EPA-450/1-73-002
ATMOSPHERIC EMISSIONS
FROM THE PULP AND PAPER
MANUFACTURING INDUSTRY
L.S. ENVIRONMENTAL PROTECTION AGENCY
-------
-------
EPA-450/1-73-002
ATMOSPHERIC EMISSIONS
FROM THE PULP AND PAPER
MANUFACTURING INDUSTRY
Cooperative Study Project
National Council of the Paper Industry
for Air and Stream Improvement, Inc.
and
Environmental Protection Agency
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Water Programs
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
September 1973
-------
This report is published by the Environmental Protection Agency to report
information of general interest in the field of air pollution. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - as supplies permit - from the Air
Pollution Technical Information Center, Environmental Protection
Agency, Research Triangle Park, North Carolina 27711, or from the
Superintendent of Documents.
Publication No. EPA-450/1-73-002
-------
PREFACE
To provide reliable information on the air pollution aspects of the pulp
and paper industry, the National Council of the Paper Industry for Air and
Stream Improvement, Incorporated (NCASI), and the Office of Air Quality
Planning and Standards of the U.S. Environmental Protection Agency
(EPA) entered into an agreement in April 1967. A cooperative program was
established to study atmospheric emissions from the various industry
processes and publish information about them in a form helpful to air
pollution control and planning agencies and to the pulp and paper industry
management. Direction of this study was vested in a NCASI-EPA Steering
Committee composed at the time of completion of the following
representatives:
EPA NCASI
*Stanley T. Cuffe *Isaiah Gellman
John L. McGinnity Peter Wrist
Joseph J. Sableski Malcolm L. Taylor
Mr. Edwin J. Vincent of EPA and Mr. Rusell O. Blosser of NCASI were the
principal investigators during much of this project and authored much of
this report. Before joining the steering committee, Mr. Joseph J. Sableski of
EPA and Dr. Isaiah Gellman of NCASI also served as principal
investigators.
Information in the report describes the nature and range of atmospheric
emissions during normal operating conditions and the performance of
established devices and methods employed to limit and control these emis-
sions.
* Principal representative.
iii
-------
ACKNOWLEDGMENTS
Many companies and individuals in the pulp and paper industry have
been helpful in providing plant-visit and questionnaire data for this study.
The sponsors also wish to acknowledge the contributions of the source
testing personnel of the Emission Measurement Branch, Office of Air
Quality Planning and Standards, Environmental Protection Agency, and of
the Division of Chemistry and Physics, Office of Research and Monitoring,
Environmental Protection Agency.
iv
-------
TABLE OF CONTENTS
LIST OF TABLES viii
LIST OF FIGURES xi
ABSTRACT xii
GLOSSARY xiii
CONVERSION FACTORS, BRITISH TO METRIC UNITS xvi
SUMMARY xvii
Production of Chemical Pulp xvii
Chemical Pulping Processes
Kraft (Sulfate) Process xviii
Sulfite Processes xviii
Semichemical Processes xviii
Soda Process xviii
Kraft Process, Specific Emission Sources and Controls xviii
Types of Emissions xviii
Digester Relief and Blow xix
Multiple Effect Evaporators xix
Recovery Furnace Systems xx
Black Liquor Oxidation Systems xxi
Smelt Dissolving Tanks xxi
Lime Kilns xxii
Brown Stock Washers xxii
Emission Ranges xxii
Semichemical Sulfite Process xxiii
Sulfite Process xxiii
Steam and Power Generation xxiv
INTRODUCTION 1
Background 1
Sources of Information 1
Questionnaire Surveys 2
Field Sampling Program 2
Literature 3
Pulp and Paper Manufacturing Industry 3
Current Production 4
Industry Growth Trends 5
KRAFT (SULFATE) PULPING PROCESS 7
Introduction 7
Process Description 7
Raw Materials and Process Characteristics 7
Emission Sources 9
Gaseous Emissions 9
Particulate Emissions 10
-------
Emission Control Systems — General 10
Gaseous Emissions 10
Particulate Emissions 12
Specific Emission Sources and Controls 12
Digester Relief and Blow 12
Emissions 13
Control Techniques 13
Emission Data 14
Multiple Effect Evaporators 17
Emissions 17
Emission Data 17
Control Techniques 18
Kraft Recovery Furnace Systems 19
Introduction 19
Composition and Control of Emissions from Kraft
Recovery Furnaces 24
Composition and Control of Emissions from Flue Gas
Direct Contact Evaporators 30
Emission Data 34
Black Liquor Oxidation Systems 44
Designs, Application, and Performance 44
Emission Data 45
Control Techniques 48
Smelt Dissolving Tanks 48
Emissions ...., 49
Control Techniques 49
Emission Data 49
Lime Kilns 53
Emissions 54
Control Techniques 54
Emission Data 55
Brown Stock Washing Systems and Other Miscellaneous
Sources 58
Emission Data 58
Control Techniques 61
SEMICHEMICAL SULFITE PULPING PROCESSES 63
Introduction 63
Raw Materials 63
Process Description 63
Product Yield 64
Emissions 64
Sulfur Dioxide Absorption Tower 64
Blow Tank 65
Chemical and Heat Recovery Furnace 65
Emission Data 66
Control Techniques 67
SULFITE PULPING PROCESS 69
Introduction 69
Raw Materials 69
vi
-------
Process Description 69
Product Yield 70
Emissions 70
Absorption Towers 70
Blow Pits 70
Recovery Units 71
Multiple Effect Evaporators 72
Control Techniques 72
Sulfiting Tower 72
Blow Pit 72
Recovery Furnace Systems 73
Emission Data 74
STEAM AND POWER GENERATION 77
Introduction 77
Fuels Used 77
Steam Usage 80
Types of Boilers 80
Emissions 82
Control Techniques 82
New Technology 82
APPENDIX A: DETAILED EMISSION DATA FROM
QUESTIONNAIRE SURVEY 87
APPENDIX B: SAMPLING AND ANALYTICAL PROCEDURES
AND EQUIPMENT 95
Summary of Procedures 95
Gas Analysis 95
Questionnaire Data 95
Special Studies 95
EPA Field Investigation 96
Particulate Sampling and Analysis 96
EPA Mobile Field Sampling Laboratory 96
Gas Analysis 97
Instrumentation 97
Dilution System 98
Gas Chromatographic-Flame Photometric System 102
Particulate Sampling and Analysis 108
Sampling Site and Traverse Points 108
Sampling Train 108
Analysis 108
APPENDIX C: ODOR SURVEY Ill
Introduction Ill
Equipment and Procedures Ill
Results 112
Evaluation 113
REFERENCES 117
vii
-------
LIST OF TABLES
Table Page
1 Summary of U.S. C'hemical Pulp Mill Production, 1969 5
2 Odor Thresholds of Some Malodorous Sulfur Compounds 9
3 Gaseous Emissions from Digester Relief and'Blow, Ques-
tionnaire Data 15
4 Sulfur Compound Emissions from Continuous Digesters,
Summary of Questionnaire Data 16
5 Composition of Gas Streams Vented to Lime Kiln, EPA Test
Results 17
6 Emissions from Multiple Effect Evaporators, Summary of
Questionnaire Data 19
7 Gaseous Emissions from Multiple Effect Evaporator, Ques-
tionnaire Data 20
8 Control Techniques for Multiple Effect Evaporator Emis-
sions, Summary of Questionnaire Data 21
9 Effect of Furnace Firing Rate and Air Supply on TRS Emis-
sion for Kraft Recovery Furnace 28
10 Analysis of Direct Contact Evaporator Function in Altering
Furnace Gas TRS Content 33
11 Particulate Emissions from Recovery Furnaces Controlled by
Electrostatic Precipitators, Averaged by Decile Groups 34
12 Particulate Emissions from Recovery Furnaces with Venturi
Scrubber Systems, Questionnaire Data 36
13 Particulate Emissions from Recovery Furnaces with Second-
ary Scrubbers after Electrostatic Precipitators,
Questionnaire Data 37
14 Gaseous Emissions from Recovery Furnaces without Black
Liquor Oxidation, Questionnaire Data 39
15 Gaseous Emissions from Recovery Furnaces with Black
Liquor Oxidation, Questionnaire Data 41
16 Recovery Furnace System Particulate Emissions, Summary
of EPA Test Results 42
17 Recovery Furnace System Gaseous Emissions, Mill A without
Black Liquor Oxidation, EPA Test Results 43
18 Recovery Furnace System Gaseous Emissions, Mill B with
Black Liquor Oxidation, EPA Test Results 43
19 Recovery Furnace System Gaseous Emissions, Mill C with
Black Liquor Oxidation, EPA Test Results 43
20 Reduced Sulfur Emissions from Black Liquor Oxidation
Systems, Special Studies Data 46
viii
-------
Table Page
21 Gaseous Emissions from Black Liquor Oxidation, Ques-
tionnaire Data 47
22 Emissions from Black Liquor Oxidation Tanks, EPA Test
Results 48
23 Effectiveness of Smelt Tank Particulate Control Devices,
Questionnaire Data 50
24 Smelt Tank TRS Gaseous Emissions, Summary of Ques-
tionnaire Data 51
25 Smelt Tank TRS Emissions, Special Studies Data 52
26 Smelt Tank Gaseous Emissions, EPA Test Results 53
27 Lime Kiln Scrubber Efficiency, Summary of Questionnaire
Data 56
28 Gaseous Lime Kiln Emissions, Summary of Questionnaire
Data 57
29 Lime Kiln Particulate Emissions, EPA Test Results 57
30 Lime Kiln Gaseous Emissions, EPA Test Results 57
31 Brown Stock Washer System TRS Emissions, Special Studies
Data 59
32 Brown Stock Washer System TRS Emissions, Roof Vents
and Under Vents, Special Studies Data 60
33 Brown Stock Washer System TRS Emissions, Total System,
Special Studies Data 60
34 Brown Stock Washer System TRS Emissions, Alternate Use
of Fresh and Condensate Wash Water, Special Studies
Data 60
35 Brown Stock Washer System Gaseous Emissions, EPA Test
Results 61
36 Extent of Controls Used in Semichemical Pulping Opera
tions, Questionnaire Data 66
37 Sulfur Dioxide Emissions from Semichemical Processes,
Questionnaire Data 67
38 Extent of Controls Used in Sulfite Pulping Operations,
Questionnaire Data 74
39 Sulfur Dioxide Emissions from Sulfite Process, Question-
naire Data 74
40 Types of Pulp and Paper Mills Reporting Information for
Power Boiler Questionnaire Survey 78
41 Fuel Consumption in Pulp and Paper Mill Power Boilers 78
42 Fuel Usage Data for Pulp and Paper Mill Power Boilers 79
43 Characteristics of Fuels Burned in Power Boilers at Pulp and
Paper Mills 79
44 Steam Use Distribution at Pulp and Paper Mills, Power
Boilers Only 80
45 Types of Power Boilers Used in Pulp and Paper Industry 81
46 Power Boiler Flue Gas Characteristics 83
47 Power Boiler Particulate Control Equipment Data 83
48 Emission Data from Power Boilers Fired with Coal Only 83
IX
-------
Table Page
49 Emission Data from Power Boilers Fired with Bark/Wood
Plus Other Fuels 85
A-1 Particulate Emissions from Recovery Furnaces Controlled by
Electrostatic Precipitators 88
A-2 Particulate Emissions from Smelt Tanks 90
A-3 Gaseous Emissions from Smelt Tanks 91
A-4 Particulate Emissions from Lime Kilns 92
A-5 Gaseous Emissions from Lime Kilns 94
C-l Odor Panel Screening Tests 112
C-2 Example of Odor Panel Response 112
C-3 Odor Panel Results, Mill B 113
C-4 Odor Panel Results, Mill C 114
-------
LIST OF FIGURES
Figure Page
1 Kraft Pulping Process 8
2 Multiple Effect Long-tube Vertical Evaporators 18
3 Typical Kraft Recovery Furnace System Options 22
4 Effect of Solids Firing Rate on Reduced Sulfur Emissions and
Steam Generation Efficiency 27
5 Observed Frequency of Total Reduced Sulfur Concentrations
in Exit Gases from Recovery Furnaces with Good Combus-
tion Controls 29
6 Observed Frequency of Total Reduced Sulfur Concentrations
in Exit Gases from a Recovery Furnace with Limited Com-
bustion Controls 30
7 Total Sulfur Increase across the Direct Contact Evaporator,
Sodium Sulfide Concentrations from 0 to 20 g/liter 32
8 Total Sulfur Increase across the Direct Contact Evaporator,
Sodium Sulfide Concentrations from 0 to 1 g/liter 32
9 Particulate Emissions from Recovery Furnaces with Elec-
trostatic Precipitators, Questionnaire Data 35
10 Smelt Tank Particulate Emissions, Questionnaire Data 50
11 Lime Kiln Particulate Emissions, Questionnaire Data 56
B-l Arrangement of Gas-Sampling System of Mobile Sampling
Van 99
B-2 Sample Dilution System and Related Equipment 100
B-3 Flame Photometric Detector 103
B-4 Gas Chromatographic-Flame Photometric Detector for
Low-molecular-weight Sulfur Compounds 104
B-5 Chromatogram of Low-molecular-weight Sulfur
Compounds 105
B-6 Gas Chromatographic-Flame Photometric Detector for
High-molecular-weight Sulfur Compounds 106
B-7 Chromatogram of High-molecular-weight Sulfur
Compounds 107
B-8 Minimum Number of Traverse Points 109
B-9 Particulate Sampling Train 110
C-1 Example of Method of Estimating Dilution Level for 50
percent Response 114
XI
-------
ABSTRACT
This report contains information on the nature and quantities of the
atmospheric emissions from chemical pulping operations, principally the
kraft process. The information was gathered in a cooperative study by the
National Council of the Paper Industry for Air and Stream Improvement,
Inc. (NCASI), and the Environmental Protection Agency (EPA). Principal
sources of information were a comprehensive questionnaire sent to all the
pulp mills, special NCASI studies reported in Technical Bulletins, other
literature sources, and a field sampling program conducted by EPA.
Control techniques are described and emission ranges reported for each of
the operations involved in the chemical pulping processes.
xii
-------
GLOSSARY
ABBREVIATIONS
acfm
ADP
Btu
°C
cfm
cm3
dscfm
EPA
°F
ft3
g
gal.
gpm
gr
hr
i.d.
Ib
max.
min.
min
ml
NCASI
NSSC
ppb
ppm
scf
sec
T
TRS
actual cubic feet per minute
air dried pulp (assumed to contain 10 percent moisture)
British thermal units
degrees Celsius (centigrade)
cubic feet per minute
cubic centimeters
dry standard cubic feet per minute
U. S. Environmental Protection Agency
degrees Fahrenheit
cubic feet
grams
gallons
gallons per minute
grains
hours
inside diameter
pounds
maximum
minimum
minute
milliliter
National Council of Paper Industry for Air and Stream
Improvement
neutral sulfite semichemical
parts per billion by volume
parts per million by volume
pounds per square inch gauge
standard cubic feet
seconds
tons
total reduced sulfur (expressed as an equivalent amount of
hydrogen sulfide)
xiii
-------
CHEMICAL SYMBOLS
CaCO3
CaO
Ca(OH)2
CH3SH
(CH3)2S
(CH3)2S2
C02
H2O
H2S
H2S03
N2
Na
Na2CO3
NaOH
Na2S
Na2SO3
Na2S203
02
SO2
Calcium carbonate
calcium oxide
calcium hydroxide
methyl mercaptan
dimethyl sulfide
dimethyl disulfide
carbon dioxide
hydrogen
water
hydrogen sulfide
sulfurous acid
nitrogen
sodium
sodium carbonate
sodium hydroxide
sodium sulfide
sodium sulfate
sodium thiosulfate
oxygen
sulfur dioxide
DEFINITIONS
Black liquor
Green liquor
Heavy (strong) liquor
Oxidation efficiency
Recovery furnace
Smelt
Liquor recovered from the digesters.
Liquor made by disolving smelt in weak wash
liquor.
Black liquor that has been concentrated in
preparation for recovery.
Percentage of sodium sulfide in the black liquor
that is oxidized by air introduced into the liquor.
Combustion unit used to recovery the spent
chemicals from the digestion liquor and to
produce steam.
Molten chemicals from the recovery furnace,
consisting mostly of sodium carbonate and
sodium sulfide.
XIV
-------
Sulfiditity
Weak wash liquor
White (cooking) liquor
Weak liquor
Percentage of sodium sulfide to total alkali in
white liquor, obtained by the formula
Na2S
Na2S + NaOH
xlOO
where the sodium compounds are expressed as
sodium oxide.
Liquid stream resulting from washing of the
lime mud.
Liquor made by causticizing the green liquor
with lime. White liquor is ready for use in the
digesters.
Black liquor as recovered from the digesters
prior to concentration (see "heavy liquor").
xv
-------
CONVERSION FACTORS,
BRITISH TO METRIC UNITS
Multiply
British thermal units
cubic feet
degrees Fahrenheit3
feet
gallons
grains
inches
inches of water
pounds (mass)
pounds per square inch
tons
By
1.06 x 103
2.83 x 10'2
5/9
3.05 x 10"1
3.79 x 10~3
6.48 x 10'5
2.54 x 10"2
2.49 x 102
4.54 x 10~1
6.89 x 103
9.07 x 102
To obtain
newton-meters
cubic meters
degrees Celsius (centigrade)
meters
cubic meters
kilograms
meters
newtons per square meter
kilograms
newtons per square meter
kilograms
To obtain Celsius (centigrade) temperature (t ) from Fahrenheit
temperature (tf), use the formula tc = (tf-32)/1.8.
xvi
-------
SUMMARY
The pulp and paper manufacturing industry consists of two well defined
segments, pulping and paper making. Pulping is the conversion of fibrous
raw materials such as wood, cotton, or used paper into a material suitable
for use in paper, paperboard, and building materials. The principal source
of the fibers is wood. The fiberous material ready to be made into paper is
called pulp.
Wood pulp is prepared either mechanically or chemically. Mechanical
pulp is produced by grinding or shredding wood to free the fibers. In
chemical pulping processes, the wood fibers are freed by dissolving the
binding material (lignin) in chemical solutions. Mechanical pulping and the
paper making process itself produce negligible air pollution, except for the
boilers that produce steam and electric power to run the mills. Therefore,
this report is concerned only with the chemical pulping processes, power
boilers, bark-burning boilers, and combination boilers. Further, emphasis
is on the kraft process, which accounts for 81.5 percent of annual chemical
pulp production.
PRODUCTION OF CHEMICAL PULP
In 1970, the production of pulp by chemical pulping processes and the
number of mills involved were as follows:
Production,
Process Number of mills million tons
Kraft 116 29.6
Sulfite 38 3.0
Semichemical 50 3.6
Soda 5 0.2
Total 209 36.4
xvii
-------
CHEMICAL PULPING PROCESSES
Kraft (Sulfate) Process
In the kraft process, the digesting liquor is a solution of sodium
hydroxide and sodium sulfide. The spent liquor (black liquor) is
concentrated, sodium sulfate is added to make up for chemical losses, and
the liquor is burned in a recovery furnace, producing a smelt of sodium
carbonate and sodium sulfide. The smelt is dissolved in water to form green
liquor, to which is added quicklime to convert the sodium carbonate back
to sodium hydroxide, thus reconstituting the cooking liquor. The spent lime
cake (calcium carbonate) is recalcined in a rotary lime kiln to produce
quicklime (calcium oxide) for recausticizing the green liquor.
Sulfite Processes
Sulfite cooking liquors contain sulfurous acid and the bisulfite of
calcium, sodium magnesium, or ammonia. Calcium based liquor is
prepared by absorbing sulfur dioxide in water in a tower filled with crushed
limestone. Sodium based liquor is formed by absorbing sulfur dioxide in a
solution of sodium carbonate. Magnesium based liquor is made by
absorbing sulfur dioxide in a slurry of magnesium hydroxide. Ammonium
based liquor is made by absorbing sulfur dioxide and ammonia in water.
Semichemical Processes
Semichemical pulps are produced by digesting with reduced amounts of
chemicals, followed by mechanical treatment to complete the fiber
separation. The most extensively used process is the neutral sulfite
semichemical (NSSC) process. The cooking solution is a nearly neutral
sulfite solution containing an alkaline agent such as sodium carbonate,
bicarbonate, or hydroxide.
Soda Process
The cooking liquor is a solution of sodium hydroxide. The spent liquor
can be recovered by concentration and incineration. The make-up chemical
is sodium carbonate. This process has declined to relative insignificance.
KRAFT PROCESS, SPECIFIC EMISSION SOURCES AND CONTROLS
Types of Emissions
The emissions from the kraft process include both gaseous and
particulate matter. The gaseous emissions are principally hydrogen sulfide,
methyl mercaptan, dimethyl sulfide, dimethyl disulfide, and sulfur dioxide.
XV1I1
-------
The participate emissions are largely sodium sulfate from the recovery
furnace, smelt tanks, and lime kiln, as well as calcium compounds from the
lime kiln.
Hydrogen sulfide and the organic sulfides are extremely odorous, being
detectable at a concentration of a few parts per billion. Thus odor control is
one of the principal air pollution problems of a kraft pulp mill.
Digester Relief and Blow
The gases formed during batch digestion are vented to maintain proper
cooking conditions. At the end of the cooking cycle, the contents of the
vessel are blown to a tank at atmospheric pressure, flashing off large
amounts of steam, as well as noncondensable gases. Gases formed in the
digester include hydrogen sulfide, methyl mercaptan, dimethyl sulfide, and
dimethyl disulfide.
The steam in relief and blow gases is usually condensed to recover heat.
Noncondensables are either vented to atmosphere or treated. Treatment
methods include burning, scrubbing with alkaline solutions, and chemical
oxidation. Burning is usually accomplished in a lime kiln, and is the most
effective method. Scrubbing with sodium hydroxide (white liquor) is
effective only for hydrogen sulfide and methyl mercaptan. Scrubbing with
chlorine solutions is of limited effectiveness.
Six respondents to the questionnaire presented data on relief and blow
gases. No treatment facilities were indicated. Median value for relief gas
emissions was 1.2 pounds per ton of air dried pulp (Ib/T ADP) as total
reduced sulfur (TRS). Median emission rate for blow gases was 0.08 Ib/T
ADP as sulfur. Eleven respondents discussed treatment facilities. Of these,
six respondents indicated they burned these gases in lime kilns, and five
used chlorination stage bleachery effluent for treatment. EPA field test
results at two mills indicated that virtually complete destruction of reduced
sulfur gases was obtained by burning in lime kilns.
Multiple Effect Evaporators
Emissions from evaporation arise from noncondensable vent gases and
liquid condensate. Liquid condensate is usually sewered, and
noncondensables are often vented to the atmosphere. These gases contain
high concentrations of hydrogen sulfide and organic sulfides.
Noncondensable vent gases may be combined and controlled with
digester blow and relief gases, using incineration or alkaline scrubbing.
Six mills reported emission data from evaporator vents. The median rate
was 0.37 Ib/T ADP as hydrogen sulfide.
xix
-------
Gaseous sulfur compounds can be stripped from evaporator condensate
with steam or air and the off-gases incinerated in a lime kiln or separate
incinerator.
Recovery Furnace Systems
Concentrated black liquor is burned in a furnace to produce a smelt of
sodium carbonate and sodium sulfide that is used to reconstitute cooking
liquor. Steam is produced as a by-product.
In a conventional system, the final stage in concentrating the black
liquor utilizes the furnace flue gas in a direct contact evaporator. Emissions
from the system are those originating in the furnace plus those released in
the direct contact evaporator.
Gaseous emissions from the recovery furnace include hydrogen sulfide
and much smaller amounts of organic sulfides. These emissions are very
low from a well regulated furnace but can be considerable if operation is
not optimum.
Considerable amounts of hydrogen sulfide can be released in the direct
contact evaporator by the reaction of the acidic gases in the flue gas with
the sodium sulfide in the liquor. Such emissions can be reduced by black
liquor oxidation, which converts the sodium sulfide to sodium thiosulfate, a
more stable form. The amount of emission reduction is dependent on the
degree of oxidation. Very low levels, 0 to 3 parts per million (ppm) TRS, can
be reached as the degree of oxidation approaches completeness.
Several alternate systems that do not use a direct contact evaporator
have been developed, thus eliminating this source of emissions. One system
utilizes additional stages of indirect evaporation of the black liquor plus
additional heat exchange surface in the furnace. The other system utilizes a
flue gas-to-air heat exchanger plus an air contact evaporator. The flue gas
imparts heat to a stream of air. The air then concentrates the black liquor
in a contact evaporator. This air is then used as combustion air in the
furnace. A modification of this system eliminates the contact evaporator
but retains the heat exchanger to preheat the combustion air. Additional
indirect evaporation of the black liquor is also used in this system.
Total reduced sulfur emission data were reported for 42 conventional
recovery furnace systems where black liquor oxidation was not used.
Median TRS emission rate was 5.9 Ib/T ADP. The EPA test team
measured TRS emission rates from two mills with well designed and
operated black liquor oxidation systems. Emission rates were 0.19 and
0.075 Ib/T ADP, respectively.
Particulate emissions from the recovery furnace consist primarily of
sodium sulfate and sodium carbonate caused by small particle carry-over
and sublimation-condensation.
xx
-------
The electrostatic precipitator is the most widely used particulate control
device. Secondary scrubbers are sometimes used after electrostatic
precipitators. Where used, direct contact evaporators function as
precleaners. Two-stage venturi systems function as both direct contact
evaporator and final particulate collector.
Particulate emission data from 87 kraft recovery furnace systems with
direct contact evaporators and electrostatic precipitators were reported in
the questionnaire survey. The range of emission rates was 1.3 to 95 Ib/T
ADP. The median rate was 14 Ib/T ADP, with 10 percent less than 2.5 and
44 percent less than 10 Ib/T ADP. Data for 10 venturi recovery units show a
range of 15 to 115 Ib/T ADP, with a median value of 45 Ib/T ADP.
Emissions from seven systems with scrubbers following electrostatic
precipitators had a range of 1.8 to 13 Ib/T ADP, with a median value of 2.8
Ib/T ADP.
Black Liquor Oxidation Systems
Black liquor oxidation is the practice of oxidizing the sodium sulfide in
the liquor at least to the sodium thiosulfate stage, using air or oxygen, as
represented in the followingequation:
2Na2S + 02 + H20 Na2S2O3 + 2NaOH
Sodium thiosulfate will not react with carbon dioxide and sulfur dioxide in
the flue gas to produce hydrogen sulfide as does sodium sulfide.
Oxidation can be performed on either weak or heavy (strong) black
liquor. Packed towers or bubble tray towers are used for weak liquor
oxidation. Air sparged reactors, some with mechanical mixers, are used to
oxidize heavy liquor. Oxidation of weak liquor may reduce emissions in the
evaporation process, but reversion to sulfide in subsequent evaporation
stages or in storage may cancel this benefit. Heavy liquor oxidation is
advantageous in oxidizing liquors high in resin soap, which foam
excessively when oxidized weak.
During the oxidation of black liquor, the air passing through the liquor
strips out some reduced sulfur gases. Data from the questionnaires, EPA
tests, and special studies give median values of reduced sulfur emissions of
0.14 Ib/T ADP for weak black liquor oxidation systems and 0.10 Ib/T ADP
for strong liquor systems.
Smelt Dissolving Tanks
The molten smelt from the recovery furnaces is discharged into a tank of
water to form green liquor. Particulate emissions are entrained in the vent
gases. Some reduced sulfur gases are formed by reactions in the tank.
xxi
-------
Wire mesh mist eliminators are the most widely used participate control
device. Scrubbers and cyclones are also used. Questionnaire data give a
median emission rate to the atmosphere of 1.0 Ib/T ADP. An EPA test gave
a rate of 0.8 Ib/T ADP from a smelt tank controlled by a wire mesh mist
eliminator.
Gaseous emission test data from questionnaire data gave a median
emission rate of TRS of 0.1 Ib/T ADP as hydrogen sulfide. Two EPA tests
gave an average rate of 0.03 Ib/T ADP.
Lime Kilns
Lime kilns supply quicklime, which is slaked and used to causticize the
green liquor to produce white liquor. The spent lime mud, CaCO3, is
recycled back to the lime kilns.
Particulate emissions consist of sodium salts from sublimation-
condensation of salts retained in the sludge and calcium carbonate and
calcium oxide resulting from entrainment. Hydrogen sulfide can be formed
from the reaction of carbon dioxide in the flue gas with sodium sulfide
remaining in the lime mud after washing. Other organic sulfides can
originate in the scrubbing liquor used in the particulate control device.
Impingement scrubbers and medium efficiency venturi scrubbers are
used in controlling particulate emissions. Questionnaire data yielded a
median emission rate of 2.7 Ib/T ADP at a concentration of 0.4 grain per
dry standard cubic foot. Two EPA tests gave an average emission rate of 1.6
Ib/T ADP from a venturi scrubber. Gaseous emission data from the
questionnaires gave a median emission rate for TRS gases of 0.43 Ib/T
ADP. Two EPA test gave an average emission rate of 0.23 Ib/T ADP.
Brown Stock Washers
Gaseous emissions from brown stock washers occur from two points, the
roof vents of the hood over the filter and vents of vacuum pumps, called
under vents. The emissions are predominantly dimethyl sulfide and
dimethyl disulfide.
The level of the emissions changes when condensate is used instead of
fresh water. A special study of 17 washing systems gave median results as
follows: roof vents using fresh water — 0.04 Ib/T ADP as hydrogen sulfide;
roof vents using condensate — 0.35 Ib/T ADP; undervents using fresh
water — 0.08 Ib/T ADP; undervents using condensate — 0.11 Ib/T ADP.
Emission Ranges
From all the data acquired in the conduct of the study, typical values
were selected to illustrate the range of emissions from kraft mill operations.
XXH
-------
These data are tabulated below. In most cases, the average emission rate is
approximately in the middle of the range. The emissions are to the
atmosphere, following control devices where used.
Emission rate, Ib/T ADP
Process Particulates TRS
Recovery furnaces 1 to 25 0.05 to 12
Smelt tanks 0.05 to 2 0.01 to 0.6
Limekilns 0.5 to 7 0.02 to 1
Digesters — 0* to 2
Multiple effect evaporators — 0* to 1.5
Black liquor oxidation — 0.05 to 0.2
Brown stock washers — 0.01 to 0.9
SEMICHEMICAL SULFITE PROCESS
Semichemical pulps are produced by digesting with reduced amounts of
chemicals followed by mechanical treatment to complete the fiber
separation. The most extensively used process is the neutral sulfite
semichemical (NSSC) process. The cooking solution is a nearly neutral
sulfite solution containing an alkaline agent such as sodium carbonate,
bicarbonate, or hydroxide.
Spent cooking liquor may be discharged as a liquid effluent or
concentrated and burned with or without chemical recovery. In some cases,
a fluid bed furnace recovers the chemical as sodium sulfate and sodium
carbonate, which can be used as make-up chemicals in a kraft process. In
others using the cross-recovery method, the spent liquor is combined with
kraft black liquor for recovery and reuse in the kraft process.
