Monitoring carryover

at the brownstock washers

By B. Richardson
BROWNSTOCK WASHI NG
Pulp & Paper Canada T 144 106:6 (2005) 43
ROWNSTOCK washers are used to recov-
er cooking chemicals from the prod-
uct pulp. They constitute an important
operation in the kraft chemical-recov-
ery system as optimization of washer
performance affects both the financial and envi-
ronmental success of a pulp mill. Minimization of
carry-over to the bleach plant reduces both bleach-
ing chemical costs and organic material in the
bleach plant effluent. Brownstock washing effi-
ciency is usually measured by measuring soda or
saltcake loss or the COD carryover into the bleach
plant. The operating target for saltcake carryover
in most kraft mills is 5 to 15 kg of washable saltcake
per tonne of pulp from the last brownstock wash-
ing stage to the bleach plant, regardless of whether
or not the mill operates an oxygen delignification
reactor. The correlation between black liquor
solids (BLS) loss and saltcake losses in a conven-
tional mill is 1.5 to 1.75 kg BLS/kg of saltcake loss
from the brownstock washers [1]. This relation-
ship is reduced to 1.0 to 1.25 kg BLS/kg saltcake
following an oxygen stage [1].
Targets for washing efficiency are generally
based on saltcake losses rather than on carryover
of dissolved organics, even though the detrimen-
tal effects of organic carryover on viscosity, delig-
nification efficiency, chemical consumption and
organochlorine formation in the bleach plant are
well recognized [2]. This practice is historical. It
continues because simple, robust sensors are not
available to monitor the carryover of dissolved
organics [3]. A means to monitor such variables
as TOC or COD, carried in the filtrate leaving
with the pulp mat from a washer, or some other
parameter that would predict bleach plant chem-
ical consumption, is necessary for effective wash-
ing control in modern mills with generally good
washing efficiency. To address environmental
impacts, BOD5, COD, and colour are the impor-
tant parameters to consider or monitor. From a
bleaching chemical demand perspective, COD,
BLS and soda loss are most important. From a
chemical recovery perspective, BLS and soda loss
are most important. The last two variables can be
measured indirectly using the conductivity of the
filtrate carried with the washed pulp stock. Con-
ductivity can, therefore, be used to control
brownstock washer efficiency. Conductivity is a
simple, quick test to perform manually. It can also
be measured in real time with an in-line sensor,
and the collected data stored and analyzed in the
mills control system.
Several researchers have simulated the dynamic
behaviour of a brownstock-washing system in order
to develop automatic process control strategies that
minimize the carryover of dissolved organics and
inorganics to the bleach plant [4-7]. Implementa-
tion of a multi-variable control strategy for filtrate
tank level control on a series of brownstock washers
improved the cleanliness of the washed pulp and
reduced variations in carryover [8]. This strategy
was implemented using cascade feed-forward con-
trollers on shower flows to eliminate interactions
between filtrate tanks and new tuning rules for
shower and tank level controllers [8]. The new tun-
ing rules for the PI controllers were particularly
successful in eliminating process oscillations and
improving process stability.
For effective washing control, it is necessary to
analyse the filtrate carried in the washed pulp mat
in order to eliminate the long lag times associat-
ed with seal tanks. A simple filtrate sampler insert-
ed below the repulping screw has been used to
obtain filtrate samples [3], but it has also been
found to be inadequate when washed pulp dis-
charge consistencies are higher than 10-12%. We
have identified, therefore, a commercial sampler
capable of handling high-consistency pulp (up to
15%), while still extracting a representative fil-
trate sample from the pulp mat.
This report looks at the economic impacts that
brownstock washing has on the other areas of the mill
using a case study for one mill. The approach and
analyses used are based on a paper by Compton [9].
MILL DATA
FOR CASE STUDY
The procedure outlined in the paper by Comp-
ton [9] was incorporated into a spreadsheet pro-
gram that performed all the calculations and pro-
duced the necessary graphs to determine the
optimum dilution factor for the brownstock wash-
ers. Comptons method analyses the effects of
varying the shower water flowrate on the last
brownstock washer to determine the economic
effects on mill operations. The amount of shower
water applied on the last washer is used to calcu-
B. RICHARDSON,
Paprican,
Prince George, BC
brichardson@paprican.ca
B
Abstract: Brownstock-washing efficiency affects bleach plant chemical consumption as well as the
load to the chemical recovery area. We investigated the effects of the dilution factor on mill oper-
ating costs with the aid of spreadsheet calculations to identify the optimum dilution factor. A study
of one mills washing operation showed that increasing the dilution factor on the last brownstock
washer from 2.3 to 4.2 yields a saving of about $1 million/yr for a 700-tpd kraft pulp mill.
late the amount of steam required to raise the filtrate tempera-
ture and for evaporation in the multiple effect evaporators.
