NC0000272_NCASI_Review_Color_Control_Tech_20060601

J U L 2 6 2006
S e
WATER QUALITY SECTION
ncasi ASHEVIL.LE.REGIONAL OFFICE
NATI,O %AL COUNCIL FOR AIR AND STREAM IMPROVEMENT
p z
INTERIM REPORT
REVIEW OF COLOR CONTROL TECHNOLOGIES
AND THEIR APPLICABILITY TO MODERN
KRAFT PULP MILL WASTEWATER
INTERIM REPORT
JUNE 2006
by
Eva Mannisto
EKONO, Inc.
Bellevue,Washington

Acknowledgments
This study was prepared by Eva H.Mannisto and Sean L.Smith of EKONO,Inc.under the direction of a five-mill
collaborative. Funding for the project was provided by the mills who also supported the publication of this report for
distribution as an NCASI Technical Bulletin. NCASI staff acted to facilitate the preparation of,this Interim Report
which will be superseded by its publication as NCASI Technical Bulletin No.919.
For more information about this document,contact:
Paul Wiegand
Vice President,Water Quality
NCASI
P.O.Box 13318
Research Triangle Park,NC 27709-3318
[phone]
[e-mail]
For information about NCASI publications,contact:
Publications Coordinator
NCASI
P.O.Box 13318
Research Triangle Park,NC 27709-3318
[phone]
[e-mail]
National Council for Air and Stream Improvement,Inc.(NCASI). 2006. Review ofcolor control technologies and
their applicability to modern krafl pulp mill wastewater. Interim Report. Research Triangle Park,N.C.:National
Council for Air and Stream Improvement,Inc.
©2006 by the National Council for Air and Stream Improvement,Inc.

TABLE OF CONTENTS
1.0 INTRODUCTION........................................................................................................................ I
1.0 Background......................................................................................................................... I
1.1 Measurement of Color........................................................................................................ 1
2.0 TASK I: COLOR REDUCTION TECHNOLOGIES...................................................................2
2.1 Wood Species—Impact On Color......................................................................................2
2.2 Kappa Number Reduction...................................................................................................3
2.3 Oxygen Deligniftcation.......................................................................................................4
3.0 BLACK LIQUOR LOSS CONTROL..........................................................................................5
3.1 Improved Brown Stock Washing........................................................................................6
3.2 Reduction of Other Regular Black Liquor Losses..............................................................8
4.0 BLEACHING............................................................................................................................. 11
4.1 Chlorine Dioxide Bleaching............................................................................................. 11
4.2 Hydrolysis Treatment for Removal of Hexenuronic Acids.............................................. 14
4.3 Ozone Use in Bleaching.................................................................................................... 16
4.4 Peroxide in Bleaching.......................................................................................................21
4.5 Peracid Bleaching.............................................................................................................23
5.0 COLOR MONITORING............................................................................................................25
6.0 RECOVERY OF COLORED WASTEWATERS......................................................................26
6.1 Black Liquor Spill Recovery............................................................................................26
6.2 Bleach Plant Effluent Recovery........................................................................................33
7.0 COLOR REMOVAL—SEPARATION PROCESSES..............................................................54
7.1 Membrane Technologies...................................................................................................54
7.2 Ion Exchange....................................................................................................................64
7.3 Activated Carbon and Activated Petroleum Coke Adsorption.........................................65
7.4 Electrodialysis(ED) and Electrodialysis Reversal (EDR)................................................68
8.0 COLOR REMOVAL—CHEMICAL PROCESSES..................................................................69
8.1 Lime Precipitation.............................................................................................................69
8.2 Alum Precipitation............................................................................................................70
8.3 Iron Precipitation..............................................................................................................70
National Council for Air and Stream Improvement

8.4 Polymer Precipitation........................................................................................................71
8.5 Nitric Acid Precipitation ...................................................................................................72
8.6 Electrochemical Treatment...:............................................................................................73
8.7 Summary of Chemical Treatment......................................................................................73
9.0 COLOR REMOVAL—OXIDATION PROCESSES............................................................. 75
9.1 Peroxide.............................................................................................................................75
9.2 Ozone ................................................................................................................................78
9.3 Wet Air Oxidation with Catalyst.......................................................................................83
10.0 COLOR REMOVAL—EVAPORATION AND INCINERATION....................................... 85
10.1 Evaporation and Incineration of Bleach Plant Effluent.....................................................85
11.0 COLOR REMOVAL—FUNGUS/BACTERIA/ENZYMES ................................................. 90
11.1 White-Rot Fungus Treatment............................................................................................90
12.0 TASK II: MILL-SPECIFIC REVIEW OF TECHNOLOGIES AND PERFORMANCE...... 93
12.1 Color in Study Mills..........................................................................................................93
12.2 Benchmarking...................................................................................................................97
12.3 Color Technologies in the Study Mills............................................................................101
13.0 SUMMARY AND CONCLUSIONS................................................................................... 106
REFERENCES .................................................................................................................................. 108

TABLES
Table 2.1 Kappa Targets of Participating Mills..................................................................................5
,Table 3.1 Correlation of Color to Other Parameters in Liquid Squeeze from
WashedUnbleached Pulp...................................................................................................6
Table 3.2 Correlation of Color to Other Parameters in Black Liquors...............................................8
Table 4.1 Impact of ECF Conversion on Effluent Color in the Study Mills .................................... 12
Table 4.2 Hexenuronic Acid Content of Various Pulps.................................................................... 15
Table 4.3 Comparison of Acid Hydrolysis Sequence with Other Bleach Sequences....................... 16
Table 4.4 Full-Scale Kraft Mill Ozone Bleaching Installations........................................................ 17
Table 4.5 Ozone Bleaching Laboratory Study Results on Softwood ZD and DZ Combinations..... 19
Table 4.6 Comparison of Effluent Data for Ozone and ECF Bleaching, kg/ADt.............................20
Table 4.7 Ozone Bleaching Laboratory Study Results on Hardwoods.............................................20
Table 4.8 Comparison of Bleach Sequences.....................................................................................21
Table 4.9 Examples of Full-Scale Peroxide Bleaching Installations................................................23
Table 4.10 Full-Scale Peracid Bleaching Installations .......................................................................24
Table 4.11 Effect of ePaa and dPaa-Stage on Bleaching Effluents...................................................25
Table 6.1 Example of Color in Acid and Alkaline Bleach Plant Effluents
in This Study(ADT=shT)................................................................................................34
Table 6.2 ESP Ash Treatment Processes for Chloride and Potassium Removal..............................43
Table 6.3 Information about the IP(Franklin, VA)Mill Closed Cycle Bleach Plant.......................48
Table 6.4 Pilot Plant Results of Eka Chemicals Ultrafiltration Concept..........................................52
Table 6.5 Simulated(GEMS)Impact of PC+on Bleach Plant.........................................................53
Table 7.1 Results of OF and NIT of Eop Filtrates..............................................................................58
Table 7.2 Information about Full-Scale OF Plants...........................................................................60
Table 7.3 Bleach Plant Effluent Characteristics for Weyerhaeuser's67
Grande Prairie,Alberta Mill.............................................................................................67
Table 8.1 Full-Scale Installations of Coagulation as Tertiary Treatment.........................................74
Table 9.1 TAML Catalyst Pilot Plant Results on Bleach Plant Alkaline Effluent............................77
Table 9.2 Bleach Plant Effluent Characteristics for Norske Skog's
ElkFalls, British Columbia Mill.......................................................................................79
Table 9.3 Whole Mill Effluent Characteristics for Tenneco Packaging's
Valdosta, Georgia Mill .....................................................................................................80
Table 10.1 Bleaching Effluent Evaporation Installations...................................................................89
National Council for Air and Stream Improvement

Table 12.1 Summary of Color Balance Data in Four Papergrade Bleached Kraft Mills.....................93
Table 12.2 Summary of Mill A Color Data(ADT=Air Dry Short Tons of Bleached Pulp)-
2005 Data..........................................................................................................................94
Table 12.3 Summary of Mill B Color Data(ADT=Air Dry Short Tons of Bleached Pulp)-
2005 Data..........................................................................................................................95
Table 12.4 Summary of Mill C Data(ADT=Air Dry Short Tons of Bleached Pulp).......................96
Table 12.5 Summary of Mill D Data(ADT=Air Dry Short Tons of Bleached Pulp).......................97
Table 12.6 Successful External Technologies...................................................................................103
Table 12.7 Unsuccessful External Technologies...............................................................................104
Table 13.1 List of Color Reduction Technologies Included in the Evaluation ................................107
FIGURES
Figure 2.1 Effluent Color in Two HWD/SWD ECF Mills...................................................................3
Figure 2.2 Bleach Plant Effluent Color as a Function of Unbleached Kappa Number........................3
Figure 2.3 Example of a Yield—Kappa Diagram.................................................................................4
Figure 3.1 Washing Loss as a Function of Dilution Factor..................................................................7
Figure 4.1 Example of Color Reduction by Increased Substitution...................................................I I
Figure 4.2 Impact of Kappa Factor on Bleach Filtrate Color.............................................................13
Figure 4.3 Contribution to the Kappa Number in Kraft Pulp.............................................................14
Figure 4.4 Example of an Acid Hydrolysis Stage(Domtar, Espanola, Ontario)................................15
Figure 4.5 Typical Arrangement of Ozone Medium or Low Consistency Stage (Espanola Mill).....18
Figure 4.6 High Consistency Ozone Stage.........................................................................................18
Figure 6.1 Daily Color for Three Mills...............................................................................................27
Figure 6.2 Example of a Simulation of a Fiberline Spill Collection System......................................32
Figure 6.3 Simulated Spill System Improvements..............................................................................33
Figure 6.4 Processes Involved in the Proposed Progressive System Closure at the
Skookumchuk Mill Based on the Literature......................................................................37
Figure 6.5 Simulated Impact of the Progressive System Closure.......................................................38
Figure 6.6 Fiberline Filtrates at the Blue Ridge Hardwood Line........................................................40
Figure 6.7 Fiberline Filtrate at the Blue Ridge Pine Line...................................................................40
Figure 6.8 Minerals Removal Process by Ion Exchange....................................................................41
National Council for Air and Stream Improvement

Figure 6.9 Impact of a Modest Recycle of Eop Filtrate on NaCI Content of White Liquor..............42
Figure 6.10 Crystallization CRP Process.............................................................................................45
Figure 6.11 Ion Exchange Process.......................................................................................................45
Figure6.12 ESP Dust by Leaching......................................................................................................47
Figure 6.13 Bleach Plant Filtrate Recycle at International Paper, Franklin, VA.................................48
Figure 6.14 Bleach Filtrate Recovery at Metsa-Rauma Mill ...............................................................49
Figure 6.15 Laboratory Simulation of Alkaline and Acid Filtrate Recovery.......................................50
Figure 6.16 EKA Nobel Partial Bleach Plant Closure.........................................................................52
Figure 7.1 Membrane Size Classifications.........................................................................................54
Figure 7.2 Ultrafiltration of Oxygen Delignification Effluents at the Nym6lla Sulfite Mill
inSweden..........................................................................................................................57
Figure 7.3 Ultrafiltration of Alkaline Filtrates from an ECF Bleach Plant........................................59
Figure 7.4 Color Removal Using NF270 Membrane.........................................................................61
Figure7.5 Membrane Bioreactor.......................................................................................................63
Figure 7.6 USFilter's PACT System Process(Single-Stage,Aerobic)..............................................66
Figure 7.7 Color Removal at Different Activated Coke Dose and Activation Periods......................67
Figure7.8 Electrodialysis...................................................................................................................69
Figure 8.1 Color Removal Process Used at Skookumchuck..............................................................72
Figure 9.1 Post Color Treatment Efficiency ......................................................................................75
Figure 9.2 Post Treatment Color Removal Percent Efficiency..........................................................76
Figure 9.3 Color Removal from Alkaline Bleach Plant Effluent.......................................................79
Figure 9.4 Color Removal versus Contact Time for Whole Mill Effluent.........................................80
Figure 9.5 Ozone and Biological Filtration at a Paper Mill...............................................................82
Figure 9.6 Super Critical Water Oxidation Process Developed by MODEC.....................................83
Figure 9.7 Color Removal with Pd-Pt-Ce/Alumina Catalyst.............................................................84
Figure 10.1 Pre-Evaporation of Bleach Plant Effluent.........................................................................85
Figure 10.2 Evaporation of Acid Bleach Filtrates, Carryover in Condensates....................................88
Figure 11.1 Color Removal in an Immobilized Fungal Bioreactor......................................................91
Figure 11.2 Color Removal by Algae Treatment.................................................................................92
Figure 12.1 Color Sources and Effluent Color in Four Study Mills....................................................98
Figure 12.2 Approximate Sources of Color from Hardwood Pulp Production....................................99
Figure 12.3 Approximate Sources of Color from Softwood Pulp Production.....................................99
National Council for Air and Stream Improvement

Figure 12.4 Bleach Plant Effluent Color at Varying Kappa...............................................................100
Figure 12.5 Benchmarking of Final Effluent Color............................................................................100
Figure 13.1 Effluent Color Benchmarking.........................................................................................106
National Council for Air and Stream Improvement

REVIEW OF COLOR CONTROL TECHNOLOGIES AND THEIR APPLICABILITY
TO MODERN KRAFT PULP MILL WASTEWATER
1.0 INTRODUCTION
1.1 Background
At the request of five U.S. bleached kraft pulp mills,NCASI contracted with EKONO Inc.to
undertake a review of color control technologies and their applicability to modern kraft pulp mill
wastewater. The study was to build on an earlier color reduction study carried out in 1995 (Baird
1995).
The ultimate goal of the current study is to help kraft pulp mills identify potential opportunities to
reduce effluent color. The specific objectives for each task are outlined below.
Task I: Develop a comprehensive list of different in-mill (both"brown"and bleached) and effluent
treatment color reduction technologies and describe the technologies,their applicability to modern
bleached kraft pulp mills, and their impact on color loads.
Task I includes a technology review based on literature and other information covering
• in-process measures, such as loss control,recovery of colored wastewaters, practical
minimum color losses associated with pulping liquor,modification of process operating
parameters(pulping and bleaching conditions), bleaching chemicals (CIOz,HZOZ, ozone,
etc.), and other operational practices
• treatment of color in wastewater, including wastewater from separate processes and
combined mill effluent wastewater through separation techniques (including management
of color residuals); using membranes, adsorption/ion exchange materials and color
removal/destruction techniques; using chemicals/polymers/catalysts,
evaporation/incineration, etc.
The detailed discussion was reserved primarily for technology or process operational practices
offered or adopted in the last 10 years. Data generated during this project are referred to as data from
the study mills.
Task 1I: Conduct a mill-specific review of the color control measures and benchmark the effluent
color of the participating mills relative to other mills.
Task II includes a) a mill-specific review of in-process loss control measures and end-of-pipe
treatment at each of the five mills participating in this project, including a review of successful and
unsuccessful in-process and end-of-pipe measures used to control color; and b)a benchmarking of the
current mill performance with respect to color contributed by bleach plant sources, and pulping liquor
sources.
1.2 Measurement of Color
Color is a property of the specific effluent sources in a mill. This report refers to the so called"true
color,"which in the U.S. is measured using the NCASI method(previously NCASI 253, modified to
NCASI 71.01 since 2000). The color of the effluent is measured by spectrophotometry at 460 r)m
after(pre-filtration using a 1 µm filter), adjusting the pH to 7.6 and filtering the sample on a 0.8 µm
filter.The standard solution to which the color is compared is a solution of platinum and cobalt salts;
therefore, color is expressed in platinum cobalt units(PCU). The measurement of the color is
sensitive to the pore size of the filter and to the pH, so standardized conditions are needed.
National Council for Air and Stream Improvement

2 Interim Report
Although color is a characterization of the effluent and not a quantity, if is customary to calculate
color as a mass(commonly expressed as lb color/d or kg color/d)based on the measured flow and
PCU concentration. Color permit limits are also varyingly expressed: as mass(lb color/d, kg color/d),
or as an effluent concentration, PCU, or as an increase in river color, PCU.
By treating color as a quantity, a mass balance can be developed for the colored streams in a mill. In
this effort many mills have documented a balancing problem. It is generally believed that the color of
the in-mill sources changes in nature on its way to the effluent treatment plant, increasing the color of
the sources. This so-called color amplification has been attributed to the presence of sulfide, green
liquor dregs, pH,mixing issues, anaerobic conditions, etc.
Similarly, mills treating the effluent in aerated stabilization basins have documented a reversion of the
color(or amplification)during the effluent treatment process.
The effect of sulfide on color reversion(amplification) in aerated stabilization basin influents and
effluents was the subject of a recent master's thesis(Esty 2005). Sulfide exposure experiments
undertaken with different effluents yielded as much as 100%color reversion in some tests,whereas in
other tests the results suggested that sulfide had relatively no effect on color reversion. The
differences appeared to be associated with the initial color of the wastewater; lighter colored samples
yielded higher color increase upon sulfide exposure than darker ones. It was concluded that the
underlying differences in color reversion appear to be related to the wastewater composition,given
the extreme variability of the wastewaters. When different humic functional groups were tested,the
catechol and anthraquinone solutions exhibited the greatest effect on sulfide color reversion.
The amplification or reversion of color was not included in this study.
2.0 TASK I: COLOR REDUCTION TECHNOLOGIES
2.1 Wood Species—Impact on Color
Since hardwoods contain less lignin than softwoods, the unbleached kappa number(or K number)is
lower for hardwood pulps than softwood pulps if the pulping and oxygen delignification processes
are similar. Therefore,the color of the effluent is lower for hardwoods than for softwoods.
A potential color control technology in multiple species mills is maximizing the hardwood portion, if
the product allows. An example of the difference in effluent color is illustrated in Figure 2.1.
Figure 2.1 shows the monthly average effluent color in two bleached kraft mills as a function of the
share of hardwood pulp production in each facility. Both mills are equipped with oxygen
delignification for both softwood and hardwood,but have different age and effluent treatment
processes. The lines indicate that softwood operation generates about 28-40 lb/ADT more color than
hardwood operation.The difference is clearly mill-specific, depending on process systems and
running rates when alternating between hardwood and softwood.
National Council for Air and Stream Improvement

Interim Report 3
60
� • Mill1
4 50
rn Mill
40
o`
0 30
20
w 0 0 m
10
0
0 20 40 60 80 100
%Hardwood Pulp Produced
Figure 2.1 Effluent Color in Two HWD/SWD ECF Mills
2.2 Kappa Number Reduction
2.2.1 Impact on Bleach Plant Effluent Color of the Kappa Number
The kappa number(K number) of unbleached pulp is an approximate measure of the amount of lignin
in the pulp.
Since lignin contains color-causing compounds,the color of the bleach plant effluent has been found
to vary with the kappa number of the unbleached pulp entering the first bleach stage. Figure 2.2
illustrates the color of the bleach plant effluent in the sequence DEoDED for two different wood
species,based on work by Liebergott(Liebergott 1992). Between kappa numbers 17 and 25,the
bleach plant effluent color—unbleached kappa lines have a slope of around 5.6 lb color/ADT/kappa
unit. A reduction of the kappa number,for example by 8 units from 25 to 17, has the potential to
reduce the effluent color by about 45 lb/ADt.
160
140 •Hem10c
•k mcelBalsam
0 120 Br
a
a 100
`0 8
0 0
U 6
0
4
0
2
0
0
1 1 2 2 3 3
0 5 0 Kapp 5 0 5
a
Figure 2.2 Bleach Plant Effluent Color as a Function of Unbleached Kappa Number(based on
Liebergott 1992)
National Council for Air and Stream Improvement

4 Interim Report
The color/kappa relationship is exponential, so a reduction from a kappa level of 17 to a lower kappa
would not result in the same color reduction as from kappa 25.
No similar documentation was found for hardwood pulp. The color/kappa relationship may be
different for hardwoods than for softwoods. Hardwoods have a higher content of hexenuronic acids,
which also are measured with the kappa measurement. The contribution to color of the hexenuronic
acids is most likely different from lignin-originated color, although not much data is available on
color of hexenuronic acids(see Section 4.2).
2.3 Oxygen DeligniScation
Oxygen delignification is the most widely used method in lowering the kappa number. Reduction of
the kappa number in the cooking process has also been studied and used, especially in the early
1990s. Studies of the pulp yield in laboratory and mill conditions have confirmed that cooking to low
kappa results in a reduced yield compared to oxygen delignification to the same kappa, as illustrated
in Figure 2.3 (Axegard and Stigsson 1998).
Yield,%on wood
60
48 ------------i---------,_ ........ %-delignn'ication....
Bleaching
i
46 -..._..... ---::�:- i_. .---------
....I:.......
••""' - Cooking
40
0 5 10 15 20 25 30
Kappa number
Figure 2.3 Example of a Yield—Kappa Diagram
With respect to kappa optimization between cooking and oxygen delignification,the trend is to target
a somewhat high digester kappa and take a large kappa drop in the oxygen stage. This is done in
order to maximize the total pulp yield. In newer and modified oxygen delignification installations,
the trend is to have a two-stage oxygen delignification system with or without intermediate washing.
A lower kappa means lower wood yield,i.e., a higher amount of dissolved wood material to remove
during pulp washing. More biofuel energy can be produced in the recovery boiler and less chemicals
will be consumed in the bleaching. However,the kappa target is very mill-specific and is determined
mainly by the process and final pulp quality and other requirements. Too low digester kappa
deteriorates the pulp strength and impairs its paper making properties. Kappa targets of the mills
participating in this study are shown in Table 2.1.
National Council for Air and Stream Improvement

Interim Report 5
Table 2.1 Kappa Targets of Participating Mills
Lines w/ Lines with
_ oxygen delignification oxygen delignification
Softwood pulp
Digester kappa 26—28 25—36
Unbleached kappa 16 -22
Hardwood pulp
Digester kappa 16 -20 15 -22
Unbleached kappa 10- 14
Use of polysulfide and/or anthraquinone can compensate the yield loss when lowering the kappa to
bleaching(Jiang et al.2002). The feasibility of polysulfide anthraquinone(PSAQ) cooking and the
impact on product qualities and papermaking, however, is mill-and product-specific. One mill has
installed PSAQ cooking process together with oxygen delignification and documented a reduction of
effluent color by about 50-60 lb/ADt(Pagoria 2005).
All the study mills that have implemented oxygen delignification have also noticed significant
effluent color reductions.
In addition to reduced bleach plant effluent color, a reduction of the kappa number also impacts the
following factors.
• Environmental parameters: lower BOD, COD, AOX per ton of pulp
• Chemical cost: lower bleach plant chemical costs
• Wood cost: lower yield—higher wood costs
• Biofuel: increased steam amount from biofuel in the recovery boiler(unless the boiler is
steam limited)
• Pulp strength: Too low kappa, especially from cooking, can damage the pulp and its strength
properties.
3.0 BLACK LIQUOR LOSS CONTROL
All kraft mills lose some of the black liquor from the liquor loop. There are in principle two types of
losses:
• stationary or continuous losses that are mainly determined on the equipment design and
efficiency; and
• diffuse losses and temporary discharges (i.e., spills, leaks, overflows).
The stationary losses include continuous losses such as washing losses,reject losses,knotter system
losses,brown stock screen room white water losses,carryover of liquor in condensate streams, and
evaporator boil-out losses. These controllable process losses are discussed in Section J.
The temporary discharges include spills, leaks, overflows,wash liquids and similar discharges that
can happen accidentally, or they can be part of normal mill shutdown/start-up procedures. These
types of losses are discussed in Section 6.1.
National Council for Air and Stream Improvement

6 Interim Report
The technologies for loss control and for recovery of lost black liquor are overlapping, since the
accidental losses can be controllable with the spill collection systems.
3.1 Improved Brown Stock Washing
Brown stock washing in modern mills is a completely countercurrent process, including brown stock
screening and possible post oxygen delignifrcation washing. The carryover of black liquor or oxygen
delignifrcation stage liquor to the bleach plant in the pulp off the last washer in the brown stock fiber
line, prior to any dilution, represents the washing loss.
The washing loss is typically measured as the amount in the pulp sheet of salt cake(Na2SO4), COD or
conductivity, and more seldom as color in the pulp sheet.Available data for the relationship between
color and other parameters are summarized in Table 3.1.
Table 3.1 Correlation of Color to Other Parameters in Liquid Squeeze from Washed Unbleached
Pulp
SWD/Oz HWD/No0, HWD/Oz SWD/O,
Color/COD 0.9 3 0.5 0.6
PCU/ppm (1.5-5.5) (0.3-0.9) (0.4-0.9)
Color/Na2SO4 1.1
PCU/ppm
Color/Dry Solids 0.8 3.5 0.7 0.9
PCU/ppm (2.1-4.9) (0.4—1.0) (0.6—1.2)
Color/ 1.0 7.7 1.8 1.2
Conductivity (3.9—9.8) (1.0—3.0) (0.7—1.8)
PCU/µS/cm
The data in Table 3.1 indicate that the color of the carryover to bleaching varies. For all oxygen
delignified pulps, the relationships between color and other parameters are about the same order of
magnitude. The washing loss of the hardwood pulp without oxygen delignifrcation seems to carry
more color than the oxygen delignified pulp. Obviously these relationships are mill- and-process
specific but may be useful when investigating color sources and reduction technologies.
The color of the washing loss carried in to the bleaching process will to some extent be bleached in
the first Do stage and then discharged in the bleach plant effluent with the wash filtrates. The extent
of this"color bleaching"of the carryover liquor was not found in the literature. However, a typical
engineering estimate is 50-70%reduction of the color originating from the black liquor in the
bleaching process.
Independent of the extent of the bleaching of black liquor originated color,the carryover to bleaching
will have a certain impact on the bleach plant effluent color. Therefore, a reduction of the carryover
reduces bleach plant effluent color.
The washing efficiency and thereby the carryover to bleaching is determined by the washing
equipment and the dilution factor in the pulp washing. In addition, the wood species, liquor charges,
and kappa drop in the oxygen delignifrcation impact the result. One guideline for the carryover to
bleaching is provided by the European Union's BAT document(EU-IPST 2001)where the following
data are given:
National Council for Air and Stream Improvement

Interim Report 7
Washing loss with conventional drum washers 10-16 Ib COD/ADt
Washing in a modern line with presses 4-8 lb COD/ADt
The washing loss can vary significantly in a given washing configuration,especially if the dilution
'factor varies significantly. For example, when the production rate varies, the carryover can show
significant variations.
Figure 3.1 illustrates the carryover as liquor solids as a function of the dilution factor for two washing
configurations with different washing efficiencies,E values(EKONO 2006).
Example of Carryover from Pulp Washing
70
0
60 a '1_____�___ s-Current
I 1
50 ____ ____i___ �Atltl one washing _
stage
O
z,40 J_____
I 1 I I
U
N I 1 I
a
rn i i I I
V
Q
J 0
0 1 2 3 4 5
Dilution Factor, m31ADt °LD
W EKONO
Figure 3.1 Washing Loss as a Function of Dilution Factor
Because the carryover-dilution factor relationship is logarithmic,the washing loss can vary
significantly if the dilution factor is not kept constant. If for example the dilution factor is allowed to
decrease from 2.5 in/ADt to 1.5 T/ADT the washing loss may increase by 11- 17 Ib/ADT as liquor
solids(BLDS)or as about 9-14 Ib/ADT as black liquor color. On the other hand,the washing loss
decreases only by 2-4 lb BLDS/ADT when the dilution factor increases from 2.5 to 3.5 T/ADT. The
stabilization of the dilution factor and the pulp washing process overall can be implemented with
advanced controls and paper design.
In summary,the best available technology(BAT)for carryover of black liquor to the bleach plant is
4-8 lb COD/ADt(4-8 lb washable Na2SO4/ADt or about 4-8 lb color/ADt)when washing with clean
water. This level of washing loss is estimated to contribute 2-4 lb color/ADt to the bleach plant
effluent color.
The reduction of the carryover of washing losses to the bleach plant has these benefits:
• lower bleach plant effluent color,estimated to be about 0.5 lb color/lb BLDS carried over to
the bleach plant, assuming that the black liquor color is reduced by 50%in the bleach plant;
• lower bleach plant effluent BOD,COD, and AOX;
• lower bleach plant chemicals demand, i.e., about 0.85 lb active chlorine/lb COD(332); and
• more bio-energy from the recovery boiler
National Council for Air and Stream Improvement

8 Interim Report
Possible cost items may include increased steam costs in the evaporation plant if the dilution factor in
pulp washing increases.
3.2 Reduction of Other Regular Black Liquor Losses
Black liquor lost to the sewer contributes to color. Black liquor losses are typically measured as dry
solids,conductivity, sodium, COD,and less often as color, although some mills measure liquor losses
as color. Examples of correlation of the color of black liquor to other black liquor parameters are
shown in Table 3.2.
Table 3.2 Correlation of Color to Other Parameters in Black Liquors(EKONO 2006)
Color/ Color/COD Color/Na2SO4 Color/Dry
Conductivity PCU/ppm PCU/ppm Solids
PCU/µS/cm PCU/ppm
Dilute black liquor 2
Specialty pulp mill 4
Pre-OZ filtrate squeeze 0.7 1.2 0.6
Pre-02 washing filtrate 1.2 1.1 1.1 0.82
3.2.1 Effluent-Free gnat Handling
The term "knots"frequently describes the rejects of the first stage of screening after cooking. This is
where the screen hole or opening is around 0.25-0.5 in. Knots include good chips that did not get
impregnated, as well as the"biological"knot(for example,where a branch connected the tree trunk).
Biological knots are very dense and have an extremely high lignin-to-fiber ratio. Their fiber yield is
thus low. In addition,they require more energy and white liquor consumption than normal wood.
Biological knots primarily become dissolved solids in the black liquor rather than pulp. Knots also
contain rocks, bullets, and tramp metal. Chip thickness screens, air density separators, and chip-
conditioning devices have all improved the capability of reducing the traditional"knot"reject stream.
(Bucher 1999).
Knots can be handled in different ways,most often by recooking, landfilling or burning. From a color
discharge point of view, it is important that the knots are handled without losing black liquor to the
sewer.
The knotters typically operate in series of two stages with the knots rejected from the second stage.
When the knots are returned to the digester(s)the knots are dewatered(e.g., in a knot drainer)and the
dewatering liquor returned to the washing line,so there is no effluent from the knot handling in those
cases. Vibratory knotters are used in many mills as the secondary knotting stage. These open
knotters are often a source of spills and leaks, even if the knots are recooked.
Some black liquor may accompany the knots as they are removed from the fiber line.
In one mill the knots were sluiced out of the process with decker brown water. By replacing the
sluice water with paper machine white water,the mill estimated the COD discharge to decrease by
about 3 Ib/t(Genco 2005). The black liquor color discharge would have decreased by roughly the
same amount(EKONO's estimate).
National Council for Air and Stream Improvement

Interim Report 9
One of the study mills in this survey has installed recovery sumps on the knot reject bins and recovers
the liquor drained from the knots to the brown side spill collection tank. This measure, together with
other improvements in spill collection and internal color control in the digester and brown side sewer,
reduced the color by about 8000 lb/d on the hardwood line (about 10 Ib/ADt of HW pulp).
3.2.2 Closed Screen Room
In modern mills the screen room water system is part of the brown stock washing, and is carried out
in a closed, completely countercurrent system. This is true whether or not oxygen delignifrcation is
used in the fiber line. The only stream leaving the system may be the screen room fine reject(in
addition to knots removed earlier in the fiber line). See Section 3.2.3.
During so-called "open screening,"the filtrate from the last washing stage(normally the decker
following the screening) would not be taken counter-currently back to the brown stock washing, but
would partly overflow to the sewer. The closing of the water cycle may require new screening
equipment and larger filtrate tanks and may be costly in older mills. However,most modernized
papergrade bleached kraft mills have implemented closed screening. The benefits include water and
energy savings as well as lower effluent load of black liquor. The carryover with the pulp to
bleaching may,however, increase and thereby increase the bleach plant chemicals demand,
depending on the efficiency of the washing following the screen room. All the study mills reported
the closing of the water systems in the screen room as a significant color reduction technology. One
study mill has documented a color reduction of 35%by implementing closed screening.
3.2.3 Effluent-Free Reject Handling
Most mills purge a small stream of reject from the screen room in order to discharge impurities. The
color load associated with that reject stream depends on the amount and consistency of the reject and
the concentration of the liquid discharged with the reject. Options to reduce color discharge with
reject streams include the following.
• Upgrading of screening equipment. In one mill,the primary and secondary screen baskets
were replaced with slotted baskets on a hardwood pulping line. This allowed the mill to shut
down the brown stock cleaners and eliminate the cleaner reject stream. The resulting
reduction in black liquor originated COD was around 9500 Ib/d(Genco 2005) or as color,
roughly 12 lb/t HW pulp.
• Replacement of brown white water with less colored water for dilution of rejects prior to
sewering. In one mill,the brown white water was replaced by paper mill white water at an
estimated COD reduction of about 5800 lb/d or about 10 Ib/ADt S WD pulp or approximately
10 lb color/ADt(Genco,2005). One mill employing oxygen delignification and a
modernized closed screening system discharges about 1.5-2 lb color/ADt pulp with the reject
when using combined condensate for diluting the feed to the last reject screening stage
(EKONO 2006).
• Washing and dewatering of the final reject with recovery of the wash filtrate and dewatering
liquids. After dewatering,the reject can be burnt or landfrlled. There would be no effluent in
such a system. One of the study mills documents the installation of reject presses to have
resulted in equipment problems and is considering abandoning the pressing due to unreliable
operation. The color associated with the reject is reportedly around 1500-2000 Ib/d;
however,the black liquor color is typically reduced by 50%in their effluent treatment plant,
so the impact on the final effluent would be lower. Another study mill successfully operates
a press on the screen room reject.
National Council for Air and Stream Improvement

