NC0000272_Workscope_StudyAquaticRes_19950601

it
I�
WORKSCOPE FOR A STUDY
OF THE AQUATIC RESOURCES AND
WATER QUALITY OF THE PIGEON RIVER
FOR CHAMPION INTERNATIONAL CORPORATION
Prepared for:
IIChampion International Corporation
Main Street
Canton, North Carolina 28716
II Prepared by:
EA Engineering, Science, and Technology, Inc.
11019 McCormick Road
Hunt Valley, Maryland 21031
(� and
I� 444 Lake Cook Road, Suite 18
Deerfield, IL 60015
I�
June 1995

1. INTRODUCTION
This document contains EA Engineering, Science, and Technology's workscope for conducting
a biological and water quality survey of the Pigeon River in 1995. EA originally surveyed the
aquatic resources of the Pigeon River in 1987 (EA 1988). The 1987 synoptic survey consisted
of the following elements: water quality, habitat, hydrography, hydrology, light attenuation,
periphyton, benthos, and fish. These elements will be repeated in 1995 except that the
hydrography and hydrology elements will rely on current USGS data or data collected in 1987.
In addition to repeating the elements of the 1987 study, a new element, fish health assessment,
has been added to the 1995 study. A tissue study to determine body burdens of dioxin will be
conducted concurrent with the synoptic survey.
The objectives of the 1995 synoptic study are:
(1) describe current biological and water quality conditions in the Pigeon River,
(2) compare these conditions with those documented in 1987, and
(3) document the extent to which recent improvements at the mill have resulted in
improved biological or water quality conditions in the Pigeon River.
1

rI
' 2. WORK SCOPE
I
f 2.1 STUDY AREA
The study area will encompass both the North Carolina and Tennessee portions of the Pigeon
River. The mainstem of the Pigeon River will be sampled from RM 64.5 upstream of Canton,
(I NC to RM 19.3 in Tennessee. Three tributaries will also be sampled along this reach of the
ll river. Biological sampling stations are presented in Table 1 and water quality stations are
presented in Table 2.
Biological 1 Stations,
[' All biological stations sampled in 1987 (EA 1988) will again be sampled in 1995 except that
the station at RM 65.5, which originally was sampled only for fish, has been eliminated and,
the downstream-most station sampled in 1987 (RM 7.8) also has been eliminated (Table 1). A
mainstem station has been added at RM 54 to better assess possible near field impacts of the
Waynesville STP. A mainstem station has also been added at RM 55.5. This location is
r' upstream of the Waynesville STP and Richland Creek, but downstream of the Clyde STP.
`` This station will allow better resolution of differences among these point sources and will
allow a direct comparison with data collected by TVA biologists from this site in 1990 (Saylor
et al.1994).
Water Quality Stations
I� Eighteen water quality stations will be sampled (Table 2), 16 in North Carolina and the two
mainstem Pigeon River stations in Tennessee.
h �
2.2 TECHNICAL COMPONENTS
(� The workscope is divided into eight technical components: light attenuation, habitat, water
quality, benthic macroinvertebrates, fish, periphyton, dioxin tissue study, and fish health
assessment. Methods used to measure or evaluate each of these components are presented
below. The methods described are consistent with those used during the 1987 synoptic survey
(EA 1988). In some cases, minor modifications of the methods used in 1987 have been made
to allow them to conform to the standard methods recommended by North Carolina DEM
(1995).
2.1.1 Light Attenuation Model
In the 1987 study, the light attenuation model was applied on a daily basis for a 3 year June-
October period at four stations in both the North Carolina and Tennessee portions of the
Pigeon River. Each station included a pool and riffle transect. Data requirements for the light
attenuation model include color, turbidity, light extinction coefficients and water depth at each
station. During the 3 year period, water depths at each transect were calculated based upon
USGS flow data. Daily true color data were available from Canton Mill's Pigeon River
monitoring program at the North Carolina stations and weekly color data were available in
Tennessee. The light extinction coefficients used in the model were based upon field
measurements of light attenuation over the water column.
2

Table 2. Water Quality Sampling Stations Along the Pigeon River and Tributaries
River
Location M&
Above Canton, River Road bridge (first) 64.5
Canton mill effluent 63.3
Fiberville bridge 62.9
Above Clyde, right bank 59.0
Clyde STP effluent 57.1
Below Clyde, Hyder Mountain Road bridge 55.5
Richland Creek, bridge at creek mile 0.2 54.9
Waynesville STP effluent 54.8
Above Crabtree Creek, local bridge off Route 209 52.2
Crabtree Creek, private bridge at creek mile 0.15 49.8
Riverside, Ferguson Bridge 48.3
Jonathan Creek, bridge at creek mile 0.9 46.0
USGS Hepco Gauging Station, left bank 45.1
Fines Creek, left bank at creek mile 0.3 42.7
New Hepco Bridge 42.6
Cataloochee Creek, right bank at creek mile 1.0 38.1
Below powerhouse, bridge at I-40 exit to Waterville 24.7
Above Cosby Creek, local bridge at Groundhog Creek 19.3
4

For the 1995 study, the June-October data for 1993, 1994, and 1995 will be analyzed using
the light attenuation model. The 1995 study will use a subset of the pool and riffle transects
used in the previous report. In this manner, improvements in river conditions will be directly
observable. The data requirements and application of the light attenuation model are discussed
in more detail below.
Hydrology - North Carolina
In the 1987 study, drainage area scaling relationships were developed based upon daily USGS
flows at Canton (RM 64.9) and Hepco (RM 45.1) for all stations between the Canton Mill and
Walters Lake Dam. For the current study, daily USGS flow records at Canton and Hepco will
be obtained for the period from 1991 through the completion of the field surveys. A staff gage
will be installed at each station used in the field program. A comparison between staff
readings and extrapolated USGS flows will confirm that the stage-discharge relationships
established in 1987 are still appropriate. These relationships will be used to provide daily
cross-sectional average depths at each transect as a function of river flow.
Hydrology - Tennessee
The hydraulic regime in Tennessee is complex because of flow variations caused by the
Carolina Power & Light (CPL) powerhouse. In the 1987 study, a flow model was developed
that could estimate hourly flows at the stations in the Tennessee portion of the river based upon
reported hourly USGS flows at Newport. During the 1995 study, hourly USGS Newport
flows will be obtained from 1991 to the completion of the 1995 field program. Hourly CPL
powerhouse flows will also be obtained for a minimum 3 month period coinciding with the-
field survey program. The flow model will be executed for this 3 month period in order to
compare predicted and actual hourly CPL powerhouse flows as re-confirmation of the model.
The flow model will then be used to generate an hourly record for the 1993-1995 June-October
period which will be converted to transect depth using the previously established relationships.
A staff gage will also be installed at each station in the Tennessee portion of the River.
Readings tabulated at the time of data collection will assist in quantifying the hydraulic regime
and in verifying the phase relationship between stations of the propagation of the flow front.
Light Extinction Coefficients
Two field surveys will be performed at all stations where the light attenuation model is to be
applied. Data collection will include turbidity, suspended solids, true color, and apparent
color. The color measurements will be made by the platinum-cobalt method (NCASI 1971) as
performed in 1987. Field measurement will also include light attenuation measurements
throughout the water column using a Lamda Li-170 photometer or equivalent. Multiple
regression techniques will be used to calculate turbidity attenuation coefficients and color
extinction coefficients from the field-measured attenuation coefficients.
5

' Application of the Light Attenuation Model
The light attenuation model will be executed for the most recent 3 year June-October period
(i.e., 1993-1995) based upon daily color and flow data. Currently the mill measures color 2
times a week at the mill intake, RM 42.7 (North Carolina) and at RM 24.7 (1-40, Tennessee).
Daily color data are collected at the plant discharge. The daily color and turbidity values for
model input will be determined based on the available data.
In 1987 it was observed that the decrease in color along the North Carolina portion of the
Pigeon River was highly correlated with increasing flow due to increasing drainage area.
Therefore the 1993-1995 June-October daily color data set will be constructed at each station
using the daily discharge color values and dilution flows scaled from the USGS data. This
daily color data set will be verified with the twice weekly values collected at RM 42.6. Any
observable systematic difference will be included back in the scaling procedures.
In the Tennessee portion of the Pigeon River, the currently available twice weekly color values
will be analyzed as a function of flow and the upstream color loading for the period 1993-
1995. The light attenuation model for the Tennessee portion of the Pigeon River will use an
appropriate color relationship as determined by this analysis.
The results of light attenuation modeling in the 1987 report were presented primarily as
frequency of occurrence tables of percent of incident light reaching the river bottom. As a
result, the 1993-1995 attenuation modeling results will be presented in a similar manner for the
same pool and riffle stations. This will allow improvements in river conditions to be directly
discernable.
2.2.2 Habitat
In 1987, EA evaluated habitat using methods described by Ball (1982) and Platts et al. (1983)
because NC DEM had not established stream habitat evaluation procedures. NC DEM (1995)
has recently developed such procedures for coastal plains streams. We will determine whether
these procedures can be used or modified for the mountain ecoregion in which the Pigeon
River is located. If they can not be modified for the mountain ecoregion, we will either use
the evaluation protocol we used in 1987 or use one of the evaluation procedures developed
recently such as the Qualitative Habitat Evaluation Procedure (Rankin 1989). Regardless of
the procedure used (all are relatively similar), the following habitat variables will be evaluated:
Substrate
Cover
Embeddedness
Velocity
Channel morphology
Riffle/Pool quality
Bank Stability
Riparian vegetation
6

Each variable will be scored, these scores will be summed, and the resultant composite scores
compared among locations and relative to the score at the mainstem reference location (RM
64.5).
2.2.3 Water Quality
During July 1987, Champion and EA conducted a synoptic survey of water quality in the
Pigeon River from Canton to the French Broad River (EA 1988). Water quality parameters
measured during 1987, which also will be measured during the 1995 study, are listed in Table
3. In 1987, samples were collected from 13 mainstream station, eight tributaries, and four
treatment plant effluents (i.e., 25 stations). Sampling took approximately 2.5 days, and the
entire survey was conducted twice over a 5-day period. In 1995, water quality samples will be
collected from the 16 North Carolina and two Tennessee stations listed in Table 2.
Prior to initiating the sampling program, EA will review North Carolina water quality
standards and prepare a brief SOP for Champion and the field crew. The plan will provide
specific sampling locations, sample collection procedures, in situ monitoring procedures, and
QA/QC procedures.
The 1995 water quality survey will be conducted over a 4 day period (2 surveys, each
expected to take 2 days). Temperature, pH, conductivity, and dissolved oxygen will be
measured in the field using a Hydrolab water quality analyzer or equivalent. All other samples
will be preserved immediately (per EPA specifications) and will be analyzed within EPA-
recommended holding times.
The water quality survey will be conducted during summer, base-flow conditions to assess the
effects of effluent discharges and tributaries on the river without the masking effects of
stormwater runoff. Contemporary conditions will be compared to conditions prior to the
modernization of Champion's Canton Mill by comparing 1995 survey results with 1987
results. Other acceptable available data will also be included in these comparisons. In the
final report, we will compare the contemporary water quality data to state standards, as well as
present a 1987 vs 1995 parameter by parameter data comparison, emphasizing water quality
change over time.
2.2.4 Benthic Community
EA studied the benthic communities of the Pigeon River and Richland Creek during July 1987,
as part of the Pigeon River synoptic survey (EA 1988). Samples were collected from nine
mainstem river stations and one tributary (Richland Creek) station. Survey methods and
results are presented and summarized in EA (1988).
Benthic field and laboratory survey methods utilized in 1987 will be used as a technical
template for the 1995 study in order to allow a comparison of the past and present benthic
community. Benthos samples will be collected from all six of the original North Carolina
mainstem stations, two of the three original Tennessee mainstem stations, and new mainstem
7

Table 3. Water quality parameters measured during the 1987 Pigeon River Synoptic Study and for the 1995 study
Organics Metals Inorganic Nonmetals Physical Determinations
Biochemical oxygen demand* Aluminum Alkalinity True color**
Chemical oxygen demand Calcium Chloride Apparent color**
Surfactants (MBAS) Magnesium Total hardness Total filterable residue
(Total dissolved solids)
Zinc Ammonia-Nitrogen Total non-filterable residue
(Total suspended solids)
Total Kjeldahl nitrogen Total volatile non-filterable
CO residue (total volatile
suspended solids)
Nitrate + nitrite-nitrogen Turbidity
Phosphorus Temperature
Sulfate pH
Conductivity
Dissolved oxygen
* Collected by EA but analyzed by Pace Environmental, Asheville in order to meet requested holding time.
** Collected by EA but measured by Canton Mill.

stations at RM 55.5 and RM 54 (Table 1). Three tributaries (Richland Creek, Jonathan
Creek, and Fines Creek) will also be sampled.
As in the 1987 study, stations will be sampled qualitatively with kick nets, sweep nets, and
fine mesh samplers. The methods used in 1987 are consistent with those currently
recommended by NC DEM (1995). A total of ten qualitative samples will be collected from
each sampling station; two kick net samples, three sweep net samples, three fine-mesh samples
(two rock/log and one sand/fines), one leaf-pack , and one visual search (considered as one
sample). Kick net samples (1 m2 net) will be collected from riffle areas of differing velocities;
sweep net samples will be obtained from bank areas and macrophyte beds; and fine mesh
samplers will be used to sample smaller invertebrates such as Chironomidae larvae.
Organisms collected in qualitative samples will be visually picked in the field in proportion to
their abundance (however it will not be necessary to remove all organisms). Samples will be
preserved and labeled immediately following field sorting, and subsequently transported to the
laboratory for identification. All benthic macroinvertebrates (taken both quantitatively and
qualitatively) including Oligochaetes and Chironomidae larvae will be identified to the lowest
feasible taxon.
NC DEM (1995) methodologies do not require collection of quantitative samples such as those
collected by the Hess sampler used in 1987. Although not required, triplicate Hess samples
will also be collected at each mainstem location to allow comparison with 1987 quantitative
data.
The summer benthic survey is intended to describe the community composition, abundance,
and distribution during summer, base-flow conditions. The approach to assessing the
community will focus on the benthos as a food base for higher trophic levels and we will use
the NC DEM bioclassifrcation scheme to determine the quality of the Pigeon River benthic
community. The analysis of the food base will include the interpretation of density (or
abundance) trends and biomass estimates as a measure of standing crop using data from the
Hess samples. The bioclassification analysis will be based on the qualitative samples to assess
quality of the community at each sampling location and will follow procedures described by
NC DEM (1995). Contemporary community conditions will be compared to conditions
encountered prior to improvements at Champion's treatment plant by comparing results of the
1995 benthic survey to 1987 survey results. In the final synoptic report, EA will provide a
comparison of present and past (i.e., 1987) benthic community data, and will discuss and
emphasize community change over time (particularly since the Canton Mill effluent
reductions/improvements). Comparisons will also be made to recently collected NC DEM
data from the Pigeon River if any such data are available.
2.2.5 Fish Community
EA studied the fish communities of the Pigeon River and Richland Creek during July 1987, as
part of the Pigeon River synoptic survey (EA 1988). Fishes were collected from ten mainstem
river stations and one tributary (Richland Creek) station. Survey methods and results are
presented and summarized in EA (1988).
9

Fisheries field and laboratory methods utilized in 1987 will be used as a technical template for
the 1995 study in order to allow a comparison of the past and present fish community. Fish
will be collected from eight of the original ten mainstem stations. The most upstream (i.e.,
RM 65.5) and downstream (i.e., RM 7.8) of the original stations will not be sampled.
Mainstem locations will be added at RM 54 and at RM 55.5 (Table 1). Fishes will also be
collected from the following tributaries: Richland Creek, Jonathan Creek, and Fines Creek.
The 1995 fish community survey will be conducted over an approximate 10-12 day period. As
in the 1987 study, fish collection techniques will be standardized to provide a representative
qualitative sample and to allow a comparison of fish communities along the length of the
Pigeon River study reach. A pram-mounted electroshocker will be used to sample all available
habitats within a 150-200m reach at each station. We have found that our pram-mounted
electrofisher is considerably more effective in streams the size of the Pigeon River than the
backpack units recommended by NC DEM. Where necessary, a small electrofishing boat will
be used to sample deeper areas. Where available, approximately equal areas of both riffle and
pool habitats will be sampled within each sample reach. The stream margin and backwater
areas (i.e., potential nursery/refugia areas) at each station will be sampled with a 3.05 x 1.22m
straight seine with 0.32 cm (1/8") square mesh or similar seine. Seining will be conducted to
complement the electrofishing. Finally, a combination kick-seining/electrofishing method may
be used to sample riffle-dwelling species that may not have been taken using the other
collection methods. Similar to the approach used by Saylor et al. (1994), fish sampling at
each location will cover all habitat types and will continue'until no new species are being
collected.
All fish will be identified to species and counted. Sportfish will be individually weighed and
measured up to a maximum of 15 individuals per station. Forage, rough, and nongame fishes
will be batch weighed only. Incidence of parasites, disease, and other morphological
anomalies will also be noted. As needed, specimens will be preserved and returned to the
laboratory for confirmation of identification, whereas all others will be released on site.
The 1995 fisheries survey is intended to describe community composition, abundance, and
distribution during summer, base-flow conditions. As in the 1987 study, fisheries data will be
tabulated to characterize individual fish community attributes (i.e., abundance, distribution,
species richness) and species-specific parameters (i.e., coefficients of condition or relative
weight for sportfish, evidence of successful reproduction). Fish community data will also be
incorporated in North Carolina's Index of Biotic Integrity (IBI) in order to characterize the
biotic condition of the surveyed length of the Pigeon River. Results of the IBI analysis will
determine whether the biotic condition of the sampling locations approximate, deviate
somewhat, or deviate strongly from conditions expected in an unperturbed system. The IBI
components currently used by NC DEM have been modified somewhat from those used in
1987. IBI scores in 1995 will be calculated using current (NC DEM 1995) guidance and, if
the original data can be retrieved, the 1987 scores will be recalculated using current scoring
procedures. The relative health of the study area will be discussed by summarizing and
synthesizing the community-level and IBI statistics. Contemporary conditions will be
compared to conditions encountered prior to the modernization of Champion's Canton Mill.
10

This will be accomplished by comparing results of the 1995 fisheries survey to 1987 survey
results and to results presented by Saylor et al. (1994). In the final report, EA will provide a
comparison of present and past fish community data, and will discuss and emphasize
community change over time (e.g., since the Canton Mill modifications).
2.2.6 Periphyton Methods
Periphyton collections will be made at all ten Pigeon River mainstem biological sampling
stations (Table 1). Methods of collection, sample analyses, data analyses, and interpretation
will be the same as used during the 1987 synoptic survey.
Sample Collection
Natural substrates (flat rocks) will be used for the collections. One month prior to the period
scheduled for the synoptic survey, eight smooth flat rocks will be collected from the banks of
the river at each station.
One side of each rock will be thoroughly cleaned with a wire brush and the other side of the
rock will be painted red with fast-drying acrylic paint. A Li-Cor submarine photometer,
Model Li-170 or equivalent, will be used to determine the depth of substrate placement. Two
substrates will be placed at a depth receiving greater than 30-35 percent of incident light. Four
substrates will be placed at a depth receiving 10-30 percent incident light and two substrates
will be placed at a depth receiving approximately 10 percent incident light. It is anticipated
that at some stations the river will not be deep enough to place substrates at the lower light
levels. At all stations rocks will be placed at measured distances along a shore to shore
transect. All substrates will be placed in riffle areas in an attempt to eliminate velocity as a
significant variable.
One month after substrate placement, during the period of the synoptic survey, each station
will again be visited for substrate (sample) retrieval. At each station, the goal will be to
retrieve one substrate from the high-light depth, two substrates from the intermediate-light
depth, and one substrate from the low-light depth (i.e., half of the substrates). If a thorough
search of a particular depth at a particular station does not yield a marked substrate, a natural
substrate of similar characteristics will be taken from the same depth for analysis.
Once located, an epilithic bar clamp sampler (EBCS) will be attached to the substrate and,
with flexible seal in place, the rock and EBCS will be removed from the river. The flexible
seal will then be removed and a trimmed paint brush will be used to scrub the rock surface to
completely loosen the attached periphyton. A large bulb pipette (i.e., turkey barter) will be
used to transfer the periphyton slurry from the well of the EBCS to a labeled 4-oz glass jar.
The sample will be kept on ice until return to the filtration site. Standard water quality
parameters (DO, pH, specific conductance, and temperature) will be taken at each station
using a Hydrolab Model 4041 Water Quality Analyzer or equivalent. A light profile will also
be taken using a submarine photometer.
11

Field Subsamnling for Chlorophyll a Phaeophytin a. Biomass and Periohyton Identification
At the filtration site, each sample will be brought up to a volume of 100 ml using a graduated
cylinder and distilled water following which the sample will be shaken vigorously for one
minute or more to homogenize the sample. A 2-5 ml subsample (depending on sample
thickness) will then be removed using a pipet and filtered through a GF/C glass-fiber filter
under low (< 10 in. Hg) vacuum. During the final phase of filtering, 0.2 ml of a 1 percent
mgCO3 suspension will be added to buffer the sample. After all water has passed through the
filter, the filter will be wrapped in bibulous paper, labeled, and placed in a dark jar filled with
desiccant.
A 10-30 ml subsample for biomass (AFDW) will be removed using a pipette and filtered
through a pre-rinsed, dried, and pre-weighed GF/C glass-fiber filter under low (< 10 in. Hg)
pressure. Each filter will be placed in a small plastic petri-dish, labeled, and placed in a dark
jar containing desiccant. Both chlorophyll a and biomass samples will be frozen on dry ice
during transport to the laboratory.
The remaining sample will be returned to its original glass jar and preserved with "M'" algal
preservative (American Public Health Association 1985) for algal identification and
enumeration.
Laboratory Analysis of Chlorophyll a, Phaeophytin as and Biomass
The samples for chlorophyll a and phaeophytin a analysis will be delivered to EA Labs where
the samples will be inspected, compared with the chain-of-custody records, logged, and
assigned laboratory accession numbers. Determinations of chlorophyll a and phaeophytin a
will be made according to American Public Health Association (1985) Method 1002G. All
appropriate quality control and quality assurance criteria will be employed during the analysis.
The sample for biomass measured as ash-free dry weight (AFDW) determination will be
delivered to EA Labs where biomass (AFDW) will be determined using American Public
Health Association (1985) Method 1002H.
Taxonomic Analysis
The inverted microscope method (Lund et al. 1985) will be used to identify and enumerate
periphyton collected from optimal light regimes at each location. Each preserved sample will
be mixed, and an aliquot will be transferred into a settling chamber designed for use on a Zeiss
Standard UPL inverted microscope. The volume of aliquot settled generally will be from 0.1
to 1.0 ml depending upon algal density and the amount of silt and detritus present. Algae will
be identified and enumerated at 1,000 x magnification under oil immersion, and reported as
number of cells (except some filamentous algae which will be reported as 10-jum lengths).
Analysis of each sample will be considered complete when approximately 400 reporting units
have been counted. However, a minimum of one radius and a maximum of two diameters of
the settling chamber will be examined for all samples. Periphyton densities (reporting
12

units/mm2) and total taxa identified at each location will be reported for all samples analyzed.
2.2.7 Fish Tissue Collections
Concurrent with the 1995 synoptic survey, EA will also collect fish for dioxin analyses as
required by the Mill's NPDES permit. Procedures will follow those described in the 1990
Study Plan (EA 1990) and updated during recent surveys (EA 1994). Key elements of the
tissue collections are described below.
Sampling Stations
Collections will be made at:
Station 1 RM 64.5
Station 2 RM 59.0
Station 3 RM 52.3
Station 4A RM 41.5
Station 4B RM 39.0
Station 5 RM 19.0
Except for Stations 4A and 4B (Walters Lake), these stations will be sampled as part of the
synoptic survey.
Equipmen
EA will use a 12-ft electrofishing boat and a Coffelt VVP-2C pram electrofishing unit as the
primary sampling gears at each station. If necessary, gill nets will be used to collect the
desired complement of fishes at the Walters Lake stations. The field crew will also be
equipped with a variety of other sampling gears (e.g., seines, trotlines, etc.) that will be used
if the primary techniques fail to yield the required number of species of fish.
Fish Collection Objectives and Sample Preparation
Collection methods, objectives, and sample preparation will follow EA Study Plan 11370.01
(February 1990) entitled "Study Plan for Dioxin Monitoring in Fish Tissue".
The goal of the fish collection effort will be to collect four composite samples at each of the
six sampling stations. Station-specific objectives include the collection of two sport fish fillet
composites, one bottom-feeder fillet composite, and one bottom-feeder whole body composite. .
Each composite should contain fillets or whole bodies from approximately 5 to 10 individual
specimens.
EA will prepare each sample in the field as per the preparation methods described in the study
plan. Following the filleting, wrapping, bagging, and labeling process, all samples will be
frozen and held on dry ice. Chain-of-custody forms will be prepared for the composite
13

samples collected at each station. All samples will be shipped via overnight courier to the
analytical lab chosen by Champion. As in the past, we will expect Champion to contract
directly with the analytical lab to analyze the samples collected.
Repo Pr-Qparation
EA will prepare a report separate from the synoptic survey report that will detail the sampling
methods used and effort expended at each tissue station, as well as a description of each fish
submitted for analysis. Analytical results will be reported as actual concentrations of TCDD
and TCDF isomers and the 2,3,7,8-TCDD Toxic Equivalent Values. Field and laboratory
data sheets will be included as a report appendix.
2.2.8 Fish Health Assessment
In 1992, Adams et al. (1993) evaluated the health of the Pigeon River fish community using
the Health Assessment Index (HAI). Adams et al. (1993) collected redbreast sunfish from
locations 10 and 70 km downstream of the Canton Mill effluent and compared HAI values at
these stations with reference stations on the Little River and Little Pigeon River. Adams et al.
(1993) found fish in the Pigeon River to be less "healthy" compared to the two reference
streams. However, Adams et al. did not collect any fish from upstream of the mill.
Furthermore, the fish at the two Pigeon River downstream sites were "healthier" than in an
average TVA reservoir. Moreover, the Adams et al. (1993) samples were collected prior to
the recent improvements at the mill, so any recent improvement in fish health would not be
apparent in their data. Furthermore, without data from upstream of the mill, it is not possible
to determine whether there actually was any decline downstream of the mill.
To address these issues, EA will collect approximately 20 redear sunfish at each mainstem
location and evaluate their health using the HAI. Each fish will be evaluated and scored on
site according to the 14 variables identified by Adams et al. (1993, Table 1). Mean scores will
be calculated for each station and station means will be compared statistically. We will also
collect approximately 20 redear sunfish from the two Walters Lake Stations for further
comparisons. Although we are specifying approximately 20 fish per station, we anticipate that
the number will vary somewhat among stations. If redbreast sunfish are common at a station,
we may collect more than 20 specimens, conversely if they are uncommon as few as 5-10 may
be collected. If a thorough search at a station yields fewer than 5 specimens, that station will
be deleted because of the enhanced probability of anomalous results associated with such a
small sample size.
2.3 REPORTS
Study results from all project components will be provided in a comprehensive synoptic report.
Included in the document will be text describing:
• methods for all tasks;
14

• results of light attenuation studies;
• habitat conditions at each sampling station;
• water quality survey results and compliance with state standards;
• benthic survey results and contemporary benthic community structure, standing crop,
community function, and health;
fish survey results and contemporary fish community composition, abundance,
distribution, and condition;
• periphyton biomass and composition;
• biological integrity of the study area;
• a comparison of 1987 water quality to the prevailing Pigeon River water quality;
• a comparison of 1987 Pigeon River periphyton, benthos and fish communities to
contemporary biotic communities, emphasizing community change over time (e.g.,
since the Canton Mill effluent reductions/improvements).
comparisons to other recent water quality or biological studies;
• other factors (e.g., agricultural runoff, deforestation, municipal STPs, and operation of
the CP&L dam) that may affect water or biological quality in the Pigeon River.
A draft synoptic document will be submitted to Champion for review by 22 November
(assumes field collections are finished by late August), and a final document will be prepared
following receipt of Champion's comments on the draft version.
A separate report will be prepared describing the results of the fish tissue collections.
15

3. REFERENCES
Adams, S.M., A.M. Brown, and R.W. Goede. 1993. A quantitative health assessment index
for rapid evaluation of fish condition in the field. Trans. Am. Fish. Soc. 122:63-73.
American Public Health Association. 1985. Standard Methods for the Examination of Water
and Wastewater. American Public Health Association, Inc. New York.
Ball, J. 1982. Stream Classification Guidelines for Wisconsin. Wisconsin Dept. of Natural
Resources Technical Bulletin.
EA Engineering, Science, and Technology, Inc. 1988. Synoptic Survey of Physical and
Biological Condition of the Pigeon River in the Vicinity of Champion International's
Canton Mill, prepared for Champion International Corporation, Stamford, CT.
Prepared by EA Engineering, Science, and Technology. Sparks, MD.
EA Engineering, Science, and Technology, Inc. 1990. Study Plan for the Monitoring of
Dioxin in Fish Tissue. EA Report No. 11370.01, prepared for Champion International
Corporation, Canton, North Carolina.
EA Engineering, Science, and Technology, Inc. 1994. Results of 1994 Dioxin Monitoring in
Fish Tissue. EA Report No. 11370.02, prepared for Champion International
Corporation, Canton, North Carolina.
Lund, J.W.G., C. Kipling, and E.D. LeCren. 1958. The inverted microscope method of
estimating algal numbers and the statistical basis of estimations by counting.
Hydrobiologia. 11:143-170.
North Carolina Division of Environmental Management. 1995. Standard operating
procedures: Biological monitoring. Biological Assessment Group. Raleigh, NC.
Platts, W.S., W.F. Megahan, and G.W. Minshall. 1983. Methods for Evaluating Stream,
Riparian, and Biotic Conditions. General Technical Report INT-138. U.S.
Department of Agriculture, U.S. Forest Service, Ogden, UT.
Rankin, E.T. 1989. The qualitative habitat evaluation index (QHEI): rationale, methods,
and application. Division of Water Quality Planning & Assessment, Ecological
Assessment Section. Columbus, OH.
Saylor, C., A. McKinney, and W. Schacher. 1994. Case study of the Pigeon River in the
Tennessee River drainage. TVA Biological Report 19.
16

