United States
Environmental Protection
Agency
Technology Transfer
v>EPA
Seminar Publication
Pollution Control
in the Forest
Products Industry
-------
-------
Technology Transfer
EPA 625/3-79-010
Seminar Publication
Pollution Control
in the Forest
Products Industry
August 1979
Environmental Research Information Center
Cincinnati OH 45268
-------
Acknowledgments
This seminar publication contains material from two pollution control seminars
for the forest industry. The U.S. Environmental Protection Agency's Environmental
Research Information Center and the Forest Products Research Society sponsored
the two seminars in Dallas, Texas (September 1977) and Portland, Oregon (April
1978) to bring together managers, engineers, and technical specialists working
in the forest products industry.
This report was prepared by Edward C. Jordan Co., Inc. in Portland, Maine. After
reviewing speaker papers, Edward C. Jordan rearranged the material presented at the
two seminars by subject material rather than by the speakers' presentation. Although
more material was extracted from some presentations than others, the sponsors
would like to thank all speakers for their excellent contributions.
The information in this publication was prepared by the following:
James L. Veatch, Attorney, Legal Branch, Enforcement Division, U.S. EPA
Region VI- David C. Bray, Technical Advisor, Air Programs, U. S. EPA Region X,
James H. Himel, Rudolph H. Moyer, and Jonathan D. Cook, GEOMET, Inc.;
John W. Robinson, Manager Environmental Services, Kirby Lumber Corporation;
David C. Junge, Ph.D., Department of Mechanical Engineering, Oregon State
University; Michael W. Guidon, Process Engineer, Georgia-Pacific Corporation;
Allan H Mick, Senior Environmental Engineer, Georgia-Pacific Corporation;
Gary Grimes, Environmental Coordinator, SWF Plywood Company; and
Clifford T. McConnell, President, McConnell Industries, Inc.
Notice
The mention of trade names or commercial products in this publication isfor illustration purposes
and does not constitute endorsement or recommendation for use by the U.S. Environmental
Protection Agency.
-------
Contents
Part I
Air Pollution Control
Page
Chapter 1. Introduction 1
Chapter 2. The Clean Air Act 3
Chapter 3. Northwest Forestry Burning 7
Chapter 4. Wood Waste Combustion 11
Chapter 5. Veneer Dryer Emissions 19
Chapter 6. Direct-Fired Drying 21
Chapter 7. Case Study: Pilot Studies for Particulate Control of Hogged Fuel
Boilers Fired With Salt Water-Stored Logs 25
List of References 31
Bibliography 33
Part II
Water Pollution Control
Chapter 8. Introduction 35
Chapter 9. The Clean Water Act 37
Chapter 10. Nonpoint Sources and Water Quality Research for Forest
Management 41
Chapter 11. Forest Chemicals 47
Chapter 12. Sediment Contributions from Southern Forest Management
Practices 53
Chapter 13. Wood Preserving 57
Chapter 14. Case Study: Wet Decking with Water Recycle 71
Chapter 15. Case Study: Wastewater Handling at Boise Cascade's Elgin,
Oregon Site 73
Chapter 16. Case Study: Waste Management at Masonite Hardboard
Mills 75
iii
-------
Chapter 17. Case Study: Wastewater Purification at Pacific Wood Treating
Corporation
List of References.
Bibliography
77
79
81
-------
List of Tables
Table 1.
Table 2.
Table 3.
Table 4.
Table 5.
Table 6.
Table 7.
Table 8.
Table 9.
Table 10.
Table 11.
Table 12.
Table 13.
Table 14.
Table 15.
Table 16.
Table 17.
Table 18.
Table 19.
Page
Prevention of Significant Deterioration Area Classifications and
Allowable Increases 4
Emission Ranges (Dry Weight Basis) 7
Statewide Emissions for Oregon and Washington 3
Paniculate Emissions from Logging Slash (Mass Basis) 9
StackSamplingData(BeforeScrubbers)KirbyLumberCorporation. . . 12
Graphic Display of the Design of the Experimental Program 14
Analysis of Heat Use and Fuel BalanceforSWF Direct-Fired Dryer 23
Veneer Dryer Emission Data Comparison 23
Hogged Fuel Boiler Steam Dryer Analysis 24
Paniculate Testing Results 26
Electro-Tube Test Data 28
Baghouse Test Data 29
Settlement Agreement Pollutants 38
Specific Chemical Substances to be Examined 39
Timber Products Processing Industry Categories 40
Major Environmentally Related Trends in Silviculture and Harvest
Management >. 42
Summary of Present Research in the Pacific Northwest to Control
Nonpoint-Source Pollution from Silviculture 45
Recommended Concentration Maxima for Silvicultural Chemicals
by Stream Class and User Group. Potable Waters Include Safety
FactorsforWildlifeahdAquaticOrganismsasWellasHumans .... 49
Guidelines for Applying Chemicals by Aircraft, and Water
Monitoring in Silvicultural Practices 51
-------
Table 20.
Table 21.
Table 22.
Table 23.
Table 24.
Table 25.
Reduction in Content of Selected Wastewater Parameters Due to
Quality of Effluent from Each Stage of a Multi- Phase Waste
Chlorophenol Concentration in Creosote Wastewater Treated
Operation Costs of Evaporative Systems at Twice Today's Energy
Costs
60
62
62
70
70
70
-------
List of Figures
Page
Figure 1. Trend of Acres Slash Burned from 1972-77 10
Figure 2. Trend of Tons Burned from 1 972-77 10
Figure 3. Trend of Tons/Acre Burned from 1972-77 10
Figure 4. Comparison of the Ash Carryover Rates in Combustion of Dry Bark
Fines vs. Dry Bark Coarse Fuel: Maximum Overfire Air 15
Figure 5. Comparison of the Ash Carryover Rates in Combustion of Dry Bark
Fines vs. Dry Bark Coarse Fuel: Maximum Underfire Air 15
Figure 6. Comparison of the Ash Carryover Rates in Combustion of Dry Bark
Fines vs. Wet Bark Fines: Maximum Underfire Air 16
Figure 7. Comparison of the Ash Carryover Rates in Combustion of Dry Bark
Fines Using Maximum Underfire and Maximum Overfire Air 16
Figure 8. Comparison of the Ash Carryover Rates in Combustion of Dry Coarse
Bark Using Maximum Underfire and Maximum Overfire Air 16
Figure 9. Plot of Exhaust Gas Opacity Versus Percent Excess Air 17
Figure 10. Plot of Exhaust Gas Opacity Versus Percent Excess Air 18
Figure 11. Transport Phenomena of Air Emissions from Veneer Dryers 19
Figure 12. Hog Fuel Boiler Layout 25
Figure 13. Particle Size Distribution HFB 26
Figure 14. Diagram of a Boulton Process Wood Treating Plant 58
Figure 15. Diagram of a Boulton Process Wood Treating Plant Using Waste
Heat for Evaporation of Wastewater 66
Figure 16. Evaporation Rate of Waterfrom a Cooling Tower as a Function of the
Difference Between Cooling Waterlnletand Outlet Temperatures at
a Constant Heat Input 66
Figure 17. Cooling Tower Evaporation Rate for Several Liquid-to-Gas
Ratios-
-------
Figure 18.
Figure 19.
Figure 20.
Spray Cooling Pond Evaporation Rate Over One Year 68
Cooling Tower Evaporation Rate Over One Year 69
Diagram of the Ultrafiltration-Reverse Osmosis Wastewater
Treatment System at Pacific Wood Treating Corp.
77
-------
Chapter 1
Part I
Introduction
Air Pollution Control
The forest products industry manages some of the
nation's most pristine land areas. At the same time, the
industry depends heavily on organic fuels—especially
wood waste—for energy needed to manufacture its
products. These manufacturing processes represent a
highly visible single influence on the air quality of
the regions in which they are located. Air quality
standards affecting the industry are strict, and in many
cases they are becoming stricter as the States and
Federal government actively pursue the goals of the
Clean Air Act.
This publication first briefly reviews the Clean Air
Act as it applies to the forest products industry. Also
presented are examples of the kinds of research and
technologies being applied by the industry to meet its
air pollution control responsibilities. :
The information presented here has been assembled
from seminar proceedings. As such, this publication
does not prescribe how individual members of the
forest industry can meet their specific air pollution
control requirements, but rather offers a perspective
on some of the issues and technologies associated
with up-to-date air quality management, as seen by
industry members, researchers, and regulatory
officials.
The forest products industry and State and Federal
regulatory agencies face the same air quality chal-
lenge, which involves technology, policy, and the
economy. The demand for wood-based products is rising
sharply. At the same time, air quality standards being
faced by industry are tightening. Where clean air
already exists, nondegradation is a key goal. Where
national air quality standards are not currently being
attained, improvements are mandated. Compliance
programs must balance concerns for clean air with
concerns for the industry's ability to produce competi-
tively an affordable product, one that plays a vital
role in the nation's economic growth.
This publication is part of a communication process
intended to foster a cooperative effort in adminis-
tering and implementing necessary air pollution control
programs. Additional reading materials relating to the
subjects covered in this publication are listed in the
Appendix.
-------
-------
Chapter 2
The Clean Air Act
Background
The Clean Air Act in its present format, amended 1977,
has as its origin the Clean Air Act of 1970. At that
time Congress established the basic programs and
philosophies that are in effect today. These programs
generally can be broken down into three areas: (1) the
attainment and maintenance of National Ambient Air
Quality Standards (NAAQS), (2) limitations on
the emissions of certain pollutants from stationary and
mobile sources, and (3) a permit program for new
sources. In conjunction with these programs, Congress
established some statutory deadlines for such things
as attainment of the NAAQS and emission levels
for new automobiles.
The primary responsibility for implementing these
and most other requirements of the Clean Air, Act
belongs to the individual states. The strategy and
mechanism for accomplishing this is the State Imple-
mentation Plan (SIP), containing regulations and pro-
cedures developed by each state and approved by
EPA for implementing the Act. Most states had approved
SIPs in effect by 1972.
By 1976 it was evident that many deadlines had
already been missed and that others were very likely to
be missed. Congress recognized that something had
to be done and began an extensive revision of the Act,
not only to relieve the deadlines, but also to deal
with the many problems and deficiencies that had
been identified during the 6 years since the last major
revision. The result of almost 2 years worth of lobbying
and debate resulted in the 1977 Amendments, which
specified significant changes from the Clean Air Act
of 1970. .
National Ambient Air Quality Standards
To date, EPA has established NAAQS for:
six pollutants—sulfur dioxide (SO2), total suspended
particulates (TSP), carbon monoxide (CO), nitro-
gen oxides (NOX), photochemical oxidants (Ox), and
hydrocarbons (HC). Similar standards for lead (Pb)
are expected shortly. Under the 1970 Act, all areas
of the country were required to attain the NAAQS by
July 1975. Many areas of the country (primarily urban)
have not yet attained the standards. In view of this,
the 1977 Amendments require that each state identify
those areas that have not yet complied with the
standards and develop strategies that will ensure the
attainment and maintenance of the primary standards
by December 31, 1982 and the secondary standards as
soon as possible. Congress has established strong
sanctions for those areas for which SIP revisions are
not approved by July 1979, including not permitting new
sources to locate in or near the nonattainment area
and establishing limitations on Federal highway funding
and other Federal funds.
A result of this effort to attain and maintain the
NAAQS may be further control of emissions from the
forest products industry. Each state and local agency will
be taking a much closer look at the sources that may
be causing or contributing to the nonattainment prob-
lems. This may well result in emissions limitations
for sources that were not previously regulated, as well
as stricter regulations for sources with existing
emission limitations. Sources that are likely to be
considered are pulp mills, hogged fuel boilers, saw mills,
material handling operations, and prescribed burning.
New Source Review
The Clean Air Act also establishes a New Source
Review procedure. This is basically a permit program
that allows a new source to be constructed or operated
only if it will be in compliance with various require-
ments of the Act. There are four programs that comprise
New Source Review: preconstruction review for
compliance with the NAAQS, New Source Performance
Standards (NSPS), National Emission Standards for
Hazardous Air Pollutants (NESHAPS), and the
Prevention of Significant Deterioration (PSD).
The preconstruction review program simply ensures
that any new source will not cause or contribute to
violations of the NAAQS. This review can be com-
plicated if the area in or near which a source proposes
to locate is already exceeding the standards. Congress
established two procedures, called the offset policy,
to assure that new growth is not prevented but also that
clean air is still attained. An offset occurs when
emissions from one or more sources are reduced so that
the overall air quality is improved, even though a new
-------
source of emissions has located in the area. These
emission reductions either can be obtained by the new
source proposing to locate, or the SIP can have a
formalized schedule for reducing emissions from exist-
ing sources to accommodate growth and still attain the
NAAQS by 1982. New sources in the nonattain-
ment area also will be required to comply with the lowest
achievable emission rate (LAER) by installing the
ultimate in control technology.
The NSPS cover emission limitations, equipment,
and design standards for specific categories of new sta-
tionary sources. EPA has promulgated NSPS for
24 source categories, including hogged fuel boilers cap-
able of firing fossil fuels for backup. Congress also
has required that EPA identify and promulgate standards
within 4 years for all source categories that contribute
significantly to air pollution.
NESHAPS are emission limitations and design,
equipment or work practice standards for sources emitting
pollutants that are especially hazardous to human
health. Currently there are standards for sources of
asbestos, mercury, beryllium, and vinyl chloride.
Standards for benzene are imminent. These standards
apply to both existing and new sources; new sources
must obtain a permit prior to construction.
The fourth and final portion of the New Source
Review program is PSD, which is likely to be the
most restrictive requirement for new sources. The PSD
section has two requirements: (1) that any new source
install and operate the Best Available Control
Technology (BACT) for reducing emissions; and (2)
that all sources locating in any given area, together,
not degrade air quality more than a specified increment.
The increments, shown in Table 1, vary from very
little degradation in Class I areas (currently national
parks and wildernesses, national memorial parks,
and international parks) to the greatest allowable degra-
dation in Class III areas (there are currently no Class
III areas). In no case will the air be allowed to degrade
beyond the NAAQS. All major stationary sources
that can potentially emit over 250 tons (226.796 Mg)
per year of any pollutant will need a PSD permit
before construction can begin. It should be noted that
a source locating in or near a nonattainment area
may have to satisfy both the offset and PSD provisions.
There is an increased emphasis on New Source
Review, especially the PSD and offsets programs. It is
evident that the air quality considerations included
in a permit to construct a new source or to modify or
expand an existing source have become more stringent.
Industrial growth in or near nonattainment areas and
near Class I areas will require the best control tech-
nologies. There probably will be some impact on the
forest products industry, because the resource (the
forests), and hence the industrial development, is almost
always in the vicinity of Class I areas. Air quality
considerations will be an important, if not deciding,
factor in future land management decisions.
Table 1. Prevention of Significant Deterioration Area
Classifications and Allowable Increases
National Ambient Air
Quality Standards
Pollutant
jug/rrr
Primary Secondary
Class lb
Paniculate matter:
Annual geometric mean
24-hour maximum
Sulfur Dioxide:
Annual arithmetic mean
24-hour maximum
3-hour maximum
Class II
Paniculate matter:
Annual geometric mean
24-hour maximum
Sulfur Dioxide:
Annual arithmetic mean
24-hour maximum
3-hoUr maximum
Class lllc
Paniculate matter:
Annual geometric mean
24-hour maximum
Sulfur Dioxide:
Annual arithmetic mean
24-hour maximum
3-hour maximum
5
10
2
5
25
19
37
20
91
521
37
75
40
182
700
75
260
80
365
75
260
80
365
75
260
80
365
150
1300
60
150
1300
60
150
1300
"Maximum allowable increase in micrograms per cubic meter
(but not to exceed NAAQS).
blnitially included most national parks, national wilderness areas,
national memorial parks, and international parks.
"Currently, there are no Class III areas.
NOTE: Presently, there are numerical increments only for total
suspended particulates and sulfur dioxide. EPA is, however, required to
study and promulgate numerical increments for other criteria pol-
lutants by August 1 979.
One portion of the PSD program, the new national
visibility goal, could potentially impact the forest
products industry, particularly with respect to prescribed
burning. Congress has created a national goal to
remedy existing and prevent future impairment of
visibility in the mandatory Class I areas. The strategy
to accomplish this goal will include: (1) any major
stationary source that may contribute to impairment
of visibility must install the Best Available Retrofit
Technology (BART) if it was built between 1962 and
1977; and (2) the states must develop a long-term
-------
strategy to accomplish the national goal. This portion
of the Act may have an impact on such sources as
pulp mills and hogged fuel boilers, as well as pre-
scribed burning.
Compliance and Enforcement
The 1977 Amendments substantially strengthened
the enforcement features of the Act by including
procedures for levying civil and criminal penalties of
up to $25,000 per day of violation and imprisonment
for up to 1 year, or both, for violation of Clean
Air Act requirements.
Congress also has established a requirement for
levying a penalty against sources that are not comply-
ing with applicable emission limitations, compliance
schedules, or other requirements of the Act or SIP.
The penalty will be equivalent to the cost of installing
and operating the necessary control equipment for
compliance and will be no less than the economic value
which the delay in compliance might have for the
owner of the source. In general, there no longer will
be any economic justification for not installing control
equipment, because the cost for complying will be the
same as the penalties levied.
Congress now requires that any Federally owned
source or Federal activity must comply with both the
substantive and procedural requirements of the
state and local agencies, such as the states' Smoke
Management Plans. This is a change in the concept
of Federal supremacy. It has an immediate impact on the
forest products industry. This requirement applies to
all activities on Federal lands, such as those managed by
the U. S. Forest Service or Bureau of Land Management.
Summary
The Clean Air Act, as amended in August 1977, is
a strong vehicle for attaining and maintaining clean
air in this nation. The emphasis during the next few years
will be on improving those areas not yet attaining
standards and maintaining or improving air quality in
areas that are cleaner than the national standards.
The responsibility for implementing the requirements of
the Act rests primarily with the state and local agencies,
and the strategies they choose will reflect both Fed-
eral requirements as well as local conditions. It appears
likely that many 'of the requirements and strategies
will impact the forest products industry. Those programs
that may have the greatest potential for impact are the
PSD and visibility requirements, because of the close
proximity of the national parks and wildernesses to
the national forests and other forest lands.
-------
-------
Northwest Forestry Burning
Emissions from Forestry Burning
Until recently, monitoring for emissions from open
burning were limited principally to industrial pollutants,
as governed by the National Ambient Air Quality
Standards. Emission factors were determined by relating
the quantity of effluent produced to the weight of fuel
burned. Emission factors reported for the major effluents
from forest fires are highly variable, since differences
in fire behavior produce widely different emission
patterns. Emissions are affected by the type and
conditions of the material burned. Arrangement' of
material, fuel loading, and burning technique also have
pronounced effects. A single emission factor cannot
adequately reflect these variations.
Applying a range of factors whenever possible, as
suggested by McMahon and Ryan,1 yields results that
are less misleading than those obtained using single
factors. The emission ranges for gases (Table 2) repre-
sent the best estimate of expected normal field emissions
that can be made from existing data. The range for
particulate emissions was obtained from a limited
number of field measurements of emissions from burning
western logging slash.2
Compared with other carbonaceous fuels the sulfur
content of forest fuels is low; therefore, sulfur oxide emis-
sions generally are considered negligible. Radke et al.3
measured the gas in five plumes and detected no sig-
nificant concentrations of gaseous sulfur.
Table 2. Emission Ranges (Dry Weight Basis)1-2
Effluent
Emissio.p Range
(Ib/ton of fuel)
Carbon dioxide (CO,) .
Water (H20)
Carbon monoxide (CO) .
Particulates (TSP) ......
Hydrocarbons (HC)
Nitrogen oxides (NO ). .
2000-3500
500-1500
i?0-500
M 17-67*
. I 1 0-40
-ii 2-6
*Best available emission range; revised in a private communication
by D. Sandberg, March 1978.
Emissions of trace elements from forestry burning
have not been reported. However, Darley and Lerman4
have made some measurements of smoke from labora-
tory and field burning of agricultural refuse. Quantitative
data relating emission rates of some trace emissions
to burning conditions are beginning to be reported.5
Total estimated emissions of CO, TSP, HC, and NO
in Oregon and Washington for forestry burning and "
other source categories are presented in Table 3.
High and low emission factors are presented for forestry
burning. Other source categories presented in the table
are wildfires, field burning, open burning, and all other
sources combined. This table indicates that forestry
burning contributes substantially to total particulate
emissions in the Pacific Northwest. Forestry burning
also may contribute significantly to total HC and
CO emissions.
The CO emission factors generally quoted were
derived from nondispersive infrared measurements
evolved from test fires in the burning tower described
by Darley et al.6 Field measurements of CO in fires,
adjacent areas, and smoke plumes by Countryman7
and Fritschen et al.,8 showed concentrations rapidly
diminishing with distance from the source. Such
measurements provide information on CO concentra-
tions near the fire zone, but they do not directly relate the
mass of CO to the quantity of fuel consumed.
Hydrocarbon concentrations generally have been
determined by flame ionization measurements of total
hydrocarbons and nonmethane hydrocarbons. However,
Hall9 indicated that flame ionization includes essentially
all volatile organic compounds; the hydrocarbons
designation, as applied to wood smoke, is misleading.
In emission studies from laboratory burning of logging
slash,10 the fire stage greatly influences the composi-
tion of hydrocarbon effluent; simple flame ionization
measurements did not reflect changes in the composition
of the hydrocarbon emission. An example of the com-
plexity of the volatile organic emissions is included in
a report by McMahon and Ryan.1
Nitrogen oxide emissions from burning forest fuels
have not been studied extensively. The factors quoted
in various reviews are mainly from Gerstle and
Kemnitz11 and Boubel et al.12 Oxidation of atmospheric
nitrogen requires temperatures greater than 2700°F
-------
Table 3. Statewide Emissions for Oregon and Washington3'"
Source
Category/State
Total
Suspended
Particulates0
Nitrogen
Oxides
Hydrocarbons
Carbon
Monoxide
Tons Per Year
Oregon:
Forestry burnings 22 53Q 2 65Q 13,253 26,507
Low estimate 88795 7.952 53,012 662,653
Highostimata " 13,045 3,069 18,417 107,432
Wildfiros, 4100 480 4,300 24,000
Field burning"1 • 4^320 1 363 5|511 24,365
Othsr Open burning 98614 194,421 276,564 1,084,731
Othersources ns'oys 199^333 305,292 1,240,528
Total nonforestry ' '•,
Washington:
Forestry burnings 29,766 3,502 17,509 35,019
low/estimate 117312 10,506 70,037 875,461
High estimate • 3'g61 90g 5;451 31,3oo
Wildfires • 2'200 200 2,600 ' 13,000
Field burning11 5'464 1 082 9>213 49,088
Other open burning 142636 352,275 368,042 1,699,740
Othersources 154161 354,466 385,306 1,793,628
Total nonforestry
Oregon and Washington:
Forestry burnings 52296 6,152 30,762 61,526
Lowestimato. 206107 18,458 123,049 1,538,114
High estimate 16^go6 3 g78 23i868 139,232
Wildfires..... 6'300 680 7,400 37,000
Field burning" : g'?84 2 445 14,724 73,453
Other open burning 236250 546,696 644,606 2,784,471
Other sources 269 240 553,799 690,598 3,034,156
Total nonforestry ; '_
•Emission figures taken from /Var/o.a/ Emissions Report ,,975J: Netiona, Emissions Date Systems ,NEDS, of tne Aerometnc and Emissions
Reporting System (AEROS). U.S. EPA, April 1977, except as otherwise noted.
"Sulfur dioxide emissions are not included, since forestry burning does not emit significant sulfur dioxide.
CTSP emissions of this table do not include fugitive dust emission due primarily to agricultural tilling (Oregon estimated at 156,776 tons
during 1976).
«Fie.d burning estimates are taken from Source Assessment Aoricuitvr*, Open Burning. EPA-600/s-77-107a, Ju,y 1977. and correspond to the
year 1973.
(1500°C), considerably higher than the temperatures
achieved in most prescribed fires. However, some nitro-
gen oxides are formed in these fires, possibly through
involvement of hydrocarbon-free radicals, as indicated
by Ay and Sichel.13 Oxidation of fuel also is possible.
In either case, NOX production should be fuel-dependent,
with needles, foliage, and duff producing relatively
greater quantities than woody fuels.
The emission of particulate matter from fires has
been studied more extensively than the emission of
gases. Open burning generates particles with organic
content that range in diameter from about 0.002 /xm
to macroscopic sizes. Generally, only particles with
aerodynamic diameters smaller than approximately
10 pan remain airborne long enough to impact on air
quality. Larger particles fall out of the atmosphere
rapidly and can be detected only within short distances
from prescribed fires. Multistage impactors with
final filters have been used to fractionate particles
from burning forest fuels on the basis of aerodynamic
properties. Sandberg and Martin14 found the distri-
bution of particle sizes shown in Table 4 in smoke
from laboratory burning of Douglas fir logging slash.
Emission factors are being updated as laboratory-
scale burning facilities are made more representative
of field conditions. The burning tower described by
Darley et al.,6 in some cases including the modifica-
tions indicated by Darley and Biswell,15 approximates
field conditions better than most laboratory installations.
The results obtained in numerous studies by various
-------
Table 4. Particulate Emissions from Logging Slash
(Mass Basis)14
Aerodynamic Particle
Diameter
> 5.0 ^ . .
1-5.0/x
0.3-1 .Oju.
<0.3/x
' Average
Percent
- f 8
, ', in
< 13
69
investigators using the burning tower have been sum-
marized by Wayne and McQueary.16
The reliability of emission factors obtained by
extrapolating laboratory results also is being improved
through correlation with data obtained from field
measurements. However, many more emission meas-
urements from actual prescribed fires will need to be
performed before complete emission patterns can
be reliably projected from burning prescriptions.
Differences in fire behavior and fuel conditions,
between laboratory and field situations, must be taken
into account when extrapolating laboratory data to
obtain emission factors of field fires. For example,
Sandberg2 measured particulate emissions of 6 to 24
Ib/ton (2.99 to 11.9 kg/Mg) in laboratory fires and
28 to 107 Ib/ton (13.97 to 53.38 kg/Mg) in field fires
where western logging residue was burned. The emission
factor of 17 pounds of particulate matter per ton of
fuel (8.48 kg/Mg), which has been used by Yamate17
as the basis for estimating atmospheric emissions
from forest fires, is compatible with the laboratory values
but well below the range of the field emissions. !
Potential Air Quality Impacts
The initial plume characteristics, meteorology, and the
terrain affect the transport, dispersion, deposition,
and transformation of forestry burning emissions and
hence their impact on air quality. Complex terrain airflow
and dispersion models can be used to predict the pollu-
tant concentrations that result from large smoke plumes
of slash fires. The available literature does not reveal
any modeling studies that specifically determine the
impact of slash burning activities on the smoke-sensitive
regions of the Pacific Northwest. However, the LIRAQ
model developed at Lawrence Livermore Laboratories18
and the complex terrain model by Fosberg19 are cur-
rently undergoing validation in the Willamette Valley
region of Oregon by the Oregon Air Resources Center
at Oregon State University.
The presence of potentially harmful emissions from
forestry burning has been documented. The health
and environmental effects of CO and NOX, which are
also industrial pollutants, have been cited in numerous
studies. Some hydrocarbon and volatile organic emissions
are unique to open burning. Their effects have not been
studied as extensively as those of the industrial pollutants.
Ethylene, one of the major hydrocarbon emissions,
can injure susceptible plants; however, smoke exposure
from prescribed fires is usually too short to cause
serious plant damage, except in cases of poor ventilation,
when smoke stagnates at ground level in areas adjacent
to a fire. Feldstein et al.20 estimated that the ethylene
concentration 1 to 2 miles (1.6 to 3.2 km) downwind
from a fire burning 2000 tons (1818 Mg) of landclearing
debris was 0.5 to 2 ppm and persisted for approxi-
mately 3 hours. Susceptible vegetation in the exposed
area could have suffered ethylene damage in this time,
though none was reported.
Relatively small quantities of potentially photo-
chemically reactive compounds, such as olefins,
diolefins, and substituted aromatics, are released by
forest fires. Downwind photochemical formation of
ozone and other oxidants from these and nitrogen oxides
is dependent on various plume and meteorological
parameters. Evans et al.21 reported increased ozone
concentrations in the upper layers of smoke plumes from
field fires in Australia. Radke et al.3 measured above-
ambient concentrations of ozone in plumes from
burning logging residue in Washington State.
Trace constituents emitted by forest fires have greater
potential for producing adverse health effects than the
major effluents. The volatile oxygenated organic
compounds, such as acids, ketones, alcohols, aldehydes,
and furans, produced by fires are partially absorbed/
adsorbed by condensing smoke particles. The particle-
bound vapors retain activity longer than they would in the
gaseous state and can be transported long distances
from the fires. Polycyclic Organic Materials (POM)
are formed by pyrosynthesis in all inefficient combustion
processes. Some of these compounds have been identified
as carcinogens and others are potentially carcinogenic.
Available data shows no direct evidence of adverse
health impacts from forestry burning in the Pacific
Northwest; however, as mentioned earlier, forestry burn-
ing has been shown to be a significant source of par-
ticulate matter in this region. Particulate emissions
from forestry burning also can obscure visibility. Smoke
intrusions into urban areas contribute to particulate
haze from industrial and transportation sources. Forestry
burning also may adversely affect the aesthetic visibil-
ity of pristine landscapes. Smoke plumes can elicit
negative reactions in rural areas, since smoke and haze
are commonly associated with urban air pollution.
Trends in Burning
The need for prescribed burning is expected to
decline on a long-term basis.a By the year 2020, most
old-growth timber will have been harvested, leaving com-
mercial stands in harvest cycles of 60 to 100 years.b
"State of Oregon, Interim Task Force on Forest Slash Utilization
Final Report. December, 1977.
""Personal Communication, J. Todd. USDA Forest Service
October, 1977.
-------
These second-growth stands are expected to create less
slash than old-growth timber, thereby decreasing the
need for slash disposal. On a short-term basis, State,
Federal, and private industry sources foresee an increase
in slash burning. More burning is expected because
of favorable productivity, cost incentives, and improved
burning technology and methodology.
Between 1972 and 1977, the number of acres of slash
burned in the Pacific Northwest increased both in
Washington and Oregon (Figure 1); however, the trend
of total tons burned varied. As Oregon increased,
Washington decreased and the region as a whole
remained constant (Figure 2). The proportional amount
of slash burned measured in tons per acre decreased for
both Washington and Oregon (Figure 3). These trends
are independent of two abnormal data sets. The 1974
80 I—
150 r—
100 —
<*.
60
10
- Oregon
__ Washington
_L
72 73 74 75 76 77
Year
Figure 1. Trend of Acres Slash Burned from 1972-77
5.0 i—
4.0
H 3.0
« I
I!
2 & 2,0
1.0
_____ Oregon
— — Washington
_L
_L
_L
72 73 74 75 76
Year
Figure 2. Trend of Tons Burned from 1972-77
10
77
70
60
50
40
30
20
10
\
\
\
u \
_—— Oregon
— — Washington
\
\
72
73
74 75
Year
76
77
Figure 3. Trend of Tons/Acre Burned from 1 972-77
data for Oregon used in these figures were incomplete
because of a computer malfunction. The 1976 data
for both states reflects an unusually high level of burning
because of extremely favorable weather.conditions.
Although there appears to be little change in the total
tons of slash being burned in the region, the downward
trend of the amount of slash burned per acre may
reduce the impact of forestry burning on air quality
in the area of the burning.
Conclusion
Forestry burning produces potentially harmful emis-
sions; however, the air quality impacts from forestry
burning cannot be accurately predicted due to the lack
of definitive ambient air monitoring data in Washington
and Oregon. The only active air monitoring installations
are located in urban areas where industrial and trans-
portation emission sources contribute to pollutant
levels.
In order to arrive at meaningful conclusions regarding
the current impact of forestry burning on air quality, it
will be necessary to conduct a comprehensive air
quality survey to collect data and to provide a basis for
definition of specific regions vulnerable to such impacts.
-------
Chapter 4
Wood Waste Combustion
Wood waste, once just a nuisance in most forest products
industry operations, is now considered a valuable
energy resource. As a natural byproduct of the indus-
try's resource utilization, wood waste is an attractive
alternative to more expensive fossil fuels. For many
years, the industry has used wood-fired boilers tp
produce process steam and electricity, while at the same
time solving a large-scale wood waste disposal problem.
In the Pacific Northwest, for example, there are
approximately 1000 industrial wood-fired boilers in
operation. Junge22 reports that annual use of wood
fuels in Oregon alone is equivalent to more than 19
million barrels (3.02 million m3) of oil.
The combustion of wood residues in hogged fuel
boilers creates a particulate emission composed of fly
ash, or the unburned remains of wood or bark fuel,
including sand or dirt picked up during the logging
operation. If not controlled, this fly ash can result in
visible stack emissions. Most states regulate both the
particulate emission rate and the visible emissions,
or opacity, of stack effluents. Although the regulations
vary from state to state, most requirements can be
met with air pollution control devices presently avail-
able. The particles emitted from hogged fuel boilers are
relatively large and thus relatively easy to collect.
Existing Emission Control Devices
There are a number of control devices presently in
use that meet various state air pollution control regula-
tions. These include: :
• Low-pressure wet scrubbers [2- to 6-in (5.08 tp
15.24 cm) H2O pressure drop]
• Multiclones in series [4- to 6-in (10.16 to 15.24
cm) H2O presure drop per unit]
• Multiclones in series with a low-pressure wet scrubber
• Variable-throat venturi scrubbers [10- to 15-in
(25.4 to 38.1 cm) H2O pressure drop]
• Fabric filtration (baghouse) :
• Multiclones in series with a portion of the exhaust
gas separated for wet scrubbing.
Cyclone separators and spray towers are alternate
methods of control; however, as air pollution laws
are tightened, these inevitably will give way to one of
the above methods to comply with regulations.
