HomeMy WebLinkAboutNC0000272_Addtl.ColorRemovalOpp.Report_20080225 olso STgTE
s,
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
o z WASHINGTON, D.C. 20460
yrq�PROIE�
February 25,2008 OFFICE OF
Memorandum WATER
Subject: Additional Color Removal Opportunities, Blue Ridge Paper Product's(BRPP)
Canton,NC Bleached Kraft Paper Mill,2008 NPDES Permit Renewal
From: Technology Review Workgroup(TRW)
Donald Anderson,Chair, EPA { .4 Af✓`t i�
Karrie-Jo Shell, EPA Region IV
Marshall Hyatt, EPA Region IV
Paul Davis,Tennessee DEC
David McKinney,Tennessee DEC
Roger Edwards, North Carolina DNR
Sergei Chemikov,North Carolina DNR
To: North Carolina Division of Water Quality and the NC Environmental
Management Commission's NPDES Sub-Committee
Since the 1997 NPDES Permit Settlement Agreement for the Canton Mill,the Technology
Review Workgroup(TRW) has examined the progress made at the facility in relation to reducing
the color content of the MilI's effluent and the potential for additional color reduction at the Mill.
The focus of this examination includes a review of reports and data submitted by Blue Ridge
Paper Products (BRPP), information and data gathered attendant to a visit to the Mill on
February 8,2007 by EPA's Technology Team (Tech Team) and TRW members, and
consultation among the EPA Tech Team and the TRW's members.
The TRW notes that the Tech Team submitted a Memorandum, publicly released by EPA
Region 4 and dated September 5,2007,including findings and recommendations for further
color reduction based upon its data gathering and analyses. The TRW also notes comments
submitted from BRPP, separate from the State of North Carolina comments submitted by letter
dated September 14, 2007, which took exception to some of the findings and recommendations
in the Tech Team Memorandum. Further, a meeting of the TRW was held in Asheville,NC on
October 23,2007,to discuss and attempt to come to resolution on these comments. Also in
response,the Tech Team prepared responses to these comments and an Addendum to the Tech
Team Memorandum. All of these documents are included as attachments to this TRW
memorandum. These documents represent an appropriate evaluation of the potential for further
color reduction at the Mill over the upcoming permit cycle (estimated to be 2008 through 2013).
The original Tech Team report,BRPP and NC DWQ comments and Tech Team Responses to
NC DWQ comments,TRW meetings and discussions, addendum to the Tech Team report,and
other public comments form the basis for the TRW's attached recommendations to the Division
Internet Address(URL)• hnpl/www.epa.gov
Recycled[Recyclable•Printed with Vegetable Oil Based Inks on 100%Posiconsumer,Process Chlorine Free Recycled Paper
2
of Water Quality and the Environmental Management Commission's NPDES Sub-Committee as
guidance for developing the effluent limitations for color and related special conditions of the
draft NPDES Permit renewal. The TRW also notes that public interest organizations separately
provided their comments to the State of North Carolina.
The TRW recognizes and concludes that not all of the options outlined in the Tech Team
Memorandum can be predicted to achieve specific color reductions with complete accuracy.
Consequently this Memorandum from the TRW provides,where appropriate, estimates of an
aggregate range of likely color reduction from the application of a group of specific technologies
that are available. In making these recommendations, the TRW relied on the following basic
tenets:
• The highest priority for additional color reduction rests with additional in-mill
improvements,particularly considering the success already achieved by continuing
improvements in leak and spill prevention and control (Best Management Practices—
BMPs) and process modifications and changes.
• As in-mill technologies are exhausted, segregated waste stream pretreatment and end-of-
pipe treatment technologies,while they may reflect significant economic,technical, and
non-water quality environmental impact issues at this mill, must also be considered
carefully as supplementary options.
i
In presenting these attached recommendations to the Division of Water Quality and the
Environmental Management Commission's NPDES Sub-Committee, the TRW wishes to
acknowledge the excellent success achieved since the 1997 Settlement Agreement and the
continued effort of BRPP to improve on this record. Very difficult and unusual circumstances
occurred in 2004,with back-to-back historical flooding of the Pigeon River and associated
significant costs ($39 million) for repairing damage and replacing equipment at the BRPP mill
after 21 days of being out of operation. The TRW also recognizes that significant additional
expenditures have been made by BRPP over the last permit term for environmental projects
($25.9 million), the largest portion for air pollution controls (e.g.,MACT I and MACT II
compliance) and including color reduction projects ($5.9 million), all building on previous
projects and expenditures. All of this has occurred during a continuing and difficult period of
industry-wide transformation, capacity shrinkage, and mill closings. Nonetheless, the BRPP mill
management has committed to surviving these challenges. These efforts and expenditures have
resulted in substantial and commendable progress made to date by BRPP through additional and
improved best management practices,process and related technologies, and incremental
improvements in treatment system performance which have reduced the long term average color
discharges. It is also noteworthy that some of these technologies and practices have been
implemented by BRPP in addition to those identified in the previous Tech Team and TRW
reviews: The TRW notes that through these efforts the long term average end-of-pipe color
discharge from the mill has been reduced from approximately 42,300 lbs/day in 2001 to
approximately 37,100 lbs/day in 2006. While the Canton Millis among world leaders in the
pulp and paper industry as measured by the quality of its treated effluent, the commitment of
BRPP, the availability of other potential color reduction technologies, and the continued interest
3
and participation in this process by the regulatory agencies, stakeholder public and
environmental interest groups, and the general public makes additional improvement during the
upcoming permit cycle appropriate.
If there are questions concerning the attached recommendation,please feel free to contact the
TRW.
Attachments: 1) TRW Recommendations
2) Memorandum from EPA Tech Team to the TRW, entitled,Additional
Color Removal Technologies for Blue Ridge Paper Products, Canton,NC,
dated September 5, 2007
3) NC DWQ comments on draft Tech Team report, dated September 14,
2007
4) Tech Team Responses to NC DWQ and(Indirectly) to BRPP Comments
5) Addendum to Memorandum from Tech Team to the TRW
6) Available upon request- electronic files with:
➢ Submissions from BRPP in response to Tech Team requests
➢ BRPP comments, dated June 4,2007, on first draft Tech Team
Memorandum
➢ Telecon with EPA Tech Team on July 25, 2007 (agenda and notes)
➢ BRPP comments dated July 26, 2007 on second draft Tech Team
Memorandum
➢ Data for color discharge,production,various Tech Team analyses
of data, etc.
4
TRW Recommendations
The options needing further study should be incorporated in the permit as an additional
increment of color reduction to be demonstrated by the Mill. BRPP should be required to
evaluate the technologies identified below and to develop an implementation plan that would
either utilize these technologies when technically, operationally, or economically feasible, or
identify other options that would result in similar increments of color reduction. The TRW
recommends that it be allowed to review and comment on the justification for any item among
the following recommendations found by BRPP to be technically, operationally, or economically
infeasible.
1. The following suite of items should be implemented, upon further expedited evaluation if
necessary to refine detailed design and operating parameters, during the upcoming permit
cycle:
A. further improvements in leak and spill prevention and control (BMP's) covering
all process lines, including probable color-generating sources (e.g., sulfide
containing) among white and green liquors in the recovery cycle
B. process optimization (enhanced extraction stages, reduced bleaching chemical
use, etc.); and
C. addition of second stage oxygen delignification on the softwood/pine fiber line
The time necessary for BRPP to implement these items in logical sequence should
realistically reflect the Mill's ability to design, fund, and install them at the earliest
possible date. For example, an updated and detailed evaluation of the addition of second
stage oxygen delignification on the softwood fiber line should identify necessary
adjustments to upstream pulp digestion (e.g., kappa number targets),bleaching (e.g.,
bleaching chemical usage rates,kappa factors) and downstream brightness/strength and
other relevant process control and product quality parameters, designing and costing, and
refining color reduction projections.
2. The following items also should be evaluated and implemented as appropriate during the
upcoming permit cycle:
A. increasing filtrate recycle and use of the existing BFR process for the hardwood
fiber line
B. reducing black liquor carryover by further evaluating in detail and adjusting
operating conditions in the direct contact evaporators (DCEs)
C. reducing impact of Chloride Removal Process (CRP)purge on treated effluent
color by gathering data sets over as long a period as possible,preferably at full
scale,with and without the CRP purge to better inform the treatability of this
source of color. If CRP color is found not to be removed in the treatment system,
further study should assess:
5
o securing whatever additional reductions are possible based on any
demonstrated technology that works and can be economically applied to
this waste stream, either within the mill and sewer system, or chlorine
dioxide pretreatment to reduce color in the CRP purge stream prior to
introduction to the treatment system, or through end-of-pipe activated
sludge biological wastewater treatment system
o avoiding release of the CRP purge during periods of low flow in the
receiving stream
D. better understanding and controlling the physical and chemical mechanisms
underlying"sewer generated color;"this will require a sustained effort going
forward beyond the upcoming permit term given that process changes and BMP
improvements to be made during the upcoming permit term will more than likely
further change the chemistry and mechanisms underlying"sewer generated color"
E. improving color removal by the Mill's wastewater treatment plant via
o better equalizing and further optimizing treatment by using polymers
and other chemicals to pretreat highly-colored segregated wastewaters
including streams that are diverted to the extra primary clarifier, or by
other operational and/or treatment means not yet identified
o investigating use of polymers or other chemicals upstream of the
secondary clarifiers, especially during periods of high influent color
and/or low river flow
3. Contrary to past practice for lowest certainty options, the TRW does not consider it
necessary for BRPP to study further the Ozone- Chlorine Dioxide(ZD)process change
within the upcoming permit cycle. However,the TRW does consider the ZD process an
available option beyond the upcoming permit cycle if further reductions in color
discharges are found to be necessary.
4. The permit should continue to require BRPP to report on or identify any"breakthrough"
color reduction technologies developed at the Mill or that become available within the
industry or the research community.
5. While the current permit does apply color limits at the effluent discharge point, the
permit issued for the upcoming term should clearly reflect that the color limits and
sampling point for color compliance is the Mill's end-of-pipe discharge point to the
Pigeon River.
6. The permit to be issued for the upcoming permit term should include an immediately
effective maximum day effluent limitation for color based upon current operations and
data representing recent long term performance,as it relates to capturing and better
controlling day-to-day effluent variability. Some TRW members raised short term
variability as a significant concern that should be addressed in the upcoming permit
cycle.
6
7. After implementing and putting into operation additional technologies and practices per
the above recommendations, but no later than the end of the upcoming permit term,the
permit should require an effluent target range of 32,000—37,000 lbs/day as an annual
average. Data on effluent color loadings should be statistically evaluated to develop a
revised annual average within this range, along with 30-day average and maximum day
effluent limitations for color,by the end of the upcoming permit term which capture
actual day-to-day variability at the end-of-pipe immediately prior to discharge to the
Pigeon River. The Division of Water Quality should apply those revised effluent
limitations in the permit through formal notification.
8. The permit should require a formal evaluation with periodic update reports as well as any
necessary permit reopeners regarding color reduction efforts.
ATTACHMENTS
MEMORANDUM
DATE: September 5,2007
SUBJECT: Color Removal Strategies For Blue Ridge Paper Products,Inc
FROM: EPA Technology Team'
TO: Technology Review Workgroup
Purpose of this Review
Blue Ridge Paper Products Inc.(BRPP)has requested renewal of their National Pollution
Discharge Elimination(NPDES) discharge permit NC0000272. This permit allows discharge to the
Pigeon River of industrial and other wastewaters from the pulp and paper mill BRPP operates in
Canton,NC. In their May 2006 Color Compliance Report(Blue Ridge,2006a),BRPP requested
that North Carolina reissue the NPDES permit for the Canton Mill with an annual average effluent
color limit of 39,000 lbs per day. This is a reduction from the current 42,000 lb per day annual
average limit which became effective January 2004. BRPP's requested limit was based on the
mill's 2005 annual average effluent color,39,000 lbs per day. EPA Region 4 requested that the
EPA Technology Team("Tech Team")support EPA's review of the color limits included in the
draft permit renewal developed by North Carolina Department of Environment and Natural
Resources. The Tech Team last evaluated color discharges at the mill in 2001 (EPA Tech Team
2001).
Members of the Tech Team visited the Canton mill on February 8,2007 with members of the
Technology Review Workgroup (TRW)to observe and collect information and data on the status of
technologies implemented and color discharges at the mill. This memorandum incorporates
analysis of data provided by BRPP prior to and from that visit, and other data subsequently
provided by BRPP at the request of the Tech Team.
This memorandum reviews the process improvements previously analyzed by the Tech Team,the
status of their implementation at BRPP,and identifies additional color reduction activities that
BRPP could implement during its next permit term. This memorandum also reviews the variability
of the mill effluent color discharges and suggests final effluent color limits for incorporation in the
revised permit.
Background and History
BRPP operates a bleached papergrade kraft pulp and paper mill in Canton,NC,which it purchased
from Champion Intemational Corporation in May 1999. BRPP is owned 55 percent by KPS
Special Situations Fund,L.P. and 45 percent by the employees through an employee stock
ownership plan.2 BRPP announced on June 14,2007 that it is being acquired by The Rank Group.
The New Zealand-based Rank Group has considerable holdings in the paper industry. Rank Group
recently bought Evergreen Packaging,the former North American beverage packaging division of
International Paper. Operations at the mill began in 1908,but the mill has been extensively
'EPA Tech Team members are Abmar Siddiqui, EPA/EAD;Karrie-Jo Shell,EPA Region 4;Donald
Anderson,EPA/EAD;Betsy Bicknell,ERG,contractor;Neil McCubbin,subcontractor to ERG;Dan Bodien,
subcontractor to ERG.
z BRPP announced on June 14,2007 that it is being acquired by The Rank Group. The New Zealand-based
Rank Group has considerable holdings in the paper industry. Rank Group recently bought Evergreen
Packaging,the former North American beverage packaging division of Intemational Paper.
Memorandum
September 5,2007
Page 2
modernized,with the most recent major project completed in 1993. The mill currently operates an
810 tpd hardwood pulping line and a 600 tpd softwood(pine)pulping line. After cooking,pulp
from each line is further delignified in single-stage oxygen delignification(OD)systems,both
installed in 1993. Hardwood pulp is subsequently bleached with a DED sequence(BRPP stopped
adding oxygen to the hardwood E stage after 2001). Pine pulp is bleached with a DEoD sequence
(BRPP stopped adding peroxide to the pine E stage after 2001). Target brightness is 86 ISO,an
increase by one point from the target brightness in 2001. Up to 80 percent of the filtrate flow from
the pine bleach line is returned to the recovery cycle using the unique Bleach Filtrate Recycle
(BFRTm)process,developed by Champion. This 80 percent closure rate is an improvement from
2001,when the pine line closure averaged 73.7 percent. Up to 25 percent of the hardwood line
bleach plant E-stage filtrate flow is also recovered. BRPP produces 300,000 tons per year of
uncoated paper including offset,tablet,and envelope grades.The mill also produces 275,000 tons
per year of bleached paperboard used for liquid packaging and paper cups, including FDA-
approved grades for milk and juice cartons.
Basic tenets of this review,which are consistent with the Tech Team's approach beginning with the
original Settlement Agreement(with Champion International) and the 2001 TRW review with
BRPP,are:
• The first and highest priority again is focused on available in plant process changes and
best management practices(BMPs)as the most cost-effective approach to color
reduction in order to maximize the likelihood of success
Process changes deemed to be of highest and reasonable certainty,thereafter
lowest certainty technologies
BMPs that hold promise to further reduce generation and discharge of color
• After it is clear that in plant process changes and BMPs alone are not be sufficient,in
plant segregated stream pretreatment and end-of-pipe treatment options are carefully
considered
The Tech Team recognizes that very difficult and unusual circumstances occurred in 2004,with
back-to-back historical flooding of the Pigeon River and associated significant costs($39 million)
for repairing damage and replacing equipment at the BRPP mill after 21 days of being out of
operation. The Tech Team also recognizes that significant additional expenditures have been made
over the last permit term for environmental projects($25.9 million),the largest portion for air
pollution controls(e.g.,MACT I and MACT II compliance)and including color reduction projects
($5.9 million), all building on previous projects and expenditures. All of this has occurred during a
continuing period of industry-wide transformation, capacity shrinkage,and mill closings.
Nonetheless,the BRPP mill management has committed to surviving these challenges. The Tech
Team notes that these efforts and expenditures have resulted in substantial and commendable
progress made to date by BRPP through expenditures for additional and improved best management
practices,process and related technologies,and treatment technologies which have reduced the long
term average color discharges. It is also noteworthy that some of these technologies and practices
have been in addition to those identified in the previous Tech Team/TRW reviews.
Process Improvements Previously Analyzed by Tech Team
As a result of its 2001 analysis of the BRPP mill discharges and operations,the Tech Team
identified five mill improvements that it judged capable of reducing the discharge of color in the
mill effluent. BRPP and its consultant also identified several process optimization projects that
would reduce color discharges.The status of BRPP's implementation or evaluation of each of these
improvements is summarized in Table 1,discussed below.
Memorandum
September 5,2007
Page 3
Table 1. Review of Process Improvements Identified in 2001
Predicted.Color
Reduction(Final Color
Effluent,lbs/day Reduction
Improvement ann.avg): Implementation Status _ :Achieved Comments
Process 1,400 Predicted reduction in Consultant recommends reduced C102
optimization C102 use(27%on HW, use and adding oxygen and peroxide to
projects 17%on SW)not made. E stages to reduce color generation
C102 use increased on
SW,decreased 5%on
HW Total
BFR reliability 1,000—1,200 Pine line filtrate recycle reduction Further improvements to BFR
improvement increased from 74 to from 2001: reliability may not be feasible
80%. 6,000 lb/day,
Improved black >5,000 Spill collection annual Additional improvement to black
liquor leak and improvements made; average liquor control possible: further
spill collection untreated color(influent eliminate overflows and pretreat
control(BMPs) to treatment)variability diverted high-color wastewater.Also
unchanged improve control of losses of white and
green liquors to reduce or eliminate
sulfide-based color
Ozone/C102 3,000—6,400 Inadequate evaluation BRPP concerns for this technology are
stage for (lab studies);not installed noted.Notwithstanding inadequacies
hardwood line of lab studies,this technology
None considered low certainty in this case
and further detailed study not
warranted at this time.
2nd Stage OD for 1,100—1,400 Incomplete evaluation Detailed engineering evaluation needed
pine line (lab studies);not installed for costing and design and color
None reduction,followed by
implementation.BRPP concerns for
additional study are noted,but still
considered high certainty.
Color treatment 52,750 Evaluated(lab studies); CRP purge stream now averages 8,745
of CRP purge not installed lb/day. Detailed testing needed for
stream extended period to detemune what
fraction of this color is removed in the
None W WTP and strategies for reducing
color released in the purge. Any color
that is not removed in W WTP could be
reduced with segregated stream
pretreatment or other options.
Process Optimization Projects
In 2001, Liebergott&Associates and GL&V Pulp Group prepared an extensive analysis of the mill
fiberlines that included identification of options for effluent color reduction. Dr. Liebergott was
retained for this work by a coalition of environmental groups. This report,Bleach Environmental
Process Evaluation and Report(BEPER)(GL&V 2001)presented 16 recommendations for
incremental improvement to fiberline operations. These included suggestions for improving
process control,OD performance,reducing the amount of C102 used on both bleach lines,and
enhancing the extraction stages on both bleach lines with oxygen and/or hydrogen peroxide. The
BEPER recommendations were evaluated, and BRPP implemented those deemed to be technically,
economically, and operationally feasible.
Memorandum
September 5,2007
Page 4
Process control and OD performance have been improved. However,on the hardwood line,the D,
stage kappa factor has not been decreased as previously recommended and use of oxygen in the
extraction stage was discontinued. As a result,the total C102 charge on the hardwood line as of
May 2006 was only 5 percent less(not the predicted 27 percent less)than the C102 charge in 2000.
Similarly,on the softwood line,the D, stage kappa factor was not decreased as previously
recommended and the use of peroxide in the extraction stage was discontinued. As a result,the
total C1O2 charge on the softwood line as of May 2006 was 11 percent more(not the predicted 17
percent less)than the charge in 2000. While BRPP has noted the increased portion of higher
brightness pulps since 2001 as a reason for these changes, among other trade-offs cited,the Tech
Team continues to believe these changes are still viable and should remain as high certainty options
that should be pursued for further color reduction.
In preparation for renewal of their NPDES permit,BRPP retained Liebergott&Associates and
GL&V Pulp Group to analyze their fiber lines,review the implementation of the 2001
recommendations,and identify further options for color reduction. In the 2006 update to the
BEPER(GL&V 2006),the consultants repeated their recommendations for BRPP to reduce the
kappa factor in the D, stage of each bleach line and replace the C102 bleaching power with oxygen
and/or peroxide in the extraction stages. Increasing the operating temperature of the hydrogen
peroxide extraction stage to reduce effluent color also has been identified as a viable option during
this process.
In addition to reducing the color of the bleach plant effluents,use of a lower kappa factor means
BRPP would use less C1O2 to bleach,which will reduce the total chloride content of bleach plant
filtrates. Lower chloride content can also be expected to make it possible for the mill to reduce the
CRP purge flow and the color it contributes to the mill effluent. The Tech Team also believes that
lower chloride content will make it possible for the mill to recycle more filtrates within the
hardwood fiber line,notwithstanding BRPP concerns(e.g.,adverse effects to washing, carryover to
D stage,possible increased chemical usage and color). Further,it also may be possible to introduce
a portion of these hardwood filtrates to the BFRTM process,which to this point has been dedicated
solely to recovering softwood fiber line filtrates,and reduce the total colored filtrates discharged to
the sewer.
BFR Reliability Improvement
As reported during the Tech Team's February 2007 site visit,BRPP spent$1.5 million over the last
five years to improve the reliability of the BFRT"l system. These expenditures included replacing
tank liners for the three existing sand filters(now high grade stainless steel),replacing and/or
upgrading existing process piping with piping with improved metallurgy,installing a fourth(new)
multimedia filter,and installing a third ion exchange softener. With these improvements in more
reliable metallurgy and unit process redundancy,BRPP has increased the BFRTM closure rate from
73.7 percent in 2001 to 79.2 percent in 2006. Mill representatives stated that the present closure
rate of approximately 80 percent of the pine line bleach plant effluent represents the maximum
amount attainable without incurring unmanageable corrosion and scaling problems. Further
increases in closure and increased recycle rates above 80 percent may be possible but are likely to
be a difficult challenge because of the corrosion and scaling problems associated with current mill
bleaching filtrate chemistry. Scaling from hardness minerals accelerates above closure rates of 80
percent(Bodien,2007).
Memorandum
September 5,2007
Page 5
Improved Black Liquor Leak and Spill Control(BMPs)
BRPP continued efforts intended to improve its management of leaks,spills,and intentional
diversions of black liquor over the last six years. These efforts include:
• Interconnecting the pine line and hardwood line spill collection sumps so that
tankage in either line can be used interchangeably for spills;
• Repositioning sewer conductivity probes from sumps to in-line to more accurately
and reliably identify high conductivity wastewaters;
• Diverting up to one hour of total mill flow to off-line 1-million-gallon clarifier,
during high color releases,thus providing some equalizing of color discharge to
wastewater treatment plant;
• Prior to process line outages,improved prior planning for and capture of high-color
process liquors and black liquors and better managing their timed release to
treatment system;
• Continued operator training;and
• Implementing two hour testing for color at the W WTP with one hour testing during
outages or semi-annual shutdowns.
BRPP has stated that these improvements in BMPs have resulted in reduced color variability in
influent to the pritnarP treatment unit,which BRPP asserts is the best measure of color reduction
through in plant changes and 13MPs. Data provided by BRPP are presented in Table 2 and depict
Primary(total)Influent Statistics by year.
Table 2. Primary(Total)Influent Statistics,by year
can - Relative Percent Decrease
True Color Standard Standard Deviation from Baseline =
Obs/day) Deviation(SD) (SD/mean%) (2001)
2001 57,725 20,619 35.7% 0.00%
2002 54,780 17,195 31.4% 5.10%
2003 55,550 19,424 35.0% 3.77%
2004 49,466 18,786 38.0% 1 14.3%
2005 45,175 22,297 49.4% 21.7%
2006 38,454 14,015 36.40/. 33.4%
Primary effluent,which is the total load to secondary treatment, is directly measured with a
composite sampler. Data provided by BRPP and compiled by the Tech Team are presented in
Table 3 (below)showing the mean, standard deviation,and relative standard deviation(standard
deviation as a percent of the mean)by year,for the years 2001 to 2006. Prior to 2006,the acid
sewer mixed with other mill wastewater prior to the treatment system,resulting in"sewer generated
color." As of January 1,2006,the acid sewer was separated from the other mill sewers and now
enters the treatment plant after the primary clarifier("primary effluent'). The primary effluent
sampler is located downstream of the mixing point of the primary clarifier overflow and acid sewer.
Thus,the statistics for 2006 presented in Table 3,represent the mixture of primary effluent and
Memorandum
September 5,2007
Page 6
newly rerouted acid sewer,including any color generated from the mixing of the two streams.
Some portion of the color generation is immediately measurable in primary effluent samples taken
from the channel leading to the aeration basin. It is also likely that additional color generation from
this mixing occurs,after the primary effluent sampling location within the aeration basin of the
secondary treatment system,with additional time for any chemical reactions to proceed to
completion.
Table 3. Primary Effluent(Influent to Secondary Treatment)Statistics,by year
Mean Relative _ T
, r True Color Standard'° ;Standard Deviation e'-Number4daysa;_
(Ibalday) ;a beviafion,{S* (SD i0eart-"/o) >100,0001b1day
2001 62,008 19,561 31.5% 13
2002 59,956 18,680 31.2% 16
2003 59,646 18,468 31.0% 10
2004 65,206 26,674 40.9% 40
2005 63,838 24,158 37.8% 28
2006 65,512 25,427 38.8% 36
Table 3 presents the number of days for which the primary effluent color exceeded 100,000 lbs/day.
Comments received from BRPP assert that primary effluent is not the most appropriate measure of
progress. BRPP further stated that color loads to the treatment system have been reduced. In spite
of improvements made over the last permit term,including the acid sewer relocation,all clearly
made in good faith by BRPP,the Tech Team respectfully disagrees with these assertions. As
shown in Table 3,evaluation of available data does not appear to support BRPP's assertion that
there has been progress in reducing the frequency at which it experiences high color loads into the
secondary treatment system.
BRPP provided the Tech Team with notes describing mill events that were related to high primary
influent color in 2006(Blue Ridge 2007a). These events included,among others,planned mill
outages,unplanned outages,a CRP slurry tank overflow, and a release from the evaporator related
to an equipment failure. BRPP did not report one single event or one single type of event that
caused high color in the primary effluent. Further,not all high primary effluent color resulted in a
high final effluent discharge(e.g.,BRPP reports that color associated with high turbidity is
effectively removed in.the secondary treatment system). However,from analysis of the data,the
Tech Team concludes that the Canton mill can further reduce primary effluent loads through
continuing efforts to minimize unplanned spills and leaks and planned discharges of high-color
streams during fiber line disruptions. BRPP has reported some success through recent efforts in
detailed scheduling of planned outages,contingency planning for unplanned outages, and
continuing efforts to minimize process operation variability. The Tech Team wishes to
acknowledge these efforts and their importance. While clearly challenging,these efforts must be
further developed and consistently implemented to minimize high color discharge risks all agree are
associated with these fiber line disruptions,both planned and unplanned.
The Tech team continues to believe that the information presented in Table 3,among other
available performance data,holds valuable indicators and clues to the sources and solutions to
reducing the overall performance and variability in effluent color discharged to the Pigeon River.
On the other hand,BRPP has asserted that primary effluent information and other data collected
within the Mill's biological treatment system are not indicative of the facility's true performance.
Memorandum
September 5,2007
Page 7
BRPP believes that secondary effluent showed improvement,and data provided by BRPP are
presented in Table 4(below).
Table 4 Secondary Effluent Statistics,by year
-,mean=- >'. Rotative w
True Color �9tandard- Standard Deviation�' - Number days
Qbslday7; • Deviation(SD)- ' "(SD/mean=O >100,000 Iblday,
2001 42,676 10,925 25.6% 3
2002 41,166 9,928 24.1% 0
2003 44,627 11,043 24.7% 1
2004* 41,463 32,568 76.6% 4
2005 39,092 10,092 25.8% 0
2006 37,058 8,959 24.2% 2
* -2004 data were affected by historic floods in September 2004
Ozone/Chlorine Dioxide Stage for the Hardwood Bleach Line
BRPP engaged the Pulp and Paper Research Institute of Canada(PAPRICAN)to investigate
potential modifications to the hardwood pulp bleaching process and determine effluent color
reduction that would result from these modifications. The PAPRICAN report entitled"Bleaching
Evaluation for Effluent Color Reduction"(Audet et al,2003)was provided to the Tech Team in
early 2007.
PAPRICAN stated that their objective was to evaluate whether modifications to the hardwood
bleaching sequence at Canton specified by Mr. Johnnie Pearson(BRPP process engineer)could
"generate effluents with a color reduction target of 25 percent."
BRPP provided PAPRICAN with oxygen delignified hardwood pulp. PAPRICAN bleached this
pulp in their laboratory,investigating various combinations of chlorine dioxide, ozone and
hydrogen peroxide.
The report concluded that ozone could replace some of the chlorine dioxide used in bleaching
(known as a"ZD"stage),while producing pulp of equal or slightly better quality than the control
sequence. This is consistent with the open literature on ozone/chlorine dioxide combinations.
PAPRICAN also concluded that replacing some of the chlorine dioxide in bleaching with ozone
increased the color in the bleach plant effluents. This contradicts the experience reported by
Domtar(formerly E.B.Eddy)at their Espanola mill,where a dramatic reduction in effluent color
was observed when that mill installed an ozone pulp bleaching system(Munro and Griffiths,2000).
The Tech Team notes that the Espanola mill has no color discharge limits and it installed ozone to
reduce its bleaching costs.
The Tech Team found significant deficiencies in PAPRICAN's analysis of the laboratory results.
The 2006 Liebergott/GLV report agrees with this finding. The concentration of color in the
effluent for each stage was added to obtain the total concentration for each tested bleach sequence.
This approach neglects two points:
1) Mixing effluents produces reactions that may increase or decrease the
concentration of color in the combined effluent.
Memorandum
September 5,2007
Page 8
2) Filtrate volume affects the measured concentration(e.g.,lower volume will
concentrate the filtrate to a higher color). PAPRICAN's report does not present the
filtrate volumes,so a comparison of concentrations may be misleading.
Liebergott, et. al.,reviewed the PAPRICAN report and identified the same deficiency with regard
to effluent mixing. They also noted that the quantity of chlorine dioxide added to the ZD stage was
too high,which would result in higher effluent color(GL&V 2006,p 18).