10 Interim Report
• Depending on the ultimate reject disposal, a refining stage may be included in the reject
handling, especially if the reject can be used for other grades (e.g., corrugated medium). In
such a case the rejects would be washed after refining and the wash liquids recovered in the
liquor cycle.
In summary, a good target for the color discharge related to reject handling is 0-2 lb/ADt, depending
on how the rejects are disposed of
3.2.4 Minimization of Carryover in Black Liquor Flash Vapors
Vapors flashed off from black liquors often carry with them droplets or mist of black liquor. The
amount of carryover varies widely, depending on factors such as liquor properties, foaming tendency,
solids levels, soap content,process temperature and pressure, vapor velocity, efficiency of drop or
mist separation, and elimination equipment.
Mills normally have safeguards in the form of conductivity meters and alarms in order to detect
carryovers and counteract such situations.
Potential process points for black liquor carryover into condensing vapor streams that can enter the
sewer systems include
• digester relief vapors (batch digester relief, steaming vessel relief);
• digester flash tank or blow tank vapors; and
• black liquor evaporator vapor dome or vapor side vapors.
In normal conditions, the color from liquor carryover into the condensate streams is low because of
effective entrainment separators and mist eliminators.
The carryover(conductivity) is also limited by the way the condensates are handled or reused. If
stream stripping is employed,the condensates taken to the stripping have to be clean from black
liquor in order to avoid foaming and to maintain the stripping efficiency of the volatile compounds.
A normal conductivity limit is around 300 µS/cm(approximately the same in color units). Above this
limit the condensates would be recovered to the spill system or sewered.
A similar limit is typically used if the condensates are reused in the pulp washing or recausticizing
area in order to eliminate foaming or total reduced sulfur(TRS) emissions.
Contaminated condensate can have a color level at about 400 PCU. Clean condensates (e.g.,
condensates from effects 2-4 in a five-effect evaporation plant)typically have a color concentration
below 100 PCU. Condensates can therefore contribute with a black liquor originated color of around
1-3 lb/ADT.
3.2.5 Evaporator Boil-Out Procedures
Evaporator surfaces can scale for many reasons and by many compounds(by burkeite, which is a
sodium carbonate-sulfate mixed salt, by calcium carbonate,by aluminum silicate, by soap or fiber,
etc.). The evaporator surfaces need to be regularly cleaned to maintain efficient heat transfer. Of
course it is also important to try to prevent the formation of scaling through various measures
(Gullichsen 2000), but even if scale prevention is implemented,scaling may occur.
The scaling with burkeite occurs mainly in effects with>40% liquor solids concentration. This
scaling is typically removed by washing with weaker liquor(switching of effects).
National Council for Air and Stream Improvement

Interim Report 11
Other scaling, such as CaCO3 scaling,requires boil-out with water, condensate or acid washes. Liquor
losses can occur if the boil-out procedures are not carefully managed. The unit to be washed is first
drained to liquor storage. The wash and rinse liquids can be fully recovered or recovered until a set
conductivity value is reached, after which the rinse liquids are sewered. The set point of the
,conductivity,together with the amount of rinse liquid sewered,determines the black liquor lost in this
way.
A normal conductivity cut-off value for sewering the boil-out liquids may be around 2500 µS/cm (or
approximately 2500-5000 PCU of color). Sewering of, for example, 800 gpm of rinse liquid at that
concentration means a color discharge of 1000-2000 Ib/h(24000-48000 lb/d), a significant color
contribution if prolonged.
4.0 BLEACHING
4.1 Chlorine Dioxide Bleaching
Bleaching with chlorine dioxide or ECF bleaching is standard BAT for papergrade bleached kraft.
Bleach sequences for dissolving kraft,however,may include hypochlorite or chlorine bleaching.
Bleach Plant Effluent Color as a Function of
CIOz-substitution(based on Liebergott)
500
450
•Kappa 13-0
6
400 ____I_____ ___I____ .Kappa16.6 ___
I
Q 3e0
I I
zoo ____L.____150
,_____I____y_
U
100
50
0
0 20 40 60 80 100 120
S.Wlatian.%
Figure 4.1 Example of Color Reduction by Increased Substitution
All papergrade mills participating in this study converted to ECF within the last 10 years and have
documented significant reductions in effluent color with the conversion to ECF bleaching. The
impact of increased substitution based on laboratory studies is illustrated in Figure 4.1,based on data
by Liebergott(Liebergott 1992).
Table 4.1 shows the approximate effluent color reduction experienced from the conversion to ECF
bleaching in three mills, where the conversion impact could be approximately separated from other
improvement measures.
The percent reduction has of course varied depending on the C102 substitution degree before the
conversion to ECF and on other measures that were implemented in the same time frame.
National Council for Air and Stream Improvement

12 Interim Report
Table 4.1 Impact of ECF Conversion on Effluent Color in the Study Mills
Color before ECF Color after ECF Reduction, %
Mill 46lb/T 28lb/T 41%
Mill 2 100000 lb/d 70 000 lb/d 30%
Mill 450001b/d 36000lb/d 25%
The operation of the Do (or D 100)stage may offer some opportunities to impact bleach plant effluent
color, i.e.,by varying the kappa factor and the temperature. One option to optimize the Do stage is to
reduce the kappa factor(underbleach) in the Do stage and increase the bleaching power of later stages
by reinforcing the extraction stages with oxygen and peroxide and if necessary, increasing the CI02
charge in the D, and DZ stages. Another option is to increase the kappa factor(overbleach) in the Do
stage. The C1O2 required to reach a target kappa number is a function of pH, chloride ion
concentration, the type of pulp, and extraction stage conditions. Pulp mill bleaching conditions such
as pulp consistency, temperature, retention time, and pulp cleanliness are extremely varied. In
addition,the objectives of bleaching, in terms of target brightness, cleanliness, and strength differ
among products and mills. Because of this variety, bleaching conditions must be optimized for each
product produced at each mill.
The impact of the kappa factor on the ECF bleach effluent Do and Eop stage color was documented
by Wohlgemuth, Lam, and Willis(1997) for a conventional brown stock softwood pulp of kappa 29.7
in mill and laboratory trials.
National Council for Air and Stream Improvement

Interim Report 13
As shown in Figure 4.2 the color of the Do stage effluent was highest around a kappa factor of 0.18-
0.20 and declined if the kappa factor was increased to more than 0.25 or decreased to below 0.18.
The color of the Eop stage effluent.steadily decreased with increasing kappa factor. The increase of
the kappa factor would result in somewhat higher bleaching costs.
DoEop filtrate colour vs KF Bleaching savings vs KF
Um w.m
um
sawmassww
am sea
Passona,o weewmDama31 41T aTT 02, it
k+Pw hnw 4PPa factor
Do eolaer n KF Eap colour ve Kappa Factor
1� C.Y1
sm Som
T
am
g o
9 ism ! asoa
No
AN
I I I am
am
I I I
aN W. an oio am aw
° wow Fader
atl ¢lo Us as Gw IQaT„EY
W .fa% ,a .o,oex ,own
pw Pieter ,�y,
Figure 4.2 Impact of Kappa Factor on Bleach Filtrate Color(Wohlgemuth, Lam, and Willis 1997)
Conventional chlorine dioxide bleaching(Do stage)is usually conducted at 50-70 eC for 30-60
minutes. One option that has been studied to improve D stage bleaching is to raise the reaction
temperature to 85-95eC and extend the reaction time to 90-150 minutes in a so called DHa, stage or
commonly denoted as D*. Most work with hot C1O2 bleaching has focused on hardwood pulps
because they seem to give better results in a hot D stage than softwood pulps. (Lachenal and Chirat
2000). This is generally considered to stem from the fact that hardwood pulps have higher
hexenuronic acid(Hex A) content than softwood pulps. Some recommendations for a hot D stage
include the use of a hot acid stage before the hot D stage (i.e.,ADho,stage), optimization of the pH to
improve performance, and the addition of sulphamic acid or formaldehyde to the hot D stage to
reduce organically bound chlorine in the pulp.Although the hot D stage could be used to replace any
D stage(Do, Di, D2), it generally gives the best results when used to replace the Do stage. One
proposed five-stage bleaching sequence is HotDo-Eop-Dl-HotD2-P. The final peroxide(P)stage is used
to increase pulp brightness and improve the brightness stability of the pulp. One drawback with the
hot D stage(D*) is a slightly negative impact on the pulp viscosity and the energy required to reach
high temperature. Hot C102 bleaching has been shown to significantly lower the AOX discharges
National Council for Air and Stream Improvement

14 Interim Report
from bleaching. Part of the AOX reduction comes from the fact that less C102 is required to reach the
same brightness level and part from the fact that the elevated reaction temperature and extended
reaction time in a hot D stage cause the AOX to be degraded to chloride ions (Lindstrom and Ragnar
2004; Ragnar 2005). The impact on effluent color from a D* stage was not found in the literature.
The first D* stage began operation in 2002(Ragnar 2005).
4.2 Hydrolysis Treatment for Removal of Hexenuronic Acids
Hexenuronic groups(Hex A) are formed in kraft pulp cooking when methanol groups are split off
from xylan while transforming the glucuronic acid components to unsaturated hexenuronic acid units.
The hexenuroninc acids contribute to the kappa number as exemplified in Figure 4.3.
20 10
16 g
Olig
16 OHerA 6 pli9nln
14 oomerswmures _ 0 OHexA
❑otherstmdums
S 12 - 6
10 t 5
E E
9 4
s 6 x 3
4 2
2 1
0 I U] N-� I UM 0
Unbleached 02-delignifed OD-01eached ODE- Unbleached 02-delignlfied OQ-bleached OQ(OP)-
bleached bleached
Spruce Kraft Pulp Birch Kraft Pulp
Figure 4.3 Contribution to the Kappa Number in Kraft Pulp (based on Sevastyanova, Li, and
Gellerstedt 2002)
The impact on hardwood pulps is more significant than on softwood pulps and more significant for
oxygen delignifred pulps than for pulps without oxygen delignification. The hexenuronic acids
consume bleaching chemicals(all except oxygen and peroxide). The hexenuronic groups also
provide anionic sites to bind transition metals and heavy metals to the pulp,the presence of which
contributes to pulp color reversion and aging of the pulps, and if peroxide bleaching is employed,to
the peroxide decomposition by free radicals. Removal of the Hex A groups can thereby also reduce
the peroxide consumption.
In general, softwood pulps contain fewer hexenuronic acid groups than hardwood pulps(Figure 43).
Therefore,the removal of hexenuronic acid groups from hardwood kraft pulps prior to bleaching can
provide substantial reductions in chlorine dioxide consumption, but the savings for softwood pulps
are less. In addition, the various cooking methods impact the hexenuronic acid content of the pulp as
can be seen in Table 4.2.
National Council for Air and Stream Improvement

Interim Report 15
Table 4.2 Hexenuronic Acid Content of Various Pulps (Jiang, Van Lierop, and Berry 2000)
Pulping Process Wood Type Kappa Number Hexenuronic Acid
Mmol/g
Conventional(batch) Hardwood 18.2 76
Conventional(batch) Softwood 25.9 54
Conventional Softwood 25.5 22.7
Lo-Solids Softwood 21.0 15.3
Soda Softwood 25.0 0.5
Soda-AQ Softwood 21.0 3.7
Polvsulfide Softwood 24.2 4.6
The impact of the Hex A groups can be reduced by introduction of a selective hydrolysis step in the
bleaching sequence,generally as the first stage following the last brown stock washer. The selective
hydrolysis would be performed at conditions to be optimized for different pulps. Testing has been
carried out at pH 3-5, temperature 85-115°C for 1-5 hours (Lindstrom and Ragnar 2004; Ragauskas
2000). Full-scale installations exist in several mills outside North America and at Domtar's Espanola,
Ontario mill on oxygen delignified hardwood pulp as an A Z/D E Dn D sequence (Semeniuket al.
2002). See Figure 4.4.
fA. Wnt
� i ♦ '°'q! i. �e pac
Nam
A Z ! D E Dn D
Sequence O-A-ZO-E-Dn-D Chemical Savings 20%
Wood Species Birch,Aspen,Maple Final Brightness >90%,ISP
Post 02 Kappa 8:0 Final Viscosity unchanged
COD in Pulp -5 kgfr Tensile unchanged
02 consumption 3—6 kgIADT TOX in pulp 50.70%d reduction
C102 R-R 2.5 to 3.5 Dirt zero
Figure 4.4 Example of an Acid Hydrolysis Stage (Domtar,Espanola, Ontario)
The impact on effluent color of an acid hydrolysis stage is likely to vary depending on the sequence
and bleach plant water configuration. Ragauskas determined the color of the effluent from a
hydrolysis stage treating well-washed HW kraft pulps (Kappa range 11.6-14.2)for 5 hours, at 2%
consistency, pH 3 and 100°C, and found a color varying between 120-220 PCU(approximately 10-20
Ib/ADT) depending on the kappa drop in the acid hydrolysis stage (Ragauskas 2000). No information
about the color of the effluents from subsequent bleaching was available from this study.
The effluent color of complete bleach sequences with and without acid hydrolysis was studied by
Colodette et al. (2002,2005) on eucalyptus pulp. The laboratory study included bleach plant
National Council for Air and Stream Improvement

16 Interim Report
effluents without filtrate recovery and with filtrate recovery to brown stock and the recausticizing
area. Although the study was for a non-North American wood species,the results are included here
for their conceptual value.
Their study on open bleaching included the sequences D (EoP)D D,Aho,(Eop) D(PO) and Dho,(Eop)D
(PO), with a washing between each stage. The bleaching was performed on a pulp with kappa
number 8.3 and with a comparatively high washing loss as carryover COD=26 kg/t(52 lb/ADT). The
reported data are summarized below(Colodette et al. 2002).
Table 4.3 Comparison of Acid Hydrolysis Sequence with Other Bleach Sequences
Unit D(EoP)DD Aho,(EoP)D(PO Dna,(Eop)D(PO).
Brown stock kappa 8.3 8.3 8.3
Carryover kg COD/t 26 26 26
Total CI02 % 1.876 0.845 0.85
Total H2O2 % 0.333 1.333 0.9
Effluent AOX g/t 1112 277 554
Effluent COD kg/t 35.2 39.6 38.1
Effluent BOD kg/t 10.8 18.2 16.8
Effluent Color kg/t 34.1 44.4 33.8
Final viscosity mPa.s 13.4 11.5 11.4
Brightness A.D%ISO 90 89.9 90.0
The data in Table 4.3 indicate that the bleach sequence D(EoP)DD generated the lowest colored
effluent and produced pulp with highest viscosity. The relatively high effluent color in the acid
hydrolysis sequence may partly have been due to the high COD carryover, which would have been
"bleached"in the D(EoP) DD but not in the acid hydrolysis stage. Based on the data in Table 4.3,
one can conclude that if the bleach plant filtrates are not recovered, an acid hydrolysis stage may
increase the influent color if the pulp is washed after the hydrolysis stage.
The concept of recovering bleach plant effluents from sequences including an acid hydrolysis stage is
included in Section 6.2.3.4.
4.3 Ozone Use in Bleaching
4.3.1 Process and Applications
Ozone bleaching has received considerable attention during the last ten years as an opportunity to
reduce bleaching with chlorine containing chemicals.
Ozone gas is a basic industrial chemical that has been widely used,for example,for chemical
synthesis. Ozone must be produced on site because it is unstable with a half-life in water of 20-30
min at 20°C. It is produced by passing oxygen through a discharge gap between two electrodes.This
discharge causes dislocation of the oxygen molecules,some of which subsequently recombine in the
form of ozone. The oxygen carrier gas is typically recovered and returned to the ozone reactor.
Ozone can be produced at up to 13-14%concentration in oxygen gas; when exceeding a
concentration of 15%,the gas mixture will be dangerous to handle. Ozone is one of the strongest
oxidizing agents known and has, like oxygen,some selectivity toward lignin attack over cellulose, but
it requires optimized process conditions in order for it to not cause any serious fiber degradation.
National Council for Air and Stream Improvement

Interim Report 17
For industrial-scale bleaching, both medium-and high-consistency systems have been used. Low
consistency ozonation tests have been run at the hardwood mill in Espanola, Ontario, using the
sequence OA(ZD)EDnD. The tests showed equal kappa reduction and ozone consumption and lower
COD and power consumption when compared to the medium consistency stage(Epiney et al.2002).
Because of the considerably lower capital costs and the developments in mixing technology,the
medium consistency approach may have been more popular than the high consistency operation.
High consistency operation may offer an opportunity to more effectively separate the water systems.
Several new installations have been high consistency. No low consistency ozone stage is currently
not in operation.
The use of ozone has been investigated for pulps without or with preceding oxygen delignification.
However, ozone sequences in use currently all follow an oxygen delignification stage See Table 4.4.
The prerequisite for ozone bleaching include metals removal in an acid washing and/or chelating
stage, i.e., a D or a Q stage. The ozone bleaching stage itself is simple, consisting of a mixer or
mixers for ozone, a short reactor(2-5 min) and a washing stage or it is followed by the next bleaching
stage. The pH is typically around 3 and temperature at 40-50'C.
There are many full-scale installations using ozone bleaching, some of which are listed in Table 4.4.
Examples of medium and high consistency ozone stages are shown in Figures 4.5 and 4.6.
Table 4.4 Full-Scale Kraft Mill Ozone Bleaching Installations
Approximate
Start- Capacity
Mill up Type, Sequence (ADt/d)
IP,Franklin,VA,USA 1992 ECF,HC O(QZ)EoD 900
Stora,Skoghall.Sweden 1992 ECF OZDEnDED,Pilot plant 200
MoDo,Husum, Sweden 1993 TCF OQPZP 1000
Metsii-Botnia,Kaskinen,Finland b 1993 TCF/ECF OPZ/ODPZ 1200
UPM,Pietarsaari,Finland"' 1993 ECF/TCF OQXZQO(PE)PZP(TCF) 1000
O(ZD)O(PEp)DEPD
SCA, Ostrand, Sweden 1995 TCF,HC Q(OP)(ZQ)(PO) 1250
Metsa-Rauma,Rauma,Finland 1996 TCF (00)(ZQ)P(ZQ)(PP) 1000
Stora-Enso,Wisconsin Rapids,WI, 1996 ECF,HC OZEoDD
Votorant,Lacarei,Brazil 1996 ECF
Votorant,Luis Antonia,Brazil 1997 ECF/TCF
ZP Rosenthal,Germany 1999 ECF/TCF, Q(OP)DQZ(PO)P(ECF) 900
HC Q(OP)QZ(PO)P(TCF)
Domtar Inc.,Espanola,ON 1999 ECF OA(ZD)EDnD 1000
Burgo Ardennes,Belgium 2000 ECF,HC ODZEO,DD 1030
Votorantim Celulose a Papel(VCP), ECF,HC OAZEDP 2100
Jacarei Mill,Brazil
Sappi Ngodwana SW OA(ZD)(EO)D
Ruzumberok, Slovakia 2005 HC
Oli,Nichinan,Japan HC
'HC=high consistency b Main product is ECF pulp. ` Ozone is not used in all sequences
National Council for Air and Stream Improvement

18 Interim Report
0,Writ to destruct unit
10,i residual
or Gas
Or Delignifed Pulp
from Astage wash press
10j inlet -
Dilution water
P.T
To Do stage
0,on r O,rml.r
Pulp and rdlrete Pulp and filhate
sample sample
Figure 4.5 Typical Arrangement of Ozone Medium or Low Consistency Stage(Espanola Mill)
(Epiney et al.2002)
Spent Gas
�• to recycling
-
From 02-post washing
H2S0q
Chelant �-�--
Ozone "-
ti
To 02-post washing
Figure 4.6 High Consistency Ozone Stage (Gullichsen 2000)
National Council for Air and Stream Improvement

Interim Report 19
4.3.2 Risks
The main risk associated with the ozone technology is reduction in yield and potentially the pulp
strength.
'A recent study documented a yield loss of 0.5%and 0.8%for the sequences(DZ)EpDED and
(ZD)EpDED, respectively, compared to the conventional DEpDED sequence on softwood pulp
(Huber et al. 2006).
4.3.3 Environmental Impact
Ozone breaks down the lignin into smaller compounds. Thus,without filtrate recovery,BOD and
COD could rise in the untreated effluent. Treated effluent BOD and COD might not increase,
however, because the lower molecule weight compounds break down more easily in the ASB. AOX
would be reduced due to lower chlorine dioxide consumption.
Metso's ZE-Trac process (former Union Camp's"C-free"process)includes recovery of the ZE stages
back to brown stock washing. Scandinavian mills using ozone bleaching are also exploring partial or
total recovery of the filtrates. In such a case,the effluent discharges in the bleaching effluent can be
greatly reduced(see Section 6.2.3).
Table 4.5 shows data for three different North American softwood pulps for different combinations of
DZ and ZD sequences(Colodette et al. 1999). The effluent data exclude the oxygen delignification
stage. As shown,the effluent color was in many cases slightly higher for the ozone containing
sequences than for the conventional DEOpDD sequences. The main environmental impact from the
ozone stage was a reduction in the AOX discharge. The viscosity of the pulps bleached in the ozone-
containing sequences was lower than the conventionally DEOpDD bleached pulps.
Table 4.5 Ozone Bleaching Laboratory Study Results on Softwood ZD and DZ Combinations
(Colodette 1999)
Total CIO,
Kappa O, CI02 Repl. Bright Bright. COD
Factor kg/OD kg/O kg/kg .% Revers. Vise. kg/O Color AOX
Sequence Ist Stage t Dt 03 ISO %ISO mPa.S Dt kg/ODt kg/ODt
Western Canadian
Spruce/Pine(Kappa 18.9)
OD(Eup)DD 0.24 0 26.8 - 90 87.2 19.9 46.1 38.5 0.83
O(DZ)(Eo,)DD 0.11 3 21.2 1.87 89.6 86.1 14.6 44.8 44.8 0.48
O(ZD)(E"e)DD 0.11 3 21.2 1.87 90 86.9 17.2 46.7 43.8 0.58
OD(OP)(ZE)D 0.19 3 16.7 3.37 90.2 87.3 18.8 45.5 42.5 0.52
Southern Pine A(Kappa 12.9)
OD(EOP)DD 0.24 0 20.1 - 90 87.2 15.9 30 27.8 0.53
O(DZ)(EOP)DD 0.11 3 16.8 1.1 90.3 86.5 12.2 27.3 28.1 0.2
O(ZD)(EOP)DD 0.11 3 16.8 1.1 90 87.1 13.4 29.4 29.3 0.32
OD(EOP)(ZE)D 0.14 3 14.9 1.73 90 87 14.7 29.2 28.7 0.28
Southern Pine B(Kappa 11)
OD(EOP)DD 0.24 0 16.5 - 90.3 87.3 17.1 27.5 23.1 0.47
O(DZ)(EOP)DD 0.11 3 13 1.17 90 86.4 12.7 26.8 24.8 0.17
O(ZD)(EOP)DD 0.11 3 13 1.17 90 86.9 14.1 27.3 25.4 0.27
_ OD(EOP)(ZE)D 0.11 3 12.2 1.43 90.3 87.1 15.2 27.4 24.2 0.23
National Council for Air and Stream Improvement

20 Interim Report
Laboratory tests by Liebergott for softwood pulp with kappa 29.5 gave the results included in Table
4.6 for impact of an ozone stage on color(Liebergott 1995).
Table 4.6 Comparison of Effluent Data for Ozone and ECF Bleaching, kg/ADt
AOX COD BOD Color
D100 Eop 1.8 44 16 40
ODIoa Eop 0.5 23 6 14
D/Z5q Eo 0.17 42 13 38
OD/Z25 Eo 0.10 31 8 19
Based on the two tables above,the use of ozone did not reduce the bleach plant effluent color
compared to the conventional ECF sequence,DEopDD, when the bleach filtrates were not recovered.
Table 4.7 documents similar data for hardwood(Eucalyptus)pulp for the sequences O-DEoPDD,
OQOP-ZEDD (Z-ECF) and OQOP-ZQPO(TCF). The effluent data include stages starting from the
D or Z, i.e.,the study assumed that the effluents from the 0(single stage oxygen delignification)and
OQOP(double stage oxygen delignification)stages were recovered.
Table 4.7 Ozone Bleaching Laboratory Study Results on Hardwoods (Colodette et al. 1998)
Effluent Kraft Kraft- Kraft- Modified ModifiedModified
from Kraft AQ Polysulfide PSAQ Kraft Batch Kraft
Broom Stock Kappa 18.2 17.4 17.9 17.2 17.2 16.9
ODEooDD(ECF Sequence)
Kappa after O 10.9 10.4 10.5 9.8 10.4 10.4
Color Ib/ADT DEopDD 13.8 12 16.4 17.2 17.6 17.8
COD Ib/ADT DEopDD 34.4 41 44 41 40.6. 36
(OQ1(OP)(ZF)DD(Z-ECFscQA
Kappa after OQOP 10.3 10.0 10.0 9.5 9.9 9.8
Color Ib/ADT (ZE)DD 11.6 11.8 13 12.8 13.6 13.8
COD Ib/ADT (ZE)DD 34.8 34.4 31.8 30.6 31.8 31
001(OPl(ZO)PO(TCF Sea)
Kappa after OQOP 10.3 10.0 10.0 9.5 9.9 9.8
Color Ib/ADT (ZQ)(PO) 8.2 7.8 8 8.2 8 8.4
COD Ib/ADT (ZQ)(PO) 6T6 65.2 66.6 64.2 63.6 65.8
The data in Table 4.7 indicate that the color from hardwood bleaching using the Z-ECF sequence is
only slightly lower than using ECF for the conventional kraft cooks,taking into account the
difference in kappa before the D or ZE stage. The color from the TCF sequence was lower. Another
feature was the difference in color for the different cooking processes, indicating a higher color from
pulp cooked in modified processes compared to conventional cooks. The COD of the TCF sequence
effluent was about double that of the ECF sequence in this study. Other parameters differed between
the bleach sequences as indicated in Table 4.8.
National Council for Air and Stream Improvement

Interim Report 21
Table 4.8 Comparison of Bleach Sequences (Colodette et al. 1998)
Parameter Unit ECF Z-ECF TCF
Pulp Yield %in final bleaching 95.6 95.4 95.1
Viscosity dm3/kg 983 960 877
Tear Index at 40°SR mNm-/g 9.61 9.99 9.37
Tensile Index at 40"SR mNin=/g 98.0 93.5 100.6
PFI rev at 40'SR 2537 2659 2987
Relative Chemical Cost 110 119 159
Relative Steam Cost 100 131 147
Effluent AOX kg/t 0.251 0.116 0
Munro et al. reported the impact of a ZD stage on color. Mill data from the Domtar Espanola,
Ontario mill documented a significant reduction in color with the modernization and installation of an
ozone stage on the hardwood line (Munro or al.2001). No data for the bleach plant color was
however made available,only color in the total mill effluent, including softwood and hardwood line,
where a 27%reduction in color was reported.
Based on the information in Tables 4.5-4.7 it can be concluded that the benefit of using ozone is most
significant for the AOX discharges. The situation changes if the bleach filtrates are recovered (see
Section 6.2.3).
The chelating agents such as ethylenediaminetetraacetic acid(EDTA)and diethylene-
triaminepentaacetic acid(DTPA)normally have to be used in connection with ozone and peroxide
bleaching because of their ability to suppress the activity of the dissolved transition metal ions
without precipitation("Q"). These metal ions are able to catalyze the decomposition of the bleaching
agent hydrogen peroxide into radicals. Totally chlorine free(TCF)bleaching is currently only
possible by treating the pulp with Q before the ozone and hydrogen peroxide stages. Increased
concentrations of Q are therefore found in wastewaters generated from the production of TCF pulps.
Although EDTA is non-toxic to mammals at environmental concentrations, there is some concern
about the potential of EDTA to remobilize toxic heavy metals out of sediments and the difficulties in
biodegrading this substance.
4.4 Peroxide in Bleaching
In recent years, hydrogen peroxide has been used extensively in ECF bleaching sequences as a
reinforcing chemical in alkaline extraction stages. In TCF bleach sequences, oxygen delignification
is usually necessary prior to hydrogen peroxide stages in order to obtain desired brightness levels.
Also, in the absence of an acidic D stage,the purging of transition metals(which deactivate peroxide)
has to be accomplished some other way.
4.4.1 Peroxide in the Eop Stage
The use of peroxide in sequences like ODEopDD or ODEopDEpD represents the conventional ECF
bleaching technology. A recent survey showed that both alkali and peroxide were used in the
extraction stages in 90%of the industry (Pryke, Kanters, and Tam 1999,2000). About 0.2-0.5%
peroxide is normally added, in many cases to a pressurized"tube"with short retention time.Peroxide
application in the E stage in addition decreases the color of the alkaline filtrate by 10-30%.
National Council for Air and Stream Improvement

22 Interim Report
Several options exist for optimizing the EoP stage for reduced environmental impact. These options
include optimizing the temperature,reaction time, and hydrogen peroxide charge,metals management
through chelation or acid stages,and the addition of magnesium or other peroxide bleach stabilizers.
Increased hydrogen peroxide charge was used at the Leaf River mill to meet the color limit under
extreme low river flow conditions(Smith and Walley 2002). The mill established an internal target
for the effluent color corresponding to 500 pcu(-21 kg/ADt,42 lb/ADT). This target was met by
modifying the bleach plant operation. Hydrogen peroxide was added to the two extraction stages,
changing the bleaching sequence from DEoDED to DEoPDEpD. The showers on the first two
chlorine dioxide stage washers were also changed,replacing the fresh water with additional caustic
extraction filtrate.This took all the flow that was going to the caustic sewer and recycled it back
through the extraction stage. The main costs were doubling the hydrogen peroxide usage. Increased
depositing of barium sulfate on the first chlorine dioxide washer was another result. Annual cleanup
of the deposits will be required.
4.4.2 Peroxide in TCF or ECF Light Bleaching
Peroxide can also be used as the dominant bleaching chemical in TCF and Low CIOZ ECF(ECF-
light)sequences.
Several mills have used various peroxide bleaching sequences,some examples of which are seen
below.
• OQPP
• OQPZP
• OQPPP
• OQPDEpD
• OQPoPQP
The last sequence listed,TCF sequence OQPoPQP,was used at the Louisiana Pacific(currently
Evergreen Pulp Inc.)mill in Samoa, California. Brightness levels on the order of 85+were achieved
while maintaining pulp strength equal to that of the original chlorine compound bleached product.
The total effluent color with the open bleach sequence was reportedly about 200 PCU for an effluent
flow of around 100 m3/t(15 MGD)corresponding to 40 lb color/T(Louisiana Pacific 2000).
Currently,the mill produces unbleached pulp(Yolton and Patrick 2005).
Another development in peroxide bleaching has been pressurized peroxide(PO or EoP). Typical
peroxide stages operate at atmospheric conditions at temperatures between 80-90°C for long
residence times(four hours or more),while the pressurized stages can have residence times of 15-60
min.
Two examples of ECF-light bleach sequences which incorporate pressurized peroxide stages are
OQ(PO)DND and ODEOpD(PO). The benefit of the first sequence is that the(PO)filtrate can be
recycled to post-oxygen washing with fewer chloride issues,while with the second,the need for a Q
stage is eliminated.
Both non-pressurized and pressurized peroxide stages are now well established both in ECF and TCF
bleaching. Table 4.9 lists examples of peroxide bleaching applications.
National Council for Air and Stream Improvement

Interim Report 23
Table 4.9 Examples of Full-Scale Peroxide Bleaching Installations
Mdl Bleaching Sequence Wood type
Munksjo,Aspa,Sweden° TCF OQPPpp°
SCA,Ostrand,Sweden TCF OQ(PO)(ZQ)(PO) SW,HW
Stora-Enso,Norrsundet,Sweden' TCF OQEopQ(PO)` Sw
Sodra,Monsterlis,Sweden TCF OQ(OP)(PaaQ)(OP) SW,HW
Sodra,Morrum,Sweden TCF OQEoPQ(PO) SW,HW
Sodra,Varo,Sweden" TCF OQEopQ(PO) SW
Rotmeros,Vallvik,Sweden° TCF OQPaaQ(PO)PPp° SW
Stora-Enso,Kemi,Finland' ECF OQo(PO)DEpD SW,HW
Stora-Enso,Uimaharju,Finland° TCF OQPPaaP` Hw
Metsa-Rauma,Rauma,Finland TCF (00)(ZQ)P(ZQ)(PP) Sw
Sunila,Kotka,Finland' TCF (00)QEopCaP° SW
UPM-Kymmene,lappeenranta,Finland' TCF (00)QEOPQ(PO)P° SW
UPM-Kymmene,Lappeenranta,Finland' ECF (00)(ED)DEoPD SW
UPM-Kymmenc,Pietmsaari,Finland' ECF ODEopDP SW/HW
UPM-Kymmene,Pietarsaari,Finland' TCF OQZ/QOPZ/or SW/HW
Evergreen Pulp,Samoa,CA TCF' QPoPPP SW
Stora-Enso,Skoghall,Sweden ECF O(OP)DQ(OP) SW,HW
Votorantim Celulose a Papel(VCP),Jacarei Mill, ECF OAZEDP HW
ZellstoffStendal,Germany ECF OQ(OP)D(PO) SW
CIA Suzan Bahia Sul,Brazil ECF D,(PO)D�
°Main product is ECF pulp. also uses Pas. `Sequence when producing TCF;main product is ECF dcurrently unbleached
Note that there is some overlap in the definition of the two processes. Many of these mills can
produce both ECF and TCF pulp. Only the SSdra mills and the SCA Ostrand mill in Sweden and the
Metsa-Rauma mill in Finland produce TCF only. The other mills produce mainly ECF pulp,but can
produce TCF on demand.
The risks associated with peroxide bleaching processes are lower pulp strength and reduced optical
brightness.
4.4.3 Environmental Impact
Peroxide bleaching reduces effluent color if it is used early in the bleaching process. Therefore,
peroxide bleaching could have a significant impact on color,while a(PO)stage at the end of an ECF
bleach sequence would have almost no effect. However, in the early stages of the bleaching peroxide
can also impair the pulp strength and yield. Since nearly all installations are in Scandinavia where
effluent color has not been an issue Of concern,there are unfortunately no data available to allow
precise estimation of color reduction(e.g.,when comparing conventional ECF bleaching and
peroxide based TCF bleaching).
If the peroxide stage effluents were recovered back to brown stock,the color reduction potential
would be even higher. AOX will be reduced as the C102 usage is lowered in an ECF sequence.
However,the BOD and COD will be increased as a result of a higher use of peroxide, if the filtrate of
the peroxide stage is not recovered.
In conclusion, use of peroxide stage in an ECF bleaching sequence reduces the bleach plant effluent
color,and more so if the peroxide or Eop-stage filtrate is recovered.
4.5 Peracid Bleaching
4.5.1 Process and Applications
Peracids can be used in both oxygen delignification and bleaching stages. Peracids have a high
selectivity for lignin and can be used as a partial or complete replacement for chlorine dioxide.
Peracids include peracetic acid (Paa) or CH3C000H, and Caro's acid or CaP=(Ca)H2SO5.Peracid is
currently used in several Nordic mills. Peracetic acid is manufactured as a distilled product from
National Council for Air and Stream Improvement