IdATER UJRLITY PLANNING Fax:[phone] Jun 21 '95 13:41 P.C11/01
State.of-North'GdEonnr . -
:Depalt'tnent;bf•EriufF 9r 1t ,taxna
lth asmftta!?memu75T1 �+, � 1
Hea and'Naturell l7esoun :,:
Dhrlslon of Environmental Monal i
James S. Hunt,Jr„iGovemor
°0P Jonathan B. Howes,Secretary
A. Preston Howarq,Jr„P:E.,Dlret ^°.g
June 9,1995
MEMORANDUM
TO Metuters,Variance Review Committee
FROM: Den Loftin
SUBJECT: July 19. 1995,Variance Review M eCting
Please be advised•that.the Epviimmmental Mauagement Co ion,Variance Review
Committee has solGdulctl ameeting•for July 19,1995-from 9: am to 3.00 pm. The meeting
will be held:at Champion International's facility incanton,.NC. The agenda will include a
meeting with.Chariapio4 for.the first hour,a tour of the facility for the next 2-3 hours,lunch and
then a wrap-up session for questions and discussion. The weetmg should adjourn by 3:00 pm.
'In tercets of travel pla cs for the meeting,Water Quality staff from the central office in Raleigh
have reserved a mini:v for any merubers'who would like a ride to Asheville. Downtown
accommodations are'.a able at the Radisson lintel(704-Z52 211)and the Best Western(704
253-1851). In addition an interest has• cen expressed by the ' bers.to share dinner together
on Tuesday evening,?Iy 18.
I have enclosed for your revieva Emiatest report prepared by C , pion International on color
removal technology(Mri130,;1'995). The reporthas a brief ary of the original variance
process,a deseriptionli ft modernizationptrocess and vplor to otion results so far, and a
discussion of the B1cac I Filtrate Recycle(BFR)process.currenty under construction_
If you have any questions or if you.are interested is either ridin in thelmi19)73 or shading dinner
on Tuesday evening�prase get in touch with Beth McGee by p one aG(919)733-5083 (exc 575),
fax at(919)715-5633,cur EMad at beth @ dem.ebnr state.nc.us" I look forward to meeting with
you in July.
DUbm
cc: . EMC Membors
Linda Bray Rinier
Frank Crawley
Regional Super0sors '
Derric Brown,Cbatapion International Corporation
Preston Howard,DEM
Steve Tedder,DEM WQ
Fritz Wagner,EPA
P.O.FW 29535;RcIoTgh:Norlh Ccrokna27626-05M TelepMne 919.733 7015 FAX [phone]
An EquO Opp«hruly AfltrmaiW Ac w Employer WL rocyclaW 10%peer-aeasumer PaPer

Champion International
Corporation
Stamford, CT
Comments on the USEPA
Proposed TCDD Effluent
Limit for Champion
International Corporation's
Canton, North Carolina
Pulp and Paper Mill
Prepared by: Dr. Paul D. Anderson
ENSR Consulting and Engineering
(Formerly ERT)
April 1989 .
Document Number 1360-010-000

I
TABLE OF CONTENTS
Page
EXECUTIVE SUMMARY
1. INTRODUCTION 1-1
2 . DERIVATION OF THE PROPOSED EFFLUENT LIMIT OF
0 . 1 PPQD 2-1
3 . REVIEW OF THE CANCER POTENCY OF 2 , 3 , 7, 8-TCDD 3-1
4 . REVIEW OF ALLOWABLE RISK LEVELS 4-1
5 , REVIEW OF BIOCONCENTRATION/BIOACCUMULATION FACTORS 5-1
5 . 1 Theory of, and Differences Between Bio-
concentration and Bioaccumulation Factors 5-1
' 5 . 2 Derivation of a Site-specific Bioaccumulation
Factor 5-3
6 . REVIEW OF FISH CONSUMPTION RATES 6-1
6 . 1 Review of Regional Freshwater Fish Consumption
Rates 6-1
6 . 2 Site Specific Fish Consumption Rate 6-5
6 . 3 Loss of Dioxin During Preparation of Fish 6-6
i� 7 . SUMMARY AND CONCLUSIONS 7-1
REFERENCES
5006H 1360-010-000

LIST OF TABLES
Table Page
3-1 Comparison of Non-Threshold Cancer Potency
Estimates 3-5
3-2 Risk Specific Doses and Reference Doses for
2 , 3 , 7, 8-TCDD 3-13
s 3-3 Effect of Selected Alternative CPFs on the
Proposed Dioxin Effluent Limit 3-15
5-1 Summary of Dioxin and Furan Analysis of Fish
Fillets from the Pigeon River 5-4
6-1 Regional Freshwater Fish Consumption Breakdown 6-3
7-1 Effect of Alternative Assumptions on the Dioxin
Effluent Limit at an Allowable Risk Level of
1 x 10-6 7-2
I
5006H 1360-010-000

LIST OF ILLUSTRATIONS
Figure Page
3-1 Graph of Dose-Response Curves for Threshold
and Non-Threshold Effects 3-3
3-2 Cancer Potency Factor 3-14
y
5006H 1360-010-000

- EXECUTIVE SUMMARY
This document contains comments on the USEPA' s proposed
dioxin effluent limit of 0 . 1 part per quadrillion- (ppgd) for
Champion International ' s, Canton, North Carolina Mill . The
comments review the basis for the proposed limit and present
recent scientific and site-specific evidence supporting an
increase in the proposed dioxin effluent limit .
The proposed dioxin limit is based on the USEPA' s dioxin
Ambient Water Quality Criterion for consumption of fish only at
a one in one million excess lifetime cancer risk. Three key
parameters about which assumptions were made in developing the
criteria, that are either outdated or not applicable to the
Pigeon River are: the carcinogenic potency of dioxin; the
bioconcentration factor of dioxin; and, the rate of fish
consumption from the Pigeon River .
Dioxin has not been shown to cause cancer in humans ,
therefore, the USEPA uses data from animal experiments to
extrapolate to effects in humans . The USEPA assumes that
dioxin' s mechanism of causing cancer has no threshold, i . e. ,.
some risk exists for all doses except zero. A large body of
evidence does not support the method used by the USEPA. The
USEPA also extrapolates from animals to humans based on body
surface area, as opposed to body weight . Scientific evidence
contradicts extrapolation based on surface area but does
support extrapolation based on body weight, especially for
compounds like dioxin. The U. S. Food and Drug Administration
(USFDA) recognizes this and extrapolates based on body weight .
At a minimum, the proposed effluent limit sho.uld be based on a
potency factor derived using body weight extrapolation.
Given the ever increasing evidence that dioxin' s mechanism
of causing cancer has a threshold, i .e. , below some dose there
is no risk of contracting cancer, the USEPA and USFDA potency
factors which assume no threshold, should not be used.
Instead, the proposed dioxin effluent limit should be based on
4975H 1360-010-000

tl -
a mechanism of carcinogenesis that incorporates a threshold.
Canada, the Province of Ontario, and several European Countries
have taken this approach.
The USEPA used a bioconcentration factor (BCF) of 5000 to
derive the proposed effluent limit . Dioxins physiochemical
properties indicate that in rivers, use of a BCF is wrong . A
bioaccumulation factor (BAF) should be used instead. Because
BAF ' s vary from site to site, the most realistic BAFs are based
on site-specific data . The comments presorted here derive a
site-specific dioxin BAF of 1005 for the Canton Mill on the
Pigeon River . This BAF should be used in the proposed
criterion instead of a BAF of 5000 , because it reflects actual
conditions at the site.
The USEPA' s proposed dioxin limit assumes people eat 6 . 5
grams of fish from the Pigeon River per day for their entire
lifetime. This generic rate of consumption was based on a
consumption survey of estuarine fish and shellfish (neither of
which are native to the Pigeon River) and freshwater fish and
thus should not be used when deriving the proposed effluent
limit . Based on available literature the comments derive a
regional per capita consumption rate of 1. 5 grams of freshwater
fish per day. The comments also derive a range of consumption
rates specific to the Pigeon River. The rates exclusive to the
Pigeon River should be used to derive an effluent limit for the
Canton Mill .
The comments also gresent evidence that dioxin is lost
during cooking . The dioxin effluent limit should account for
cooking loss .
The dioxin effluent limit of 88, 266 parts per quadrillion
recommended by these comments is supported by recent scientific
and site-specific data, is protective of the public health at a
1xl0-6 excess lifetime risk level and should be the dioxin
effluent limit for the Canton Mill .
4975H 1360-010-000

r
1 . INTRODUCTION
i �
This document contains comments on the USEPA' s proposed
dioxin effluent limit for Champion International Corporation ' s
Mill in Canton, North Carolina . The comments begin with a
discussion of the basis for the proposed 0 . 1 parts per
quadrillion (ppqd) dioxin effluent limit and then identify the
key assumptions used to derive that limit . For each key
assumption the comments support a change in that assumption
when applied to the Canton Mill . The impact of the more
appropriate alternative assumptions on the proposed dioxin
effluent limit is also shown. The comments conclude by
presenting, several alternative, site-specific, scientifically
defensible effluent limits . All of these are protective of the
public health.
Note that in the comments, 2 , 3 , 7, 8-TCDD is refered to
interchangeably as 2, 3 , 7, 8-TCDD, dioxin or TCDD. Similarly
2 , 3 , 7, 8-TCDF is referred to as 2,3 , 7, 8-TCDF, furan or TCDF.
The comments focus on the properties of dioxin as opposed to
furan, because most of the information available is for dioxin
and also because most of the cancer risk from effluents is
usually associated with dioxin. The comments , also do not
focus on dioxin toxic equivalents for the same reason they do
not focus on furans i .e. the dioxin concentration accounts for
the majority of the dioxin toxic equivalent concentration. The
amount of dioxin in water is expressed as parts per quadrillion
(ppqd) in the comments in part because effluent detection
limits are about 10 ppqd. Similarly, dioxin levels in fish are
expressed as parts per trillion (ppt) . Detection limits are
about 1 ppt in fish.
1-1
4820H 1360-010

2 . DERIVATION OF 'THE PROPOSED EFFLUENT LIMIT OF 0 . 1 ppgd
The draft permit derives the proposed effluent limit based
on the following relationship:
(TCDD River Concentration) x (Average River Flow) _
(TCDD Effluent Concentration) x (Effluent Flow)
An allowable effluent concentration can be solved for
given the flow rate of the river, the flow rate of effluent,
and an allowable TCDD concentration in the river, by
rearranging as follows:
L (Effluent Conc. ) _ (River Conc._) x (River Flow)
(Effluent Flow)
To derive the proposed 0 . 1 ppgd effluent limit , the USEPA
assumed the average flow of the Pigeon River is 319 cubic feet
per Pigeon second (cfs) , the average flow of the effluent,
following modernization of the mill, will be 45 cfs , and the
allowable concentration of 2, 3 , 7, 8-TCDD in river water is 0 . 014
ppqd. This concentration is equal to the USEPA Ambient Water
Quality Criterion for consumption of fish only at a 1 x 10-6
excess lifetime cancer risk level .
The USEPA has recognized that excess lifetime cancer risk,
the adverse health effect that is the basis for the water
quality criterion, is a chronic effect and therefore the
average river flow is appropriate . The USEPA has also used the
expected effluent flow for the mill once the modernization is
complete and appropriately matched this flow to annual average
river flow. The USEPA further recognized that people do not
use the Pigeon River as a drinking water supply and therefore
the Ambient Water Quality Criterion for consumption of fish
only is the appropriate criterion to use for establishing the
proposed TCDD effluent limit . The water quality criterion for
consumption of fish only, however, contains several assumptions
that are not appropriate for the Canton, North Carolina mill .
2-1
4821H 1360-010

The criterion has as its basis the following fundamental
formula used to estimate excess lifetime cancer risk:
Risk = Dose x Potency
In deriving the criterion of 0 . 014 pq/1 the USEPA assumed
an allowable excess lifetime cancer risk is 1 x 10-6
(one in
one million) and that the potency of dioxin is 156, 000
(mg/kg-day)-1. Hoth of these assumptions have a significant
impact on the proposed dioxin effluent limit . Alternative
values should have been used in setting the proposed effluent
limit . The basis for the alternative values is discussed in
LL more detail in subsequent sections .
The dose of TCDD from eating fish is a function of how
much fish people eat, how much TCDD is in the fish people eat,
and how much people weigh. when deriving the criterion, the
USEPA assumed the concentration of TCDD in fish was the product
of a bioconcentration factor (BCF) and the concentration of
TCDD in river water. The USEPA assumed people living in the
Pigeon River area (assumed to be the residents of Haywood and
Cocke Counties in this document) eat 6 . 5 grams of fish per day
for 70 years, that people weigh 70 kilograms for 70 years and
that the BCF for TCDD is 5, 000 . Since 1984 , when the USEPA
published the criterion in final form, a great deal of
additional information has become available and indicates that
using BCFs to estimate TCDD fish concentration is incorrect,
that the average amount of freshwater fish eaten by United
States residents is less than 6 . 5 grams per day and that
dioxin ' s potency may be substantially lower . The new
information is discussed in subsequent sections of the comments .
Given all the assumptions made by the USEPA the equation
to estimate excess lifetime cancer risk becomes :
2-21
4821H 1360-010

1 x 10-6 = 0 . 0065 (kg fish/person-day)
x 5000 ( liters water/kg fish)
x X (mg TCDD/liter water)
x 1/70 (person/kg body weight)
x 156, 000 (mg/kg-day)-1
Rearranging the equation to solve for the unknown water
concentration results in:
X (mg TCDD/liter water) = 1 x 10-6
x 1/0 . 0065 (person-day/kg fish)
x 1/5000 (kg fish/liters water)
x 70 (kg body weight/person)
x 1/156, 000 (mg/kg-day)
and solves to 0 . 014 ppgd, which is the USEPA Ambient Water
Quality Criterion for dioxin assuming fish ingestion is the
only exposure pathway, an allowable excess lifetime cancer risk
is 1 x 10-6, as well as all the other USEPA assumptions . The
remainder of the comments review key assumptions used to derive
the USEPA national criterion that should be changed to
accommodate both site-specific data and more recent scientific
data .
The next section (Section 3) of the comments reviews
available information about the carcinogenic potency of dioxin
and shows the effect more recent estimates of dioxin cancer
potency have on the proposed TCDD effluent limit . Section 4
reviews the basis for alternative allowable excess lifetime
cancer risk levels and shows what effect these have on the
proposed TCDD effluent limit . Section 5 discusses why
bioaccumulation factors (BAFs) and not bioconcentration factors
(BCFs) are more appropriate for estimating dioxin
concentrations in fish. Section 5 also presents evidence that
BAFs vary from site to site and uses data collected from the
Pigeon River to derive a site-specific BAF. Section 6 reviews
more recent data on rate of fish ingestion and also derives a
2-3
4821H 1360-010

site-specific fish consumption rate based on the amount of
edible food fish produced by the Pigeon River . Section 6 also
presents data that indicate dioxin is lost during cooking .
Section 7 summarizes the findings in the comments and shows the
effect of combinations of alternative assumptions on the
proposed TCDD effluent limit .
2-4
4821H 1360-010

3 . REVIEW OF THE CANCER POTENCY OF 2 , 3 , 7, 8-TCDD
The USEPA Ambient Water Quality Criterion published in
final form early in 1984c (USEPA 1984) , and which forms the
basis of the proposed effluent limit for the Canton Mill, is
derived using a cancer potency factor (CPF) of 156 , 000
(mg/kg-day) -1 . Since the development of the bases of that
criterion in the late 1970s and early 1980s , a great deal more
data have been collected about the potential carcinogenicity of
TCDD. That CPF, used by the USEPA, assumes that dioxin' s
mechanism of carcinogenisis has no threshold, i .e. any exposure
to dioxin potentially poses some level of risk.
Recent evidence indicates that dioxin ' s mechanism of
carcinogenisis has a threshold (Friess 1989 ; CanTox 1989) .
This section reviews why CPFs derived from animal data are used
when assessing excess lifetime cancer risks from dioxin; the
difference between non-threshold and threshold models ; the
derivation of the current USEPA CPF; the status of USEPA' s
review of that CPF; and, also the basis for- the U. S. Food and
Drug Administration (FDA) and Moolgavkar-Knudson-Venson (MKV)
potency factors . The section concludes by presenting how
potential risks are determined from threshold models,
summarizes several allowable doses, and discusses the effect
selected alternative CPFs have on the proposed TCDD effluent
limit . It should be noted, that the relative difference in
potency between TCDD and the other dioxin and furan isomers is
assumed to stay the same regardless of the potency of TCDD.
To date, the conclusive evidence that exposure to TCDD
causes an increased incidence of cancer in humans has not been
demonstrated (USEPA 1988; Friess 1989) . Lacking direct
evidence, estimates of the potential risk to humans have been
based upon experimental data derived from laboratory animals
(Can 'pox 1989 ; USEPA 19.85) . Two types of dose extrapolation
models can be used to estimate the potential risks to human
health from animal data . First, non-threshold models , such as
that employed by the USEPA to derive the water quality
3-1
4833H 1360-010-000

criterion, assume that any dose, no matter how small, has
associated with it an increased risk of cancer . Second,
threshold models assume a threshold dose exists below which
there is no risk of contracting cancer (see Figure 3-1) . Both
types of model have been used to assess the excess cancer risk
following exposure to TCDD.
Experimental data suggest that TCDD acts as a promoter,
and thus , that a threshold may exist for induction of cancer
(Friess 1989 ; CanTox 1989) . TCDD has been found to be inactive
in most mutagenicity tests, and the few, weak positive results,
have not been reproducible in other laboratories (CanTox
1989) . No evidence exists that TCDD binds to, or covalently
reacts with, DNA. Many researchers have concluded that TCDD is
not a genotoxic agent, and therefore, that it is not a direct
acting carcinogen (Scheuplein 1984 ; Kociba 1984; Kimbrough et
al . 1984 ; USEPA 1986; NCASI 1987 Friess 1989 ; CanTox 1989) .
Several foreign governments (Ontario 1985 ; Denmark 1984 ; and
Germany 1987) have concluded that TCDD is an indirect acting
carcinogen and recommend that the carcinogenic risk of TCDD be
estimated with a threshold model (NCASI 1987) . Much of the
following discussion regarding the derivation of alternative
potency estimates is taken from NCASI (1987) .
Non-threshold models
The United States uses a non-threshold model in order to
estimate the potential risk to human health from carcinogens .
Strictly speaking, a carcinogen is an agent that by itself, can
cause cancer, including causing the initial mutation of DNA
(NCASI 1987) . Initiators are agents that produce this
irreversible mutational change in the cells genetic material .
Promoters are agents which cause the initiated cells to . produce
a tumor . Promoters only act on previously initiated cells and
the effects may be reversible if discontinuation of exposure
occurs before a tumor is formed. Generally in the United
States , regulatory agencies make no distinction between
3-2
4833H 1360-010-000

NO THRESHOLD
F
U
W
LL
LL
W
LL
O
}
U
Z
W
LU
W
Q
LL
DOSE
THRESHOLD
H
U
w
LL
LL
W
LL
O
}
U
Z
W
O
THRESHOLD DOSE
w
LL
DOSE
No Effect Doses
Figure 3-1 Graph of Dose-Response Curves for Threshold and Non-Threshold Effects
0
u
V
n
3-3

initiators and promoters in their models , but classify both
types as carcinogens and consider them to act by - a
non-threshold mechanism. Although this generalization may be
appropriate for some chemicals it is not appropriate for dioxin
for which threshold data exist .
The approach used to estimate the carcinogenic risk to
human health assuming no threshold involves two general
extrapolations . The first is to extrapolate from high dose
experimental data to low doses that more closely approximate
typical human exposures . In the second step, doses in animals
are converted or extrapolated to doses in humans . Choice of
how to extrapolate from high to low dose has the potential to
affect CPFs to a much greater extent than animal to human
extrapolation. However, to date in the United States, the
animal to human extrapolation is the source of most of the
differences among federal regulatory agency potency estimates
(Anderson 1988) . The upper 95% confidence limit of the slope
of the dose-response curve derived following these two
extrapolations is called the Cancer Potency Factor (CPF) and
has units of (mg/kg/day)-1.
Estimates of the cancer potency of TCDD derived from
non-threshold methods are summarized in Table 3-1 and described
below in the following subsections .
USEPA Cancer Potency Fa or Linearized Multistage
As a matter of science policy, and contrary to
accumulating scientific evidence, the USEPA uses the linearized
multistage model, a non-threshold, high-to-low dose
extrapolation model, to estimate human carcinogenic risk from
TCDD. It is an exponential model approaching 100% risk at high
doses, with a shape at low doses that is described by a
polynomial function. The USEPA assumes different species are
equally sensitive to the effects of TCDD if they absorb the
same dose per unit body surface area . Thus, the USEPA
extrapolated doses in rats to a human dose on the basis of body
3-4
4833H 1360-010-000

TABLE 3-1
COMPARISON OF NON-THRESHOLD CANCER POTENCY ESTIMATES
Cancer Potency
Estimate (a)
Extrapolation
Agency Model (SA) BW
EPA (1984) Linearized 1.56 x 105 2.9 x 104
Multistage
EPA (1988) None 1 x 104 (b)
FDA Linear -- 1.75 x 104
Extrapolation
-- [MKV] [1.7 x 1033
(a) Cancer potency estimates which were derived using the surface
area scaling factor were recalculated using the body weight
scaling factor, because the body weight factor correlates
more closely between species for TCDD: (Crouch and Wilson
1979; Crump et al. 1980)
(SSA) Calculated using surface area scaling factor
(BW) Calculated using body weight scaling factor
(b) This estimate was not calculated, but rather derived by
concensus opinion of the EPA Agency Workgroup (see text) .
[MKV] Indicates a tentatively derived value by Thorslund using the
Moolgavkar-Knudson-Venson model (1987) .
3-5
4833H 1360-010-000

surface area, which results in a higher CPF than when
extrapolated on the basis of• body weight (NCASI 1987; Anderson
1988) . Several analyses have been performed which compare the
accuracy of using a surface area versus body weight scaling
factor for extrapolation from animals to humans . Both Crouch
and Wilson (1979) and Crump et al . (1980) concluded that
scaling factors based on body weight have the closest
correlation between species suggesting the USEPA' s approach is
not scientifically justified and overestimates potential risk.
The USEPA used the data of Kociba et al . (1978) , and
combined tumor incidence from various sites ( liver, lung,
hard-palate, and nasal tubinate) . Kociba (1984) argued that
the tumors in areas other than the liver were not systemic, and
probably resulted from inhalation of some of the food and
prolonged direct contact of the TCDD treated food with the
mouth and respiratory tract, i . e. only the liver should be used
for assessing dioxin' s systemic effects in rats . The tissue
slides from Kociba et al . (1978) were reexamined by Dr . R.
Squire at the request of the USEPA and these data were also
used by the USEPA in determining the CPF for TCDD. Dr . Squire
identified a different number of tumors at some of the sites ,
but the effect of this on the CPF of dioxin is small (i .e. less
than 5%) . The USEPA uses a CPF derived from data limited to
animals that survived more than one year as evaluated by both
Kociba et al . (1978) and Squire (USEPA 1985) because of better
"statistical fit" between the observed data and the modeled
dose-response curves . The CPF of 1. 56 x 105 (mg/kg/day)-1
used by the USEPA is the geometric mean of two CPFs, one
calculated using the results of Kociba et al . (1978) and the
other using Squires ' interpretation of Kociba et al . data
(NCASI 1987) .
Another finding of Kociba et al . ( 1978) , often not
mentioned, is that the frequency of tumors at other sites
examined by the pathologists also changes . Forty-four sites
were examined in male rats and 40 in female rats . In males, 3
sites showed a significant increase (p<0 . 05, Fisher exact test)
3-6
4833H 1360-010-000

at higher doses, and 3 showed a significant decrease at higher
doses . In addition, total tumors were higher in controls (139)
than in other dose groups (40 , 49 , and 60 in low, medium and
high, respectively) .
A similar pattern occured in female rats . Tumors at 4 of
40 sites increased significantly (p<0 . 05, Fisher exact test)
and at 4 of 40 they decreased significantly at higher doses .
Total tumors were highest in the controls (230) . The low
medium and high dose groups had a total of 96 , 102 , and 120
tumors, respectively.
Typically, sites where tumor incidence decreases with
increasing dose are not included in potency derivations in
order to be protective of the public health and because
background tumor rates at those sites are often very high in
laboratory rodents . Sites with background tumor rates in
animals far higher than background rates observed in the same
sites in humans are typically excluded, because mechanisms of
carcinogenesis may be different at those sites between
species . It is nevertheless important to remember that some
sites in the Kociba et al . (1978) study did have a significant
decrease in tumor incidence at high doses and that total tumors
were highest in rats that received no dioxin. Because
background tumor rates differ between rats and humans at some
sites, these sites are easily identified as potentially having
different mechanisms of carcinogens . Other sites as well,
however, may have different mechanisms of carcinogenesis ,
including those sites used to derive the CPF. This is
especially true for the liver of older rats . Livers in such
rats not only have a higher background tumor rate than do human
livers , but because they contain many initiated cells they also
appear to be especially sensitive to substances that promote
the formation of tumors (CanTox 1989) .
Another assumption made by the USEPA in deriving the
current CPF, is use of only data from female rats because these
data result in higher CPFs (USEPA 1985) . The USEPA derived
CPFs of 14 , 700 and 17, 300 based on data from male rats only and
3-7
4833H 1360-010-000

a CPF of 156 , 000 based on data from only female rats . Again,
selecting data that result in the highest CPF may be typical
public health practice, however, it should be noted that the
current CPF is based on only data from female rats and that
data from male rats indicated dioxin is about 10 times less
potent . As will be shown below, other US and foreign agencies
(and even the USEPA recently) recognize some of the
uncertainties and shortcomings of the way the current CPF is
derived and use lower estimates of dioxin' s carcinogenic
potency.
EPA' s Proposed Reevaluation of TCDDs ' Carcinoaenic Potency
Because recent scientific evidence suggests that the
current USEPA CPF is high, the USEPA' s Agency Workgroup
(Workgroup) has re-examined the scientific basis and methods
used for its estimation of the cancer potency of TCDD, and has
recommended that the CPF be reconsidered (USEPA, 1988) . The
Workgroup felt that the methodology utilized by the USEPA
overestimated TCDD' s cancer potency and should be adjusted so
that it is similar to the value used by other agencies because
the literature suggests that TCDD is a promoter . The Workgroup
evaluated several different approaches to estimating TCDD ' s
cancer potency, but concluded that "there does not appear to be
compelling scientific reasons for regarding any one of them as
a most appropriate approach" . It recommended that the CPF be
reduced by a factor of 16, in which case the risk specific dose
would be 1 pg/kg/day. The Workgroup feels this estimate of
TCDD' s potency is consistent with the available data and
theories, and that it represents a reasonable science policy
position. This value was not specifically derived by any one
model . The Workgroup felt that there is currently no
definitive scientific basis in this value, but believes that it
is grounded in rational, science policy. They felt that an
order of magnitude estimate of the potency, as opposed to some
more precise estimate of the risk-specific dose, helps to
3-8
4833H 1360-010-000

convey the 'notion that the numerical expression is only a rough
estimate .
FDA Potency Estimate
The United States Food and Drug Administration (FDA) has
primary responsibility for regulating contaminants in
foodstuffs, e. g . compounds in fish eaten by people . The Food
and Drug Administration (FDA) also used the data of Kociba et
al . (1978) , although it did not exclude all of the data from
the first year of exposure. The FDA extrapolated from rats to
humans on the basis of body weight, which results in a 5 . 4-fold
decrease in the CPF when compared with the EPA CPF. The FDA
utilized a linear extrapolation model and derived a CPF of 1. 75
X 104 (mg/kg/day) -1 (NCASI 1987; Anderson 1988) .
MKV Potency Estimate
The MKV (Moolgavkar-Knudson-Venson) Model is a two-stage
mathematical dose-response model originally applied to the
epidemiology of lung and breast cancers (Moolgavkar and Knudson
1981) . In a new application, it has recently been used to
describe TCDD' s ability to act as a tumor promoter (Thorslund
1987) . This model assumes that TCDD acts exclusively as a
tumor promoter. It is assumed that a promoter increases the
net growth rate of preneoplastic, naturally initiated stem
cells . The model has the ability to increase the growth rate
to a fixed upper bound. The difference between this upper
bound and the background growth rate is the maximum increase in
growth rate possible. The fractional• part of the change in the
maximum increase in growth rate due to a given dose is
calculated and is used to predict the tumor dose-response
relationship of TCDD.
Two approaches can be used to describe the dose-dependent
changes in cell growth. The first growth rate model assumes
that the changes follow first order kinetics which result in a
3-9
4833H 1360-010-000

negative exponential model that contains only one parameter
requiring estimation. This model is fitted to the most
extensive data set available in order to estimate the three
unknown parameters in the tumor dose-response model . A second
approach assumes that the rate of change of TCDD-induced growth
rate is proportional to the product of cellular TCDD levels and
the number of TCDD receptors . This model for predicting
increased cell growth rate is log-logistic in form and has only
two parameters, the slope and the intercept .
TCDD potency estimates derived using the MVK Model
(Thorslund 1987) use the increased hepatic tumor rates in
female Sprague-Dawley rats (Kociba et al . , 1978) and do not
include tumors observed at other sites in the same study. This
model used the Squire pathology analysis only, and does not
combine this analysis with the Kociba et al . (1978) analysis,
which was the approach taken by the USEPA in deriving its CPF
(USEPA 1984) . In an effort to adjust for high non-tumor
related mortality in the highest dose group, all animals that
died during the first year (prior to when the first tumor was
observed) , were excluded. The USEPA also eliminated these
animals . This model extrapolates the equivalent human dose
from a rat dose on the basis of surface area, which is similar
to the USEPA' s approach, and also leads to a greater estimate
of risk, when compared to body weight-based human equivalent
doses . The cancer potency estimate derived using the MKV model
and the surface area scaling factor is 1. 7 x 103
(mg/kg/day) -1
which is approximately two orders of magnitude
lower than the 1985 USEPA estimate and one order of magnitude
lower than the FDA estimate .
In summary, at a minimum the USFDA CPF should be used to
derive the dioxin effluent limit and as data supporting the MKV
model become available, it should replace the USFDA potency.
Moreover, accumulating data suggest that the no threshold
.assumption is wrong for dioxin and a threshold model should be
employed.
3-10
4833H 1360-010-000

Threshold models
The basic approach taken by regulators when dealing with
chemicals acting by a threshold mechanism is to derive a safe
level for exposure, based upon the No-Observed-Adverse-Effect
Level (NOAEL) . The NOAEL is determined from either
experimental or less frequently, epidemiologic data, and is the
highest dose in the most sensitive species , at which no toxic
response is observed. An acceptable daily intake (ADI) is
calculated from the NOAEL after applying appropriate
uncertainty factors (UF) . A OF is used to account for
intraspecies variation (human to human) , interspecies variation
( laboratory animal to human) , and to extrapolate from
subchronic to chronic exposures . A OF of 10 is used in each
case and for each applicable source of uncertainty, the UFs are
multiplied together to obtain the estimated total UF. For
instance, to estimate a chronic ADI from subchronic mouse data,
a OF of 1000 is usually used. The USEPA, when calculating
ADIs , or their equivalents, also utilizes a Modifying Factor
(MF) which incorporates professional judgement concerning the
quality of the data used to derive the ADI .
ADI = NOAEL ADI - NOAEL
OF OF*MF
The use of a threshold model to predict human carcinogenic
risk is appropriate for TCDD since it acts as a tumor promoter,
which is thought to be a reversible, threshold effect (NCASI
1987; Friess 1989 ; CanTox 1989) . Doses of 1 ng TCDD/kg/day
failed to produce tumors in rats, and this level has been
accepted as a NOAEL (Kociba et al . 1978; NCASI 1987; Friess
1989 ; CanTox 1989) for carcinogenesis . This NOAEL was derived
in a chronic, animal experiment . Thus the appropriate OF is
100 and an ADI can be calculated. as follows :
t
ADI = 1 na/ka/day = 0 . 01 ng/kg/day = 10 pg/kg/day
100
3-11
4833H 1360-010-000