In response to EPA's question about how the PAPRICAN results were used to estimate impacts on
final effluent color,BRPP responded:
The pilot study reactors were batch and not continuous and did not include
filtrate recycle. The pilot studies included bleach stage filtrate color
concentrations, but there was no filtrate flow rate data from which to calculate a
production-normalized filtrate color mass. For these reasons, we did not attempt
to calculate secondary effluent color impacts using pilot study data for individual
bleach plant color streams.
In the PAPRICAN study, the individual bleach stage colors were compared
directly and in total. By both means, the color of the ZD stage was higher than
the baseline DEoD. It is also well known that when individual bleach stage
filtrates are mixed the resultant effluent color is very difficult to predict (Blue
Ridge,2007a)
The Tech Team concludes that BRPP's investigation of the potential for hardwood pulp ozone
bleaching to reduce the mill's effluent color was inadequate. Notwithstanding the above-noted
inadequacies and upon reflection,the Tech Team believes that further laboratory trials appear not to
be necessary because this application of ZD technology may not be appropriate for this mill'at this
time. Therefore,ZD technology is considered a technology option of lowest certainty for
application at this mill at this time.
Second Stage Oxygen Delignification for the Pine Bleach Line
BRPP contracted with Andritz/Ahlstrom("Andritz")to study, among other things,the addition of
an additional stage of oxygen delignifrcation in the pine(softwood)bleach line at Canton. Andritz,
a well established supplier of pulping and bleaching technology and equipment,maintains the
Pmyn's Island Technical Center,which conducted the tests. The Andritz report, entitled,
"Laboratory CKand Lo-Solids Cooking with O-Do-Eop-D Bleaching Sequences on Softwood
Furnish from Blue Ridge Paper,Blue Ridge,NC"(Andritz Ahlstrom'Sales 2001)was provided to
the Tech Team in early 2007.
The text of the Andritz report states that"BRPP requested the sales department of Andritz-
Ahlstrom to investigate alternative methods of pulping and bleaching that would reduce their waste
products and improve the quality of their hardwood and softwood paper products." As indicated by
this statement of work,testing of second-stage oxygen delignifrcation was a minor part of the work
Andritz did for BRPP. Further,very few of the data in the report are useful for analysis of effluent
color improvement.
Memorandum
September 5, 2007
Page 9
Specifically,Andritz compared single stage and two stage oxygen delignification of samples of
pulp provided by the BRPP mill. The tests showed that a second stage of oxygen delignification
could reduce the kappa number of the unbleached pine pulp by 22 percent.
Andritz did not bleach the pulp after the two stage oxygen delignification,so the report provides no
information on the impact of the additional oxygen delignification stage on effluent characteristics,
including color.
In response to EPA's question about how the Andritz results were used to estimate impacts on final
effluent color,BRPP responded:
*The Andrtiz-Ahlstrom study of second stage 02 for pine showed delignification
ranging from 42.7 percent for the single stage, 48.6 percent for the 0-0 and 55
percent for the 00 stage. With improvements that BRPP made on the existing
single stage pine 02, the deliginification increased from 40 percent to 45 percent.
With BFR in place on the pine fberline, the effluent color reduction from this
improvement in 02 deliginification has been very difficult to identify. The
inability to quantify the effect on effluent color of improved 02 deliginification
made the capital cost to install a second stage 02 on pine not economically
feasible. (Blue Ridge,2007a)
BRPP's analysis of the benefits of adding a second oxygen delignification stage to the pine line is
incomplete. BRPP's consultants note that even with the percent delignification currently achieved
on the pine line, a second stage could achieve an additional 20 to 25 percent delignification(see
GL&V 2006,p 118).
By making a 20 to 25 percent reduction in the kappa number of the pine pulp before bleaching,20
to 25 percent of the colored material currently discharged from bleaching to the BFRTM would be
recovered and burned in the mill's recovery boiler. In addition, the quantity of chlorine dioxide and
caustic required in bleaching would be reduced by approximately 20 to 25 percent,reducing the
load on the BFRTM system. This would allow an increased proportion of the bleach filtrates to be
recycled through the BFRTM system. BRPP believes that a kappa number decrease or
delignification efficiency increase would not equate to a similar decrease in color. Although with
the information available it is not possible to rigorously calculate the benefit of this change,the
Tech Team would expect on the order of 1000 lbs/day reduction in bleach plant color discharge.
In 2001,the Tech Team recommended that BRPP conduct a detailed study to develop an
engineering design leading to installing an additional oxygen delignification stage for the nine
(softwood)pulping/bleaching line.The Tech Team estimated that this process change would reduce
effluent color by 1,100 to 1,400 lbs/day. Liebergott, et. al.,(GL&V 2006)estimated that
implementing a second oxygen delignification stage on the pine fiber line would reduce color
discharge by 1142 lbs/day.They estimated the capital cost of adding an additional oxygen
delignification stage would be in the order of$2 to $3 million and the chemical cost savings
approximately$1.2 million/year,providing a reasonable payback. Mill staff has mentioned a three
year pay-back,which is consistent with this estimate. However,the mill has further asserted from
their more recent estimates that the cost could be$3 to 5 million owing,presumably at least in part,
to increases in the general prices of stainless steels(not as specifically fabricated and estimated for
this technology). The BRPP oxygen delignification systems were installed in 1993. Since that
time,the use of two-stage oxygen delignification rather than the traditional single-stage systems has
become well established in the industry because it normally further reduces mill operating costs.
BRPP believes that the previous evaluation of adding a second stage oxygen delignification stage
Memorandum
September 5,2007
Page 10
should be repeated in part because of differences cited in delignification efficiencies, and concerns
for possibly lower-than-estimated color reductions. Notwithstanding these concerns,the Tech
Team continues to believe this technology has progressed beyond being considered"reasonable
certainty"in 2001 to"highest certainty"at this time. Therefore,the Tech Team again recommends
that priority should be given to a detailed evaluation for identifying necessary adjustments to
upstream pulp digestion(e.g.,kappa number targets),downstream bleaching(e.g.,bleaching
chemical usage rates)and brightness/strength parameters, designing and costing, and refining color
reduction projections. In order to satisfy BRPP's recent concerns,an update from the results of this
work may need to be shared with the TRW,prior to being implemented at the earliest possible date.
Andritz also evaluated modifications to the mill cooking process. These would require complete
replacement of the digester systems at Canton,which would cost(at least)several tens of millions
of dollars, if they are feasible at all within the mill's space constraints. The Tech Team concluded
that while theoretically possible,modifications to the mill cooking process do not merit further
analysis at this time.
Treatment of CRP Purge Stream for Color Removal
The Chloride Removal Process(CRP)purge is a low flow,highly concentrated stream. Color is
typically 41,000 platinum cobalt units(pcu) in a stream that discharges at 10 gpm(15,000 gal/day
or 3 tank trucks/day).BRPP reported that in 2006 the CRP purge contributed approximately 8,745
lb/day(23 percent)of the total mill color loading to the treatment system but only 0.05 percent of
the discharge flow. By comparison,in 2001 the CRP contributed approximately 5,000 lb/day(13
percent)to the treatment system loading(EPA Tech Team,2001). Thus,in the last five years there
has been a significant increase in color contributed by the CRP,both in lb/day and in percent of
total mill load. BRPP reported on its investigations into technologies for reducing CRP purge color
in its March 2005 report,"Chloride Removal Process(CRP) Color Reduction Technology
Assessment" (Blue Ridge 2005). In this report,BRPP points out that the CRP purge is a very
concentrated material(360,000 mg/L or 36 percent total dissolved solids)that is discharged to
sewer at 165'F. Handling the material is difficult because it will crystallize as it cools. Also,it is
quite corrosive due to the high chloride content.
As described in their 2005 report,BRPP assessed more than nine alternatives for disposal or
treatment of the CRP purge and concluded that none was technically feasible. Although consistent
performance was not demonstrated, C1O2 bleaching was a low cost,potentially effective means of
reducing the CRP purge color prior to introduction to the treatment system. BRPP found that C1O2
bleaching could remove 75 to 90 percent of the color in the CRP purge stream,with some concern
about consistency. This is a significant reduction in a now prominent source of color.
Although CRP purge is a relatively low flow stream,BRPP estimated that off-site solidification and
land disposal would cost more than$3.6 million/year,which does not include the additional cost of
loading facility infrastructure.
During the Tech Team's 2007 mill visit,BRPP stated that there was no apparent decrease in
secondary effluent color when CRP wastewater was not flowing into the treatment system. In
response to a Tech Team request,BRPP presented data correlating the days on which the CRP
process was shut down with secondary effluent discharge color,for August 2006 to January 2007
(Blue Ridge 2007a). During this period,there were nine widely dispersed pairs of days when the
CRP process was down for part of the day. BRPP believes that CRP color is removed in the
secondary treatment system,based on their analysis of variance(ANOVA)comparing"down"days
with CRP operating days and"general observations."
Memorandum
September 5,2007
Page 11
In response to a Tech Team follow-up request for clarification,BRPP later asserted that these data
allowed no definitive conclusion,and that there is no relationship of presence or absence of CRP
purge to secondary effluent color based upon general observations. The Tech Team does not agree
with this belief or rationale for the following reasons. First,with one exception,the periods of CRP
shutdown are only a day or so each,which is insufficient time for the W WTP to stabilize with the
change in raw effluent characteristics. Second,the times of shutdown and startup of the CRP do
not correspond with the effluent sampling times in the mill system,so that there is no direct
correlation in time with the effluent sample.Further,the fact that all CRP"off'days are in pairs
suggests that CRP was down for a period that spanned parts of two mill sampling days.
The simple mathematical average of treated effluent color discharge during"CRP off'days for
August to December 2006 is 39,995 lbs/day. The average when the CRP process was operating is
36,958 lbs/day. This suggests that the presence of the CRP purge stream in the wastewater
treatment system causes a reduction in effluent color. This defies common sense. Also,the
difference in effluent color between the"CRP on"and"CRP off'days is 7 percent,while day to
day variations in color discharge values are frequently over 20 percent,tending to subsume and
confound any analysis of the impact of changes in the CRP purge stream. BRPP asserted that color
performance data from this period likely varied due only to normal day-to-day performance
variation.
In short,because there are so few consecutive days without CRP purge,the data from the period
August to December 2006 are not sufficient for establishing the extent to which the CRP purge
contributes to final effluent color. In order to explore this further,BRPP could conduct a full-scale
mill trial in which the CRP purge is excluded from the wastewater treatment plant for a period of at
least three weeks. This can be achieved in two alternative ways, or perhaps a combination of both.
The first is to shut off the CRP purge for at least three weeks, and allow the chloride concentration
to slowly build up in the recovery system. Early literature published by Champion on the BFRTm
process showed that the response of the chloride and potassium content in the mill's white liquor to
operation of the CRP was very slow,requiring a month or more to show significant changes. This
is consistent with experience in other mills,and is due to the fact that the quantity of chlorides and
potassium removed by CRP is small relative to the inventory in the mill's black,green and white
liquor system. The second approach would be to store the CRP purge stream, either in the spare
clarifier or off site for a period of at least three weeks. The impact of excluding the CRP purge
from the wastewater treatment plant would be observed by routine analysis of effluent color. If the
effluent color without the CRP purge is unchanged,BRPP could conclude that the CRP purge
contributes little to the final effluent color. In this case,further research into reducing CRP color
would be a low priority. On the other hand, if the effluent color without the CRP purge decreased,
BRPP could conclude that the CRP purge contributes to the final effluent color and further research
into reducing CRP color should be undertaken.
BRPP mill staff asserted that a full-scale shut off of CRP purge to the treatment system for such a
period of time(e.g.,at least three weeks)would not be operationally feasible,but offered no
specific facts or reasons why this would be the case. Nonetheless, given BRPP's concerns for
operational feasibility at full scale,the Tech Team is open to another reasonable approach that
could be devised. Such an approach would need to gather the data necessary over a sufficient
period of time to better identify and quantify the underlying color loads to the treatment system,
both with and without the purge from CRP. Moreover,it would be important to identify and
quantify to the extent possible any changes in the downstream color generation and removal
processes that are occurring with the current acid sewer introduction point just upstream of the
aeration basin in the secondary activated sludge wastewater treatment system.
Memorandum
September 5,2007
Page 12
BRPP also could investigate approaches to prevent color from accumulating in the CRP. The
source of the color in the CRP purge stream is carryover of black liquor particles in the direct
contact evaporators(DCE)in the recovery boiler systems. This carryover could be eliminated if the
two traditionally designed recovery boilers(which are 34 and 42 years old)were replaced by one
modem boiler. This would probably represent a capital cost of over$100 million, and would very
substantially reduce the energy costs at the mill,since today's recovery boilers are much more
efficient than the DCE/recovery boiler systems of the vintage installed at Canton. Analysis of all
the economics and long term life of the mill would be necessary to evaluate such a major
investment.
On a more modest level,it may be feasible to reduce black liquor carry over by adjusting operating
conditions in the existing DCE's. The Tech Team is not aware of any experience with this in other
mills, or research,but the situation at Canton strongly suggests that at least some investigation and
trials of modifications to the operations is warranted. Any of the measures discussed previously in
this document to reduce C102 use could reduce the quantity of chloride to be removed in the CRP
purge stream. If this is reduced,the color discharge also would be reduced.
BRPP should also further investigate C1O2 treatment and other treatment of color in the segregated
CRP purge stream and other approaches for excluding the CRP purge stream from the mill
discharge.
Additional Color Reduction Strategies
The Tech Team has identified the following color reduction activities that should be for improving
the color removed by the Canton Mill wastewater treatment plant:
• Investigate in further depth color formation when acid wastewater is mixed with
mill wastewater in the current configuration,with and without the CRP purge, and
identify techniques to reduce this effect;
• Maintain addition of polymer and/or other treatment chemicals to the high-color
wastewater diverted to the extra primary clarifier,and investigate improving
equalization and pretreatment performance; and
• Further investigate adding polymer and/or other treatment chemicals to aeration
basin mixed liquor prior to introduction to the secondary clarifiers.
These strategies are discussed below.
Color Formation When Acid Wastewater Is Mixed With Mill Wastewater
BRPP provided the Tech Team with daily color data for the"Low Lift"(mill sewer),acid sewer,
and primary effluent sampled after the acid sewer is added in the discharge channel of the primary
clarifier leading to the aeration basin of the secondary activated sludge biological treatment system.
Figure 1 (below)presents the total mill color calculated by adding the mill sewer lb/day to the acid
sewer lb/day(the lower(blue line) on the figure). For comparison,the primary effluent sampled
after the acid sewer is introduced is also shown on the figure(the upper red line on the figure).
Thus,the figure depicts the impact of mixing the acid sewer with the rest of the mill effluent'.
3 The color of the mill sewer wastewater maybe reduced somewhat by treatment in the primary clarifier. This color
reduction is neglected in the calculated untreated color,so Figure 1 is probably a slight underestimate of the impact of
mixing the acid sewer with the rest of the mill effluent.
Memorandum
September 5, 2007
Page 13
150,000
Total untreated color,before and after mixing,2006 (lbs/day)
125,000
100,000
75,000
50,000
25,000
1Jan 31Jan 141ar 314.r 30-Apr 30-May 29Jun 294ul 28-Aug 27Sep 21-00 2641 v 26-0ec
Figure 1: Impact of Mixing Acid Sewer and Mill
The Tech Team observed:
• The quantity of color formed on mixing the acid sewer and mill sewer is striking.
On average,the quantity of color formed is 78 percent of the total color in the two
streams. In other words,the simple mixing of these streams forms nearly half the
total color discharge from the mill.
The variation in quantity of color formed by mixing the effluent streams from day-
to-day is dramatic, as is evident in the graphs. When expressed as a percentage of
primary input color,the value ranges from essentially zero on some days to a
maximum of 480 percent.
• There is no obvious correlation between the values on any one day,or series of
days close to one another;however, statistical analysis has not been attempted.
Notwithstanding these observations,the Tech Team understands and appreciates BRPP's efforts
during the last permit term to reduce color formation through the acid sewer relocation project. The
Tech Team suggests that BRPP build on this project and the above color reduction strategies to
better understand the mechanisms of and the follow-up strategies for reducing color formation in
the current sewer configuration. BRPP could investigate, among other things,how sulfides
introduced by leaks,spills, and unplanned discharges from white and green liquor systems into the
mill wastewater contribute to effluent color. In particular,BRPP could investigate whether sulfides
in the mill wastewater contribute to color formation when acid sewer is added to mill effluent prior
to biological treatment. Approaches to this investigation should include:
• Laboratory experiments on effluent with varying degrees of sodium sulfide
addition; and
• Daily measurement of the sulfide content of the effluent from the primary clarifier
influent at the low lift pump sampling station for a period of at last three months to
determine correlation of sulfide concentration with color formation.
In most mills, losses of sulfides can be reduced. Well known sources include imbalance in the
weak wash system, as well as spills of white,black, and green liquor. BRPP should further
investigate losses from the green and white liquor systems and technologies that can reduce these
losses.
Memorandum
September 5,2007
Page 14
Improving WWTP Performance Using Chemicals
Several bleached kraft mills around the world have successfully reduced color to well below 5
lb/ton pulp by installing tertiary treatment systems. These systems use polymers and other
chemicals to precipitate color bodies,then remove the,precipitate in tertiary clarifiers. One of the
newest of these mills is Celco in Valdivia, Chile. This mill has excellent in-plant effluent control, a
conventional secondary wastewater treatment plant,followed by a small tertiary system.Effluent
from this mill is depicted in Figure 2(below)beside the effluent from BRPP.
ah
u r
f
Figure 2: Comparison of BRPP Canton Mill Effluent(February 2007) and Valdivia Mill
Effluent(November 2005)
The Tech Team notes that treatment systems like the one operated at Valdivia have a capital cost in
the tens of millions of dollars,and can create issues with solid waste disposal. BRPP also
expressed the opinion that this comparison was not appropriate. However, some more modest
control measures using variations of this technology also exist. For example, since March 2004,
Glatfelter Inc., Spring Grove,PA has supplemented its in-plant color control strategies by using a
commercially available polyaluminum chloride polymer to enhance color removal in the
wastewater treatment plant. The polymer is added in the discharge flume from the aeration basin,
just upstream of the mixing box feeding the four secondary clarifiers. The polymer reduces pH and
complexes with organic compounds that produce color. Colored material settles out with the
secondary sludge which is held in the former stabilization basin before dewatering. Glatfelter has
not reported sludge dewatering problems.Polymer used for additional color removal reportedly
costs on the order of$2,000 to$3,000/day. Moreover,neither the Valdivia mill nor the Glatfelter
mill employs the BFRTM system as applied at BRPP.
As further in-plant process changes and improved BMPs are implemented and color loads are
reduced in magnitude(longterm average and variability) and change chemical composition,BRPP
could conduct further laboratory trials of commercially available polymers and other wastewater
treatment chemicals. In some cases this may entail repeats of previous tests, for example for
polyamine,but under potentially different circumstances and wastewater chemistry than previously
tested. In addition to those previously evaluated,it could be helpful to assess adding one or more of
these chemicals upstream of the secondary clarifiers. During these trials BRPP would determine
the potential for reducing color discharge and estimate the costs. Such analysis would consider
Memorandum
September 5,2007
Page 15
seasonal use of the additive chemicals(i.e., adding the chemicals during periods when river flow is
low and the mill discharge has the greatest impact on the river color). The Tech Team
acknowledges BRPP concerns regarding significant issues that may occur,such as possible effluent
toxicity, difficulty in sludge dewatering, and cost.
If the laboratory trials are successful,BRPP could institute a full-scale trial for one month,and
report results to the TRW.
Pretreatment of Diverted High-Color Wastewater
BRPP currently diverts up to one hour of total mill flow at current flow rates to their off-line 1-
million-gallon clarifier,during high color releases. The purpose of this flow diversion is to equalize
color contributions to the wastewater treatment plant. Diverted high color wastewater is batch
pretreated by polyamine and returned to the wastewater treatment system. The Tech Team believes
that BRPP should further investigate options for increasing the performance of equalization and
pretreatment capacity beyond that which currently exists. This would probably be an important
contribution to both further reducing color variability,an increasingly higher priority endeavor as
long term averages are reduced,and to removing color before being introduced into the secondary
activated sludge biological treatment system.
Summary: Additional Identified Color Load and Variability Reducing Activities that BRPP
can Pursue in the Next Five Years
Since installing the BFR process and other improvements(OD,BMPs, etc.),BRPP is to be
commended for the important progress in reducing their annual average color discharges,which
needs to be continued. However,high color discharges continue to be experienced for short periods
(e.g., daily) and,these discharges become more evident as the annual average discharge is reduced.
When these discharges coincide with periods of low river flow(typically in the late summer),they
can contribute to an elevation in river color that could be noticeable to the citizens who use the
river. Thus,reducing the impacts of the peaks in color discharged from the mill requires not just
reducing the annual average color discharged but also reducing variability measured by the daily
color discharges. Mill process changes that reduce wastewater color are generally preferred to end-
of-pipe treatment because they may have lower capital costs and may benefit the mill by reducing
operating costs and improving process efficiencies. However,mills in environmental regulatory
jurisdictions with severe restrictions on their color discharges have been required to implement end-
of-pipe color removal technologies since the 1970's. Although the Tech Team continues to
maintain the highest priority for in-mill improvements such as process changes and optimization,
increased black liquor recovery and further improvements in BMPs, external color removal
technologies should continue to be considered carefully in the mix of options for further controlling
the color of BRPP's discharges. These color removal technologies include,with first priority,
treatment of segregated low-volume concentrated wastestreams(e.g.,the CRP purge)and thereafter
end-of-pipe wastewater treatment.
The Tech Team identified strategies focused primarily on in-mill process improvements but also
including color treatment that BRPP can use to further reduce its effluent color discharges. These
strategies are summarized below.
• Continue to improve the performance of BMPs to further substantially reduce and
ultimately eliminate discharges of highly-colored wastewaters directly to the
wastewater treatment system through further improvements in-
Memorandum
September 5, 2007
Page 16
o managing and controlling planned and unplanned releases of highly
colored process liquors through regular mill staff meetings
o interconnected collection sump capacity within the mill available to both
fiber lines for capture and recovery of leaks,spills,and planned diversions
of black liquor and other highly-colored wastewater
o increasing use of short-term testing to supplement advanced real-time
process monitoring,rapid communication among mill staff, identifying and
immediately repairing failed equipment/parts,regular operator training,
and
o moving forward with the planned CRP sump and containment project to
eliminate unplanned releases of this highly-colored material to the mill
sewer.
• On the pine bleaching line, implement the use of peroxide fortification of the Eo
stage and decrease the target kappa factor as recommended in BEPER 2001 and by
Liebergott/GL&V 2006. Evaluate the use of high temperature for the peroxide-
fortified extraction stage.
• On the hardwood bleaching line,implement the use of oxygen and peroxide
fortification of the E stage and decrease the target kappa factor as recommended in
BEPER 2001 and 2006. Evaluate the use of high temperature for the peroxide-
fortified extraction stage.
• Complete an expedited and detailed evaluation of and install an additional oxygen
delignification(OD)stage for the pine pulping/bleaching line.
• As C102use is reduced through second stage OD on the pine line and other options
on both fiber lines,further investigate increasing the amount of hardwood filtrates
recycled and investigate introducing for the first time some of these hardwood
filtrates to the BFRTm system, and thus potentially further reduce the purge rate
from the CRP,now a significant portion of the total color discharged from the mill
processes.
• Continue to evaluate the impact of the CRP purge on treated effluent color to
determine if the CRP color,now a significant portion of the total color discharged
from the mill processes,is removed by the treatment system. Such an effort may
take special efforts to accomplish and for a meaningful period of time. If CRP
color is not removed in the treatment system,investigate approaches to prevent
color from accumulating in the CRP, such as reducing black liquor carryover by
further adjusting operating conditions in the direct contact evaporators. Also,if
CRP color is not removed in the treatment system,further investigate C102
pretreatment of the CRP purge to reduce its color prior to being introduced into the
mill wastewater treatment system.
• If the CRP purge color is found not to be removed in secondary treatment,avoid
releasing the CRP purge during periods of low stream flow(or truck it off site);
• Continue to investigate and implement strategies for improving color removed by
the Canton Mill wastewater treatment plant:
Memorandum
September 5,2007
Page 17
— Further analyze color formation when acid wastewater is mixed with mill
wastewater in the current configuration,particularly in the activated sludge
aeration basin, and identify other techniques,such as minimizing sulfide
releases to the mill sewer from white and green liquor leaks,spills,and/or
diversions,to reduce this effect
— Maintain or further increase the performance for pretreating highly-colored
wastewaters prior to introducing them to the wastewater treatment system,
including further optimizing adding polymer and other pretreatment
chemicals to the highly-colored wastewater diverted to the extra primary
clarifier and/or any additional facilities that may be provided;
— Further investigate treating total mill biological system effluent prior to
discharge using polymers or other wastewater treatment chemicals
upstream of the secondary clarifiers,particularly during periods of high
influent color and/or low river flow;and
• Curtail pulp production during periods of low stream flow;this should continue to
be considered an option of last resort,given that best performance has been noted
by the mill to be during extended periods of steady production and greater risk of
elevated color during process shutdown and startup. See Low Flow Contingency
Plan,December 1, 1998.
Suggested Final Effluent Color Limits for Blue Ridge Pulp and Paper
Substantial and commendable progress has been made to date by BRPP through expenditures for
additional and improved process and related management practices and treatment technologies
which have reduced the long term average color discharges. Some of these technologies and
practices have been in addition to those identified in the previous Tech Team/TRW reviews. This
progress in reducing the annual average of color discharges has made increasingly important and
apparent the importance of reducing day-to-day effluent variability along with the other
recommendations of the Tech Team going forward from this critical point in time. Therefore,the
Tech Team recommends to the TRW and the North Carolina Department of Environment and
Natural Resources that the existing permit limits be expanded to include a maximum day discharge
limit, and not rely solely on the annual average and maximum 30-day average limits to protect the
quality of the Pigeon River. Daily maximum and maximum 30-day average permit limits would
enhance the consistency of day-to-day in-stream water quality,and would be consistent with limits
for other parameters controlled in the present BRPP permit(BOD5,TSS,AOX,etc.). The Tech
Team also believes that these effluent limitations for color should continue to be applied
unambiguously at the end-of-pipe discharge of the mill to the river. However,the current permit is
not clear that this is the point of application of effluent limits for color. See Table A(1),which
prescribes end-of-pipe monitoring,but no color limits are included. See the TRW's 2001
Memorandum, at item no. 8. BRPP has commented that inclusion of a daily maximum end-of-pipe
effluent limitation for color would lead to administrative efforts focusing on any violations of such
a daily maximum limit that may occur, and thus diverting limited mill staff and resources away
from executing process controls and BMPs and wastewater treatment measures. The Tech Team's
recommended range of end-of-pipe permit limits and the derivation of these limits are presented in
Table 5, and discussed,below.
Memorandum
September 5, 2007
Page 18
Table 5. Tech Team Recommended Range of End-of-Pipe Color Permit Limits
Limtt - Rangeo[12ecommeniled Limits(1 /day)
Annual Average 32,000 to 37,000
30-day(Monthly)Average 44,800 to 51,800
Daily Maximum 83,840 to.96,940
Basis for Recommended Annual Average for Color
BRPP provided daily measurements secondary effluent color(lb/day)for every day in 2006.
Examination of the daily measurements for 2006 identified two days(July 7 and 8)with measured
discharge greater than 100,000 lb/day. BRPP reported that this elevated color discharge resulted
from"CRP slurry tank overflow for approximately 20 minutes,the first time the mill experienced
this type of event." Because these discharges were so high and from a unique source,they were
omitted from the calculation of the annual average. The.annual average for 2006,without July 7
and July 8, is 36,695 lb/day,which rounds to 37,000 lb/day.
• This annual average load is less than the 39,000 lb/day suggested by BRPP.
However,it is based on the mill's 2006 performance and does not include any
reductions that may be achieved by the process changes outlined in this
memorandum. As a result of analysis of the BRPP mill discharges by the Tech
Team and others in support of the 2001 permit,the permit's interim color goal was
32,000 lb/day with a range up to 39,000 lb/day.
• As described earlier in this memorandum,the Tech Team recommends that BRPP
implement several key process improvements and investigate others in order to
reduce the annual average toward the interim goal of 32,000 lb/day,which the Tech
Team recommends be carried forward from the 2001 permit as the goal for this
permit term.
Basis for Suggested Daily Maximum and 30-Day Average Limits for Color
The statistical analysis used for the development of EPA's Cluster Rules is documented in
Statistical Support Document for the Pulp and Paper Industry: Subpart B (EPA, 1997). This
document describes EPA's development of,among other things,the variability factors that were
used to calculate NSPS for BOD5 for the Bleached Papergrade Kraft and Soda(BPK) subcategory.
These variability factors are reproduced in Table 6.
Table 6. Bleached Papergrade Kraft NSPS Variability Factors
Variability Factors
Analyte - '1-I)ay 30-Day(Monthly)i
BOD, 2.62 1.4
Source: U.S.EPA 1997. Table 2-4.
The BOD5 variability factors shown above were developed using daily monitoring data for the best
performing(in terms of production normalized BOD5 load)BPK mills. The monitoring data
represent the effluent from well-operated wastewater treatment systems. The calculated variability
factors account for the autocorrelation of the daily loads and the log-normal distribution of the
measurements.
Memorandum
September 5,2007
Page 19
Notwithstanding BRPP's comments to the contrary,the Tech Team believes that it is reasonable
and appropriate to use the variability factors developed for the BPK NSPS for BOD5 to calculate
BRPP daily maximum and 30-day average permit limits for color because:
• The factors were developed using monitoring data from BPK mills with well-
operated secondary treatment and BRPP is a BPK mill with well-operated
secondary treatment;
• Color, like BOD5,is monitored at the effluent of the secondary treatment plant;and
• Color,like BOD5,is removed from effluent by a combination of biodegradation
and absorption/adsorption to biomass.
Applying the BOD5 variability factors to the 2006 annual average daily color discharge results in
the following daily maximum and 30-day average limits:
Daily Maximum: 37,000 lb/day x 2.62=96,940 lb/day
30-Day(Monthly)Average: 37,000 lb/day x 1.40=51,8001b/day
The Daily Maximum and 30-day (Monthly)average limits for the long term average goal of 32,000
lb/day are derived in the same manner.
Comparing the 2006 effluent monitoring data to these limits finds that two days exceeded the daily
maximum,July 7(104,504 lb),and July 8 (101,223 lb),during the unexpected CRP tank overflow.
Further,the 30-day(Monthly)average(51,800 lb/day),as calculated for compliance purposes in the
current permit,was not exceeded for 2006.