24 Interim Report
acetic acid and hydrogen peroxide. It is mainly used in TCF bleaching sequences(TCFP,,) where
higher brightness has been achieved, as well as improved strength, when compared to peroxide- or
ozone-based sequences. The recirculation of Paa-stage effluents could be a possibility in order to
further reduce effluent color.
All full-scale applications include a chelating stage prior to the Paa stage because the hydrogen
peroxide in the peracid solution generates hydroxyl radicals. Without chelation,the peroxide
consumption would increase. In the Rottneros Vallvik mill in Sweden,the peracetic acid bleaching
stage is operated at 60'C for a 3-hour retention time, and is followed by a second chelating stage with
no intermediate washing. The Vallvik mill does not have biological effluent treatment. Effluent
color, however, is not reported. The COD discharge in 2005 was 41 kg/ADt(82 Ib/ADT)and the
AOX discharge was 0.11 kg/ADt bleached pulp(0.22 Ib/ADT) (Rottneros 2004).
Bleaching with peracids is a proven technology. It is practiced in full scale at several Nordic mills,
listed in Table 4.10. Many of the mills listed in Table 4.10 predominantly produce ECF pulp, but
have the capability to make TCF pulp on demand. Mill-scale trials with peracids have been
conducted at many more mills, including Munksjo/Aspa in Sweden and the Stora-Enso mill at
Norrsundet in Sweden. The kraft mill in Stendahl, Germany also has the option of using a Pau stage
in place of its D stage, but this is not currently done.
The bleach sequence at Sodra in Monsteras, originally Q1, OP, Z(Q), PO,was modified in 2000 by
replacing the ozone stage with a Paa-stage. (Sodra Cell, 2002). The peroxide stage is a Prepox stage.
Bleach plant washing is done in a press(after pre-bleaching), two vacuum filters in parallel(after the
Q stage)and three Kvaemer wash presses following the OP, PaaQ, and PO stages. The bleached pulp
is then screened, cleaned, and dried in two Flakt dryers.
The Sodra V9r6 mill also used Paa in bleaching(Sodra Cell 2002).
Table 4.10 Full-Scale Peracid Bleaching Installations
Mill Peracid used Sequence Wood type
Rottneros,Vallvik,Sweden' Peracetic OQPaaQ(PO)PPP Softwood
Stora-Enso,Kemijarvi,Finland' Peracetic (XQ)(00)PPaaQP Softwood
Stora-Enso,Uimaharju,Finland' Peracetic OQPPaaP Hardwood
Sunila,ICotka,Finland' Caro's Acid (00)QEOpCaP Softwood
Stora-Enso,Oulu,Finland' Peracetic ECF main sequence SWD/HWD
Sodra,Viirb Peracetic TCF Softwood
Sodra,M6nsteras Peracetic TCF SWD/HWD
Zellstoff Stendal,Germany' Peracetic OQ(OP)Paa(PO)or Softwood
OQ(OP)D(PO)
Main product is ECF pulp with some chlorine dioxide use.
4.5.2 Risks
Peracid bleaching may not be compatible with all pulp quality parameters. It has been applied in TCF
bleach lines with the purpose of improving brightness and strength. An environmental risk would be
increased BOD and COD in the effluent caused by the acetic acid, unless Paa stage filtrates are
recirculated back to brown stock washing. As shown in Table 4.10, all known sequences utilize the
chelating agents (Q) in order to accomplish bleaching to high brightness levels.
National Council for Air and Stream Improvement

Interim Report 25
4.5.3 En vironmental Impact
The Pau treatment can be carried out with both equilibrium (ePaa) and distilled acid (dPaa), although
the optimum pH differs(ePaa pH 4-5; dPaa pH 5-8). The main difference is the higher effluent load
,when using ePaa as illustrated in Table 4.11.
Table 4.11 Effect of ePaa and dPaa-Stage on Bleaching Effluents(Vuorenvirta, Panula-Ontto, and
Fuhtmann 1998)
Stage/Paa charge Unit CODc, TOC Oxalic Acid
ePaa/10 kgBDT 27.9 8.6 0.08
P kgBDT 7.9 3.5 0.08
ePaa+P,total kgBDT 35.8 12.1 0.16
dPaa/10 kgBDT 13.5 5.7 0.03
P kgBDT 7.2 2.7 0.07
dPaa+P,total kgBDT 20.7 8.4 0.10
dPaa/5 kgBDT 9.9 3.8 0.02
P kgBDT 7.7 3.1 0.08
dPaa+P,total kgBDT 16.6 6.9 0.10
In addition to the reactions of the peracids with lignin, peracids have been found to react with the
hexenuronic acid(HexA) by oxidizing it to smaller compounds. The peracids have been reported to
actually degrade HexA faster than lignin. Removal of the HexA prior to peracetic acid treatment in
an A-stage (acid hydrolysis)would result in more selective bleaching and reduced peracid
consumption.
Since most of the installations and trials have been carried out in Nordic countries where effluent
color is typically not regulated,there is very little information available on the impact on color. Early
laboratory studies have indicated that peracids applied to effluent reduce color to the same extent that
peroxide does,which would suggest that a Paa stage would behave similarly to a peroxide stage.
If Part effluents were recirculated back to brown stock washing,the effluent color would be reduced.
Because of the acetic acid used,effluent BOD and COD would rise unless the effluents were
recirculated back to brown stock washing.
5.0 COLOR MONITORING
Monitoring of the color is essential for color control.
Although color is a property of the specific effluent sources in a mill, it is customary to calculate
color as a mass load so that a mass balance could be developed for the colored streams in the mill and
for the total treatment plant influent. Many mills have documented a balancing problem, and it is
generally believed that the color of the in-mill sources changes in nature on its way to the effluent
treatment plant, increasing the color of the sources. This so-called color amplification has been
attributed to the presence of sulfide,green liquor dregs,pH, mixing issues, anaerobic conditions,etc.
In essence, it is largely unknown what causes the color amplification or how extensive it is.
It is, however, essential for color control to establish an in-mill color balance as detailed as possible to
understand the color sources. After that,an appropriate color measurement and monitoring system
can be set up so that color sources can be monitored on a sufficiently frequent basis.
National Council for Air and Stream Improvement

26 Interim Report
Most mills monitor the bleach plant filtrates and black liquor sources separately. For black liquor,
conductivity is frequently used as a substitute for color. In areas where the color losses are due only
to the presence of black liquor,the color mass can be estimated when the typical color/conductivity
for black liquor is known and the effluent flow(see Table 3.1)..
An essential part of the color monitoring would be directed to monitoring spills and spill handling
systems. Monitoring for control of black liquor spills is described in the following section.
Monitoring of washing losses and other regular black liquor originated sources as well as kappa
number variability should also be part of the color monitoring. Variation in washing losses and kappa
numbers are reflected in the bleach plant effluent loads.
Blue Ridge Paper Products (Blue Ridge 2006)is using an extensive in-mill monitoring system to
manage and detect black liquor and bleach filtrate losses. Their monitoring system includes
• 2 In color monitoring of influent color(1 hr monitoring during upsets and outages);
• in-mill daily color source balance with color measurements in about a dozen in-mill effluent
streams;
• alarms for sewer conductivity, liquor tank level, and filtrate tank level;
• redundant overflow alarms for black liquor and spill containment tanks;
• real time sewer conductivity and color trend data for mill sewers available at process operator
control stations;
• action levels for process operators for color in influent and mill sewers from each production
area; and
• daily process control trend charts for color in mill sewers and waste water treatment.
The detailed color balance developed on a daily basis is essential for maintaining low color loads at
that mill.
6.0 RECOVERY OF COLORED WASTEWATERS
6.1 Black Liquor Spill Recovery
The Cluster Rule promulgated by the U.S.Environmental Protection Agency(EPA) in1998 includes
requirements for mills to implement"best management practices"(BMP) plans.These,in effect,are
spill control systems for black liquor,turpentine, and soap. To comply with this rule,the mills must
record variability of effluent before treatment and take corrective action to reduce peaks.The rule
leaves decisions on control criteria to mill management. The chosen control criteria typically are
conductivity,COD, or TOC, but some of the study mills in this survey have chosen to use color as
BMP control parameter.
The prime target for most spill control programs is black liquor. Soap and turpentine can cause
catastrophic spills that drastically reduce the performance of the biological treatment, and these types
of spills have to be prevented by enclosure in accordance with the BMP.
Black liquor spills, however, are difficult to completely prevent,so black liquor loss(spill)control
programs include both prevention of spills and recovery of spills.
National Council for Air and Stream Improvement

Interim Report 27
6.1.1 Contribution of Spills to Effluent Color
Task I1 of this report addresses the contribution of black liquor losses to the influent color in the
actual study mills in this survey.
Data illustrating influent color variability in three mills published by McCubbin(McCubbin 2001 a) is
shown in Figure 6.1.
ua -
Wt D89y color EdirflueiittoWW7Pf6rMiBS
.No,.
M
tdt
Tm
Daly cdaatkAteittleVeA Plvr k9la
ida
a•
w
m ,
ft! Wily colar#inNuvAluVMW PM1L
10C
ra
ten_ in
Figure 9:Daily color flow for three different mills
Figure 6.1 Daily Color for Three Mills(McCubbin 2001a)
National Council for Air and Stream Improvement

28 Interim Report
"The graphs represent six months of daily color flows for three mills, expressed as
kilograms of color per average ton of pulp production at the treatment plant influent.
The mills are similar. The principal difference is the execution of the spill control
programs. All mills have two fiber lines. The mills have paper machines built before
1940 and are located on severely restricted sites with far from optimal layout of the
equipment and sewer systems.
All three mills have complete coverage with spill recovery sumps, and a number of
conductivity monitors with data display and alarms on the operator's control panels.
Management at all three mills believes they have excellent control of spills. In the
author's view,the performance of the systems shown ranges from average to
excellent. Clearly the staff at mill S is the most effective at controlling spills."
(McCubbin 2001)
Assuming that the peaks in the charts in Figure 6.1 are due to spills, one can draw the conclusion that
significant spills could double or even triple the daily average influent color compared to the normal
operation in the mills included in that survey.
Some variability in the influent daily color is caused by variability in the process operation. Variable
production rate, kappa number, brown stock washing efficiency, and bleaching conditions impact the
influent color, but likely not to the extent shown in mill C and L in Figure 6.1.
6.1.2 Black Liquor Spill Management
Spill discharges consist of foam, leaks,tank overflows,failures in pump seals, unauthorized draining
of equipment for repair, etc. Spill in the form of foam is probably the most difficult to handle,
because of the consistency. Foam can occur in large volumes, causing overflows and problems
transporting the spill through the usual spill collection system. Combined with the fact that spill is by
definition diffuse,the exact amount of spill discharge is often difficult to estimate.
A good spill management system includes measurements and monitoring, spill prevention, spill
collection, and lastly, spill equalization to reduce impact on effluent treatment and level out
unavoidable spills.
6.1.3 Spill Prevention
Important features for spill prevention include the following.
• Operating philosophy. Steady production and process conditions always result in lower
process discharges compared to variable conditions. This of course requires appropriate
instrumentation and control,but also the correct set points for e.g.,production and operating
conditions. Both from production and discharge point of view it is often more beneficial to
target e.g., a manageable production level that can be steadily maintained than to target too
high production that causes equipment overloads, downtime and variability, overflows, and
uneven operation, all of which can result in a lower average production.
• Operator and maintenance personnel training. According to McCubbin(McCubbin 2001 a)
the principal difference between the three mills in Figure 6.1 is operator skill and attitude.
Education and training of process operators in understanding the process operating
parameters that impact effluent color is important for the prevention of spills and reduction of
variability. Training and education of types and sizes of spills and the impact of spills on
color is an essential part of spill management to reduce the frequency and magnitude of the
spills.
National Council for Air and Stream Improvement

Interim Report 29
"Training programs should be tailored to the mill's specific systems, and to the level of
knowledge of the personnel. Training should explain the key parameters, and how the
department worked in affects the effluent discharge. Each piece of equipment or operating
procedure that can generate spills should be identified and explained,and corrective action
defined. In all cases,training should encourage feedback from operators and maintenance
people, since they know many local details well. A good spill control system provides
operators with continuous, rapidly updated data on key factors of plant operation. Operators
must understand these data,be able to diagnose causes, and take corrective action. Initial
training requires several hours of class time for each student, with a couple of hours as a
refresher each year for most operators and maintenance personnel. Continuous feedback
helps workers learn from mistakes. Mills where spills are controlled successfully normally
have a report of all major incidents at daily production meetings, and advise all operators in
relevant departments about what happened and how to avoid repeat incidents."(McCubbin
2001b)
• Maintenance of equipment. Many mills perform daily rounds to check condition of
equipment,piping, and pumps for leaks.
• Instrumentation and control. Reliable instrumentation is very important for spill prevention.
Level measurements with alarms on liquor tanks alert the operators about potential situations,
so that overflows can be prevented. Conductivity or temperature measurements and alarms
on overflow pipes and in locations where equipment failures might occur are important to
alert the operating personnel about spills so that preventive actions can be taken quickly.
• Cascading of tank overflows and adequate filtrate and spill collection tank volumes. Most
large spills occur when tanks overflow.
6.1.4 Spill Collection and Recycling
One important feature of a spill control system is to collect the spills, e.g.,tank overflows,before they
reach the floor drains. A modern brown stock line with oxygen delignification typically has at least
six liquor filtrate tanks that are coupled for countercurrent washing via the wash filter showers. When
imbalances in these long systems occur, the result may be an overflowing tank. Rather than
overflowing to the sewer and to the spill collection system, some mills have cascaded the overflows
countercurrently to the preceding liquor(or fiber)tank. That way, any extra dilution is avoided and
the overflows are recovered closer to the source. Where tank overflows are a problem,an overflow
detector is useful. Some mills monitor temperature in the overflow pipe, since this will detect foam
overflows that fail to show on the tank level monitor.
The spill recycling system with drains,sumps, and spill tanks should collect spill and directly recycle
it back to the process, preferably as close to the spill source as possible(Lundstrom, Berglin, and
Annergren 2002). Recovery pumps are often installed in all sumps that receive spill. These pumps
are activated automatically, when highly contaminated spill is collected in the sumps so that the spill
can be transferred to a spill collection tank. Conductivity is often used to estimate the contamination
in spill. If a large amount of water is recycled, it can cause overload in the evaporators,to the point
that operators have to deactivate the spill recovery system(McCubbin,2001b). Uncontaminated
water also dilutes the contamination in spill, which can cause significant amounts of discharge to pass
the recovery pumps undetected. It is,therefore, important not to mix uncontaminated water into the
spill recycling system. Other important factors in the spill recycling system are outlined below.
• Sufficiently large spill tanks with the capacity to collect even large spills.
National Council for Air and Stream Improvement

30 Interim Report
• Sufficiently large pipes that can quickly regulate any imbalance in the system.
• Appropriate number of spill sumps throughout the mill. Simple,single line mills typically
require 3 to 6 sumps, though some mills require a dozen or so. Actual requirements are very
site-specific, and usually involve some compromise between the ideal configuration and the
costs of retrofitting.
• Continuous measurement of conductivity in each sump and operating area of interest.
Locations should be selected so that they will serve to locate spills in a reasonably small area
(such as an evaporator set or the digester department)all under the control of one operator.
• An appropriate choice of conductivity limit in the sumps to activate spill recovery pumps.
• Recovery sumps on knot reject bins.
• Mechanical pump seals in black liquor areas to avoid clear water dilution of color materials
that prevents efficient recovery.
Specific conductivity is the most successful parameter in current use for continuous monitoring for
black liquor spills. It will also detect spills of white and green liquor,and usually soap (because some
black liquor normally travels with soap). Spills have to be monitored instantaneously by conductivity
or other continuous sensor, at multiple points,to initiate appropriate corrective responses. A longer-
term assessment is also useful, particularly to benchmark against other mills. COD and color are
useful for such assessments.
The conductivity limit for activating spill pumps or rerouting an in-mill sewer or floor drain to a spill
collection system eventually determines the black liquor losses to the effluent. In North America,
most mills consider that an effluent stream should be pumped back to the black liquor system when
the conductivity exceeds 5000 micromhos/cm(µmhos). This corresponds to a black liquor
concentration of about 0.5%, although some mills have set points for recovery at 2500 µmhos
(McCubbin 2001a). A lower limit was employed in a Swedish mill where the usual limit was 150
mS/m or 1500 µmhos. That limit is increased to 175 mS/m (1750 µmhos)during shutdowns or in
cases of large temporary spill events,to ensure that as much high contaminated spill as possible is
recycled (Lundstrom, Berglin, and Annergren 2002).
When taking into account the typical ratio color/conductivity of black liquor(1-4 PCU/µmhos) (see
Table 3.1),the color contribution of a spill can be approximately calculated, if the volume of the
black liquor spill is known.
The recycling of a spill also has an economical value,because chemicals and energy are lost with the
black liquor solids and need to be replaced. On the other hand,the water with the spill has to be
evaporated which consumes energy. To minimize the cost it is again important to separate clean
waters from black liquor floor drains that are connected to spill sumps.
6.1.5 Example of an Optimization of a Spill Collection System
The spill collection system in each mill has to be optimized individually for the specific
configuration, layout, floor drain, and sewer layout.
A Swedish investigation studied the spill system in an existing mill (Monsterds, 750000 ADtlyear
bleached market kraft pulp)and established an improved concept based on a dynamic simulation of
the fiber and liquor systems (Lundstrom, Berglin, and Annergren 2002).
A schematic structure of the existing fiber line and spill recycling system is shown in Figure 6.2(Part
1). The filtrate tanks in this mill are connected such a way that the filtrate in one tank can be directly
National Council for Air and Stream Improvement

Interim Report 31
transported to the next tank. This structure reduces the risk of overflow of filtrate from these tanks.
The maximum allowable level in the blow tank is—80%. At higher levels the operators reduce the
inflow from the spill system to reduce the risk of overflow. The total volume of the spill tanks on the
brown stock side(Tanks 1,2, and 3)was 230 1113 or 60,000 gallons.
Three typical types of spills were simulated:
• leaking pipe(10 Us or 160 gpm leak for 16 hours at the end of the brown side);
• emptying a wash filter(filter in the screening position plugs and has to be empted fast);and
• a major tank overflow(a 500 in or 130 000 gallon overflow from the blow tank flushed
down with water so that the pulp in the floor drain had a 1%consistency).
The present spill recycling system in this mill was good,but dimensions were not sufficient to cope
with all the simulated spill cases. Suggestions for improvements to the spill recycling system that
were identified in the study are shown in Figure 6.2 (Part 2). The suggested improvements are
outlined below.
• Replacing of Spill Tanks 2 and 3 with a new 2000 m3(510000 gallon)tank. The size was
chosen to ensure that all highly contaminated spills from the simulated cases were collected.
• New pipes from both sump 1 and spill tank 1 to the new spill tank.
• A seal that automatically shuts down the inflow of spill when the volume in the blow-tank
exceeds 80%,to reduce the risk of overflow. To quickly control changes in the spill
recycling system,the dimensions of the pipe between the blow tank and the new spill tank
were such that a flow rate of 150 Us (-2400 gpm)could be handled.
• A separate system for clean water or water containing low levels of contaminants,to reduce
the dilution of the spill recycling system.
With these improvements,the simulation showed that even a significant blow tank overflow could be
collected so that the average daily COD discharge from black liquor remained around 1 kg/ADt, (2
Ib/ADT)compared to an increase by 12.4 kg COD/ADt(25 Ib/ADT) in the existing mill case.
The simulated improvements in COD discharges are shown in Figure 6.3.
National Council for Air and Stream Improvement

32 Interim Report
C.'" ADU_3 10000no
81.tank HC-tame
ld f
0A �T 4c;
(2)
1-7
FitnateTa nk Nftiate tank Bu er ank Filinate tank Ftin,ate tank
D.J.
Cnd,
S'iU fiom S,ill(i
bf.uh,i.nt
pflM k 1
Tmatmam
C.'et., 30.3 joanit 100nn3 135.0
Figure 1. A schematic figure of the fiber line and the spill recycling system in the mill. The location of the
three spill cases can be seen in the figure. Numbers refer to the three spill cases: (1) leaking pipe, (2)
emptying of a wash filter, and(3)tank overflow.
cdjaNY
elves lank He-lows
L �t bi in g
1 1 L-
Filt1m4eaFORnAnTa Fiftat.1.0 n ,lank Ffltmta tank
..........
Main.
Maw nnftnr:l OIL!j II
bleachiplant
PaVy2000m3 I Spilinank4
S l S�ratmnin,
Biological Fit t2nkl
Treatment C,-t,30-a
Figure 2: Proposed changes at the mill to improve the spill recycling process.
Figure 6.2 Example of a Simulation of a Fiberline Spill Collection System
(Lundstrom,Berglin, and Annergren 2002)
National Council for Air and Stream Improvement

Interim Report 33
Table 3:Comparison or tho original and the irapmvod spill rucvcling syslot,,
'Avvaago COD disehmp during 24h to biological treatment(3:F/A 0
ConOnnous Leuk orn plpt' FJnpfyiitrifa z Tnik
6Dill,-' 'x7nnslt Qltes _oecFlloR
Size ortlie scenario Nornul day sawn Medium Large
Or[glrml spill recycling system
Dilution factor (m'/ADi) 2.6 3.8 2.8 2.7
Avvra c COD disdtar • l.0 0.7 IA 12.4
proposed split rnryde system
Dilution factor WAD) 2.7 3.8 2.6 4_7
Mvraac COD disel +o' 0.8 0.5 1 1.0 1 2
Figure 6.3 Simulated Spill System Improvements (Lundstrom,Berglin, and Annergren 2002)
The study concluded that a spill recovery system that would reduce the black liquor spill discharges
from the fiber line to a mean level below 1 kg COD/ADt or about 2 lb color/ADT could probably be
constructed(at an estimated cost of 5.1 million SEK or about 0.6 million USD,2001) but not without
negative effects. Recycling of spill results in either a high dilution factor of material entering the
evaporators or deterioration in washing efficiency, followed by a larger amount of carry-over to the
bleach plant. To recycle as much organic matter as possible,the flow of clean water into the spill
system has to be kept low. To reduce the amount of water entering the spill system, a separate system
for uncontaminated water(seal water,cooling water etc.) can be built. Large temporary spill events
must also be avoided by spill prevention techniques to ensure a low total discharge from a mill.
6.1.6 Spill Diversion (Sewer Diversion)
Diversion of spills is often utilized in cases where spills escape the in-mill recovery systems and enter
the effluent stream. To reduce the impact on effluent treatment and a risk for exceeding e.g., a daily
color limit, it is beneficial to have a back-up where the colored waters can be stored and slowly bled
into the treatment plant, or even slowly recovered into the process if the evaporation plant capacity
allows. The sewer diversion is typically managed via conductivity guards ahead of the diversion
system. Diversion for effluent color is most useful if the mill has a segregated sewer system so that
black liquor-containing sewers are separately monitored from other sewers. That way the
conductivity from elevated black liquor content can be separated from other high conductivity losses
(white and green liquor losses, ESP dust purges, demineralizer backwashes, etc.). Because the
effluent stream at the treatment plant influent often has a high flow and also contains non-colored or
less colored effluent streams,the back-up volumes should be large(million gallons).
One study mill treats the diverted black color-containing stream in a spare primary clarifier batch
wise with polyamine,which separates the black color from the effluent. The sludge from the spare
clarifier is mixed with the primary and secondary sludge.
6.2 Bleach Plant Effluent Recovery
6.2.1 General
The recovery of bleach plant effluent in combination with the black liquor recovery cycle has been
the target of a long-time development effort. The introduction of the ECF and TCF technologies,
chloride and non-process elements removal processes, has enabled partial recovery of the bleach plant
effluents.
National Council for Air and Stream Improvement

34 Interim Report
Kraft pulp bleaching alternates between acid and alkaline conditions. Modern bleach plants cascade
the water systems countercurrent to produce an acid effluent and an alkaline effluent from the first
two bleaching stages. The alkaline filtrate is compatible with the brown stock area from a pH point of
view; therefore, most bleach plant effluent recovery concepts include the use of alkaline filtrate in the
brown stock area. The first acid filtrate in the bleaching process contains the major part of metals and
minerals that are released and washed out from the pulp in the acidic conditions, and represents a
purge point for non-process elements that enter the process with the wood (green liquor dregs being
another major purge point). Recycle of the acid filtrate is not common. It is known to be in
continuous operation only at the Blue Ridge mill, as a partly closed system, and in the TCF bleach
line at the ASSI Doman,Fr6vi where the acid effluent(Q) is evaporated in a separate evaporation
plant(see Section 10).
6.2.2. Recycle of Filtrates in ECFBleacking
6.2.2.1 Color in Filtrates
Table 6.1 summarizes effluent color data from the alkaline and acid stages of some of the ECF mills
participating in this study.
Table 6.1 Example of Color in Acid and Alkaline Bleach Plant Effluents in This Study(ADT=shT)
Acid Effluent Alkaline Effluent
Wood type Kappa Total Color
Ib/ADT Flow, Color, Flow,m3/ADT Color,
m3/ADT Ib/ADT Ib/ADT
HW 12.8 24.4 23.6 16.4 15.1 8.0
HW 17.4 37.6 21.5 27.0 4.5 10.6
SW 17.8 36.1 8.5 15.4 6.0 20.7
SW 16 55.4 27.1 22.9 18.8 32.5
SW 28 62.5 29.9 29.3 10.7 33.2
SW 22.6 67.5 43.3 26.1 12.3 41.4
HW 16.9 40.1 32 37 1 3.1
As shown in Table 6.1,the alkaline effluent contained only about a third of the bleach plant color
load in the two hardwood bleach lines, while in the softwood lines the alkaline filtrate contained 50-
60%of the bleach plant color. Note that the data in Table 6.2 do not take into account any
amplification of the color measured in the in-mill streams compared to influent color. This
phenomenon may cause an increased color load of the bleach plant effluent and may impact the acid
and the alkaline streams differently.
National Council for Air and Stream Improvement

Interim Report 35
6.2.2.2 Filtrate Recycle
The recycling of chloride containing filtrates from ECF bleaching is practiced in some mills. The
most extensive known recycle is at the Blue Ridge mill in Canton,North Carolina,where the Bleach
Filtrate Recycle (BFR)process for recycle of both alkaline and acid filtrate is employed on the pine
line and alkaline(EoP) filtrate recycle on the hardwood line. Another kraft mill using partial recovery
of the EoP filtrate is the Tembec Inc. mill in Skookumchuk, British Columbia. That mill has
announced a long-term plan for complete recovery of the EoP filtrate.
To achieve a significant color reduction from the Fop filtrate recycle, a substantial portion of the
alkaline filtrate needs to be returned to the liquor loop.
The presence of chloride in the alkaline filtrate is large enough to require chloride removal from the
liquor loop at substantial degrees of filtrate recycle in order to protect equipment from chloride
corrosion and especially the recovery boiler from plugging. With low degrees of recycle(< 10-30%),
a sufficient chloride amount could be removed by purging recovery boiler ESP dust, depending on
the mill chloride balance, but with more recycle of alkaline filtrate the installation of a chloride
removal process becomes necessary (Baxter et al.2004).
Removal of chloride from the liquor cycle can be motivated even without recovery of the bleach plant
filtrates, depending on the chloride intake to the liquor loop from other sources. Chloride removal
processes are discussed in Section 6.2.2.7.
The optimum process concept for recycle of the alkaline bleach filtrate is mill-specific and has to be
developed for each mill individually,taking into account equipment configuration in the bleaching,
brown stock, and recovery areas, chloride and potassium and other non process element balances,
color balances, washing efficiencies, effluent flows and composition, etc.
The following are risks of E-stage effluent recirculation:
• build-up of chlorides, potassium and other NPE in the liquor system;
• increased foaming and corrosion in brown stock washing;
• higher recovery system loading;
• disruption of the sodium-sulfur balance; and
• new technology at very high recycle degrees in ECF.
The concept proposed for the Tembec Inc. Skookumchuk mill and the concept in operation at the
Blue Ridge mill are reviewed below.
6.2.2.3 EoP Filtrate Recycle at the Tembec Inc.Skookumchuk Mill
One of the mills using EoP filtrate recycle is the Tembec mill in Skookumchuck, British Columbia.
This mill is a softwood market kraft mill with oxygen delignification and ECF bleaching.
Skookumchuck has been recycling 10-15%of its EoP filtrate back to the unbleached fiber line since
August 1996. The EoP filtrate has been returned back to the liquor balance tank between the decker
and the two-stage atmospheric diffusion washer and occasionally to the second post oxygen washer
filtrate tank. The chloride content of the liquor loop has been managed, albeit the white liquor
chloride concentration has increased.
The Skookumchuck mill has been working with Paprican of Canada to develop a progressive system
closure concept(Baxter et al. 2004). One of the main objectives of the progressive system closure
concept at Skookumchuck is to achieve 100%EoP filtrate recycle back to the second post oxygen
National Council for Air and Stream Improvement

36 Interim Report
washer. Limitations to increasing EoP filtrate recycle include brown stock washing operation, black
liquor evaporation capacity, chloride levels in the recovery cycle, and increased bleach plant chemical
consumption. On the other hand, sodium and dissolved solids would be recovered.
To achieve 30% EoP filtrate recycle,computer modeling done by Skookmmchuck has shown that 3%
of the ESP dust must be sewered to prevent increased chloride levels in the white liquor. To achieve
100%EoP filtrate recycle,the following would be required at Skookumchuck.
• Chloride removal system for treatment of ESP dust to remove excess chloride that
accumulates in the recovery cycle. The planned system is an ion exchange system(PDP; see
Section 6.2.2.7)
• Reuse of black liquor evaporator condensate that is being displaced by the EOP filtrate. The
displaced condensate can be used to replace fresh water and reduce mill effluent flow.
Possible uses for the condensate are the bleach plant or pulp machine. Further treatment of
the condensate maybe required.
• Use of EOP filtrate on the second set of DO showers. This strategy is expected to lower the
volume and increase the concentration of the EOP filtrate being recycled to brown stock
washing.
To combat the side effects of filtrate recycle at Skookumchuck,several other projects would be
required to facilitate system closure. These include
• improvements in post oxygen washing;
• improvements in brown stock washing control;
• improvements in bleach plant control;
• water conservation;
• evaporator and recausticizer upgrades;
• installation of a chlorine dioxide generator acid purification and evaporation system; and
• installation of a sodium thiosulphate removal system to help control the mills chemical
inventory.
Figure 6.4 shows a block diagram of the processes involved,based on the description of the system
given in the literature(Baxter et al. 2004).
National Council for Air and Stream Improvement

Interim Report 87
Eop
Bleached
Pulp
Chips BSW and Pulp
Pulping 02 dell D E Bleach Plant Machine H2O Do
Filtrate
OWL Thiosuilate CI02
removal
process
Na2S203 Bleach Chemicals C_ondensate
Hi0 preparation Polishing
Recovery
GAP CI02 generator
Hi0 Na2SO4d/HzSO
POP NaCIO3
H2SO4
NaCl
Figure 6.4 Processes Involved in the Proposed Progressive System Closure at the Skookumchuk
Mill Based on the Literature(Baxter et al.2004)
The simulated impact of the proposed progressive system closure on the effluent color and dissolved
organic material and on the chloride concentration of the white liquor is shown in Figure 6.5,based
on the published material. The mill effluent color was simulated to decrease by 50%with 100%Eop
filtrate recycle. The effluent flow volume at the mill was about 40 m3/ADt,so the resulting total mill
color discharge was simulated to be 40 m3/ADt * 450 PCU or 18 kg/ADt(36 lb/ADt),without the
use of chemical treatment. Without 100%recycle, the mill effluent color would be about twice that
amount or 72 lb/ADt(without chemical treatment).
The implementation of the progressive system closure at the Skookumchuk mill is reportedly
currently on hold,so there is no information available with respect to the actual color reduction
achievable with the proposed system in operation.
National Council for Air and Stream Improvement

38 Interim Report
m 25 30
x 20
15 _ ___. 25 -__ __ __ _______ _ ____
______ _ __ ___
m
10
5 __ ___.—
c y
0 0
O Base 15%Eop 30%Eop 100%Eop .2
Case recycle recycle recycle+ O 10 ______
PDP
1200
U 1000 - __ 0
'0 800 -- ----.- ------------ Base 15%, 30% 100 100
u 600 -- ----- ----------- Case Eop Eop % %
a 400
W200 -- ----- ---- -- Chloride concentration in white liquor with increasing Eop
0 filtrate recycle rate and after implementation of the PDP system for the
processing of40%arch ESP dust.3%of the ESP dust is assumed to be
Base Case BSWISpIII 100%Eop sewered in the 30%recycle case for chloride control.
Control recycle
Figure 6.5 Simulated Impact of the Progressive System Closure(Baxter et al. 2004)
6.2.2.4 Eop Filtrate Recycle in Blue Ridge Hardwood Line
The Blue Ridge Paper Products mill currently recycles Eop filtrate on the hardwood line to reduce
effluent color. The recycling is possible because the chloride removal process is in operation at the
mill as part of the BFR process. A schematic of the hardwood line filtrate systems is included in
Figure 6.6. The estimated color reduction due to the partial For,filtrate recycle on the hardwood line
is 5-7 lb/ADT of hardwood pulp(Blue Ridge 2006).
6.2.2.5 Bleach Filtrate Recycle (BFR)Process
The Blue Ridge Paper Products mill developed their bleach filtrate recycle(BFR)process to reduce
the effluent color from the pine line bleaching. The process enables recycle of the filtrates from the
first two stages of bleaching back to the kraft recovery cycle. There are three key components to their
process:
• oxygen delignifrcation combined with 100%C102 substitution(OD 100), in order to
minimize the impact of dissolved solids carryover to bleaching;
• a minerals removal process (MRP)to purge metals from the fiber line by treating the first D
stage effluent; and
• A chloride removal process(CRP)to remove chloride from the recovery cycle by treating the
precipitator catch.
In the BFR installation the Eop filtrate is split in half,with part being used as wash water on the first
D stage and part sent to the post-oxygen washing decker seal chest.
National Council for Air and Stream Improvement