Gaseous emissions in the NSSC process are essentially limited to sulfur
dioxide, except that in those cases where kraft-type green liquor is sulfited
hydrogen sulfide may be emitted. Emissions sources are blow tanks, spent
liquor evaporators, and the liquor burning or chemical recovery furnace.
Absorbers may be controlled by extra absorption stages. Blow gases can be
controlled by venting to the absorber. Recovery furnace processes control
sulfur dioxide absorption. Nonrecovery burning processes vent sulfur
dioxide to the atmosphere. No significant emission data were obtained from
the questionnaires.
SULFITE PROCESS
Sulfite cooking liquors contain sulfurous acid and the bisulfite of a base
such as calcium, sodium, magnesium, or ammonium. Recovery processes
can be used with all bases except calcium, for which scaling is excessive.
*Noncondensables incinerated in a lime kiln.
xxiii
-------
Sources of sulfur dioxide emissions are blow pits, digester relief vents,
absorbers, and recovery furnaces. Relief and blow gases can be controlled
by extra stages or scrubbers. Recovery furnaces are controlled by an
absorption or scrubbing system. Nonrecovery burning processes vent sulfur
dioxide to the atmosphere.
Twenty sulfite mills returned questionnaires. Fifteen of the 20 used some
control on absorbers, 15 of 16 controlled digester relief gases, and 4 of 18
controlled blow gases. No significant emission data were give.
STEAM AND POWER GENERATION
Pulp and paper mills generate steam in industrial type boilers.
Questionnaire information was provided by 288 mills, representing 66
percent of the total pulp and/or paper produced. Fuels consumed by these
mills were: coal — 26,000 tons/day, oil — 3,450,000 gallons/day, gas — 498
million cubic feet/day, and bark/wood waste — 24,300 tons/day.
Coal and oil used had mean sulfur contents of 1.9 and 1.8 percent, giving
a total sulfur dioxide emission rate of 1470 tons per day.
Particulate emission data were reported for 17 boilers fired with coal, ah
controlled by cyclone type collectors. Average emission rate was 0.39 grain
per standard cubic foot, or 18 pounds per ton of coal.
Particulate emission data were reported for 26 boilers fired with
bark/wood waste plus other fuels. All these boilers were controlled by
cyclone collectors. Average emission rate was 0.45 grain per standard cubic
foot. This is equivalent to 23 pounds per ton of bark/wood burned.
Data are also presented for the single electrostatic precipitator that has
been installed on a combination coal/bark fired boiler. Emissions from this
boiler were reported as 0.0052 grain per standard cubic foot.
XXIV
-------
ATMOSPHERIC EMISSIONS
FROM THE PULP AND PAPER
MANUFACTURING INDUSTRY
INTRODUCTION
BACKGROUND
Information in this report describes the nature and range of atmospheric
emissions from pulp and paper manufacturing operations during normal
operating conditions as determined from survey questionnaires completed
in the early phases of the study. In the interim, the Enviromental Protection
Agency (EPA), in cooperation with three mills, conducted a program of
stack sampling and analysis specifically for this study. Data from this study
are also included. The period from initiation of this study until preparation
of the report also represents a period during which strides were made by the
industry in a series of special studies to define the reduced sulfur emission
control capabilities of the kraft recovery furnace system, which includes the
kraft recovery furnace and the contact evaporator. This technology is
included as a portion of the report since it represents a significant element
in the performance of process operations and control devices and the
methods employed to limit and control the emissions from these sources.
Also included is information on emissions from miscellaneous sources such
as black liquor oxidation vents, brown stock washer systems, and lime
kilns. This information was generated in special studies reported in the
literature, the above referenced field sampling programs conducted by
EPA, and a special study conducted by the National Council of the Paper
Industry for Air and Stream Improvement (NCASI).'
SOURCES OF INFORMATION
Various sources of information about pulping operations and
atmospheric emissions were utilized in compiling this report. The principal
sources were:
1. Questionnaire surveys conducted in 1968.
2. Field investigations, including source sampling conducted in
', 1969, 1970, and 1971.
3. The literature.
-------
The emission data presented in this report were developed using
different sampling and analytical procedures. Those gaseous emission data
gathered in special studies and field investigations were collected using
analytical procedures considered to be much more sensitive and precise
than those used earlier, when these procedures had not been developed, to
collect data for the questionnaire surveys. Current definitions of particulate
matter and the procedures used for sampling may also be more sensitive
than those used for collection of the particulate data for the questionnaire
surveys. A summary of the sampling and analytical procedures used during
collection of the data is presented in Appendix B.
Questionnaire Surveys
Three questionnaire surveys were conducted:
1. Kraft pulp industry survey.
2. Acid sulfite and rionintegrated neutral sulfite semichemical
(NSSC) pulp industry survey.
3. Steam and power boiler survey.
The comprehensive kraft questionnaire form was sent to 116 mills.
Returned forms with usable information were received from 80 of these
mills. Very few of the mills were able to answer all of the questions on the
form. The data on each process or operation were tabulated and
summarized. These tables are found in the sections where the operations
are discussed and in Appendix A.
The sulfite questionnaire form was sent to 60 mills. Forms were received
from 34 of these mills. The amount of information obtained from this
survey was rather limited. The information is summarized and discussed in
the section on sulfite processes.
The power boiler survey was sent to 450 mills, and replies were received
from 288 of these mills. The information is summarized and discussed in
the section on power boilers.
Field Sampling Program
The objectives of the field sampling program were to obtain data to
verify the reasonableness of the emission data reported en the
questionnaires and to obtain emission data from well controlled mills.
A mobile source sampling laboratory was developed by the Emission
Measurement Branch of the Office of Air Quality Planning and
Standards. A continuous heated gas sampling line transfers a filtered
sample to a dynamic dilution system. This system provides dilution levels
PULP AND PAPER INDUSTRY EMISSIONS
-------
up to six orders of magnitude. The diluted samples are delivered to various
instruments. The instruments installed included two gas chromatographs,
employing a flame photometric detector; a total sulfur analyzer, also
employing a flame photometric detector; coulometric titrator; an oxygen
analyzer; a carbon dioxide analyzer; a carbon monoxide analyzer; and
particulate sampling equipment. Details of the equipment are given in
Appendix B.
Field sampling was conducted at three mills. Only limited data were
obtained at Mill A, as this was the initial operation of the equipment. More
extensive information was obtained at Mills B and C. These mills were
selected because they had good air pollution control programs and the
results were expected to typify the emission levels from well controlled
plants. The results are reported under the description of process emissions
and their control in the appropriate sections of the report. Limited odor
surveys were conducted at Mills B and C. Results were rather inconsistent
and are reported in Appendix C.
Literature
Several literature sources provided information for this report.
Published literature in text books2'3 and the final report, "Control of
Atmospheric Emissions in the Wood Pulping Industry,"4 of a study
conducted for EPA served as the basis for materials on process
descriptions. Information on emissions, particularly that referring to
definitions of control technology capabilities and miscellaneous source
emissions, was derived from several sources. These included technical
journals and papers presented at technical meetings but as yet
unpublished. NCASI Technical Bulletins, as well as data collected as part
of ongoing special study programs but not yet published in Technical
Bulletin form, also served as sources of information.
PULP AND PAPER MANUFACTURING INDUSTRY
The pulp and paper manufacturing industry is reported to be the ninth
largest manufacturing industry in the United States, accounting for nearly
4 percent of the value of all manufacturing. It consists of two well defined
segments, pulping and paper making. Pulping is the conversion of fibrous
raw materials such as wood, cotton, or used paper into a material suitable
for use in paper, paperboard, and building material. Wood is the dominant
source of fibers for paper production. Pulp is produced by two general
methods: mechanical pulp is produced by grinding or shredding the wood
to free the fibers; chemical pulp is produced by cooking wood chips in
chemical solutions that dissolve the lignin binding material. Since the air
pollution aspects of the chemical pulping processes are much more
significant than those of the mechanical processes, this report will only be
concerned with the chemical processes.
Introduction
-------
The kraft process is the leading chemical pulping process. A solution of
sodium hydroxide and sodium sulfide is used as cooking liquor. The spent
liquor is concentrated and burned in a furnace to recover the chemicals.
The sulfide process utilizes a cooking liquor made by absorbing sulfur
dioxide gas in a solution or slurry of one of the following base chemicals:
sodium carbonate, magnesium hydroxide, ammonia, or calcium carbonate.
Chemical recovery may or may not be practiced. Semichemical pulps are
produced by digesting with reduced amounts of chemicals, followed by
mechanical treatment to complete the fiber separation. The soda process
utilizes a solution of sodium hydroxide for digestion.
Practically all of the wood pulp produced in the United States is
consumed in this country, with only approximately 7 percent exported.5
Imported wood pulp is equivalent to approximately 10 percent of the pulp
manufactured in this country. Wood pulp accounts for approximately 80
percent of the raw fiberous material for paper and board manufacturing in
the United States, the remaining 20 percent consisting generally of recycled
paper. In addition, significant quantities of filler and coating materials are
employed to achieve desired product properties. About 90 percent of the
wood pulp produced is used on site for the manufacture of paper,
paperboard, and building paper products.
The nature of air quality protection measures in the pulp and paper
manufacturing industry closely parallels the nature and growth of its
chemical pulping industry. The air quality protection problems of paper
and paperboard manufacturing, and mechanical pulping as well, are
minor, consisting almost entirely of those from the combustion of fuel used
to generate steam and power.
Current Production
The production of wood pulp in the United States in 1969 was 41 million
tons (air dry). Of this, 27.6 million tons of paper grade pulp and a small
amount (less than 1 million tons) of dissolving and alpha cellulose pulp were
made by the kraft process Paper grade pulp produced by the sulfite process
amounted to 2.3 million tons, with less than 1 million tons of alpha and
dissolving pulp made by this process. Semichemical pulp production was
3.4 million tons. Other grades of pulp produced (mechanical, defibrated,
exploded) amounted to 5.3 million tons. The capacity for wood pulp pro-
duction in 1969 was 45.6 million tons.6
There were reported to be 209 chemical pulp mills at the end of 1970.
One hundred and sixteen were kraft mills, 38 sulfite mills, 50 Semichemical
pulp mills, and 5 soda mills. These data are summarized in Table 1.
Industry Growth Trends
The per capita consumption of paper and paperboard in the United
States is a relative indication of growth in the U. S. wood pulping industry
since the amount of imported wood pulp used in these products through the
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table 1. SUMMARY OF U.S. CHEMICAL
PULP MILL PRODUCTION, 1969
Process
Kraft
Sulfite
Semichemical
Soda
Total
Number of
mills
116
38
50
5
209
Annual production,
million tons
29.6
3.0
3.6
0.2
36.4
years has been small. The consumption of paper and paperboard has shown
an annual growth rate of about 3 percent. In 1970, the pulp and
paperboard consumption was 56.8 million tons, or 556 pounds per person.
It has been estimated that chemical pulp production will double between
1970 and 1985, rising by about 35 million to 70 million tons annually.4 As
shown earlier, over 80 percent of the current annual chemical pulp
production is produced by the kraft process. This or a more dominant
position for the kraft chemical pulping process is projected through 1975.4
While forecasts are subject to the hazards of a changing economy, there is
no reason to suspect that chemical pulping will not experience a continuing
favorable growth rate. Neither is there reason to suspect a major
rearrangement in the position of processes used for the manufacture of
chemical pulp during the previously referenced 15-year projection. Radical
changes in technology, however, although not now expected, could result in
such a shift by the end of the 15-year period.
Introduction
-------
-------
KRAFT (SULFATE) PULPING PROCESS
INTRODUCTION
Process Description
The kraft pulping process (Figure 1), which came into being in 1879, was
a modification of the caustic soda system in that sodium sulfide was added
to the cooking liquor. The introduction of the spray type recovery furnace in
the period 1928 to 1934 brought about a tremendous increase in the use of
krat't pulp; recovery of cooking chemicals from kraft spent liquor is essen-
tial for the kraft process to be competitive in cost with other processes. The
recovery of chemicals is accomplished by spraying concentrated spent
liquor (black liquor) into the recovery furnace, where the organic com-
pounds are burned and an inorganic smelt of sodium sulfide and sodium
carbonate is formed. To make up for chemicals lost in the operating cycle,
salt cake (sodium sulfate) is usually added to the concentrated spent liquor
before it is sprayed into the furnace.
The smelt of sodium sulfide and sodium carbonate flows from the
furnace and is dissolved in water to form green liquor. This solution is
reacted with quicklime to convert the green liquor to cooking liquor (white
liquor), which is a solution of sodium hydroxide and sodium sulfide. The
calcium carbonate created by this reaction is settled out, dewatered, and
burned in a lime kiln. The resultant calcium oxide is returned tor reaction
with the green liquor to close the chemical recovery cycle.
Raw Materials and Process Characteristics
The presence of caustic soda in the cooking liquor permits the pulping of
practically all wood species. The other active chemical, sodium sulfide, has
a buffering action that allows digestion to take place at a lower (OH)~ion
concentration, thus reducing damage to the fibers and producing pulps
that are stronger than those made from NSSC or sulfite processes. Small
amounts of sodium sulfide react with lignin and carbohydrates in the wood
to form odorous compounds that may cause a reduction of air quality.
Cooking chemicals (caustic soda and sodium sulfide) are expensive
relative to chemicals used in some other pulping processes. Thus their
recovery is an economic necessity. During the recovery process, steam is
produced from the combustion of the organic materials, adding to the
economic benefits of the recovery system.
-------
WOOD CHIPS
EVAPORATOR
GASES
STEAM
Figirel. Kraft pulping process.
The kraft process produces a dark-colored pulp that normally represents
from 45 to 50 percent of the initial weight of the wood used. Because of its
dark color, the unbleached pulp is usually used only in board, wrapping,
and bag papers. For use in the manufacture of white papers, the pulp must
be treated further in a bleach plant.
PULP AND PAPER INDUSTRY EMISSIONS
-------
Emission Sources
Gaseous Emissions
The kraft mill odor problem arises from the use of sodium sulfide as one
component of the disgesting liquor. In the digesters, the sulfide ion from
the sodium sulfide combines with various organic side-chain radicals from
the cellulose and the lignin of wood chips to form such organic sulfides as
methyl mercaptan, CH3SH; dimethyl sulfide, (CHj^S; dimethyl disulfide,
(CH3)2S2; and small amounts of similar ethyl sulfide compounds.7 In
addition, hydrogen sulfide is formed in considerable amounts. These gases
are released with the digester relief and blow gases, as well as other sources.
These sulfides are extremely odorous, being detectable at concentrations
as low as 1 part per billion (ppb). Table 2 shows odor thresholds for some of
the compounds mentioned above.8-9
Table 2. ODOR THRESHOLDS OF SOME
MALODOROUS SULFUR COMPOUNDS
Compound
H2S
CH3SH
(CH3)2S
(CH3)2S2
so2
Odor threshold, ppm
0.00473
0.002 1a
0.00 10a
0.0056b
0.47a
aOdor threshold defined as the concentration at
which all panel members detect odor.
DOdor threshold defined as the median concentra-
tion detected by the individual panel members.
The residual sodium sulfide and other sulfur compounds in the spent
cooking liquor (black liquor) can be the source of additional emissions. In
the multiple effect evaporators, the sodium sulfide reacts with dissolved
lignin to produce additional amounts of the gases mentioned above. These
gases are released from the noncondensables vents of the evaporator
condenser. Other unit processes that handle black liquor in a manner
permitting its contact with ventilation air, such as brown stock washing
systems and black liquor oxidation systems, are also sources of reduced
sulfur emissions. Scrubbers designed to control particulate emissions
sometimes use process water containing residual sulfur compounds and are
Kraft (Sulfate) Pulping
-------
a source of gaseous sulfur compounds. The kraft recovery furnace system,
which in most cases includes a direct contact evaporator, is the largest
potential source of reduced sulfur emissions. Sulfur dioxide is also a
potential emission. Smelt tanks can also be a source of reduced sulfur
compounds. In the direct contact evaporator, carbon dioxide in the flue gas
from the recovery furnace reacts with sodium sulfide in the black liquor to
produce hydrogen sulfide. In the lime kiln, carbon dioxide in the
combustion gases can react with sodium sulfide remaining in the wet lime
mud after incomplete washing to produce hydrogen sulfide.
Particulate Emissions
Particulate emissions occur primarily from the recovery furnace, the
lime kiln, and the smelt dissolving tank. They are caused mainly by the
carry-over of solids plus the sublimation and condensation of inorganic
chemicals. The sublimation and condensation produce a fume that initially
is probably submicron in size but has a tendency to agglomerate. In
addition, particulate emissions occur from power boilers and boilers fired
with bark in combination with other fuels.
Particulate emissions from the recovery furnace consist primarily of
sodium sulfate and sodium carbonate. These emissions may be carried up
by the furnace draft or formed by the vaporization-condensation step. The
high flue gas velocity ma)' cause the carry-up of small droplets of black
liquor that have been sprayed into the furnace. These droplets should burn
in the oxidizing zone, but :he resulting fine particles may be carried out of
the furnace.
Particulate emissions from the lime kiln consist principally of sodium
salts, calcium carbonate, and calcium oxide. The sodium salts result
primarily from the sublimation-condensation of salts that are retained in
the sludge because of incomplete washing. Calcium carbonate and calcium
oxide emissions result from entrainment.
Particulate emissions from the smelt dissolving tank are primarily
caused by the entrainment of particles in the vent gases. Because of the
violent reactions taking place in each of these tanks, it is reasonable to
expect that the turbulence of the dissolving water will splash droplets
containing both dissolved and undissolved inorganic salts above the
surface. These droplets may be carried out by the vent gases if they are not
of sufficient weight to drop back into the liquid.
Emissions Control Systems - General
Gaseous Emissions
Three general principles are utilized in controlling gaseous emissions
from kraft mill operations: effluent treatment, process control, and process
change. Combustion, absorption, and liquid phase oxidation are examples
of effluent treatment. Process control involves the manipulation of process
10 PULP AND PAPER INDUSTRY EMISSIONS
-------
variables to minimize emissions. Process change includes the alternation,
elimination, or addition of unit process equipment or operations.
Combustion involves the thermal oxidation of reduced sulfur
compounds in pulp mill noncondensable gases to sulfur dioxide. For
example, noncondensable gases from the multiple effect evaporators and
digesters can be vented to the inlet of the combustion air fan of the lime
kiln. In some cases, a separate incinerator is used if the distance from the
source to the lime kiln makes installation costs prohibitive. Sulfur dioxide is
considered less objectionable than the compounds from which it was
formed. In the lime kiln, most of the sulfur dioxide reacts with lime in the
kiln or is absorbed in the scrubber that controls particulate emissions from
the kiln. The volume of the lime kiln combustion air limits the volume of
emissions that can be handled in this manner. There is also a limited
practice of combustion of gases from digestion and evaporation as well as
those from brown stock washer system vents in the recovery furnace.
Absorption usually involves scrubbing the gas stream with an alkaline
process liquor, such as sodium hydroxide, lime mud weak wash, or white
liquor. Absorption is limited to gases containing little or no carbon dioxide,
such as evaporator or digester noncondensable gases and those from smelt
tanks. Effectiveness is limited largely to hydrogen sultide and methyl
mercaptan removal. Packed towers and sprayed mist pads are the types of
scrubbers usually used for absorption.4
Liquid phase oxidation is used to convert reduced sulfur compounds to
less odorous or more stable substances. Oxidizing agents used are chlorine,
atmospheric oxygen, and molecular oxygen.
Chlorination can be used on streams containing sulfur gases such as
those from digester relief and blow condensers and multiple effect
evaporator vents. A portion of the required chlorine is frequently available
in the chlorination stage washer effluent from the bleach plant, if present.
The dimethyl sulfide is absorbed and oxidized to sulfone. The dimethyl
disulfide is absorbed and oxidized to methyl sulfonyl chloride. This
technique is of limited effectiveness.
Black liquor oxidation is accomplished by the use of atmospheric
oxygen, or occasionally tonnage oxygen. This operation oxidizes the sodium
sulfide in the liquor to sodium thiosulfate. The purpose is to prevent the
formation of hydrogen sulfide by carbon dioxide and sulfur dioxide in the
recovery furnace flue gases.
Process control as a means of minimizing reduced sulfur emissions is
applicable to both the recovery furnace and lime kiln. Proper manipulation
of process feed rate and air supply both result in minimizing emissions from
these sources. The thoroughness of lime mud washing, which is reflected in
the amount of residual sulfide, also affects ^missions from the lime kiln.
Kraft (Sulfate) Pulping 11
-------
Process and equipment change — An example of a process and
equipment change is black liquor oxidation. This process, described earlier,
is designed to minimize emissions from the flue gas direct contact
evaporator by stabilizing sulfur to prevent its loss from black liquor
brought in contact with flue gas.
Another example of an equipment and process change designed to
minimize reduced sulfur emissions is the elimination of the flue gas direct
contact evaporator. This is accomplished in one of two systems. Both
extend the heat recovery systems commonly employed on recovery furnaces.
One uses heated air from an indirect heat recovery unit for direct contact
concentration of black liquor. This air is used as combustion air in the
furnace. The second utilizes additional steam produced from an added
economizer section to carry evaporation in a forced circulation concentra-
tion stage beyond that point normally accomplished in multiple effect
evaporation.
Particulate Emissions
In the kraft pulping process, particulate emissions are controlled by
electrostatic precipitators, scrubbers, cyclone collectors, and wire mesh
demister pads.4
The high-voltage electrostatic precipitator is the dominant type of
collector used to control recovery furnace particulate emissions. Most
precipitators are designed for collection efficiencies of 90 to 99+ percent. In
some instances, a low- or medium-energy scrubber is installed after the
precipitator. Such scrubbers can be effective because of their agglomer-
ating effect.
Scrubbers may be used for particulate emission control from several
sources. A venturi recovery system using black liquor as a scrubbing
medium serves as a primary particulate collection device as well as a flue
gas direct contact evaporator. Lime kiln particulate emissions are
exclusively controlled by scrubbers. Some smelt tank vent particulates are
controlled with scrubbers. Secondary scrubbers are installed behind the
primary collection devices, electrostatic precipitators, or venturi recovery
units on kraft recovery furnaces.
Cyclone collectors, and wire mesh demister pads and scrubbers are used
to control particulate emissions from smelt tanks. A liquid spray, usually
lime mud weak wash, is used, in most cases, with these devices.
SPECIFIC EMISSION SOURCES AND CONTROLS
Digester Relief and Blow
The batch digestion of wood chips takes place in large cylindrical mild-
steel vessels averaging about 4000 cubic feet (ft3) in volume. The chips are
cooked at temperatures ranging from 170 to 175 degrees Celsius (°C) and
12 PULP AND PAPER INDUSTRY EMISSIONS
-------
pressures ranging from 100 to 135 pounds per square inch gauge (psig). At
the end of the cooking cycle, the contents of the digester vessels are blown to
a tank at atmospheric pressure, flashing off large amounts of steam as well
as noncondensable gases. Digesters are also equipped with a
noncondensable relief-vent system for use during the cook. Most kraft
pulping is done in batch digesters; although increasing numbers of
continuous digesters are being employed in the industry.
Emissions
The objectionable odors from the digester are a result of the chemistry of
the digestion process. The active ingredients in the cooking liquor normally
consist of about three-fourths sodium hydroxide and one-fourth sodium
sulfide in water solution. Sodium sulfide provides sulfide ions that combine
with organic components of the wood to produce organic sulfides.
The organic sulfides formed in the digester include methyl mercaptan,
dimethyl sulfide, and dimethyl disulfide. Other sulfides in smaller amounts,
as well as terpenes, may also be present.
Control Techniques
As previously stated, the flashed material from the blow tank, as well as
the relief stream, consists of both steam and malodorous noncondensable
and condensable organic materials, including turpentine. The steam is
usually condensed to recover heat. Some of the odorous materials are also
condensed in this process. The noncondensable fraction containing hydro-
gen sulfide and organic sulfides can then be vented to a control device.
In most cases, relief gas is piped to a vapor-liquid separator such as a
cyclone, to recover entrained cooking liquor, before going to the condenser.
If turpentine recovery is economical, as it is in many southern mills that
cook softwood (pine) chips, the vapor from the cyclone is condensed and
sent to a decanter, which separates the liquid turpentine for storage, the
water being sewered. Noncondensables from the condenser are either
vented to atmosphere or treated.
The blow gases are condensed in a direct-contact condenser or in a
surface condenser. Noncondensable gases are vented from this condenser.
The condenser cooling water recovers the heat in the blow steam and may
be used in some other process such as the cooking liquor make-up system.
Methods used to dispose of noncondensable malodorous sulfur
compounds include burning, scrubbing with an absorbent, and chemical
oxidation. The methods most prevalent in the industry are burning and
scrubbing.
Burning of the gases is accomplished in the lime kiln or in a special
separate incinerator.10 In burning, care must be taken to avoid the
occurrence of explosive mixtures. In a batch digestion system, the problem
of preventing large surges of gas to the burning device arises. Gas surge
Kraft (Sulfate) Pulping 13
-------
capacity is provided by using either large spherical tanks equipped with a
movable nonporous diaphram11 or conventional gas holders. Burning can
be a very effective technique for disposal of these gases. Data in the section
on lime kilns illustrate this point.
Scrubbing the gas stream with a sodium hydroxide solution offers a
partial control method for digester emissions. Effectiveness is limited to
hydrogen sulfide and methyl mercaptan. Some mills use such scrubbers for
preliminary treatment of gases before burning them. Three objectives are
achieved: (1) some sulfur is recovered, (2) steam is condensed, and (3)
turpentine vapors and mists are removed, mitigating an explosion hazard.
Scrubbing with chlorine solutions is practiced in some mills. In the case
of bleached kraft mills, chlorine-containing effluent from the bleach plant
may be used to scrub the gases. It is necessary that residual chlorine be
present at all times in these effluents to maintain the effectiveness of this
technique, which is of limited effectiveness at best.
Emission Data
Questionnaire Data— Of 80 respondents to the questionnaires, 6 presented
data on the composition and amount of sulfur compounds in the relief and
blow gas streams from batch digesters. No treatment facilities were
indicated.
The data are presented in Table 3. The emission rates on a weight basis
were computed on the assumption that the average flow rate multiplied by
the duration and the number of cooks per day was a measure of the total
daily gaseous effluent volume. The validity of this assumption is not known.
For units 3 and 5, the average flow rate of blow gases is the rate ahead of the
condenser. The gas concentrations appear to be taken after the condenser.
Hence, the weight rates of sulfur gases computed for these mills is probably
incorrect. Disregarding these two values, the median value of total reduced
sulfur (TRS) from blow gases was derived as 0.40 pound per ton of air-dried
pulp (Ib/T ADP). The median value for TRS from relief gases is 1.03 Ib/T
ADP as hydrogen sulfide.
Two plants submitted sample data on the combined relief and blow gas
noncondensable streams from continuous digester systems prior to any
treatment. These data are summarized in Table 4.
Of 80 respondents to the questionnaires, 16 indicated they treated their
relief or blow gas noncoridensables in some way, and eight indicated they
did not. The remaining respondents did not indicate whether or not they
employed control techniques for these sources. Of those providing
description of their treatment method, six burned the gases in lime kilns,
five used chlorination stage bleachery effluent for treatment, one used a
catalytic afterburner, one scrubbed with black liquor, and one vented the
gases into the black liquor oxidation tank. Data adequate to evaluate the
operation were not given in any of these questionnaires.
14 PULP AND PAPER INDUSTRY EMISSIONS
-------
<
Z
(A
LLJ
g
00
a
z
UJ
CC
OC
UJ
k
LLJ
u
Q
^
O
oc
LL
V)
CO
D
O
UJ
CO
ro
CO
CM)
X
CA
«
CO
cc
I-
E
i
sulfur gases.
Reduced
"c\
81
2 c
3 0
a*:
w ||
5£S
> .A re
< re *•
Production
rate.
a.
Q
h-
13
re
J3
Q.
C.
CM
CO
~co
X
o
CM
CO
""«
X
0
CO
CO
X
O
CO
CM
X
E
a
a
re
•o
c
fc
u-
3
•6
re
ro
oT
Q
1-
'c
D
Source
CM
O
0
r-.
CM'
CD
cn
CO
cn
o
0
o
o
CM
o
CD
in
r*.
O
in
in
CM
CM
-
"CD
CC
CD
O
00
CM
O
1
1
s
1
1
in
o
in
ca
O
co
S
§
CM
CM
t/i
CD
CO
cn
CO
0
CM
§
1
'
CM
cn
CO
CM
°
0
cn
in
o
in
CO
00
CM
CM
in
CO
CO
co in
CO CD
cn CM
CD CM
o r-
O CO
o «-
CD" «-"
CM
O CD
CM
§o
in
cn"
CM "~
CM cn
•3- CM
00 |
in oo
•* in
o o
CD in
CO
O O
cn o
CD q-
m
0 0
CM •*
in r^
•a- in
§
o
o
1
CO
1
1
in
CO
CM
1
CM
O
CO
CM
CM
O
O
co"
in
cn
CD
CD
o
CD
«- cn
CO CD
O O
CO
c> r--
oo in
CO
in in
CO CO
in"
1 S
CO
' i
15 8
00 S
1 S
CM
1 §
CD
S $
CM CO
CO
o f2
CM T-
0 0
§co
CD
in" T-"
0 0
CD CM
CM in
CM •*
to
QJ
a
CO
o
CM
CM
0
in
in
o
CM
CM
CD
cn
CD
o
CM
o
CM
CO
in
o
CM
o
s
in"
in
CM
CM
-
o
o
r-
CO
0
8
CM
cn
CM
o
CM
CM
cn
CD
1
00
in
•2
CO
in
o
8
in"
0
in
in
o
cn
cn
§
co"
CO
CO
o
CO
o
CM
O
cn
in
CM
CO
CM
CM
O
8
CM"
in
in
oo
-
>erature.
Q>
V)
CD
Kraft (Sulfate) Pulping
15
-------
Table 4. SULFUR COMPOUND EMISSIONS FROM
CONTINUOUS DIGESTERS, SUMMARY
OF QUESTIONNAIRE DATA
Compound
H2S
so2
CH3SH
(CH3)2S
(CH3)2S2
Emission rate,
Ib/T ADP
0.0021 and 4. 17
—
0.13
0.303
1.04
Number
of mills
2
—
1
1
1
EPA Test Results—Two systems for treatment of noncondensable gaseous
emissions from digester relief and blow and from multiple effect evaporator
vents were sampled by the EPA test team.