Low-Pressure Wet Scrubbers
The best installation for low-pressure wet scrubbers
is following a primary collector, such as the multi-
clone. One of the first wet scrubbers of this type was
installed at the Hoerner Waldorf Corporation's
Missoula, Montana mill. The boiler at this mill is a
1963 Springfield upgraded from 126 million Ib/hr
(15.86 Mg/s) to 150 million Ib/hr (18.88 Mg/s),
operating at 600 psig (4.14 MPa) and 750°F (399°C).
The boiler was originally equipped with a conventional
multiclone collector and emitted fly ash at a rate of 0.74
grains per standard cubic foot (gr/scf) (1.70 g/m3),
corrected to 12 percent CO2. More stringent regulations
required that additional controls be installed. Hoerner
Waldorf installed two parallel, Size 56, Type D,
Turbulaire stainless steel scrubbers following the
multiclones. Designed gas volume was 155 million
acfm(73.1 thousandmVs) at490°F(254°C) and5-inch
(12.7 cm) H2O pressure drop. Performance tests
after startup indicated that the emission rate was reduced
to 0.016 gr/scf (0.036 g/m3) or about 16 percent of the
maximum allowed. Overall collection efficiency of
the system is reported to be 99.5 percent.
Kirby Lumber Corporation installed a similar type
scrubber in 1972 at its Silsbee, Texas mill. Primary
multiclones were installed originally on 100-million-
Ib/hr (12.59 Mg/s) boilers emitting particulate matter,
with mass median particle size diameter of approxi-
mately 9 ^m and a total emission rate from the multi-
clones of approximately 250 Ib/hr (113.4 kg/hr). The
boilers, installed in the early fifties, were equipped
for tangential gas firing at a rate of 120 million Ib/hr
(15.11 Mg/s) and with traveling grates for wood
firing at a rate of 100 million Ib/hr (12.59 Mg/s).
Texas adopted more stringent regulations in January
1972, to become effective December 31, 1973. These
regulations required that Kirby's boilers emit no more
than approximately 40 Ib/hr (18.2 kg/hr) of particulate
matter and that the stack have an opacity no greater
than 30 percent.
Emission rates for the boilers prior to scrubber instal-
lation are shown in Table 5. After installation of
carbon steel scrubbers, the emission rate from each
stack was 7.4 Ib/hr (3.4 kg/h), giving an overall removal
11
-------
Table 5. Stack Sampling Data (Before Scrubbers) Kirby
Lumber Corporation
Stack 1 Stack 2
FSue Gas Rate (acfm)
flue Gas Rate (scfm)
Emission Rate (!b/hr)
Dry Bulb Tamp, Flue Gas (°F) .
Wot Bulb Temp. Flue Gas (°F).
72,400
45,000
263
401
145
62,800
41,100
192
350
136
efficiency of 97 percent. The regulation limit for each
stack was 41 and 33 Ib/hr (18.6 and 15.0 kg/hr);
thus, Kirby was able to reduce the emission rate from
each scrubber to approximately 18 percent of that
allowed by regulations. The cost of the 2-boiler scrubber
installation was $50,000 in 1972.
Since these initial scrubber installations, a number of
scrubbers have been installed. In 1974, an Alabama
plywood plant installed a 110,000-lb/hr (13.8 kg/s)
hogged fuel boiler, whose scrubber is preceded by a
mechanical collector. Temple Industries has two such
installations: one at Diboll, Texas and the other at
Pineland, Texas. Both include single multiclone
collectors followed by a wet scrubber.
Multiclones (Two in Series)
Theoretically, it should not be advantageous to install
a multiclone in series with the same size tubes and
with the same pressure drop, since each particle of a
given hydrodynamic diameter has a specific collection
efficiency. In practice, however, the mechanical
action in the cyclones contributes to a certain amount
of size reduction and subsequent re-entrainment. For
example, a particle of a given size might be collected with
an efficiency of 85 percent, but may break up into
smaller particles that have individual efficiencies of,
for example, 50 percent. To overcome this problem, two
multiclones have been installed in series and have
proven quite effective in meeting state regulations.
Kirby Lumber installed two multiclones in series on a
small boiler, reducing particulate emissions to 24
percent of that allowed by Texas regulations. At a Bon
Wier, Texas installation, a 120,000-lb/hr (15.1 kg/s)
boiler uses multiclones in series.
Although dry collectors offer less operating problems
than wet scrubbers, caution must be exercised in
selecting the proper system. Dirty, wet fuel could present
problems in achieving rated efficiency. Wet scrubbing
has a slightly higher collection efficiency, but pre-
sents problems with sludge disposal, as well as corrosion
if coal firing is contemplated.
If series multiclones will not meet regulations, their
operation may be improved by treating a portion of
the total gas stream (i.e., approximately 20 percent
of the stack gas), scrubbing, and then returning the clean
gas to the stack. Additional Induced-Draft (ID) fan
capacity must be provided for this operation.
Venturi Scrubbers
Venturi scrubbers have been used on hogged fuel
boilers in the 150-million-lb/hr (18.85 Mg/s) range.
A single multiclone was replaced with a parallel,
variable-throat venturi scrubber. Multiclone emissions
averagedO.597 gr/scf(1.37 g/m3) [276 lb/hr(125 kg/hr)].
Average scrubber discharge is 78,500 acfm (37.04
m3/s), containing 0.0035 gr/scf (0.008 g/m3) [1.74 Ib/hr
(0.79 kg/hr)], which represents a 99.9+ percent
collection efficiency. Pressure drop is reported to be
approximately 10 in (25.4 cm) H2O through the venturi
and 20 in (50.8 cm) H2O through the entire control
system. Cost of this system is estimated at $4.50
per scf, as compared to $1.16 per scf for the low-
pressure type.
Fabric Filtration (Baghouse)
Bark boiler emissions at Simpson Timber Company's
Sheldon, Washington mill are controlled by the use of
a baghouse. This application was utilized to control
the high sodium chloride content and objectionable
opacity of the plume caused by burning logs saturated
with salt water. Although the initial and operating costs
of fabric filtration are quite high, its principal advan-
tage over wet scrubbers is the ability to handle and
dispose of collected dust. A recent quote for a bag-
house installation on a 100,000-lb/hr (12.59 kg/s)
boiler was approximately $200,000.
Dry Scrubber
A dry scrubber has been operating for about 4 years
at Weyerhaeuser's lumber mill in Snoqualmie Falls,
Washington. The gas flow rate is 55 million acfm
(26-thousand m3/s) at 350°F (176°C) and a pressure
drop of approximately 5 in (12.7 cm) H2O. The
average emission rate from this dry scrubber is 0.05
gr/dscf(0.115g/m3).
The dry scrubber consists of a cylindrical vessel
containing two concentric, louvered, cylindrical tubes.
The annular space between the tubes is filled with
pea-sized gravel. Particulate-laden gas enters the filter
through appropriate ducting and is distributed to the
filter face by the plenum section formed between
the outlet louvered cylinder and the vessel wall. Dirty
gases pass through the filter at velocities ranging from
100 to 250 fpm (0.508 to 1.27 m/s). The gravel media
is continuously changed by a downward flow carrying
the gravel and particulate matter into a shaking
mechanism that separates the particulate from the
gravel. The gravel media is returned to the top.
Summary
Particulate emissions from hogged fuel boilers can
be controlled with relatively low-cost equipment, such as
series multiclones or a multiclone in series with a
low-pressure-drop wet scrubber. Other devices, such
as venturi scrubbers, fabric filters, and dry scrubbers,
12
-------
are considerably more expensive but may fill special
needs. In selecting a method of control, each case
must be evaluated with consideration given to: :
• Cleanliness and moisture content of fuel :'.
« Proximity to highly populated area
• Particle residence time in furnace :
• Availability of solid waste disposal area
• Governmental regulations. '
Particulate emissions from hogged fuel boilers: can be
reduced through proper fuel characteristics and1 care-
ful combustion control. Research in these areas has
been conducted recently at Oregon State University,
as described in the following paragraph.
Fuel and Combustion Research
Introduction
Present technology affecting the design and operation
of wood-fired boilers is limited. As a result, a large
percentage, of both new and older facilities are 'exper-
iencing difficulty in operating efficiently and in
meeting environmental restrictions.
The technology gap focuses about the rate of; com-
bustion of the fuel and factors that affect combustion
rates. For example, it is known that moisture influences
combustion (dry fuels burn faster than wet fuel), but
there is little data for predicting the specific influence of
moisture. Similarly, the size of the fuel can be ex-
pected to have a significant effect on the combustion
process (small size will burn more rapidly than large
size), but there are no comparative numbers to indicate
how important size is in completing the combustion
reaction. This makes it difficult to respond to such often
asked questions as, How small should wood fuel be
before it goes to the boiler? and, How dry should it be
for good combustion to occur?
Another important question that engineers, de-
signers, and operators ask is, How much fuel can be fed
to the boiler and still keep within the emission stand-
ards? This problem is usually approached by rule-
of-thumb estimates, such as, the maximum fuel 'flow
rate should result in 1,000,000 Btu/hr/ft2 (3150 kW/m2)
of grate area. Such rules of thumb do not take into
account the variations in fuel size, moisture, inlet air
temperatures, species of wood, ratios of overfire to
underfire air, etc. These parameters are not considered
because the data is not available in the literature.
Recognizing that there are many unanswered questions
regarding the combustion of wood fuels, Oregon State
University obtained funding from the Energy Re-
search and Development Administration (ERDA),
Division of Conservation Research and Technology in
April 1976, (reorganized into the Department of
Energy in 1977) to study combustion of wood residue
fuels and the factors that influence combustion.
Design of the Experiments
The overall experimental program was to study the
influence of normal industrial process combustion
variables on the combustion of wood residue fuels.
These variables were to include fuel species, fuel size,
fuel moisture, percent of excess air, and percent of
undergrate (underfire air) and overfire air.
To study the effects of these variables, the amount
of unburned carbon and noncombustible ash carried out
of the combustion chamber per unit of time was deter-
mined for each firing condition. This parameter
was used as a measure of the completeness of the com-
bustion reaction in the test facility. High levels of
combustible carbon carryover indicate poor combustion
conditions; low levels of carryover indicate better
combustion conditions.
Carryover was determined gravimetrically by weigh-
ing the material collected by the cyclone separator.
The rate of carryover was expressed in pounds per
hour [Ib/hr (0.126 g/s)]. The amount of unburned carbon
in the total carryover was determined by laboratory
analyses.
Data was collected for 79 combustion test condi-
tions using Douglas fir bark. Replicate experimental
runs were made for 29 of the 79 test conditions. The
overall experimental program is displayed graphically in
Table 6.
Data Collected
For each test run, data was collected for independent
and dependent variables. The independent variables
included fuel-related species; size range; moisture
content; density; feed rate; temperatures at forced draft
fan exit, undergrate air duct, front overfire air duct,
rear overfire air duct, and bypass air duct; gas (air) flow
rates at forced draft fan exit, front overfire air duct,
rear overfire air duct, bypass air duct, and fuel con-
veyor air cooling duct; and weather conditions.
The dependent variables on which data was collected
for each test run included combustion zone tempera-
ture 18 in (45.72 cm) and 60 in (152.4 cm) above
the grate, at the inlet to the air preheater and at the exit
from the air preheater; ash carryover rate from the
discharge of the cyclone; percent excess air as determined
by oxygen analyses; and opacity from the gas dis-
charge of the cyclone.
Results of the Experiments
The highlights of the experiments are presented in
Table 6 to show the impact of variations within the
parameters of fuel size, fuel moisture level, fuel feed rate,
and amounts of overfire and underfire air. Shown is
the rate of carryover of ash (including both combustible
and inorganic ash) as a function of levels of excess
air. Increasing levels of carryover indicate decreasing
levels of combustion efficiency. Low levels of carry-
over indicate more complete combustion of the wood
residue fuels.
1-3
-------
.
a>
r~
to
"i.
CM
10
t CO
Q CD
co £;
CD <°
co S2.
io I
F
o
o
o
CO
_co
JD
_ CO
0 '5
.- Jr
14
-------
The levels of excess air are measured downstream
from the combustion zone and are indicated on a
percentage basis. Thus, 0 percent would indicate that
stochiometric combustion conditions existed in the com-
bustion zone. A 50-percent level of excess air indi-
cates that 50 percent more air was supplied to the
combustion zone than was needed to complete the com-
bustion of the fuel for the conditions stated. For each
element presented in the graphs, an interpretation of the
data is provided in the following paragraphs to aid
in understanding the results of the experimental data.
Fuel Size
Figure 4 illustrates a comparison made between the
combustion characteristics of large dry wood fuel
and small [1A in (0.63 cm) or less] dry wood fuel.
The combustion regime is limited undergrate air, com-
parable fuel feed rates, and inlet air temperatures.
This figure indicates that the effect of fuel size under
these combustion conditions is negligible. Based on
related data, the critical parameter under these test con-
ditions is the utilization of high levels of overfire air
and limited undergrate air. It also should be noted that
increasing levels of excess air have limited effect on the
completion of the combustion reaction.
As shown in Figure 5, fuel size also appears to be
I
Q
60 i—
50
40
•30
20
10
3/8-3/4-in Bark Size I
38% Moisture :
Feed Rate = 689 Dry Ib/hr
Vt-'m Minus Bark Size
41% Moisture
Feed Rate = 656 Dry Ib/hr,
50
100
Percent Excess Air
150
200
Figure 4. Comparison of the Ash Carryover Rates in
Combustion of Dry Bark Fines vs. Dry Bark Coarse Fuel:
Maximum Overfire Air "
60
50
40
30
20
10 —
' %-%-m Bark Size
38% Moisture
Feed Rate = 689 Dry ib Air/hr
— —- Bark Fines
41% Moisture
Feed Rate = 643 Dry Ib Air/hr
I
50
150
200
100
Percent Excess Air
Figure 5. Comparison of the Ash Carryover Rates in Com-
bustion of Dry Bark Fines vs. Dry Bark Coarse Fuel:
Maximum Underfire Air
of little impact when undergrate air levels are increased.
With high levels of undergrate air, comparable fuel
feed rates, and comparable inlet air temperatures, the
effect of fuel size is relatively small, perhaps even
negligible. The influence of excess air under these
combustion conditions is pronounced, with the most
complete combustion taking place in the range of
75- to 125-percent excess air.
Moisture Content
Figure 6 illustrates a comparison made between the
combustion characteristics of wet versus dry fines utiliz-
ing undergrate air as the primary combustion air source.
The fuel feed rates and inlet air temperatures are
comparable.
This mode of combustion shows a combined influence
of undergrate air and fuel moisture levels. The. dry
(41 percent moisture) fuel appears to increase in com-
bustion efficiency as the levels of excess air are
increased. Wet (56 percent moisture) fines decrease in
combustion efficiency with increased levels of excess air.
The difference in response undoubtedly reflects the
longer residence time requirement for the 'wet fuels to
go through the combustion process. From this it
can be concluded that moisture levels in fuel principally
affect the smaller size fractions of the fuel. This
15
-------
I
cc
S
80
70
60
50
40
30
20
10
, Bark Fines
56% Moisture
Feed Rate = 571 Dry !b Air/hr
Bark Fines
41% Moisture
Feed Rate = 643 Dry Ib Air/hr t
50
100
Percent Excess Air
150
200
Figure 6. Comparison of the Ash Carryover Rates in
Combustion of Dry Bark Fines vs. Wet Bark Fines: Maximum
Underfire Air
effect is very pronounced under conditions of high
levels of undergrate air.
Amounts of Overfire and Underfire Air
Several test runs compared the use of primarily
undergrate air versus primarily overfire air. Carryover
rates under these two modes are shown in Figure 7
for dry (41 percent moisture) small-sized fuel particles.
As can be seen, the location of the incoming combustion
air has a strong influence on the combustion reac-
tion. The effect is particularly pronounced under these
specific fuel feed rates and combustion conditions.
Using larger-sized fuel particles, operating in the same
modes, carryover rates were again shown_to be lower
for the maximum overfire air mode. As illustrated
in Figure 8, low levels of undergrate air result in
IT
.c
In
I
50
40
30
20
10
_— Maximum Underfire Air
_._ Maximum Overfire Air .
Feed Rate = 643 Dry Ib Air/hr
41% Moisture
L_ N
50 100 150
Percent Excess Air
200
Figure 7. Comparison of the Ash Carryover Rates in
Combustion of Dry Bark Fines Using Maximum Underfire
and Maximum Overfire Air
50
40
30
20
10
_-^^ Maximum Underfire Air
_„__ Maximum Overfire Air
Feed Rate = 670 Dry Ib Air/hr
38% Moisture
50
100
Percent Excess Air
150
200
Figure 8. Comparison of the Ash Carryover Rates in
Combustion of Dry Coarse Bark Using Maximum Underfire
and Maximum Overfire Air
16
-------
better combustion of the dry bark fuel. In this case the
carryover at 50-percent excess air for overfire air is
roughly one fourth the carryover rate for undergrate air.
Exhaust Gas Opacity
For most (but not all) of the experimental test con-
ditions, opacity of the exhaust gas from the cyclone
was measured by eye. The results are summarized on
Figures 9 and 10. Figure 9 is a plot of exhaust gas
opacity versus percent excess air for all of the tests in
which high levels of overfire air were used in conjunction
with low levels of underfire air. The data shown repre-
sents all fuel sizes, moisture levels, etc. Only one
data point exceeded 20-percent opacity.
By contrast, Figure 10 is a plot of exhaust gas
opacity versus percent excess air for those tests in which
low levels of overfire air were used in conjunction
with high levels of underfire air (typical of firing prac-
tices in most spreader-stoker boilers). In this case the
opacity levels exceeded 20-percent opacity in 34 percent
of the tests conducted. Seven of the tests had opacity
levels approaching 40-percent opacity, and two tests
had opacities reaching 80 percent.
Opacity is a commonly used measure of pollu-
tant emissions from woodfire boilers (as well as other
sources of pollutants). Because of the emphasis .by
regulatory agencies on this parameter, these results are
indeed important and serve to emphasize the benefits
of firing boilers with proper locations for combustion
air input.
100
80
60
40
20
O
O
oo
O
O
11 I
oo
Summary of Significant Research Results
Research on the combustion characteristics of wood
residue fuels was performed in a water-wall-cooled,
spreader-stoker combustion test facility equipped with a
fixed pin-hole grate. Douglas fir bark was used ex-
clusively, but with varied fuel size, fuel-moisture
content, fuel-feed rates, levels of excess air, and loca-
tion of inlet combustion air. Results of the experiments
are summarized as follows:
• There is a significant interaction effect on combustion
efficiency between fuel size, fuel moisture content,
levels of excess air, and location of the inlet air
for combustion.
• The effect of fuel size on combustion efficiency is
limited. Test data indicates that fuel size fractions
less than 1A in have generally lower emissions
of unburned solid carbon than larger size fractions.
• Fuel moisture effects are minimal on combustion
efficiency for some modes of operation, but may
be highly significant for small size fractions
coupled principally with undergrate air feed. The
effects of moisture are negligible when the under-
grate air flow is limited to low levels and the majority
of the combustion air flow is provided as overfire air.
• The carryover of unburned carbon from the combus-
tion zone is roughly linearly related to fuel feed
rates (heat release rates) when air is supplied as
undergrate air. For normal operation of spreader-
stoker combustion systems, one can expect carryover
Data From All Tests With High Levels of Overfire Air
and Low Levels of Underfire Air.
Data is Shown for All Fuel Sizes and Moisture Levels.
Vertical Lines are Tests Exhibiting Variable Opacity Readings.
1° pfc I
I 1
20
40
60
80 100
Percent Excess Air
120
140
160
180
Figure 9. Plot of Exhaust Gas Opacity Versus Percent Excess Air
17
-------
g
O
M
s
100
so
60 —
40 —
20 —
Data From All Tests With High Levels of Underfire Air
and Low Levels of Overfire Air.
Data Shown for All Fuel Sizes and Moisture Levels—32 Tests.
Vertical Lines are Tests Exhibiting Variable Opacity Readings.
—
i
0 ®«
O
, fo
0 20 40
O
O
•>
f , 0 1,
0
\ I
I I II I
60 80 100 120 140 160 1£
Percent Excess Air
Figure 10. Plot of Exhaust Gas Opacity Versus Percent Excess Air
concentrations from the combustion zone of 3 gr/scf
(6.9 g/m3) at heat release rates of 0.52 million
Btu/ft2 (5905 MJ/m2) of grate area. The concen-
tration may approach 5 gr/scf (11.43/m3) at
heat release rates of 0.72 million Btu/ft2 (8177
MJ/m2) of grate area.
Wood fuels are roughly 80-percent volatile on
a dry basis. It is predicted in theory and now shown
in practice that combustion efficiency and stack gas
opacity can be improved by taking advantage of
this characteristic of the fuel by burning it in two-
stage combustion; that is, by strictly limiting the flow
of undergrate air to levels approaching 10 percent
of the total air flow and making up the remainder
of the combustion air by overfire air. Thus, the solid
carbon portion of the fuel is allowed to burn to CO on
the grate with very low vertical gas velocities.
The volatile portion of the fuel is burned above the
grate at the level of the overfire air inlet ports. This
mode of operation permits combustion at relatively
low levels of excess air. This finding is perhaps
the most significant of the research effort to date.
18
-------
Chapter 5
Veneer Dryer Emissions
At the point of discharge from the exhaust stack, veneer
dryer emissions consist of hot air, a small amount of
wood fiber, water vapor, and gaseous organic com-
pounds that have been distilled from the wood veneer.
The character and quantity of these steam-distilled
organics are related to the wood species, the drying tem-
perature, the humidity in the dryer, and the type, of
dryer.
. In the early 1970's, state and regional air pollution
control agencies formed a committee to establish a
source test procedure for determination of particulate
emissions from veneer dryers. This test procedure
addressed only the distilled organic gases that con-
• densed at standard temperature [70°F (22°C)]; these
were classified as particulates. Subsequent tests on local
veneer dryers demonstrated that, for the most part,
veneer dryers met current emission standards fori exist-
ing sources. Control of dryer emissions was subsequently
directed toward reducing the visible blue-haze ;
emissions to 10- to 20-percent capacity. The noncon-
densible fraction of these organic emissions was not
addressed until recent studies showed that photo-
chemical oxidants were exceeding air quality alert
levels in many parts of the country.
Photochemical oxidants are formed by chemical reac-
tions in the atmosphere between reactive hydrocarbons
and oxides of nitrogen in the presence of sunlight.
EPA's strategy for achieving air quality standards for
photochemical oxidants is to reduce emissions of
hydrocarbon precursors. According to the EPA strategy,
reducing hydrocarbon emissions from industry sources
and motor vehicles would limit the amount of hydro-
carbons available for sunlight-induced reaction and thus
reduce photochemical oxidants.
A closer look at what is being emitted from a Veneer
dryer stack reveals that initially direct condensation
of the less-volatile organic gases occurs, producing
particles that make themselves evident as a visible
blue-haze plume (Figure 11). These condensible hydro-
carbon emissions are unique to wood products and are
produced almost exclusively by veneer dryers. The
condensible organics average about 70 percent of the
total organic emissions from veneer dryers. The life-
time of these primary particles is short; in the absence
of chemical reactions, a major portion of the condensed
organics quickly revaporizes. The fate of those veneer
dryer emissions, which remain permanently in particle
form, can be predicted, if the particle size distribu-
tions and the chemical reactivity of the particles is
known. If the chemical reactivity is low, these particles
can be treated as any other stable group of particles
with a similar size distribution. Thus, their lifetime
in the local atmosphere can. be estimated and models
can be used to determine their impact on Total Sus-
pended Particulate (TSP).
The volatile component and the revaporized primary
particles have two possible fates. Some of them may
•.'•/': Secondary '•' '•.'.' '.
;W'.' /Smnnl '•• \ j. •
Persistent Primary
o Particles
Figure 11. Transport Phenomena of Air Emissions from
Veneer Dryers
19
-------
follow the hydrocarbon-to-oxidant pathway, possibly
increasing the oxidant concentration in the area; the
rest may follow the hydrocarbon-to-particle pathway,
producing secondary particles. These secondary
particles are quite distinct from the primary particles
evident in the veneer dryer plumes. They are pro-
duced slowly, over a period of hours to days and over
an entire airshed. They become visible and are
associated with the characteristic aerosol haze in rural
areas, such as the Smokey Mountains in North Caro-
lina or the Blue Ridge Mountains of Virginia. They
also can contribute to urban smog.
Of the possible fates, Washington State University
researchers Dr. Dagmar Cronn and Dr. Malcolm
Campbell23 believe that the majority of veneer dryer
emissions take the route to particle formation. Campbell,
working with related hydrocarbons, observed that more
than 80 percent of the hydrocarbons were rapidly
converted to secondary particles, leaving less than 20
percent available for oxidant production. Similar conclu-
sions have been reached by Gay and Arnts, researchers
with the Environmental Protection Agency, Environ-
mental Sciences Research Laboratory, Research
Triangle Park, North Carolina.24
On the other hand, in a statement published in the
Federal Register*5 on July 8, 1977, EPA stated that
in the analysis of available data, very few volatile
organic compounds are of such low photochemical
reactivity that they can be ignored in oxidant control
programs. Reactivity scales are measures of the
rate at which hydrocarbon compounds react in the
ambient air. EPA has classified terpenoids as highly
reactive. The classification specifically assumes that all
of the hydrocarbon is consumed in producing ozone.
There is evidence that this is not true for wood-related
hydrocarbons, which actually consume oxidants
by reacting directly with ozone in the atmosphere. In
fact, Gay and Arnts were able to account for only a
small percentage of the heavy, wood-related hydrocar-
bons that they reacted with nitrogen oxides; the bulk
of the hydrocarbons was apparently being converted
into particles.
At the request of the American Plywood Association,
the team at Washington State University submitted
a research proposal with the following goals and
objectives:
Project A: Physically and chemically characterize the
gaseous and particulate emissions from
four or five veneer dryers of varying types
and/or feedstocks.
Project B: Determine the lifetime of the primary par-
ticle plume above.veneer dryers and the
fraction of it that is persistent.
Project C: Determine the yields of particles and oxi-
dants from a typical veneer dryer, during a
2-day period of residence of the emissions in
the atmosphere.
In their summary, they felt the following data was
needed to determine the significance of wood product
hydrocarbon emissions in local air pollution.
• Information is needed on the chemical breakdown
of hydrocarbons between particulate and gas phases
at the stack, using veneer dryers as a source.
Older studies of this subject, with categorizations
into condensible and non-condensible fractions, are
probably misleading. The size distribution and
chemical stability of the persistent primary particles
need to be determined.
• The ratio of gaseous hydrocarbons going respec-
tively to oxidant formation and secondary particle
formation is urgently needed. It is expected that
almost all wood product hydrocarbon gases will end
up as secondary particles.
20
-------
Chapter 6
Direct-Fired Drying
Tremendous amounts of energy are required to extract
water from wood products to make them suitable for use
as building materials. Drying is one of the most
energy-intensive processes in the forest products
industry. This fact, coupled with rising fuel costs, has
led many mills to turn toward direct-fuel drying as
a more efficient means of heat utilization. With
direct-fired drying, the off-gases of combustion are
applied directly to the drying process, without using a
secondary heat transfer medium.
There are more than 100 direct-fired drying systems
in operation, most using wood waste fuel but having
provisions for auxiliary fossil-fuel backup. Typical
systems have been adapted to lumber dry kilns, plywood
veneer dryers, rotary dryers, and flash tube dryers.
It is possible to understand direct-fired dryer systems
better through an analysis of system similarities. For a
majority, the following apply.
• Some form of fuel preparation is required.
• A portion of dryer emissions is recirculated.
• Dryer emission hydrocarbons and wood particles are
destroyed in the recirculation process.
• At least the equivalent of combustion air require-
ments plus moisture vapor picked up in the drying
process is exhausted to the atmosphere.
• A combined effluent of dryer emissions and burner or
combustion emissions is exhausted.
• Less heat energy input than for other systems is
required to accomplish the same drying production
rates.
With direct-fired drying, the process equipment in
effect serves as the particulate collector for the products
of combustion. It is essential, therefore, that the com-
bustible element of the fuel be consumed completely in
burning and that the fuel contain a minimum of
noncombustibles which would result in particulate
emissions.
The ideal fuel for such direct heating systems should
be dry, clean, and sized to permit rapid and complete
combustion prior to entering the drying process. A
typical direct-fired system for a lumber dry kilri, veneer
dryer, or rotary dryer would use a clean, dry fuel,
such as planer mill shavings, sanderdust, plywood trim,
particleboard trim, or dried sawdust. Typical direct-
fired systems require that the fuel be reduced to
Ys in (0.32 cm) and finer and that the moisture content
be less than 20 percent. The small-sized fuel particles
and the low moisture content ensure that the process of
combustion completely consumes the fuel particles and
that only the noncombustible element in the fuel is
introduced into the process.
As the products of combustion are passed through the
drying process, a large percentage of the dry ash
remains on the surface of the material being processed
in the dryer. In the case of high-temperature pine lumber
kilns on a 24-hour schedule, there is a fine gray residue
that remains on the surface of the wood. This residue
is completely removed in the planer mill operation and
does not result in any degree of pollution. On veneer
dryers, because of the relatively short time the veneer is
exposed to the heat process, there is no noticeable
buildup of ash on the surface. With systems installed
on rotary dryers for particleboard, there has been no
noticeable color change in the particleboard as a result
of being direct-fired with wood residue.
One of the immediate and obvious advantages of
direct firing is the resultant efficiency of the heat
generated, since there is no loss of heat from steam
conversion nor is stack heat lost to the atmosphere.
Direct firing on some processes also adds some intangible
pollution benefits, particularly in the case of rotary
dryers and veneer dryers. A typical veneer dryer,
direct-wood, waste heating system involves recirculating
kiln atmosphere back through a heating chamber, where
the products of combustion from the burner are intro-
duced and mixed with the recirculated atmosphere into
a chamber controlled at 1100°F (590°C). This tempera-
ture, plus the presence of flame from the burner, helps re-
duce the amount of blue-haze emission normally
associated with direct-fired softwood veneer dryers. The
same is true on rotary dryers, where a percentage of
the stack emissions can be recycled through the heating
system.
With direct-fired drying systems, the fuel required for
the burner system must be processed, handled, and
stored. In a typical system, the dry-wood residue is
introduced into a live bottom surge bin that feeds a con-
trolled amount of fuel to a hammermill. The final
hammermill should have a screen with approximately
21
-------
54-in (0.32 cm) openings. In grinding planer shavings,
ply-trim, or particleboard trim the resultant grinding
produces a fuel that to a large extent is finer than No. 6
mesh; only 25 percent is finer than No. 40 mesh and
only a small percentage will pass No. 100 mesh. This fuel
size can be successfully collected in a high-efficiency,
long-cone cyclone without the need for secondary
filters. However, with sanderdust as fuel, normally it is
necessary to use the cyclone on the storage bin as a
primary collector, and a bag-type filter as secondary
collector to meet existing pollution requirements.
A typical system includes a low-pressure blower
system from the hammermill to a high-efficiency cyclone
on top of the storage bin. The fuel is fed into the
burner in a high-pressure system. The burner becomes
the collector at the end of the high-pressure system.
Existing systems for direct firing have used dry,
clean residues; the rough bark and other wet residues
have been left to steam generation. This probably will
continue to be the trend in most instances, since a dry,
clean, direct-fired wood residue system is by far the
simplest and least expensive installation for producing
heat from wood residue.
Systems are being developed that can use green fuels
and convert the resultant heat energy into clean hot air
for the drying process. Some of these systems process
green fuel through dryers, producing pellets from the
dry bark and using these pellets for boiler feed sys-
tems similar to those now being used for burning clean,
dry waste. Other systems are being developed that
will use unprocessed green hogged fuel to generate
direct-fired heat for the heating process.
A recent installation in a lumber mill in east Texas
uses an underfeed stoker system to burn hogged fuel
on a grate. A regenerative fly ash collector unit is
installed in the discharge stream from the furnace
chamber to remove all but the finest particulate matter
in the combustion gases. This allows the combustion
gases to be mixed with recirculated kiln atmosphere
and to be introduced into the dry kiln to provide the
necessary heat.
Another interesting development is the burner gasifier
unit. In this system the green hogged fuel is fed into
the burning chamber through an underfeed pipe. Pile
burning occurs on a pinhole grate and the underfire
and overfire air in the burning chamber are controlled
to approximately 1200°F (648.8°C). The volatiles are
introduced into a secondary cyclonic burning chamber,
where additional combustion air is introduced and
rapidly mixed. The secondary burning increases dis-
charge temperatures to approximately 2200°F (1204°C).
Emission tests indicate that the discharge gases on green
hogged fuel at 60-percent moisture (wet basis) result
in 0.156 gr/scf (0.358 g/m3) of exhaust gas corrected
to 12 percent CO2. If the hogged fuel in the unit is
only 35-percent moisture (wet basis), the resultant grain
is 0.053 gr/scf (0.12 g/m3) corrected to 12-percent
CO2. This particular unit is under development, both
for direct-firing processes and for introducing the
hot gases into existing boiler systems where stack
emission problems are particularly critical.
Where new equipment is being added to a mill,
or where expensive pollution-control equipment must
be added to bring older boilers into compliance, direct
firing of appropriate equipment should be considered.
In many instances, relieving some steam require-
ments from boilers can reduce considerably the pollution
equipment required. It is also possible that future tech-
nology may indicate that alternative ways of burning
green hogged fuel should be considered, rather than
attacking the pollution problem at the stack with costly
pollution control equipment. Wood gasification units,
fuel preparation of sizing, and drying are all very possible
and practical approaches both to direct firing and
to firing of steam boilers. All of the processes that
are currently being developed for direct firing will have
future application for clean firing of boilers, as well
as providing direct-fired process heat.