Interim Report gg
The first D stage filtrate is also split, with part going directly to the post-oxygen washing decker
upper shower, and part being treated in the MRP. Treated D filtrate is reused as wash water on the
post-oxygen washing decker lower shower and the first D stage shower. Only filtrate from the final
D stage would enter the sewer at complete closure of the system. Currently, the degree of closure is
around 75%. A schematic of the water systems in the pine line is shown in Figure 6.7.
The recycling of the bleach filtrates to this extent, including recirculation of both acid and alkaline
chloride containing filtrates, is possible because the mill has installed a minerals removal process
(MRP)and a chloride removal process(CRP). Minerals removal processes are discussed in Section
6.2.3.6 and chloride removal processes in Section 6.2.3.7. The mill maintains a certain redundancy in
these processes to achieve high uptime.
The BFR project reduced the final effluent color by about 12 lbs/ADT total pulp or about 28 lb/ADT
of softwood pulp (15000-20000 lb/d). Other environmental benefits would include reduced BOD,
COD, and AOX.
The BFR process generates two new waste streams:the ion-exchange backwash from the MRP,and
the chloride purge stream from the CRP. These streams have to be sewered in order to remove non-
process elements and chloride from the system. The recovery boiler is equipped with a direct contact
cyclone evaporator. Liquor droplets carried over into the flue gases have therefore resulted in a
colored purge from the chloride removal process. The color of the purge is about 5000 Ib/d(3.5
Ib/ADT) in the influent, but most of this color is likely to be removed in the effluent treatment plant.
Colored purge would probably not be an issue in a recovery boiler without direct contact evaporator.
The operating costs have been 10%higher due to filtrate recycle. Chemicals and energy would be a
major part of the increased costs. The MRP requires a sodium chloride regenerant. Recovery loop
makeup requirements would be impacted by the addition of sodium from the bleaching effluents and
by the amount of sodium lost with the purge from the chloride removal process (as sodium chloride,
carbonate and sulfate, in addition to the same salts of potassium). The net effect on the sodium-sulfur
balance and chemical makeup requirements would be very mill-specific,depending on the actual
sulfur-sodium-chloride-potassium balances.
The evaporation in the CRP process requires steam in an amount that depends on the selected
chloride removal process. There would also be a minor increase in power demand.
The main process risks with the BFR process include potential corrosion, scaling, and pulp quality
issues.
National Council for Air and Stream Improvement

40 Interim Report
Blue Ridge Paper Canton Mill-No.1 Hardwood Fiber Line and Bleach Plant Filtrate Flow
B
65W Y S 3
asW 1l 1 11
Ww nc n.lm Fa 62
u<Ya n%-aY
's e 3
WWI MW2 Oil
{ta
nbae nlem olzL olm Pa I2
TmY Tn1 Ne i Tet
Tw
aaV a roF..p®on
maw
uwa
Figure 6.6 Fiberline Filtrates at the Blue Ridge Hardwood Line
Blue Ridge Paper Canton M1111I-No.2 Pine Fiber Line and Bleach Plant Filtrate Flow
n „
caw E 3
caw rou i i
2 C8° 1 1
O.Im T
0.<Yn af I
cuM1 0.3
3
L
n.lm
FJuue
CFWI ®W3 Ce%V] P 02 Tmk
f ilmm riNue rilirn[ r'i= a� Flo-ae M.TvnY 3knY lknk T" Osckn Fit." Tak Tmk
no-m
Tan%
nlnP — —
nm To 6vapmamn
alxw
Figure 6.7 Fiberline Filtrate at the Blue Ridge Pine Line
National Council for Air and Stream Improvement

Interim Report 41
6.2.3.6 Removal of Non-Process Elements
Ion Exchange Methods
The minerals removal process(MRP)installed at the Blue Ridge mill is an ion exchange process and
'is shown in Figure 6.8.
Removal of non-process elements is necessary if the acid bleach filtrates are extensively recovered, as
is the case in the Blue Ridge mill, in order to avoid scaling and other harmful impact due to build-up
of non-process element concentrations in the fiber line.
In this process the De stage(DI00)filtrate pumped to the MRP is first sent through a fiber filter. The
fiber is returned to the Da stage,and the filtrate drops into a storage tank. It is further filtered in
several parallel sand filters in order to remove other suspended solids down to a size of one micron,
dropping to a second storage tank.
It is then pumped through parallel ion-exchange beds to a treated filtrate tank,from which it is
returned to the process.
The ion-exchange beds must be regenerated with NaCl. One bed can be regenerated while the others
continue to treat the filtrate.
Blue Ridge Paper Products, Canton
Schematic of Metals Removal Process
Bleach Filtrate (MRP)
from D1 Stage
Fiber
Return I
ZFlltorodT
t
Fiber
Strainer %Troatod
Regenerant
_____ waste- >
Untreated Treated D7 Filtrate
to Bleach Plant
Caron, Delaney, 1997
Figure 6.8 Minerals Removal Process by Ion Exchange (Blue Ridge 2006)
National Council for Air and Stream Improvement

42 Interim Report
"Chin Kidnev"
The main source for metals non-process elements(NPEs) in the bleach plant acid effluent is the
wood.Normally, a large part of the NPEs contained in the wood will stay attached to the fibers until
they reach an acidic stage in the bleach plant. The other part of the NPE follows the black liquor to
eventually exit primarily with green liquor dregs (Wohlgemuth et al. 2001). It has been proposed to
use a"chip kidney"to purge metals. The metals would be extracted in an acid treatment stage(acid
hydrolysis)of the chips. This process removes metals effectively(Backlund and Radestrom 2005),
but it also creates a new effluent that contains the metals and some wood extractives. Before the
"chip kidney" can be regarded as an acceptable process, a suitable method for handling the acid
treatment effluent has to be developed. Precipitation methods have been studied for this. However,
the"chip kidney"process is in the development phase.
6.2.2.7 Chloride Removal Processes
Even a modest recycle of the Eop stage effluent to the brown stock washing has been found to
increase the white liquor sodium chloride concentration. An example is provided in Figure 6.9 from a
mill that tried the recycle for an extended period of time. At a level of 10%Eop filtrate recycle,the
sodium chloride concentration increased by about 4 g/l NaCl (2.5 g CIA).
NaCl in White Liquor vs. Eop filtrate recycled
12 12
510 --- -- -•-----}----- - ---- 10 c
•
o r • • o
= 8 ------- ------ ---- - • - w
• • m
r ; 4) W
4
CLO
2 O
Z �___ _ __________ ________r________r_______ W
i
i
0 0
200 400 600 800 1000 1200
Days of recycle
• NaCl in WL Eop filtrate — 91 ]EIsoivo
A Dawl.,C...
Figure 6.9 Impact of a Modest Recycle of Eop Filtrate on NaCl Content of
White Liquor
National Council for Air and Stream Improvement

Interim Report 43
The tolerable level of chloride in the liquor loop depends on the mill and equipment and recovery
boiler operating conditions. Typical maximum allowable chloride levels are
• 2-3 g/l chloride in white liquor;
• 0.2-0.3 %chloride of black liquor solids; and
• 1.0-1.7%chloride in the recovery boiler ESP dust.
Recirculation of a significant amount of the Eop filtrate to brown stock cannot be tolerated without the
installation of a chloride removal system.
There are several patented processes for the removal of chloride from the kraft pulp mill. Table 6.2
shows a list of ESP ash treatment processes for chloride removal, the expected chloride and potassium
removal efficiencies, and the expected sulfate and carbonate recovery efficiency. The efficiencies are
defined as percent removed or recovered from the precipitator ash.
Chloride removal processes currently in operation in full scale include leaching processes(e.g., in
Aracruz, Brazil)and crystallization processes (e.g., in Blue Ridge and Soporcel's Figueira da Fez mill
in Portugal).
Table 6.2 ESP Ash Treatment Processes for Chloride and Potassium Removal
Process Chloride Potassium Sulfur Sodium
Process,Developer Removal Removal Recovery Recovery
Efficiency Efficiency' Efficiency' Efficiency'
CRP(Sterling Chemicals), Crystalli- 85-90% 80-90% 85—90% —80%
ARC(Andritz), zation
PDR(Eka Chemicals)
PDP,Paprican Ion >92% --2% —98% —80-90%b
Exchange
ALE(Andritz) Leaching- 80% 80% —70% 70%
filter
Leaching(Kvaemcr/Metso) Leaching- 90% 82% 68% 63%
centrifuge
Bipolar Membrane Membrane 60% low 95%+
Electrodialysis(BME),
Paprican
Aqueous-Aqueous 60% 90%
Separation of Chloride w/
Polyethylene Glycol
'efficiencies defined as percent removed or recovered from treated ash
c depending of the ratio K/Na in the ESP dust
Conceptually,the leaching processes involve less equipment and process operations than e.g.,
crystallizing and ion exchange processes,but on the other hand experience higher losses of sulfur and
sodium.
The ion exchange process removes chloride selectively. Potassium and sodium are purged according
to their molar ratio in the ESP dust,which leads to very low removal of potassium. This may
beneficial, but may also be an undesired feature in mills which experience build-up of potassium in
National Council for Air and Stream Improvement

44 Interim Report
their liquor loop. The plugging and scaling of recovery boiler heating surfaces are, however,more
related to chloride than to potassium in the recovery boiler carryover and in the ESP fine dust.
Another difference between the ion exchange process and the crystallizer or leaching processes is that
the cleaned dust stays in solution(around 28%) in the ion exchange process and therefore has to be
evaporated to firing solids concentration when returned to the black liquor system. This consumes
steam for evaporation.
Again,the selection of the optimal chloride removal process becomes a mill-specific issue depending
on the equipment, recovery boiler conditions and sulfur/sodium/chloride/potassium balances, energy
prices, etc.
Crystallizing process
The concept for the crystallizing process used at Blue Ridge is shown in Figure 6.10. Dust from the
recovery boiler ESP is dissolved in water to a concentration of about 26%solids(corresponding to
85%saturation of sodium sulfate) and stored in a pre-evaporator feed tank. The solution is
crystallized in a vacuum evaporator/crystallizer, separated in a hydroclone, and dewatered on a
vacuum rotary drum filter. Condensate from the evaporator is returned to the ESP to dissolve the ash.
The concentration of the slurry leaving the crystallizer is maintained below the saturation point of
potassium sulfate and potassium chloride. It is pumped to a hydroclone to separate the sodium sulfate
crystals,which are dewatered on a rotary vacuum drum filter and dropped to an agitated black liquor
slurry tank. The black liquor slurry at high consistency is returned to heavy liquor storage. The
solution leaving the hydroclone(containing fine crystals) is returned to the crystallizer. A portion of
the filtrate from the vacuum drum is sewered to control the chloride and potassium levels in the
liquor. The rest is pumped back to the precipitator ash-dissolving tank.
The concept of the ARC process by Andritz and the PDR process by Eco-Tech utilize the same unit
operations, i.e., dissolving and recrystalizing of the ESP dust followed by washing and dewatering of
the crystals on a filter.
Ion Exchange Process
The ion exchange process, PDP, developed by Paprican and marketed by Noram Engineering and
Eco-Tech is illustrated in Figure 6.11 (Noram 2006).
The dust is first dissolved in warm water in an agitated tank, forming a nearly saturated sodium
sulfate/sodium carbonate solution(28%) with chloride and potassium. There is an insoluble portion
that must be removed since suspended solids tend to clog the ion exchange resin bed. Pulse filtration
(Eco-PulseTM Pressure Filter)is used to separate the suspended solids. The filter elements of the
pressure filter are periodically cleaned of insoluble solids using a short, low-pressure back-pulse.
During this sequence,the flow of liquor to the Pulse Filter is stopped and recycled back to the
dissolving tank(Brown et al. 1998).
National Council for Air and Stream Improvement

Interim Report 45
Chloride Removal Process - CRP
Condense
• Diagram of the Process
for Removing Chloride Preclphatar
and Potassium Ash Crystallizer
from the Kraft Rester
Recovery Boiler q
Wash
Black
Strong F
Liquor—
Saltcake 9allouke
Filter
Mix
To Recovery Tank
Roller
Chloride and
Potassium
Caron, 1995 to Disposal
Figure 6.10 Crystallization CRP Process (Blue Ridge)
PDP Precipitator Dust Purification
System
Water
Water
ESP NaCl
Dust Waste
.0_
NPE to black liquor
Sludge evaporators
M- Z,otvl FWKed Sulfate
ank
an
Figure 6.11 Ion Exchange Process (Noram 2006)
National Council for Air and Stream Improvement

46 Interim Report
The salt separation unit(SSU) utilizes short bed/water regeneration ion exchange technology.The
chloride is removed with an ion exchange process utilizing a short bed column. The fact that the ion
exchange occurs in a very narrow zone makes it possible to reduce the length of the column to just
longer than the zone. Using fine mesh resins allows this zone to be reduced further.A special
amphoteric ion exchange resin with both cationic and anionic exchange groups on each resin particle
is used. It allows both cations and anions to be removed simultaneously.The special resin also allows
sodium chloride to be eluted from the process with water,a key feature in the economic feasibility of
the process. The process involves an upstroke and downstroke in the column. In the upstroke,the
solution to be treated containing a mixture of sodium chloride and other chemicals is introduced into
the column. Sodium chloride is picked up by the resin. The cleaned solution is collected from the top
of the column. On the downstroke,water is passed countercurrently through the column. The resin is
regenerated by elution of the sodium chloride from the resin. Waste sodium chloride is collected
from the bottom of the column and the process is repeated. The waste solution, containing mainly
sodium chloride, is discharged. The cleaned solution of the ESP dust has a concentration of about
28%solids and is recirculated to the evaporation plant. Stem is thus needed to concentrate the dust to
firing solids concentration(Brown et al. 1998).
This process can be applied to several other chloride-bearing streams such as the oxidized white
liquor and the spent scrubber liquor.It is reportedly possible to remove more than 90%of chloride
while retaining over 90%of valuable components.
Leaching Processes
Ash leaching is a simple process schematically shown in Figure 6.12. The ESP dust is diluted with
water to about 28%consistency. After that the solution, in which the chloride and potassium are
enriched,is separated from the solids fraction by filtering(Andritz)or by centrifuge(Kvaemer).
The solids fraction at about 90%consistency is returned to the recovery boiler mix tank.The filtrate
or the centrifugate is sewered for purging the chlorides and potassium.
National Council for Air and Stream Improvement

Interim Report 47
ESP dust Water
Treated ESP
CTPuroe dust to mix tank,
@ 90%solids
Figure 6.12 ESP Dust by Leaching
6.2.3 Recycle of Chloride-Free Bleach Filtrates
Recovery of alkaline filtrates is more easily implemented where the preceding bleach stages do not
use chlorine dioxide. Mills that operate TCF or ECF-light sequences recycle some or all of the
alkaline filtrate to brown stock. Regardless of the sequence(ECF, TCF, ECF-light, etc.),the issues
with non-process elements present in the acid filtrates remain.
6.2.3.1 Lrternational Paper,Franklin, VA
The International Paper mill at Franklin, Virginia operates a high consistency ozone stage,the ZEtrac
process in an ECF-light sequence O(A)ZED on pine pulp. The mill recycles the filtrates from the Z
and the E stages according to the concept shown in Figure 6.13 (Griggs 1997).
National Council for Air and Stream Improvement

48 Interim Report
WASH WATER
Z ........... g ........... p ......�
POST O, ACID
n9 �RESS—
TO BROWN-
STOCK
WASHING
2.1 m'/ADt 7.3 m'/ADt
Figure 6.13 Bleach Plant Filtrate Recycle at International Paper, Franklin, VA(Griggs 1997)
The last wash stage after the oxygen delignification stage is a wash press where Z-stage filtrate is
used as wash water. Following that press the pulp is pressed again in an"acid press." Some water
(Z-stage filtrate or water) is obviously added between the presses to allow a discharge of 2.1 m3/ADt
of acid filtrate from the acid press to sewer. Based on the water balance data indicated in Figure 6.13,
most of the"acid press"filtrate was obviously sewered together with non-process elements removed
in the"acid press." Clean water is used as wash water on the E-stage and D-stage. All filtrate from
the D-stage(7.3 m3/ADt)is sewered, as well as the filtrate from the"acid press",2.1 m3/ADt. Thus,
in this scheme, the problems with chloride and other non-process element build-up have been
significantly reduced by sewering the D-stage and the "acid press"stage filtrates.
The literature documents the following performance data for the ZEtrac F-bleach line.
Table 6.3 Information about the IP(Franklin,VA)Mill Closed Cycle Bleach Plant
Griggs 1997 Bicknell et al. 1995
m3/t kg/t Ib/T m3/t kg/t Ib/T
Flow 9.4 11
AOX 0.05 0.10 0.085 0.16
BOD 4.4 8.8 6.94 14.0
COD 11 22 11 22
Color 3.1 6.2 5.75 11.5
6.2.3.2 Mead-Bolnia,Rauma Mill,Finland
The Metsa-Botnia mill in Rauma, Finland operates the only Finnish 100%TCF mill with the bleach
sequence(ZQ) (PO)(ZQ)(PO)(PO). According to a recent announcement,the mill will start
producing ECF pulp also in 2007 (Brantestrom 2006). Figure 6.14 shows the filtrate recovery
concept is conceptually similar as at IP,Franklin F-line: alkaline filtrate is recovered to brown stock
washing and acid filtrate is purged to the sewer.
National Council for Air and Stream Improvement

Interim Report 49
The Metsa-Rauma mill has a joint effluent treatment plant with UPM-Kymmene's mechanical pulp
and paper mill on the same site. The Metsa-Botnia mill reported its share of the total effluent
discharges after biological treatment as follows (Metsa-Botnia 2006):
Effluent flow 19.6 m3/t
BOD 0.16 kg/t 0.32 Ib/T
COD 4.3 kg/t 8.6 Ib/T
Color is not reported by the Metsa-Rauma mill.
Metsil-Botnia-Rauma Bleach Filtrate Recycle
White water
Wash < -- �--� Condensate
yl 1
li}{LYt�I`: putt x.ZV�4' POI06,
7M31t 1 112 Stage DD Fractional Washing
Matra Rauma,eteeeb Rec)cO,Seplember 1999
Figure 6.14 Bleach Filtrate Recovery at Metsa-Rauma Mill(Gleadow et al. 2003)
6.2.3.3 Slvedisk TCFMills
Two of the Sbdra mills in Sweden(Varb and Monsteras) have converted to 100%TCF operation.The
Monsteras mill studied closed cycle technology extensively in the late 1990s, but concluded that it
was not successful. The mill decided to install biological effluent treatment to meet the COD limit
(S6dra Cell 2002).
The recovery of the acid filtrate was studied at the Varb mill using e.g.,the Netfloc process(Kemira)
as a non-process element removal "kidney."These studies were not continued after the Swedish EPA
ordered the mill to build a biological effluent treatment plant. The mill had originally planned to
implement closed cycle operation.
All three S6dra mills currently operate very efficient biological treatment plants. Unfortunately,
effluent color is not reported by any of these mills, as color monitoring is not required in the Nordic
countries.
The SCA mill in Ostrand,Sweden has a ZEtrac bleach plant. The bleaching sequence was reported to
be O Q OP ZQ OP (SCA 1996). Kappa for softwood pulp from the digester was 23 and from the Oz-
stage 12. The brightness of the pulp is reported to be 90% ISO and viscosity 560-570 dm3/kg. The
bleach plant effluent was reported to be 7 m3/ADt.The COD discharge from the bleach plant has been
reported to be 20-25 kg/ADt. Data for effluent color is not available. Precipitation of metals in the
National Council for Air and Stream Improvement

50 Interim Report
first acid bleaching stage has been reported as a problem when closing up the plant(370). When the
water system is closed to a certain degree,calcium oxalate and gypsum start to precipitate in the
piping system. When this happens the mill has to open the water system. To improve the metal
management, the mill was going to install a green liquor filter and a new dregs filter. Recent
information suggests "a totally closed bleach plant would be a very difficult target to reach."The mill
has therefore installed efficient biological treatment(Rodden 2005b).
6.2.4 Laboratory Simulation ofAlkalfne and Acid Filtrate Recovery
A laboratory simulation was carried out for the recycling of both acid and alkaline stage effluents for
the Ahel(EOP)D(PO)bleach sequence according to the concept shown in Figure 6.15 (Colodette et al.
2002). In this concept both unbleached and bleached pulp were assumed to be pressed to high
consistency to allow a tight water balance.The simulation was carried out on oxygen delignified
eucalyptus pulp with a kappa number of 8.3 and a carryover to bleach plant of 10.2 kg COD/t.
O-Prom Aa„-Vacuum Filter (EOP)-Vacuum Filter D-Vacuum Filter (PO)-Press
CF-1.67 W=1.67 DF- 1.67 DFv 1.67 DF- 1.67
t-o (MP) 9A (4 CC a 5 SOP) 9.0 (d CC-5 YM9 9.0 (PC) 4.0 10.7 (VAV)
I
1 Csly 70 Caly 73 2 Calm 12 Caty 10
2d1 7.77 7.37 7d7 2.3]
lu
x
Csl= 1.$ 3 Cst. 7.5 _ Caty 1.5 Col• 4
65.7 65.7 65.7 2a0 ,';
6.67 167 1.67 1.8T 1'
i Cat- 10 Cat- 10 Cat• 10 ,Cst= to Col- to
6.67
I
56.7 67Z 567 67.3 56.7 67.3 15.0 25.7
I
Rocauallolzing 4.0 to(193P) 9.0 to D
3.89 10.7 10.7 10.7 107
5.0 to A„
I
0.11 9.0
Filtralo.mlit
Who 6ew0r 1311uont 9.11
Consis to nry(CaQ=% Roraustleixing@iluont 3.89
Volume•m�ll Prd4)a Waohing 4.00
dlutlon Factor(OF)-Wit While Walor(WW) 9.00
Clean Condonseto(CC) 8.00
Figure I -A6o,(EOP)D(PO)Bleaching Filtrate Strategy,
Figure 6.15 Laboratory Simulation of Alkaline and Acid Filtrate Recovery(Colodette et al. 2002)
National Council for Air and Stream Improvement

Interim Report 51
In the study, part of the AHo,stage effluent was returned to the recausticizing cycle to be used for lime
mud and dregs washing. The minerals in the acid filtrate would be discharged mainly with green
liquor dregs. The Eop stage effluent was recycled to the post OZ delignification wash press and the
AHo,drum washer. Most of the D-stage filtrate was sewered.
The use of the acid filtrate from the acid hydrolysis stage on the post-oxygen washer was not
considered viable for the following reasons:
• increased oxidized white liquor consumption in the oxygen delignification;
• risk for lignin precipitation;
• increased calcium oxalate formation in the oxygen delignification;
• accumulation of heavy metals possible leading to a selectivity loss in the oxygen
delignification; and
• contamination of the pulp with by-products from the hydrolysis.
With the acid hydrolysis filtrate used as a water source in the white liquor making,the color in the
total bleach plant effluent would be 2 lb/T. With the acid filtrate going to sewer,the total bleach
plant effluent color was about 12 lb/T. The chloride-containing effluent would not be recycled, so a
chloride removal process would not be necessary.
The application of the acid hydrolysis filtrate as make-up water in the recausticizing was not
simulated. This concept represents new technology and poses several challenges.
• The water balance of the recausticizing area has to be able to receive the make-up water, i.e.,
the lime kiln should have an electrostatic precipitator rather than a scrubber,which would
supply most of the water make-up to the weak wash.
• The filtrate has to be filtered to remove fibers.
• The non-process elements (NPEs)introduced with the acid filtrate will increase the NPE
content of the lime mud and green liquor. NPEs will likely accumulate in the lime mud,thus
impacting the kiln and recausticizing operations(Colodette et al.2002)
• Potential odor and TRS emission issues may arise from using an acid filtrate for water make-
up in the recausticizing area.
• Increased formation of dead load compounds may result from the sulfur present in the acid
filtrate.
In conclusion, the proposed laboratory concept represents an interesting opportunity to further reduce
color but generates several unanswered questions.
6.2.5 ECFBleach Plant Closure by Eka Chemicals(Ultrafiltration)
Eka Chemicals has studied a concept for partial bleach plant closure(PC'Tm),which is illustrated in
Figure 6.16. Results from operating the process in pilot scale in a Scandinavian mill were reported
(Bryant et al. 1998).
It was concluded that the OF and partial closure of the alkaline filtrates can reduce the COD by as
much as 60% and the AOX by as much as 40%. Since the permeate, now containing mainly low
molecular weight material, is taken to biological treatment,the efficiency of the biological treatment
increases.
National Council for Air and Stream Improvement

52 Interim Report
The Eka Chemicals' concept was tested extensively in pilot plant scale at Stora Norrsundet mill,
where combined Eop and EP filtrates from sequence D(Eop)D(EP)D were treated. The cut-off was
4000 Dalton and the flux rate was about>180 I/m3 h.
UNBLEACHED BLEACHED
CHIPS DIGESTER PULP BLEACH PULP
BROWN STOCK AREA PLANT
WHITE BLACK ALKALINE
LIQUOR LIQUOR FILTRATE
RECOVERY MEMBRANE UNIT
RECAUST AREA INE,AFIItTRATION} ACID
EFFLUENT
OF PERMEA
TREATED
ESP DUST
ClandK EFFLUENT
Removal TREATMENT
SYSTEM
CHLORIDE TREATED
SOLUTION EFFLUENT
Figure 6.16 EICA Nobel Partial Bleach Plant Closure (PC)
The bleach plant was not operated as a closed cycle but only a small fraction of the effluent was
filtered in the membrane unit. The permeate combined with the acid filtrate was treated filrther in a
pilot biological treatment unit.
The ultrafiltration of the alkaline effluents reduced the bleach plant loads and improved the treatment
efficiency of the biological treatment as shown in Table 648.
Table 6.4 Pilot Plant Results of Eka Chemicals Ultraftltration Concept(Bryant et al. 1998)
Bleach Plant Discharge Removal in Biological Treatment
1 2 1 2
kg/ADt(lb/ADT) kg/ADt(lb/ADT) % %
AOX 0.52(1.04) 0.41 (0.82) 69.9 76.9
COD 27.3(54.6) 15.3 (30.6) 60.1 77.6
TOC 11.1 (22.2) 6.7(13.4) 64.3 78.6
Color 14.4(28.4) 5.5(11.0)
BOD7 8.4(16.8) 5.2(10.4) >96 >96
NOTES: 1 =Untreated acid filtrate+untreated alkaline filtrate
2=Untreated acid filtrate+OF treated alkaline filtrate
It was concluded that the OF improved the biological treatability considerably.
A simulation of the application of Eka Chemicals PC'process was carried out for the Leaf River Pulp
Mill (Herstad et al. 1998, 1999). The simulation included both a OF process for the alkaline filtrate
National Council for Air and Stream Improvement

Interim Report 53
and a chloride/potassium removal process for the recovery boiler ash. Different points of adding the
OF concentrate to the fiber line were tested. Table 6.5 summarizes the main data for the bleach plant
effluent at different points of adding the OF concentrate in the fiber line, compared to the base case
(no ultrafiltration, or recycle).
Table 6.5 Simulated(GEMS) Impact of PC'on Bleach Plant(Herstad et al. 1999)
Unit COD Color AOX
Base process,no UF,no closure kg/ADt(lb/ADT) 53.4(106.8) 46(92) 0.59(1.18)
OF concentrate to post Oz-stage kg/ADt(Ib/ADT) 40 (80) N/A 0.55 1.1
decker bottom showers ( )
OF concentrate to pre OZ-stage brownstock washer kg/ADt(lb/ADT) 35 (70) 20(40) 0.48 (0.96)
OF concentrate to weak black liquor(A) kg/ADt(lb/ADT) 34(68) 20(40) 0.46(0.92)
Based on the simulation, about 1 kg of chloride/tp(21b/ADT)would be returned to the liquor cycle
with the OF concentrate. The application of the Eka Chemicals chloride removal process would
eliminate any chloride buildup in the cycle, and actually result in less chloride in the cycle than the
base case mill. Based on these simulations, it was concluded that the Eka Chemicals PC'process, if
applied at Leaf River mill
• would reduce bleach plant effluent AOX and COD after effluent treatment by 33%and 53%
respectively (20%and 37%before treatment). Color would be reduced by 60%.
• would not change evaporation loading. Total solids to the RB would be about the same
because less ESP dust is recirculated to heavy black liquor.
• would decrease the chloride and potassium in the ESP dust 54%and 89%respectively with
partial closure and chloride removal.
• would increase the bleach plant chlorine dioxide use by around 1 kg/t while the sodium
make-up requirement would decrease by about 2.9 kg NaOI Ut(48).
The application of the membrane process at the two mills participating in the above studies has not
gone forward. The membrane technology is still developing. No applications for ECF or TCF bleach
plant filtrates are known to be in operation. The only full-scale operating membrane treatment for a
fiber line filtrate is at the Stora-Enso Nymolla mill in Sweden where the oxygen delignification stage
filtrate is treated. The Nymolla mill is a Mg based sulfite mill. The oxygen delignification stage uses
sodium hydroxide as the base. Therefore,joint recovery of the cooking and oxygen delignification
spent liquors is not possible. The installation at the Nymolla mill and the experiences with membrane
filtration is discussed in more detail in Section 7.
National Council for Air and Stream Improvement

54 Interim Report
7.0 COLOR REMOVAL—SEPARATION PROCESSES
7.1 Membrane Technologies
The following sections describe various membrane techmologies which could be used for treating
colored wastewaters. Depending on the pore size, the membrane processes are divided into
microfiltration(MF), ultraflltration(UF), nanofrtration(NF), and reverse osmosis (RO). The pore
size or cut-off value of the membranes in these processes decreases in named order. A membrane
with a higher cut-off value has a higher filtration capacity but generally lower removal efficiency.
Figure 7.1 gives a comparison of membrane sizes.
Separation M
Process ,
rt Milk, ratelns E CQam men t eron'eCd
ReLttmi Gelagn Red Blood CwJls Fatilllce0es
StEeCf Motalbn Baciefla
Entl0loxln.P lopen: gp�yatod
CoutRlcm 011JEMUIS1 n carbon„ d
fihafCit'fal syna ua virus:
Dyes Ind[p Dya- cryptosporhllum
..Lactose ': .`C0liddsl. G 0 Huniao'Ha1F
auli�l.. $itICA t
Mtcrans 0.001 o.qI , as 1.0 10 10o tooD
`t t too 20D 1,000 2i1,coo 100.000 500.000
Vr@d0h t,R114 b&11if �,
Nota:1 micron'(m[cronwtorl04xli)5!ches•1%104Angstrartiunits' +920W=Kodi0.tam6ryiieSystems
Figure 7.1 Membrane Size Classifications(from Koch Membrane Systems)
Because of the high membrane cost and their limited flux capacity,the economical feasibility of the
process is directly proportional to the volumetric flow. Therefore,of the kraft mill effluents, the Eop
filtrate with a low volume and high concentration is typically considered most feasible for treatment
with membrane separation. Treatment of total mill effluent would be extremely costly.
7.1.1 Microfrttration (MF)
Microfrltration has the largest pore size of any of the membrane technologies discussed above,
approximately 0.03 to 10 microns and a molecular weight cut-off of greater than 100,000 daltons.
The typical operating pressures for MF systems are 10 to 100 psi. Microfiltration membranes usually
come in two flow types: cross-flow and dead-end flow. With cross-flow membranes,the fluid runs
parallel to the membrane and passes through the membrane based on the pressure differential. In
dead-end flow,the fluid flows perpendicular to the membrane. Because of its large pore size, MF is
often used as a pre-treatment step for reverse osmosis and nanofrltration. To prevent fouling, frequent
National Council for Air and Stream Improvement