From a curve of washing efficiency versus dilution factor, one
can determine black liquor solids lost to bleach plant sewers.
The loss of black liquor solids to the bleach plant is then used
to calculate the amount of chemicals required for make-up and
bleaching, the additional BOD load on the effluent treatment
system and the fuel value of the lost organics for the production
of steam. Table I lists the data provided by a Canadian mill,
which allowed an economic analysis of the brownstock system. A
generic Cadsim model of the fibreline was used to determine
the current dilution factor1. Table II lists the assumed data used
to complete the economic analysis.
The case study mills brown fibreline consists of a chip bin,
Kamyr digester, atmospheric diffuser, knotters, two brownstock
drum washers, an open screen room and a brown decker. The
flow of wash liquor is counter-current to the pulp stream, start-
ing at the brown decker with a mixture of fresh hot water and
condensate. Filtrate from this decker is used on showers of the
last brownstock washer and so on to the first brownstock drum
washer, and then to the atmospheric diffuser. The computer
model determined that the case study mill is currently operating
the last brownstock washer at a dilution factor of about 2.3.
RESULTS, DISCUSSION
Effect of Increasing the Dilution Factor:
Costs: Changing the dilution factor not only affects brown-
stock washer efficiency, but also has a major impact on evapora-
tor loading. For a mill that has additional evaporator capacity,
the cost of increasing the dilution factor is equivalent to the cost
of additional steam required at the evaporators and a slight
increase in pumping horsepower [9]. Dilution factor is the
weight of wash liquor applied minus the weight of liquor leaving
with the pulp mat divided by the weight of pulp and is based on
the shower liquor flow rate and discharge mat consistency (kg
water/kg pulp) [10]. Figure 1 shows the evaporation steam costs
versus dilution factor for the case study mill. Two assumptions
were made in developing this curve. First, the digester blow con-
sistency was set equal to the pulp discharge consistency for the
last brownstock washer. We also assumed that the evaporator sys-
tem had extra capacity. If there is no extra capacity at the evap-
orators, then evaporation capacity can be increased by adding
external heaters, evaporator bodies, or reconfiguring the bodies
to operate with lower steam economies [9]. These modifications
all have capital and/or operating costs associated with them.
For a counter-current washing system, increasing the shower
flow rate on the last washer will increase the hydraulic flow of filtrate
throughout the washing line, thereby increasing the electric power
requirements for the filtrate pumps at each stage of washing.
Shower water temperature influences brownstock-washing
efficiency, which increases as the water/filtrate temperature
increases. However, the temperature must not be so high as to
cause the filtrate to flash in the washers drop leg. The maxi-
mum operating temperature is normally between 70 and 80C
for drum washers [9]. The temperature of the incoming pulp
and the shower on the last stage of washing, and the heat loss
from the washers and filtrate tanks, determine the operating
temperature of the washer line. One way of optimizing the wash-
ing operation is to control the shower water temperature. The
case study mill currently operates its showers at 55C. Figure 2
shows the cost associated with heating the shower water from 15
to 55C as a function of the dilution factor used in the brown-
stock washing line.
Increasing the shower water temperature will lower its vis-
cosity and enable greater diffusion into the sheet for better
washing [10]. At temperatures above 60C, hemicellulose
becomes soft, opening the fibre structure and improving the
flow of washing liquor through the pulp mat [12]. Low wash
water temperatures, and those above 77C, generally result in
poor washing [13]. Mill experience has shown that the optimum
wash water temperature is typically close to 70C [14]. The
increase in washing efficiency and savings in bleach plant chem-
icals that would result from an increase in shower water tem-
perature can, however, only be estimated by a mill trial, as there
are no known calculations to determine how washing efficiency
varies with increasing shower water temperature.
Benefits: On the positive side, increasing the dilution factor
decreases the amount of black liquor solids lost to the bleach
plant sewer, thereby reducing chemical make-up costs and
organic loading on the effluent-treatment system. Additional
savings result from decreased black liquor solids losses because
these solids are kept within the chemical recovery system and
now become fuel, which is used in the recovery boiler to gener-
ate steam. Pulp mills require a specific amount of steam to oper-
ate. The required amount of steam is generated by the recovery
and power boilers. Returning black liquor solids to the recovery
system decreases the demand for hog fuel by the power boiler.
