secondary knot screens are equipped with a screw rotating inside a vertical or
inclined perforated screen cylinder. The pulp feed enters the screen near the bottom.
As the knots are transported upwards by the screw, accepted fibers pass
Fig. 6.38 The Noss Raditrim [38].
604 6 Pulp Screening, Cleaning, and Fractionation
through the screen apertures to the annular accept chamber. Shower liquid is
added to the knots, and assists separation by washing good fibers from the knots
to the accept side. A certain liquor level is maintained inside the screen, and after
the knots emerge from the liquor they dewater by gravity before being discharged
through the reject nozzle. The enclosed design of such secondary knot screens
avoids emissions to atmosphere.
The Noss Raditrim is an example of a vertical atmospheric knot screen
(Fig. 6.38). In line with other manufacturers’ screen designs, shower liquid is
introduced into the lower end of the screw shaft and becomes distributed through
holes in the shaft.
Secondary knot screens are typically fed with an inlet consistency between 1.0%
and 1.5%, and deliver knots at a consistency of 25–30%. The amount of good
fibers carried along with the knots is in the range of 10% of the total reject.
6.9.2.2 Vibratory Screens
The number of vibratory screens in use in the pulp industry is continuously
diminishing. This may be due to the fact that vibratory screening is connected to
a number of drawbacks, such as the unsuitability for fully automated control, the
rather dilute accept consistency, and the mostly uncovered design impairing vent
collection. However, if operated in the last stage of screening, the vibratory screen
has the advantage of delivering a reject stream which contains only a minor
amount of acceptable fibers.
6.9.3
Hydrocyclones
Since efficient centrifugal cleaning requires low pulp consistency and small-sized
hydrocyclones, a large number of units is required to deal with the considerable
flow rates. This has formerly resulted in long rows of cleaners with atmospheric
reject discharge. Today, the established arrangement of large numbers of pressurized
hydrocyclones is in canisters.
Figure 6.39 shows a Noss Radiclone, where the hydrocyclones are installed
radially in a pressurized cylindrical canister with vertical axis. Depending on the
cleaning capacity, one canister can hold several hundred cyclones. The feed enters
the canister centrically from the bottom and the pulp flows to the individual
hydrocyclones, where the separation takes place. The rejects and accepts from the
individual cyclones are then collected in separate compartments and leave the canister
through nozzles at the bottom. Typical cyclone diameters range from 80 to
125 mm. The pressure drop from the feed side to the accept side is between 1 and
2 bar [39]. In order to reduce the fiber loss from the last stage of cleaning, hydrocyclones
can be equipped with apex dilution.
The above-mentioned type of hydrocyclone used for the separation of heavyweight
particles is also called a forward cyclone. Cleaners for the separation of
light-weight contaminants are often termed reverse cyclones, accounting for the
6.9 Screening and Cleaning Equipment 605
inversion of accept and reject positions. Reverse cleaners can also be arranged in
canisters. The flow pattern in such canisters is similar to that illustrated in
Fig. 6.39, but the dimensions of the flow channels are adapted to the comparatively
larger apex flow rate and smaller base flow rate. In contrast to forward
cyclones, reverse cleaners do not thicken the reject, but lift the accept consistency
considerably above the feed consistency. Typical thickening factors are between
1.5 and 3.0 [40].
Larger-diameter, individual cyclones are sometimes employed for the separation
of heavy-weight contaminants to protect screen baskets or refiners from detrimental
feed components. These cyclones are typically 200–500 mm in diameter, and
may extend some meters in an axial direction. An example of a larger-diameter
cyclone separator, the Metso HC cleaner, is shown in Fig. 6.40. The cleaner can be
operated either on a continuous basis or with intermittent reject discharge as a
junk trap. HC cleaners are designed to work with feed consistencies up to 5%, but
are normally operated in the 1.5–2.5% range [41].
606 6 Pulp Screening, Cleaning, and Fractionation
Fig. 6.39 The Noss Radiclone AM [39]. Fig. 6.40 The Metso HC cleaner [41].
References 607
References
1 Gooding, R.W., R.J. Kerekes, Derivation
of performance equations for solid-solid
screens. Can. J. Chem. Eng., 1989;
67(10): 801–805.
2 Gooding, R.W., R.J. Kerekes, The
motion of fibers near a screen slot.