Comparison to Glatfelter Permit Limits
The Glatfelter,Inc,mill in Spring Grove,PA(NDPES permit number PA0008869), like BRPP,is
an older bleached kraft mill that discharges to a small,color-limited stream. As noted previously,
however,Glatfelter does not use the BFRTm process that is unique to BRPP. As described in PA
DEP's Water Quality Protection Report(Furjanic,2007), during the summer,the Glatfelter mill is
permitted to discharge 140 mg/L color(average monthly) and 280 mg/L(daily maximum). Winter
discharge limits are more stringent. The Glatfelter mill's average daily discharge is 11.9 MGD and
it produces an average of 1268 U.S. tons of paper per day. BRPP produces approximately 1640 US
tons per day,with a greater proportion of softwood to hardwood processed than at the Glatfelter
mill, and with a daily total wastewater flow of approximately 26 MGD. Therefore,the Tech Team
calculates that the recommended range of end-of-pipe permit limits for BRPP (32,000 to 37,000
lb/day annual average;44,800 to 51,800 lb/day 30-day average; 83,840 to 96,940 lb/day daily
maximum) are less stringent than the Glatfelter permit limits expressed on the same production
normalized basis (either lbs/metric ton pulp or finished paper).
Memorandum
September 5, 2007
Page 20
References
Andritz Ahlstrom Sales. 2001. Pmyn's Island Technical Center Report 2001-068 Part 1.
Laboratory and Lo-Solids Cooking with O-Do-Eop-D Bleaching Sequences on Softwood Fumish
from Blue Ridge Paper,Blue Ridge,NC. Part 1 Softwood Results. Part 2. Hardwood Results.
(December 6).
Audet,Andre,Michel Faubert,Zhi-Hua Jiang and Barbara van Lierop,PAPRICAN. 2003.
Technical Service Contract Report TSC-7447. Bleaching Evaluation for Effluent Colour
Reduction. Prepared for Blue Ridge Paper Products,Inc., Canton,NC. October.
Blue Ridge Paper Products,Inc. 2005. Chloride Removal Process(CRP) Color Reduction
Technology Assessment. (March).
Blue Ridge Paper Products,Inc. 2006a. Color Compliance Report: Canton Mill. (May).
Blue Ridge Paper Products,Inc. 2006b. Blue Ridge Paper response to questions from the
Technology Review Workgroup(TR I19 that were e-mailed on 28 Nov 2006. (December 19).
Blue Ridge Paper Products,Inc. 2007a.Response to additional questions for BRPP about data
provided to TRW Don Anderson e-mail dated 5 March 2007(March 15)
Blue Ridge Paper Products,Inc. 2007b. March 19, 2007 Additional Data Required from BRPP
(March 28).
Bodien,Danforth G. 2007. Site Visit Report, Bhie Ridge Paper Products, Canton, North Carolina.
(April 2007)
EPA Tech Team. 2001. Memorandum to Technology Review Workgroup. "Additional Color
Removal Technologies and Their Economic Impacts on Blue Ridge Paper Products,Canton,NC."
(July 25,2001).
Furjanic, Sean M. 2007. Water Quality Protection Report,P.H. Glatfelter Company, Spring
Grove Borough and Jackson Township,York County,for the Renewal of NPDES Permit No.PA
0008869.
PADEP Southcentral Regional Office(draft,February 2007).
GL&V Pulp Group,Inc. and Liebergott&Associates Consulting. 2001. Bleach Environmental
Process Evaluation and Report. Prepared for Blue Ridge Paper Products, Inc. and Clean Water
Fund of North Carolina. (June 8)
GL&V Pulp Group,Inc.,Liebergott&Associates Consulting. 2006. 2006 Update: Bleach
Environmental Process Evaluation and Report. Version containing manufacturer's proprietary
information.Prepared for Blue Ridge Paper Products,Inc. (July 7,2006)
Munro,Fred and John Griffiths. 2000. Operating Experience with an Ozone-based ECF Bleaching
Sequence,Proc.International Bleaching Conference,Halifax, Canada, 2000. TAPPI Press.
Technology Review Workgroup,2001.Memorandum to North Carolina Division of Water
Quality and the NC Environmental Management Commission's NPDES Sub-Committee.
Memorandum
September 5,2007
Page 21
"Additional Color Removal Opportunities, Blue Ridge Paper's (BRP) Canton,NC
Bleached Kraft Paper Mill, 2001 NPDES Permit Renewal." (August 3,2001).
U.S. EPA, 1997. Statistical Support Document for the Pulp and Paper Industry: Subpart B.
(November)
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 1
BRPP Comments are in blue italic text. Requested edits are in Meek-strikeout and blue text.
MEMORANDUM
DATE: May 21,2007 (Draft Final,DO NOT DISTRIBUTE until this header is
deleted)
SUBJECT: Color Removal Strategies For Blue Ridge Paper Products,Inc.
FROM: EPA Tech Teami
TO: Technology Review Workgroup
BRPP Comments—Purpose and Overall Report. We disagree with stated purpose as inconsistent with
the scope of the Technology Review Workgroup (TRW) under the 2001 NPDESpermit. The purpose of
the draft report is defined in Part I, Section A. (8),Paragraph 8 of the 2001 NPDESpermit.
By February 1, 2006, the Division of Water Quality(in consultation with the Technology
Review Workgroup)shall recommend to the NPDES Committee, considering the statistical
analysis report submitted by the permtttee and the demonstrated performance of the mill, the
lowest achievable annual average and monthly average color loading effluent limitations. If the
limits determined to be achievable are within or below the target range of 32,000-39,000
pounds per day as an annual average, the limits shall become effective on March 1, 2006, by
written notification from the Director. If the limits determined to be achievable by the Division
of Water Quality(in consultation with the Technology Review Workgroup) are above the target
range, the Permit shall be modified in accordance with North Carolina's permitting process to
reflect those limits."
Blue Ridge Paper submitted the statistical analysis report referenced in this permit condition in
December 2005. In January 2006, the Division of Water Quality deferred changing the Canton Mill
color limits until completion of the permit renewal process.
We disagree strongly with the negative tone for the majority of the draft report. The May 2007 draft
report focuses almost exclusively on 2001 TRW recommendations that did not work, overlooks the mill's
actual effluent color performance and the key positive statements in the 2006 Liebergott Report
concerning implementation of the 2001 TRW recommendations, and contains fundamental technical
errors. The draft report overlooks the fact that the Canton Mill achieved the second tier range of
effluent color performance in the December 2001 NPDESpermit. The 2006 secondary effluent color
performance was 37,058 Ibs per day equivalent to 26.2 lbs per air-dried ton of bleached pulp
(lbs/ADTBP). The mill's current performance is among the best, ifnot the best, of Kraft pulp and paper
mills in the world that use in-mill color prevention technology to achieve effluent color goals.
As documented in the May 2006 Color Compliance Report,Blue Ridge Paper aggressively pursued the.
2001 recommendations of the TRW In addition to the TRW recommendations,Blue Ridge Paper
implemented alternative techniques for achieving the color reduction goals in the NPDESpermit. We
achieved the second tier effluent performance goal for true color in the range of 32,000 to 39,000 Ibs per
day.
Specifics follow:
Purpose of this Review
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 2
Blue Ridge Paper Products Inc. (BRPP)has requested renewal of their National Pollution Discharge Elimination
(NPDES) discharge permit NC0000272. This permit allows discharge to the Pigeon River of industrial and other
wastewaters from the pulp and paper mill BRPP operates in Canton,NC. The pe ffnit expired n,,..embe.an
2006.. In their May 2006 Color Compliance Report(Blue Ridge,2006a),BRPP requested that North Carolina
reissue the NPDES permit for the Canton Mill with an annual average effluent color limit of 39,000 lbs per day.
This is a reduction from the current 42,000 lb per day annual average limit,which became effective January
2004. BRPP's requested limit was based on the mill's 2005 annual average effluent color, 39,000 lbs per day.
The Technology Review Workgroup requested that the EPA Tech Team support EPA's review of the color limits
included in the draft permit renewal developed by North Carolina Department of Environment and Natural
Resources. The Tech Team last evaluated color discharges at the mill in 2001 (EPA Tech Team 2001).
BRPP Comment—The 2001 NPDES permit did not expire. Blue Ridge Paper submitted a complete
permit application 180 days prior to the permit expiration date. The permit continues in effect as
outlined in the Standard Conditions for NPDES Permits,Part 77, Section B. 10.
Members of the Te�TRW visited the Canton mill on February 8,2007 to observe and collect
information and data on the status of technologies implemented and color discharges at the mill. This draft
memorandum incorporates analysis of data from that visit and other data provided by BRPP.
This memorandum reviews the process improvements previously analyzed by the Tech Team, the status of their
implementation at BRPP, and identifies additional color reduction activities that BRPP could implement during
its next permit term. The memorandum also reviews the variability of the mill effluent color discharges and
suggests final effluent color limits for incorporation in the revised permit.
Background and History
BRPP operates a bleached papergrade kraft pulp and paper mill in Canton,NC,which it purchased from Champion
International Corporation in May 1999.BRPP is owned 55 %by KPS Special Situations Fund,
L.P. and 45 %by the employees through an employee stock ownership plan'Operations at the mill began in 1908,
but the mill has been extensively modernized. ,Friest reseritb,in 1993. The mill currently operates a 810 tpd
hardwood pulping line and a 600 tpd softwood(pine)pulping line. After cooking,pulp from each line is further
delignifted in single-stage oxygen delignification(OD)systems,both installed in 1993. Hardwood pulp is
subsequently bleached with a DED sequence(BRTP stopped adding exygen to the hardwood E stage in 2005*.
Pine pulp is bleached with a DEoD sequence mnoo stopped adding peroxide to the pine v stage in 2005 Target
brightness is 86 ISO,an increase by one point from the target brightness in 2001.Up to 80% of the filtrate flow
from the pine bleach line is returned to the
---------------------
EPA Tech Team members are Ahmar Siddiqui, EPA/FAD;Karrie-Jo Shell,EPA Region 4;Don Anderson,EPA/EAD;
Betsy Bicknell,ERG,contractor;Neil McCubbin,subcontractor to ERG;Dan Bodien,subcontractor to ERG. BRPP
announced on March 26,2007 that it is engaged in preliminary discussions with The Rank Group concerning the sale of Blue
Ridge. The New Zealand-based Rank Group has considerable holdings in the paper industry. Rank Group recently bought
Evergreen Packaging,the former North American beverage packaging division of International Paper.
----------------------
TM
recovery cycle using the unique bleach filtrate recovery(BFR )process, developed by Champion.This 80%
closure rate is an improvement from 2001,when the pine line closure averaged 73.7%. Up to 25%of the hardwood
line bleach plant E-stage filtrate flow is also recovered. BRPP produces 300,000 tons per year of uncoated paper
including offset, tablet, and envelope grades.The mill also produces 275,000 tons per year of bleached paperboard
used for liquid packaging and paper cups,including FDA-approved grades for milk and juice cartons.
BRPP Comment- The strike through text are editorial comments and are not pertinent to the description
of the current mill process.Also, not all of the years listed are correct.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 3
Process Improvements Previously Analyzed by Tech Team
As a result of its 2001 analysis of the BRPP mill discharges and operations,the Tech Team identified five mill
improvements that it judged capable of reducing the discharge of color in the mill effluent. BRPP and its
consultant also identified several process optimization projects that would reduce color discharges.The status of
BRPP's implementation or evaluation of each of these improvements is summarized in Table 1,then discussed
below.
BRPP Comments—Revisions to table are marked below including several comments.
This table is an over simplification of a very complex process and decisions related to process
optimization during the current permit term. It overlooks conclusions from the 2006 Liebergott/GL&V
report. The trade-offs between bleaching chemical rates and otherpulp mill color control techniques
related to kappa,filtrate recycle, and brownstock washing eficiency are not reflected in this table or in the
TRW's review. Also during the 2001 NPDESpermit term, the market specification for some grades offine
Kraft paper changed from 87 bright to 92 bright. This market specification change required a change to
Canton Mill kappa and bleaching strategies. The current pulp bleach operation is very different than the
operation in 2001. Significant progress was made during the permit term in reducing bleach plant effluent
color.
ZD (ozone/chlorine dioxide) is not a viable technology for the Canton A1111(referencepage 18 of the 2006
Liebergott/GL&V Report). Recent experience of the pulp and paper industry is that ZD bleaching has
not given the color reduction benefits promised. Its application is mill and wood species specific. The ZD
process at Spring Grove has been shutdown.
More details regarding our edits to Table I are in comments that follow draft report discussions on the
table items.
Table 1.Review of Process Improvements Identified in 2001
Predicted Color - - - -
- Reduction(Final
.. ,Effluent,Ibsiday- _ r Color
ann:avg)' - r Reduction
Improvement ' Implementation Status Achieved Comments _ -
Process 1,400 Predicted reduction in BRPP consultant recommends
optimization C102 use(27%on IIW, reduced C1O2 use and adding oxygen
projects 17%on SW)not made. and peroxide to E stages to reduce
C102 use increased on color generation(GL&V 2006,pp52-
S W,decreased 5%on 53)
ITV
777
1,000—1,200 Pine line filtrate recycle Total Further improvements.to BFR
increased from 74 to reduction reliability may not be feasible
80%. from 2001:
6,000
Improved black >5,000 Some Significant spill lb/day,annual Additional improvement to black
liquor leak and collection improvements average liquor control possible: further
spill collection made;untreated-eelee eliminate overflows,and continue to
control(BMPs) Gent) pretreat diverted high-color
wastewater.Also improve control of
total influent color losses of white and green liquors to
variability was reduced reduce or eliminate sulfide-based
color.
Ozone/CI02 3,000-6,400 lsadegnate-eveluetien None
stage for (lab-studies);not installed
hardwood line to evaluate ozone bleaebiiu. This
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 4
technology is not recommended by
Dr.Liebergott(see page 18 of 2006
Liebergott/GL&V Report). ZD
process at Spring Grove is shutdown;
no color reduction benefits were seen
during operation.
god Stage OD for 1,100—1,400 Ifleomplete evaluation None This alternative is technically and
pine line Evaluated(lab studies); eeenemieally feasible but may not be
not installed a viable retrofit for the Canton Mill.
Complete additional studies.
Following 2001 recommendations,
BRPP optimized existing single stage
OD system on pine and achieved
most of benefit of a 2-stage OD
system based on average industry
experience.Eegieeering desigrequired,fellowed by
implement an. Additionally,with
Bleach Filtrate Recycle(BFR),
benefit of Vd stage 02 will be
marginal(see page 60 of 2006
Liebergott/GL&V Report).
Color treatment <_2,750 Evaluated(lab studies); None CRP purge stream now averages
of CRP purge not installed 8,745 lb/day.wee
stream extended period(a deie�iae what
net removed in nnvmn__uld be
_ed. ed._ith_ . Need
further study of strategies for
reducing color released in the purge.
Process Optimization Projects
BRPP Comments-
Blue Ridge Paper evaluated all of the 2001 TRW recommendations. All the "Highest Certainty"items
were evaluated and implemented as technically, economically and operationally feasible. These and other
color reduction initiatives identified by Blue Ridge Paper achieved the first tier color reduction to 42,000
lbs per day. The draft report should include these facts as documented in the several color progress
reports submitted under the 2001 NPDES permit.
BRPP absolutely disagrees with the statements in this section that the Canton Mill was slow in pursuing
BMP controls and other TRW recommendations. As documented in the May 2006 Color Compliance
Repo rt•Blue Ridge Paper aggressively pursued the 2001 recommendations of the EPA TRW. We
implemented those TRW recommendations that were feasible in conjunction with additional improvements
to achieve the second tier effluent performance goal of true color in the range of 3Z,000 to 39,0001bs per
day.
The 2006 Liebergott review of Canton Mill color performance was very worthwhile. Overall,Dr.
Liebergott understood and agreed with our decisions concerning process changes that were made and not
made relative to the 2001 TRW recommendations. In the executive summary to the 2006 report, Dr.
Liebergott states:
"The report concludes that Blue Ridge Paper Products has done a very goodjob implementing the
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 5
2001 recommendations and has made additional environmental improvement as a result. An
updated comparison of the Canton Mill to other Kraf mills in the USA, Canada and Finland
indicated that there are still no COD, BOD or color values lower than those achieved by the
Canton mill. Given the current state of adaptable technology and the mill's already high level of
environmental performance,future improvements are expected only to be marginally
incremental."
This key statement and the positive evaluations of Blue Ridge Paper color work during the term of the
current NPDESpermit need to be reflected on the draft report. The report also should consider important
benchmarking information provided in the 2006 NCAS1 Color Study report which was submitted to the
TRW with the 2006 Liebergott Report.
Dr.Liebergott has identified areas of investigation for further improvement. Blue Ridge Paper supports
these recommendations. It is not accurate to characterize the 2006 Liebergott/GL&V report as critical of
the Canton Mill operations. The benchmarking data that Dr.Liebergott provided during the February
2007 TRW review meeting clearly demonstrate that the Blue Ridge Paper Canton Millis operating at an
in-mill color performance level where few, ifany mills in the world, have any experience. This is also
supported by the 2006 NCASI Color Study,which was provided to the TRW The Bleach Filtrate
Recycling(BFR)process makes the Canton M1111 situation unique. Process changes for color that may
work at non-BFR, higher color level mills, don't necessarily make sense for the Blue Ridge Paper Canton
Mill. Dr.Liebergott has made this point repeatedly and has advised that BRPP needs to carefully evaluate
further process changes. He was very supportive of the Canton gill's findings related to the effect of
brownstock washing efficiency on kappa factor, chemical consumption, and overall bleach plant effluent
color, an important area of our approach.
Specific edits and comments on this section follow:
In 2001, Liebergott&Associates and GL&V Pulp Group prepared an extensive analysis of
the mill fiberlines that included identification of options for effluent color reduction. Dr.Liebergott was retained
for this work by a coalition of environmental groups. This report,Bleach Environmental Process Evaluation and
Report(BEPER)(GL&V 2001)presented 16 recommendations for incremental improvement to fiberline
operations. These included suggestions for improving process control,OD performance,reducing the amount of
C1O2 used on both bleach lines,and enhancing the extraction stages on both bleach lines with oxygen and/or
hydrogen peroxide. The Liebergott recommendations were included in the 2001 NPDES permit and were
evaluated. Those that were technically,economically and operational feasible were implemented
BRPP Comments—Dr.Liebergott was identified and retained by a coalition of environmental groups in
2001. Blue Ridge Paper agreed to allow Dr.Liebergott and these groups f dl access to the mill for
technical review of our process and color reduction opportunities. However, the Canton Mill was not the
client for the 2001 Liebergott report. The mill evaluated all of the 2001 Liebergott report
recommendations, and the majority of these recommendations were implemented. Where not
implemented, the mill learned information that in several cases led to other improvements.
eptimizatien and:..................nts ia.....ified by the TRW in 2001 (T.,..haelegy Review{IL.-L..retip 2V10 r_r v.v
2006e and Blue Ridge 2006 a) afneng the"highest ee.we:.at.e'ite..... They have implemented many..f the t]T. N
b evef a ffmeh longer pefied of time than aatieipa4ed by the Teel.Teem and the T M' Dceeess eptim zatien and
mefe 1 implementingBNIPs together tfibute to redueing variability efbet6.,.w untfeeted
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 6
Process control and OD performance have been improved. However,on the hardwood line,they did not decrease
the Di stage kappa factor as recommended and they discontinued use of oxygen in the extraction stage. As a result,
the total C1O2 charge on the hardwood line as of May 2006 was only 5%less (not the predicted 27% less)than the
charge in 2000. Similarly,on the softwood line,they did not decrease the Di stage kappa factor as recommended
and they discontinued use of peroxide in the extraction stage. As a result,the total C1O2 charge on the softwood
line as of May 2006 was 11%more(not the predicted 17%less)than the charge in 2000.
BRPP Comments-
Discontinuation of 02 in hardwood E stage—Item 7,page 14 in the 2001 Liebergott report recommended
a reduction in 02 in hardwood E stage. BRPP implemented this change and continued optimization
resulted in the elimination of 02 from hardwood E stage. During all this process work, BRPP saw no
change in E stage kappa or to effluent color.
C1O2 charge-Bleaching strategy changed in response to market conditions as discussed above. This has
affected overall bleach plant chemical use and kappa factor. As discussed at the Feb 2007 TRW meeting,
Blue Ridge Paper has identified pulp washing efficiency as a primary factor influencing kappa factor and
bleach plant chemical demand. This is the current focus of color and chemical use efforts. At the April
2007 outage, the pre-bleach washer drum was changed as BRPP advised the TRW at the Feb 2007
meeting. As a result,Pine DI stage kappa factor is now at targets recommended by Dr. Liebergott(27
Before, .24After).
In preparation for renewal of their NPDES permit,BRPP again hired retained Leibergott&Associates and GL&V
Pulp Group to analyze their fiberlines,review the implementation of the 2001 recommendations and identify
further options for color reduction. In the 2006 update to the BEPER(GL&V 2006),the consultants repeated their
recommendations for BRPP to reduce the kappa factor in the Di stage of each bleach line and replace the C1O2
bleaching power with oxygen and/or peroxide in the extraction stages. The,.,.nsialtants ale _.........mended th t
eelef.
BRPP Comment-The Liebergott recommendation is to investigate peroxide fortification of the E stage.
High temperature peroxide(PHT) is not a recommendation in Table 1 of the 2006Liebergott/GL&V
Report.
In addition to reducing the color of the bleach plant effluents,use of a lower kappa factor means BRPP would use
less C1O2 to bleach,which will reduce the total chloride content of bleach plant filtrates. T ewer ehleride eente-t
will make it possible fef the mill te feeevef more hardweed fiRfates with the BERIM preeess and reduee the
Mill tO F@dUGe the GRP... Fge fgew and the eekw it eentFibutes to the mill effluent.
BRPP Comment-BFR is not on the hardwood line. Statements above are technically incorrect. A
portion ofhardwood E stage filtrate is recycled. Mill experience shows increasing in hardwood E stage
recycle adversely affects pulp washing and causes carryover to the D stage, increased chemical
consumption and results in increased color. These offset any potential color benefit from increased E
stage recycle.
BFR Reliability Improvement
As reported during the Tech Team's February 2007 site visit,BRPP spent$1.5 million over the last five years to
improve the reliability of the BFRTh1 system,These expenditures included replacing tank liners for the three
existing sand filters (now high grade stainless steel),replacing and/or upgrading existing process piping with
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 7
piping with improved metallurgy,installing a fourth(new)multimedia filter, and installing a third ion exchange
softener. With these improvements,BRPP has increased the BFRTM closure rate from 73.7%in 2001 to 79.2%in
2006. Mill representatives stated that the present closure rate of approximately 80%of the pine line bleach plant
effluent represents the maximum amount attainable without incurring unmanageable corrosion and scaling
problems. This is because scaling from hardness minerals accelerates above closure rates of 80% (Bodien,2007).
BRPP Comment—This discussion accurately reflects our experience.
Improved Black Liquor Leak and Spill Control(BMPs)
BRPP Comments-
Blue Ridge Paper strongly disagrees with the technical analysis, conclusions and tone of this discussion.
We have a number of comments and request that the Draft Team properly consider all of our wastewater
data before making conclusions about lack of demonstrated improvement in black liquor management and
spill control.
The Canton Mill's Best Management Practices (BAP)program and action levels for black liquor leak and
spill control are based on total influent color to the wastewater treatment system. Primary influent color
is the most representative parameter to track in-mill colorperformance. This has always been the focus of
previous TRW review sessions.
Primary effluent color is monitored for process control purposes, but it is not an accurate indicator of
black liquor losses or other color material losses to the mill sewer system. Primary effluent color is
skewed by fine turbidity and also reflects the sewer generated color phenomena and possibly other effects.
Its apparent removal efficiency in biological treatment is better than total influent color—suggesting that
the fine turbidity component is significant or that the treatability of color is transformed in the primary
treatment process.
The TRW requested information on color events determined from our BAP monitoringprogram during
2006. The monitoring of influent color identifies these events. The TRW did not request data for high
primary effluent color data, and Blue Ridge Paper did not provide such information.
Blue Ridge Paper requests that the Draft Report Team present total influent and secondary effluent color
in any discussion of black liquor leak and spill control. The appropriate data tables are below. Between
2001 and 2006, average influent color was reduced by 19,271 lbs/day or 33.3 percent. Secondary effluent
color was reduced by 5,618 lbs/day or 13.2 percent.
In summary,primary influent color—the correct measure of black liquor loss and spill control—showed
definitive improvement during the term of the 2001 NPDESpermit. Secondary effluent color also showed
definitive improvement. In 2006following implementation of the acid sewer project, the contribution to
color f om all mill sewers balanced with measured influent color to wastewater treatment. This was a
major topic of discussion at the February 2007 TRW meeting at the Canton Mill. The balance of mill
sewer color loadings with measured influent color in 2006 supports influent color as the accurate measure
of BMP effectiveness. This information indicates that reduced influent color does not translate into
reduced effluent color. The Primary effluent table should be removed from this analysis. The lowest
primary effluent year was the highest secondary effluent year. More importantly, the highest primary
effluent year was the lowest secondary effluent year. Therefore, this is not an appropriate comparison of
mill colorperformance.
Table 2.I—Primary Influent(Totallrrfluent)Statistics by Year
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 8
Mean True Relative Percent ,
Color' Standard.. Number of Decrease .'
(lbs/day) Standard Deviation days>100;000 From Baseline
Deviation(SD) (SD/mean%) lb/day (2001)
2001 57,725 20,619 35.7% 4 0.00%
2002 54,780 17,195 31.4% 7 5.10%
2003 55,550 19,424 35.0% 9 3.77%
2004 49,466 18,786 38.0% 8 14.31%
2005 45,175 22,297 49.4% 7 21.74%
2006 38,454 14,015 36.4% 2 33.38%
Table 2.2-Primary Effluent Statistics by Year(Should not be included in Report)*
Mean True Relative Standard
Color(lbs/day) ;'- Standard Deviation(SD/mean Number of days>
" Deviation SD : ' .- % r a 100 000ablda
2001 62,008 i9,561 31.5% 13
2002 59,956 18,680 31.2% 16
2003 59,646 18,468 31.0% 10
2004 65,206 26,674 40.9% 40
2005 63,838 24,158 37.8% 28
2006 65,512 25,427 38.8% 36
*Percentages not given due to lack ofsignificant statistical change.
Table 2.3-Secondary Effluent Statistics by Year
Mean True Relative Percent
Color Standard Number of Decrease
(lbs/day) Standard Deviation days?100,000 FromBaseline
Deviation SD SD/mean % lblda : 2001
2001 42,676 10,925 25.6% 3 0.00%
2002 41,166 9928 24.1% 0 3.54%
2003 44,627 11,043 24.7% 1 -4.57%
2004 * 41,463* 32,568 * 76.6%* 1 4 0.43%
2005 39,092 10,092 1 25.8% 0 8.40%
2006 37,058 8959 1 24.2% 2 13.17%
*2004 data were affected by historic floods in Sep 2004
Specific edits follow:
BRPP continued efforts intended to improve its management of leaks, spills, and intentional diversions of black
liquor over the last six years. These efforts include:
• Interconnecting the pine line and hardwood line spill collection sumps so that tankage in either line can be
used interchangeably for spills;
• Repositioning sewer conductivity probes from sumps to in-line to more accurately and reliably identify
high conductivity wastewaters;
• Diversion of up to one hour of total mill flow to off-line 1-million-gallon clarifier, during high color
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 9
releases,to equalize color discharge to wastewater treatment plant;
• Prior to process line outages,improved prior planning for capture of high-color process liquors and black
liquors and better managing their timed release to treatment system;
• Continued operator training;and
• Implementing two hour testing for color at the W WTP with one hour testing during outages or semi-annual
shutdowns.
systeffi. Table 2 sh8ws the ffieaft, standard deviation,and relative staridafd deviation (s4andafd deviatien as-a
BRPP Comments-
The primary effluent table should be removed from this analysis. The lowest primary effluent year was the
highest secondary effluent year. More importantly, the highest primary Effluent year was the lowest
secondary effluent year. Therefore, this is not an appropriate comparison of mill color performance.
In 2006, the contribution to colorfrom all mill sewers balanced with measured influent color to
wastewater treatment. This further supports that influent color is the accurate measure ofBMP
effectiveness.
Tamp 1.PArnary Effluent Statistics,by year
R,I...(..l��T`�//��.ae C-619G 1�YI@HYC-SL9a(IBEfI .
r
ldy
2M 62-898 19�G, p 31-3/o . _ 4-3 . ..
2082 39,956 18,680 31.29; 46
24" 39,646 19,468 3-i.o% 4-0
24" 65,286 26#74 48:9% 49
24" 63,838 24,438 97.P-"/ 28
Prior te-2006,the asid sewer Rrixedwith other mill wastewater prier to the treatffleirt systeffi,resulting in"sewe
generffted eeler." As efiantiat-y 1,2006,the aeid sewer was separated firom the other mill sewers Eaid new enters
the treatment plant a4ef the prinrat5,Glarifiep. However,the primary effluent sampler is leeated devmstFearn of 430
Faixing point efthe primary clarifier everflew and aeid sewer. Thus,the statisties far 2006 presented in Table 2,
primary effluent,whieh is the total load to seeondary treatment,is directly measured with a eomposite sampler.
of the t.
Table 2 presents the immber of days for whiek the pmnafy e0lueat eeler exGeeded 100,000 lbs,lday.As shown,
BRPP has made ne pregress in fedusing the frequeney at whiek it experienees high oelor leads into the tFeatrnen
syster.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 10
The Teeh Team notes that paFagraph 19 of the 1998 Settlement AgFeement ineluded the fellewing previsiw.
disEuptiens through detailed seheduling of plapned outages and sentiageney plaw�ag for unplanned outa
This was again iieted in the 2001 T-eeh Team report. Despite this eenifaitment,the mill has net deefease
disehaFge of greater than 100,000 peands,the mill variability inoreased between 2001 and 2006 (see Table 2).
Ozone/Chlorine Dioxide Stage for the Hardwood Bleach Line
BRPP Continents-
ZD (ozone%hlorine dioxide) is not an appropriate technology for the Canton Mill(seepage 18 of the
2006 Liebergott/GL&V Report). The ZD process at Spring Grove was shutdown after it failed to
demonstrate color benefit.
Dr. Leibergott discussed the performance of the Domtar Espanola mill during the February 2007 TRW
meeting at the Canton Mill. The Espanola mill is cited as a successful application of ZD that is
applicable to other pulp and paper mill. The Domtar Espanola color performance is 3 to 4 times higher
than the current color performance of the Blue Ridge Paper Canton Mill(Recent Environmental
Improvements at the Espanola Mill, 2002). Additionally, the Domtar mill uses a very different wood
species than the Canton Mill. The Domtar mill does not have bleach filtrate recycling(BFR).