Several countries- have utilized this approach and the data
Of Kociba (1978) to predict human carcinogenic risk from
exposure to TCDD, although each country uses different UFs .
Ontario used a OF of 100 and calculated an ADI of 10 pg/kg/day;
Netherlands used a OF of 250 to calculate an ADI of 4
pg/kg/day; Denmark utilized a OF of 200 to calculate an ADI of
5 pg/kg/day; and Germany utilized a OF of from 100 to 1000 to
calculate an ADI of from 1 to 10 pg/kg/day (NCASI 1987) .
A summary of risk-specific doses (estimated for a one in
one million excess lifetime cancer risk) and maximum allowable
doses (also sometimes referred to as reference doses) is
provided in Table 3-2 and Figure 3-2 .
Effect of Alternative CPFs on the Proposed Effluent
Guideline
In describing the effect of the different CPFs on the
proposed dioxin effluent limit, the key assumption is that the
relationship between potency, dose and risk is linear . In
other words , a ten fold increase in dose or potency increases
risk by ten fold. Because this is a fundamental assumption of
the non-threshold approach, the consequence of using the
proposed USEPA CPF of 10, 000, which is 16-fold lower than the
CPF used to derive the proposed limit, is a sixteen fold
increase in the proposed limit to 1. 6 ppgd (Table 3-3 ) .
Use of the FDA potency increases the effluent limit to 0 . 9
ppqd (Table 3-3) . Using the Canada (Health and Welfare Canada
and Province of Ontario) allowable daily dose increases the
effluent limit by 1667 fold to 167 ppgd (Table 3-3) .
This comparison shows that use of any other CPF or
allowable dose would result in an increase (in some cases very
substantial) in the proposed dioxin effluent limit . Because
many of these other CPFs or allowable doses have been derived
subsequent to the currently used USEPA CPF, they better reflect
the current state of knowledge and are thus more representative
of the actual cancer risk posed by dioxin. It would seem
3-12
4833H 1360-010-000

TABLE 3-2
RISK SPECIFIC DOSES (lx10-6)
AND
REFERENCE DOSES FOR 2,3,7,8-TCDD
Risk Specific or
Reference Dose
Source (pg/kg/day)
Canada (Health and 10.0
Welfare Canada and
Province of Ontario)
Netherlands 4.0
N.Y. State 2.0
Federal Republic of 1.0
Germany
[Moolgavkar-Knudson- [0.6]
Venson Model]*
USEPA (proposed) 0.1
National Research 0.07
Council, Canada
U.S. Food and Drug 0.06
Administration
U.S. Centers for Disease 0.03
Control
California 0.007
U.S. EPA (Linear 0.006
Multistage Model)
*Doses and sources of doses surrounded by brackets,
represent research efforts, and not conclusions drawn by
agencies.
3-13
4833H 1360-010-000

\�o
C`
a\ N"
a �%0
Cac fP
NPO'JS' O S G cad d^' (tQP
I I 1 1 1 1 1 1 r-7T-77 l l
0 10.000 20.000 30.000 140.000 150.000 160.000 170.000
The figure presents the cancer potency factors that correspond to the t
a predicted one in one million(1 x 10-6)excess lifetime cancer risk giv FIGURE 3-2

TABLE 3-3
EFFECT OF SELECTED ALTERNATIVE
CPFs ON THE PROPOSED DIOXIN
EFFLUENT LIMIT
CPF (mg/kg-dayl-1 Agency Change in Limit Effluent Limit (onad)
156,000 USEPA - 0.1
10,000 USEPA 16-fold 1.6
17,500 USFDA 8.9-fold 0. 9
Risk Specific Dose* (pg/kg-day)
0.07 Canada (NRC)** 11.7-fold 1.17
2.0 New York State 333-fold 3.33
4.0 Netherlands 667-fold 66.7
10.0 Canada (HWC)*** 1667-fold 166.7
*The risk specific dose is the lifetime average daily dose at which excess
lifetime cancer risk is ecual to lxl0-6,
**(NRC) is the National Research Council of Canada.
***(HWC) is Health and Welfare Canada and Province of Ontario.
3-15
4833H 1360-010-000

prudent when setting an effluent limit today, that the USEPA
employ the most recent and best science available and not base
proposed effluent limits on outdated perception.
3-16
4833H 1360-010-000

4 . REVIEW OF ALLOWABLE RISK LEVELS
The proposed dioxin effluent limit for the Canton, North
Carolina Mill is based on the dioxin Ambient Water Quality
Criterion developed assuming an excess lifetime cancer risk of
lxl06 (one in one million) is allowable based on consumption
of fish exposed to dioxin in the Pigeon River . Use of
alternative levels of allowable risk would change the proposed
effluent limit . For example, had the dioxin Ambient Water
Quality Criterion based on a lx10-5 excess lifetime risk been
used (USEPA 1984) , the proposed effluent limit would have been
10-fold higher (1. 0 ppgd) .
Given the conservative nature of the exposure assumptions
used by the proposed dioxin effluent limit, and the large body
of evidence suggesting that dioxin is much less potent than
assumed in the proposed limit, use of a stringent level of
allowable risk seems inappropriate. If it seemed likely that
many people could be exposed to large amounts of dioxin or that
dioxin ' s cancer potency is larger than assumed, than using
lxl0-6 as the allowable risk level could be justified.
However, quite the opposite is true: both potential exposure
and potency have been overestimated. The USEPA has recommended
an allowable risk level of lx10-6 in this case, however use
of 1x10-5 as an allowable level of risk is not at all
extraordinary nor precedent setting . It is common at both the
state and federal level .
An excess lifetime cancer risk of lx10-5 is the excess
lifetime cancer risk used by North Carolina when they developed
the fish advisory for the Pigeon River (USEPA 1988) . Risks
below this level were considered allowable and above this level
not allowable. It seems contradictory to use different levels
of allowable risk for the same river and exposure pathway.
Travis et al . (1987) recently reviewed 132 regulatory
decisions in which the federal government decided whether to
take action or not, depending at least in part, on the
4-1
4853H 1360-010-000

magnitude of excess lifetime cancer risk. When estimated
excess lifetime cancer risks were less than lx10-6 (one in
one million) federal agencies never acted to reduce risks
further . When the expected risks were -4
p greater than 1x10
(one in ten thousand) agencies acted to reduce risks . When
risks were between lx10-6 and lx10-4 federal agencies
sometimes acted and sometimes not . The decision to act was
influenced by the number of people exposed and the costs
associated with taking action. When costs were high, or
relatively few people were potentially exposed, federal
agencies tended not to act . Given that the potential risks
estimated for the Canton, North Carolina mill are based on many
conservative assumptions that overestimate risk, and that
relatively few people are exposed, using an allowable risk
level of lx10-6 seems inconsistent with previous federal
agency actions .
It is of interest to note that excess lifetime cancer
risks of 1x10-5 , or greater, are also considered allowable by
federal drinking water standards . The Maximum Contaminant
Level (MCL) for benzene (5 ppb) has an estimated excess
lifetime cancer risk of 7.4x10-6 associated with it . This is
more than 7 times greater than the 1x10-6 risk level employed
by the proposed dioxin effluent limit . Excess lifetime cancer
risks for some other compounds with MCLs do not just approach
lx10-5, but exceed this level . Carbon tetrachloride at its
MCL (5 ppb) poses an excess lifetime cancer risk of 1 . 9x10-5
and vinyl chloride at its MCL (2 ppb) poses an excess lifetime
cancer risk of 3 . 3x10-4 .
Not- only are these risks much greater than that used to
set the proposed dioxin limit, but two of the compounds
(benzene and vinyl chloride) are known to cause cancer in
humans, unlike dioxin which is only suspected of causing cancer
in humans based on animal experiments . As discussed above, the
potency of dioxin is also likely to be lower than used in the
proposed effluent limit . Finally the exposure pathway to which
MCLs apply (drinking water) is far more certain than ingestion
4-2
4853H 1360-010-000

r of fish which is the pathway upon which the proposed effluent
limit is based. Comparison of the allowable risk levels used
to set the proposed effluent limit to the level associated with
MCLs , suggests a higher allowable risk level would be good risk
management policy, would be consistent with previous state of
North Carolina, USEPA and other federal agency action, and
would not be based solely on public misperception.
4-3
4853H 1360-010-000

5 . REVIEW OF BIOCONCENTRATION/BIOACCUMULATION FACTORS
The USEPA used a bioconcentration factor (BCF) of 5000 to
derive the dioxin Ambient Water Quality Criteria . Since 1984 ,
the year the criteria was adopted, a great deal more research
has been performed on the bioconcentration and bioaccumulation
of dioxin in fish. This research indicates that not only are
bioconcentration factors (BCFs) different from that employed by
the USEPA, but that they are also inapplicable to almost all
rivers . Because of dioxin' s physio-chemical properties ,
bioaccumulation factors (BAFs) should be used to determine
uptake of dioxin by fish. The parameters that determine BAFs
vary from river to river . Thus site-specific BAFs should be
derived for each river. Section 5 . 1 reviews BCFs and BAFs and
the differences between them. Section 5 . 2 uses site-specific
information to derive a BAF for the Canton Mill-Pigeon River
system.
5 . 1 Theory of, and Differences Between Bioconcentration and
Bioaccumulation Factors
Bioconcentration, bioaccumulation, biomagnification, BCF,
BAF are terms that often lead to misunderstanding,
misinterpretation and outright confusion about how to use and
apply them in work with aquatic organisms . The following
definitions will be used in these comments .
Bioconcentration - process by which a substance dissolved
in water can be taken up by an organism directly from
water .
BCF - a value which describes the degree of
bioconcentration.
Bioaccumulation - process by which an organism accumulates
a substance as a result of ingestion of water and food in
which the substance is dissolved (water) or to which the
substance is bound (food) .
5-1
4818H 1360-010

BAF - a value which describes the degree of
bioaccumulation.
Biomagnification - result of bioconcentration and
bioaccumulation in which tissue concentrations of the
chemical increase through two or more trophic levels .
For many water-soluble compounds uptake from water is
assumed to be 'the most important exposure pathway for aquatic
organisms . Thus, the contribution from diet can be ignored
since the concentration in the organism' s food would have to be
10, 000 to 100 , 000 times higher than the concentration in water
in order to have a comparable effect (Schuurmann and Klein,
1988) . However, for lipophilic (literally "fat loving" )
compounds such as dioxin, the uptake from water is
significantly decreased because of the reduced bioavailability
of the compound and the uptake from food becomes the major
route of exposure (Spacie and Hamelink, 1985) . In other words,
bioconcentration is inversely related to a compound' s water
solubility (Oliver and Niimi , 1985) .
Literature values of octanol-water distribution
coefficients (Kow) , a measure of how strongly a compound binds
to fat, for hydrophobic compounds can vary substantially,
sometimes by orders of magnitude especially for compounds -with
an octanol-water distribution coefficient of > 104 . Thus
calculating BCF/BAF values from Kow' s may give spurious
results . Other means of calculating BAF' s which account for
sediment characteristics like fraction organic carbon and
particle size can be used in the absence of other site-specific
data on actual body burdens and are likely to give a more
accurate representation of the BAF.
When attempting to estimate body burdens of
hydrophobic/lipohilic contaminants in organisms, a BAF will
give a far more realistic estimate of actual conditions . The
BAF may exceed the BCF (by orders of magnitude in some
instances) because of the importance of the contribution from
5-2
4818H 1360-010

food. The various means. of calculating a BAF for a compound
can give quite different results depending upon whether a
steady-state or kinetic approach is used. A kinetic approach
will more accurately represent "real-life. " A kinetic approach
makes no assumptions about steady state or equilibrium, which
may not be reasonable assumptions in riverine ecosystems .
Concentrations are not constant in the exposure media in rivers
because of seasonal and episodic events, and therefore the
uptake of dioxin by the biota will reflect the fluctuations in
the exposure media. The BCF used by the USEPA for 2 , 3 , 7 , 8-TCDD
in the criterion may significantly misrepresent actual
concentrations because the dioxin contribution in food is not
considered and uptake from water is greatly overestimated.
Section 5 . 2 uses site-specific data on concentration of TCDD
and TCDF in fish and water of the Pigeon River to derive a BAF.
5 . 2 Derivation of a Site-specific Bioaccumulation Factor
The data the USEPA collected on the concentration of
dioxins and furans in fish fillets (Table 5-1) can be combined
with the effluent flow and river flow to derive a site-specific
BAF. The site-specific BAF is derived in the same way one
would derive a BCF, however, because the biological. processes
represented by the BAF are all those that determine
bioaccumulation, and not just bioconcentration, the value is a
BAF.
To derive the BAF, the water concentration of dioxin and
furan is estimated using the concentration of dioxin and furan
in the effluent and the flow of the river and effluent . The
BAF is the ratio of the concentration of dioxin and furan in
fish to the concentration in river water . A BAF is derived
below for 2 , 3, 7, 8-TCDD and another for 2 , 3 , 7, 8-TCDF . BAFs were
not derived for the other isomers because of insufficient
data . Separate BAFs were derived for dioxin and furan because
the two compounds have been shown to have substantially
different BAFs and BCFs at other sites .
5-3
4818H 1360-010

TABLE 5=1
SUMMARY OF DIOXIN AND FURAN ANALYSIS OF FISH
FILLETS FROM THE PIGEON RIVER
Species N RM 2378-TCDD (put) 2378-TCDF (vvt) TEO (nvt)
EPA Environmental News*
SM Bass 5 33 ND (1) ND (1) ND
RB Sun. 10 33 ND (1) ND (1) ND
SM Buff 3 24.5 7.0 6.7 7.7
SM Bass 4 24.5 5.9 3.2 6.2
Drum 1 24.5 0.55 (J) 2.2 0.8 (est)
Black Buff 2 24.5 ND (2.0) 5.0 0.5 (est)
RB Sun 10 13.6 1.5 0.9 2.3
SM Buff 5 12.8 ND (1) 1.4 0.14 (est)
LM Bass 5 12.8 2.9 2.4 3.1
Carp 5 12.8 6.3 3.1 6.6
LM Bass 3 0.2 0.67 1.10 0.78
RB Sun 6 58.0 12.01 5.94 12.62
Synoptic Eitudy of the Pigeon River System**
Blgill 3 58 7.2 5.2 7.7
Carp 1 58 5.2 5.6 5.8
WT Sucker 1 58 75.7 145.38 97.1
LM Bass 5 41.5 9.6 5.8 10
4846H 1360-010
5-4

TABLE 5-1 (Continued)
Species N RM 2378-TCDD (Pnt) 2378-TCDF (ppt) TEO (Pyt)
Synoptic Study of the Pigeon River System (continued)
B Bullhead 2 41.5 10 ND (1) 10
SM Bass 1 41.5 10 5.8 11
Muskie 1 41.25 7.4 26 10
B Crappie 5 41.25 7.5 8.7 8.4
B Bullhead 2 41.25 3.4 ND (1) 3.4
LM Bass 5 41.25 11 4.8 12
LM Bass 1 41.25 11 5.6 12
B. Bullhead 5 40 7.9 ND (1) 7.9
B Trout 1 40 78 10 80
LM Bass 2 40 11 7.6 12
B Bullhead 4 39 13 ND (1) 13
B Crappie 3 39 6 7.2 6.9
R Bass 1 8.5 0.17 ND 0.17
Carp 1 8.5 22.07 16.09 Z5.1
*Data taken from: Environmental News. USEPA. Region 4. Results of Additional Dioxin
Analysis of Fish Taken from the Pigeon River in Tennessee and North
Carolina.
**Data taken from: Assessment of Dioxin Contamination of Water, Sediment and Fish in the
Pigeon River System (A Synoptic Study) USEPA. Region IV
Environmental Services Division. Report No. 001. November 15, 1988.
4846H 1360-010
5-5

The concentration of dioxin and furan in river water is
estimated using the following formula :
River Water Concentration (mg/1) = Effluent concentration (mg/1)
x Effluent flow (cfs)
x 1/River flow (cfs)
The mean annual flow of the Pigeon River is 319 cfs and
the effluent flow for the existing mill is 6.8 . 7 cfs . The
effluent has been tested once for dioxins and furans . Neither
TCDD or TCDF were found at detection limits of 18 and 24 ppgd,
respectively. This calculation of a site-specific BAF assumes
that the concentration of TCDD and TCDF is equal to 10 ppgd,
about the current method detection limit . This value is about
one half of detection limits of the one sample at the Canton
Mill . Given this assumption, the concentration of 2 , 3 , 7, 8-TCDD
in the Pigeon River is 2 . 15 ppgd and the concentration 2, 3 , 7, 8
TCDF 2 . 15 ppgd.
The concentration of TCDD and TCDF in sport fish is based
on the data presented in Table 5-1. Perusal of the data
indicates that the concentration of dioxin and furan in fish is
log-normally distributed, i .e. , most of the fish have
relatively low concentrations of TCDD and TCDF, but a few fish
have high concentrations . The geometric mean is a better
measure of central tendency than the arithmetic mean for such
distributions . The geometric mean of TCDD concentration for
all species of fish is 3 . 36 ppt and the; geometric mean
concentration of TCDF for all species of fish is 2 . 11 ppt .
Based on this information, the site-specific BAF for
2 ,3 , 7, 8 TCDD is 1563 liters/kilogram (3 . 36 x 10-6 milligrams/
kilogram divided by 2 . 15 x 10-9 milligrams/liter) . The
site-specific BAF for 2 , 3 , 7, 8-TCDF is 981 liters/kilogram (2 . 11
x 10-6 -9
milligrams/kilogram divided by 2 . 15 x 10
milligrams/liter) .
The BAF estimated using site-specific data is about 5
times smaller than the BCF used by the USEPA in the proposed
dioxin effluent limit . When the site-specific BAF is used
5-6
4818H 1360-010

instead of the USEPA BCF, the proposed effluent limit increases
to 0 . 34 ppgd. Because the processes that govern accumulation
of TCDD and TCDF in fish are highly dependent upon
site-specific parameters , a generic BCF or BAF should not be
employed, especially when site-specific data are available.
5-7
4818H 1360-010
e

6 . REVIEW OF FISH CONSUMPTION RATES
The USEPA used a fish consumption rate of 6 . 5 grams of
fish per person per day when they derived the Ambient Water
Quality Criterion' of 14 ppgt which forms the basis for the
proposed dioxin effluent limit for the Canton, North Carolina
mill . The 6 . 5 grams per day is a national average consumption
rate for freshwater fish, estuarine fish and shell fish (USEPA
1985) . Thus , the average is a composite of types of fish which
includes types of fish not found in freshwater rivers , such as
the Pigeon River in North Carolina and Tennessee.
A study by Rupp et al . (1980) suggests that the national
per capita consumption rate of freshwater fish is lower than
assumed by the USEPA and also that regional differences exist .
Results of that study are reviewed in more detail in Section
6 . 1 to derive a regional rate of fish ingestion for Tennessee
and North Carolina . Section 6 . 2 uses data collected on the
standing stock of fish in the Pigeon River and the size of the
surrounding population to derive a rate of consumption
applicable to only the Pigeon River . Section 6 . 3 presents
evidence that indicates dioxins are lost during cooking .
6 . 1 Review of Regional Freshwater Fish Consumption Rates
A great deal of recently collected data exists regarding
the saltwater recreational catch on a regional and national
level (NMFS 1985, 1986; USDA 1986) and also for some bodies of
saltwater on a local level (NOAA 1985; McCallum 1985) . The
saltwater data are clearly not applicable to freshwater
rivers . Surprisingly, little recent information is available
about the amount of freshwater fish that people eat per day.
The most applicable data were collected in 1973 and 1974 by the
National Marine Fisheries Service (NMFS) (described in Rupp et
al . 1980; and, SRI 1980) . The results of this survey have been
reviewed and form the basis of the regional rates of freshwater
6-1
4839H 1360-010=000
e

consumption reported in these comments . SRI reviewed three
other surveys of fish consumption and concluded that the NMFS
survey was the most scientifically sound (SRI 1980) . The other
surveys reviewed by SRI are not included in these comments .
NMFS Survey
The results of the NMFS survey are based on the responses
of 24 , 562 individuals who recorded the number of fish meals
they ate and the size of the meal for each species of fish over
a one month period. The results of the survey can be broken
down by age, census region of the country, sex, and species of
fish. Key results of the survey follow.
The rate of fish consumption increased with age. On a
national average basis, 1 to 11 year old children ate 0 . 18
kilograms per year, 12 to 18 year olds ate 0 . 31 kilograms per
year, and 18 to 98 year olds ate 0 . 54 kilograms per year .
Consumption was lowest for all age groups in the Northeast and
greatest in the Eastern and Western South Central portions of
the country.
A key finding of the NMFS study is that saltwater finfish
constitute the majority of a United State residents , fish
intake. Total intake was estimated by the survey to be 12 . 7
grams of fish per day. Saltwater finfish comprised 69% or 8 . 8
grams of the intake, 22% was shellfish, and only 9 . 3%, or about
1 . 2 grams per day was freshwater fish (Rupp et al . 1980) .
Strong regional differences exist for each type of fish. New
England residents eat the least freshwater fish, 0 . 034
kilograms per year or about 0 . 1 gram per day. Residents of the
East South Central part of the country eat the most freshwater
fish, about 1. 9 grams per day. Regional breakdown of
freshwater fish consumption is provided in Table 6-1.
North Carolina is in the South Atlantic Region
(consumption rate of 1. 05 grams per day) while Tennessee is in
the East South Central Region (consumption rate of 2 .24 grams
per day) . Because the Pigeon River flows from North Carolina
6-2
4839H 1360-010-000
e

TABLE 6-1
REGIONAL FRESHWATER FISH CONSUMPTION BREAKDOWN
1970s 1980s
Per Capita Per Capita
Average Average
All Ages All Ages
(am/day) (gm/day)
United States 1. 18 1 . 38
New England 0 . 22 0 . 26
Mid Atlantic 0 . 74 0 . 87
S. Atlantic 0 . 90 1 . 05
E.S. Central 1. 92 2 . 24
E.N. Central 1. 53 1. 79
W.N. Central 1. 75 2 . 05
W. S. Central 2 .30 2 . 69
Mountain 1. 26 1. 47
Pacific 0 . 90 1 . 05
Source: Rupp et al . (1980) .
4841H 1360-010-000 6-3
e

into Tennessee, an appropriate average freshwater fish
consumption rate for people living near the Pigeon River is the
average of these two regional rates, or 1 . 65 grams of fish per
day. This is consistent with Canadian findings (Friess 1989) .
Note this assumes each resident in the Pigeon River area (i . e. ,
residents of Haywood and Cocke Counties) eats 1 . 65 grams of
fish from the Pigeon River for every day of their lives . This
is an unlikely occurrence as will be shown in Section 6 . 2 in
which site-specific data are used to derive a site-specific
consumption rate.
The survey also provides some data about the breakdown of
fish consumption by species . For most rivers, species-specific
rates of consumption likely follow the distribution of species
within the river of interest, assuming consumption is
proportional to game fish abundance in the river of interest .
Nationally, the most frequently consumed species that spends at
least some proportion of its lifecycle in freshwater was
salmon, followed by trout, pike, catfish, bass , bluegill,
perch, crappie, and sunfish. For the Pigeon River, species
specific rates of consumption could be derived using data from
a standing stock survey (EA Engineering Science, and Technology
Inc. 1988) and assuming that people eat game fish in proportion
to their frequency of capture.
National Fish Consumption Trends
The NMFS survey of fish consumption was conducted in the
mid 1970s ., From 1977 to 1985 , the rate of fresh and frozen
fish consumption increased in the United States by 16 . 9% (from
3 . 5 to 4 . 1 pounds per person per year; USDA 1986) . Therefore
the rates of consumption derived above may be low by as much as
16 . 9%, assuming that trends in freshwater fish consumption
track trends in total fish consumption. The adjustment has
been made to the regional data taken from Rupp et al . (1980)
and reported in Table 6-1 .
6-4
4839H 1360-010-000

Effect of Regional Average Consumption Rate on Proposed
Dioxin Effluent Limit
Changing consumption rate changes the amount of dioxin a
person is exposed to. Note too, that changes in exposure are
linearly related to changes in risk. Therefore, any change in
consumption rate (i .e. an increase by 2 fold) will have a
linear (albeit an inverse) effect on the effluent limit . Thus
changing the assumed consumption rate of 6 . 5 grams per day to
the regional average of 1. 65 grams per day, a reduction of 3 . 9
fold, increases the proposed effluent limit by 3 . 9 fold to 0 . 39
ppgd, assuming all other assumptions remain the same. Given
that the 6 . 5 grams per day used by the USEPA includes types of
fish that do not and cannot occur in the Pigeon River, the
USEPA should at a minimum use regional consumption rates for
the types of fish found in the Pigeon River when deriving the
proposed effluent limits for dioxin.
6 . 2 Site Specific Fish Consumption Rate
Champion International Corporation has had an extensive
study performed of the fishery in the Pigeon River (EA
Engineering, Science, and Technology, Inc . 1988) . The amount
of fish from Waterville Reservoir and Pigeon River were
estimated separately and combined to determine a total amount
of fish produced. In a subsequent report, (EA Engineering,
Science and Technology, Inc. 1989) a best estimate and upper
limit of edible fish tissue produced by the Pigeon River and
Waterville Reservoir was estimated. The upper limit estimate
is 1360 kilograms of edible fish tissue per year and the more
likely best estimate is 192 kilograms per year (EA Engineering,
Science and Technology, Inc. 1989) . The upper limit was
derived assuming : the Pigeon River has 398 hectares of surface
area; there are 65 kilograms of fish per hectare; 46% of the
standing stock is edible - fish; of the edible fish 40% are big
enough to keep; the exploitation rate is 25%; and, 40% of a
6-5
4839H 1360-010-000
e

fish is edible tissue. The upper limit also assumed Waterville
Reservoir has 120 . 5 hectares of surface area; there are 350
kilograms of fish per hectare; 50% of the standing stock is
edible fish; the exploitation rate is 10%; and 40% of a fish is
edible tissue. To derive the most likely estimate of fish
produced by the Pigeon River and Waterville Reservoir the
following assumptions were changed: the exploitation rate in
both water bodies is 2% instead of 10%; the standing stock in
the Pigeon River is 35 . 6 kilograms per hectare instead of 68 . 1;
20% of the standing stock is edible fish in the Pigeon River
instead of 44%; and, 35% is edible in Waterville Reservoir
instead of 50%.
The yearly edible tissue production rates of 134 or 1245
kilograms can be converted to daily rates per person by
dividing the production rates by the number of days in a year
and the local population size. Assuming that it is only people
from Haywood and Cocke Counties (combined population size of
76 , 722) who use the Pigeon River for fishing results in daily
per capita fish ingestion rates of 7 milligrams and 49
milligrams of Pigeon River Fish per person per day. These two
estimates are 1300 times and 147 times, respectively, smaller
than the 6 . 5 grams per day used by the USEPA to derive the
proposed dioxin effluent limit . The consequence of using these
rates of fish consumption specific to the Pigeon River is an
increase in the proposed effluent limit to 14 . 7 or 130 ppgd,
respectivE:.ly. As with the key assumptions discussed
previously, the risk-based effluent limits under current
development must recognize the site-specific nature of risks
and set limits accordingly.
6 . 3 Loss of Dioxin During Preparation of Fish
It is likely that people catching and eating fish from the
Pigeon River gut the fish they catch and discard internal
organs . These discarded entrails may contain high
concentrations of TCDD. In addition, most people, if not all,
6-6
4839H 1360-010-000
e

probably cook the filleted fish they catch and eat . This
assumption is supported by a Connecticut survey of recreational
fishermen that asked about cooking method (Connecticut 1987) .
One-half of the respondents ate some of their fish fried,
one-third ate broiled fish, and one-third ate baked fish.
Other types of cooking methods recorded included barbequeing,
smoking and eating fish raw, although only 1% of the
respondents ate raw fish.
A recent investigation of TCDD loss during cooking of
restructured carp fillets found that between 35 and 67 percent
(with an average of 54 . 6 percent) of the TCDD was lost during
cooking (Stachiw, et al . 1988) . Method and duration of cooking
and size of the fillet all affected the amount of TCDD lost .
No other studies of TCDD loss during cooking could be located
in the literature, however, losses of PCB, a class of compounds
that shares many physiochemical properties with TCDD, have been
measured.
Puffer and Gossett ( 1983) found that about 25% of PCB was
lost during pan frying . Other studies (Humphrey et al . 1976)
have reported greater reductions in PCB levels, however, these
studies compared concentrations in whole raw fish to
concentrations in cooked fillets . Cordle et al . (1982) report
that up to 80% of PCB may be lost during cooking but based this
on Humphrey et al . (1976) .
Based on area studies it appears that, at a minimum, 25%
of the dioxin in fillets is lost during cooking . The USEPA
dioxin Ambient Water Quality Criterion is based on the amount
of dioxin in fish that people eat; not the amount of dioxin in
fish people catch. When these two are the same, then no
preparation loss adjustment is necessary. However, when the
concentration changes from the time fish are caught to when
they are eaten, an adjustment should be incorporated into risk
assessments and also risk-based criteria . All data indicate
dioxins, and compounds like them, are lost during cooking .
Incorporating a cooking loss of 25% into the criterion would
decrease exposure by 25% and therefore increase the risk based
6-7
4839H 1360-010-000
e

effluent limit to 0 . 13 ppgd. This is a relatively small
change, but one that is supported by data and should be part of
any risk based effluent limit under development or review.
6-8
4839H 1360-010-000
�e

7 . SUMMARY AND CONCLUSIONS
The preceding sections have identified several key
parameters used to develop the proposed dioxin effluent limit
of 0 . 1 ppgd that are either outdated (CPF and BCF) , not
consistent with other regulatory agency actions (allowable risk
level) , not based on site-specific information (BAF and
consumption rate) or were not considered (cooking loss) . In
all cases use of more apporpriate assumptions is warranted and
would substantially impact the proposed dioxin effluent limit .
Table 7-1 shows how different sets of assumptions impact the
proposed limit . The largest impact is caused by changes in
CPF, and consumption rate, followed by allowable risk level .
Cooking loss and using a site-specific BAF have the smallest
impact . Based on currently available scientific data and
widely accepted interpretation of that data, coupled with the
site-specific data available for the Pigeon River and
surrounding area, USEPA should modify the proposed limit using
the information contained in these comments and the cited
references and set a new dioxin effluent limit of 88, 266 ppq.
7-1
4854H 1360-010-000
e

