United States
Environmental Protection
Agency
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completi
1. REPORT NO. 2.
EPA-600/R-96-067
3. I
4. TITLE AND SUBTITLE
Sources and Factors .Affecting Indoor Emissions from
Engineered Wood Products; Summary and Evaluation
of Current Literature
5. REPORT DATE
June 1966
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Sonji Turner, Cybele Martin, Robert Hetes, and
Coleen Northeim
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P. C. Box 12194
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
CR822003-01
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 1/94 - 8/95
14. SPONSORING AGENCY CODE
EPA/600/13
ib. supplementary notes ^pp^jQ project officer is Elizabeth S. Howard, Mail Drop 54, 819/
541-7915.
m g> & DCTQ A
The report summarizes information from the first two phases of a five-
phase cooperative effort between Research Triangle Institute and the Indoor Environ-
ment Management Branch of the TJ. S. Environmental Protection Agency's Air Pollu-
tion Prevention and Control Division to apply pollution prevention techniques to re-
duce indoor air emissions from engineered wood products. Research objectives of
the effort are to characterize indoor air emissions from engineered wood products
and to identify and evaluate pollution prevention approaches for reducing indoor air
emissions from these products. The research's five phases are to: (1) evaluate exis-
ting data and testing methodologies; (2) convene research planning meetings; (3) se-
lect high-priority emissions sources; (4) evaluate high-priority emissions sources;
and (5) develop and demonstrate pollution prevention approaches for reducing indoor
air emissions from high-priority sources. Information presented here will be used
to select reconstituted wood components with various finishing and resin systems for
initial emissions screening. (NOTE: Reconstituted engineered wood components--
e.g., particleboard and medium-density fiberboard--are common to several types
of consumer wood products—e.g., residential and ready-to-assemble furniture and
kitchen cabinets.)
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFtERS/OPEN ENDED TERMS
o. COSATI Ficld/Gioup
Pollution Organic Compounds
Wood Products Volatility
Composite Materials
Particle Boards
Fiberboards
Polymers
Formaldehyde
Pollution Control
Stationary Sources
Engineered Wood Pro-
ducts
Wood Finishing
Volatile Organic Com-
pounds (VOCs)
13 B
11L 20 M
11D
11G
07D
07 C
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
93
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 <9-73)
-------
EPA-600/R-96-067
June 1996
SOURCES AND FACTORS AFFECTING INDOOR
EMISSIONS FROM ENGINEERED WOOD PRODUCTS:
Summary and Evaluation of Current Literature
by:
Sonji Turner
Cybele Martin
Robert Hetes
Coleen Northeim
Research Triangle Institute
P.O. Box 12194
Center for Environmental Analysis
Research Triangle Park, NC 27709-2194
EPA Cooperative Agreement No. CR822002-01
EPA Project Officer: Elizabeth M. Howard
National Risk Management Research Laboratory
Research Triangle Park, NC 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, D.C. 20460
-------
FOREWORD
The U. S. Environmental Protection Agency is charged by Congress with pro-
tecting the Nation's land, air, and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions lead-
ing to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA's research
program is providing data and technical support for solving environmental pro-
blems today and building a science knowledge base necessary to manage our eco-
logical resources wisely, understand how pollutants affect our health, and pre-
vent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for
investigation of technological and management approaches for reducing risks
from threats to human health and the environment. The focus of the Laboratory's
research program is on methods for the prevention and control of pollution to air,
land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites and groundwater; and prevention and
control of indoor air pollution. The goal of this research effort is to catalyze
development and implementation of innovative, cost-effective environmental
technologies; develop scientific and engineering information needed by EPA to
support regulatory and policy decisions; and provide technical support and infor-
mation transfer to ensure effective implementation of environmental regulations
and strategies.
This publication has been produced as part of the Laboratory's strategic long-
term research plan. It is published and made available by EPA's Office of Re-
search and Development to assist the user community and to link researchers
with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
EPA REVIEW NOTICE
This report has been peer and administratively reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Information
Service, Springfield, Virginia 22161.
-------
ABSTRACT
Reconstituted engineered wood components (e.g., particleboard and medium-density
fiberboard) are common to several types of consumer wood products (e.g., residential and
ready-to-assemble furniture and kitchen cabinets). The selection of resins used to bind the
components, coatings, and laminates applied to the components to produce the final products
affects emissions of formaldehyde and other volatile organic compounds from the products to
the indoor environment. Research Triangle Institute is collaborating with the Indoor
Environment Management Branch of the U.S. Environmental Protection Agency's Air
Pollution Prevention and Control Division on a cooperative agreement entitled, "The
Application of Pollution Prevention Techniques to Reduce Indoor Air Emissions from
Composite Wood Products." The research objectives are to characterize indoor air emissions
from engineered wood products and to identify and evaluate pollution prevention approaches
for reducing indoor air emissions from these products.
The research has a five-phase approach: (1) evaluate existing data and testing methodologies;
(2) convene research planning meetings; (3) select high-priority emissions sources; (4) evaluate
high-priority emissions sources; and (5) develop and demonstrate pollution prevention
approaches for reducing indoor air emissions from high-priority sources. This report
summarizes information from the first two phases of research. Information presented here will
be used to select reconstituted wood components with various finishing and resin systems for
initial emissions screening.
ii
-------
Table of Contents
Section Page
Abstract ii
List of Figures vi
List of Tables vi
List of Acronyms viii
Conversion Table ix
1.0 Introduction 1
2.0 Sources of Indoor Emissions from Engineered Wood Products 2
2.1 Resins Used to Manufacture Engineered Wood 2
2.2 Finishing Materials 4
2.3 Emissions Rates 4
2.4 Effects of Physical Parameters on Emissions 6
2.5 Health Effects Related to Indoor Air Quality 7
3.0 Engineered Wood Product Classifications and Industry Statistics 8
3.1 Plywood Panels 9
3.1.1 Structural Plywood 9
3.1.1.1 Industry Outlook 9
3.1.2 Hardwood Plywood 12
3.1.2.1 Wall Paneling 12
3.1.2.2 Industrial Hardwood Plywood 14
3.2 Engineered Lumber 14
3.2.1 Industry Outlook 17
3.3 Reconstituted Wood Panels 17
3.3.1 Particleboard 17
3.3.1.1 Industry Outlook 18
3.3.2 Oriented Strandboard 20
3.3.2.1 Industry Outlook 20
3.3.3 Hardboard 20
3.3.4 Medium-Density Fiberboard 22
3.3.4.1 Industry Outlook 22
3.3.5 Cellulosic Fiberboard 24
4.0 Engineered Wood Manufacturing Process Descriptions 24
4.1 Plywood Manufacture 24
4.1.1 Debarking 24
4.1.2 Heating the Blocks 24
4.1.3 Cutting Veneer 25
iii
-------
Table of Contents (continued)
Section Page
4.1.4 Veneer Storage and Clipping 25
4.1.5 Veneer Drying 27
4.1.6 Lay-up and Pressing 27
4.1.7 Finishing 28
4.2 Reconstituted Panel Manufacture 28
4.2.1 Wood Reduction 30
4.2.1.1 Oriented Strandboard 30
4.2.1.2 Particleboard 30
4.2.1.3 Fiberboard (Cellulosic Fiberboard, MDF,
and Hardboard) 30
4.2.2 Drying 30
4.2.3 Screening and Air-Classifying 31
4.2.4 Blending 31
4.2.4.1 Particleboard and Oriented Strandboard 31
4.2.4.2 Medium-Density Fiberboard and Dry Process
Hardboard 32
4.2.5 Mat Forming 32
4.2.5.1 Cellulosic Fiberboard, Wet and Wet/Dry
Process Hardboard 32
4.2.5.2 Particleboard, Oriented Strandboard, Medium-Density
Fiberboard, and Dry Process Hardboard 33
4.2.6 Hot Pressing 33
4.2.7 Finishing 33
5.0 Emissions During Manufacture of Engineered Wood Products 38
6.0 Source Management/Pollution Reduction: Approaches and Issues 40
6.1 Pollution Prevention 40
6.1.1 Alternatives to Wood Feedstock 40
6.1.2 Alternative Resins 41
6.1.2.1 Low Mole Ratio Urea-Formaldehyde Resins 41
6.1.2.2 Use of Formaldehyde Scavengers with UF Resins 42
6.1.2.3 Melamine-Fortified Urea-Formaldehyde Resins 45
6.1.2.4 Phenol-Formaldehyde Resins 46
6.1.2.5 Methylenediphenyl Diisocyanate 47
6.2 Novel Adhesives 47
6.2.1 FAREZResin 48
6.2.1.1 Environmental Effects 48
iv
-------
Table of Contents (continued)
Section Page
6.2.1.2 Availability 49
6.2.2 Methyl Glucoside 49
6.2.2.1 Future Availability . . . . 50
6.2.3 Lignin Adhesives 50
6.2.3.1 Adhesive Utilization 50
6.2.3.2 Availability 51
6.2.4 Tannin 51
6.2.4.1 Adhesive Utilization 51
6.3 Source Management Approaches 52
6.3.1 Use/Maintenance 52
6.3.2 Use of Laminates or Veneers 52
6.3.3 Bakeout 53
7.0 Conclusion 53
References 54
Appendix A Chemistry of Formaldehyde-Based Resins A-l
Appendix B Summary of Indoor Air Emissions Research B-l
v
-------
List of Figures
Fieure Page
3-1 Various types of plywood construction 10
3-2 U.S. shipments of hardwood plywood and structural plywood 11
3-3 U.S. shipments of structural plywood 1987 to 1993 13
3-4 U.S. production of hardwood plywood wall paneling 15
3-5 U.S. shipments of industrial hardwood plywood 16
3-6 U.S. industry shipments of particleboard 1978 to 1993 19
3-7 U.S. structural panel market 21
3-8 U.S. shipments of medium-density fiberboard 23
4-1 Cutting action on a lathe and slicer 26
4-2 Reconstituted wood panel process flow 29
4-3 Blending system diagram 32
4-4 Two types of mat forming machines 34
4-5 Various types of mat construction 35
4-6 Schematic of multiopening board press 36
4-7 Schematic of continuous press 37
6-1 Urea-formaldehyde resin mole ratios 42
6-2 Formaldehyde emissions versus F:U mole ratio 43
6-3 Emissions and properties 43
6-4 Estimated F:U molar ratios with scavenger solution 44
6-5 Combination blending formaldehyde emissions, desiccator results 45
6-6 Combination blending formaldehyde emissions, large-scale test chamber results ... 46
A-l Manufacturing route to isocyanatc A-7
B-l TVOC concentrations B-ll
B-2 TVOC emissions rates B-15
List of Tables
Table Page
2-1 Commonly Used Adhcsives 5
3-1 Engineered Wood Products 8
3-2 U.S. Production Estimates of Engineered Lumber 17
3-3 U.S. Shipments of Reconstituted Wood Panel 1985 to 1993 18
5-1 Reported Releases of SARA Section 313 Chemicals from Plywood Plants for 1991 . 38
5-2 Reported Releases of SARA Section 313 Chemicals from Reconstituted
Wood Panel Plants for 1991 39
vi
-------
List of Tables (continued)
Table Page
B-l Results of analysis of Air Around Materials B-2
B-2 Typical Predicted Emissions Rates for Sources in 400-m2 Office Area B-3
B-3 Sources of Indoor Organic Compounds B-3
B-4 Concentration Measurements of Organics in a Ventilated Building B-5
B-5 Range of Formaldehyde Emissions Rates for Seven Composite Wood Products . . B-5
B-6 VOC and Formaldehyde Emissions from Various Engineered Woods B-7
B-7 Percent of Higher Aldehydes (Hexanal) and Terpenes of Total VOC B-7
B-8 NPA Target Compounds B-8
B-9 Sample Container Contaminants B-9
B-10 Test Specimens B-10
B-l 1 Formaldehyde Emissions Data from Engineered Wood Products B-12
B-12 Formaldehyde Emissions Data from Particleboard and MDF B-13
vii
-------
LIST OF ACRONYMS
ACH
Air changes per hour
APA
American Plywood Association
APPCD
Air Pollution Prevention and Control Division
ASHRAE
American Society of Heating, Refrigerating and Air-Conditioning Engineers
ASTM
American Society for Testing & Materials
CARB
California Air Resources Board
DNPH
Dinitrophenylhydrazine
EPA
Environmental Protection Agency
EPCRA
Emergency Planning and Community Right-to-know Act
ER
Emissions rate
ERG
Environmental Resource Guide
FEV
Forced expiratory volume
FID
Absorbent flame ionization detector
FVC
Forced vital capacity
GC/MS
Gas chromatography-mass spectrometry
GTRI
Georgia Tech Research Institute
HEM
Human exposure model
HPLC
High-performance liquid chromatography
HPVA
Hardwood Plywood and Veneer Association
HUD
Department of Housing and Urban Development
HVAC
Heating, ventilation, and air conditioning
IAB
Indoor Air Branch
IAQ
Indoor air quality
IVOC
Individual volatile organic compound
LEM
Low-emitting materials
MDF
Medium density fiberboard
MDf
Methylenediphenyl diisocyanate
MEK
Methyl ethyl ketone
MIK
Methyl isobutyl ketone
MMSF
Million square feet
NAAQS
National Ambient Air Quality Standards
NCASI
National Council of the Paper Industry for Air and Stream Improvements, Inc
NEM
NAAQS Exposure Model
NIOSH
National Institute of Occupational Safety and Health
NOAEL
No observed adverse effects level
NPA
National Parlicleboard Association
NRC
National Research Council
OSB
Oriented strandboard
PB
Particleboard
PF
Phenol formaldehyde
PMDI
Popolynier methylenediphenyl diisocyanate
PPM
Parts per million
PRF
Phenol resorcinal formaldehyde
PTFli
I'olytetrafluorethylene
QA
Quality assurance
RTA
Ready to assemble
RTI
Research Triangle Institute
SARA
Superfund Amendments and Reauthorization Act
SBS
Sick Building Syndrome
SYP
Southern yellow pine
TGNMO
Total gaseous nonmethane organics
TRI
Toxics Release Inventory
TVOC
Total volatile organic compound
UF
Urea-formaldehyde
voc
Volatile organic compound
WB
Waferboard
viii
-------
CONVERSION TABLE
To Convert from:
Length
foot (fl)
inch (in)
Area
square foot (ft2)
square inch (in2)
Area/Volume
square foot/cubic foot
(a2/fp)
Volume
ft3
cubic inch (in3)
Time
minute (min)
hour (h)
Mass
U.S. ton (t)
pound-mass (Ib„.)