Dr.Liebergot identified issues with the PAPRICAN work. Blue Ridge paper reported these to the TRW
as part of color project documentation.
Edits and comments for this section follow:
BRPP engaged the Pulp and Paper Research Institute of Canada(PAPRICAN)to investigate potential
modifications to the hardwood pulp bleaching process and determine effluent color reduction that would result
from these modifications.The PAPRICAN report entitled"Bleaching Evaluation for Effluent Color Reduction"
(Audet et al,2003)was provided to the Tech Team in early 2007.
PAPRICAN stated that their objective was to evaluate whether modifications to the hardwood bleaching
sequence at Canton specified by Mr.Johnnie Pearson(presumably a BRPP representative) could"generate
effluents with a color reduction target of 25%."
BRPP Comment-Mr.Pearson was the BRPP process engineer who oversaw this work.
BRPP provided PAPRICAN with oxygen delignified hardwood pulp.PAPRICAN bleached this pulp in their
laboratory, investigating various combinations of chlorine dioxide,ozone and hydrogen peroxide.
The report concluded that ozone could replace some of the chlorine dioxide used in bleaching(known as a
"DZ"stage),while producing pulp of equal or slightly better quality than the control sequence. This is
consistent with the open literature on ozone/chlorine dioxide combinations.
PAPRICAN also concluded that replacing some of the chlorine dioxide in bleaching with ozone increased the
color in the bleach plant effluents.This contradicts the experience reported by Domtar(formerly E.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 11
B. Eddy)at their Espanola mill,where a dramatic reduction in effluent color was observed when that mill
installed an ozone pulp bleaching system(Munro and Griffiths,2000). The Tech Team notes that the Espanola
mill has no color discharge limits and it installed ozone to reduce its bleaching costs.
The Tech Team fsund-a-sigai€ieaxt agrees with findings in the 2006 Leibergott/GLV report that there were
deficiencies in PAPRICAN's analysis of the laboratory results. The concentration of color in the effluent for
each stage was added to obtain the total concentration for each tested bleach sequence. This approach neglects
two points:
1) Mixing effluents produces reactions that may increase or decrease the concentration of color
in the combined effluent.
2) Filtrate volume affects the measured concentration(e.g.,lower volume will concentrate
the filtrate to a higher color).PAPRICAN's report does not present the filtrate volumes,
so a comparison of concentrations may be misleading.
Liebergott et.al. reviewed the PAPRICAN report and identified the same deficiency with regard to effluent
mixing. They also noted that the quantity of chlorine dioxide added to the ZD stage was too high,which would
result in higher effluent color(GL&V 2006,p 18).
In response to EPA's question about how the PAPRICAN results were used to estimate impacts on final effluent
color,BRPP responded:
The pilot study reactors were batch and not continuous and did not include filtrate recycle. The pilot
studies included bleach stage filtrate color concentrations, but there was no filtrate flow rate data
from which to calculate a production-normalized filtrate color mass. For these reasons, we did not
attempt to calculate secondary effluent color impacts using pilot study data for individual bleach plant
color streams.
In the PAPRICAN study, the individual bleach stage colors were compared directly and in total.By
both means, the color of the ZD stage was higher than the baseline DEoD.It is also well known that
when individual bleach stage filtrates are mixed the resultant effluent color is very difficult to predict.
(Blue Ridge,2007a)
The Teeh Team eansludes thRt BR-421i's inves9gatien ofthe potential for haFdwood pulp ozone bleaehing to
Fedtlee the mill'q effluent seloF was inadequate. FuAher laberatery 4fials,designed and implemented to provide
eorabined at the BRPP mill are needed to evaluate this tephiielegy,
BRPP Comment- While there were issues with the PAPRICAN work, Dr.Liebergott does not
recommend ZD technology for the Canton Mill(page 18 of the 2006 Liebergott/GL&V report).
Second Stage Oxygen Delignification for the Pine Bleach Line
BRPP Comments-Following Dr.Liebergott's 2001 suggestions, Blue Ridge Paper made significant
process improvements in our existing,single stage 02 delig system on pine. We don't agree that 2"d
stage 02 is now a "highest certainty"color reduction technology. Blue Ridge Paper does agree to
conduct another lab scale study for this potential process modification.
In the 2006 report,Dr.Leibergott warns that it is proven difficult to accurately forecast the benefit to
secondary effluent color of individual in-mill process changes. Two stage 02 deliginifncation on pine
may not be the most cost effective use of limited capital for additional color reduction. Typical
performance for 00 is 40 to 50%delignification efficiency[TAPPI Bleach Plant Operations 2007]. The
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 12
current single-stage 02 delig achieves 4501o. [TAPPLBleach Plant Operations 2007]. The 63%
efficiency for 00 stated in the 2006 Liebergott report is a single example for a mill with a different wood
species and higher incoming kappa than the Canton Mill.
Specific edits and comments follow:
BRPP contracted with Andritz/Ahlstrom("Andritz")to study,among other things,the addition of an additional
stage of oxygen delignification in the softwood bleach line at Canton.Andritz,a well established supplier of
pulping and bleaching technology and equipment,maintains the Pruyn's Island Technical Center,which
conducted the tests. The Andritz report, entitled, "Laboratory CK and Lo-Solids Cooking with O-Do-Eop-D
Bleaching Sequences on Softwood Furnish from Blue Ridge Paper,Blue Ridge,NC"(Andritz Ahlstrom Sales
2001)was provided to the Tech Team in early 2007. -
The text Pf tl�e-4—nd-Fit-A Fopeft states that"BRPP requested the sales depaFtmant of Andritz AhlstFem to
investigate altemative metheds e9pulping and bleashing that watild reduee their waste pmduets and impreve the
quality of their haf",eed and saftwood papeF predtiots." As indieated by this statement of wer4E,testin94
seeend stage exyges delignifleatien was a minef part ef the werk AadritE did fin:BRPP. FeEther,very few efthe
Specifically,Andritz compared single stage and two stage oxygen delignification of samples of pulp provided by
the BRPP mill. The tests showed that a second stage of oxygen delignification could reduce the kappa number of
the unbleached pine pulp by 22%.
Andritz did not bleach the pulp after the two stage oxygen delignification, so the report provides no information
on the impact of the additional oxygen delignification stage on effluent characteristics, including color.
In response to EPA's question about how the Andritz results were used to estimate impacts on final
effluent color,BRPP responded:
The Andrliz-Ahlstrom study ofsecond stage 02 for pine showed delignification rangingfrom
42.7%for the single stage, 48.6%for lire 0-0 and 55%for the 00 stage. With improvements
that BRPP made on the existing single stage pine 02,the deliginifrcation increased from 40%
to 45%. With BFR in place on the pine fiberline, the effluent color reduction from this
improvement in 02 deliginifrcation has been very difficult to identify. The inability to quantify
the effect on effluent color of improved 02 deliginifrcation made the capital cost to install a
second stage 02 our pine not economically feasible. (Blue Ridge,2007a)
BRPP's aAalysis evaluation of the benefits of adding a second oxygen delignification stage to the pine line is
ineemplete should be conducted again. BRPP's consultants,note that even with the percent delignification
currently achieved on the pine line,a second stage could achieve an additional 20%to-25%delignification(see
GL&V 2006,p 118). Page 117 of the Liebergott Report predicts an 8%improvement in delignification
efficiency at the Canton Mill.
By making a 20 to 0appanwneFe 0
r
6 the .,..p:op
of thehardweed bleach filtFates to be Feeyeled-threugh the BFR3:m system.. A kappa number decrease or
delignification efficiency increase would not equate into a similar color reduction. Although with the
information available it is not possible to rigorously calculate the benefit of this change,the Tech Team would
expect in the order of 1000 Ibs/day reduction in bleach plant color discharge.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 13
BRPP Comment-BFR is not on the hardwood line. Statements above are technically incorrect. A
portion of hardwood E stage filtrate is recycled. Mill experience shows increases in hardwood E stage
recycle adversely affects pulp washing and causes carryover to the D stage, increased chemical
consumption and results in increased color. These offset any potential color benefit from increased E
stage recycle.
Liebergott et al(GL&V 2006)estimated that implementing a second oxygen delignification stage on the pine
line would reduce bleach plant(Note 4,Page 6)color discharge by 1142 lbs/day.They estimated the capital cost
of adding an additional oxygen delignification stage would be in the order of$2 to$3 million* and the
chemical cost savings approximately$1.2 million/year,providing a reasonable payback. Mill staff has
mentioned a three year pay-back,which is consistent with this estimate. The BRPP oxygen delignification
systems were installed in 1993. Since that time,the use of two-stage oxygen delignification rather than the
traditional single-stage systems has become well established in the industry because it normally reduces mill
operating costs.
BRPP Comment- *Based on steep steel price increases since the 2006 Liebergott report, the capital
cost estimate is now on the order of$3 to 5 million. Potential payback is overstated. Blue Ridge Paper
agrees that second stage 02 delig on pine warrants further evaluation.
The Tech Team identified the following color reduction activity for BRPP: develop an engineering design
leading to installation of an additional oxygen delignification stage for the pine(softwood)pulping/bleaching
line. In 2001,the Tech Team estimated that this process change would reduce effluent color by 1,100 to 1,400
lbs/day. As noted above,even after the recent improvements to pine line delignification efficiency,GL&V
estimated that this change would reduce discharge by from the bleach plant on the order of 1000 lbs/day.
Andritz also evaluated modifications to the mill cooking process. These would require complete
replacement of the digester systems at Canton,which would cost(at least) several tens of millions of dollars,
if they are feasible at all within the mill's space constraints. The Tech Team concluded that modifications to
the mill cooking process do not merit further analysis.
Treatment of CRP Purge Stream for Color Removal
BRPP Comments—The suggested experiment to shut down CRP for an extended period is operationally
infeasible. The CRP process is an essential part ofBFR and must achieve high uptime to sustain a high
BFR closure rate. Without BFR, the Canton Mill secondary effluent color will be at a minimum of
15,000 to 20,000 lbs per day higher than the current level ofperformance. We will continue to look at
options for alternative handling of the CRP purge stream.
Specific edits and comments follow:
The Chloride Removal Process(CRP)purge is a low flow,highly concentrated stream. Color is typically 41,000
platinum cobalt units(pcu) in a stream that discharges at 10 gpm(15,000 gal/day or 3 tank trucks/day).BRPP
reported that in 2006 the CRP purge contributed approximately 8,745 lb/day(23%)of the total mill color loading
to the treatment system but only 0.05%of the discharge flow. By comparison,in 2001 the CRP contributed
approximately 5,000 lb/day(13%)to the treatment system loading(EPA Tech Team,2001). Thus, in the last
five years there has been a significant increase in color contributed by the CRP,both in lb/day and in percent of
total mill load'. BRPP reported on its investigations into technologies for reducing CRP purge color in its March
2005 report,"Chloride Removal Process(CRP)Color Reduction Technology Assessment" (Blue Ridge 2005).
In this report,BRPP points out that the CRP purge is'a very concentrated material(360,000 mg/L or 36%total
dissolved solids)that is discharged to sewer at 165*F. Handling the material is difficult because it will
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 14
crystallize as it cools. Also,it is quite corrosive due to the high chloride content.
BRPP Comment- The increase in CRP purge contribution to total mill color loading is related to other
color improvements including increased BFR closure rate.
As described in their 2005 report,BRPP assessed more than nine alternatives for disposal or treatment of the
CRP purge and concluded that none was technically feasible. Although consistent performance was not
demonstrated, C1O2 bleaching was a low cost,potentially effective means of reducing the CRP purge color prior
to discharge to the treatment system. BRPP found that C1O2 bleaching could remove 75 to 90%of the color in
the CRP purge stream.
BRPP Comment- Color removal from C1O2 bleaching of CRP during pilot scale trials ranged from zero
to 90 percent and was very inconsistent. Foaming was extreme. Process control was very di cult.
Based on the extended pilot scale testing, C107 bleaching of CRP is not considered operationally
feasible.
Although CRP purge is a relatively low flow stream,BRPP estimated that off-site solidification and land
disposal would cost more than$3.6 million/year.
During the Tech Team's 2007 mill visit,BRPP stated that there was no apparent decrease in secondary effluent
color when CRP wastewater was not flowing into the treatment system. In response to a Tech Team request,
BRPP presented data correlating the days on which the CRP process was shut down with secondary effluent
discharge color,for August 2006 to January 2007(Blue Ridge 2007a). During this period,there were nine
widely dispersed pairs of days when the CRP process was down for part of the day. BRPP concluded that GRP
eeler is removed in the seeendafy tfeatment system,based ea aii ANOVA analysis somparing"demW'days
with nBn ,...,._..«:__ days The data analysis conducted did not indicate a difference in secondary effluent color
between days when CRP purge was being sewered and days when the CRP process was down. These data
were not conclusive.
The T-eeh Team does not agfee with this eenelusion fef the fellewing reasons. Finfl, F4he
change in ram,efPueat ehaFaetoristies. Seeendly,the times of shutdaA%and stai:vap of the GRP de not
earrespend with the effluent sampling times in the mill systeff4 se that there is no direct c rr elation in time Ivith
the effluent sample.Further,the faet that all GRP"Eiff!'days are in pairs,suggests that GRP was dev�%fer-a
period that spafaed pafts eftwe mill sampling days.
BRPP Comment- There was no definitive conclusion. BRPP has reported the general observation
supported by limited statistics that CRP purge on and off does not appear to influence secondary effluent
color.
Also,The simple mathematioal aveFage a f tfealed offluent e elef dis eharge duFing"GRP off'days for August te
December 2006 is 39,995 lbs/day. The average when the GRP preeesswas epefating is 36,958 lbs/day.This
suggests that CRP purge stfeam eauses a reduetien in effluent colon,whiE;h defies common sense.
diff�renee in effluent eelef between the "GRP aft"and"GRP eff'days is 790G,while day to day var-iafielis in eale
.,6.......s in the CRP.......tee..f..eam �
in shei�,beeause there are se fiew eenseeutive days withetit a GRP purge,the data ffem the period August te
eater in eFde«to e.,..le..e this fb fther BBDB should eendeet--C.411 ..vale Fnill t.:el ill which the GRP..uEge i&
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 15
white ..low, month o a to sh,... sig 'F 6
liquor Y � J )requiring� b
This is consistentwith experience in otheF mills,and is due to the faet fliat the quantity of oblerides an
petassium removed by CRP is small relatiVe to the inventery in the mill's blaek,green and white liqueF systeffi.
The seeend appreaeb would be to store the CRP purge stfeam, either in the spare elarifier er eff site fey a period-
ef at lest thFee weeks,
The impaet ef&Eeluding the GRP purge from the wastewatef tFeatmeat plant would be obseFved by reut
the GRP pufge seatFibutpq 1i#1A tA 4is final effluent e9lar. In this ease, Ah-th8f FOsearGh inte reduoing GRP eele
On the other hand, if the e9hient eeler wishous the GRP pufge deereased,BRI212 would eenehide that the
Feeevery boiler systems. This eaFfy ever eauld be eliminated 9the two tfaditionally designed reeev@Fy beilevq
(whish aFe 3 4 and 42 yeaFs old)weFe Foplaeed by one modem boiler. This would prebably represent a eapital
oest ef ever$100 Faillion,and would very substantially reduee the energy eosts at the mill, sines today's boilers
ffe Hraeh fnere effiei&4 than these efthe vk-itage installed at Ganten.Analysis ef all the eeeiieftaies and long
teEm life of the mill would be neeessary to evaluate sueh aft investmen .
On a ffiefe faedest level,it may be feasible to reduee blaek liquef eaffy ever by adjusting epefating eendit
intheelkistifigDCB,q. 4:1se:PAA� Team is flat awaFe 4 any expeFiABAA vith t�iq ifi other mills, er researeh,but
OpffatiOHS is warFanited. Any of the measures diseessed pFeviausly in this doeument to 2-aSd-6AHld
Feduee the e removed in the GRP ping.stf..am TC is is reduee.l the solar.1:....1 arge
quantity )
also would be redueed.
BRPP Comment- Champion did an extensive investigation ofDCE carryover ofsolids into CRP and did
not identify any viable solutions.
BRPP could also further investigate C1O2 treatment and other treatment of the CRP color and other
approaches for excluding the CRP purge from the mill discharge.
Additional Color Reduction Strategies Not P_,..,:......ly Analyzed by Teen Team
BRPP Comment-All of the items below were discussed in the past by the TRW. Seepage 3 of July 25,
2001 TRWreport.
Mill wastewater tFeatmen!plan :
The following color reduction activities have been previously identified and evaluated for the Canton Mill. They
should continue to be evaluated.
1) Investigate color formation when acid wastewater is mixed with mill wastewater,and
identify techniques to reduce this effect;
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 16
BRPP Comment-The Canton Mill has been a leader in the pulp and paper industry on the phenomena
of sewer generated color. The mill's extensive sewer color balance data first identified sewer-generated
color as an issue. The acid sewer project completed in January 2006($1.5 million capital cost)was a
specific measure to reduce the sewer generated color effect. Secondary effluent color has been reduced
as a result.
2) Add polymer and/or other treatment chemicals upstream of the secondary clarifiers;and
BRPP Comment- Tlae Canton Mill has evaluated end-of-pipe treatment of this type in detail. There are
many problems with this type of treatment approach.
3) Add polymer and/or other treatment chemicals to the high-color wastewater diverted to the
extra primary clarifier.
BRPP Comment-The Canton Mill has been doing this for years and will continue to do so.
These strategies are discussed below.
Color Formation When Acid Wastewater Is Mixed With Mill Wastewater
BRPP Comment-Blue Ridge Paper will continue to study and evaluate opportunities to continue to
reduce sewer-generated color.
BRPPBRPP id d 1. T h T with daily solar data fef the "Low r ifV' (mill sewer), aeid se and.. i
r. ............. ....... ........ ........ ..J ...... ....... .,.. ...,..—'ter � 1
effluent sampled after the aeid seweF is added. Figure 1 presea4s the total fflill eeleF eaktilated by adding4he
min_,.er lb/day to the a"id sewer lt,la_"y(the lower(blue fine) a_ the figure). v eempaFisen,the
effitient sampled after the aeid sewer is added is also shevm on the figere (the iiPper(fed lifle) on the figur,+.
Thus,the a...._e depicts the i«.. aef._f_...i.6ng the acid ",....eF with the_e of h ill' fn
uen
The Teen Team ebsB.n.ad
of..olo«fe fmod o g the aeid d ill striking On average, 6
Y J sewer n m n o
quantitya tal polar in tl.a Vye st feams in other ,m....an the simple m
of__these streams fefms nearly half the total .elegy diseha ge fro h ill
• The variation in quantity of eeler f4med by mixing the effluent s4feams fr-em day te day is dramatie,-as
is evident in the 6 Y Ypereentage efprifnafy iBpet eelan,
-:ally oar.. .... ..eme days to a of 4801
wawa. ..J ...,....,.,uv,..v "J" u
• There is..o obvious eaffelati:n between the values __ any ene day, e_series e f says elese to efle
The Teoh Team suggests that BRPP itwe B&6iHg 418 GO1OF fOFM-Rd V'hAB the Agid 90WAF iq
mixedwith the rest Rf t� -mill effl-ijent. -R-P62-12 eeuld investigate, amoag other giings,how sulfides intreElUeed b
to effluent eeler. In paFfieulEff-,BRPP eetild invesfigate whetheF sulfides in the mill wastewater eentribule4a
eolor feEmatioa when aeid seweF is added to mill effluent prier te bielegieal treatment. Af)proashes to
I aberate is oirefflueR4 with varying degrees e sediums xl.Qe adaifien—aripump a
....piing"t"tien for a period of at last thie,.......n4hs te det,...._.i_e ....ffel..ti.._l ofsulfide
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 17
in most Faills, losses ef sulfides san be redueed. Well lmeA%satiFees inelude imbalaflee in thoweak Wash
system,as vvell as spills efwhite,blark,and green liquor. BRPP should fitEther investigate losses fEem the green
and white li"of systems and teehnelegies that ean reduee these losses.
160,000
Total untreated color,before and after mixing,2006 (lbs/da})
125,000
100,000
75,000
50,000
2s,000
1J. 31dan 1-hler 31A1ar 30Apr W0.M 29Jun 29 Ul 20Aug 27Sep 29-0ct 2644ov 26dec
'Ph�PAIAF AA13A Mill WMW WaSleWatff may be Fedueed somewhat by tFeatment in the primary elaFifieF.This eeleF Feduetion is Begleeted
so Figure 1 is prebably ft Slight tffideFeStiffiat2 efthe impaet ef naiiking the aeid sewer with the Nst ef the
mill effluent.
� '
Figure 1; impaet of Mking Acid Sewer and Mill
Improving WWTP Performance Using Polymers
BRPP Comments—
Blue Ridge Paper currently uses polyamine to batch treat color material diverted to the spare clarifier.
This is the only use ofpolyanline that we have found effective. Polyamine is effective for treatment of
brown color—color from•om black liquor solids—but is not effective for treating bleach filtrate color. It is
also not effective for dilute brown color. The majority of color in the Canton Mill wastewater effluent is
related to bleach plant filtrates.
The use ofpolyamine in the secondary treatment system is operationally infeasible, at this time,for a
number ofreasons:
• The polyamine adversely affects sludge dewatering.
• The polyamine can result in chronic toxicity in the treated wastewater effluent
• The polyamine builds zip in activated sludge causing problems with settling and overall
treatment effectiveness
• High cost treatment
In 2005, the Canton Mill did an extended pilot scale trial of tertiary color treatment using polymer
coagulation and microfiltration (Pall Filtration study in Figure 1 of the May 2006 Color Compliance
Report). A number of dfferent polymers were tested including polyamine and polyaluminum chloride.
There was no significant redaction in secondary effluent true color from polymer-added filtration. The
reason for this result is the absence of brown color in the Canton Mill effluent. Polymer coagulation is
not effective on bleach plant filtrate color, the majority of the color in the Canton Mill treated secondary
effluent.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 18
bodies,then remeve t4e preeipita4e in teFtiary elaFifiers. One of the newest of these mills is Pelge iR 1441di*'
installing teAiary treatment systems. These systems use peb�mers and other ehemioals to pFeeipitate
plant,Chile. This mill has exeallent in plant effluent eeatT-el, a eewventienal secondary wastewater tfeatment
fellowed by a small teftiafy system.Effluent from this mill is depieted in Fig'are 2,flext to effluent ffem B
v —
}3
k,
Figure 2: Comparison of BRPP Canton Mill Effluent(February 2007) and Valdivia Mill Effluent
(November 2005)
BRPP Comment-This is an inappropriate comparison with different size containers and different
contrasting backgrounds. This is not an appropriate comparison and not appropriate in a technical
report on color reduction technology.
Tf satment systems like the one ap erated 9 3.laldivia have a eapital oost in the tons of milliefis of dellaFs, an d
ean ereate issues with solid wasto disposal. However-, same mere modest eentFel Fneasufes using variatie
.,
Spring GFeve,PA has supplem
its in plant eelar Gentfel strategies by using a eamAnereially available pelyalamintim 61410fide palyffl@F40
enhance eeler Fameval in the wastewater tfeatment plant. The polymer is added in the discharge flume fieffi the
and eemplexes�Mth arganie eampaunds that produee ealor. Colored material sett4es etA with the seeaadaFy
sludge m4ieh is hold in the feEmer stabiliza4ion basin b8fff@ dOWat8fiHg. Qlatfeltei:have not Fepei4ed sludge
dewatei4ng pFeblems. Pelymer used far additional eeler removal repoi4edbx easts an the order of$2,000 to
$3,00941ay.
BRPP Comment-Representatives from the Valdivia, Chile Mill visited the Canton Mill in summer of
2006 to review our success with in-mill color prevention,specifically bleach filtrate recycling. The
motivation for their visit was to identify alternatives to their very expensive and sludge intensive tail-
pipe color treatment system.
BRPP could conduct laboratory trials of commercially available polymers and other wastewater treatment
chemicals,adding them upstream of the secondary clarifiers. During these trials BRPP would determine the
potential for reducing color discharge and estimate the costs. Such analysis would consider seasonal use of the
additive chemicals(i.e., adding the chemicals during periods when river flow is low and the mill discharge has
the greatest impact on the river color).
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 19
If the laboratory trials are successful,BRPP could institute a full-scale trial€eeene meuth,and report results
to the Technical Review Workgroup.
Pretreatment of Diverted High-Color Wastewater
BRPP currently diverts up to one hour of total mill flow to their off-line 1-million-gallon clarifier,during high
color releases. The purpose of this flow diversion is to equalize color discharge to wastewater treatment plant.
Diverted high color wastewater is batch treated by polyamine,then drained back into the wastewater treatment
system.
An days........ear of high eele-leads(>100 nnn lb/,day)into the...,eendaFy tFeat..eat system ODOR eeWd treat
system.
Summary: Additional Identified Color Reduction Activities that BRPP can pursue in the Next Five Years
BRPP Comments-BRPP disagrees with the assertion that we have not adequately implemented the
strategies identified by the TRW.
As required by the NPDESpermit, Blue Ridge Paper evaluated all of the 2001 TRW recommendations,
implemented those that were feasible, and developed alternative technical approaches to achieve the
color reduction goals in the permit. Upon detailed examination, the "Reasonable Certainty"items in
the 2001 NPDESpermit were not technically, economically or operationally feasible for the reduction
of effluent color. Additionally, while the mill achieved significant influent color reduction, the actual
effluent color reductions from the Highest Certainty items were less than forecast by the TRW The
projects implemented by Blue Ridge Paper that achieved the greatest effluent color reductions—the
hardwood brownstock washing improvements, and the acid sewer project—were not envisioned by the
TRW The Canton Mill met the color reduction goals in the 2001 NPDESpermit using a combination of
projects identified by the TRW and additional projects identified by the mill. The many color reduction
initiatives pursued by Blue Ridge Paper that went "above and beyond permit requirements"should be
recognizedfor both the effort involved andfor the success ofthose initiatives that worked.
Edits and specific comments follow. Details for edits are in our comments for specific sections of the
draft report above.
Mill process changes that reduce wastewater color are generally preferred to end-of-pipe treatment because they
may have lower capital costs and may benefit the mill by reducing operating costs. However,mills in
environmental regulatory jurisdictions with severe restrictions on their color discharges have been required to
implement end-of-pipe color removal technologies since the 1970's. Although the Tech Team continues to
maintain the highest priority for in-mill improvements such as process changes and optimization and increased
black liquor recovery, external color removal technologies must now also rise to a higher priority than
previously and be considered carefully in the mix of options for further controlling the color of BRPP's
discharges. These color removal technologies include both end-of-pipe wastewater treatment and treatment of
segregated wastestreams(i.e.,the CRP purge).
The Tech Team identified numereus the following strategies,focused primarily on in-mill process
improvements,that BXPP ean use to further reduce its effluent color discharges. na..st of these e._ategi..s were
identified by the Teeh Team and OtheFS dUFiHg the development of the 2001 permit. However,BRPP has
not adequately implemented them. These strategies are summarized below.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 20
1)On the pine bleaching line,irnplernent evaluate the use of peroxide fortification of the Eo stage
and decrease the target kappa factor as recommended in BEPER 2001 and 2006.
2) On the hardwood bleaching line,implement evaluate the use of oxygen and peroxide fortification
of the E stage and decrease the target kappa factor as recommended in BEPER 2001 and
2006.
3)
BRPP Comment-This is not technically possible. D stage hardwood is not recovered
in BFR.
4) Continue to improve the control of leaks, spills,and capture of black
liquor and other highly-colored wastewater:
• Move forward with the planned CRP sump and containment project to eliminate unplanned
releases of this highly-colored material to the mill sewer.
• Eliminate discharges of highly-colored wastewaters directly to the wastewater treatment
system. These wastewaters are generated during planned mill outages or diverted to the off-
line clarifier during mill upsets.Maximize recovery of these wastewaters and treat the
wastewaters that cannot be recovered to reduce color prior to metering them to the treatment
system.
5) Fully im estigafe the peten6a!&F hara..eed pulp e e 1.1.,aebing«......auee the f..al's effluent
as eembined at the
BRPP Comment-Dr.Liebergott does not recommend ZD technology for the Canton
Alill (page 18 of the 2006 Liebergott/GL&V report).
6) install Evaluate an additional oxygen delignification stage for the pine pulping bleaching line.
,A s-E192-use on the pine line is reduced, ineFease the ......,unt of har&,eed D stage filtrate
BRPP Comment—Increasing DI stage filtrate recovery is not technically possible. D
stage hardwood is not recovered in BFR.
7) Continue to evaluate the impact of the CRP purge on treated effluent color. i€RI eolef is net
sentaot evaporators.Also,if GR-P eelor is net removed in the tfeannent plant,AR4her investigate
Q92 tFeannent of the CRP purge to reduee its eeler prier to disehafge to the mill wastewate
treatment systera.
8) Continue to investigate strategies for improving the color removed by the Canton Mill
wastewater treatment plant:
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 21
leaks, spills, andlor diversions,to Feduee this affeet;
A.iiiiti Aa of polym@F and etl.er#eatmeSt ehemiealq to the highly ..elared...e..telyater diyeAed to
Effluent Color Variability
BRPP Comments—Several of the strategies listed are already used by Blue Ridge Paper and are areas
of continued evaluation and improvement. Other recommendations are impractical. Additionally,
several of the recommendations have been evaluated before and determined to be technically,
economically or operationally infeasible. The section should be deleted.
Blue Ridge Paper agrees that reducing the frequency and magnitude of high color events is an
important focus. BRPP also believes that we have demonstrated significant progress in this area. Our
primary color control strategy is to capture highly-colored materials in the process. During the current
permit term, we increased storage volume for black liquor materials by inter-connecting the pine and
hardwood spill systems. We made strategic improvements to spill sumps, conductivity meters and other
systems based on investigation of color events. With careful planning of outages, we minimize losses of
highly colored materials to the mill sewer.
The Canton Mill does treat highly colored material that is diverted to the spare primary clarifier. This
diversion has been a permit requirement since 1997(ref Low Flow Contingency Plan dated December
1, 1998). BRPP adds polyamine to treat this material in batch and then slowly feeds the chemically
treated material back into the wastewater treatment system. We also have action levels for adding
polyamine to primary treatment to treat highly-colored materials that are not diverted. These
recommended strategies are already in place and are effective.
Variability in daily color during the permit term has improved as outlined in the tables for primary
influent and secondary effluent color listed below. The two days greater than 100,000 lb/day in 2006
were related to a single process problem that has not occurred since CRP startup in the mid-1990s. An
engineering control for this problem is planned for installation in 2007.