Figure 3 shows the equivalent fuel savings for hog fuel that result
from returning more black liquor solids to the recovery system
as a function of the brownstock washer dilution factor.
BROWNSTOCK WASHI NG
44 106:6 (2005) T 145 Pulp & Paper Canada
Item Value Units
Chlorine dioxide use 0.6 kg ClO
2
/kg equiv. saltcake [9]
Caustic use 1.0 kg NaOH/kg equiv. saltcake [9]
Black liquor fuel value 12,780 kJ/kg of dry BLS
Hog fuel cost $2.11 $/kJ
Recovery boiler
thermal efficiency 70 %
Recovery & hog
boiler efficiency* 75 %
Impact of BL losses kg BOD / kg of black
on effluent treatment 0.175 liquor solids [11]
* Weighted average based on steam production
TABLE II. Assumed data.
Item Value Units
Multiple Effect Evaporators:
Number of effects 6
Steam economy 4.9 kg H
2
O removed/kg steam used
Black liquor solids 1,680 kg/odt pulp
Evaporator product solids 43 %
Steam cost $8.82 $/1,000 kg
Last Brownstock Washer:
Mat discharge consistency 11 %
Shower water temperature 55 C
Incoming water temp 15 C
White Liquor Characterization:
EA 81.6 g/L as Na
2
O
AA 96.0 g/L as Na
2
O
TTA 112.0 g/L as Na
2
O
Digester Operations:
Pulp yield 45.0 % by wt.
Active alkali charge 16.0 % on pulp
Other Parameters Required:
Equivalent saltcake makeup cost* $0.00
Chlorine dioxide cost %1.10 $/kg
Caustic cost $0.35 $/kg
Nutrient cost $8.60 $/1000 kg BOD
*Excess sodium sesquisulphate from the chlorine dioxide generator is avail-
able for make-up and the saltcake make-up cost is, therefore, negligible.
TABLE I. Data provided by the mill.
Decreasing the discharge of black
liquor solids decreases the load on the
effluent-treatment system. Saving in waste
treatment costs can, thus, be achieved
when there is a decrease in the concentra-
tion of black liquor solids discharged from
the mill to the effluent-treatment system,
Fig. 4. The calculated savings are based on
the predicted reduction in required nutri-
ent chemicals. A reduction in black liquor
losses may also affect the number of aera-
tors operating in the lagoon. As this is site-
specific, no allowance for a reduction in
aeration horsepower has been incorporat-
ed in this analysis.
The brownstock-washing efficiency has a
major impact on operating costs for a
bleached mill as it affects the consumption
of oxygen in the delignification system,
chlorine dioxide in the first bleaching stage
and caustic in the second stage [9, 15]. The
cost of the case study mills bleaching chem-
icals per tonne of pulp varies as a function
of the dilution factor, Fig. 5.
Over-all Cost / Benefit Analysis: Figure 6
is reproduced from the Compton paper
[9]. This graph can be developed for a
specific mill by obtaining a pulp sample
from the vat of the last brown washer. In
the lab, the pulp sample is washed with
various amounts of water (simulating vary-
ing dilution factor), and the amount of
remaining saltcake and black liquor solids
in the resulting washed pulp mat is mea-
sured. Alternately, the data for the graph
can be obtained from the mills washer by
varying the shower flow rate. The shape of
the curve in Fig. 6 greatly influences the
over-all economics and the final optimum
dilution factor. The calculated effects of
varying the dilution factor on chemical
makeup, hog fuel (or equivalent fuel val-
ue of black liquor), effluent treatment
and bleaching chemical costs are directly
affected by this figure.
The various costs and benefits that
result from varying the dilution factor (as
illustrated in Figs. 1 to 5) are summed and
plotted in Fig. 7, disregarding any impact
that improved brownstock washing might
have on the aeration requirements in the
effluent treatment system. Operating the
brownstock washers at the optimum value
given by Fig. 7 will minimize the operating
costs. Figure 7 shows that the optimum
operating point for the brownstock wash-
ers at this mill is with a dilution factor of
about 4. There is very little to gain with
higher dilution factors. The case mill is
currently operating at a dilution factor of
about 2.3. Compton [9] and Korhonen
[13] calculated similar optimum operat-
ing values, 4.5 and 4.0, respectively for
bleached kraft mills.