TABLE 7-1
EFFECT OF ALTERNATIVE ASSUMPTIONS ON
THE DIOXIN EFFLUENT LIMIT AT AN ALLOWABLE RISK LEVEL OF 1 X 10-6
Cancer
Potency Bioaccumulation Ingestion Effluent
Factor Factor Rate Cooking Limit
(mg/kg/day)-1 (liters/kilogram) (gms/day) Loss (oocrd)
(156,000] [5,000] [6.5] [0] [0.1]
156,000 (1) 1563 6.5 0.25 0.4
17,500 (2) 1563 6.5 0.25 3.8
10,000 (3) 1563 6.5 0.25 6.7
(10.0) (4) 1563 6.5 0.25 665
156,000 (1) 1563 1.65 0.25 1.7
17,500 (2) 1563 1.65 0.25 15.0
10,000 (3) 1563 1.65 0.25 26.2
(10.0) (4) 1563 1.'65 0.25 2,621
156,000 (1) 1563 0.049 0.25 56.6
17,500 (2) 1563 0.049 0.25 504
10,000 (3) 1563 0.049 0.25 883
(10.0) (4) 1563 0.049 0.25 88,266
156,000 (1) 1563 0.007 0.25 396
17,500 (2) 1563 0.007 0.25 3,531
10,000 (3) 1563 0.007 0.25 6,179
(10.0) (4) 1563 0.007 0.25 617,859
[ ] The numbers surrounded by "[ ]" are the proposed permit assumptions
used by USEPA in the permit calculations.
(1) Current U.S. EPA Cancer Potency Factor
(2) Current U.S. FDA �Cancer Potency Factor
(3) Proposed U.S. EPA Cancer Potency Factor
(4) Canada Health and Welfare and Province of Ontario risk specific
dose at a 1 X 10-6 excess lifetime cancer risk assuming dioxin
has a threshold.
4891H 1360-010-000
7-2
e

REFERENCES
Anderson, P.D. 1988 . Scientific Origins of Incompatibility in
Risk Assessment . Stat . Science, 3 : 320-327 .
Brett, J.R. , J.E. Shelborne and C.T. Schoop . 1969 . Growth
rate and body composition of fingering sockeye salmon
Oncorhynchlis nerka in relation to temperature and ration
size . J. Fish. Res. Bd. Can. 26 : 2363-2394 .
Can Tox. 1989 . Biological Risk Assessment for 2 , 3 , 7, 8-Tetra-
chlorodibenzo-p-dioxin. Ontario, Canada .
Connecticut (State of Connecticut) . 1987 . A survey of fishing
and fish consumption patterns among Connecticut Sports
Fishermen. Abstract . Department of Health Services .
Cordle, F. , R. Locke, and J. Springer . 1982 . Risk assessment
in a federal regulatory agency: An assessment of. the risk
associated with the human consumption of some species of
fish contaminated with polychlorinated biphenyls (PCBs) .
Env . Health Persp. , 45 : 171-182 .
Crouch, E. and R. Wilson. 1979 . Interspecies comparison of
carcinogenic potency. J. Toxicol Environ. Health
5 : 1095-1118 .
Crump, L. S . , R.B. Howe, and M.B . Fiering . 1980 . Approaches to
carcinogenic, mutagenic, and teratogenic risk assessment .
Report prepared by Science Research Systems , Inc . , for
U.S. Environmental Protection Agency Contract Bo .
68 : 01-5975 .
Czuczwa, J.M. , and R.A. Hites . 1985 . Dioxins and Dibenzofurans
in Air, Soil and Water . In: Dioxins in the Environment,
ed. by M.A. Kamrin and P.W. Rodgers , Hemisphere Publishing
Corp . , N.Y. , N.Y.
Denmark. 1984 . Formation and emission of dioxins especially in
connection with waste incineration. Miljostyrelson,
Strandgrade 29 , 1401 Copenhagen. (Cited in NCASI , 1987) .
EA Engineering, Science and Technology, Inc. 1988 . Synoptic
Survey of Physical and Biological Condition of the Pigeon
River in the Vicinity of Champion International ' s Canton
Mill .
EA Engineering, Science and Technology, Inc . 1989 . Estimation
of the Annual Amount (Weight) of Edible Fish Available
(Maximum Sustainable Yield) from the Pigeon River from
Canton, North Carolina to its Confluence with the French
Broad River.
4822H 1360-010

REFERENCES (Continued)
Freeman, R.A. and J.M. Schroy. 1985 . Environmental Mobility of
TCDD. Chemosphere 14 ( 6/7) : 873-876 .
Friess , S. L. 1989 . Issues Related to the Assessment of
Potential Human Carcinogenic Risks Associated with
Investigation of Total Dioxin Equivalents (TEQ) in Pigeon
River Water and Fish.
Germany, (Federal Environmental Agency and Federal Health
Office) . Report on Dioxins-Update to November 1984 .
(Cited in NCASI, 1987) .
Humphrey, H.E.S. , H. Price, and M. Budd. 1976 . Evaluation of
changes if the level of polychlorinated biphenyls (PCB) in
human tissue. Final Report on FDA Contract 233-73-2209 .
Cited in: Maxim, L .D. and L. Harrington. 1984 . A review
of the Food and Drug Administration risk analysis for
polychlorinated biphenyls in fish. Reg . Toxicol . and
Pharm. , 4 : 192-219 .
Kimbrough, R.D. , H. Falk, P. Stehr, and G. Fries . 1984 .
Health implications of 2 , 3 , 7, 8-tetrachlorodibenzo-p-dioxin
(TCDD) contamination of residential soil . In: Public
Health Risks of the Dioxins . Ed. W.W. Lowrance. Los
Altos, CA: William Kaufman, 121-150 . (Cited in NCASI ,
1987) .
Kleeman, J.M. , J.R. Olson, S .M. Chen, and R.E. Peterson.
1986a . Metabolism & disposition of
2 , 3 , 7, 8-Tetrachlorodibenzo-p-dioxin in Rainbow Trout .
Toxicol . Appl . Pharm. 83 : 391-401 .
Kleeman, J.M. , J.R. Olson, S.M. Chen, and R.E. Peterson.
1986b. 2 , 3 , 7, 8-Tetrachlorodibenzo-p-dioxin metabolism and
disposition in yellow perch. Toxicol . Appl. Pharm.
83 : 401-411 .
Kociba, R.J. 1984 . Summary and critique of rodent
carcinogenicity studies of chlorinated dibenzo-p-dioxins .
In: Public Health Risks of the Dioxins . Ed. W.W.
Lowrance . Los Altos , CA: William Kaufman. (Cited in
NCASI , 1987) .
Kociba, R.J. , D.G . Keyes, J.E. Beyer, R.M. Carreon, C.E. Wade,
D.A. Henber, R.P. Kalnins , L.E. Frauson, C.N. Park, S.D.
Barnard, R.A. Hummel, and C.G. Humiston. 1978 . Results
of a two-year chronic toxicity and oncogenicity study of
2 , 3 , 7, 8-tetrachlorodibenzo-p-dioxin in rats . Toxicology
and Applied Pharmacology, 46 :279-303 .
4822H 1360-010
a

REFERENCES (Continued)
Kuehl , D.W. , P.M. Cook, A.R. Batterman, D.L. Lothenbach, and
B.C. Butterworth. 1987. Bioavailability of
polychlorinated dibenzo-P-dioxins and dibenzofurans from
contaminated Wisconsin River sediment to carp . Chemosphere
16 : 667-679 .
Lyman, W. J. , W.F. Reehl, and D.H. Rosenblatt . 1982 . Handbook
of Chemical Property Estimation Methods . McGraw-Hill Book
Co . , N.Y.
McCallum, M. 1985 . Recreational and subsistence catch and
consumption of seafood from three urban industrial bays of
Puget Sound: Port Gardner, Elliot Bay, and Sinclair
Inlet . Washington Department of Social and Health
Services , Olympia, WA. 59 pp . Cited in: Guidance Manual
for Health Risk Assessment of Chemically Contaminated
Seafood. Tetra Tech, Inc. 1986 .
Mill, T. 1985 . Prediction of the Environmental Fate of
Tetrachlorodibenzodioxin. IN: Dioxins in the
Enron en . ed. by M.A. Kamrin and P.W. Rodgers,
Hemisphere Publishing Corp. , N.Y. , N.Y.
Moolgavkar, S .H. and A.G . Knudson, Jr . 1981. Mutation and
Cancer : A Model for Human Carcinogenesis, Jour . Nat .
Cancer Inst . , 66 : 1037-1052 .
Muir, D.C.G . 1988 . Bioaccumulation and effects of chlorinated
dibenzodioxins and furans in fish, shellfish and
crustacea . A brief review Technical Report . Dept .
Fisheries and Oceans . , Winnipeg, Ontario .
National Research Council Canada (NRCC) . 1981 . Polychlorinated
Dibenzo-p-Dioxins : Criteria for Their Effects on Man and
His Environment, NRCC No. 18574 , Ottawa, Canada .
NCASI (National Council of the Paper Industry for Air and
Stream Improvement, Inc. ) 1988 . U.S . EPA/Paper Industry
Cooperative Dioxin Screening Study, Tech. Bull . No . 545 .
NCASI (National Council of the Paper Industry for Air and
Stream Improvement, Inc. ) 1987 . Descriptions of
Univariate Statistical Models for Use in Environmental
Data Analyses and Means for Predicting Their
Goodness-of-Fit . Tech. Bull . No . 530 .
NCASI, (National 'Council of the Paper Industry for Air and
Stream Improvement, Inc . ) . 1987 . Dioxin: A Critical
Review of its Distribution, Mechanism of Action, Impacts
on Human Health, and the Setting of Acceptable Exposure
Limits . Technical Bulletin Number 524 . New York, NY.
4822H 1360-010
0

EFFLUENT DISCHARGE COLOR REDUCTION PROJECT
CHAMPION INTERNATIONAL CORPORATION
CANTON MILL
CANTON, NORTH CAROLINA
FEBRUARY 1987

CHAMPION INTERNATIONAL CORPORATION
CANTON, NORTH CAROLINA MILL
FEBRUARY 1987
EFFLUENT DISCHARGE COLOR REDUCTION PROJECT
-SUMMARY
Champion International Corporation proposes to modernize
it's Canton, North Carolina facility to improve simultaneously
the mill 's cost competitiveness and effluent discharge to the
Pigeon River. The modernization project calls for Champion
to install new equipment and control technology which will reduce
its effluent color by approximately 50% from its current average
level. A reduction of this magnitude will make the Canton Mill
the lowest color discharger of all bleached kraft pulp and paper
mills in North America. It is already one of the lowest color
mills 'in the industry today.
Champion is committed to the implementation of this
project, costing in excess of $200 million, provided that an
acceptable NPD,ES permit agreement can be reached with state and
federal agencies. Following agreement on permit specifications,
a five-year engineering and construction period would be required
to complete the mill modernization program. The color reduction
portion of the project, however, could be completed within the
first three-year period .
This report summarizes the mill' s historical efforts
regarding effluent discharge and provides details on the proposed
project technology. It also identifies proposed effluent
discharge color standards and reviews Pigeon River modelling
studies .
BACKGROUND
The Canton Mill began operation in 1908 and has a capacity
of 1900 tons per day of bleached kraft paper products including
milk carton stock, envelope paper, copier paper, and other
printing and writing paper products. Pulp for the six paper
machines is produced from hardwoods and pine by the kraft
cooking process. The pulp is bleached in sequential stages using
chlorine, sodium hydroxide, calcium hypochlorite, and chlorine

2
dioxide bleaching reagents. The mill' s effluent flows through a
modern primary and secondary wastewater treatment plant and
discharges into the Pigeon River at approximately 44 million
gallons per day. The annual average river flow is approximately
200 million gallons per day.
Since the late 1950 's, Champion has taken major steps to
improve the quality of its discharge to the Pigeon River . In
1959, a trickling filter was installed to reduce organic matter
in the mill' s effluent. This unit was followed by the
installation of three primary clarifiers in 1960 and 1965, and an
activated sludge secondary treatment system in 1970. Through
19861 Champion has spent nearly $25.0 million for equipment to
improve the quality of mill effluent (Table I) .
Concurrent with the installation of a modern wastewater
treatment system, Champion has also pursued conservation
technologies to reduce its effluent flow. Since 1973, fresh
water usage has declined nearly 20% from approximately 54 million
gallons per day to 44 million a day as a consequence of numerous
recycling and conservation practices.
Champion ' s concern for the issue of color reduction, due to
the small size of the Pigeon River , has also been demonstrated
during the past two decades. During the 1970 ' s, Champion
participated with the U.S. Environmental Protection Agency in
research involving color reduction technologies. Champion
pioneered the concept of ultrafiltration technology and completed
two pilot studies in 1973 and 1980 . From this work, and later
developments in membrane technologies during the 19801s, it was
anticipated that significant color reduction could be achieved in
a cost-effective manner using ultrafiltration. In 1984 and 1985,
Champion invested nearly $1.0 million in a demonstration
facility. After operating the facility for one year, it was
conclusively demonstrated that this technology was neither
technically nor economically feasible. This conclusion, plus the
need to strengthen and improve the mill ' s economic viability, led
to a comprehensive review of- all possible avenues to reduce
effluent color in a cost-effective manner. This effort was
completed in late 1986 and the results formed the basis for the
current project.
PROJECT BASIS
The two major strategic issues directly affecting the
economic viability of the Canton Mill are its ability to control
manufacturing costs and the need to resolve the issue of water
quality in the Pigeon River. These issues are inseparable and
their association is supported by the factors listed below.
(1) The mill is experiencing a significant, continuing
decline of earnings and the financial performance measures of
Percent Return on Sales and Capital Employed are currently at

3
TABLE I
ENVIRONMENTAL CAPITAL EXPENDITURES
FOR
WASTEWATER TREATMENT AND SOLID WASTE DISPOSAL
1959 Dow-pack Trickling Filter $ 873 ,650
1960 One 125 Ft. Primary Clarifier 483 , 874
and Equipment
1965 Two (2) 200` Ft. Primary Clarifiers 2,187 ,264
Screens, Pumps and No. 4 Sewer
1970 Secondary Waste Water 4,610,659
Treatment Plant
1975-76 Sludge Dewatering System 2,264,099
Passavant Presses
1977 Water Treatment - Modernization 1, 480 , 835
Sludge Removal
1977 One 150 Ft. Secondary 1,390 ,625
Clarifier and Equipment
1979-80 Sludge Removal - 3 ,530 ,503
Tait-Andritz Belt Presses
1973-80 Landfills 1,732,282
1965-82 Miscellaneous Additions-Electrical 2,469, 426
Pumps, Piping, Motors, Landfill
Equipment, etc.
1983 Landfill Property, Dissolved Oxygen 428,754
Monitoring System, Computer, Sewer
Monitoring System & Misc.
1984 Waste Treatment Plant Electrical 645,498
Modifications, Front End Loader ,
Misc.
1985 Storm Water Collection-New Chip Sys . 204,606
Oxygen Enrichment System 1,469,365
Waste Treatment Plant Improvements 316,422
Sludge Trucks 206,926
1984-85 Ultrafiltration Demonstration 578,000
Facility
Total $24, 872,788

4
their lowest point during the last twenty years . These
significant trends are threatening the economic viability of the
mill.
(2) During the past two decades, the mill has spent
approximately 26% of its total capital budget to purchase
equipment necessary to meet environmental regulations. Compared
to other pulp and paper mills, Canton ' s expenditures were
significantly higher in the late 1970 ' s and early 19801s . These
incremental capital investments are directly associated with the
mill' s efforts to comply with stringent environmental standards .
(3) During the ten year period from 1976-1985, the mill ' s
annual operating expenses for environmental controls increased
from $5 million to $19 million, an increase of 16% per year.
This is the single largest cost escalation rate in the mill
today, and much higher than the inflationary rate during this
time period .
(4) The Canton Mill is already a low color discharger
compared to an industry survey and ranks with the best and newest
bleached kraft mills in the country. Additional costs to remove
incremental color from its effluent will add significantly to
its already high capital and operating cost base .
These factors led Champion to seek new color reduction
technologies that would provide for the economic viability of the
mill.
COLOR REDUCTION TECHNOLOGY REVIEW
The generation of colored organic matter from wood pulp is
principally due to the pulp bleaching process which removes
lignin from wood fiber . Although most of the lignin compounds
are recovered in the mill and used as a fuel to generate steam, a
small percentage enter the bleaching operation and discolor the
non-recyclable effluents from this process operation. The
inability to recycle these streams in Canton' s existing operation
is due to equipment limitations and the sequential arrangement of
chlorine based oxidizing agents. The mill currently uses
chlorine, calcium hypochlorite, and chlorine dioxide as bleach
reagents to bleach both pine and hardwood pulps.
The new technologies presently available for in-mill color
reduction substitute oxygen for chlorine to selectively remove
lignin just ahead of , the pulp bleaching operation. This practice
is called oxygen delignification, and the process change allows
maximum filtrate recycle of a non-corrosive combustible material
to the mill ' s chemical recovery operation (See Figure 1) . As a
consequence, less lignin enters the pulp bleaching operation
resulting in a major reduction in color discharged from this
source.

FIGURE 1
CANTON MILL
CURRENT MILL DIGESTER PULP MANUFACTURING PROCESS
BROWN BROWN
STOCK SCREEN
STOCKEBLEACH
WASHER THICKENER
PLANT
TO PAPER
MACHINES
C PS
TE WEAK
LIQUOR BLACK BROWN FILTRATE
LIQUOR
CHEMICAL WASTE - COLOR
RECOVERY TREATMENT
SYSTEM SYSTEM STREAMS
PROPOSED MILL
DIGESTER OXYGEN
DELIGNIFICATION
BROWN BROWN ;'.•'
STOCK SCREEN STOCK PULP
WASHER ROOM WASHER :i : "•: WASHER BLEACH
TO PAPER
tTREAT
NT MACHINES
M
L
A
,t r ,
�Fl
CHIPS WHITE WEAK R
LIQUOR BLACK BROWN OXYGEN MENTFILTRATE FILTRATE
REDUCED `�
MODIFIED CHEMICAL FILTRATE WASTECOLOR
RECOVERY SYSTEM RECLAMATION TREATMENTSTREAMS
SYSTEM SYSTEM

6
To implement this process change, Champion proposes to
install two new oxygen delignification systems to process both
pine and hardwood pulps. In addition, it will also install a new
pine bleach plant and modify an existing one to process hardwood
pulp. These changes will necessitate, the installation of two
improved pulp screening, washing , and recycle systems just
ahead of the bleaching operation in order to increase recycle of
unbleached pulp filtrates. These filtrates will then be
processed through six stages of evaporation and the organics
burned to carbon dioxide and water in the recovery boiler
operation. Modifications to the evaporators and recovery boilers
will also be completed to handle the incremental load of
chemicals being recycled in the mill. The inorganics will be
reclaimed from the boiler smelt and reused as make-up chemicals
for subsequent cooking of pine and hardwood chips. All of these
process operations will also require the installation of computer
control capability to optimize the balance of chemicals, steam,
pulp, and filtrate within the pulp mill system.
In addition to the major changes proposed above, Champion
also intends to install several filtrate reclamation systems that
will eliminate a portion of the flow and color from reaching the
waste treatment system. These smaller waste streams will also be
processed through the evaporators and eventually reach the mill' s
chemical recovery cycle. Several new large storage tanks will
also be added in the pulp mill..to better, balance chemical
inventories in the chemical recovery cycle.
As a final step, Champion also intends to install a
chemical mixer to selectively treat the caustic extraction stage
effluents with an oxidizing agent to remove additional color .
Champion proposes to use chlorine in this application and will be
treating a non-recyclable alkaline stream. The net effect of the
above process changes are documented in the following section.
Table II summarizes the color reduction technologies that were
evaluated by Champion for possible application at Canton.
EFFLUENT- CHARACTERISTICS
The test method used by Champion for color measurement was
developed by the National Council for Air and Stream Improvement
(NCASI) for the pulp and paper industry and adopted by EPA for
the measurement of mill effluent color (Federal Register Vol. 39,
No. 104 (1872-18754) May 291 1974) . Color concentration
measurements are expressed as milligrams per liter as platinum
cobalt color units. Discharge rates to the Pigeon River are
expressed in pounds per day, which is the product of color
concentration and flow to the river .
The generation of individual colored effluent streams
within the mill is .almost entirely due to the pulp processing
' operations. Color contributions from any one area are tied to
both scheduled and unscheduled repair outages, wood specie mix,

TABLE II
COLOR TECHNOLOGY REVIEW'
REMOVE/REDUCE THE SOURCE TREAT THE PROBLEM
(In-Mill) (End-of-Pipe)
I . Pulp Delignification/Bleaching Process I. Select Stream Treatment Methods
A. Extended Delignification Pulping A. Oxidation. of Extraction '
B. Washer Efficiency Studies Stage Filtrates
C. Oxygen Delignification Studies 1. Calcium hypochlorite oxidation
1. % Lignin reduction vs. color (2 levels)
2. Oxygen stage carryover (3 levels) 2. Chlorine oxidation
D. Alternate Bleaching Studies 3 . Chlorine dioxide oxidation
1. Short cycle vs. conventional (2 levels)
sequence 4. Hydrogen peroxide oxidation
2. Non-oxygen bleaching baseline (2 levels)
3. Chlorine dioxide substitution 5. Ozone oxidation
for chlorine (4 levels) 6. Ozone-UV activated oxidation
4. Caustic extraction with oxygen 7 . Potassium permanganate
5. D4 and sodium hypochlorite oxidation
(2 levels) 8. Sodium peroxide oxidation
6 . D4 and hydrogen peroxide 9. Sodium hypochlorite oxidation
(4 levels) B . Separation Processes on Extrac-
E'. Mill Scale Sodium Hypochlorite tion Stage Filtrates
Trial (49 days) 1. Carbon absorption/acid
regeneration
II. Color Containment Studies 2. Evaporation
A. Lithium Tracer Studies 3. Microfiltration
B. Screen Room Closure 4. Percolation through flyash
C. Filtrate Containment/Recycle 5. Precipitation by alum/air
D. Mechanical Pump Seals flotation
6. Precipitation by polymer/
air flotation
7 . B5 and B6 combined
B. Precipitation by ferric
chloride/air flotation
9. Precipitation by lime/gravity
separation
10. Ultrafiltration
11. Water Chem, Inc. (proprietary
chemistry)

TABLE II
COLOR TECHNOLOGY REVIEW
(Continued)
REMOVE/REDUCE THE SOURCE TREAT THE PROBLEM
(In-Mill) (End Pipe)
III . Characteristics I . C. Disposal Processes For
A. Commercial Installations Concentrated Color Streams
B. Replacement Capital Dollars 1. Wet air oxidation
C. Operating Cost Reduction 2. Dedicated incinerator
D. Strong Industry Support 3. Ozone destruction
E. Few, if any, Environmental 4. Solid waste landfill
Side Effects 5. Lime kiln incineration
F. State-of-The-Art Technology 6 . Additive for adhesives
D. Treatment of Combined Mill
Wastewater
1. Biological augmentation
2. Wastewater aeration and
oxidation
3 . Precipitation by alum,
polymers, lime, calcium,
and ferric ion
4. WaterChem, Inc. (proprietary
chemistry)
II . Characteristics,
A. Mostly Developing Technologies
B. New Capital Dollars
C. Operating Cost Increase
D. Little Industry Support
E. Possible Adverse Environmental
Side Effects
F. Technical and Engineering
Risks
m
i

9
and degree of process recycle. As a consequence, daily process
changes often cause a wide range of color variation in the
numerous .mill sewers that feed the waste treatment plant. These
swings are only partially dampened as the effluent moves through
the waste treatment system. The variability and long-term ,
averages for 1986 effluent discharge rates are summarized in
Table III . Figures are listed for two time periods: a full year
of operation representing 365 days and a partial year of
operation representing 300 days. The partial year figures are
' proposed as more typical of normal operations because- they
exclude three abnormal events: (1) A cold mill shutdown from
April 28,, 1986, through May 1, 1986; (2) a production curtailment
from August 4, 1986, through August 22, 1986, due to the record
drought in Western North Carolina; and (3) a sodium hypochlorite
bleach trial starting on November 20, 1986, and extending through
year-end. The bleach trial was initiated to further evaluate
the color reduction mechanisms already in place in the mill:
Removing these isolated events from the daily figures provides a
more representative base of typical process operations .
Consequently, these figures have been selected as the current
base for calculating color standards for future mill operation.
The effect of using the more representative data base is to lower
' the present average discharge rates and results in a more
stringent proposed color limitation.
TABLE III
CANTON MILL
EFFLUENT DISCHARGE COLOR
(Lbs/Day)
Time Period Average Discharge Standard Deviation
1986-365 Days 356,008 120, 944
1986-300 Days (1) 344,735 (1) 74,315 (1)
(1) Excludes three abnormal events during the year associated
with a cold mill shutdown, a production curtailment due ,
to the summer drought, and a sodium hypochlorite bleach
trial.

10
The establishment of new color specifications for the
restructured Canton Mill hinges on the combined application of
process technologies discussed previously. A 50% color
reduction is achievable on both long-term average and short-term
variability as a result of the modernization program. These
reductions, once achieved , would rank Canton as the lowest color
mill in North America for bleached kraft mills of similar size
and wood specie utilization. The new permit specifications
proposed for the mill are listed in Table IV.
TABLE IV
CANTON MILL
PROPOSED COLOR DISCHARGE LIMITS
(Lbs/Day at Point of Discharge)
1986 Base Proposed
Maximum Allowable (300 Days) Limits
Annual Average (1) 344;735 172,368
Monthly Average (2) - 258,945
(1) The "Annual Average Discharge" is defined as the arithmetic
mean of the total mass of all reported daily discharges
during a calendar year divided by the number of days the
tests were reported .
(2) The "Monthly Average Discharge" is defined as the arithmetic
mean of the total mass of all reported daily discharges
during a calendar month divided by the number of days the
tests were reported.
The annual average proposed limit is 50% less than the
current annual average established for 1986 operations. The
maximum allowable monthly average represents the proposed limit
calculated at the 99th percentile rank based on an improved
variability. Clearly, these limits taken as a group impose a
strict tolerance on the mill to comply with the lowest color
discharge possible on an everyday basis. Minor variations can be
accommodated but, over the long term, the mill must comply with
the 172,368 lbs/day limit. This limit in effect guarantees a 50%
reduction in color discharged to the Pigeon River . The impact of

11
this reduction at the North Carolina-Tennessee border is
summarized in the following section.
PIGEON RIVER MODELLING STUDIES
The Pigeon River modelling studies conducted by Champion
utilized a modified dilution model to predict color
concentrations at various points downstream from the mill.
Figure 2 illustrates a diagram of the Pigeon River flow from the
mill (mileage mark 63 .5) to its confluence at the French Broad
River in Tennessee. The flow model assumes that color is
conserved and its concentration is influenced' by additional fresh
water dilution as well as retention time in Walters Lake. The
color concentration from the mill can be accurately predicted to
the Hepco sampling station (mile mark 42 .6) , just upstream from
the entrance to Walters Lake, by the following equation:
HEc = (WTPc * WTPf) + (HEf - WTPf) * Dc (1)
HEf
Where WTPc = Waste treatment plant discharge color (ppm)
WTPf = Waste treatment plant discharge flow (MGD)
HEf = Flow at Hepco sampling station
HEc = Color concentration at Hepco sampling station
Dc = Color concentration of all dilution streams,
(13
To mathematically predict the color concentrations across Walters
Lake to the North Carolina-Tennessee state line, an additional
equation was developed. This equation corrected for the observed
differences in color due to the reduction in the lake, and is
described below.
(-0 .22.4 * (log HEf) + 0.781)
F = 10 (2)
Where F = Ratio of color concentration at Hepco
sampling station (mileage 42.6) and the
I-40 bridge in Tennessee (mileage 24.7) .
The completed river model equation combining the dilution effect
from several tributaries and the reduction effect across the lake
is described in equation (3) .
State Line Color = (WTPc * WTPf) + (HEf - WTPf) * Dc (3)
(-0.224 * log (HEf) + 0 .781)
HEf * 10

FIGURE 2
Pigeon River Diagram
CHAMPION
DISCHARGE
HEPCO
CP&L POWERHOUSE
PIPELINE
alters Lake
E N WPORT
NATURAL
RIVER
BASIN FRENCH
BROAD
RIVER
63.5 42.6 State Line 24.7 5.0
RIVER MILE MARKS (not to scale)

13
Historical river flow data used to develop this equation was
obtained from the United States Geological Service (USGS) records
for a period in excess of fifty years. Color data were obtained
from Champion' s data base and averaged over monthly periods to
minimize the effect of time lag to the state border. The state
line color data at the I-40 bridge included only periods when
the CP&L powerhouse was discharging water.
Using the final equation, Figure 3 illustrates the actual
color concentration values at the state line with today' s
discharge and the predicted values with the modernized mill.
Table V summarizes the frequency distributions at specific color
concentrations of 40, 50, 60 , 80, and 100 ppm.
TABLE V
CONCENTRATION DISTRIBUTIONS
Percent of Time Below a Specific Concentration
(North Carolina-Tennessee State Line)
Color Concentration (ppm)
Mill Configuration 40 50 60 80 100
. Current Mill 5% 12% 23% 47% 64%
Modernized Mill 40% 59% 73% 90% >95%
The model predicts that the modernized mill will achieve 100 ppm
color or less at the Tennessee border greater than 95% of the time
as opposed to 64% today. Additionally, a color concentration of
50 ppm or less will be achieved 59% of the time as opposed to
only 12% of the time now.
ENGINEERING AND CONSTRUCTION TIMETABLE
Once an agreement is reached on permit limitations, it is
anticipated that the entire modernization program can be
accomplished within a five-year period. The color reduction
portion of the project, however, can be completed within three
years. Anticipating regulatory agreement, Champion has already
contracted with CRS Sirrine, Inc. , of Greenville, South Carolina,
to begin preliminary engineering and refine capital cost and
construction estimates.

FIGURE 3
COLOR MODEL FOR STATE LINE
Modeled by Dilution with Lake Influence
200
19 Current Discharge
180
170
160
E 150
a
a 140
ai 130
o . 120
> 11
p 10 Projected Discharge
6
U 90
80
-1 70
°) 60
D
N 50
40
30
20
10
0
0 20 40• 60 80 100
% Time Below Indicated Color Value

15
CONCLUSIONS
Champion International Corporation is prepared to address
the Pigeon River color issue with an in-mill modernization
project that incorporates state-of-the-art technology to achieve
a 50% color reduction. The technology selected for use is a
combination of several commercially proven approaches providing
color and operating cost reduction while avoiding adverse
environmental side effects.
The proposals in Champion ' s project will achieve a
significant improvement in color discharged to the Pigeon River .
Color concentrations at the North Carolina-Tennessee border will
be below 50 ppm nearly 60% of the time, with almost all
measurements below 100 ppm.

Champion International Corporation
Canton Mill
EFFLUENT DISCHARGE COLOR REDUCTION PROJECT
MAJOR PROCESS MODIFICATIONS
SCREEN ROOM
The unbleached pulp produced by digesting wood chips in
the presence of caustic soda and sodium sulfide is washed
with fresh water to remove soluble lignin derivatives. After
washing , the pulp is processed through several stages of
screening to remove undigested small wood particles. This
screening is normally done at low consistency (1%) after
which the pulp is thickened to 10-12% consistency before
storage and subsequent bleaching .
The filtrate generated during the thickening operation
is brown in color and can be either sewered in an open system
or recycled to pulp washing in a closed system. This
recycling results in an increase in dissolved solids. The
gravity discharge Cowan screens used at the Canton Mill are
older models which cannot function properly in the presence
of higher concentrations of dissolved solids. Therefore, new
pressure screens must be installed and modifications made to
the washing system to permit filtrate recycling and reduce
color losses to the waste treatment system. This modifi-
cation results in "closing" the screen room.
OXYGEN DELIGNIFICATON
In an oxygen delignification system, washed unbleached
pulp is mixed with caustic soda and oxygen at elevated
temperatures and pressures. In a period of approximately one
hour, about one-half of the incoming quantity of lignin is
made soluble by the reaction with oxygen. Pushing the
reaction further results in a rapid degradation of
carbohydrate chain length and a reduced pulp strength. Since
oxygen delignification does not utilize any chlorine
containing streams, its filtrate can be recycled through the
brown stock washing system to the recovery boiler. The
lignin in the liquor is burned in the boiler and the
inorganic chemicals recovered in the smelt.
The oxygen treated pulp contains less lignin and
consequently requires less chlorine to reach a required pulp
brightness. This process change significantly reduces the
discharge of chlorinated and colored organics from the
bleaching operation. Figure 1 illustrates changes in the
screen room and the proposed oxygen delignification system.