Direct-Fired Veneer Dryers at SWF Plywood
Company
System Description
SWF Plywood Company, a subsidiary of Southwest
Forest Industries, currently operates 16 veneer dryers
at five manufacturing complexes in Oregon. Ten of these
dryers are fired by nine suspension, direct wood-fired
systems. An SWF internal energy study in 1976
revealed that some 830 million ft3 (23.5 million m3)
of natural gas replacement (in Btu equivalence) was
gained through direct wood firing. Approximately
60 percent of all energy used, exclusive of electricity,
was derived from wood wastes generated in the manu-
facture of plywood.
Direct-firing concepts at SWF operations evolved
following installation in 1972 of a sanderdust burner in
conjunction with a wigwam burner to meet emission
standards. Two units were installed in 1973-74
at former Carolina Pacific Plants, and seven units
were installed in 1974-75 under the newly formed SWF
Plywood Company's direction. All nine units are
suspension type burners using finely ground fuel
with maximum size of'5/64-in (0.198 cm).1The plants
where these units are located all receive veneer that
is shipped in from company-owned or outside veneer
mills. Wood waste fuel is made readily available at each
location as a byproduct of the plywood production
process. After processing, the wood waste fuel has an
average moisture content of 2 to 4 percent and an
average heat value of 8500 Btu/lb (19.77 MJ/kg).
In plywood manufacturing, there is roughly an 18-
to 19-percent waste factor. About 180 to 190 pounds
(81.64 to 86.18 kg) of dry wood waste is produced
per 1000 ft2 (92.90 m2)'of %-in (0.95 cm) finished
plywood from standard dimensions of veneer. The fuel
mix can vary depending upon the degree of finishing
done at the plant.
22
-------
At 8500 Btu/lb (19.77 MJ/kg), the waste generation
rate described in the previous paragraph equates to
1530 to 1616 Btu/ft2 (17.4 to 18.4 MJ/m2) of 3/s-in
(0.95 cm) production. This Btu factor is very important
in the analysis of the dry fuel direct-fired system.
Source testing data gathered on an Energex unit in
December 1977 revealed that 1630 Ib/hr (0.205 kg/s)
of fuel were being used to dry 12,300 ft2 (1142.7 m2)
of 0.1-in (0".25 cm) Douglas fir veneer [%-in
(0.95 cm) basis] input rate. This analysis, shown in
Table 7, reveals that the heating value from plywood
waste is sufficient to dry the plywood completely using
a direct-fired veneer dryer.
Fuel requirements increase with the drying of
wetter, thicker woods and/or cooler ambient tempera-
tures. A review of annual SWF operations shows excess
fuel being generated in the summer months tapering
to a slight deficit (3 to 5 percent) in mid-winten
Current goals are to achieve wintertime self-sufficiency
of fuel through increasing system efficiencies.
Emission Testing
Good emissions data needs to be developed for
direct-fired systems. The data gathered should define
the system's environmental impact and also should
enable the owner to provide maximum efficiencies and
production levels and minimum capital requirements
for environmental control.
The need for data occurred when the preliminary
Oregon emission guidelines directed the SWF direct-
fired system to meet the 10-percent-average, 20-percent-
maximum, veneer-dryer opacity standard. Also in-
cluded in the guidelines was a particulate emissions
limitation of 0.1 gr/scf (0.229 g/m3) at 12-perc;ent
CO2. The required test method was EPA Method No. 5
with a backup filter. Testing locations on the unit also
Table 7. Analysis of Heat Use and Fuel Balanqe for
SWF Direct-Fired Dryer
Heat Use :
1. 1630 Ib/hr fuel X 8500 Btu/lb = 13,855,000 Btu/hr input
2. 13,855,000 Btu/hr input •*• 12,300 ft2 production =1126, Btu/hr/ft2
3. Btu in recirculating gas flow = 3,500,000 Btu/hr ,
4. Recirculation 3,500,000 Btu/hr H- 12,300 ft2 = 284 Btu/fir/ft2
5. Total Heat Use = 11 26 Btu/hr/ft2 + 284 Btu/hr/ft2 •!
= 1410 Btu/hr/ft2
Fuel Production _ . . .
1. 1 2,300 ft2/hr production X 0.85* = 10,455 ft2/hr finished plywood
2. 10,455 ft2/hr finished wood X 180 Ibs fuel/1000 ft2
= 1880 Ib/hr fuel produced ,'
Fuel Balance ti
1880 Ib/hr produced - 1630 Ib/hr used = + 250 Ib/hr surplus fuel
* Finished veneer factor i
were suggested. Data was needed to determine if the
system(s) were properly understood and defined, if the
stated test methods were applicable, and if the emissions
were typical of veneer dryer emissions for control
purposes.
Data gathered on the test unit in 1976 has provided
SWF with the information necessary to implement
a comprehensive program potentially able to reduce
former emission levels by 50 percent, while improving
fuel use efficiency and maintaining production levels.
Opacities on the test unit have not yet been brought
down to the 10-percent average level but have been
reduced significantly, with the potential identified for
further reduction prior to control system contemplation.
Table 8 is a comparison of test results showing the
emission reduction benefit already gained. Low levels
of CO2 in the most recent (1977) data revealed a
possibility for further improvement of combustion
efficiencies through the control of combustion air. Instru-
mentation has now been installed to balance fuel and
combustion air input. Fuel usage rates are being moni-
tored to check overall efficiency impacts. Increased
HC reductions also are anticipated as a result of 100-
percent recirculation of the dryer dry-end emissions
to the burner.
Analysis of the emission particulate revealed, as
expected, that emissions from direct-fired systems are
typical of the emissions from steam or gas-fired veneer
dryers: there is a definite wood combustion component
Table 8. Veneer Dryer Emission Data Comparison
October—1976
(Average Of Two Tests—
Production Rate 14,000 ft2/hr)
Inlet
Return Atmosphere
Temp (°R)
CO (%)
H 0 (%) .
Emissions (Ib/hrl
1,367
17,812
5.45
1 5.2
0 1495
22.64
760
8,670
4.3
19.8
0.183
13.60
760
13,481
4.3
19.8
0.183
21.15
Average Emission Rate
1.51 lb/1000 ft2 veneer
December—1977
(Average Of Three Tests—
Production Rate 12,680 ft2/hr)
Inlet
Return Atmosphere
Temp (°R) . .
Flow (scfm) .
CO, (%) ....
Grains/scf
Emissions
(Ib/hr)
Average Emission Rate
1,329
18,761
3.37
8.6
0.1013
16.35
801
10,111
2.3
12.1
0.0952
8.16
817
10,405
2.3
12.1
0.0952
8.49
' 0.67 lb/1000 ft2 veneer
23
-------
in the exhaust gases. Air flow velocities are sufficient
to keep the finer particles in suspension and in the
recirculation flow. Laboratory analyses have verified
that a significant amount of combustion particulate is
impinged upon the veneer surface. This takes place on
the wet veneer in the dryer green zone.
This system has three built-in emission control
features, each of which should be optimized prior to
adding on emission controls:
• Up to half of the exhaust gas is kept in recirculation
• Roughly half of the condensible HC is destroyed
from the return flow in blending, or 25 percent of total
condensible HC
• A total of 15 to 20 percent particulate is dropped
out in the dryer and/or impinged upon the veneer.
The data shows that the direct-fired system (operated
at peak efficiencies—lowest fuel usage per unit of
production) can operate within the compliance emis-
sion rates for a hogged fuel boiler-steam dryer system.
Table 9 contains theoretical emission rates (at
compliance) calculated for a hogged fuel boiler-steam
dryer unit on a pounds-of-emission per square foot of
fi-in (0.95 cm) plywood produced basis. The hogged
fuel boiler analysis is based upon industry averages
reported by the American Plywood Association in June
1975, which were: 133 pounds (60.5 kg) dry hogged
fuel required to produce 1,000 pounds (453.5 kg) of
steam; and 2,400 pounds (1088.6 kg) of steam required
to dry 1000 ft2 (92.90 m2) of %-in (0.95 cm)
Douglas fir veneer.
A comparison of Table 9 with the 1977 emission
rate of the direct-fired systems in Table 8 shows that
emissions problems caused by direct firing are not as
bad as stack opacities might indicate. The key to lower
emissions is in the unit efficiencies. The less fuel that is
required to generate needed heat results in less ash
than is created in combustion. The more dryer emissions
are recirculated from the HC-producing dryer zones,
the greater the level of hydrocarbon destruction.
When compared with boiler-steam dryer units, the
direct-fired unit produces fewer emissions in terms
of pounds per unit of production. However, to achieve
optimum results, the system must be operating efficiently
within a carefully controlled combustion regime. And
even under these conditions, exhaust opacity standards
may be difficult to meet.
The direct-fired system has the following advantages:
• Fossil fuel replacement
• Solid waste utilization for energy
• High-efficiency heat utilization
• A low emission rate per unit of production.
Other observations concerning the operation of direct-
fired systems include:
• Fewer dryer fires at higher operating temperatures
• Easier dryer cleaning, requiring only a water rinse
as opposed to the use of caustic chemicals
• Improved veneer gluing characteristics.
Table 9. Hogged Fuel Boiler Steam Dryer Analysis
Boiler-Fuel Requirements
133 Ib dry fuel 2400 Ib steam 12,300 ft2 veneer
X : X
1000 Ib steam 1000ft2veneer
hr
_ 3,925 Ib dry fuel
hr .
Assuming a 13-percent extra heat demand to remove 50 percent
moisture in fuel:8
3,925 Ib dry fuel 4,435 Ib dry fuel
- X 1.13 = -
hr
Boiler-Exhaust Rate
hr
At 50-percent excess air, 11,543 scf combustion gases is produced
from 100 Ib dry fuel:8
11,543 scf 4,435 Ib fuel 1 hr 8,532 scf
X X =
100lbfuel hr 60 min. min.
Boiler-Compliance Loading at 12-percent C02
0.1125 grains 8,532 scf 60 min. 1 Ib _ 8.3 Ib
X
scf
min. hr 7000 grains hr
Steam Dryer-HC Emission Rateb
0.56 Ib 12,300 ft2 veneer 6.9 Ib
•X
1000 ft2 veneer
hr
hr
Assuming 50-percent collection efficiency to meet 10-percent average
opacity:
6.9 Ib _ 3.45 Ib
X 0,5 —
hr hr
System Total
8.3 Ib 3.45 Ib 11.75 Ib
hr hr
11.75lb 1 hr
-X
hr
0.955 Ib
hr 12,300 ft2 veneer 1000 ft2 veneer
ajunge, David C., Boilers Fired with Wood and Bark Residues,
Research Bulletin No. 1 7, Forest Research Laboratory, Oregon State
University, I November, 1975.
bAdams, Donald F., Evaluation of Veneer Dryer Emissions, American
Plywood Association, February, 1971.
The increased air flow of the direct-fired system
raises pressure levels and necessitates tighter seals and
baffles to prevent fugitive emissions. But the more
difficult problem is exhaust stack opacity. The future
of the direct-fired dryer may need to be carefully
weighed against opacity standards and their enforcement.
24
-------
Chapter 7
Case Study: Pilot Studies for
Particulate Control of Hogged
Fuel Boilers Fired with Salt
Water-Stored Logs
Introduction
Firing salt-laden wood residue in hogged fuel boilers
results in increased particulate emissions and high visual
opacities. Georgia-Pacific Bellingham has very little
control of the salt content in the fuel because a varying
amount of hogged fuel is purchased from sources that
utilize salt water storage of logs.
Salt makes up about 25 to 50 percent of the par-
ticulate loading from the hogged fuel boilers at G-P
Bellingham and as high as 65 to 90 percent of the
particulate from boilers at other Northwest, mills. Salt
particulate is submicron in size and consequently difficult
to collect. The total grain loading must be reduced to
less than 0.05 gr/scf (0.115 g/m2) to reduce stack
opacity to 20 percent.
Hogged Fuel Boiler Facilities
The hogged fuel boiler system at G-P Bellingham
consists of four Sterling boilers built by Babcock and
Wilcox Co. in 1937 to 1940. They are of the Dutch
oven design with two cells per boiler. Each boiler has
three steam drums and one mud drum, combustion
air preheaters, but no economizers. They are rated at
55,000 Ib/hr (6.92 kg/s) of steam when hogged fuel is
fired alone, or 70,000 Ib/hr (8.8 kg/s) each when
oil is fired in combination with hogged fuel.
Boilers 1, 2, and 3 have Western Precipitation multi-
clone for particulate removal. No. 4 boiler has had its
Dutch oven blanked off and is currently fired with
natural gas or oil (no hogged fuel). All four boilers
exhaust through a common breeching to a single stack
(see Figure 12).
The primary source of hogged fuel for these boilers
is residue from salt-water-floated logs from Canadian
lumber mills. Other fuel is trucked in from local saw-
mill operations or supplied by the mill's barking
plant. Clarifier sludge is burned in the boiler about
one shift each day.
The hogged fuel has not been classified or dried.
Severe variations in moisture content of the hogged fuel
require oil to be fired at times to ensure combustion.
Emission Studies
Particulate tests were performed in January 1977
using an EPA Method 5 particulate sampling train
to collect design data for scrubber selection. Subsequently,
a particle size distribution was made using a high
resolution cascade impactor designed and built by
Phillip A. Nelson. Particulate and particle size distribu-
tion were tested at a point upstream of the main stack,
after the multiclones; 30 to 35 percent of the total
particulate is smaller than 1 ^m in size. (Refer to
Figure 13). See Table 10 for the test results from the
particulate testing. In order to use the particle size
distribution as a basis for scrubber selection, it was
necessary to test when salt content in particulate
emissions was at its worst. To simulate this condition,
salt was added to the fuel being burned at a rate of
0.5 percent by weight on a dry fuel basis.
Initial Investigation of Particulate Collectors
and Pilot Selection
Five types of particulate collectors were investigated
as possible devices for bringing the hogged fuel boilers
into compliance. These were: (1) wet scrubbers (both
low-energy impaction type and variable-throat
venturi), (2) gravel bed scrubbers, (3) dry precipitators,
(4) wet precipitators, and (5) baghouses.
16-inch Insulated
Duct to Pilot Units
Figure 12. Hog Fuel Boiler Layout
25
-------
99.99 99.9 99.8
99 98
95
I
80 70 60 50 40 30 20
Run No. 1, Sa\t Added
Run No. 2, Salt Added
Run No. 3, No Salt
.01 .05 .1 .2 .5 1 2 5 10 20 30 40 50 60 70 80 90 95 98 99 99.8 99.9 99.99
Percent Smaller Than (By Weight)
Figure 13. Particle Size Distribution HFB
Table 10. Paniculate Testing Results
Item
,
(© 70aP 29 92" Hg)
Units
grains/sof
%
acfm
scfm
°F
103lb/hr
%
%
1-11-77
No. 1,2,3
0.79
13/5
172,843
101,615
440.6
50
25.0
52.4
1-11-77
No. 1,2,3,4
0.54
10.4/6.7
204,724
120,842
437
50
44.1
NA
1-12-77
No. 1,2,3
0.30
8/1 1.5,
161,265
98,963
479
32
118.0
NA
Note: Fuel burned during all three tests was purchased hogged fuel; no barking plant refuse or clarifier sludge.
Operating conditions for all three tests: gas and oil were not being burned and the cinder reinjection fans were off.
A survey of pilot work performed at other mills
and of equipment currently under construction (or in
operation) showed that all of the collector types already
had been tested and/or selected for use except the
Union Carbide Corporation Electro-Tube precipitator.
This device was invented by Air Pollution Systems,
26
-------
Tukwila, Washington and is exclusively licensed to
Union Carbide Corporation. This device is a wetted-wall
pipe type precipitator with a patented electrode geom-
etry. It was decided to pilot test this unit, since it had
many potential advantages, such as low space require-
ments, minimal pressure drop, and very low electrical
usage.
Additional information was required for the economic
evaluation of baghouses. Two bag fabrics were
recommended: Nomex® andTeflon®-coated Fiberglas.
Fiberglas baghouses have been operated for over a
year and a half. Nomex®, however, was not being
used in a baghouse operating on hogged fuel boilers
fired with salt-laden fuel. Because there would be an
economic incentive to use Nomex® if it had a .signifi-
cantly longer bag life than the Fiberglas, a Nomex®
baghouse was pilot tested to allow Dupont to predict
Nomex® bag life. Other reasons for running this pilot unit
were to: (1) confirm an air-to-cloth (A/C) ratio of 3.25:1
(air-to-cloth is the ratio of acfm treated gas to the
fabric surface area in ft2) with the unit being cleaned
on-line; (2) measure collection efficiency; and (3) collect
enough dust to evaluate settling rates, etc.
Electro-Tube
Description of the Pilot Plant
The Union Carbide Corporation Electro-Tube
is a two-stage electrostatic device that combines a
wetted-wall electrostatic precipitator with high-
intensity particle precharging. Particles are charged at
various disks and are continuously attracted to the tube
wall by the field between the electrode and the tube
wall. The Electro-Tube is a pipe type electrostatic
precipitator with wetted walls that has proved to be
effective where the resistivity or stickiness of the
dust prohibits the use of a conventional unit. The
electrode geometry provides a more stable corona dis-
charge system that results in operating field levels of
three to five times higher than that of conventional
electrostatic precipitators. The result is a higher
charge on the particles, and consequently, a greater
force of attraction to the anode. This happens because the
center electrode is a metal rod, not a wire with
a weight attached. The rod is relatively rigid (compared
with the wire and weight). Thus, a close tolerance on
the gap between the electrode and tube can be main-
tained. The smaller that gap, the greater the electrical
field for constant electrical usage.
The Electro-Tube was tested for a maximum gas
flow rate of 800 acfm (0.377 m3/s) per tube. The wetted-
wall anode is 10 ft (3.048 m) long by 12 in (30^48 cm)
in diameter. These tube dimensions are the same
for both the pilot unit and full-size installation. Thus,
a scale-up to a full-size unit would mean simply
increasing the number of tubes for a specific application.
A 2-in (5.08 cm) diameter driving field electrode
with four different discharge electrode configurations
was used for testing.
Testing Procedure
Fourteen sets of inlet and outlet grain loading tests
were performed with varying electrode configurations,
voltage-current levels, and weir overflow rates
(Table 11). Two inlet and outlet tests each were run
at gas flow rates of 500, 625, and 750 acfm (0.236,
0.294, and 0.354 m3/s) with all other variables constant.
These tests were to determine the effect of gas flow rate
on performance.
Eleven of the 14 sets of tests were performed by
G-P personnel using an EPA Method 5 particulate
sampling train. Since only one sampling train was
available, tests were not concurrently run at the inlet
and outlet to the Electro-Tube precipitator. This intro-
duced some uncertainty in the evaluation of the Electro-
Tube. Air Pollution Systems (APS) was capable of
providing simultaneous inlet-outlet sampling; these are
identified in Table 11 as tests 12, 13, and 14.
Discussion of .the Particulate Test Results for the
Electro-Tube
Actual Electro-Tube outlet grain loadings ranged
from 0.00087 gr/scf to 0.0214 gr/scf (0.0019 to
0.049 g/m3) (see Table 11). Corrected outlet load-
ings ranged from 0.0019 gr/scf (.004 g/m3) at 12 percent
CO2 to 0.044 gr/scf (0.100 g/m3) at 12 percent CO2-
This was accomplished with near zero opacity.
Grain loadings corrected to 12 percent CO2 are
useful for compliance testing, but not for determining
the true collection efficiencies. Because of this, collec-
tion efficiencies discussed here are calculated on
the basis of uncorrected grain loadings. Calculated
efficiencies for 12 of the 14 tests ranged from 95.6 per-
cent to 99 percent.
Measured pressure drops across the Electro-Tube
(from inlet sampling point to the outlet sampling
point) ranged from 0.3-in (0.76 cm) H2O to 1-in
(2.54 cm) H2O, averaging 0.68 in (1.73 cm).
Water used for particulate removal from the
Electro-Tube ranged from 1.33 to 6 gpm/1000 acfm
(0.178 to 0.802 mVs/1000 m3/s) of treated gas. Drain
fittings were not adequate on the pilot unit and plugged.
Increasing water flow prevented a buildup of particulate
on the tube wall, and resulted in a reduced sparking
rate. It can be speculated that adequately designed
drain fittings would have allowed good operation at a
water flow rate of 2 gpm/1000 acfm (0.267 m3/s/1000
m3/s) of treated gas.
Electrical requirements for the Electro-Tube itself
would be approximately 100 hp (74.6 kW) for a system
large enough to treat the gases from G-P's hogged
fuel system. Electrical requirements of pumps and fans
are not included.
Baghouse
Description of the Pilot Plant
Mikro Pul Corporation supplied a baghouse for G-P's
pilot trial. This was a pulse-jet baghouse fitted
27
-------
Table 11. Electro-Tube Test Data
Data Set
Test No.
1
2
3,
4,.,
5...
6
7
8
g
jO
11.
12,
13
14..
Voltage Amperage Temp.
kV MA °C
750 78 4.3 165
90
750 78 4.3 183
93
750 78 4.3 167
87
750 78 4.3 164
84
500 90 3.6 164
84
625 76 2.3 166
94
625 76 2.3 166
95
750 76 2.3 178
111
750 76 2.3 171
95
500 76 2.3 102
72
500 76 2.3 116
72
790 64 0.9 192
110
680 75 1.1 185
89
739 74 1.5 192
83
Grain*
gr/sof
0.3311.
0.0439
0.7036
0.018
0.9307
0.00994
0.502
0.0096
0.3245
0.0073
0.2608
0.0060
0.2880
0.0092
0.2487
0.0045
0.0361
0.01 1 0
0.1859
0.0027
0.1664
0.0019
Uncorrected , , . ,
Uncorrected gpm
3rain Loadina Comments
jram uodumg % £ff Wej|.
gr/scf
0.1617 86.7 1.0
0.214
0.3446 97.4 1 .0
0.0088
0.4708 98.9 1.8 Plugged Multiclone
0.00503
0.254 98.1 1.8 Plugged Multiclone
0.0047
0.137 97.7 1.8
0.0031
0.1898 97.8 3.0
0.0042
0.1824 96.8 3.0
0.0058
0.1976 98.3 3.0 Data in Question
0.0035
0.0274 71.2 3.0 Data in Question
0.0079
0.1219 98.9 3.0 Salt Water Added
0.0013
0.0906 99.0 3.0 Salt Water Added
0.00087
0.2279 96.2 3.0 Salt Water Added; Boilers Not At
0.008 Maximum Steam Production
0.4167 99.7 3:0 Salt Water Added
0.0011
0.1991 95.6 3.0 Salt Water Added
0.0088
'Corrected values to 12% C02
with 100 standard galvanized support cages and
Nomex® felt bags. The pilot plant was designed for
a gas flow rate of 3000 to 3500 acfm (1.42 to 1.65 m3/s).
Testing Procedure
Nine sets of inlet and outlet grain loading tests were
performed to verify cleaning efficiencies and also to
determine the air-to-cloth ratio at which the unit could
be operated with a stable presure drop. These tests were
performed while the unit was being cleaned.
To verify feasibility of off-line cleaning during the
second phase of testing, the baghouse was operated for
a period of time with its compressed air shut off.
Air flow was then stopped and the unit was cleaned.
Pressure drop readings were recorded while the bag-
house was operating and after cleaning periods.
One bag was removed from the unit each week and
was returned for physical tests to project bag life on the
Nomex® in a full-scale installation.
Discussion of Test Results for the Baghouse
Actual outlet grain loadings ranged from 0.00046 to
0.015 gr/scf (0.0010 to 0.034 g/m3) (see Table 12).
Corrected outlet loadings ranged from 0.00077 to 0.028
28
-------
Table 1 2. Baghouse Test Data
Date <
Test h
1
2
3 ,.. .
4. ; . .
5
6
7
8
9
, i Grain Loading (
JQ ACFM gr/SDCF
' (actual)
* 2781 0 7369
'. 0.0043
3673 03849
0.0150
0.00b46
..-•-.-..... . 3432 0 4301
0.0015
, ; , 3047 0 2396
0.0060
9Q7Q O i^RV
0.0049
i 321 4 01 96
0.0099
i 2924 0 242
! 0.003
' i 2753 0 335
0.0012
Srain Loading
gr/SDCF E
@ 1 2% C02
0.0083
0 5371
0.0281
0.00077
0.00230
0 356
0.010
0.009
0.0142
0417
0.0046
0.0016
Collection
-rr- - n/ A/C Ratl"°
:fficiency % 2
. ' acfm/fr
(actual)
gr/scf (0.0017 to 0.064 g/m3) at 12 percent GO2.
Stack opacity was essentially zero during the 7-week
pilot run, except for a few hours during startup periods.
During startup, water vapor was visible from the
stack until the system was brought up to operating
temperature. >
Collection efficiencies based on actual inlet aiid outlet
grain loadings ranged from 94.9 to 99.8 percent. These
efficiencies corresponded to no visible discharge.
On-line operation of the baghouse at an air-to-cloth
(A/C) ratio of 3.3:1 or less resulted in a stable pressure
drop of ab'but 2!5-in (6.35 cm) H2O across the tube
sheet. An increasing pressure drop resulted when the
A/C ratio was increased above 3.3:1.
Off-line cleaning of the baghouse was performed at
A/C rajios as high as 5.9:1. Adequate cleaning of the
bags was achieved, which meant that pressure drop could
be maintained at this higher A/C ratio. It should
be pointed out that off-line cleaning requires additional
valving hardware to isolate modules during the cleaning
cycle.
Dupont performed physical tests on the test bags
and projected a baglife of 18 to 24 months for Nomex®
in the service of salt-laden fuel fired in a hogged fuel
boiler. Three factors are detrimental to the use of
Nomex® in this service: (1) a temperature of 385 to
400°F (196 to 204.4°C) is recommended as the
maximum continuous temperature; (2) Nomex® is
subject to chemical attack from SO2 in the flue gas; and
(3) the bag life of Nomex® seems to be shortened by
chlorides.
Summary and Conclusions
Union Carbide Corporation Electro-Tube
Overall performance of the pilot unit was excellent.
Electrical costs are only a fraction of that required
for any other conventional particulate collection device.
Pressure drop is nominally 1-in (2.54 cm) H2O, again
less than any other competing device. It is expected
that a full-scale system would perform equally well. The
principal drawback to this design involves construction
materials, since most metals experience extreme
corrosion when exposed to salt water at 400°F
(204.4°C). Consultants recommended Inconel 625,
but this material would double the cost for a baghouse.
Nomex0 Baghouse
Fabric filtration has been satisfactorily established
as a pollution control device for hogged fuel boilers
with salt-laden fuel. G-P's pilot work verified this.
Nomex® bag life is expected to be approximately
29
-------
the same as Fiberglas. However, temperature, SO2 con-
centrations, and chlorides must be controlled to allow
Nomex® to be used for this application.
In summary, it was demonstrated that both
Electro-Tube and Nomex® baghouse devices would
reduce particulate emissions to required levels.
Each had its unique advantages and disadvantages,
and individual peculiarities. Final selection of a
full-sized installation would, of course, require site-
specific considerations.
30
-------
Air Pollution Control
List of References
'McMahon, C. K. and P. W. Ryan. 1976. "Some
Chemical and Physical Characteristics of Emissions
from Forest Fires." 60th Annual Meeting of the
Air Pollution Control Association. Portland,
OR.
2Sandberg, David V. 1974. "Measurements of
Particulate Emissions from Forest Residues in Open
Burning Experiments." Ph.D. Thesis. College
of Forest Resources, University of Washington.
3Radke, L. F., J. L. Stitch, D. A. Hegg, and P. V. Hobbs.
1978. "Airborne Studies of Particles and Gases
from Forest Fires."/. Air Pollution Control
Assoc., v. 28, no. 1:30-34.
4Darley, E. F. and S. L. Lerman. 1975. Air Pollutant
Emissions from Burning Sugar Cane and
Pineapple Residues from Hawaii. EPA 450/3-
75-071.
5McMahon, C. K., and S. N. Tsoukalas. 1977. "Poly-
nuclear Aromatic Hydrocarbons in Forest Fire
Smoke." Proceedings of the Second International
Symposium on Polynuclear Aromatic Hydro-
carbons. September 28-30, 1977.
6Darley, Ellis F., F. R. Burleson, E. H. Mateer et al.
1966. "Contribution of Burning of Agricultural
Wastes to Photochemical Air Pollution." J. Air
Pollution Control Assoc., v. 16, no. 12:685-90.
7Countryman, C. M. 1964. "Mass Fires and Fire
Behavior." USDA Forest Service. Pacific Forest
and Range Experiment Station Resource Paper
PSW-19. Berkeley, CA. ".
8Fritschen, L., Harley Bovee, Konrad Buettner,
Robert Charlson, Lee Monteith, Stewart Pickford,,
James Murphy and Ellis Darley. 1970. Slash Fire
Atmospheric Pollution. Pacific Northwest
Forest and Range Experiment Station and Cali-
fornia Univ., Riverside, Statewide Air Pollution
Research Center RP-PNW-97. 42 p.
9Hall, J. A. 1972. Forest Fuels, Prescribed Fire,
and Air Quality. USDA Forest Service. Pacific
Northwest Forest and Range Experiment Station.
Portland, OR, 46 p.
10Sandberg, D. V., S. G. Pickford, and E. F. Darley.
1975. "Emissions from Slash Burning and the
Influence of Flame Retardant Chemicals." /. Air
Pollution Control Assoc., v. 25:278-81.
uGerstle, R. W. andD. A. Kemnitz. 1967. "Atmos-
pheric Emission from Open Burning." /. Air
Pollution Control Assoc., v. 17, no. 5:324-27.
12Boubel, R, W., E. F. Darley and E. A. Schuck.
1969. "Emissions from Burning Grass Stubble and
Straw."/. Air Pollution Control Assoc.,
v. 19:497-500.
13Ay, J. H. and M. Sichel. 1976. "Theoretical Analysis
of NO Formation Near the Primary Reaction
Zone in Methane Combustion." Combustion and
Flame, v. 25:1-15.
14Sandberg, David V. and Robert E. Martin. 1975.
Particle Sizes in Slash Fire Smoke. USDA Forest
Service. Pacific Northwest Forest and Range
Experiment Station Research Paper, PNW-199.
7 p.
15Darley, Ellis F. and H. H. Biswell. 1973. "Air
Pollution from Forest and Agricultural Burning."
/. Fire and Flammability, v. 4, no. 4:74-81.
16Wayne, L. G. and M. L. McQueary. 1975. Calcu-
lation of Emission Factors for Agricultural
Burning Activities. EPA 450/3-75-087.
17Yamate, George. 1973. Development of 'Emission
Factors for Estimating Atmospheric Emissions
from Forest Fires. IIT Research Institute Final
Report. Chicago, IL. 147 p.
18Bass, A., A. Q. Eschenroeder, and B. A. Egan. 1977.
The Livermore Regional Air Quality Model
(LIRAQ): A Technical Review and Market
Analysis. ERT Report P2348-1.
19Fosberg, M. A. 1976. Estimating Airflow Patterns
Over Complex Terrain. USDA Forest Service
Research Paper RM-162.
20Feldstein, M., D. Duckworth, H.C. Wohlers and
B. Linsky. 1963. "The Contribution of the Open
Burning of Land Clearing Debris to Air Pollution."
/. Air Pollution Control Assoc., v. 13, no. 11:542-46.
21Evans, L. F., I. A. Weeks, et al. 1977. "Photo-
31
-------
chemical Ozone in Smoke from Prescribed Burning
of Forests." En viron. Sci. and Tech., v. 11, no.
9:896-900.
22Junge, David C., Boilers Fired With Wood and
Bark Residues, Research Bulletin No. 17, Forest
Research Laboratory, Oregon State University, ;
November 1975.
23Cronn, D. R. and Campbell, M. J., "Study of the
Physical and Chemical Properties of Atmospheric
Aerosols Attributable to Plywood Veneer Dryer
Emissions," Research proposal submitted to
the American Plywood Association, January 27,
1978.
24Gay, B. W. and Arnts, R. R, "The Chemistry of
Naturally Emitted Hydrocarbons" presented at the
Research Triangle Institute Meeting, Research
Triangle Park, NC, 1976.
^Federal Register, "EPA Recommended Policy on
Control of Volatile Organic Compounds," July 8,
1977.
32
-------
Air Pollution Control
Bibliography
Adams, Donald F., "Veneer Dryer Emissions and
Control Systems," Environmental Protection Agency
Workshop, Atlanta, GA, June 29, 1977. !
Adams, T. N., Particle Burnout in Hog Fuel Boiler
Furnace Environments, Proc. TAPPI Environmental
Conference, Atlanta, GA, April 26-28, 1976.
Bray, David, "Overview of Standards and Regulations,"
EPA-FPRS Pollution Control Seminar for the
Pacific Northwest Forest Industry, Portland, OR,
April 4-6, 1978.
Cronn, D. R. and Campbell, M. J., "Air Pollution
in the Medford-Ashland Area," August 10, 1977.
Duval, Keith, "A Review and Survey of Hydrocarbon
Emission Sources in the Medford AQMA," EPA
Contract 68-01-4160, with Pacific Environ-
mental Services, Inc., May 1977.
Grimes, Gary, "Direct Fired Drying—The Hybrid
Unit," EPA-FPRS Pollution Control Seminar for the
Pacific Northwest Forest Industry, Portland,
OR, April 4-6, 1978.
Guidon, Michael, "Pilot Studies for Particulate
Control of Hog Fuel Boilers Fired with Salt Water
Stored Logs," EPA-FPRS Pollution Control
Seminar for the Pacific Northwest Forest Industry,
Portland, OR, April 4-6, 1978.
Junge, D. C. and Kwan, K. T., "An Investigation
of the Chemically Reactive Constituents of Atmos-
pheric Emissions from Hogged Fuel Fired Boilers
in Oregon," Forest Projects Journal, v. 24, no. 10
pp 25-29, 1974.