The best effluent color performance occurs when the pulp mill is running steady at normal production
rates. Elevated color tends to occur when the pulp mill is slowed back or when bleached pulp
production rate is up and down due to equipment or other process problems. The greatest risk of black
liquor color losses and spills occurs during-process shutdowns and startup, and is the reason we do not
take outages during low flow/recreation periods as detailed in the Low Flow Contingency Plan
required by the permit. The idea of curtailingproduction during periods of low stream flow would likely
have the opposite affect desired.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 22
Table ZI—Primary Influent(Total Influent)Statistics by Year
-Mean True Relatiie" Percent
Dolor -Siandaid Nomber oV -,-Decreas6"Obs/day)," 'Standard -` ,-,Deviation- .' days>100,00k4i�omBaseline
Deviation-(S%, (Sb/niean %l-' Ab/da�-v�, (2001y-,_,
2001 57,725 20,619 35.7% 4 0.00%
2002 54,780 17,195 31.4% 7 5.10%
2003 55,550 19,424 35.0% 9 3.77%
2004 49,466 18,786 38.0% 8 14.31%
2005 45,175 22,297 49.4% 1 7 21.74%
2006 38,454 14,015 36.4% 2 33.38%
Table 2.2—Secondary Effluent Statistics by Year
a n Me u- Tr e Relative Percent
Coloi Standard = Number of Decrease',_
-St'iuidaia - "'De Deviation FromBaseline(lbs/day)' ation days>�10C�j()00
Deviation(SD) 4 (SD/meau,% lb/da (2001)
2001 42,676 10,925 25.6% 3 0.00%
2002 41,166 9928 24.1% 0 3.54%
2003 44,627 11,043 24.7% 1 4.57%
2004 41,463 * 32,568 76.6% * 4 0.43%
2005 39,092 10,092 25.8% 0 8.40%
2006 37,058 8959 24.2% 2 13.17%
2004'data were affected by historic floods in Sep 2004
Sinee BRPP installed the BFR preeess they have made gradual progress in redeeing theif annual aver-age
ElisehaFges.However, they eentinue to experieftee many days per year e�high seler disehafges. When these
diseharges eeineide with periods of low Fiver gow(t5Tieally in the late suffhmeF), they eaias@ an elevation in
P:ver eoloF that is notieeable to the eitizenswho use the river. Thus,reduotion of the impaets of the eelor
Eliseharged 49m the mill requires netjiast a reduptioB ofthe average aanual eelar diseharged but also a
reduetion of the daily mw6murn eelor diseharges. The Teeh Team identified qeveval strategies feF reduoing
daily maximmm eeler diseharges and their ifnpaet on the fiver. These strategies are listed bele :
pF@t+p,&t thp agid spy,&F tqwv to FAillimize and reduee the vafiabilky of the generat4oa of"seweF gone-a
eoler;"
• T-feat wastewater di-,,ei4ed to the eff line primary elafifief;
Segregate the highly eelefed wastewatef released duFing planned and unplanned diversiens and tfeat i
Yam'or to meten:ag it into the mill wastewatef tfeatment sysiefa;
Pfm4de additional eEraalizatien(inereasing the tank velume available to diNei4 mid tfeat 14ghly eelered
wastewatef);
T-Feat total mill effieefAusing pebfners or other vffistewater tFeatment ehemieals,paizfiettlaf4y dufin.g
periods e�hig4 inAtlent ARInr and./Ar law river iqow;
if the GRP puEge eoloF is Aund not to be removed in seeondary tFeatment,avoid Feleasing the GR-p
pur-ge dafing p@Fiadlq A 10*qtFZ.. flaw(aF tmek it 99 site),and
Gui4ail pulp produetion dufing periods of jew stfeam Aew.
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 23
Suggested Final Effluent Color Limits for Blue Ridge Pulp and Paper
BRPP Comments— The 2001 NPDES permit includes annual average and monthly average color limits.
The compliance period is every calendar year and every calendar month. The reporting period is
monthly based on daily composite samples. Waste load factors used for BOD and other conventional
water pollutants are not appropriate for color limits. In addition, the Tech Team has no responsibility
or mandate to determine water quality standards based on effluent limits for the mill; that activity will
be done by the state as part of the permit development process.
In the December 2005 report required by the NPDES and submitted to Alan Klimek of the NC Division
of Water Quality,Blue Ridge Paper stated its commitment to continued color performance improvement
and recommended an annual average secondary effluent true color limit of 39,000 lbs per day and a
monthly average true color effluent limit of 52,000 lbs per day. We continue to support an annual
average effluent color goal in the range of 32,000 to 39,0001bs per day.
Natffal Resetwees expand the eidsting pefmit limitr to k4slude maximum daily and maximum 3 0 day aveFage
disehafger,and not rely solely aft the amnial aN,@Fage limit to preteet the quality ef the Pigeon RWer. Daily an
Monthly peffflit lifnits'A'auld.he congi,4Bnt with lifnits for other pa,-ameteFr lin4ited in�Le present peFmit as,
-ther pulp and papef mills nationwide.
BRPP Comment—The preceding and following paragraphs are not only incorrect(i.e. the effluent
limits for the mill have always been at the end-of-the-pipe), they also include non-technical hyperbole
such as "consistent with--limitations for other pulp and paper mills nationwide", "critical point in
time"and wrongly implies that there has been a problem demonstrating compliance with permit limits
and that somehow adding additional parameters "will simplify and facilitate compliance monitoringMefeever,the T-eE;h Team also believes ".
0
ly that these erPient limitatianq gbould be applied a4 the And. of
pipe diseharge of the ff�ll to the fiver(not downstFearn of the diseharge at a peifil that ifieludes in stre
afid the inefeasingly impertant role ef fediiehig effluent variability aleng with the other reeeffmiefidatiens efth
Teoh Team going ferward 4om this eritioal point in time,this leeatien fer deffienstfating eemplianee with
peffait limitswill simplify and faeilitate eamplianee fneaitarii4g. End efpipe pemiit limits also will be
Memerandtun
,lo.:...,ti.... ..r♦H.,.... 1:...its ........e..te.i:..Table 1 ,1:........sed 1..,low.
,and
Table 3.Teeh Tea— R.Peommeaded Range of End of Pipe Color PeFMit Limits
a . _ - .
Anal-Average c . . . .. 32,BB9-te37990 ..
J94a 44 90Bte-3-1,890
Baily Malki�xn 83,840 to 96,910
Basis for-Reeommended Annual Average
BRPP provided daily measufamei#s seeendary effluent eeler(jb,qa59 fef e,,,efy day ift 2006. Examinatien ef the
lWday. BRPP repefted that this elevated eoleF diseharge resulted ffem"GRP slufry tarA-�everflem,for 294nm-
"Ep—
soufee,they v.eav vaauuvu as vaarw
i
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 24
e....,,.d average a.r 2006 ...:the..«1*a and July a.
36,695 lhide., rounds to 37,000 lb/day This e ,el e e lead is less than the an nnn lb/day
stiggested by BRPP Heweyer it is based on the mill's 2006 perfe,.. anee and does not:,.hide
..e1. sis of the BRPP Fail! di6ehaFges by the Tesh Team and ..there:n sti....art of the 2001
't t e 't s inter•.:^ ter e^ 1 32 nnn lh de..with t„ 4(1 nnn 1T.iday n
p^cxx -pci' .v c oc cr-boar::,ss.,v.,v-xv�vo�-�))4ar$iairgosijr.v,/vo.T. rco
deseribed eadief in this memo,BRPP has yet te implement several of these key preeess
T.«ele.,.eating the use of peroxide and oxygen Oerti C.eetion eF the Ea stage and deereasing the target
hiereasing reesNefy Bf blaek liquer leaks, spills, and ifftefifienal diveEsions(net simply metering these
xd-
TInstell:ng 2 stage a ygen deligni fi. tie.. n the pine line and r additional hardwood D stage
�-l`- 3cuxxxx]gZ .....t,..�:�j.V.... ....b.... Ga[.cxx cxx -r'l2xaV2Ia�j 6I36-rccovi.-Lnr��Qigvxvar[xm'v-vPoo .l[�agc
filtfates.
Deeause RAR h.. not b.f,./e,..fe.red these-pnoee,.e :—..,.e...ent.. the Teeh T-ei .e ....,.,.inmt.nds that the eoW
fps INS ueMlif te.,.... P«the L.....t....... ..........ee e.,/e«/:..,Ft ee,a6.,.e t..be 32 000 /A/d....
BRPP Comment—The following discussion, which suggests that color at the Canton Mill should have
daily limits based on BOD statistics for the pulp and paper industry, should be eliminated. There is no
need for daily color limits as discussed above. The annual and monthly color limits are protective of
the North Carolina Water Quality Standard.
Basis fOF Suggested Daily Maximum and 30 Day A-YeFage himits
The statistieal analysis used for the developffient of EPA's GlusteF Rules is daeumen4ed in Statkdeal Supper
development ef-, ameng other things,the variability faeteFs tha';A,AFA uged.tA paleulaw NSIDS F—pnns For the
Bleaelsed Pap@Fgade Kraft and Seda(BP-K) subeatagory. These variability faeters are repredueed in Table 4.
B Asa a-a
emm1eu fee.TTU.S. A BD 1997. Table 2 A
The BOD3:ariability sheaf.above shea above e developedusing daily g data f r the beet peFfaiiiiin
.
3 ivied\ nn=nil.s. The m ..:tering deterepresent the effluent c e......ell
daily leads and the log normal distribu fie., a the The Teoh Team believes that it is reasonable tA use the vafiability fastors developed fer the BPK DISPS for
measurements-.
The faetnfq weFe developed using faeniteFing data fiam-BPK-mills*,4 well opeFated seeondai�,
Celer, like BODs '
Geier,like T 0--
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 25
Applying 'e BQP4 *ariability FaCtOFS W the 2006 annual average daily 001eF disehaFge Fesults in the Allewing
naily MaAanimi 3 7,000 lt,ia.,..., 2.62—96 nnn l,.i a y
30 Bay Average: 3q,099 lb{day?i 1.400--51,900 m
(The Paily Maximum and 30 day avefage jimits for the lang teFm aveFage goal of 3-2,000 !Wday are derived in
Comparing the 2006 effluent menitoFiag data te these liffiRs finds that M,e days exeeeded the daily ffiweiffydm:
Jul. 7 104,504 ib
july$ 101,2-23
Ftffther, the 30 day avefage (51,800 lb/day) was exeeeded for apt3foximately eight 30 day averaging periods.
'u`JTp 62,275
AugtlS 33,399
5354
Comparison to Glatfelter Permit Limits
BRPP Comments—There are several mistakes in the evaluation of Spring Grove effluent color and
comparison to the Canton M111. The Table 6 comparison is not correct and should be removed from the
report. Our actual effluent color performance, using pollution prevention approaches, is similar to the
Spring Grove Mill.
Bleached pulp production, not paper production, is the appropriate normalizingfactor for comparison
between mills. Otherfactors that must be considered are the wood species and the mix of hardwood and
softwood pulp that is produced by a mill. The ratio ofpulp production to paper production at the
Glatfelter Spring Grove is lower than for the Blue Ridge Paper Canton Mill. On an air-dried ton of
pulp basis, the Spring Grove summer 30-day average effluent limit is equivalent to—22 Ibs/ADTBP,
which is close to the Canton Mill color performance. On an effluent color concentration basis,
performance of Spring Grove and the Canton Mill are similar.
The Glatt ltef inn mill in o_..:.g n..eye PA(TMPES ...Fmit number u A nnnoo<n)is like noun .. older
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 26
290 mgq� (daily maikiffww).Winter dissbarge limits"o lower. The mill's eweFage daily diseharge is 119
MQD and it preduees an aN,@Fage of 1269 U.S. tons of paper per day.44aere�re, en a produstion noFmali-zed
basis,the Glatfelter mill's average monthly permitted discharge is 11 lbAon paper(22 lbAon daily maxiaw*.
BRPP preduees appreikimately 1640 US tons pef day. Applying the Glatfelter lifnits to BRPP wetild resel
belem,the Teeh Team feeeffffnended fange ef end of pipe permit liffiRs feF BR42P (4 4,900 to 5 1,800 lb/day
Table 6. Comparison of BRPP Reeemmended Range of Permit Limits and Glatfelter Permit Limits
References
Amwakk%fege 19.5 e 22 cc
an day Average 44 :17 to 1i
Daily P 4ffid�m 2-2 51 to-59
Andritz Ahlstrom Sales. 2001. Prayn's Island Technical Center Report 2001-069 Part 1. Laboratory and Lo-
Solids Cooking with O-Do-Eop-D Bleaching Sequences on Softwood Furnish from Blue Ridge Paper,Blue
Ridge,NC.Part 1 Softwood Results.Part 2. Hardwood Results. (December 6).
Audet,Andre,Michel Faubert„Zhi-Hua Jiang and Barbara van Lierop,PAPRICAN. 2003. Technical Service
Contract Report TSC-7447. Bleaching Evaluation for Effluent Colour Reduction.Prepared for Blue Ridge
Paper Products,Inc.,Canton,NC. October.
Blue Ridge Paper Products,Inc.2005. Chloride Removal Process (CRP) Color Reduction Technology
Assessment. (March).
Blue Ridge Paper Products,Inc.2006a. Color Compliance Report: Canton Mill. (May).
Blue Ridge Paper Products,Inc.2006b.Blue Ridge Paper response to questions from the Technology Review
Workgroup (TRW) that were e-mailed on 28 Nov 2006. (December 19).
Blue Ridge Paper Products,Inc.2007a.Response to additional questions for BRPP about data provided to
TRW Don Anderson e-mail dated 5 March 2007(March 15)
Blue Ridge Paper Products,Inc.2007b.March 19, 2007 Additional Data Required from BRPP(March 28).
Bodien,Danforth G.2007. Site Visit Report,Blue Ridge Paper Products, Canton, North Carolina. (April
2007)
EPA Tech Team. 2001.Memorandum to Technology Review Workgroup. "Additional Color Removal
Technologies and Their Economic Impacts on Blue Ridge Paper Products, Canton,NC."(July 25,2001).
Furjanic, Sean M. 2007,Water Quality Protection Report,P.H. Glatfelter Company, Spring Grove
BRPP Comments on Draft EPA"Draft Team"Report, 10 May 2007,Page 27
Borough and Jackson Township,York County,for the Renewal of NPDES Permit No.PA 0008969.
PADEP Southcentral Regional Office(draft,February 2007).
GL&V Pulp Group,Inc. and Liebergott&Associates Consulting. 2001.Bleach Environmental Process
Evaluation and Report.Prepared for Blue Ridge Paper Products,Inc. and Clean Water Fund of North Carolina.
(June 8)
GL&V Pulp Group,Inc.,Liebergott&Associates Consulting. 2006.2006 Update: Bleach
Environmental Process Evaluation and Report.Version containing manufacturer's proprietary
information. Prepared for Blue Ridge Paper Products,Inc. (July 7,2006)
Munro,Fred and John Griffiths. 2000. Operating Experience with an Ozone-based ECF Bleaching
Sequence,Proc.International Bleaching Conference, Halifax, Canada, 2000.TAPPI Press.
Technology Review Workgroup,2001.Memorandum to North Carolina Division of Water Quality and
the NC Environmental Management Commission's NPDES Sub-Committee. "Additional Color
Removal Opportunities,Blue Ridge Paper's (BRP) Canton,NC Bleached Kraft Paper Mill, 2001
NPDES Permit Renewal." (August 3, 2001).
U.S. EPA, 1997.Statistical Support Document for the Pulp and Paper Industry: Subpart B. (November)
Agenda
EPA Tech Team Report Discussion
Blue Ridge Paper Products Inc.
July 25,2007
I. Introduction
II.Areas of Agreement:
• CRP color studies
• Leaks and spills
• Outage planning
• Acid sewer color studies
• Color treatment
• Spill capacity
III.Unresolved Issues:
• Second-stage oxygen deliginification on the pine fiberline
The requirement to install second stage oxygen delignification is not appropriate at this time.
Laboratory studies and engineering analysis to evaluate the color reduction benefit and cost of
2-stage OD on pine against the criteria of technical, economic and operational feasibility are
appropriate.
• Peroxide fortification on the pine and hardwood fiberlines extraction stages and hardwood
extraction stage oxygen fortification
The 2006 Liebergott report recommends evaluation of peroxide fortification of the pine Eo stage
and evaluation of oxygen and peroxide fortification of the hardwood E stage. Consistent with
the 2006 Liebergott report,full-scale trails to evaluate the color reduction benefit of EoP on
bleach plant filtrate and final effluent color are appropriate.
• Further trials to evaluate potential of ozone/chlorine dioxide(ZD)bleaching for the
hardwood fiberline
We do not agree with the Tech Team recommendations on ZD. Dr. Liebergott does not
recommend ZD for the Canton Mill.We believe that further evaluation of ZD bleaching is not
warranted at this time.
• Daily color limit
Daily color performance is important and we continue to work hard to minimize color variability.
However, a daily color limit is not appropriate.Also,under both EPA and North Carolina water
quality procedures for applying aesthetic water quality standards a 30-day average is listed as the
appropriate time period.Therefore,a monthly average limit is appropriate.
IV. Other issues and concerns
V. Follow-up communications
JUL-25-2007 WED 12�40 PH ERG, INC FAX NO. 17032637280 P. 01
J
WERE
FAX COVER SHEET
Date: 7
Time: d,t 3 5 A M
TO: UDA AwN�w Y%,
ORGANIZATION: EpAZE AJ
FAX NUMBER:
ACCOUNT NO: O f7Z -•Utl aS Of D
FROM: 5 « tSi c-kne.<<
Eastern Research Group,Inc.
14555 Avion Parkway, Suite 200
Chantilly,VA 20151-1102
(703)633-1600(tel)
(703)263-7280(fax)
(703)263-7281 (fax)
IL pages follow this cover sheet
- aP* weJd& 4e Fe1A - sae �s
w 1Ik CR&f;4k5 skam,- etec c W4 ,
-IA Imes 4nc BRtf cay A-s a S. %�e- Warms -b k^M
wham -r Rw �uil4 cam- �4hn �ech�"pttu� ,-e�x�
Ls� YM I�JhoceJ i F ►T V)0- is 1\0+ redA"Je OLAtA�
l' 11 oaot-a snav& eAvaA vd - , Qu..
JUL-25-2007 WED 12:40 PM ERG, INC FAX NO. 17032637280 P. 02
AM&A W^
SSr.-16Z 1
-B),c.
--------------
---- ---------- ...........
-� L
,wed aft
Q/e4kq�—
p0
4 Ll
5 �
------------------
AJcAa y ma�t; -k)ma& wz"u �fkow 6;
ATLOA a;t cn� p b-a
-----------
Al Ale.= b
Y0
------------rxo"' WOV4---- L-V-
-------------
tmA'. Y, 4rlol; icao4'
nlo'
L44�/ Al
.................
VIM)-
.......-------------
c-
to. IK04 ""'q
JUL-25-2007 WED 12:41 PH ERG, INC FAX NO. 17032637280 P. 03
7
.......... Qm1 ---------------
C�'�
---- -----------
-----------
tz--a
.. " Ej Iv [cc* cc4m 0 Ec4d'c'-'
cn. .......... ..... .... ....... ......—
8-3- .--5S .......
----------
u�- /ro
V1
L 4 M'd
5yj-hoo /(�l �la'p nLI
-Tv
D -iwz-3 -------
.... ---off-
Q�6fa, '�r
A6-
.............
............
C'Lo�
qdL
PA...
JUL-25-2007 WED 12;41 PM ERG, INC FAX NO, 17032637280 P. 04
-5C4U 64aa,)
kow zcfulj � ko�
------------
------------
..................
/Do oo "
7.,7
Ow a' `vc ------- -- -----
tj
-----------
Soo
'te Wr
s C,
............
P44/L
-4
........... ..
JUL-25-2007 WED 12:41 PH ERG, INC FAX N0, 17032637280 P. 05
_.......... . . .. . .... .. . . ._._4P:�.c.Y.UQ. -C1,n,-C.((.X�QP�..
oC'....
.. .... ...... ... . . ....... . ......... trc-Ll- J -fr¢l,Al_ Cl}PllP_.. Y�..-4{'J'[L�{�___._�'�.
i _...,
f `J/ de
---- )^t^a°1 -..j ��^o:ar..h'.��-ie?lw�_a�± :b-c✓` 0 deai<v, ✓y�qa.9L L.d-* 3 -------------
: c.'1�c�-----------
-------'---•-----' I" 2D '�t�L•w ' ✓Hd � k >� uv., cF�l�e�eo� co-i�
,...,..1��.'-�- add_ �k, r..,..� ���a.��J-.•�-J-._ .._,._..
ala5_...-
_.... __.:.- - -- ,___......... -- --- — hauu� 1 d� a�.{ y p 1 tr✓JoiU2 p
--- - . ..:;._&yLtf'S '�Vtip .. WC�G "c0���{ P�t �Q(C��1L .5�lcnU'✓gRL nvm�
......-.....
........
n '<x'n
c ; mp^ a.Re�-,.1 f..lool 42 c uo, rf ..l. creccX ,x�t
JUL-25-2007 WED 12:42 PM ERG, INC FAX NO, 17032637280 P. 06
4L
416 dA&e'�-1
..........
eAff-�
ICILIO
RT '4_6AL Ag L.
--- —------
.......... ------ -----—------.............. .......
------- -----
--------—----------- ......
�
JU[-25-2007 QED 12|42 PH ERG/ {@C FAX NO. 17032637280 P. 07
....^-'^-~ .^'.^^_--=`- `'--^'^^^-~�_�__�
�
_.�.------_---'-_-'-_-�
�
�
-'-----_---`-�-.`^_-_-_-___--^--__-.--
�
�
-`--_-`~_-''---_�--_`'_----_--'---'--_-... -.....
-........ ...... -` ... ..........-'-' -----........~`'_-_'_--_ --
_--- -- � ^�---�-----�� -',---------`----_-_-.---'_- --_--- � '^ --
--,--___�--`-�---_-..........-'`'=. -`-__' -'=�-,-��^- _'--_�'--_-``~_ .-'------
---------- '--^-- ---_-.--,
� ��_-__-_--_-------__--_-_'___-_ -- -___-_-_-_''-------
/
-'----------^ ------ ����- �����`
----^`---^-_-'_-�-- - _'~----�_- _-' -^-_`--�-_-,_---~-- -- ---'_�-__' -'__�~-� --_-� _-��-'`
�_--_`-.,_--�-_-_-'_--_'�'-_- _--_-_-----_'''-.-'--___---,'------'--'-__--_'--_------_-�'--_--.-_-
! �
--_--_�--^-`-_-_'-_--_ �--_�_^-_-'_'-_------_�-_---''___---�'-`----', '---'-'--' -,_-_ _ ' -�_--'
� ----'-- -_-'------- _-----'-_-'''----_ -_'---'-'--__-'-_'' ---- _--_'----_
�
___- --'�
-`-
�
-. . ..........'--'-'�'--- '--� - ...... -----� _- � -^- - - `_-- --^----_-----'_----'~� ---_� �` ' -~,�
|
i
Blue Ridge Paner Comments July 25,2007
MEMORANDUM
DATE: June 29,2007 (2nd Draft Final,DO NOT DISTRIBUTE until this header is
deleted)
SUBJECT: Color Removal Strategies For Blue Ridge Paper Products,Inc.
FROM: EPA Tech Team' -
TO: Technology Review Workgroup
Purpose of this Review
Blue Ridge Paper Products Inc. (BRPP)has requested renewal of their National Pollution Discharge
Elimination(NPDES)discharge permit NC0000272. This permit allows discharge to the Pigeon River of
industrial and other wastewaters from the pulp and paper mill BRPP operates in Canton,NC. In their
May 2006 Color Compliance Report(Blue Ridge,2006a),BRPP requested that North Carolina reissue
the NPDES permit for the Canton Mill with an annual average effluent color limit of 39,000 lbs per day.
This is a reduction from the current 42,000 lb per day annual average limit which became effective
January 2004. BRPP's requested limit was based on the mill's 2005 annual average effluent color,
39,000 lbs per day.The Technology Review Workgroup requested that the EPA Tech Team support
EPA's review of the color limits included in the draft permit renewal developed by North Carolina
Department of Environment and Natural Resources. The Tech Team last evaluated color discharges at
the mill in 2001 (EPA Tech Team 2001).
Members of the Tech Team visited the Canton mill on February 8,2007 to observe and collect
information and data on the status of technologies implemented and color discharges at the mill. This
draft memorandum incorporates analysis of data from that visit and other data provided by BRPP.
This memorandum reviews the process improvements previously analyzed by the Tech Team,the status
of their implementation at BRPP,and identifies additional color reduction activities that BRPP could
implement during its next permit tern. The memorandum also reviews the variability of the mill effluent
color discharges and suggests final effluent color limits for incorporation in the revised permit.
Background and History
BRPP operates a bleached papergrade kraft pulp and paper mill in Canton,NC,which it purchased from
Champion International Corporation in May 1999.BRPP is owned 55 %by KPS Special Situations Fund,
L.P. and 45 %by the employees through an employee stock ownership plan?BRPP announced on June
14,2007 that it is being acquired by The Rank Group. The New Zealand-based Rank Group has
considerable holdings in the paper industry. Rank Group recently bought Evergreen Packaging,the
former North American beverage packaging division of International Paper. Operations at the mill began
in 1908,but the mill has been extensively modernized,with the most recent major project completed in
1993. The mill currently operates a 810 tpd hardwood pulping line and a 600 tpd softwood(pine)pulping
'EPA Tech Team members are Ahmar Siddiqui, EPAIEAD;Karrie-Jo Shell,EPA Region 4;Donald Anderson,
EPA/FAD;Betsy Bicknell,ERG,contractor;Neil McCubbin,subcontractor to ERG;Dan Bodien,subcontractor to
ERG.
z BRPP announced on June 14,2007 that it is being acquired by The Rank Group. The New Zealand-based Rank
Group has considerable holdings in the paper industry. Rank Group recently bought Evergreen Packaging,the
former North American beverage packaging division of International Paper.
Memorandum
May 10,2007
Page 2
line. After cooking,pulp from each line is further delignified in single-stage oxygen delignification(OD)
systems,both installed in 1993.Hardwood pulp is subsequently bleached with a DED sequence(BRPP
stopped adding oxygen to the hardwood E stage after 2001).Pine pulp is bleached with a DEoD sequence
(BRPP stopped adding peroxide to the pine E stage after 2001). Target brightness is 86 ISO,an increase
by one point from the target brightness in 2001. Up to 80%of the filtrate flow from the pine bleach line
is retuned to the recovery cycle using the unique bleach filtrate recovery (BFRTm)process,developed by
Champion. This 80%closure rate is an improvement from 2001,when the pine line closure averaged
73.7%.Up to 25%of the hardwood line bleach plant E-stage filtrate flow is also recovered. BRPP
produces 300,000 tons per year of uncoated paper including offset,tablet,and envelope grades. The mill
also produces 275,000 tons per year of bleached paperboard used for liquid packaging and paper cups,
including FDA-approved grades for milk and juice cartons.
Basic tenets of this review,which are consistent with the Tech Team's approach beginning with the
original Settlement Agreement(with Champion International)and the 2001 TRW review with BRPP,are:
• The first and highest priority again is focused on available in plant process changes and best
management practices(BMPs)as the most cost-effective approach to color reduction in order to
maximize the likelihood of success
•• Process changes deemed to be of highest and reasonable certainty,thereafter lowest
certainty technologies
•• BMPs that hold promise to further reduce generation and discharge of color
• After it is clear that in plant process changes and BMPs alone may-are not be sufficient,in plant
segregated stream pretreatment and end-of-pipe treatment options will be are carefully considered
The Tech Team recognizes that very difficult and unusual circumstances occurred in 2004,with back-to-
back historical flooding of the Pigeon River and associated significant costs($39 million) for repairing
damage and replacing equipment at the BRPP mill after 21 days of being out of operation. The Tech
Team also recognizes that significant additional expenditures have been made over the last permit term
for environmental projects($25.9 million),the largest portion for air pollution controls(e.g.,MACT I and
II compliance)and including color reduction projects($5.9 million),all building on previous projects and
expenditures. All of this has occurred during a continuing period of industry-wide transformation,
capacity shrinkage, and mill closings. Nonetheless,the BRPP mill management has committed to
surviving these challenges. The Tech Team notes that these efforts and expenditures have resulted in
substantial and commendable progress made to date by BRPP through expenditures for additional and
improved process and related technologies,treatment technologies, and management practices which
have reduced the long term average color discharges. It is also noteworthy that some of these
technologies and practices have been in addition to those identified in the previous Tech Team/TRW
reviews.
Process Improvements Previously Analyzed by Tech Team
As a result of its 2001 analysis of the BRPP mill discharges and operations,the Tech Team identified five
mill improvements that itjudged capable of reducing the discharge of color in the mill effluent. BRPP
and its consultant also identified several process optimization projects that would reduce color discharges.
The status of BRPP's implementation or evaluation of each of these improvements is summarized in
Table 1, discussed below.
Memorandum
May 10,2007
Page 3
Table 1. Review of Process Improvements Identified in 2001
Predicted Color
Reduction(Final Color - -
Eftlueut,'lbs/day - - Reduction -
Improvement -arm.avg).- Implementation Status Achieved - Comments = _
Process 1,400 Predicted reduction in BRPP consultant recommends
optimization C102 use(27%on HW, reduced C102 use and evaluate adding
projects 17%on SW)not made. oxygen and peroxide to E stages to
C102 use increased on reduce color generation(GL&V 2006,
SW,decreased 5%on HW pp52-53)
BFR reliability 1,000—1,200 Pine line filtrate recycle Total Further improvements to BFR
improvement increased from 74 to 80%. reduction from reliability may not be feasible
2001: 6,000
Improved black >5,000 Spill collection lb/day,annual Additional improvement to black
liquor leak and improvements made; average liquor control possible: further
spill collection untreated color(influent to eliminate overflows and pretreat
control(BMPs) treatment)variability diverted high-color wastewater.Also
unchanged improve control of losses of white and
green liquors to reduce or eliminate
sulfide-based color
Ozone/CIOr 3,000-6,400 Incomplete Inadequate
stage for evaluation(lab studies);
hardwood line not installed None evaluate ozone bleaebing;BRPP
No further study warranted.
2"a Stage OD for 1,100—1,400 Incomplete evaluation(lab
pine line studies);not installed
None
fellowed by implementation.
Laboratory studies and engineering
analysis for cost and color reduction
benefit me required.
Color treatment 5 2,750 Evaluated(lab studies); CRP purge stream now averages
of CRP purge not installed 8,745 lb/day.Additional l%ul1-seal,
stream testing fer-extended-peried to
determine what fraction of this color
is removed in the W&TP should be
None undertaken. Any of this color that is
not removed in W WTP should eeuld
be-continued to be evaluated for
reduction through-redueed-vAA
pretreatment. Need further study of
strategies for reducing color released
in the purge.