Champion International Corporation
Canton Mill
BLEACH PLANT MODIFICATIONS
There are three existing bleach plants at the Canton
Mill, two for processing pine and one for hardwood pulp. The
current plans are to install a new pine bleach plant and
modify an existing one for hardwood, with both having a
(C+D) Eo D sequence. The new bleach lines will be preceded
by two oxygen delignification systems and have computer
controlled production rate and bleaching capability. Figure
2 illustrates a potential design of a modern short-sequence
bleaching facility.

FIGURE 1
CANTON MILL
CURRENT MILL PULP MANUFACTURING PROCESS
DIGESTER
BROWN SCREEN BROWN
STOCK ROOM STOCK
WASHER THICKENER TO BLEACH PLANT
C WHITE WEAK
LIQUOR BLACK BROWN FILTRATE
LIQUOR
CHEMICAL = TO SEWER
RECOVERY .
SYSTEM
PROPOSED MILL
DIGESTER OXYGEN
DELIGNIFICATION
BROWN BROWN
STOCK SCREEN STOCK ti; ;;ti;• PULP
WASHER ROOM WASHER :lf'; WASHER TO BLEACH PLANT
ti;ti
dtir:k•:ti
i'
S WHITE WEAK
LIQUOR' BLACK BROWN OXYGEN
LIQUOR FILTRATE FILTRATE
CHEMICAL
RECOVERY
SYSTEM .

FIGURE
CANTON M%[ PAPER MILL
PROPOSED BLEACHING OPERATION RECYCLE WATER i
} ST BEkNf \
. gon4E \
WASHER :r WASHER WASHER © \
\
FROM OXYGEN '
v LIGmmQr ON
WASHER 2 q !
� a . . ..
{CHLORIN } OXIDATIVE CHLORINE
* : EXTRA- DIOXIDE .
} CHLORIN
�nowN
aIOGK .
Ll
COLOR .
TREATMENT

it t
ON
IF
Fix-
v -; :+

gV Io1vlq!JaQ spun$ 'oang ap{I V
Pun 'gnsuo3uoo/J 'u00{7 mma8ul
SI 6dolompal u38,(xo Jo piluaiod amin3 •Z
Xdl 'Igvpu'oyS Iagul
Pilo ')Idl 'JPIOW vulJsUgD
6I Su!goealq ua8Axo;o main sallpoginbV .q
&_LS 'SlagzugqS sm7
Puv 'Ii.LS 'pv;68uu;1 MUM
ZI Sucgaealq ua26xo jo s»adse leivaumonnug 'S
gV p"1a11!g 'uvms vjug
Pun 'gV IIaJoQoW 'wnssoi vialg
II dlnd pagaealq-ua26xojo Alquno {
gV Y!i,7a1 VJS 'j!a,unl imunD
L shadse 3cuiouoja i!agl pue sucals6s 8u!q:)ealg g
IjlS 'uaJ88un(7 uals
q uoyeagluggap uaSAxo jo ,Clls!uiagrxalduio:)aql •Z
S uotimpollul 'I
gV lo1vlq!jaa spung 'oang ap 11v
Pilo '1insuo3uoo.c3 'uooij Ivwasul
5a8ed fummunS
'salalpn 8uimollof ag; u! podal s!g; u!
paz!.cniuiuns s! pun 586I 'I HOW pun 0961 'I isn8n y ccaam;aq palanpuoa svm. Rpnis aq L
:coiv.cq!fap spunS 'oang ap fly pun '11nsuo3uoojD 1110013 lnuiasul
Jo uoisicuadns ayI .capon Ino pa!Um aq p1nogs„'1vi1ua1od aminf sl! pun n8olougla;
uaSRxo fo sn;nis IuaclnD_ pall!lua 'fipn1s a Inch pap!jap aaofaxat 1 snm If 'Rcaucdolaaap
ssa.7o.id gs!pamS Jo auolsaauco_' n sv a1o.c 5ii8oloutlaal ua8fixo Jo aan;nd snoauo.ua
pun Su!pvalsicu Rlaucacl.ra tin oiul padolaaap aingap agi 'sjaaias-go pauuofu!-11am of
-patio!Isanb aaam ssaupunos lo-mucouola
pun Ina!ugaal sp pun 'RSolougial ua8fixo Jo slaaf fa Inluamuogaua aa!i!sod agl
•Eg6I Suunp ssaid filpnp gs!pamS ag; pun slncunol apn�l gjoq ui pagsllgnd a�am RSolou
-gaal ua8fixo Jo si.7af fa 1n!a.yauaq Inogv signoQ •8u!g_iva1q uaBRxo of painqui;n alacn
'aauap!aa a!f!irvus ainlosgv Inog;lm. 'ga!gm slaa f fa lnivacuuo.c!aua pamogs anblugaal
8ulgivalq uaSRxo aril Sulsn sppiu dlnd ap!sjno slua!dhai ay; u!pauuof iad salpn;s auuoS

Summary
It is impossible to give a clear-cut recommendation of any In general, the introduction of oxygen bleaching therefo
single process alternative to achieve optimum total economy in clearly rer.esents a highly significant environmental pri
the manufacture of bleached sulfate pulp. Today it is often a tective r.,easure.
matter of expanding or modernizing existing plants where The few investigations carried out to date regarding the it
existing processand plant space isa major consideration.How- fluenceof oxygen bleaching onthedischarge from theplantai
ever, even in the case of new investments, in addition to plant inconclusive.This is most probably due to variations in wasl
space and localization, such parameters as environmental ing losses. If this process were better controlled, systemat
demands and prices of timber, utilities, chemicals and capital investigations should here also show that the introduction e
also play a role. oxygen bleaching is favorable from an environmental slant
The technical and economical evaluation showed that both point.
oxygen bleaching alternatives, 0(CD)(EO)D and Thedecelopmentofthecookinglechnologyhasaconsiderabl
0(CD)(EO)DED, as well as alternatives with modified cook- influence on the possibility of improving the economical an
i
ingst Kappa 25,are l relatively equal value frome a e used in evironmental effects of the bleaching process when an oxyge
cal standpoint when low levels of chlorine dioxide are used in stage is used as thefirst stage after cooking.The introduction c
the chlorine stage. As the amount of chlorine dioxide is in- high sulfidity,atmospheric cooking and computer control has
creased, the alternative with modified cooking and oxygen generally speaking, increased the potential for delivering
bleaching tends to be the most economical.The conventional high-viscosity pulp with stable Kappa number from the di
method with aerated ponds is definitely less economical. gester to the oxygen stage.
Three-stage bleaching appears economically advantageous
over a five-stage process. Higher chlorate prices strengthen During recent years it has been demonstrated that it is possib)
economic potential of the oxygen sequence. to activate the unbleached pulp using certain pretreatment
Thus, the comparison shows that the so-called modified (for example, NO-NO2, 03 or green liquor) to improve th
cooking process should not be regarded as a viable alternative selectivity of a subsequent oxygen stage.
to oxygen bleaching(regardless of whether it is carried out to Oxygen-reinforced alkali stages have been shown tobeboth at
Kappa 25 or 30), but rather as a complement. effective and economical complements in the bleaching se
A careful comparison shows that there are no significant dif- quence,especially as a first alkali stage.This type of treatmen
ferences in quality between oxygen-bleached and non-oxygen- may also be introduced as a first stage in special cases.
bleached sulfate pulps, assuming that the pulps are produced One of the fundamental ideas underlying the increased use o,
under controlled conditions. This finding applies to strength non-chlorine,non-corrosive chemicals in the bleach plant is tc
and optical preperiies eleanliPess and beatability. feed the waste liquor to the recovery cycle, so that the organic
Available know-how has demonstratea Gnat me introduction matter can be burned and sodium recovered.
of oxygen bleaching before a traditional(or reduced)chlorine A Swedish mill is currently recycling waste liquor from the OE
bleaching, combined with the recovery and burning of spent stage. By using oxidized white liquor in both the O and OE
liquor fro.n the oxygen stage,results in a significant reduction ytages and utilizing the flexibility of the chlorine dioxide pro-
in the discharge of BOD, COD and color and ToCI. The re- duction,the plant has succeeded in establishing an economical
duction includes chlorinated and non-chlorinated organic system with low discharges.
matter of both high and low molecular weight, includipg
Khlorophenolic compounds and chloroform.

Introduction
The sulfate process produces an un- consistency technology is progressing.It favor oxygen treatment as an alternative
bleached pulp with a relatively high re- is likely that oxygen stages at medium process (Fig. 2). The short sequence,
sidual lignin content.The lignin ishighly- cwsistenry will be-ome more common O(CD)(EO)D, is particularly interest-
cross-linked. Therefore, it is -fairly in the future. ing in this context.
complicated and expensive to bleach When oxygen delignification was intro-
sulfate pulp. duced, it was discovered that current FIGURE i.
The usual bleaching chemicals(chlorine, cooking technology was not satisfac- MARKET PRICES FOR
alkali and chlorine dioxide) have in- tory. In many mills, the Kappa number BLEACHING CHEMICALS, SEK/tonne
creased greatly in price in recent years varied from 25 to 40. Today, computer
and will presumably continue to do so in control, atmospheric presteaming and SEKN NaC10, _
the future (Figure 1).Oxygen is an inex- chip equalizing systems make it possible
pensive chemical and its energy require- to maintain the Kappa number within 3000
ments during production are signifi- narrower limits(28-321.The economic
cantly less than those of chlorine, alkali result depends to a great extent on 2500
and chlorine dioxide. whether the unbleached pulp can be pro-
The National Swedish Environment duced at a high final viscosity (greater z000
Protection Board's long-term aim for the than 1200) with less variations in the cl•
control of the discharge of bleach plant Kappa number.The industry has a great lsoo
waste l of
a di formulated in bleach
plant
as deal to learn in this respect. if cooking is
carried out successfully, a Kappa num-
follows: ber of about 14 can be achieved follow- O.
— Chlorinated organic compounds ing the oxygen stage without any 500
shall not he discharged. weakening of the bleached pulp. De-
- Waste water shall neither give rise to pending on the variations in wood raw
genetic effects nor contain accumu- material and cooking, many plants to- 1976 7e BO a2 &: 86 yea,
lating substances. day achieve Kappa numbers only in the
— Waste water shall not result in acute 18-20 range.
toxic effects on fish or other aquatic- Economically this is unsatisfactory since FIGURE 2.
organisms. with toda}'s technology the objective RELATIVE BLEACHING COSTS,
— The r shall s reduced.
in matter in waste should be a Kappa number of about 15. USS.adt
water shall be reduced.
It is evident that future developments in Relative bleaching costs.US$/ad:
The National Swedish Environment oxygen technology will be guided pri- 43
Protection Board's demands were for- marily by economical considerations. 46 (D„CKAEO)DED
mulated in the light of an existing effort We can expect more technologies adapt- ar-3led pond
to promote oxygen bleaching and El able to the needs of individual mills.
discharge cleaning, in an attempt to pressurized or non-pressurized oxygen +3
achieve a color reduction equivalent to stages, high consistency (greater than
that resulting from the ion exchange 20%) or medium consistency (8-20%) (Dr.,Ca)(EO)DED
method. The ion exchange method has, in the first stage or in the alkali extraction
g (cwc, )EDeD
however, since fallen into disuse.
stage already exist.
Pressurized oxygen stages were intro- —3 OCo D,,)EODED
duced more than ten years ago on the
basis of the possibility of recycling waste REINFORCED ALKALI STAGES (EO)
liquor from this stage to the recovery —s
system. The degree of delignification Non-pressurized oxygen stages can o(c„D„)(EoiD
achieved is determined by the level of easily beintroduced in existing bleaching
improved selectivity which can be at- plants by utilizing the alkali and hypo- 350 375 4o0 azs 460 475
tained, while maintaining the full bleaching towers.This technology is de- Prices for sodrom chlorale. Us s9pnne
strength of the pulp. veloping rapidly. At present there are
It is important that a well-functioning approximately 40 ones are continually being ins,but new
washer is installed after the oxygen
stage. Today, there are more than 30
installations of this type in the world,the SHORT SEQUENCE
majority of which operate at high pulp
consistency. If chlorine dioxide further increases in
In general it may be said that medium price, economic advantages alone will

The complex chemistry of
oxygen delignification
Today,some knowledge is available con- principle react ions%•.hich occur with lig- In a recently initiated project, STFI
cerning a number of different reaction nin rriginating from the initially formed intends to clarify the molecular kinetics
mechanisms between different lignin p',enoxy radicals.The two types of reac- of these reactions,primarily with respect
structures and oxygen.Knowledge now tions possible are indicated as a.and It.in to such typical structures in residual lig-
needs to be acquired concerning the the figure. In a., a reaction takes place nins as vinyl ether, and stilbene and
molecular kinetics of the reactions — between aromatic structures through cross-linked structures. Comparative
how reactive different types of residual radical coupling(i.e., two phenoxy Tad- studies of reactivity and the speed of deg-
lignin structures are towards oxygen. icals condense and form biphenolic radation in relation to such factors as
Process kinetics investigations earlier structures). This type of reaction in- pH, temperature and oxygen pressure
carried out at STFI have shown that volves a risk of additional cross-linking are also under investigation. With the
during oxygen bleaching. dissolution of in the lignin, making degradation more help of these data,it ishoped to beable to
lignin (measured as a decrease in Kappa difficult. However, it has been shown explain what happens with the lignin
number) has a rapid initial phase fol- that these structures can also be degra- during the two different bleaching
lowed by a slower final delignification. dated to acids, which perhaps compen- phases indicated by the kinetics of the
One current reseaTch project has sought sates for this risk. process.
to collect kinetic data at a molecular Reaction b. is probably referable for
level,from which it is hoped to obtain an p y p
explanation of these two lignin dissolu- lignin degradation and dissolution.This REACTION OF PHENOXY
tion phases. reaction is dominant under alkaline con- RADICALS
ditions (pH>13). Cvclohexadienone
The chemistry involved in oxygen hydroperoxideinterme liates are formed
bleaching is very complex for two rea- through the reaction of phenoxy radicals O O
sons.One is that the structure of residual a� W 0 0_ orl.e
with hydrophenoxy radicals (HOO') or
lignin is m bleached is relatively varied superoxide anions (O'Z) formed from O ,R
and to some extent unknown.The other the oxygen. Depending on the position OW
reason is that oxygen can appear and be in the aromatic ring in which the hydro- ° y oo"O1A
converted to a number of different re- peroxide group is situated,the lignin can e I oo"
active oxidizing species which undergo e °A`
be degraded in three different ways.Side o 0
different types of reactions. In general, chain elimination means that the lignin + 1
there are three fundamentally different structure is broken up between the aro- a,0era,;a. 2)nng 31&IM,.
types of reaction processes between lig- matic hydrocarbons, forming various tlmmairo. rp ring a n..
nin and oxygen in an alkaline medium. aldehydes and organic acids from the arm;.co,
One is the alkaline hydrolysis reactions side chain as well as paraquinones from
which normally also occur during alka- the aromatic part. Paraquinones are During oxygen bleaching importaw drpadabl!
line cooking processes.Here,the lignin is colored structures which can be trans- hydroperoaidchacrinediwesac)orardr:are-
split with the help of hydroxide ions and formed into colorless structures with the artier:b.. bu rearhon beta een the oipgrn and
becomes more water soluble through the the phe ncra .radialscan.formed)rev,der 5o r'lignic
help of the small amounts of hydrogen The vhrnoxy.aa;lah Can. hJseoer.also.ea:r
formation of phenolia groups.The pres- peroxide formed during the bleaching. will;each other in a cmnprtitier reaction via
enee of oxygen also initiates both ionic reaction a See text.
and radical reactions.Ionic oxidation re-
actions occur from ionized species, for
example, hydrogen peroxide anions, I IGN9N IS DISSOLVED MORE
which may form in small quantities EASILY
during the reaction of oxygen in an alka-
line medium and play a role in the re- The ring-opening reaction makes the lig-
moval of color-forming structures such nin more hydrophilic and more easily
as quinones. The formation of radical dissolved.This is due to the fact that the
species is important for the oxygen's break-up of the aromatic ring produces
initial reactions because, depending on water soluble dicarboxylicacids,suchas
the pH level, such radicals may give rise muconic acid or,succinic acid, i al.The
to chain reactions originating from phe- third reaction is demethylation, arising
nolic structures in the lignin. from intermediates where the hydro-
peroxy group has become attached close
to the methoxy group. The methoxy
ADDITIONAL CROSS-LINKING group is then eliminated and methanol is
MAKES.DEGRADATION MORE formed at the same time as the ring opens
DIFFICULT in certain cases.This has less affect on the
degradation of lignin, but may increase
The figure on this page summarizes the its solubility.

Bleaching systems and
their economic aspects
It is impossiblr to mAke a clear-cut rec- sistency gives a lower investment cost A standard allowance has then been
ommendation of any one procef,alter- but higher steam requirements.The cost made to cover all additional costs.
native for achieving optimal t-,tal econ- comparison is not affected by the choice. Investment costs for CI02 production
orny in the production of bleached sul- An evaluation has been made for all include the cost of an SVP(113)faciliiy.
fate pulp. Conditions vary significantly alternatives with both high and low Those cases which involve a high chlo-
from case to case. Today, it is often a chlorine dioxide levels in the prebleach- rine dioxide level in the chlorine stageare
question of expanding or modernizing ins stage. assumed to demand an extra filter stage
existing plants where process and plant Comparisons have been made on the before the chlorine stage. Other process
space is a major consideration. How- assumption that batch cooking is used. solutions are conceivable.
ever, even where new investments are No significant change in the differences
involved, such parameters as environ- between the alternatives should arise if The difference in investment costs for
mental demands and prices of wood, a continuous cooking process were different capacity requirements in the
utilities, chemicals and capital also play applied liquor recovery cycle for varying
a role in addition to plant space andloca- amounts of transferred solids has not
tion. The choice does not become any The transfer of substance to the chlorine been taken into account. To simplify
easier in the light of current develop- stage has been estimated using a com- matters,it has been assumed that the cost
ments, including such new processes as puter program (GEMS). In cases where differences are balanced by energy
modified cooking, which may prove modified cooking is applied,thewashing revenues from the solids.
themselves Worthy of competing for a effect due to digester displacement has Investment costs associated with chlo-
place in the production chain. been taken into account. rine dioxide preparation are based on
One of the questions which then arises Table 2 shows estimated consumption tenders for a couple of plants of different
concerns the role of oxygen bleaching in levels of bleaching chemicals for the sizes (SVP (R3) type). Costs have been
this context. In order to illustrate the various alternatives. In all cases, the calculated for other capacity levels.
economic aspects of this process, four consumption levels are measured at final Table 5 presents a picture of the different
different process alternatives based on bleaching with both very high and total economies of the various alterna-
the production of bleached softwood limited chlorine dioxide levels in the tives when the production and capital
pulp with similar market demands have chlorine stage. costs are weighed together. Capital costs
been selected and their production and Estimated chemical costs were based on are estimated on the basis of a 15-year
capital costs compared, under condi- the following prices(SEK/tonne): annuity at an interest rate of 15 per cent.
tions described in"Assumptions used in The differences are illustrated in block
comparative calculations,•on page S. C12: 1,085 diagrams for the two capacity levels.
C102: 2,200(active 02;
Table 1 shows the process alternatives excl. capital costs) The comparison shows that both alter-
and gives a brief description of their NaOH: 1,280 natives with oxygen bleaching as well as
discharge characteristics. In all alterna- 02: 850 the alternatives with modified cooking
tives, 3-stage, (DC)(EO)D, and 5- Ox. white liquor: 770 to Kappa 25 are more or less equivalent
stage, (DC)(EO)DED, processes have from an economic standpoint when low
been selected for final bleaching. Chemical charges to the digesters are levels of chlorine dioxide are used in the
assumed to be the same in all cases. chlorine stage.As theamount of chlorine
Two of the alternatives include oxygen dioxide increases, the alternative with
bleaching.The others include a conven- Table 3 shows the other differences in modified cooking and oxygen bleaching
tional line with cooking to Kappa 32 production costs. Wood consumption tends to be the most economical for the
supplemented by external waste water has been judged to be 1.8 per cent higher higher capacity. The conventional
treatment in an aerated pond and a in the alternative with modified cooking method using an aerated pond is con-
modified cooking line with extended to Kappa 25 than in the other altema- siderably more costly. Three-stage
delignification to Kappa 25 without Lives. In alternative 1 it is assumed that bleaching appears to be economically
additional discharge reduction. While external waste water treatment requires more favorable than a five-stage process.
the selection of Kappa 25 seems ex- an extra half man per shift, while the
tremely low, it may prove to be an other alternatives require approximately Thus, an essential conclusion drawn
acceptable level in the long term. the same manpower. Maintenance costs from the comparison is that the so-called
are estimated to be 3 per cent of the rela- modified cooking process should not be
The two alternatives applying oxygen tive investment costs. seen as an alternative to oxygen bleach-
bleaching include a conventional line ing, but rather as a complement.
with cooking to Kappa 32 and oxygen The differences in investment costs have
bleaching to Kappa 18. and a line with been calculated for two capacity levels,
modified cooking to Kappa 32 and 600 and 1,000 tonnes bleached pulp per
oxygen bleaching to Kappa 15. Oxygen day (Table 4). Investment costs refer to
bleaching is carried out at a high pulp new installations and are based on ten-
consistency. Bleaching at medium con- ders for machines and other equipment.

ASSUMPTIONS UbhU 11v kvlvi �.�,
CALCULATIONS FOR THE TABLES
Washing press: research laboratory and are, thus, not
PROCESS ALTERNATIVE, (TABLE I) Dewatering efficiency 0.50 linked to any special calculations for this
report.
The assurrrrtions listed below and used Discharged solids 0 30—
in the calculatio•s have provided the content g'
basis for a limv.ed description of the dif- 0.35 k kg CONSUMPTION OF:TABLE CHING
ferent process alternatives from an en- In the alternatives with modified cook-
vironmental standpoint only, not from ing account has been taken of the wash- Chemical consumption:
an economic standpoint. ing effect associated with two displace- Consumption data are based on practi-
Assumptions used for GEMS calcula- ment processes in the digesters, in the cal experience gained at SCA's research
liquor exchange for heat recovery and in laboratory. An adjustment has been
lions: the liquor exchange For extended made with respect to substance carried
Chip solids content: 0.50 kg/kg delignification. over from the unbleached line in this spe-
Alkali charge: 18 y
Alkali concentration: 330 kg'm3 Substanre to "chlorinr stage. For alkali to the oxygen stage,it hasbeen
Sulfidity: 35 'r Na contents (according to SCAN) have
Liquor-to-wood ratio: 3.8 m3/ton been obtained through GEMS calcula- assumed that oxidized white liquor is
tions. used to 100 per cent. The price has been
Pulp yield after cooking: The following relation between COD assumed to be 60 per cent of the price of
At Kappa 32 0.475 kg%kg and Na2SO- has been used for calcula- NaOH. ]n calculating the price, the fol-
lowing factors were taken into account:
At Kappa 25 0.465 kg/kg
lion of COD:
Substance level • Reduction heat in the recovery boiler
Alkali charge in oxygen bleaching: kg Na nee level
04 , kg COD'kg Na2SOa • Oil to the lime kiln
At final Kappa 20 38.0 kg/tloo - -- _ k - • Pump energy
At final Kappa 18 20.0 kg%tlop Unbleached line, without oxygen a Capital cost of equipment.
At final Kappa 15 23.6 kg/tloo bleach1ng 1.5
q 13 RELATIVE PRODUCTION
.
Pulp yield in oxygen COSTS (TABLE 3)
bleaching: 0.97 kg/kg Unbleached line,with oxygen bleaching
7 1.0 This table records the relative cost differ-
Washing filter: 6 0.9 ences between the alternatives.
E-factor: 1.8 Consumption of chemicals, electricity
Discharged solids Bleach plant discharges: p
content: 0.12— The reported discharge values are based and steam, and manpower requirements
0.14 kg/kg on practical experience gained at SCA's with respect to the aerated pond, are
TABLE 1 —PROCESS ALTERNATIVES, BLEACHED SOFTWOOD PULP, 90% ISO
Kappa Substance tochlorinestage Bleach plant discharge
number - kg't,at DF— 2.5 Sr kg lK
Alternative Unbleached 0=bleached Bleach sequence Na,SO+r' COD D COD Color ToCI
32 — (DC)(EO)D 12 18 25 60 130 5.1"
1.BK—F—F—S—F—P+Pond (DC)(EO)DED 85 60 80 2.3
25 (DC)(EO)D 9 12 25 60 100 5.2
2.BK—F—F—S—F—P Modified (DC)(EO)DED 85 60 70 2.5
15 43 70 3.8
3.BK—F—F—S—F—P—Ox—F—P 32 18 (DO(EO)DED 7 7 25 43 65 3.5
85 43 So 1.8
32 ]5 (DC)(EO)D 15 37 60 3.2
4.BK—F—S—F—P—Ox—F—P Modified (IXI(EO)DED 7 7 25 37 55 3.0
85 37 45 1.3
BK— Batch digester;F=washing filter stage;S = screening:P = displacement press stage:Ox — oxygen bleaching
1)According to SCAN.
2)Regards discharge following aerated pond.Reduction in the pond approx.COD ZS�.n.Color OEM,ToC125 m.

Per- Chemical consumption,kgadta,
Kappa number O, Bleaching renege 01 NaOH(o.. CI7 CIO, NaOH Cost
Alternative unbleached bleached stages D white liq.) Art.Cl SEK/L ,
1. BK-F-F-S-F-P+Pond 32 - 5 25 5 - 55.6 31.1 38 199
- 5 85 5 - 11.2 66.7 28 235
- 3 25 5 - 62.6 32.7 33 205
i
- 3 85 5 - 13.1 72.3 23 247
2.BK-F-F-S-F-P 25 Modified - 5 25 5 - 42.7 25.5 31 261
- 5 85 5 - 8.5 52.0 24 290
- 3 25 5 - 47.7 27.3 28 166
- 3 85 5 - 9.5 52.7 21 200
3. BK-F-F-S-F-P-Ox-F-P 32 18 5 15 20 17 28.8 37.2 30 147
18 5 85 20 17 5.7 38.1 23 170
18 3 25 20 17 26.3 224 25 352
18 3 85 20 17 5.7 42.1 20 377
4 BK-F-S-F-P-ox-F-P 32 Modified 15 5 35 24 21 23.0 15.2 27 130
15 5 85 24 21 4.7 30 c 20 151
1C 3 25 24 21 21.0 10.6 22 142
15 3 85 24 21 4 7 33.5 17 357
taken from the SSVL report, ••Environ- ment costs are estimated from budgeted In the alternatives with high percentages
mentally safe manufacture of bleached machinery costs provided by Sunds of D in the chlorine stage(85 per cent), it
pulp_' Defibrator. has been assumed that a filter stage is in-
Wood consumption per tonneofbleached Additional costs were then calculated for cluded before the chlorine stage.
pulp is assumed to be the same in all construction, electricity, instrumenta- Investment costs for chlorine dioxide
alternatives except alternative 2.where it tion and engineering. The amounts preparation are based on budgets for a
is 1.8 per cent higher. therefore represent the total investment couple of plant sizes. Costs for Other
asts have been calculated cost of a new installation.The following capacity levels have been calculated
Maintenance co
M per cent the relative investment total investment costs have been according to the following equation:
estimated:
cost, except for the aerated pond,where 600 t/d 1000 t/d The investment cost, M. SEK = 15 +
1 per cent has been used. M. SEK M. SEK capacity (kton 002/yr.)X 4.
OxH-F-P 60 75 The equation is designed to provide the
total investment cost for an SVP-type
RELATIVE INVESTMENT COSTS F 15 25
(TABLE 4) 2 bleaching stages 45 60 facility.
Additional cost of
Apart from the aerated pond, invest- "modified cooking" 20 30
TABLE 3 - RELATIVE PREDUCTION COSTS, SEK tyo. BLEACHED SOFTWOOD PULP, 90% ISO
Wood Personnel
Bleaching yield Electricity Steam SEK
Bleaching b chemi- SEK Chemicals SEK 0.15' SEK 50 Gco,000.yr. Mainte.
Alternative stages D cals 250'm' pond k11'h tonne post nance Total
1.BY-F-F-S-F-P+Pond 5 25 190 - 10 15 2 2 11 239
5 85 235 - 10 15 2 2 15 279
3 25 205 - 30 30 - 2 6 233
3 85 247 - 10 10 - 2 10 279
2.BK-F-F-S-F-P 5 25 161 23 - 5 2 - 12 203
5 85 190 23 - 5 2 - 15 235
3 25 366 23 - - - - 6 195
3 85 200 23 - - - - 9 232
3.BK-F-F-S-F-P-Ox-F-P 5 15 147 - - 18 5 - 16 18o
5 85 270 - - ]8 5 - 19 212
3 25 152 - - 13 3 - 31 179
3 85 177 - - 13 3 - 14 207
4.BK-F-S-F-P-Ox-F-P 5 15 139 - - 16 5 - 16 176
5 85 151 - - 36 5 - 39 191
3 25 142 - - 11 3 - 11 167
3 85 157 - - 11 3 - 14 185
9

A W N u awy a w N 1
ry m m to m G
0088080 oS B8oa"+6 $ x f x x Rm >. x x x
n M T T it if
• u Z N T T T � N T T T
d l I I I I I
Ln •let N i N ;
N ' Er
ro I I I 0� I I I
rTi
` TI 7 '' •IR s i
m
W wUU w WGU wwGU S
o0
' f ° 7 W wUU W W UU w W UU W W UU A S 0> ml,1PU UN�U mNUN pl, O� r
° a 7 U VV�� Vr UI en VI UI 4
nn m 3
N �n UUen uN,°n°�1 voOi lJn rmn l'In U �N Ora G•n•n•n p. pp. ��JJ��..// yy _
I N$ 1"I�
ra a
(1r ���J� rNau IS"6"6"6' I I I I
unp i
I N
ra
S S � S (n W W a W ayA W aA as n p• 1..1 rn I pU U I Ota (uji I Sa I S� � m
O K
�Y N � mS NIPS ;' I >r15 G' 1875 C
j A O U T tNJ G U 7 N, u G O r n uG 4 r U 7 O ti °41 VIL n ✓ 6 1
MSam
�
o w v -
o.
2 e m GOT•O u O O•O O G V 1 J Y O d 09
C
w 'JOtwnU GOOD UGOO U>r S>r OJ
q I Vr
b UUU•n a1
v_ ro o F'oGorTla w!wa°' o.°yn lOdJ.`e. mmM4: ^p n.
w {� p.
I� i
- u O �1 OOIJ �O J.0 y •n •J J P srOn r)iJ OSG•J COOS O