Junge, David C., "Combustion Characteristics of
Douglas Fir Bark in Spreader-Stoker Boilers,"
EPA-FPRS Pollution Control Seminar for the Pacific
Northwest Forest Industry, Portland, OR,
April 4-6, 1978. ;
Himel, James H. et al., "The Impact of Forestry
Burning Upon Air Quality," EPA-FPRS Pollution
Control Seminar for the Pacific Northwest Forest
Industry, Portland, OR, April 4-6, 1978.
Kreisinger, Henry. "Combustion of Wood Waste
Fuels," Mechanical Engineering, v. 61:115-120
February 1939. :;
McConnell, Clifford T., "Control of Emissions from
Direct-Fired Process Heating Systems," EPA-FPRS
Pollution Control Seminar for the Southern
Forest Industry," Dallas, TX, September 28-29
1977.
Mick, Al, "Fate of Veneer Dryer Emissions," EPA-
FPRS Pollution Control Seminar for the Pacific
Northwest Forest Industry, Portland, OR,
April 4-6, 1978.
Pacific Northwest Industry Society and Air Pollution
Control Association, "Source Test Procedures
for Determination of Particulate Emissions
from Veneer Dryers," S-8 Source Test Committee,
September 1972.
Radian Corporation, "Examination of Ozone Levels
and Hydrocarbon Emissions Reduction," Novem-
ber 18, 1977.
Ritchey, J. R., "Venturi Wet Scrubber for Particulate
Control on a Bark Boiler," Proc. First Annual
Symposium on Air Pollution Control in the
Southwest, Texas A&M University, College Station,
TX, November 5-7, 1973.
Robinson, J. W., "Wet Scrubber Application to
Hogged Fuel Boilers," Proc. APCA, June 15-19,
1975, Boston, MA.
Robinson, John, "Control of Emissions from Wood
Waste Steam Generation Systems," EPA-FPRS
Pollution Control Seminar for the Southern Forest
Industry, Dallas, TX, September 28-29, 1977.
Semrau, K. T., C. W. Marynowski, K. E. Lunde,
and C. E. Lapple, "Influence of Power Input on
Efficiency of Dust Scrubbers," Industrial and
Engineering Chemistry, v. 50:1615, November 20,
1958.
Semrau, K. T., "Practical Process Design of Particulate
Scrubbers," Chemical Engineering, v. 84, no. 20,
September 26, 1977.
U.S. Environmental Protection Agency, "Control
of Volatile Organic Compounds," EPA Workshop,
Atlanta, GA, June 1977.
Veach, James L., "Summary of Applicable Air
Pollution Control Guidelines and Programs," EPA-
FPRS Pollution Control Seminar for the Southern
Forest Industry, Dallas, TX, September
28-29, 1977.
Wailing, J. C., "Scrubber Controls Hogged Fuel
33
-------
Boiler Fly Ash Emissions," Pulp & Paper, June
1971.
Ward, D. E., C. K. McMahon, and D. D. Wade. 1974.
"Particulate Source Strength Determination
for Low-Intensity Prescribed Fires." In Specialty
Conference on Control Technology for Agricultural
Air Pollutants. Air Pollution Control Associa-
tion. Memphis, TN.
34
-------
Chapter 8
Part II
Introduction
Water Pollution Control
The forest products industry is a resource-intensive
industry relying on the ability of land and water to
produce timber. The industry's impact on water quality
results from its land management and timber harvesting
practices and from its manufacturing processes. In
both areas, the industry must comply with pollution
control regulations enforced by the U.S. Environmental
Protection Agency and state-level environmental
control agencies acting under the guidelines of the
nation's Clean Water Act.
To assist the industry in planning and implementing
pollution control measures, the U.S. Environmental
Protection Agency and the Forest Products Research
Society sponsored pollution control seminars held
in Dallas, Texas on September 28-29, 1977, and
in Portland, Oregon on April 4-6, 1978. This publica-
tion contains materials presented at the seminars
and is intended to serve as a reference to the industry
and others concerned with the approaches and tech-
nologies that will enable the industry to meet its pollution
control obligations.
The papers presented by regulatory officials,
researchers, and industry personnel are reviewed in
this publication. As such, no attempt is made to provide
a comprehensive picture or cookbook approach to
water pollution control problems in the industry. Rather,
the contents of this publication represent a range
of thinking and activities relating to water quality
management by the forest products industry.
Appended is a list of references available on the
subjects covered.
35
-------
-------
Chapters
The Clean Water Act
The Clean Water Act of 1977 (PL 95-127) is an
amended version of the Federal Water Pollution Control
Act Amendments of 1972 (PL 92-500). Its objective
is to restore and maintain the chemical, physical, and
biological integrity of America's waters. The Act
sets a national goal of eliminating man-caused pollutant
discharges into navigable waters by 1985, with an
interim goal of swimmable, fishable water quality
by 1983.
Pollutants targeted by the Clean Water Act,are
broadly categorized as either point sources or nonpoint
sources. Point sources are pollutants discharged at
discrete locations, such as pipe and channel outfalls.
Nonpoint sources are more pervasive and include
such pollutants as overland runoff and underground
seepage leading to contamination of surface and
groundwaters.
Point sources of pollution are controlled through the
National Pollutant Discharge Elimination System
(NPDES), a permits program administered by the U.S.
Environmental Protection Agency (EPA) through
state-level environmental agencies. The Clean Water
Act also established procedures for setting effluent limi-
tations and standards for point sources.
Point Source Control
When the PL 92-500 was signed into law in October
1972, there were 27 categories of point sources for
which effluent limitations and standards were to be estab-
lished by the EPA Administrator. The first regulations
affecting the primary wood products industry (timber
products processing) appeared in the Federal Register
in April 1974 and January 1975. Pretreatment regu-
lations for existing sources in the timber products
processing point source category appeared in the Federal
Register in December 1976.
These regulations set forth effluent limitations for
existing sources, new source performance standards, and
pretreatment standards for 15 subcategories ofithe
industry. Except for the subcategories that include wood
preserving plants, the pollutant parameters addressed
are biochemical oxygen demand (BOD), suspended
solids, and pH. For the wood-preserving segment of the
industry, chemical oxygen demand (COD), phenols,
oil and grease, and pH are the parameters.
After these regulations were promulgated, a civil
action suit was brought by the Natural Resources
Defense Council (NRDC), Environmental Defense
Fund (EDF), the National Audubon Society, and
Businessmen for the Public Interest, Inc. in the U.S.
District Court for the District of Columbia. As a result
of this action, a Settlement Agreement was made on
June 7, 1976, which has serious ramifications for
industry and the EPA. The Agreement charges EPA
with the task of developing and promulgating effluent
limitations guidelines and standards for 65 toxic
substances; this is to be done for 21 major industry
categories, of which timber products processing
is the first on the list.
Table 13 lists the 65 chemical substances for which
effluent limitations and guidelines and new source
performance and pretreatment standards are to be
developed. Since the list of the 65 chemical substances
is somewhat generic, EPA was faced with the problem
of how to identify specific chemical substances
that would meet the requirements for addressing the
65 substances listed in the Settlement Agreement.
Table 14 lists 129 specific representative chemical sub-
stances. Table 15 is a list of those Standard Industrial
Classification (SIC) codes in the timber products
processing industry that are to be addressed in carrying
out the work under the Agreement.
The timber products processing industry was the first
studied by EPA under the Settlement Agreement. The
work that is being done for the timber products processing
category by EPA under the Settlement Agreement is
divided into three major components: (1) a technical
study, (2) an economic impact analysis study, and
(3) environmental impact studies. The schedule set forth
in the Agreement called for EPA to retain a contractor
for technical studies of the timber products processing
industry no later than June 30, 1976. This was
done. The technical work was completed by Environ-
mental Science and Engineering, Inc. of Gainesville,
Florida. Their report was delivered to EPA on July 29,
1977 and has been circulated to other Federal offices,
states, industry, and public interest groups for review
37
-------
Table 13. Settlement Agreement Pollutants
Aconophthono
Acrolom
Acrylonitrlle
Aldfin/Dioldrin
Antimony & Compounds
Arsenic & Compounds
Asbestos
Benzene
Benzldlna
Beryllium & Compounds
Cadimium ft Compounds
Carbon Totrachlorido
Chlordane
Chlorinated Benzenes
Chlorinated Ethanes
Chloralkyl Ethers
Chlorinated Napthalene
Chlorinated Phenols
Chloroform
2-Chlorophonol
Chromium & Compounds
Copper & Compounds
Cyanides
DDT & Metabolites
Dichlorobenzene
Dichlorobenzidine
Dichloroethylenes
2,4-Dichlorophenol
Dichloropropane & Dichloropropene
2,4-Dimethylphenol
Dinitrotoluene
Diphenylhydrazine
Endosulfan & Metabolites
Endrin
Ethylbenzene
Fluoranthene
Haloethers
Halomethanes
Heptachlor & Metabolites
Hexachlorobutadiene
Hexachlorocyclohexane
Hexachlorooyclopentadiene
Isophorone
Lead & Compounds
Mercury & Compounds
Napthalene
Nickel & Compounds
Nitrobenzene
Nitrophenols
Nitrosamines
Pentachlorophenol
Phenol
Phthalate Esters
Polychlorinated Biphenyls
Polynuclear Aromatic Hydrocarbons
Selenium & Compounds
Silver & Compounds
2,3,7,8-Tetrachlorodibenzo-P-Dioxin
Tetrachloroethylene
Thallium & Compounds
Toluene
Toxaphene
Trichloroethylene
Vinyl Chloride
Zinc & Compounds
and comment. An economic analysis of the industry
is being conducted by Arthur D. Little of Cambridge,
Massachusetts.
Under the current schedule in the Settlement
Agreement, EPA must propose regulations affecting >
the timber products processing category by September
30, 1978 and must promulgate final regulations by
March 30,1979.
The information gathered to date indicates that ,
considerable effort is being made by the wet process
hardboard, insulation board, and wood preserving
segments of the industry to conserve water and to recycle
and reuse process wastewater. In addition to pollution
control, some of the incentives for re-examining
water and wastewater management practices in the
plants have been the limited availability of fresh
water, higher energy and sludge disposal costs, and,
for indirect dischargers, industrial cost recovery fees
and increased user charges. Some of the plants are
finding that the cheapest wastewater control option is
complete recycle of wastewater or no discharge.
An economic impact analyst will be studying the
impact of each level of pollution control technology
applied to the industry, including such impacts as
potential plant closures, employment, profitability,
return on investment, and international competition.
Studies of the environmental impact of the pollutants
being discharged by the industry will focus basically on
those water bodies in various geographical regions of
the country that might be most severely impacted by j
the industry's discharges. As EPA approaches its
rulemaking deadlines, it will continue to seek comments
from the industry. Communications have been estab-
lished with the industry through the efforts of the
American Wood Preservers Association, the American
Wood Preservers Institute, the National Forest
Products Association, the American Board Products
Association, the Southern Furniture Manufacturers
Association, and the Forest Products Research Society.
Nonpoint Source Control
Section 208 of the Clean Water Act is designed to
produce areawide waste treatmentmanagementplanning
to assure adequate control of sources of water pollution
in each state. Activities not covered by the point source
control program are addressed by the nonpoint source
planning aspects of Section 208.
Point and nonpoint sources of pollution can be
differentiated by comparing their control methods. Most
nonpoint sources are best handled by using land man-
agement techniques rather than by treating the effluent.
In other words, it's more practical and economical
to treat nonpoint sources where the rain hits the ground
than to collect the runoff and send it to a treatment plant.
Nearly all nonpoint source pollutants are storm-
generated and most are predominantly rural in nature.
EPA has arrived at eight categories of nonpoint sources:
agriculture, silviculture, construction, mining, hydro-
logic modifications, practices primarily affecting
groundwater, residual waste disposal, and urban runoff.
While construction and urban runoff are major factors
in urban areas, the remaining categories are more
prevalent in rural areas.
EPA realizes that the magnitude of the nonpoint
source problems is immense—possibly greater than the
total problem caused by point sources. It is estimated
that 15 percent of the nation's waters are failing to
38
-------
Table 14. Specific Chemical Substances to be Examined
1. acenaphthtne3
2.'acroleina
3. acrylonitrile8
4. benzene3
5. benzidine8
6. carbon tetrachloride8
Chlorinated benzenes8
7. chlorobenzene
8. 1,2,4-trichlorobenzene
9. hexachlorobenzene
Chlorinated ethanes8
10. 1,2-dichloroethane
11. 1,1,1-trichloroethane
12. hexachloroethane
13. 1,1-dichloroethane
14. 1,1,2-trichloroethane
15. 1,1,2,2-tetraohloroethane
16. chloroethane
Chloroalkyl ethers3
17. bis (chloromethyl) ether
18. bis (2-chloroethyl) ether
19. 2-chloroethyl vinyl ether
Chlorinated napthalene3
20. 2-chloronaphthalene
Chlorinated phenols8
21. 2,4,6-trichlorophenol
22. panachlorometa creosol
23. chloroform8
24. 2-chlorophenol8
Dichlorobenzenes3
25. 1,2-dichlorobenzene
26. 1,3-dichlorobenzene
27. 1,4-dichlorobenzene
Dichlorobenzidine3
28. 3,3'-dichlorobenzidine
Dichloroethylenes8
29. 1,1-dichloroethylene
3O. 1,2-trans-dichloroethylene
31. 2,4-diohlorophenol8
Dichloropropane and dichloropropene8
32. 1,2-dichloropropane
33. 1,3-dichloropropylene
34. 2,4-dimethylphenola
Dinitrotoluene8
35. 2,4-dinitrotoluene
36. 2,6-dinitrotoluene
37. 1,2-diphenylhydrazinea
38. ethylbenzene8
39. fluroanthene8
Halpethers3
40. :4-chlorophenyl phenyl ether
41. 4-bromophenyl phenyl ether
42. .bis (2-chloroisopropyl) ether
43. bis (2-chloroethoxy) methane
Halomethanes3
44. •methylene chloride
45. methyl chloride
46. ' methyl bromide
47. bromoform (tribromomethane)
48., dichlorobromoethane
49. |:trichlorofluoromethane
50.': dichlorodifluoromethane
51.!' chlorodibromomethane
52.'! hexachlorobutadiene8
53.' hexachlorocyclopentadiene8
54. isophorone3
55. napthalene8
56. ! nitrobenzene3
Nitrophenols3
57. 2-nitrophenol
58. 4-nitrophenol
59. 2,4-dinitrophenol3
60.:' 4,6-dinitro-o-creosol
Nitrosamines8
61. .N-nitrosodimethylamine
62. 'N-nitrosodiphenylamine
63. N-nitrosodi-n-propylamine
64. pentachlorophenol3
65J phenol8
Pthalate esters8
66. 'bis (2-ethylhexyl) phthalate
67. butyl benzyl phthalate
68. di-n-butyl phthalate
69. :di-n-octyl phthalate
70. diethyl phthalate
71. dimethyl phthalate
Polynuclear aromatic hydrocarbons3
72. 1,2-benzanthracene
73. benzo (a) pyrene (3,4-benzopyrene)
74. 3|4-benzofluoranthene
75. :11,12-benzofluoranthene
76. chrysene
77. acenaphthylene
78. 'antracene
79. 1,12-benzoperylene
80. 'fluorene
81. phenanthrene
82. 1,2,5,6-dibenzanthracene
83.i:indeno (1,2,3-cd) pyrene
84. pyrene
85. 'tetrachloroethylene8
86.- toluene3
87.I trichloroethylene8
88. vinyl chloride (Chloroethylene)3
Pesticides and metabolites
89. aldrin3
90. dieldrin8
91. chlordane8
DDT and metabolites3
92. 4,4'-DDT ~
93. 4,4'-DDE (p,p'-DDX)
94. 4,4'-DDD (p,p'-TDE)
Endosulfan and metabolites3
95. a-endosulfan-Alpha
96. b-endosulfan-Beta
97. endosulfan sulfate
Endrin and metabolites8
98. endrin
99. endrin aldehyde
Heptachlor and metabolites8
100. heptachlor
101. heptachlor epoxide
Hexachlorocyclohexane8
102. a-BHC-Alpha
103. b-BHC-Beta
104. r-BHC (lindane)-Gamma
105. g-BHC-Delta
Polychlorinated biphenyls (PCB's)8
106. PCB-1 242 (Arochlor 1242)
107. PCB-1254 (Arochlor 1254)
108. PCB-1221 (Arochlor 1221)
109. PCB-1 232 (Arochlor 1232)
110. PCB-1248 (Arochlor 1248)
111. PCB-1260 (Arochlor 1260)
112. PCB-101 6 (Arochlor 1016)
113. toxaphene3
114. antimony (Total)8
115. arsenic (Total)8
116. asbestos (Fibrow)3
117. beryllium (Total)8
118. cadmium (Total)8
119. chromium (Total)"
120. copper (Total)3
121. cyanide (Total)8
122. lead (Total)3
123. mercury (Total)3
124. nickel (Total)8
125. selenium (Total)8
126. silver (Total)8
127. thallium (Total)8
128. zinc (Total)3
129. 2,3,7,8-tetrachlorodibenzo-
p-dioxin (TCDD)3
8Specific compounds and chemical classes as listed in the consent decree.
Ambiguous compounds or classes of compounds are underlined.
39
-------
Table 15. Timber Products Processing Industry Categories approach, EPA has recognized four important facts:
SIC Coda
Title
2411 ... Logging Camps and Logging Contractors (Camps Only)
2421 ... Saw Mills and Planing Mills, General
2426 ... Hardwood Dimension and Flooring Mills
2429 .,. Special Purpose Sawmills, Not Elsewhere Classified
2431 ... Millwork
2434 ... Wood Kitchen Cabinets
2435 ... Hardwood Veneer and Plywood
2436 ... Softwood Veneer and Plywood
2439 ... Structural Wood Members, Not Elsewhere Classified
2491 ... Wood Preserving
2499 ... Wood Products, Not Elsewhere Classified (Furniture Mills)
2661 ... Building Paper and Building Board Mills (Hardboard Only)
meet water quality standards because of nonpoint
source pollution and another 35 percent are equally
degraded by a combination of point and nonpoint
sources. The effects of these nonpoint sources will
become more evident as pollution attributed to point
sources is further reduced through applied control
technology.
EPA's approach to nonpoint source pollution control
has been to establish, by regulation, the concept of
Best Management Practices (BMP's). These are
defined as:
... a practice, or combination of practices, that is determined
after problem assessment, examination of alternative practices,
and appropriate public participation to be the most effective,
practicable (including technological, economic, and institu-
tional considerations) means of preventing or reducing the
amount of pollution generated by nonpoint sources to a level
compatible with water quality goals.
Thus BMP's are broadly defined in terms of objectives
rather than specific control activities. In adopting this
• Essentially there is no one right way to prevent,
manage, or control nonpoint source pollution
problems.
• The same problem-solving techniques cannot be
applied everywhere and produce the same results due
to such uncontrollable factors as climate and
geography.
• The physical environment together with legal,
political, and institutional conditions differs from
area to area—to such an extent as to make rigid
Federal requirements for control inappropriate and
unworkable.
• Successful management of nonpoint sources and of
the 208 program must involve the continual integra-
tion of technical, administrative, and regulatory
functions and skills.
For the most part, effective land management
techniques are known and many are in daily use through-
out the country. These will form the options from
which BMP's will be chosen. As an example, the Soil
Conservation Service, Forest Service, Soil and
Water Conservation Districts, Extension Services,
and a host of other Federal, state, and local agencies
have worked for many years to develop land management
techniques to keep the soil on the land. When these
techniques are evaluated from the slightly different
viewpoint of keeping the soil out of the water, they
become BMP's.
Section 208 recognizes that technology alone may not
be sufficient to achieve the goals of the Clean Water
Act. In order to go beyond a technological assault
on water pollution, major public policy questions must
be addressed. It was with this foresight that Congress
placed the local emphasis on Section 208 planning,
involving locally elected officials and requiring extensive
public participation in plan development.
40
-------
Chapter 10
IMonpoint Sources and Water
Quality Research for Forest
Management
Introduction
Research devoted to developing methods for controlling
nonpoirit source pollution from silvicultural activities
is a very recent scientific endeavor. It grew out of an
applied research area known as watershed management,
which, in turn, had its origins in some very fundamental
questions about how forests affected climate and
streamflow.
Concern about the interaction between forests and
water qan be found throughout history, mostly regarding
the influence of forests on floods. The theme "Pro-
tecting the forest is protecting the river" has been
carried through time up to the nation's earliest efforts
to regulate forest use for flood protection. A principal
reason for establishing National Forests was to
secure and protect the forested headwaters of the nation's
river systems for flood-control purposes. j;
Controlling soil erosion through reforestation of
abandoned farmland became an important research issue
in the 1920's and 1930's. This concern was extended
to investigations of the effect of timber harvesting on
soil loss in steep terrain. Most of this research began in
the eastern U.S. and the focus was on maintaining
soil productivity, not water quality. As a result, soil losses
generally were quantified in terms of tons lost per acre
rather than in the concentration units typical of
today's water quality standards.
Research that focused specifically on the relationship
between silviculture and water quality did not begin
until the late 1950's. In the Northwest, the research
on the H. J. Andrews Experimental Forest and in the
Alsea Basin Aquatic Resources Study were pioneering
efforts to establish such relationships. ;
Today, within the framework of the Clean Water Act
and its focus on nonpoint sources, there is increased
concern regarding the impact of forest management
practices on water quality. Attempts to regulate such
practices have revealed a dearth of available information
on which to base and justify control recommendations
and policies. To expand upon available information,
a number of studies have been initiated recently.
The EPA and the U.S. Forest Service have been
involved principally in research programs presently
underway. But the efforts to date have only begun to
establish meaningful or usable information. Most experts
agree that further research is needed—and most point
to the specific need for cross-fertilization of research to
develop better understanding of the linkage mechanisms
that translate actual forest management practices
to site-specific water quality impacts. Such understand-
ing will not come easily.
The problems surrounding nonpoint source studies
are related to their general context: forestry-related
nonpoint sources are created within the context of a
partially controllable system that is variable over space
and time. With respect to water quality, forest lands
can be characterized as follows:
• Day-to-day variations in water quality in undisturbed
watersheds are substantial, particularly during
periods of changing flows.
• Fluctuations in undisturbed systems may be as great
as those in apparently similar, but disturbed, sys-
tems.
• Most water quality constituents identified as
pollutants (with the exception of certain introduced
chemicals) also occur naturally within the system.
« The pollutants discharged enter the water in a
diffuse manner and, in many cases, at intermittent
intervals.
« The extent of pollution is related, at least in part, to
certain uncontrollable climatic events, as well as
geographic and geologic conditions, and may
differ greatly from place to place.
• The pollutants arise over an extensive area of land
and are in transit overland before they enter
receiving waters.
These factors, as well as natural variations in site
characteristics and the cyclic nature of most silvicultural
activities, contribute to the complexity of assessing
man-induced pollution and then selecting appropriate
controls.
EPA's Environmental Research Laboratory in
Athens, Georgia, has been charged with the responsi-
bility for developing total management and technological
systems to control nonpoint source pollutants associated
with forest products industry operations. However,
EPA stresses that it is relying heavily on outside
expertise and interagency agreements (particularly with
the Forest Service) to meet this challenge.
41
-------
Examples of Recent Research
One of the first steps taken in the EPA Athens
Laboratory research program was to assess what is
already known about the forest nonpoint source problem
and, from that, to identify priority research needs.
This assessment and identification was done for EPA
under an interagency agreement with the U.S. Forest !
Service. It resulted in a report entitled Non-Point \
Water Quality Modeling in Wildland Management: '>
A State-of-the-Art Assessment.26'21 That document,
published in two parts as a part of the EPA's Ecological
Research Series, presents an assessment and review
of: (1) the generalized cause-effect relationships
between major types of forest management activities
and uses and water quality parameters, such as
sediment, temperature, and dissolved oxygen; (2) the
magnitude of the potential problem associated with
each of the majortypes of activities and uses; (3) manage-
ment techniques available to reduce the problem; and
(4) the usefulness and reliability of existing nonpoint
source predictive models in planning effective forestry
nonpoint source controls. It also includes an evaluation
of the data base available for model development
and testing.
One of the key points identified during the project was
that nonpoint pollution is directly related to the
time and space variability of the hydrologic cycle and
existing terrain and that the relationship is site-dependent.
Studies made during the course of the project also
identified some major weaknesses in available predict-
ive models.
The interagency agreement with the U.S. Forest
Service also resulted in the development of a method-
ology for evaluating alternative erosion control ;
strategies for silvicultural activities. As part of the
study, an assessment was made of the erosion potential
associated with common silvicultural activities on a
nationwide basis. The project report, entitled
Silvicultural Activities and Non-Point Pollution
Abatement: A Cost-Effectiveness Analysis Procedure,28
assesses the erosion potential of common silvicultural
activities and provides a method of identifying costs
and effectiveness of alternative erosion control practices.
To use this or similar cost evaluation procedures for
water quality management, one must be able to relate
man-induced accelerated soil erosion on the land to
defined water quality goals for the receiving waters. One
must also know the purpose of the activity or use, the
appropriate practices for the conditions, activities
and uses involved, and the local dollar costs for
implementing those practices.
A private research consulting firm, Development ,
Planning and Research Associates, conducted a study
for EPA to determine and assess those current and
emerging trends in U.S. silviculture and agriculture that
will have the most significant environmental implica-
tions in 1985 and in 2010. Individual assessments
were conducted for silviculture and harvest management,
Table 16. Major Environmentally Related Trends in
Silviculture and Harvest Management
Rank
Trend
1
2
3
4
5
6
7
8
Pest Control
Source: Upper, S. G. Environmental Implications of Trends in Agricul-
ture and Silviculture. U.S. Environmental Protection
Agency, Athens, GA. EPA-600/3-77-121. October 1977.
range and pasture management, livestock produc-
tion, and irrigated and nonirrigated crop production.
The most significant environmentally related trends
were rated and rank-ordered in an evaluation work-
shop, comprising a panel of experts representing
the Federal government, private industry, and academia.
These experts assessed each trend and its environ-
mental implications at both the regional and national
levels. Each subtrend was then rated according to
its extensiveness and the intensiveness of environ-
mental effect for 1976 and 2010.
The silviculture panel identified regional differences
within trends and their implications. Trends and
their implications were assessed on a regional basis by
forest types. The top trends related to silviculture
and harvest management are identified in Table 16.
It shows that the most important trends relate to
access to timber resource, site preparation, log extraction,
timber utilization, fire control, growth enhancement,
stand conversion, and pest control. This study
suggests that:
• BMP's will be highly variable, as the problem varies
between and within regions.
• There is still much to be learned about the significance
of aesthetics and ecological effects associated
with forest management operations.
• Relative to agricultural nonpoint sources, forests
pose less of a pollution problem on a national scale.
• Increased wood production can be achieved in an
environmentally acceptable manner, with perhaps
even some net improvement in environmental
quality.29-30
WRENS—Water Resources Evaluation,
Nonpoint Sources, Silviculture
A more recent Forest Service project was aimed at
development of a handbook on procedures for identifying
the technical suitability of alternative control strate-
gies for a specific site and activity. This document,
commonly referred to as WRENS, addresses nonpoint
42
-------
source pollution in terms of site-specific relationship
to existing terrain and the hydrologic cycle. The
handbook describes a procedure for identifying and
assessing this relationship. It also explains how
to identify technically suitable control alternatives for
site-specific silvicultural activities.
The WRENS handbook should prove usefuljto
planners and managers. Information flow charts help
the program planner define the technical assessments
needed. The analytical procedures, example appli-
cations, and references help guide technical consultants
through the assessments. And the step-by-step
procedural guides and examples of control alternatives
for specific conditions help project designers and
managers to identify suitable controls for the particular
activity and site conditions. While the specific analytical
methods presented are not the only ones, they were
carefully chosen according to the capabilities of the
science and state of the art.
Silvicultural activities that can be evaluated using the
described procedures include timber harvesting.
associated transportation systems, and cultural practices,
such as site preparation and timber stand improve-
ment. These activities can be analyzed quantitatively
with respect to water temperature and inorganic
sediment, including that derived from surface erosion
and mass wasting. The handbook procedures also
allow qualitative assessment of nutrients, dissolved
oxygen and organic matter, and introduced chemicals.
Underlying these analyses is an assessment of the
hydrologic cycle, including possible modifications from
the silvicultural activity.
The handbook is process-oriented. It leads the user to
consider the kinds and magnitude of on-the-ground
changes that may result from the activity or use. Those
changes are then assessed in terms of how they modify
soil detachment and infiltration potential, volume
and concentration to surface runoff, and similar factors
that affect the pollutant generating and transport
processes.
The handbook also contains a section on control
opportunities. Controls described in the WRENS
document are related to the modifications to vegetation,
soil cover, drainage patterns, and other factors that
may be affected, not only by the type of activity, but also
by where, when, and how it is carried out. Thus,
suitable preventive and mitigative alternatives for the
specific site characteristics and conditions created
can be identified.
The WRENS project has revealed that, despite the
considerable knowledge gained from research, many of
the linkage mechanisms needed to develop a fully
integrated predictive system suited to quantitative analy-
ses are still missing. For example, documented
procedures for linking soil erosion rates to instream
sediment loads are still incomplete. The handbook
presents theoretically sound techniques to estimate values
for the major mechanisms that are missing; however,
some of these techniques have not been thoroughly
tested. In addition, it should be noted that the handbook
covers only the technical aspects of nonpoint source
water pollution control. It does not address the
economic, social, and institutional considerations
that also form an integral component of Best Manage-
ment Practices.
Other Studies
Colorado State University is evaluating the effective-
ness of buffer strips along streams as a candidate forest
BMP for the control of direct surface runoff related
pollutants. This project will specify the abilities of buffer
strips to retain sediment and sediment-related pollu-
tants as a function of sediment properties and buffer ties
and buffer media, based on hydraulic and sediment
transport theory.
The U.S. Forest Service and Colorado State Univer-
sity are cooperatively developing a general resource
planning model for the evaluation of forest management
alternatives. In the forest management phase, the
U.S. Forest Service is identifying those components
of forest management that must be understood and
predicted in order to evaluate the environmental impact
of management options. These components include
timber growth, cutting systems and road layout, nutrient
cycling, and site productivity. Models for predicting
the influence of management alternatives on the forested
watershed and habitat itself also will be developed.
In the engineering phase, Colorado State University
is developing a predictive model of the watershed re-
sponse to precipitation for a forested watershed for
which biological/physical properties are defined (based
on the forest management phase). This will be accom-
plished via further development and refinement of
existing conceptual transport models forwater, sediment,
and associated pollutants (organics, nutrients, and
temperature). An optimization routine (goal program-
ming) will be developed to link the management model
to the transport model to enable evaluation and deter-
mination of the best integrated management system
for forestry within a watershed, taking into account envi-
ronmental constraints.
At the state level, much of the recent work on forestry-
related nonpoint source pollution has been initiated as
part of the nation's Section 208 planning program. In
Oregon and Washington, for example, state envi-
ronmental regulatory agencies have been assessing their
existing regulations concerning management practices
and the applicability of these regulations toward meeting
the Federally mandated requirement for establishing
BMP's.
A demonstration project by the Washington State
Department of Ecology as part of its 208 program
was designed to evaluate the effectiveness of an existing
regulation program as compared with development
of a completely new program. This project also sought
to further define the extent of existing and projected
nonpoint source problems, using a combination of
primary research and subjective assessments.
43
-------
Dr. George W. Brown of Oregon State University,
who chairs the Forest Practices Act Technical Work
Group, conducted an informal survey of the recent and
ongoing studies of silviculture's impact on water quality
in the Pacific Northwest. Results of his survey, as '
shown in Table 17, show about $2.3 million being
spent on studies to be completed over the next 3 to 5 years.
Most of this research is Federally financed. The
largest individual budgets are for Forest Service Experi-
ment Stations, which also provide additional grant
support to universities. Other monies come from the
Office of Water Resources Technology (OWRT) or
from EPA through state agencies in Idaho or Wash-
ington as part of their 208 effort.
Sedimentation is the primary pollutant being studied.
Of the 52 projects listed, 35 deal specifically with
erosion, sedimentation, and the impact on aquatic
habitat The predominance of interest in sediment results
from the complexity of the process, the number of
sources available, and wide variation in processes
within the region.
Detailed descriptions of the studies by the program
leaders makes additional generalization possible. Most
of the studies are field investigations; i.e., the research
focuses on obtaining basic data from field experiments,
rather than modeling. There is a fairly common con-
sensus that existing data is insufficient to derive meaning-
ful models.
There has been a great deal of coordinated program!
planning between research organizations. This is
to be expected when one organization, such as the
Forest Service, carries the major share of the research
load. But it is important to recognize that the universities
also participate in this planning effort. Slope stability
research, at Oregon State University for example,
is being done jointly with the Pacific Northwest and
Pacific Southwest Forest Experiment Stations.
The majority of the field research is done as watershed-
level experiments. This is particularly true of the work
by the Forest Service Experiment Stations and the work
at the University of Washington. Watershed-level
research is expensive because of needed long-term
calibration periods prior to treatment, travel to remote
locations, and data processing.
It is interesting to note that most of the research
presently being conducted falls into two categories:
predicting behavior of physical systems and predicting
man's impact on the system. These two categories are
essential building blocks for further research areas,
such as: predicting response of the aquatic system and
obtaining better decision models. Much work remains to
be done to link these categories of research together.
Needed Research
Further research will be needed to develop more
meaningful information on which to base regulations
for improved forest management practices for water
quality control. This research can be categorized into
four areas:
• Predicting the behavior of natural physical systems.
« Predicting man's impact.
• Predicting the response of aquatic biota.
• Developing better decision models.
Predicting Behavior of Natural Physical Systems
The major research need in this category continues
to be prediction of sediment transport in small streams.