Mill B operated both batch digesters and a continuous digester, with a
combined capacity of about 1150 T/day. Relief and blow gases are collected
in a vaporsphere similar to that described by Morrisonl'The gases from the
vaporsphere are scrubbed in a rock-filled packed tower utilizing weak wash
liquor. The scrubber removes some of the sulfur gases and most of the
turpentine vapors. The multiple effect evaporator gases are scrubbed in a
packed tower with white liquor. Tests by mill personnel indicate that the
scrubber removes essentially all the hydrogen sulfide and methyl
mercaptan, which represent about 96 percent of the sulfur compounds from
this stream. The outlet gases from the two scrubbers are combined and
vented into the combustion air fan of the lime kiln. This stream was
sampled by the EPA test team. The plant flow meter indicated an average
flow rate of 20 cubic feet per minute (cfm). Results are given in Table 5.
Mill C operated only a continuous digester. The condensate from the
blow heat recovery system is stripped of sulfur gases by aeration. This off-
gas, together with the noncondensable gases from the digester relief and the
multiple effect evaporator vents, is piped directly to the lime kiln for
burning. The total volume of this gas stream had been measured by plant
personnel at 380 standard cubic feet per minute (scfm) at a mill production
rate of 500 T ADP/day. The composition of the stream was sampled by the
EPA test team. Results are given in Table 5.
Tests run on the emissions from this lime kiln revealed only traces of
organic sulfur compounds, as noted in the section on results of EPA tests of
lime kilns later in the report. This indicates that burning these gases is a
very effective treatment method.
16
PULP AND PAPER INDUSTRY EMISSIONS
-------
TableB. COMPOSITION OF GAS STREAMS
VENTED TO LIME KILN, EPA TEST RESULTS
Reduced sulfur gases, ppm
Mill
B
C
H2S
291
so2
2500
CH3SH
21,100
6,000
(CH3)2S
11,500
6,500
(CH3)2S2
6,100
500
TRSasHzS
ppm
44,400
13,790
Ib/hr
4.8
28.1
Ib/T ADP
0.19
1.36
Multiple Effect Evaporators
The kraft process utilizes multiple effect evaporation to concentrate
weak black liquor (spent cooking liquor) washed from the pulp. Removal of
large amounts of water from the liquor is necessary to facilitate combustion
of the dissolved organic material in the recovery furnace. The liquor is
concentrated in the multiple effect evaporators from a solids content
between 12 and 18 percent to one between 40 and 55 percent.12
Most kraft mills utilize long-tube vertical shell-and-tube type
evaporators. The weak black liquor is fed to the tube side of the latter
evaporation effects and steam is supplied to the shell side of the first effect.
As shown in Figure 2, the liquor proceeds through the tube side of each
effect from last to first, being heated in each by condensation of the vapor
driven off the boiling liquor in the tube side of each preceding effect.
Evaporated water vapor from the last effect of the set is condensed in
one of two types of condensers: direct contact barometric condensers or a
surface condenser with steam ejectors. The condenser must remove vapor
fast enough to create a vacuum in the vapor space of the last effect. Each
type of condenser is equipped with a small steam ejector to remove
noncondensables.
Emissions
Emissions from the multiple effect evaporator system occur from the
condenser and the various steam ejectors. The shell side of each effect is
vented through a relief valve for noncondensables. Common practice is to
vent all noncondensables to the condenser. It is possible, however, to vent
noncondensables from those effects that are above atmospheric pressure
directly to the atmosphere.
Emission Data
Information in questionnaires and the number of units reporting the
composition of effluent gases from the evaporator hot well vent stack are
presented in Table 6.
The questionnaire data on which the above table is based may be found
in Table 7. No data were reported on the composition of condensate
Kraft (Sulfate) Pulping
17
-------
VAPOR
\
PROCESS
STEAM
/
LIQUID
RETURN
\
CONDENSATE
THICK BLACK LIQUOR
WEAK BLACK LIQUOR
Figure 2.
feed).
Multiple effect long-tube vertical evaporators (backward
streams. The large variations in the data result because it is difficult to
measure the volume of this stream and because analytical techniques were
not standarized. While this stream is referred to as "noncondensables," it
contains a high and variable percentage of water vapor. Accurate
measurement of the volume flow rate is very difficult.
Control Techniques
Emissions from evaporation arise from noncondensable vent gases and
liquid condensate. Liquid condensate is usually sewered, and
noncondensables are often vented to the atmosphere.
18
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table 6. EMISSIONS FROM MULTIPLE EFFECT
EVAPORATORS, SUMMARY OF QUESTIONNAIRE DATA
Compound
H2S
CH3SH
(CH3)2S
(CH3)2S2
Emissions
Concentration, ppm
Range
0 to 44,000
5 to 211
10 to 196
10 to 1200
Median
1055
59
22
50
Rate, Ib/T ADP as H2S
Range
0 to 5.9
0.002 to 0.1 16
0.0002 to 0.095
0.0003 to 1 .23
Median
0.29
0.011
0.012
0.033
I
Number
of
mills
10
6
6
6
Noncondensable vent gases may be combined and controlled with
digester blow and relief gases, using such methods as incineration or
alkaline scrubbing. (See digester control techniques section earlier in
report.) Two mills reported utilization of control techniques for vent gas
emissions other than burning. Both of these mills used packed scrubbers.
Table 8 summarizes data from these mills. High removal efficiency on
sulfur dioxide and hydrogen sulfide are shown, but effectiveness on removal
of organic sulfur gases is not reported.
Pilot plant investigation of the feasibility of stripping malodorous gases
from kraft mill condensate streams (including evaporator condensate) was
conducted in 1958 by a major pulp and paper company.'3 Steam stripping
in an eight-stage bubble-cap type fractionating column accomplished sig-
nificant separation of these gases from condensate effluent streams. Pre-
heated condensate feed was passed downward through the column counter-
current, to steam injected at the bottom. The resulting overhead streams
consisted of noncondensable gases (which could be disposed of by the
previously mentioned methods) along with relatively pure water. Removal
of 95 percent or more of the hydrogen sulfide, mercaptans, and dimethyl
disulfide was accomplished.
Malodorous gases may also be removed from condensate streams by air
stripping. The condensate is aerated in a closed, agitated tank, and off-
gases are piped to a lime kiln. Tests on an installation of this type treating
condensates from digestion and evaporation showed 75 percent removal of
dimethyl sulfide and 85 percent removal of methyl mercaptan. 14
Kraft Recovery Furnace Systems
Introduction
The first two kraft recovery furnace systems shown in Figure 3 consist ot
two to four separate processes, depending upon whether primary and/or
Kraft (Sulfate) Pulping
19
-------
O
LU
DC
2
Z
O
D
a
03
CC
CC
o
a.
>
LU
t
O
CC
U-
(/}
2
O
V)
(A
UJ
CO
3
O
UJ
CO
<
C3
0)
2
CD
CA
(0
CO
CC
1-
I
of
ai
|
3
v>
1
1
c\
o
CO
s i s
Oc 2
c +^
o> to
Q> g."
fl) *^
>
O
o a)*
•§ s
o
Q-
0.
Q
<
1-
£
>
(0
""Q
™"
E
Q
a
CM
co
~co
I
o
CO
~ro
X
o
"co
X
o
co
CM
X
E
Q.
a.
a.
o
E
y
* >* «3- f
^f oo «~ co o o
1^ ^ CO ^ CO CO
I- CO «- CO CO O
t- CM co "S- in co
O) CO CM
t- in CM o
d ^ d d
CM TT
§252"'
T- CM
G' ^) O) ^^
8 " 2 *~
CM
1 2 1 1
CM
' S i '
| in co r^
§00 co o
CM
O i-
CM"
1 1
§000
CO 00 n^
LO C2 00 00
^f ^5 ^^ CO
O) o r** r*-
^
o in o o
CO t O O
o «— r^* r^
-Q 03 -Q
<0 P** CO 00
§ "5
c t
° £
o o
„ (J
cu
"o3 !>
m <
ro ^
20
PULP AND PAPER INDUSTRY EMISSIONS
-------
TableS. CONTROL TECHNIQUES FOR
MULTIPLE EFFECT EVAPORATOR EMISSIONS,
SUMMARY OF QUESTIONNAIRE DATA
Mill
34
60
Equipment
Packed scrubber
(fluid, weak white
liquor at 36gpm)
Packed scrubber
(fluid, weak black
liquor at 35 gpm)
Gas flow
rate,
dscfm
110
496
Concentration, ppm
Inlet
H2S= 44,000
S02 = 24
H2S= 126
CH3SH = 73
Outlet
H2S = 2000
so2 = o
H2S=0
CH3SH= 17
secondary control devices are employed. These are (1) the kraft recovery
furnace, (2) the flue gas direct contact evaporator, (3) the primary
particulate emission control device, and (4) the secondary particulate
emission control device. In some situations, the flue gas direct contact
evaporator has served the dual purpose of a black liquor evaporator and
particulate emission control device. In recent years, it has also been
eliminated or modified in its manner of use in a limited number of new
installations, as illustrated in the last three diagrams in Figure 3, being
replaced with extended multiple effect evaporation or operated with hot air
rather than flue eas as a source of energy for evaporation. The emissions
from the , kraft recovery furnace system therefore always consist of those
from the kraft recovery furnace, as well as those from the flue gas direct
contact evaporator, when one is used, since gas flows from the furnace
through the evaporator prior to its discharge. Under certain conditions,
some constituents of the recovery furnace gases are absorbed in the
alkaline black liquor in the direct contact evaporator.15
Kraft Recovery Furnace — The primary function of the kraft recovery
furnace is recovery of chemicals from black liquor, although steam is
produced from the heat of combustion of organic residue in the liquor.
Concentrated black liquor is sprayed into the lower part of the furnace,
which is designed for operation in a reducing atmosphere near the bottom
and an oxidizing atmosphere in the remainder. Essentially all of the
recovered chemicals are removed from the bottom of the furnace as a
molten smelt consisting principally of sodium sulfide and sodium
carbonate. Particulate matter, normally consisting principally of sodium
sulfate with some sodium carbonate present, is carried from the reducing
zone, as are gaseous sulfur compounds.
Flue Gas Direct Contact Evaporators —Until recently, concentrated black
liquor from the multiple effect evaporators in the kraft recovery process was
almost always further concentrated in a contact evaporator prior to its
combustion by bringing recovery furnace flue gas in contact with black
Kraft (Sulfate) Pulping
21
-------
A. CONVENTIONAL FLUE GAS DIRECT CONTACT SYSTEM
RECOVERY (J-FLUE
FURNACE \ DIRECT C
-? •« 1 EVAPOR
B VE-NTURI RECOVERY UNIT SYSTEM
/7\
^^^^^^^^^^^
RECOVERY
FURNACE !v BLACK LIQUOR
-?* -*-
X ^ PARTIHIIATh ^
GAS*-1! CONTROL
DNTACT BLACK DEVICE
ATORTTn7inT
1 J '
~*7\ VENTURI )\
1 [ \_ RECOVERY [ \
1 1 UNIT 1
^» .in— ^^—
I _I
C. B & W HIGH SOLIDS SYSTEM WITH MO DIP
ADDED
y X---, ECONOMIZER
/ f^\i- 1 l
_BLACK I
"'~: ~LIOUO~R
ECT CONTACT EVAPORATOR
~ PRIMARY
PARTICULATE
CONTROL
DEVICE
CONTROL
DEVICE
iQPTIONALi
SECONDARY
PARTICULATE
CONTROL
DEVICE
(OPTIONAL!
SECONDARY
PARTICULATE
CONTROL
DEVICE
(OPTIONAL)
FORCED j
C IRCULATION U^
EVAPORATOR
D. CE SYSTEM WITH NO FLUE GAS DIRECT CONTACT EVAPORATOR
^~ ~^.
/ \ j
-«MX~-?°X AIF
x 1 CONT
,-ir*- ,-*^ EVAPOF
1PT "~| <^OH -^O-L/^ "\
n\j \ I ^-* /^-jA_^v^AIO
ATOR \. G*^°-AIR
l~~ t*^-s\
PARTICULATE
CONTROL
DEVICES
! J BLACK LIQUOR j
I
E. CE SYSTEM WITH NO DIRECT CONTACT
/ \
/ ^\ fFp^
vi— di
-o^>- o* :
^| ^ BLACK LIQUOR i
"^ I
EVAPORATOR
\
I
*-OHI "fc-CT^N
~*~i ^—A
PARTICULATE
CONTROL
DEVICE
FORCED 1
CIRCULATION *^m
EVAPORATOR ^^
Figure 3. Typical kraft recovery furnace system options.
22
PULP AND PAPER INDUSTRY EMISSIONS
-------
liquor. These evaporators are of three forms. Cascade evaporators provide
for the contact of black liquor with recovery furnace flue gas through the
use of rotating wheels, the bottom portions of which move through a vat of
black liquor that is carried into a moving stream of hot combustion gas
from the recovery furnace. In cyclone evaporators, black liquor is sprayed
into the hot gas stream and then separated from the gas stream by the use
of a cyclone. The third type of contact evaporator used is commonly called
the venturi recovery unit. In these units, black liquor is introduced into a
stream of recovery furnace flue gas at a venturi throat and then separated in
a centrifugal separator. Venturi recovery units differ from other types of
contact evaporators so far as emissions are concerned principally in that
particulate removal efficiencies are higher. Cascade and cyclone
evaporators are commonly believed to remove from 40 to 50 percent of the
particulate matter leaving the recovery furnace, while venturi recovery units
have been designed for greater than 90 percent capture of particulate
matter. While capturing particulate matter and sulfur dioxide present in
recovery furnace exhaust gas, the contact evaporator is a potential source of
reduced sulfur compounds, the amount depending on the residual sulfide
in the black liquor fed to the evaporator.
Recovery Systems without Flue Gas Direct Contact Evaporators — Both
manufacturers of kraft recovery furnaces in the United States have
participated in recent years in the development of innovated designs for
kraft recovery furnace systems that eliminate the use of a flue gas direct
contact evaporator. Both recover the bulk of the heat formerly used for
evaporation of black liquor in the flue gas direct contact evaporator. The
Babcock and Wilcox (B&W)* system shown in Figure 3C uses an extended
economizer section for this heat recovery. The Combustion Engineering
Company (CE) design shown in Figure 3D uses a flue gas-to-air heat
exchanger to heat ambient air, which is subsequently used to evaporate
black liquor in a conventional contact evaporator. In the CE design, the
exhaust from the contact evaporator serves as a portion of the combustion
air for the recovery furnace. A modification of this system (Figure 3E)
eliminates the contact evaporator but retains the heat exchanger to preheat
the combustion air. Additional indirect evaporation of the black liquor is
also used in this system.
While both these designs eliminate the flue gas direct contact evaporator
as a potential source of reduced sulfur emissions, the particulate load on
subsequent particulate emission control devices is almost doubled. Sulfur
dioxide formed in the recovery furnace by the burning of black liquor is
maintained at reasonable levels by the contact evaporator. Its control in
those systems with no flue gas direct contact evaporator is dependent on
currently ill-defined furnace and other process operating variables.
*Mciition of commercial products or company names does not imply endorsement by EPA or
NCASI.
Kraft (Sulfate) Pulping 23
-------
Composition and Control of Emissions from Kraft Recovery Furnaces
Both gaseous sulfur compounds and particulate matter are generated in
and emitted from kraft recovery furnaces. In essentially all cases, the
gaseous sulfur compounds leaving the kraft recovery furnace contain sulfur
dioxide and varying amounts of hydrogen sulfide. Organic sulfur
compounds consisting of methyl mercaptan, dimethyl sulfide, and dimethyl
disulfide may be present, although their presence is usually contingent on
the manner in which the recovery furnace is operated and is not common.
When present, these gases seldom represent more than a small percentage,
less than 10 percent, of the hydrogen sulfide present. The particulate
matter emitted from the recovery furnace consists of sodium sulfate and
sodium carbonate and may contain small amounts of sodium chloride. The
presence of the latter depends on whether the wood used for pulping has
been stored in saline water and on the chloride levels in make-up chemicals
used in process.
Particulate Emissions -The particulate matter levels in kraft recovery
furnace flue gas before it reaches a control device normally range from 8 to
12 grains per standard cubic foot (gr/scf) or 200 to 450 Ib/T ADP. The
actual emission level is a function of control device efficiency, which is a
function of the system design. In conventional recovery furnace systems,
those with flue gas direct contact evaporators, recovery furnace particulate
emission control consists of (1) the contact evaporator, (2) a primary control
device, which is either a precipitator or a venturi recovery unit, and (3)
possibly a secondary scrubber. Where no flue gas direct contact evaporator
is used, particulate emission control in current designs is by precipitators
alone, although nothing precludes the use of secondary scrubbers. Emission
ranges for existing conventional draft recovery furnace systems are
described in a following section. Currently, high-efficiency electrostatic
precipitators with design collection efficiencies of greater than 99 percent
are commonplace in new and replacement installations.
Gaseous Emissions—The kraft recovery furnace is one of the two largest
potential emission sources of reduced sulfur compounds in the kraft
recovery furnace system, the second being the contact evaporator. Control
of reduced sulfur emissions from the kraft recovery furnace depends on
operational control of the furnace. Depending on the mode of furnace
operation, reduced sulfur concentration in the furnace exhaust gas may be
as low as 1 ppm or as high as several hundred ppm. This range of possible
reduced sulfur emission levels is independent of the presence or absence of
a flue gas direct contact evaporator. Therefore, new recovery furnace
system designs that eliminate the flue gas direct contact evaporator only
eliminate one potential source of reduced sulfur emission. Those
operational factors that account for control of reduced sulfur emissions
from the recovery furnace apply regardless of furnace age.
Historically, there has been little concern for the sulfur dioxide
generated in a kraft recovery furnace. In conventional kraft recovery
24 PULP AND PAPER INDUSTRY EMISSIONS
-------
furnace system designs, concentrations are substantially reduced when
furnace exhaust gas passes through the contact evaporator. After the
contact evaporator, they are characteristically between 50 and 150 ppm,
well below those resulting from the combustion of fossil fuels containing 0.5
percent sulfur, i.e. less than 300 ppm. Concentrations of sulfur dioxide in
recovery furnace exhaust gas range from less than 50 to as high as 700 or
800 ppm. The factors responsible for this range of concentration are not
well identified. Blue and Llewellyn, 'Sand results of currently unpublished
studies as well, indicate that sulfur dioxide generation in kraft recovery fur-
naces is a function of several variables. One is cooking liquor sulfidity, an
indirect measure of the soda and sulfur ratio in black liquor fed to the fur-
nace. Others include smelt bed depth, manner in which liquor is sprayed
into the furnace, ratio of primary to secondary combustion air, and possibly
temperature within the furnace itself. Blue and Llewellyn show sulfur
dioxide emission levels of about 50 to 100 ppm from a kraft recovery fur-
nace during performance tests. Actual operating experience with several
other furnaces has failed to define the reasons for sulfur dioxide emissions
of 5 to 10 times this level. As stated previously, the higher concentrations
are of limited practical concern except in those recovery furnace system
designs that eliminate the contact evaporator.
Several operating and design variables that have some effect on, or
relationship to, the control of reduced sulfur compounds emitted from the
kraft recovery furnace have been identified. Among these have been the
quantity and manner of introduction of combustion air, rate of solids
(concentrated black liquor) feed, turbulence in the oxidation zone, oxygen
content of the flue gas, spray pattern and droplet size of the liquor fed the
furnace, and smelt bed disturbance.15-17 The impact of these variables is
independent of the absence or presence of a contact evaporator.
The presence of adequate oxygen throughout the oxidation zone of the
recovery furnace and its thorough mixing with the products of combustion
are major factors in assuring that reduced sulfur compounds lost from the
smelting (reducing zone) in the bottom of the furnace are oxidized to less
odorous forms of sulfur such as sulfur dioxide. The amount of combustion
air and its distribution between the various points for its admission, i.e.
primary, secondary, and tertiary combustion zones of the furnace, have
been found to be factors in assuring that satisfactory conditions of
combustion exist for a minimum emission of reduced sulfur. Combustion
conditions may also be enhanced by adjusting engineering variables, such
as inlet air velocity, to improve the turbulence and mixing in the oxidation
zone. Some designs also provide for introduction of combustion air
tangential to the walls of the furnace to enhance mixing and promote
complete combustion.15'17
A partial measure of an adequate supply of air to support combustion in
the recovery furnace and hence contribute to the factors accounting for a
minimum reduced sulfur emission is residual oxygen content of the flue
gas. Special studies on 26 recovery furnaces showed that some residual
Kraft (Sulfate) Pulping 25
-------
oxygen content was necessary to achieve minimum reduced sulfur emission
levels. However, a residual oxygen content in the flue gas was not a
guarantee of minimum reduced sulfur emission rate. The point of
introduction of combustion air, degree of mixing, and leaks are all probably
responsible for a range of residual oxygen contents, all of which were
associated with minimum reduced sulfur emissions in different furnaces. 15
These and other studies have demonstrated that minimum reduced sulfur
emissions are not commonly observed, however, unless residual oxygen
content of the flue gas is in the range of 2.5 to 4.5 percent.
Physical disturbance in the smelting zone, created either by excessive
impingement of combustion air on the surface of a high smelt bed or smelt
sloughing from the walls, has been identified as a factor that may increase
emission levels of reduced sulfur compounds.17 The size of liquor droplets
sprayed into the furnace is also considered to be of some significance,
reduced sulfur emissions increasing as droplet size decreases. Some control
of smelt bed height is possible in operation of the furnace. Smelt sloughing
from the walls, however, is not controllable with current design. Spray
droplet size is partially controlled by altering viscosity through temperature
control of the black liquor feed.
One of the most significant, but not the sole, factor in control of reduced
sulfur emissions from the kraft recovery furnace is the rate of concentrated
black liquor feed. Detailed investigations of recovery furnaces indicate that
an optimum liquor firing rate can be defined for each furnace. The
optimum rate, coupled with proper control of previously described
operating variables, will result in a minimum reduced sulfur emission rate.
The investigations that lead to this conclusion involved tests, lasting from a
few to several hundred hours, of more than 20 recovery furnaces. 15,17
An example of the interrelation of liquor feed rate, TRS emission level,
and steam generation efficiency for one furnace is shown in Figure 4.15 The
geometric increase in reduced sulfur emissions once a certain liquor feed
rate (34,000 Ib/hr in this case) is exceeded is typical of furnaces fired above
the critical level for minimum TRS emission. This minimum emission rate
for individual furnaces was also observed to bear a relationship to the ratio
of air to solids fired (Ib/lb). This is illustrated in Table 9,15 where reduced
sulfur emissions are shown to increase substantially when the air-to-solids
ratio fell below 4.25. Like oxygen content, this ratio covered a range for
different furnaces, usually falling between 3.5 and 4.5.
No well defined relationship between solids firing rate at minimal
reduced sulfur emissions rate and rated furnace capacity was found in these
studies. In only one case did maximum firing rate for a minimum reduced
sulfur emission level and manufacturer's rated solids firing capacity
coincide. Commonly, minimal emissions rate occurred at 1.15 times rated
capacity, on one occasion, at 1.4 times rated capacity, and never below
rated capacity.15 Several factors can account for these differentials,
including divergence in heat value of that liquor actually burned from
26 PULP AND PAPER INDUSTRY EMISSIONS
-------
liquor characteristics used for design purposes. Probably most important,
however, has been a changing definition of rated capacity throughout the
years and the fact that it was not originally conceived with the idea that it
was related to emissions.15
The above referenced studies showed that there are a series of primary
operating variables for recovery furnaces whose control is required for a
minimum emission of TRS. Among these factors are: (1) liquor firing rate,
(2) available oxygen for combustion, (3) air-to-solids ratio, and (4) probably
the ratio of primary to secondary and tertiary air.
Under the most favorable control of primary operating variables, there
have also been observed a series of secondary operating variables, such as
(1) smelt falling from the walls, (2) a smelt bed of sufficient height to
prohibit good mixing of the products of pyrolysis with air, and (3) plugged
ash hoppers and bridged liquor feed guns, which can superimpose a
condition resulting in a temporary increase of emissions that are variable,
of limited magnitude, and usually of short duration. There has been no
evidence that sulfide content of the liquor being burned bears any
relationship to the reduced sulfur emissions from a recovery furnace. This is
not to be confused with sulfur compounds generated or stripped in the flue
gas contact evaporator.
25
30 35 40
DRY SOLIDS FIRED, Ifl3 Ib/hr
3.5
3.0 =
2.5 i
2.0
Figure 4. Effect of solids firing rate on reduced sulfur emissions
and steam generation efficiency.
Kraft (Sulfate) Pulping
27
-------
Table 9. EFFECT OF FURNACE FIRING RATE
AND AIR SUPPLY ON TRS EMISSION
FOR KRAFT RECOVERY FURNACE
Dry solid
firing rate,
Ib/hr
20,000
20,000
19,300
24,200
24,000
24,400
24,200
23,100
23,500
25,200
30,000
Air, Ib/lb
dry solids
5.40
5.40
5.25
4.50
4.50
4.42
4.40
4.35
4.25
4.16
3.33
Oxygen,
percent
6.2
6.4
5.8
5.0
4.8
5.3
4.8
4.2
4.7
2.8
1.6
Average TRS
as HUS, ppm
1.4
1.0
1.7
1.5
1.4
1.3
1.5
1.9
2.6
52.0
560.0
Ability to maintain maximum control of the above operating variables is
somewhat dependent on age and furnace appurtenances. Application of
best available operation control technology for new recovery furnaces may
therefore result in different emission levels than those from older furnaces.
Emission levels observed during extended study periods characterize the
reduced sulfur emissions control possible on recovery furnaces built during
the period 1955 to 1968, which usually have more refined combustion and
firing controls. As illustrated in Figure 5, these furnaces showed total
reduced sulfur emissions below 1 ppm 65 percent of the time, below 2 ppm
80 percent of the time, below 4 ppm 90 percent of the time, and below 16
ppm 99 percent of the time.15 Llewellyn reports similar reduced sulfur
emissions from a new recovery furnace using a Babcock and Wilcox high
solids system; daily average TRS emissions ranged from 0.5 to 8.8 ppm with
a median of 2 ppm, which represented 0.064 Ib/T ADP. The mean of
monthly mean TRS emission rates for this furnace over a 1-year period was
0.022 Ib/T ADP, and monthly averages ranged from 0.011 to 0.098 Ib/T
ADP. These values are contrasted with those from performance tests, which
are usually of short duration, on the same furnace; performance tests
indicated an average of 0.4 ppm, or about 0.01 Ib/T ADP, and a maximum
of 1.1 ppm, or approximately 0.03 Ib/T ADP.
Only limited data are available on the degree of reduced sulfur emission
control possible with older recovery furnaces, which are characterized
within the structure of existing knowledge as those constructed without (1)
the refined metholds of combustion air measurement, air distribution to
28
PULP AND PAPER INDUSTRY EMISSIONS
-------
0.1 1 10 "30" 50 70 90 99 99.9
PERCENT OF TIME TRSVINDICATED CONCENTRATION
Figure 5. Observed frequency of total reduced sulfur
concentrations in exit gases from two recovery furnaces
with good combustion controls.
various zones of the furnaces, and what is currently believed to be its proper
method of introduction, (2) refined control of the liquor feed apparatus,
and (3) indicators of adequate combustion conditions such as flue gas
oxygen analyzers, combustible meters, and reduced sulfur monitors.
An 84-hour study on one furnace built in 1947, which is routinely fired
for minimum emission level by controlling liquor feed rate and maintaining
residual oxygen content of 5 percent in the flue gas, is illustrative of the
possibility for controlling emissions from furnaces not equipped with the
more up-to-date firing control devices. As shown in Figure 6,15 sulfur
emission, which was essentially all hydrogen sulfide, was below 5 ppm 50
percent of the time, below 15 ppm 90 percent of the time, and below 30 ppm
98 percent of the time. Other data, gathered in studies of shorter duration
on furnaces of a similar vintage, revealed emission levels of less than 15
ppm of TRS.15 This indicates that the emission control possible with fur-
naces of this era is neither fully explored nor adequately understood.
When recovery furnaces are operated in conjunction with flue gas direct
contact evaporators that are fed highly oxidized black liquor, their reduced
sulfur emissions are absorbed. A combination of best furnace operation
and very high degree black liquor oxidation may therefore represent the
control necessary for a nonobjectionable operation except in extreme
situations.
Kraft (Sulfate) Pulping
29
-------
30
25
20
tSi
= 15
tS)
-------
defined on the basis of the percentage of the total sulfur that is oxidized by
the process, and the reversion of sulfur compounds to sulfide prior to the
time the liquor enters the direct contact evaporator has been neglected.
Residual sodium sulfide levels of the black liquor entering the evaporator
must be no higher than 0.1 gram per liter (currently the minimum.
measurable amount) if emissions are to be near zero (i.e., less than 0.2
ppm). However, since initial sodium sulfide concentration can range from 8
to 35 g/liter and percent conversion can cover a correspondingly wide
range, percent conversion is not an adequate measure of the performance of
a black liquor oxidation unit. Improved engineering designs that permit
continuous high-level-performance black liquor oxidation have led to the
realization of full reduced sulfur emission control capabilities at the flue
gas direct contact evaporator; and recently developed techniques to isolate
the contact evaporator for studies of factors controlling emissions on a
continuous, real-time basis have demonstrated their efficiency.
Data obtained in a series of special studies, as well as observations of
currently existing installations, illustrate the potential for controlling the
reduced sulfur emissions from the contact evaporator through control of
residual sulfide content by black liquor oxidation and pH adjustment of the
liquor entering the contact evaporator.15 These studies showed that
reduced sulfur emission control was not of maximum benefit until the
residual sodium sulfide content of the black liquor entering the contact
evaporator approached zero. Illustrative of the relationship of residual
sulfide in black liquor and reduced sulfur contributions of the contact
evaporator are the data in Figure 7.15 These data, which are from two
contact evaporators, illustrate a general increase in contact evaporator
contribution of reduced sulfur from almost zero to about 275 ppm as
sodium sulfide concentration in the liquor increases from less than 0.05 to
24 g/liter.15
A similar relationship is shown in Figure 815 for sodium sulfide concen-
trations of less than 1 liter in the liquor fed to one of these evaporators. TRS
contributions fell from 22 ppm, at sodium sulfide levels of 0.8 g/liter to zero
or less as a result of absorption of reduced sulfur from the recovery furnace
flue gas at sodium sulfide concentrations of zero. Once the sodium sulfide
levels fell below 1.0 g/liter, the flue gas direct contact evaporator contribu-
tion of dimethyl sulfide fell to zero from 3 ppm. Methyl mercaptan contri-
butions did not reach zero until the sodium sulfide levels in the feed liquor
were essentially zero.l5
At a third installation where high-degree black liquor oxidation is
practiced, the reduced sulfur contribution of the contact evaporator over a
30-hour period ranged from 0.1 to 3.1 ppm while residual sodium sulfide
levels in the black liquor ranged from 0.06 to 0.2 g/liter.15 As a result of
more effective and continuous high-performance black liquor oxidation at
this mill, resulting in only trace amounts of residual sulfide in the black
liquor entering the flue gas direct contact evaporator, emissions from this
source were shown to range from zero to less than 1 ppm. 15
Kraft (Sulfate) Pulping 31
-------
300
Cs
a:
a:
200
100
0
0
20
4 8 12 16
Na2S IN STRONG LIQUOR, grams/liter
Figure 7. Total sulfur increase across the direct contact
evaporator, sodium sulfide concentrations from 0 to 20 g/liter.