Quality of
oxygen-bleached pulp
BACKGROUND If bleached pulp with ontimum strength Sweden can be differentiated from those
properties and'or yi6d is desired, de- derived from wood from southern
Our intention has been to investigate adjusted ion in o the the
e oxygen stage must be number Sweden.
whether there is a difference in strength adjusted to the
viseulp following viscosity-Kappa-number
One forest productscompany in Sweden
relation propertiesbetw•een fullybleached sulfate standard delignification rate of 50% in markets under the same name pulp pro-
pulps that are oxygen bleached and those the oxygen stage is a good starting point duced at mills which both do and do not
that are not. if the pulp has a ven good viscosity- apply oxygen bleaching.This is a further
Reference sources, practical experiences Kappa-number relation from the digest- example of the fact that comparable pulp
gained from laboratory and plant in- er.if the pulp shows a poorer relation,a qualities can be manufactured.
vestigations at MoDo and Billerud, and lower rate of delignification is recom-
analyses performed by these companies mended. COMPETITIVE TECHNIQUES
of competitors'pulps have been used in
the comparison. In the production of softwood sulfate
EXPERIENCES pulp,oxygen bleaching is today the only
technically viable method available with
VISCOSITY CAN BE USED AS A Over the years, a large number of lab- which to achieve low lignin contents
MEASURE OF THE STRENGTH oratory investigations have compared (Kappa number <28) entering the
OF THE PULP the qualities of oxygen-bleached and chlorine bleaching plant and at the same
non-oxygen-bleached pulp. In general. time achieve high pulp quality.
Bleached pulps from the same wood lot the conclusions are that the two different
exhibit the same pulp strength at a given bleaching processes produce pulps with
viscosity whether or not an oxygen stage similar strength properties. An example SUMMARY
is included.Examples of this observation is a summary of 23 investigations carried
date back as far as 1976 at STFI W.The out at MoDo's research laboratory (2). There are no significant differences in the
same conclusion is drawn for bleached Delignification varied between 44 and 72 quality of oxygen-bleached and non-
pulps in more recent publications, in- per cent, with an average as high as 55 oxygen bleached sulfate pulps,when the
eluding(2),and is supported by practical per cent. Despite this, no significant dif- pulps are produced under controlled
experience. ferences could be noted between the conditions. This applies to strength and
For purposes of comparison,the viscosity oxygen bleached
leacpulp e an regard oxygen en optical properties, cleanliness and
bleacheof bleached pulp can thus be used as a strength properties. beatability.
measure of the strength of the pulp. For
viscosities greater than 850-9D0 Oxygen' bleaching was installed in
dm3/kg, only a marginal increase in Husum's softwood sulfate line in 1977. REFERENCES
pulp strength is obtained with an in- During the seven months prior to, and
crease in viscosity. eleven months following, the introduc- 1. Alfthan. C—1, 146nu?, ist, W:"Viskositet
tion of oxygen bleaching, the quality at sour matt pa massastyrka,'•(Viscosity as a
both the pulp and the paper mill was measure of pulp strength)STFI
DEGREE OF DELIGNIFICATION IN carefully watched (2). Total testing data B-meddelande or 370(2976).
THE OXYGEN STAGE consisted of approximately 3,700 meas- 2, larnieson, A: "The strength of oxygen
urements. No significant changes could bleached pulps,"speech presented at
The precondition for achieving a pulp of be observed in the strength properties of Eucepa, Helsinki, 19so.
optimum quality is that the pulp shall the pulp or the fine paper.It should also
not be unnecessarily damaged. This be noted that the ash content of the paper
applies to oxygen-bleached as well as was approximately 10 per cent higher
non-oxygen-bleached pulps and, not during the period of oxygen bleaching.
least, to cooking and final bleaching. These results were also confirmed by
This is a well-known fact among pulp Billerud's investigations and practical
manufacturers. experience.
If the pulp is excessively delignified in the Both MoDo's and Billerud's analyses of
oxygen stage, its yield and strength Swedish, and in certain cases Finnish,
properties are jeopardized. "hen this competitive pulps show that it is impos-
happens, the bleached pulp shows a sible,based on such variables as strength
lower viscosity (and strength). and optical properties, cleanliness and
The same phenomenon occurs if the de- beatability, to discern whether a pulp is
lignification is brought too far in the oxygen bleached or not.However,pulps
digester. produced from wood from northern

Environmental aspects of
oxygen bleaching
CURRENT KNOW-HOW able attention in environme;ital contexts greatly. However, a number of ques-
bec:.ase of its elassificatic•n as being car- tions, including questions of the mate-
When oxygen bleaching was first intro- e,iogenic). However, most of the rial's biological and chemical degrada-
duced for industrial application, BOD, organic material in the C and E effluents tion under recipient-like conditions and
COD and color were the most important is high-molecular. It is chlorinated and of the capacity of specific compounds to
has a considerably lower aromatic con- bioaccumulate in organisms, remain to
parameters for characterizing the en-
vironmental effects of bleaching plant tent (<10 ro) than the residua] lignin in be answered in order to estimate the risk
discharge. unbleached sulfate pulp(10, 15). posed by the release of such discharges.
Laboratory investigations have also This applies especially to the behavior of
It was established at an early stage that shown that C- and E-stage discharges bleach plant waste liquor in the re-
the use of oxygen bleaching substantially from conventional bleaching of soft- cipient. Noteworthy in this context are
reduces the chlorine content of the wood sulfate pulp show a mildly acute recent laboratory results showing that
bleach plant effluent. As a result, a toxic effect on fish (6, 11, 121 (96 hours high-molecular material from C-and E-
sizable portion of the effluent can be re- LC 50 concentration = 15-50% V.-V. stage effluentsfrom thebleachingof soft-
cycled into the process, correspondingly rainbow trout). This effect is caused by wood sulfate pulp can be degraded with
reducing the discharge to the recipient.If chlorinated phenols, catechols and bacteria (isolated from receiver sedi-
a change is made from a conventional guaiacols and by various chlorinated men!! and that chlorinated anisoles and
CEHDED to a OCEDED sequence, and non-chlorinated resin and fatty veratroles are then formed which have a
BOD, COD and color discharges are re- acids. C and E discharges at sublethal higher bioaccumulation potential than
duced by between 30 and 70 per cent for concentrations from conventional other compounds from bleach plant
softwood sulfate pulp(1-7).The size of bleaching of softwood sulfate pulp also waste liquor(13).
the reduction depends,among other fac- affect fish. Laboratory tests have dem- In the case of discharges from thebleach-
tors, on the Kappa number of the un- onstrated that such discharges may ing of hardwood sulfate pulp and sulfite
bleached pulp and on the degree of de- affect reproduction, and have behav- pulps. available knowledge of chemical
lignification in the oxygen stage. For ioral, physiological, histological and compositions and biological effects is
hardwood sulfate pulp, the correspond- certain latent effects. The majority of less extensive than for softwood sulfate
ing reductions are a consequence of these can be observed at concentrations pulp,and needs to be improved in future
changing the bleaching sequence from as low as about 5 per cent of the 96 hours studies.
CEDED to OCED (3, 5, 8). LC So value for the discharge (6). The
test for ecloparasites in the gills offish is
probably the most sensitive. The effects EFFECT OF OXYGEN BLEACHING
CONVENTIONAL BLEACH PLANT of C-and E-stage discharges from the ON THE CHEMICAL COMPOSITION
EFFLUENTS: CHEMICAL conventional bleaching of softwood AND BIOLOGICAL EFFECTS OF
COMPOSITION AND BIOLOGICAL sulfate pulp can be detected using this THE.DISCHARGES
EFFECTS test, at concentrations 25 to 30 times
lower than the 96 hours LC 50 concentra- Available knowledge of the effect of oxy-
In recent years, there has been a change tion (6). Today, it is not known which gen bleaching on the chemical composi-
in standards for evaluating bleach plant specific compounds are responsible for tion and biological effects of the dis-
effluents (9). Earlier, BOD and color the effects at the sublethal level. These chargefroma subsequent,reducedtradi-
were the primary variables, but today a are not necessarily the same as those tional "chlorine bleaching"is not so ex-
more comprehensive method of evalua- responsible for acute toxicity. tensive as the knowledge concerning
tion has been developed, based on a Laboratory investigations in recent waste liquors from conventional bleach-
gradually increasing number of chemical years have also determined that C- and ing. As mentioned above, the oxygen
and biological tests. As a result, avail- E-stage effluents from the conventional bleaching does, however, result in a re-
able knowledge of chemical composition bleaching of softwood sulfate pulp(and duction in the amount of residual lignin
and various biological effects,especially other pulp types) contain compounds which is transferred to the chlorination
with respect to discharges from the con- that are mutagenic and or carcinogenic. and subsequent bleaching stages.This is
ventional bleaching (CEHDED) of soft- Anumber of such compounds have been the primary reason for the sizable reduc-
wood sulfate pulps,has increased exten- identified (9, 10). However, the content tions in BOD, COD and color which ac-
sively(6,10,11).To date,approximately levels are so low that the risks associated company oxygen bleaching.New results
250 separate compounds have been iso- with the release of these compounds into have shown that this reduction is ac-
lated in C- and E-stage effluents from the recipients—w•ithtoday's know-how companied by a similar reduction in total
such pulps(10).The compounds consist — are judged to be negligible or non- organic bound chlorine(ToCI) (6, 14).
of chlorinated and non-chlorinated all-
phatic and aromatic acids, chlorinated existent. With reference to the qualitative organ-
phenols, catechols and guaiacols and Available knowledge of the chemical is/chemical composition of C-stage
a number of chlorinated and non- compositions and biological effects of (possibly with some addition of D)and
chlorinated neutral compounds includ- discharges from conventional bleaching E-stage effluents, today's know-how
ing chloroform (which has drawn not- of softwood sulfate pulp has increased suggests that these do not differ to any

_from non-oxygen bleaching. 1f igatedandtheoxygenbleachedpulprnn- point.
�luents
-the organic material in the effluent is di- tained significlintly higher washing loss-
vided into fractions (including material es than the non-oxygen-bleached pulp.
with high molecular weight, water- Logically, this may seem to be the most REFERENCES
soluble material with low molecular important explanation oftheunexpected
weight, ether-soluble material with low result. New investigations of acute tox- 1. Croon, 1.;Andreuu,D.H.Tappi 54(71)
molecular weight - which can be fur- Jami p971).
irity confirm this (9)and indi- 2. Jamieson.A.;Smedman. L.Tappi 56(6)
ther divided into sub-fractions con- Bate that the introduction of oxygen 107(1973).
taining acids, lipophile acids, lipophilic bleaching should not only reduce BOD, 3. Carpenter, W,1.:McKean, W.T.;
phenols and neutrals - and finally a COD, color and ToCI at comparable Berger, H.F. Gellman, L.Tappi 5901)
fraction containing easily volatile mate- washing efficencies, but also reduce the 8] (1976).
rial) it is found that the relativ- sizes of acute toxicity of the discharge. 4. "Morid i btervinningssystem,••
the fractions are largely ind:pendent of (Chloride in recovery systems)final
whether or not oxygen bleaching is used In a recently performed study of re- report, the Swedish Pulp and Paper
cipients outside the Monsteras sulfate Association, Stockholm3977.
This suggests that the substantial reduc- pulp mill, it was concluded that bladder 5. Wong.A.; Pralines. S.
Pulp
and
in the discharge of organic material wrack had disappeared from the area Canada 79R. )41 (rs.8) Pulp and Paper
around the mill's discharge pipe (16). Canada nlig t 11 erknin.
expected from a well-conducted oxygen 6. Alilji+vanlig tilh•erkning a+•bleU mas-
The bladder wrack began to disappear at
bleaching must affect all compounds, or sa,"(Environmentally safe production
at least all classes of compounds,existing the same time as the plant was redesigned of bleached pulp)final report, the
in C- and E-stage effluents. This con- to accommodate among other things, Swedish Pulp and Paper Association.
elusion is supported by the observation oxygen bleaching and bleaching with Stockholm, 1982.
that the amounts of chlorophenolic com- chlorine dioxide in the final bleaching 7. Belt. P.B.:Joyce.T.;Chang H-in.
pounds, among others. are largely the stage. Several conceivable reasons were presentation at the Tappi Annual
same in both types of bleaching se- given, but an important recommenda- Meeting. 1982, Proceedings 391.
tion was that no further oxygen bleach- 8. Jones A.R. Tappi 66W)42(1983).
quences but are lower for the alternative ing be installed until further notice (16). • Annergren, G. presentation at SPCI,
with oxygen bleaching (6). Other sup- However, it has since been resolved that pulp section's autumn meeting. Gavle,
porting evidence includes results show- Sweden, Nov.30,1983.
ing that the chemical compositions of the damage to the bladder wrack cannot be 30. KringStad, K.P.;Lindstrom. K.
materials w•ithhigh molecular w�eightare attributed to any known or unknown Environmental Science Technology,
also of largely the same type (15). The compound formed during oxygen under publication. .
observation that only very small bleaching. The most probable cause of IL Krirrgstad. K.P.;Stockman.. L.G.:
amounts of chloroform (80ve ye reduc- the damage was the presence of a rela- 5trbmberg. L.M.presentation at XXI
Lion) are formed f the bleaching se- tively high chlorate content in the dis- EUCEPA Conference:"Chemical
charge(26). This is a result of the use of processes in pulp and paper technology,"
quences with oxygen bleaching does not chlorine dioxide in the plant and of the Torremolinos, May 1984.
conflict with this conclusion, since this is fact that the plant's while water system is 12. I-alden, CC.;Howard. T.E. Tappi
a result of the fact that the bleaching se- tightly closed.The problem has nothing 600)122(3977).
quence applying oxygen bleaching has 13. Neilsen, A.H.;Allard,A.-S.:
no hypochlorite stage, which is the to do with oxygen bleaching alone. H-vnning. P.A.:Rembrrger,M.;
major source of the chloroform (6). Landner, L. Applied Environmental
Microbiology 45(3)774 (3983).
On the basis of the observation that the SUMMARY If. Lindstrom, K.;Osterberg.F. presenta-
introduction of oxygen bleaching sub- tion at SPCI,pulp sections autumn
stantially reduces the discharge of BOD, Available data show that the introduc- meeting. Gavle, Sweden, Nov.301983.
COD, color and ToCI from the bleach tion of oxygen bleaching before tradi- 15. Lindstrom, K.:Osterberg, F.
plant, and that this reduction affects all tional (and reduced)chlorine bleaching. STFI, Stockholm 1984,private
classes of compounds, it is also very communication.
combined with recovery and combus- 36. Lind:all, B.;Ahn,A. Message 83:5
likely that the biological effects of the tion of spent liquor from the oxygen from Kalmar Institute of Technology.
discharge will be reduced correspond- stage,results in a significant reduction in Institute of Natural Science and Tech-
ingly. Few investigations have been the discharge of BOD, COD, color and nology, Kalmar, Sweden, 1983.
made of the biological effects of the ToCI. The reduction includes chlo- 17. Lehtinerr. K.-J.;Mattson,J.]X7.,
discharges from the bleaching sequences rinated and non-chlorinated organic ma- Karlskrona 1983,private communica-
containing oxygen bleaching. The re- terials of both high and low molecular tion.
sults available so far are inconclusive. weight, including chlorophenolic com-
Within the framework of the project pounds and chloroform. In general, the
"Environmentally safe production of p
introduction of oxygen bleaching there-
bleached pulp,•a comparison was made fore clearly represents a highly signifi-
of the acute toxicity, reproduction dis- cant measure from an environmental
turbances, behavioral changes and standpoint.
various physiological and histological
sublethal changes (fish) and effects on With respect to the influence of oxygen
the planktonic algae in bleach plant bleaching on the bleach plant discharges,
discharges,where softwood sulfate pulp the few investigations carried out so far
was bleached both with and without are inconclusive. This is probably a re-
oxygen bleaching(6).The results of this sult of variations in washing losses. If
comparison showed that oxygen bleach- such process parameters were better
ing did not seem to reduce the effects, controlled, systematic investigations
which were as large as those due to the would probably also establish that the
discharges from the conventional se- introduction of oxygen bleaching is

Authorities' view of
oxygen bleaching
1':TRODUCTION Swedish pulp and paper industry concen- nomical possibilities" to reduce bleach
[rated its environmental protection efforts plant discharges thrr dgh various measures
The following summary is the result of a primarily on the environment in and (5).
stud), of the concession decisions in around the mills. Large resources were During recent years,joint research carried
Sweden during the past 5-10 years for all invested in an attempt to reduce the dis- out by Swedish industry has shown that
planes manufacturing bleached sulfate
charge of oxygenc'onsumi g
ng substances. the rate of deli nification and the transfer
pulp.The intention has been to find possi- The first oxygen stages started operation in of organic substances to bleach plants are
ble trends in the authorities'viewof oxygen the beginning of the 1970s and represented two primary factors affecting conventional
bleaching. an answer to the demands of authorities to parameters and toxicity in bleach plant
reduce the discharge of BOD7, lignin and discharges.During the current decade, the
THE CURRENT STATUS OF INTERNAL color. Often, conditions were formulated Environment Protection Board has also
AND EXTERNAL MEASURES IN so that companies could choose between begun to an increasing extent to insist on a
an oxygen stage and external purification reduction in the Kappa number in cooking
S%NTDISH BLEACHING PLANTS 0) in conventional bleaching and in oxygen
stages,as well as on conditions for washing
Of the fifteen sulfate mills which manufac- During the mid-1970s, the National losses(6).
ture bleached pulp,eight have today oxy- Swedish Environment Protection Board In addition to demands for oxygen stages
gen stages for bleaching sof twood pulp.Of regarded oxygen bleaching in combination yg
these mills, one also bleaches with conventional final bleaching as a and aerated ponds, recent requirements
hardwood pulpl+$ban initial oxygen stage. ..standard requirement"in new mills.Inthe have also been specified for Kappa number
One additional sulfate mill is planning to opinion of the Environment Protection reductions,certain proportions of chlorine
instal an oxygen stage for bleaching soft- Board, oxygen bleaching was a first step dioxide in chlorine stages and improved
wood pulp. towards a closed bleach plant(2).External washing (7). It is interesting to note that
The portion of chlorine dioxide in the chlo- purification methods for bleach plant one mill with an existing aerated pond has
waste liquors and ion-exchange and lime- received permission to instal extended
rine stage for softwood pulp varies be- sludge methods were regarded as an ac- cooking in combination with better wash-
tween a few per cent and 20 per cent.The ceptable alternative to oxygen bleaching Ing and an increase in the level of chlorine
percentage for the bleaching of hardwood for existing installations(3). dioxide, as an alternative to ox.9en
pulp varies between a few per cent and 90 bleaching(8).
per cent. During the latter half of the 1970s,environ-
mental concerns shifted to an increasing In summary, it may be said that there was
All fifteen sulfate mills which manufacture extent towards the long-term effects of dis- a noticeable trend that mills were either or-
bleached pulps have sedimentation ponds. charges in time and space.To an increasing dered or themselves undertook to instal
However, at one mill, all or part of the extent, discussions began to be concerned oxygen stages during the 1970s.Duringthe
bleach plant waste water is taken directly ]with the reduction of the discharge of chlo- latter half of the decade, demands in-
past the pond. Five of the mills currently rinated organic substances.As a result,the creased to encompass oxygen stages in
have aerated ponds. A decision to con- Environment Protection Board's demands combination with additional internal and
struct a biological purification system has were increased to encompass oxygen external measures.
been reached at another. In certain cases, bleaching and measures to reduce these
all the bleach plant wastewater is not led to substances.Among the most important of CONCESSION DECISIONS
the aerated pond. One mill uses external these measures was the exchange of chlori-
purification of bleach plant effluents in the neforchlorinedioxideinthefirstbleachin ;: o--
form of adsorption on lime sludge ("the g 1. �' " )°^`F,1 73
stage as well as the external treatment of s{utsl.ar April 1.1074
lime sludge method")and another uses ion
bleach plant discharges through aerated Korsnas April 1.1974
exchange treatment. ponds, and the ion-exchange and fenox 2. Sl utsl.a: June 10.1977
One mill has both an oxygen stage and an methods. Mill closure measures through Karlsborg December 13.1978
aerated pond. At another mill, with oxy- recycling of portions of the waste liquor 3, Ostrand November 20.1973
gen bleaching,a decision has been made to from the first alkali extraction stage were 4. Varo June 25.3980
also demanded b the Environment C',,,um October October
1980
82
instal biological purification. At the mill }• Gmvbr. October 2.]962
planning to instal an oxygen stage, an Protection Boa/o(4). Giu.6, April 6.1982
aerated pond is already,in use.At another Husum. June 29,1982
mill with an aerated pond,permission has A noticeable trend during this period was 5 M6mte,A, May 2o.1077
in that the National Swedish Franchise Board Mbnstera• April 29.1961
been granted to instal extended cookingcombination with improved washing and for Environment Protection deferred final Ost'.cl January
I.S.30.1980
an increase in the use of chlorine dioxide in demands and requirements with respect to Varo June 25.19&
the chlorine stage, as an alternative to bleach plant discharges. At the time, the hlorrum October 28 1980
Franchise Board for Environment Pro- Nomundet October 7.1961
introducing oxygen bleaching. tection derided to await the results of the 6. (strand April 20,1980
' rs Ocober2 9
SwedishPul and Paper Nor .ndet October 7,3081
TRENDS IN OXYGEN BLEACHING ongoing project, "Environmentally safe Husum June 29.1962
INSWEDEN manufacture of bleached pulp:'During the 7. Husum June 20.3982
period of deferment, mills were usually S. Karlsborg December 14,1982
During the 1960s and in the early 1970s,the ordered to investigate"technical and eco- KadsborF January 12.1984

I 'UIUl C.1JUM1 LLIV 1 V1
oxygen technology
DELIGNIFICATIONK7THAN sp_nt liquor to the closed liquid and re- its aggressive character,which reveals itself
ALKALINE OXYGEN STAGE covery system of the pulping process. in difficulties in controlling pulp quality
Pretreatment with green liquor, peroxide, !:his applies to the process with high
ozone or nitric oxides in acid media gives concentration), and its cost, which, as a
A basic precondition for the use of an positive results.The best results for sulfate result of the low energy and conversion
alkaline oxygen stage in a delignification pulps are achieved with nitric oxides. In yield, has remained at a high level.Ozone
operation is that the development of the Sweden, Professor 011e Samuelson, to- production technology has developed
selectivity for the delignification is better gether with MoDo, has carefully studied gradually; important progress has been
than in the final phase of the sulfate cook- the treatment of unbleached sulfate pulp achieved in recent years.It can thereforebe
ing. The viscosity and Kappa number of with nitric oxides prior to an oxygen stage. expected that an introduction of ozone in
the cooked pulp are determining factors for certain positions will be economically
the technical and economic results of the Amazingly good results have been feasible within the near future. Low con-
process in the oxygen stage.A high intrin- reported in terms of pulp yield and espe- centration methods(around 3 per cent pulp
sic viscosity(<1200 dm'/kg)for a Kappa cially for pulp viscosity after the oxygen consistency)appear to be easier to control
number around 30.with small variations, stage. It appears that pulp quality can be with regard to the viscosity of the pulp.
is a precondition for a good result.The de- maintained at a high level down to Kappa Ozone may pose an attractive future alter-
velopment of cooking technology greatly numbers of about 30. It also appears pos- native as a complement to oxygen bleach-
furthers the possibilities of improving eco- sible to increase the degree of delignifica- ing. primarily as an activating pre-stage.
nomic and environmental effects by using tion to 75 per cent from the current average
an oxygen stage as the first stage following of 50 per cent.
cooking. The introduction of high so]- These results have been so promising that a CLOSED SYSTEMS
fidity,atmospheric steaming and computer group of Swedish companies — MoDo,
control has, generally speaking, increased Sunds Defibrator, AGA and Kema Nobel Thebasic idea underlying theincreased use
the possibility of being able to deliver a — have decided to finance a pilot plant of non-chlorine, non-corrosive chemicals
high-viscosity pulp with stable Kappa project to facilitate the development of in the bleaching plants is to recycle waste
number from the digester to the oxygen equipment and to evaluate the process liquors to the recovery cycle. It is then
stage. from the technical and economic perspec- possible to burn the organic matter and re-
Modified cooking processes have been de- tive.A thorough analysis and characteriza- cover sodium.
veloped and tested on a large scale. It is tion of the effluent from the bleach plant c A Swedish pulp mill is currently recycling
possible to cook to around Kappa 25 while will also be conducted as well as of the the liquor from the stage to the maintaining viscosity.The total yield will health hazards in the environment within re-
covery cycle. By using oxidized white
be somewhat lower and the consistent the mill. liquor in both the O and OE stages, and
quality of the pulp has not as yet been using the flexibility of the chlorine dioxide
ensured on a full scale. However, the production, the plant has succeeded in in-
know-how gained can contribute to the HYDROSTATIC OXYGEN STAGES troducing an economical system with]ow
production of a pulp with Kappa numbers es.
of 28-3owith consistently better viscosity. Oxygen-reinforced alkali extraction stages discharges.
Therefore it is very likely that the current have been shown to be both an effective A bleach plant for a brightness of 90%ISO
cooking-oxygen process can be further de- and economical complement in the bleach- with two pressurized oxygen stages and
veloped using improved washing methods ing sequence, especially as the first alkali satisfactory washing systems may be
now under development so that Kappa extraction stage. By raising the height of designed with only chlorine dioxide as a
numbers of 12-19 can be achieved with- the tower,the effectivity of the delignifica- bleaching agent. The bleaching plant is
out adversely affecting the quality of the tion process can be increased.This variant closed to a high degree with a water con-
pulp. may also be introduced as a first stage in sumption of only 5 m3/ton of pulp. A
This means(contrary to what has been re- special cases such as with hardwood sulfate bleed of 0.5-1 m3/ton is assumed. This
ported in certain contexts) that the im- pulp, sulfite pulp, or other pulps where a waste liquor may be treated before it is
proved cooking process should not be lower delignification(around 35 per cent) released to the recipient. If the brightness
regarded as an alternative to oxygen is acceptable. target is sufficient for the pulp's end use,
bleaching, but as a complement, which A sequence of the (EO)(CD)(EO)D type application of a hydrostatic extraction
improves the positive effects of the oxygen may be an attractive alternative. stage may be a viable alternative to the
stage on the production economy and the pressurized oxygen stage.
environment. It has also been demonstrated that the
replacement of the hypostage with an
In recent years, it has been demonstrated alkali oxygen stage is favorable for both
that it is possible to activate the unbleached birch and pine sulfate pulp(for example,a
pulp using certain pretreatments so that the (CD)E(EO)DED sequence).
selectivity of a subsequent oxygen stage is
greatly improved. Pretreatment with
chlorine prior to theoxygen stage consider- OZONE
ably improves the selectivity of the oxygen
stage. Because of corrosion problems, The introduction of ozone into the bleach-
however,it is hardly possible to recycle the ing process has so farbeen handicapped by
]5

• • OFTIM • '
-.•� ! 1��[ may.. �� .' _ -+_
\\��`\\v`
a.\\�\!•�=:a L\t`\\\\�� �U/1H11�1��!%i�IIIIIi i - x
+'.-• a .. .oaf's .���I M' A . -
• Y..Y _ YC a -
--• •� Ya Pti t e T ,.� ix
` � ke
r • r •r • •• r • r r r �• • r • ••/• r • i /
•r • •r •r rr• r • 'r r
• r r r r r r
•r • rr • • •r • r

Some studies performed in the recipients outside pulp mills using the oxygen bleaching
technique showed environmental effects which, without absolute sciratific evidence,
were attributed to oxygen bleaching. Doubts about beneficial effects of oxygen tech-
nology were published in both trade journals and the Swedish daily press during 1983.
The positive environmental effects of oxygen technology, and its technical and
economical soundness were questioned.
To well-informed observers, the debate developed into an extremely misleading and
erroneous picture of oxygen technology's role as a cornerstone of Swedish process
development. It was therefore decided that a study, entitled "Current status of oxygen
technology and its future potential,"should be carried out under the supervision of
Ingemar Croon, CroonConsult, and Alf de Ruvo, Sunds Defibrator.
The study was conducted between August 1, 1983 and May 1, 1984 and is summarized
in this report in the following articles.
Summary Page 4
Ingemar Croon, CroonConsult, and
Alf de Ruvo, Sunds Defibrator AB
1. Introduction, 5
2. The complex chemistry of oxygen delignification 6
Sten Lfunggren, STFI
3. Bleaching systems and their economic aspects 7
Gunn&Tlirnvik, SCA Teknik AB
4. Quality of oxygen-bleached pulp 11
Greta Fossum, MoDoCell AB, and
Brita Swan, Billerud AB
5. Environmental aspects of oxygen bleaching 12
Knut Kringstad, STFI, and
Lars Stromberg, STFI
6. Authorities'view of oxygen bleaching 14
Christina Molde. IPK, and
Inger Strbmdahl, IPK
7. Future potential of oxygen technology 15
Ingemar Croon. CroonConsult, and
Alf de Ruvo, Sunds Defibrator AB

: Summary
It is impossible to give a clear-cut recommendation of any In general, the introduction of oxygen bleaching therefore
single process alternative to achieve optimum total economy in clearly rer.esents a highly significant environmental pro-
the manufacture of bleached sulfate pulp. Today it is often a tective r.,easure.
matter of expanding or mcidemizing existing plants where The few investigations carried out to date regarding the in-
ever,even in hss and plant space is a major consideration.How- fluence of oxygen bleaching on the discharge from the plant are
ever, even in the case of new investments, in addition to plant inconclusive. This is most probably due to variations in wash-
space and localization, such parameters as environmental ing losses. If this process were better controlled, systematic
demands and prices of timber, utilities, chemicals and capital investigations should here also shoes• that the introduction of
also play a role. oxygen bleaching is favorable from an environmental stand-
The technical and economical evaluation showed that both point.
oxygen bleaching alternatives, O(CD)(EO)D and O(CD)(EO)DED, as well as alternatives with modified cook- The development of the cooking technology has a considerable
influence economical and
ing to Kappa 25,are of relatively equal value from an economy- eviuence cal effects of the bleaching process when on the possibility of improving the econom an oxygen
cal standpoint when low levels of chlorine dioxide are used in stage is used as the first stage after cooking.The introduction of
the chlorine stage. As the amount of chlorine dioxide is in- high sulfidity, atmospheric cooking and computer control has,
creased, the alternative with modified cooking and oxygen bleaching tends to be the most economical. The conventional generally speaking, increased the potential for delivering a
method with aerated ponds is definitely less economical. pester to the oxygen stage.
Three-stage bleaching appears economically advantageous
over a five-stage process. Higher chlorate prices strengthen During recent years it has been demonstrated that it is possible
economic potential of the oxygen sequence. to activate the unbleached pulp using certain pretreatments
Thus, the comparison shows that the so-called modified (for example, NO-NO2, 03 or green liquor) to improve the
cooking process should not be regarded as a viable altemative selectivity of a subsequent oxygen stage.
to oxygen bleaching (regardless of whether it is carried out to Oxygen-reinforced alkali stages have been shown to be both an
Kappa 25 or 30), but rather as a complement. effective and economical complements in the bleaching se-
A careful comparison shows that there are no significant diE- quence, especially as a first alkali stage.This type of treatment
ferences in quality between oxygen-bleached and non-oxygen- may also be introduced as a first stage in special cases.
bleached sulfate pulps, assuming that the pulps are produced One of the fundamental ideas underlying the increased use of
under controlled conditions. This finding applies to strength non-chlorine,non-corrosive chemicals in the bleach plant is to
and optical piceperFies cleanliness and beatability. feed the waste liquor to the recovery cycle, so that the organic
Available know-how has demonstrates tnat the introduction matter can be burned and sodium recovered.
of oxygen bleaching before a traditional (or reduced)chlorine A Swedish mill is currently recycling waste liquor from the OE
bleaching, combined with the recovery and burning of spent stage. By using oxidized white liquor in both the 0 and OE
liquor fro,,i the oxygen stage, results in a significant reduction %tages and utilizing the flexibility of the chlorine dioxide gro-
in the discharge of BOD, COD and color and ToCI. The re- duction, the plant has succeeded in establishing an economical
duction includes chlorinated and non-chlorinated organic system with low discharges.
matter of both high and low molecular weight, including
elorophenolic compounds and chloroform.