Figure 8 shows the incremental savings
per oven-dry tonne of pulp in the case
study mill as a function of the dilution fac-
tor. Maximum savings are achieved at a
dilution factor of 4.4, which produces
over-all savings of $5 per oven-dry tonne
of pulp when compared to current opera-
tions at a dilution factor of 2.3. Maintain-
ing a dilution factor above 4 would be sat-
isfactory, as there is very little to gain at
higher dilution factors. However, in order
to operate at the optimum dilution factor,
the current evaporator system must be
able to handle the extra evaporation load
and weak black liquor flow.
Economics of Brownstock Washing Con-
trol: Brownstock washing is operated to
supply the bleach plant with clean pulp
and to separate the cooking chemicals
from the pulp, while minimizing the
amount of wash water used. At most mills,
the last brownstock washer typically oper-
ates with a fixed shower flowrate, while the
incoming pulp mass flowrate, as well as the
amount of black liquor solids carried with
it, can vary. Simply controlling the shower
flow rate with pulp production provides a
significant improvement towards minimiz-
ing mill-operating costs. The measure-
ment of filtrate conductivity provides the
operator with an indication of the cleanli-
ness of the pulp mat, and allows for a bet-
BROWNSTOCK WASHI NG
Pulp & Paper Canada T 146 106:6 (2005) 45
FIG. 1. The cost of steam for the evaporation of black
liquor as a function of the dilution factor.
FIG. 2. The cost of heating BSW shower water from 15 to
55C as a function of the dilution factor.
FIG. 3. The fuel value of incremental recovered black liquor
solids as a function of the dilution factor.
FIG. 4. The effect of reduced black liquor losses on effluent
treatment costs.
ter or more elaborate control strategy for
the brownstock-washing system and addi-
tional benefits relative to filtrate tank level
control [8]. Installation of a simple filtrate
sampler and a flow controller would
enable a mill to optimize the shower flow
rate and minimize the carryover of black
liquor solids to the bleach plant.
A Canadian mill has successfully tested
such a system. Use of a filtrate sampler
could also help the mill avoid possible
over-use of shower water, thereby mini-
mizing the volumetric flow of weak black
liquor to the evaporators and maximizing
black liquor density. Trials at another
Canadian mill have also shown that the
conductivity of the filtrate carried with the
pulp mat can be used to minimize
defoamer use. Based on the $5/odt of
pulp savings estimated for the case study
mill, a filtrate sampler, conductivity sensor
and a control package should pay for itself
in less than one month.
Washing control might also be applied
beneficially to bleach plant washing stages
such as D0; it might be possible to moni-
tor the washing efficiency of this stage by
measuring the UV absorbance of the fil-
trate from the pulp mat and use that mea-
surement to minimize COD load to sub-
sequent stages [15, 16]. This technique
might also be extended to monitor the
purity of contaminated condensates used
for pulp washing and the COD and colour
loading resulting from the sewering of
extraction stage filtrates [16].
CONCLUSIONS
Brownstock-washing efficiency has signifi-
cant impacts on a kraft mills operating
costs, as it affects bleaching chemical
costs, the loading on multiple effect evap-
orators and effluent treatment system,
and chemical make-up costs. The effect of
the brownstock dilution factor on mill
operating costs was investigated using a
laboratory-developed washing curve, mill
operating data and a series of spreadsheet
calculations. Calculations based on the
case study mill data, indicate that increas-
ing the dilution factor on the last stage of
brownstock washing from 2.3 to 4.2 would
yield over-all savings of $5/odt or about
$1 million/year for a 700-tpd kraft pulp
mill. A complete brownstock washing con-
trol package would, therefore, have an
expected payback period of less than one
month. Operating a washing system at its
optimum dilution factor will minimize
operating costs, but it is only practical if
the resulting volume of black liquor can
be handled at the evaporators and pulp
cleanliness meets the mills specifications.
LITERATURE
1. SHACKFORD, L.D., OSWALD, J.M. Flexible brown-
stock washing eases implementing oxygen delignifica-
tion. Pulp and Paper 61(6): 136-141 (June 1987).
2. SHACKFORD, L.D. Unravelling the mystery of wash-
ing efficiency determination. Proc., Tappi Pulping
Conference, San Francisco, CA 77-89 (October 1997).
3. TURNER, P.A., LIVINGSTONE, L. A method to
monitor the cleanliness of washed pulp directly. Pulp
Paper Can. 96(6): 71-75 (June 1995).