Memorandum
May 10,2007
Page 4
Process Optimization Projects
In 2001,Liebergott&Associates and GL&V Pulp Group prepared an extensive analysis of the mill
fiberlines that included identification of options for effluent color reduction. Dr. Liebergott was retained
for this work by a coalition of environmental groups. This report,Bleach Environmental Process
Evaluation and Report(BEPER)(GL&V 2001)presented 16 recommendations for incremental
improvement to fiberline operations. These included suggestions for improving process control,OD
performance,reducing the amount of C102 used on both bleach lines, and enhancing the extraction stages
on both bleach lines with oxygen and/or hydrogen peroxide. The BEPER recommendations were
evaluated, and BRPP implemented those deemed to be technically,economically,and operationally
feasible.
Process control and OD performance have been improved. However, on the hardwood line,the Di stage
kappa factor has not been decreased as previously recommended and use of oxygen in the extraction stage
was discontinued. As a result,the total CI02 charge on the hardwood line as of May 2006 was only 5%
less(not the predicted 27% less)than the charge in 2000. Similarly,on the softwood line,the D, stage
kappa factor was not decreased as previously recommended and the use of peroxide in the extraction
stage was discontinued. As a result,the total CIOZ charge on the softwood line as of May 2006 was 11%
more(not the predicted 17%less)than the charge in 2000. While BRPP has noted the increased portion
of higher brightness pulps since 2001 as a reason for these changes, among other trade-offs cited,the
Tech Team believes these changes may-ere still be viable and should remain as options for additional
color reduction.These process modifications should continue to be evaluated and
considered items of reasonable certainty.
In preparation for renewal of their NPDES permit,BRPP retained Liebergott&Associates and GL&V
Pulp Group to analyze their fiberlines,review the implementation of the 2001 recommendations, and
identify further options for color reduction. In the 2006 update to the BEPER(GL&V 2006),the
consultants repeated their recommendations for BRPP to reduce the kappa factor in the D, stage of each
bleach line and evaluate the replacement of the C102 bleaching power with oxygen and/or peroxide in the
extraction stages. Increasing the operating temperature of the hydrogen peroxide extraction stage to
reduce effluent color also has been identified as a viable option to evaluate during this process.
In addition to reducing the color of the bleach plant effluents,use of a lower kappa factor means BRPP
would use less CIOZ to bleach,which will reduce the total chloride content of bleach plant filtrates.
Lower chloride content can also be expected to make it possible for the mill to reduce the CRP purge flow
and the color it contributes to the mill effluent. The Tech Team also believes that lower chloride content
could will make it possible for the mill to recycle more filtrates within the hardwood fiber line,
notwithstanding BRPP concerns(e.g., adverse effects to washing,carryover to D stage,possible increased
chemical usage and color). Further,it also may be possible to introduce a portion of these these
hardwood filtrates to the BFRT process,which to this point has been dedicated solely to recovering
softwood fiber line filtrates,and reduce the total colored filtrates discharged to the sewer.
BFR Reliability Improvement
As reported during the Tech Team's February 2007 site visit,BRPP spent$1.5 million over the last five
years to improve the reliability of the BFRTnl system. These expenditures included replacing tank liners
for the three existing sand filters(now high grade stainless steel),replacing and/or upgrading existing
process piping with piping with improved metallurgy,installing a fourth(new)multimedia filter, and
installing a third ion exchange softener. With these improvements in more reliable metallurgy and unit
process redundancy,BRPP has increased the BFRTm closure rate from 73.7%in 2001 to 79.2%in 2006.
Mill representatives stated that the present closure rate of approximately 80%of the pine line bleach plant
Memorandum
May 10,2007
Page 5
effluent represents the maximum amount attainable without incurring unmanageable corrosion and
scaling problems. Further increases in closure and increased recycle rates above 80%may be possible but
are likely to be a significant and very difficult challenge with the corrosion and scaling problems
associated with current mill bleaching filtrate chemistry. This is because scaling from hardness minerals
accelerates above closure rates of 80%(Bodien,2007).
Improved Black Liquor Leak and Spill Control(BMPs)
BRPP continued efforts intended to improve its management of leaks, spills,and intentional diversions of
black liquor over the last six years. These efforts include:
• Interconnecting the pine line and hardwood line spill collection sumps so that tankage in
either line can be used interchangeably for spills;
• Repositioning sewer conductivity probes from sumps to in-line to more accurately and
reliably identify high conductivity wastewaters;
o Diverting up to one hour of total mill flow to off-line 1-million-gallon clarifier,during
high color releases,thus providing some equalizing of color discharge to wastewater
treatment plant;
a Prior to process line outages, improved prior planning for capturing high-color process
liquors and black liquors and better managing their timed release to treatment system;
• Continued operator training; and
• Implementing two hour testing for color at the W WTP with one hour testing during
outages or semi-annual shutdowns.
BRPP has stated that these improvements in BMPs have resulted in reduced color variability in in uent
to the Primary treatment unit,which BRPP asserts is the best measure of color reduction through in
plant changes and BMPs. Data provided by BRPP are presented in Table 2 and depict Primary(total)
Influent Statistics by year.
Table 2. Primary(Total)Influent Statistics,by year
Mean Relative "
_Trot Color-" 'Standard _° Standard Deviation Percent Dcetcasc
"s(Ibslday), Devianlon_(SD). ' '(SD/mean_%) fromMaseline.(2001)
2001 57,725 20,619 35.7% 0.00%
2002 54,780 17,195 31.4% 5.10%
2003 55,550 19,424 35.0% 3.77%
2004 1 49,466 18,786 38.0%
2005 45,175 22,297 49.4% 21.7%
2006 38,454 14,015 36.4% 33.4%
Primary effluent,which is the total load to secondary treatment,is directly measured with a composite
sampler. Data provided by BRPP and compiled by the Tech Team are presented in Table 3 showing the
mean,standard deviation,and relative standard deviation(standard deviation as a percent of the mean)by
year,for the years 2001 to 2006. Prior to 2006,the acid sewer mixed with other mill wastewater prior to
Memorandum
May 10,2007
Page 6
the treatment system,resulting in"sewer generated color." As of January 1,2006,the acid sewer was
separated from the other mill sewers and now enters the treatment plant after the primary clarifier. The
primary effluent sampler is located downstream of the mixing point of the primary clarifier overflow and
acid sewer.Thus,the statistics for 2006 presented in Table 3,represent the mixture of primary effluent
and acid sewer, including any color generated from the immediate mixing of the two streams. It is
possible that additional color generation from this mixing occurs after sampling within the secondary
treatment system with additional time for any chemical reactions to proceed to completion.
Memorandum
May 10,2007
Page 7
Table 3. Primary Effluent(Influent to Secondary Treatment)Statistics,by year
Mean Relative
-
True Color Standard- Standard Deviation- Number of days
_ Obs/day) Deviation(SD). _ (SD/mean%) >100,0001b/day
2001 62,008 19,561 31.5% 13
2002 59,956 18,680 31.2% 16
2003 59,646 18,468 31.0% 10
2004 65,206 26,674 40.9% 40
2005 63,838 24,158 37.8% 28
2006 65,512 25,427 38.8% 36
Table 3 presents the number of days for which the primary effluent color exceeded 100,000 lbs/day.
Comments received from BRPP assert that primary effluent is not the most appropriate measure of
progress. BRPP further stated that color loads to the primary treatment system have been reduced. In
spite ef impFevements made ever the last pemik tefm, ineluding the aeid sewer relaeatien, all rlearly
inade in good faith by BRI111,the TeehTeain Fespeetfully disagFees with these asseFtinn; Assh—g-y.'ain
Table 3, aAlailable data does net appear to suppeg that theFO has been pEogress in FeduGing the&equeney
BRPP provided the Tech Team with notes describing mill events that were related to high primary
influent effluent color in 2006(Blue Ridge 2007a). These events included,among others,planned mill
outages,unplanned outages, a CRP slung tank overflow,and a release from the evaporator related to an
equipment failure. BRPP does not track and did not report one single event or one single type of event
that caused high color in the primary effluent. Further,not all high primary effluent color resulted in a
high final effluent discharge(e.g.,BRPP reports that color associated with high turbidity is effectively
removed in the secondary treatment system). However,from analysis of the data,the Tech Team
concludes that the Canton mill can further reduce primary effluent loads through continuing efforts to
minimize unplanned spills and leaks and intentional discharges of high-color streams during fiber line
disruptions. BRPP has reported some success through recent efforts in detailed scheduling of planned
outages and contingency planning for unplanned outages and continuing efforts to minimize process
operation variability. The Tech Team notes that these efforts,though challenging,must be further
developed and consistently implemented wherever possible to minimize high color discharge risks all
agree are associated with these fiber line disruptions,both planned intentional and unplanned
unintentional.
Data provided by BRPP,which
shows that secondary effluent has improved significantly over the past five years,are presented in Table
4. This shows that reliance on primary effluent information and other data collected within the Mill's
biological treatment system is not indicative of the true facility performance.
Table 4 Secondary Effluent Statistics,by year
Mean T Relative
True Color Standard =+ 'Standard Deviation Number of days .
,ObsMay) Deviation(SD) (SD/mean%). >100,000 lb/day`
2001 42,676 10,925 25.6% 3
Memorandum
May 10,2007
Page 8
2002 41,166 9,928 24.1% 0
2003 44,627 11,043 24.7% 1
2004* 41,463 32,568 76.6% 4
2005 39,092 10,092 25.8% 0
2006 37,058 8,959 24.2% 2
* -2004 data were affected by historic floods in September 2004
Ozone/Chlorine Dioxide Stage for the Hardwood Bleach Line
BRPP engaged the Pulp and Paper Research Institute of Canada(PAPRICAN)to investigate potential
modifications to the hardwood pulp bleaching process and determine effluent color reduction that would
result from these modifications.The PAPRICAN report entitled"Bleaching Evaluation for Effluent Color
Reduction"(Audet et al,2003)was provided to the Tech Team in early 2007.
PAPRICAN stated that their objective was to evaluate whether modifications to the hardwood bleaching
sequence at Canton specified by Mr.Johnnie Pearson(BRPP process engineer)could"generate effluents
with a color reduction target of 25%."
BRPP provided PAPRICAN with oxygen delignified hardwood pulp.PAPRICAN bleached this pulp in
their laboratory, investigating various combinations of chlorine dioxide, ozone and hydrogen peroxide.
The report concluded that ozone could replace some of the chlorine dioxide used in bleaching(known as
a"DZ"stage),while producing pulp of equal or slightly better quality than the control sequence. This is
consistent with the open literature on ozone/chlorine dioxide combinations.
PAPRICAN also concluded that replacing some of the chlorine dioxide in bleaching with ozone increased
the color in the bleach plant effluents. This contradicts the experience reported by Domtar(formerly E.
B.Eddy) at their Espanola mill,where a dramatic reduction in effluent color was observed when that mill
installed an ozone pulp bleaching system(Munro and Griffiths,2000). The Tech Team notes that the
Espanola mill has no color discharge limits and it installed ozone to reduce its bleaching costs.
The Tech Team found significant deficiencies in PAPRICAN's analysis of the laboratory results. The
2006 Liebergott/GLV report agrees with this finding. The concentration of color in the effluent for each
stage was added to obtain the total concentration for each tested bleach sequence. This approach neglects
two points:
1) Mixing effluents produces reactions that may increase or decrease the concentration of
color in the combined effluent.
2) Filtrate volume affects the measured concentration(e.g., lower volume will concentrate
the filtrate to a higher color). PAPRICAN's report does not present the filtrate volumes,
so a comparison of concentrations may be misleading.
Liebergott, et. al.,reviewed the PAPRICAN report and identified the same deficiency with regard to
effluent mixing. They also noted that the quantity of chlorine dioxide added to the ZD stage was too
high,which would result in higher effluent color(GL&V 2006,p 18).
In response to EPA's question about how the PAPRICAN results were used to estimate impacts on final
effluent color,BRPP responded:
Memorandum
May 10,2007
Page 9
The pilot study reactors were batch and not continuous acid did not include filtrate
recycle. The pilot studies included bleach stage filtrate color concentrations, but there
was no filtrate flow rate data from which to calculate a production-normalized filtrate
color mass. For these reasons, we did not attempt to calculate secondary effluent color
impacts using pilot study data for individual bleach plant color streams.
In the PAPRICAN study, the individual bleach stage colors were compared directly
and in total. By both means, the color of the ZD stage was higher than the baseline
DEoD. It is also well known that when individual bleach stage filtrates are mixed the
resultant effluent color is very difficult to predict. (Blue Ridge,2007a)
The Tech Team concludes that BRPP's investigation of the potential for hardwood pulp ozone bleaching
to reduce the mill's effluent color was incomplete inadequate. The Tech Team believes that further
laboratory trials,designed and implemented to provide reliable information on the production normalized
color generated by each stage and by the effluents as combined at the BRPP mill,are not needed to
evaluate this technology.
Second Stage Oxygen Delignifcation for the Pine Bleach Line
BRPP contracted with Andritz/Ahlstrom("Andritz")to study,among other things,the addition of an
additional stage of oxygen delignification in the softwood bleach line at Canton. Andritz,a well '
established supplier of pulping and bleaching technology and equipment, maintains the Pruyn's Island
Technical Center,which conducted the tests. The Andritz report, entitled,"Laboratory CK and Lo-Solids
Cooking with O-Do-Eop-D Bleaching Sequences on Softwood Furnish from Blue Ridge Paper,Blue
Ridge,NC"(Andritz Ahlstrom Sales 2001)was provided to the Tech Team in early 2007.
The text of the Andritz report states that`BRPP requested the sales department of Andritz-Ahlstrom to
investigate alternative methods of pulping and bleaching that would reduce their waste products and
improve the quality of their hardwood and softwood paper products." As indicated by this statement of
work,testing of second-stage oxygen delignification was a minor part of the work Andritz did for BRPP.
Further,very few of the data in the report are useful for analysis of effluent color improvement.
Specifically,Andritz compared single stage and two stage oxygen delignification of samples of pulp
provided by the BRPP mill. The tests showed that a second stage of oxygen delignification could reduce
the kappa number of the unbleached pine pulp by 22%.
Andritz did not bleach the pulp after the two stage oxygen delignification,so the report provides no
information on the impact of the additional oxygen delignification stage on effluent characteristics,
including color.
In response to EPA's question about how the Andritz results were used to estimate impacts on final
effluent color,BRPP responded:
The Andrttz Ahlstrom study of second stage 02 for pine shoved delignification
ranging from 42.7%for the single stage, 48.6%for the 0-0 and 55%for the 00 stage.
Willi improvements that BRPP made on the existing single stage pine 02, the
delignnification increased from 40% to 45%. Willi BFR in place on the pine fiberline,
the effluent color reduction from this improvement in 02 delignnification has been very
difficult to identify. The inability to quantify the effect on effluent color of improved
02 deliginffication made the capital cost to install a second stage 02 on pine not
economically feasible. (Blue Ridge,2007a)
Memorandum
May 10,2007
Page 10
BRPP's analysis of the benefits of adding a second oxygen delignification stage to the pine line is
incomplete. BRPP's consultants note that even with the percent delignification currently achieved on the
pine line, a second stage could achieve an additional 20 to 25%delignification(see GL&V 2006,p 118).
By making a 20 to 25%reduction in the kappa number of the pine pulp before bleaching,20 to 25%of
the colored material currently discharged from bleaching to the BFRTM could potentially would be
recovered and burned in the mill's recovery boiler. In addition, the quantity of chlorine dioxide and
caustic required in bleaching would be reduced by approximately 20 to 25%,reducing the load on the
BFRTM system. If actually effective tThis may would allow an increased proportion of the bleach filtrates
to be recycled through the BFRTM system. BRPP believes that a kappa number decrease or
delignification efficiency increase would not necessarily equate to a similar decrease in color. Although
with the information available it is not possible to rigorously calculate the benefit of this change,the Tech
Team would expect in the order of 1000 lbs/day reduction in bleach plant color discharge.
Liebergott,et. al., (GL&V 2006)estimated that implementing a second oxygen delignification stage on
the pine line would reduce color discharge by 1142 Ibs/day. They estimated the capital cost of adding an
additional oxygen delignification stage would be in the order of$2 to$3 million and the chemical cost
savings approximately$1.2 million/year,providing a reasonable payback. Mill staff has mentioned a
three year pay-back,which is consistent with this estimate,but have further asserted from their more
recent estimates that the cost could be$3 to 5 million owing,presumably at least in part,to increases in
the general prices of stainless steels(not as specifically fabricated and estimated for this technology). The
BRPP oxygen delignification systems were installed in 1993. Since that time,the use of two-stage
oxygen delignification rather than the traditional single-stage systems has become well established in the
industry because it normally further reduces mill operating costs. BRPP believes that the previous
evaluation of adding a second stage oxygen delignification stage should be repeated in part because of
differences cited in delignification efficiencies.Nonetheless,the Tech Team believes this technology has
potential to move gregressed beyond being considered"reasonable certainty"in 2001 to"highest
certainty"at this time. and should be implemented at the earliest Possible date
The Tech Team identified the following color reduction activity for BRPP: conduct laboratory studies to
evaluate the color reduction benefit and engineering analysis to determine the cost of develop an
angineefing design least.._ the installation of an additional oxygen delignification stage for the pine
(softwood)pulping/bleaching line.In 2001,the Tech Team estimated that this process change would
reduce effluent color by 1,100 to 1,400 lbs/day. As noted above,even after the recent improvements to
pine line delignification efficiency, GL&V estimated that this change would reduce discharge by the
order of 1000 Ibs/day.
Andritz also evaluated modifications to the mill cooking process. These would require complete
replacement of the digester systems at Canton,which would cost(at least)several tens of millions of
dollars, if they are feasible at all within the mill's space constraints. The Tech Team concluded that
modifications to the mill cooking process do not merit further analysis.
Treatment of CRP Purge Stream for Color Removal
The Chloride Removal Process(CRP)purge is a low flow,highly concentrated stream. Color is typically
41,000 platinum cobalt units (pcu)in a stream that discharges at 10 gpm(15,000 gal/day or-3tank
tFueksfday).BRPP reported that in 2006 the CRP purge contributed approximately 8,745 lb/day(23%)of
the total mill color loading to the treatment system but only 0.05%of the discharge flow. By comparison,
in 2001 the CRP contributed approximately 5,000 lb/day(13%)to the treatment system loading(EPA
Tech Team,2001). Thus,in the last five years there has been a significant increase in color contributed
Memorandum
May 10,2007
Page 11
by the CRP,both in lb/day and in percent of total mill load. BRPP reported on its investigations into
technologies for reducing CRP purge color in its March 2005 report,"Chloride Removal Process(CRP)
Color Reduction Technology Assessment" (Blue Ridge 2005).In this report,BRPP points out that the
CRP purge is a very concentrated material (360,000 mg/L or 36%total dissolved solids) that is
discharged to sewer at 165'F. Handling the material is difficult because it will crystallize as it cools.
Also, it is quite corrosive due to the high chloride content.
As described in their 2005 report,BRPP assessed more than nine alternatives for disposal or treatment of
the CRP purge and concluded that none was technically feasible. Although consistent performance was
not demonstrated, C1O2 bleaching was a low cost,potentially effective means of reducing the CRP purge
color prior to discharge to the treatment system. BRPP found that C1O2 bleaching could inconsistently
remove up to 75 to 90%of the color in the CRP purge stream. This is _ _ignifi,ant_...,Ue fiBH ift.. ..OW
Although CRP purge is a relatively low flow stream,BRPP estimated that off-site solidification and land
disposal would cost more than$3.6 million/year excluding loading facility infrastructure.
During the Tech Team's 2007 mill visit,BRPP stated that there was no apparent decrease in secondary
effluent color when CRP wastewater was not flowing into the treatment system. In response to a Tech
Team request,BRPP presented data correlating the days on which the CRP process was shut down with
secondary effluent discharge color,for August 2006 to January 2007(Blue Ridge 2007a). During this
period,there were nine widely dispersed pairs of days when the CRP process was down for part of the
day. BRPP believes eoncluded that CRP color is removed in the secondary treatment system,based on an
ANOVA analysis comparing"down"days with CRP operating days and general observations. However,
in response to a Tech Team follow-up request for clarification BRPP later indicated confirmed these data
allowed no definitive conclusion,and that there is no relationship of presence or absence of CRP purge to
secondary effluent color based upon general observations.
The Tech Team does not agree with BRPP's•belief this een elesiaa or rationale for the following reasons.
First,with one exception,the periods of CRP shutdown are only a day or so each,which is insufficient
time for the W WTP to stabilize with the change in raw effluent characteristics. Second,the times of
shutdown and startup of the CRP do not correspond with the effluent sampling times in the mill system,
so that there is no direct correlation in time with the effluent sample. Further,the fact that all CRP`off'
days are in pairs,suggests that CRP was down for a period that spanned parts of two mill sampling days.
The simple mathematical average of treated effluent color discharge during"CRP off'days for August to
December 2006 is 39,995 Ibs/day. The average when the CRP process was operating is 36,958 lbs/day.
This suggests that CRP purge stream causes a reduction in effluent color,which defies common sense.
Also,the difference in effluent color between the"CRP on"and"CRP off'days is 7%,while day to day
variations in color discharge values are frequently over 20%,tending to subsume and confound any
analysis of the impact of changes in the CRP purge stream. The resultant statistical color performance
data from this period likely varied just due to the normal variation in day to day color performance.
In short,because there are so few consecutive days without a CRP purge,the data from the period August
to December 2006 are not sufficient for establishing the extent to which the CRP purge contributes to
final effluent color. In order to explore this further,BRPP should conduct additional evaluation of the
effect of CRP purge on final effluent color levels.These evaluations could include a Pall Seale mill trials
to determine the impact of CRP on effluent color in whieh the CRP purge is exoluded fiom
Memorandum
May 10,2007
Page 12
purgethe mill's white liquer to opefatien efthe GRP was very slow,Fequiring a month er uieFe te show
signifieant Ghanges. This AABqistent with mpelielwe in other Filills,and is due to the fast that the
quantity ef shlerides and potassium Ferneved by GAP is small relative to the inventefy in the mill's blaek,
the spare elarifier eF e9site fer a period ef at least flifeeweeks. Nenetheless, BRPP mill staffasseFt tha
a shutdem% fef sueh a period ef time would net be opefationalb,feasible,bet eff�Eed fie speeifie faets oF
reasens why this would be the ease.
Follow the evaluation,Tthe impact of excluding the CRP purge
stream could be related to final would be ebsen,ed by reutine analysis e effluent color. If the effluent
color without the CRP purge is unchanged,BRPP could conclude that the CRP purge contributes little to
the final effluent color. In this case, further research into reducing CRP color would be a low priority.
On the other hand, if the effluent color without the CRP purge decreased,BRPP could conclude that the
CRP purge contributes to the final effluent color and further research into reducing CRP color should be
undertaken.
BRPP could investigate approaches to prevent color from accumulating in the CRP. The source of the
color in the CRP purge stream is carryover of black liquor particles in the direct contact evaporators
(DCE)in the recovery boiler systems. This carry over could be eliminated if the two traditionally
designed recovery boilers(which are 34 and 42 years old)were replaced by one modern boiler. This .
would probably represent a capital cost of over$100 million,and would very substantially reduce the
energy costs at the mill,since today's recovery boilers are much more efficient than the DCE/recovery
boiler systems of the vintage installed at Canton. Analysis of all the economics and long term life of the
mill would be necessary to evaluate such a major investment.
On a more modest level,it may be feasible to reduce black liquor carry over by adjusting operating
conditions in the existing DCE's. The Tech Team is not aware of any experience with this in other mills,
or research,but the situation at Canton strongly suggests that at least some investigation and trials of
modifications to the operations is warranted. Any of the measures discussed previously in this document
to reduce C1O2 use could reduce the quantity of chloride to be removed in the CRP purge stream. If this
is reduced, the color discharge of the CRP also would be reduced.
BRPP could also further investigate C1O2 treatment and other treatment of the CRP color and other
approaches for excluding the CRP purge from the mill discharge.
Additional Color Reduction Strategies
The Tech Team has identified the following color reduction activities that should be for improving the
color removed by the Canton Mill wastewater treatment plant:
1) Investigate in further depth color formation when acid wastewater is mixed with mill
wastewater in the current configuration,and identify techniques to reduce this effect;
2) Further investigate adding polymer and/or other treatment chemicals upstream of the
secondary clarifiers; and
3) Maintain addition of polymer and/or other treatment chemicals to the high-color
wastewater diverted to the extra primary clarifier,and investigate adding additional
equalization/treatment tankage.
Memorandum
May 10,2007
Page 13
These strategies are discussed below.
Memorandum
May 10,2007
Page 14
Color Formation When Acid Wastewater Is Mixed With Mill Wastewater
BRPP provided the Tech Team with daily color data for the"Low Lift"(mill sewer),acid sewer,and
primary effluent sampled after the acid sewer is added in the discharge channel of the primary clarifier
leading to secondary activated sludge biological treatment system. Figure 1 (below)presents the total
mill color calculated by adding the mill sewer lb/day to the acid sewer lb/day(the lower(blue line)on the
figure). For comparison,the primary effluent sampled after the acid sewer is added is also shown on the
figure(the upper(red line) on the figure). Thus,the figure depicts the impact of mixing the acid sewer
with the rest of the mill effluent'.
150,000
Total untreated color,before and after mixing,2006 (Ibs/day)
125,000
100,000 111
75,000 —
50,000
25,000
1Jan 31Jan 1-Mar 314ar 30-Apr 30-May 29Jun 29Jul 28-Aug 27Sep 27-0ct 264lov 26-0Oo
Figure 1: Impact of Mixing Acid Sewer and Mill
The Tech Team observed:
• The quantity of color formed on mixing the acid sewer and mill sewer is striking. On
average,the quantity of color formed is 78%of the total color in the two streams. In
other words,the simple mixing of these streams forms nearly half the total color
discharge from the mill.
• The variation in quantity of color formed by mixing the effluent streams from day-to-day
is dramatic,as is evident in the graphs. When expressed as a percentage of primary input
color,the value ranges from essentially zero on some days to a maximum of 480%.
• There is no obvious correlation between the values on any one day,or series of days
close to one another.
Notwithstanding these observations,the Tech Team understands and appreciates BRPP's efforts during
the last permit term to reduce color formation through the acid sewer relocation project. The Tech Team
suggests that BRPP build on this project to better understand the mechanisms of and the follow-up
strategies for reducing color formation in the current sewer configuration. BRPP could investigate,
among other things,how sulfides introduced by leaks,spills,and unplanned discharges from white and
green liquor systems into the mill wastewater contribute to effluent color. In particular,BRPP could
3 The color of the mill sewer wastewater may be reduced somewhat by treatment in the primary clarifier. This color reduction is
neglected in the calculated untreated color,so Figure 1 is probably a slight underestimate of the impact of mixing the acid sewer
with the rest of the mill effluent.
Memorandum
May 10,2007
Page 15
investigate whether sulfides in the mill wastewater contribute to color formation when acid sewer is
added to mill effluent prior to biological treatment. Approaches to this investigation should include:
• Laboratory experiments on effluent with varying degrees of sodium sulfide addition; and
• Daily measurement of the sulfide content of the effluent from the primary clarifier
influent at the low lift pump sampling station for a period of at last three months to
determine correlation of sulfide concentration with color formation.
In most mills, losses of sulfides can be reduced. Well known sources include imbalance in the weak wash
system, as well as spills of white,black,and green liquor. BRPP should further investigate losses from
the green and white liquor systems and technologies that can reduce these losses.
Improving WWTP Performance Using Polymers
Several bleached kraft mills around the world have successfully reduced color to well below 5 lb/ton pulp
by installing tertiary treatment systems. These systems use polymers and other chemicals to precipitate
color bodies,then remove the precipitate in tertiary clarifiers. One of the newest of these mills is Celco in
Valdivia, Chile. This mill has excellent in-plant effluent control, a conventional secondary wastewater
treatment plant, followed by a small tertiary system. [This comparison isn't useful or appropriate.]
Effluent€rem this mill is depicted in FigHre 2,next to effluent c_em nnnn
id
Figure 2: Gompar-ison of BRFP Ganton AUR Effluent(February 2007) and VuldiAa Mill Effluen
n
rt
('-'vvci"vca--2o05)
[This whole section should be removed or significantly edited.l
Treatment systems like the one operated at Valdivia have a capital cost in the tens of millions of dollars,
and can create issues with solid waste disposal.
was not appfepfiate. However, some more modest control measures using variations of this technology
also exist. For example,since March 2004,Glatfelter Inc., Spring Grove,PA has supplemented its in-
plant color control strategies by using a commercially available polyaluminum chloride polymer to
enhance color removal in the wastewater treatment plant. The polymer is added in the discharge flume
from the aeration basin,just upstream of the mixing box feeding the four secondary clarifiers. The
polymer reduces pH and complexes with organic compounds that produce color. Colored material settles
out with the secondary sludge which is held in the former stabilization basin before dewatering.
Memorandum
May 10,2007
Page 16
Glatfelter have not reported sludge dewatering problems.Polymer used for additional color removal
reportedly costs on the order of$2,000 to$3,000/day.It is noted that neither of these mills have installed
the pollution prevention BFR system such as has been done at BRPP.
BRPP could conduct further laboratory trials of commercially available polymers and other wastewater
treatment chemicals in addition to those already evaluated,adding them upstream of the secondary
clarifiers. During these trials BRPP would determine the potential for reducing color discharge and
estimate the costs. Such analysis would consider seasonal use of the additive chemicals(i.e.,adding the
chemicals during periods when river flow is low and the mill discharge has the greatest impact on the
river color). The Tech Team acknowledges BRPP concern regarding issues that may occur,such as
possible effluent toxicity,difficulty in sludge dewatering, and cost.
If the laboratory trials are successful,BRPP could institute a full-scale trial for one month, and report
results to the TRW.
Pretreatment of Diverted High-Color Wastewater
BRPP currently diverts up to one hour of total mill flow to their off-line I-million-gallon clarifier,during
high color releases. The purpose of this flow diversion is to equalize color contributions to the
wastewater treatment plant. Diverted high color wastewater is batch treated by polyamine and returned to
the wastewater treatment system. The Tech Team believes that BRPP should further investigate options
for providing additional tankage that would be important in increasing the equalization and treatment
capacity beyond that which currently exists,now limited to about one hour of wastewater volume at
current flow rates. This would probably be an important contribution to both further reducing color
variability, an increasingly higher priority endeavor as long term averages are reduced, and to removing
color before being introduced into the secondary activated sludge biological treatment system.