Interim Report 55
backwashing aids in keeping the membrane clean and helps maintain membrane flux. Microfiltration
can also be used with biological treatment in membrane bioreactors. Most mill-scale implementations
of microfiltration are for the recovery and reuse of internal water streams(coating kitchen effluent)or
as pre-treatment stages for other membrane processes.
7.1.2 Ultrafiltration (UF)
Membrane filtration is a separation method that separates dissolved molecules based on their size and
charge. High molecular weight substances such as lignin compounds, which are the primary sources
of color and COD in effluents, can be concentrated in a smaller stream for separate treatment or
destruction. The permeate, which contains low molecular weight components(mainly water), is
sewered or may be reused in the mill to some extent.
OF is a pressure-driven process where colloids,particulates,and high molecular mass soluble
compounds are retained by size exclusion. OF has a pore size of approximately 0.005 to 0.1 microns
and a molecular weight cut off of 10,000 to 100,000 daltons. The typical operating pressures for OF
systems are 5 to 150 psi. The OF membranes are generally manufactured in either flat-sheet or
tubular form. The main system components include the membrane units,pumps, and cleaning
system.
Ultrafiltration is considered proven technology but it has not gained acceptance in the pulp industry.
The OF or RO treatment processes would require extensive pre-treatment processes for solids
removal, which would most likely include coagulation and filtration. This would increase the
equipment cost and land required for these treatment processes. There are no known installations of
kraft mills using OF to treat the whole mill effluent or bleach plant, although OF is used in paper
mills to treat the white water in the paper mill circulation system and coating kitchen effluent.
Several risks related to membrane applications are discussed below.The modularity of the membrane
installation makes it more reliable than a single unit. Membrane fouling and abrasion gradually
diminish the flux, necessitating membrane replacement after some time. Typically, membrane
lifetime has been around 18 months. Short membrane lifetimes or unpredictable fouling tendencies
make a membrane unreliable.
Recirculation of membrane concentrate into the liquor loop or reuse of the permeate may have a
considerable impact on liquor quality and production processes. Some impacts might be increased
ash and calcium content of pulp and increased chloride concentration in the black liquor. The
increased chloride content in the liquor could lead to recovery boiler fouling. Return of the
concentrate to the weak black liquor would increase the load on evaporation and the recovery boiler,
possibly reducing production.
Impact of pH
It is well known that membrane filtration of acidic bleach plant filtrates results in severe fouling
problems, destroying the capacity of the membrane(Sierka, Cooper, and Pagoria 1996; STFI 2002).
This is explained by the fact that organic acids(e.g., fatty acids)are protonized in acidic filtrates and
strongly adsorbed onto the membrane. Therefore, ultrafiltration of the acidic bleach plant effluents
would not be practical.
Because of the high membrane cost and their limited flux capacity,the economical feasibility of the
process is directly proportional to the volumetric flow. Therefore, of the kraft mill effluents,the Eop
filtrate, which is alkaline and has a low volume and high concentration, has typically been considered
most feasible to treat with membrane separation. Depending on the bleaching conditions and the
kappa number, the first E stage (El)effluent can have a higher content of higher molecular weight
and colored organic compounds than the other bleach stages. Since the volume of this stream is much
National Council for Air and Stream Improvement

56 Interim Report
less than the total bleach plant effluent, it would be much more economical to treat while still
removing a large portion of the bleach plant COD,AOX, and color.
Flux
The surface load or flux(liters/tnZh)defines the dimensions of the plant. Typically, an expected flux
for OF would be 150-300 I/mZh, and for NF 50—70 I/mZh. Studies of EoP filtrates in an OD EoP DED
sequence confirmed that the flux on OF membranes increases with decreased incoming COD(STFI
2002). Another study found that the parameter with the largest impact on flux is conductivity, i.e.,
the concentration of the low molecular weight inorganic substances. The retention of organic
substances varied depending on the type of filtrate and the water management system of the mill.
A study where EoP filtrate was filtered through a OF membrane with 4000 D cut off found that the
flux varied between 160-200 1/mZh, depending on the applied water pressure, at volume reduction
factors at least up to 12(Bryant et al. 1998).
The Finnish Lappeenranta Technical University has studied the DOW 270 membrane, a rather
hydrophilic NF membrane with fairly high retention and permeability, in high shear filters called CR
filters. (Manttari,Nystrom, and Pekuri 2004). Metso PaperChem(OptiFilter CR)supplies the CR
filter process. On paper machine white water the flux rates of 80-100 I/mZh could be obtained at a
recovery below 70%. No corresponding studies on EoP filtrate using this specific membrane could be
found in the literature.
Fouling
The membrane technology is developing with respect to resistance to fouling. A study used
membranes made of polyether sulfone in a trial to treat the EoP filtrate from a D(Eop)DED sequence.
They found that membranes that were modified in order to minimize their hydrophobicity reacted
more positively to washing than unmodified membranes. They also concluded that a considerable
amount of the fouling caused by a cake or gel layer could be reduced if the shearing forces at the
membrane surfaces increased (STFI 2002).
Various physical or chemical methods can be employed to clean the membranes.
The Stora-Enso mill at Nymolla in Sweden has a OF plant with membranes from PCI in operation, to
treat the alkaline effluent from an oxygen delignification stage. Figure 7.2 shows the OF
configuration at the Nymolla mill. A study was performed to evaluate the cleaning and operation of
the ultrafiltration plant(Greaves 1999). Laboratory tests were carried out in order to find out what
the fouling layer consists of and to find the most appropriate cleaning procedure. Production data
were analyzed to establish if the membranes were cleaned unnecessarily often. It was found that the
fouling was mainly of organic origin and was seen as brown precipitation on the surface of the
membranes. Analysis of the membrane surface with energy dispersive X-rays (EDX)showed that
there was also some inorganic fouling, mostly silicon, iron, and magnesium. Cleaning tests were
carried out on a small laboratory test unit from PCI using fouled membranes from the mill. Both acid
and alkaline chemicals were used to clean the membranes. The current cleaning procedure was not
optimal and some improvements were suggested. Production data showed that the membranes were
often cleaned when there was no decline in flux. This leads to high wear on the membranes and
shortens their operational lifetime. Less frequent cleaning, especially during the first stages, of the
OF plant was therefore recommended. It was found that many of the membrane tubes were clogged
with a clay-like precipitation consisting of fibers, organic material and precipitated magnesium
hydroxide. This clay completely stopped the flow through the tube,thereby reducing the active
membrane area of the OF plant. The 2003 EMAS report issued by the Nymolla mill notes that the
National Council for Air and Stream Improvement

Interim Report 57
COD discharge of the mill increased because of bad performance of the OF plant due to plugged
membranes. The problems have reportedly been corrected.
WASH WATER
O Q P
UNBUUA DPULP
ISMD$ DCEACH
tma
w
UQUORSTO
CONCENTRATE Acnvt SLUDGE
MMMOILEN
Figure 7.2 Ultrafrltration of Oxygen Delignification Effluents at the Nyrnblla Sulfite Mill in Sweden
(Wickstr6m 1997)
One novel technique that has been studied in the literature is to use dissolved gas flotation to remove
the suspended solids and colloidal material prior to OF treatment(Pokhrel and Viraraghavan 2004).
In laboratory experiments,E] effluent was treated with both dissolved air flotation (DAF)and
dissolved COZ flotation followed by UF. The dissolved COZ flotation proved slightly better than
DAF, but both were effective in removing a large portion of the TSS in the El effluent. The dissolved
COz flotation reduced membrane fouling more than DAF.
Treatment Efficiencies
Studies have been carried out on EoP stage kraft bleach plant filtrates from oxygen delignified pulp
(Faith, Pfromm, and Sokal 2000; Bryant et al. 1998; EKONO 2006).
Table 7.1 summarizes the reduction efficiencies that have been measured when filtering the EoP
filtrate from softwood bleach plants,sequence DE(OP)DED.
National Council for Air and Stream Improvement

58 Interim Report
Table 7.1 Results of OF and NF of Pop Filtrates
OF OF OF+NF OF OF+NF
in series in series OF
Effluent type Eop Eop Eop Eop Eop Eop
filtrate filtrate filtrate filtrate filtrate
Bleach Sequence D(Eop)DED DEDED DEDED DEDED DEDED DeopDED
Filter/Membrane 1) 2) 2) 3) 3) 4)
Flux I/m2h 200-400 50- 80 200-400 50-80 160-200
Removal
Lignin 70%
COD 71% 55% 91% 26% 85% 50%
BOD 16% 28%
Conductivity 2% 38% 2% 15%
Color —75%
TOC 49% 90%
AOX 51% 96% 50%
DS 0% 0%
P 27% 93% 27% 91%
N 41% 68% 41% 65%
Monovalent Ions very low
Divalent Ions 70-80%
Na 8% 44% -2% 18%
Cl 1% 3% 0% 1%
K 14% 43% 0% 17%
Mn 67% 100% 100% 100%
Fe 100% 100% 100% 100%
SO4 3% 98% -13% 90%
NOTES: 1) polysulfone membranes
2) CR filters using ultrafiltration membranes C30,PS 100,PES259(cutoff 4000 D),and nanofiltration
membrane DESAL 5(cutoff 200-500 D). Tests with a low volume reduction factor,VRF=2-4(VRF=
feed flow/concentrate flow);
3) same as 2)but high volume reduction factor,VRF>7;and
4) PCI—filter with ES404 polyethersulfone membrane(cutoff 4000 D),VRF=6.
Based on Table 7.1, it can be concluded that OF reduced lignin and thereby the color of the filtrate by
70-75%. The COD and AOX were reduced by 50-55%. BOD,however,was only reduced 20-30%.
OF removes the high molecular weight colored compounds that are hard to biodegrade, leaving the
smaller, easily biodegradable compounds. Soluble chloride was not removed at all by UP,and
sodium very little, 0-8%. Divalent ions, e.g.,Fe and Mn, were also removed quantitatively.
When the OF permeate was treated with NF (or the Eop filtrate as such with NF),more organic
material was removed. COD,TOC, and AOX(and color)were removed to>90%. Soluble chloride,
being a very small ion, was still going through the membrane almost quantitatively, while the NF
membrane removed 20-40%of the sodium. Note that AOX(organic halogen)was retained in the
concentrate.
Thus, if OF or OF+NF or NF were used,the soluble chloride concentration in the feed,permeate,
and concentrate would be about the same. The amount of chloride recovered would be determined by
National Council for Air and Stream Improvement

Interim Report 59
the amount of concentrate, i.e., the volume reduction factor and the chloride attached to the high
molecular mass as AOX.
One literature source concludes that the optimum VRF may be between 5-7 or 80-85%permeate of
the feed(Bryant et al. 1998). If a larger fraction is removed as permeate, the cost of removal
increases due to lower flux. On the other hand, more chloride is recovered with the concentrate at
lower permeate fractions. This impacts the recovery cycle.
The most viable application of a membrane process is to treat the EoP filtrate by using a OF type of
membrane. However,the membrane pore size must be established in lab and pilot tests. Figure 7.3
shows a possible configuration for OF treating bleach plant effluent.
F
OF
Unit
To OF OF TO
Brownstock Concentrate Permeate Waste
Fiherllne Treatment
Figure 7.3 Ultrafiltration of Alkaline Filtrates from an ECF Bleach Plant(Herstad et al. 1998)
The permeate,which contains most of the original inorganic dissolved compounds(e.g., chlorides)as
well as about half of the organics, would be directed to the treatment plant.
Separate incineration and evaporation of the concentrates would be possible,but it is more likely that
the OF concentrate would be returned to the liquor loop for evaporation and incineration. To prevent
unacceptable build-up of chlorides in the white liquor, it would most likely be necessary to purge
chloride in a chloride removal process.
Treatment of bleach plant EoP filtrate by full-scale ultrafiltration processes has been applied in at least
two Japanese mills and one Swedish kraft mill for a number of years. To our knowledge, all three of
these ultrafiltration installations were discontinued after several years of operation.
Although ultrafiltration is considered proven technology it has not gained acceptance in the pulp
industry.
Table 7.2 summarizes the information known about full-scale installations in chemical pulp mills
(kraft and sulphite). The only OF plant that is known to be operating on effluent in a pulp mill is the
Stora-Enso sulfite mill in Nymolla, Sweden, where oxygen delignification effluent is treated.
A sulphite mill in Norway operates a OF plant on fermented spent sulphite liquor prior to vanillin
production.
National Council for Air and Stream Improvement

Table 7.2 Information about Full-Scale OF Plants
Supplier,
Mill Treated stream surface area, Feed Cut-off Flux Reduction
oper.period, m3/d gpm Daltons Um2/h VRF Achieved
status
Borregaard Industries Calcium sulphite PCI Remove low m01.
Sarpsborg sulphite mill liquor,after fermentation 1200 220 20,000 material prior to
Norway 1984...... the vanillin
production unit
MoDo,Domsjo Pre-bleach caustic PCI
Na-based sulphite mill deresination stage effluent 728 m2 2880 530 100,000 150 23.5 COD 70%
Sweden 1985- 1988 pitch 80-85%
shut down
MoDo,Husum About 40%of E, filtrate in Flootek COD 50%
bleached kraft mill O(CD)(EO)D bleaching 200 m2 1200 220 25,000 200-250 15 AOX 50%
Sweden 1989- 1993 BOD7 35%
shutdown Na 20%
Cl 7%
Stora+Enso,Nymolla Oxygen delignif filtrate PCI COD 46-55%
bleached sulphite mill prior to bleaching 4650 m2 7200 1320 4,000 70 SW 50 Na 5%
Sweden (2lines: SW,HW) 1995-.... 87 HW TOC 50%
in operation
Sanyo-Kokusaku Pulp Co. One-third of El stage Ultrasep
bleached kraft mill effluent 672 m2 900 165 47-66 83 COD 82%
Iwakuni,Japan 1981—unknown Color 94%
shut down
Taio Pulp Company El stage effluent Nitto-Denko
bleached kraft mill 1480 m2 1740 320 8,000 98 16.5 COD 79%
Japan 1990—unknown
shut down

Interim Report 61
7.1.3 Nanofillration (NF)
NF has a pore size of approximately 0.001 microns and a molecular weight cut-off of 1,000 to 10,000
daltons. No full-scale NF plants are known to operate in kraft mills.
Nanofiltration has a lower permeate flux than ultrafiltration,which results in cleaner product,but NF
is more sensitive to fouling and must be preceded by a pre-treatment step. The pre-treatment step is
often ultrafiltration or a biological process.
Generally,NF has been conducted with conventional spiral wound membranes, but recently high-
shear cross-rotational (CR) membranes have been developed. CR membranes are more resistant to
fouling and may not require the rigorous pre-treatment step needed by spiral wound membranes. In
CR filters, the high-shear or turbulence is generated on the surface of the membrane by a rotating
blade. Laboratory tests for a high-shear CR membrane(named NF270)were conducted using the
effluent from an activated sludge plant in a mechanical pulp and paper mill (Manttari,Nystrom, and
Pekuri 2004). At volume reduction factors of 2 and 13,the NF270 membrane retained about 85%of
the TDC, 60% of the conductivity,and most of the color(see Figure 7.4). Performance was enhanced
by neutral pH conditions and temperatures below 45'C. NF appears better at removing color than
UF, because salts pass through the OF membranes.
Like the other membrane technologies,NF could be used to treat the whole mill effluent. However,to
treat the whole mill effluent, a pre-treatment step would be required to remove suspended solids to
prevent membrane fouling. Also, using NF would create a highly concentrated stream that would
need to be disposed of. There are no known installations of kraft mills using NF to treat the whole
mill effluent.
Conductiviry
l0 10000 Calaur
$ •
• e 1000
6 V
Z • n'
100
4
• e u
0 2 • t0 °
90° 00 0 0
0 1
0 5 10 15 0 5 10 IS
(a) VRF,- (b) VRF,-
300 TIC 1200 TOC
250 1000
• •
200
800
B 150 E 600 •
V • ° ° Ci •
t- 100 • F 400
e
so °0 0 ° ° 200 •s
0
00 5 10 15 0 5 l0 15
(c) VRFr (d) VRF,-
Fig.6.Conductivity(a),colour(b),total dissolved inorganic carbon(c)and total dissolved organic carbon(d).concentration in the concentate and the
permeate streams as a function of volume reduction factor(VRF)in naaofiltration of discharge water of an activated sludge plant((•)concentrate,(0)
TFC ULP rnembrane and(e)NF270 membrane).
Figure 7.4 Color Removal Using NF270 Membrane(Manttari,Nystrom, and Pekuri 2004)
National Council for Air and Stream Improvement

62 Interim Report
7.1.4 Reverse Osmosis(RO)
Reverse osmosis is similar to ultrafiltration in that the effluent is treated by passing it through a
membrane that rejects molecules that are larger than the pore size. The difference is that in reverse
osmosis the pore size is much smaller, with the result that high-pressures (10-100 Bar)must be used
to force the effluent through the membrane.
Reverse osmosis is used,for example, in desalinization of seawater and has the potential to remove
almost all impurities and produce clean water for reuse.
To treat the whole mill effluent, a pre-treatment step would be required to remove suspended solids to
prevent membrane fouling. There are no known installations of kraft mills using RO to treat the
whole mill effluent or colored bleach plant effluents. RO has been used in sulfite mills for water
reuse and spent liquor treatment.
In 1998,the Irving pulp and paper mill in Saint John,New Brunswick, started up an RO system for
the treatment of condensates from the 5th evaporation effect(Dube et al. 1999). The mill is a
bleached kraft mill that produces about 900 tons/day of market pulp. The mill did not have secondary
treatment and instead used in-process measures to meet its environmental requirements. The RO
treatment of the condensate from the 5th evaporator effect was implemented to reduce the effluent
toxicity,which was not removed by condensate stripping. The RO treatment removed about 89%of
the COD from the stream and about 88%of the BOD. No data were given for color removal. The
clean permeate is returned and used as wash water on the No. 2 post oxygen delignification washer
and the concentrate (about 1% of the flow)is either burnt in the bark boiler or sent to the high solids
crystallizer and eventually burnt in the recovery boiler.
7.1.5 Membrane Bforeactors
The membrane bioreactor(MBR)is a hybrid biological treatment system that combines the biological
activity of a free-floating reactor,such as an AST,with the advantage of membrane separation to
achieve reductions in suspended solids,COD,BOD, and toxicity. In this process,membranes are
placed in an aerated biological reactor. Effluent is fed to the reactor,where it is biologically
degraded. The effluent passes through the membrane and this permeate is discharged. The
accumulated sludge is then withdrawn for disposal. The advantages of a membrane bioreactor
include higher loading, smaller size, insensitivity to sludge settling characteristics, and effective
removal of solids, COD,and toxicity. In addition,the sludge retention time(SRT) and the hydraulic
retention time(HRT) can be varied independently. Disadvantages include membrane life span,
restricted flux rates, and initial investment cost.
In one pilot study, an MBR treating CTMF interstage refining effluent and operating with a hydraulic
retention time of 36 hours and a sludge retention time of 15 days achieved removals of 68%, 87%,
and 98%for COD, BOD, and suspended solids, respectively(Dufresne et al 1998). Color removal
was not reported. Figure 7.5 shows the configuration used for this study.
National Council for Air and Stream Improvement

Interim Report 63
1
2 a
3
I
ir
II
T AL
r
i
i
4 i
+t
r
' r i
aaoe�o �
t
S i
i r r
a t 6
S ooa a now _ No
1
I
I
2.Rhematic dlogmm of the hollow nherintemal membrane bioreactor.1.prmtum
gauge,L vacuum pump 3.flod for level controt 4.membranes,S.aemters,&sludge
waste,1.backwmhlntr 8 wastewater entry,9 air entry,10.effluent sewage
Figure 7.5 Membrane Bioreactor(Dufresne et al. 1998)
Laboratory studies of the treatment of secondary condensates in a bioreactor, consisting of a reactor
and a ceramic tabular OF membrane,were carried out at the Western Pulp Ltd.Kraft mill in
Squamish, British Columbia. The reactor was operated at 60oC and fed with evaporator condensate.
The removal efficiencies were 99%for methanol, 91%for TOC,and 99%for reduced sulfur. Color
was not reported(Bdrub6 2000a,2000b,2001).
A membrane bioreactor combines both biological treatment and membrane separation. The system
consists of a reactor tank,ultrafiltration membrane, and aeration system. The reactor tank can be
filled with packing media, which serve as a carrier for biological growth. In this process,the
untreated effluent is fed to an aerated reactor tank with the appropriate hydraulic retention time
(HRT).A side stream is then withdrawn and passed through an ultrafiltration membrane. The
permeate from the membrane is discharged as treated effluent and the concentrate from the membrane
is returned to the reactor. A portion or all of the permeate from the membrane can be recycled to the
reactor if needed.
This process is used in a variety of industrial applications, but there is no full-scale implementation
treating kraft mill effluent. This process has been proposed as a form of condensate polishing to treat
evaporator condensates,but it could be used to treat other streams.
National Council for Air and Stream Improvement

64 Interim Report
7.2 Ion Exchange
Ion exchange resins, such as weak basic resins based on a phenol formaldehyde matrix, have been
found suitable for treatment of pulp and paper mill effluent.
The ion exchange treatment processes include the following steps.
• pH adjustment and pre-treatment(pH requirement varies with the resin; usually it is on the
acid side towards pH 2 -4,but may be up to pH 9)
• Effluent treatment step passing the effluent through the column until breakthrough capacity is
reached. When the resin is saturated the column needs to be eluted
• Elution stage,where the pollutants are removed usually with caustic in a concentrated form
• Activation stage where sulfuric acid is passed through the resin to reactivate it.
In the kraft industry,the ion exchange process research has concentrated on treatment of bleach plant
effluent for removal of color and chlorinated organic compounds. One example is the Uddeholm
non-polluting bleach plant developed in Sweden in 1975-1980 (Billerud AB, Skoghall mill).
However,the process is currently not in use at the Skoghall mill(Fitch 1981).
Based on information from Billerud, Sweden,the ion exchange technology worked very well in the
bleach plant. There were no problems in incinerating the ion-exchange elutate. However, because of
the high costs of the ion exchange resin, the costs of the process became very high. Therefore, the
company decided to shut down the process and moved toward oxygen delignification and other
methods of cleaning effluents.
Other developers of ion exchange systems include Rohm and Haas (Fitch 1981) and Dow Chemicals
(Fitch 1981).
Their processes have been tested in pilot scale but no full-scale installations are known.
Risks
The risks with the ion exchange process are a)resin life time and plugging problems, and b) disposal
of elutate and associated chloride problems.
Treatment Efficiencies
The EPA tested the Billerud non-polluting bleach plant in pilot scale in 1980 (Fitch 1981). About 25
different organic components and 13 heavy metals were measured, as were the typical parameters
(COD, color,pH, and chloride).
Based on these results, the bleach plant effluent load reduction is estimated to be as follows:
• COD 65-70%reduction
• Color 90%reduction
• Chlorinated phenols 90-99% reduction
• Zn 40-90%reduction
• BOD 25%reduction
• Toxicity almost complete elimination
National Council for Air and Stream Improvement

Interim Report
65
• AOX very high reduction.
Though promising,this technology has not been developed further for pulping effluents since the
1980s and does not appear to be an active area of investigation.
7.3 Activated Carbon and Activated Petroleum Coke Adsorption
Activated carbon has been used for removing organics from wastewater for many years. The
effectiveness of activated carbon in removing dissolved and colloidal material by adsorption is
primarily due to its extremely high surface area. Pore size distribution and surface chemistry also
determine the overall effectiveness.
Activated carbon treatment in the pulp and paper industry has been proposed primarily for color
reduction. High molecular weight organic compounds are typically amenable to carbon adsorption,
whereas colloidal compounds or strongly polar organic compounds(amino acids, hydroxyl acids,
sulfates, and sugars)are refractory to carbon treatment. Powdered activated carbon has a higher
surface area than granular carbon, which may improve color removal.
The PACTTM, or powdered activated carbon treatment process,has been developed by USFilter's
Zimpro division, which owns the trademark(http://www.usfilter.com/en/). Figure 7.6 shows
USFilter's PACT process.According to the PACT trademark,the process consists of"wastewater
treatment systems comprising aeration contact tanks,aerobic digesters,air diffusers, clarifiers,
clarifier drive mechanisms,scum collectors,waste sludge airlift pumps, scum removal airlift pumps,
flow measuring weirs, aeration blowers, recycle pumps, froth control pumps, polymer feed systems,
carbon eductors and/or motor control centers sold as a unit." The PACT process is also used at
DuPont's Secure Environmental Treatment(SET) commercial and industrial wastewater treatment
facility located at its Chambers Works site in Deepwater,New Jersey. One potential problem with
the PACT process was pointed out in a recent study(Kennedy et al. 2000), which determined that
PACT-treated wastewaters were toxic to Ceriodaphnia dubia. The study concluded the toxicity was
a result of ingested PAC. Tertiary filtration was recommended. Another bench-scale study also
recommended tertiary filtration when powdered activated carton was used in activated sludge
treatment at 500-1000 ppm dosages(Narbaitz et al. 1996).
National Council for Air and Stream Improvement

66 Interim Report
' `" t f- r �� � ^' 7 _ r s'h 9.ir�.r at• �;w ti 4 'Tr.' Y�5 y' � -'- �'>.` , ,' 1, €'^ :e � ti� c _ r• �r z
s r e r ' VIRGIN ri -�•` s�` 1 .c 4 � '
STO7 RALGEE
M1 t ,C of x`t 3 STORAGE, ` RAG
"cf TANKS yFILTPATKIWV p•
t S.A
0FnomAL
` .; •}Y t y.!1 s I � 1 -'' o-��bg^ zz J. "s -r s
'co HTACTAERATIOIJ
b - r.` +• Ze a •r .t, 7C. �a �4EFFWINT':
xCMBOH
_ a
OVERFLOW
. r r y,l vn .r rai.':ki� � t.y,c•. � ,4Y•.✓'+M1'_Y.:.
v n '} r y TO PEOENEiIATION r3 i
l � r C•F y t*'4 .OR
Figure 7.6 USFilter's PACT System Process (Single-Stage,Aerobic)
Powdered activated carbon can also be combined with ultra- or microfiltration to remove dissolved
species(Zhou and Smith 2002). The PAC is added to the recirculation loop of the membrane system.
PAC may also reduce the fouling of the membrane due to the surface scouring effect of the PAC
particles.
Recent studies have focused on less expensive alternatives. One investigation used delayed
petroleum coke, a waste by-product from the oil sand industry (Shawwa, Smith, and Sego 2001; see
below).
Another novel source of activated carbon is coconut jute carbon(Singh 2006). Coconut shell husk
was used to produce activated carbon, which was then tested on pulp and paper mill effluent to
determine color removal. Color removal varied between 30-90%depending on pH,time, initial color
concentration, and amount of activated carbon added. Color reduction improved with increased time
(maximum at 100 minutes), increased adsorbent concentration, and higher initial color concentration
and reduced pH(maximum at pH=2). Dosage was 0.6 g/100 ml or 6000 ppm.
_Color Removal from Bleach Plant Effluent using Activated Petroleum Coke(Shawwa Smith
and Sego 2001)
One laboratory study addressed the removal of color from bleaching effluent using activated carbon
obtained from delayed petroleum coke.The study found that there was an abundant and cheap supply
of delayed petroleum coke available in western Canada. The petroleum coke was ground into
powdered form and activated in a two-stage process. The carbonization stage was carried out at 8500C
followed by a steam activation stage that lasted 1-6 hours. The activated petroleum coke was tested
using bleach plant effluent. The activated petroleum coke was shown to be effective in removing
color and AOX. Applications of 2,500 mg/I of powdered activated petroleum coke resulted in about a
30%reduction in color and AOX.Applications of 15,000 mg/l of powdered activated petroleum coke
resulted in about a 90%reduction in color and AOX.
National Council for Air and Stream Improvement

Interim Report 67
The petroleum coke was subjected to various activation periods(2,4, and 6 hours)and then used to
treat bleach plant effluent. The characteristics of the bleach plant effluent are presented in Table 73.
Table 7.3 Bleach Plant Effluent Characteristics
Parameter Unit Value
COD mg/I 2126
DOC mg/1 575
AOX mg/1 as Cl 80.2
UV254 nm(cm') 13.11
Color465 nm(mg/1 Pt-Co) 2300
pH 2.1
Batch testing was then conducted to generate adsorption isotherms using the bleaching effluent. The
color removal, based on a 2-6 hour reaction time, is shown in Figure 7.7.
Based on the figure,the activated carbon produced using a 4-hour activation period provided the
maximum color reduction.Dosages of up to 2500 mg/1 produced a maximum color removal of 33%.
Higher color removal required higher activated carbon dosages. The use of activated petroleum coke
appeared to remove the recalcitrant portion of the organic matter in the bleach plant effluent. This
would make the remaining effluent more susceptible to biological treatment.
Sludge handling was not addressed in this study.
2500
E- 2 hours
2000 4 hours
_6 6 hours
ob 1500
0
U 1000
500
0
100 500 1000 5000 10000
Activated-Coke dose (mg/L)
Figure 7.7 Color Removal at Different Activated Coke Dose and Activation Periods(Shawwa,
Smith, and Sego 2001)
National Council for Air and Stream Improvement

68 Interim Report
Powdered Activated Carbon Addition to Total Mill Effluent
The Appleton Papers mill in Roaring Springs,Pennsylvania,has conducted a full-scale trial with
powdered activated carbon addition. he mill is an integrated hardwood kraft mill producing about
300 Tpd of paper. The mill's wastewater treatment system (4.5 MGD, color concentration 250-300
PCU) (Simmers 2005) consists of a bar screen, and primary clarification(2 clarifiers), followed by
two aerated lagoons in series.The overflow from the second lagoon goes to two aeration tanks and a
re-aeration tank before going to final clarification(3 clarifiers). The mill conducted tests with a
variety of coagulants, oxidative processes, and carbon absorption to determine their effectiveness at
color removal. For coagulants,the mill tested ferric chloride, ferrous sulfate, ferric sulfate, lime
combinations, and polymers. For oxidative processes,the mill tested hydrogen peroxide, ozone,
ultraviolet light treatment, and ultraviolet light with hydrogen peroxide. The target was to reach a
color concentration of<200 PCU(Simmers 2005).
For carbon absorption, the mill tested various granular and powdered activated carbons on final
clarifier effluent. The most promising results were obtained with powdered activated carbon,which
consistently reduced the effluent color to 150-200 PCU. The mill conducted a full-scale test adding
powdered activated carbon to the final clarifiers. The powdered activated carbon was slurried with
water and added at the aeration tanks. During the trial the feed rate varied from 0.5 to 1.0 Tpd of
carbon(27-56 ppm)(Simmers 2005). It was found that the addition of powdered activated carbon
reduced color and BOD. The sludge from the clarifiers was dewatered by a press and then
incinerated in the mill's power boiler. The amount of powdered activated carbon in the effluent was
not determined. The mill is now considering a full-scale implementation of powdered activated
carbon addition. When used in the biological process,the carbon becomes part of the biological
sludge and is dewatered and handled in the same processes as the other sludges.
One of the study mills in this survey undertook mill-scale trials with activated carbon added to
different locations throughout the effluent treatment process. The sludge handling during these trials
performed as normal, and no special difficulties were experienced.
7.4 Electrodialysis (ED)and Electrodialysis Reversal(EDR)
In the electrodialysis (ED)method,the electrolytes in a water solution are separated with the aid of an
electrical current and a membrane. The achievable separation result depends first on the magnitude of
applied electrical current and the available size of membrane, and secondly on the ion strength of the
solution.
An electrodialysis system consists mainly of the electrodialytic membrane stack,pumps, and
membrane cleaning system.
Electrodialysis has been proposed as a method of closing up the bleach plant. One study looked at the
laboratory treatment of bleach plant acid stage effluent using ED to remove non-process elements and
then return the treated filtrates to brownstock washing(Tsai et al. 1999). Figure 7.8 shows the
configuration proposed in this study. Electrodialysis could also be combined with other membrane
processes such as the treatment of nanofrltration permeate by ED and then reuse in the mill(De Pinho
et al. 1996). Electrodialysis is a commercial technology used for desalination of water and the
treatment of various types of industrial effluent, but there are no known applications in kraft mills.
National Council for Air and Stream Improvement

Interim Report 69
Weak black liquor NaOH,Or
To chemical recovery Mg,storm
(Wag",inotganics,llgnin)
4 _
Brown pulp 02 t
(from digester) wash deligni- --i wash '
brown sto, £icatlon r
I t
washing L-- ---- -- ----- - -� t
A -=- - - - -->'
� •-- — -- --_---i t...Water
D wash E wash D wash
J
! _1
clo: Acidic effluent i Naox, ----- - ----
11'0 (coming most mcWs, QOr
Stmm transitim morals, H20 H,0
chloride) Storm Bleached
la Pm$u,400inor to pulp
rake up rgaNc hWli s
To eledradialysis: (m PaPer-
RemoveNPE's,thenregcle To xroa making)
Figure 1: Settrmatieofa generiebleached kraft pulp bleaching operation with water recycling(D:chlorine dioxide
bleaching stage.E:caustic extraction stagc).
Figure 7.8 Electrodialysis(Tsai 1999)
The electrodialysis reversal (EDR)process operates on the same basic principle as ED, where an
electrical current supplies the driving force to transfer electrolytes through an ion-change membrane.
However,the EDR process includes polarity reversal that provides self-cleaning of the membrane
surfaces as part of the process. The EDR process can be used to remove COD and color. EDR has
been used commercially;however,there are no known applications in the pulp and paper industry.
8.0 COLOR REMOVAL-CHEMICAL PROCESSES
Chemical coagulation and solids separation of biologically treated effluent is a method to reduce
dissolved residual components such as color,COD,AOX and nutrients. By adding organic
coagulants or Me3+(Fe,Al)salts, larger dissolved organic molecules can be precipitated out of the
solution. Floc formation and ability to settle are often enhanced by adding a separate organic polymer
prior to solids separation. The treatment involves the addition of a coagulant and a polymer,pH
control (if metal salts are used),separation of the formed solids in a dissolved air flotation unit or in a
conventional clarifier, and sludge disposal.
The addition of chemicals to the effluent treatment,especially in larger quantities, has to be evaluated
from all aspects, including the toxicity of the specific chemical used in order to avoid potential
effluent toxicity issues.
8.1 Lime Precipitation
Precipitation with lime has been used to reduce effluent color. Lime treatment has been effective in
industry trials for color removal(Ganjidoust et al. 1996; Roux and Bohmer 2000). Other parameters
would be reduced as well, including COD(60-70%),P, and organic solids. Lime treatment typically
National Council for Air and Stream Improvement