0.2 0.4 0.6 0.8
Na?S IN STRONG LIQUOR, grams/liter
1.0
Figure 8. Total sulfur increase across the direct contact
evaporator,sodium sulfide concentrations from 0 to 1 g/liter.
32
PULP AND PAPER INDUSTRY EMISSIONS
-------
Walther and Amberg recently reported contribution from a flue gas
direct contact evaporator of 0 to 1 ppm TRS when liquor entering the
contact evaporator was at pH 12 and contained 0.2 to 0.4 g/liter of sodium
sulfide. This study served to verify the work of other investigators who had
found that significant reductions in reduced sulfur emissions from the flue
gas direct contact evaporator can be made by adjusting the normal liquor
pH upward. At similar residual sulfide levels with no pH adjustment,
emissions from the contact evaporator ranged from 0 to 12 ppm. 18
Other benefits of highly oxidized black liquor feed to a contact
evaporator in controlling reduced sulfur emissions have also been
demonstrated. In one case, a cascade evaporator followed an older recovery
furnace that was fired for nominal emission but had an unstable TRS
emission level. The liquor, subjected to high degree oxidation, usually
showed zero to trace quantities of sodium sulfide. During a 71-hour study,
residual sodium sulfide concentration normally ran at 0.05 g/liter or less
and never exceeded 3.5 g/liter, the high value being due to a brief
mechanical failure in the oxidation system. Data collected at this
installation illustrate the potential of the contact evaporator to act as an
equalizer when furnace emission levels are erratic. The evaporator acted as
an absorber 21 percent of the time, it produced no change 7 percent of the
time, and there was some contribution of reduced sulfur from it 72 percent
of the time. The TRS changes, their magnitude, and their frequency are
shown in Table 10.15
Table 10. ANALYSIS OF DIRECT CONTACT EVAPORATOR
FUNCTION IN ALTERING FURNACE GAS TRS CONTENT
Condition
Absorption (-)
No change
Release (+)
Total
Difference in TRS
as H2S, ppm
>21.0
1 1 to 20
6 to 10
1 to 5
0
Oto 5
6 to 10
1 1 to 20
>21.0
Duration of condition
hours
2
8
3
2
5
13
9
18
11
71
percent
2.8
11.2
4.2
2.8
7.0
18.3
12.7
25.5
15.5
100.0
Total
duration,
percent
21.0
7.0
72.0
100.0
Kraft (Sulfate) Pulping
33
-------
In summary, information at hand indicates that proper black liquor
oxidation can control the reduced sulfur emissions from contact
evaporators to the level of 0 to 3 ppm, a level almost identical to that from
well operated recovery furnaces. Additional reduced sulfur absorption from
erratic recovery furnace discharge and sulfur dioxide absorption from
recovery furnace flue gas are other results of feeding the direct contact
evaporator a highly oxidized black liquor.
Emission Data
Questionnaire Data — Particulate emission data from 87 kraft recovery
furnace systems equipped with flue gas direct contact evaporators and
electrostatic precipitators. 10 with venturi recovery units, and 7 with
electrostatic precipitators ahead of secondary scrubbers were received in
the questionnaire survey. Reduced sulfur emission data were received from
42 recovery furnace systems that did not employ black liquor oxidation and
from 20 that did. These data are covered in the following sections.
Particulate emissions — Particulate emission data from 87 kraft recovery
furnace systems equipped with flue gas direct contact evaporators and
electrostatic precipitators were reported in the questionnaire survey. The
particulate emission levels reported are arranged by decile in Table 11. The
mean emission level in the lowest decile was 2.1 Ib/T ADP, while that in the
highest decile was 75.2 Ib/T ADP. The range of particulate emission levels
Table 11. PARTICULATE EMISSIONS FROM
RECOVERY FURNACES CONTROLLED BY ELEC-
TROSTATIC PRECIPITATORS, AVERAGED BY
DECILE GROUPS
Emission decile
First (lowest)
Second
Third
Fourth
Fifth
Sixth
Seventh
Eighth
Ninth
Tenth (highest)
Average
emission rate,
Ib/T ADP
2.1
3.3
4.8
6.8
12.4
17.0
18.4
28.4
46.3
75.2
34 PULP AND PAPER INDUSTRY EMISSIONS
-------
reported was from 1.3 to 95 Ib/T ADP, the median was 14 Ib/T ADP, and
44 percent of those reporting had particulate emissions less than 10 Ib/T
ADP. The concentration and particulate emission rates generally reflected
both the age of the precipitators and their design efficiencies, which ranged
from 75 to 99.5 percent. The complete emission data from furnaces
equipped with electrostatic precipitators are presented in Table A-l in
Appendix A. These data are plotted in Figure 9.
60
40
30
I 20
t: is
10
in
i 5
12 5 10 20 30 50 70 80 90
PERCENT OF EMISSION RATES=MNDICATED VALUE
95 98 99
Figure 9. Particulate emissions from recovery furnaces with
electrostatic precipitators, questionnaire data.
Data on particulate emissions from 10 recovery furnace systems
equipped with venturi recovery units show a range of 15 to 115 Ib/T ADP.
The median was 45 Ib/T ADP compared with a median of 14 Ib/T ADP
from those systems equipped with precipitators. These data are shown in
Table 12.
The particulate emissions from seven kraft recovery furnace systems
equipped with electrostatic precipitators ahead of secondary scrubbers
ranged from 1.8 to 13.1 Ib/T ADP. These data are shown in Table 13.
Kraft (Sulfate) Pulping
35
-------
(A
HI
0<
ocO
=) 01
U. QC
o2
OH
LU (/)
CC LU
1°
CO to
tO *• 13
O
C >-
O co
•a . ~a
°- u & '??•
"5 s *• 9t
5-D2Q
o <
a H
05
CMOOCMOO'tCO'51-OCO
OTOOOOOI^I^OOOOCDCO
OOO'-OO'-'-'-CN
tQCD»— «— OOOI^OJCN'—
coO'-'d-cMfMCNTr'-r^-
T-'-tN'-CMCMCNCSIOItN
Sooooooooo
ooooooooo
ot^-oomc^joiooom
T-COO'tCDCOCDOOr^O)
r^COCOOOOI^CMCMCN
SOo>CMe3)enr^-LOCDin
Cn^j-OOCMCMCMCOC^CO
36
PULP AND PAPER INDUSTRY EMISSIONS
-------
CO
cc
UJ
m
ca
D
CC.
o
(0
cc
0"
m UJ
oc
la
ai U
DC UJ
is
II
CO CC
CO I-
5°
2 UJ
I- CC
'u
it
%
ollection
o
1
w
O)
C*
o
1
i
0
c
8
re
I
r
re
O.
CD
2
5
_o
•s
g
CO
tr.
£.
o.
S>*
S
o
S
'o.
'o
£
o
1
o
is
0)
Ul
.
O
u
3
o
0.
O
$
—
CO
C3
c
o
1
0.
^
£
u
ectrostati
Ul
1
.0
£5
03
•o
Q)
£
S
CO
a.
O
h-
^
a>
c5
|
3
O
C/3
•ecipitatoi
a
«
3
O
3>
3
o
s
c
re
o7
Q
^"
1-
1
>
re
aT
O
1-
'E
D
oo co in oo en CN «-
«- CN CM CM CN Tf CO
^~
CM «- CO CO
1 oS r*' 1 i< in 1
O) O C7> O3
in
oo o co «*' CM in o
en r» CD co in in in
in o o co
1 r^ CM 1 in CD 1
C7) O3 Ol O3
K tSi^ScooS
O O O O O «- CO
o o o o o o o
in in
CO 00 *- CM CM CO
eo •— CM in «— CM r»
«- o o o o o o
CM 00
1 o cq | in CM |
I--' CM CM' CM
00 CO <* •* CM »J- CM
CM r«» in i-^ co o r^
CM CM T- ,- CM CM <-
§OOOOOO
O O O O O O
in «* r^ o *t *-
Soo to oj co" o> o"
r^ <* co «- in in
*~
T^1 <* ^" O O t—
-------
Gaseous emissions—Reduced sulfur emission data were received for 42
recovery furnace systems where black liquor oxidation was not used. The
reduced sulfur emission rates from these systems ranged from 35 to 1300
ppm, representing 1.5 to 62 Ib/T ADP, with a median of 5.9 Ib/T ADP.
Sulfur dioxide emission data were reported for 33 of these systems. These
ranged from 0 to 575 ppm, representing from 0 to 55 pounds of sulfur
dioxide or 27.5 Ib/T ADP as sulfur. The median sulfur dioxide emission
rate was 20 ppm or 2.4 Ib/T ADP. These data are presented in Table 14.
Reduced sulfur emission data were received for 20 recovery furnace
systems operated with black liquor oxidation systems. The range of reduced
sulfur emissions from these furnace systems was from 0.2 to 25.9 Ib/T ADP,
with a median of 3.7. This median compared with 5.9 Ib/T ADP for those
systems without black liquor oxidation. The lowest emission rate reported
was associated with the black liquor of lowest sodium sulfide content from
the black liquor oxidizer, containing 0.2 g/liter. The low emission rate of
0.2 Ib/T ADP indicates a combination of low recovery furnace emissions
and low flue gas direct contact evaporator emissions for a low total system
emission level. In two other situations reported, the emissions from recovery
furnace systems receiving the same oxidized black liquor were 3 (units 13
and 17) and 20 (units 4 and 15) times as great in one recovery furnace
system as in the other. This served to illustrate earlier discussions that
pointed out the need for detailed knowledge of the recovery furnace
contribution in evaluating the control capability of black liquor oxidation.
The emission data for recovery furnace systems with black liquor oxidation
are shown in Table 15.
EPA Test Results —As mentioned earlier, EPA conducted tests at three
mills, designated Mills A, B, and C. Particulate measurements were made
only at Mills B and C. Gaseous emissions were sampled at all three mills.
The data obtained are discussed in the following sections.
Particulate emissions — The EPA test team made particulate
measurements on two recovery furnace systems. The emissions from Mill B
were controlled with an electrostatic precipitator having a manufacturer's
rating of 97.5 percent. The average of three tests showed a precipitator
operating efficiency of 95.5 percent and an emission rate of 4.5 Ib/T ADP.
At Mill C, the particulate control devices include a 96 percent efficiency
rated precipitator followed by a wet scrubber guaranteed to collect 80
percent of all particles larger than 2 microns. The tests showed the
precipitator to have a 97.4 percent collection efficiency, the scrubber a 51.3
percent efficiency, and the overall particulate collection system a 98.7
percent efficiency. The emission rate was 3.3 Ib/T ADP. These data are
summarized in Table 16.
Gaseous emissions—Gas analyses were made at one mill (Mill A) not
practicing black liquor oxidation. Results in Table 17 show that reduced
sulfur emissions from the furnace itself were only 2.2 ppm (0.06 Ib/T ADP)
but that there was a substantial increase of 500 ppm of TRS across the
direct contact evaporator.
38 PULP AND PAPER INDUSTRY EMISSIONS
-------
*
z
o
r-
Q
X
o
CC
0
a
j
v^
0
i
GO
K~
D
o
r—
5
E2
O tf
c
^3 IT
U uj
1
o
CC
LL.
CO
z
o
CO
CO
s
LU
o
LU
CO
o
^
_«
CO
1—
GO
X
trt
(D
DC
H
E
a
«.-
si
S>
1
u
D
T3
(U
CC
CM
O
00
W
2 c
5 5
£ = 1
M- 3? T3
O £ re
o °-
'•w o
re *-
CC
,
3 |I
c *j
« Q)"
at!
Jv
s 1
UJ .0
o
£• 5
(2 •§
o
Q.
0.
Q
~
_g
£
£
£
CM
oo
"To
u
oo
-£^
CO
X
U
X
X
CJ
00
CM
E
a
re
oT
0
*~
LL
D
g
u
™
™
aT
Q
K
+-
c
iDC^LOCOCM LOCOO*"^ LOCO
»— CNCMCMCO^-^'^'^-LDlOtOr-^r^COO
«-
or^Lnc\cO(n^or^or^LncDOOcsi
t— t— If- T—
CM (N T- <— T— CNCNCN<— CO
I 1 1 1 1 1 1 I 1 II *"• °^ 1
llillltllo^-o*-
CO CO
1 1 1 I 1 1 01 1 o o o 1 Icn^i
to i i i o r^ ^" co oo o i i i i to i
O) *~ ^" O O CO 00 I""-- O * — O O * — CO O
CNJ 00 O) O CO P*- O LO ] Lf) CN CO O d ^ ^"
T-CSI'— T— «— CN CM <— T— CO
ooor-oioro-ooo LO coo
CO ^ *"" ^- »- ^
COCNCMLDCOCOCOCNOOOrN^-OOCNOCO
tOLOOOOOCOOOOOOOOOO
o o o o o o o o o o o o o o o o
^•COOOOCMOOCNOO*— *O CO O) O
r\j «— r\j r*j i— *— TJ- CM «— ^— <— *-
oorsicop^or^OLnoLntooiOLn^-co
t— OlCNLn^— CN^rxCNJCN»— COCMCNCTJCN
CO^-CDCOCNJCOCOCO^LOCOCOOJOJ^CO
*"~ CNJ 00 ^" LD CO r*« CO O) O *"• ON] CO ^J" Lfl CO
Kraft(Sulfate) Pulping
39
-------
z
0
1-
Q
X
cc
0
D
C)
^T
^t
o
J
00
i_
0
X
— p
$ "c
co ,9
Lu o
o
i<
D°
LL LU
>- 5E
GC **f
LU S
^ ^
81
LU |—
CC CO
~ ^
O
CC
LL
CO
O
CO
LU
CO
D
O
LU
(0
^
C3
^
5
S
O_
CO -D
CS~
CD _^
SI
a
a
CM
CO
CM
To
E o
a
•>" CO
S M
g, ^
3 O
±:
•g x
O CO
3 C*l
1 "
CO
CM
X
CM E
O a.
8 a
07
= § >
1 1 -s
5= = £ S:
•»- X « Q
o S "> <
0 a ,_
CO *-
CC
,
ra
s I s u.
tn c 3 o
t +••
0)
*J
V> QJ
S S E
•5 5 S
UJ ^
c >
a ° 5
tx •§ D
o <
Q. 1-
'c
_CNCN(D(Drao502
.-.-CD^OOCXD^co
*— •— CNCNCNCN»— ^5
COCOCn^coCDCMC^oo
COCM^C'1CO°CMCO|V"
CM
to •- •- "~
r^ CD | CD CM CM O 1 O
r- CM
•-OOOCDCDCMnO
<— o 1^5 co 01 co 01
1 | «- in co
"- CM
enooooooco'-cn
CM^OOOOOOO
LD CM CM *— CM CM CMCM
ooooooooo
SOOOQOOQO
OOOOOOOO
LOCMCMOCMCMr^CMCM
>- «- CM CM CM •- «-
COOOOOr^P^O'-CD
"-oocM^j-^rrfcococo
r^oooio»— cMco^tm
r-'-'-CMCMCMCNCMCM
aB(B10Nr.N
oo ^r co ^ in r~ o
CO CM CO i§ 5 S S
o o in o oo i- co
S § S 2 2 ^ co
«-«-•-
1 1 1 ^ 1 ^
•-
1 O O 1 CD (**•• CO
1 CN ' CNJ CO CO
o
| O O | CN O LO
CN i- «f CN
O O LO O O O O
^" O CN LO
111 "~ in
CXI O CO CD O O5 00
o o in o o o CM
r^« o O) o co LD oo
•- CO CM CM CM r- CM
§000000
O O O Q O Q
0 O O O 0 O
O o en CM o in r^
CM r— r- CM r- «—
r~ a> *- in in in r».
00 CM CM CO CM CM CO
Sr^ oo cn o »— CM
CM CN CM CO CO CO
40
PULP AND PAPER INDUSTRY EMISSIONS
-------
z
O
1—
O
X
o
cc
o
D
g
^
o
^
CQ
I
I-
§
CO (TJ
o **
<<
1Q
£a
ol
LLJ C/}
11
CC
LL
_
0
CO
to
n i
V)
U
0
II 1
yj
U
LO
0)
.a
E
0}
a)
u
i
3
LL
C
3
c
- —
ra
'x
O
o
3
U
m
'cv
X
V)
re
CO
H
E
a
of
(U
(U
O)
5
"S
u
•o
cc
c.
g
is
1 =
__ *rf
!5 c:
s a>
O" hf
'"i fl»
J Q.
c 2
o 3
Oxidati
temperal
h.
.§
f,
CO
ra
Z
Q.
O
l_
£
O.
e/T1
CM
O
CM
fo
I
o
i
to
X
o
CO
CM
X
IE
Q.
a
oxidized
T3
1
LL
o
a
0
a
c
'c
3
CNCOO^cocDr^«~co csir^OtDCO LDOO
T— T— CN
^cog^^lgoor.^og^ u,gs
«— <— CMCO'— «- «— ^t r~-
O
V ^ ^ o
o
OLO OiOiCN | i_ *~ i-
• * • 1 ^ ' U_ Q .^ ^_
^J CO ll) ^J T^ r^ r— ^— CN ^~ ^ \- UJ CM '
V o
«- o
10 o^^^^* oco r*«
«— COO^CN — CDOCO^-'O Icdo in— CN
in CM — m4-'*~coi^
C) 0
• o
MSSi2^S^§SS|8?58^§g
o
in — *" co
— .^—.^-'--l .i O -_•
8 i 8 i S S i 8 i 8 i i8i 8oi
^
CM
i 1^1 i ii i O i
col(£>lrocolcQlcMl loi LO^I
«— >— «— <— «— in in —CM
CM
o in
CO CO
f~ ^ f*^ f~i ^3 ^^ 10 irt
° 1 co I r- co 1 to 1 — 1 Ij^l ^21
o "~ *~ *~ *~ ^ ""in
M- CO
^ | M. | ^ ° | M. \ ^ | |^| W. 0 |
0 "~
00 °
, ^ , °» 8 , , i i ^ , « o i
o'^icoioiini i .i i •« ^ i
in —
-CMCOTtincDI-.COOO-CMCO* LOtDI^
Kraft (Sulfate) Pulping
41
-------
Table 16. RECOVERY FURNACE SYSTEM PARTICULATE
EMISSIONS, SUMMARY OF EPA TEST RESULTS
Measurement or calculation
Equivalent production, T ADP/day
Stack flow rate, scfm
R*i"fi(*i rvf voliimp 1~n nroHiiPtion av^iiii
PidLKJ LJI VUILJIIIC HJ |_HllUull*llUII,
T ADP/day
Electrostatic precipitator inlet loading, gr/scf
Electrostatic precipitator inlet loading, Ib/hr
Electrostatic precipitator outlet loading, gr/scf
Electrostatic precipitator outlet loading, Ib/hr
Scrubber outlet loading, gr/scf
Scrubber outlet loading, Ib/hr
Electrostatic precipitator collection efficiency.
percent
Scrubber collection efficiency, percent
Overall collection efficiency, percent
Emission rate, Ib/ T ADP
MilIB
957
133,300
1 QQ
loa
3.71
4,330
0.162
185
—
-
95.7
—
95.5
4.5
MilIC
560
172,000
on~7
oU/
4.18
6,160
0.107
158
0.052
77
97.4
51.3
98.7
3.3
Gas analyses were made at two mills practicing black liquor oxidation.
At Mill B, where strong black liquor oxidation is practiced, reduced sulfur
emission rates were 0.005, 0.11 and 0.19 Ib/T ADP, respectively, following
the recovery furnace, flue gas direct contact evaporator, and electrostatic
precipitator. At Mill C, where weak black liquor oxidation is practiced, the
reduced sulfur emissions were 0.014, 0.023, 0.035, and 0.075 Ib/T ADP,
respectively, at the recovery furnace, contact evaporator, electrostatic
precipitator, and wet scrubber. Each of these points was sampled on a dif-
ferent day; hence direct comparisons are not valid. The test data from these
two studies are shown in Tables 18 and 19.
The sulfur dioxide concentrations in the furnace exit gases differed
substantially, being 1 ppm at Mill B and 239 ppm at Mill C. The reduced
sulfur concentrations in the furnace exit gases at both mills were low. The
contribution of reduced sulfur from the contact evaporators differed
substantially at the two mills, being about 0.1 Ib/T ADP at Mill B and only
0.01 Ib/T ADP at Mill C. No information was collected on the sulfide
content of black liquor entering the flue gas direct contact evaporator at
these mills. The data support earlier observations concerning the very low
reduced sulfur emission levels of the recovery furnace and flue gas direct
contact evaporator when high-degree black liquor oxidation is practiced.
42
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table 17. RECOVERY FURNACE SYSTEM GASEOUS
EMISSIONS, MILL A WITHOUT BLACK LIQUOR
OXIDATION, EPA TEST RESULTS
Sample
location
Direct contact
evaporator
inlet
Electrostatic
precipitator
inlet
Electrostatic
precipitator
outlet
so2
ppm
1 4
-
Reduced sulfur gases, ppm
H2S
1.2
500
300
CH3SH
1.0
-
(CH3)2S
-
-
(CH3)2S2
-
3
TRS as H2S
ppm
2.2
503
300
Ib/hr
2.3
525
313
Ib/T ADP
0.06
148
88
Table 18. RECOVERY FURNACE SYSTEM GASEOUS
EMISSIONS, MILL B WITH BLACK LIQUOR OXIDATION,
EPA TEST RESULTS
Sample
location
Direct contact evaporator
inlet
Direct contact evaporator
outlet
Electrostatic precipirator
outlet
Number
of
samples
6
7
9
so2,
ppm
1 0
Tr
012
Reduced sulfur gases, ppm
H2S
0 19
1 7
45
CH3SH
ND
1 4
1 4
(CH3I2S
ND
037
083
(CH3I2S2
Tr
027
040
TRSasH2S
torn
0 19
40
75
Ib/lir
021
44
7 7
Ib/T ADP
0005
0 12
020
Tr - trace, ND - not detected
Table 19. RECOVERY FURNACE SYSTEM GASEOUS
EMISSIONS, MILL C WITH BLACK LIQUOR OXIDATION,
EPA TEST RESULTS
Sample
location
Direct contact evaporator
inlet
Direct contact evaporator
outlet
Electrostatic precipitator
outlet
Scrubber outlet
Number
of
samples
6
8
8
-
so2,
ppm
239
752
732
408
H2S
Tr
008
021
1 0
Reduced sulfur gases, ppm
CH3SH
ND
ND
ND
Tr
(CH-j)^
ND
ND
ND
ND
(CH3)2S2
0 12
0 15
0 19
0 13
TRSasH2S
;>pm
024
038
059
1 26
Ib/hr
034
054
084
1 79
Ib/T ADP
0015
002
004
008
Tr — trace, ND — not detected
Kraft (Sulfate) Pulping
43
-------
Black Liquor Oxidation Systems
Black liquor oxidation is the practice of oxidizing the sodium sulfide in
either weak or strong black liquor, using either oxygen or air, to sodium
thiosulfate or possibly higher oxidation stages, as represented in the
following reaction
2Na2S + 2O 2 + H2O —• Na2S2O3 + 2NaOH
Sulfur present in the latter form is not displaced by the acid components
of recovery flue gas, carbon dioxide and sulfur dioxide, as it passes through
the direct contact evaporator. Increase in the proportion of hydrogen
sulfide and methyl mercaptan present in undissociated form as a result of
passage of acidic flue gases through the liquor is thereby prevented, and
their stripping in the form of hydrogen sulfide and methyl mercaptan is
avoided. The benefits of black liquor oxidation are outlined in the section
on kraft recovery furnace systems. To be most effective, the black liquor
oxidation must reduce sodium sulfide levels to 0.1 g/liter or less in the black
liquor entering the flue gas direct contact evaporator. During the oxidation
of black liquor, some reduced sulfur compounds are stripped by the air
passing through it. This source of emissions is commonly classified as a
miscellaneous emission source. The improvement in emission levels at the
flue gas direct contact evaporator resulting from black liquor oxidation
favors its use even though the emissions from the black liquor oxidation
system may not be treated.
Designs, Application, and Performance
Black liquor oxidation systems that use air are of three types, (1) packed
towers, (2) bubble tray towers, and (3) air-sparged reactors, which may be
equipped with mechanical mixing devices to enhance oxygen transfer. One
existing black liquor oxidation system uses tonnage oxygen. Except in rare
instances, air is the more economical source of oxygen. A new oxidation
system designed for use with air in series with existing units to reduce
residual sodium sulfide levels of about 3 g/liter or less to almost zero
represents a new concept in black liquor oxidation application.20
Packed towers and bubble trays have found application almost
exclusively in the northern and northwestern United States where weak
black liquor (direct from the brown stock washing system) of low foaming
potential permits satisfactory use. Improvements in foam-breaking
equipment have led to at least three recent installations of bubble trays for
high-foaming-potential black liquor from the pulping of pine. Air-sparged
reactors are also used for oxidation of low-foaming-potential weak black
liquor.
When satisfactorily designed and maintained, all weak liquor oxidation
systems have been demonstrated to be capable of producing a liquor with
only trace amounts of residual sodium sulfide. One of the benefits once
44 PULP AND PAPER INDUSTRY EMISSIONS
-------
assigned to weak liquor oxidation, namely reduction of sulfur losses at the
multiple effect evaporators, may be overbalanced from an emissions
standpoint by sulfur reversion to sulfides in the multiple effect evaporators
and subsequent storage. Another benefit attributed to weak black liquor
oxidation is a reduction in the amount of dissolved sulfur compounds in the
evaporator condensate.
In general, some of the benefits once claimed for weak black liquor
oxidation have been minimized if not negated by subsequent developments:
1. The development of methods for collecting low molecular weight sul-
fur compounds from the multiple effect evaporators and returning
them to process.
2. The development of a control technique to adequately dispose of the
high molecular weight sulfur compounds from the multiple effect
evaporators.
3. Continued absence of demonstrated practical improvements in evap-
orator condensate quality where weak black liquor oxidation is
practiced.
Strong black liquor oxidation has found most extensive and almost
exclusive use in the oxidation of black liquor that is high in resin soap, such
as that in the southern United States. To substantially reduce the foaming
potential of the high resin soap content liquor, the liquor is partially
evaporated, to about 25 percent solids content, and a high percentage of the
soaps is removed. The liquor may be oxidized after soap removal and then
further concentrated in multiple effect evaporators to about 50 percent
solids content, and the 50 percent solids liquor may be oxidized. Heavy
black liquor oxidation systems are currently exclusively air-sparged
systems. The designs for most of these systems are similar, with (1) air
introduced through a series of nozzles, (2) provisions for deaeration storage
time after oxidation to permit satisfactory pumping of the liquor, and (3)
usually provisions for foam collection and breaking in the exhaust air.
Residence time of 3 or more hours is usually provided.21 Like weak liquor
oxidation systems, when properly designed and operated they produce a a
liquor containing almost no sodium sulfide. Sulfide reversion has been
observed to occur if the liquor is permitted to stand in storage for periods in
excess of 4 to 6 hours.
The selection of black liquor oxidation systems should be made on the
ability of the system to produce a very low sodium sulfide concentration in
the liquor entering the direct contact evaporator.
Emission Data
Special Studies - The emissions from 11 weak and 4 heavy liquor oxidation
systems measured in special studies1 are shown in Table 20. Nine of the 11
weak liquor oxidation systems had reduced sulfur emissions ranging from
Kraft (Sulfate) Pulping 45
-------
0.08 to 0.13 Ib/T ADP, one 0.22 Ib/T ADP, and the other 0.02 Ib/T ADP.
The median was 0.12 Ib/T ADP. The reduced sulfurs present were
essentially all the organic sulfur compounds of dimethyl sulfide and
dimethyl disulfide. The heavy black liquor oxidation systems had TRS
emission rates ranging from 0.01 to 0.18 Ib/T ADP. The median emission
rate for heavy liquor oxidation systems, 0.10 Ib/T ADP, was about the same
as the median for weak liquor oxidation units.
Questionnaire Survey — Emission data from six black liquor oxidation
systems were received in the questionnaire survey. These data are
summarized in Table 21. The emissions from four weak liquor oxidation
systems ranged from 0.004 to 0.73 Ib/T ADP, and those from the heavy
liquor oxidation systems were 0.01 and 0.054 Ib/T ADP. The median
emission level was about (> times greater from weak than from heavy liquor
oxidation systems. The organic sulfur compounds, methyl mercaptan,
dimethyl sulfide, and dimethyl disulfide were the principal sulfur
compounds present.
Table20. REDUCED SULFUR EMISSIONS FROM BLACK
LIQUOR OXIDATION SYSTEMS, SPECIAL STUDIES DATA
Type
of
system
Weak
Heavy
Oxidation
efficiency,
percent
83
87
90
90
50
60
60
80
80
80
85
97
97
95
99
99
98
TRS
Percent
H2S and CH3SH
0
0
0
0
—
2
2
6
2
0
0
25
0
0
—
—
—
80 |
Percent
(CH3)2S and (CH3)2S2
100
100
100
100
—
98
98
94
80
100
100
75
Emission rate,
Ib/T ADP
0.1
0.08
0.13
0.08
0.12
0.12
0.09
0.22
0.11
0.02
0.06
0.02
100 0.01
100 0.10
—
0.03
0.09
0.18
0.14
46
PULP AND PAPER INDUSTRY EMISSIONS
-------
z
9
x
o
DC
O
OH
i§
10
HI
(A
3
O
LJJ
V)
CM
to
JD
(C
&
<
•*,&
S>-
10
V) 5
EC .S
1— ~
1
Q.
C/T*
CM
i ^
1 H
<"" M
SCO
_ CM
CQ -^
0* CO
1 §
" I
1 1,
1 %
' J1
N|
co a
•s2 I 1
l!|l|
5 s S " <
5 a »-
8 M'u.
Oc 2 o
aj m
1 si
v) n o
co *-
*°l
ro o in
1^ O «- CN
o o o o
ro
oo tfi oo ^r
co ^- ^r CM
m i-
LT) CD O ID
co (o m «-
CO CM
O5 CO CM O
CO 00
r« o to o
r- rr CM
o o o o
CM •**• CO
o o o m
in
CD 1 r- |
i- CD O «-
CM CN «-
So o m
f r~ in
o o o o
cSSSS
in CD a> in
o o o o
•*r o co o
r- «- •* in
^
(V
01
?
s^
0 O
O O
in oo
^t K
CM
O5 CO
CM
0 0
(O O
o co
CO O
1 0
oo r-~
<- -
i8
-------
EPA Test Results—The EPA test team sampled the emissions from two
mechanically agitated air-sparger black liquor oxidation systems. The
reduced sulfur emissions of 0.13 Ib/T ADP from the weak liquor oxidation
system were about 3 times those of the heavy liquor oxidation system. As
found previously, the bulk of the reduced sulfur compounds were organic
sulfurs. The data from these tests are summarized in Table 22.