Temperature
°F
Density
lb.,/ft1
FIowrate/Area
lhn/(h • ft2)
Velocity (linear)
ft/'min
ft/s
To:
meter (m)
centimeter (cm)
millimeter (mm)
cm
m
square centimeter (cm2)
square meter (m2)
square millimeter (mm2)
cm2
m2
m2/nv
cubic centimeter (cm1)
m3
second (s)
s
min
Mg
kg
kg
g
°C
kg/m3
g/m'
g/cm3
kg/(s • m2)
m/s
cm/s
nv's
cm/s
Multiply bv:
0.3048
30.48
25.4
2.54
2.54 x JO"2
929
0.0929
0.06452
6.452
6.452 x 10"4
3.28
0.02817
16.387
1.6387 x 10"5
60
3600
60
0.9018
907.18
0.4536
453.6
5/9(°F - 32)
16.02
16,020
0.01602
0.00136
5.08 x 10"3
0.5080
0.3048
30.48
IX
-------
1.0 INTRODUCTION
Existing approaches for improving indoor air quality (IAQ) have frequently focused on control
strategies, such as increased or improved ventilation. Although these approaches can result in
improved IAQ, they may also be inefficient. It is widely accepted in the IAQ research
community that source management is the method of choice to reduce indoor air pollution.
Source management in the context of IAQ refers to practices that: (1) reduce the overall source
strength; (2) render the source more amenable to control; or (3) eliminate the source entirely.
The U.S. Environmental Protection Agency's (EPA's) Air Pollution Prevention and Control
Division (APPCD) is responsible for much of EPA's indoor air engineering research and seeks to
integrate IAQ and pollution prevention (source reduction) into a strategic approach to indoor
source management. APPCD's Indoor Environment Management Branch (IEMB) Pollution
Prevention/IAQ research objective is lo employ accepted pollution prevention techniques (e.g.,
process modification and product reformulation) to reduce indoor air contamination through the
development of "low-emitting materials'" (LEM). An LEM is a material designed to emit fewer
emissions when used in the same manner as another material in the same indoor environment.
APPCD is investigating the application of pollution prevention approaches to reduce indoor air
emissions from specific indoor sources. The APPCD has selected engineered wood products for
the initial phase of their efforts. Engineered wood products can be a significant source of indoor
air contamination. Engineered wood products comprise basic building materials, such as
underlay, flooring, cabinets, and common furnishings, and are ubiquitous to most indoor settings.
The Research Triangle Institute (RTI) is collaborating with IEMB on a cooperative agreement
entitled, "The Application of Pollution Prevention Techniques to Reduce Indoor Air Emissions
from Composite Wood Products."' The objectives of this research are to characterize indoor air
emissions from engineered wood products, then to identify and evaluate pollution prevention
approaches such as the development of LEM for use indoors. This research will be conducted in
five phases: (1) evaluate existing data and testing methodologies; (2) convene research planning
meetings; (3) select high-priority emissions sources; (4) evaluate high-priority emissions sources;
and (5) develop and demonstrate pollution prevention approaches for reducing the indoor air
emissions from the selected high-priority sources.
This report summarizes information collected in the first two phases of this project. The
information presented here, along with information from outside technical advisors, will be used
to select products for further research. The technical advisors include trade association
representatives, industry representatives, and technical assistance providers. Extensive source
characterization will be carried out on the selected products. Pollution prevention approaches
will be identified and applied. These improved products will be evaluated through quantitative
emissions measurements to determine the technical and economic feasibility, overall pollution
prevention potential, and indoor air quality benefits of the pollution prevention approaches.
-------
2.0 SOURCES OF INDOOR EMISSIONS FROM ENGINEERED WOOD
PRODUCTS
Research over the past two decades has indicated that engineered wood products can be sources
of a large number of organic compounds, particularly formaldehyde. Indoor emissions from
engineered wood products can arise from the raw engineered wood (both the wood and resin),
the finishing materials applied to the boards for decorative purposes, and the glues used to adhere
pieces of engineered wood together. In addition to acting as sources of organic compounds,
engineered wood products may also behave as adsorbers and re-emitters of these compounds that
are emitted from other sources, such as carpeting, paints, and environmental tobacco smoke
(Neretnieks et al., 1993).
Emissions data are abundant in the literature on formaldehyde emissions associated with resins
used to bond engineered wood, primarily urea-formaldehyde (UF) resins. Some emissions data
also exist on formaldehyde and other organics emitted naturally from the wood. Currently, two
major research studies are being conducted to characterize volatile organic compounds (VOC)
from various types of raw engineered wood. One of these research studies is being conducted at
the U.S. Department of Agriculture's Forest Products Laboratory in Madison, Wisconsin. This
research is characterizing emissions from different wood species and UF resins used to construct
particleboard (PB). The other research study is being conducted by Forintek, a renowned
Canadian forest products laboratory. Forintek is characterizing emissions from different species
of wood used to construct oriented strandboard (OSB), another type of engineered wood.
At present there is very little data on how finishing materials contribute to emissions from
engineered wood products. Finisliing materials in this context mean any type of material added
to raw engineered wood such as paper, wood veneer, ink prints, and vinyl. Some of these
finishes may contribute to a product's overall emissions. For example, many paper finishes that
are applied to raw engineered wood are impregnated with resins; these resins are added during
the paper making process to make the paper more durable. In addition, many papers and other
types of overlays are coated with materials, usually resins, to give them a protective finish. Thus,
an evaluation and understanding of the indoor air emissions associated with finishing materials is
a critical step in the development and evaluation of pollution prevention opportunities such as
low-emitting materials.
2.1 Resins Used to Manufacture Engineered Wood
Both the resins and the wood used to manufacture engineered wood products are sources of
organic emissions. Urea-formaldehyde resins are the most commonly used adhesives for
engineered wood manufacture in the United States. One of the major disadvantages of
urea-formaldehyde resins is their susceptibility to hydrolytic degradation in the presence of
moisture and/or acids. Hydrolysis of UF resins occurs when resins are exposed to water or to
humid conditions. The resins absorb moisture resulting in the slow release of formaldehyde from
2
-------
UF resins. The hydrolysis reaction mechanism is essentially the reverse reaction of the UF resin
formation reaction (NRC, 1981).
Phenol-formaldehyde (PF) resins are the second most commonly used resin. Phenolic resins are
used because of their desirable physical and chemical properties; they resist hydrolysis. Phenolic
resins are relatively inexpensive compared to alternative resins, such as melamine, but they are
more costly than urea-formaldehyde resins.
Methylenediphenyl diisocyanate (MDI) resins are the third most commonly used type in the
United States. MDI is used in the production of approximately one third of the U.S. market of
OSB and in the manufacture of other specialty products. The other two thirds of the U.S.
production of OSB is made with PF resins. Currently, MDI binders are used in at least 18 OSB
facilities in North America. In the United States. MDI binders are used to produce engineered
structural lumber products and MDF. MDF produced with MDI binders is used to make
exterior-grade products and low-formaldehyde-emilting interior-grade products. In Europe, MDI
binders are used to produce both PB and MDF, resulting in a reduction of formaldehyde
emissions from the finished panel (ICI, personal communication, 1994). Products made with
MDI emit little formaldehyde (Meyer et al., 1986). When compared with the other three resins
discussed. MDI has adhesive properties equivalent to, if not better than, phenolic formaldehyde
resins, and its water resistance is superior (Meyer, 1979). Higher price, the tendency of the
adhesive to adhere to manufacturing equipment, and worker exposure issues have limited MDI
applications. These adhesives are expensive and require expensive manufacturing processes
(Meyer etal., 1986).
Melamine-formaldehyde resins are used extensively in Europe and Scandinavia, mainly to
produce exterior board products and interior-grade products that meet European indoor emissions
standards. Melamine resins are well established in the European market and have been popular
in producing interior-grade plywood. Melamine products emit more formaldehyde than phenolic
resins but significantly less than UF products (Meyer, 1979).
Resorcinol-formaldehyde resins are waterproof and form the strongest wood bond of the
previously mentioned formaldehyde-based resins (Pizzi, 1983); however, they are the most
expensive (Pizzi, 1983; Sellers et al., 1988). Resorcinol-formaldehyde resins are the most
durable and strongest of the four formaldehyde resins mentioned (Koch et al., 1987). They are
often described as the "ideal adhesive" due to their durable, waterproof bonding and short curing
times at ambient temperature (Koch et al., 1987); however, resorcinol-formaldehyde resins are
veiy expensive ($ 1.45 per pound in 1985 dollars). To obtain some of the quality of a resorcinol
resin at a lower cost, phenol-resorcinol-formaldehyde (PRF) resins have been developed. These
highly durable resins are produced by combining phenol with formaldehyde under mildly
alkaline conditions, followed by a resorcinol addition and completion of the synthesis (Sellers
et al., 1988).
Often, more expensive formaldehyde-based resins, such as melamine- and resorcinol-
formaldehyde resins, are blended with other formaldehyde-based resins (Koch et al., 1987;
3
-------
Blomquist et al., 1981). Melamine-urea-formaldehyde and phenol-resorcinol-formaldehyde are
two examples of these blended systems. These are combined to provide improved performance
using the advantageous properties of each constituent in the system (Koch et al., 1987;
Blomquist et al., 1981) while reducing expense. Melamine-formaldehyde resins (Meyer et al..
1986), melamine-urea-formaldehyde resins, PF resins, and MDI are used largely in the European
engineered wood markets (WPPF, 1994). Table 2-1 shows typical applications and mixed cost
per pound for some of these resins, and a discussion of the chemistry of each of these resinous
products is provided in Appendix A.
2.2 Finishing Materials
Types of finishing materials used on engineered wood materials include laminates, edge-bands,
adhesives for attaching laminates and edge-bands, conversion varnish coatings, paints, stains, fire
retardants, and preservatives. Some of these finishing materials are sources of organic emissions
while others, such as laminates or veneers, may help reduce emissions (the usefulness of
laminates and veneers for reducing emissions is discussed in Section 6.3.2). Appendix B
presents summaries of several studies of indoor emissions associated with finished and
unfinished engineered wood materials made with various resins.
2.3 Emissions Rates
In 1984, the Department of Housing and Urban Development (HUD) established formaldehyde
product standards for plywood and PB materials bonded with a resin system or coated with a
surface finish containing formaldehyde when installed in manufactured homes* (Federal
Register, 1984). Many plywood and PB manufacturers changed their products to comply. Most
of this section focuses on recent indoor air studies and presents studies that evaluate various
sources of indoor air emissions and how the emissions rates are affected by various parameters/
Appendix B presents emissions rate information from several studies, and Section 2.4 presents
the effects of several physical parameters, such as temperature and humidity, on indoor air
emissions from engineered wood products.
One study, presented in Appendix B, concluded that new building materials produced high
emissions levels, but with effective ventilation (the ventilation rate was 2.5 air changes per hour)
The HUD safety standards (24 CFR 3280.308) for certified plywood and PB used in manufactured home
construction require that formaldehyde emissions not exceed 0.2 ppm (0.246 mg/m3) from plywood and 0.3 ppm
(0.369 mg/mJ) for PB, as measured by the specified air chamber test. The specified air chamber test is the
Large-Scale Test Method FTM 2-1983. Individual engineered wood products are tested in accordance with the
following loading ratios: plywood—0.29 fl2/ftJ (or 0.369 nr/m') and PB—0.13 ft2/ftJ (or0.43 m2/m3) (24 CFR
3280.406). Using the operation conditions specified in the Large-Scale Test Method FTM 2-1983 and the
formaldehyde emissions rate equation, formaldehyde emissions rates are 0.13 mg/m2 • h (2.66 x 10"s 1 b/ft2 • h)
for plywood and 0.43 mg/m2 • h (8.81 x 10"8 lb/fl2 • h) for PB.
+ The reader is referred to Godish (1988) for an evaluation of additional indoor air studies performed before 1984.
4
-------
Table 2-1. Commonly Used Adhesives
Adhesive Resin
Typical Application
Mixed Cost/lb"
Urea-formaldehyde
Phenol-formaldehyde
IIW flooring, Type II plywood
(decorative), PB, and fiberboard;
interior exposure; furniture; wood
veneering
Structural plywood, truss components.