Summary: Additional Identified Color Reduction Activities that BRPP can Pursue in the Next Five
Years
Mill process changes that reduce wastewater color are generally preferred to end-of-pipe treatment
because they may have lower capital costs and may benefit the mill by reducing operating costs-.;
however,some mills have chosen to use end-of-pipe technologies to achieve color reductions. It is noted
that throughout this color reduction process the Mill has continued to evaluate other color reduction
technologies, including end-of-pipe. These technologies have never been demonstrated to be
operationally,technically or economically feasible.However-,mills in a ffi,ifea ffiental regalatofy
jut--isdietiens with severe resttietiens oft theiF eoloF disehaEgas have bean required te implement end of
pipe 91or reEn val teGlinalegie.. Since the 1 non'. Nonetheless,Aalthough the Tech Team continues to
maintain the highest priority for in-mill improvements such as process changes and optimization and
increased black liquor recovery,external color removal technologies should also rise to a higher pi Pity
than previously and continue to be considered carefully in the mix of options for further controlling the
color of BRPP's discharges. These color removal technologies include both end-of-pipe wastewater
treatment and treatment of segregated wastestreams (i.e.,the CRP purge).
The Tech Team identified strategies,focused primarily on in-mill process improvements,which that
BRPP can use to further reduce its effluent color discharges. These strategies are summarized below.
1) On the pine bleaching line,conduct full scale trials with implement the use e peroxide
fortification of the Eo stage and decreased the target kappa factor as recommended in
BEPER 2001 and by Liebergott/GL&V 2006 to determine the impact to effluent color.
Also,Revaluate the use of high temperature for the peroxide-fortified extraction stage.
Memorandum
May 10,2007
Page 17
2) On the hardwood bleaching line,conduct full scale trials with implement the use e
oxygen and peroxide fortification of the E stage and decreased the target kappa factor as
recommended in BEPER 2001 and 2006. Also,Bevaluate the use of high temperature for
the peroxide-fortified extraction stage.
3) Continue to improve the control of leaks, spills,and intentional diversions of black liquor
and other highly-colored wastewater:
— Move forward with the planned CRP sump and containment project to eliminate
unplanned releases of this highly-colored material to the mill sewer.
— Further substantially reduce and ultimately eliminate discharges of highly-
colored wastewaters directly to the wastewater treatment system. These
wastewaters are generated during planned mill outages or diverted to the off-line
clarifier during mill upsets.Increase interconnected tankage available to both
fiber lines in order to maximize recovery of these wastewaters and treat the
wastewaters that cannot be recovered to reduce color prior to metering them to
the treatment system.Regular training and advanced realtime coordination
among mill staff(identifying and correcting problems,failed equipment/parts)is
critical to the success of these efforts and is ever more important in reducing day-
to-day variability which is becoming more apparent as long term averages are
decreased.
4)
5) install an Evaluate an additional oxygen delignification(OD)stage for the pine
pulping/bleaching line by conducting laboratory studies and engineering analysis to
evaluate the color reduction benefit and cost of 2-stage OD on pine.
6) As 00—i—ise is redueed thfough seeend stage OD on the pine line and other aptiens on
beth fiber lines, investigate inereasing the amount of hardwerid D stage ffitfate recyried
and intFedueing fer the fifst time same of these hardweed filtFates in BFR,aild thus
7) Continue to evaluate the impact of the CRP purge on treated effluent color to determine if
the CRP color,now a significant portion of the total color discharged from the mill
processes, is completely removed in the secondary treatment system.If CRP color is not
removed in the treatment plant,investigate approaches to prevent color from
accumulating in the CRP, such as reducing black liquor carry over by adjusting operating
conditions in the direct contact evaporators.Also, if CRP color is not removed in the
treatment plant, further investigate C1O2 treatment of the CRP purge to reduce its color
prior to the mill wastewater treatment system. Such an effort may take special efforts to
accomplish.
8) Continue to investigate and implement strategies for improving color removed by the
Canton Mill wastewater treatment plant:
Memorandum
May 10,2007
Page 18
Further analyze color formation when acid wastewater is mixed with mill
wastewater in the current configuration,and identify other techniques,such as
minimizing sulfide releases to the mill sewer from white and green liquor leaks,
spills,and/or diversions,to reduce this effect;
— Further analyze addition of polymer(s) and other treatment chemicals upstream
of the secondary clarifiers; and
— Maintain or further increase the hydraulic capacity for capturing and treating
highly-colored wastewaters prior to introducing them to the wastewater treatment
system,including the further optimization of adding polymer and other treatment
chemicals to the highly-colored wastewater diverted to the extra primary clarifier
and/or any additional facilities that may be provided.
Memorandum
May 10,2007
Page 19
Effluent Color Variability
Since installing the BFR process and other improvements(BMPs, etc.),BRPP is to be commended for the
continuing and important progress in reducing their annual average color discharges. However,the Tech
Team believes there are additional areas where further color reduction strategies can contribute to reduced
effluent variability.Neweven many days_... eaf of high .,.,,._diseh.._,.,... ..en fi....,..,,be ,._...,._:,...,,,.a an
diseh......es a eide with periods e f lew F F flow(t5cpieally in the late s eF) they eause an elevation
peaks in ealer 0
The
Tech Team identified several strategies for reducing daily-rnwEimum color variability diseberges and their
ffivaet on the :er. These strategies are listed below:
• Segregate and reuse highly colored waste streams during normal operations and pretreat
the remaining acid sewer flow to minimize and reduce the variability of the generation of
"sewer generated color,"now occurring in the secondary activated sludge biological
treatment system;
• Continue and improve treating highly colored non-reusable wastewater diverted to the
off-line primary clarifier, including increasing the wastewater volume which can be
treated;
• Segregate highly-colored wastewaters released during planned and unplanned diversions
/mill outages and recover/reuse those for which it is feasible and treat the remaining
wastewaters prior to metering into the mill wastewater treatment system;
• Increase the tank volume available for equalizing highly-colored wastewater prior to
pretreatment and introduction into the wastewater treatment system;
• Further investigate treating total mill biological system effluent prior to discharge using
polymers or other wastewater treatment chemicals,particularly during periods of high
influent color and/or low river flow;
• If the CRP purge color is found not to be removed in secondary treatment, avoid
releasing the CRP purge during periods of low stream flow (or truck it off site); and
• Curtail pulp production during periods of low stream flow;this should continue to be
considered an option of last resort,given that best performance has been noted by the mill
to be during extended periods of steady production and greater risk of elevated color
during process shutdown and startup. See Low Flow Contingency Plan,December 1,
1998.
Memorandum
May 10,2007
Page 20
Suggested Final Effluent Color Limits for Blue Ridge Pulp and Paper
Substantial and commendable progress has been made to date by BRPP through expenditures for
additional and improved process and related technologies,treatment technologies,and management
practices which have reduced the long term average color discharges. Some of these technologies and
practices have been in addition to those identified in the previous Tech Team/TRW reviews. This
the kWei4anee of reduoing day to day effluent variability along with the ether Feeemmendations efthe
T-Oeh TOUR g6ifig&FWaFd&afn this efitioal point in time. T-h@F'e&re,the Tesh Team reseffiffiends to
pand the existing
pefmit limits te inelude a maximum day disehafge limit-,and net fely solely on the annual average an
mwEimam 90 day avefage limits to preteet the quality efthe Pigeon River. Daily ffiffitifflum ail
quality,and wetild be eensistefit with KwAts for other pammeters eentrelled in the pFesent BRPP peffnit
EB9BT88,A01, ete.). The Teeh Team also believes that these effluent lifnitgtiAns sh uld-be applip�
Gendifien A.5)prevides for m2a4gring eeler(without limits)at the end of pipe,and demmstFeafn of the
disehafge at Fiberville Bfidge (and ether leeatiens fin4her dewastFeam) at whish peints in stream dilution
is ineluded and dependent upen Pigeon River flow. This reeemmeadatien was ineluded in the q4MI's
2001 AilemeFandum-, at item no. . The Tech Team's recommended range of end of pipe permit effluent
limits and the derivation of these limits are presented in Table 5,and discussed below.
Table S. Tech Team Recommended Range of Effluent End of Pipe Color Permit Limits
• �:= Limit - _ ,`_. Range of Recommended Limits(lb/day)'
Annual Average 32,000 to 37,000
30-day Average 44,800 to 51,800
Daily-Ataximum 93,840-te-96,948
Basis for Recommended Annual Average
BRPP provided daily measurements secondary effluent color(lb/day)for every day in 2006.
Examination of the daily measurements for 2006 identified two days(July 7 and 8)with measured
discharge greater than 100,000 lb/day. BRPP reported that this elevated color discharge resulted from
"CRP slurry tank overflow for—20 min,first time the mill experienced this type of event." Because
these discharges were so high and from a unique source,they were omitted from the calculation of the
annual average.
Annual average for 2006,without July 6 and July 7:
36,695 lb/day,rounds to 37,000 lb/day
• This annual average load is less than the 39,000 lb/day suggested by BRPP. However,it
is based on the mill's 2006 performance and does not include any reductions that may be
achieved by the process changes outlined in this memorandum. As a result of analysis of
the BRPP mill discharges by the Tech Team and others in support of the 2001 permit,the
permit's interim color goal was 32,000 lb/day with a range up to 39,000 lb/day.
• As described earlier in this memorandum,the Tech Team recommends that BRPP
implement several key process improvements and investigate others in order to reduce
Memorandum
May 10,2007
Page 21
the annual average color performance toward the overall permit interim goal of 32,000-
37,000 lb/day.
Basis for Suggested Daily Ma*imum and 30-Day Average Limits
The statistical analysis used for the development of EPA's Cluster Rules is documented in Statistical
Support Document for the Pulp and Paper Industry: Subpart B(EPA, 1997).This document describes
EPA's development of,among other things,the variability factors that were used to calculate NSPS for
B0135 for the Bleached Papergrade Kraft and Soda(BPK)subcategory. These variability factors are
reproduced in Table 6.
Table 6. Bleached Paper-grade Paper grade Kraft NSPS Variability Factors
Vari_ab34ty Fac_tots,-
Analytc. .. ., 3-Ba5`_ _ = 30-Day
GODS 2,62 1.4
Source: U.S.EPA 1997. Table 24.
The BOD5 variability factors shown above were developed using daily monitoring data for the best
performing(in terms of production normalized B0135 load)BPK mills. The monitoring data represent the
effluent from well-operated wastewater treatment systems. The calculated variability factors account for
the autocorrelation of the daily loads and the log-normal distribution of the measurements.
Notwithstanding 13111212s eefliments to , The Tech Team believes that it is reasonable and
appropriate to use the variability factors developed for the BPK NSPS for BOD5 to calculate BRPP daily
maidam t and 30-day average permit limits for color.because-
Xhe Astem vxeFe developed using meniteFing data ftem BPK mills wM well opera
• Color-,like BOD ;and
• Geler 1REe B0_'_
abserption4dseEption to biomass.
Applying the B0135 variability factors to the 2006 recommended annual average daily color discharge
results in the following daily maxiin tl 30-day average limits:
37 000 96,940 llarly��feximaan: 2:62
30-Day Average: 37,000 lb/day x 1.40=51,8001b/day
(The Daily MwEimum and 39 day average lifaits fer the long tofm aver-age geal of 32,000 lbAlay aFe
GempaFing the'2006 effluent menitefing data te these limits finds that twe days aiweeded the daily
guar:
Nly:� 1h
duly 101229lb
Memorandum
May 10, 2007
Page 22
peri«ed.. Th,,.., perieds aFe listed in T..1,1.,7
AAY-; Q,119-5
Aa)lg GS R
66�64
c]�on
�2�l2tnE3G'P-E8 �$
wl_.IlO�cmnGTLv4.._26 424
(avefage monthly) and 280 m&q.(daily maximurn). Winter diseharge lifaits aFe more strillgent. The
older bleaebed iffaft mill that diseharges to a small, e9lef !Lnaited stfeam. Glatfelter,
paper per day. BRPP pfeduees appfoximately 1640 US tens per day,with a gfeater prepet4ion e
seftweed to hardweed pFeeessed than at the GlatfelteE mill, and with a daily total wastewatef fle-A,e
reeemmeaded range of end of pipe..e.mit limits c„_nnnn inn onn ,51 onn 11,i a y,.,enthl...,..,,_..,....
Memorandum
May 10,2007
Page 23
References
Andritz Ahlstrom Sales. 2001. Pruyn's Island Technical Center Report 2001-068 Part 1. Laboratory and
Lo-Solids Cooking with O-Do-Eop-D Bleaching Sequences on Softwood Fumish from Blue Ridge Paper,
Blue Ridge,NC. Part 1 Softwood Results. Part 2. Hardwood Results. (December 6).
Audet,Andre,Michel Faubert,Zhi-Hua Jiang and Barbara van Lierop,PAPRICAN. 2003. Technical
Service Contract Report TSC-7447. Bleaching Evaluation for Effluent Colour Reduction. Prepared for
Blue Ridge Paper Products,Inc., Canton,NC. October.
Blue Ridge Paper Products,Inc. 2005. Chloride Removal Process (CRP) Color Reduction Technology
Assessment. (March).
Blue Ridge Paper Products,Inc. 2006a. Color Compliance Report: Canton Mill. (May).
Blue Ridge Paper Products,Inc. 2006b. Blue Ridge Paper response to questions from the Technology
Review Workgroup (TRW) that were e-mailed on 28 Nov 2006. (December 19).
Blue Ridge Paper Products,Inc. 2007a.Response to additional questions far BRPP about data provided
to TRW Don Anderson e-mail dated 5 March 2007(March 15)
Blue Ridge Paper Products,Inc. 2007b. March 19, 2007 Additional Data Required from BRPP(March
28).
Bodien,Danforth G. 2007. Site Visit Report, Blue Ridge Paper Products, Canton,North Carolina.
(April 2007)
EPA Tech Team. 2001. Memorandum to Technology Review Workgroup. "Additional Color Removal
Technologies and Their Economic Impacts on Blue Ridge Paper Products,Canton,NC."(July 25,2001).
Fudanic, Sean M. 2007. Water Quality Protection Report,P. H. Glatfelter Company, Spring Grove
Borough and Jackson Township, York County,for the Renewal of NPDES Permit No. PA 0008869.
PADEP Southeentral Regional Office(draft,February 2007).
GL&V Pulp Group, Inc. and Liebergott&Associates Consulting. 2001. Bleach Environmental Process
Evaluation and Report. Prepared for Blue Ridge Paper Products, Inc. and Clean Water Fund of North
Carolina. (June 8)
GL&V Pulp Group, Inc.,Liebergott&Associates Consulting. 2006. 2006 Update: Bleach
Environmental Process Evaluation and Report. Version containing manufacturer's proprietary
information.Prepared for Blue Ridge Paper Products, Inc. (July 7,2006)
Munro,Fred and John Griffiths. 2000. Operating Experience with an Ozone-based ECF Bleaching
Sequence,Proc. International Bleaching Conference, Halifax,,Canada, 2000. TAPPI Press.
Technology Review Workgroup,2001.Memorandum to North Carolina Division of Water Quality
and the NC Environmental Management Commission's NPDES Sub-Committee. "Additional
Color Removal Opportunities,Blue Ridge Paper's (BRP) Canton,NC Bleached Kraft Paper
Mill, 2001 NPDES Permit Renewal." (August 3, 2001).
U.S. EPA, 1997. Statistical Support Document for the Pulp and Paper Industry: Subpart B. (November)
of W ATF
\G� qQG Michael F.Easley
Q�nA Governor
NCDENR William G. Ross,Jr.,Secretary
North Carolina Department of Environment and Natural Resources
Coleen H.Sullins, Director
Division of Water Quality
James D. Giattina,Director September 14,2007
Water Management Division
EPA Region IV,Atlanta Federal Center
61 Forsyth Street
Atlanta,GA 30303-8960 Subject: Response to EPA Technology Team
Memorandum on Color Removal Strategies
For Blue Ridge Paper Products,Inc.
Dear Mr. Giattina: NC NPDES Permit No.NC0000272
The North Carolina Division of Water Quality(NCDWQ)received the memorandum from the EPA Technology
Team entitled"Color Removal Strategies For Blue Ridge Paper Products,Inc". We appreciate the opportunity to
review the report and provide EPA Region IV written comments. NCDWQ has a number of serious objections
and concerns regarding the report that we would like to bring to your attention. NCDWQ also wishes to present
these comments as part of the written record to be considered by the upcoming Technology Review Workgroup
(TRW). As you are aware, the TRW will review the EPA Technology Team(Tech Team)report as part of its
decision-making process and subsequent recommendations for Blue Ridge Paper Product's NPDES renewal.
Specific concerns with the Tech Team's report are outlined in Appendix A attached to this letter. We would
appreciate you sharing these comments with members of your staff who will be involved in the NPDES renewal
for Blue Ridge Paper Products, Inc. (BRPP),as well as members of the TRW.
NCDWQ's most serious objections to the report are with respect to the Tech Team's recommendation of effluent
color limits and the Tech Team's recommendation of immediate implementation of certain perceived color
reduction technologies. It is NCDWQ's understanding that the Tech Team's responsibility was to address
technically,economically, and operationally feasible measures in the manufacturing process that may reduce
color in the effluent. The recommendation of effluent color limits was never the goal or responsibility of the Tech
Team and NCDWQ strongly objects to numeric effluent limits as part of the report. Discussion and
recommendation of any effluent color limits is part of the TRW,and NCDWQ will make its recommendation
regarding any differing effluent color limits within the scope of that workgroup,and as part of the NPDES
renewal. However, because the Tech Team has already recommended effluent color limits, it will be more
difficult for all parties that comprise the TRW to review the Tech Team report and objectively recommend any
differing color limits or possible reductions. It is likely that many members of the TRW(and members of the
general public) will view these recommendations as absolutes, when the recommendations should not have been
made at all,and have now been made based on a technically flawed report.
In addition to the recommendation of effluent color limits,NCDWQ also objects to the Tech Team's
recommendation to immediately implement certain technologies to reduce color. This also was not the goal of the
Tech Team,as NCDWQ understood it. Some of these technologies have not been carefully tested and evaluated
as to their ultimate effect on effluent color(especially in regard to the complexities of the BRPP mill). While
NCDWQ does not object to evaluation of some of these technologies, immediate implementation by BRPP may
not prove to be effective or provide significant reduction.Discussion of implementation should have been held
with members of the TRW and should not have been part of the Tech Team report.
N.C.Division of Water Quality/NPDES Unit Phone:(919)733-5083
1617 Mail Service Center,Raleigh,NC 27699-1617 fax:(919)733-0719
Internet:www.newaterquality.org DENR Customer Service Center..1 800 623.7748
Mr.Giattina
EPA Region IV
Page 2
Again,NCDWQ strongly objects to the Tech Team's report,as it strayed significantly from its goal of providing
only factual information regarding the technically, economically,and operationally feasible methods of color
reduction for pulp and paper mills. Instead, the Tech Team proceeded to make recommendations of which
technologies BRPP should implement and when,as well as what effluent color limits BRPP should receive with
renewal of the NPDES permit. We hope that EPA Region IV reviews the Tech Team report and takes this into
consideration prior to convening the TRW. Although it is too late for NCDWQ to have influence over the Tech
Team report, we respectfully request that this letter be shared with all parties of the TRW in equal measure to the
flawed Tech Team report,and that this letter with comments be considered along with that report. Should you
have further questions,please contact Charles Wakild,P.E.,Deputy Director,at(919) 733—5083.ext. 204,Dr.
Sergei Chemikov,ext. 594,or Susan A.Wilson,P.E.,ext.510.
SFere
C . Sullin
w/Appendix
cc: Mr.Donald Anderson,EPA Headquarters
Mr.Marshall Hyatt,EPA Region IV
Ms.Karrie-Jo Shell,EPA Region IV
Mr.Roger Edwards,ARO/Surface Water
NPDES File
Central Files
Mr.Giattina
EPA Region IV
Page 3
Appendix A
NCDWQ has made an effort to clarify some of the inaccuracies in the Tech Team's report.Likely,due to the
short time frame between notification of the finalized report and release to the public,NCDWQ may not have
caught every flaw or misstatement. Should further issues with the report be found,NCDWQ representatives will
bring this to the attention of the TRW.
Again,NCDWQ believes the Tech Team has inappropriately made recommendations on effluent color limits and
recommendations regarding the implementation of color removal technologies. NCDWQ's representatives will
provide information on our understanding of the goal of the Tech Team and the goal of the TRW during the
meeting of that workgroup.
I) Data in the discussion of the Primary Effluent Color(Table 3)represents a phenomenon well known to
pulp and paper experts as sewer-generated color.Primary Effluent data cannot be used to evaluate the
effectiveness of color reduction efforts because it reflects sewer-generated color.The best indicators of
mill performance are primary influent color,and most importantly,secondary effluent color.Both of
these streams have shown a significant decrease in color over the term of the permit.The Tech Team's
claim that mill efforts did not result in reduced color loads to the treatment systems is incorrect and
appears to be in contradiction to the existing data.Sewer-generated color is not a measurement of the
mill's efforts to reduce color.The Tech Team's conclusion that the"Canton mill can further reduce
primary effluent loads through continuing efforts to minimize unplanned spills and leaks"is incorrect.
Reduction in leaks and spills is not reflected in primary effluent,it is reflected in primary influent and
secondary effluent.
2) Although this part of the Tech Team report may seem minor,the comparison of the two effluents on page
14 of the report is not relational because it does not provide equivalent conditions,and it violates the most
basic scientific principles.This comparison is also inappropriate because it compares a sample from a
facility that uses tertiary treatment with a facility that uses in-plant controls. The Tech Team must
acknowledge that BRPP uses in-plant controls to reduce color load—a preferred option over end-of-pipe
treatment,which frequently generates more pollution load in other media. BRPP has been a leader in the
industry with implementation of in-process color reduction.
3) NCDWQ disagrees with the indication that variability factors developed for BOD can be used for color in
developing a daily maximum value. The color phenomenon in pulp and paper is very complicated and has
substantial unknowns. One example of this is sewer-generated color,which describes creation of an
additional color due to the mixing of the two streams with different pH levels. This generated color
significantly exceeds simple summation of color in two streams; BOD does not exhibit similar behavior.
BRPP's experience with color reduction efforts clearly indicates that some measures to reduce color,that
were recommended by experts, resulted in no or very little color reductions,while others were very
successful. The same cannot be said about BOD. On one hand the Tech Team acknowledges complicated
and unpredictable color behavior, on the other hand the Team suggests that color should be treated like a
conservative parameter.
4) The comment to apply the color limit at the end of the pipe demonstrates a lack of understanding of the
permit, since NCDWQ already applies a color limit at the end of the pipe.NCDWQ has provided an
explanation about this issue to the Tech Team on several occasions and it is unclear why the Tech Team
makes a recommendation that has already been implemented.
5) The Tech Team recommends that BRPP spend approximately$5,000,000 to install 2"a stage
delignification, which might potentially result in only 1,100 Ibs of color reduction. This reduction is only
an estimate, which might not be realized. Removal of 1,100 Ibs of color is equivalent to about 4%of the
total color output,which is well within daily variation.This removal would also translate into only 1 unit
of color at the Fiberville bridge just below the discharge.
Mr.Giattina
EPA Region IV
Page 4
Such a change is likely not distinguishable by the naked eye and is well within daily color variability.
Even under tightly controlled laboratory conditions this small difference would not be measurable
considering color variability. Therefore,a recommendation to spend$5,000,000 to receive such a small
reduction is inappropriate.NCDWQ will make this continent as part of the TRW as well.
In addition,NCDWQ believes the cost savings that might be realized are significantly overestimated.The
Liebergott report(mill consultant)estimates the cost savings to be on the order of$500,000.The Tech
Team miscalculated the savings by trying to take double credit for saving the chemicals and increasing
the yield.However, such an approach would not result in color reduction and thus cannot be
recommended as a.color reduction measure.It is not possible to achieve three things at the same time:
reduce consumption of chemicals,increase yield and reduce color.
6) The Tech Team's recommendation to shut down the Chloride Removal Process(CRP)for an extended
period of time would result in major corrosion, scaling and pluggage due to the accumulation of chlorides
and potassium in the equipment. This option is not operationally feasible. NCDWQ feels that the Tech
Team may lack an understanding that CRP is an integral part of Bleach Filtrate Recycling(BFR)process
and,although CRP itself contributes to color output,the entire BFR process(including CRP)results in a
significant reduction of color on the order of 20,000—25,000 lbs/day.
7) Statements in the Tech Team report in regard to the CRP contribution to the effluent color appear to be
incongruous.The report states that"day to day variations in color discharge values are frequently over 20
percent". At the same time surprise is expressed by the fact that during time when CRP was down the
average color was reduced by 7%. Since 7%reduction is far below the variation of 20%,the scientific
conclusion must be that there is no statistically significant difference between days when CRP is
operational and days when CRP is not operational.The report statement that"This defies common sense"
is puzzling at best. Studies conducted by a graduate student from Duke University demonstrated that color
removal at the treatment plant acts primarily on brown color.The CRP produces brown color,which
might explain why shutting it down did not result in any significant difference in the effluent color.
8) The Tech Team assertion about the effectiveness of C102 treatment of CRP purge is inaccurate.Although,
it was about 80%effective in some cases,the report failed to mention that it was about 0%effective in
other cases. If the statement is left in the report"as is",it could create an impression that BRPP,for some
reason,refuses to implement an inexpensive and effective method to reduce color,which is false.
NCDWQ believes this treatment option should be studied to determine its feasibility and will recommend
this as part of the TRW. One of the successful CRP treatment studies conducted by BRPP where 002
was used for 20 days demonstrated average color removal at approximately 79%level.However,this
reduction did not manifest itself in any significant decrease of color in the effluent, which questions the
need to treat CRP color.The color phenomenon is very complicated,and there is a need to carefully study
it before making any categorical recommendations to implement some technological measure.
9) The Tech Team's recommendation to add a daily maximum color limit is inappropriate. As stated
previously,NCDWQ does not believe that recommendation of permit limits was the goal of the Tech
Team. It is also necessary to stress that the color standard is an aesthetic standard that is based on 30Q2
flow.Therefore, a monthly average limit is the most appropriate.The daily maximum limits are typically
placed in the permit for a parameter that exhibits acute toxicity or where there are standard effluent
guidelines for a daily maximum limit.Present color level in the mill's discharge is not toxic and there are
no Federal effluent guidelines for color. NCDWQ believes there is no need for color to be limited on a
daily basis and this will be emphasized during meetings with the TRW.
Mr.Giattina
EPA Region 1V
Page 5
10) The Tech Team's recommendation to reduce the kappa factor is not feasible,as the market currently
requires paper and paperboard with higher brightness values, which requires a higher degree of bleaching.
Therefore,there is a market driven need to maintain a higher kappa factor.
If the kappa factor were reduced,the mill would be forced to do more bleaching at the second stage in
order to produce marketable products.The filtrate from the second stage is not recycled,and shifting
more bleaching to the second stage would result in more color production.
11)The Tech Team recommends BRPP to investigate possible ways to reduce black liquor carryover in the
direct contact evaporator.The name of the equipment used in the process is a"direct contact evaporator",
which already implies that black liquor is in direct contact with the exhaust and it is physically impossible
to reduce carryover because of the existing'direct contact.The proposed solution to install a new boiler is
extremely expensive(about$100,000,000)and is not economically feasible.NCDWQ disagrees with this
recommendation and will reiterate this as part of the TRW.
12)The comparison to the Glatfelter Permit(Spring Grove,PA) limit is inappropriate in the technical report.
There are several significant differences between the mills, the critical difference being that the Glatfelter
mill purchases a substantial amount of pulp to produce paper.Therefore,any comparison based on the
amount of paper produced is inaccurate. Since color is predominantly produced during the pulping
process,any technical comparison should be based on the amount of pulp production by the two mills.
Taking into account the above stated issues,the color performance of both mills appears to be similar.
Although again, this portion of the report does not maintain NCDWQ's understanding of the Tech
Team's goal on the technically,economically,and operationally feasible measures in the manufacturing
process that will reduce the color in the effluent.
13) The Tech Team claims that the mill consultant recommended adding oxygen and peroxide to E stages,
when in fact it was recommended to evaluate such additions,not to implement it immediately.
14) The Tech Team claims the need to use oxygen in the extraction stage on the hardwood line. BRPP has
used oxygen for an extended period of time and did not observe any effect on the effluent color.After
oxygen use was discontinued,effluent color also did not change.NCDWQ will object to this
recommendation within the scope of the TRW.
15) The Tech Team claims that the consultant recommended replacement of CIO,bleaching power with
oxygen and/or peroxide.This is not true,as the consultant only recommended evaluation of this option,
not replacement.
16) The Tech Team's insistence on the need to recycle more filtrate from the hardwood line is unfounded.
This proposal is not operationally feasible since the hardwood line has no MRP(metal removal process),
which is necessary to successfully recycle bleach filtrate.The construction of MRP and BFR for the
hardwood line would cost about$30,000,000 and is not economically feasible.NCDWQ will object to
this within the scope of the TRW.
17) The Tech Team's use of the following phrase—"BRPP believes that secondary effluent showed
improvement"is improper(page 7)and should be stated definitively,not as a belief. The existing data
clearly indicates that a decrease in effluent color from 42,676 to 37,058 was indeed realized.
18) The Tech Team's remark on page 16 of the memo regarding mills that are required to implement end-of-
pipe color removal technology fails to mention that those mills do not have extensive in-mill process
controls, which has been emphasized at BRPP.
Mr.Giattina
EPA Region IV
Page 6
19) The Tech Team's selected strategy to"... ultimately eliminate discharges of highly colored wastewaters"
is unrealistic. Continued reduction,when feasible,is certainly NCDWQ's goal for colored wastewater
discharges, but any reduction or elimination must be weighed along with the technically and
economically feasible issues.
20) The Tech Team's recommendation to avoid releasing CRP purge streams during low flow condition is not
operationally feasible. CRP is an integral part of the BFR,which produces significant reductions in color
discharges.NCDWQ objects to this recommendation and will do so as part of the TRW.
21) The paragraph on page 17 that contains the recommendation to curtail production during low flow
contradicts itself. In the beginning the memo acknowledges that color performance is best during long
periods of steady production.Then it recommends reducing production,which has been shown to increase
color load to the receiving stream.NCDWQ objects to this recommendation and believes all plants retain
better performance under steady operation.
22) The Tech Team recommends further trials to evaluate potential of ozonetchloride dioxide(ZD)bleaching
for the hardwood fiberline.This recommendation contradicts Dr.Liebergott's conclusion in the report
dated July 7, 2006.It also has a high capital cost($10,000,000) and operating cost($1,500,000—
$1,700,000). Another concern with this recommendation is its effect on pulp strength,which is of special
importance for BRPP since it produces milk and juice cartons.The Glatfelter Spring Grove mill that
employed this technology determined that there is little or no color benefit.Due to the high cost and
unlikely benefit,this technology is not economically feasible and should not be evaluated.NCDWQ will
reiterate this within the scope of the TRW.