There are still no acceptable methods for determining
natural or background levels of suspended sediment,
bedload, or turbidity. The problem is one of coping
with high variability. Unless researchers are able to
overcome this problem, rational development of water
quality standards, sampling methods to judge compliance,
and evaluation of compliance will be extremely diffi-
cult. A second problem is to predict erosion, whether
it be surface erosion in semi-arid range and forest water-
sheds or mass wasting in mountainous terrain. While
several estimates of erosion rates have been obtained
from watershed studies, there are no predictive models
that relate erosion to fundamental processes. Without
such models, rational, effective methods for control
will be impossible to develop, as will development of
hazard or risk rating techniques to guide managers.
Predicting Man's Impact
As mentioned previously, watershed studies have
produced several point estimates of silvicultural impacts
on water quality. But good techniques do not exist
for extrapolating these estimates very far spatially or
for predicting the impact of new technology or methods.
Part of the problem is that the variability described
is compounded by the additional variability imposed
by man. But even more fundamental is the fact that most
watershed studies have not been designed to yield
good estimates of water quality changes. They have
focused on erosion of soil and have provided estimates
of soil loss in tons per acre per year, not concentration
values so necessary for judging water quality impair-
ment. As a result good predictive methods for judging
silvicultural effects are still lacking.
Predicting Response of Aquatic Biota
Forest hydrologists agree that there is a major research
need for better methods of predicting the response
of aquatic organisms to altered water quality, especially
from sediment. However, in predicting responses, diffi-
culties result from the large array of highly variable
organisms that must be evaluated in a fluctuating envi-
ronment. At a 1978 EPA workshop on sediment and
water quality, researchers31 noted that "the sources,
44
-------
Table 1 7.'< Summary of Present Research in the Pacific Northwest to Control Nonpoint-Source Pollution from Silviculture
Research Organization
(Source of Information)
Study Topics
Total
Budget ($)
Source"
Humboldt State University
X-ray Diffraction Techniques for Determination of Suspended Sediment Source
Areasb
(not provided)
University of Idaho
(Dr. George H. Belt)
Oregon State University
Pacific Northwest Forest and
Range Experiment Station,
Corvallis, OR
(Dr. Logan Norris)
Pacific Northwest Forest and
Range Experiment Station,
Wenatchee, WA
(Dr. Arthur R. Tiedemann)
Pacific Southwest Forest and
Range Experiment Station,
Arcata, CA
(Dr. Raymond M. Rice)
Hydrologic Properties of Three Coast Range Soils'5
Scour and Fill of Spawning Gravels
Best Management Practices for Silviculture11
Road Standards and Water Quality*1
Influence of Topographic Characteristics on Sedimentation from Metamorphosed
Granitic Soilsb
Stream Protection During Harvesting15
Overland Flow Following tractor Harvesting
Sediment Transport and Deposition in Small Streams'1
Dissolved Oxygen Exchange in Turbulent Streams'1
Forest Roads and Slope Stability
Streamflow Characteristics as Influenced by Weather and Land Management
Stream Sedimentation in Small Watersheds as Influenced by Natural Events
Stream Water Quality (Natural Chemicals)
Plant Succession on Harvested Areas
Subsurface Hydrology Effects on Slope Stability
Soil Mantle Creep
Role of Plant Roots in Maintaining Soil Strength
Relation Between Mass Soil Movements, Land Management Practices, and
Intense Storms :
Movement, Persistence, and Fate of Herbicides in the Forest
Behavior of TCDD in the Forest and Aquatic Environments
Entry of Fertilizer to Forest Streams
Persistence of DDT in Eastern Oregon and Washington
Behavior and Impact of Fife Retardant Chemicals in Forest Environments
Role of Streamside Vegetation in Maintenance of Anadromous Fish Habitat
Role of Intermittent Streams in Anadromous Fish Production
Impact of Grazing Systems on Anadromous Fish Habitat
Role of Large Debris and Effects of its Removal on Fish Habitat
Factors Affecting Stream Carrying Capacity for Fish
Effects of Deforestation by Tussock Moth and Salvage Logging on Streamflow
and Quality
Effects of Harvesting Grand-fir on Soil and Streams
Harvesting and Nonpoint-Source Pollution
Effects of Sheep Grazing on Water Quality and Streams
Effects of Fire and Fertilization on Stream Water Chemistry
Effects in Recreation on Microbial Concentrations in Streams and Lakes
Effects of High Elevation Livestock Grazing on Microbial Concentrations
Rehabilitating Landslide Scars
Timber Harvest and Dissolved Organics
Channel Morphology and Sediment Transport
Large Organic Material and Fish Habitat
Clearcutting and Slope Stability
Logging and Erosion
Roads and Erosion
Clearcutting and Subsurface Hydrology
Timber Harvesting and Flood Peaks
Stream Sampling Methodology
(not provided)
15,000 USFS
100,000 Idaho
46,000 U. Idaho
15,000 U. Idaho
90,000
10,000
150,000
30,000
20,000
50,000
30,000
75,000
5,000
55,000
85,000
20,000
20,000
105,000
25,000
30,000
20,000
20.000
75,000
30,000
70,000
20,000
15,000
40,000
20,000/yr
30,000/yr
15,000
40,000
36,000
35,000
6,500
13,000
90,000
13,000
125,000
46,000
14,000
80,000
13,000
33,000
OSU
USFS
OWRT & Wey-
erhaeuser
OSU/OWRT
OSU
45
-------
Table 17. Summary of Present Research in the Pacific Northwest to Control Nonpoint-Source Pollution from
Silviculture—Continued
Research Organization
(Source of Information)
University of Washington
(Or, David D. Wooldridge)
Study Topics
Effects of Clearcutting on Dissolved Nutrients in the Olympics'3
Effects of Clearcutting in the Olympics on Streamflow and Water Qualityb
Effects of Newly Constructed Forest Roads on Water Quality
Log Hauling and Sediment Production13
Managing Buffer Strips'"
Total
Budget ($)
23,000
290,000
47,000
26,000
61,000
Source8
USFS
DNR, OWRT
OWRT, DOE
DOE
DOE
Washington State University
(Dr. Donald R. Satterlund)
Plant Indicators of Slope Stability
25,000
WSU
•Source of funds shown for University research projects only. j'
""Project lasts for more than one year; budgets indicate total funds allocated.
transport, and relationships of sediment with biological
populations will require substantial investigation |
before final workable criteria can be implemented. •
The role of sediment in aquatic ecology is highly variable
and the literature is deficient in attempts at synthesis
and in crossing disciplines."
Better Decision Models
Research is necessary to supply resource managers
with the information needed to make better decisions and
to predict the consequences of their actions. Planners
must be able to compare relative consequences of
alternative silvicultural practices and policy makers
need to be able to assess the costs of altering silvicultural
practices and the benefits to water quality.
Research programs should be structured to enable
decision makers to predict the interaction between ;
decisions in various resource areas. Multidisciplined |
research must seek to define the linkages between j
resources and how these are altered through manage-
ment decisions.
Much of the existing information about the impact
of silviculture on water quality comes from watershed
experiments, where results were measured in streamflow
or sediment at the outlet. The factors producing the
observed change usually are not quantified, leaving
researchers to make some general inferences about cause
and effect. Research must now take the next step:
integrated, process-level studies of system function.
For example, instead of measuring sediment concentra-
tions at the outlet of a small watershed after Clearcutting,
sediment production will be defined as a function of
soil disturbance, soil characteristics, terrain, and
climatic variables. The transport and deposition of sedi-
ment in the stream channel will be defined as a function
of the hydraulic factors involved. These conditions
then will be related to the variables that influence aquatic
organisms. From such activities, research can hope
to predict response to another silvicultural operation
in another watershed. This type of experimentation
is illustrated by the study conducted by the U.S. Forest
Service and Oregon State University on the Andrews
Experimental Forest.
Once general relationships are developed from
process-watershed studies, these relationships can
be used to test hypotheses on a large scale. No other
technique will do the job as well. Finally, the research
must include the full range of physical, biological, and
economic criteria by which decisions are judged.
46
-------
Chapter 11
Forest Chemicals
Several groups of chemicals used in forest management
can potentially affect water quality. These biologically
active chemicals fall into four broad categories:
fertilizers, herbicides, insecticides, and rodenticides.
These materials are used to (1) enhance productivity,
(2) focus the productive effort in desirable species,
and (3) protect the desirable species from consumption;
however, their use must be managed to prevent them
from inadvertently harming the aquatic ecosystem.
Chemicals and Use Patterns
The variations in patterns of usage and dosage of
major classes of chemicals used in forest management
affect the likelihood of encountering pollution problems.
Their biological impact in water is determined by the
range of biological activity and inherent chronic and
acute toxicities.
Fertilizers generally are very low in toxicity. The
forms used in forests are all found in nature; the
baseline levels in soils are usually much larger than
the amounts applied. Fertilizer poses a potential pollu-
tion problem only because the materials are soluble
at the time of application, leading to a brief period
during which nutrients may move into water. The likeli-
hood of contamination actually taking place is tied to
deposition directly in open water or to the occurrence
of high-intensity rainfall immediately after application.
When contamination does occur, it poses no special
water quality problems unless the water is trapped in
an impoundment where algal bloom can result from
an increase in nutrient concentration over an extended
period.
Fertilizers are applied at high rates but at low
frequency and in widely dispersed locations. No
large-scale contamination of river systems has been
encountered from fertilization. Forest fertilization
is also done in established stands, where nutrients
are utilized by vegetation shortly after application and
are not lost from the forest system. There is no evidence
that application of nitrogen fertilizers has ever led
to water concentrations of nitrate approaching the 10
ppm (nitrogen equivalent) water standard set by EPA.
As a result, there appears to be no special reason for
being concerned about fertilization nor any special
reason for modifying existing fertilization methods.
Herbicides are much more widely used than fertili-
zers in forests. These chemicals were quite specific
in their effects on plants, and some of the most widely
used materials in forests are registered for use in aquatic
weed control at much higher concentrations than are
encountered in forest watersheds.
Silvicultural herbicides usually are applied by
helicopter in small units no more than once or twice in
the development of a timber crop. Roughly 0.2 per-
cent of the commercial forest land in the United States
is treated in any given year.
Aerial applications of herbicides and their effects on
water quality have been studied extensively. At no
time in studies of silvicultural use of herbicides has
water contamination reached concentrations known to
affect the most sensitive aquatic plants or fauna. The
only contamination known to be potentially harm-
ful is that of picloram when it enters water upstream
from irrigated potatoes or tobacco, and thus far, no
damage to crops has been recorded as the result of
water contamination. Despite widespread publicity,
2,4,5-T and silvex do not pose a special problem,
even though they contain traces of TCDD (2,3,7,8-
tetrachlorodibenzo-p-dioxin). These materials have
been the subject of very sophisticated toxicology
research, and the technological base for their continued
use is stronger than that of any other pesticide.
Insecticides and rodenticides are of special interest
because of their ability to injure animals at low dosages.
For rodenticides, the amounts are so small and the
applications so confined to baits and burrows that
there is very little concerns for water quality. Insecticides,
conversely, are applied to large areas in a diffuse
pattern. The chance for incidental contamination of
waterways is, therefore, greater than for other types of
chemical applications. Moreover, the spectra of
activity of insecticides indicates that for most materials,
very low levels of contamination in water may result
in biological impact. The persistent organochlorine
compounds have the greatest potential for long-term
effects.
47
-------
Insecticides are subject to strict administrative
control. Most aerial application projects are coopera-
tively scheduled over areas large enough to bear the
overhead costs of monitoring and careful supervision.
Despite this, most of the insecticides in use are potentially
seriously harmful to fish or aquatic insect populations
if inadvertently applied directly to open water or if
spilled directly into a stream in significant quantities.
In summary, there is abundant evidence that aerial
application of insecticides rates the highest priority
for water pollution control in the use of silvicultural ;
chemicals. Among these, the organochlorine compounds
warrant special consideration to keep them from entering
waterways.
Water Quality Targets
The presence of trace contamination of a chemical
in streamwater does not imply that harm will result.
Aerial application of any chemical normally results
in minute quantities appearing in water for a matter
of hours, or perhaps a few days. The alternatives to such
applications, however, often have impacts on water
quality that last for much longer periods, in the form
of siltation, large dumps, or organic litter into stream
channels, and so forth. Some of these impacts have ••
serious implications for water users as well as aquatic
fauna. It is imperative, therefore, that rules be estab-
lished for safe use of chemicals. Safety is ensured by pre-
venting biologically significant amounts of chemicals
from entering water. Cleanup is clearly impractical.
The degree of tolerance by known species of
aquatic organisms or water users to concentrations
of chemicals in water must be known before rules gov-
erning the safe use of chemicals can be determined.
Secondly, operating guidelines must ensure that
water quality standards are not violated while land
management goals are being met. ;
The toxic principles of chemical action determine
the approach taken in setting limits on water concentra-
tion. Some chemicals are acutely toxic, meaning that
they produce symptoms quickly or not at all. Some
are chronically toxic, meaning that symptoms are
likely to be delayed until deposits are accumulated, or
until some metabolic function has been decreased
causing detectable changes to occur. In general,
the persistent compounds, especially the fat-soluble
organochlorine insecticides, are the most likely to be
chronically toxic. Virtually all of the herbicides and
organophosphorus insecticides are in the acutely toxic
category. These chemicals are usually eliminated
quickly and non-lethal effects are transient. The
principal concern for the acutely toxic materials is short-
term evidence of lethal or severe intoxication, whereas
chronically toxic materials must be evaluated over
much longer periods and must be studied for signs !
of accumulation through food chains. Food chain magni-
fication appears to be largely a function of fat solubility
and is not a problem for pesticides other than organo-
chlorine insecticides presently registered for use in
forests.
Chemicals also can be classed according to their
degrees of toxicity. Acutely toxic does not imply
a high degree of toxicity, but merely that toxic symptoms
show up quickly, if at all. Degree of toxicity is an inherent
property of a compound once it has entered into a
metabolic system. Acutely toxic materials are evaluated
according to acute oral feeding or exposure levels
that produce some measurable symptom in a population
of test organisms. Typical determinations for rodents
are lethality tests, in which a dosage that kills half the
animals is known as the LD50 (lethal dose for 50
percent of a population). Typical determinations for
fish is exposure to water having various concentrations
of toxicant at various life stages. A typical expression
of toxicity is LC50 (lethal concentration for 50 percent
of a population).
Test data for most of the silvicultural chemicals
is sufficient to determine at which levels of water concen-
tration one can anticipate harmful effects on aquatic
insects, fish, plants, and on animals using the water
for drinking. Available data also shows at which point
chemicals are likely to affect irrigated crops, either by
directly affecting the crop, as with a herbicide, or
by depositing an illegal residue.
All tests contain an uncertainty factor determined
by random variation within test organisms. There
is also uncertainty resulting from using one species of
test organism to draw inferences about responses
of others. Test data for many species demonstrate the
degree of variation among species and show which
groups are the most sensitive. Having an array of
data decreases the likelihood of overlooking potential
effects on any major group and virtually eliminates the
likelihood of human or fish sensitivity remaining
undetected. Levels of exposure where no effect occurs
can be ascertained with adequate precision with these
methods, especially for the acutely toxic substances.
Maximum concentrations for all silvicultural
chemicals other than dinoseb were established for three
classes of streams and for irrigation or potable use.
These target maxima were based on a substantial
margin of safety below the lowest concentration known
to affect any organism likely to be exposed. Data for
insects, fish, birds, and mammals were considered.
Target water quality standards were given safety factors
that provide for maximum exposures 10 to 1,000 times
lower than the lowest concentrations known to have
caused injury to fauna. The safety margin provides
much larger factors for chronically toxic and persistent
chemicals than for acutely toxic and quickly degraded
materials.
Table 18 lists the recommended target standards
for silvicultural chemicals. It provides for a graduation
in allowable concentration downward with increasing
size of stream and discriminates between potable
standards, which provide for the safety of all aquatic
organisms, and irrigation standards, which take into
-------
Table 18. Recommended Concentration Maxima for Silvicultural Chemicals by Stream Class and User Group. Potable
Waters Include Safety Factors for Wildlife and Aquatic Organisms as Well as Humans.
Class Chemical
Fertilizer Nitrate
Phosphate
Herbicide Amitrole
Ammonium ethyl
Carbamoyl
Phosphonate
Arsenicals
(Organic)
Dalapon
Dicamba
Dinoseb
Picloram
Silvexb
Triazines
2,4- Db
2,4,5-Tb
TCDD
Insecticide Carbaryl
Diazinon
Disulfoton
Endosulfan
Endrin
Fenitrothion
Guthion
Lindane
Malathion
Phosphamidon
Trichlorfon
Most Sensitive
Test Species Test Basis &
Affected Concentration
Man
Algae
Daphnia
Bluegill
Man
Daphnia
Bluegill
Bass
Chinook
salmon
Daphnia
Bluegill
Bluegill
Coho salmon
Stonefly
Daphnia
Stonefly
Rainbow trout
Coho salmon
Atlantic
salmon
Stonefly
Brown trout
Daphnia
Daphnia
Stonefly
No effect.
1 0 mg/l
Growth'
response var.
LC50 48 hr.
3 mg/l
LC5Q 48 hr,
670 Mg/l
No effect.
0.12 mg/l
LC50 48 hr.
1 1 .0 mg/l
LC50 96 hr.
23.0 mg/l
!
LC50 48; hr.
19.?! mg/l
LC50 48: hr, ,
1 .2 mg/l
LC50 48^ hr.
1 .0 mg/l
LC50 48 hr.
1.0 mg/l
LC50 48 hr.
1 .4 mg/l
No effect
96 hr,
0.000000056
mg/l
LC5048hr,
0.0048 mg/l
LC5049;hr,
0.0009 mg/l
LC5048;hr,
0.005. mg/l
LC6096:hr,
0.0003 mg/l
LC50 96 .hr.
0.0005 mg/l
Behavior" test
1 mg/I
LC50 96 hr,
0.0015 mg/l
LC50 48 hr.
0.002 mg/l
LC50 96 hr,
0.0018 mg/l
LC50 48 hr.
0.0088 mg/l
LC50 96 f,r.
<1 0 cfs
Potable
10a
Irrig.
10a
Criteria, ppm 24-hr
Mean Stream Class &
User 10 cfs- Navigable
Navigable
Potable
10a
inadequate basis
0.15
5
0.1
0.5
0.2
0.5
0.06
0.05
0.05
0.06
0.1
5
0.1 a
0.1
0.004
inadequate
0.001
0.02
0.05
0.05
0.02
0.03
1
0.05a
0.1
0.05
data
0.05
0.03
0.03
0.05
0.03
Irrig.
10a
Potable
10a
Irrig.
10a
for recommendation
0.01
1
0.1 a
0.02
0.002
0.0005
0.02
0.03
0.02
0.02
0.000000006 for all
0.001
0.0001
0.001
0.00003
0.00005
0.025
0.0003
0.0001
0.0005
0.0005
0.002
0.001
0.0001
0.001
0.00003
0.00005
0.025
0.0003
O.OOO1
0.0005
0.0005
0.002
0.0005
0.00005
0.00025
0.00001
0.0000 1a
0.01
0.0002
0.00005
0.0002a
0.0005
0.0005
0.0005
0.00005
0.00025
0.00001
0.015
0.5
0.05a
0.10
0.01
0.005
0.01 a
0.01
0.01
0.01
water
0.0002
0.00001
0.00024
0.000003
0.00001" 0.000005
0.01
0.0002
0.00005
0.0002a
0.0005
0.0005
0.005
0.00007
0.0000 1a
0.0001
0.0002
0.00005
0.01
0.5
0.1 a
0.02
0.001
0.0001
0.01 a '
0.01
0.005
0.01
0.0002
0.00001
0.00025
0.000003
0.000005
0.005
0.00007
0.00001 '
0.0001
0.0002
0.00005
"As listed in QCW.
bThe phenoxy herbicides may occur in water as esters or other forms. The given criteria for potable water may be increased by a factor of 10 for forms
other than esters. Criteria for irrigation use are for total phenoxy herbicide.
49
-------
account the special sensitivity of certain crops to some
herbicides. The rationale for decreasing levels with
increasing size of stream is that concentration peaks
move more slowly downstream in large streams than in
small creeks, and more total exposure occurs in larger
watercourses having a specified maximum observed
concentration.
It is noteworthy that insecticides, as a group,
have much lower tolerance limits than herbicides in
potable water. The differences between these groups are
much greater than the relative differences in nominal
application rates. For this reason more attention
must be given to application methods for insecticide
work than for herbicide.
Chemical Behavior in Forest Applications
Water contamination can result if chemicals are
applied directly to water or if runoff later carries them
to streams. For those chemicals with post-treatment
mobility, wide buffer zones help provide for maximum
Ueup in soils and organic matter prior to runoff entry
into streams. For those chemicals that do not move
readily, and this includes nearly all pesticides, the '<
critical factor is in preventing direct application to water
at levels exceeding the accepted criteria.
The limited mobility of most forest chemicals
is attributable to their tendency to adsorb to organic
material and soil colloids. Most forest soils have rela-
tively high cation exchange capacities and low base
saturations. This opportunity to fix pesticides in-situ
is substantially greater than actual amounts applied.
As a result, most chemicals never penetrate forest
soils substantially below the duff layer. Forest soils
normally have high infiltration rates, so surface runoff of
chemicals in solution is rare. In short, little migration
occurs in solution, either through soil or over it.
Disturbed soils are far less stable than those without
recent history of machine activity. Scarified or tilled ]
soils subjected to intense rainfall are capable of
losing many tons of silt per acre in a single storm. When
such soils are treated with a chemical, the adsorbed
materials will move in association with the silt. The
degree to which adsorbed chemicals affect stream life
is related to the quantity reaching the stream, decreased
by the tendency for the material to remain attached
to soil. Indirect contamination of this type is more
difficult to evaluate than direct contamination, because
the silt tends to form deposits that continue to release
small quantities of contaminant as desorption occurs.
In particular, this process is especially critical when the
chemicals are absorbed and retained by stream biota over
an extended period. The persistent organochlorine com-
pounds, and endosulfan used on Christmas trees
in particular, warrant close attention in this situation.
Silt mobility also is a consideration in the evaluation
of nonchemical alternatives for vegetation control.
Direct application to open water accounts for most
stream contamination. Accuracy of aircraft guidance,
and technology of nozzles and solvent systems are
the most useful controls over direct placement of
chemicals in water. This form of contamination is a
brief concentration spike that cannot provide chronic
exposure. If the peak is not harmful, no effects occur.
Thus, elimination of harmful peaks is the first line of
water protection.
An aircraft releases chemicals through nozzle
systems that break the spray into droplets of various
sizes. Applications requiring heavy coverage and precise
targeting, such as herbicides, are delivered in relatively
high volumes of liquid, through large-orifice nozzles
that emit large drops. Conversely, insecticides are
generally applied in general treatments in which
very fine sprays are delivered over large areas. Very low
volumes of total liquid per acre are required for
effectiveness in large projects, yet a deficiency of drop-
lets per square inch of foliage can allow too many insects
to survive. The droplet size therefore is reduced to
increase the density of droplets per unit of foliage and
to increase the uniformity of coverage between aircraft
swaths. Unfortunately, the very technology that
contributes to effective insect control also enhances
the difficulty of precise targeting of the spray. Thus,
effective insecticide applications near streams are likely
to deposit significant amounts of material in the water.
Except for certain of the most selective of the insec-
ticides and biological agents, such deposits are likely
to have some effects on aquatic insects or fish, depending
on the specific chemical.
Herbicides are more amenable to technical spray
modification without loss of effectiveness. Spray
nozzle configurations that increase uniformity of droplet
size and decrease the proportion of fine droplets with
high drift potential are available. Spray thickeners,
emulsification agents, and foams may be used to decrease
fine droplet movement away from the target zone. The
use of helicopters, rather than fixed-wing aircraft,
also improves precision.
For many species of vegetation requiring control,
there are several herbicides available. Some herbicides
usually have lower impact on aquatic systems than
others, and unusually sensitive areas may be treated
with some of them by helicopter without having direct
impact on water quality. There are some opportunities for
substitution of insecticides, as well, but margins of
selectivity are not as great.
Herbicides applied with conventional cone-nozzle
systems delivering 10 gallons (0.038 m3) of water per
acre (4,04.6 m2) usually will show a rather precise swath
boundary. In the absence of wind drift, deposits
50 ft (15.24 m) from the edge of a spray project will
approximate 5 percent of the nominal application
rate, or less. Insecticides, on the other hand, are applied
from greater distances above the canopy of vegetation
in smaller droplets, their swaths are wider, and swath tails
extend further from the target boundary.
The width of a chemically free buffer zone along a
stream is often expressed in terms of swath widths
so
-------
of the aircraft. The effective swath width (ESW) of a
spray aircraft is the maximum distance permissible
between successive swaths without having a measurable
decrease in dosage between swaths. The actual swath
width is much wider, in view of the movement of fine
droplets to distances of several boom lengths either
side of the aircraft. In practice, every swath consists of a
principal application centered on the flight line, plus
minor deposits from adjacentswath tails. The elimination
of such tails improves ability to avoid minor deposits in
water.
Guidelines for Protection of Water Quality
Table 19 provides alistofforestmanagementpractices
involving the application of chemicals and outlines
the rules for buffer strip treatment and monitoring
to meet the water quality and productivity goals.
Methods used to reduce impact of chemicals (Priority I)
include designation of buffer zones of widths in accord-
ance with the potential hazard posed by the chemical.
The rationale behind recommendations for buffer-
strip widths is based on the earlier described 20-fold
decrease ofcontamination with each herbicide swath
width away from the stream and five-fold decrease
per swath of low-volume insecticides with winds less
than 5 mph (8.04 kmph). Based on experience with
various pesticides, the proposed criteria for water
concentrations will be met with a margin of safety when
registered rates of application are applied as recom-
mended. In those exceptions where buffer strips are
defined in terms of absolute width, the problem being
addressed is the physical movement overland or through
the soil in subsurface flow, a group of processes not
affected by application technology.
To achieve the second priority, meeting forest produc-
tion goals without compromising water quality, identi-
fication of practices that have adverse impacts near
water and substitution of less harmful practices is
emphasized. Exceptional conditions under which un-
treated buffer zones are recommended are identified
so that unnecessary loss of productivity can be
avoided.
Monitoring will be needed to insure that the recom-
mendations are (1) being observed and (2) effective
in maintaining water quality. Monitoring for validation
of practices will be the responsibility of state and
Federal water resources agencies, and operational
quality control will be the responsibility of the operator.
Monitoring of insecticides by users, in particular,
will be necessary on a limited scale to provide a record
of the consequences of chemical activity at the point
of maximum potential trouble. The intensity of monitor-
ing is specified in Table 19.
In conclusion, biologically important direct impacts
from herbicides, rodenticides, and fertilizers will
not occur when used as prescribed. Virtually all com-
mercial forest lands adjacent to streams may be managed •
in ways that include the use of such chemicals without
impairing water quality. Insecticides also may be
used, but substantial buffer zones and greater operational
quality control and monitoring are in order. By follow-
ing these rules, management goals may be achieved
without major extra cost and without resorting to non-
chemical tools having adverse impacts.
Table 1 9. Guidelines for Applying Chemicals by Aircraft, and Water Monitoring in Silvicultural Practices
Practice
Chemical Used
Minimum Distance Between
Nearest Water and Center
Line ,of Nearest Swath
Treatment of Buffer
Suggested Location and
Frequency of Water
Sampling
Fertilization Urea
Phosphorus
Forest Site Preparation Amitrole
Ammonium ethyl
Carbomyl phos-
phonate
Atrazine
Dalapon
% of an effective swath width Apply by ground rig.
(ESW).
% ESWa Exceptions: up-
stream from lake or im-
poundment.
Vz ESWa Exceptions: within
a mile1 of potable users,
50-foot buffer.
ESW3
i ESWa Exceptions: scarred
areas, SO feet.
ESW
Apply by ground rig.
(1) Apply by ground rig.
(2) Apply substitute chemical.
(3) Plant buffer zone with
tolerant tree species.
Can be treated.
Do not disturb soil within
buffer zone.
Do not disturb soil within
50 feet of creek.
Composite, Day 1, at potable
user site, if within 1 mile
downstream from project.
None
Composite, Days 1 & 2, at
potable user site if within
1 mile downstream.
Composite, Day 1 at potable
user site if within 1 mile of
project downstream.
None
None
51
-------
Table 19. Guidelines for Applying Chemicals by Aircraft, and Water Monitoring in Silvicultural Practices-Continued
Practice
Chemical Used
Minimum Distance Between
Nearest Water and Center
Line of Nearest Swath
Treatment of Buffer
Suggested Location and
Frequency of Water
Sampling
Phenoxys
Picloram
ESW
100 feet (200 feet when
applied during period of
rainfall surplus).
Can be treated. Composite, Day 1, at intake
if potable user within 1
mile of project downstream.
Can be treated with substitute Composite, weekly at irriga-
chemical within prescribed tion user if within 5 miles
limits. of project, and crops in-
clude potatoes, tobacco or
legumes. Sample after
spraying, again in
sequence after effective
rainfall.
Forest Insect Control:
Biological
Chemical
Bacillus thuringiensis None3
Nuclear polyhedrosis None3
Carbaryl
Diazinon
Disulfoton
Endosulfan
Fenitrothion
Guthion
Malathion
Phosphamidon
Trichlorfon
1 ESW3 or 100 feet, which-
ever is greater.
Roden Control(Seeding):
Chemical Endrin
Same as Carbaryl.
Same as Carbaryl.
4 ESWa or 300 feet, which-
ever is greater.
1 ESWa
3 ESWa or 200 feet, which-
ever is greater.
3 ESW3 or 200 feet, which-
ever is greater.
i
1 ESW" or 100 feet, which-
ever is greater.
1 ESW3 or 100 feet, which-
ever is greater.
%ESW
Can be .treated.
Can be treated.
May treat with biological
agent.
Same as Carbaryl.
Same as Carbaryl.
May treat with carbaryl diazi-
noh, fenitrothion or phos-
phamidon to 1 ESW from
water.
Same as Carbaryl.
Same as Carbaryl.
Same as Carbaryl.
Same as Carbaryl.
Same as Carbaryl.
Can be treated by hand.
None
None
Composite each day of spray-
ing immediately downstream
from project and above
potable user, and 2 days
after. Sample at water in-
take, if within 2 miles of;
project. Filter samples.
Same as Carbaryl.
Same as Carbaryl.
Sample as with organophos-
phorus insecticides, but
sample also after each
heavy rain for next month.
Sam! as Carbaryl.
Same as Carbaryl.
Same as Carbaryl.
Same as Carbaryl.
Same as Carbaryl.
None
•For definition and discussion of ESW see page 55. Designation of "None" or V4 ESW under Buffer Strip Width implies only that buffer strip width is
•Mho discretion of the operator and that direct impact on water quality is not at issue. Even without a buffer strip, the aircraft should never be operated
within a half-ESW of streams that are likely to have fish in them at time of chemical application. For those insecticides requiring one or more effective
swath widths, the proposed buffers are for helicopters with droplet size of 200 ;U.MMD. |f droplets are smaller or large fixed-wing aircraft are used,
buffers should be 200 feet (61 m) plus the given swath numbers. .Helicopters may be used in conjunction with large aircraft.
52
-------
" Chapter 1 2
Sediment Contributions from
Southern Forest Management
Practices
Introduction
By the year 2000 the demand for wood from the 12
southern states is expected to double. Southern forests,
which now produce almost one-half of the nation's
wood, will need to produce 70 percent more softwood
and 40 percent more hardwood. This expected increase
in production will call for the conversion of 20 jnillion
acres (80 Gm2) of low-quality upland hardwoods to pine,
the reforestation of 10 million acres (40 Gm2) of un-
stocked or poorly stocked forest land, and the planting
or natural regeneration of lands being harvested. To
reach these goals, intensive management must be imple-
mented on more southern forest land. Concurrently,
the water resource must be protected or improved. Thus,
the primary challenge facing southern wood producers
is to increase production at competitive costs while
minimizing nonpoint source pollution—particularly
sediment contributions.
Baseline Levels and Water Quality Standards
The Clean Water Act requires that states review and
revise water quality standards and then develop water
management plans. To establish a baseline from which to
set these standards, the quality (natural background
level) of water flowing from undisturbed forest lands and
its natural variability must be determined. However,
the National Commission on Water Quality (1976)
has been unable to describe this natural water quality;
i.e., water quality at the time of the European coloniza-
tion of the continent. Thus, appropriate natural back-
ground levels for water quality are those produced by
present-day, not-recently-disturbed forest lands.
Past research has indicated a concentration of
0.0066 tons/acre-in (about 58 nig/1 or ppm) as the
average annual base rate for sediment in stormflows from
small, undisturbed catchments of southern pine. Average
annual sediment concentration appeared an adequate
expression of the natural background level because
concentrations were not correlated with individual
stormflow volumes or with annual or average annual
water yields.32 However, concentrations for individual
events, due to natural variations, may be higher by a
factor of 10 or more. The higher rates result primarily
from the occasional flushing of inorganic sediment
and organic debris accumulated in channel networks.
Data collected for 88 years from 10 catchments
supporting loblolly pine planted to control erosion
and mature shortleaf pine-hardwoods show that
sediment concentrations for the mature pine-hardwoods
(n = 30) did not differ statistically from those for the pine
plantations (n = 58). The overall mean for both types
was 0.007 tons/acre-in (61.51 mg/1). The fact that
pine established on severely eroding lands can, in
less than 20 years, reduce sediment concentrations to
levels of mature pine-hardwood types suggests a high
recovery potential after the initial rotation. It seems
possible to predict a prompt and complete recovery even
after severe disturbances.