PB, OSB, and WB; exterior exposure;
boat building; furniture
0.08
0 10 (plywood)
0.34 (OSB, liquid)
0.55 (WB,
spray-dried)
Melamine/Urea-formaldehyde
Isocyanates
Melamine-formaldehyde
Phenol-resorcinol
formaldehyde
Resorcinol-formaldehyde
Plywood (decorative), flat-bed stock, 0.30
and end joints in laminating; interior
and limited exterior exposure; furniture
WB, OSB, and PB; interior and exterior 0.70
exposure
Laminated beams (moderate in use in 0.40
U.S.), end joints in laminating truck
decking, and Type 1 HW plywood
(decorative); limited exterior exposure
Bridge and pier components, boat 1,30
building, laminated beams, and truck
decking; interior and exterior exposure
Laminates, ship components, outdoor 1.45
furniture, and firerated panels; extreme
exterior service
Sources: Sellers et al.. 1988;Pizzi, 1983.
IIW = llardwood.
OSB = Oriented strandboard.
PB -- Partieleboard.
WB = Wafcrboard.
a In 1985 dollars.
5
-------
these emissions could be reduced. This study covered measurements taken for an 8-month
period (Saarela and Mattinen, 1993). The maximum concentration of formaldehyde reached
0.122 mg/m3 in the new office building investigated for the Saarela study. Another indoor
analysis suggested that building materials may be the main source of organic compounds in the
indoor environment (Melhave. 1982). The total average concentration for the most frequently
identified compounds in this study was 72.96 //g/m3, and the average arithmetic mean emissions
rate for all the identified compounds was 9.5 mg/m2 • h (Molhave, 1982). The Levin et al., study
(1989) presented predicted emissions rates, and the Meyer study (1983) presented emissions rate
ranges for several engineered wood products. The remaining studies analyzed samples of
various engineered wood products (Sundin, 1992; GTRI, 1993; GEOMET, 1994: Tlwgersen
et al., 1993). The National Particleboard Association (NPA) had a two-part preliminary study
conducted by two laboratories, GEOMET Technologies and Georgia Tech Research Institute.
The GEOMET study (1994) analyzed emissions from finished and unfinished engineered wood
products. For most of the finished samples of Southern Yellow Pine (SYP) substrates, the total
volatile organic compounds (TVOC) concentrations (with ranges from 156 //g/m3 at 24 hours to
2,520 //g/m3 at 120 hours) were lower than unfinished SYP substrates (with ranges of 2,880
//g/m3 at 24 hours to 918 //g/m3 at 120 hours). The results of these analyses are presented in
Appendix B. In addition to measuring organic emissions for water-damaged chipboard, which
had a mean formaldehyde concentration of about 0.0475 mg/m3, the Th0gersen et al. study
(1993) evaluated microbiological growth from water-damaged chipboard samples and revealed
significant growth of fungi following water damage. Although not discussed in Appendix B,
EPA performed earlier research on a variety of consumer products and building materials that
presented emissions rate data and discussed the effect of temperature and air exchange on the
emissions rate (Tichenor and Mason, 1988).
Reported information concerning organic and formaldehyde emissions rates and concentrations
was dependent upon the design and objectives of each study. Emissions data for each individual
study were collected using different analytical test methods. Because methods used to collect
emissions data were study-dependent and researchers presented emissions data differently,
comparative conclusions could not be drawn between the studies presented in this report.
2.4 Effects of Physical Parameters on Emissions
The indoor environment is not static. Temperature, humidity, and air exchange rates vary
continually over the course of a day, from day to day, and seasonally. These environmental
variations influence indoor concentrations. In addition, many other parameters affect indoor
concentrations, including types and quantities of materials used in construction and furnishing,
volume-to-surface ratio of those materials, kind and method of air control (ventilation), indoor
activities, and structural tightness of energy-efficient buildings.
Formaldehyde emissions from engineered wood increase with increasing temperatures and
decrease with decreasing temperatures. Formaldehyde emissions are sensitive to absolute and
relative humidity. Diurnal variations of formaldehyde concentrations coincide well with changes
6
-------
in indoor temperature. Emissions are highest during summer months when temperature is high,
and emissions are lowest during winter months when temperature is low. Formaldehyde
emissions also correlate linearly with average ambient temperatures under controlled indoor
environmental conditions. Outdoor temperature changes alter the pressure differences between
the inside and outside of the structure. These pressure differences are one major factor
contributing to air infiltration into the indoor environment. Formaldehyde concentrations are
significantly affected by ventilation. When indoor temperature and humidity are held constant,
formaldehyde concentrations will be higher under closure conditions than when windows and
doors are open (Godish, 1988). Little information is available on the effects of physical
parameters on nonformaldehyde emissions.
When wood is exposed to air in which the humidity is variable, it changes in moisture content
and shrinks or swells accordingly. Engineered wood materials made with UF are not water
resistant and consequently can absorb water. This absorbed water can act in two ways: (1) it can
react with free formaldehyde present in the wood from incomplete cross-linking of the UF resin;
and (2) it can cause slow hydrolysis of the UF resin, resulting in a slow release of formaldehyde.
2.5 Health Effects Related to Indoor Air Quality
A multitude of organic compounds with a high degree of variability and relative concentrations
found from one microenvironment to another characterize the indoor air environment. The effort
required to evaluate each possible indoor microenvironment and each possible indoor air
pollutant is prohibitive. As a result, TVOC has become a measurement of exposure and indoor
air quality complaints. Many studies have attempted to correlate indoor TVOC levels with
indoor air complaints, and to use TVOC as a measure for acceptable indoor air quality.
Formaldehyde is one of the most studied organic compounds regarding toxicity. It has been
shown to have a wide range of effects in both humans and experimental animal species. The
EPA has classified formaldehyde as a "Probable Human Carcinogen" (Group Bl) following its
guidelines for carcinogenic risk assessment (U.S. EPA, 1994). Available epidemiologic data
suggest limited evidence for carcinogenicity in humans; however, sufficient evidence of
genotoxicity and carcinogenicity in experimental animals supported these data. Formaldehyde
has also been associated with eye and nose irritation, changes in respiratory function, and
respiratory irritation. A thorough evaluation of the health and environmental effects was beyond
the scope of this project.
7
-------
3.0 ENGINEERED WOOD PRODUCT CLASSIFICATIONS AND INDUSTRY
STATISTICS
Engineered wood products arc distinct from solid wood in that they are composed of wooden
elements of varying sizes held together by an adhesive bond (Table 3-1). In general, the
manufacturing process involves some type of wood size reduction, followed by drying (except
for wet process boards), adhesive application, and pressing at elevated temperatures.
Table 3-1. Engineered Wood Products
Product Type
Wood Form after Size
Reduction
Primary
Adhesive(s)
Manufacturing
Process
Plywood panels
-Structural plywood
-Hardwood plywood
Engineered lumber
Reconstituted wood
panels
-Oriented strandboard
-Particleboard
-Medium-density
fiberboard
-Ccllulosic fiberboard
-Hardboard
Veneer
Veneer
Veneer and lumber
Wood strands of uniform
size
Finely ground wood
particles of various sizes
(fluffy, dust-like texture)
Wood fibers of uniform size
(fluffy, dust-like texture)
Wood fiber of uniform size
(fluffy, dust-like texture)
Wood fibers of uniform size
(fluffy, dust-like texture)
PF
UF
PF
PF, MDI
UF
UF
Starch or
asphalt
PF
Dry
Dry
Dry
Dry
Dry
Dry
Wet
Dry, wet, or wet/dry
PF=Pheno!-Forma1dehyde.
UF=Urea-Formaldehyde.
MDI-Methylenediphenyl diisocyanate.
8
-------
3.1 Plywood Panels
Engineered wood panels made with wood veneers are classified as plywood. Veneer is produced
by cutting or peeling thin sheets of wood from a log. In plywood manufacture, veneers are
bonded together with a synthetic adhesive resin to form a laminate. (A laminate is any object
built up of thin layers.) The number of veneers or plies in a panel varies by product. The outside
plies of a plywood panel are called the face and back. The center ply or plies are called cores.
The center plies are layered with the grain of each sheet perpendicular to the previous one.
Plywood may be made entirely of veneer, or the core materials may be lumber, PB, plastic,
metal, or other materials (Figure 3-1).
Plywood panels are classified into two groups according to end use: structural plywood and
hardwood plywood. Structural plywood panels are used primarily as a sheathing product in the
construction of residential homes and nonresidential buildings; other applications include
materials handling, concrete forming, and construction of transportation equipment, furniture,
and fixtures. Hardwood plywood is used for decorative applications such as wall paneling and
industrial applications such as furniture manufacture. The structural plywood industry is
significantly larger than the hardwood plywood industry (Figure 3-2).
3.1.1 Structural Plywood
Structural plywood panels are made primarily from softwoods. Softwoods arc coniferous or
needle-leaved trees (pine, fir, spruce, hemlock), as opposed to hardwoods, which are deciduous
or broad-leaved trees (oak, ash, maple, walnut). The term "softwood" has only a general
reference to actual wood hardness. Structural panels may use either variety, but are more
commonly manufactured of softwoods (APA, 1993).
Structural plywood panels are constructed entirely of veneers (typically 1/10 inch thick) that arc
bonded with PF adhesive resins. Phenol formaldehyde resins arc waterproof, which allows
structural panels to be used in exterior applications. Panels are typically manufactured into 4- by
8-foot sheets that may be sanded. No finishes (e.g., liquid coatings, paper coalings) are applied
to the panels.
3.1.1.1 Industry Outlook
Since the mid-1980s, timber harvests from publicly owned lands have declined by more than 50
percent (Carliner, 1994). Industry attributes the decline to new land management policies by the
Federal Government that have reduced the amount of land available for harvesting. Heavy
restrictions have been placed on national forests in the Pacific Northwest that contain twice as
much timber as all the other national forests combined (Evergreen Magazine, 1994).
9
-------
Various types of plywood
construction
All veneer
construction
3-fayer
3 ply
3-fayer
4 ply
Alternative constructions
Comply
S4ay»f
5 ply
lumber core plywood
Partlcleboard
3-laytr
Slayer
Tact*
Crotsband
Figure 3-1. Various types of plywood construction (Haygreen amd Bowyer, 1989).
Reprinted with permission.
10
-------
T3
c3
O
CQ
tt-H
o
-4—»
1)
vo
oo
ON
oo
oo
ON
O (N "3"
On On On
On ON ON
Figure 3-2. U.S. shipments of hardwood plywood and structural
plywood(USDC, 1988; APA, 1993).
Notes
1. Data unavailable on shipments of hardwood plywood after 1988
2. Hardwood plywood data are reported as square feet of surface measure, irrespective of panel width. The pre
mix of hardwood plywood panels ranges from 5/16 " to 3/8" in width (Groah, 1994); therefore, an average width o
11/32" was used to convert surface measure to cubic feet.
-------
Harvesting restrictions have negatively impacted the structural plywood industry. The shortage
of timber has increased log prices, resulting in increased manufacturing costs for plywood
production. Rising lumber prices throughout the 1990s have also resulted in higher log costs.
Lumber and plywood operations now compete for the same log. While plywood prices are held
in check by substitution of oriented strandboard (see Section 3.3.2), lumber prices are more free
to rise because there are fewer substitutes for lumber. The replacement of structural lumber is
only in the beginning stages with products such as laminated veneer lumber (LVL) and glulam
beams (see Section 3.2). Plywood is therefore less able to compete with lumber as a common
log resource (Roberts, 1994).
The combination of log shortages, rising manufacturing costs, and competition from OSB have
resulted in plywood production curtailments and mill closures, particularly in the West. There
have been 112 plywood and veneer plant closures in Washington, Oregon, and California
between 1980 and the end of 1993 (Wood Technology, 1994). Shipments of structural plywood
from the West have decreased 53 percent since 1987 (Figure 3-3). Total U.S. shipments of
structural plywood have decreased 15 percent since 1987.
3.1.2 Hardwood Plywood
Hardwood plywood is categorized into two types of products: prefinished wall paneling and
industrial hardwood plywood. Prefinished wall paneling is bonded with UF adhesive resins,
which are nonwaterproof. Industrial hardwood plywood is bonded with either UF or PF adhesive
resins, depending on the end use application (i.e., PF adhesive resins are used where the bond
must be waterproof).
3.1.2.1 Wall Paneling
Hardwood plywood wall panels are primarily 3-plv and 1/9 inch to % inch thick. All wall panels
are prefinished. There are two types of prefinished wall paneling: (1) natural finished wall
paneling and (2) decorative finished wall paneling. Plywood used for natural finished wall
paneling is constructed in the United States from species such as oak, birch, walnut, elm, cherry,
and pecan and is finished to retain its natural look. Plywood panels used for decorative finished
wall paneling are imported from Indonesia. The imported panels arc unfinished and are
decorated (e.g., painted, laminated, etc.) in the United States.
12
-------
o 400 -
IS
3
u
c 300 -
200
100
0
~—\ f
1 1-
1
3 North & South*
~ West
¦ Inland
—Total
1987 1988 1989 1990 1991 1992 1993
Figure 3-3. U.S. shipments of structural plywood 1987 to 1993 (APA, 1993).
* Shipments from the North are very small and were combined with shipments from the South to avoid
disclosure
-------
Industry Outlook. There has been a substantial decline in the use of prefinished wall paneling
since the late 1970s. The total U.S. market for prefinished wall paneling has declined from
around 110 million cubic feet in 1978 to 35 million cubic feet in 1992 (Figure 3-4). According
to the Hardwood Plywood and Veneer Association (HPVA), the loss of market share by
hardwood plywood wall paneling is due to the following factors:
• Customer preference for products such as gypsum wallboard that can be repainted
(in different colors, if desired) so that the appearance of a room can be more easily
changed from time to time (HPVA, 1991).