Introduction
The sulfate process produces an un- consistency technology is progressing.It favor oxygen treatment as an alternative
bleached pulp with a relatively high re- is likely that oxygen stages at medium process (Fig. 2). The short sequence,
sidual lignin content.The lignin is highly- consistency will be.ome more common O(CD)(EO)D, is particularly interest-
cross-linked. Therefore, it is -fairly in the future. ing in this context.
complicated and expensive to bleach When oxygen delignification was intro- FIGURE 1.
sulfate pulp. duced, it was discovered that current
The usual bleaching chemicals(chlorine, cooking technology was not satisfac- MARKET PRICES FOR
alkali and chlorine dioxide) have in- tory. In many mills, the Kappa number BLEACHING CHEMICALS, SEK/tonne
creased greatly in price in recent years varied from 25 to 40. Today, computer
and will presumably continue to do so in control, atmospheric presteaming and sEK/t NaC10, -zn
the future(Figure 1).Oxygen is an inex- chip equalizing systems make it possible `
pensive chemical and its energy require- to maintain the Kappa number within 3000
ments during production are signifi- narrower limits(28-32). The economic
cantly less than those of chlorine, alkali result depends to a great extent on 2500
and chlorine dioxide. whether the unbleached pulp can be pro-
The National Swedish Environment duced at a high final viscosity (greater 2000
than 1200) with less variations in the c(,
Protection Board's long-term aim for the Kappa number.The industry has a great 1500
control of the discharge of bleach plant deal to learn in this respect. If cooking is
waste water was formulated in 1977, as carried out successfully, a Kappa num- 1000
follows: ber of about 14 can be achieved follow- o,
— Chlorinated organic compounds ing the oxygen stage without any soo
shall not be discharged. weakening of the bleached pulp. De-
- Waste water shall neither give rise to pending on the variations in wood raw
genetic effects nor contain accumu- material and cooking, many plants to- 1976 76 so 62 84 86 year
lating substances. day achieve Kappa numbers only in the
— Waste water shall not result in acute 18-20 range.
toxic effects on fish or other aquatic FIGURE2.
organisms. Economically this is unsatisfactory since RELATIVE BLEACHING COSTS.
g with toda}'s technology the objective
— Theamountsofligninmatterinwaste should be a Kappa number of about15. USS/acit
water shall be reduced.
It is evident that future developments in Relative bleaching costs.USS/adt
The National Swedish Environment oxygen technology will be guided pri- +6
Protection Board's demands were for- marily by economical' considerations. +6 (D.,C,)(EO)DED
mulated in the light of an existing effort We can expect more technologies adapt- ae-3ted pond
to promote oxygen bleaching and El able to the needs of individual mills.
discharge cleaning, in an attempt to Pressurized or non-pressurized oxygen +0
achieve a color reduction equivalent to stages, high consistency (greater than
that resulting from the ion exchange D,ocx,)(Eo)DeD
method. The ion exchange method has, 20%) or medium consistency (8-20%)
however, since fallen into disuse. in the first stage or in the alkali extraction cW c, )EDeD
stage already exist.
Pressurized oxygen stages were intro- _3 O(CmD,o)EODED
duced more than ten years ago on the
basis of the possibility of recycling waste REINFORCED ALKALI STAGES (EO)
liquor from this stage to the recovery —6
system. The degree of delignification Non-pressurized oxygen stages can O(C650,5)(EO)D
achieved is determined by the level of easily be introduced in existing bleaching
improved selectivity which can be at- plants by utilizing the alkali and hypo- 350 375 400 426 450 475
tained, while maintaining the full bleaching towers.This technology is de- Prices for sodium chiorate. Us snonne
strength of the pulp. veloping rapidly. At present there are
approximately 40 installations, but new
It is important that dwell-functioning i
continually being introduced.
washer is installed after the oxygen ones are
stage. Today, there are more than 30
installations of this type in the world,the SHORT SEQUENCE
majority of which operate at high pulp
consistency. If chlorine dioxide further increases in
In general it may be said that medium price, economic:advantages alone will

chemistry o
The complex f
oxygen dengnif ication
Today,some knowledge is available con- principle reactions v.hich occur with lig- In a recently initiated project, STFI
cerning a number of different reaction nin rriginating from the initially formed intends to clarify the molecular kinetics
mechanisms between different lignin p',enoxy radicals.The two types of reac- of these reactions,primarily with respect
structures and oxygen. Knowledge now tions possible are indicated as a.and b.in to such typical structures in residual lig-
needs. to be acquired concerning the the figure. In a., a reaction takes place nins as vinyl ether, and stilbene and
molecular kinetics of the reactions — between aromatic structures through cross-linked structures. Comparative.
how reactive different types of residual radical coupling(i.e., two phenoxy rad- studies of reactivity and the speed of deg-
lignin structures are towards oxygen. icals condense and form biphenolic radation in relation to such factors as
Process kinetics investigations earlier structures). This type of reaction in- pH, temperature and oxygen pressure
carried out at STFI have shown that volves a risk of additional cross-linking are also under investigation. With the
during oxygen bleaching, dissolution of in the lignin, making degradation more help of these data,it is hoped to be able to
lignin (measured as a decrease in Kappa difficult. However, it has been shown explain what happens with the lignin
number) has a rapid initial phase fol- that these structures can also be degra- during the two different bleaching
lowed by a slower final delignification. dated to acids, which perhaps compen- phases indicated by the kinetics of the
One current research project has sought sates for this risk. process.
to collect kinetic data at a, molecular Reaction b. is probably referable for
level,from which it is hoped to obtain an P y p
explanation of these two lignin dissolu. lignin degradation and dissolution. This REACTION OF PHENOXY
reaction is dominant under alkaline con- RADICALS
tion phases. ditions (pH>13). Cvclohexadienone
The chemistry involved in oxygen hydroperoxide intermediates are formed
bleaching is very complex for two rea- through the reaction of phenoxy radicals O O r a""a'
sons.One is that the structure of residual
with hydrophenoxy radicalsa Meo oMe °(HOO') or o_ 0-
lignin to be bleached is relatively varied superoxide anions (O•z) formed from R
and to some extent unknown.The other the oxygen. Depending on the position O oMe
reason is that.oxygencan appear and be in the aromatic ring in which the hydro- ° y oo"
converted to a number of different re- peroxide group is situated, the li nin can c I I C
active oxidizing species which undergo be degraded in three different ways. Side o oMe o oMe
different'types of reactions. In general, chain elimination means that the lignin l { +
there are three fundamentally different structure is broken up between the aro- ,l a;aapha;n. zt ring. 3)aemethy
types of reaction processes-between lig- matic hydrocarbons, forming various erminauon opening canon
nin and oxygen in an alkaline medium. aldehydes and organic acids from the acids,Co,
One is the alkaline hydrolysis reactions side chain as well as paraquinones from
which normally also occur during alka- the aromatic part. Paraquinones are During oxygen bleaching fmp:rtant deg-adabir
line cooking processes.Here,the lignin is colored structures which can be trans- hydropci-micirirttrrmrdia:r>are fornted:tare-
split with the help of hydroxide ions and action b. by reartion bola r the oxygen:and
formed into colorless structures with the rile phenoxv radicals formed from dte hilm1 .
becomes more water soluble through the help of the small amounts of hydrogen The phrnoxy radicals can. hose-:er. also react
formation of phenolia groups.The pres- peroxide formed during the bleaching. with each oils,in a comprtitive reaction :is
enee of oxygen also initiates both ionic reaction a. See text.
and radical reactions.Ionic oxidation re-
actions occur from ionized species, for
example, hydrogen peroxide anions, LIGNIN IS DISSOLVED MORE
which may form in small quantities EASILY
during the reaction of oxygen in an alka-
line medium and play a role.in the re- The ring-opening reaction makes the lig-
moval of color-forming structures such nun more hydrophilic and more easily
as quinones. The formation of radical dissolved.This is due to the fact that the
species is important for the oxygen's break-up of the aromatic ring produces
initial reactions because, depending on water soluble dicarboxylic acids,such as
the pH level, such radicals-may-give rise muconic acid or, succinic acid, i al. The
to chain reactions originating from phe- third reaction is demethylation, arising
nolic structures in the lignin. from intermediates where the hydro-
peroxy group has become attached close
to the methoxy group. The methoxy
ADDITIONAL CROSS-LINKING group is then eliminated and methanol is
MAKES DEGRADATION MORE formed at the same time as the ring opens
DIFFICULT in certain cases.This has less affect on the
degradation of lignin, but may increase
The figure on this page summarizes the its solubility.

Bleaching systems and
their economic aspects
It is impossiblr to make a clear-cut rec- sistency gives a lower investment cost A standard allowance has then been
ommendatian of any one prcce-,alter- but higher steam requirements.The cost made to cover all additional costs.
native for achieving optimal rat econ- comparison is not affected by the choice. Investment costs for CIOz productio,
omy in the production of bleached sul- An evaluation has been made for all include the cost of an SVP (R3)Facility.
fate pulp. Conditions vary significantly alternatives with both high and low Those cases which involve a high chlo-
from case to case. Today, it is often a chlorine dioxide levels in the prebleach- rine dioxide level in the chlorine stage are
question of expanding or modernizing ing stage. assumed to demand an extra filter stage
existing plants where process and plant Comparisons have been made on the before the chlorine stage. Other process
space is a major consideration. Ho", assumption that batch cooking is used. solutions are conceivable.
ever, even where new investments are No significant change in the differences The difference in investment costs for
involved, such parameters as environ- between the alternatives should arise if different capacity requirements in the
mental demands and prices of wood, a continuous cooking process were liquor recovery cycle for varying
utilities, chemicals and capital also play applied. amounts of transferred solids has not
a role in addition to plant space and loca-
tion. The choice does not become any The transfer of substance to the chlorine been taken into account. To simplify
easier in the light of current develop- stage has been estimated using acorn- matters,it has been assumed that the cost
menu, including such new processes as puler program (GEMS). In cases where differences are balanced by energy
modified cooking, which may prove modified cooking is applied,the washing revenues from the solids.
themselves worthy of competing for a effect due to digester displacement has Investment costs associated with chlo-
place in the production chain. been taken into account. rine dioxide preparation are based on
One of the questions which then arises Table 2 shows estimated consumption tenders for a couple of plants of different
concerns the role of oxygen bleaching in -levels of bleaching chemicals for the sizes (SVP (R3) type). Costs have been
this context.' In order to illustrate the various alternatives. In all cases, the calculated for other capacity levels.
economic aspects of this process, four consumption levels are measured at final Table 5 presents a picture of the different
differert process alternatives based on bleaching with both very high and total economies of the various alterna-
the production of bleached softwood limited chlorine dioxide levels in the tives when the production and capital
pulp with similar market demands have chlorine stage. costs are weighed together. Capital costs
been selected and their production and Estimated chemical costs were based on are estimated on the basis of a 15-year
capital costs compared, under condi- the following prices(SEKi tonne): annuity at an interest rate of 15 per cent.
tions described in "Assumptions used in The differences are illustrated in block
comparative calculations;' on page 8. C12: 1,085 diagrams for the two capacity levels.
C102: 2,200 (active C11;
Table 1 shows the process alternatives excl. capital costs) The comparison shows that both alter-
and gives a brief description of their NaOH: 1,280 natives with oxygen bleaching as well as
discharge characteristics. In all alterna- 0 . 850 the alternatives with modified cooking
tives, 3-stage, (DC)(EO)D, and 5- Ox. white liquor: 770 to Kappa 25 are more or less equivalent
stage, (DC)(EO)DED, processes have from an economic standpoint when low
been selected for final bleaching. Chemical charges to the digesters are levels of chlorine dioxide are used in the
assumed to be the same in all cases. chlorine stage.As the amount of chlorine
Two of the alternatives include oxygen dioxide increases, the alternative with
bleaching. The others include a conven- Table 3 shows the other differences in modified cooking and oxygen bleaching
tional line with cooking to Kappa 32 production costs. Wood consumption tends to be the most economical for the
supplemented by external waste water has been judged to be 1.8 per cent higher higher capacity. The conventional
treatment in an aerated pond and a in the alternative with modified cooking method using an aerated pond is con-
modified cooking line with extended to Kappa 25 than in the other alterna- siderably more costly. Three-stage
delignification to Kappa 25 without tives. In alternative 1 it is assumed that bleaching appears to be economically
additional discharge reduction. While external waste water treatment requires more favorable than a five-stage process.
the selection of Kappa 25 seems ex- an extra half man per shift, while the Thus, an essential conclusion drawn
tremely low, it may prove to be an other alternatives require approximately acceptable level in the long term. the same manpower. Maintenance costs modified cooking process should not from the comparison is that the so-called
are estimated to be 3 per cent of the rela-
be
The two alternatives applying oxygen tive investment costs. seen as an alternative to oxygen bleach-
bleaching include a conventional line ing, but rather as a complement.
with cooking to Kappa 32 and oxygen The differences in investment costs have
bleaching to Kappa 18, and a line with been calculated for two capacity levels,
modified cooking to Kappa 32 and 600 and 1,000 tonnes bleached pulp per
oxygen bleaching to Kappa 15. Oxygen day (Table 4). Investment costs refer to
bleaching is carried out at a high pulp new installations and are based on ten-
consistency. Bleaching at medium con- ders for machines and other equipment.

ASSUMPTIONS USED IN COMPARATIVE
CALCULATIONS FOR THE TABLES
Washing press: research laboratory and are, thus, not
PROCESS ALTERNATIVE, (TABLE 1) Dewatering efficiency 0.50 linked to any special calculations for this
Po
The assurr�)tions listed below and used Discharged solids . 0.30—
in the calculatio is have provided the content 0.35 kg'kg CONSUMPTION OF"BLEACHING
basis for a lirr '.ed description of the dif-
ferent process alternatives from an en- In the alternatives with modified cook-
vironmental standpoint only, not from ing account has been taken of the wash- Chemical consumption:
an economic standpoint. ing effect associated with two displace- Consumption data are based on practi-
ment processes in the digesters, in the cal experience gained at SCA's research
Assumptions used for GEMS calcula-
liquor exchange for heat recovery and in laboratory. An adjustment has been
lions: the liquor exchange for extended made with respect to substance carried
Chip solids content: 0.50 kg/kg delignification. Over from the unbleached line in this spe-
Alkali charge: 18 %
Alkali concentration: 110 kg/m3 Substance to "chlorine stage; cial case.
Sulfidity: 35 % Na contents (according to SCAN) have For alkali to the oxygen stage,it has been
Liquor-to-wood ratio: 3.8 m3/ton been obtained through GEMS calcula- assumed that oxidized white liquor is
. tions. used to 100 per cent. The price has been
Pulp yield after cooking: The following relation between COD assumed to be 60 per cent of the price of
0.475 kg/kg NaOH. In calculating the price, the fol-
At Kappa 32 kg/kg and Na2SO4 has been used for calcula- lowing factors were taken into account:
0.4 5
At Kappa 25 g g tion of COD:
Substance level • Reduction heat in the recovery boiler
Alkali charge in oxygen bleaching: kg Na nee level
kg COD'kg Na,SO4 • Oil to the lime kiln
At final Kappa 20 '18.0 kg/t100 — --— • Pump energy
At final Kappa 18 20.0 kg%1100 Unbleached line, without oxygen . • Capital cost of equipment.
At final Kappa 15 23.6 kg/tlp0 bleaching
12 1'S RELATIVE PRODUCTION
q 1.3 COSTS (TABLE 3)
Pulp yield in oxygen k 0.97 /k oxygen bleaching: g g Unbleached line,with ox en bleaching
7 1.0 This table records the relative cost differ-
Washing filter: 6 0.9 ences between the alternatives.
E-factor: 1.8 Consumption of chemicals, electricity
Discharged solids Bleach plant discharges: P
content: 0.12— The reported discharge values are based and steam, and manpower requirements
0.14 kg/kg on practical experience gained at SC.A's with respect to the aerated pond, are
TABLE 1 —PROCESS ALTERNATIVES, BLEACHED SOFTWOOD PULP, 90% ISO
Kappa Substance to chlorine stage Bleach plant discharge
number kg4w at DF =2.5 kg toe
Alternative Unbleached OZ bleached Bleach sequence Na,SO t' COD D COD Color ToCI
(DC)(EO)D 25 60 130 5.111
1. BK—F—F—S—F—P+Pond 32 (DC)(EO)DED 12 18 B5 6p 80 2.3
25 — (DC)(EO)D 32 25 60 100 5.2
2. BK—F—F—S—F—P- Modified (DU(EO)DED 9 85 60 70 2.5
15 43 70 3.8
3.BK—F—F—S—F—P—Ox—F—P 32 18 (DC)(O)DED 7 7 25 43 65 3.5
85 43 50 1.8
15 37 60 3.2
4. BK—F—S—F—P—Ox—F—P 32 Modified 15 (DC)(ODED 7 7 25 37 55 3.0
85 37 45 1.3
BK m Batch digester;F— washing filter stage;S—screening;P=displacement press stage;Ox = oxygen bleaching
1)According to SCAN.
2)Regards discharge following aerated pond.Reduction in the pond approx.COD 25%,Color 0%,ToC 25°k.

Per- Chemical consumption.kg adtao
Kappa number O, Bleaching centage O, NaOH(ox. Cl, CIO, NaOH Cost
Alternative unbleached bleached stages D white licl.) Act.Cl SEK/t90
1. BK-F-F-S-F-P+Pond 32 - 5 25 5 - 55.6 31.1 38. 399
- 5 85 5 - 11.1 66.7 28 235
- 3 25 5 - 62.6 32.7 33 205
- 3 85 5 - 13.1 72.3 23 247
r
2. BK-F-F-S-F-P 25 Modified - 5 25 5 - 42.7 25.5 31 161
- 5 85 5 - 8.5 52.0 24 190
- 3 25 5 - 47.7 27.1 28 166
- 3 85 5 - 9.5 57.7 21 200
3. BK-F-F-S-F-P-Ox-F-P 32 18 5 15 20 17 28.8 17.2 30 147
18 5 85 20 17 5.7 38.1 23 170
18 3 25 20 17 26.3 22.4 25 152
18 3 85 20 17 5.7 42.1 20 177
4. BK-F-S-F-P-Ox-F-P 32 Modified 15 5 15 24 21 23.0 15.2 27 130
15 5 85 24 21 4.7 30.c 20 151
15 3 25 24 21 21.9 10.6 22 342
15 3 85 24 21 4.7 33.8 17 157
taken from the SSVL report, "Environ- ment costs are estimated from budgeted In the alternatives with high percentages
mentally safe manufacture of bleached machinery costs provided by Sunds of D in the chlorine stage(85 per cent), it
Pulp:' Defibrator. has been assumed that a filter stage is in-
Wood consumption per tonne of bleached Additional costs were then calculated for cluded before the chlorine stage.
pulp is assumed to be the same in all construction, electricity, instrumenta- Investment costs for chlorine dioxide
alternatives except alternative 2,where it tion and engineering. The amounts preparation are based on budgets for a
is 1.8 per cent higher. therefore represent the total investment couple of plant sizes. Costs for other
Maintenance costs have been calculated cost of a new installation.The following capacity levels have been calculated
as 3 per cent the relative investment total investment costs have been according to the following equation:
estimated:
cost, except for the aerated pond,where The investment cost, M. SEK = 15 +
1 per cent has been used. M.
t/d M.ioo SEK
capacity (kton C]02/yr.) X 4.
P M. SEK M. SEK
OxH-F-P 60 75 The equation is designed to provide the
total investment cost for an SVP-type
RELATIVE INVESTMENT COSTS F 15 25
(TABLE 4) 2 bleaching stages 45 60 facility.
Additional cost of
Apart from the aerated pond, invest- "modified cooking" 20 30
TABLE - RELATIVE PREDUCTION COSTS, SEK t90. BLEACHED SOFTWOOD PULP, 90%o ISO
-- - lh'ood Personnel
Bleaching yield Electricity Steam SEK
Bleaching m chemi- SEK Chemicals SEK 0.15: SEK SO, 900,000•yr. Mainte-
Alternative stages D cals 250lm' pond kWh tonne post nance Total
1. BK-F-F-S-F-P+Pond 5 25 199 - 30 15 2 2 11 239
5 85 235 - 10 15 2 2 15 279
3 25 205 - 10 10 - 2 6 233
3 85 247 - 30 30 - 2 10 279
2. BK-F-F-S-F-P 5 25 161 23 - 5 2 - 12 203
5 85 190 23 - 5 2 - 15 235
3 25 166 23 - - - - 6 195
3 85 200 23 - - - - 9 232
3. BK-F-F-S-F-P-Ox-F-P 5 15 147 - - 18 5 - 16 186
5 85 170 - - 18 5 - 19 212
3 25 152 - _- 13 3 - 11 179 '
3 85 177 - - 13 3 - 14 207
4. BK-F-S-F-P-Ox-F-P 5 15 139 - - 16 5 - 16 176
5 85 151 - - 16 5 - 39 191
3 Z5 142 - - 11 3 - 11 167
3 85 157 - - 11 3 - 14 185
9

T U T T T M U T YI 'lll T
g m w III to vl to Cn •1,1 to to to
o tm' a t0 ro �n •I.1 YI m I I I I 711
� � •I I I I m
p y}
t»
v CO m
'� O � WWNN wWNN WWUN WwNU mS S
n p (n r n O $
EI W � O Q WWNU WWNN WWNN WWNN qd � g1Jmum1Jmu mNm�J m�Jm1� n
a o � = OD NInNN NtnNN NNN In 411nNril od9 O
N wr ���°,G Ninnn �I4"YI°In 4m1 IN.INIn' Oo yY viwq� PG$$$ o00o I I I I �m
A rTh'"� .
On oo�o HN�n oo�t6' 1 1 1 1 Z5n
vl0 a w
g o N
. � � �_ m •p N o v w N.o �.+•p w •o•4•o o N 1n � n u^ x
K En
v m � m !. m 5�
g m 11' 'n m •` ' ry UI' I •°JI°S S I (,1 S v I N S CI 1 tml S 25,°° O
O' N a •o M m{J mina.N. .u,m o Ut En
m w u o �• 90 vlrnpp. J
O $N u A t d p4 t4� ONOO O OTi 2
A O
VV I i i o
mOP.O u4Ui r10 O.O O W V NLp d 55��d � �
I _ Doan 4"Ooo tNll tWiloo tell tn$u 255N 'J
w
0
� J m � n `J tan tail tail n v O
d N Q �nNPP V P V P qPm V •Oq OppY n O• d
�
O d O NlP/IUN 06
u aNNEWr I O
1 �
N'-' W1� 1JuylJ nuWtJ uNN NU W;J _
N OJ NO.MN • qY V Ouqu 1\IO 1 n�
N Vq I d 0077 tOi�tgn0 VNi rn tail tno toilH ntn I EEL
w
a
N u _
�" •O� 0•4. •Om00 O V� O N uu p a.+pp u pp. ^' p_N pu. 5Qj
�"'O VumtJ G•p O1J NOYu b NO^�n rtOnpp OUIPnO N000 x
a•

Quality of
oxygen-bleached pulp
BACKGROUND If bleached pulp with optimum strength Sweden.can be differentiated from those
properties and'or yi%�d is desired, de derived from wood from southern
Our intention has been to investigate tdtion in the
ixygeKappa-nu n stage must
be Sweden.
adjusted to
whether there is a difference in strength cooking.A One forest products company in Sweden
relation of the pulp following c
properties between fully bleached sulfate standard delignification rate king A markets under the same name pulp pro-
pulps that are oxygen bleached and those the oxygen stage is a good starting point duced at mills which both do and do not
that are not. apply oxygen bleaching.This is a further
if the pulp has a very good viscosity-
example of the fact that comparable pulp
Reference sources, practical experiences Kappa-number relation from the digest-
gained from laboratory and plant in- er.If the pulp shows a poorer relation, a qualities can be manufactured.
vestigations at MoDo and Billerud, and lover rate of delignification is recom-
analyses performed by these companies mended.
pf competitors' pulps have been used in COMPETITIVE TECHNIQUES
the comparison. In the production of softwood sulfate
EXPERIENCES pulp,oxygen bleaching is today the only
technically viable method available with
VISCOSITY CAN BE USED AS A Over the years, a large number of lab- which to achieve low lignin contents
MEASURE OF THE STRENGTH oratory investigations have compared (Kappa number <28) entering the
OF THE PULP the qualities of oxygen-bleached and chlorine bleaching plant and at the same
non-oxygen-bleached pulp. In general, time achieve high pulp quality.
Bleached pulps from the same wood lot the conclusions are that the two different
exhibit the same pulp strength at a given bleaching processes produce pulps with
viscosity whether or not an oxygen stage similar strength properties. An example SUMMARY
is included.Examples of this observation is a summary of 23 investigations carried
date back as far as 1976 at STFI(1). The out at MoDo's research laboratory (2). There are no significant differences in the
same conclusion is drawn for bleached Delignification varied between 44 and 72 quality of oxygen-bleached and non-
pulps in more recent publications, in- per cent, with an average as high as 55 oxygen bleached sulfate pulps, when the
cluding(2), and is supported by practical per cent. Despite this, no significant dif- pulps are produced under controlled
experience. ferences could be noted between the conditions. This applies to strength and
For purposes of comp arison, the viscosity oxygen bleached and non-oxygen optical properties, cleanliness and
of bleached pulp can thus be used as a bleached pulps with regard to thestrength properties. beatability.
measure of the strength of the pulp. For
viscosities greater than 850-900 Oxygen ' bleaching was installed in
dm3/kg, only a marginal increase in Husum's softwood sulfate line in 1977. REFERENCES
pulp strength is obtained with an in- During the seven months prior to, and
crease in viscosity. eleven months following, the introduc- 1. Alfthai. C-1, Homgrist. W: 'Viskositet
tion of oxygen bleaching, the quality at som matt pa massastyrka,'•(Viscosity as a
both the pulp and the paper mill was measure of pulp strength)STR -
DEGREE OF DELIGNIFICATION IN carefully watched (2). Total testing data B-meddelande nr 370(1976).
THE OXYGEN STAGE consisted of approximately 1,700 meas- 2. Jamieson, A: 'The strength of oxygen
urements. No significant changes could bleached pulps,"speech presented at
The precondition for achieving a pulp of be observed in the strength properties of Eucepa, Helsinki, 1980.
optimum quality is that the pulp shall the pulp or the fine paper. It should also
not be unnecessarily damaged. This be noted that the ash content of the paper
applies to oxygen-bleached as well.as was approximately 10 per cent higher
non-oxygen-bleached pulps and, not during the period of oxygen bleaching.
least, to cooking and final bleaching. These results were also confirmed by
This is a well-known fact among pulp Billerud's investigations and practical
manufacturers. experience.
If the pulp is excessively delignified in the Both MoDo's and Billerud's analyses of
oxygen stage, its yield and strength Swedish, and in certain cases Finnish,
properties are jeopardized. When this competitive pulps show that it is impos-
happens, the bleached pulp shows a sible,based on such variables as strength
lower viscosity (and strength). and optical properties, cleanliness and
The same phenomenon occurs if the de- beatability, to discern whether a pulp is
lignification is brought too far in the oxygen bleached or not.However,pulps
digester, —'— produced from wood from northern