J. Pulp Paper Sci., 1989; 15(2): 59–62.
3 Yu, C.J., J. DeFoe, Fundamental study of
screening hydraulics. Part 1: Flow patterns
at the feed-side surface of screen
baskets; mechanism of fiber-mat formation
and remixing. Tappi J., 1994; 77(8):
219–226.
4 Walmsley, M., Z. Weeds. Concentration
and flow variations in a wood pulp fiber
separator. In APCChE Congress and
CHEMECA. Melbourne, 2002.
5 Yu, C.J., B.R. Crossley, L. Silveri, Fundamental
study of screening hydraulics.
Part 3: Model for calculating effective
open area. Tappi J., 1994; 77(9): 125–131.
6 Gooding, R.W., R.J. Kerekes, Consistency
changes caused by pulp screening.
Tappi J., 1992; 75(11): 109–118.
7 Olson, J., G. Wherrett, A Model of Fiber
Fractionation by slotted screen apertures.
J. Pulp Paper Sci., 1998; 24(12):
398–403.
8 Olson, J.A., Fiber length fractionation
caused by pulp screening, slotted screen
plates. J. Pulp Paper Sci., 2001; 27(8):
255–261.
9 Olson, J., B. Allison, N. Roberts, Fiber
length fractionation caused by pulp
screening. smooth-hole screen plates.
J. Pulp Paper Sci., 2000; 26(1): 12–16.
10 Gooding, R., J. Olson, N. Roberts.
Parameters for assessing fiber fractionation
and their application to screen
rotor effects. International Mechanical
Pulping Conference. Helsinki: TAPPI,
2001.
11 Ammala, A., Fractionation of thermomechanical
pulp in pressure screening. University
of Oulu, 2001.
12 McCarthy, C., Various factors affect
pressure screen operation and capacity.
Pulp Paper, 1988; 62(9): 233–237.
13 Pinon, V., R.W. Gooding, J.A. Olson,
Measurements of pressure pulses from
a solid core screen rotor. Tappi J., 2003;
2(10): 9–12.
14 Feng, M., J. Gonzalez, J.A. Olson, C.
Ollivier-Gooch, R.W. Gooding, Numerical
simulation and experimental measurements
of pressure pulses produced
by a pulp screen foil rotor. J. Fluids Eng.,
2005; 127: 347–357.
15 Amand, F.J.S., B. Perrin. Fundamentals
of screening: effect of rotor design and
fiber properties. Tappi Pulping Conference.
Orlando, FL, USA, 1999.
16 Rautjarvi, H., A. Ammala, J. Niinimaki,
Effect of entrained air on the performance
of a pressure screen. Tappi J.,
2000; 83(9).
17 Gooding, R.W., D.F. Craig, The effect of
slot spacing on pulp screen capacity.
Tappi J., 1992; 75(2): 71–75.
18 Kumar, A., R.W. Gooding, R.J. Kerekes,
Factors controlling the passage of fibers
through slots. Tappi J., 1998; 81(5):
247–254.
19 Friesen, T., et al., Pressure screen
system simulation for optimal fractionation.
Pulp Paper Can., 2003; 104(4):
T94–T99.
20 Olson, J.A., S. Turcotte, R.W. Gooding,
determination of power requirements
for solid core pulp screen rotors. Nordic
Pulp Paper Res. J., 2004; 19(2): 213–217.
21 Slack, M. Application Challenge Cyclonic
Separator. 2nd QNET-CFD Workshop.
Lucerne, Switzerland, 2002.
22 Li, M., et al., Characterization of hydrocyclone-
separated eucalypt fiber fractions.
J. Pulp Paper Sci., 1999; 25(8):
299–304.
23 Vomhoff, H., K.-J. Grundstrom, Fractionation
of a bleached softwood pulp and
separate refining of the earlywood- and
latewood-enriched fractions. Annual
General Meeting. Baden-Baden, Germany:
ZELLCHEMING, 2002.
24 Statie, E., et al., A computational study
of particle separation in hydrocyclones.
J. Pulp Paper Sci., 2002; 28(3): 84–92.
25 Nelson, G.L., The screening quotient: a
better index for screening performance.
Tappi, 1981; 64(5): 133–134.
26 Olson, J.A., et al., Fiber length fractionation
caused by pulp screening. J. Pulp
Paper Sci., 1998; 24(12): 393–397.