• The establishment of an Indonesian cartel, which has resulted in significant
increases in the price of imported hardwood plywood blanks during the late 1970s
and 1980s. Currently, Japan and China are the largest markets for Indonesian
plywood, with the United States a distant third (IIPVA, 1991).
• A decline in promotion and advertising by some of the major manufacturers
accompanied, in some cases, by their withdrawal from the hardwood plywood
paneling market (HPVA, 1991).
• Formaldehyde release from wall paneling, either real or perceived (HPVA, 1991).
3.1.2.2 Industrial Hardwood Plywood
Industrial hardwood plywood panels are commonly made using 3, 5, and 7 plies. The panels
vary in thicknesses; V2 inch and 3/4 inch thick panels are common. Industrial hardwood plywood
panels are unfinished, i.e., coatings, laminates, etc., are not applied (HPVA, 1991). The
unfinished panels are used in the manufacture of furniture, cabinets, and specialty panels. Some
hardwood plywood industrial panels are made with a PB core (15 percent of the market) or an
MDF core (15 percent of the market). Veneer core (65 percent of the market) is the predominant
type used in industrial panels (HPVA, 1991). Unlike wall panels, production of industrial
hardwood plywood panels has remained fairly constant over the years (Figure 3-5).
3.2 Engineered Lumber
As mentioned in Section 3.1.1.1, glulams beams and LVL are emerging as substitutes to lumber.
Glulam is short for glued-laminated structural timber ~ large beams fabricated by bonding layers
of specially selected lumber with PF adhesives. End and edge jointing permit production of
longer and wider structural wood members than are naturally available. Glulam timbers are used
with structural wood panels for many types of heavy timber construction.
Laminated veneer lumber is constructed of veneers that are bonded together with a PF adhesive
resin to form a laminate; the veneers are layered with the wood grain along the long axis of the
beam. The thickness of the veneers varies from 1-3/8 inches to 1-1/2 inches Laminated veneer
14
-------
120 -
T3
03
O
X>
<4-1
o
-4-»
0)
<+-1
o
3
3
o
C
o
100 - „
80 -
60
40 ¦
20 4-
I Natural finished wall paneling
] Decorative finished wall paneling
¦ Total wall paneling
1
1
oo
ON
r-
r-
ON
ON
II i i
ci
oo oo oo
ON ON ON
Figure 3-4. U.S. production of hardwood plywood wall paneling (Groah, 199
Notes:
1. Shipments of hardwood plywood are reported as sq. ft of surface measure, irrespective of panel thickness. The presen
hardwood plywood panels ranges in thickness from 0.313 in to 0.375 in (Groah, 1994). An average thickness of 0.344 in
to convert surface measure to cubic feet.
-------
Os
t3
c3
O
S3
o
-(-J
u
,IU
o
is
3
o
£3
O
30 -
25 -
20
15 +
10 1
5 -
oo
r-
o\
CTs
r-~-
OS
O
OO
OS
•00
OS
-------
lumber is manufactured to typical lumber sizes (2 by 4, 2 by 6, etc.). The beams can be
manufactured up to 30 feet long using end joints or finger joints. Another application of LVL is
in the construction of wood I-joists. (An I-joist is a small beam that resembles the letter "1".)
Laminated veneer lumber is used to construct the top and bottom of the joist and OSB is used to
construct the center.
3.2.1 Industry Outlook
As seen in Table 3-2, the production of glulam beams and LVL is increasing rapidly and is
expected to continue. By 2003, the North American output of LVL is expected to reach 98
million cubic feet (Blackman, 1994).
Table 3-2. U.S. Production Estimates of Engineered Lumber"
1992
1993
1994
1995
1996
1997
Glulam beams (million board feet)
271
288
308
352
392
391
I-joists (million linear feet)
225
350
385
440
495
450
Laminated veneer lumber (million cubic feet)
18
25
29
33
39
37
Source: APA, 1993.
•"Production figures for 1993-1997 arc estimated.
3.3 Reconstituted Wood Panels
Engineered wood panels made with wood strands, particles, and fibers are classified as
reconstituted wood panels. Reconstituted wood panels include PB, OSB, cellulosic fiberboard,
MDF, and hardboard. Table 3-3 lists shipments of reconstituted wood panels by U.S.
manufacturers on a volume basis (cubic feet of board) from 1985 to 1993. Particleboard, OSB,
and MDF represented 84 percent of reconstituted panel shipments by U.S. manufacturers in
1993.
3.3.1 Particleboard
Particleboard is a panel product made from finely ground wood particles of various sizes that are
bonded together with a synthetic adhesive resin under heat and pressure in a hot press. Other
materials may also be added during manufacturing to improve certain properties such as fire
resistance and dimensional stability. The wood particles come primarily from planer shavings,
sawdust, plywood trimmings, and other process residuals. Most PB in the United States is
bonded with UF adhesive resins. Particleboard panels bonded with UF adhesive resins arc used
for interior applications that do not require water resistant materials.
17
-------
Table 3-3. U.S. Shipments of Reconstituted Wood Panel (million ft3) 1985 to 1993
PB
OSB
MDF
Hardboard
Cellulosic Fiberboard
1983
188
42
38
NA
NA
1984
200
64
40
NA
NA
1985
208
83
43
66
NA
1986
225
110
49
61
NA
1987
232
127
56
57
49
1988
239
144
59
53
49
1989
241
160
61
54
50
1990
238
169
59
52
51
1991
236
175
60
51
47
1992
249
208
67
55
54
1993
265
219
101
55
54
Source: NPA, 1994b; AFA, 1994.
NA = Not available
Particleboard panels arc manufactured in a variety of sizes and densities, depending on the
specific end use application. Panel sizes range from 3 to 9 feet in width, '/i inch to 2 inches in
thickness, and almost any length that is transportable. Particleboard panels are used for industrial
applications, floor underlay, mobile home decking, door core, shelving, and stair treads. The
most common use of PB panels is for industrial applications. Industrial-grade panels represented
80 percent of PB shipments by U.S. manufacturers in 1993. In comparison, PB used for floor
underlay made up 7 percent of the 1993 shipments, mobile home decking (4 percent), door core
(3 percent), shelving (2 percent), and stair treads (0.3 percent). A large volume of industrial
grade PB is used as a core stock material for furniture (domestic, institutional, office), kitchen
and vanity cabinets (sides, backs, drawers, doors), doors (solid core flush doors, bifolds, sliding
doors), games (table tennis, pool tables), and other goods. Industrial-grade PB is also used as a
substrate for laminated panel construction such as countertops, desktops, wall paneling, and
shelving. High-pressure laminates; thermofused, resin-saturated papers; vinyl films; hot transfer
films: decorative papers; and wood veneers are the types of overlay materials most commonly
applied to PB substrates (NPA, 1994a).
3.3.1.1 Industry Outlook
Shipments of industrial PB by U.S. manufacturers have been increasing steadily since the early
1980s and in 1993 were at record levels (Figure 3-6). Statistics for PB panels manufactured in
the United States are compiled by the NPA. The NPA also compiles statistics lor MDF panels
manufactured in the United States. Statistics are compiled from member companies who
18
-------
300 -
2 200
a 100
•Total
• Industrial
¦ Underlayment
¦ Manufactured Home Decking
1995
Figure 3-6. U.S. industry shipments of particleboard 1978 to 1993 (NPA 1994b)
-------
represent more than 85 percent of the total U.S. manufacturing capacity for U.S. PB and MDF.
Most of the PB manufactured by NPA members is unfinished, i.e., there are no coatings or
laminates applied to the board at the plant; of the 265 million cubic feet of PB shipped in 1993, 3
percent was laminated and 1 percent was coated (NPA, 1994b). Particleboard is laminated and
finished by end users such as furniture and cabinet manufacturers as well as producers in the
laminating industry who sell finished PB to furniture and cabinet manufacturers.
3.3.2 Oriented Strandboard
Oriented strandboard is made from strands of wood bonded together with a waterproof adhesive
under heat and pressure in a hot press. The strands of wood are sliced from small diameter, fast
growing trees; tree species used are generally aspen in the northern U.S. and pine and soft
hardwoods in the southern United States. Strand dimensions vary depending on the slicing
machinery and wood species. Typical strand dimensions are 3 inches long, 3/16 to 2 inches
wide, and 20/1000 to 28/1000 inches thick (about the thickness of a business card). The strands
of wood in the surface layers are aligned in the long panel direction, while the inner layers are
random or cross aligned. The OSB process was developed in the early 1980s as an improvement
to the waferboard process in which strands of wood were randomly distributed throughout the
panel. Two-thirds of the U.S. production of OSB is made with PF adhesives, and the remaining
third is made with MDI adhesives.
Oriented strandboard is used in structural applications as a replacement for sheathing-grade
plywood. The most common applications are for wall and roof sheathing and floor decking in
wood frame construction. Other applications include materials handling (crates and pallets), web
material for wood I-beams, siding, and other specially products. The strength of OSB comes
from the uninterrupted wood fiber, interleaving of the long strands, and degree of orientation of
the strands in the surface layers. The waterproof adhesives combined with the strands provide
internal strength, rigidity, and moisture resistance.
3.3.2.1 Industry Outlook
As illustrated in Figure 3-7, OSB has been rapidly increasing its share of the U.S. structural panel
market. By late 1996, the capacity of OSB is expected to increase by 45 percent in the U.S. and
by 130 percent in Canada; total North American capacity is expected to rise by 70 percent
(Roberts, 1994). A key factor in the growing OSB demand is the continuing decline of structural
plywood production. The substitution of OSB for structural plywood has been relatively easy
since it competes on price at equal or near equal performance. In the future, the falling price of
OSB due to new capacity will also play a key role in encouraging its substitution.
3.3.3 Ilardboard
IIardboard is made with wood fibers bonded with a synthetic adhesive resin under heat and.
pressure with a waterproof adhesive. Phenol-formaldehyde adhesives are used to bind hardboard
20
-------
Structural Plywood
Oriented Strandboard
700
-3 600
o
X 500
C
a
400 -
3
c 300 —
e
200 -
100
oc
G\
o
t
(N
cn
in
vO
E^-
OC
ON
O
y—t
r-
oo
OO
OO
oo
oo
oo
oo
OO
oc
OO
as
ON
o
a\
ON
ON
OS
ON
ON
ON
O
ON
a*
OS
ct\
ON
T-—1
r—i
r—i
r—«
T—<
T-H
*—H
r—-<
T—<
r—i
«—i•
»«"H
f—¦*
•p—<
Figure 3-7. U.S. structural panel market (APA, 1993).
-------
in the United States. Hardboard is manufactured by a wet, dry, or wet per dry process (see
Section 4.2.5.1). Raw materials used to generate wood fibers in the dry hardboard process are
primarily dry planer shavings from lumber operations. Green chips are used to generate wood
fibers for wet and wet/dry hardboard processes (green means that the wood has never been
dried). Most green chips are chipped from lumber slabs and forest harvesting residues, such as
branches and tops. (In the lumber process, logs are peeled and then cut to length with squared
sides; lumber slabs are the part of the tree that remains after the lumber has been cut.) A small
amount of green chips are chipped from roundwood
Hardboard has a density ranging from 40 to 70 lb/fit3 and is categorized into three product groups:
basic hardboard, hardboard siding, and pretlnished wall paneling. Basic hardboard and siding
make up the largest volume of hardboard products. Basic hardboard is used in a wide variety of
applications including floor underlayment, furniture, case goods, truck and head liners, and door
skins and faces; door skins and faces are the fastest growing market for basic hardboard
(Wagner, 1994).
3.3.4 Medium-Density Fiberboard
Medium-density fiberboard is made with wood libers bonded together with a synthetic adhesive
resin under heat and pressure in a hot press. Urea-formaldehyde adhesive resins are the primary
type of resin used to manufacture MDF in the United States. Raw materials used to generate the
fibers come from dry planer shavings, plywood trim, and sawdust. The density of MDF ranges
from 40 to 60 lb/ft3.
Medium-density fiberboard is used in the manufacture of furniture, cabinets, and general
millwork applications. It is increasingly being used as a substitute for kiln-dried dressed lumber
in applications such as window frames, door jambs, and decorative moldings. While nearly all
MDF is made with non-water-resistant UF adhesives aimed at interior applications, there is
continuing development and interest in exterior-grade panels and overlay systems that allow
MDF panels to be used outside (Roberts, 1994). Due to the fine texture and homogeneous nature
of MDF, it machines clean and is easily painted to produce high-quality finishes. Because MDF
is made and sold as a panel, many furniture and cabinet components can be made from a single
piece. The same components made from natural wood require labor intensive jointing and
assembly operation.
3.3.4.1 Industry Outlook
The MDF industry is growing rapidly; U.S. shipments of MDF were at record levels in 1993
(Figure 3-8). The United States currently accounts for about 90 percent of North America's
MDF capacity (Roberts, 1994). By late 1996, U.S. capacity is expected to increase by 30
percent, while Canadian capacity is expected to increase to over five times its current level; total
North American capacity will rise by 75 percent (Roberts, 1994).
22
-------
120 -
100
¦4-J
CD
<4-*
0
T
1978 1980 1982 1984 1986 1988 1990 1992 1994
Figure 3-8. U.S. shipments of medium density fiberboard (NPA, 1994b).
-------
3.3.5 Cellulosic Fiberboard
Cellulosic fiberboard is composed of wood fibers bonded together with either starch or asphalt.