To provide an interim reference, the figure 0.00"7
tons/acre-inch can be considered as the natural back-
ground level for undisturbed, well-stocked southern
pine types in the Coastal Plain. The standard error of
this mean is ± 0.0008 tons/acre-in (7 mg/1). Because
of the rough terrain, highly erosive soils, and past abuse
of the catchments cited, 0.007 tons/acre-in may
be slightly higher than the mean for the southern Coastal
Plain pinery; however, this rate falls within the limits
of values for a wide range of undisturbed eastern forest
types.33'34-35'36
To further define natural background levels,
Stanley J. Ursic of the Southern Forest Experiment
Station compiled 22 years of data for each of two other
forest land cover types: (1) abandoned lands with a
dense cover of native grass representing lands requiring
reforestation to pine; and (2) poorly stocked upland
hardwood stands typical of those being converted to pine.
Basal area per acre of the hardwood stands averaged
52 ft2 (4.83 m2), of which 23 ft2 (2.140 m2) were cull.
Average annual sediment concentrations between the
two types did not differ statistically. However, the
average concentration for these covers differed signifi-
cantly from that of the pine types and represented
a discrete population of erosion potential. The annual
sediment concentration for the unstocked lands and
depleted hardwoods averaged 0.026 ton/acre-ih
(228.5 mg/1), almost four times the rate of the pine covers.
Results of studies at the Southern Forest Experiment
Station are confirming differences in sediment con-
centrations among forest land covers. Results forecast
53
-------
an improvement in water quality following conversion
of low-quality upland hardwoods to pine. Compared
to 5-year pretreatment means, average annual sediment
concentrations were reduced an average of 64 percent
during the fourth through the seventh years after hard-
woods on two of the small catchments cited were
deadened with chemicals and underplanted with pine
(Ursic, unpublished data). The long-term improvement
of water quality demonstrated in this study37 indicates
that initial impairment is only temporary and will be
more than compensated for over a rotation.
Channel Contributions I
Since there is little, if any, overland flow from well-
stocked forests, most of the sediment yield results from
channel erosion. Because of wide variations in channel
stability and conditions, sediment contributions from
channels should be evaluated apart from those caused
by forestry activities and should be considered when :
developing water quality standards to prevent plac-
ing needless restrictions on forest lands. On the
upper reaches of a drainage network, sediment losses are
largely a natural phenomenon, and channels in hilly ,.
terrain may drain just a fraction of an acre. But the
quality of water from forested lands in the Coastal Plain
progressively worsens as it continues downstream. ;
One main factor is the degrading of silted channels
and floodplains caused by previous farming. Channels
occupying 1 percent of a watershed area and eroding
at the rate of 0.01 in/yr (0.025 cm/yr) contribute
an equivalent of 0.02 tons (18.14 kg) per watershed acre
(4,046 m2), but Coastal Plain channels often eroded
at much higher rates. For example, the average
annual sediment concentration from an 88-acre
(356,000 m2) Coastal Plain catchment to rehabilitate
the loblolly pine was found to be 13 times as great as
the suggested base rate (0.007 tons/acre-in) for
small headwater areas. An additional 0.09 tons/acre-in
(790.9 mg/1) was contributed by the larger sandbed
channels. Sediment contributions from some Coastal
Plain channels that drain several thousand acres
in mixed land use may be so great as to obscure the effects
of any or all forest land management practices. r
Harvesting Impacts
Among harvesting options, clearcutting can have
the greatest short-term impact on nonpoint source
pollution. However, over the long term, total or average
annual impact over a rotation may be equal or greater
for a silvicultural system that includes more frequent
cutting. Where temporary roads are the primary
sediment source, frequent reentry could make the impact
greater and the nonpoint source control costs higher.
The cutting of trees has little direct effect on water
quality, since basic hydrologic processes are essentially
unaltered. Any adverse impact results primarily from
soil disturbances caused by removing the trees and
by the rapid oxidation of the forest floor following
removal of the canopy. Skidding causes the most serious
disturbances, but in the South, skidding seldom
disturbs more than 20 percent of clearcut area.38 In one
study, erosion was measured from 20 trails created by
uphill skidding of hardwood logs on a variety of soils,
slopes, and slope lengths.39'40 During the first year,
soil moved from the trails at concentrations averaging
0.124 tons/acre-in (1090 mg/1). Although soils were
severely compacted, vegetation rapidly invaded
the trails, and concentrations during the second year
averaged 0.023 tons/acre-in (202 mg/1) (a five-fold
reduction). These rates seldom apply to more than
20 percent of a clearcut area, and they represent soil
movement (i.e., a relocation on slopes) and not
nonpoint source pollution, which involves the delivery
of particulate matter into a channel. The bulk of these
deliveries can be prevented.
Similarly, rapid oxidation of the forest floor can result
in localized soil movement, but in the humid subtropical
areas of the South, invasion of herbaceous vegetation
quickly compensates for reductions of the forest
floor. In the skid trail study cited, bare soil on trails
was reduced from 30 to 11 percent during the first year
and to 3 percent after 2 years. The average amount of
bare soil on four small catchments of loblolly pine alter
clearcutting the tree-length skidding similarly declined
from 10 percent 1 year after logging, to 7 percent after
2 years, and to 5 percent after 3 years.
The Southern Forest Experiment Station, Forest
Hydrology Laboratory, has been conducting a
cooperative study on the Natchez Trace State Park
and on the Pine Tree Branch Watershed (TVA) in west
Tennessee. Using small catchments to obtain integrated
impacts of harvesting loblolly pine plantations, re-
searchers replicated four treatments twice at each
location. Treatments consisted of clearfelling and
planting with and without seeding an annual grass on
disturbed areas, strip-cutting on contour to remove
about three-fourths of the volume, and undisturbed pine
plantations (controls).
Landings and haul roads were excluded from this study
because of the small size of the catchments. Common-
sense practices were applied to create disturbances
that maximized on-site detention of rainfall and moving
sediment. Trees were felled away from the channels and
skidded tree-length, roughly on contour, to the nearest
spur ridge and then to the landing. Most importantly,
the channels were not disturbed. These measures
were taken with little, if any, increase in logging costs,
although they required working closely with the feller
and skidder operator.
From the first 2 years of the study 27 sets of runoff
events from the Pine Tree Branch watershed were
examined. The average discharge-weighted con-
54
-------
i
centration of sediment for the controls [0.009 tons/acre-
in (79 mg/1)] was somewhat higher than the suggested
base rate for pine types, partly as a result of disturbances
during installation of measuring devices. An average
was taken for the six harvested catchments, and this
average was compared to the mean for the controls.
Average sediment concentration from the harvested
areas for the 27 events was about six times higher than
the controls. Much of this increase was from one catch-
ment with a deeply incised channel. The increased
sediment delivery resulted from enlargement of small,
subsurface, pipe-like channels that dumped sediment
directly into the main channel. Average sediment con-
centration, excluding this anomalous catchment,
was three times that of the controls. It appears that, after
clearcutting pine, water quality is comparable to that
cited for unstocked lands and lands occupied by low-
quality hardwoods [0.026 tbns/acre-in (228 mg/1)].
These particular catchments represent an extreme
erosion hazard for Coastal Plain conditions. Had
practices that did not consider water quality been used,
sediment concentrations could have been considerably
higher. However, this preliminary data indicates that
southern pines in the Coastal Plain can be clearcut
without serious or long-lasting impairment to water
quality. Where mitigative controls are called for,, sedi-
ment concentrations after harvesting can be reduced
effectively by placing temporary, low-cost structures to
hold sediments in the small ephemeral channels. Small
brush dams, for example, effectively trap the bulk
of larger particles.41
Although temporary and permanent road systems
used to manage forested tracts have a high potential for
producing sediment, this problem can be handled
with current technology. Almost all serious or continuing
damage is preventable.42-43 It is a costly business, but
there are no easy alternatives except to design harvesting
systems requiring fewer roads.
Regeneration Impacts
Intensive mechanical site preparation, which is used
when converting hardwoods to pine or preparing sites
for pine regeneration, presents a serious nonpoint
source problem. Site preparation practices include tree-
crushing, root-raking, chopping, discing, bedding,
chipping, and other techniques used singly or in combina-
tion. Some restrictions may be needed, because such
practices can bare three-fourths or more of the soil.
This kind of exposure and soil compaction can result
in sediment-producing overland flows and in large
increases in stormflow volumes. However, the magni-
tude and duration of the impacts of such practices
have not been adequately quantified.
The Southern Forest Experiment Station Forest
Hydrology Laboratory is currently measuring the
impacts of several widely used site preparation practices.
As expected, stormflow volumes have increased, but
the effect on sediment concentrations has not been ade-
quately defined. One study involving three methods
of intensive mechanical preparation indicates that
sediment yields during the first year after treatment are
comparable to those considered acceptable for agri-
cultural lands. Sediment concentrations after site
preparation are expected to decline rapidly; increased
water yields are expected to persist longer because of soil
compaction.
In a recent study of site preparation practices in the
North Carolina Piedmont, root-raking was com-
pared with broadcast burning. Root-raking, which creates
a greater nonpoint water pollution source hazard than
burning, reduced the site index of loblolly pine 14 ft
(4.26 m) and volumes at age 20 by 11 cords per acre
(.01 m3/m2). The chairman of a recent workshop on site
preparation concluded that the trend is toward less
intensive practices and that the overriding consideration
was not nonpoint source control, but rapidly escalating
operational costs.44
55
-------
-------
Chapter 13
Wood Preserving
Introduction
The wood-preserving industry in the United States
comprises about 400 commercial treating plants,
350 of which are equipped with pressure retorts. Most of
these plants are concentrated in two distinct regions.
The larger region extends from East Texas to Maryland
and corresponds roughly to the natural range of the
southern pines, the major species utilized. The second
concentration of plants is located along the Pacific Coast,
where Douglas fir and western red cedar are the species
of primary interest to the industry. Less than 25 percent
of the plants, and a much smaller percentage of produc-
tion capacity, are located outside these two regions.
Creosote in its various forms, pentachlorophenol-
petroleum solutions, and various inorganic salt formula-
tions based on copper, chromium, and arsenic are
the preservatives used by the industry. All plants use at
least one of the three and 80 percent employ two or
more in their operations. Many plants treat with two or
three preservatives and a fire retardant. Copper-8-
quinolinilate and tributyl tin oxide are also used by a
limited number of plants.
Stock may be treated'seasoned or unseasoned. In the
latter case, the wood is conditioned in the retort prior
to treatment. Both Boultonizing and steaming are, used
for this purpose, the former process being employed
principally with the West Coast species—notably
Douglas fir—and the latter with the southern pines.
Because of the relatively large volume of condensate
generated by these operations and its high phenol content
and oxygen demand, the conditioning processes account
for an important part of the pollution problem in the
industry. : *
Significant progress has been made by this industry
since 1972 in installing the technology required to meet
current BPT and BAT effluent guidelines. However,
the level of technology that will be required to meet
revised BAT discharge standards, including probable
restrictions on priority pollutants, is unknown at this
time. Wastewater treatment and disposal technology
currently used by the wood-preserving industry ate
summarized here. Technology not used but which may
have application to the industry or a segment thereof
is also discussed. :
Wastewater Characteristics
Wastewater characteristics of the wood-preserving
industry are unique among wood-base industries. This
uniqueness stems not so much from wastewater volume,
which generally is quite small, as from the number
and concentration of conventional and priority pollu-
tants that it contains.
Among the several factors influencing both the
concentration of pollutants and volume of effluent, total
conditioning time, whether by steaming or Boultonizing,
is the most important. Water from conditioning accounts
for most of the loading of pollutants in a plant's effluent.
With the exception of single-pass cooling water, con-
ditioning water accounts for the largest volume. There
is also a large volume of rainwater that falls on or in
the immediate vicinity of the retorts and storage tanks
(an area of about 1 acre for the average plant). Con-
taminated rainwater presents a treatment and disposal
problem at most plants, but can be especially trouble-
some for plants in areas of high rainfall.
Wastewaters resulting from treatments with inorganic
salt formulations are low in organic content, but
contain traces of heavy metals used in the preservatives
and fire retardants employed. The presence and
concentration of a specific ion in wastewater from
such treatments depend upon the particular formula-
tion and the extent to which the waste is diluted by
washwater and stormwater.
In addition to conventional wastewater parameters,
such as oil and grease and chemical oxygen demand,
EPA has identified 129 elements and compounds
(see Chapter 9, Table 14) that present a real or potential
health hazard because of their toxicity and actual or
suspected mutagenicity, tetrogenicity, or carcinogen-
ity. Designated priority pollutants, a number of these
materials have been identified in effluents from the
wood preserving industry. Except for the metals, which
are associated with inorganic salt-type preservatives
and fire retardants, most of the priority pollutants
are constituents of creosote.
The volume of wastewater discharged from a given
wood-preserving plant varies with the conditioning
method used, type of preservatives, the extent to which
contaminated and uncontaminated waste streams
57
-------
are segregated, the volume of water that is recycled, and
the volume of rainwater that falls on or in the immediate
vicinity of the retorts and preservative storage tanks.
Data from 18 Boulton and 41 steaming plants gave ,
average discharge values of 6,785 and 7,375 gpd
(25.6 and 27.9 m3/d), respectively. When only
western Boulton plants and closed-steaming plants were
compared, the average discharge volumes were, in order,
3,425 and 3,215 gpd (13.06 and 12.2 m3/d). Nation-
ally, it is estimated that the average two-retort plant
generates 8,000 gpd (30.3 m3/d) of wastewater, includ-
ing 4 000 gpd (15.1 mVd) of contaminated rainwater.
Boulton Conditioning Process
Preconditioning is accomplished in the Boulton
process by heating the stock in a preservative bath under
reduced pressure in the retort. The preservative serves
as a heat transfer medium. Water vapor from the
wood during the Boulton process passes through a con-
denser to an oil-water separator where low boiling
fractions of the preservative are removed and returned
to treatment solutions. The Boulton cycle may have a
duration of 48 hours or longer for large poles and piling.
The step may be accomplished either by pressure or non-
pressure processes. Figure 14 shows a Boulton wood
preserving plant. Steps in the process are described ,
in the following paragraphs.
• Water in Wood into Cylinder. The amount of water
that comes in with the wood depends upon the
moisture content of the wood. Green wood of small
diameter may have a moisture content in excess
of 100 percent on a dry weight basis. Larger diameter
wood contains heartwood, which has a moisture
content of about 35 percent. The stock may be
partially dry and therefore have a moisture content
between 35 and 100 percent.
Water Entering with Wood
= 20 lb/ft3 (320 kg/m3) for large diameter poles
[10 to 12 in (25.4 to 30.48 cm)]
= 32 lb/ft3 (512.6 kg/m3) for wood with 100-
percent moisture content
From an operational standpoint, the quantity of
water in the sapwood, the outer 2 in (5.08 cm) of the
wood, is most important. Conditioning mainly re-
moves water from the sapwood; little water is removed
from the heartwood.
Preservatives to Cylinder. The preservative is then
added to the cylinder and heated with steam coils at
the bottom of the cylinder. Steam condensate is re-
turned to the boiler.
' Vapors. After the oil has been heated, a vacuum is
drawn on the cylinder for 10 to 40 hours for Douglas
fir and 6 to 12 hours for oak, depending upon the
initial moisture content of the wood. The oil transfers
heat to the wood and vaporizes the water. Between
4 and 20 lb/ft3 (64 and 320 kg/m3) of wood are
removed.
• Air and Vapors. The vacuum pump removes mostly
non-condensible gases and water vapor from the sys-
tem. The quantity of water expelled is unknown.
Wood In
Cooling Water
Cylinder Drippings
ana1 Rainwater
Work Tank
Wastewater
Figure 14. Diagram of a Boulton Process Wood Treating Plant.
68
-------
• Preservatives to Work Tank. After the treatment
process, the preservative is drained from the cylinder
and transferred to the work tank. '
• Wood Out. The treated wood is removed from the
cylinder. The average moisture content of the stock,
which is dependent upon the original moisture con-
tent of water removed as a vapor, will be around
35 percent, or about 6 to 12 lb/ft3 (96 to 192 kg/m3).
The moisture in the sapwood will be between 2 and
4 lb/ft3 (32 and 64 kg/m3).
• Cylinder Drippings and Rainwater. Cylinder drip-
pings and rainwater often are added to the flow from
the cylinder and fed to the oil-water separator.
In some plants they are fed to a separate oil-water
separator to prevent cross-contamination of pre-
servatives. Rainwater can vary between 0 and 11.3
lb/ft3 (0 and 181 kg/m3) for conditions of no rain to
1-in (2.54 cm) rainfall in 24 hours, and depending
upon the area drained toward the treating cylinder.
This flow can be evened out by sufficient rainwater
holding capacity. The minimum area in which
rainwater is collected includes the immediate cylinder
area, the area where the wood removed from the
cylinder drips extra preservatives, and the. tank farm.
Rainfall will produce a yearly average rainwater
flow of 2 lb/ft3 (32 kg/m3).
• Recovered Oils. Oils recovered in the oil water
separator are returned to the work tank.
• Wastewater. The quantity of wastewater produced
depends upon the amount of green wood condi-
tioned and rainfall. Wastewater generated will be
between 3 and 22 lb/ft3 (48 and 352 kg/m3).
!
Current Treatment and Control Technology
With some minor exceptions, the wood preserving
industry has already installed, or has available to it, the
technology required to meet current effluent guidelines.
The level of technology that will be needed to meet new
BAT guidelines, which are presently being reviewed, and
to conform to probable restrictions on priority pollutants
is unknown at this time. If it exceeds that which is either
currently in use or has been indicated for the industry
by appropriate studies, a higher level of technology will
be required. Technology currently used by the industry
is summarized in the following paragraphs. This sum-
mary is followed by a review of technologies currently
unused, or used by a single plant, but which may have
application to the whole industry or a segment thereof.
Reduction in Volume of Wastewater
The volume of process wastewater generated by
the average plant that employs steam conditioning has
been reduced almost 70 percent—from 13,000 to
4,000 gpd (49.4 to 15.2 m3/d)—since 1973. Factors
that have contributed to this reduction include the
following:
• Adoption of closed steaming as a replacement for
open steaming. ':
• Replacement of barometric-type with surface-type
condensers.
• Recycling of barometric cooling water.
• Predrying of a higher percentage of production than
previously, thus reducing total steaming time.
• Segregation of contaminated and uncontaminated
waste streams.
• Inauguration of effective plant maintenance and
sanitation practices.
• Recycling of coil condensate.
• Recycling of process water.
Because cylinder condensate resulting from steam
conditioning is the most important source of wastewater,
the adoption of closed steaming, which permits this
waste stream to be recycled, has contributed most to
a reduction in total pollution loading and, at some plants,
wastewater volume.45 Predrying of stock preparatory
to preservative treatment, a practice that negates the
need to steam condition, has had a commensurate
effect at some plants.
Installation of facilities to recycle cooling water for
barometric condensers has reduced the total volume of
wastewater at plants using a single-pass system to a
fraction of its original volume. The practice of recycling
this water, or, alternatively, installing surface-type
condensers, was quickly adopted by the industry when it
became evident that it was not economically feasible
to treat the large volume of contaminated water gen-
erated by a single-pass, barometric-condenser system.43
Reuse of process water is not widely practiced in the
industry. There are, however, noteworthy exceptions
to this generalization. Process wastewater from
. salt-type treatments is so widely used as makeup water
for treating solutions that the practice is now common
industrywide. Of 184 plants treating with salts that
were questioned in 1974, 160 indicated that they
have achieved zero discharge of direct process waste-
water through a combination of water conservation
measures, including recycling.46 All except three of
the salt plants responding to a recent EPA questionnaire
stated that they had achieved a zero discharge.
Condensate from steam coils is one of the main
sources of uncontaminated water at wood-preserving
plants. While previously this water was frequently
allowed to become mixed with process wastewater, most
plants now segregate their waste streams, thus.re-
ducing the total volume of polluted water. Many plants
now recycle their coil condensate for boiler feedwater,
a use for which it is admirably suited because of its purity
and high BTU content.
Primary Treatment
Oil-Water Separation. Because of the deleterious effects
that oil has on all subsequent steps in a treatment regime,
efficient oil-water separation is basic to effective waste-
water treatment. Oil, whether free or in emulsified form,
59
-------
accounts for a significant part of the oxygen demand of
wood-preserving effluents and, in the case of wastewater
from pentachlorophenol processes, serves as a carrier
for concentrations of the chemical that far exceed its
solubility in oil-free water. In a real sense, control of
oils is the key to wastewater management in this industry.
API-type oil-water separators are extensively used
by wood-preserving plants for primary oil separation
from wastewater. This process is preceded and followed
at many plants by a rough oil separation and secondary
oil separation, respectively. The rough oil separation
occurs either in the blowdown tank or in a surge tank
preceding the API separator. Secondary separation
usually occurs in another API separator operated in
series with the first, or it may be conducted in any vessel
or lagoon where the detention time is sufficiently long to
permit further separation of free oil. A few plants |
achieve almost complete removal of free oils by filtering
the wastewater through an oil-absorbent medium.
This practice may be unnecessary if the wastewater is
to be chemically flocculated.
Oil content of wastewater entering the blowdown !
tank may range as high as 10 percent, with 1 to 5 percent
the usual range. Depending upon the efficiency of ;,
rough separation, the influent to the primary separator
will have a free oil content in the range of less than
200 mg/1 to several thousand mg/1. Removal efficien-
cies of 60 to 90 percent can be achieved, but the results
obtained are affected by temperature and oil content :
of the influent and by separator design, especially
with regard to detention time. Data published by the
American Petroleum Institute47 shows that 80 percent
removal of free oils is normal in the petroleum industry.
Secondary separation should remove up to 90 percent
of the residual free oil, depending upon the technique
used. i
Chemical Flocculation. Because oil-water emulsions
are not broken by mechanical oil-removal procedures,
chemical flocculation is required to reduce the oil :
content to a satisfactory level for wastewaters contain-
ing emulsions. Lime, ferric chloride, various poly-
electrolytes, and clays of several types are used in the
industry for this purpose. Automatic meteringpumps and
mixing equipment have been installed at some plants
to expedite this process, which is usually carried out
on a batch basis. COD reductions of 30 to 80 percent or
higher are achieved—primarily as a result of oil removal.
Average COD removal is about 50 percent.
Influent oil concentration varies with the efficiency
of mechanical oil separation and the amountof emulsified
oil. The latter variable in turn is affected by type of pre-
servative—either pentachlorophenol in petroleum,
creosote, or a creosote solution of coal tar or petroleum—
conditioning method used and design of oil-transfer
equipment. Pentachlorophenol preservative solutions
cause more emulsion problems than creosote or its
solutions, and plants that steam condition—especially
those that employ open steaming—have more problems
than plants that use the Boulton conditioning method.
Plants that use low-pressure, high-volume oil transfer
pumps have less trouble with emulsions than those that
use high pressure, low-volume equipment.
Typically, influent to the flocculation equipment
from a creosote process will have an oil content of less
than 500 mg/1, while that from a pentachlorophenol
process will have a value of 1000 mg/1 or higher. For
example, analyses of samples taken from the separator
outfalls at 10 plants revealed average oil contents of 1470
and 365 mg/1 for pentachlorophenol and creosote
wastewater, respectively. The respective range of values
was, in order, 540 to 2640 mg/1 and 35 to 735 mg/1.
Average separator effluents for three steaming plants
sampled in conjunction with a recent study gave oil
and grease (O & G) values of 1690 and 935 mg/1 for
pentachlorophenol and creosote separators, respectively.
Flocculated effluent generally has an oil content of less
than 200 mg/1 and frequently the value is less than
100 mg/1 (Table 20). Significant reductions in penta-
chlorophenol content (PCP) also are achieved by
flocculation. In some instances, the reduction in this
parameter is such that the final value is less than the^
solubility of the chemical in water—about 17 mg/1 at 68° F
(20°C). This result is attributed to the selective ad-
sorption of pentachlorophenol by the oil phase, which
is removed during flocculation.
Table 20. Reduction in Content of Selected Wastewater Parameters Due to Flocculation Treatments
Plant No
7
14
Before Flocculation
(mg/liter)
O&G
2,640
24,450
735
270
3,010
35
i COD
: 12,627
147,555
15,695
22,915
1 5,345
1,170
PCP
47.8
860
26.7
' 5.7
After Flocculation
(mg/liter)
O&G
165
90
245
45
120
20
COD
6,645
4,645
10,515
20,390
2,565
530
PCP
5.2
134
9.0
2.71
% Reduction
O&G
94
99+
67
83
94
43
COD
47
97
33
11
84
69
PCP
89
84
66
53
60
-------
Filtration. Many plants that flocculate their wastewater
filter it through sand beds to remove the sludge. When
properly conducted, this procedure is highly efficient in
removing both the solids resulting from the process as
well as small amounts of residual oil. The solids that
accumulate on the bed are removed periodically along
with the upper inch or so of sand. ',
A common mistake that renders a filter bed almost use-
less is to discharge incompletely flocculated wastewater
on it. The residual oil effectively prevents percolation
of the water through the bed, thus necessitating the
replacement of the oil-saturated sand. This has
happened frequently enough at some plants so that
the sand filters have been abandoned and replaced by a
decantation process to remove the solids. At many plants
decantation is part of the design of the flocculation
system. Solids removal is expedited by use of vessels
with cone-shaped bottoms. Frequently, the solids are
allowed to accumulate from batch to batch, a practice
that reduces the amount of flocculating agents required.
Secondary Treatment
Biological Treatments. Effluents from the wood pre-
serving industry are amenable to biological treatment.
Reviews of the literature on this subject were presented
in both the Effluent Guidelines Development Docu-
ment and in the Pretreatment Standards Draft Devel-
opment Document. Data presented in those documents
show that the conventional wastewater parameters
—COD, total phenols, oil and grease, arid suspended
solids—can be reduced to acceptable levels by treatment
regimes that include biological treatment. Furthermore,
some preliminary evidence indicates that flocculation
followed by biological treatments effectively remove
the priority pollutants from these wastewaters.
Considerably less data on pentachlorophenol re-
moval rates are available for industrial installations
than for other phenolic pollutants. The reason for this is
that there is no standard method for analyzing for this
chemical, and the methods available to the average plant
(e.g. the safranin method) are notoriously inaccurate.
Moreover, neither the effluent guidelines nor most
EPA discharge permits or state standards require
separate monitoring for pentachlorophenol. There is,
therefore, an information gap that can be filled only
in part by data from studies based on bench-scale equip-
ment and those from a single study that involved full-scale
equipment.48
Approximately 10 percent of the wood preserving
plants have biological wastewater treating facilities.
The number of facilities by type reported by a sample
in 1977 of 196 plants is as follows:
Trickling filter 1
Activated sludge 3
Aerated lagoon 10
Soil irrigation 8
Not included in this tabulation are several plants
that use oxidation ponds alone. While it is true that some
plants do obtain effective wastewater treatment with
oxidation ponds, this term has come into such misuse
that any body of water, including sumps, may at one
time or another be described as an oxidation pond.
The characteristics of the influent to a biological
treatment unit vary with the type and effectiveness of
pretreatment and amount of dilution water added to the
process waste. For dilution water, a few plants mix their
process waste with coil condensate, boiler blowdown,
etc., after flocculation and before biological treatment.
For this reason, it is difficult to compute treating
efficiency.
Average removal efficiency in 1977 of the treatment
systems for five plants for which both raw and treated
wastewater parameters were available is as follows:
Parameter Efficiency (%)
COD
Phenols .....
O&G
94.8
98.2
90.5
Average pollution loadings at the plant outfalls
are given in the following for seven plants:
Parameter Residual Loaf "9
(ib/1000ft3)
COD
Phenols
O&G
2.10
0.003
0.200
This data shows that current BAT standards can be
met by treatment regimes that include biological treat-
ment. However, there is a question of whether these
standards can be met consistently by single-phase
biological treatments. All of the plants for which data
is given in the tables employ biological treatment that
includes a high degree of redundancy. Multiphase
biological treatment, in combination with appropriate
primary treatment, may be required to meet current BAT
guidelines. Whether this level of treatment alone will
meet future criteria for priority pollutants is unknown
at this point.
The effluent quality that can be achieved by a multi-
phase treatment regime consisting of oil separation,
flocculation-filtration, and biological treatment is
shown in Table 21. A single-stage biological treatment is
assumed to be used. The type of biological treatment-
activated sludge, aerated lagoon, or soil irrigation—is
not really important, since the level of reduction shown
can be obtained using any one of the three.
Potential Treatment and Control Technologies
Technologies currently used or proposed for use in
industries related to wood preserving and which may
have application to that industry-are reviewed in the
following paragraphs. A sufficient body of data is not
available to provide more than an indication of the
61
-------
Table 21. Quality of Effluent from Each Stage of a
Multi-Phase Waste Treatment System
Raw
Parameter Waste
Load
Oil
Separa-
tion
Floccula-
tion and
Filtration
Bio-
logical
Treat-
ment
(mg/liter)
COD
Pentachlorophenol . . .
Oil and Grease
21,670
100
2,985
6,500
135
25
895
3,250
127
17
200
500
7.5
2.5
20
potential use of these methods with wood-preserving [
wastewaters. :
i
Chemical Oxidation
Chlorination and other forms of chemical oxidation,
are not economically viable treatments for wood- j
preserving wastewater when applied before or in lieu
of biological treatment. Inadequate dosage, with conse-
quent formation of chlorophenols where none existed!
previously, is another problem associated with
Chlorination (Table 22). It is reasonable to assume that
organic priority pollutants, in addition to phenolic '
compounds, can be adequately oxidized chemically !
Table 22. Chlorophenol Concentration in Creosote
Wastewater Treated with Chlorine
Residual
Ca(OCI)2 Phenols
as by
ECD Analysis |.
(mg/liter) "
pH Chlorine Alpha
(gm/liter) Method 2,4-Dichloro- 2,4,6-Tri-
(mg/liter) P^nol chlorophenol
4.S. . . .
7.0
9.5. . . .
0
0.5
1.0
1.5
2.0
3.0
5.0
0.5
1.0
1.5
2.0
3.0
5.0
0.5
1.0
1.5
2.0
3.0
5.0
438.5
256.1
30.8
0
0
0
0
300.0
101.5
7.7
0
0
0
315.4
101.5
11.5
0
0
0
-
161.0
0
0
0
0
0
122.0
0
0
0
0
0
198.0
0
0
0
0
0
-
910.7
6.7
1.5
il.O
0.3
D.3
316.0
35.0
|e.4
!2.8
;1.5
1.3
264.0
27.0
25.0
3.7
3.8
:.1'9
when adequate dosages are employed; however,
definitive data to support this supposition is not
available.
The use of cholerine and hypochlorites as a treatment
to oxidize phenols, in wastewater is widely reported
in the literature. A review of the pertinent literature,
with emphasis on the use of chlorine in treating wood-
preserving wastewaters, was published in a recent EPA
document.49 An unpublished literature review of
other chemical oxidants, including potassium permanga-
nate, hydrogen peroxide, and ozone, that have been
proposed for use with phenol-bearing wastewaters is
also available.50
Membrane System
This term refers to both ultrafiltration, which is
employed primarily to remove suspended and emulsified
materials in wastewater, and to reverse osmosis (RO),
which removes all or part of the dissolved substances,
depending upon the molecular species involved, and
virtually all of the suspended substances. Both technolo-
gies are currently used as part of the wastewater treat-
ment system of many diverse industries51'52 and may
have potential application in the wood-preserving industry.
Ultrafiltration treatment of oily waste basically
involves passingthe waste under apressure of 30 to 50 psi
(0.21 to 0.3 5 MPa) over a membrane cast onto the ins ide
of a porous fiberglass tube. The water phase of the waste
is forced through the membrane and discarded, reused,
or further processed by other means. The oil and other
solids not in solution remain in the tube. The process
in effect concentrates the waste. Volume reductions
on the order of 90 to 96 percent have been reported.52
Results obtained in pilot- and full-scale operations of
ultrafiltration systems have been mixed. Goldsmith
et al.52 operated a pilot unit continuously (24 hr/d)
for six weeks, processing waste emulsions containing 1
to 3 percent oil. The permeate from the system, which
was 95 percent of the original volume, contained 212
mg/1 ether extractables—primarily water-soluble
surfactants. A 4,000-gpd (15.2 mVd) system installed
based on the pilot plant data produced a permeate con-
taining 25 mg/1 ether extractables. No significant re-
duction in flux rate with time was observed in either
the pilot- or full-scale operation.
Ultrafiltration tests of a pentachlorophenol waste-
water were conducted by Abcor, Inc. in cooperation
with the Mississippi Forest Products Laboratory.53
The samples contained 2,160 mg/1 oil and had a total
solids concentration of 3,900 mg/1. Flow rate through the
system was 25 gpm (0.095 m3/m) at a pressure of 48 psi
(0.33 MPa). A 26-fold volumetric concentration,
representing a volume reduction of 96.2 percent, was
achieved. Two membrane types were tested. Both
showed a flux decline on the order of 55 to 60 percent
with increasing volumetric concentration. A detergent
flush of the system was found to be necessary at the end of
each run to restore the normal flux values of 35 gal/
ft2/d (1.4 m3/m2/d). Oil content of the permeate
62
-------
was 55 mg/1. While this value represents a reduction
of over 97 percent, it does not meet the requirements for
stream discharge. COD was reduced 73 percent.
The principle of RO is similar to that of ultra-
filtration. However, higher hydraulic pressures [400 to
600 psi (2.75 to 4.14 MPa)] are employed and the
membranes are semipermeable and are manufactured to
achieve rejection of various molecular sizes. Efficiency
varies, but rejection of a number of salts in excess of 99
percent have been reported.54 For organic chemicals,
rejection appears to be a function of molecular size and
shape. Increases in chain length and branching are re-
ported to increase rejection.55 Phenols are removed
to the extent of only about 20 percent by cellulose
acetate membranes, while polyethylenimine membranes
increase this percentage to 70 but achieve a lower
flux rate.52 In case studies that have been cited, RO was
found to be competitive with conventional waste treat-
ment systems only when extremely high levels of treat-
ment were required.56
Removals of 83 percent TOC and 96 percent TDS
were reported for RO in which cellulose acetate mem-
branes at 600 psi (4.14 MPa) were used.57 Flux rates
in this work of 34 to 3 6 gal/ft2/d(1.36 to 1.44 m3/m2/
d) were achieved. However, in other work, pretreat-
ment by carbon adsorption or sand filtration was found
to be necessary to prevent membrane fouling.58 Work
by the Institute of Paper Chemistry59 indicates that
membrane fouling by suspended solids or large molecular
weight organics can be controlled in part by appropriate
pretreatment, periodic pressure pulsations, and
washing of the membrane surfaces. In this and other60
work, it was concluded that RO is effective in concen-
trating dilute papermill waste and produces a clarified
water that can be recycled for process purposes.