608 6 Pulp Screening, Cleaning, and Fractionation
27 Panula-Ontto, S., Fractionation of
unbleached softwood kraft pulp with wedge
wire pressure screen and hydrocyclone.
Helsinki University of Technology,
2003.
28 Olson, J., et al., Fiber fractionation for
high porosity sack kraft paper. Tappi J.,
2001; 84(6).
29 Weckroth, R., et al. Enhanced pulp
screening using high-performance
screen components and process simulation.
APPW, Durban, South Africa:
TAPPSA, 2002.
30 IMPCO HI-Q Fine Screen (product leaflet).
GL&V Pulp Group: Nashua, USA,
2001.
31 DeltaScreen (product leaflet). Metso
Paper: Sundsvall, Sweden, 2001.
32 IMPCO HI-Q Knotter (product leaflet).
GL&V Pulp Group: Nashua, USA, 2001.
33 DeltaCombi (product leaflet). Metso
Paper: Sundsvall, Sweden, 2001.
34 Radiscreen-F Fine Screen (product leaflet).
Noss: Norrkoping, Sweden, 2002.
35 Fredriksson, B., Increased screening
efficiency with belt dilution. 90th PAPTAC
Annual Meeting. Montreal: PAPTAC,
2004
36 McMinn, T., A. Serres. Intermediate
deflocculation and dilution device (ID2):
a new technological decisive step in the
screening processes. APPW. Durban,
South Africa: TAPPSA, 2002.
37 AFT VariProfile (product leaflet). AFT:
Montreal, Canada, 2003.
38 Raditrim Secondary Knotter (product
leaflet). Noss: Norrkoping, Sweden,
1999.
39 Radiclone AM80 (product leaflet). Noss:
Norrkoping, Sweden, 2002.
40 Radiclone BM (product leaflet). Noss:
Norrkoping, Sweden, 2000.
41 HC Cleaners (product leaflet). Metso:
Valkeakoski, Finland, 2002.
609
1147
III
Recovered Paper and Recycled Fibers
Hans-Joachim Putz
Handbook of Pulp. Edited by Herbert Sixta
Copyright © 2006 WILEY-VCH Verlag GmbH &Co. KGaA, Weinheim
ISBN: 3-527-30999-3
©2006 WILEY-VCHVerlag GmbH&Co .
Handbook of Pulp
Edited by Herbert Sixta
1
Introduction
During the era before the introduction of industrialized paper production 200
years ago, the most common fiber furnish was secondary fibers recovered from
used textiles. These were rags based on hemp, linen, and cotton. Only after the
invention of mechanical woodpulp in 1843 and chemical woodpulp during the
second half of the nineteenth century was paper production no longer as reliant
on recycled material as in the previous 2000 years.
Before industrialized paper production and the invention of the paper machine
in 1799, stationery or writing paper made from rags was recycled to produce lowgrade
board. As early as 1774, Claproth in Gottingen, Germany, improved the processing
of used, hand-made writing papers. His process removed optically disturbing
inks or printing ink. Today, we call this method “deinking”.
With growing industrialization and gross national product, the global paper production
increased significantly from almost 44 million tons in 1950 to 339 million
tons in 2003. The data in Tab. 1.1 indicate that between 1960 and 2000, for a doubling
of the paper production worldwide, in the CEPI countries (all EU countries
plus Czech Republic, Hungary, Norway, Slovak Republic, and Switzerland) or in
Germany, an approximate period of 20 years was necessary, whereas between
1950 and 1960 only a 10-year period was required for the first doubling of paper
production. In all of these time periods no doubling appeared in the USA where,
Tab. 1.1 Development of paper production between 1950 and 2003, in million tons [1–6].
Country Year
1950 1960 1970 1980 1990 2000 2003
Germany 1.6 3.4 6.6 8.8 12.8 18.2 19.3
CEPI 10.5 20.5 36.7 40.7 63.1 90.8 95.2
USA 22.1 31.3 47.6 56.8 72.2 85.8 80.2
World 43.8 74.4 129.3 171.7 240.8 324.0 338.8
1149
Handbook of Pulp. Edited by Herbert Sixta
Copyright © 2006 WILEY-VCH Verlag GmbH &Co. KGaA, Weinheim