23) The Tech Team's statement on page 6 that`BRPP did not report one single event or one single type of
event that caused high color in the primary effluent"is incorrect. BRPP provided information on the
cause of some elevated primary effluent color days in the March 2007 response to additional questions.
Technology Team Responses to NC DWQ Comments, dated September 14,2007,
and (Indirectly) to BRPP Comments, dated June 4, 2007,
from Teleconference on July 25, 2007, and dated July 26,2007, on
Technology Team Memorandum Regarding Color Removal Strategies
for the Blue Ridge Paper Products Mill, Canton, NC
Unless noted, all responses are based on the Technology Team (TT) report to the
Technical Review Worlrgroup (TRW), as publicly released by EPA Region 4 on
September 5, 2007 and per Addendum dated December 19,2007
Comment 1) . . . Sewer generated color is not a measurement of the mill's efforts to
reduce color. "The Tech Team's conclusion that the Canton mill can further reduce
primary effluent loads through continuing efforts to minimize unplanned spills and leaks
is incorrect. Reduction in leaks and spills is not reflected in primary effluent, it is
reflected in primary influent and secondary effluent!"
Response:
• It is important to control mechanisms resulting in sewer generated color
(location of mixing acid and alkaline wastewater streams, now occurring in
primary effluent and also at least in part in the aeration basin of the
secondary activated sludge biological treatment) and further reduce sources
of leaks and spills contributing to this phenomena; the TT report noted (pg.
13) that the simple mixing of these streams forms nearly half the total untreated
color discharge from the mill to the treatment system
• The TT acknowledged that since 2001 the mill made improvements in spill
collection (BMPs,pg.5, 13),including a project to reroute the acid sewer to
address sewer generated color; the TT report also recommended additional
improvements to BMPs and black liquor control(pg. 15, 16) to build upon
further initiatives already planned by BRPP
• BRPP would not invest in these initiatives if they did not believe, as the TT
does,that there would be reduction in final effluent color
• Thus, the TT considered it important to note that variability in influent to
secondary activated sludge biological treatment(primary effluent) and the
color now also generated in part in the secondary treatment system (see also
pg. 12, 13) "holds valuable indicators and clues to the sources and solutions
to further improving the overall performance [Long Term Average-LTA]
and reducing the variability of effluent color discharged to the Pigeon River"
(pg . 6 and the Addendum to the TT Report)
• The TT also notes that a sustained effort to further study and control this
phenomena will be required beyond the upcoming permit term given that
process changes and BMP improvements to be made during the 2008—2013
permit term will more than likely further change the mechanisms underlying
"sewer generated color"
Comment 2) "The comparison of the two effluents on page 14 of the report is not
relational because it does not provide equivalent conditions, and it violates the most basic
scientific principles. This comparison is also inappropriate because it compares a sample
from a facility that uses tertiary treatment with a facility that uses in-plant controls." The
Tech team must acknowledge that BRPP uses in-plant controls to reduce color load—a
preferred option to end-of-pipe treatment, which frequently generates pollution load in
other media.
Response:
• The photographic comparison (Celco,Valdivia, Chile), and another example
at P. H. Glatfelter,were offered by the TT as evidence of potential for further
improvements at BRPP (pg. 14,15) using improvements in end-of-pipe
treatment technologies, both secondary and tertiary
• The TT also noted (pg 14, 15) the differences in both in-plant and end-of-pipe
technologies between these mills and BRPP,BRPP's comments concerning
the comparison, and other caveats and concerns which the TT noted may
become important if similar end-of-pipe technologies are pursued
• Both the current and all past TT reports have included basic tenets of its
review(TT report pg. 2; see also 2001 TT report, pgs 5-7, 12; and TRW
recommendations to NC DWQ, 2001, pg 2; see also Summary Report
12/02/1997,pgs 6, 10), notably
➢ the importance and preference for in-plant controls and process
changes to prevent the generation of pollutants before they must be
treated at end-of-pipe, as well as
➢ the potential for multi-media issues that may occur with end-of-pipe
treatment(e.g., generation of sludges that may be difficult to dewater,
potential effluent toxicity possibly attributable to coagulants, etc.)
• The current TT report also includes multiple acknowledgements of BRPP's
efforts and expenditures reflecting these basic tenets (pg.4, 5, 6, 13, 14, 15,
16)
Comment 3)NCDWQ disagrees with the indication that variability factors developed for
BOD can be used for color in developing a daily maximum value. The color
phenomenon in pulp and paper is very complicated and has substantial unknowns. On
the one hand the Tech Team acknowledges complicated and unpredictable color
behavior, while on the other hand the Team suggests that color should be treated like a
conservative parameter.
Response:
• The TT has appreciated the complexity of color generation in pulp and paper
effluent particularly at this mill since 1997, and it exercised its best
professional judgment (BP.l) to derive reasonable and appropriate suggested
color limits (see pg. 18, 19) and used the variability factors developed for the
Bleached Papergrade Kraft(BPK) subcategory(presented in EPA's
2
Statistical Support Document for the Pulp and Paper Industry, Subpart B,
EPA. 1997) for BOD5to calculate suggested BRPP daily maximum and 30-
day average permit limits for color as follows:
➢ The factors were developed using monitoring data from BPK mills
with well-operated secondary treatment; BRPP is a BPK mill with
well-operated secondary treatment
➢ Color,like BOD5,is monitored at the effluent of the secondary
treatment plant
➢ Most importantly, color,like BOD5' is reduced by a combination of
biodegradation and absorption/adsorption to biomass
➢ The TT notes that BRPP representatives have indicated their belief
that brown color is likely subject to removal processes in the
treatment system, e.g., adsorption/absorption and biodegradation
➢ The TT has never asserted that color, from its various sources
including that generated from mixing of different wastewater streams,
is a conservative pollutant
• An acceptable alternative would be gathering effluent color monitoring data,
after technologies actually implemented by BRPP are installed and
operating, and a statistical analysis to develop variability factors to calculate
maximum day and maximum month WQBELs for color to be applied at the
end of pipe (see response to comment#4); results of this should be shared
with the TRW
Comment 4) The comment to apply the color limit at the end of the pipe demonstrates a
lack of understanding of the permit,since NCDWQ already applies a color limit at the
end of the pipe.
Response:
• The TT acknowledged that the current color limits were applied at the end-
of-pipe, although ambiguously.
• The TT report further notes: ". . .the current permit is not clear that this is
the point of application of effluent limits for color. See [from the 2001
NPDES permit] Table A(1), which prescribes end-of-pipe monitoring, but no
color limits are included(emphasis added). See also the TRW's 2001
Memorandum, at item no.8."
• The TT report states, at pg, 17, "Daily maximum and maximum 30-day
average permit limits would enhance the consistency of day-to-day in-stream
water quality, and would be consistent with limits for other parameters
controlled in the present BRPP permit(SODS,TSS,AOX, etc). The Tech
Team also believes that these effluent limitations for color should continue to
be applied unambiguously at the end-of-pipe discharge from the mill to the
river(emphasis added)."
3
Comment 5) The Tech Team recommends that BRPP spend approximately$5,000,000 to
install 2"d stage delignification,which might potentially result in only 1,100 pounds of
color reduction. This reduction is only an estimate,which might not be realized.
Response:
• In 2001 BRPP estimated the capital cost of adding an additional oxygen
delignification stage would be in the order of$2 to $3 million and the
chemical cost savings approximately$1.2 million/year,providing a
reasonable payback; mill staff has mentioned a three year pay-back,which is
consistent with this estimate
• BRPP has further asserted from their more recent estimates that the cost
could be$3 to $5 million,but BRPP offered no details to substantiate this
increase; the TT presumes this increased cost is a rough estimate and not as
specifically fabricated and costed for installation at BRPP (pg. 9),with only a
small part of this increase in cost possibly attributable to increases in the
general prices of stainless steels
• In 2001 the TT report recommended that BRPP conduct a detailed study to
develop an engineering design leading to installing an additional oxygen
delignification stage for the pine(softwood) pulping/bleaching line, and
estimated that this process change would reduce effluent color by 1,100 to
1,400 lbs/day(see 2001 TRW memorandum,item no. 4)
• The TT noted (see pgs. 8—10) that Liebergott, et. al., (GL&V 2006; BRPP's
consultant) estimated that implementing a second oxygen delignification
stage on the pine fiber line would reduce color discharge by 1142 lbs/day
• In 2007 the TT noted and remained confident that
➢ this technology progressed beyond being considered"reasonable
certainty"in 2001 to "highest certainty" at this time, and thus will
contribute to further reduction of color
➢ adjustments in upstream pulp digestion would increase pulp yield and
reduce cost for bleaching chemicals (e.g., C102 and caustic, thus
increased savings will be realized compared to earlier estimates by the
TT),while also reducing chloride and color contained in the purge
stream from the Chloride Removal Process (CRP); the only
unknowns are the actual magnitudes of these adjustments
• Therefore,the TT again recommended that priority should be given to
expedited and detailed evaluation and implementation at the earliest possible
date during the upcoming permit term
• In order to satisfy BRPP's recent concerns concerning differences in
products and processes and potentially higher costs since 2001,
➢ the updated detailed evaluation needs to identify necessary
adjustments to upstream pulp digestion (e.g., kappa number targets),
bleaching (e.g., reducing bleaching chemical usage rates (kappa
factor; see response to comment#10), enhanced extraction stages (see
responses to comment#13,#14,#15), etc.) and downstream
4
brightness/strength parameters, designing and costing, and refining
color reduction projections
➢ results of this updated evaluation should be shared with the TRW,
with provision that any or all of the technologies implemented be
conditioned upon the results of this updated and expedited evaluation
Comment 6)The Tech Team's recommendation to shut down the Chloride Removal
Process (CRP) for an extended period of time would result in major corrosion, scaling
and pluggage due to the accumulation of chlorides and potassium in the equipment. This
option is not operationally feasible.
Response:
• The TT has noted,from data and facts presented by BRPP,that CRP color is
now a significant portion of the total color discharged from the mill processes
(pg. 10)
• This mill has operated in the past for weeks at a time without the CRP
running
• Early literature on the BFR process published by Champion indicated that
weeks are needed for significant build up of chlorides that could likely cause.
operational problems (pg. 11) due to the small mass of chlorides purged daily
through CRP relative to the large inventory of chlorides in the recovery loop
• During August 2006 to January 2007,the mill shutdown the CRP (for parts
of each day) for". . . nine widely dispersed pairs of days." (pg. 10, 11). The
TT believes that the mill should extend the period of CRP shutdown(e.g.,
three weeks) in order to allow sufficient time for the treatment system to
respond to the decrease in color from the CRP purge contributed to the
influent and any changes in color now generated from the relocated acid
sewer at the influent to and within the activated sludge system
• BRPP offered no specific facts or details, aside from the assertion in these
comments,why there would be operational difficulties in conducting a full-
scale mill trail in which the CRP purge is excluded from the system
s Nonetheless, the TT is open to another reasonable approach that could be
devised(pg. 11); such an approach would need to gather the data necessary,
preferably at full scale, and over a sufficient period of time to better identify
and quantify the underlying color loads to the treatment system,both with
and without the purge from CRP
Comment7) Statements in the Tech Team report in regard to the CRP contribution to the
effluent color appear to be incongruous. The CRP produces brown color,which might
explain why shutting it down did not result in any significant difference in the effluent
color.
5
Response:
• See response to Comment 6) above for background regarding discharges
from the CRP;which contributes wastewater with very high color:(--41,000
pen)
• BRPP representatives believe that brown color is likely subject to removal
processes in the treatment system
• The TT noted,from data supplied byBRPP, that day-to-day variations in
color discharges tend to subsume and confound analysis of the impact of
changes in the CRP purge stream on effluent color
• Notwithstanding these data and the expected removal processes in the
treatment system,the TT also noted that color loads , though not a major
difference in the limited data being considered (pg. 11),was the reverse of
what would be expected when CRP was operating(i.e.,higher color loading)
versus color discharge when CRP was not operating(i.e., lower color
loading); see the Addendum to the TT Report
Comment 8)The Tech Team's assertion about the effectiveness of C102 treatment of
CRP purge is inaccurate. Although,it was about 80% effective in some cases,the report
failed to mention that it was about 0% effective in other cases. If the statement is left in
the report"as is,"it could create an impression that BRPP; for some reason,refuses to
implement an inexpensive and effective method to reduce color, which is false.NCDWQ
believes this treatment option should be studied to determine its feasibility and will
recommend this as part of the TRW.
Response:
• The TT acknowledges (pg. 10)variability in results (from as low as 0 percent
to as much as 80 percent)(see also the Addendum to the TT report), and that
• NC and BRPP support the TT's recommendation (pg. 10, 16) to further
study this segregated stream pretreatment technology, and the TT
recommended implementation as appropriate during the upcoming permit
term, given that CRP color is now a significant portion of the total color
discharged from the mill processes
Comment 9)The Tech Team's recommendation to add a daily maximum color limit is
inappropriate. The DM limits are typically placed in the permit for a parameter that
exhibits acute toxicity or where there are standard effluent guidelines for a DM limit.
Response:
• NPDES permit regulations(40 CFR 122.45(d)(1))for continuous discharges
require daily maximum effluent limitations,including those necessary to
achieve water quality standards (WQBELs); it is not impracticable to
calculate such limits
6
• The TT notes that BRPP's progress in reducing the annual average of color
discharges has made increasingly apparent the importance of reducing day-
to-day effluent variability (pg. 15, 17)
• Daily maximum permit limits would further focus efforts on reducing day-to-
day process and wastewater variability, and thus enhance the consistency of
day-to-day final effluent and in-stream water quality, and would be
consistent with limits for other parameters controlled in the present BRPP
permit(BOD5, TSS, AOX, etc.)
• The TT notes (pg. 19) the suggested range of values for a daily maximum
effluent limitation for color was exceeded only for two days during 2006 for
which data were available; both days were attributable to unexpected and
first-time CRP tank overflows
• To address this overflow, the TT recommended that BRPP continue to move
forward with the already planned CRP sump and containment project to eliminate
unplanned releases of this highly-colored material to the mill sewer
• Thus,it is not impracticable to achieve such a daily maximum limit(see
responses to comments#3 and#4)
Comment I O) The Tech Team's recommendation to reduce the kappa factor is not
feasible, as the market currently requires paper and paperboard with higher brightness
values,which requires a higher degree of bleaching. If the kappa factor were to be
reduced,the mill would be forced to do more bleaching at the second [bleaching] stage in
order to produce marketable products. The filtrate from the second stage is not recycled,
and shifting more bleaching to the second stage would result in more color production.
Response:
• In preparation for the current renewal of their NPDES permit,Liebergott&
Associates and GL&V(2006)repeated their recommendations, also made in 2001,
for BRPP to reduce the kappa factor in the Di stage and replace some of the CIO
bleaching power with oxygen and/or peroxide in the extraction stages(previously
discontinued by BRPP),for both bleach lines(see also comments 13—15 and
responses); increasing the operating temperature of the hydrogen peroxide
extraction stages to reduce effluent color also was identified as a viable option
during this review process
• Recommendations by the consultant in 2006, also embraced by the TT(see pgs 3,4,
and 16),recognized changes in market-driven requirements for higher brightness
pulp for certain paper products produced since 2001 and the resulting need for
adjustments in both the chemicals used and their application rates in the bleaching
sequence
• In addition to allowing for higher pulp brightness products while reducing the color
of the bleach plant effluents,use of a lower kappa factor with the above process
changes and optimizations means BRPP would
use less CI02 and also caustic to bleach,thus with increased unit costs for
chemicals cost savings have been under-estimated from previous TT
estimates
➢ reduce the total chloride content of bleach plant filtrates, and thus also
7
➢ reduce the amount of color CRP purge contributes to the mill effluent(pg.4)
• All of the above considerations are contingent upon further evaluating and,if
feasible,implementing on an expedited basis second stage OD on the pine line,
enhanced extraction stages,and other process ; see response to comment#5
Comment 11) The Tech Team recommends BRPP to investigate possible ways to reduce
black liquor carryover in the direct contact evaporator. The name of the equipment used
in the process is a"direct contact evaporator,"which already implies that black liquor is
in direct contact with the exhaust and it is physically impossible to reduce carryover
because of the existing direct contact. The proposed solution is to replace the existing
[recovery] boiler,which is extremely expensive and not economically feasible.
Response:
• The TT noted that replacement of the recovery boiler is an existing option,in
theory, but also noted it". .. would probably represent a capital cost of over
$100 million. . . .Analysis of all the economics and long term life of the mill
would be necessary to evaluate such a major investment"
• In light of this fact,because DCE carryover typically is not an issue at most
mills, the TT instead recommended "On a more modest level,it may be
feasible to reduce black liquor carry over by further evaluating in detail and
adjusting operating conditions in the existing DCE's." (pg. 12) The Tech
Team will provide suggestions if requested by BRPP.
Comment 12) The comparison to the Glatfelter Permit limit is inappropriate in the
technical report. There are several significant differences between the mills,the critical
difference being that the Glatfelter mill purchases a substantial amount of pulp to produce
paper. Since color is predominantly produced during the pulping process, any technical
comparison should be based on the amount of pulp production by the two mills.
Response:
• The TT considered and acknowledged the differences in processes and pulp
production between the two mills,including the amount of purchased pulp
employed at the Glatfelter mill
• Notwithstanding these differences, substantial differences were noted in
production normalized end-of-pipe color discharge loadings between the two
mills (pg. 14, 19)(see also the Addendum to the TT report)
Comment 13) The Tech Team claims that the mill consultant recommended adding
oxygen and peroxide to E stages,when in fact it was recommended to evaluate such
additions, not to implement it immediately.
8
Response:
• The consultant's report does state the recommendation to "evaluate the use
of oxygen and peroxide fortification of the extraction stage at reduced Dl
factors."
• The TT discusses (pg. 3,4)and also recommends the changes be
implemented (pg. 16); see also response to Comment 10, above.
Comment 14) The Tech Team claims the need to use oxygen in the extraction state on the
hardwood line. BRPP has used oxygen for an extended period of time and did not
observe any effect on the effluent color.
Response:
• See response to comment 10 and 13, above.
• The Tech Team continues to believe these changes are still viable for both
fiber lines and should remain as high certainty options that should be
implemented during this permit cycle for further color reduction. (pg. 4)
Comment 15)The Tech Team claims that the consultant recommended replacement of
CIOZ bleaching power with oxygen and/or peroxide. This is not true, as the consultant
only recommended evaluation of this option not replacement.
Response:
• See response to comment 10, 13 and 14, above.
Comment 16) The Tech Team's insistence on the need to recycle more filtrate from the
hardwood line is unfounded. This proposal is not operationally feasible since the
hardwood line has no metal removal process,which is necessary to successfully recycle
bleach filtrate. The construction of MRP and BFR would cost about$30 million and
would not be economically feasible,
Response:
• Installing a second stage OD system on the softwood fiber line (see response
to Comment#5 and#10 regarding again evaluating this technology,
enhanced extraction, and other process improvements, and reaffirming it is
appropriate to be implemented) would reduce the load on the BFR system,
and thus the TT noted (pg.4),it may be possible to process an increased
portion of these hardwood filtrates through the existing BFRTm process
(notably CRP) and reduce the total color load discharged to the sewer
• The BFR system as originally installed was adequately sized (CRP
component; MRP was not adequately sized,plus reliability problems which
have since been overcome) to accommodate some hardwood fiber line
filtrates (Eo) and thus the original TT summary report included a (then)
near-term recommendation for partial BFR(using CRP) on the hardwood
9
fiber line(see Summary Report 12/02/1997,pgs.4, 5-6); recent addition of
ion exchange media could possibly reduce or eliminate the MRP bottleneck
• Thus, notwithstanding further recent comment from BRPP and NC to the
contrary, a second separate and complete BFR process for the hardwood
fiber line, including both the Metals Removal Process (MRP) and Chloride
Removal Process (CRP),would not be necessary and thus was not considered
by the TT
Comment 17) The Tech Team's use of the following phrase—'BRPP believes that
secondary effluent showed improvement' is improper(page 7) and should be stated
definitively,not as a belief.
Response:
• The TT's statement was based on an interpretation of the data presented by
BRPP in Table 4, pg 7; the report also acknowledges improvements in
performance(pgs. 2,4, 5, 6, 13, 15, 17); this statement is consistent with
BRPP's comment,but could be more explicit; see Addendum to TT report
Comment 18)The Tech Team's remark on page 16 of the memo regarding mills that are
required to implement end-of-pipe color removal technology fails to mention that those
mills do not have extensive in-mill process controls, which has been emphasized at
BRPP.
Response:
• The TT report(pg. 14) acknowledges more than one important caveat
regarding differences among the mills, including that". . . neither the Valdivia
Thi
mill nor the Glatfelter mill employs the BFR system as applied at BRPP."
• However, the TT report also notes that both of the other two mills do have in
place extensive and very effective in-plant controls (e.g., BMPs) (pg. 14)
Comment 19) The Tech Team's selected strategy to "...ultimately eliminate discharges
of highly colored wastewaters"is unrealistic. Continued reduction,when feasible, is
certainly NCDWQ's goal for colored wastewater discharges,but any reduction or
elimination must be weighed along with the technically and economically feasible issues.
Response:
• The TT recommended(pg 15, 16) strategies included continuing". . . to
improve the performance of BMPs to further substantially reduce and
ultimately eliminate discharges of highly-colored [untreated] wastewaters
directly to the wastewater treatment system (emphasis added). . . "
• It was implied,but not explicitly stated, that this recommendation for highly
colored wastewaters focused on both
10
➢ further reducing leaks and spills (e.g., moving forward with BRPP's
planned CRP sump project), and
➢ segregated stream pretreatment as necessary to ensure reduced color
loadings introduced into the wastewater treatment system
• The TT
➢ did not envision completely eliminating highly colored untreated
wastewaters to the treatment system, though this is always a worthy goal
➢ notes that all projects are evaluated for technical and economic feasibility
Comment 20) The Tech Team's recommendation to avoid releasing CRP purge streams
during low flow condition is not operationally feasible. CRP is an integral part of the
BFR, which produces significant reductions in color discharges.
Response:
• The TT notes that(see response to Comment 6) -
➢ this mill has operated in the past for weeks at a time without the CRP
running
➢ given the small daily mass of chlorides generated by CRP operation
relative to the large inventory of chlorides in the recovery loop,weeks are
needed for significant build up of chlorides that could likely cause
operational problems
• Alternatively,also given the modest volume of CRP purge wastewater
(approximately three tank truck loads per day) it could be feasible to avoid
discharging this wastewater during those infrequent periods of very low flow
conditions in the Pigeon R.
• The TT notes
➢ this is an option for contingency only, and need not be among the first
options considered
➢ it will be important to evaluate offsetting revenues (or costs) for use of the
black liquor solids at another off-site facility, or the cost of disposal at an
off-site hazardous waste treatment and disposal facility
Comment 21) The paragraph on page 17 that contains the recommendation to curtail
production during low flow contradicts itself. In the beginning the memo acknowledges
that color treatment is best during long periods of steady production. Then it
recommends reducing production,which has been shown to increase color load to the
receiving stream.
Response:
• The TT only reiterated past consideration of this option, a requirement of the
original Settlement Agreement,to curtail pulp production during periods of
low stream flow
• The TT also noted as it did previously,however,that". . . this should
continue to be considered an option of last resort,given that best
11
performance has been noted by the mill to be during extended periods of
steady production and greater risk of elevated color during process
shutdown and startup. See Low Flow Contingency Plan, December 1, 1998."
(pg. 17)
Comment 22) The Tech Team recommends further trials to evaluate potential of
ozone/chloride dioxide(ZD)bleaching for the hardwood line. . . .Due to the high cost
and unlikely benefit,this technology is not economically feasible and should not be
evaluated.
Response:
• As noted on pg. 8, "The Tech Team concludes that BRPP's investigation of the
potential for hardwood pulp ozone bleaching to reduce the mill's effluent
color was inadequate."
• As also noted on pg. 8, "Notwithstanding the above-noted inadequacies and
upon reflection,the Tech Team believes that further laboratory trials appear
not to be necessary because this application ofZD technologu may not be
appropriate for this mill at this time (emphasis added). Therefore,ZD
technology is considered a technology option of lowest certainty for
application at this mill at this time."
• Contrary to past practice by BRPP for lowest certainty options, the Tech
Team agrees this technology need not be studied within the 2008 -2013
permit term; however, the Tech Team does consider the ZD process an
available option for in-process changes and modifications in subsequent
permit cycle(s) if further reductions in color discharges are found to be
necessary in the future
Comment 23) The Tech Team's statement on page 6 that`BRPP did not report one single
event or one single type of event that caused high color in the primary effluent"is
incorrect. BRPP provided information on the cause of some elevated primary effluent
color days in the March 2007 response to additional questions.
Response:
• The TT noted (pg. 6)the following reasons the mill presented for high color
days: planned mill outages, unplanned outages, a CRP slurry tank overflow,
and a release from the evaporator related to an equipment failure. However,
in the TT's opinion, these events by themselves did not account for high color
and variability of the primary effluent. See the Addendum to the TT report
12
ADDENDUM to Tech Team Report, released September 5,2007 by EPA Region 4
December 19, 2007
1. The first sentence, last paragraph on pg 6, is being amended to read as follows:
"The Tech Team continues to believe that the information presented in Table 3,
among other available process and wastewater treatment systems performance data,
holds valuable indicators and clues to the sources and solutions to further improving
the overall performance and reducing the variability of effluent color discharged to
the Pigeon River."
2. On pg. 11, second paragraph, the third and fourth sentences are being amended to
replace them as follows:
"The Tech Team notes that the difference in treated effluent color loads,though not
major based upon the limited data considered, was the reverse of what would be
expected when the CRP was operating (i.e.,higher color loading) compared to
when the CRP was not operating(i.e., lower color loading)."
3. On pg. 10, the second paragraph under the heading Treatment of CRP Purge
Stream for Color Removal, the third sentence is amended to read as follows:
"BRPP found that C102 bleaching could remove 75 to 90 percent of the color in the
CRP stream,but with variability noted to range on occasion to as low as 0 percent
removal."
4. On pg. 19, under Comparison to Glatfelter Permit Limits, the end of this
paragraph is amended by replacing the last sentence with the following:
"The annual average color discharge, expressed in kg/Metric Ton(kg/MT), for the
Glatfelter mill is approximately 10.4 kg/MT pulp produced on site,while at BRPP
the long term average color discharge is 13.1 kg/MT pulp produced on site. The
Glatfelter 30-day (monthly) average and maximum day effluent limitations
(included in its NPDES permit) are substantially more stringent than those
suggested by the Tech Team for BRPP. However,the Tech Team believes,
consistent with similar efforts by BRPP in the past, it is appropriate for BRPP to
undertake data gathering and develop statistically derived effluent limitations as an
alternative to the effluent limitations suggested by the Tech Team. Such an effort
would follow implementation by BRPP of those technologies found to be
technically feasible and economically achievable during the 2008—2013 permit
term. The following table presents a comparison of actual long term performance
and suggested BRPP and actual PH Glatfelter end-of-pipe effluent limitations for
color.
Effluent Limitations and Performance
P H Glatfelter and Bide Ridge Paper Products
BRPP PH Glatfelter 3
Parameter Low High Winter Summer Winter Summer
Kg/MT" KWMT2 Kg/MT KgtMT PCU PCU
Long 13.14 10.4 NA
Term Ave.
Ann. Ave. 22.7 26.2 NA NA NA NA
Max.Niax. 31.8 36.7 11.6 13.2 123 140
Monthly
(30-day)
Ave.
Max.Day 59.0 68.7 23.3 26.5 246 280
NA Not Applicable
1 Source:Memorandum from EPA Technology Team to Technology Review Workgmup,"Color Removal Strategies for Blue Ridge
Paper Products,Inc.,"released by EPA Region 4 on September 5,2007,and as amended by this Addendum December 19,2007
2 Range of end-of-pipe effluent limitations for color suggested by EPA Technology Team;
3 Source:"Water Quality Protection Report Amendment,"P.H.Glatfelter Company,Spring Grove Borough and Jackson Township,
York County,for the Renewal of NPDES Permit No.PA 0008869,Pennsylvania Department of Environmental Protection,
Southcentral Regional Office,June 1,2007.
4 Actual long term performance;for PHG,the above values in kg/MT were calculated from wastewater flow,effluent limitations in
PCU,and on-site pulp production(excluding purchased pulp)for comparison with BRPP
5 Effluent limitations in NPDES permit;units:PCU
11
5. On pg. 15, the bullet at the bottom of the page is amended to read as follows:
"Continue to improve the performance of BMPs to further substantially reduce, or
eliminate, if technically feasible and economically achievable, discharges of
untreated wastewaters directly to the wastewater treatment system through further
improvements in—"
6. The next to last bullet on pg. 16 is amended to read as follows:
"If the CRP purge color is found not to be removed in secondary treatment, avoid
releasing the CRP purge during periods of low stream flow(or truck it offsite); this
should be considered as a contingency after consideration of possible offsetting
revenues or costs, such as off-site use of black liquor solids or hazardous waste
treatment and disposal facility costs."
7. On pg. 6, in the second complete paragraph below Table 3, the third sentence is
amended to read as follows:
"BRPP did not report any one single event or types of events that by themselves
account for high color in the primary effluent."
8. At the bottom of pg. 13,prior to the start of the next section entitled Improving
WWTP Performance Using Chemicals, the last paragraph is being amended by
adding the following sentence"
"The Tech Team also notes that a sustained effort to further study and control this
phenomena will be required beyond the upcoming permit term given that process
changes and BMP improvements to be made during the 2008 —2013 permit term
will more than likely further change the mechanisms underlying `sewer generated
color."'
9. The first sentence at the top of pg. 7 is amended to read as follows:
`BRPP stated that the color reduction projects completed since 2001 and ongoing
management of mill processes and wastewater treatment systems have resulted in
secondary effluent improvement, as shown by data provided by BRPP,presented in
Table 4 (below).
10. The List of References is amended by adding an additional items as follows:
"Water Quality Protection Report,"P. H. Glatfelter Company, Spring Grove
Borough and Jackson Township,York County, for the Renewal of NPDES Permit
No. PA 0008869, Pennsylvania Department of Environmental Protection,
Southcentral Regional Office,October 23, 2006.
"Water Quality Protection Report Amendment,"P. H. Glatfelter Company, Spring
Grove Borough and Jackson Township,York County, for the Renewal of NPDES
Permit No. PA 0008869,Pennsylvania Department of Environmental Protection,
Southcentral Regional Office,June 1, 2007.