Recycling of process wastewater, following ultra-
filtration and RO treatment, is the objective of a
treatment installation currently being constructed by
Pacific Wood Treating Corp., Ridgefield, Washing-
ton61 (also see Case Study, later herein). The concen-
trated waste will be incinerated and the permeate
from the system will be used for boiler feedwater. The
system cost is estimated at more than $200,000.
Data on the use of RO with wood-preserving waste-
water has been provided by the cooperative work
between Abcor, Inc. and the Mississippi Forest
Products Laboratory referred to previously.53 In this
work, the permeate from the UF system was processed
further in an RO unit. Severe pressure drop across the
system indicated that fouling of the membranes occurred.
Permeate from the system had an oil content of 17 mg/1,
down from 55 mg/1, and the COD was reduced by.
75 percent. Total oil removal and COD reductions in
both the UF and RO systems were 99 percent and 92
percent, respectively.
Carbon Adsorption
Adsorption isotherms have been developed for
wood-preserving wastes from several plants to determine
the economic feasibility of employing carbon adsorption
in lieu of conventional secondary treatments.62 The
wastewater used for this purpose was usually pretreated
by flocculation and filtration to remove oils. Theoretical
carbon usage rates obtained from the isotherms ranged
from 85 to almost 1000 lb/1000 gal (120 kg/m3) of
wastewater. With few exceptions, however, the
experience has been that while activated carbon (AC)
does an excellent job in removing phenolics from wood-
preserving wastewater, other organics, principally
water-soluble wood sugars, greatly increase carbon
exhaustion rates.
Use of AC to treat wastewater from a plant producing
herbicides was described by Henshaw.63 With the
exception of wood sugars, this waste was similar to wood-
preserving effluents, especially as regards COD
(3600 mg/1) and phenolic materials (210 mg/1). Raw
wastewater was piped directly to an AC adsorber and
the carbon was regenerated thermally. Flow rate and
loading rate were not revealed, but the effluent from
the system had a phenol content of 1 mg/1. Cost of the
treatment was reported to be about $0.36/1000 gal
(0.095/m3).
The effect of high organic content on carbon usage
rate is well known in industry. Recent work to develop
adsorption isotherms for 220 wastewater samples
representing 75 Standard Industrial Classification
(SIC) categories showed a strong relationship between,
carbon usage rate and organic content of the samples,
as measured by TOC.64 Usage rates as high as
1500 lb/1000 gal (179.9 kg/m3) were reported for
wastewater samples from the organic chemicals industry.
For petroleum refining, the values ranged from 0 2 to
141 lb/1000 gal (0.024 to 16.9 kg/m3) depending
upon the TOC of the waste.
Use of AC in wastewater treatment in oil refineries is
common.65'66'67 Because oil refineries and wood pre-
serving are similar in terms of wastewater characteristics,
a few of the more pertinent articles dealing with AC treat-''
ment of refinery wastewater are summarized here.
Workers dealing with treatment process methodology
emphasized the necessity of pretreatment of AC
column influent.68 Based on these reports, suspended
solids in amounts exceeding 50 mg/1 should be removed.
Oil and grease in concentrations above 10 mg/1 should
likewise not be applied directly to AC. Both materials
cause head loss and can reduce adsorption efficiency by
coating the carbon particles. This is apparently
more critical in the case of oil and grease than for sus-
pended solids.
Common pretreatment processes used by the industry
include chemical clarification, oil flotation, and
filtration. Adjustments in pH are frequently made to
enhance adsorption efficiency. An acid pH has been
shown to be best for phenols and other weak acids.
Flow equalization is, of course, necessary for most
treatment processes.
Efficiency of adsorption varies among molecular
species. In a study of 93 petrochemicals commonly
63
-------
found in that industry's wastewater, adsorption was
found to increase with molecular weight and decrease
with polarity, solubility, and branching.67 However,
molecules possessing three or more carbons apparently
respond favorably to adsorption treatments.64
One author66 studied the relative efficiency of lignite
and bituminous coal carbons and concluded that lignite
is better for refinery wastes because it exhibits a greater
surface area.
Several authors have discussed the economics ot
carbon adsorption. Paulson of Calgon Corp. reports
that direct operating costs based on BOD removal
from refinery wastewater range between 4.8 and 10.9
cents/1000 gal (1.26 and 2.88 cents/m3) for primary
effluents and between 5.3 and 9.4 cents/1000 gal (1.40
and 2.49 cents/m3) for secondary effluents. These
estimates were based on a carbon usage rate of 9.9
lb/1000 gal (1.19 kg/m3). Much higher operating
costs for AC systems have been reported by Rosfjord;
et al.69 Costs of $0.40 to $6/lb ($0.88 to $13.24/kg) of
phenol removed were cited. The efficiency of the carbon
regeneration facilities were listed as an important
consideration in computing operating costs. Recovery
of chemicals, particularly phenols, from carbon beds
by either reactive or solvent methods was reported to be
less costly than thermal regeneration.
According to Hutchins,56 it is most economical
to discard carbon at usage rates lower than 350 to
400 Ib/d (158.5 to 181 kg/d) and to regenerate
thermally at higher usage rates. _ j
Limited data on AC treatment of wood-preserving
waste strongly indicates that the process is not eco-
nomically viable when applied as a secondary treatment.
The high cost of carbon adsorption when employed
in this capacity is primarily due to the presence of
high wood sugar levels in the effluents from this industry.
Use of the process as a tertiary treatment following
biological oxidation—assuming a high oxidation
efficiency of the sugars—may be economically viable.
Adsorption by Other Media |
Polymeric adsorbents have been recommended |
for use under conditions similar to those where carbon
adsorption is indicated.70 Advantages cited for these
materials include efficient removal of both polar and
nonpolar molecules from wastewater, ability to tailor-
make an adsorbent for a particular contaminant, small
energy inputs for regeneration compared to carbon, ;..
and lower cost compared to carbon where carbon ;
depletion rates are greater than 5 lb/1000 gal (0.601
kg/m3). Data on efficiency of polymeric adsorbents
were not presented. Clay minerals, such as attapulgite
clay, have been recommended for use in removing certain
organics and heavy metals from wastewater.71
New Preservatives ,
An obvious solution to the pollution problems con-
fronting the wood-preserving industry is to develop ;
new preservatives that are both environmentally
compatible and equally as effective as those presently
in use. This solution is not practical. There are many
candidate wood preservatives currently being screened;
however, some of these cannot be registered because
of the same objections made against preservatives
currently being used. Other new preservatives
undoubtedly will prove to be ineffective, and none
of them has been tested sufficiently to warrant commer-
cialization. Furthermore, no single preservative can
replace fully the system of preservatives now used by
industry, regardless of how good it may be.
At least one or two of the new preservative formula-
tions being studied look promising, based on laboratory
tests. Field testing has begun, but it will take 5 years
to complete and additional service tests will be required.
The consensus is that the prospects are poor for finding
replacements for creosote, pentachlorophenol, and
the salt formulations within the next decade.
New Processing Technology
A new processing technology has been developed to
reduce the volume of wastewater that must be treated.
The process has been successfully tested at the
pilot-scale level, but not on a commercial scale. In the
process, the pentachlorophenol is deposited in-situ
following treatment of wood with a water-soluble salt
of that chemical. Wastewater generated during
treatment can be recycled to prepare new batches of
treating solution.
Wastewater Disposal-Evaporation Processes
Evaporation has been most widely accepted by
the industry as the principal method of wastewater dis-
posal. Some applications of evaporation by the
industry are both economical and effective, while others
are not. Evaporation methods that require the use of
process steam are not considered economical. On the
other hand, methods that depend on simple surface
evaporation from lagoons may be economical but
are not effective, since such installations usually
are located in regions where rainfall equals or exceeds
lake evaporation.
The use of one or more evaporation lagoon equipped
with spray equipment is the most effective and eco-
nomical method. By proper sizing of the lagoons to
account for seasonal differences in both rainfall and.
evaporation, a zero discharge of process wastewater
can be achieved. Actually, such installations serve a
dual purpose. In addition to evaporating all or most
of a plant's effluent, significant reductions in conven-
tional wastewater parameters are achieved with only
a minimal accumulation of sludge, since an evaporative
lagoon functions in much the same fashion as an
aerated lagoon. Cooling towers that utilize waste heat
for wastewater evaporation are equally as effective
and much more efficient in terms of load requirements.
A closer look at selected evaporation processes, particu-
64
-------
larly with respect to Boulton processing plants in
the west, is presented in the following paragraphs,
along with energy and cost data.
Forced Evaporation
Equipment for forced evaporation can take several
forms. All require considerable quantities of energy.
The most common in use in the wood treating industry is
an open pan heated by gas, oil, or steam coils. :
Roughly 1000 Btu/lb (1055 kJ/lb) of water are
required to evaporate water, depending upon the
temperature at which the evaporation takes place.
A Boulton plant treating 2000 ft3/d (56.63 m3/d)
(1 retort) of green stock [at 22.6 Ibs water/ft3 (362
kg/m3) stock] requires 45 million Btu (47.47 million kJ)
per day to evaporate the wastewater. At an energy cost
of $1.4/million Btu ($1.32/million U) based on a
regulated natural gas price of $1.45/1000 ft3 ($0.051/
m3) and an evaporation efficiency of 58 percent,
evaporation costs are $119 per day or $38,000 per year.
Similarly, a plant treating dry stock will have an
evaporation energy cost of $26.40 per day or $9240 per
year. As energy costs increase, the cost of forced
wastewater evaporation also will increase. If waste
wood is being used as boiler-feed then forced evapora-
tion costs will be much lower.
The energy saving advantages of using pre-dried
stock are obvious. Most treating plants will use a mixture
of pre-dried stock, the dryness of the stock depending
upon the season, storage yard capacity, and product
demand. Any shift toward using dryer stock will reduce
the quantity of water to be evaporated and thereby
the cost
Evaporation Containment Ponds
Evaporation containment ponds can be used in
areas where land is available at a low enough cost to be
economical and where the yearly evaporation rate
exceeds the yearly rainfall plus wastewater generated.
In eastern Oregon and Washington, pan evaporation
is 30 to 40 in (76.2-to 101.6 cm) per year, and in
California, 40 to 50 in (101.6 to 127 cm) per year. Rain-
fall in eastern Oregon, Washington, and California
is less than 20 in (50.8 cm) per year, so evaporation
containment ponds are feasible there.
Adding a spray system to an evaporation pond will
increase the rate of evaporation by exposing greater
quantities of water to the air. This sytem will redtice
the size of the evaporation pond required to evaporate
rain and wastewater. The size of an evaporative contain-
ment spray pond depends heavily on climatic conditions.
Use of Waste Process Heat for Evaporation
Waste process heat is produced whenever part
of the process is being cooled, for instance in condensers
and compressor coolers. The waste heat is removed
from the process by cooling water. This heat can be
used to evaporate treated wastewater in spray cooling
ponds or cooling towers. At the 1972 Annual Meeting
of the American Wood Preservers Association, Charles
Best described four such systems that have been
in operation since before 1971. Figure 15 shows how
such a system would be connected.
Heat is applied to the preserving cylinder via steam
coils while a vacuum is used to evaporate water from the
stock. The heat contained in the water vapor that
leaves the cylinder is transported to the condenser,
where it enters the cooling water. The water is cooled
in a cooling tower or spray pond, and the heat is trans-
ferred to the air in the form of water vapor. The amount
of heat absorbed by the cooling water is slightly more
than is required to evaporate the water. Additional
heat is obtained from oil condensation in the condenser
and cooling of the condensate to below the boiling
temperature. The following heat and water balance
demonstrates that there is sufficient heat available to
evaporate all of the process water. There is insufficient
heat available to evaporate rain water.
Vapors from retort:
7.5 lb/ft3 (120 kg/m3) for partially dried stock
20 lb/ft3 (320 kg/m3) for green stock
Energy from retort contained in vapor at 200°F (93 3°CV
8594 Btu/ft3 (320 MJ/m3) for partially dried stock'
22,916 Btu/ft3 (853.5 MJ/m3) for green stock
Rain water addition:
2 lb/ft3 (32 kg/m3) (yearly average)
Wastewater to be treated-
9.5 to 22 lb/ft3 (152.2 to 352.4 kg/m3)
Heat added to cooling water [vapor condensed to
100°F(37.7°C)]:
8034 Btu/ft3 (299.2 MJ/m3) forpartially dried stock
21,557 Btu/lb3 (802.8 MJ/m3) for green stock
Heat added to cooling water via wastewater at 60°F
(15.5°C):
266 Btu/ft3 (9.9 MJ/m3) for partially dried stock
616 Btu/ft3 (22.9 MJ/m3) for green stock
Water evaporated from cooling tower at 1065 Btu/lb
(2477 kJ/kg):
7.8 lb/ft3 (125 kg/m3) for partially dried stock
20.8 lb/ft3 (333 kg/m3) for green stock
Cooling System Design
Design of a cooling tower or spray cooling pond
must take into consideration optimization both of cooling
and of evaporation performance. The cooling water tem-
perature must be low enough so that existing condensers
can be used and process vacuums can be maintained.
An additional advantage to lower cooling water tem-
peratures is that evaporation is enhanced. These
lower temperatures are achieved by cooling facilities
larger than normally would be considered if cooling
were the sole objective.
65
-------
Wood In
Wood Out
Treating Cylinder
Preservatives
to Work
Tank
Preservatives
to Cylinder
Cooling
Water
Cylinder
Drippings
and Rainwater
Recovered Oils
Work Tank
Oil-Water
Separator
Wastewater
Air and
Vapors
Polishing
Oil Removal
Water
Vapor
1 Sludge
Evaporator
Figure 15. Diagram of a Boulton Process Wood Treating Plant Using Waste Heat for Evaporation of Wastewater.
Cooling systems designed for maximum evaporation
should operate with the hot water input temperature
as low as possible to avoid heat loss. Air leaving the
cooling system will be heated to a temperature near |
the hot water temperature. When the exit air temperature
is above the atmospheric air temperature, energy
escapes into the air and less is available to evaporate
water. Figure 16 illustrates the amount of evaporation
achievable from a cooling pond with a constant heat
input, but with varying flows and thereby different cool-
ing ranges. The cooling range is the difference between
the hot and cold water temperatures entering and ;t
leaving the cooling pond/tower. ;
Also shown in Figure 16 is the volume of wastewater
evaporation corresponding to the heat input from a one-
retort wood treating plant (does not include rain
water). In order to evaporate all of the wastewater, a
cooling range of less than 10°F (5.6°C) is required.
Figure 16 was developed for the month of March
in Portland, Oregon. During other months the evapora-
tion rate would be different. Low cooling water
temperatures are attained by high cooling water flow
rates through condensers or by cooling water recircula-
tion through the cooling tower.
Evaporation from a cooling tower is closely tied
to the liquid-to-gas (L/G) ratio. When gas flow rates
are large compared to liquid flow rates (low L/G), more
evaporation takes place. Figure 17 shows the amount of
2000 i-
1500
.2 1000
500
Process Water Input Rate
Evaporation Rate
10
20
30
40
50
Range, °F
Figure 1 6. Evaporation Rate of Water from a Cooling
Tower as a Function of the Difference Between Cooling
Water Inlet and Outlet Temperatures at a Constant
Heat Input.
evaporation that may be expected from a one-retort
wood treating plant at several L/G ratios. Cooling
towers normally are designed to operate with an L/G
ratio of about 1, and spray cooling ponds with an L/G
ratio of between 0.10 and 1.0.
-------
5000 r—
4000 —
3000 —
= 0.18
I 1 1 I
2000 —
1000 —
March
April
May
June July Aug.
. Time of Year
Figure 17. Cooling Tower Evaporation Rate for Several Liquid-to-Gas Ratios.
Sept.
Oct.
Nov. Dec. Jan.
When a cooling tower is designed with a low '
L/G ratio, the exiting cooling water temperature,
within limits, will be lower than when high L/G ratios
are used. The lower exiting cooling water temperature
will result in a lower entering cooling water tempera-
ture and more evaporation will result.
Cooling Pond Design and Operation
o A cooling pond was designed for a maximum of
5 F (2.78° C) rise in cooling water temperature:in the
condenser when green stock is being conditioned.
The resulting cooling water flow through the condenser
for one retort treating 1600 ft3/d(45.30 m3/d) was
650 gpm (0.04 m3/s). The cooling pond dimensions.
were 100 ft by 134 ft (30.48 by 40.84 m) and with 17
spray nozzles spaced 2 ft (61 cm) apart, resulting in L/G
of 0.24 for a 5-mph (8.04 km/h) wind. The summertime
maximum cold water temperature is 70°F (21°C)
Figure 18 shows the evaporation attainable from
such a pond located in Portland, Oregon, over a year's
time when the condenser is being operated at 78-percent
capacity. Also shown in Figure 18 is the amount of
process water that requires evaporation, and the
amount of process water and rain water that falls on the
retorts and vicinity, tank farm, and in the cooling pond.
As can be seen from the graph, in the winter the cooling
pond will be unable to evaporate all of the process
and rain water; however, evaporation during the summer
months will be in excess of the process and rain
water collected. The yearly average evaporation for the
pond will be in excess of the water addition.
Storage must be considered to accommodate the
winter rain water. The unevaporated process and
rain water will be 20,000 ft3 (566.33 m3) for the
winter season for an average year. The cooling pond used
in this example has a surface area of 13,400 ft2 (1244.9
67
-------
4000 i—
3000 -
2000 -
i
1000 -
Jan.
Feb.
March
April
May June , July Aug.
Time of Year
Sept.
Oct.
Nov.
Dec.
Jan.
Figure 18. Spray Cooling Pond Evaporation Rate Over One Year.
m2) in which the accumulated rain would cause a
level rise of 8 in (20.32 cm). A cooling pond with 2 ft
(61 cm) of freeboard could accommodate storage of
excess rainwater even for exceptionally wet years.
In drier climates, such as east of the Cascades or in;
California, evaporation will exceed rainwater
encountered.
Cooling Tower Evaporation Performance
A cooling tower was designed to cool 650 ^
(0 04 m3/s) of condenser water heated 5°F (2.78 C)
from a retort treating 1600 ft3 (45.30 m3) green stock
per day. The L/G ratio was 0.5, somewhat lower than
normal cooling tower design. The summertime j_
maximum cold water temperature was 75 F (23.8 C).
Figure 19 shows evaporation from the tower over \.
the span of one year when the tower was operated at
78-percent capacity. Also shown is the quantity of
process and rainwater that needed to be evaporated.
As can be seen from Figure 19, a cooling tower of
the design described is unable to evaporate all of the
process and rainwater accumulated during the winter
months. Excess evaporation takes place during thq
summer. Over an entire year, there is more evaporation
than wastewater input.
To accommodate the excess process and rainwater
during the winter months, storage capacity needs to
be built. Excess water collected over a winter averages
16 400 ft3 (464 39 m3). To accommodate this water
in wet years, a 200,000- to 300,000-gal (757.08
to 1135.6 m3) tank will be needed. An alternative is to
install excess fan horsepower in the cooling tower to
increase the L/G as needed.
Retrofitting to Existing Plants
An evaporation system with a spray cooling pond
or cooling tower of sufficient size and proper cooling
rate to evaporate process and rainwater from a
wood treating plant is apt to be limited by the size of the
existing condenser. The small cooling ranges required
for good evaporation dictate high cooling water flows
through the condenser and thus small temperature
gains. To circumvent installation of a larger condenser,
the normal quantity of cooling water can be used through
the condenser and the bulk of the cooling water can be
recirculated through the cooling pond or tower, mixing
the hot condenser water with it.
68
-------
4000 ,—
3000
2
o
0.
2000
1000
Process and
, Rainwater
Evaporation
Process Water
J L
J L
Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. Jan.
Time of Year
Figure 19. Cooling Tower Evaporation Rate Over One Year.
Evaporative Process Operational Problems
Volatile Compounds. Regulatory agencies are pri-
marily concerned with the quantity of volatile toxic
compounds introduced into the air by evaporation of
wood-treating wastes, a number of which are known
to exist. The fate of these compounds in the several
evaporative processes is unknown. Many of the com-
pounds are known to be volatile. As a result, EPA is
initiating a project to measure these volatile compounds.
To avoid evaporation of these toxic compounds
into the air, good oil-water separation and secondary
polishing oil removal must be provided. Many of the
volatile compounds are insoluble in water and will be
removed in the oil phase. However, such compounds as
phenols that cannot be removed by oil-water separation
will be passed on to the cooling tower and evaporated to
the air.
Corrosion. Corrosion in operating systems has
not been a problem. Since all the water added to the
cooling water system is essentially distilled water, there
is no build up of salts and no need for a blowdown.
Sludge accumulates in the system and needs to be dis-
posed of periodically. Sludge accumulation may be
about 4000 gal/yr (15.14 m3/yr).
Fouling. Fouling of condenser tubes can be a serious
problem. Preservatives and dissolved wood compounds
tend to polymerize with time and heat and coat the
insides of pipes and condenser tubes. In one plant,
fouling made it necessary to remove the compressor
from the recirculated cooling water system. Plants
using barometric condensers will not encounter
fouling problems.
Pretreatment of wastewater prior to disposal into
the cooling pond or tower may be necessary to prevent
fouling. One preserving plant uses a sequence of gravity
oil and water removal, lime flocculation, filtration, and
an aerated lagoon to treat wastewater. Another plant
adds its wastewater, after gravity oil and water separa-
tion, to its cooling water. The cooling water passes
through a spray cooling pond, an aerated lagoon, a
holding pond, and then back to a surface condenser.
Comparative Economics
Economics is the guiding factor in selecting a
wastewater disposal system. With energy being a
69
-------
strong factor in the economics of one such system, costs
are difficult to assess. Two scenarios are presented ]
in the following paragraphs: one at approximately
today's energy costs and one at double today's energy
costs.
Table 23 shows approximate capital costs and
Tables 24 and 25 show approximate operating costs
for each of the systems. As can be seen, the system i-
using a cooling tower is the most expensive in terms bf
both capital and operational costs. The high operating
cost results from the high amortization expenses
caused by large capital costs. As energy becomes more
expensive, though, the cost of wastewater evaporation
by use of cooling tower and forced evaporation become
more comparable.
Forced evaporation is the second most expensive
system and will become even more expensive as energy
prices continue to rise. Operational costs are highest
when green stock is being treated and more tolerable
when pre-dried stock is being treated. The least expensive
method for disposal of treating plant wastes is evapora-
tion by spray cooling ponds.
Table 24. Operation Costs of Evaporation Systems
Tray evaporation:
Interest and Capital Payments .
Depreciation (15 yr)
Energy
Maintenance
$3,418
1,800
9,200-38,000
1,000
$15,400-45,200
Cooling tower:
Interest and Capital Payments .
Depreciation
Energy
Maintenance
Spray pond:
Interest and Capital Payments .
Depreciation
Energy
Maintenance
$32,868
16,700
10,000
7,500
$67,100
$ 7,296.79
3,700
5,000
1,700 .
$17,700
Table 23. Capital Costs of Evaporation Systems
Tray Evaporator, 200 ft2 S 26,000
Cooling tower, 650 gpm
Storage tank, 300,000 gal.
Flocculation-filtration
$160,000
65,000
25,000
Total $250,000
Spray pond (140 X 100 ft X 6 ft Land)
630,000 gal:
5,000
Materials
Labor.
Pumps
Piping
Misc. Elec
Engineering
Flocculation-Filtration
Total
5,000
1 0,000
3,000
2,000
2,500
3,000
$ 30,500
I.
25,000
$ 55,500
Table 25. Operation Costs of Evaporative Systems at
Twice Today's Energy Costs
Tray evaporation:
Interest & Payments ,
Depreciation
Energy
Maintenance
$3,400
1,800
18,400-76,000
1,000
$24,600-82,200
Cooling tower:
Interest & Payments
Depreciation
Energy
Maintenance
Spray pond:
Interest & Payments
Depreciation
Energy
Maintenance
$32,900
16,700
20,000
7,500
$77,100
$ 7,300
3,700
10,000
1,700
$22,700
70
-------
Chapter 1 4
Case Study: Wet Decking with
Water Recycle
The Pinecrest Lumber Company, a wholly owned
subsidiary of Green Bay Packaging Incorporated,
is located near Menifee, Arkansas, about 35 miles
(56.32 km) northwest of Little Rock. It is a chip-and-saw
operation supplying chips to Arkansas Kraft Corpora-
tion, another Green Bay Packaging subsidiary, which
operates an unbleached kraft linerboard mill. Pine-
crest started operations in November 1973. About
one-quarter of its raw materials come from 150,000
acres (607 km2) of Arkansas Kraft woodlands and the
rest from private lands. The majority comes from .within
a 50-mile (80.46 km) radius of the mill. The predominant
species is shortleaf pine, with about 5 percent loblolly.
Forest planting operations are predominantly loblolly.
The nominal capacity is 140,000 board-feet (fbm) per
day (330.4 m3/d) in a two-shift operation. Logs are
mechanically barked in a 26-in (66 cm) ring barker. A
chip-and-saw unit reduces one dimension to 4 in
(10.16 cm), followed by a vertical saw arbor for conver-
sion of the center cant to two-by-fours, which make
up 80 percent of production. Additional chips are
generated in a chip-and-saw, two-side edger. Two-in
(5.08 cm) lumber is produced in widths to 12-in (30.48
cm) and 1-in (2.54 cm) lumber to 6 in (15.24 cm). A
high-temperature, gas-fired, 24-hr kiln completes the
processing. ;
Dry storage of logs reduces lumber quality by allowing
the growth of blue stain. In addition, bark beetles do
severe damage by boring. The pulp mill prefers never-
dried chips to produce higher strength pulp and to prevent
yield losses in byproduct turpentine and tall oil. For
these reasons, the logs are kept wet during storage by
wet decking. Logs have been stored wet for up to
14 months, with no adverse results. Logs have been
successfully stored in this manner up to 4 years, accord-
ing to the experience of nearby operations.
Log lengths are from 10 to 20 ft (3.04 to 6.08 m) and
average about 9-in (22.86 cm) in diameter at breast
height (dbh). As many as six rows may be formed side-
by-side with no intermediate access aisle. Six rows
can be well covered with four rows of sprinklers. As many
as 11 rows can be accommodated using the entire deck,
but 8 rows is the usual maximum.
Yard inventorying of logs is highly variable, depend-
ing on woodlands conditions. During periods of heavy
rainfall, which restricts logging, the wet deck is
used to supply the mill. To avoid double handling, the
preferred operation is to bypass storage, but that balance
cannot be maintained. It is not practical to sprinkle logs
if less than two rows are stored, since the spray pattern
converts the adjacent access areas to mud. Concrete
pads are being considered seriously to eliminate the
mud problem and to assure front-end loader maneuver-
ability without having to dry out the aisles between
piles. If possible, first in-first out storage is practiced.
Wet decking of stored logs was started in April 1976.
Logs are piled 12 to 15 ft (3.66 to 4.57 m) high in 300-ft
(91.44m) rows. Drainage from the 2.8-acre(l 1,330 m2)
wet deck flows to a 1.3-acre (5261 m2) sedimentation
pond about 5 ft (1.52 m) deep, with a capacity of 7.5
acre ft (9250 m3). Excess drainage following settling
of solids overflows to a second 1.1-acre (4452 m2)
pond having a depth of 5.5 ft (1.67 m) and a 5.7-acre ft
(7031 m3) capacity.
Water is pumped from either pond, as chosen, to feed
the sprinkler system. The sprinkler system was designed
according to recommendations of the sprinkler manu-
facturer, based mainly on irrigation criteria. The
pump is a 30 hp (22 kW), high-head centrifugal pump
with a capacity of 350 gpm (0.022 m3/s) at 100 psi
(689 kPa). The normal operating range is 40-100
psi (276-689 kPa). The water is pumped to the wet deck
area via 300 ft (91.44 m) of a 4-in diameter (10.2 cm)
pipe. The sprinkler pipe on the long piles is 2-in diameter
(5.08 cm), quick-couple, aluminum irrigation pipe
with 3/ie-in or W-in (0.48 or 0.64 cm) rainbird units
every 20 or 40 ft (6.09 or 12.2 m). With an approx-
imately 60 ft (18.28 m) spraying diameter, there is
ample overlap to cover the entire area. Pipe length is
chosen to fit the pile contour. The sprinkler orifice
diameter may be changed to improve spray coverage. As
many as six rows totalling 1,800 ft (548.6 m) of 2-in
(5.08 cm) pipe have been used. The rainfall equivalent
of 150 gpm (0.01 m3/s) on 1.5 acres (6071 m2) is
5.33 in (13.53 cm) per day.
Orifice plugging is not a serious problem. A wire cage
with V2-'m (1.27 cm) spacing protects the foot valve,
which itself has a 3/ie -in (0.47 cm) screen. These
screens have not required cleaning to date. It was found
that sprinkler pluggage was essentially eliminated
71
-------
by terminating each line with a cap having a J/£-in >
(1.27 cm) diameter hole through which a flushing flow
is continuously passed.
A few comments should be made about the history
of regulations on wet decking. The EPA published
a Development Document72 in August 1974, which •
included a study of wet decking by Environmental
Science and Engineering, Inc., of Gainesville, Florida.
They examined practices at eight West Coast mills
and six mills from the Southeast and South. Among
these 14 installations, they found that recycling was
more commonly used in the Southeast and South than in
the West. The sprinkler flow-to-log inventory ratio
varied by a factor of 15, comparing maximum to mini-
mum, showing the great variability in these operations.
They chose for a typical Southern wet deck, with recycle,
a hydraulic flow of 150 gpm (0.57 mVmin) onto
4,600 yd3 (35,169.5 m3) of logs and a deck area of
4 acres (1618.8 m2). Although Pinecrest can be con-
sidered in this flow range, the company's stored log
volume would often be three to four times this 4,600
yd3 (3517 m3) figure. Presumably Pinecrest's larger
inventory would increase the amount of water lost by
evaporation in comparison with EPA's typical mill.!
The Development Document recommended that Best
Practical Control Technology Currently Available
(BPT) for the wet storage subcategory should call for
"no discharge of process wastewater pollutants between
May 1 and October 31, except a volume of water
equal to the difference between the mean precipitation
for a given month and 10 percent of the annual lake
evaporation; no discharge of process wastewater
pollutants between November 1 and April 30, except
a volume of water equal to the mean precipitation that
falls on the drainage area of the wet storage facility."
When a discharge was allowed, the debris could
not exceed 1 in (2.54 cm) in diameter and the pH was
limited to the range 5.5 to 9.0.
The same limitations were recommended for Best
Available Technology Economically Achievable
(BAT) and, for wet decking only, for New Source
Performance Standards (NSPS). These proposed ,
regulations were published in the Federal Register on
August 26,1974. !
Following study of comments on the proposed [
regulations, final wet decking regulations were promul-
gated in the Federal Register on January 16, 1975.
The limitation on debris diameter was kept, but the
pH range was changed to 6.0 to 9.0 and the flow limita-
tion was dropped.
Water Balance
Rainfall records from the U.S. Weather Bureau
show that the Little Rock area has had about 48 in
72
(121.9 cm) of annual rainfall. The Climatic Atlas indi-
cates that the mean annual lake evaporation has been
43 in (109.2 cm). (Using these figures, the originally
proposed regulations would have meant no allowable
discharge in the months of June through October.)
Various estimates of evaporation rates from a sprayed
wet deck have been in the neighborhood of 30 percent
of the applied water to 0.1 gal/ft2/d (0.004 m3/m2/d)
with an equal gallonage lost to absorption and
percolation.
Approximate water balances were calculated using
both estimates and the following assumptions:
150 gpm (0.564 mVmin) sprinkler flow
12 month operation
3 acres (12,141 m2) of log deck, one-half under spray
2.5 acres (10,118 m2) of ponds
50 in (127 cm) of annual rainfall
43 in (109.2 cm) of annual evaporation from ponds
Case I: 30 percent of applied spray lost
Case II: 0.2 gal/ft2/day (0.008 m3/m2/d) lost
For Case I, there would be an annual loss prediction
of 19.5 million gal(73,815 m3)or30gpm(0.114m3/min).
For Case II, the annual loss prediction was 0.3 million
gal (1135 m3) or 0.6 gpm (0.0027 m3/min). An existing
on-site well could supply the deficit in Case II but not
Case I. There was actually more short-term concern
for water shortage than for pond discharge.
In view of the predicted deficit, EPA, Region VI, was
provided an Affadavit of No Discharge. The affadavit
was filed in December 1976, after 7 months of operation.
The water level in December was essentially the same
as when spraying had started. The 1976 rainfall at Little
Rock was about 10 in (25.4 cm) below normal although
no records were available to document the actual
on-site precipitation. It was estimated that three or
four successive years, each with 10 percent higher than
normal rainfall, could cause the ponds to crest and.
1 year with 25 percent higher than normal rainfall could
be expected to produce similar results.
Little Rock rainfall records for the previous 98 years
were grouped in sets for each 5 percent greater than
the mean, or roughly 2.5-in (6.35 cm) intervals above
48 in (121.9 cm). The records show that 125 percent
of the average annual rainfall has been reached, or
exceeded, in 13 years out of 98,150 percent was reached
or exceeded in 2 years out of 98, and the highest
year out of the 98 was 157 percent of the average.
Pinecrest has an inadequate data base on which to
speculate about the future of the ponds. Records are now
being kept of pond height in an attempt to relate spraying
history and rainfall to observed pond height.