70 Interim Report
requires higher dosage than alum treatment and generates more sludge. A considerable amount of
research was devoted to the development of lime treatment processes in the late 1960s and 1970s.
Basically,there were"minimum lime"and"massive lime" processes. A few installations were built
in the U.S. (Baird 1995). Essentially, the processes were similar to the raw water lime treatment.
Rebumt lime was used as the precipitating agent. In the massive lime process,the precipitated sludge
containing lignin, other organic material, and the used lime was regenerated in a dedicated lime kiln
or it was combined with the mill's lime sludge. In other processes,separate sludge handling was
attempted. The large dosages of lime that were needed to achieve an acceptable treatment efficiency
resulted in very large quantities of sludge. The difficulties involved in the sludge dewatering
eventually led to the discontinuation of the efforts to develop lime treatment of effluents.
A more recent study on lime precipitation looked at the effectiveness of lime precipitation prior to
biological treatment(Roux and Bbluner 2000). Both lab-and pilot-scale studies were conducted on
spent OZ delignification liquor, foul condensates, and total mill effluent. Color removal of about 80%
was achieved. The optimum pH range was 10.5-12.5 with retention times of 1-2 hours. Lime
precipitation worked best when the calcium concentration was above 1000 mg/l.
One of the study mills in this survey attributes about 30%color removal in the effluent treatment
system to the presence of lime in the influent. The mill adds 40 tpd of lime (about 320 ppm)to
neutralize the mill effluent. Any precipitate is removed in the primary and secondary clarifiers with
the sludges and taken to landfill. That mill does not experience any special issues with the sludge
handling.
8.2 Alum Precipitation
8.2.1 Industry Applications
One of the study mills adds an aluminum-based salt to its activated sludge process for color control.
The addition is controlled depending on the season, and varies between 40 and 88 ppm. The color
reduction in the activated sludge process increased by about 30%when the Al-salt was added.
Most of the installations utilizing alum treatment are tertiary treatment installations,using aluminum
and/or iron salt with or without polymer addition as a coagulant(mecltanical/recycled paper mills).
The main drawback of the coagulation/precipitation methods has been the disposal of the formed
sludge. The volume of sludge can be significant, especially if metal coagulants are used. Also,the
dewatering characteristics of the sludge are typically poor.
Arauco's Valdivia bleached kraft pulp mill in Chile uses aluminum sulfate for tertiary treatment of its
effluent(Rodden 2005a). The tertiary treatment is mainly for the removal of color. The chemical
sludge from the tertiary treatment is currently landfrlled,although future plans call for composting the
sludge.No information about the equipment for handling of the tertiary sludge as found in the
literature. The primary and secondary sludge is dewatered on belt presses and incinerated.
8.3 Iron Precipitation
One example of tertiary treatment using iron is the FennoTriox process by Kemira. FennoTriox uses
both oxidation and coagulation to remove pollutants. The FennoTriox process uses ferrosulphate(a
waste) in combination with hydrogen peroxide(Fenton reaction)in a chemical effluent treatment
plant. The reactions between the metal salt and peroxide will produce radicals that can oxidize
organic compounds in the effluent.
In the FennoTriox system, all chemicals are introduced into a reactor with a mixing zone and a
reaction section. The flocs generated in the reactor are removed in the flotation unit following the
reactor. The retention time is 10-30 minutes in the reactor and 3-5 minutes in the flotation unit. The
National Council for Air and Stream Improvement

Interim Report 71
removal of organics and nutrients is claimed to be more efficient than with conventional coagulation.
The sludge from this process(0.2-0.5 kg/m3 effluent) could be landfilled or burned.
Laboratory testing of biologically treated kraft mill effluent(following AST)resulted in the following
reduction efficiencies with the FennoTriox process(Syvdpuro 1994):
• COD 92%
• AOX 88%
• Chlorinated phenolics 57%
• N 85%
• P 98%
Other laboratory testing of the FennoTriox process showed color removal rates of 90-95%.
No full-scale applications of the FennoTriox process are known. Tertiary treatment with ferric—
aluminum sulfate(AVR)without hydrogen peroxide is done in a few Scandinavian mechanical pulp
and paper mills for phosphorus control.
8.4 Polymer Precipitation
Polymers have been used in primary, secondary, and tertiary coagulation and flocculation treatment
of industrial effluents.
Two of the study mills in this survey use polyamine to remove color. One mill adds polyamine to a
spare clarifier that contains diverted highly colored effluent. The other mill adds polyamine to a
process effluent stream that contains mainly black liquor.About 4700 lb/d is added to 17000 gpm of
effluent(23 ppm)to remove about 100 t/d of color.
One polymer treatment process used in the pulp and paper industry is the Smurfit-Stone Container
Corporation's color removal process. The process is covered by U.S. patent 4,738,750,which states
"A system and method for converting pulp and paper mill waste water into a
decolored, neutral pH effluent and a solid suitable for use as fuel in a furnace. The
treatment system is used following primary and secondary treatment of pulp and
paper mill waste waters typically found in the industry.After secondary biological
treatment,the waste waters are pumped to a coagulation tank where the waste water
is brought in contact with a polyamine coagulant which coagulates lignins,degraded
sugars, and other compounds which typically discolor this water.The coagulation
particles are increased in size by addition of an acrylamide polymer in a flocculation
tank to improve the hydrophilic characteristics of the coagulant.The waste water is
then mixed with a dissolved air and water solution under pressure. Upon dissolution
of the dissolved air at atmospheric pressure the air is absorbed by the flocculated
matter in the aeration tank and the flocculated matter is caused to migrate towards the
area of less pressure,i.e., the surface of the tank.The flocculated matter accumulates
on the surface of the flocculating tank and can be skimmed from the top,dried and
ultimately burned in a furnace."
This process uses an organic polyamine as a coagulant targeting color precipitation and
polyacrylamide as a flocculant. Since the formed sludge is almost purely organic(lignins),
incineration in the recovery boiler has become possible while increasing heat recovery and reducing
sludge disposal costs. In bleached kraft mills,however, the incineration of chlorine-containing sludge
National Council for Air and Stream Improvement

72 Interim Report
could lead to increased chloride levels in the white liquor loop, resulting in corrosion and boiler
plugging problems. In recovery-limited mills,adding solids to the black liquor would also potentially
reduce the pulp production capacity.
Smut-Stone Container Corporation has used this process at their mills in Missoula, Montana and
Hodge, Louisiana. The corporation also has a patent(U.S. patent 4,724,045)for the removal of color
from alkaline pulp and paper wastewaters using a polyacrylamide coagulant.
The Stone process is also used at the mill in Skookumchuck, British Columbia(Hodgson et al. 1997;
Stevenson 1995). The Skookumchuck mill has a river-based color limit. With the Stone process, they
report color removal rates of 55-80%depending on effluent characteristics and polymer type.The
mill has also reported reductions in other parameters, including BOD,suspended solids, COD(22%
removal), and AOX(23%removal). The sludge from the flotation unit is returned to the black liquor
evaporators where it is combined with the black liquor and incinerated in the recovery boiler. Figure
8.1 shows the process used at the Skookumchuck mill. Because of the high cost of the polymer
treatment, the mill has studied the recycling of the Eop filtrate as an option to reduce effluent color
(see Section 6.2.2.3).
Figure 8.1 Color Removal Process Used at Skookumchuck(Hodgson et al. 1997)
r. From T6 Cuft -
SYmn
a i,
� A salmon
G�agufant � i .,. �,FlaatWn GuiEm.
.. .k • Lt, d, ?�
"FNcatant �. . j • ,,.
> •Man i,.:';x ,, i , - . �f
:JSlalpitoNWk i5 -•
',7Y1ky�U _.< Black Lkryx T•nk
8.5 Nitric Acid Precipitation
Nitric acid has been used in laboratory and pilot tests for color precipitation(Roux and Bohmer
2000). The effluent was acidified to a pH of 3 and then a polymer was added.The precipitate was
removed using a dissolved air flotation(DAF)unit. The tests were conducted using untreated whole
mill effluent. It was found that the color in effluents containing high amounts of black liquor was
easily removed by precipitation with about 80%color removal. The DAF unit was required because
the precipitate formed from nitric acid treatment did not settle well. In addition,the acidification of
the effluent caused the formation of H2S, which could cause odor problems. This technology has
been tested in pilot scale only and there are no know commercial applications of this technology. Use
of nitric acid would add to the nitrogen load on the effluent, which could be a potential problem.
National Council for Air and Stream Improvement

Interim Report 73
8.6 Electrochemical Treatment
Electrochemical treatment involves the use of sacrificial iron electrodes. A current is applied across
the electrodes, which creates iron ions at the cathode and hydrogen and hydroxide ions at the anode.
The iron ions then form flocs of Fe(OH)2 onto which the color is adsorbed. The flocs can then be
removed by clarification. Electrochemical treatment has been used industrially in the textile and
carpet industries to remove dyes and heavy metals from wastewaters. The sludge recovered from the
process would have to be dewatered and disposed of. In laboratory tests of mill effluent color,
reduction ranged from 60-90%for electrochemical treatment(Springer, 1995)and up to 95%for
electrochemical treatment combined with a polymer(Orori et al. 1995). Toxicity was also reduced in
one study. Operating costs included energy, sludge disposal, iron for the electrodes and acid for
cleaning the electrodes.
8.7 Summary of Chemical Treatment
Table 8.1 presents data on some installations where chemical treatment is used as tertiary treatment.
National Council for Air and Stream Improvement

Table 8.1 Full-Scale Installations of Coagulation as Tertiary Treatment
State/ Targeted Chemical
Prov. Source m3/d Parameter
Stone Container Missoula MT USA UBK/I O%BK Stone 1988 7 mo/yr 38,000 Color Polyamine—
Corp. Stone Process
Stone Container Hodge LA USA UBK/NSSC Stone 1985 dry 53,000 Color Polyamine—
Corp. season Stone Process
Crestbrook Industries Skookumchuck BC Canada BK Stone 1994 cont. 38,000 Color Polyamine—
Stone Process
Arauco Valdivia Chile BK Tertiary 2005 Color AI-compound
Study Mill USA BK Activated Color Similar to PAC
Sludge
add'n
SCA Ostrand Sweden CTMP Fennotriox 1995 intermit. 3,600 COD,P FeSO4/H201
M-Real Kirkniemi Finland CTMP/GWD Tertiary Cont. 10,000 COD,P AVR°
Stora Billerud Fors Fors Sweden CTMP/GW Tertiary cont. COD,P AVRI°
AB
Holmen Paper AB Braviken Sweden TMP/GW Tertiary COD,P AVR°
Holmen Paper AB Hallsta Sweden TMP/GW Tertiary COD,P AVR°
Stern Feldmuhle Hylte Sweden TMP/GW/SI Tertiary COD,P AVR"
Hvltp AR
United Paper Mills 7amsankoski Finland TMP Tertiary only COD,P
Ltd. emerg.
Savon Sella Ltd. Kuopio Finland NSSC Tertiary On
occasion
a ferric aluminum sulfate

Interim Report 75
9.0 COLOR REMOVAL—OXIDATION PROCESSES
Recent research has focused on the use of advanced oxidation processes(AOPs)to treat either whole
mill effluent, bleach plant effluent, or effluent from specific bleaching stages. AOPs include Fenton
and photo-Fenton reactions, peroxide(H202)treatment, ozone treatment, ozone with H202 or UV
light or both, and heterogeneous photocatalytic processes(such as titanium dioxide photocatalysis).
Some of these processes will be reviewed below.
9.1 Peroxide
9.1.1 Peroxide Treatment ofEop Filtrate
Peroxide treatment has been used commercially to treat various wastewater streams.
Treatment of the EoP-Filtrate with hydrogen peroxide has been used at the mill in Grande Prairie,
Alberta since 1997 (Wohlgemuth, Lam, and Willis 1997).
Figures 9.1 and 9.2 show the effect of peroxide treatment on EoP filtrate for non-oxygen delignified
pulp. Chemical oxidation with peroxide(HZOZ) oxidizes certain structures of organic compounds and
hydrolyzes large molecules into smaller units,thus reducing color. Some components are fully
oxidized to carbon dioxide and water. Color removal is not proportional to peroxide dosage. With
higher dosage the chemical becomes less efficient. The required dosage increases exponentially as a
function of color removal,which enables one to determine the maximum reduction achievable with a
reasonable dosage.
Effluent colour vs Post Treatment H2O2
3s00 ,
3000
_ 2500
7
it 2000
1500
U
1000
S00
0
0 200 400 Goo 000 1000 1200
H2O2(ppm)
KF=0.20 0 KF=0.22♦KF=0.27 Eap EU02- 0.4z
Figure 9.1 Post Color Treatment Efficiency(Wohlgemuth, Lam, and Willis 1997)
National Council for Air and Stream Improvement

76 Interim Report
60
Removal %
50
40
30
20
10
0
0 150 300 450 600 750 9W 1050 1200 1350
H2O2 Charge, ppm
�KF=0.16-F KF=0.2 . KF=0.2
Figure 9.2 Post Treatment Color Removal Percent Efficiency(Wohlgemuth 1999)
The major advantage of this technology is the low capital cost involved. Also,the process is easy to
operate and the influent color can be controlled to a desired level.
The peroxide treatment system at the mill in Grande Prairie included mixing of the aqueous peroxide
solution with E-stage filtrate leaving the seal tank. The mixture was allowed to react in a retention
tank for 0.5-1 hour before being discharged to the sewer. The system included the following main
equipment: a) peroxide delivery system retention tank, and b) level and dosage control systems.
This system is currently not in use after installation of oxygen delignifrcation.
The peroxide treatment technology is simple,involving primarily the mixing of chemicals and
retention of the mixture. Up to 30-50%reduction of E-stage color has been reported to be achievable
(Wohlgemuth, Lam, and Willis 1997). Peroxide treatment can also reportedly reduce BOD and COD
(Robinson 1994).
9.1.2 Enhanced Peroxide Treatment with Catalyst of Er Stage Effluent
This process is similar to peroxide treatment except that a catalyst is added to the reactor to enhance
the treatment. Catalysts(TAML)are currently being developed at Carnegie Mellon University
(Wingate et al.2001, 2004).
TAML (tetra amido macrocyclic ligand) iron(III)catalysts are one-time use hydrogen peroxide
activators. By using TAML catalysts with hydrogen peroxide,color and AOX can be removed from
E-stage effluent.
National Council for Air and Stream Improvement

Interim Report 77
Bench-scale and pilot-scale tests have been conducted on softwood (pine) and hardwood(eucalyptus)
E-stage effluent at a kraft mill in New Zealand. The TAML catalysts function best under alkaline
conditions and with efficient agitation to achieve the maximum color removal. The pilot plant tests
consisted of two vessels, a 200 L vessel with a hydraulic retention time of one hour, and an 800 L
vessel with a hydraulic retention time of four hours.
The vessels were operated in series with a flow rate of 3.3 Urnin and chemical addition to the first
vessel. It was determined that a reaction time of one hour provided suitable color removal. Table 9.1
shows the results of the pilot testing. As can be seen in the table,the TAML catalyst is more effective
for softwood (pine)effluents than for hardwood (eucalyptus) effluents.
Table 9.1 TAML Catalyst Pilot Plant Results on Bleach Plant Alkaline Effluent'
Chemical Application Parameter Color AOX
0.5 µM catalyst,6.5 mM HzOZ,pine Influent 18t4 0.41f0.11
(0.23 mg/I,catalyst, 190 mg/I H102) TAML treated 10±3 0.34t0.05
Removal(%) 46 -
1 µM catalyst, 13 mM H201,pine Influent 23±3 0.38f0.10
TAML treated 8tl 0.31t0.09
Removal(%) 67
2 µM catalyst,22 mM 1420,pine Influent 25t3 0.38t0.04
TAML treated 6±1 0.26t0.04
Removal(%) 78 32
2 µM catalyst,22 mM H202,eucalyptus Influent 2.9f0.04 0.22t0.09
TAML treated 1.6t0.3 0.12f0.06
Removal(%) 45 -
°Conditions: pH 11.8, 1hr,60°C
In previous work,the authors of the study noted that the addition of hydrogen peroxide alone caused
an increase in the color at 400 nm wavelength due to the formation of a finely divided precipitate.
The TAML catalyst is short lived and degrades after about 10 minutes at the reaction conditions.
Therefore,Microtox toxicity testing was also done on the pilot plant effluent to determine if the
catalyst degradation products impacted effluent toxicity and no toxicity was found.
Experiments have also been undertaken using the TAML in the Ep stage in the process (Horwitz
2005).
A summary of the chemical dosages for about 30%color reduction using TAML/H2O2 color was
given in the 2005 NCASI Effluent Color Management workshop(NCASI 2005):
Eop filtrate treatment (825 ADT/d @ Eop filtrate 1.8 MGD):
13.5 lb FE-TAML/d(0.016 lb/ADT)and 0.5 T HZOZ/d(0.06%)
Eop tower treatment (300 ADT/d)
2.5 lb FeTAML/d(0.008 lb/ADT)and 0.65—0.75 %H202 on pulp
Currently,this technology is not commercially viable because the TAML catalyst is not produced on
an industrial scale. When the TAML was added to the Eop tower, the catalyst use was lower than
National Council for Air and Stream Improvement

78 Interim Report
when the Eop filtrate was treated. The TAML may thus be more effective when added in the
bleaching process.
9.2 Ozone
9.2.1 Ozone Treatment
Process and Applications
Ozone is a gaseous oxidant with a high oxidation potential,second only to fluorine,preferentially
attacking compounds which consist of carbon-carbon double bonds(color causing structures) and
organic functional groups. Any unreacted 03 decomposes rapidly to 02. The reactivity of ozone is
not dependent on temperature and favors lower pHs. Therefore, ozone treatment may be applied on
the total effluent as well as on any particular stream.
A small paper mill in Germany called Biittenpapierfabrik,which produces about 5,000 tons/yr of
colored paper, uses ozone to decolorize their effluent. The ozone also gives a 10-20%reduction in
COD and reduces AOX. This allows the mill to reuse part of their effluent for process water(Webb
2002).
The ozonation process includes the ozone generation and ozone dissolving into effluent. Ozone is
generated from dry air or pure oxygen by a high voltage electric discharge. The generator outlet gas
stream may contain up to 12%ozone. The gas is dissolved into the effluent by means of diffusers or
other equipment.
Ozone could be added to the D or Eop stage effluent or the total mill effluent. The gas would be
dissolved into the effluent by means of diffusers or other similar equipment in an airtight tank with a
detention of approximately 10-20 minutes. Any residual ozone in the off-gases would be destroyed
before venting to the atmosphere.
The scope includes the following main equipment: a) ozone delivery system, b)retention tank, and c)
off-gas treatment unit.
The production of ozone is proven technology. Ozone treatment of water for disinfection purposes is
a well-established technology employed at several municipal water and wastewater treatment plants.
In the pulp and paper industry the D and E stage and total mill effluents have been targets for much
lab-and pilot-scale testing with ozone. However,no full-scale ozone effluent treatment has been
installed on kraft mill effluents.
The potential environmental risks are related to these following:
• color reversion in ASB;
• possible BOD increase;
• environmental impact of 03 oxidation products; and
• residual ozone air emissions.
Impact on Color when Treatine Bleach Plant Alkaline Filtrate
A 60%reduction of E-stage filtrate color would be expected, assuming no color reversion in
secondary treatment. Literature results are varied on color reversion,with some studies showing
reversion, some no change, and some further color reduction in the treatment system. In one study
ozone was applied to the alkaline bleach plant effluent from a bleached kraft mill (Bijan and Mohseni
2003). The characteristics of the bleach plant effluent are presented in Table 9.2.
National Council for Air and Stream Improvement

Interim Report 79
Table 9.2 Bleach Plant Effluent Characteristics (Bijan and Mohseni 2003)
Parameter Unit Value
BOD5 mg/I 282.2
COD mg/I 1586.3
TC mg/I 676.6
pH 11.03
Color C.U. 1542.7
Figure 9.3 shows the color removal from the alkaline bleach plant effluent for bench scale testing.
Color removal of up to 70Va was achieved in the testing.
1aa0
i
1600
1400 •
_ 1200
c
` 1000
c 800
U
600 ♦
400 —
200
0
0 0.1 0.2 0.3 0.4 0.6 0.6 0.7 0.8 0.9 1
Ozone dosage(mg WmL wastewater)
Figure 5: Color removal of alkaline bleach plant effluent during ozonation(7'=20°C, pH= 11,
inlet gas flow rate= 185 mL/min,03 concentration in the input gas=0.1 1 mg/mL)
Figure 9.3 Color Removal from Alkaline Bleach Plant Effluent(Bijan and Mohseni 2003)
Impact on Color When Treating Combined Mill Effluent
Ozone treatment can reduce up to 80%of whole mill effluent color with an inlet ozone concentration
of up to 3.0%by weight according to results from pilot-scale tests with ozone at a softwood kraft pulp
mill (Zhou and Smith 1996). Bench-scale testing used effluent from a stilling basin following an
aerated lagoon. The characteristics of the whole mill effluent are presented in Table 9.3.
National Council for Air and Stream Improvement

so Interim Report
Table 9.3 Whole Mill Effluent Characteristics (Zhou and Smith 1996)
Parameter Unit Value
COD mg/1 485
BOD5 mg/1 1 I
TOC mg/1 192
AOX mg/I 7.77
Color TCU 943
pH 7.62
Total Mn mg/l 0.62
The color removal for an inlet gas flow rate of 1500 mUmin and an ozone concentration of 3.0%by
weight is shown in Figure 9.4.
1000 20
-a- effluent color
800 -o-off-gas ozone 16
0
�-n
F=V- 600 12 w
0 0W
U 400 8
200 Qr,: 1500 mUmin 4
CG..n: 3.0 % (%v/w %)
0 0
0 3 6 9 12 15
Time, min
Figure 9.4 Color Removal versus Contact Time for Whole Mill Effluent(Zhou and Smith 1996)
Pilot testing on whole mill effluent using ozone was also conducted at the Tenneco Packaging
containerboard mill in Valdosta, Georgia(Lovell, Stein,and Schmadel 1997). Color removal of
about 90%was achieved.
National Council for Air and Stream Improvement

Interim Report 81
BOD
The breakdown of organic molecules increases the potential of BOD formation(Zhou and Smith
1996), by one estimate increasing the raw BOD load to treatment by 5-10%. This low molecular
weight BOO may, however, be treated more easily with biological treatment, so the net impact on
BOD may be low.
COD and TDS
As a result of oxidation, COD and total dissolved solids may be reduced by approximately one-third
of the ozone dosage.
9.2.2 Ozone/UV Treatment
The ozone/UV treatment processes use UV photons to activate ozone molecules to form hydroxyl
radicals.The hydroxyl radicals then oxidize pollutants in the wastewater. In laboratory experiments,
bleach plant and total mill effluent exposed to sunlight have been decolorized and dechlorinated(i.e.,
AOX and COD have been reduced), and high molecular weight material (HMW)has been degraded
(Sonnenberg et al. 1994).Both UV and visible light contribute to this process. In a UV or ozone/UV
process,effluent would be exposed to ultraviolet radiation in an exposure/reaction chamber.This
process is still in the research phase. Initial results indicate that pH, light intensity, exposure time and
BOD/COD ratio are important process parameters. In the ozone/UV process, 03 consumption is
much higher than with 03 treatment alone. This makes the 03/UV process less economically attractive
compared to 03 treatment alone or ozone/peroxide or peroxide/UV processes(Zhou and Smith 2002).
9.2.3 Ozone with Biological Filtration
This process is applied at a paper mill in Germany(Schmidt and Lange 2000). The tertiary treatment
step consists of ozone treatment followed by biological filtration. The ozonation stage consists of two
reactors in parallel. Ozone is mixed with the wastewater and flows into the reactor. The off-gasses
from the reactor are used as such, i.e.,no off-gas destruction, in an AST. The wastewater from the
ozone reactor flows to a holding tank with a 1.5 hour retention time that allows the remaining ozone
to dissipate and protects the biofilter from ozone shock. The biofilters consist of three rotating filter
units with air and wastewater feed to the bottom and a design speed of 4.5 m/h. The system was
designed to minimize COD and BOD discharges.
The mill's entire treatment system consists of primary clarification, equalization,AST,secondary
clarification, ozonation, and biofiltration. The entire system operates at approximately 98%BOD
reduction efficiency and approximately 85%COD reduction efficiency. The treatment volume is
small, 14,000 m3/d (3.7 MGD),and there is no pulp mill at the site. Figure 9.2 shows the treatment
system at the mill.
National Council for Air and Stream Improvement

82 Interim Report
Qa •110fiG m'/d ,Qe' SAAOm'(A r ohs u6.
CO -n.owkp'd. COD•roriii,ksld.'
HOB —Il000kPH, .DOD- sits k�ld '
'SSN 31AOOl'�d' R 3T7 NO oNm-
meow
aed;mraudonl. - - '
rn
PAL3t4
1yv,CR '6edv etodge S'i H3� '
efllueN _ twmus'sm) ' �..
P11 S —
. tmuv.
muiog, omne
.. � iCrriar _ :'reiitloe
J t
(Wench) -
AruO.
A^'66v6 'azere.
11m6n: :.A_: icon vir 'amtka.
_
C58 t306mtl
Figure 9.5 Ozone and Biological Filtration at a Paper Mill (Schmidt and Lange 2000)
9.2.4 Heterogeneous Pkotocatalyst and Ozone Treatment
Heterogeneous photocatalysis involves the irradiation of a semiconductor anode, such as titanium
dioxide(Ti02),with ultraviolet light with a wavelength less than 390 run. The light source can be
either sunlight or UV lamps. This process is thought to create hydroxyl radicals that react with the
organics in the effluent. Several variations of the heterogeneous photocatalysis process have been
experimented with, including combined 03/UV heterogeneous photocatalysis treatment and
photocatalysis with Ti02. Although 03/1JV experiments were based on bleach plant effluent,this
process could be used on the whole mill effluent as well. In the 03/UV heterogeneous photocatalysis
process,the ozone increases the production of hydroxyl radicals and thus improves the treatment
efficiency. The combined 03/UV heterogeneous photocatalysis provided the same treatment
efficiency as ozone and UV heterogeneous photocatalysis applied in sequence, but the time required
for reaction was much lower(Torrades et al. 2001).
Other laboratory experiments using a Ti02 slurry and an Hg vapor lamp as a light source were
conducted with a variety of mill obtained effluents after primary treatment(Almquist and Boyd
2004). This research looked at photocatalysis only(without ozone) and its effect on COD, BODs and
toxicity(Microtox test). After an 8-hour treatment,the kraft mill effluent COD was reduced by 40%,
BOD by 30% and toxicity by 35%. The color of the effluent was also reduced, but not measured.
Reductions were larger for mills using other types of pulping processes. Other studies(Torrades et al.
2001)have shown that the combined ozone-photocatalysis process is more effective than each
process individually or in sequence.
This process is still mostly laboratory scale(Zhou and Smith 2002). For this process to be
commercially viable, several areas will need to be addressed, including reactor design and catalyst
immobilization.
National Council for Air and Stream Improvement

Interim Report 83
9.3 Wet Air Oxidation with Catalyst
9.3.1 Process and Applications
Under conditions of moderate temperatures (200-650°C) and elevated pressure(100-250 bar)
'dissolved and solid organics can be oxidized in the presence of water and oxygen in a process called
super critical water oxidation(SCWO). Depending on temperature, pressure and residence time,
organic compounds are broken down into smaller molecules and eventually to carbon dioxide, water
and inorganic acids. Figure 9.3 shows an example of super critical water oxidation(Cooper et at.
1996).
Experiments have been conducted on the use of wet air oxidation (WAO)with a catalyst(Zhang and
Chuang 1999). By using a catalyst like Pd-Pt-Ce/alumina, much milder reaction conditions, 130-
170°C and 15 bar for 3 hours,can be used. Without a catalyst, oxidation would not occur under these
conditions.The milder conditions also allow the use of carbon steel process equipment(Zhang and
Chuang 1999). Figure 9.4 shows color removal in laboratory tests of combined Do and Eop stage
effluent.
LIQUID
OXYGEN
WASTE TANK
WASTE BYPRODUCT 91LIQUID
TANK CO2STORAGE OXYGEN
PUMP
RECIRCULATION 8
HOMOGENIZATION OXYGEN
02-0O2 VAPORIZERS MED.PRESS.
SEPARATOR SEP.
HP
FEED
PUMP VENT
COMPRESSOR MAKEUP 02
LOW PRESS.
02 A-CUI.SULATOR CCOLER HIGH SEP.
RESS,
PREHEATER REACTOR COOLER SEP.
BOILER AQUEOUS
OJSOUD EFFLUENT
HEATRECOVERY SEP.
RECIRC. LOOP
PUMP
SOLIDS
STORAGE
Figure 9.6 Super Critical Water Oxidation Process Developed by MODEC(Cooper et al. 1996)
National Council for Air and Stream Improvement

84 Interim Report
14
100
13
60
• • �- 72
a 11
`m
a 60 I 10
m A n
if r 4
0 40 • -ice
U
i
20 \Ox
6
E
0 5
0 20 40 60 80 100 120 140 160 180 200
Run Time, min
FIGURE 4. Color and pH versus reaction time profile at 443 K and
1.5 MPa with 1.0 g of catalyst.
Figure 9.7 Color Removal with Pd-Pt-Ce/Alumina Catalyst(Zhang and Chuang 1999)
Several catalysts have been investigated, including various iron and zinc oxide catalysts. These have
suffered from several problems, including catalyst deactivation due to dissolution in acidic effluents,
which would require an additional step to remove the leached metal ions. To combat this,the
combined bleach plant effluent, which has a higher pH,can be treated. This reduces the leaching of
metal ions from the catalyst.
This process results in a significant reduction in color and TOC. Because most of the color removal
occurs in the first hour of reaction,a short treatment time could be used as a preliminary step
followed by biological treatment. Super critical water oxidation has also been proposed as a
treatment for the concentrate from ultrafiltration or reverse osmosis(Hauptman, Gaims, and Modell
1994).
9.3.2 Technical Feasibility
Wet air oxidation is a commercial process with a few installations in the pulp and paper industry. The
Zimpro(wet air oxidation)process is employed at the Weyerhaeuser mill in Rothschild, Wisconsin
for sludge conditioning, a mild version of wet oxidation. At the Stora-Enso mill in Kimberly,
Wisconsin(Jortama 2003)wet oxidation enables the recovery of filler from waste activated sludge.
Research is ongoing for using wet air oxidation on pulp mill wastes and effluents, including at least
one mill that has investigated its impact on kraft mill bleaching effluent(Cooper et al. 1996). As the
National Council for Air and Stream Improvement

Interim Report 85
largest units currently in operation can handle only a few liters per minute,this technology is still
several years away from full-scale availability for bleaching effluent.
9.3.3 Risks
'The fundamental reliability issue is the scaling problems experienced on heating surfaces. Also,
plugging caused by inorganic precipitation in the reactor has created problems. Chlorides in the
effluent will set requirements on the materials needed to avoid corrosion.
10.0 COLOR REMOVAL—EVAPORATION AND INCINERATION
10.1 Evaporation and Incineration of Bleach Plant Effluent
The bleach plant effluents contain the major part of a Icraft mill's color, COD, dissolved organic
material, and total dissolved solids discharge. The bleach plant effluent typically also contains metals
such as manganese, iron, calcium, aluminum,silica,phosphorus, nitrogen,and other trace elements.
Several groups have developed separate evaporation and incineration of ECF bleach plant effluent.
Separate evaporation of TCF effluent is also being developed,with incineration of the concentrate
either in the recovery boiler or in a separate incinerator. Condensates from the evaporation process
are high in methanol and need to be treated to be suitable for reuse or to avoid air emissions. Figure
10.1 shows how bleach plant effluent evaporation might be integrated into the process.
Figure 10.1 Pre-Evaporation of Bleach Plant Effluent(Algehed, Stromberg, and Berntsson 2000)
IAeke-up wafer
Wood 4 Cooking' 02 .E Bleach Dryina 5
h
Pulp
•.'_-�:<e r:..plant `s�{ a,' � «:� ti.r
White 81xk Bleach ptarl
liquor I'ar olAuenl
Recovery" Black Ilquor con Vo a --
area;i -evaporation '- '^ Pre-
evaporation
'Ccndensat
oean candgnsate `,cleaning:< Canden5ata
<,4Reciplent,
ry.edn faldenShcB
Rmprocl l
National Council for Air and Stream Improvement

86 Interim Report
Two sample bleach sequences(ECF: DEOpDD and TCF: O-Q-Paa-PO-P) are briefly evaluated below:
10.1.1 Process Description
The process would involve bleach plant effluent minimization through filtrate recycle, collection of
the'effluent to be treated,pH adjustment of the effluent,pre-evaporation, concentration, incineration,
condensate treatment, and incinerator residue handling for disposal or for recovery of chemicals to the
bleach plant. After treatment,the evaporator condensate could be used as wash water in the bleached
fiber line.
The scope would typically include the following main equipment:
• bleach plant piping modifications to allow increased water recycle;
• effluent storage tank;
• pre-evaporation plant and concentrating unit(>40%ds);
• cooling tower for cooling water recirculation;
• incinerator;
• condensate handling;and
• incineration residue handling,possibly with chemicals regeneration.
10.1.2 Evaporation of ECF Bleach Plant Effluent
The following concepts have been studied:
• Evaporation of the whole bleach plant effluent(Blackwell et al. 1991)
• Evaporation of the EOP filtrate only (Dahl and Niinimaki 2000). Pilot trials at the Stora
Gruv6n mill have been performed with actual ECF bleach filtrate.
• Taking the EOP filtrate to the brown stock washing, and taking only the DO filtrate to
evaporation(Dahl et al. 1997). This concept would introduce chloride into the current
recovery cycle along with some of the dissolved material in the EOP filtrate.
Pre-evaporation for the bleach plant effluent
A few suppliers have developed evaporation plants for chloride-containing effluent, including Andritz
(trademark Zedivap) and Hadwaco (Finnish company owned by Hackman, Finland,now USFilter)
(Fagemas,McKeough, and Kyllonen 1999; Koistinen 1996). Both of these companies use
evaporation principles used in desalination plants.
The Hadwaco plant applies MVR(mechanical vapor recompression) and FF(falling film)principles
(Koistinen 1996). To withstand corrosion in a chloride-containing environment, the heating surface
is made of high density polyethylene(HDPE). The low heat transfer of HDPE is increased by
making the surface very thin(0.02-0.04 mm). This very thin plastic surface results in low AT(2.5°C)
and low pressure drop. At low dry solids concentrations,the energy demand becomes low(8-10
kWh/t water). The vapor recompression can then be accomplished by a blower type compressor. Up
to 8%dry solids concentration can then be achieved. The use of plastic material limits the upper
operating temperature to about 55°C, so a warmer effluent would flash off the excess heat. The
system is available in modules with capacity about 3.5 1/s, each covering 33.2 mZ ground area.
National Council for Air and Stream Improvement