4. NASE, T., SJOBERG, K.-E. Advanced control of a
drum washing plant. P& Pulp Paper Can. 90(9): 108-
111 (September 1989).
5. EDWARDS, L., PEYRON, M., MINTON, M. Models
for cross-flow pulp washing calculations. Pulp Paper
Can. 87(1): 67-70 (January 1986).
6. HAN, Y., EDWARDS, L. Optimization of filter wash-
er operation and control. Tappi J. 71(6): 101-4 (June
1988).
7. TURNER, P.A., ROCHE, A.A., MCDONALD, J.D.,
VAN HEININGEN, A.R.P. Dynamic behaviour of a
brown stock washing plant. Pulp Paper Can. 94(9): 37-
42 (September 1993).
8. TURNER, P.A., ALLISON, B.J., OEI, J.K. Brown
stock washer control, Part II: Filtrate tank level con-
trol. Pulp Paper Can. 94(11): 45-48 (November 1993).
9. COMPTON, R. Economic analysis of brown stock
washing systems. Proc., Tappi Pulping Conference,
San Francisco, CA 257-272 (October 1997).
10. LEDDY, J.E. Evaluating the performance of rotary
drum vacuum washers. Appita J. 51(6): 315-322
(November 1998).
11. WONG, A., WEARING, J.T., ULOTH, V.C. Toxici-
ty of brownstock filtrates is related to the degree of
washing. Pulp & Paper 53(10): 96-100 (September
1979).
BROWNSTOCK WASHI NG
46 106:6 (2005) T 147 Pulp & Paper Canada
FIG. 5. The cost of bleaching chemicals required as a result
of increased carryover of black liquor solids as a function
of dilution factor.
FIG. 6. Typical washing efficiency curve for the last brown
stock washer [9].
FIG. 7. The effect of dilution factor on total operating costs
for the case study bleached kraft pulp mill.
FIG. 8. Incremental saving per oven-dry tonne of pulp ver-
sus dilution factor.
12. GREN, U. Washing of Cellulose Fibres. Goteburg,
Sweden: Chalmers University of Technology (1973).
13. McSWEENEY, J.M. Control and Instrumentation.
Tappi Brown Stock Washing Short Course, City, 17-20
(October 1982).
14. KORHONEN, O. Brown stock washer operating
variables. Proc., CPPA, TS, Pacific Coast and Western
Branch Spring Conference, Harrison Hot Springs, 46-
50 (May 1979).
15. VIIRIMAA, M., DAHL, O., NIINIMKI, J. ALA-
KAILA, K., PERMKI, P. Identification of the wash
loss compounds affecting the ECF bleaching of soft-
wood kraft pulp. Appita J. 55(6): 484-8 (November
2002).
16. VAN HEEK, R., DICK, D. An International report.
A spectrophotometric method for the monitoring of
chemical oxygen demand in chlorine dioxide bleach-
ing stage filtrates. Paprican Report No. PPR 1399.
Montreal Paprican (December 1998).
BROWNSTOCK WASHI NG
Pulp & Paper Canada T 148 106:6 (2005) 47
Reference: RICHARDSON, B. Monitoring carryover at the brownstock washers. Pulp & Paper
Canada 106(6): T144-148 (June, 2005). Paper presented at the 90th Annual Meeting in Montre-
al, QC, on January 26 to 29, 2004. Not to be reproduced without permission of PAPTAC.
Manuscript received on October 16, 2003. Revised manuscript approved for publication by the
Review Panel on October 28, 2004.
Keywords: UNBLEACHED PULPS, KRAFT PULPS, ENTRAINMENT, WASHERS, MONI-
TORING, DILUTION, EFFICIENCY, OPTIMIZATION, ENERGY CONSERVATION, CHEMICAL
CONSUMPTION, BLEACH PLANTS, COST CONTROL, PROCESS CONTROL
Rsum: Lefficacit du lavage de la pte brune a une incidence sur la consommation de pro-
duits chimiques latelier de blanchiment ainsi que sur la charge la zone de rcupration des
produits chimiques. Nous avons valu les effets du facteur de dilution sur le cot dexploitation
de lusine laide dun tableur, afin de dterminer le facteur optimal pour la dilution. Ltude des
oprations de lavage dune usine a indiqu que laccroissement du facteur de dilution de 2,3 4,2
la dernire pile laveuse de pte brune permet dconomiser denviron 1 million $ pour une usine
produisant 700 tonnes de pte kraft par jour.