The raw materials used to generate the wood fibers are primarily green chips. A few plants in the
U.S. manufacture cellulosic fiberboard from bagasse and newsprint. Cellulosic fiberboard is a
low-density board ranging from 10 to 30 lb/ft3. Ninety percent of the 1993 shipments of
cellulosic fiberboard were comprised of exterior wall sheathing, roofing, or other specialized
exterior panels; the remaining 10 percent were used in a variety of industrial applications
(Wagner, 1994).
4.0 ENGINEERED WOOD MANUFACTURING PROCESS DESCRIPTIONS
4.1 Plywood Manufacture"
The following sections describe the manufacturing processes involved in making plywood. The
major steps include the following: debarking, heating the blocks, cutting veneer, drying veneer,
lay-up and pressing, and finishing.
4.1.1 Debarking
The most widely used type of debarker in medium to large mills is the ring debarker. In this
machine the log passes through a rotating ring that holds a number of pressure bars. These
press against the log and tear off the bark. Large units of this type can debark logs at speeds
of up to 200 lineal ft/min.
A rossing head type of debarker is sometimes used in mills where high production rates are
not needed. The rossing head is a rotating cutterhead, similar to the head on a lumber planer,
that rides along the log and cuts off the bark as the log is rotated. This debarker is also suited
to situations where crooked or stubby logs must be debarked.
4.1.2 Heating the Blocks
When logs are cut to the length required for rotary veneer cutting, they are called blocks.
Almost all hardwood and many softwood blocks are heated prior to cutting the veneer.
Heating softens the wood and knots, making it easier to cut. It also improves surface quality,
reducing roughness. Some of the dense hardwoods must be heated to produce satisfactory
veneer. Softwood veneer from some species can be produced from cold logs because of their
lower density and because the roughness limitations are less critical than for high-quality
hardwoods. However, even in the softwood industry the advantages of heating the blocks
generally outweigh the cost of the process. Baldwin (1975) lists four advantages of heating
softwood logs:
** Sections 4.1.1 through 4.1.6 were reprinted with permission from Forest Products and Wood
Science: An Introduction by J.A. Haygreen and J.L. Bowyer, Second Edition, Iowa State University
Press, 1989.
24
-------
1. Higher yields of veneer can be obtained from the logs. The reduction of cutting imperfections
increases the yield an average of 3 - 5 percent.
2. The grade of the veneer is improved. Studies by Lutz (1960), Grantham and
Atherton (1959), and the American Plywood Association (sic) have found that grades
of veneer are upgraded from 4 to 25 percent.
3. Labor costs are reduced. Veneer from heated peeler blocks tends to hang together in
a more continuous ribbon as it comes from the lathe. This reduces handling.
4. The amount of adhesive used can be reduced. Glue spreads can be lighter because of
the improved surface.
A variety of methods are used to heat logs. Steaming, soaking in hot water, spraying with hot
water, or combinations of these methods are all suitable to some situation. Dense hardwoods are
usually heated by soaking at temperatures up to 200 °F (93 °C). Regardless of the heating medium,
the objective is to heat the log to a suitable temperature as deeply into the log as veneer will be cut.
4.1.3 Cutting Veneer
Two major methods for producing veneer arc slicing and peeling. Most veneer is produced by
peeling (rotary cutting), which is accomplished on a veneer lathe. Slicing is used for producing
decorative veneers from high-quality hardwood and is seldom used with softwoods. The cutting
action on a lathe and on a slicer are very similar and are illustrated in Figure 4-1. In either case,
the wood is forced under a pressure bar diat slightly compresses the wood as it hits the cutting
edge of the knife.
On the veneer slicer a cant of wood called the flitch is rigidly dogged, i.e., clamped, to a carriage
that oscillates, cutting on the down stroke. Before each cutting stroke the knife and pressure bar
move forward the thickness of the veneer to be cut. In a rotary lathe they move forward
continuously as the block rotates.
4.1.4 Veneer Storage and Clipping
In modern mills the green veneer must be handled gently and rapidly as it comes from the lathe.
The veneer is peeled at from 300 to 800 lineal ft'min. A series of trays is used in many softwood
plywood plants to handle these long ribbons of wood. Trays are often about 120 ft long, long
enough to handle the veneer that comes from a typical 15-in. block.
Clippers are high-speed knives that chop the veneer ribbons to usable widths. In hardwood veneer
mills, clipping may be done manually to obtain the maximum amount of clear material from the
flitch. In softwood mills and in some hardwood mills, clipping is often done automatically at
speeds of up to 1500 lineal ft'min. The clipper will cut the veneer to about 54 in. (the panel width
plus an allowance for shrinkage and panel trimming) if possible. However, if open defects are
present, the veneer may be clipped to less than full panel width. Automatic clippers detect open
defects with scanners that can be overridden by the operator when it is desirable to do so.
25
-------
Two means of cutting
veneer
Nosebar
SUCER
Veneer
Roffer bar
LATHE
Chuck
Veneer block
Vertical
pressure
compressed
Roller
Veneer
Knife
rotation
A - loose side
B - tight side
Horizontal pressure
Veneer
Figure 4-1. Cutting action on a lathe and slieer (Haygreen and Bowyer, 1989).
Reprinted with permission.
26
-------
4.1.5 V cnccr Drying
Veneer driers consist of a means of conveying the veneer through a heated chamber where
temperatures range from 150 to 260°C. In older roller driers, air is circulated in a manner
similar to that in a dry kiln. This type of drier is still in wide use for hardwood veneer. Most
plants built in recent years utilize jet driers. These are also called impingement driers since a
curtain of air at velocities of 2000-4000 fpm is directed against the surface of veneer. The
high velocity produces turbulent air on the surface of the veneer. This eliminates the laminar
boundaiy that slows down heat and moisture transfer under ordinary drying conditions.
Developments such as the use of microwave energy, use of high temperature preheaters, and
increased drying temperature (up to 800 °F, or 427 °C) may find use for some types of veneer
drying. In most softwood veneers, however, temperatures over about 400 °F (204 °C) have
adverse effects on glueability.
4.1.6 Lay-up and Pressing
Almost all adhesives used in the plywood industry in the United States today are
thermosetting (cured by heat) synthetic resins. These have almost completely replaced the
blood and soybean tlour protein glues that were used in the past for interior (nonwatcrproof)
grades of plywood. The two most important types of resins used are phenol-formaldehyde,
which is used for interior and exterior grades of softwood plywood and for exterior grades of
hardwood plywood, and urea-formaldehyde, which is used to manufacture interior grades of
hardwood plywood. The basic components of these resins are formaldehyde, which is
derived from methanol, urea, and phenol.
Only rarely are pure or "neat" resins used as adhesives for plywood. In most cases they are
mixed with fillers or extenders such as Furafil and fine flour produced from wood, bark, or
nutshells. Furafil is a chemical lignocellulose by-product of furfiiryl alcohol production that
can be produced from corn cobs, rice hulls, and oal hulls. Starch and animal blood are also
used as extenders to modify the viscosity, control the penetration into the wood, and control
other characteristics of the adhesive mix such as the tack (stickiness).
The process of applying adhesives to the veneers, assembling veneers into a panel, and
moving the panels in and out of the press are often the most labor-intensive steps in
manufacture. Veneer is highly variable in width, length, and quality, which makes it a
difficult material to handle with automated systems. Yet major advances have been made to
increase automation in this stage of manufacture.
One advancement has been in the application of adhesive to the veneer. The old method is to
pass veneer through rubber-faced grooved rollers that apply glue by contact to the top and
bottom surfaces. One person is required to feed the roller glue spreader while an offbearer
places the veneer onto the panel being laid up. If veneer is extremely rough, the glue spread
will not be uniform and skips may occur.
Newer means of glue application, spray and curtain-coaters, have distinct advantages in terms
of uniformity of the glue spread and are suited to automated lay-up systems. These methods
overcome the problem of poor glue spread on rough stock. In these systems the veneers travel
on a belt conveyor under the spray or curtain. A curtain-coater consists essentially of a box
27
-------
with a slot in the bottom through which the adhesive flows in a continuous sheet or curtain.
Glue not deposited on a piece of veneer passing through the curtain is pumped back up into
the box.
Recently, two new methods of adhesive application, liquid extrusion and foamed resin
extrusion, have been used successfully. These systems lay down continuous beads of resin
on the veneer.
The actual assembly of veneer into plywood panels can also be mechanized — at least in
larger plants producing standard-size panels. Although equipment has been developed to do
this almost automatically, most mills use systems that are partially manual and partially
mechanized. For example, the full-size 4 x 8-ft veneers for the two faces may be handled by
machine, but the narrower strips of veneer used in the core may be assembled manually.
Most softwood plywood plants prepress the loads oflaid-up panels prior to final pressing in
the hot-press. This is done in a cold press at lower pressure. The purpose is to allow the wet
adhesive to "tack" the veneer together. This permits easier loading of the hot-press and helps
prevent shifting of the veneers during loading.
Pressing of the panels is usually done in the multiopening presses. Such presses can produce
20 - 40 4 x 8-ft panels at each pressing cycle, which may take 2-7 minutes. The purpose of
the press is twofold: to bring the veneers into close contact so that the glue line is very thin
and to heat the resin to the temperature required for the glue to polymerize. Adhesives made
from phenol-formaldehyde resins typically require temperatures of240°F (115.5°C) in the
innermost glue line for approximately 90 seconds to cure properly. Resin systems must be
carefully tailored to the specific conditions in a plant. Press time and temperature can be
modified; i.e., a shorter press time may be possible if press temperature is increased.
4.1.7 Finishing
Stationary circular saws trim up to 1 inch from each side of the pressed plywood, producing
even-edged sheets. About 20 percent of annual plywood production is sanded. Those that are
sanded may be sanded on both sides or only one. Sheets move through enclosed automatic
sanders while pneumatic collectors above and below the plywood continuously remove the
sanderdust. Sawdust in trimming operations is also removed by pneumatic collectors. The
plywood trim and sawdust are sold to reconstituted panel plants.
4.2 Reconstituted Panel Manufacture
The following sections describe the major processes involved in the manufacture of reconstituted
wood panels. These processing steps include wood reduction, drying, adhesive blending, mat
forming, and pressing (Figure 4-2). The drying and pressing steps are not required by all
products.
28
-------
WOOD REDUCTION
QSB
Logs are debarked
and cut into strands
PB
Wood residues are
ground into fine particles
of varying sizes
MDF
BINDER APPLICATION
(for MDF)
DRYER
Fiberboards
Wood residues are digested in a
team cooker and mechanically
separated into fibers
Hardboard Hardboard Cellulosic
Dry Wet & Wet/DryFiberboard
WET
FORMING
(Binder Applied]
BINDER
APPLICATION
(for OSB and PB)
HardboardHardboard Cellulosic
Wet Wet/Dry Fiberboard
DR
FORA
Y
yiING
PRESSING
OVEN
Hardboard
Wet/Dry
PRESSING
Figure 4-2. Reconstituted wood panel process flow.
29
-------
4.2.1 Wood Reduction
4.2.1.1 Oriented Strandboard
The raw materials or "furnish" used to manufacture OSB are specially produced from green
roundwood at the plant. Logs entering OSB plants are cut to 100-inch lengths by a slasher saw.
The logs are debarked and carried to stationary slasher saws, where they are cut into 33-inch
lengths called blocks. The blocks are then sent to a waferizer, which slices them into strands
approximately 0.028 inches thick (Vaught, 1989). The strands are then conveyed to a storage bin
to await processing through the dryers.
4.2.1.2 Parfieleboard
The furnish used to manufacture PB can be either green or dry process wood residues. Green
process residues include planer shavings from surfacing green lumber, green sawdust, and
plywood trim such as veneer clippings, edgings, and trimmings. Dry process residues include
shavings from planing kiln-dried lumber, mill ends from kiln-dried lumber, sawdust, sanderdust
and plywood trim. The wood residues are ground into particles of varying sizes using
mechanical refiners and hammermills.
4.2.1.3 Fiberboard (Ccllulosic Fibcrboard, MDF, and Hardboard)
The furnish used to manufacture fiberboards consists of the same type of green or dry process
wood residues used to manufacture PB. Fibers are generated by first cooking the wood residues
in a moderate pressure steam vessel (digester). During this step the wood changes both
chemically and physically; becoming less susceptible to the influences of moisture and less
brittle as the lignin softens. This semiplastic wood is then "rubbed" apart into fiber bundles
instead of being mechanically "broken" apart as in the PB process. The fibers are all the same
size, therefore, they need no screening.
4.2.2 Drying
In the manufacture of cellulosic fiberboard, wet process hardboard, and wet/dry process
hardboard, the furnish is not dried because the forming process uses water (see Section 4.2.5).
In the manufacture of OSB, PB, MDF, and dry process hardboard, the furnish is dried to a very
low moisture content to allow for moisture gained by adding resins and other additives. The
furnish is not dried further after blending; however, some evaporation may occur from the heat in
the furnish as it comes from the dryer and from air exposure when conveyed from the blender to
the forming station. Furnishes are generally no wanner than 100°F when blended to avoid
precuring and drying out of the resin (Maloney, 1977).
Rotating drum dryers requiring one to three passes of the furnish are the most common type of
dryer used at reconstituted wood plants. The dryers are fired by wood wastes from the plants
30
-------
(such as bark from OSB plants and fines from PB plants) and occasionally by oil or natural gas.