Control Techniques
There is no existing emission control system for black liquor oxidation
exhaust emissions. A logical method is combustion since there is no known
effective chemical absorption or oxidation process. If thermal oxidation
were to be practiced, a separate incineration device, the recovery furnace, or
a power boiler would represent logical approaches since the volume of
exhaust gas exceeds that which can be handled in the lime kiln.
Table 22. EMISSIONS FROM BLACK LIQUOR
OXIDATION TANKS, EPA TEST RESULTS
Measurement or calculation
Production rate, T ADP/day
Effluent volume, cfm
Ratiolof volume
to production, /fnp/;r~
SO2, ppm
H2S,ppm
CH3SH, ppm
(CH3)2S, ppm
(CH3)2S2, ppm
TRS, ppm
TRS, Ib/hr
TRS, Ib/T ADP
Heavy liquor
957
18,500
19
Trace
Trace
1.2
0.79
10.8
23.6
1.76
0.047
Weak liquor
560
17,000
30
1.4
9.0
21.3
7.1
1.9
41.2
3.0
0.15
Smelt Dissolving Tanks
The smelt dissolver is a large tank (3000 to 5000 ft3 or 22,400 to 37,400
gallons, measuring about 25 feet in diameter by 10 feet high) located below
the recovery furnace hearth; in it, molten sodium carbonate and sodium
sulfide smelt that accumulates on the floor of the furnace are dissolved in
water to form green liquor. It is equipped with an agitator to assist
dissolution, and a steam or liquid shatterjet system to break up the smelt
stream before it enters the solution. The dissolved sodium carbonate in the
48
PULP AND PAPER INDUSTRY EMISSIONS
-------
green liquor is later causticized, forming (with the dissolved sodium sulfide)
white cooking liquor, and completing the recovery cycle. After clarification
(settling out of suspended solids), the white liquor is ready for use as the
cooking chemical in the digesters. Contact of the molten material with the
water causes the evolution of large volumes of steam, which must be vented
at about 200 degrees Fahrenheit (°F) (dry bulb).
Emissions
Particulate matter (finely divided smelt, from submicron size to several
hundred microns in diameter) is entrained in the vapor that leaves the tank.
Because of the presence of a small percentage of reduced sulfur compounds
in the smelt, some of these odorous materials escape the tank with the
flashed steam.
Control Techniques
A widely used smelt tank control device is the mist eliminator. This
usually is a wire mesh that is supported in the vent stack and intermittently
backwashed with liquid to return the particulate matter to the smelt tank.
Wet scrubbers, pack towers, and cyclones are also used to remove
entrained material. Some means is usually provided, regardless of the
control device used, whereby large volumes of steam unexpectedly flashed
in a short time can by-pass the control device to prevent rupture of the tank
or ductwork.
On the questionnaires, nine mills reported sufficient data to compute
the efficiency of their smelt tank control system. These data are given in
Table 23.
Emission Data
Questionnaire Data—Smelt tank particulate emissions reported by 17 kraft
mills ranged from 0.05 to 2.38 Ib/T ADP as sodium oxide, with an
approximate median value of 1.0 Ib/T ADP. These data are tabulated in
Table A-2 in Appendix A. Frequency distribution of the particulate
emissions is shown in Figure 10. Concentrations ranged from 0.016 to 0.582
gr/dscf (60 °F. 1 atmosphere), with a median value of 0.30 gr/dscf.
Gaseous emission data were reported by 18 mills. These data are
tabulated in Table A-3 in Appendix A. Total reduced sulfur compounds
range from 0.013 to 3.70 Ib/T ADP as sulfur, with a median value of 0.09
Ib/T ADP. Data on individual compounds are summarized in Table 24.
Special Study - A special study was conducted by NCASI personnel in 1970
and 1971.1 The reduced sulfur contributions from 20 smelt tank vents are
summarized and reported in Table 25. Some of these units were equipped
with spray showers, demister pads, or packed towers for particulate control.
Kraft (Sulfate) Pulping 49
-------
Table 23. EFFECTIVENESS OF SMELTTANK
PARTICULATE CONTROL DEVICES. QUESTIONNAIRE DATA
Control device
Pad entrainment
separator
Pad plus shower
scrubber
Pad plus packed
scrubber
Packed scrubber
Gas flow
rate, dscf
2,100
2,700
5,050
5,600
9,000
9,090
9,800
37,000
6,350
Particulate
concentration.
gr/dscf
Inlet
0.39
0.412
0.72
5.94
4.73
1.65
1.58
1.6
1.0
Outlet
0.11
0.094
0.16
0.581
0.311
0.482
0.06
0.13
0.016
Collection
efficiency.
percent
71.8
77.2
77.8
90.2
93.4
70.8
96.2
91.9
98.4
Emission
rate.
Ib/T ADP
0.052
0.15
0.63
2.3
1.2
1.58
0.41
1.20
0.05
2.4
2.2
2.0
1.8
| 1-6
Hr 1.2
1—
I 0.8
g 0.6
0.4
0.2
12 5 10 20 30 40 50 60 70 80 90 95 98 99
PERCENT OF EMISSION RATES VINDICATED VALUE
Figure 10. Smelt lank particulate emissions, questionnaire data.
50
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table 24. SMELTTANK GASEOUS EMISSIONS,
SUMMARY OF QUESTIONNAIRE DATA
Compound
so2
H2S
CH3SH
(CH3)2S
(CH3)2S2
Emissions
Concentration, ppm
Range
2 to 385
0 to 337
0 to 400
Oto 150
Oto25
Median
92.2
37.2
47.2
27.2
7.5
Rate, Ib/T ADP as H2S
Range
0.005 to 2.1
0 to 1 .7
Oto 2.0
Oto 1.7
Oto 0.37
Median
0.08
0.05
0.05
0.16
0.17
Number of
units
8
16
12
11
10
These were operated on either lime mud washer filtrate water, fresh water,
or contaminated condensate.
As shown in Table 25, the mass emission rate of TRS from those units
operated without a particulate control device, with spray showers, or with
demister pads varied from negligible to 0.03 Ib/T ADP with the exception
of one unit at Mill 4. The data reported for Mill 4 indicated the operation of
demister pads with fresh water to be responsible for a slight absorption of
reduced sulfur. The same conclusion was drawn from a similar study
conducted at Mill 5. However, the use of lime mud washer filtrate water for
demister pad shower water at Mill 5 resulted in a small contribution of
TRS.
Only two vents equipped with packed towers for particulate control have
been monitored to date. The values reported for Mill 2 indicated a potential
for the evolution of soluble sulfides from either the scrubbing solution or
the particulate matter collected. This was attributed to the practice of
scrubbing solution recirulation and the absorption of carbon dioxide. This
is substantiated to some degree by the results reported at Mill 5 upon the
use of a shower water rich in soluble sulfides. Current particulate
regulations specific to smelt tank vent emissions dictate the use of packed
towers or possibly fine demister pads equipped with high-volume showers
and recirculation. Consequently, the necessity of further studies on smelt
tank vents equipped with high-efficiency particulate control devices is
indicated, to determine the effects of shower water source, recirculation
rate, and sulfide content of the particulate collected on reduced sulfur
emission rate.
The study did indicate that the concentration of reduced sulfur in the
emission could be maintained at 10 ppm or less. The compounds present
were mainly hydrogen sulfide and methyl mercaptan. It appears treatment
needs in the future would be predicated on emission appearance rather
Kraft (Sulfate) Pulping
51
-------
Table25. SMELT TANK TRS EMISSIONS,
SPECIAL STUDIES DATA
Mill
1
2
3
4
5
6
7
8
9
10
11
Production
rate,
T ADP/day
130
250
145
215
215
435
435
435
435
420
420
520
520
520
155
310
400
-
240
400
375
600
350
350
Water
source
1
2
2
2
2
-
2
-
2
—
1
-
2
—
2
3
-
2
2
2
1
1,3
1,3
Control device
or test point
No control device
Packed tower
Spray showers
Spray showers
Spray showers
After demister
Before demister
After demister
Before demister
After demister
Before demister
After demister
Before demister
After demister
Showers
Showers
After demister
No control device
After demister
After demister
After demister
Packed tower
After demister
After demister
TRS
Concentration,
ppm
1.0 to 2.5
10 to 40
ND to 0.6
1 .0 to 20.0
3.0 to 26.0
10 to 35
20 to 66
1.5 to 3.0
4 to 9
ND
ND
2.0 to 4.0
0.8 to 1.8
ND
4 to 6
5 to 8
2 to 5
ND
4 to 6
4 to 6
2 to 6
ND
1 to 1.5
ND to 2.5
Rate, Ib/T ADP
asH2S
0.01
0.12
Ng
0.04
0.04
0.08
0.11
0.01
0.02
ND
ND
Ng
Ng
ND
0.02
0.03
0.01
ND
0.02
0.01
0.01
ND
Ng
Ng
aNg - negligible; ND - not detected.
1 - lime mud washer filtrate, 2 - fresh water; 3- contaminated condensate.
than possible ground level concentrations of reduced sulfur that might
occur at most mill locations. Venting to the main stack for increased
dilution and despersion appears to be an effective control technique.
EPA Test Results — Two smelt tanks serving one recovery furnace were
tested for particulates by the EPA test team. Total effluent volume
averaged 28,000 scfm. Emissions were controlled by wire mesh mist
eliminators. Measurements were made only after the mist eliminator.
Average emissions for the two stacks totaled 34.8 Ib/hr or 0.77 Ib/T ADP.
Gaseous emissions at two mills were sampled by the EPA test team.
Results are summarized in Table 26.
52
PULP AND PAPER INDUSTRY EMISSIONS
-------
Lime Kilns
The lime kiln is an essential element on the closed-loop system that
converts green liquor to white liquor. The kiln supplies calcium oxide
(quicklime, CaO), which is wetted (slaked) by the water in the green liquor
solution to form calcium hydroxide, Ca(OH)2for the causticizing reaction.
The reaction in the causticizer
Ca(OH)2 + Na2CO3 — 2NaOH + CaCO3
provides sodium hydroxide for cooking liquor and precipitated lime mud,
which is recycled through the kiln. The lime sludge enters as a 55 to 60
percent solid-water slurry. The reaction which takes place in the kiln is
CaCO,, + heat — CaO 4- CO0
o Z
Most lime kilns used by pulp mills are the rotary type, usually ranging
from 8 to 13 feet in diameter and 125 to 400 feet in length. The kilns are of
steel construction and are inclined at an angle of about 10 degrees to the
horizontal. Lime mud is fed in at the elevated end and contacted by hot'
gases resulting from the combustion of natural gas or fuel oil and proceding
through the kiln in the opposite direction. Large motors (several hundred
horsepower) turn the entire kiln at low speeds, causing the lime to proceed
downward through the kiln toward the high-temperature zone (1800 to
2000 °F) to discharge at the lower end. As the lime melt and mud move
along, they dry in the upper section, which may be equipped with chains or
baffles to give the wet mud better contact with the gases. As the lime moves
down farther, it agglomerates into small pellets and finally is calcined to
calcium oxide in the high-temperature zone near the burner.
Rotary kilns are capable of producing the large quantities of quicklime
required by kraft mills (40 to 400 T/day), but heat losses through the long
kiln are considerable.
Fluidized bed calciners are presently being used by a few manufacturers,
but their production rate at this time is not as great (25 to 150 T/day).
Table26. SMELTTANK GASEOUS EMISSIONS,
EPA TEST RESULTS3
Mill
B
C
so2,
ppm
Tr
0.17
Reduced sulfur gases, ppm
H2S
3.7
2.0
CH3SH
1.1
ND
(CH3)2S
2.3
ND
(CH3)2S2
005
Tr
TRS as H2S
ppm
7.0
2.0
Ib/hr
1.42
0.27
Ib/T ADP
0.036
0.011
ND - not detected, TR - trace.
Kraft (Sulfate) Pulping
53
-------
Emissions
The rolling and tumbling of the lime in the rotary kiln and the
vaporization of sodium compounds (carried into the kiln with the lime mud)
in the high-temperature zone and their later condensation are responsible
for the formation of most of the particulate matter carried by the kiln
exhaust gases. This emission constitutes not only an air pollution problem
but also a loss of usable chemical.
The lime dust is made up of particles of sizes ranging from 1 to over 100
microns in diameter, while the soda fume consist of very small particles,
most less than 1 micron in diameter. Therefore, the lime dust is removed
from the exhaust gas'quite easily; but the soda fume is very difficult to
remove.
Some odorous sulfur gases are emitted from lime kilns. Hydrogen sulfide
is produced in the kiln by the reaction
Na2S + H20 + C02 — Na2C03 + H2S
Sodium sulfide entering the system as an impurity in the lime sludge and
sulfur bearing process water used as scrubbing liquid in the lime kiln may
be sources of reduced sulfur emission.
At some mills, the odorous noncondensable gases from the digesters,
blow tanks, and multiple effect evaporators are incinerated in the lime kiln.
Burning of these reduced sulfur gases forms sulfur dioxide, which
apparently reacts with either the lime in the kiln or the alkaline scrubbing
solution in the scrubber, resulting in no significant increase in total sulfur
emission.
Control Techniques
Odorous emissions of hydrogen sulfide can be reduced by adequate
washing of the lime mud precipitate from the causticizer to remove
adherent white liquor containing sodium sulfide. However, a small amount
of sodium compounds (about 0.25 percent) is intentionally left in the lime to
avoid the formation of large loose balls or rings of lime adhering to the
inner surface of the kiln in the agglomeration zone.
Sufficient excess air must enter the kiln for complete combustion of the
fuel and for oxidation to sulfur dioxide of some hydrogen sulfide formed in
the kiln by the reaction of water, carbon dioxide, and sodium sulfide.
Several types of control equipment are available for the reduction of lime
kiln particulate emissions. The water scrubber,usually of the impingement
or venturi type, is used exclusively in the kraft industry. The impingement
type is a fairly low pressure drop unit (5 to 6 inches of water) in which the
exhaust gas flows through wetted baffles with water sprays between them.
Nozzles and baffles tend to be obstructed by lime buildup, which may
54 PULP AND PAPER INDUSTRY EMISSIONS
-------
reduce efficiency and cause maintenance problems. Maximum solids
content of the scrubbing medium is usually less than 2 percent.
Venturi scrubbers are generally more efficient (97 to 99 percent) than
impingement type scrubbers and operate with a pressure drop of 10 to 20
inches of water. They can use a scrubbing medium of up to 30 percent
solids since they have less tendency to sludge up, and thus do not require
cleaning as frequently as impingement scrubbers. Water flows down the
edges of the vertical venturi, from large orifices, to the throat where an
annular ridge causes it to splash outward into the throat. Particulates in the
gas flowing through the throat in the same direction as the water are caught
by the curtain of water. The extreme turbulence in this area prevents the
buildup of sludge. Increasing the water and/or gas velocity increases the
pressure drop across the scrubber as well as the efficiency.
If the scrubbing liquor contains sodium sulfide, as it does in some
installations, hydrogen sulfide may be formed in the scrubber from the
reaction of sodium sulfide, carbon dioxide, and water in the same manner
as it is in the direct contact evaporator. Other reduced sulfur compounds in
the scrubbing liquor may also be partially stripped from solution.
Other sources of particulate emissions associated with kilns are open
conveyors, elevators, slaker vents, and storage facilities for lime. Enclosing
these sources and venting them through the kiln control device or a
separate air cleaner appears to be the most effective means of control.
Emission Data
Particulate stack emission data for 66 lime kilns were reported by 35
mills on the questionnaire. Rates ranged from 0.08 to 43 Ib/T ADP. The
median concentration was 0.4 gr/dsct. Stack flow rates ranged from 5800
to 24,500 dscfm, with a median value of 13,800 dscfm or 33 dscfm/T ADP.
These data are tabulated in Table A-4 in Appendix A. The distribution of
emission ranges is plotted in Figure 11.
Particulate concentrations at the inlet to the scrubber were reported for
15 kilns. These data were separated by scrubber types, efficiencies were
computed, and results are tabulated in Table 27.
Gaseous emission data from 22 lime kilns were reported in the
questionnaires by 13 mills. These data are tabulated in Table A-5 in
Appendix A. Emissions of TRS compounds ranged from 0.015 to 4.0 Ib/T
ADP hydrogen sulfide as sulfur. The median value was 0.43 Ib/T ADP.
Values for individual compounds are summarized in Table 28.
Particulate tests on lime kilns were run at two mills by the EPA test
team. Both kilns were controlled by venturi scrubbers with pressure drops
of about 10 inches of water. Results are summarized in Table 29.
Kraft (Sulfate) Pulping 55
-------
12 5 10 20 30 50 70 80 90
PERCENT OF EMISSION RATES * INDICATED VALUE
95 98 99
Figure 11. Lime kiln particulate emissions, questionnaire data.
Table 27. LIME KILN SCRUBBER EFFICIENCY,
SUMMARY OF QUESTIONNAIRE DATA
Impingement scrubbers
Paniculate concentration.
gr/dscf
Inlet
3 50
14.05
14.29
9.22
16.02
11.78
14.81
Averages
11.94
Outlet
0.46
0.43
0.58
1.05
0.88
1.56
0.53
0.78
Collection
efficiency.
percent
86.9
96.9
95.9
88.6
94.5
86.8
96.4
92.2
Venturi scrubbers
Particulate concentration.
qr/dscf
Inlet
4.68
10.00
6.33
9.30
13.80
12.14
6.05
2.55
8.11
Outlet
0.16
1.00
0.23
0.13
0.12
0.14
0.38
0.37
0.32
Collection
efficiency.
percent
96.5
90.0
96.4
98.6
99.1
98.9
93.7
85.5
94.8
56
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table28. GASEOUS LIME KILN EMISSIONS,
SUMMARY OF QUESTIONNAIRE DATA
Compound
so2
H2S
(CH3)SH
(CH3)2S
322
Emissions
Concentration, ppm
Range
Oto 140
0 to 500
Oto 90
0 to 245
Oto 11.4
Median
33.8
107.9
14.0
27.0
5.4
Rate, Ib/T ADP as H?S
Range
0 to 2.35
0 to 4.0
0 to 0.36
0 to 0.46
Oto 0.21
Median
0.3
0.5
0.07
0.05
0.03
Number of
units
13
24
10
14
10
In the EPA testing program, two lime kilns were sampled for gaseous
compounds. The water used for scrubbing the gases from both kilns
contained only traces of organic sulfur compounds. Both kilns were used to
incinerate digester and evaporator noncondensable gases. Only traces of
organic sulfur gases were detected. The hydrogen sulfide and sulfur dioxide
results are given in Table 30.
Table 29. LIME KILN PARTICULATE EMISSIONS,
EPA TEST RESULTS
Mill
B
C
Gas flow
rate,
scfm
17,610
16,500
Particulate loading
Concentration,
gr/scf
Inlet
7.89
Outlet
0.322
0.285
Rate,
Ib/hr
Inlet
957
Outlet
48.7
40.1
Scrubber
efficiency,
percent
95.9
Emission
rate,
Ib/T ADP
1.42
1.7
Table 30. LIME KILN GASEOUS EMISSIONS,
EPA TEST RESULTS
Mill
B
B
C
C
Compound
so2
H2S
so2
H2S
Emissions
Concentration, ppm
Range
Trace
7.2 to 79.9
1.2 to 48.8
30 to 146
Mean
_
25.0
20.4
67.8
Rate, Ib/T ADP as H2S
Range
_
0.04 to 0.5
0.027 to 0.44
0.1 5 to 0.72
Mean
_
0.16
0.18
0.34
Number of
tests
8
8
10
12
Kraft (Sulfate) Pulping
57
-------
Brown Stock Washing Systems and Other Miscellaneous Sources
Emission inventories and special studies have identified the significant
emission sources from brown stock washing systems to be those associated
with the washing of black liquor from, or screening of, pulp. Such sources
include the brown stock washer hood and knotter vents, the exhausts of
vacuum pump systems used on brown stock washers, and the brown stock
filtrate tanks, which serve to collect the black liquor and air from brown
stock washers operated with a barometric leg. Washer hood vents are
usually mechanically exhausted, although a limited number of older
systems are vented by natural draft and a smaller number yet have no hoods
or vents whatsoever. Knotters may be vented singularly, vented in a
common washer hood vent system, or operated without a vent system. Few
knotter systems are vented singularly. Depending on the system design,
vents from other processes where black liquor is violently agitated, such as
the salt cake mix tank, could be a miscellaneous source emission.
Lime kiln slaker vents are also commonly listed as a miscellaneous
source, although the emission of reduced sulfur, if any is present, is no more
than a trace. Both the smelt tank vent and black liquor oxidation tower
vents are also commonly characterized as miscellaneous sources. These two
processes and their emissions are described elsewhere in this report.
In the past, these miscellaneous sources have been characterized as
minor emission sources based on the relative amount of reduced sulfur
compounds emitted from them as compared to those from uncontrolled
major emission processes such as the recovery furnace system. With
application of the technology available for control at major emission points,
there can be a major rearrangement in the significance of miscellaneous
emissions. Under the conditions of controlled major sources, they may
represent the largest source of reduced sulfur emissions.
Emission Data
The data in Table 31 represent the reduced sulfur emissions from 17
mechanically vented brown stock washing systems.1 The emissions are
characteristically dominantly dimethyl sulfide and dimethyl disulfide from
the roof vent system and almost exclusively these two compounds from the
undervent system (vacuum pump exhausts and filtrate seal tank vents). As
shown in Table 32, the median TRS emission level from the roof vent of 14
of the above systems using fresh water on the washers was 0.04 Ib/T ADP,
compared to 0.35 from five systems using condensate as a washing medium.
The median TRS emissions from the undervents on these systems was 0.08
and 0.11 Ib/T ADP, respectively. As shown in Table 33, the median TRS
emission for the total system was 0.10 when using fresh water and 0.40 Ib/T
ADP when using condensates as a washing medium in these systems. The
most significant difference in emissions between systems using fresh water
and those using condensates was at the roof vents, as shown in Table 32.
The difference in median levels was 0.31 Jb/T ADP, or 9 times greater when
condensates were used.
58 PULP AND PAPER INDUSTRY EMISSIONS
-------
CO
Z
O
CO
01
CO
OC
UJ <
co<
>Q
OC UJ
II
15
O tu
I- Q.
COCO
Z
O
cc
CO
ES
*
CO
CM CM
sa
-1- m
c x
S o
fi £ 8
cc tr to
J2 CC
CM CM
"CO^CM
X -X,
5?
H
c x
g O
a. CM
Q X
CO
cc
3 i
*rcq«-ppCMpcM'-«-p'-ppCN
o' o' d o' d o d o' d o' o' o' o' o' o" o o o' o'
ooooooomoooooo
OOOOOOOI^OI^OOOO 1
o
lO
oooooooinoooooo
CM CO
I I I
00
LTJ LO
odd ddodddddddddddd
8OOOOOOOOOOOOO
OOOOOOOOtO)'-'-^-
QO
| |OO
OOOOOOOOOOOOOO
•- to «- a> a> CD
OO
«— r~ r- m in o 10 ^ «— ^ «— «— oo i— coioo
•-CMO ip'-p'-ppppppp^pcoS
odd ddodddddddddddd
8 ^ .c ^ ^ j:
"g U. U. LL LL
O
O
t/» tf> t/» v)
Q) QJ O QJ
^.c-c-cSSJIls!
iriuli"g"g"g"°
o o o o
o o o o
Kraft (Sulfate) Pulping
59
-------
Table 32. BROWN STOCK WASHER SYSTEM TRS EMISSIONS,
ROOF VENTS AND UNDER VENTS,
SPECIAL STUDIES DATA
Number
of
systems
13
5
Wash
water
Fresh
Condensates
TRS as H?S, Ib/T ADP
Roof vents
Max.
0.08
0.65
Min.
0.01
0.03
Median
0.04
0.35
Mean
0.05
0.32
Under vents
Max.
0.63
0.30
Min.
0
0.05
Median
0.08
0.11
Mean
0.13
0.15
TABLE 33. BROWN STOCK WASHER SYSTEM TRS
EMISSIONS, TOTAL SYSTEM, SPECIAL STUDIES DATA
Number
of
systems
14
5
Wash
water
F resh
Condensate
TRS as H2S, Ib/T ADP
Max.
0.84
0.90
Min.
0.015
0.14
Median
0.10
0.40
Mean
0.17
0.46
The bulk of the stripping of reduced sulfurs from condensates in these
systems occurs in the washers (Table 34). The difference in median
discharge levels from the undervent system when using the two different
wash waters was 0.03 Ib/T ADP.
A comparison of TRS emission levels from two systems operated
alternately on fresh water and condensates is shown in Table 34. Roof vent
Table 34. BROWN STOCK WASHER SYSTEM TRS
EMISSIONS, ALTERNATE USE OF FRESH AND
CONDENSATE WASH WATER, SPECIAL STUDIES
DATA
System
number
5
5
14
14
Wash
water
Fresh
Condensate
Fresh
Condensate
TRS as H2S, Ib/T ADP
Roof vents
0.01
0.15
0.08
0.41
Under vents
0.02
0.09
0.18
0.22
System
0.03
0.24
0.26
0.63
60
PULP AND PAPER INDUSTRY EMISSIONS
-------
losses increased 0.14 and 0.33 Ib/T ADP, representing increases of 15 and 5
times, when condensates were substituted for fresh water at the washers.
The undervent system increases were less, being 0.07 and 0.04 Ib/T ADP,
respectively. System emission increases were 0.21 Ib/T ADP, or 8 times, and
0.37 Ib/T ADP, or 2.5 times, respectively, for the two systems when
condensates were substituted for fresh water.
Walther and Amberg report a reduced sulfur emission of 0.6 Ib/T ADP
from a brown stock washer system using fresh water.22 This is higher than
the median of 0.10 Ib/T ADP found in these 14 systems but less than the
maximum found, which was 0.84 Ib/T ADP.
The EPA test team made stack measurements at brown stock washer
systems and knotter hood vents at two mills. The data are given in Table 35.
Table 35. BROWN STOCK WASHER SYSTEM GASEOUS
EMISSIONS,EPA TEST RESULTS a
Mill
B
B
C
C
C
Operation
Brown stock
washer, west
Brown stock
washer, east
Brown stock
washer
Washer seal
tank
Knotter hood
Gas flow
rate,
scfm
20,000
10,000
34,600
760
15,300
so2,
ppm
Tr
Tr
025
Tr
020
Reduced sulfur gases, ppm
H2S
2.2
0.9
Tr
268
030
CH3SH
39
3.5
Tr
189
0.24
(CH3)2S
56
11 1
2 1
382
14 1
ICH3)2S2
26
TRS as H^
ppm
16.9
1 6 ! 18.7
08
502
25
3.7
769
19.6
Ib/T ADP
005
003
003
013
0.06
aTr- trace
Control Techniques
There is no known feasible absorption or chemical oxidation system for
brown stock washer system vent gases. A logical control method is thermal
oxidation, and the only control system for these sources in the United States
uses this method. In this case, the vent gases are burned in the recovery
furnace using a specially designed injection system. The large volumes,
which range from 35,000 to 50,000 cfm per washing system, limit the
existing process equipment suitable for handling such a volume to the
recovery furnace or possibly a steam or power boiler. The procedure is not
attractive for application in any but new recovery furnace systems due to
major engineering and construction changes required. The reasonably low
gas volumes from most undervent systems suggest the possibility of burning
these in the lime kiln. In those cases where this can be done in conjunction
with the use of fresh water in the washers, a reduction of approximately 90
percent emission level for this source would be accomplished.
Kraft (Sulfate) Pulping
61
-------
-------
SEMICHEMICAL SULFITE PULPING PROCESSES
INTRODUCTION
Raw Materials
Semichemical pulps can be made by any of the commerical cooking
processes by reducing cooking time, temperature, or the amount of
chemical charged to the digester. Many types of wood can therefore be
pulped using these processes. Semichemical pulping process is used for the
most part to produce pulp from hardwoods for use in making corrugated
board. About 9 percent of the wood pulp produced in the United States is
made in this manner.
Process Description
As pointed out above, Semichemical pulps can and are manufactured by
variations in other commerical chemical pulping processes. The process
most extensively used is the neutral sulfite semichemical (NSSC) process
however. Semichemical pulping is a two-stage process that uses a mild
chemical treatment of the chips to weaken the intercellular bonding by
partial removal of the hemicellulose and lignin, followed by mechanical
treatment to separate the individual fibers.
For the manufacture of NSSC pulps, chips, usually hardwood, are
cooked in batch or continuous digesters with a nearly neutral solution of
sulfite containing a small amount of alkaline agents such as carbonate,
bicarbonate, or hydroxide. After cooking, they are blown to a blow tank.
The cooked chips are further processed in disk refiners, washed free of
cooking liquor, and converted to board or paper. Some NSSC pulp is
bleached prior to use, but the practice is limited due to high bleaching
costs. Since a number of NSSC mills were converted from other processes,
many variations in operating conditions and equipment exist.
The cooking liquor may be prepared either by adding fresh chemicals
(sulfite and carbonate) to water or spent liquor or by absorbing sulfur
dioxide generated in a sulfur burner in a sodium carbonate solution.
The spent cooking liquor may be discharged as a liquid effluent, burned
with heat recovery but no chemical recovery, or burned with heat and
chemical recovery. Prior to burning, the liquor is concentrated in multiple
effect evaporators similar, if not identical, to those used in the kraft
63
-------
process. Chemical recovery may be combined with the kraft recovery
process, a process commonly referred to as cross recovery, with soda and
sulfur from the semichemical process replacing salt-cake make-up in the
kraft process. In cross recovery schemes, the spent cooking liquors are
mixed prior to burning.
Semichemical liquor may also be burned alone in a conventional kraft
furnace for recovery of chemicals. Green liquor from smelt formed is
sulfited directly or indirectly to produce new cooking liquor. There are
only a limited number of Ihese systems. The nature and control of emissions
from these systems may differ significantly from other semichemical
processes, however, since the smelt from which cooking liquor is produced
may contain sodium sulfide. The liquor may also be burned under oxidizing
conditions in fluidized combustion, forming sodium sulfate, which may be
used for make-up in the kraft process.
Product Yield
Pulps that have a yield ranging from about 60 to 80 percent are generally
classed as semichemical pulps. Those pulps in the lower semichemical yield
range that require only moderate mechanical treatment may be called high
yield chemical pulps.
EMISSIONS
Because of the difference in the chemical attack on the lignin when using
sulfite liquors, such compounds as methyl mercaptan and dimethyl sulfide
are not formed during digestion. The NSSC process should,therefore, be
free from these odorous compounds. In addition, the absence of sulfide ions
from the cooking liquor will virtually eliminate hydrogen sulfide as a
possible emission. Exceptions to this rule might be expected in those
systems where liquor is burned in such a manner that the smelt contains
sodium sulfide. Atmospheric emission sources from an NSSC mill will
include sulfur dioxide absorption towers, if they are used, blow tanks, spent
liquor evaporators, and the liquor burning or chemical recovery furnace. In
the case of spent liquor recovery in a kraft mill recovery system, the NSSC
liquor is said to have an effect upon the emissions from the kraft recovery
system.4 This is not documented however.