-------
Chapter 15
Case Study: Waste water
Handling at Boise Cascade's
Elgin, Oregon Site
Boise Cascade acquired a studmill at Elgin, Oregon
in 1961. The mill was modernized and a plywood plant
was built on the site.
The logs in the storage area were being sprinkled
from Phillips Creek, which borders the log yard. The
creek was inadequate, since it goes dry in late July, just
when the water was really needed. With the additional
storage of logs required because of the additional ,
production, a new water source was urgently needed.
Accordingly, a well was drilled and an artesian aquifer
encountered. A pump capable of delivering 1500 gpm
(5.67 m3/min) at 100 psi (689 kPa) was installed.:
This arrangement provided a satisfactory water source
for the logs, but a wastewater problem developed,
since the contaminated water could not be routed to
Phillips Creek. The waste stream included not only log-
yard runoff from summer sprinkling, but miscellaneous
waste streams from processing that flowed year around.
It was decided at that time to purchase a section of
flat land, divert the wastewater to this area and flood
irrigate the land. The irrigation benefited both a hay crop
and cattle grazing.
As time passed, stricter regulations became effective
and Boise Cascade could no longer flood the flat land
because of runoff to the Grande Ronde River. To
avoid this, it was decided to install a sump pit, catch the
runoff, and recycle it back to the log decks. The logs
were sprinkled from April 1 to December 1 from the re-
cycled runoff. From December 1 to April 1 the mill's
permit allowed a discharge to Phillips Creek. Subsequent
to the enactment of the Water Pollution Control Act
Amendments of 1972, even stricter permit requirements
were imposed. The plant had to face the question of
proper disposal of process wastewater, log-pond over-
flow, storm runoff, and water from numerous surface
springs without dumping any contaminated water into an
adjacent freshwater stream. There was also the
problem of year-round excess water in the log yard.
The sources of clear and contaminated wastewater
were freshwater springs in the log yard; plant site and
log pond; surface storm runoff; equipment cooling water;
plant process wastewater, including boiler blowdown
water; and log-pond overflow water. Because the waste
discharge permit issued by Oregon's Department of
Environmental Quality forbids contaminated water
discharge, each of these water sources had to be con-
sidered individually. The most difficult problem was
Phillips Creek, which runs along the west and south
sides of the log yard. Many years ago the stream ran along
the north side of the present log yard, but heavy runoff
caused flooding and the creek changed to the present
stream bed. It was suspected that the source of many
of the freshwater springs in the log yard and around
the plant site came from the adjacent stream.
A series of test holes were dug in various parts of the
log yard and the data gathered from them indicated that
the stream was indeed providing the unwanted water
to the surface springs. From the test-hole data a French
drain was designed that would intercept the fresh clear
spring water and return it to the freshwater stream
downstream of the log yard. The Department of Environ-
mental Quality had no objection to returning uncomtami-
nated spring water to the stream.
The 1500-ft long (457.2 m) French dram was con-
structed across the log yard using concrete pipe set in
clean gravel at the depth of the hardpan, about 8 to 10 ft
(2.43 to 3.0 m) below the ground surface with a drainage
slope toward the stream. A 1 Vi-in (3.81 cm) space was
left at each pipe joint to allow water inflow. The open
joints were covered with wire mesh to keep gravel
and dirt from entering the pipe. The pipe was covered
with clay to a depth of 2 ft (61 cm) above the pipe and
firmly compacted; the clay provided a seal so that con-
taminated log yard surface water could not leach
down into the uncontaminated fresh water. The balance
of the French drain trench was filled with gravel from the
excavation. After a short flush-out period, the French
drain stabilized to a clear freshwater flow of about
200 gpm (0.757 m3/min). It was not long before it
was noted that previously wet areas in the log yard had
dried up. A lowering of the water table downstream
from the French drain was also noted.
Collection of contaminated wastewater was
accomplished through two collection systems. The first
system collected log yard runoff, storm runoff, plant
wastewater, surface spring water around the sawmill,
and all the machinery cooling water. Because of the
scattered nature of the cooling water, no attempt has yet
been made to collect and dispose of it separately. Two
collection culverts transported all of this wastewater
to an 11,000-gal (41.6 m3) collection sump located at
the southeast corner of the plant site.
Due to the special nature of boiler blowdown
water, a separate collection sump, pump, and pipeline
73
-------
were used to move this wastewater to the lagoon system.
A collection sump near the boiler house provided
a base for the two 25-hp (18.64 kW), 4-in (10.2 cm)
sump pumps. The two pumps, piped in parallel, pump
the blowdown water through a pipeline to the wastewater
settling lagoon. Only one of these pumps is used at any
one time; the other one is available on a standby basis.
Since there are several opportunities for debris
to enter the wastewater collection system, a rotating
screen was installed to pick up and dispose of any
material that could prove harmful to the pumps. A pump
station was built at the collection sump so that the
accumulated wastewater could be moved to the settling
and storage lagoons. Two, 1500-gpm (5.68 m3/min)
sump pumps, piped in parallel to a common manifold, and
each powered by a 40-hp (29.83 kW), 1800-rpm :
electric motor, provide the necessary pumping capacity
to move the wastewater through the 1100-ft (335 m)
pipeline. Only one of the pumps is used at any one time.
The second pump, on a standby status, will start auto-
matically should the first pump fail. A signal light,
mounted on top of the pump house, informs operating
people in the log yard a quarter mile away if any equip-
ment hi the pump station fails.
A pipeline over 1100 ft (335.28 m) long was installed
to lift the wastewater more than 60 ft (18.28 m) from
the collection sump to the settling lagoon. The pipeline,
12-in (30 cm) PVC pipe, had to cross the fresh water
stream. The stream crossing was accomplished
with the use of two 60-ft (18.28 m) Douglas fir logs.
The settling lagoon is about 6-ft (1.82 m) deep and con-
tains 17 acre ft (20,970m3). All the contaminated waste-
water from the collection sump and boiler blowdown
sump end up in the settling pond. The lagoon acts some-
what like a pretreatment pond or clarifier, in that all the
settleable solids are deposited in this pond. This lagoon
also acts as a water disposal site, in that it has a large
evaporating surface. Maintenance on the settling lagoon
consists of draining and dredging every 8 to 10 years.
From the settling lagoon the wastewater passes
through a 120°F (48°C) weir into the storage lagoon.
Associated with the weir is a float recorder/totalizer
with a 60-day chart. With this recorder, the daily and
seasonal variations in the wastewater production can be
monitored.
The wastewater is used for sprinkling on the log yard
for about 7 to 8 months of the year. Since production of
wastewater continues over the full year, a storage
lagoon became necessary. A site was chosen adjacent to
the already existing settling lagoon. The storage lagoon
has a capacity of over 136 acre ft (167,800 m3), or
over 44 million gal (166,540 m3). About 55,000 yd3
(42,050 m3) of material was excavated from the lagoon
site and placed in dikes in the construction of this facility.
The storage lagoon was constructed entirely from
materials found on the site. The dike core was formed
from clay and silty clay soils to secure a waterproof seal
for water containment The surface of the dike was
covered with topsoil taken from the bottom of the lagoon.
The topsoil provides a good medium for revegetation.
The inside of the dike was covered with 2 ft (61 cm) of
gravel to prevent wind and wave action erosion. Filling
of the storage lagoon begins at the end of each water
disposal season.
In addition to storage, the lagoon provides a very
large evaporation surface. For example, with a waste-
water temperature of 50°F (10°C), an air temperature
of 80°F (26.6°C), a lagoon surface area of 19 acres
(76,900 m2), a 5 mph (8 km/h) breeze, and a relative
humidity of 30 percent, as much as 11,500 gph (43.47
m3/hr) can be evaporated. This sytem is successful, since
during the summer evaporation exceeds precipitation in
this dry region by about 25 percent. If the relationship
were reversed, the system likely would be unsuccessful.
This condition is somewhat unique, since in most
timber growing areas and their associated manufacturing
centers the annual precipitation substantially exceeds
the annual evaporation.
Agricultural irrigation is the second wastewater
disposal method used in this system. A 50-acre site
adjacent to the storage lagoon provides an excellent
area for wastewater disposal. A 25-hp (18.64 kW) motor
driving a 4-in (10.16 cm) centrifugal pump moves up to
864,000 gpd (3283 m3/d) of wastewater from the
storage lagoon to the fields during the disposal season.
This is a fortunate situation, since the vast majority
of Boise Cascade facilities do not fit this situation.
Wet log-deck irrigation is the third and primary method
of wastewater disposal. Wastewater is taken from the
storage lagoon through a 10-in (25.4 cm) pipe to a 6-in
(15.24 cm) centrifugal pump driven by a 100-hp
(74.5 kW) motor. The pump increases the pressure to
about 95 psi C655 kPa) and delivers the wastewater
to the log-deck irrigation sprinklers. As the water
leaves the pump it is filtered to remove any debris that
might clog a sprinkler head.
The 40-acre (161,880 m2) log deck provides an excel-
lent wastewater disposal site. It not only provides a
large evaporation area, but also furnishes the sprinkling
water for the log deck. Unfortunately, not all of the
1.8 mgd (6804 m3/d) of wastewater applied to the log
decks evaporates. The balance of the water drains to the
return culvert and is recycled through the recycle system.
The last major area of modification necessary to
provide this self-contained wastewater disposal system
was the elimination of outflows from the log pond. All
surface inflows to the pond were intercepted and diverted
or eliminated. The French drain dries up the underground
springs that had previously fed into the bottom of the
pond. The log pond was then no longer a wastewater
source.
It has been estimated that the additional annual opera-
tional and maintenance costs of the self-contained
wastewater disposal system over that of disposing of
excess wastewater to rivers and streams will be about
$18,000 to $20,000 per year. This cost represents
the use of additional power, maintenance pumps, and
pipes and dredging.
74
-------
Chapter 16
Case Study: Waste
Management at Masonite
Hardboard Mills
Hardboard manufacturers use water to-convey fiber,
clean equipment, scrub air emissions, trim mats, seal
vacuum pumps, dry panels, clean fiber, and cool panels.
Water is a component of the dynamic force that
separates fiber, and water is the vehicle in some of the
paints that coat the finished product. Water is essential
to the manufacture of wet process hardboard and its
use and discharge must be carefully controlled.
An essential step in processing the type of hard-
boards manufactured at Laurel, Mississippi and
Ukiah, California mills involves the use of great quan-
tities of water to wash the fiber to ensure removal of a
large portion of the water-soluble extractives found
naturally in wood. The most important of these extrac-
tives are wood sugars.
The Laurel, Mississippi plant pioneered studies
to use wood sugars recovered from fiber washing opera-
tions as a source of carbohydrates for animal feed.
This early work, performed at Mississippi State
College, failed to produce a solution and Masonite
utilized the wood sugars to produce alcohol. The alcohol
plant at Laurel also did not provide an answer and faced
early closure.
In 1950 the Ukiah, California plant opened with
a waste disposal system consisting of Vulcan evaporators
to reduce the wash water or liquor to 45 percent :
solids, a sludge burning boiler, and irrigation fields.
Subsequently, attempts were made to start a modified
activated sludge system. While such a system was recog-
nized as a viable treatment for municipal waste, it
was not found to be applicable to industrial effluent with
wide variability of loading. BOD loads ranging from
500 to 5000 mg/1 kept the treatment system in an upset
condition and forced continued use of the sludge burning
boiler.
Recognition that use of the wood sugars was a better
solution to disposal than discharge caused renewed
interest in the production of foodstuff from process water.
In the early 1960's this interest led to a pilot plant
at Laurel, Mississippi for the production of wood sugar
molasses. Success brought about a full-scale production
plant at the Laurel site and establishment of a pilot
plant for technology transfer at Ukiah.
Transfer technology failed at Ukiah, as it had
in earlier attempts to duplicate production equipment.
It became necessary, therefore, to establish a new ,
method for producing the desired wood sugar molasses.
New studies of the liquor, evaporation methods, and
storage and feeding techniques were instigated. Finally,
in 1970, a three-stage evaporator plant went into
production of the cattle feed supplement Masonex
at the Ukiah mill.
The production of Masonex as a byproduct of the
manufacture of hardboard is paradoxical, since the
manufacture of a readily saleable product is the result
of the disposal of a pollutant Liquors drawn off the fiber
washers generally range from 4.5 to 5.5 percent
dissolved solids. Evaporation of excess water is accom-
plished in a multiple-effect process that utilizes three
Swenson falling-film evaporators in tandem to raise
the solids to approximately 12 percent. Solids are
then increased to 45 percent in Vulcan evaporators.
Finally a Parkson unit concentrates Masonex to the
desired consistency of 60 percent solids. At this point
a caustic must be added to ensure safe storage in steel
tanks.
Masonex is used as a liquid supplement in prescribed
feed mixtures at commercial feed lots. Tests indicate
increased palatability of the mixture and faster rate
of gain when such a supplement is used.
The final Masonex product from Ukiah differs from
that produced at Laurel. Although both products provide
identical results under test feedings, the Ukiah
product has a much lower viscosity, 700 centipose,
while the Laurel product is near 5000 centipose at the
same percent solids. The West Coast product also has a
shorter storage life. These differences appear to
be the result of the use of different wood species at the
two mills.
In response to the California Water Quality Control
Act, the Ukiah mill developed a plan in 1969 for com-
plete control of the quality and quantity of waste-
water discharge to the Russian River. The expansion of
the evaporator plant in 1970 and the successful pro-
duction of the cattle feed supplement was the first step.
Operation of the enlarged evaporator permitted
Masonite to improve excess process water by reducing
its solids content. The evaporative process reduces
the quantity and BOD loading of the water to be treated.
From the earliest days of operation, a combina-
75
-------
tion of factors including wood species and materials
handling and disposal systems have permitted Ukiah
to operate with a process water system that is essentially
closed. Process water used for conveying fiber is
circulated through a large tank that acts as a primary
clarifier. Accumulated sludge is removed from the tank
and returned to the furnish system for the production
of hardboard through a process pioneered but found
not to be applicable at the Laurel mill.
In 1971 work began again in Ukiah on a biological
treatment system. A pilot plant, consisting of Doughboy
pools and other unsophisticated equipment, shed
some light on the possibility of dividing the waste
streams from the mill to allow increased treatment ;
where needed. The pilot biological plant operated
approximately 9 months, during which time the design
concepts for a proposed treatment plant were evaluated.
In 1973 final design of the plan began and a startup date
in early 1975 was set to conform with the requirement
of the Regional Water Quality Control Board.
The wastewater treatment facility that went into
full operation in May 1975 consists of a dual biological
treatment system for wastewater streams from the mill.
A strong-side stream, consisting of evaporator con-
densate and chip pile runoff, is treated in an activated
sludge system. A weak-side stream consisting of
stormwater runoff, cooling, and washdown water is
treated in an aerated pond system.
The strong-side stream averages 600 mg/1 BOD with
very low solids. Following nutrient addition of aqueous
ammonia and phosphoric acid, the water enters
either of two aeration basins where it is retained for
2V4 days. During this period the total suspended solids
(TSS) will rise to approximately 2500 mg/1. Wastewater
from the aeration basins flows by gravity to a clarifier
where a half day of retention will effect a 99 percent
or better reduction in BOD while dropping the TSS
to near 15 mg/1. The aeration basins receive 50 percent
of the flow from these clarifiers as culture. The remaining
50 percent is mixed with the weak-side stream either
before or after treatment of the weak-side, depending
on the need for further reductions of loading.
The weak-side stream has an average flow of 1.2 mgd
(4542 m3/d) of cooling and washdown water and
stormwater runoff. Typical loading of this influent is
125 mg/1 BOD and 80 mg/1 suspended solids. Floating
grease and oils are collected prior to entry into the
treatment system by a cable moving across the surface
of the influent ditch.
Chlorinated sanitary wastewater is added to the weak-
side stream prior to entry into a 9-million-gal (34,068
m3) aeration pond. If the discharge from the strong-side
needs additional treatment, it will be added here also.
Water from the aeration pond flows by gravity into an
11-million-gal (41,639 m3) aeration and settling
pond for the remainder of its total 14- to 16-day retention
in the weak-side treatment system. During this
period, Masonite normally experiences a 70 to 80
percent reduction in BOD and TSS.
The discharge from the weak-side is mixed with
discharge from the strong-side, and the treated effluent
is recycled to the hardboard mill for reuse as cooling
water, vacuum pump seals, and washdown at a rate
of over 800 gpm (3.02 m3/min). Excess water is
distributed to managed irrigation fields or discharged
to the Russian River during the non-recreation
season from October 1 to May 14.
While upset conditions resulting from accidental
spills of process water and extreme weather conditions,
both drought and deluge, have caused temporary
operational difficulties, Masonite's wastewater treat-
ment facility has been able to treat mill effluent continu-
ously beyond levels experienced elsewhere in the
industry. The success of this facility is attributed to a
unique combination of wood species, manufacturing
processes within the mill itself, and seasonal variations in
temperature and meteorological conditions, along
with the solid research effort that preceded construction.
76
-------
Chapter 1 7
Case Study: Wastewater
Purification at Pacific Wood
Treating Corporation
Pacific Wood Treating Corporation uses ultrafiltration
and reverse osmosis to purify wastewater for use as
boiler feedwater (see Figure 20). As designed, waste-
water enters one of three oil-water separators: one for
creosote accumulater water, one for pentachloro-
phenol accumulator water, and one for door drippings
and rainwater. Oil and sludge from door drippings
and rainwater separator are fed to the sludge tank, while
oil from the other two separators is returned to the
respective work tanks. Water from the separators is Com-
bined and fed to a cartridge type filter for further oil
and sludge removal. The filtered water is then fed to an
ultrafiltration unit for further oil and solids removal.
The concentrated waste from the ultrafiltration unit is
fed to the sludge tank. Dissolved solids are removed from
the filtered water in a reverse osmosis unit and this
concentrate also is fed to the sludge tank. Following
removal of dissolved solids the water is fed to the
boiler-feed deaerator and then into the boiler.
Pacific Wood Treating Corporation disposes of the
waste sludge in a wood-waste-fired boiler which is
equipped with a baghouse filter for air pollution control.
Wastewater From:
Creosote Retort
Permeate
Added to
Sawdust
Fuel to Bciiler
•*»• Steam to Process
Figure 20. Diagram of the Ultrafiltration-Reverse Osrnosis Wastewater Treatment System at Pacific Wood Treating Corp.
77
-------
To date, little information has been gathered on
the operation of the ultrafiltration-reverse osmosis
treatment scheme because of fouling and membrane
disruption in the unit To meet immediate zero dis-
charge permit limitations, Pacific Wood Treating ;
has modified its treatment sequence. The first modifica-
tion was designed to reduce the membrane fouling
problem by flocculating the troublesome components
from the wastewater. Four pounds of zinc chloride
and enough caustic soda to raise the pH to 8 to 9 were
added to each 1000 gal (3.785 m3) of wastewater
in a storage tank. The resulting rapidly settling floe is
removed from the bottom of the storage tank. The
zinc hydroxide floe pulls down iron and copper, as well
as organic materials with it. An estimated 90 percent
of the fouling materials are removed by the floe.
The floe that is removed from the bottom of the tank
is incinerated with sludge from other points in the treat-
ment process. Upon incineration the zinc enters the
flue gas, where it is helpful in reducing acidity, resulting
in longer bag life in the baghouse.
After oil ultrafiltration the wastewater is sent to
the boiler feedwater softener where it replaces about
half of the boiler water makeup. The result has been
longer water softener regeneration cycles, and less boiler
blowdown has more than paid for the zinc chloride
and caustic chemical costs.
The steam produced by the boiler has a phenolic odor
to it. Because of the high pH in the boiler (11.5 to 12)
most of the phenols end up in the boiler blowdown.
The boiler blowdown is added to the sludge and inciner-
ated. The system has not been operating long enough
to produce data on boiler fouling and corrosion
problems.
78
-------
Water Pollution Control
List of References
26U.S. Forest Service, Non-Point Water Quality
Modeling in Wildland Management: A State-
of-the-Art Assessment, v. I—Text, prepared
for EPA Environmental Research Laboratory,
Athens, GA, EPA 600/3-77-036, April 1977.
27U.S. Forest Service, Non-Point Water Quality
Modeling in Wildland Management: A State-
of-the-Art Assessment, v. II—Appendices,
prepared for EPA, Environmental Research
Laboratory, Athens, GA, EPA 600/3-77-078, July
1977.
28U.S. Forest Service, Silvicultural Activities and
Non-Point Pollution Abatement: A Cost-
Effectiveness Analysis Procedure, prepared for
EPA, Environmental Research Laboratory,
Athens, GA, EPA 600/8-77-018, November 1977.
29Unger, S. G., Environmental Implications of Trends
in Agriculture and Silviculture, v. I—
Trend Identification and Evaluation, Development
Planning and Research Associates for EPA,
Environmental Research Laboratory, Athens, GA,
EPA 600/3-77-121, October 1977.
30Development Planning and Research Associates,
Environmental Implications of Trends in Agri-
culture and Silviculture, v. II^Environmental
Effects of Trends. Prepared for EPA, Environmental
Research Laboratory, Athens, GA (In preparation
1977).
31Scaramelli, A. B. and F. A. DiGiano, "Wastewater
Treatments: Physical and Chemical Methods,''
Journal Water Pollution Control Federation, v. 47,
no. 6:1249, 1975,
32Ursic, S. J. and P. D. Duffy, "Hydrologic Performance
of Eroded Lands Stabilized with Pine," In Miss.
Water Resour. Conf. Proc., Miss. State Univ.
Mississippi State, MS, p. 203-216, 1972.
33 Anderson, Henry W., MarvinD. Hoover, and Kenneth
G. Reinhart, Forests and Water: Effects of Forest
Management on Floods, Sedimentation, and Water
Supply. USDA Forest Service, GTR. PSW-18,
Pacific Southwest Forest and Range Experiment
Station, Berkeley, CA, 1976, 115 p.
34Douglass, James E., "Southeastern Forests and the
Problem of Non-Point Sources of Water Pollution,"
In Non-Point Sources of Water Pollution, Pro-
ceedings, Southeastern Regional Conference,
Virginia Water Resources Center, Blacksburg, VA,
1975, p. 29-44.
35Lull, Howard W., "Runoff from Forest Lands."
In Man's Impact on Terrestrial and Oceanic
Ecosystems, W. H. Matthews, F. F. Smith,
and E. D. Goldbergs [eds.] MIT Press, Cambridge,
MA, 1971, p. 240-251.
36Patric, J. H., "Soil Erosion in the Eastern Forest,"
J. For., v. 74:671-677, 1976.
37Ursic, S. J., Hydrologic Effects of 'PrescribedBurning
and Deadening Upland Hardwoods in Northern
Mississippi, USDA Forest Service, Res. Pap.
SO-54, Southern Forest Experiment Station,
New Orleans, LA, 1970.
38Dickerson, B. P., Logging Disturbance on Erosive
Sites in North Mississippi. USDA Forest Service,
Res. Note 50-72, 4 p. Southern Forest Experiment
Station, New Orleans, LA, 1968.
39Dickerson, B. P., "Stormflows and Erosion after
Tree-Length Skidding on Coastal Plain Soils,"
Transactions, American Society of Agricultural
Engineers 18:867-868, 872, 1975.
40Dickerson, B. P., "Soil Compaction after Tree-
Length Skidding in Northern Mississippi," Journal
Soil Science Society of America, v. 40:965-966,
1976.
41Ursic, S. J., Planting Loblolly Pine for Erosion
Control in North Mississippi. USDA Forest
Service, Res. Pap. SO-3, Southern Forest Experi-
ment Station, New Orleans, LA, 1963.
42Kochenderfer, J. N., "Erosion Control on Logging
Roads in the Appalachians," USDA Forest Service,
Res. Pap. NE-158, Northeastern Forestry
Experiment Station, Upper Darby, PA, 1970,28 p.
43Thompson, W. S., "Status of Pollution Control
in the Wood Preserving Industry," Proc. American
Wood-Preservers' Assoc. 69:39-46, 1973.
44U.S. Forest Service, "Proceedings, Site Preparation
Workshop," U.S. Dep. USDA Forest Service,
Southeastern Area—State & Private For., Southern
Forest Experiment Station, and Kisatchie National
Forest, Alexandria, LA, 1977.
45Thompson, W. S., "Pollution Abatement by In-
plant Process Changes and Sanitation," In Pollu-
tion Abatement and Control in the Wood Preserving
Industry, W. S. Thompson, ed., Mississippi State
79
-------
University, Forest Products Laboratory, Missis-
sippi State, MS, 1970, 322 pp.
^Thompson, W. S., "Status of Pollution Control in the
Wood Preserving Industry in 1974. Proc. American
Wood-Preservers' Assoc., 1975. i
47American Petroleum Institute. Manual on Disposal
of Refinery Wastes, v. I. Wastewater Contain1
ingOil.(6thed.), 1959, 99 p. '
48White, J. T., T. A. Bursztynsky, J. D. Crane, and
R. H. Jones, "Treating Wood Preserving Plant
Wastewater by Chemical and Biological Methods,"
Grant No. 12100 HIG, Office of Research and
Development, U.S. Environmental Protection
Agency, Corvallis, OR, 1976, 97 pp.
49U.S. Environmental Protection Agency, Development
Document for Proposed Effluent Limitation ;
Guidelines and New Source Performance
Standards for the Plywood, Hardboard, and Wood
Preserving Segments of the Timber Products In-
dustty, EPA 440/1-73-023, 1973. '
s°Thompson, W. S., "File Report—Chemical Oxidation
Treatments," Unpublished Data, Mississippi
Forest Products Laboratory, Mississippi State,
Miss. 71:94-104,1973.
51Lin, Y. H. and J. R. Lawson, "Treatment of Oily and
Metal-Containing Wastewater," Poll. Eng., v. 5,
no. 11:45-49,1973.
"Goldsmith, R. L., D. A. Roberts, and D. L. Burre,
"Soluble Oil Waste Treatment Using Ultra-
filtration," Presented at 46 Annual Conference of
the Water Pollution Control Federation, Cleveland,
OH, 1973.
53Abcor, Inc., The Concentration ofPentachlorophenol
Wastes by Abcor Membrane System. Report sub-
mitted to Fullco Lumber Co., Haleyville, AL, 1974.
54Merten, U. and D. T. Bray, "Reverse Osmosis for
Water Reclamation," Advances in Water Pollution
Research, v. 3 Proc. 3rd Int'l. Conf., Water Poll.
Control Fed., Washington, D.C., 1966.
5SDruveI, W. A. and T. Helfgott, "Removal of Waste-
water Organics by Reverse Osmosis." Journal
Water Pollution Control Federation, v. 47:57,
1975.
56Hutchins, R. A., "Thermal Regeneration Costs,"
Chemical Engineering Program, v. 71, no. 5:80,
1975.
57Boen, D. F. and G. L. Jahannsen, Reverse Osmosis
for Treated and Un-Treated Secondary Sewage
Effluent. EPA Report 670/2-74-007,1974. |
58Rozelle, L. T., "Ultrathin Membranes for Treat-
ment of Waste Effluents by Reverse Osmosis,"
Appl. Polymer Symp. 22, p. 222. Chem. Abs. 80,
63554f, 1974.
59Morris, D. C., W. R. Nelson, and G. O. Walraven,
Recycle ofPapermill Wastewaters and Applica-
tion of Reverse Osmosis, ORM, EPA Program
No. 12040 FUB, 1972.
60Wiley, A. J., G. A. Dubey, and J. K. Bansol, "Reverse
Osmosis Concentration of Dilute Pulp and Paper
Effluents," Pulp Manufacturers' Research League
and Inst. of Paper Chem., EPA Proj. No. 12040
EEL, 1972.
61Pacific Wood Treating Corp. "Wood Preserving
Waste Recycle and Incineration System," EPA
Proj. Control No. S805179 01, 1976.
62Fang, H. H. and E. S. Chain, "Removal of Alcohols
and Aliphatic Acids in Aqueous Solution by
NS-100 Membranes," Journal Applied Polymer
Science, v. 19:1347, 1975.
63Henshaw, T. B., 1971. "Adsorption/Filtration
Plant Cuts Phenols from Effluent," Chemical
Engineering, May 1971, p. 47.
64Hager, D. G., "Industrial Wastewater Treatment
by Granular Activated Carbon," Industrial
Water Engineering, Jan/Feb 1974, p. 14.
65Baker, D. A., "Petroleum Processing Wastes,"
Journal Water Pollution Control Federation,
v. 47 no. 6:1476, 1976.
66DeJohn, P. B. and A. D. Adams, "Activated Carbon
Improves Wastewater Treatment," Hydrocarbon
Proc. October 1975, p. 104.
67Salo, Ernest O. and R. L. Rulifson, Sediment
and Water Quality: A Review of Literature In-
cluding a Suggested Approach for Water Quality
Criteria, EPA Region X, EPA 910/9-78-048,
1978.
68Suhr, L. G. and G. L. Gulp, "State of the Art--
Activated Carbon Treatment of Wastewater,"
Water and Sewage Works, Ref. No. R104, 1974.
69Rosfjord, R. E., R. B. Trattner, and P. N. Cheremisinoff,
"Phenols: A Water Pollution Control Assess-
ment," Water and Sewage Works, v. 123, no, 3:96,
1976.
70Stevens, B. W. and J. W. Kerner, "Recovering
Organic Materials from Wastewater," Chemical
Engineering, February 1975, p. 84.
"Morton, S. C. and E. W. Sawyer, "Clay Minerals
Remove Organics, Viruses and Heavy Metals from
Water," Water and Sewage Works, 1976.
72U.S. Environmental Protection Agency, Development
Document for Effluent Guidelines and New Source
Performance Standards for the Wet Storage,
Sawmills, Particleboard, and Insulation Board
Segment of the Timber Products Processing Point
Source Category, EPA 440/1-74-03 3, August
1974.
80
-------
Water Pollution Control
Bibliography
Bodein, D. B., Plywood Plant Blue Wastes Disposal.
Federal Water Pollution Control Administration,
Northwest Region, Pacific Northwest Water
Laboratory, U.S. Department of Interior, 1969.
Dallons, V., Comparison of Pollution Loads Resulting
from Use of Different Wood Species in the Pro-
duction ofHardboard. Industrial Environmental
Research Laboratory, EPA, Cincinnati, Corvallis
Field Station, 1976.
Environmental Science and Engineering, Inc.
Technical Review of the Best Available Technology,
Best Demonstrated Technology, and Pretreatment
Technology for the Timber Products Processing
Point Source Category. Prepared for the U.S. ',.
Environmental Protection Agency, July 1977. ,
Glass, George Graham, Jr., "The Effects from
Rootraking on an Upland Piedmont Loblolly Pine
(P. Taeda L.) Site," North Carolina State University
School of Forest Resources, Tech. Rep. v. 56,
1976, 44 p.
Gran, G., Wastewater from Fibreboard Mills. .
In. Pure and Applied Chemistry, v. 29, 1972.
Kremen, S. S., "Reverse Osmosis makes High :
Quality Water Now," Environmental Science •
and Technology, v. 9:314, 1975.
MacDonald, R. G. and J. N. Franklin, The Pulping
of Wood, Second Edition, v. I, McGraw-Hill,
New York, 1969.
Miskovsky, Milan C. and Matthew B. Van Hook,
"Regulation of Forestry Related Nonpoint Source
Pollution under the Federal Water Pollution '
Control Act Amendments of 1972," Nat. Resour.
Lawyer, v. 9:645-671, 1977.
National Commission on Water Quality, "Staff
Report to the National Commission on Water Quality,
Part I, Introduction and Summary of Findings,"
U.S. GPO Supt. of Documents, Washington, D.C.,
1976, 79 p. ;
Newton, Michael and J. A. Norgren, Silvicultural,
Chemicals and Protection of Water Quality, Oregon
State University for EPA, EPA 910/9-77-036,
1977.
Pair, Claude H., Walter W. Hinz, Crawford Reid,
and Kenneth R. Frost, "Sprinkler Irrigation,"
Editor's Press, Hyattsville, MD, 1969.
Schreiber, J. D., P. D. Duffy, and D. C. McClurkin,
"Dissolved Nutrient Losses in Storm Runoff from
Five Southern Pine Watersheds," Journal of
Environmental Quality, v. 5, no. 2:201-205, 1976.
Sopper, W. E., "Effects of Timber Harvesting and
Related Management Practices on Water Quality in
Forested Watersheds," Journal of Environmental
Quality, v. 4, no. 1:24-29, 1975.
Thompson, Warren S., and Mississippi Forest
Products Laboratory, Draft Development Docu-
ment for Pretreatment Standards for the
Wood Preserving Industry, Prepared for the
U.S. Environmental Protection Agency, July
1976.
U.S. Environmental Protection Agency, Draft
Development Document for Effluent Limitations
Guidelines and New Source Performance Stand-
ards, Timber Products Processing Industry,
Phase I, June 1973.
U.S. Environmental Protection Agency, Draft
Development Document for Effluent Limitations
Guidelines and New Source Performance Stand-
ards, Timber Products Processing Industry,
Phase II, December 1973.
U.S. Environmental Protection Agency, Aerobic
Secondary Treatment of Plywood Glue Wastes,
R2-73-195, April 1973.
U.S. Forest Service, "The Impact of Timber Harvest
on Soils and Water," Report on the President's
Advisory on Timber and Environment, 1973.
U.S. Forest Service, An Analysis of Research and
Development Needs on Non-Point Source Water
Pollution on Forest and Rangelands, for EPA,
Environmental Research Laboratory, Athens, GA,
January 1977.
Ursic, S. J., "Harvesting Southern Forests: A Threat to
Water Quality?" In Non-Point Sources of Water
Pollution, Proc., Southeast. Reg. Conf., Va.
Water Resour. Cent., Blacksburg, VA, p. 145-151,
1975.
81
11. S. GOVERNMENT PRINTING OFFICE: 1979-659-511
-------
------- |