Dryer inlet temperatures may be as high as 1,600 °F with a wet furnish. However, dry planer
shavings require that dryer inlet temperatures be no higher than 500 °F because the ignition point
of dry wood is 446 °F.
4.2.3 Screening and Air-Classifying
In PB manufacture, screening removes the lines (which would absorb too much resin if not
removed) from the dryer exhaust and classifies particles by size for face and core layer
(Ilaygreen and Bowyer, 1989). The PB industry commonly uses inclined or horizontal vibrating
screens or gyratory screens (Moslemi, 1974).
Air classifiers, which separate particles by particle surface area and weight, may be used alone or
in conjunction with screening equipment. Air classifiers perform best if the feed is limited to
particles with uniform widths and lengths. The classifier can then efficiently separate particles of
different thicknesses due to the weight difference among particles of approximately equal surface
area (Maloney, 1977). Undesired material is sent to a fuel preparation system for the dryer
burner. The screened particles are stored in dry bins until they are conveyed to the blender
(Vaught, 1989).
4.2.4 Blending
4.2.4.1 Particieboard and Oriented Strandboard
After drying, the furnish is blended with a synthetic resin, wax, and other additives added via
spray nozzles, simple tubes, or atomizers. Waxes are added to the furnish to retard the
absorption of water into the board. The wax, a mixture of petroleum hydrocarbons, is typically
added to the furnish as an emulsion (wax emulsions are dispersions of very small wax particles in
water). The dispersion is stabilized with various chemicals known as emulsifiers. The amount
of wax added to the furnish ranges from 0.25 up to 1 percent or more of the board weight. Resin
may be added to the furnish as received (usually an aqueous solution); mixed with water, wax
emulsion, catalyst, or other additives; or added as a spray-dried or finely divided powder
(Maloney, 1977). The amount of resin added to the furnish varies anywhere from 2 to 14
percent, depending on the specific end use of the product.
Particles are blended in short retention time blenders in which the furnish passes through in
seconds. The blenders consist of a small horizontal drum with high-speed, high-shear impellers
and tangential glue injection tubes. As the wood furnish enters the drum, the impellers hurl the
furnish at high speeds, which effectively combines it with the resin.
Strands are blended using long retention time blenders that are very large (several feet in
diameter and many feet in length). The furnish takes several minutes to pass through these
blenders. The blenders are large rotating barrels that are lilted on Iheir axes. As the strands are
31
-------
fed into the drums, they are sprayed with resin. The tumbling action of the strands through the
drums serves to blend the strands with the resin.
4.2.4.2 Medium-Density Fiberboard and Dry Process Hardboard
After refining, the fibers are discharged through a blow valve into a blowline. In the blowline,
the fibers are sprayed with a resin. The resin can be injected immediately after the blow valve or
anywhere along the blowline (Figure 4-3).
Figure 4-3. Blending system diagram (Frashour, 1990). Reprinted with permission.
4.2.5 Mat Forming
Mat forming is the spreading of the furnish particles into a uniform mat. Mat formation may be a
wet, dry, or wet/dry process.
4.2.5.1 Cellulosic Fiberboard, Wet and Wet/Dry Process Hardboard
Cellulosic fiberboard and wet and wet/dry process hardboard are formed by a wet process. In
the wet forming process, fibers are mixed with water and adhesive and this water-fiber mixture is
then metered onto a wire screen. Water is drained away with the aid of suction applied to the
underside of the wire, and the fiber mat along with the supporting wire is moved to a prepress
where excess water is squeezed out.
J
HCTINCR
rVRNACC
BLOW VALVE (fr?)
32
-------
4.2.5.2 Particleboard, Oriented Strandboard, Medium-Density Fiberboard, and Dry
Process Hardboard
Particleboard, OSB, MDF, and dry process hardboard are formed by a dry process. The dry
forming process uses air to distribute the furnish onto a moving caul (tray), belt, or screen
(Figure 4-4). Particleboard mats are often formed of layers of different sized particles, with the
larger particles in the core, and the finer particles on the outside of the board (Figure 4-5). In PB
and fiberboard manufacture, the particles and fibers are distributed in a random orientation.
Oriented strandboard is produced by deliberate mechanical or electrostatic orientation of the
strands. In mechanical orientation processes, mats are produced by dropping long slender flakes
between parallel plates or disks. In electrostatic orientation, particles align with an electrical
field when dropped between charged plates (Haygreen and Bowyer, 1989).
4.2.6 Hot Pressing
All reconstituted wood panels, except for cellulosic fiberboard, are hot pressed to increase their
density and to cure the resin. For cellulosic fiberboard, the mat is simply brought to desired
thickness using a press roll and then dried in an oven. Wet/dry process hardboard is also dried in
an oven before being hot pressed. Most plants use multiopening platen presses (Figure 4-6).
Typical multiple-opening presses have 14 to 18 openings (Maloney, 1977).
The last 10 years have seen the introduction of the continuous press (Figure 4-7). Though more
popular in Europe, the continuous press is currently being used in two PB and two MDF plants in
the United States. The forerunner of the continuous press was the single opening platen press
which was very popular worldwide for many years, but gradually has been phased out in the
United States.
The hot pressing time needed to compress the mat to final board thickness once the press platens
make contact with the mat surfaces is a function of the mat thickness, platen temperature, mat
moisture content, and the type of resin and additives used. Phenol-formaldehyde resins need a
longer curing time than do UF resins (Maloney, 1977).
4.2.7 Finishing
Primary finishing steps include cooling or hot stacking, grading, trimming/cutting, sanding, and
shipping. Cooling is important for IJF-resin-cured boards since the resin degrades at high
temperatures after curing. Boards bonded using PF resins may be hot-stacked to give additional
curing time (Maloney, 1977). Secondary finishing steps include filling, painting, laminating, and
edge finishing (Maloney, 1977). However, the vast majority of reconstituted panel
manufacturers do not apply secondary finishes to their panels. Panels are primarily finished by
end users such as cabinet and furniture manufacturers. Panels are also finished by laminators
who sell finished panels to furniture and cabinet manufacturers.
33
-------
Two types
of formers
Fan Air flow
Furnlth .
.«!c. .... v. •<¦
^¦(ffiFan
-$8®S
,' J*}/"
'
Partlcfa mate'*
:. sM. .
M¦>*-• •-,T.-S4 •¦>•*'•
* Bait
8>
Rollir dlicharga head
Caatlng rollar
¦SMft
(V) . Moving bait
B. Savaral forming haada can bt utad In sariaa
to product a 3« or s-layar mat
Figure 4-4. Two types of mat forming machines (Haygreen and Bowyer, 1989).
Reprinted with permission.
34
-------
A. SINGLE LAYER
B. 3-LAYER
C. 5-LAYER
&
1$X£±S£l
•••*»•~ «»•~•«•• • *£?••• ••••/••••»«•*,«»V• '.•/•!!•~~•»*•'.•#••• *«V•/•!*•!
0. GRADUATED
Figure 4-5. Various types of mat construction (Moslcmi, 1974). Reprinted with permission.
35
-------
simultaneous
CLOSING
OE VICE
COLUMNS
TOP PLATEN
OR CROWN
platens
OR PLATES
MOVING
table or
PLATEN
RAMS
Figure 4-6. Schematic of multiopening board press (Suchsland and Woodson, 1987).
36
-------
Figure 4-7. Schematic of continuous press (Wolff, 1996). Reprinted with permission.
-------
5.0 EMISSIONS DURING MANUFACTURE OF ENGINEERED WOOD PRODUCTS
Table 5-1 summaries the 1991 TRI forms filed by plywood plants. Section 313 of Emergency
Planning and Community Right-to-Know Act, (EPCRA) (Title III of the Superfund Amendments
and Reauthorization Act [SARA] of 1986) requires certain facilities manufacturing, processing,
or otherwise using listed toxic chemicals to report their environmental releases of such chemicals
annually. Reported air releases were much greater than releases to water or land.
Table 5-1. Reported Releases of SARA Section 313 Chemicals from Plywood Plants for 1991
Total Releases (103 lb)
Total Transfers (10' lb)
Compound
Air
Water
Land
POTW
Other Off-site
Acetone
378
0
0
0
5
Ammonium sulfate solution
0
0
0
0
0
Barium compounds
<1
0
4
0
0
Dichloromethane
0
0
0
0
2
Diphenylmethane diisocyanate
10
0
0
0
0
Formaldehyde
86
0
2
<1
0
Glycol ethers
24
0
0
0
0
Methanol
362
0
0
0
<1
Methyl ethyl ketone
79
0
0
0
3
Methyl isobutyl ketone
14
0
0
0
16
Phenol
34
0
4
<1
0
Sulfuric acid
<1
0
0
0
0
Toluene
101
0
0
0
5
Xylenes
76
0
0
0
7
Source: NCASI, 1993.
SARA -- Superfund Amendments and Reauthorziation Act
Table 5-2 is a summary of 1991 toxics release inventory (TRI) forms filed by reconstituted wood
panel plants. Air releases were much greater than releases to water or land. Formaldehyde
releases to the air exceeded the releases of all other compounds combined. Since these
compounds are emitted during product manufacturing and may be a product or byproduct of the
manufacturing process, they can be emitted in the indoor environment.
38
-------
Tabic 5-2. Reported Releases of SARA Section 313 Chemicals from Reconstituted Wood
Panel Plants for 1991
Total Releases (103 lb) Total Transfers (103 lb)
Compound
Air
Water
I ,and
POTW
Other Off-Site
Acetone
284
0
0
0
3
Ammonia
78
99
50
63
<1
Ammonium nitrate solution
<1
0
0
0
0
Ammonium sulfate solution
0
<1
1
0
5
Asbestos
0
0
17
0
0
Barium compounds
<1
0
J
0
0
Chlorine
0
<1
0
0
0
Diphenylmethane diisoeyanate
4
0
<1
0
0
Ethyl benzene
57
0
0
0
15
Ethyl glycol
27
0
0
0
0
Formaldehyde
3506
<1
51
<1
4
Glycol ethers
40
0
0
0
11
Lead
0
0
<1
0
0
Manganese compounds
0
0
0
48
0
Methanol
622
0
0
0
0
Methyl ethyl ketone
369
0
0
0
39
Methyl isobutyl ketone
117
0
0
0
28
n-Butyl alcohol
110
0
0
0
2
Nitric acid
<1
0
0
0
0
Phenol
184
<1
o
0
0
Polychlorinated biphcnyls
0
0
0
0
22
Styrene
0
0
<1
0
2
Sulfuric acid
29
0
0
0
0
Tetrachloroethylene
<1
0
0
0
0
Toluene
799
0
0
0
22
1,1,1 -Trichloroethane
0
0
0
0
2
1,2,4-Trimelhylbcnzene
13
0
0
0
1
Xylenes
431
0
0
0
76
Zinc compounds
0
0
<1
0
0
Source: NCASI, 1993.
SARA - Superfund Amendments and Reauthorization Act.
39
-------
6.0 SOURCE MANAGEMENT/POLLUTION REDUCTION: APPROACHES AND
ISSUES
Some general management techniques for providing good indoor air quality include source
management, pollution prevention, source removal/substitution, source modification, increased
local ventilation, air cleaning, and general ventilation. Traditional approaches for managing
indoor air pollution have relied on the use of ventilation system operating and design changes.
These approaches can result in improved IAQ. However, they can also result in substantial
increases in energy consumption (for increased ventilation) and transfer of the pollutants to
another medium and are inconsistent with the environmental protection hierarchy that was
established by the Pollution Prevention Act of 1990.
Pollutants are managed most effectively at their source. The major approaches are chemical
substitution, process changes, and product redesign. Strategies vary depending on whether
emissions are the result of offgassing from construction or finished materials (e.g., organics in
wood products) or product operation or use. When offgassing is the major emissions source,
substituting low-emitting materials may be an alternative. When product use or operation is the
major source of emissions, knowledge of the relevant process is essential in developing
emissions controls or redesigning the product to minimize emissions. In either case, chemical
substitution may be warranted if a product or equipment is found to emit chemicals that have
been shown to cause serious human health or environmental effects.
6.1 Pollution Prevention
6.1.1 Alternatives to Wood Feedstock
As discussed in Section 3.1.1.1, harvesting restrictions from publicly owned lands have curtailed
production of plywood and lumber, particularly in the West. Production of PB and MDF has
also been affected by the harvesting restrictions. Over 80 percent of the wood used to
manufacture PB/MDF are residuals from lumber or plywood manufacturing such as chips and
shavings (McCredic, 1993). Curtailment of the production of plywood and lumber affects the
supply of chips and shavings available for the production of PB/MDF. Western board
manufacturers have been especially hard hit by the drop in timber supply. Oregon alone has 11
PB/MDF plants that account for more than 25 percent of the U.S. production of these products.
The shutdown of plywood and lumber mills has put a premium on the remaining supply of wood
residuals (McCredie, 1993). PB/MDF manufacturers that were historically able to satisfy their
wood supply requirements within a 100-mile radius of the plant now routinely travel 200 to 300
miles (McCredie, 1993). This results in increased competition for available supples and
increased transportation costs (McCredie, 1993). While no Oregon PB/MDF mills have been
permanently shutdown for lack of wood, some production curtailments have occurred
(McCredie, 1993).