Interim Report 87
The Zedivap plant operates along the same principles(FF) but the heating surface is a metallic surface
coated with thin plastic on both sides. The temperature limit of the operation is 80-85°C.The
wastewater stream itself or another hot waste stream(> 60-65°C)may be used as energy sources.
Depending on the situation, up to 10-12 evaporation stages may be feasible. It can also operate as a
VRC unit, (8-10 kWh/t water), in which case the specific loading may be higher and the unit smaller.
Depending on the effluent properties,the effluent may have to be sweetened to 4%dry solids. In this
case it would be evaporated to 10-15% dry solids.The heating elements can be stacked, reducing the
surface demand.
Concentration of the effluent
Before the pre-evaporated filtrate can be incinerated, it has to be further concentrated to 40-50%
solids.This can be done in a similar plant as the pre-evaporation, but needs a higher AT due to higher
boiling point rise, and thus more energy(20-30 kWh/t water). A conventional evaporation plant with
the adequate metallurgy(titanium for ECF mill)could also be used.
Incineration
The incineration of the chloride-containing concentrate represents a risk for generating dioxins and
other hazardous air pollutants. Combustion tests have been performed at pilot scale using the
CONOX oxidation unit developed in Finland by the Combustion Chemistry Research Group working
in cooperation with Abo Akademi(Koistinen 1996). A mean residence time of 2 seconds at about
1000°C was considered necessary for complete oxidation of all organic matter.
The ultimate handling of the incineration residue is still not satisfactorily solved. In an ECF mill,the
incineration product will largely be NaCl,and Na2CO3, together with the metals,phosphorus,etc.
present in the bleach plant effluent. It could be possible to clean the residue and recover the sodium
and chloride from it by an electrolytic process.
The incinerator would need adequate flue gas cleaning, e.g., a scrubber and ESP.
Instead of incinerating the evaporation residue on site, it could be potentially sold as such to an
outside chemicals manufacturer,who would process it back into chemicals(NaC1O3 and NaOH).
Condensate treatment
The volatile compounds in the effluent will boil over in the evaporation, including methanol and
volatile chlorinated organic compounds such as chloroform, if present. At higher dry solids
concentrations, hydrochloric acid has also been found to boil over in evaporation tests.
Figure 10.2 shows the carryover in the condensates from the evaporation of acidic bleach plant
effluents as measured in laboratory tests. At high solids concentrations the condensate became more
colored and had a higher concentration of chloride.
Prior to reuse,the condensate has to be cleaned by steam stripping and possibly other methods.The
stripper off-gases have to be disposed of at least thermally, but also with a scrubber to remove the
chlorine compounds.
No full-scale installations of ECF bleach plant evaporation and incineration are in use at this point.
The pilot unit that was operating at Gruv6n, Sweden is not in use any longer.
National Council for Air and Stream Improvement

88 Interim Report
700 300
600
p E. 500 250
U o --�---__ _ i e E E 200
_ l
c e 300 € A 'a 150 — ..l-_—
_ 200 9 100 I 1
'o E u 0 W
QV 10 T ¢ q E 50
0
1 2 3 4 5 1 2 3 4 5
Total solids of concentrate,% Total solids of concentralo,%
—COD.mOA TTOC.m04�Cnbu.Igyll -i—FoI--Od.n94'4 hbftml rr94—a—Acett ociQ mB4l
120 3
100 120
I 2.9 n f I
� u7 + I I A o 100
v E so
S a 60 —' - — ¢' U 1 i
80
s u c 27 7i d $ 60
♦....... ♦ 26 o -6 e 40
c 20 —I 0 -.I 25 E 0 1
1 2 3 4 5
1 2 3 4 5
Total solids of concentrate,% Total solids of concentrate,%
—Chbri MrIll —Cand-Wly. ,V- ...e...pH
�•fb�Ca—ar—Ng
Figure 10.2 Evaporation of Acid Bleach Filtrates, Carryover in Condensates
(Dahl,Niinimaki, and Kuopanportti 1998)
10.1.3 Evaporation ofTCFBleacbingEffluent
Evaporation and incineration of effluent from a TCF plant would need the same basic processes as the
ECF bleach filtrate evaporation. Since no chloride is present, conventional material can be used,
relying on less unproven technology.
The incineration residue would not contain chloride but Na2SO4 (if H2SO4 is used in the QA stage for
pulp acidification)and Na2CO3 or Na2S and Na2CO3 if incinerated in a reductive boiler.
A full-scale evaporation unit is in operation in a Swedish TCF line in a linerboard mill in which the
concentrate is mixed with black liquor from the unbleached process and burned in the recovery boiler
(Pekkanen and Kiiskila 1996;Valttila et al. 1995).
10.1.4 Summary ofBleacll Plattt Effluent Evaporation
Table 10.1 lists the pilot- and full-scale applications of bleaching effluent evaporation.
National Council for Air and Stream Improvement

Interim Report 89
Table 10.1 Bleaching Effluent Evaporation Installations
Location Bleach sequence Filtrate evaporated Scale Equipment
Stora,Gruvtin, O-DEOP-D-Ep-D 50150 mix of acid and 300 t/d Hadwaco—pilot
Sweden alk.effluent
studies only
Assi Doman, 00-Q-Paa-Po-P Q stage 1400 t/d Zedivap—full
Frbvifors, Sweden scale, in operation
In the only mill-scale installation, only part of the total kraft pulp is bleached. Obviously,the
accumulation of NPE and other potential problems related to the closure of the bleach plant water
system would be diminished when only less than 50%of the total kraft production employs closed
cycle technology.
10.1.5 Risks
ECF effluents
• Materials failure risk.
• Chlorinated compounds in condensates and flue gas.
• Incineration residue that is contaminated with chloride may be more difficult to dispose of or
use for recovery of chemicals.
• Build-up of chloride levels in bleacher/dryer if the bleach plant evaporator condensates are
reused on the pulp dryer(related to possible HCI carryover in the condensates).
• If the concentrate is incinerated in the recovery boiler, the accumulation of chlorides and
other NPEs poses a high risk for the mill operation.
TCF Effluents
• Incinerator residue handling in the case of separate incineration(chemicals recovery,
destruction). Mixing with other black liquor prior to incineration is the preferred method for
handling the concentrate.
• Non-process element build-up in recovery if the evaporator concentrate is burned with the
black liquor.
10.1.6 Environmental Impact
The system would eliminate the bleach plant effluent discharges to the sewer, and potentially make a
closed cycle around the bleach plant. However,the real possibilities of achieving this have to be
examined further as technology and experience develop.
The evaporation plant condensate will contain volatile compounds present in the bleach filtrates
(methanol etc.), and also chlorides, depending on the pH. Tests have also indicated that color bodies
may be carried over into the condensates (Dahl,Niinimaki,and Kuopanportti 1998; Dahl and
Niinimaki 2000). The condensates have to be cleaned before they can be reused.
National Council for Air and Stream Improvement

90 Interim Report
10.1.7 Lupact on Chemicals
In order to minimize carryover in the condensate,the pH of the feed to the evaporator should be
slightly alkaline. Acid waters(such as if Do effluent is treated alone) have to be adjusted to alkaline
pH.;
There could be a potential to recover bleach plant chemicals from the incinerator wastes if that
technology would develop.
10.1.8 Water Requirements
The need for cooling water in the evaporation plant will depend on the system configuration. In
principle, all energy that is fed to the evaporator, including the energy in ingoing water less energy in
outgoing condensate and liquor, has to be removed from the system with cooling water. If heat is the
energy source,the cooling demand is significant but lower in a MVR or VRC unit.
11.0 COLOR REMOVAL—FUNGUS/BACTERIA/ENZYMES
11.1 White-Rot Fungus Treatment
Various strains of white-rot fungus have been known to be able to degrade lignin as a"secondary
metabolism,"meaning lignin would be metabolized if a certain growth factor becomes limited. This
ability is not lost when the lignin becomes chlorinated in pulp bleaching,so the fungus is of interest
in removing chlorolignins and chlorinated phenolics in bleach plant effluents as a means of reducing
effluent color and toxicity.
Several parallel research programs have been pursued and the one most advanced at this point is the
"MyCoR"process at North Carolina State University. In this process the fungus is immobilized on a
series of flat disks, which are mounted in parallel on a rotating horizontal shaft such that portions of
each disk are alternately submerged in the waste operation. Either 1- or 2-day retention time would
probably be required (Pellinen,Joyce, and Chang 1988).
Since the fungus cannot use lignin as an energy source, it must be supplied one. If the stream to be
treated does not contain enough energy sources, such as hemicellulose,suggested possible additives
are glucose, xylose, cellulose, or possibly primary sludge. In addition,the pH should be between 3
and 5,the temperature between 28-40°C, and nitrogen should be the limiting nutrient.
Recent work conducted by researchers in Mexico used a two-stage process in the laboratory to
simulate the treatment of weak black liquor spills(water with weak black liquor) (Caffarel-Mendez et
al. 2004). In this experiment no additional carbon source was used. The treatment process consisted
of an anaerobic first stage,which was a methanogenic fluidized bed reactor,and an aerobic second
stage, which was an upflow reactor packed with wood cubes containing immobilized Trametes
versicolor white-rot fungi. Hydraulic retention times (HRTs)for stage 1 were varied between 0.5 and
5 days and the hydraulic retention times for stage 2 were 2.5 and 5 days. The fungal reactor was able
to sustain removal activity for 95 days without carbon addition.The overall process had a COD
removal efficiency of 78%and a color and liginoid removal efficiency of 75%. The aerobic fungal
treatment(stage 2)removed color and ligninoids more efficiently than the anaerobic stage(stage 1),
while the anaerobic stage was better at COD removal.
Other white-rot fungal treatments include the MyCoPor(Messner et al. 1990)and immobilized fungal
fluidized bed bioreactor.
National Council for Air and Stream Improvement

Interim Report 91
The MyCoPor process is a trickling filter that immobilizes the fungus on the surface of polyurethane
foam cubes. Both the MyCoR and MyCoPor use passive immobilization, i.e., adhesion of cells to a
solid support.
,The immobilized fungal fluidized bed bioreactor uses entrapment of the fungus in urethane foam
(Pallerla and Chambers 1996). This active immobilization results in a media with a high resistance to
deterioration by mechanical action and pH. This reactor is effective at removing color and AOX,
with removal efficiencies of 70%and 50%, respectively. Figure 11.1 shows the color reduction
during the experiment using a continuous fluidized bed bioreactor. The reactor was fed a mixture of
60%D-stage effluent and 40%E-stage effluent obtained from a mill with a OD(EO)DED bleaching
sequence. The reactor functions best at a pH of 5, and due to the porous nature of the urethane foam,
transport processes are non-diffusion limited. As with other white-rot fungus treatments, an energy
source for the cells must be supplied. In this case glucose at 8 g/1 would be recommended along with
nitrogen and phosphorus. Currently this technology is in the pilot-scale phase.
-
tI
861
O
72LU
G ;
= 58�
o I
u 44;
I
30'--
0 5 10 IS 10 25 30 35
TIME, days
Figure 11.1 Color Removal in an Immobilized Fungal Bioreactor(Pallerla and Chambers 1996)
11.1.1 Technical Feasibility
This process has only been tested at small scale,with the largest existing unit treating about 75 1/d.
Technology to take advantage of the decolorizing ability of the white-rot fungus has not developed to
the point where it can be utilized industrially.
11.1.2 Environmentallmpact
Rough removal efficiencies for one-day retention time are listed below.
• Color 50-80%
• Chlorinated phenolics — 100%
• AOX 45-70%
National Council for Air and Stream Improvement

92 Interim Report
• BOD 30%
• COD 14-30%.
Tests for acute toxicity of effluent treated with this process have been done on daphnia and Microtox;
complete removal of toxicity was demonstrated. Dioxin has also reportedly been reduced by white-
rot fungi.
Laboratory experiments have also been conducted by researchers in Turkey on the use of algae to
treat wastewater from kraft mills (Dilek,Tarlan, and Yetis 2002). The experiments were conducted
on wastewater from a mill pulping red pine and using a CEHDED bleaching sequence. The tests
were conducted in jars maintained at a constant temperature and the wastewater was treated with a
nutrient medium and a light source to stimulate algae growth. After a treatment time of 42 days the
COD had been reduced by about 55%,the color by about 80%, and the AOX by about 65%. Figure
11.2 shows the reduction of pollutants during the experiment. The long treatment time, the
temperature requirement,and the light requirement make this technology impractical for a mill
operating in a northern climate.
1.01, 50
S l 0.8 40
U xK
X 0.6 30
Q 0.4 20
O e
0.2 10
0 1 0
0 10 20 30 40 50
Tme,d
Fig. 1. Algal growth and COD, AOX and color removals under 3.4
kIx light intensity and 230 mg/l initial COD((M): COD; (•):color; (�)
AOX; (x) algal biomass).
Figure 11.2 Color Removal by Algae Treatment(Dilek,Tarlan, and Yetis2002)
National Council for Air and Stream Improvement

Interim Report 93
12.0 TASK II: MULL-SPECIFIC REVIEW OF TECHNOLOGIES AND PERFORMANCE
This study was supported by a group of bleached kraft mills. It included a review of successful and
,unsuccessful color technologies applied or tried at those mills. It also included a benchmarking of the
color discharges from the mills.
12.1 Color in Study Mills
Table 12.1 summarizes the color data for the four papergrade mills.
Table 12.1 Summary of Color Balance Data in Four Papergrade Bleached Kraft Mills, Ib/ADT
Mill A Mill B Mill C Mill D
Bleaching 12.6 42.9 42.4 25.2
CRP Waste 3.5
Brown Side Color 11.4 9.2 7.2 8.9
Paper Machines 1.1
Total Measured 28.6 52.1 49.6 34.1
Unknown/ampliflcatio
n 4.7 28 ? ?
Total Influent 33.3 80.1 51.2
Final effluent 28.3 66.3 20.1 23.5
In these mills the identified"brown color"amounted to 7-10 lb/ADT.
When using best available technology(BAT)the"brown color"target(excluding washing loss)could
be
• From spills: 2 kg CODA or about 4-5 lb color/ADT(22)
• From condensates: 2 Ib/ADT(EKONO's estimate)
• From rejects and knots: 0-2 lb/ADT(EKONO's estimate)
• Total 6-91b/ADT
A more detailed summary of the color data for four mills participating in the study is shown in Tables
12.2 through 12.5.
Table 12.2 shows the data for Mill A based on 2005 average data. The in-mill color sources are
extensively monitored, so that only about 14%of the color in the combined influent has not been
identified. The mill largely attributes the unknown color to amplification of the color in the sewer
system. The color is reduced in the activated sludge plant by an average of about 15%. The mill
attributes the reduction mainly to reduction of black liquor originating color rather than bleach plant
color. The mill estimates the black liquor color to be reduced by 50% due to sorption of colored
lignin on the bacterial sludge.
National Council for Air and Stream Improvement

94 Interim Report
Table 12.2 Summary of Mill A Color Data(ADT=Air Dry Short Tons of Bleached Pulp)-
2005 Data
Unit SWD HWD Total Total color,
Ib/d
Bleached Kraft Share %of total mill output —79
Hardwood %of total bleached 58
Digester Kappa kraft Dula _35
Pre-Bleach Kappa 16-I7lest) 10(est)
Bleach Sequence ODEoD ODEoD
Kappa Factor (No 1 Stage) 0.255 0.229
Bleach Plant Effluent
Acid Sewer
Flow m3/ADT(GPM) 2.4(257) 7.0(1028)
Color Ib/ADT(PCU) 8.3(1566) 5.7(367) 9371
Alkaline Sewer
Flow m3/ADT(GPM) 1.2(125) 2.5(361)
Color Ib/ADT(PCU) 7.5(2940) 4.6(853) 8117
Total Bleach Plant Effluent Color Ib/ADT 15.8 10.3 12.6 17489
Non-Bleach Color Sources
Brown Stock Ib/ADT 4.4 6100
Digester Area Ib/ADT 0.8 1149
Condensates Ib/ADT 2.0 2821
Evaporation Plant Ib/ADT 0.6 818
Recovery Boiler Ib/ADT 7.1 9789
Paper Mill&Other Ib/ADT 1.1 1426
Total Non Bleach Sources Ib/ADT 16.0 22103
Total known Sources Ib/ADT 28.6 39591
(Unmeasured Color/Amplification Ib/ADT 4.7(-14%) 6522
Measured Primary Influent Color Ib/ADT(PCU) 33.3(215) 46113
Final Effluent
Flow m3/ADT(MGD) 70.1(25.6)
Color Ib/ADT(PCU) 28.3(183)
Color Limit(Annual Average) Ib/d 42000 39128
Color Limit Normalized, lb/ADT-2005 prod'n 30.5
Approximate
°5000 lb/d(3 Ib/ADT)is due to CRP,i.e.,the bleach plant filtrate recovery process.
Table 12.3 shows the color sources for mill B,based on 2005 data. In this mill the difference
between known in-mill sources and combined influent is 35%, likely due both to unmeasured color
and to possible color amplification in the sewer system.
The color is reduced in the effluent treatment plant by 17%, of which 2%in the primary clarifier and
15%in the activated sludge plant.
National Council for Air and Stream Improvement

Interim Report 95
Table 12.3 Summary of Mill B Color Data(ADT=Air Dry Short Tons of Bleached Pulp)-
2005 Data
Unit SWD HWD Total Tolalcolor
Ib/d
Bleached Kraft Share %of total mill 80
Hardwood %of total bleached 78.5
Digester Kappa kraft Dula 28 17.4
Pre Bleach Kappa
Bleach Sequence DEopPD DEopD
Kappa Factor (No 1 Stage) 0.26 0.22
Kappa Factor (whole BP) 0.44 0.37
Bleach Plant Effluent
Acid Sewer
Flow m3/ADT(GPM) 30(1260) 21(3310)
Color Ib/ADT(PCU) 29(445) 27(570) 29414
Alkaline Sewer
Flow m3/ADT(GPM) 10.6(450) 4.5(700)
Color lb/ADT(PCU) 33.1(1410) 10.6(1055) 16501
Total Bleach Plant Effluent Color Ib/ADT 62.5 37.6 42.9
Non-Bleach Color Sources Ib/ADT
Digester Area lb/ADT N/A
Brown Stock Ib/ADT 6.1 6500
Contaminated Condensate lh/ADT N/A
Evaporation Plant Ib/ADT N/A
Recovery Ib/ADT 3.1 3269
Paper Mill&Other Ib/ADT N/A
Total Measured Non-Bleach Color Ib/ADT
Total Known Color Sources Ib/ADT 52.1 55684
Unmeasured Color/Amplification b/ADT 48(35%)
Combined Influent Color(Measured) Ib/ADT(PCU) 80.1(480) 85725
Secondary Influent Color Ib/ADT(PCU) 78.5(462) 84065
Final Effluent
Flow m3/ADT(MOD) 77(21.8)
Color lb/ADT(PCU) 66.3(390) 70965
Color Limit(max.downstream in 75(monthly)
river)
225(daily)
Table 12.4 shows the color sources for mill C. The bleach plant color is monitored regularly,but
other sources are not. Data from a,color survey in year 2000 was used to break down the black
liquor-originated color;therefore,the current level of black liquor originated color and the so called
color amplification cannot be assessed. The data indicate that the bleach plant effluent contributes
over 80%of the combined influent color as an average.
Mill C takes a high color reduction in the effluent treatment system. The biological treatment by
itself reduces the color by some 40-50%. The mill also has to add chemicals (alum based)to meet the
current color limit. The total color reduction in the effluent treatment system is 55-65%when
chemicals are added.
National Council for Air and Stream Improvement

96 Interim Report
Table 12.4 Summary of Mill C Data(ADT=Air Dry Short Tons of Bleached Pulp)
Unit SWD HWD Total Total color
Ib/A
Bleached Kraft Share %of total mill _80
output
Hardwood %oftotal bleached 55
kraft pulp
Digester Kappa —25 —20
Pre-Bleach Kappa 16-17 lest) 10-11
Bleach Sequence ODEopD ODEopD
Kappa Factor (No 1 Stage) 0.27 0.25
Kappa Factor (Total) 0.5 0.5
Bleach Plant Effluent
Acid Sewer
Flow m3/ADT(gpm) 24.4(1443) 21.3(1560)
Color Ib/ADT 22.9(425) 16.4(350) 13933
Alkaline Sewer
Flow m3/ADT(gpm) 14.7(871) 3.6(265)
Color Ib/ADT(PCU) 32.5(1000) 8(1000) 13653
Total Bleach Plant Effluent Color Ib/ADT 55.4 24.4 38.2 27586
Other Measured Color Sources(2000 data) Ib/ADT
Other Measured Color Sources,Total Ib/ADT 6.5 4709
Total Known Color Sources Ib/ADT 44.9 32295
Other b/ADT(%) 1.2 1006
Combined Influent Color(Measured) IVADT 46.1 33301
Secondary Influent Color Ib/ADT 43.3 31300
Final Effluent
Flow m3/ADT(MGD) 65.5(12.5)
Color_summer Ib/ADT(PCU) 16.9(133) 13873
Color_winter" Ib/ADT(PCU) 19.2(117) 12206
Color Limit in final effluent ppm,monthly avg 140(so)
123(wi)
Color Limit in Final Effluent-Normalized Ib/ADT 2005 prod 22(an)
(Approximate) 19 8(wi)
Su=summer
W i=winter
Table 12.5 details the color balance for Mill D. In this mill, all black liquor-containing color sources
are continuously monitored in a sewer separate from the bleaching paper making. However, the
combined influent is not monitored separately, only by adding the separate influent streams. The
black liquor-sourced color is 8.9 Ib/ADT.
This mill also measures a color reduction in the effluent treatment system of about 30%. The mill
attributes this to the use of lime used for neutralization of the influent.
National Council for Air and Stream Improvement

Interim Report 97
Table 12.5 Summary of Mill D Data(ADT=Air Dry Short Tons of Bleached Pulp)
Unit HWD SWD HWD Total Total
Ib/A
Bleached Kraft Share %of total mill _q5
output
Hardwood %of total
-82
Digester Kappa bleached kraft 15 36 16.9
Pre-Bleach Kappa 1 L8 22.6 16.9
Bleach Sequence ODED ODEDD DEDD
Kappa Factor (No 1 Stage)
Kappa Factor (Total)
Bleach Plant Effluent
Bleach Plant Effluent-Special Test
2006
Acid Sewer
Flow m3/ADT(gpm) 3.8(665) 39.2.(2750) 32(4400)
Color lb/ADT 3.9(460) 23.4(273) 37(510)
Alkaline Sewer
Flow m3/ADT(gpm) 5.3(930) 11.1(1060) 1(105)
Color lb/ADT(PCU) 5.7(487) 37.5(1052) 3.1(1740)
Total Ib/ADT(PCU) 9.6(491) 61.9 40.1
Total Bleach Plant Effluent Color- lb/ADT(PCU) 10.0(491) 38.1(516) 25.2 --59801
2005 Average
Total Non-Bleach Color lb/ADT
Flow m'/ADT(MGD) 8.3(4.72)
Color Ib/d 21112
Color Ib/ADT(PCU) 8.9(536)
Combined Influent Color(By Addition, lb/ADT 38.9 80912
Not Measured)
Final Effluent
Flow m'/ADT 53A(30.3) 55891
Color lb/ADT 23.5(221)
Color Limit in Final Effluent ppm,monthly 800
Color Limit in Final Effluent lb/ADT 2005 85.2 202.000
(Normalized,Approximate) prod'n
12.2 Benchmarking
Figure 12.1 summarizes the color from the mills in bar charts as sources to the influent and as final
effluent. The data has been normalized based on the total pulp produced on each site in air dry short
tons(ADT).
The data in Figure 12.1 are for the average production at the mills. Figures 12.2 and 12.3 show the
color sources separately for softwood and hardwood, assuming that the black liquor-originated color
is the same for both wood species and the bleach plant effluent color differs as reported by the mills
and detailed in Tables 12.1 through 12.4. Mill F and Mill G in these charts are mills outside the
survey,based on literature and EKONO file data.
The data in Figures 12.2 and 12.3 indicate that, depending on the specific mill conditions, hardwood
production would generate 30-45%less color than softwood.
Figure 12.4 is an attempt to separately benchmark the bleach plant effluent color in comparison with
literature data. All bleach lines in this study are included as well as other data from EKONO files.
National Council for Air and Stream Improvement

98 Interim Report
As shown,there is a wide spread among the data, depending on measurements and specific mill
conditions.
Figure 12.5 shows a benchmarking of the final effluent color of 30 bleached kraft mills in North
America and South America in EKONO's database. The mills include both market pulp mills and
integrated wood free mills. In this case the effluent color was normalized based on the total mill
output of paper(and market pulp, if any). Three of the study mills are among the lowest six
dischargers of color. When known,the share of hardwood pulp is indicated on the charts. The
highest discharges of effluent color occur among the softwood market pulp mills.
Color Sources and Effluent Color
so
®Unknown/Amplification
80
Oldentified BL color
70 ------------ -- - ❑Paper Mill __
0 CRP Waste
60 ------- ------------- o BP
o El Final Effluent
$ 50 ----' --
a
MIM
$ 40 -------------------- -------- --------------------
0
U
x _
20 10 r _________
------- II -- ---------
0 1 - � -
58% HWD 78 % HWD 55 % HWD 83 % HWD
Mill A Mill B Mill C Mill D
Figure 12.1 Color Sources and Effluent Color in Four Study Mills
National Council for Air and Stream Improvement

Interim Report gg
Color Sources of Influent Color, Hardwood
120
0 Unknown/Amplification
11Identified BL color
100 -------------- _____ ❑Paper Mill
W CRP
O BP
80 -------- ------- --------------
Q
60 ------------—
`o
0
U
S'
20 ------- -------71 ------- -------
Mill A Mill B Mill C Mill D/02 Mill D/No 02
Figure 12.2 Approximate Sources of Color from Hardwood Pulp Production
120 Color Sources of Influent Color, Softwood
0 Unknown/Amplification
17 Identified BL color
100 ------------ ------- ----- ❑Paper Mill
03 CRP
O BP
80 - - ------- -----
a Y
60
0
40 _____ _____ _____
>
20 �
0
Mill A Mill B Mill C Mill D Mill F MIII G
Figure 12.3 Approximate Sources of Color from Softwood Pulp Production
National Council for Air and Stream Improvement

100 Interim Report
ECF Bleach Plant Effluent Color as a Function of
Unbleached Kappa number (Paper Grade Pulp)
100
OMiIIA/02/BFR Q''sWD _
OMM A/02 ___ __ _____�__________,
80 �__________
0 9MillD/02 O'HWD
*Mill D/02 ---
70 ♦Mill D/No 02
_-_______y_________y----------
F 0Mill C/02 -'
I
aOMHIB 0
so OMills
i
C
O Mill H/02
0 40 -----1-------- -- ---'----------'----------
U •Mill H/No 02 i
i
30 ----- ------
:0
10 - - - - --- ------------------------------
0 0 5 10 15 20 25 30
Kappa
Figure 12.4 Bleach Plant Effluent Color at Varying Kappa
Benchmarking of Final Effluent Color
180
3 160 --------------------------
-----------------------------
0" 140
-------------------------
a 7t
120
i
Z 100 ----------------------------------------- 1 + t-
f
60 ------------------------------
0PPa ° r � iI
40
O 20
U 1Eiiiftl rtllE 0 i a
x x x x x x x x x x x x x x
cl d ¢ m m
Figure 12.5 Benchmarking of Final Effluent Color
National Council for Air and Stream Improvement

Interim Report 101
12.3 Color Technologies in the Study Mills
A questionnaire was sent out to the participating mills to collect data on successful and unsuccessful
color technologies that have been implemented or tried in the mills. The technologies were discussed
in follow-up interviews with the mills.
12.3.1 In-Process Technologies
Successful in-mill color technologies that were listed and commented on by the participating study
mills include the following:
• Maximize hardwood(possibly, but market-driven)
• Kappa to bleach plant(reduced by installing oxygen delignification,raised in digester for cost
savings, including yield).
• Oxygen delignification(high cost)
• "Mini 02" (Modeled in one mill—pending decision)
• New washer line
• Addition to existing washer line; lay-on roll additions, added stages, replace shower bars
• DF control on last washer before bleach plant; study of filtrate balance and washer
performance during good color period to set targets
• Vibratory knotters replacement, pressure knotters installation
• Knot filtrate recovered:
• In-process, post-discharge knotter drainage recovery being installed
• Knotter bins: small source of color,therefore not much reduction
• New screening equipment-closed screening(35%reduction in color)
• ECF conversion
• Kappa factor change
Study recommendations and process optimization
• Eop—peroxide reinforced extraction(no benefit if the Eop filtrate is recovered)
• Effluent-free reject handling
• Reject press project: Mechanical reliability and operational issues persist, SE color
reduction<500 lbs/day
• Reject press-successful
• Larger flash tanks, drop separators
• Evaporator boil-out procedures—effluent-free
Capture wash water in spill collection
• Equipment drainage before and during S/D—collected or sewered
• BMP program; collect and return process colored materials
• Pump seal and leak daily rounds
• BMP inspection program,mechanical seal project underway
• Overflow alarms
• Conductivity probes in sewers
• Daily/hourly color balance/alarm
In operation
• Operator education
• Conductivity cut-off limit
• Conductivity set points vary by process area
• Spills collection
• capture and recover color events when possible, or batch treat with chemicals
• Eop filtrate recycle to brown stock washing w/o pretreatment—requires chloride removal
National Council for Air and Stream Improvement

102 Interim Report
e bleach filtrate recycle process
• (BFR; ion exchange+chloride removal)
• Chloride buildup and washer scaling are ongoing operational problems
• Mechanical reliability of minerals removal process also an issue.
• Mechanical seals
• On pumps in black liquor areas to avoid clear water dilution of color materials
Unsuccessful in-mill color technologies include:
• OZ-two stages
• Pulp strength deterioration, insignificant savings at BFR closure rates>80%
• COz use for acidification of e.g., washwater
Excessive foaming
• Closing screen room existing line
in specialty mill-product quality constraints
• Ozone bleaching
Studied in detail, color increased, freeness decreased, and high operating cost
• Peroxide treatment of effluent streams
• Hwd Eo filtrate—no benefit with Eo filtrate recycle in place
• CRP-Severe foaming; high operating costs; no benefit to secondary effluent color,
increased cost of production
12.3.2. External Color Technologies
Table 12.6 summarizes the external color technologies that were judged as successful from a color
reduction point of view. Many of the technologies are,however, not justifiable from economic
feasibility or technically related issues,such as sludge handling,etc. The most promising external
technologies could be:
• Polymer treatment for brown color(applied in two mills). With appropriate sewer
segregation the resultant sludge could be combined with black liquor,burnt in the recovery
boiler and yield savings in make-up chemicals.
• Aluminum-based treatment in the activated sludge treatment(AST) plant(applied in one
mill). Possibly AST combined with activated carbon and sludge burning in the recovery
boiler.
The reported color reduction in the biological treatment(without chemical aid) in the mills using
activated sludge process varied between 15-40(50)%. An improved understanding of the color
reduction in the activated sludge process could provide an opportunity to reduce color for some mills.
Only one of the study mills has an aerated stabilization basin. That mill, like many other mills with
the ASB process, experiences a color reversion, i.e.,an increase of the color in the effluent treatment
system.
Table 12.7 summarizes the technologies that were unsuccessful from a color removal point of view or
that were technologically not reliable.
National Council for Air and Stream Improvement

Table 12.6 Tested Successful External Technologies as Reported by the Study Mills(Y=Yes)
Treated Stream Ap�licabilit anal sn _
CE
"o, a v e z' c l a C. ra III 3
M1i L I VCi
Y Y 1 !.—�I —_I1 2 Costly
Activated Carbon t i :—-- Y F I I E Costly
Y Y { Y j 3 1 5 i 2 Too costly due to too much AC required and
sludge disposal issues
Lime Treatment I Y Y I I Y I I 7 2 If time is reused in the process,could have
t roduct uality unpact:
30% Yes Lime used for neutralization prior to effl.trot.
Polymer precipitation i Yes i -- _I I 7in mill sewer containing black 100 T/d Y I 5 7 i 5 5 Successfully reduces 100 tons of color/day in black
liquor _ _ ' liquor trot
AI-based them.add Y Y i 5 !i 3 5 i 2°effluent with recycled sludge
Peroxide treatment ' Y 1 Y I q 2 5 2 For HWD Eo filtrate—not applicable if Eo filtrate
i i is recovered __
Treatment of Hwd Eo,successful pilot,cost of
Enhanced peroxide y y 4 0 3 4 TAML catalyst very high and prohibitive,may
treatment of E-stage effluent - revisit if TAML cost comes down
(TAML catalyst) Y Y 3 j_ I 5 2 ? Not yet ready for full scale operation due to
z ��� i
`}-- I manufacturing issues.
i Y I ! Y i 5 ! 7 _ 4 j —30%reduction in Ep filtrate _
Not recommended for full scale operation due to
cost/sludge issues;also concerns about effluent
Polyamine Y 3 Y 3 0 3 5 toxicity,works fine for batch treatment of high
color events captured in spare primary clarifier,30
to 65%color reduction
Scale: 0=no reduction.....5=very significant reduction
'Scale: 0=not technically feasible(technology not ready for full scale) ......5=technology ready and in use
Scale:0=not economically justified(high costs) ......5'=positive payback
°Scale:0=negative impact.....5=no or improving impact on quality