Sulfur Dioxide Absorption Tower
Sulfur dioxide absorption towers generally are countercurrent packed
towers using soda ash or another alkaline solution as the absorbing
medium. Chemical absorption takes place according to the following
reaction
Na2CO3 + SO2 —Na2SO 3 + CO2
64 PULP AND PAPER INDUSTRY EMISSIONS
-------
This reaction leads to the emission of carbon dioxide. Sulfur dioxide may
also be absorbed in water according to the following reaction
SO2 + H2O — H2SO3
Nearly total absorption of sulfur dioxide in the tower is feasible in a
properly designed and operated tower. Emissions are, therefore, dependent
on the design and operating conditions of the individual towers. Smelt that
is formed when burning the cooking liquor in a converted kraft furnace
under a partial reducing atmosphere and then dissolved to form green
liquor contains sulfide. If sulfited, it will release hydrogen sulfide.23
Blow Tank
When the cooked pulp is blown into the blow tank, large amounts of
steam and gases escape from the pulp and spent liquor. Sulfur dioxide is
the major gaseous emission. Recently, recovery systems have been installed
to recover the sulfur dioxide from the blow pit gases. These installations are
based upon the absorption of sulfur dioxide by scrubbing either in the vent
stack or by routing the gas to the sulfur dioxide absorption tower. Again,
the emissions will be determined by the design and operation of the system.
Chemical and Heat Recovery Furnace
In those situations where semichemical liquor is burned for the single
purpose of destroying organic matter with no chemical recovery, a water
quality protection measure, the residual sulfur present in the liquor can be
expected to be emitted as sulfur dioxide. The mills that follow this practice
all produce pulp using an ammonia pulping base.
Where semichemical liquor is burned in conjunciton with black liquor
form the kraft process in cross recovery, there is no evidence that the
conditions controlling emissions from the furnace are different than when
kraft black liquor alone is burned. The potential exists for a pH reduction
when the two liquors are mixed in cross recovery systems. This might have
an effect on the reduced sulfur emissions from the contact evaporator,
although there is no evidence that such a situation occurs.
In theory, the sulfur emissions from fluidized bed combustion systems
should be extremely low. Short-term studies show this to be so. The
emissions measured on two different systems amounted to no more than 0.1
Ib/T ADP, of which 70 percent of the sulfur emission was reduced sulfur,
principally hydrogen sulfide, and 30 percent was sulfur dioxide.
When semichemical liquor is burned in a partial reducing atmosphere,
some sodium sulfide is reported to exist in the green liquor, white or
cooking liquor, and spent cooking liquor. Hydrogen sulfide stripping in a
flue gas contact evaporator would be anticipated if one were used. It has
been shown that proper and controlled amounts of combustion air can
Semichemical Sulfite Pulping Processes 65
-------
control the reduced sulfur emissions from these furnaces. Anticipated
emission rates of reduced sulfur from the furnace by this control technique
are 0.25 Ib/T ADP. It has also been demonstrated that the ratio of soda to
sulfur present in the cooking liquor is related to the sulfur dioxide emission
from these furnaces.23
EMISSION DATA
Questionnaires were sent to 60 acid sulfite and nonintegrated neutral
sulfite mills. Eleven were returned by the nonintegrated semichemical mills,
ten from soda base and one from ammonia base mills.
Information on the use of methods to control sulfur dioxide emissions
from absorbers and from digester relief and blow gas are summarized in
Table 36. Absorbers were considered controlled if a second absorber was
used or if a scrubber was used after the absorber.
Table 36. EXTENT OF CONTROLS USED IN SEMICHEMICAL
PULPING OPERATIONS, QUESTIONNAIRE DATA
Operation
Absorbers
Digester relief gas
Digester blow gas
Number
of
mills
6
11
11
Controls used
Yes No
3 3
3 8
3 8
Relief and blow gases were controlled by venting to the absorber or to a
separate scrubber. Three of six mills using absorbers provided controls, and
three of the eleven reporting used digester and relief gas control.
The data provided in the questionnaires on the amounts of sulfur
dioxide emissions are shown jn Table 37. The data are so sparse that a
summary is not warranted.
The information on types of cooking liquors used and the types of heat
recovery or chemical recovery furnaces is summarized in the following
paragraphs.
Ten of the mills use sodium-base liquors. Three of these use fluidized
bed recovery furnaces. The chemicals are recovered in the form of sodium
sulfate and sodium carbonate and can be used as make-up chemicals in a
kraft process.
66 PULP AND PAPER INDUSTRY EMISSIONS
-------
Table 37. SULFUR DIOXIDE EMISSIONS FROM SEMICHEMICAL
PROCESSES, QUESTIONNAIRE DATA
Mill
number
21
25
28
30
31
Emission rate, Ib/T ADP
Absorption
Before
control
0.33
ND
0.365
0.298b
0.1 92b
After
control
ND
0.021
ND
-
Digester relief
Before
control
_
-
_
0.775b
-
After
control
_
-
_
_
-
Furnace
Before
control
—
-
_
26.5
ND
After
control
—
-
_
2.4
7.1
Type
mill
NH3
NH3
Na
Na
Na
aNo data reported for digester blow or evaporators.
ND — process controlled, but no data reported.
No controls.
Two of the mills using sodium-base liquors use recovery systems
designed by the Institute of Paper Chemistry. In this system, the spent
liquor is concentrated in multiple effect evaporators and burned in a kraft
type recovery furnace. A smelt of sodium carbonate and sodium sulfide is
dissolved to make green liquor. This liquor is sulfited in an absorbing tower
using sulfur dioxide from a sulfur burner. This produces cooking liquor
containing sodium sulfide, sodium carbonate, and sodium hydrosulfide.
Hydrogen sulfide and sulfur dioxide are emitted from the sulfiting tower.
One of these mills reported total emissions of 20.1 Ib/T ADP of hydrogen
sulfide and organic sulfur compounds and 5.21 Ib/T ADP of sulfur dioxide.
The other mill reported total emissions of 24.7 Ib/T ADP of hydrogen
sulfide and organic sulfur compounds and 13.5 Ib/T ADP of sulfur dioxide.
One of the mills reporting uses ammonium-base cooking liquors. This
mill concentrates the liquor to 45 percent solids and burns it in a pulverized
coal-fired boiler. The resulting sulfur dioxide is vented to the atmosphere.
CONTROL TECHNIQUES
Little information is available on control of emissions from new NSSC
mills. The limited information on emission control from semichemical
pulping is referred to in other parts of this section. Control of emissions
from those unit operations similar to sulfite sources (absorption tower, blow
tank) will be described in the following section.
Semichemical Sulfite Pulping Processes
67
-------
-------
SULFITE PULPING PROCESS
INTRODUCTION
Raw Materials
Sulfite pulp can be made from several types of wood, but soft woods are
generally used. The specific type of wood used depends both on the final
product desired and the cooking base employed.
Process Description
Sulfite pulping is an acid chemical method of dissolving the lignin that
bonds the cellulose fibers together. Many of the older mills use a sulfurous
acid-calcium bisulfite solution for the cooking acid. Calcium-base spent
liquor, because of problems associated with evaporation and chemical
recovery and its low chemical cost, is not normally recovered. To
satisfactorily recover the spent liquor, several other acid bases have been
developed, the most important being sodium, magnesium, and ammonium.
Because sulfite pulp is used in a wide variety of end products, pulping
conditions will vary considerably between mills. These products can include
pulp for making high-grade book and bond papers and tissues, for
combining with other pulps, and for making dissolving pulp for producing
cellophane, rayon, acetate, films, and related products.
The pulping operation involves cooking wood chips with a low resin
content in the presence of an acid within a digester. The heat required for
cooking is produced by the direct addition of steam to the digester or by the
steam heating of the recirulated acid in an external heat exhanger. The
cooking liquor, or acid, is made up of sulfurous acid and a bisulfite of one
of the above four bases. The sulfurous acid is usually produced by burning
sulfur or pyrites and absorbing the sulfur dioxide in liquor. Normally, part
of the sulfurous acid is converted to the base bisulfite to buffer the cooking
action. During the cooking action, it is necessary to vent the digester
occasionally as the pressure rises within the digester. These vent gases
contain large quantities of sulfur dioxide and, therefore, are recovered for
reuse in the cooking acid.
Upon completion of the cooking cycle, the contents of the digester,
consisting of cooked chips and spent liquor, are discharged into a blow pit
or tank. During this operation, some water vapor and fumes escape to the
atmosphere from the tank vent. The pulp then goes through a washing
69
-------
stage, where the spent liquor is separated from the fibers. The washed pulp
is either shipped or kept within the plant for further processing.
Product Yield
Wood cooked by the sulfite process has a pulp yield ranging from well
below 50 percent for dissolving pulps up to 60 percent, and generally is
about 50 percent or below. A pulp yield above 60 percent generally calls for
classifying the pulping operation as semichemical even though it may be an
acid sulfite or bisulfite cooking process.
EMISSIONS
Sulfur dioxide is the principal atmospheric emission from the sulfite
processes. The main sources of sulfur dioxide release are the absorption
towers, blow pit or dump tank, multiple effect evaporators, and liquor
burning or chemical recovery systems.
Absorption Towers
Industrial absorption towers for sulfite processes are usually packed
towers or venturi absorbers. In the case of packed towers, sulfur dioxide gas
is introduced in the bottom of the tower while an alkaline solution of the
desired base (ammonia, etc.) is introduced at the top of the tower. Where
calcium is the base, limerock (CaCOs) is used as packing into the tower.
Sulfur dioxide reacts with water to yield an acidic solution.
Acid fortification towers are also absorption towers. Weak cooking
liquor is passed through the tower for the purpose of absorbing additional
sulfur dioxide. This replenishment of sulfite in the liquor offsets the sulfite
lost through mill emissions as sulfur dioxide or combined as lignosulfonic
acids in the spent pulping liquor.
The quantity of sulfur dioxide gas delivered to the absorption tower will
depend upon the desired pH and strength of inorganic chemicals in the
cooking liquor. The amount of sulfur dioxide emitted will depend on the
design and operating conditions of the individual towers.
Blow Pits
There are three methods of discharging the digester; hot blowing, cold
blowing, and flushing. In a hot blow, the pressure in the digester is relieved
to a predetermined level and the contents are then blown into a blow pit. In
cold blowing, the pressure in the digester is relieved to a low level and the
contents are then pumped into a dump tank below the digester. Spent
liquor is introduced into the bottom cone of the digester to reduce pulp
consistency and aid discharge. In the flushing system, after the digester has
been relieved, spent liquor or hot water is pumped into the digester for
70 PULP AND PAPER INDUSTRY EMISSIONS
-------
several minutes at a high rate. The blow valve is then opened and the pulp is
discharged while the flushing liquid continues to enter the digester.
The three types of digester discharge affect the amount of sulfur dioxide
that is emitted to the atmosphere. Gases that leave the digester during relief
are sent to the accumulators, where they fortify the cooking liquor. The
gases that pass through the accumulators are sent to other areas in the
process, where they may be absorbed. However, blow tanks and dump
tanks are usually vented to the atmosphere. Gases carried from the digester
to these tanks may, therefore, be sources of emissions. Recent installations
of gaseous control devices on blow pit gases will reduce these emissions.
A review of the digester discharge system shows that for the hot blow
style, the pressure is only partially relieved before the blow is made. The
digester gases that were not relieved will be sent to the blow pit during the
blow. Significant quantities of sulfur dioxide are therefore emitted in this
style of discharging if no recovery is practiced. In the cold blow and flushing
style of discharging the digester, the pressure is almost fully relieved, and
the relief gases are routed to the accumulators. That fraction of the gases
that remains in the digester may then escape from the system when the pulp
is discharged to the dump tanks. Ranges of sulfur dioxide emissions that
might be expected from the blow pit vent stack without scrubbing are: 4
Potential source SC>2 emissions, Ib/T ADP
Blow pit, hot blow 100 to 150
Dump tank 10 to 25
Recovery Units
Practices in the recovery of the base used in sulfite pulping differ widely
from mill to mill. Because of the variety of chemical and physical properties
exhibited by the base chemicals — calcium, sodium, magnesium, and
ammonia—different processes have been developed to satisfy the handling
and recovery problems peculiar to each base. In some instances, no attempt
is made to recover the chemical or sensible heat of the spent liquors, or in
some cases, only the heat is recovered. No attempt is made to recover
chemicals from calcium base liquor.
Spent liquor from several magnesium sulfite processes can be burned in
a heat and chemical recovery system in which the inorganic salts break
down into magnesium oxide and sulfur dioxide. These chemicals can then
be recombined directly to produce magnesium bisulfite acid for cooking.
The absorption towers absorb the sulfur dioxide with a solution of
magnesium hydroxide. Sulfur dioxide enters the towers from the recovery
furnace as well as from the fortification towers and/or the digester.
Magnesium hydroxide readily absorbs the sulfur dioxide. Absorption
Sulfite Pulping Process 71
-------
efficiencies of the venturi systems range from 95 to 98+ percent. Sulfur
dioxide emissions from the absorption system range from 10 to 25 Ib/T
ADP. 4
A process for the recovery of pulping chemicals from ammonium-base
bisulfite spent cooking liquor has recently been reported to be in
operation.24 The spent cooking liquor is concentrated to about 53 percent
solids in multiple effect evaporators. The concentrated liquor is then
burned in two conventional oil-fired boilers at a rate of 60 gallons per
minute (gpm). About 3 gpm of No. 6 fuel oil is also fired with the liquor as
the liquor itself will not support combustion. The combustion gases
containing sulfur dioxide are absorbed to produce cooking liquor. The
absorbers are of the sieve tray type. One absorber has six trays, and
emissions are reported to be 65 ppm sulfur dioxide. The other absorber has
four trays, and emissions are reported to be 118 ppm. Total sulfur dioxide
released is about 1 T/day, which is about 18 Ib/T ADP.
Multiple Effect Evaporators
Multiple effect evaporators, which concentrate spent liquors, may be a
source of sulfur dioxide emissions. Such emissions are evolved because of
the high-temperature and low-pressure conditions in the effects. The type
of condenser has an effect on the sulfur dioxide emissions. Adequate
contact between the sulfur dioxide and the cooling water will remove a large
portion of the gas. Sulfur dioxide emissions from the multiple effect
evaporators are in the range of 5 to 10 Ib/T ADP.4
CONTROL TECHNIQUES
Little information is available on control of emissions from sulfite mills.
Those unit operations that are similar to kraft sources can use control
systems similar to those already described.
Sulfiting Tower
The pressure accumulator is usually vented into the acid storage tank
because the two tanks are nearby. The acid storage tank is then vented to
the absorption tower. This system effectively prevents emissions from these
items of equipment, thereby leaving the acid absorption tower as the only
significant source of emission in the acid system.
The efficiency of absorption of most sulfiting towers exceeds 90 percent.
Some mills have placed a second absorption tower in series with the
sulfiting unit for scrubbing exhaust gases. The increase in design efficiency
and the required operating conditions are unknown.
Blow Pit
Flash steam, sulfur dioxide, and inert gases are released in the blow pit
during a digester blow. These gases then exit from the blow pit, and after
72 PULP AND PAPER INDUSTRY EMISSIONS
-------
the steam is condensed, the noncondensable gases, mainly sufur dioxide,
are generally absorbed in a packed tower. The recovered sulfur dioxide is
reused in the process, and the condensate creates a source of hot water. The
system is about 95 percent efficient in absorbing sulfur dioxide from the
blow pit stacks, where the sulfur dioxide concentration is approximately 4
percent during the initial stage of the blow. This produces recovery water
containing 0.85 percent sulfur dioxide. The recovery water is subsequently
used directly in the fortifying of the cooking acid, or the sulfur dioxide is
stripped from the recovery water and used in the manufacture of cooking
acid.
Recovery Furnace Systems
Significant amounts of particulate matter will be generated if sodium or
magnesium based liquor is burned in a recovery furnace, but probably not
if ammonia based liquor is burned. Sulfur dioxide will be generated in all
cases where liquors are burned. The degree of recovery of sulfur dioxide
from the recovery furnace for reuse in the liquor making process is a
function of (1) the efficiency of liquor collection in the pulp washing
process, (2) the efficiency of the sulfur dioxide collection systems on the
acid-making system absorbers, digesters, recovery furnace, etc., and (3) the
required liquor characteristics for the type of pulp being manufactured.
The interrelationship between the above factors and the amount of sulfur
dioxide in the tail gas from the recovery furnace system is best understood if
it is recognized that the cation base and sulfur may not exist in
stoichiometric amounts in cooking liquor. Depending on the type of pulp
being manufactured and the cation base used, a certain amount of free
sulfur dioxide exits in the liquor. This results, in some situations, in an
excess of sulfur dioxide in fresh and spent liquor over that which can
combine with the cation in a recovery furnace sulfur dioxide recovery
system. Whether there is sufficient cation added in the case of ammonia or
recycled and added in the case of magnesium to capture a major portion of
the sulfur dioxide in the recovery furnace tail gas and return it to reuse is
mainly dependent on the amount of free sulfur dioxide in the liquor going
to the recovery furnace. Assuming that sulfur dioxide recovery from
miscellaneous sources is equal in all cases, the liquor collection system
efficiency dictates how much sulfur (free and combined sulfur dioxide) and
how much cation go to the furnace. The amount of cation carried in the
system dictates whether sufficient cations are present to capture the sulfur
dioxide from the furnace in the tail gas scrubber. The sulfur dioxide
emission rates from sulfite recovery furnace systems do not therefore
directly reflect the performance of the sulfur dioxide absorbers used on the
furnace but may reflect insufficient cation to capture the sulfur dioxide
present. This makes it necessary, for example, to distinguish between sulfur
dioxide emission capability of the bisulfite systems referred to earlier as
compared to sulfite systems. To obtain equivalent sulfur dioxide emission
rates from different systems may therefore require special external
treatment schemes in one case and not another.
Sulfite Pulping Process 73
-------
EMISSION DATA
Questionnaires were sent to 60 acid and nonintegrated semichemical
pulp mills. Twenty were returned from mills manufacturing acid sulfite
pulp.
The information on the control of sulfur dioxide emissions from
absorbers, digester relief gas, and blow gas, is summarized in Table 38.
Absorbers were considered controlled if a second absorber in series with the
main absorber was used or if a scrubber was used after the absorber. Relief
and blow gases were considered controlled if vented to the absorber or to a
separate scrubber. Fifteen of twenty used some control on absorbers, fifteen
of sixteen reporting controlled digester relief gas and four of eighteen
controlled digester blow gas.
Table 38. EXTEIMT OF CONTROLS USED IN SULFITE
PULPING OPERATIONS, QUESTIONNAIRE DATA
Operation
Absorbers
Digester relief gas
Blow gas
Number
of
mills
20
16
18
Controls used
Yes
15
15
4
No
5
1
14
The data given in the questionnaires on the amounts of sulfur dioxide
emissions are summarized in Table 39. The information on types of cooking
liquors used and the tjpes of heat or heat and recovery boilers is
summarized in the following paragraphs.
Table 39. SULFUR DIOXIDE EMISSIONS FROM SULFITE PROCESS,
QUESTIONNAIRE DATA
Mill
number
4
5
12
13
14C
15
17
25
27
34
Sulfur dioxide. Ib/T A DP
Absor
Before
control
7 1
35b
013
93
-
ND
03
ND
ption
After
control
004
ND
ND
9.05
021
0.09
1 29
Digester relief
Before
control
-
096
_
_
_
-
After
control
-
ND
-
-
-
-
Digester blow
Before
control
ND
0335b
-
-
-
-
After
control
0.193
-
-
_
_
-
Evaporators
Before
control
-
44.83
_
-
_
-
After
control
-
ND
_
-
-
-
Furnace
Before
control
-
ND
ND
-
-
-
-
After
control
-
5
9
-
_
_
-
3NO - process controlled, but no data reported
^Jo controls.
cTwo furnaces
74
PULP AND PAPER INDUSTRY EMISSIONS
-------
Seven of the mills used magnesium-base cooking liquor. Of these, two
operated recovery furnaces. Magnesium oxide and sulfur dioxide are
recovered in these systems and recombined to make the cooking liquor.
Eight of the mills reporting used calcium-based cooking liquors. Two of
these mills concentrated their spent liquor and burned it in conventional
boilers. Presumably, no use is made of the resulting calcium sulfate.
Six of the mills use ammonia-base cooking liquors. One of these mills
burns its spent cooking liquor. It concentrates the weak liquor to 60 percent
solids in a quadruple effect evaporator and burns it in a Combustion
Engineering Special Liquor Burner.
Sulfite Pulping Process 75
-------
-------
STEAM AND POWER GENERATION
INTRODUCTION
Pulp and paper mills require large amounts of steam for process
heating, utility heating, driving equipment, and genereating electricity. A
portion of the steam is generated in recovery furnaces, but much of it is
generated in conventional industrial boilers.
A separate questionnaire form was sent to all mills for information on
power boilers. Information was requested on the following subjects:
1. Mill production, capacity, and type.
2. Steam generation and use.
3. Boiler and emission control equipment.
4. Fuel types, quantities, and composition.
5. Emission data.
6. Automatic monitoring units.
7. Analytical procedures.
The number, types, and capacities of pulp and paper mills that reported
some information are summarized in Table 40. The 288 mills reporting
represent 66 percent of the total pulp and paper capacity of the United
States.
Fuels Used
The fuels used by the pulp and paper mills are coal, oil, natural gas, and
bark/wood waste. The amounts of each fuel used by each of the mill
categories are given in Table 41. The portions of the total energy supplied
by each fuel are: coal — 35 percent; oil — 27 percent; gas — 26 percent;
and bark/wood — 12 percent. Most mills burned combinations of fuels:
31.9 percent of the plants reporting used bark plus other fuels, 39.9 percent
used coal plus other fuels, 54.7 percent used oil plus other fuels, and 35.5
percent used gas plus other fuels. A further breakdown of fuel usage is
presented in Table 42.
The characteristics of the fuels are tabulated in Table 43. Coal used had
an average sulfur content of 1.9 percent and ash content of 8.1 percent. Oil
averaged 1.8 percent sulfur, and bark/wood averaged 2.9 percent ash.
77
-------
Table 40. TYPES OIF PULP AND PAPER MILLS REPORTING
INFORMATION FOR POWER BOILER QUESTIONNAIRE SURVEY
Production
category3
C
C+P
M
M+P
C+M+P
P
Total
Number of mills
reporting
15
91
6
13
22
141
288
Total U.S. capacity (1968)
Percent of U.S. capacity
reporting
Total nominal capacity of mills reporting.
T/day
Chemical pulp
8,367
58,683
12,712
79,762
114,500
70
Mechanical
pulp
878
2,965
7,014
10,857
21,000b
52
Paper/
paperboard
58,042
4,980
18,511
19,974
101,507
165,000b
62
aC —chemical pulp; M — mechanical pulp; P — paper and/or paperboard.
b Estimated.
Table 41. FUEL CONSUMPTION IN PULP AND PAPER MILL
POWER BOILERS
Production
category3
C
C+P
M
M+P
C+M+P
P
Total
Number
of mills
reporting
12
80
5
11
18
124
250
Approx heat value of
fuel used, 1010 Btu/dayc
Percent of total energy
consumption
Annual average fuel consumption
Coal, T/day
1 ,700 (2)
12,600(37)
280 (2)
1,190(7)
3,110 (4)
7,170(50)
26,050(112)
68
35
Oil,gal./day
134,000(7)
1,773,300 (48)
2,710 (2)
142,960 (5)
493,420(12)
902,540 (67)
3,448,930(141)
52
27
Gas, 103ft3/day
33,000 (5)
312,270 (32)
750(1)
18,140(5)
100,220 (10)
33,650 (23)
498,030 (76)
50
26
Bark/wood,
T/day
2,800 (8)
16,890 (50)
80(1)
93 (4)
3,960 (11)
480 (1)
24,303 (75)
22
12
aC — chemical pulp, M — mechanical pulp; P — paper and/or paperboard.
Figures in parentheses represent number of mills reporting use of the indicated fuel.
cBasis: coal - 13,000 Btu/lb; oil - 150,000 Btu/gal.; gas - 1000 Btu/ft3, bark/wood
4500 Btu/lb.
78
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table 42. FUEL USAGE DATA FOR PULP AND PAPER
MILL POWER BOILERS
Fuel usage
Coal
Coal + oil
Coal + gas
Coal + bark
Coal + oil + gas
Coal + oil+ bark
Coal + gas + bark
Coal + oil + bark + gas
Oil
Oil + gas
Oil + bark
Oil +gas + bark
Gas
Gas + bark
Bark
Number of
mills reporting
66
6
3
11
6
10
3
4
67
21
16
20
17
23
0
Percent of
total
24.17
2.19
1.10
4.03
2.20
3.66
1.10
1.47
24.5
7.69
5.86
7.32
6.23
8.42
-
Table 43. CHARACTERISTICS OF FUELS BURNED IN
POWER BOILERS AT PULP AND PAPER MILLS
Production
category3
C
OP
M
M+P
C+M+P
P
Total
Number of
mills reporting
15
90
6
13
22
129
275
Overall mean average
Range
Heating value
Coal.
Btu/lb
12,100
(2)
13,100
(39)
14,400
(3)
12,500
(7)
13,300
(4)
13,200
(52)
13,100
(107)
10,500 to
14,700
OiL
Btu/gal
149,000
(7)
148,000
(60)
149.000
(3)
147,000
(71
149,000
(12)
149,000
(61)
149,000
(150)
122,000 to
155,000
Gas
Btu/ft3
1030
(7)
1030
(44)
1050
(2)
1010
(6)
1040
(14)
1020
(33)
1030
(1061
-
Bark/wood.
Btu/lb
4240
(9)
4810
(53)
6200
(1)
3270
(3)
4140
(12)
3150
(2)
4590
(80)
-
Ash content,
percent
Coal
125
(2)
80
(37)
52
(3)
86
(7)
9 1
(3)
8 1
(54)
8 1
(1061
35to
2 1
Bark/wood
1 2
(7)
3 1
(34)
34
-
-
(0)
26
(6)
6 9
(2)
29
(50)
0 1 to 20
Sulfur content,
percent*1
Coal
20
(2)
1 7
(39)
1 1
(3)
22
(7)
1 9
(4)
20
(51)
1 9
(106)
05 to 10
Oil
1 7
(7)
20
(51)
23
(2)
20
(6)
20
(13)
1 6
(57)
1 8
(136)
0 1 to 3 9
aC - chemical pulp, M - mechanical pulp, P - paper and/or paperboard
Figures in parentheses represent number of mills reporting the indicated da
Steam and Power Generation
79
-------
Steam Usage
The steam use distribution, by production category, is summarized in
Table 44. The overall average use for all plants was as follows:
Process heating
Equipment drives
Utility heating
Electricity generation
TOTAL
68 percent
17 percent
13 percent
10 percent
108 percent
The total is greater than 100 percent because some mills use the exhaust
steam from equipment drives and electricity generation for process and
utility heating, thus using some steam twice.
Table 44. STEAM USE DISTRIBUTION
AT PULP AND PAPER MILLS, POWER BOILERS ONLY
Production
category3
C
C+P
M
M+P
C+M+P
P
Total
Number of
mills reporting
14
91
6
13
22
138
284
Average steam use distribution, percent on weight basis
Electricity
generation
Average
6
9
3
9
12
11
10
Range
Oto23
Oto67
Oto20
Oto25
OtoSO
Oto99
Equipment
drives
Average
26
18
5
11
18
17
17
Range
3 to 90
OtoSO
Oto24
Oto47
0 to 66
0 to 90
Process
Average
75
72
46
68
66
66
68
Range
50 to 99
30 to 99
Oto97
40 to 98
24 to 90
Oto99
Utility heating
Average
6
8
36
14
6
16
13
Range
Oto30
0 to 40
Oto99
2 to 28
Oto20
Oto90
Total
113
107
100
102
102
110
108
aC — chemical pulp; M — mechanical pulp; P — paper and/or paperboard.
Types of Boilers
The types, ages, and firing methods of the boilers are summarized in
Table 45. The average age for all boilers was 23 years.
The firing methods for the 397 coal-fired boilers included the following:
Pulverized
Spreader stoker
Underfed stoker
Traveling grate
Cyclone furnace
Vibrating grate
26.0 percent
24.7 percent
22.4 percent
22.1 percent
2.8 percent
2.0 percent
80
PULP AND PAPER INDUSTRY EMISSIONS
-------
CC
C/3
Q
Z
OC
LU
a.
Q.
o
z
D
Q.
O
LU
0}
CC
LU
O
ffl
CC
LU
O
0.
u.
O
CO
LU
CD
1
£
O)
2 1
*-
Ui
"5 £
r .i
i*f
c £ o
1 = £
•g -°
£ 1
O- c
fl
-= o
S ^
^ "S
O) t_
•i JS
E 1
SL
flj J2
fc S
2"
er size,
> steam/hr
o ^
03 CO
o
Number of
Production
«
CC
S,
0>
<
1
o
6
*™
3
o
u.
m
0
0
CD
<
1
CC
3,
(0
4
V
g1
CO
CC
ft
0)
<
mills reporting
categorya
O lf> CM O ,-, O
>- r- o «- S in
CO ^- CM CM J£J CM
O O ° O 0 O
~ +• o *- S ~
r^ CM ^ o ... o
oo <- m >- "* CM
co o CM r^ *- co
r- O *— tO CO If)
CO If) 00 CM CM CM
«- o | •* t | en
•» in CM o co co
«- ID | CO «- CM
CD CD ^t O) *- (^ 00
CM ID — 1 O -
CJ
1
LU
J_-
J«i
o
CD
•D
tu
CD
1
o>"
TO
O)
c
i
">
o
oT
2
CT
en
*O)
>
CO
1
CO
•D
(U
^ f
™ ?
5 i
u 1
£<
a .
-S fe
-D -a
C i_
ca o
CD -D
CC ^
?s
- c
Q. S
3 -a
°- 0
Is
1 a
8 §
E <"
1 fc
"!
D O
a -D
T5 0
O -C
1 1
f 1
ra"^
- a
-9-^0
D O -O
a
c/>
co
i
T3
'fe I
Steam and Power Generation
81
-------
EMISSIONS
The sulfur emissions from the burning of coal and oil in power boilers
were computed from the total fuels used (Table 41) and the average sulfur
content (Table 43). As indicated in Table 46, coal-fired boilers emitted a
total of 988 T/day sulfur dioxide and oil-fired boilers a total of 483 T/day
sulfur dioxide, for a total of 1471 T/day.
CONTROL TECHNIQUES
On the questionnaires, 320 boilers were reported as having particulate
control devices. The data are summarized in Table 47. On all but four
boilers, the control devices were multiple cyclone collectors. The other four
devices consisted of two scrubbers and two electrostatic precipitators.