40
-------
One potential alternative fiber source for engineered panels is agricultural fibers. Large volumes
of agricultural fiber arc generated each year in the United States. Sources of this fiber include
bagasse, cereal straw, corn stalks and cobs, cotton stalks, kenaf, rice husks, rice straw, and
sunflower hulls and stalks (Youngquist et al., 1993). The volume of agricultural fiber generated
in the United States is so great that if 75 reconstituted wood panel plants were to switch entirely
to agricultural fiber, and, on average, each plant required 135 million tons of fiber annually, more
than 30 times as much agricultural fiber would be available as would be consumed (Youngquist
et al., 1993). Recently, a plant was built in North Dakota which manufactures particleboard from
wheat straw and MDI resins (Galbraith, 1995).
Technically, agricultural fiber may be substituted for wood fiber in the manufacture of
engineered products. Whether this is pollution prevention, however, requires a comparison of
indoor air emissions and manufacturing emissions from composite agricultural products and
engineered wood products. Unfortunately, emissions data from composite agricultural products
were not found in the literature to make such a comparison. Consequently, substitution of
agricultural fiber for wood fiber is presented in this report only as a "potential" pollution
prevention option.
6.1.2 Alternative Resins
Several resin and additive approaches are used by engineered wood manufacturers to reduce
formaldehyde board emissions from UF bonded products such as PB and MDF (manufacturers
are concerned primarily with formaldehyde because of the IIUD standard). Included are low
molar ratio UF resins, the use of formaldehyde scavengers with UF resins, melamine-fortified
UF resins, UF resins, and MDI resins. While these alternative resins and additives result in
products with significantly lower formaldehyde emissions, only low molar ratio UF resins and
formaldehyde scavengers have made a significant penetration in the PB and MDF industry.
6.1.2.1 Low Molar Ratio Urea-Formaldehyde Resins
The formaldehyde-to-urea (F:U) molar ratio is the ratio of the number of moles of formaldehyde
to number of moles of urea in UF resin adhesives. For example, a F:U molar ratio of 1.15 has
1.15 moles of formaldehyde for each mole of urea. A wide range of molar ratios is used in UF-
bonded products: for PB, when a single resin is used throughout the board, the F:U molar ratio
can fall anywhere within the range set by the lace/core systems (see Figure 6-1); MDF products
use resins with F:U molar ratios higher than PB resins; hardwood ply wood products use the
highest F:U molar ratios. The nature of the product and process dictates the F:U molar ratio of
resin used. The molar ratio directly impacts the ultimate strength the resin will produce in the
board, i.e.. certain products require higher molar ratio resins to attain an adequate level of bond
strength (Rammon, 1990).
41
-------
2 r
1.8 '
o
to
& 1.6
u
2 1.4 1
g r
"i i
i - ¦ ; ¦— . -—
Particle board Particle board MDF HDWD
Face Core Plywood
Figure 6.1. Urca-formaldchydc resin mole ratios (Rammon 1990).
Reprinted with permission.
The F:U molar ratio relates to formaldehyde board emissions (Gollob, 1990). The higher the
mole ratio, the higher are formaldehyde board emissions and vice versa (Figure 6-2) (Gollob,
1990). However, there are limits to the amount of formaldehyde emissions reduction that can be
achieved through lowering the resin molar ratio (Gollob, 1990). As the F:U molar ratio is
lowered to somewhere around F:U = 1.00 to F:U = 1.10 (the range is different for different
mills), the board properties tend to drop off significantly (Figure 6-3) (Gollob, 1990).
Productivity and operating efficiency also drop off significantly (Gollob, 1990).
6.1.2.2 Use of Formaldehyde Scavengers with UF Resins
Another method for reducing formaldehyde board emissions is to use a formaldehyde scavenger
with a standard F:U molar ratio resin. Formaldehyde scavengers fall into two distinct categories;
scavengers that are incorporated prior to pressing the panel and scavengers that are incorporated
after pressing. The basic premise behind the two application methods is essentially the same;
both allow the use of a high mole ratio IJF resin with all its intended benefits while achieving
acceptable emissions by scavenging the excess formaldehyde.
Prepressing Scavengers. Most plants incorporate scavengers prior to pressing the panel. The
most prevalent type of prepressing scavenger is chemical urea. Urea is most commonly applied
as a 40 percent solution, although some plants use dry powdered urea. Application rates are
42
-------
Resin Mole Ratio
Figure 6-2. Formaldehyde emissions versus F:U mole ratio (Outman, 1991).
Reprinted with permission.
Fo rmaldc Kj"dc
Board Emissions
Board-Prnpertiies
05>0 1JOO l.lO 1.20
F/M IVIoIax* Ratio
1 -30
Figure 6-3. Emissions and properties (Gollob, 1990). Reprinted with permission.
43
-------
generally 0.25 to 1.0 percent urea based on dry wood weight. These levels usually result in
formaldehyde reductions of 15 to 50 percent depending on many plant characteristics (Graves,
1990). Figure 6-4 illustrates the estimated F:U molar ratio for urea solutions for the percent
liquid resin on wood. The data assume a 40 percent urea solution for the scavenger with
calculations set up for F:U = 1.20 molar ratio. However, an equivalent chart can be constructed
for any molar ratio. For example, if there were 9 percent liquid resin on wood without a
scavenger solution, then the molar ratio would be F:U - 1.20 (Gollob, 1990). As the percent
urea solution increases, and is considered part of the resin's F:U molar ratio, the net effect is a
decrease in the F:U molar ratio.
1.22
0% Urea Solution
2
18
0.50% Urea Solution
16
0.75% Urea Solution
1.14
1.00 Urea Solution
12
.1
1.08
9
10
11
12
13
14
8
15
16
Percent Liquid Resin on Wood
Figure 6-4. Estimated F:U molar ratios with scavenger solution (Gollob, 1990).
Reprinted with permission.
Another type of prepressing scavenger is a scavenging wax emulsion where the scavenging
chemicals are added to the normal wax emulsion. (Wax emulsions are used in engineered
boards to retard the absorption of water [see Section 4.2.4.1].) This approach eliminates the need
for a separate metering and storage system for the scavenger but does not provide flexibility in
scavenger level for different products or conditions (Graves, 1990).
A relatively new prepressing scavenger method is combination blending. Combination blending
is the process whereby two liquid IJF resins are used in combination to reduce formaldehyde
emissions without a loss in board properties or in production efficiency (McAlpine, 1990). The
two resins consist of a standard IJF resin and an ultra-low F:U mole ratio scavenger resin.
Typically, the scavenger resin is combined with the standard resin just before application to the
44
-------
wood by way of an in-line mixer, although good results have also been achieved using separate
application of the two components. Long-term contact of the two components results in a
mixture that behaves just like a low mole ratio resin (Graves, 1993).
In one study that evaluated the effectiveness of combination blending, formaldehyde emissions
from the panel were significantly reduced without any detrimental effect on board properties
(McAlpine. 1990). For the highest level of scavenger resin blended, the study measured a 48
percent reduction in formaldehyde board emissions from desiccator tests (Figure 6-5), and a 63
percent reduction in formaldehyde board emissions from large-scale test chambers (Figure 6-6)
(McAlpine, 1990).
1
S
0.8
a
O)
3
0.6
c*
>
M
0.4
o
td
u
• p4
0.2
0J
o
0
0.82
0.67
0.56
0.42
0.53 ]
0 45 /I 0.43 |
^ yi
0/0 5/5 10/10 15/15 20/20 25/20 30/20
% Scanvenger Resin (Face/Core)
Figure 6-5. Combination blending formaldehyde emissions, desiccator results
(McAlpine, 1990). Reprinted with permission.
Postpressing Scavengers. Postpressing treatments are much less common than prepressing
treatments but can be very effective (Graves, 1990). The most well known of these is gassing the
panels with anhydrous ammonia. Other techniques that have been tried include the application
of liquid ammonia or ammonia salt solutions to the board surface prior to stacking. All three
methods utilize the reactivity of ammonia toward formaldehyde forming a relatively stable
compound (Graves, 1990).
6.1.2.3 Melamine-Fortified Urea-Formaldehvde Resins
Melamine-fortified UF resins have been relatively popular in Europe for many years but have not
been widely accepted in North America (Graves, 1993). The appeal of this approach is improved
45
-------
0.18
0.16
a
0.14
Ph
ex
0.12
CD
3
0.1
<0
>
0.08
o
0.06
H
W
0.04
J
0.02
0
; 0.16
0.14
k I I/
0.12
0.11
0.09
0.09
0.06
,ZI71
0/0 5/5 10/10 15/15 20/20 25/20 30/20
% Scanvenger Resin (Face/Core)
Figure 6-6. Combination blending formaldehyde emissions, large scale test chamber
results (McAlpine, 1990). Reprinted with permission.
resistance to hydrolysis contributed by the melamine. Although the amount of melamine used is
quite small, generally less than 10 percent based on resin solids, the impact on cost of the resin is
significant due to the high cost of melamine. Prices published in the Chemical Marketing
Reporter (April 2, 1990) list urea at $0.10 a solid pound and melamine at $0.40 a solid pound
(Gollob, 1990). Thus, urea resins can be fortified with melamine to get a more stable resin, but
there is a big increase in the raw materials cost (Gollob, 1990). Successful production trials using
mclamine-fortificd UF resins were carried out at a particular plant in die United States in 1983
and yielded a 40 percent reduction in formaldehyde board emissions with no adverse effects on
production or board quality (Graves, 1993). Although proven effective, little acceptance of this
approach was forthcoming due to the higher cost of the resin. There have been a few instances
where a melamine-fortified resin was used to overcome hot stack degradation problems caused
by inadequate board cooling (Graves, 1993). Recently, there has been a resurgence of interest in
these resins, particularly to produce panels that will meet the European El formaldehyde
emissions requirements (Graves, 1993). Production trials have been largely successful, but
production volume remains small (Graves, 1993).
6.1.2.4 Phenol-Formaldehyde Resins
Formaldehyde emissions from panels made with phenolic resins are minimal (Graves 1993).
Phenol formaldehyde-bonded boards that have been tested, primarily OSB, have had
formaldehyde levels that were at, or only slightly above, background levels (Graves, 1993).
46
-------
Traditionally the two drawbacks that have prevented wide use of PF resins in the manufacture of
PB have been resin cost and cure speed. Resin costs would increase about two to three times if
phenolic resin were used to replace UF resin, and press cycles would have to be extended 50 to
75 percent (Graves, 1993).
There is one PB producer using phenolic resin on a regular basis (Graves, 1993). In addition to
low-formaldehyde board emissions, the board is much stiffer and machines more smoothly than
UF-bonded PB (Graves, 1993). Cost and productivity issues could be mitigated by using PF
resin only in the surface layer with a conventional UF resin in the core (Graves, 1993). Even so,
there is insufficient phenol supply to sustain a large-scale switch to PF resins by the PB industry
(Graves, 1993).
6.1.2.5 Methylenediphcnyl Diisocyanate
Methylcnediphenyl diisocyanate has seen limited use in the PB and MDF industry. It has
demonstrated low formaldehyde emissions levels and fast cure speed but has yet to be viewed as
an economic alternative to UF resins. In addition, there are concerns about workplace exposures
and press platen sticking. Methylenediphcnyl diisocyanate has been used in the core layer of PB
with a PF resin in the face layer (as commonly practiced in the OSB industry) with some success
(Graves, 1993). In addition, it is being used to produce a specialty "no formaldehyde added
MDF"' to satisfy a niche market (Graves, 1993).
6.2 Novel Adhesives
The three major adhesives used in the wood panel industry (UF, PF, and MDI) are all
synthesized from petroleum-derived chemicals. The wood panel industry consumes about
25 percent of the total U.S. adhesives production (Koch et al., 1987). Over the past decade, the
price of the raw materials used in these adhesives has grown rapidly. The price of phenol, for
example, has more than doubled since 1985 ($0.45 per solid pound in 1990 compared to $0.19
per solid pound in 1985). The price of methanol (the basic raw material used to produce
formaldehyde) increased from $0.68 per gal in June of 1994, to $1.30 per gal in October 1994
(Wellons, 1994). These rapid price increases are a result of increased competition for these raw
materials.
The major competition for phenol has come from demand for bisphenol A and caprolactam.
Bisphenol A is the feed stock component for polycarbonates that is used to produce automotive
plastics, compact discs, and computer discs. Caprolactam is a basic raw material for the
production of Nylon 6 that is used in the fast growing stain resistant carpet market. Phenol
demands for both bisphenol A and caprolactam are expected to exceed PF resin demand by 1995
(White, 1990). In addition to causing higher prices for PF adhesives, the demand for phenol is
reaching world capacity. In 1988, the consumption of phenol equaled 96 percent of the world
capacity (White, 1990).
47
-------
The increase in demand for methanol has been fueled both by general economic improvement in
the United States and abroad, and most important, by the growth of methyl tertiary butyl ether
(MTBE) as an oxygenate and octane booster in gasoline (Wellons, 1994).n The MTBE market
is forecasted to grow from the 1993 level of 4.4 million metric tons to over 8 million metric tons
in 1997 (Wellons. 1994). Other major methanol markets, formaldehyde and acetic acid, are
forecasted to grow equally to or above the rate of the gross national product (Wellons, 1994).
The significant price increases and potential scarcity of raw material supplies for UF, PF, and
MDI adhesives have created interest in renewable raw material sources. Research and
development funds are being expended by chemical raw material suppliers, forest products
companies, and wood adhesive/binder suppliers to search for renewable raw material sources to
replace entirely, or least partially, petroleum-derived chemicals in the manufacture of wood
adhesives. Some of the current research is presented below. Naturally derived adhesives are
included in this report as pollution prevention because of their potential to use renewable
resources, which in many cases are byproducts of other processes. An effort was made to collect
information on availability, cost, properties, composition, environmental effects, and stage of
product development for each product. Any information found on these topics is included with
each adhesive. However, for most of the adhesives, the above information was not found.