Table 12.7 Tested Unsuccessful External Technologies as Reported by the Study Mills(Y=yes)
Treated Stream ( Armlicabili anal sis
i C
I I
Uua 1 E 9 w t°. � e j N
I E S u 1 o E } u ° 1 ,c n e :' '
OWL ! U } II L
q i 4 i � 1 0 H G u w ,�,
i ; i ! V i � F w '_' e 9 r•i •G a E u e
ii Limited and inconsistent color removal,
Microfiltration(MF) Y 1 I t Y 3 1 2 I 3 ? requires pretreatment chemicals,high
j I i capital and operating cost,pilot
I ! I equipment had control problems
? I $ I Membrane fouling and failure,high
3 Ultrafiltration(UF) Y ; 1 Y 3 2 Y 2 1 1 2 3 capital and operating cost,not reliable
! _1_ _ _ _ i i for routine use
Reverse osmosis(RO) I Y 1 �_ �? 1 j 2 Expensive
It iIi ! EE ' Massive amounts of lime and lime
l 1 I ! f sludge if treating total wastewater
influent,high capital and operating cost.
Lime addition to 5B sewer from
I
Lime precipitation ! Y Y i 2 Y 1 0 3 1,2&4 i recovery had no statiscally significant
j i effect on SE color. Massive lime doses
I during kiln wash will remove color 6y
co-precipitation in primary clarifers.
S (( Not practical or economical for
eve!jday treatment of color.
E I I Causes pH problems at WTP,massive
AI-based precipitation i Y Y 2 ! Y 1 3 0 3 1 1 &2 ; amounts of sludge that dewaters poorly,
high capital and operating cost
, t
I Electrochemical Test stream not defined ; I Y 0 1 T
treatment
Scale:0=no reduction....-5=very significant reduction
Scale:0=not technically feasible(technology not ready for full scale) ......5=technology ready and in use
Scale: 0=not economically justified(high costs) ......5=positive payback
"Scale:0=negative impact.....5=no or improving impact on quality

1 Treated Stream _ Amilicabili analysis
— — —
C A
U fro
W E Cn e E O `_' '_' w o �• J:' II - '- II � w y v
w w` U m Fw W �
u �
t C H N C•• c U y o
Peroxide treatment with - Y ' 0 Y t I 1 0 0 ! q "Poor Man's TAML"-trial on Hwd Eo
iron activator was a bust,no color removal
Capital intensive with high operating
i Ozone treatment ( Y Y 3 Y 0 0 3 2 cost,effluent toxicity w/out ozone
' a 111 111 destruction step
I+ Y 0 Y 0 3 3 4 Fungus would not grow,trial was
White-rot fungus terminated
treatment - --- —
} ) Not sustainable in a lagoon
( 5 r 4
Y Mostly 1 1 2
environment.
CIO,treatment j -- —� ? CI02 added to hdwd Eo filtrate—no
' benefit with Eo filtrate recovery
Y(contro I Pilot scale !
CIO,treatment of CRP I Y i I i I trial on CRP process control issues,high chemical
waste i I I (color) foaming) I 0 2 3 i purge sheam costs,foaming
i ! I
I Other-UV-Peroxide Y ? I
? ! i 0 I � 4
1 i I
Scale: 0=no reduction.....5=very significant reduction
'Scale:0=not technically feasible(technology not ready for full scale) ......5=technology ready and in use
Scale:0=not economically justified(high costs) ......5=positive payback
Scale:0=negative impact.....5=no or improving impact on quality

106 Interim Report
13.0 SUMMARY AND CONCLUSIONS
At the request of five U.S. bleached kraft pulp mills,NCASI contracted with EKONO Inc.to
undertake a review of color control technologies and their applicability to modern kraft pulp mill
wastewater. The study was to build on an earlier color reduction study carried out in 1995 (Baird
1995).
The ultimate goal of the current study is to help kraft pulp mills to identify potential opportunities to
reduce effluent color. The specific objectives for each task follow.
Task I:Develop a comprehensive list of different in-mill (both"brown"and bleached)and effluent
treatment color reduction technologies and describe the technologies, their applicability to modem
bleached kraft pulp mills, and their impact on color loads
Task II: Conduct a mill-specific review of the color control measures and benchmark the effluent
color of the participating mills relative to other mills.
The color technologies addressed in this study are summarized in Table 13.1 together with an
assessment of their impact on color and some of the risks associated with them. Many of the
technologies are applied or have been tested in the study mills.
The color of the effluents from the study mills producing papergrade pulp is among the lowest
compared to a database of 30 bleached kraft mills, as shown in Figure 13.1.
Benchmarking of Final Effluent Color
180
160 -- --------------------------
------- --------
V
0 140 -- --------------------------------------------
IL
120 - -- - --- -
LL100 ----------------------------------------------
80 ------------------------- - -
60 -------
i 40 ----
UO I f
20 I ! I
0
Figure 13.1 Effluent Color Benchmarking
National Council for Air and Stream Improvement

Table 13.1 List of Color Reduction Technologies Included in the Evaluation
Technology Scale or Implementation Color reduction Risks
Reduce kappa number Full scale Major Product quality
Improved washing Full scale Major
Improvements in knotter and reject systems Full scale Moderate
Closed screen room water cycle Full scale Major
Reduced carryover of black liquor in Flash vapors Full scale Moderate to high
Improved evaporator boil-out and wash procedures Full scale Moderate to high
Convert to ECF bleaching Full scale Major
Use oxidizing bleach chemicals Full scale
Hexenuronic acid removal—w/o filtrate recovery Full scale No
Ozone—w/o filtrate recovery Full Scale No(lab data) Product quality
Eop w/o filtrate recovery Full scale Moderate Colorreversion
Peroxide(POP) No data—likely moderate Product quality,color reversion
Peracids w/o filtrate recovery Full scale No data—likely moderate Product quality,color reversion
Spill prevention Full scale Major
Spill collection and control Full scale Major
Spill system adequate sizing Study Major
Operator education Full scale Major
Eol•filtrate recirculation to brown stock washing 10—15%recycle-Full scale Low Corrosion/RB plugging
100%recycle-Study Potentially high Chloride removal process necessary,scaling
Bleach Filtrate Recycle(BFR)process with chloride and minerals removal 75%recycle in full scale Major Corrosion,scaling
Recycle of chloride free bleach filtrates-alkaline Full Scale Potentially high
Recycle of chloride free bleach filtrates-acid Study Potentially high Scaling,NPE accumulation
Membrane treatment of alkaline bleach filtrate Kraft-installations SID Potentially high N/A High costs—fouling,developing for kraft
Sulfite 02 filtrate-Full Scale Fouling problems
Ion exchange Kraft installation SID Potentially high High costs,fouling,No data for ECF
Electrodialysis Lab No data Cost
Activated carbon No full scale yet Moderate to high Carbon cost,sludge,toxicity
Chemical precipitation
Of brown color Full scale Potentially high Chemicals cost,Sludge
Of mill effluent Full scale High Chemicals cost,Sludge,Toxicity
Oxidation processes
Peroxide treatment of Eop filtrate Was operating in full scale Moderate Chemicals cost,color reversion
Peroxide TAML Lab and Pilot Potentially high Chemicals cost,color reversion,TAML supply in
industry scale
Ozone Lab Moderate to high at high dosages Ozone cost,toxicity,ozone
Ozone/UV Lab No data(probably significant) Ozone cost,toxicity,ozone
Ozone/Photocatalyst Lab No data(probably significant) Ozone cost,toxicity,ozone
Wet air oxidation Lab Significant at long residence times Full scale undeveloped
Evaporation and incineration of bleach plant effluent Full Scale for TCF No data
Studies for ECF Significant Chloride,corrosion,materials,condensate
composition
Fungustalgae/bacteria/enzymes Lab Moderate to high—no effect Sensitive to external factors
on some tests.

Interim Report 108
REFERENCES
Author I Name Journal j Month Year Pages
Air Liquide i Air Liquidewebmte i i............:.............�_
i Algehed,L; { Energy efficient pre-evaporation of bleach
Stromberg,J.; I plant filtrates:an economic evaluation of " TAPPI Journal 9 2000 : 55
Bemtsson,T. : various options
Almquist,C.B.;Boyd, I The Application of Photocamlysis on TO2
L.K. for Degrading COD in Paper Mill ! TAPPI Journal ! 9 i 2004 9-15
fWastewater
Axegard,P.;Stigsson, ( Efficient utilization of resources in the Breaking the Pulp Yield Barrier 11998 22-237
L. Eco-Cycle bleached kraft pulp mill Symposium,TAPPI Press
R3deslr6m,R. Wood Fuel Upgrading with a Chip Kidney VanneCork Service AB 2005
j Baird,J.K. { Color reduction in pulp and paper mill Master's Thesis,Pennsylvania 12 t 1995 I
wastewater treatment plant effluent ( State University, I
j
Baxter,G;Bums,B.;
Cloutier,J-N.; if Progressive System Closure at Tembec's j I
Paleologou,M; 1 Skookumchuck Kraft Pulp Mill,Part 1. 1 PAPTAC 90th Annual Meeting ? I C67j 2004 C6�-
Thompson,R; Project Description and Analysis t
f Towers,M.
Fate and Removal Kinetics on
Bdmbc,P.R.;Hall, Contaminants Contained in Evaporator pppTAC 86th Annual Meetin 2 2000 B67-
E,R. Condensate During Treatment for Reuse in g B71
a High Temperature Membrane Bioreactor j
I
! Removal of Methanol from Evaporator
4 Bdmbd,P.R.;Hall, g a High
Condensate Using h TemPerature i I � Z —1 t ! 1
I E.R Membrane Bioreactor: Determination of i Pulp&Paper Canada 3 2000 T7
Optimal Operating Temperature and t 776
I
{) � System Cost ,
i Cost Comparison Between a Hig-
tttii BBmbd,PA.;Hall, Temperature Membrane Bioreactor and a
1 E.R Steam Stripper for the Treatment of Foul TAPPI Journal 6 2001 62 j
iEvaporator Condensate for Reuse i
Bicknell,B;S en el, Comparison of Pollutant Loading from P
aper
P g ECF,TCF,and Ozone/Chlorine Dioxide International Non-Chlorine 3 I995
D.;Holdworth,T. Bleaching Bleaching Conference
! Using ozone to reduce recalcitrant '
Bijan,L.;Mohseni, ' compounds and to enhance 5th International Conference on r j
I Fate and Effects of Pulp and
M. biodegradbility of pulp and paper 6
j effluents Paper Mill Effluents f
I
I
i Blackwell,B; Options for Concentrating Kraft h Bleac I !
j Kirkpatrick,B.; ` TAPPI International Chemical A15_
Sm th,S;Simpson, Plant Effluent and Oxidizing the ` 4 1995
t y P Concentrate:PreliminaryProcess Design ` Recovery Conference I A35
Blue Ridge Paper Blue Ridge Paper Products:NPDES ! 2006
Products documents J
1 Brown,C.,Russer,A. ' i proceedings of the 1998 TAPPI I _
i Paleologou,M., Chloride removal from haft liquor using International Environmental 1998 r 1249
i Thompson,R.,and ion exchange technology Conference - I 1260
j lemon,R. I
Branlestrbm,K Eka Chemicals to start chlorine dioxide PAPERNET NEWS-Edition
1 i production in Rauma 7,(Nordisk Paper Journal) 5 2006 May
National Council for Air and Stream Improvement

Interim Report 109
Author i Name ! Journal Month Year Pages
—
Ultrafillration of alkaline fillmles to i I
Bryan[,et al. International Pulp Bleaching 229_ i
1 maximize partial closure of ECF bleach 1 6 1 1998 2
' plants I Confemece,Helsinki Finland 37
Economical options for handling kraft !
Bucher;W. Pulp and Paper 1999 59-63
fiberline rejects help recover fiber costs t
Caffarel-Mendez,S.;
Espana-Garcia,F.; i
Estrada-Vazquez,C.; !
j Integrated Biological Treatment of '
Ortega-Clemente,A.; Recalcitrant Effluents from Pulp Mills i Water Science and Technology j 2004 ! 156
Poggi-Vamldo,H.M.; '
Rinderknecht-Seijas,
N.
Colodette l.L.;Costa,
M.M.;Landim,A.; A Novel Bleaching Technology Adapted TAPPI Fall Technical
Silva,C.M.; _ to Partial Bleach Plant Closure Conference 9 2002
Ventorimrim=G. I
Colodette,J.L.;Costa, i
M.M.;Landim,A,; I A Novel Bleaching Technology Adapted paperija Pon 11 2005 �2
Silva,C.M.; to Partial Bleach Plant Closure I qqg
Wit
torim,G. i t
Effect of the pulping process on pulp International Pulp Bleaching
Colodette,J.L.;et al. ( bleachability with ECF,Z-ECF and TCF Conference 1998 61-70
bleaching sequences
Colodette,J.L.;Shah, Alternative Method to Apply Ozone in 7qI-
A,
Campos de Brito, : TAPPI Pulping Conference l 1 1999
. Salvador E.
ECF Bleaching: The Z3 Technology 757
Cooper,S.P.;Fols[er, ( Treatment of Lagoon Sludge,Primary
S A Him tratms Clarifier Sludge,and Bleach Plant Effluent CPPA 82nd Annual Meeting I 11gg6 B55-
E.G. Pttnann, ; by Supercritical Water Oxidation i B60
Evaporation of Acidic Effluent from ECF Dahl,O.;Niinimaki, TAPPI International 841-
( Bleaching,Reuse of the Condensate and 5 ''. 2000
J.
i Further Processing of the Condensate Environmental Conference f 852
:
Dahl,O.;Niinimaki; 1 Evaporation of Acidic Effluent from ECF TAPPI International ' 1323_
Kuo an ort6,H. Bleaching:Pilot Tests 4 1998 1123
P P_ _ g: 1 Environmental Conference
Dahl,O.;Tim,T.; Evaporation of Acidic Effluents from ECF
Kuopanportti,H.; I Bleaching:Effect of Feed pH on the I 1121-
Niinimaki,J.; ! Condensate and Re-Usability of the TAPPI Pulping Conference 10 1997 1129
Kiiskila,E. Condensate t
Geraldes,V.M.;Rosa, -
M.J.;Afonso,M.D.; -
Figueim,H.;Taborda, j
F.;Almeida,G.; Water Recovery from Bleached Pulp i j 117_
Journal (2 1996 TAPPI
Ganho,R.;Creusen, ! Effluents i i 124
R.;Hanemaaijer,I.; : y
Gaeta,S.;Amblard,
P.;Gavach,C.; j a
Guizard,C.
Dilek,F.B.;Tarim, Effectiveness of Algae in the Treatment of }
E.;Yetis,U. Wood-Based Pulp and Paper Industry ± Bioresource Technology 2002 1-5
water
Dube,M.;McLean, ! Reverse Osmosis Treatment Effects on - { A59-
R.;MacLatchy,D.; PAPTAC 85th Annual Meeting 1 1999
Savage,P. i Effluent Quality ± A62
Dufresne,R:; ? Comparison of Performance Between I
Lavallee,H.-C.; Membrane Bioreactor and Activated ( 131-
Lebrun,R.-E.;Lo S.- I Sludge Systems for the Treatment of TAPPI Journal 4 ! 1998 1135
N. Pulping Process Wastewaters
National CouncilforAir and Stream Improvement

Interim Report 110
Author _ _ i Name i Journal------- Month Year Pages
I EKONO rEKONO File Data
----_-- -- -_--_—_� -__-__
Epmey,M.;Fisher,S.; Successful Low Consistency Ozone rTAPPI Fall Technical
Kingsley,M.;Munro, i 9 2002 I —
F. Bleaching Mill Trial I Conference
j Est ,J.M I The Effect of Sulfide on Pulp and Paper Thesis,Auburn University, i
! y I Wastewater on Color Reversion } Aubum,Alabama. 12005
I Integrated Pollution Prevention and
EU-IPST Control(IPPC): Reference Document on I ! I
Best Available Techniques in the Pulp and European Commission 112 2001
Paper Industry
Fagemas,L.; Evaporation of Effluents from Pulp and I
' The Second Biennial Johan 129_
McKeough,P.; Paper Industry end Treatment of 9 1999
Ky1lUnen Evaporation Concentrates I Gullichsen Colloquium — Iq�
Faith,F.;Pfromm, ' International Pulp Bleaching • 209
1 Ultrafiltration of El Stage Filtrate + Conference,sponsored by
I P.H.;Sokol,J.C. PAPTAC 1210
Evaluation of ion exchange technology for U.S.Environmental Protection ! !
Fitch,J. i toxic and non-conventional pollutant ± Agency,Washington,DC. 1981 I
reduction in bleach plant effluents EPA/600/2-8I/006
�
Ganjidoust,H.;� Effect of Synthetic&Natural Coagulant I I 305-
7atsumi,K.; Sth IAWQ Symposium on 16 1996 I
on Lignin Removal from Pulp and Paper 311
Give shi,T.; I I Forest Industry Wastewaters
Gholian, I I I I
RN. Wastewater i I ---— I
Final Report,International Paper XL-2 Submitted to the U.S. l
Genco,l.M. Effluent Improvements Environmental Protection 2005
Agency and Maine DEP
Pulp Mill Process Closure:A Review of Gleadow,P.;Stratton, I NCASI Technical Bulletin No. I
S.C. Global Technology Developments and 860 5 2003 I-108
Mill Experiences in the 1990s
The Use of Ultrafiltration for COD ' TAPPI Intemational i 1167-
Greaves,R. Reduction in Pulp Mill Effluent-A Case I 4 1999 I
Study from Sweden Environmental Conference 11171 j
Union Camp Route to Bleach Plant CPPA TECH 97 Process 1 j
Griggs,B. Closure Closure
Course,Montreal,QC, 1997
C
Gullichsen,J. Liquorcal Pulping:Evaporation of Black Fa
pet Oy,Helsinki,Finland 2000
in
Hauptmann,E.G.; Strategies for Treating Mixtures of Bleach
Plant Effluents and Waste Water B71-
M.
Gaims,S.A.;Modell, Treatment Sludges by Supercritical Water { CPPA 80th Annual Meeting 2 1994_ B77�
Oxidation
I
Herstad-Svard,S.;
i lour,P.;Bryant,P.S.; ) Increasing the BioVeatability of ECF i 1 t65-
I I Bleaching Effluents by Ultrafiltration and TAPPI Pulping Conference 1 10 11998
Basta,J.;Aggarwal, i Partial Closure of Alkaline Filtrates 1176
Herstad Svard,S.; Increasing the Biotreatability of ECF 1
Jour,P.'Bryant,P.S.; TAPPI International 319-
Basta,J.;A anvil, Bleaching Effluents by Alkaline
FiltratsUltrafiltration and 14 1999
gg � Environmental Conference 33I
P Partial Closure of Alkaline Filtrates I 1
Higgs,R. Leaf River Flows into Deep Waters I Paper 10 1991 22
I Iron-TAML°Catalysts for Decolorization !. NCASI Effluent Color
Horwitz,C I of Alkaline Extractiob Stage Effluents 1 Management Workshop 12005
I Hodgson,A.T.; i • ! !
Effect of Tertiary Coagulation and
. Hitzroth,A.J. TAPPI Environmental -
i Stevenson,B.;Duff, ! Flocculation Technologies on Bleached ; Conference ! 1997 3313
07
07
S.J.B. I Kraft Mill Effluent Quality
National Council for Air and Stream Improvement

Interim Report 111
I Author _ i Name I Journal _ i Month_ Year Pages 1
Huber,P.;Jiang,Z.-
How does ozone in chlorine dioxide stages PAPTAC 92nd Annual I C219-
H;van Lierop,B.; 1 2 2006
Be R. affect bleached yield of kraft pulp? I Meeting _ ! i C224 t
m' ---
fI How much of the yield incuease from - 1
Jian ,et al. E modified pulping processes is retained Journal of Pulp and Paper 193_ {
g P P gP Science i 2002 197
during bleaching? l I
r Jiang,Z.-H.;Van ! Hexenumnic Acid Groups in Pulping and i 1 167-
Lierop,B=Berry,R. _ Bleaching Chemistry TAPPI Journal I 12000 175
[Iorm P Implementation of a novel pigment University of Oulu,Academic
recovery process for apaper mill I Dissertation 12 i 2003
nnedy,K.J.; !raham,B;Droste, ' Microtox and Ceriodaphnia dubia toxicity i 205- r
RL;Fernandes,L.; of BKME with powdered activated carbon Water SA 4 t 2000
Narba_tz,R. treatment 1 216
eTreatment of Pulp and Paper Industry
Effluent Using New Low Cost `TAPPI Minimum Effluent j 253-
Kois[inen,P. I Evaporation Technology with Polymeric Mills Symposium I ' 1996 1260
Heat Transfer Surfaces
i
Lachenal,D.;Churn, High temperature chlorine dioxide TAPPI Journal 8 2000 r 96
C. 1 bleaching of hardwood kraft pulp
Workshop on Emerging I
Liebergott,N. O Pulping and Bleaching
Ozone Bleaching:New Developments 1992
Technologies,North Carolina �
State University, NC
_ }
Wo Pulpinrksg and Bleachinghop on Emerging
Liebergott,N. Ozone bleaching:New Developments i 1995�
Technologies,North Carolina
State University Raleigh,NC
Lieber ott,N.et al. ' Modifying the bleaching process to
decrease AOX formation
g E Pulp&Paper Canada j 3 1991 1, 84-89
! � � �
i
Lindstrom,M.E.;
I A Comparison of Emerging Technologies: j t !
Ra ar,M. — Ho[Chlorine Dioxide Bleaching Versus Paper ja Puu ; 2 2004 39-44 f
Hot Acid Treatment f j
Closed-cycle totally chlorine free bleached
Louisiana Pacific, California Energy 3,N000 ion,
Samoa : kraft pulp production at Louisiana Pacifies 1 5 1 2000 1
Samoa pulp mill Final Report May 3,2000 I
I
Lovell,J.;Stein,R; Pilot Testing ofOzonation for Color TAPPI Environmental ! 1997 1 331-
i Schmader,R. Removal of Kraft Mill Effluent ( Conference 1337
Lundstrom,A.; ++ SPCI 7th international
Lundstrom, I Spill-a significant discharge that can be Conference on New Available
reduced Techniques,Stockholm, 2002 96-I00
Annergren,G. Sweden
I
IManuari,M.; ' NF270,A New Membrane Having i 1
i Promising Characteristics and Being I ! 107_
! al of Membrane Science ) 2004
.Nystrom,M.;Pekuri, Journal
116 !
Suitable for Treatment of Dilute Effluents I
{ I from the Paper Industry i
McCubbin,N. I Spill Control Part l:Assessing your Solutions!For People, 11 i 2001
siNation Processes and Paper,TAPPI
Solutions!For People,McCubbin,N. ; Spill Control Part II Processes and Paper,TAPPI f 12 2001 1
1 Messner,K.;Enter, [ 1 In Biotechnology in Pulp and
G.;Jaklin-Farcher,S.; r Paper Manufacture, !,
Treatment of bleach plant effluent the i
Boskosysky,P.; ! Applications and Fundamental
mycopor system ! I I
s
Regensberger,U.; i Investigations,Kirk,T.K., '
Blada,A. ; Chang,H.M.
Metsd-Botnia = Metsa Bomia,Rama mill EMAS report i 1
}j Metso Paper 1 Metso Paper website i
Munro,et al. 1 Operating experience with an ozone based Pulp a:Paper Canada 9 2001 35-39
ECF bleaching sequence {
National Council for Air and Stream Improvement

Interim Report 112
Author 1 Name j Journal Month _ or .' Pages
1 Narbad _
z RM.;�� —__ .�_._---..._.-- '. ---_ _. _
Droste R L
Fernandes L; f PACT"Process for Treatment of Kraft SthdAWQ Symposium on i 297_
' Kennedy,K.J.;Ball, Mill Effluent Forest Industry Wastewaters 6 1996 304
I D. I
NCASI Color NCASI Color workshop,2005
I Workshop p•
Information from Nomm Engineering
Nomm Engineering )
website �
Norberto de Pinho,
Mnhalma,M.;JoAo Rosa,M.;
Minhal 3 1
i Integration of Flota[ion/Ulhafiltration for I t 1115-
1 TAPPI Pulping Conference 10 1998
Taborda,F. 1 Treatment of Bleached Pulp Effluent 1124
1
Orori,B.O.,ENdgni, Decolorization of a Pulp and Paper Mill
L.,Rajah,M.S., Effluent in Webuye Kenya by a pulp&Paper Canada 3 2005 T50-
Situma,M.L.,and Combination of Electrochemical and
Ofusu-Asieda,K. Coagulation Methods I i T55
Pagoria,P. , Grande Prairie Effluent Color Growth II NCASI Effluent Color 2005 '
f Management Workshop
New Urethane Prepolymer Immobilized
Pallerla,S.; Fungal Bioreactor for Decolorization and TAPPI Journal 5 1996 161-
Chambers,R.P. Decolornation of Kraft Bleach Plant 61
Effluents
i
Pekkane M.; 1 Options to Close the Water Cycle of Pulp
n' ' and Paper Mills by Using Evaporation and TAPPI Minimum Effluent 1 1996 229
Kiiskila,E. • MiIIsSymposium 7 I
1 Condensate Reuse r 243 I
Pellinen,J.;Joyce, Dechlorination of high-molecular-weight • l0
T.W.;Chang,H-m. chlorolignin by white-rot fungus A TAPPI Journal 9 1988 194
chrysosporium _�
} i
Pokhrel,D.; 1 Treatment of Pulp and Paper Mill Science of the Total t 2004 37-56
Vimraghavan,T. Wastewater-A Review Environment j t
i ; I
!! Pryke,D.C,Kanters, j ECF Bleaching practices in Canada,Part 1. 1 I '
PAPTAC 86th Annual
Results of PAPTAC Bleaching Committee i 1999
C.7.,Tam,T. Meeting, 170
survey for softwood kraft pulps
1
ECF Bleaching practices in Canada,Part 1.
Pryke,D.C,Kanters, International Pulp Bleaching !
t CJ.,Tam,T. Analysis of PAPTAC Bleaching 2000
1 1 Committee survey for softwood kmft pulps 1 Conference
Influence ofhexenumnic acids on kraft Georgia Tech University,IPST I
Rag auskas'A. bleaching Publication 2000
I � 2-stage bleaching of oxygen delignified
Ragnar,M. ? HW kraft pulp to full brightness: pulp and Paper Canada 1 4 2005 s T97-
i Practically possible by changing DO to D' € TI01
` Robinson,S. Hydrogen Peroxide Finds Increased Use in
Solving
Mill Environmental Problems Pulp&Paper 11 1994 ! 115
Rodden,G. Valdivia has the world at its fingertips Pulp&Paper International i 1 2005a
Redden,G. SCA Ostrand builds a"natural ecosystem" Paper Loop ! 4 20056
Rottneros ? Rottneros website
Roux,A.;Bohmer, ; Improved Biological Treatment Through I TAPPI International , 815- j
V.J. 1 Chemical Pre-Treatment i I -
{ Environmental Conference 5 I{ 2000 � 832
SCA Ostrand increases the capacity and Svensk Pappers[idning II I996
f decreases the it
I
National Council for Air and Stream Improvement

Interim Report 113
Author I Name _ Journal _ _ Month ! Year ; Pages
Treatment of Paper Mill Effluent By the
i Schmidt,T.;Lange S. I Use of Ozone and Biological Systems: ` TAPPI International 5 2000 1765
I Large Scale Application at Lang Paper, ` Environmental Conference 776
Ettringen(Germany) I
The relationship between kappa number
Sevastyanova O,Li, I Journal of Pulp and Paper
i and oxidizable Strnetures in bleached kraft ; Science
{ L;Gellerstedt,G. pulps ? 8 2002
Semeniuk,S.; _--_- -
C. .-
Chandrasekman,S.; I Recent Environmental Improvements at I TAPPI International I 1
i Smith,K.;Munro,F.; j the Espanola Mill ! Environmental Conference 4 2002 I
Craig,D.
Shawwa,A.R.;Smith, Color and Chlorinated Organics Revmoval ? 745
D.W.;Sego,D.C. from Pulp Mills Wastewater Using ' Water Research 3 1 2001
Activated Petroleum Coke 1749 I
Sierka,R A:;Cooper Ultraflltration and Reverse Osmosis j 5th IAWQ Symposium on pg_
S P.;Pagoria,P S $ Treatment of an Acid Stage Wastewater Forest Industry Wastewaters 6 1996 180
I I
Simmers,S i Color Treatment Evaluations NCASI Effluent Color ' 2005 I
— ---- Management Workshop
I ! Investigations on reduction of colour from t. Journal of Water Supply: ' I
Singh,T.S. pulp and paper mill effluent by activated Research and Technology- 2006 i I-10
I coconutjute carbon j AQUA
rSmith,R.A.;Walley, ' High and Dry:Leaf River Survives Solutions!For People,
l 12002 75-78
T.B. _---, I Drought Through Planning,Teamwork I Processes and Paper
j Socha Cell Sedm,M6nsteras and Vard mills,EMAS I i.
reports for 2002,2003 ! 2002 i
—_-__ --�—
I Sonnenberg,L.B.; Photolysis of Chlorinated Organic I I !
TAPPI International 359-
I Eames,R.C.;Watts, I Compounds in Bleach Plant Effluents q j 19gq I
Environmental Conference 367 1 S.J.;Contra,M.A. Exposed to Sunlight I I 1 I
in _ 1...___
Springer,A.M.;Hand, 1 Electrochemical Removal of Color and ? TAPPI Journal 12 1995 85-92
V.C.;Jarvis,T.S. [ Toxicity from Bleached Kraft Effluents
Stevenson,Brian; s -------
Smyth,Sean;Caffrey, Removal of Colour from Kraft Mill CPPA Western Branch Spring j 5 1995 ! Session
Andrew;Acketl, I Effluent-A Canadian First Conference 5.2
Ch_a_r_lie i _ i
1 Ecocyclic Pulp Mill,Final Report 1996-
E STFI i 2002 STFI,Stockholm,Sweden I
G------
Advanced Effluent Treatment-
; Syvapurq T. j Envirotech Sympo'94:Finland `: 3 1994 taper ?
FenanoTriox _ 111 j
Torrades,Fq Peral J.;
Removal of Organic Contaminants in
Perez,M.;Domench, I
X.;Hortal,J.A.G.; Bleached Kraft Effluents Using I TAPPI Journal 16 2001 i 63
I Heterogeneous Photocatalysis and Ozone
Riva,M.C. I
I Tsai S.-P.;Henry ! Recycling of water in bleached kraft pulp TAPPI International 831_ j
M P, o,A T + mills using electrodialysis I Environmental Conference 1 4
Pfromm, 1999 838
,P._.__.
I US Filter 1 US Filter Website
Valhila,011i; --
Pollaneo,Taina; Experience in Evaporation ofTCF Bleach ! —�—
i Kiiskila,Eikki; i Plant Filtrates I TAPPI Pulping Conference 10 1995 :' 69
Kuusio,Mario I
van Lierop,B, High Temperature Extraction;
Faubert,M,Sacciadis j Understanding the Interaction of Variable Tappi Fall Technical I 2004
i S,Jiang,Z-H ! through Modeling i Conference 1
Vuorenvina,K., I --
Panula-Ontto,S.; I Peracetic acid in bleaching of softwood International Pulp Fuhrtnanq A. I Bleachin
P ( j 55225Conference,Helsinki,Finlarp nd I- !
National Council for Air and Stream Improvement

Interim Report 114
Author Name + Journal _ ' Mo_nth ' Year Pages
Vuorinen,T.; r ( "I
'a erstr6m;Buchert, 1 Selective Hydrolysis of Hexenuronic Acid j I
g ' Groups and its Application in ECF and , Journal of Pulp&Paper 5 1999 I55
1.;Tenkaneq M.; I ; Science(JPPS) j ! 162
Teleman,A. TCF Bleaching of Kraft Pulps I —
Webb,L. Kidney Technology Brings Success Pulp&Paper international 4 2002 28-32
j ----�
---I I TAPPI Minimum Effluent j
Wickstrom,P. Membrane filtration in the closed-cycle I
Mills Symposium i 1997 79-83
I bleach plant I i
I Wingate,K.G.;
Robinson,M.J.; Colour and AOX Removal from Bleached TAPPI International - I
' Sturbridge,T.R.; j Kraft Mill Wastewaters Using Catalytic Environmental Health and 4 ! 2001 I
I Collins,T.1.;Wright, 1 Peroxide Activators Safety Conference
i L.J.
Wingate,K.G.: t 3
Application of TAML Catalysts to
Sluthridge,T:R.; i I 255- I
Wright,L.J.;Horwitz, I Remove Colour from Pulp and Paper Mill i Water Science and Technology 2004 Effluents 260
C.P.;Collins,T.J. j
Optimizing Conditions for Maximum I j Wohlgemuth,G;Lam I I B169-
S.;Willis,D. I Efficiency and Minimum Effluent Colour i CPPA 83rd Annual Meeting i I 11997 Ij BI74 '
I I at Weyerhaeuser,Grande Prairie i
Optimizing Conditions at the
Wohlgemuth,G.; Weyerhaeuser,Grande Pairie Waft Bleach Pulp&Paper Canada 7 1999 T211
Lam,S.;Willis,D. Plant for Maximum Bleaching Efficiency nil
and Minimum Effluent Color
Wohlgemuth,G.; -----
Mannisto,F.; Reduction of manganese in brown stock- I TAPPI Pulping Conference i 2001
Yaraskawitb,I.;Jack, laboratory study and mill trial results i
D.
Yohon,J.and Patrick, West Coast Pulp Making:Evergreen Pulp Nov/
K. becomes the industry's first Chinese- PaperAge Dec 2005 24-28
owned mill complex in the U.S.
Treatment of Combined Bleach Plant
} Zhang,Q.;Chuang, I Effluents via Wet Oxidation over a Pd-Pt- j Environmental Science and 1999 13641-
K.T. Ce/Alumina Catalyst Technology I lj 3644
t th. g { --. .._......_..... ..
Zhou,H.;Smith, Process Parameter Development for Sth IAWQ Symposium on 6 1996 267
D.W. Ozonation ofKraft Pulp Mill Effluents Forest Industry Wastewaters _ 274
Zhou,H.;Smi Advanced technolo ies in water and I
I D.W. wastewater treatrnent i J.Environ.Eng.Sci. t 2002 _-- 264
National Council for Air and Stream Improvement