From the above group, emission data were reported for 43 boilers. Of
these, 17 were fired with coal only. The emission data for these boilers are
tabulated in Table 48. The average concentration to the collector was 2.24
gr/ft3, with an average outlet concentration of 0.39 gr/ft3, for an average
collection efficiency of 78 percent. This concentration is equivalent to an
emission rate of 181b/T of coal fired. Multiplying the rate by the total
amount of coal used from Table 41 gives a total emission rate of 470,000
Ib/day of particulates for the mills responding to the questionnaire,
assuming that the average emission rate can be applied to all the boilers.
Emission data were reported for 26 boilers that were used to burn bark
and wood waste. All but two of these boilers also burned other fuels. The
data are tabulated in Table 49.
The average fuel utilization for this group, on a Btu basis, is as follows:
bark/wood — 48.5 percent; oil — 31.0 percent; coal — 11.3 percent; and
gas — 9.2 percent.
Particulate emissions from all of these boilers were controlled by cyclone
collectors. The average emission concentrations were 2.74 gr/scf at the
collector inlet and 0.45 gr/scf at the collector outlet. Total emission from
the 26 boilers was 71'77 Ib/hr from the burning of 7337 T/day of
bark/wood, plus auxiliary fuels. This is an emission rate of 23 Ib/T of
bark/wood. Assuming that this rate is valid for the other boilers burning
bark/wood, the total particulate emission from the respondents is 560,000
Ib/day.
NEW TECHNOLOGY
Since the questionnaires were returned, the installation of an
electrostatic precipitator to control fly ash emission from a coal and bark
fired boiler has been reported by Nachbar.25 This is believed to be the first
installation of an electrostatic precipitator to control emissions from a
combination coal and bark fired boiler. Previously, it was widely believed
that the carbon content of the char would cause it to be difficult to collect in
an electrostatic precipitator. A pilot study showed that precipitability was
"medium to good."
82 PULP AND PAPER INDUSTRY EMISSIONS
-------
Table 46. POWER BOILER FLUE GAS CHARACTERISTICS
Production
category3
C
C+P
M
M+P
C+M+P
P
Total
Flue gas
Flow,
103acfm
Mean
110
102
124
35
92
41
Range
43 to 311
4 to 470
1 to 460
8 to 103
2 to 504
Temperature,
°F
Mean
428
463
453
427
387
477
Range
200 to 750
70 to 100
365 to 550
300 to 600
275 to 550
Total SO-, emissions,
T/d2avb-c
Oil-fired boilers
18.8 (7)
248 (48)
0.4 (2)
20.0 (5)
690(12)
126 (67)
483 (141)
Coal-fired boilers
64 (2)
478 (37)
10.6 (2)
45.2 (7)
118 (4)
272 (50)
988 (112)
Atmospheric
particulate
emissions,
gr/scfc
1 08(5)
0.44 (26)
0.21 (1)
0.42 (3)
0.22(1)
aC - chemical pulp, M - mechanical pulp, P - paper and/or paperboard.
Emissions calculated from data on average fuel consumption and sulfur content reported in Tables 41
and 43 and assuming 8 pounds per gallon of fuel oil
°Figures in parenthesis represent number of mills reporting data.
Table 47. POWER BOILER PARTICULATE CONTROL
EQUIPMENT DATA
Production
category3
C
C+P
M
M+P
C+M+P
P
Total
Percent
Number of mills
reporting
15
91
6
13
22
139
Control equipment.
number of boilers
Yes
23
171
4
15
28
79
320
33
No
21
231
6
28
60
311
657
67
Pressured drop.
inches of water
Average
2.6
27
-
23
3.6
2.4
Range
06 to 40
0.2 to 6.0
-
05 to 3.0
1.0 to 100
0.3 to 5.9
Rated efficiency.
percent
Average
929
856
92.0
842
87.6
80.3
Range
85 to 98
•38 to 99
92
70 to 90
80 to 95
20 to 96
Fly ash reinjection.
number of boilers
Yes
19
90
4
4
13
52
No
25
321
6
39
75
335
aC — chemical pulp, M — mechanical pulp, P - paper and/or paperboard.
Table 48. EMISSION DATA FROM POWER BOILERS
FIRED WITH COAL ONLY3
Mill
number
031
031
096
107
113
119
178
272
273
134
178
275
Average
Boiler
number
6
8
2
4
2
3
19
18
4
6
5-6b
9°
25C
2
3
1-5
7
Collector rating
Pressure drop,
inches of water
1 5
1 7
25
25
23
30
-
-
26
3
-
-
22
23
39
3
1 7
Efficiency,
percent
90
90
92
92
91
88
85
85
93
85
85
95
97
86
90
85
93
Flow,
103dsclm
43
104
78
121
62
111
110
128
75
25
140
140
183
15 5
166
208
90
Fly ash
reinjection
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Yes
Yes
Particulate concentration,
gr/dscf
Inlet
221
1 58
-
-
-
-
-
-
1 29
029
_
489
493
255
021
-
224
Outlet
073
068
008
065
024
028
0 19
059
04
0035
025
0 1
043
066
028
0033
028
039
Collection
efficiency,
percent
62
57
-
-
_
-
-
-
69
88
_
91
86
89
84
-
78
Emission
rate,
Ib/hr
269
603
534
674
128
264
183
656
257
8
300
120
674
88
40
59
216
298
Controlled by centrifugal collectors except as noted
bSc rubber
cElectrostatic precipitator
Steam and Power Generation
83
-------
The precipitator was put into operation in November 1969. Numerous
test gave the following average results:
Gas flow rate 387,500 acfm
Inlet dust loading 1.03 gr/scf
Gas velocity, full force 3.78 ft/sec
Collection efficiency 99.5 percent
Precipitator operation and performance are reported as being excellent.
Normally there is no visible emission. Condensed water vapor is visible only
during the coldest winter days.
84 PULP AND PAPER INDUSTRY EMISSIONS
-------
o
0
--V
CC
m
X
i-
>
Q
LU
CC
CO
5 W
_J —I
O 3
OQ LL
CC CC
III III
0 £
Q. O
2 ^
O J
CC Q.
2
O
z
o
(0
CO
§
LU
.
§
0)
.a
(0
c
o
1 2
UJ
§ z
s S
"5 it
O a
- = it*
I SS5
s 8 * &
o
*£ +
to »•
•c
n
_>
«-
1
V
3
» "g '»
o £ _g
8 3 ffi
a!
£
a
•5
CD
I Q
* «
a i
£
'5
m
=
' .c
" r
. 8
a
m
O
F
. t!
n1
o
i
u
- a
0
>-
•o
3
5
_
o
£
m
" «
§
£
a>
i
O
£
C
fn
f
i
f
CMOr--ocooo ^•cncofoocMr^cO'— CNCMLnococococo o>r^
oOLncocN^-co 1 loomcQinocN^rr-cocococNCOintoco en r^.
^ ^
5 ^ ^ ^ CM - - w ° - 5 co 5 co - co ° " co [o ^t to en S S ^t S
OOO'-'-OOOOOOOOOOOOOOOOOOOOO 0
ocn o »- «- coo ^r
OlO CNOCMCO *— T— CM T— *3-(N
in
I co in T— T— T— co (Dtor^coO'— oofo^tocncooiricocsi ^-
SS^SSS8|SSS£SS|S££SSSooooo
o
CO
OOOCDOCDinOOOOOOOOOOOOOOOOOOO ^
OOOOOOOOOCDOOOOClOO^rcOC31OOOOOO "^
•^t co co <^- co en
ino)CNCocO'-or-.co incMcocooorococooOLncoooo *-
c^Lnro^^ CN.-'CO co ^r CD co - CN ^ in r^ ^ CD co
in
rSinS^^CNCNr^co^McSSIncooMrMro^ScN^foco^ co
| OCOCNCMCMCOOOCNOCOCNCMrOinin03-S;inT3-[^CDCOCD
&n&c>a>a>a>c>caa>cnc>c>3>3>^c>a)cQ
cotnininco in.too m COCN CD'—
CNCNJcNJCNJfvi^rcocococN ' 'co'co'co 'COCNCOCMO 'co^i-dcN
r^^-T-T-ror-r-^-Lnr^roco^inncNro^^CNCoracQCNro^-
*— co CNJ CD r^co^- co in ^— r-. cocn too cvicsicDm ^- cu-2
ro^-r-vOi o *— ^r cocooi'— J^ *~ inco ^e^SS pS -rf £
5
Steam and Power Generation
85
-------
-------
APPENDIX A: DETAILED EMISSION DATA
FROM QUESTIONNAIRE SURVEY
87
-------
Table A-1. PARTICULATE EMISSIONS FROM RECOVERY
FURNACES CONTROLLEDBY ELECTROSTATIC PRECIPITATORS
Unit
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Pulp production
rate,
T ADP/day
599
870
632
664
232
362
637
657
375
1766
900
821
902
340
188
865
621
225
328
902
621
95
358
390
815
281
160
600
380
298
225
163
350
492
263
213
626
569
315
330
510
149
379
108
213
139
492
225
265
350
332
317
200
328
250
359
Gas flow
rate,
dscfm
119,000
88,700
135,000
117,000
62,500
73,800
65,900
110,000
52,000
60,500
118,000
144.001)
71,800
42,000
29,000
146,000
78,000
47,700
57,003
91,800
85,300
49,700
71,400
40,300
138,001}
61,700
46,500
114,000
61,500
86,000
54,000
38,000
61,800
103,800
70,300
94,500
96,000
72,6013
97,000
89,000
105,001)
64,500
78,000
24,200
43,000
69,003
114,500
93,000
43,700
73,700
71,003
85,303
31,003
57,503
62,203
83,003
Ratio of flow rate
to production,
dscfm
T ADP/day
198
102
214
176
270
204
103
167
139
130
131
175
80
124
154
164
125
209
174
102
137
520
200
103
170
220
295
190
162
390
240
233
176
210
267
443
153
128
308
270
206
433
206
224
202
568
232
413
165
210
214
270
155
175
249
231
Paniculate
concentration,
gr/dscf
Inlet
644
6.4
40
232
-
259
207
-
-
3.4
6.42
4 14
361
2.53
-
55
407
3.88
-
3.03
526
1 55
-
322
5.0
3454
427
2.45
2.83
-
3102
3.26
4.04
-
308
2.41
_
3.099
5.12
4.4
-
271
3.68
1 25
4.1
1.48
1 96
-
395
1 42
-
269
323
4 1
-
-
Outlet
00327
0.076
004
006
004
0.0524
011
007
0.094
017
0105
0167
0176
0.124
011
0.1
0.143
0.0955
0 12
0.211
0.157
00424
0.133
0.23
0148
0118
0092
0.15
0178
01
0124
0.139
0 188
0168
0 14
0.1
03
0373
0.17
02
0317
0152
0327
0307
0365
0148
0.33
019
049
040
040
0325
0.57
0.51
0362
04
Collection
efficiency.
percent
995
98.8
990
97.5
-
980
94 7
-
_
950
98.4
960
95 1
95 1
-
98.2
965
97 5
-
93 1
97 1
978
92.9
97 1
96.6
97.9
93.9
93.8
_
960
954
Emission rate,
Ib/T ADP
Inlet
263
134
176
842
_
109
44!
-
_
553
173
706
592
64
-
191
105
169
-
635
149
167
"
685
1743
170
255
958
944
-
153
156
954 147
-
95 5
95.8
_
87 9
96.7
95.5
-
944
91 1
755
91 1
90.0
83.2
—
87 6
720
-
880
824
876
-
-
-
170
221
-
81 4
325
245
-
241
156
58
170
151
939
-
134
746
-
149
103
148
-
-
Outlet
1 33
1 59
1 76
2 17
221
221
234
241
27
28
283
284
289
314
249
3 5
370
4 15
429
442
444
457
464
489
52
534
549
587
593
594
612
665
684
737
772
917
948
98
108
11 24
134
135
139
142
15 1
151
158
162
166
1735
176
180
182
184
186
19 1
88
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table A-1. PARTICULATE EMISSIONS FROM RECOVERY
FURNACES CONTROLLED
BY ELECTROSTATIC PRECIPITATORS (Con't)
Unit
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
Pulp production
rate,
TADP/day
225
210
260
225
287
280
318
395
150
206
390
316
160
335
380
130
130
379
171
239
165
437
437
250
120
208
139
560
239
192
Gas flow
rate,
dscfm
42,000
70,000
38,000
61,000
73,000
100,000
1 1 3,000
63,400
93,000
48,500
76,700
79,000
54,600
66,500
76,500
50,000
50,000
86,400
35,000
65,000
72,000
120,500
1?0,500
74,700
63,000
58,000
62,200
107,000
86,500
82,800
Ratio of flow
rate to produc
tion. dscfm
T ADP/day
186
333
146
270
256
357
356
160
620
235
196
250
340
200
201
285
385
228
205
272
437
276
276
300
525
280
450
190
350
430
Paniculate
concentration,
gr/dscf
Inlet
284
50
297
-
60
-
623
-
334
4.26
4 14
3.46
305
40
4.0
-
Outlet
051
03
073
0.394
0413
03
0819
0721
019
0.5
068
0.552
0469
087
0815
0.5
05
089
3 96 1 05
-
-
9.95
9.95
-
3.22
253
1 35
-
28
215
08
0.5
0.89
089
0.99
060
1 04
0.725
0366
1 2
1 05
Collection
efficiency,
percent
82 1
940
-
868
_
950
_
884
85 1
840
867
865
71 5
-
87 5
875
-
73.7
_
91.1
91 1
_
81 4
589
470
_
58.0
503
Emission rate,
Ib/T ADP
Inlet
109
343
-
166
_
442
_
206
_
162
173
213
243
125
_
316
316
_
167
_
_
565
565
348
145
124
_
209
191
Outlet
195
206
21 9
22
22
221
234
238
242
245
276
283
330
357
358
395
395
41 8
44 1
448
448
505
505
61
648
648
666
151
895
949
Appendix A
89
-------
Table A-2. PARTICULATE EMISSIONS FROM SMELT TANKS
Unit
1
2
3
4
5
6
7
8
9
Pulp production
rate,
T ADP/day
400
902
358
312
388
294
383
254
500
10 I 308
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
266
165
280
139
130
130
822
493
316
298
210
232
569
225
318
225
318
288
225
Gas
flow rate,
d scf rn
6,350
2,100
2,700
3,450
3,290
9,800
3,100
2,400
14,000
2,650
5,050
6,200
1 5,000
2,620
10,000
10,000
37,000
9,000
6,700
6,400
10,000
10,400
9,090
5,000
1 1 ,040
7,900
10,900
5,600
5,900
Ratio of flow rate
to production,
dscfm
T ADP/day
16
2.3
7.5
11
8.5
33
8.1
9.5
28
8.6
19
38
54
19
77
77
45
18
21
21
48
45
16
22
35
35
34
19
26
Partial late concentration,
gr/dscf
as Na2O
Inlet
1.0
0.39
0.41
-
-
0.23
_
-
-
-
0.72
-
-
_
-
—
1.6
4.73
-
-
-
—
1.65
-
—
-
-
5.94
-
Outlet
0.16
0.11
0.094
0,09
0.14
0.06
0.3
0.3
0.1
0.35
0.16
0.10
0.09
0.26
0.07
0.07
0.13
0.31
0.33
0.33
0.15
0.17
0.48
0.38
0.26
0.29
0.30
0.58
0.44
Emission rate,
Ib/T ADP
as Na20
Inlet
3.27
0.19
0.64
-
-
1.58
-
-
-
-
2.82
-
-
-
-
—
14.8
17.8
-
-
-
—
5.43
-
—
-
-
23.7
-
Outlet
0.05
0.052
0.15
0.21
0.25
0.41
0.50
0.58
0.57
0.62
0.63
0.77
0.99
1.01
1.10
1.10
—
1.2
1.44
1.46
1.47
1.57
1.58
1.74
1.86
2.09
2.11
2.3
2.38
90
PULP AND PAPER INDUSTRY EMISSIONS
-------
•o tn
X cs
•H in
5 ra
a.
a
S
o' o" do" o d d d d o" d o' d o' d o <-' co'
co
CSI
<— «~ «— oo
CO
*
Z
<
UJ
CO
o
cc
u.
CO
Z
g
CO
UJ
CO
o
UJ
CO
CO
3)2S2
CO
0
I
v>
co
o
00
CM
I
LO! II 1
oloool IOLOOOO
oo CM in ro
Icn
CNIO
s
«-«-«-CN'-»-CMCN
•-«-
coooo
* „- E
O M **-
.•E
C
D
§OOOOOOOOOOOOOOOOO
oOLnomoooocNooooooo
COCO»-OCOCDO^l-COCOOOC35OOOO
o" t-" Tt m" LO" a>" •* CD" co" o" in o" «-* oo" t" ID" •<«•" co"
i- «- >- CM ^t-^
-------
Table A-4. PARTICULATE EMISSIONS FROM LIME KILNS
Unit
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Pulp production
rate,
T ADP/day
294
520
498
804
793
498
294
308
422
338
900
706
475
350
700
950
735
280
593
873
225
420
200
270
262
560
540
437
355
550
320
686
262
363
540
490
309
Exit gas
flow rate,
dscfm
10,600
13,800
12,000
24,000
1 5,300
17,200
8,700
12,000
9,530
9,530
11,700
6,300
19,200
10,000
18,700
23,800
7,000
10,000
18,600
13,700
10,000
11,500
11,300
7,500
8,270
14,500
1 5,000
11,700
15,810
27,800
5,800
32,000
8,520
22,200
18,000
15,502
5,100
Ratio of flow rate
to production,
dscfm
T ADP/day
36
28
24
30
19
35
30
39
25
28
13
9
40
29
28
25
9
36
31
16
44
27
56
28
32
26
28
27
45
50
18
47
32
61
33
32
16
Exit
concentration,
gr/dscf
0.01
0.02
0.02
0.02
0.03
0.02
0.02
0.05
0.09
0.09
0.21
0.31
0.07
0.10
0.12
0.13
0.36
0.10
0.14
0.28
0.11
0.19
0.11
0.22
0.22
0.29
0.3
0.37
0.23
0.21
1.05
0.27
0.40
0.22
0.4
0.43
0.83
Emission
rate,
Ib/T ADP
0.074
0.019
0.10
0.12
0.12
0.14
0.12
0.40
0.42
0.52
0.55
0.57
0.583
0.594
0.660
0.67
0.71
0.73
0.88
0.90
0.97
1.07
1.28
1.265
1.438
1.552
1.727
2.06
2.12
2.15
2.28
2.57
2.67
2.73
2.75
2.79
2.82
92
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table A-4. PARTICULATE EMISSIONS FROM LIME KILNS (Con't)
Unit
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
Pulp production
rate,
T ADP/day
509
490
486
437
551
1100
540
642
473
315
275
275
335
347
232
465
298
86
315
290
500
232
747
166
284
400
473
284
232
Exit gas
flow rate,
dscfm
4,830
12,300
4,830
13,669
9,800
29,500
21,500
19,980
22,900
12,300
18,700
1 8,700
14,750
4,830
8,500
26,400
11,000
15,300
23,000
29,200
1 2,600
17,080
13,900
10,000
10,000
12,100
28,170
15,100
12,000
Ratio of flow rate
to production,
dscfm
T ADP/day
9
25
10
38
18
27
40
31
48
39
68
68
44
14
37
57
37
180
73
100
25
74
19
84
49
30
60
53
52
Exit
concentration,
gr/dscf
1.5
0.58
1.5
0.49
0.88
0.59
0.4
0.53
0.38
0.48
0.29
0.29
0.46
1.5
0.57
0.37
0.59
0.13
0.34
0.25
1.0
0.38
1.56
0.51
0.98
1.2
0.65
1.28
4.0
Emission
rate,
Ib/T ADP
2.93
3.00
3.07
3.14
3.22
3.26
3.28
3.38
3.78
3.86
4.06
4.06
4.17
4.30
4.30
4.33
4.86
4.76
5.16
5.18
5.19
5.75
5.97
6.33
7.11
7.48
7.96
14.02
53.1
Appendix A
93
-------
'
1
CD
«-'3-ioco«^CMCNCNicO'-oco^fCQcocoG>coir>ro'*
do dciociddddddddddd'--'-•«-'cri'd-
io
ir)coroioo>Jooooco^r«-oi^oococococNi«ooo
*~ ' «-»^^r-r-«-oocO'-'-'-oomir5co^a)rs»
CiOOLO«— COOCO
t— t— CM CO CN LO 00
CM
CO
Z
UJ
O
DC
u.
CO
CO
CO
O
loolcNCN-lcDCN
^ CN
1 1 1 1
o
l lininl I liol 1 1
3
•o
1
o
V)
IcooroLOoo'*
t- CM
m^-
O to LD O O O MOO
T- «- co co "". f cooomioojCMOOcDOOco
co K»-«-t-ininmoocoooo
t- i-.-CMOOCNUICO
UJ
CO
o
UJ
CO
o
in
«?E
8 a
.oOr-^-mini^o.iows. .0.0.
II '-I I 10 t- CO »- 1 1 1 00 1 00 1
™ f-
; I ct
o t3 E 5
£ -ilg
^ 2 re <
.2 a i-
« 8
cc
OOCOOOOOOOCOCNCO
^-Ot^CNI
1=
I I |u.
i- 8. 3 °
coocsooooocoocooor-ooooooomo
minr^Lnr-r^i^cMininooooromLncocomcomcoco
Is
^D ^D ^D C3
^~ O tO O
§ 1
? afS:
2 a
£ "2
od
94
PULP AND PAPER INDUSTRY EMISSIONS
-------
APPENDIX B: SAMPLING AND ANALYTICAL
PROCEDURES AND EQUIPMENT
SUMMARY OF PROCEDURES
A variety of sampling and analytical methods were used by the mills
replying to the questionnaires, by the National Council of the Paper
Industry for Air and Stream Improvement (NCASI) in their special studies,
and by the Environmental Protection Agency (EPA) in their field
investigations. In the case of gas analysis, significant improvement in
precision and sensitivity of gas analysis occurred over the approximate 4-
year period during which the data were collected. In the case of particulate
analysis, varying efficiencies of the media used to collect particulate matter
and the incorporation of a gas washing and condensation step may account
for differences in amount of materials referred to as particulate matter in
this report.
Gas Analysis
Questionnaire Data
Essentially all the gas emission data reported in the questionnaires were
developed during a period when the analytical procedures available for
hydrogen sulfide included its collection in (1) a strong caustic solution and
subsequent analysis for sulfide ion by potentiometric titration, or (2) zinc
acetate and subsequent analysis for sulfide using a methelene blue
colorometric procedure, or (3) cadmium chloride and subsequent analysis
for sulfide ion by iodometric titration. The latter two procedures are
described in References 26 and 27. Methyl mercaptan was also collected in
cadmium chloride and subsequently analyzed by iodometric titration after
separation of the mercaptide precipitate from the cadmium solution.2The
method used for alkyl sulfides and disulfides involved collection in benzene
and subsequent titrimetric and spectrophotometric measurement of the
alky] sulfides present. A modification of the West-Gaeke method 28 was
commonly used during this period for sulfur dioxide determination.
Special Studies
All special studies on recovery furnace systems were conducted using an
instrumental coulometric titration (Barton titrator). The procedure is fully
described in the basic references cited in the text. Special studies on
miscellaneous sources were conducted using a flame photometric
chromatographic detector, a flame ionization chromatographic detector,
95
-------
and continuous analysis of the sulfur dioxide formed upon oxidation of the
reduced sulfur compounds present. The sampling, handling, and
instrumental methods are described in detail in the basic references cited in
the text.
EPA Field Investigation
A special mobile field sampling laboratory developed by EPA was used
for these investigations. This facility and the sampling and analytical
procedures used are described later in this appendix.
Particulate Sampling and Analysis
It was common practice during the period that particulate emission data
reported in the questionnaires were being generated to use the in-stack
alundum thimble collection procedure described in Reference 29 and other
sources. In their field investigations, EPA employed procedures, described
later in this appendix, developed for compliance testing for new stationary
sources and published in the Federal Register.30 The relative particle
collection efficiency of the alundum thimble method and the EPA method
has not been fully established, but the limited data available suggest that
the EPA procedure has the greater efficiency.
EPA MOBILE FIELD SAMPLING LABORATORY
A mobile source sampling laboratory was developed by the Emission
Measurement Branch of the Office of Air Quality Planning and Standards,
Environmental Protection Agency. The equipment was mounted in a self-
propelled vehicle to facilitate sampling at various locations.
A unique continuous gas sampling system was devised for the mobile
laboratory. The gas sample from the source is filtered to remove
particulates and delivered through an electrically heated Teflon sampling
line at a temperature above its dew point. At the mobile laboratory, the
sample is passed through a dynamic dilution system where purified air is
used as the diluent. The dilution system provides simultaneous dilution
levels up to six orders of magnitude.'Teflon components are utilized where
possible; and where stainless steel is used, the quantities of gas handled
overwhelm the minor surface adsorption effects. The appropriately diluted
samples are delivered to various instruments.
Particulate samples are obtained using a modified sample train
developed by EPA. The gas meters and draft gauges are located within the
mobile laboratory, while the filters, cyclone, and impingers are located with
the particulate probe at the sampling site. Two complete sampling trains
can be used to simultaneously sample the inlet and outlet of a control
device. Pitot tube readings are recorded on a multipoint recorder after
being transduced by a transmitting differential pressure manometer.
96 PU LP AND PAPER INDUSTRY EMISSIONS
-------
Remote temperature readings are obtained by thermocouples and are also
recorded on a multipoint recorder.
In addition to two electrically traced Teflon gas sampling lines, four 250-
foot umbilical cables were prepared. The umbilical cables were fabricated
by hand-wrapping the necessary power lines, thermocouple leads, pitot
lines, communication lines, and sample lines. All the umbilical cables are
reel-mounted to facilitate handling. A voice powered telephone system with
electrical buzzers provides communications between all sampling points
and the mobile laboratory.
Gas Analysis
Instrumentation
The continuous gas monitoring instrumentation provides on-site
analysis for oxygen, carbon monoxide, carbon dioxide, hydrocarbons, and
various sulfur compounds. Dynamic dilution permits the use of sensitive
laboratory instruments for actual source testing while avoiding problems
associated with condensation of moisture in the sample. Instruments
installed included two gas chromatographs, a total sulfur analyzer, a
Barton titrator, an oxygen analyzer, a carbon dioxide analyzer, and a
carbon monoxide analyzer.
Oxygen analysis is performed continuously on the process gas stream by
a Beckman Model F3 oxygen analyzer. The F3 analyzer measures the mag-
netic susceptibility of oxygen. Three ranges were provided: 0 to 5, 0 to 10,
and 0 to 25 volume percent. The instrument accuracy is +1 percent of full
scale.
Carbon monoxide and carbon dioxide are analyzed by Beckman
infrared analysis instruments. A Beckman Model IR 315 is used for
carbon dioxide with three ranges: 0 to 1, 0 to 10, and 0 to 20 volume
percent. Carbon monoxide is analyzed by a Beckman Model IR 315 A with
ranges of 0 to 500 parts per million (ppm), 0 to 2 percent, and 0 to 10
percent by volume. Both instruments have accuracies of +1 percent full
scale and sensitivities of 0.5 percent full scale. The instruments have three
panel-mounted sections: an analyzer section, an amplifier/control section,
and a constant-voltage transformer.
Total sulfur in the process stream is analyzed by a Melpar sulfur dioxide
analyzer. The instrument oxidizes sulfur compounds in a hydrogen flame
and uses a flame photometric detector to measure the sulfur dioxide. The
output of the flame photometric detector is recorded on a log/linear
recorder. Sensitivity of the instrument is 0.01 ppm, and response is linear
between 0.01 and 10 ppm. The dilution system extends the useful range of
the Melpar to allow analysis of much higher concentrations. Total sulfur is
reported as sulfur dioxide concentrations in parts per million by volume.
The measurement provides a basis for determining material balances, peak
sulfur loadings in the source, and checks on the chromatographs.
Appendix B 97
-------
The gas chromatograph systems installed in the mobile laboratory were
developed by the Division of Chemistry and Physics, National
Environmental Research Center, Research Triangle Park, North Carolina.
These systems are described in detail later in this appendix.
Dilution System 31
This section describes the dynamic serial dilution system developed for
the mobile sampling van, and its function in the dilution of stack gases.
Purpose —Stack sampling of gaseous pollutants within a given industry is
often complicated by widely varying moisture and pollutant concentrations.
The dewpoint temperature is usually greater than the operating
temperature of many instruments and consequently must be lowered. In
most cases, this has been accomplished with a scrubber or condenser. In
order to follow large fluctuations in a pollutant's concentration, individual
instruments may have analytical and/or electronic parts in duplicate or
even triplicate.
The term "sample conditioning" refers to the process by which the stack
gas is rendered acceptable for analysis. Normally this term would include
only the removal of moisture and particulates from the sample. In this case,
however, the design of the van's entire sample-conditioning system was
primarily governed by the need to selectively and continuously identify
various gaseous pollutants with instruments whose analytical capabilities
are limited to concentrations more typically encountered in ambient air.
This situation meant that some means of diluting the stack gas would be
needed. If dry air was used to dilute the sample and lower the moisture
content, the sample could be kept above its dewpoint until it had been
analyzed and the moisture problem would be solved. Furthermore, it would
be solved without using a mechanism that could remove or alter some of the
constituents in question.
Since the van would be used later on sources in other types of industry,
the fact that the moisture problem could be nullified in a manner that
would be almost universally applicable made the dilution concept doubly
attractive.
Description—Figure B-l shows the relationship of the dilution system to the
other components of the sampling system, and Figure B-2 shows the actual
piping arrangement. In view of the quantity and type of analytical
instruments being employed, and in order to keep the entire operation as
mobile as possible, it was necessary to permanently mount both the
instruments and the dilution system in the van. Consequently, some means
of transporting the sampled portion of the stack gas to the van without
allowing condensation to form was required. The sample first passes
through a heated probe containing glass wool to remove coarse particulates.
It then flows through a heated millipore filter and thence into a 3/16-inch-
98 PULP AND PAPER INDUSTRY EMISSIONS
-------
inside-diameter (i.d.) Teflon sample line 250 feet in length. The sample line
temperature, which is monitored as the line enters the van, is maintained at
approximately 20°F above the dewpoint of the gas by varying the voltage
applied to insulated nichrome wires that have been encased in the sample
line.
A stainless-steel bellows pump, which is similarly heated, is one source
of sample line vacuum. Stack gas exiting from this pump tees to the first
stage of the serial dilution system and to an unrestricted atmospheric vent
(denoted as "V" on Figure B-2).
SAMPLE
FILTRATION
AND
TRANSFER
STACK
SAMPLE
DILUTION
I
DILUENT
AIR
CALIBRATION
SYSTEM
ANALYTICAL
INSTRUMENTS
VAN
Figure B-1. Arrangement of gas-sampling system of mobile
sampling van.