Environmental aspects of
oxygen bleaching
CURRENT KNOW-HOW able attention in environme;ital contexts greatly. However, a number of ques-
becuse of its classificatic.n as being car- tions, including questions of the mate-
When oxygen bleaching was first intro- e iogenic). However, most of the rial's biological and chemical degrada-
duced for industrial application, BOD, organic material in the C and E effluents tion under recipient-like conditions and
COD and color were the most important is high-molecular. It is chlorinated and of the capacity of specific compounds to
has a considerably lower aromatic con- bioaccumulate in organisms, remain to
parameters for characterizing the en-
vironmental effects of bleaching plant tent (<10%) than the residual lignin in be answered in order to estimate the risk
discharge. unbleached sulfate pulp (10, 15). posed by the release of such discharges.
g Laboratory investigations have also This applies especially to the behavior of
It was established at an early stage that shown-that C- and E-stage discharges bleach plant waste liquor in the re-
the use of oxygen bleaching substantially from conventional bleaching of soft- cipient. Noteworthy in this context are
reduces the chlorine content of the wood sulfate pulp show a mildly acute recent laboratory results showing that
bleach plant effluent. As a result, a toxic effect on fish (6, ]1. 12) (96 hours high-molecular material from C- and E-
sizable portion of the effluent can be re- LC 50 concentration = 15-50% V:V. stage effluents from the bleaching of soft-
cycled into the process, correspondingly rainbow trout). This effect is caused by wood sulfate pulp can be degraded with
reducing the discharge to the recipient.If chlorinated phenols, catechols and bacteria (isolated from receiver sedi-
a change is made from a conventional guaiacols and by various chlorinated mend and that chlorinated anisoles and
CEHDED to a OCEDED sequence, and non-chlorinated resin and fatty veratroles are then formed which have a
BOD, COD and color discharges are re- acids. C and E discharges at sublethal higher bioaccumulation potential than
duced by between 30 and 70 per cent for concentrations from conventional other compounds from bleach plant
softwood sulfate pulp(1-7).The size of bleaching of softwood sulfate pulp also waste liquor(13).
the reduction depends, among other fac- affect fish. Laboratory tests have dem- In the case of discharges from the bleach-
tors, on the Kappa number of the un- onstrated that such discharges may ing of hardwood sulfate pulp and sulfite
bleached pulp and on the degree of de- affect reproduction, and have behav- pulps, available knowledge of chemical
ligniffcation in the oxygen stage. For ioral, physiological, histological and compositions and biological effects is
hardwood sulfate pulp, the correspond- certain latent effects. The majority of less extensive than for softwood sulfate '
ing reductions are a consequence of these can be observed at concentrations pulp,and needs to be improved in future
changing the bleaching sequence from as low as about 5 per cent of the 96 hours studies.
CEDED to OCED (3, 5, 8). LC 50 value for the discharge (6). The
test for ectoparasites in the gills of fish is
probably the most sensitive. The effects EFFECT OF OXYGEN BLEACHING
CONVENTIONAL BLEACH PLANT of C-and E-stage discharges from the ON THE CHEMICAL COMPOSITION
EFFLUENTS: CHEMICAL conventional bleaching of softwood AND BIOLOGICAL EFFECTS OF
COMPOSITION AND BIOLOGICAL sulfate pulp can be detected using this THE DISCHARGES
EFFECTS test, at concentrations 25 to 30 times
lower than the 96 hours LC 50 concentra- Available knowledge of the effect of oxy-
In recent years, there has been a change tion (6). Today, it is not known which gen bleaching on the chemical composi-
in standards for evaluating bleach plant specific compounds are responsible for tion and biological effects of the dis-
effluents (9). Earlier, BOD and color the effects at the sublethal level: These charge from a subsequent,reduced tradi-
were the primary variables, but today a are not necessarily the same as those tional "chlorine bleaching" is not so ex-
more comprehensive method of evalua- responsible for acute toxicity. tensive as the knowledge concerning
tion has been developed, based on a Laboratory investigations in recent waste liquors from conventional bleach-
gradually increasing number of chemical years have also determined that C- and ing. As mentioned above, the oxygen
and biological tests. As a result, avail- E-stage effluents from the conventional bleaching does, however, result in a re-
able knowledge of chemical composition bleaching of softwood sulfate pulp (and duction in the amount of residual lignin
and.various biological effects,especially other pulp types) contain compounds which is transferred to the chlorination
with respect to discharges from the con- that are mutagenic and!or carcinogenic. and subsequent bleaching stages. This is
ventional bleaching (CEHDED) of soft- A number of such compounds have been the primary reason for the sizable reduc-
wood sulfate pulps, has increased exten- identified (9, 10). However, the content tions in BOD, COD and color which ac-
sively(6,10,11).To date,approximately levels are so low that the risks associated company oxygen bleaching.New results
250 separate compounds have been iso- with the release of these compounds into have shown that this reduction is ac-
lated in C- and E-stage effluents fr6m the recipients—with today's know-how companied by a similar reduction in total
such pulps(20).The compounds consist — are judged to be negligible or non- organic bound chlorine (ToCI) (6, 14).
of chlorinated and non-chlorinated ali- t ten .
phatic and aromatic acids, chlorinated existent. With reference to the qualitative organ-
phenols, catechols and guaiacols and Available knowledge of the chemical icichemical composition of C-stage
a number of chlorinated and non- compositions and biological effects of (possibly with some addition of D) and
chlorinated neutral compounds includ- discharges from conventional bleaching E-stage effluents, today's know-how
ing chloroform (which has drawn not- of softwood sulfate pulp has increased suggests that these do not differ to any

great extent from.tnr lW 6aap"nuuib�.- y..•..��. . ..+.,. _..,�.._.o__ .._._ ....__.. ._ _._.__. ._..-
Fluents from non-oxygen bleaching. If gated and the oxygen bleached pulp con- point.
,the organic material in the effluent is di- tained significantly higher washing loss-
vided into fractions (including material es than the non-oxygen-bleached pulp.
with high molecular weight, water- Logically, this may seem to be the most REFERENCES
soluble' material with low molecular important explanation of the unexpected
weight, ether-soluble material with low P p 1. Croon, f.;Andrews, D.H. Tappi 54(11)
wei
g result. New investigations of acute fox- 7983(1971).
molecular weight - which can be fur- icit confirm this conclusion(9)and indi-
ther divided into sub-fractions con- y 2. 107((son. A.;Smedman, L. Tappi 56(6)
cafe that the introduction of oxygen 107(1973).
taining acids, lipophile acids, lipophflfc bleaching should not only reduce BOD, 3. Carpenter, W.L.:McKean, W(T.;
phenols and neutrals - and finally a COD, :olor and ToCI at comparable Berger, H.F. Gellman, L. Tappi 5901)
fraction containing easily volatile mate- washing efficencies, but also reduce the 81 (1976)•
rial) it is found that the relativ_ sizes of acute toxicity of the discharge. 4. "Klorid i atervinningssystem,
the fractions are largely inrizpendent of (Chloride in recovery systems)final
whether or not oxygen bleaching is used In a recently performed study of re- report, the Swedish Pulp and Paper
(6, 14). cipients outside the M6nsteras sulfate Association, Stockholm 1977.
This suggests that the substantial reduc- pulp mill, it was concluded that bladder 5. Wong, A.: Lebouris,M.; Wostra-
wrack had disappeared from the area dowski. R.;Prahacs, S. Pulp and Paper
tion in the discharge of organic material around the asap discharge pipe area f-anada 79(7)41 (1978).
expected from a well-conducted oxygen The around
th emu'swrack began r disappear ). 6• Miljbvanlig tillverkning av blekt mas-
bleaching must affect all compounds, or sa,"(Environmentally safe production
the same time as the plant was redesigned of bleached pulp)
in least all classes of compounds,existing to accommodate amongother things, P P)final report, the
in C- and E-stage effluents. This con- g Swedish Pulp and Paper Association.
clusion is supported by the observation oxygen bleaching and bleaching with Stockholm, 1982.
that the amountsof chlorophencliccom- chlorine dioxide in the final bleaching 7. Brt, P.B.:Joyce, Chang. l
pounds, among others, are largely the i.
stage. Several conceivable reasons were presentation at thee Tappi Annual
given, but an important recommends- Meeting. I982, Proceedings 397.
same in both types of bleaching se- g P 8. Jones. A.R. Tappi 66(12)42(3983).
lion was that no further oxygen bleach-
quences but are lower for the alternative i+. Amtergren. G. presentation at SPCI,
that
with oxygen bleaching (6). Other sup- ing be installed until further notice ( 'pulp section's autumn meeting. Gavle,
porting evidence includes results show- However, it has since been resolved that Sweden, Nov.30, 1983.
ing that the chemical compositions of the damage to the bladder wrack cannot be 10. Kringstad, K.P.; Lindstrom. K.
materials with high molecular weight are attributed to any known or. unknown Environmental Science Technology,
also of largely the same type (15). The compound formed during oxygen under publication.
observation that only very small bleaching. The most probable cause of 11. Kringstad. K.P.;Stockman. L.G.:
amounts of chloroform (80-90% reduc- the damage was the presence of a rela- Stromberg. L.M. presentation at XXI
tively high chlorate content in the dis- EUCEPA Conference: "Chemical
tion) are formed in the bleaching se- charge(16).This is a result of the use of processes in pulp and paper technology,"
c
quences with oxygen bleaching does not g Torremolinos, May 1984.
chlorine dioxide in the plant and of the Walden, C.C.;Howard, T.E. Tappi
conflict with this conclusion, since this is fact that the plant's white water system is
12
a result of the fact that the bleaching se- 60(1)uz(1977).
quence applying oxygen bleaching has tightly closed. The problem has nothing 13 Neilson, A.H.;Allard. A.-S.:
no hypochlorite stage, which is the to do with oxygen bleaching alone. Hynning. P.-A.;Remberger-, M.;
major source of the chloroform (6). Landner, L. Applied Environmental
Microbiology 45(3)774 (1983).
On the basis of the observation that the SUMMARY 1y. Lindstrom, K.;Osterberg, F. presenta-
introduction of oxygen bleaching sub- tion at SPCI,pulp section's autumn
stantially reduces the discharge of BOD, Available data show that the introduc- meeting, Gavle, Sweden, Nov.301983.
COD, color and ToCI from the bleach lion of. oxygen bleaching before tradi- 15. Lindstrom, K.;Osterberg. F.
plant, and that this reduction affects all yg g comSTFmunication.
Stockholm 1989, private
liana] (and reduced)chlorine bleaching, communication.
classes of compounds, it is also very combined with recovery and combus- 16. Lindvall. B.;Alm. A. Message 83:5
likely that the biological effects of the lion of spent liquor from the oxygen from Kalmar Institute of Technology,
discharge will be reduced correspond- stage,results in a significant reduction in Institute of Natural Science and Tech-
ingly. Few investigations have been the discharge of BOD, COD, color and nology, Kalmar, Sweden, 1983.
made of the biological effects of the ToCI. The reduction includes chlo- 17. Lehtinen. K.-J.;Mattson, 1. IVL,
discharges from the bleaching sequences rinated and non-chlorinated organic ma- Karlskrona 1983, private communica-
containing oxygen bleaching. The re- terials of both high and low molecular tion.
sults available so far are inconclusive. weight, including chlorophenolic com-
Within the framework of the project pounds and'chloroform. In general, the
"Environmentally safe production of introduction of oxygen bleaching there-
bleached pulp;"a comparison was made fore clearly represents a highly signifi-
of the acute toxicity, reproduction dis- cant measure from an environmental
turbances, behavioral changes and standpoint.
various physiological and histological P
sublethal changes (fish) and effects on With respect to the influence of oxygen
the planktonic algae in bleach plant bleaching on the bleach plant discharges,
discharges, where softwood sulfate pulp the few investigations carried out so far
was bleached both with and without are inconclusive. This is probably a re-
oxygen bleaching(6). The results of this sult of variations in washing losses. If
comparison showed that oxygen bleach- such process parameters were better
ing did not seem to reduce the effects, controlled, systematic investigations
which were as large as those due to the would probably also establish that the
discharges from the conventional se- introduction of oxygen bleaching is
13

Authoritiesi' view of
oxygen bleaching
1:.TRODUCTION Swedish pulp and paper industry concen- nomical possibilities" to reduce bleach
[rated its environmental protection efforts plant discharges thr-ugh various measures
The following summary is the result of a Primarily on the environment in and (5).
around the mills. Large resources were During recent years. joint research carried
study of the concession decisions in invested in an attempt to reduce the dis- out by Swedish industry has shown that
Sweden during the past 5-10 years for all charge of oxygenconsuming substances. the rate of delignification and the transfer
plants manufacturing bleached sulfate
pulp.The intention has been to find possi- The first oxygen stages started operation in of organic substances to bleach plants are
ble trends in the authorities'view of oxygen the beginning of the 1970s and represented two
oprimary
factors affecting conventional
tional
bleaching. an answer to the demands of authorities to Ptoxicity n bleach plant
reduce the discharge of BOD7, lignin and discharges.During the current decade, the
color. Often, conditions were formulated Environment Protection Board has also
THE CURRENT STATUS OF INTERNAL so that companies could choose between begun to an increasing extent to insist on a
AND EXTERNAL MEASURES IN an oxygen stage and external purification reduction in the Kappa number in cooking
SWEDISH BLEACHING PLANTS (1) in conventional bleaching and in oxygen
stages,as well as on conditions for washing
Of the fifteen sulfate mills which manufac- During the mid-1970s, the National losses(6).
ture bleached pulp, eight have today oxy- Swedish Environment Protection Board In addition to demands for oxygen stages
gen stages for bleaching softwood pulp.Of regarded oxygen bleaching in combination and aerated ponds, recent requirements
these mills, one also bleaches with conventional final bleaching as a have also been specified en Kappaquire mats
hardwood pulp with an initial oxygen stage. "standard requirement"in new mills.In the reductions,certain pnpiedforroportions of cumber
One additional sulfate mill is planning to opinion of the Environment Protection dioxide ns chlorine stages and improved
hlorine
instal an oxygen stage for bleaching soft- Board, oxygen bleaching was a first step washing n c It J6 interesting to note that
wood pulp. towards a closed bleach plant(2). External one ing (7). It existing aerated pond has
The portion of chlorine dioxide in the chlo- Purification methods for bleach plant ermission to instal extended
waste liquors and ion-exchange and lime- receivedp
rine stage for softwood pulp varies be- sludge methods were regarded as an ac- cooking in combination with better wash-
tween a few per cent and 20 per cent. The ceptable alternative to oxygen bleaching ing and an increase in the level of chlorine
percentage for the bleaching of hardwood for existing installations(3). dioxide, as an alternative to oxygen
pulp varies-between a few per cent and 90 bleaching (8).
per cent. During thelatterhalf of the 1970s.environ- summary, it be said that there was
All fifteen sulfate mills which manufacture mental concerns shifted to an increasing y may
extent towards the long-term effects of dis- a noticeable trend that mills were either or-
bleached pulps have sedimentation ponds.
charges in time and space.To an increasing dered or themselves undertook to instal
However, at one mill, all or part of the extent, discussions began to be concerned oxygen stages during the 1970s.During the
bleach plant waste water is taken directly with the reduction of the discharge of chlo- latter half of the decade, demands in-
past the pond. Five of the mills currently rinated organic substances.As a result, the creased to encompass oxygen stages in
have aerated ponds. A decision to con- Environment Protection Board's demands combination with additional internal and
struct a biological purification system has were increased to encompass oxygen external measures.
been reached at another. In certain cases, bleaching and measures to reduce these
all the bleach plant wastewater is not ledto substances.Among the most important of CONCESSION DECISIONS
the aerated pond. One mill uses external these measures was the exchange of chlori-
purification of bleach plant effluents in the ne for chlorine dioxide in the first bleaching 1. 1,allvik June 8. 1973
form of adsorption on lime sludge ("the stage as well as the external treatment of Skwskar April 1.1974
lime sludge method")and another uses ion bleach plant discharges through aerated Koesnas April 1.1974
exchange treatment. ponds, .and the ion-exchange and fenox 2. 5kutskar June 10.1977
One mill has both an oxygen stage and an methods. Mill closure measures through Karlsborg December 13.1978
aerated pond. At another mill, with oxy- recycling of portions of the waste liquor 3. Ostrand November 20.1973
une 25.1980
gen bleaching,a decision has been made to from the first alkali extraction stage were 4. MoV;irrum October 28.1980
instal biological purification. At the mill also demanded by the Environment Gruvon October 2.1952
planning to instal an oxygen stage, an Protection Boa/tl (4). Gruver, April 6.1982
Husum June 29,1962
aerated pond is already in use.At another A noticeable trend duringthis period was 977
mill with an aerated pond, permission has P 5. MSnsmonster9s May 2e.1.
been granted to instal extended cooking in that the National Swedish Franchise Board Skoghali April 15.1979
combination with improved washing and for Environment Protection deferred final O!tran ll January 15,1980
Ostrand April 30,1980
an increase in the use of chlorine dioxide in demands and requirements with respect to Varo June 25,1980
the chlorine stage, as an alternative to bleach plant discharges. At the time, the Morrum October 28.1980
Franchise Board for Environment Pro- Norrsundet October 7.1981
introducing oxygen bleaching. tection decided to await the results of the 6. Ostrand April 10.1980
ISM von October 2.1980
Swedish Pulp and Paper Association's then Norrsundei October 7.1981
TRENDS IN OXYGEN BLEACHING ongoing project, "Environmentally safe Husum June 29,1982
IN SWEDEN manufacture of bleached pulp.' During the 7, Husum June 29,1962
period of deferment, mills were usually S. Karlsborg December 14.1982
During the 1960s and in the early 1970s,the ordered to investigate"technical and eco- Karlsborg January 12.1984

ruture.PUtellull vl
oxygen technology
DELIGNIFICATION WITH AN sp.nt liquor to the closed liquid and re- its aggressive character,which reveals itself
ALKALINE OXYGEN STAGE covery system of the pulping process. in difficulties in controlling pulp quality
Pretreatment with green liquor, peroxide, !,his applies to the process with high
ozone or nitric oxides in acid media gives concentration), and its cost, which, as a
A basic precondition for the use of an positive results.The best results for sulfate result of the low energy and conversion
alkaline oxygen stage in a delignification pulps are achieved with nitric oxides. In yield, has remained at a high level.Ozone
operation is that the development of the Sweden, Professor 011e Samuelson, to- production technology has developed
selectivity for the delignification is better gether with MoDo, has carefully studied gradually; important progress has been
than in the final phase of the sulfate cook- the treatment of unbleached sulfate pulp achieved in recent years.It can therefore be
ing. The viscosity and Kappa number of with nitric oxides prior to an oxygen stage. expected that an introduction of ozone in
the cooked pulp are determining factors for certain positions will be economically
the technical and economic results of the Amazingly good results have been feasible within the near future. Low con-
process in the oxygen stage'.A high intrin- reported in terms of pulp yield and espe- cenlration methods(around 3 per cent pulp
sic viscosity(<1200 dm3/kg)for a Kappa cially for pulp viscosity after the oxygen consistency)appear to be easier to control
number around 30, with small variations, stage. It appears that pulp quality can be with regard to the viscosity of the pulp.
is a precondition for a good result.The de- maintained at a high level'down to Kappa Ozone may pose an attractive future alter-
velopment of cooking technology greatly numbers of about 10. It also appears pos- native as a complement to oxygen bleach-
furthers the possibilities of improving eco- sible to increase the degree of delignifica- ing, primarily as an activating pre-stage.
nomic and environmental effects by using Lion to 75 per cent from the current average
an oxygen stage as the first stage following of 50 per cent.
cooking. The introduction of high sul- These resul ts have been so promising tha t a CLOSED SYSTEMS
fidity,atmospheric steaming-and computer group of Swedish companies — MoDo,
control has, generally speaking, increased Sunds Defibrator, AGA and Kema Nobel The basic idea underlying the increased use
the possibility of being able to deliver a — have decided to finance a pilot plant of non-chlorine, non-corrosive chemicals
high-viscosity pulp with stable Kappa project to facilitate the development of in the bleaching plants is to recycle waste
number from the digester to the oxygen equipment and to evaluate the process liquors to the'recovery cycle. It is then
stage. from the technical and economic perspec- possible to burn the organic matter and re-
Modified cooking processes have been de- tive.A thorough analysis and characteriza- cover sodium.
veloped and tested on a large scale. It is tion of the effluent from the bleach plant A Swedish pulp mill is currently recycling
possible to cook to around Kappa 25 while will also be conducted as well as of the the liquor from the OE stage to the re-
maintaining viscosity. The total yield will health hazards in the environment within covery cycle. By using oxidized white
be somewhat lower and•the consistent the mill. liquor in both the O and OE stages, and
quality of the pulp has not as yet been using the flexibility of the chlorine dioxide
ensured on a full scale. However, the production, the plant has succeeded in in-
know-how gained can contribute to the HYDROSTATIC OXYGEN STAGES troducing an economical system with low
production of a pulp with Kappa numbers es.dischar
of 28-30 with consistently better viscosity. Oxygen-reinforced alkali extraction stages g
Therefore it is very likely that the current have been shown to be both an effective A bleach plant fora brightness of 90%ISO
cooking-oxygen process can be further de- and economical complement in the bleach- with two pressurized oxygen stages and
veloped using improved washing methods ing sequence, especially as the first alkali satisfactory washing systems may be
now under development so that Kappa extraction stage. By raising the height of designed with only chlorine dioxide as a
numbers of 12-14 can be achieved with- the tower, the effectivity of the delignifica- bleaching agent. The bleaching plant is
out adversely affecting the quality of the tion process can be increased.This variant closed to a high degree with a water con-
pulp. may also be introduced as a first stage in sumption of only 5 m3/ton of pulp. A
This means(contrary to what has been re- special cases such as with hardwood sulfate bleed of 0.5-1 m3/ton is assumed. This
ported in certain contexts) that the im- pulp, sulfite pulp, or other pulps where a waste liquor may be treated before it is
proved cooking process should not be lower delignification (around 35 per cent) released to the recipient. If the brightness
regarded as an alternative to oxygen is acceptable. target is sufficient for the pulp's end use,
bleaching, but as a complement, which A sequence of the (EO)(CD)(EO)D type application of a hydrostatic extraction
improves the positive effects of the oxygen may be an attractive alternative. stage may be a viable alternative to the
stage on the production economy and the pressurized oxygen stage.
environment. It has also been demonstrated that the
replacement of the hypostage with an
In recent years, it has been demonstrated alkali oxygen stage is favorable for both
that it is possible to activate the unbleached birch and pine sulfate pulp(for example, a
pulp using certain pretreatments so that the (CD)E(EO)DEDsequence).
selectivity of a subsequent oxygen stage is
greatly improved. Pretreatment with
chlorine prior to the oxygen stage consider- OZONE
ably improves the selectivity of the oxygen
stage. Because of corrosion problems, The introduction of ozone into the bleach-
however,iris hardly possible to recycle the ing process has so far been handicapped by
15

Champion International Corporation
Canton Mill
EFFLUENT DISCHARGE QQLQBR REDUCTION PROJECT
MAJOR PROCESS MODIFICATIONS
SCREEN ROOM
The unbleached pulp produced by digesting wood chips in
the presence of caustic soda and sodium sulfide is washed
with fresh water to remove soluble lignin derivatives. After
washing , the pulp is- processed through several stages of
screening to remove undigested small wood particles. This
screening is normally done at low consistency (1%) after
which the pulp is thickened to 10-12% consistency before
storage and subsequent bleaching.
The filtrate generated during the thickening operation
is brown in color and can be either sewered in an open system
or recycled to pulp washing in a closed system. This
recycling results in an increase in dissolved solids. The
gravity discharge Cowan screens used at the Canton Mill are
older models which cannot function properly in the presence
of higher concentrations of dissolved solids. Therefore, new
pressure screens must be installed and modifications made to
the washing system to permit filtrate recycling and reduce
color losses to the waste treatment system. This modifi-
cation results in "closing" the screen room.
OXYGEN DELIGNIFPCATON
In an oxygen delignification system, washed unbleached
pulp is mixed with caustic soda and oxygen at elevated
temperatures and pressures. In a period of approximately one
hour, about one-half of the incoming quantity of lignin is
made soluble by the reaction with oxygen. Pushing the
reaction further results in a rapid degradation of
carbohydrate chain length and a reduced pulp strength. Since
oxygen delignification does not utilize any chlorine
containing streams, its filtrate can be recycled through the
brown stock washing system to the recovery boiler . The
lignin in the liquor is burned in the boiler and the
inorganic chemicals recovered in the smelt.
The oxygen treated pulp contains less lignin and
consequently- requires less chlorine to reach a required pulp
brightness. . This process change significantly reduces the
discharge of chlorinated and colored organics from the
bleaching operation. Figure 1 illustrates changes in the
screen roam—and the proposed oxygen delignification system.

r 01400Fr• fuW
m a P- 0 + n r
of r rr rr ti H.rr r ft o ro
o rrr :C - Mm .
0' a 00 140 y 0f-
w 0 r ti rr rr 0r
� nmx° O lb 0roormi �
a aLa ti
r M
am to (DrnX0oa r °
0 to " 0 m W to rr
H. 0m
Fj a aa'a rC'raro fD
w G fD m f•• rr n it 0
rrro0r00m0lY a .
"40rt1 . 0t0 AM rt
• rr r•w (D rt m m M.
m O p O O N m °
_ 0 `Jw• 0 N `
rt M m wr•m D)
wrt wy 0 0 x r
a a o O' MNtO M• A
rrtam Ort to p
m rr M a to rr rr
Cl. H. 0 r P. f•. ray
N 0 p E W r (0D W p
r- O. rr a Al
to FA Or 11 op a a
0 O•L< r E 0 0 r 1
. rur m Om olm M.
O
rAam 011E o0 0
Op0 . ro 00'
a emu, ti• m
0 0 a pH. MN
0t0 0 0 rt p a
a am xO• n 0
(D 0 U) r rt
max O• mnr
oro o E o a a
a c H- rr O M a
N O• m pmMOart
o r o O ro O ft
wo 0* a) ro a
rrrr aom
I Ng w O
m - Rro O rrroA
,pm rr n LO c a
NNtiryaa'Crrt
0 i0 °d
°
O 0 m
m M a y
m 0
m

FIGURE 1
CANTON MILL
CURRENT MILL DIGESTER PULP MANUFACTURING PROCESS
BROWN SCREEN BROWN
STOCK ROOM STOCK
WASHER THICKENER
€ <;
TO BLEACH PLANT
C WHITE WEAK
LIQUOR BLACK BROWN FILTRATE
LIQUOR
cHEMICAL = TO SEWER
RECOVERY
SYSTEM
PROPOSED MILL
DIGESTER OXYGEN
DELIGNIFICATION
BROWN BROWN
;••• -
STOCK .SCREEN STOCK •,;•�::+;• PULP
WASHER ROOM WASHER WASHER TO BLEACH PLANT
••4 L
S WHITE WEAK
LIQUOR BLACK BROWN OXYGEN
LIQUOR FILTRATE FILTRATE
CHEMICAL
RECOVERY
SYSTEM

. . ■ � ■ ■ ; . ■ ' ' . ■ . , .
, . .
�y
�4J
i}
�E
(\
�
4
� \ \
«ate\>
� � ,

Effluent Color Sources - Present Mill
During January and February 1986, Champion conducted an extensive lithium
ion tracer survey to determine the primary sources of color in the Canton mill.
Survey Results % Raw Color Load By Source:
Caustic Extraction Stage
Bleach
Plant
Fir
Chlorine Stage
IF
•
® Unbleached Pulp Filtrates
tea;,; 3'` ga '4;a4•.:. �,, Pulp
mill
and
=k �'. .: T3.�:�a�: Recovery
" '`:i;�Z�,.3
:° a'�` Area
s
3I� � Z31r5�i3..
�� �' ���� a�3,3 Continuous Process Losses
Discontinuous Process Losses
Other
III Champion Exhibit 1
Champion International Corporation

Extended delignification
Closed screen room
Pulp mill
Oxygen
process Dell nlflcatibn Liquor Recovery
changes
In-mill Process Loss
Process Reduction
Changesto
Reduce Process Control
Improvement Chlorine dioxide
Color substitution for Cl2
Bleach Caustic extraction
Plant with of en
PfOCeSS Caustic extraction with 02
Changes +sodium h ochlorite
Caustic ext=with.02+hydro
(III Champion
v Champion International Corporation Exhibit 2A

Precipitation by GravityClarification Kiln burning of lime
lima Centrifugation stud elrecove of lime
WaterChem,Inc.
(pro data chemis
Precipitation by lertic
chbdde Dissolved Air Flolation
Press Dewatering
precipitation Precipitation by Gravity Clarificationof CEF Color acidification
Precipitation by
pol➢arnme Landfill
Precipitation by
alum+polymer
Dissolved Air Flotation Lr Add Veatmeiiu;i very
of alum sludges
CEF color Precipitation by alum
separation
processes
EIncineration
cineration N e
cated Incinerator r
Evaporation'
Perc shlion or n ash In mill
ashlbohom ash cova bofler
Treatment of Use otconcontrate as an
Bleach Plant Ca bon adaorptioNacid additive for adhesives
Caustic regeneration(ISEP)
Extraction LandfiVetll ofe Ultraviolet-acnvaled
Filtrate
Mivofiltration concon ozone oxidation
Ozone destruction of
concentrate color Ozone oxidation
UlVafiltretion Chlorine oxidation
• Wet air oxidation of
concontrete color Potassium
rman mate ozldatien Sodium hypochlorite
CEF color
oxidation
oxidation
(destruction) Ca dim hypochlodie
processes oxidation
P Sodium perozlda
ozidatlon
Chlorine dioxide
oxidation
Hydrogen peroxide
oxidation
�JII Champion
V Champion International Corporation Exhibit 2B

Wastewater Biological augmentation
Color In activated sludge
oxidation
(destruction) Precipitation by Wastewater aeration+
processes alum chlorine+peroxide
Treatment of
Precipitation by
Combined Mill amine
Wastewater
Precipitation by
Precipitation ferric Ion
Wastewater In Primary Press Dewatedng Landfill
Treatment Precipitation by
COIOr lime
separation &
disposal Precipitation by calcium
processes chloride+alum
WaterChem,Inc.
ro data chemia
Calcium ion treatrneni
in activated sludge
Polyamine in Dissolved Air Flotation Acid Recovery of Incineration In mill
treated effluent Pol amine 11recovery boiler
i......................._.............................._................................._..._......(Stone Process) ......_._...__......................._._...................................._...............
I
Champion
Champion International Corporation Exhibit 2Ci

Effluent Color Sources - Present Mill
During January and February 1986, Champion conducted an extensive lithium
ion tracer survey to determine the primary sources of color in the Canton mill.
Survey Results - % Raw Color Load By Source:
Caustic Extraction Stage
Bleach
Pliant
Chlorine Stage I
® ® Unbleached Pulp Filtrates
II is
R 't a c� rr Pulp
.
NNH mill
and
Recover
Area
R� 3. S Y
M;;,;,,, Continuous Process Losses
Discontinuous Process Losses
Other
Champion
Exhibit 1
Champion International Corporation

Extended delignification
11
Closed screen room
Pulp mill oxygen
process Deli nificatlon Liquor Recovery
changes
In-mill Process Loss
Process Reduction
Changesto
Reduce Process Control
Improvement Chlorine dioxide
Color substitution for C12
Bleach Caustic extraction
Plant with o en
Process Caustic extraction with 02
Changes +sodium h ochlorite
Caustic extraction with 02
+hydrogen peroxide
III Champion
�Vf Champion International Corporation Exhibit 2A

Precipitation by Gravity Clarification Centrifugation Kiln burning of lime
lime sludge/recovery of lime
WaterChem,Inc.
proprietary chemistry)
Precipitation by ferric
chloride Dissolved Air Flotation
Press Dewatedng
Precipitation Precipitation 6y Gravity Clarification
of CEF Color additicetion
Precipitation by
potyamine Landfill
Precipitation by
alum+polymer
Dissolved Air Flotation Acid treatmenVrecovery
of alum sludges
CEF color Predpfladon by alum
separation
processes
Incineration in a
_•_��•®-��• dodicated indneralor
Evaporation
Percolation through fly Incineration in mill
asMbottom ash recovery boiler
Treatment of Use of concentrate as an
Bleach Plant Carbon adsorptiordacid additive for adhesives
Caustic - regeneration(ISEP)
Extraction Landfll of Ultraviolet-activated
IvLcrofiltration concentrate
Filtrate ozone oxidation
Ozone desVuctlon of
concentrate color Ozone oxidation
UltrafilVaflon Chlorine oxidation
Wet air oxidation of
_ concentrate color Potassium Sodium hypochlorite
CEF color rtnan errata oxidation oxidation
oxidation
(dOStrUCflOn) Calcium hypochlarite
processes oxidation
P Sodium peroxide
oxidation Chlorine ioxi
de
oxidation
Hydrogen peroxide
oxidation
III Champion Exhibit 2B
v Champion International Corporation

Wastewater M
ation
color ge
oxidation
Preci station b on+
(destruction) p y ide
processes alum
Treatment of Precipitation by
Combined Mill I amiY ne ...
Wastewater
Precipitation by _
Precipitation fe rri c ion
in Primary Press Dewatering Landfill
Wastewater Treatment Precipitation by
color lime
separation &
disposal Precipitation by calcium
processes chloride+alum
WaterChem,Inc.
ro riet chemis
Calcium ion treatment
in activated sludge
Polyamine In Dissolved Air Flotation Acid Recovery of Incineration in mill
treated effluent Pol amine recovery boiler
1.......................................................................................................(Stone Process) .....................................................................................................1
Champion Exhibit 2C
Champion International Corporation