Appendix B
99
-------
100
PULP AND PAPER INDUSTRY EMISSIONS
-------
The output of a diaphragm pump that is connected in parallel is
monitored to determine the sample residence time in the sample line. This
time is kept at a minimum in order to reduce the possible occurrence of
interactions between the various constituents. A vacuum gauge located in
front of the diaphragm pump is checked periodically to see that the
pressure drop in the line does not exceed the value beyond which the
bellows pump will no longer deliver the quantity of stack gas required by
the dilution system.
The dilution system produces stack gas diluted to six different integral
powers often. Although most sources require use of no more than the first
two or three stages, there are occasions when high pollutant concentrations
will warrant use of the higher dilutions.
Each stage of dilution consists principally of a Komhyr model A-150
positive displacement pump, which meters the gas being diluted. Displace-
ment is approximately 150 cubic centimeters per minute (cm' /min) and is
constant within^ 1.5 percent of mean value. Since both the inlet and outlet
of this pump are indirectly vented to the atmosphere, the pressure drop
across it remains constant, and a constant mass is delivered per unit of
time. Air that is dried and then filtered through activated charcoal is used
as the diluent. The diluent air is teed into the pump's outlet line after
passing through a monitoring orifice. The quality of this air is periodically
verified by comparison with high-purity cylinder air. Assuming each stage
is diluting by a factor often, 90 percent of the resulting mixture then exits
the dilution system through an unrestricted atmospheric vent, with the
remaining 10 percent going to the next dilution pump. Thus the quantity of
gas available for analysis under these circumstances is approximately 1350
cmVmin for each power of dilution.
The entire dilution system is presently contained in a box about 18
inches square. Refinements in the pump design could reduce this size
considerably. The box is maintained at 250°F to prevent condensation in
the first stage of dilution and to ensure that the mass pumped per unit of
time will not change due to temperature fluctuations.
Each stage is initially adjusted to the required dilution ratio by
supplying it with gas from a sulfur dioxide permeation tube -12 calibration
system that has been adjusted to deliver a sulfur dioxide concentration of
10.0 ppm. The output of that stage is then monitored with a Melpar sulfur
dioxide analyzer, which employs a Microtek flame photometric detector 3-3
that has been previously calibrated with the permeation tube. For example,
if a dilution ratio of 10:1 is desired, the valve controlling the flow of diluent
to that stage is adjusted until a reading of 1.00 ppm of sulfur dioxide is
obtained. This procedure is then repeated for the remaining stages. Once
the system has been calibrated, a Bourdon tube pressure gauge that is
connected to a manifolded solenoid valve arrangement (Figure B-2) is used
to periodically monitor the diluent flows through their respective orifices.
Appendix B 101
-------
Accuracy and Reliability — The dilution system as described has been
operated for as long as 12 hours without developing any measurable
changes in the dilution ratios. Fluctuation in the barometric pressure dur-
ing the course of operation, however, will directly affect the mass of stack
gas being diluted per unit of time.
The accuracy of the system is principally dependent on the sensitivity of
the analytical instrument used in the calibration. Since the flame
photometric detector used for calibration has been found to be accurate to
within +1 percent for a given reading, the accuracy of the calibration of a
single dilution calibrated with this instrument is within +2 percent.
Utility—Since all dilution factors are simultaneously available in sufficient
quantity to meet the sample requirements of all the instruments from a
single stage of dilution, the instruments can be independently transferred
from one stage of dilution to another. Thus, the flame photometric detector
employed in the calibration of the dilution system can thereafter be used to
measure total sulfur, and two gas chromatographic systems that employ
identical flame photometric detectors can be used to identify various sulfur
compounds.
The use of the dynamic dilution system has several secondary benefits
with regard to the analytical instruments, the principal one being that the
extra analytical and electronic parts that were previously required in order
to extend the instruments' ranges are no longer necessary. Instead, the
instruments are switched from one dilution stage to another when the
pollutant concentration moves beyond the range of detection. Thus, it is
possible to confine the pollutant values delivered to an instrument to its
most sensitive detection range, and thereby eliminate the expendature of
time required to calibrate the instrument beyond that range.
Gas Chromatographic-Flame Photometric Systems34
The flame photometric detector (Figure B-3) measures sulfur
compounds by detecting the chemiluminescent emission from the excited
$2 molecules formed whenever sulfur compounds are burned in a
hydrogen-rich air flame. A narrow-band-pass interference filter between
the flame and the photomultiplier tube isolates a particular band of the 82
emission at 394 microns. The interference filter allows the virtual
elimination of interferences from non-sulfur-bearing constituents. The
background flame noise is also reduced by viewing only the
chemiluminescent emission above the flame.
Low-molecular-weight Sulfur Compound Detector — The gas
chromatographic-flame photometric system shown in Figure B-4 was
developed to measure low-molecular-weight sulfur compounds in kraft mill
effluents. The analyzer consists of a Varian 122 gas chromatographic oven,
a Meloy flame photometric detector, a Tracer power supply and
electrometer, and a modified Beckman 10-part sliding plate valve equipped
102 PULP AND PAPER INDUSTRY EMISSIONS
-------
HEATED EXHAUST
INTERFERENCE
FILTER
PHOTOMULTIPIER
TUBE
COLUMN
EFFLUENT INLET
Figure B-3. Flame photometric detector.
with a 10-cm3 Teflon sample loop, stripper column, and analytical column.
The function of the stripper column is to prevent heavier sulfur compounds
from reaching the analytical column by backflushing them to vent. The
analytical column is 36-foot by 0.085-inch-i.d. Teflon tubing packed with
30/60 mesh Teflon and coated with 6 grams of polyphenyl ether and 500 mg
of orthophosphoric acid. The 2-foot by 0.085-inch-i.d. stripper column is
packed with the same material as the analytical column.
Two solenoids and an industrial cam timer automatically actuate the
sampling valve at 10-minute intervals. The timing sequence to actuate the
valve for sample injection, foreflushing, and backflushing is as follows:
1. Valve energized for 1 minute while sample is injected into
stripper column and analytical column.
2. Valve de-energized for 9 minutes while stripper column back-
flushes heavy sulfur compounds to vent, analytical column
continues to be foreflushed, and sample loop is refilled.
The 10-port sample valve was modified to minimize sample-to-metal
contact, which can cause severe losses at levels below 10 ppm. The 1/16-
inch pipe to 1/8-inch tube fittings on the valve were drilled out so that the
Appendix B
103
-------
TJ
C
o
Q.
E
o
o
Q)
o
_OJ
o
E
o
o
Q)
O
i_
'CD
o
o
.c
Q.
CD
E
Q.
CO
O)
o
ra
o
O
CO
O)
iZ
104
PULP AND PAPER INDUSTRY EMISSIONS
-------
o
D-
LU
a:
•4x 10-9 A
xlO-9 A
0.18 ppm
C2H5SH
2 x 10'9 A
0.24 ppm
C3H?SH
TIME
Figure B-5. Chromatogram of low-molecular-weight sulfur com-
pounds.
Teflon lines would go through the fitting and into the body of the valve up
to the Teflon slider, thus making the valve essentially all Teflon. The
column exit was also fitted into the base of the detector to further minimize
sample-to-metal interaction.
A chromatogram of a sub-ppm mixture of sulfur compounds obtained
utilizing permeation tubes as a source of sulfur compounds is shown in
Figure B-5. Hydrogen sulfide, sulfur dioxide, methyl mercaptan, ethyl
mercaptan, dimethyl sulfide, and propyl mercaptan were resolved in 10
minutes on the 36-foot by 0.085-inch-i.d. polyphenyl ether Teflon column.
Chromatographic conditions were as follows:
1. Nitrogen carrier gas flow, 100 cmVmin.
2. Detector temperature, 105°C.
3. Exhaust temperature,! 10°C.
4. Column temperature, 50°C.
5. Flame conditions: hydrogen flow, 80cm3/min; oxygen flow 20
cm-Vmin.
High-molecular-weight Sulfur Compound Detector—The analytical
system developed for the heavier sulfur compounds is shown in Figure B-6.
A Chromatronia Teflon six-port gas sampling valve equipped with a 10-
cm3 Teflon sample loop was used since backflushing was not necessary.
The analytical column is 10-foot by 0.085-inch-i.d. Telfon tubing packed
with 30/60 mesh Teflon coated with 10 percent Triton-X 305. The lighter
sulfur compounds will emerge rapidly from this column as one peak,
followed by heavy sulfur compounds that elute separately.
Appendix B
105
-------
co
TJ
C
3
o
a.
E
o
O)
CD
O
E
o
Q)
0
T3
CD
O
as
CO
Q.
CO
L_
O)
o
*-•
co
o
CO
CO
CD
CO
0)
106
PULP AND PAPER INDUSTRY EMISSIONS
-------
4x10-8 A
0.22 ppm
0.19 ppm
C3H7SH
o
o.
C/i
_
0.17 ppm
C4H9SH
2 x IO-SA
0.21 ppm
(C 3 H y) 2 S
2 x 1Q-8A
0.18 ppm
i i i
j i i
TIME
Figure B-7. Chromatogram of high-molecular-weight sulfur
compounds.
A chromatogram of a sub-ppm mixture of high-molecular-weight sulfur
compounds is shown in Figure B-7. Butyl mercaptan, dimethyl disulfide,
dipropyl sulfide, and dibutyl sulfide were resolved in 10 minutes on the 10-
foot by 0.085-inch-i.d. Triton-X 305 column. Chromatographic conditions
were as follows:
1. Nitrogen carrier gas flow, 100 cm3/min.
2. Detector temperature, 105°C.
3. Exhaust temperature, 110°C.
4. Column temperature, 70°C.
5. Flame conditions: hydrogen flow, 80 cmVmin; oxygen flow 20
cmVmin.
Teflon permeation tubes gravimetrically calibrated according to the
procedure of O'Keeffe and Ortman35 were used as primary standards. The
permeation tube assembly is shown in the upper right of Figure B-6. The
instruments are calibrated by injecting aliquots of an air stream flowing
over the tubes into the chromatographic column. The concentration of the
sulfur compound is inversely proportional to the air flow over the
permeation tube.
Sampling Procedure — When kraft mill stack effluents are sampled
directly, special sampling techniques are required to reduce losses because
of high moisture content and a wide concentration range of the sulfur
compounds present (ppb to percent levels). The dynamic dilution system
Appendix B
107
-------
described earlier was used to bring the effluent samples into the inherently
limited dynamic range of the flame photometric detectors.
To determine if the chromatographic peaks represent the total volatile
sulfur introduced into the chromatographs, a Meloy total sulfur analyzer
was used to continuously monitor the diluted sample.
Participate Sampling and Analysis
The mobile van also includes facilities for particulate sampling and
analysis. The procedures employed are detailed in the Federal Register.30
The following sections briefly describe the major features of these
procedures.
Sampling Site and Traverse Points
When possible, a sampling site at least eight duct diameters downstream
and two diameters upstream from any flow disturbance such as a bend,
expansion, contraction, or visible flame is selected. For a rectangular cross
section, an equivalent diameter, equal to 2(length) (width)/(length+width),
is calculated. For a sample site meeting these criteria, a minimum of 12 tra-
verse points are sampled. For circular stacks, the traverse points are located
on perpendicular diameters; for rectangular stacks, on the centroids of
equal rectangular areas.
In some cases, it is necessary to sample at points that do not meet the
criteria mentioned in the preceding paragraph. In such cases, the number
of traverse points sampled must be increased to ensure a representative
sample. Figure B-8 shows the minimum number of traverse points for
various distances from a disturbance.
Sampling Train
The EPA particulate sampling train is shown in Figure B-9. Stack gas
velocity is determined from gas density and from velocity head as measured
by a Type S (Stauscheibe or reverse type) pitot tube. The sampling probe is
Pyrex (or stainless steel, if necessary) and employs a heating system
capable of maintaining a temperature of 250°F. The particulate sample is
collected by glass fiber filters and impingers. The first two impingers
contain water, the third is empty, and the forth contains silica gel. The
metering system includes a vacuum gauge, leak-free pump, thermometers,
dry gas meter, and related equipment as required to maintain an isokinetic
sampling rate and to determine sample volume.
Analysis
The following samples, are placed in individual containers for analysis:
1. Filter.
2. Loose particulate matter and acetone washings from all sample-
exposed surfaces prior to the filter.
108 PULP AND PAPER INDUSTRY EMISSIONS
-------
NUMBER OF DUCT DIAMETERS UPSTREAM'
(DISTANCE A)
FROM POINT OF ANY TYPE OF
DISTURBANCE (BEND, EXPANSION. CONTRACTION, ETC )
NUMBER OF DUCT DIAMETERS DOWNSTREAM'
(DISTANCE B)
Figure B-8. Minimum number of traverse points.
3. Water from the first three impingers and water washings of all
sample-exposed surfaces between the filter and the forth
impinger.
4. Silica gel from the forth impinger.
5. Acetone washing of all sample-exposed surfaces between the filter
and the forth impinger.
After appropriate sample conditioning, as detailed in the referenced
procedure, the weight of particulate in each sample and the total
particulate weight are determined.
The total particulate weight and the sample gas volume, adjusted to
standard conditions (70°F and "29.92 inches of mercury), are used to
calculate the sample concentration by two methods, referred to as the
sample concentration method and the ratio of area method. If the
concentrations determined by the two methods fall within acceptable limits,
the concentration is reported as the average of the two values. If the
concentrations do not fall within acceptable limits, the test is repeated.
Appendix B
109
-------
ro
a.
to
03
CD
CC
CL
cn
en
Q)
110
PULP AND PAPER INDUSTRY EMISSIONS
-------
APPENDIX C: ODOR SURVEY
INTRODUCTION
One of the original objectives of this study was to make a comprehensive
organoleptic assessment of the odorous emissions from major sources in a
pulp mill and to relate the organoleptic measurements to the measured
concentrations of the individual compounds in the emissions. This survey
was to be conducted by an experienced Environmental Protection Agency
(EPA) odor investigator, with a carefully selected and trained panel of odor
observers. However, because of the pressure of other assignments, the odor
investigator was unable to do more than initiate the odor survey. A greatly
curtailed odor survey was then carried out by the principal EPA
investigator. The restrictions imposed on the investigation resulted in data
with many inconsistencies. The data are reported here to illustrate a
method of odor measurement and for the limited significance that may be
attached to the results.
EQUIPMENT AND PROCEDURES
The gas sampling system in the mobile van contains a tap for collecting
odor samples. At this tap, the sample was diluted approximately 10:1.
Samples were drawn through a hypodermic needle into 100-milliliter (ml)
syringes. The samples were subsequently further diluted with odorless air in
glass syringes and presented to odor panelists directly from the syringes.
The method is basically that of ASTM Method D 1391-57.36 The odor
threshold is reported as the number of dilutions at which 50 percent of the
panel can detect the presence of an odor. Panels were composed of mill
personnel and EPA test crew members. Panels were composed of three to
six members.
Odor panel members were screened by the use of hydrogen sulfide from
a lecture bottle. A 100-ml syringe was filled from the lecture bottle and
injected with air into a 2-cubic-foot plastic carboy, producing a dilution
ratio of 567. Samples from the carboy were then diluted by multiples of 10
by using successive syringes. The results of the screening tests at Mill B are
given in Table C-1. The dilution ratios are from the carboy.
The response of some of the panelists was rather erratic, but there was
no opportunity to obtain any other panelists.
Ill
-------
Table C-1. ODOR PANEL SCREENING TESTS
Panel
member
P
H
F
A
0
E
Dilution ratio (trial number)
B (2>
_
—
—
_
-t-
-
B (4)
_
—
—
_
—
-
108 (5)
+
—
—
_
_
-
10« (7)
+
_
_
-
107 (1)
+
_
_
-
107 (6)
+
_
_
-
106 (3)
+
+
+
+
+
+
a+ odor detected; —, no odor detected; B, blank consisting of pure air.
Odor samples were taken at the times when samples for gas analysis
were being run. Because of the limited time the odor panel was available for
evaluation of samples, an odor sample was not taken for each sample
analyzed. However, at least one odor sample was evaluated from each point
at which a series of samples were analyzed for components. Samples were
taken at various times during the day. They were stored in the dark until
presented to the odor panel. It was felt that the panelists should be in
nonodorous surroundings for at least 2 hours before a panel was held. Since
the panelists were not available in the evening, it was necessary to postpone
the panel sessions until the following morning. Some tests with a limited
panel indicated no deterioration on storing overnight.
RESULTS
The response of the odor panel to one sample is presented in Table C-2.
The percent response at each dilution was plotted on log probability paper,
as shown in Figure C-1, to estimate the dilution level for 50 percent panel
response. The odor dilution thresholds of the other odor samples were
obtained in the same manner. The results are tabulated in Table C-3 and
Table C-4. The total reduced sulfur (TRS) values are the sum of the reduced
sulfur compound readings obtained from the gas chromatograph. The TRS
concentrations at the dilution threshold are obtained by dividing the stack
TRS concentrations by the dilution threshold.
Table C-2. EXAMPLE OF ODOR PANEL RESPONSE
Dilution ratio (trial number)
member
0
P
E
F
Percent
response
BI2)
+
B(9)
-
10s 131
+
105 (6)
-
10*111
+
10* (5l
-f
-f
104 (7)
+
104(10]
+
103 (41
+
103 (81
+
io2m>
-f
+
+
+
12 12 67 75 100
a+, odor detected, —, no odor detected, B, blank consisting of pure air
112
PULP AND PAPER INDUSTRY EMISSIONS
-------
Table C-3. ODOR PANEL RESULTS, MILL B
Sample
6B
6C
6D
106
9B
13A
15
11A
12A
12C
16A
16B
16C
17A
17B
18A
19
20
21
22
23
24
Source
Recovery furnace
Electrostatic precipitator inlet
Electrostatic precipitator inlet
Electrostatic precipitator inlet
Electrostatic precipitator inlet
Electrostatic precipitator outlet
Electrostatic precipitator outlet
Electrostatic precipitator outlet
Smelt tank
Smelt tank
Smelt tank
Washer vent
Washer vent
Washer vent
Lime kiln
Scrubber, outlet
Scrubber, outlet
Scrubber, inlet
Scrubber, inlet
Black liquid oxidation
Multiple effect evaporator vents
Scrubber, outlet
Scrubber, inlet
Scrubber, inlet
Vapor line to lime kiln
Odor dilution
threshold3
5x 103
3x104
3x 104
5x 103
1 x 103
1x104
1x104
5x 103
5x104
5x 103
5x 103
7x 703
7x 103
1 x 102
1 x 102
1 x JO2
1 x 102
IxlO4
5x 109
3x I07
1 x 108
1x1010
TRS emission,
ppm
4.1
3.4
4.9
4.3
5.9
6.2
3.1
3.3
8.1
3.8
14.5
10.1
16.0
7.2
11.1
15.6
31.2
13.6
14,000
14,000
36,000
46,600
TRS concentration
at odor
threshold, ppb
0.8
0.1
0.2
0.8
5.9
0.6
0.3
0.7
0.2
0.8
2.9
1.4
2.3
70
110
150
310
1.5
0.003
0.5
0.4
0.005
Dilution level detectable to 50 percent of the panel.
EVALUATION
An examination of the odor levels reported in Table 2 of the main text
for the compounds composing TRS would lead one to expect that the odor
threshold level of TRS would be in the range of 1 to 5 parts per billion.
Many of the results in Table C-3 do fall in or near this range, but those in
Table C-4 are generally well below it. The greatest deviation is in samples
from lime kilns, and the deviation is fairly consistent. No reasons can be
advanced for this deviation from expected levels.
Odor measurements in general are not capable of a high degree of
precision. Chromatographic measurements of gaseous components, on the
other hand, are capable of a considerable degree of precision. The
deviations in the calculated TRS concentrations at the odor dilution
threshold are therefore believed to be due to imprecision in the odor
measurements. The fact that the TRS mixtures from the different sources
are not the same could possibly account for some deviations. It is not
believed, however, that it could account for the large deviations.
Appendix C
113
-------
Table C-4. ODOR PANEL RESULTS, MILL C
Sample
1
2
3
4
12
5
6
14
7
g
10
11
13
15
16
17
18
19
20
Source
Recovery furnace
Direct contact evaporator inlet
Direct contact evaporator inlet
Direct contact evaporator outlet
Direct contact evaporator outlet
Direct contact evaporator outlet
Electrostatic precipitator outlet
Electrostatic precipitator outlet
Electrostatic precipitator outlet
Scrubber outlet
Scrubber outlet
Scrubber outlet
Smelt tank
Black liquor oxidation
Lime kiln
Scrubber outlet
Scrubber inlet
Scrubber outlet
Brown stock washer
Knotter vent
Brown stock seal tank
Odor dilution
threshold a
3x 102
1 x 103
4x 104
4x 104
1.4 x 104
4x 104
4x 103
5x103
3x104
9x 104
2x104
2x 102
2.6 x 105
2.5 x 103
4x 102
2.5 x 102
6x 102
6x 102
1.7x 104
TRS emissions,
ppm
0.26
0.28
0.26
0.26
1.59
0.10
0.10
1.01
0.11
1.3
2.74
2.0
TRS concentration
at odor
threshold, ppb
0.87
0.28
0.0065
0.0065
0.11
0.0025
0.025
0.2
0.0037
0.015
0.14
10
7.9 !, 0.03
85.4
33
59.7
2.65
12.9
626
34
82
240
4.5
21
38
Dilution level detectable to 50 percent of the panel.
LU
>
ODOR DILUTION THRESHOLD
1 2 3 5 10 20 40 50 60 70 80 90 99
PERCENT OF PANEL DETECTING ODOR
Figure C-1. Example of method of estimating dilution level for 50
percent response.
114
PULP AND PAPER INDUSTRY EMISSIONS
-------
The results of this survey illustrate the care that must be exercised in
organizing and conducting an odor survey. It is believed that the greatest
inadequacy of this study was the lack of opportunity to screen a
considerable number of people in order to select odor panelists who have a
consistent response to odors. It appears that mill personnel and people who
are at the mill for testing purposes do not make good panelists because they
may suffer from "odor fatigue." Securing an adequate panel is the greatest
difficulty in conducting an odor survey.
Appendix C 115
-------
-------
REFERENCES
1. Factors Affecting Emission of Odorous Reduced Sulfur Compounds
from Miscellaneous Kraft Process Sources. National Council of Paper
Industry for Air and Stream Improvement, Inc. New York, N.Y. Technical
Bulletin No. 60. March 1972.
2. MacDonald, R.G. and J.N. Franklin. Pulp and Paper Manufacture;
Volume 1, The Pulping of Wood. New York. McGraw-Hill Book Company.
1969.
3. Lee, H.N., et al. The Manufacture of Pulp and Paper. New York.
McGraw-Hill Book Company. 1937.
4. Control of Atmospheric Emissions in the Wood Pulping Industry.
Environmental Engineering, Inc., Gainesville, Fla., and J.E. Sirrine
Company. Gainesville, S.C. Contract No. CPA 22-69-18. March 1970.
5. The Statistics of Paper 1971. American Paper Institute, New York,
N.Y. 1971.
6. Paper and Paperboard Capacity 1966-1969. American Paper Institute,
New York, N.Y. 1969.
7. Hansen, G.A. Odor and Fallout Control in a Kraft Pulp Mill. J. Air
Pollut. Contr. Assoc. L2_(9):409-413, September 1962.
8. Leonardos, G. et al. Odor Threshold Determinations of 53 Odorant
Chemicals. J. Air Pollut. Contr. Assoc. 19(2):91, February 1969.
9. Wilby, F. V. Variation in Recognition Odor Threshold of a Panel. J. Air
Pollut. Contr. Assoc. ,19(2):, February 1969.
117
-------
10. Blosser, R.D. and B.H. Cooper. Current Practices in Thermal
Oxidation of Noncondensable Gases in the Kraft Industry. National
Council of the Paper Industry for Air and Steam Improvement, Inc. New
York, N.Y. Atmospheric Pollution Technical Bulletin No. 34. November
1967.
11. Morrison, J.L. Collection and Combustion of Noncondensable
Digester and Evaporator Gases. Tappi. 5_2_(12):2300-2301, December 1969.
12. Britt, K.W. Handbook of Pulp and Paper Technology. New York.
Reinhold Publishing Corporation. 1964.
13. Mattesan, M.J., L.N. Johanson, and J.L. McCarthy. SEKOR II:
Steam Stripping of Volatile Organic Substances from Kraft Pulp Mill
Effluent Stream. Tappi. 30(2):86, February 1967.
14. Walther, I.E. and H.R. Amberg. A Positive Air Quality Program at a
New Kraft Mill. J. Air Pollut. Contr. Assoc. 2Q(1):9-18, January 1970.
15. Factors Affecting Reduced Sulfur Emissions from the Kraft Recovery
Furnace and Direct Contact Evaporator. National Council of the Paper
Industry for Air and Stream Improvement, Inc. New York, N.Y. Technical
Bulletin No. 44. December 1969.
16. Blue, J.D. and W.F. Llewellyn. Operating Experience of a Recovery
System for Odor Control. Tappi. 54^(7): 1143-1147, July 1971.
17. Thoen, G.N., G.G. DeHaas, R.G. Tallent, and A.S. Davis. The Effect
of Combustion Variables on the Release of Odorous Sulfur Compounds
from a Kraft Recovery Furnace. Tappi. .5_i(8):329-333, August 1968.
18. Walther, J.E. and H.R. Amberg. The Role of Direct Contact
Evaporator in Controlling Kraft Mill Emissions. Paper Trade J. 155(8):89.
February 22, 1971.
19. Murray, F.E. and H.B. Rayner. Emission of Hydrogen Sulfide from
Kraft Black Liquor during Direct Contact Evaporation. Tappi. 48(10):588.
October 1965.
118 PULP AND PAPER INDUSTRY EMISSIONS
-------
20. Tobias, R.C., G.C. Robertson, D.E. Schwabauer, and B. Dickey. A
Non-conventional Strong Black Liquor Secondary Oxidation Treatment.
Western Kraft Corporation, Hainesville, Kentucky. (Presented at National
Council of the Paper Industry for Air and Stream Improvement, Inc.,
Central and Lake States Regional Meeting. Detroit, Mich. September 24,
1970.)
21. Survey of Current Black Liquor Oxidation Practices in the Kraft
Industry. National Council of the Paper Industry for Air and Stream
Improvement, Inc. New York, N.Y. Technical Bulletin No. 39. December
1968.
22. Walther, I.E. and H.R. Amberg. A Positive Air Quality Control
Program at a New Kraft Mill. J. Air Pollut. Contr. Assoc. 20_(1):9-18,
January 1970.
23. Galeano, S.F. and B.M. Dillard. Process Modifications for Air
Pollution Control in Neutral Sulphite Semi-Chemical Mills. Owens-Illinois,
Inc. Toledo, Ohio. (Presented at 64th Air Pollution Control Agency
Meeting.)
24. Durham, J.F. Trip Report. Environmental Protection Agency.
Research Triangle Park, N.C. January 26, 1971.
25. Nachbar, R.H. Electrostatic Precipitation of Ash from a Coal and
Bark Fired Boiler. Westvaco Corporation, Covington, Virginia. (Presented
at the Fall Meeting, South Atlantic Section, Air Pollution Control
Association. Myrtle Beach, South Carolina. September 11, 1970.)
26. Manual for the Sampling and Analyasis of Kraft Mill Recovery Stack
Gases. National Council of the Paper Industry for Air and Stream
Improvement, Inc. New York, N.Y. Technical Bulletin No. 14. October
I960.
27. A Method of Measuring the Concentration of Sulfur Compounds in
Process Gas Streams. National Council of the Paper Industry for Air and
Stream Improvement, Inc. New York, N.Y. Technical Bulletin No. 28.
December 1965.
References 119
-------
28. West, P.W. and G.C. Gaeke. Fixation of Sulfur Dioxide as
Sulfitomercurate III and Subsequent Colorimetric Determination. Anal.
Chem. 22:1816-1819, 1956.
29. ASME Power Test ("ode 27, Determining Dust Concentration in a
Gas Stream. American Society of Mechanical Engineers. New York. 1957.
30. Environmental Protection Agency. Proposed Standards of
Performance for New Stationary Sources; Appendix — Test Methods,
Methods 1 and 5. Federal Register. 42 CFR Part 466. Vol. 36, No. 159.
August 17, 1971.
31. Grimley, K.W., W.S. Smith, and R.M. Martin. The Use of a Dynamic
Dilution System in the Conditioning of Gases for Automated Analysis by a
Mobile Sampling Van. Environmental Protection Agency, Research
Triangle Park, N.C. (Presented at the 61st Annual Meeting of the Air
Pollution Control Association. St. Louis, Mo. June 1970.)
32. Scaringelli, G.P., S.A. Frey, and B.E. Saltzman. Evaluation of Teflon
Permeation Tubes for Use with Sulfur Dioxide. Amer. Ind. Hyg. Assoc. J.
22, May-June 1967.
33. Stevens, R.L., A.E. O'Keeffe, and G.C. Ortman. Absolute Calibration
of a Flame Photometric Detector to Volatile Sulfur Compounds at Sub-
Part-Per-Million Levels. Environ. Sci. Technol. 3_:7, July 1969.
34. Mulik, J.D., R.K. Stevens, and R. Baumgardner. An Analytical
System Designed to Measure Multiple Malodorous Compounds Related
to Kraft Mill Activities. Environmental Protection Agency, Research
Triangle Park, N.C. (Presented at Tappi Water and Air Conference.
Boston, Mass. April 1971.)
35. O'Keeffe, A.E. and G.C. Ortman. Anal. Chem. 38_:760, 1966.
36. ASTM Test Method D 1391-57, Measurement of Odor in
Atmospheres (Dilution Method). In: 1966 Book of ASTM Standards, Part
23. American Society for Testing and Materials. Philadelphia, Pa. 1966. p.
728-731.
^r u. s. GOVERNMENT PRINTING OFFICE: 1973—747775/301
120 PULP AND PAPER INDUSTRY EMISSIONS
-------
-------
1 t I
Q « xi
1 * 2|*
? S ?fi*l
o : 5 • • 5
1? \\l\
if II!!
^ z • * SO
•* 5 * s 2 -
1 " otyi
•*>-it sr- s ? N 3 5 >
Tl^-TT- <- 3 3 §• »
U1< H J3 /> m - Z
n^fn^-yj Q
£>-n 3*
a_3 o c_
< •> T _}
X^^ 13
no
H^rcn c_
»-<-< J
~*s> z.
>r- 3 n
z n
-4 m
™' n
^:T1
-~>n
•
j>
o m
• "0
IN} J>
-*J
-j r-
i— >
i-" ' - ,-<'
N 0 3
— CD
TJ^g
R-5
Q. nj ^
CD ^ 3"
0
m 3
CD ju
— . CD
3 1
O £j"
C ~
Q- 1 — l
= Q
10 .
CO
"t
m
r.
5
r
"T,
C-
^;1 m
m t ?
.T' A- 5
L.:-« O-o
<" -«,- §O
-- -J- 12
XT fi ^
Z >
O in -*O
i, J TO > m
i?
z
o
------- |