6.2.1 FAREZ Resin
FAREZ resin is manufactured from furfuryl alcohol, a derivative of corn. The resin is being
evaluated at the University of New Brunswick as an alternative low-VOC binder to substitute for
PF resins used for fiberglass insulation products and wood products (Rude, 1994).
6.2.1.1 Environmental Effects
Unlike PF resins, the FAREZ resin is storage stable at ambient temperatures, without
refrigeration. As delivered, the FAREZ resin contains very low amounts of volatile components.
Upon curing, especially heat curing, the FAREZ resin liberates 80 to 90 percent less total VOC
emissions and 80 to 90 percent less emissions of materials on the hazardous air pollutants
(IIAPs) list (Rude, 1994). The FAREZ system offers the same relative cure speed as the PF resin
systems. Various acidic catalyzation options are available that allow either ambient temperature
or heat curing capability.
T*Carbon monoxide (CO) is a colorless, odorless gas. It is primarily a byproduct of incomplete fuel combustion
in exacerbated at high altitudes. Because these environmental conditions reduce combustion efficiency, they
increase pollution from motor vehicle exhaust, especially during "cold starts," which last until the engine warms.
The CO nonattainment program begins with the division of nonattainment areas into classifications based on the
severity of CO pollution. The 1990 Clean Air Act Amendments (CAAA) set forth two classifications of CO
nonattainment areas (moderate and serious). Each classification is defined through a range of designated values.
Oxygenated gasoline, fuels blended to contain sufficiently high levels of oxygen, is one of the requirements set by
States to remedy all or part of a serious CO nonattainment area during winter months.
48
-------
6.2.1.2 Availability
The FAREZ resin is manufactured by the Quaker Oats Company and is currently in the
experimental stage of development. The insulation and wood products industries have shown
little interest in the resin because of its high cost; the cost of the resin is twice that of a PF resin.
However, cost analyses have been performed for the insulation, which show that using the resin
to meet future IIAP standards would be cheaper than purchasing and operating control devices
such as scrubbers. The same is likely to hold true for the wood product industry (Rude, 1994).
6.2.2 Methyl Glucosidc
Researchers at the University of Illinois have been investigating the use of methyl glucoside
(MeG), a corn-derived monosaccharide, in the manufacture of plywood adhesives. Researchers
have incorporated MeG into plywood adhesives as both an additive and as a partial replacement
for phenol in PF resins used in plywood adhesives.
MeG can be incorporated into plywood adhesives either by adding it to the glue mix used at a
mill or cooking it into the PF resins manufactured at regional plants. At the mill, MeG can be
added to the glue mix as an additional ingredient or it can replace part of the PF resin in glue
mix. (Details about the incorporation of MeG into glue mixes is published in Sellers and
Bomball, 1990.) Laboratory studies and large-scale mill trials indicate several positive properties
from the incorporation of MeG into glue mixes including lower adhesive cost, better dry-out
resistance, and better flow properties of the glue. The improved dry-out resistance allows spread
reductions of up to 12 percent, resulting in significant cost savings (Dunn, 1993).
A more direct and easier way to incorporate MeG into plywood adhesives is to cook it into the
PF resins at the regional distribution plants. The modified PF resins can then be distributed to
individual plywood mills, as opposed to attempting to optimize MeG in every plywood mill's
glue mix. Laboratory studies have shown that MeG can replace up to 50 percent of the phenol in
a base 1.7 to 2.1 formaldehyde to phenol (F:P) molar ratio resin, resulting in a resin with an F:P
molar ratio of up to 4.2; wood failure and tensile strengths comparable or superior to
conventional resins were achieved with no increases in free formaldehyde or loss in storage
stability (Dunn, 1993).
One pollution prevention benefit of adding MeG to PF resins is that the modified resin can bond
high-moisture veneer (veneer with 9 percent moisture was bonded in the above studies). The
pollution prevention advantage of using an adhesive that is capable of bonding high moisture
wood is the potential to reduce VOC emissions from both dryers and presses. Another pollution
prevention benefit of adding MeG to the PF resin is that phenol consumption is reduced, which
in turn may reduce phenol emissions at the press (emissions data were not found in the literature
to confirm this).
49
-------
6.2.2.1 Future Availability
MeG is made from corn. As the drought of the summer of 1988 demonstrated, even when yields
of corn are almost halved, com surpluses remain sufficient for all current needs (Dunn, 1993). In
normal years, American farmers produce nearly twice what is needed in the United States or can
be exported (Dunn, 1993).
The incorporation of MeG into PF resins (the easier and more direct way of incorporating McG
into plywood adhesives) has yet to be scaled-up to resin producers/plywood mills. But if the
modified resins can be demonstrated on a plant scale, incorporation of McG into plywood
adhesives will proceed at a much faster pace than at present. The long-term goal after successful
commercialization of this technology in the plywood industry is to extend this approach to other
structural panels such as OSB. OSB panels use a much higher amount of PF resin adhesive
because of the higher surface area of the wood strands being glued compared to plywood. The
use of MeG in the OSB industry could lead to greater cost savings and pollution prevention
benefits through adhesives reduction (Dunn, 1993).
6.2.3 Lignin Adhesives
Lignin is an aromatic polymer made up from phenylpropane units. Lignin makes up one of the
three major components of wood, along with cellulose and hemicellulose. In chemical paper
processes, such as kraft pulping, lignin is transformed into a soluble form and removed. In 1980,
an estimated 20 million tons of lignin (in the form of spent liquor) were produced from kraft
pulping in the United States. Only 35,000 tons (0.18 percent) of this lignin was recovered from
the spent liquor and used in manufacture of by-products (Zhao et al., 1994).
6.2.3.1 Adhesive Utilization
Lignin can be used as a feedstock for aromatic-based chemicals. Consequently, the abundance of
lignin as a waste product in pulp mills has made it a desirable raw material alternative to
nonrenewable petroleum-derived chemicals in the production of wood adhesives. Research into
utilizing lignin for adhesive production has been going on for years. A large portion of the
research has focused on the substitution of lignins for PF adhesives. While 100 percent lignin
substitution has yet to be achieved, there has been industrial success with partial substitution of
lignins into PF adhesives. Substitution of 20 to 30 percent of kraft lignin into PF resins is
common in plywood applications. Until recently, no more than 20 to 30 percent of lignin could
be substituted into PF resins because cure times increased as the amount of lignin increased.
Another drawback has been that lignin adhesives have low cross-linking and strength. However,
recently, a new approach has been developed that can substitute large amounts of kraft lignin for
PF adhesives, while actually increasing the cure speed, as well as board strength (Stephanou and
Pizzi, 1993). According to research published by Stephanou and Pizzi (1993), a different, and up
to now unknown, chemical reaction mechanism has been developed to give "truly competitive
kraft lignin-based wood adhesives of good exterior grade performance." The new approach,
50
-------
which utilizes isocyanates, has been demonstrated on a laboratory scale; exterior-grade PB panels
with excellent strength properties were prepared from adhesives containing 46 to 63 percent kraft
lignin resin solids, at press times much faster than those used for traditional synthetic PF resins
(Stephanou and Pizzi, 1993). The results of this research .. clearly indicate that excellent
lignin-based adhesives for exterior grade panel products can be manufactured. These adhesives
have none of the drawbacks which up to now have been the main stumbling blocks in the
effective industrial utilization of lignin for panel binders, namely low cross-linking and strength,
and particularly, far too slow pressing times. The adhesives produced gave panels of excellent
strength, good durability and very fast press times" (Stephanou and Pizzi, 1993).
6.2.3.2 Availability
Currently, Westvaco is the only U.S. company that operates a commercial lignin extraction
facility. Tn early 1990, the selling price of kraft lignin was approximately $0.32 per solid pound
to $0.34 per solid pound, and the selling price of phenol was approximately $0.45 per solid
pound (White, 1990). Although an adhesive manufacturer can purchase kraft lignin at a lower
cost than phenol, additional processing is required to make the lignin appropriate for adhesive
production; the additional processing results in the cost of lignin being equal to that of phenol
(White, 1990). However, if the price of phenol continues to rise and exceed the total
manufacturing cost of lignin, and assuming sufficient quantities of extracted commercial lignin
could be made available, lignin could compete as a partial raw material eoreactant with phenol in
PF wood adhesives (While, 1990). For there to be sufficient quantities of extracted lignin,
additional lignin extraction facilities must be built. The capital cost of a new commercial lignin
extraction facility compared to the capital cost of a new phenol plant is estimated to be almost
equal per pound of product produced.
6.2.4 Tannin
Tannin refers to a widely occurring group of substances derived from plants. In their most
common usage, these substances are capable of rendering raw hides into leather. Common
tannin or tannic acid is found in a variety of nuts, tea, sumac, and bark. It is a complex, dark
polyhydroxy PF compound.
As a bark product, tannins are readily available. Large quantities of bark are available at mill
sites where bark is a waste product of sawmilling and other panel making processes. On average,
bark contains 15 to 30 percent polyphenols, 20 to 30 percent lignin, 30 to 45 percent
carbohydrates and 1 to 3 percent fats and waxes (Koch et al., 1987). These values are dependent
upon wood species, location, age, and exposure.
6.2.4.1 Adhesive Utilization
The use of tannin adhesives for the manufacture of exterior-grade weatherproof PB has gained
increased industrial and technical acceptance during the past 20 years (Pizzi, 1983, 1989). Until
51
-------
recently, all industrial formulations have been based on tannins, such as wattle (mimosa extract
of commerce) and quebracho (Pizzi 1983, 1989). In countries where these tannins are produced,
wattle and quebracho tannins have been progressively displacing synthetic PF adhesives for
manufacturing exterior PB; this progression is attributed to their lower cost and excellent
performance (Pizzi 1983,1989). However, the total worldwide production of these two tannins
is only 150,000 tons per year. This is coupled with the fact that the tannin producers do not
make more than 20 to 30 percent of their total production available for adhesives application (the
rest being reserved for their traditional market, leather). For these reasons, these materials,
although industrially and commercially very successful, have had little opportunity to influence
the PF adhesives market.
A tannin that can easily be extracted and rendered available in most countries is pine tannin,
extracted from the bark of the most diffuse worldwide forestry crop, pine trees (Pizzi et al.,
1993). Significant amounts, well into the millions of tons, of this tannin could be rendered
available if a successful wood adhesive system based on it could be proven industrially (Pizzi
et al., 1993). Recently in Chile, a small industrial producer of native and sulphated pine tannins
has developed an adhesive consisting of MDL tannin, and formaldehyde for exterior-grade PB.
Industrial results of extended plant production runs of this adhesive system showed "very
encouraging results" (Pizzi et al., 1993).
6.3 Source Management Approaches
6.3.1 Use/Maintenance
Emissions from engineered wood products are due to the emissions of unreacted formaldehyde,
hydrolysis of the formaldehyde-based resins, the wood itself, and materials used in product
finishing. The rate of formaldehyde emissions has been shown to be greatly affected by
temperature and relative humidity. Higher temperatures can increase the rate of emissions, while
increased relative humidity increases the rate of hydrolysis. Therefore, keeping the wood
products dry and not exposed to high temperatures should reduce the potential for hydrolysis and
emissions.
6.3.2 Use of Laminates or Veneers
Although designed for other purposes such as aesthetics, overlays, coatings, and veneers may
serve as barriers to emissions (GEOMET, 1994). These coverings can also protect the resins
from drastic changes in relative humidity and temperature, thereby minimizing the potential for
hydrolytic attack. However, it should be noted that additional resins may be used as adhesives to
attach these covers to the wood products. Both UF and PF resins have been used for this purpose
and, as a result, increase the amount of resin present as a potential source. Other adhesives used
for attaching laminates and veneers include water-based contact cement, epoxy, and polyvinyl
acetate. It is possible that adhesives will contribute to their own emissions; however, published
adhesive emissions data were not found in the literature to verify this issue.
52
-------
6.3.3 Bakeout
Offgassing of organics is a major source of pollutants from engineered wood products. While
not pollution prevention, bakeout procedure has been suggested to increase the initial emissions
of organics, reducing the overall pool of residuals available for later emissions in the indoor
environment. Theoretically, this reduces the rate of emissions that would occur during
occupancy. During bakeout, the temperature of a new building is elevated and the ventilation is
increased prior to occupancy to accelerate the aging of the building materials and furnishings and
to increase the emissions rate of free unreacted organic compounds. (Both the temperature and
ventilation rate are elevated and increased, respectively, above normal occupancy conditions.)
Some researchers believe that this procedure is effective, but just how effective is unknown
because offgassing varies with temperatures, time, and ventilation rates (Grimsrad et al, 1988).
In addition, in most buildings you cannot increase both the temperature and ventilation
sufficiently to reduce sink effects. (Pollutant sinks represent mechanisms by which pollutants
can escape the indoor air environment. This escape can occur through deposition on room
surfaces, adsorption with those surfaces, reaction with other compounds present in the air,
reactions with surfaces, or dispersion to other rooms or outside. Adsorption initially occurs
where gas phase molecules are held to the surfaces by relatively weak forces [i.e., Van der Waals
forces and weak chemical bonds]. Because these forces are weak, desorption can occur due to a
finite mass capacity of the sink and contact with lower concentration air thereby allowing the
molecules to be re-cmitted into the indoor air environment.) Finally, this practice may be limited
by the potential damage that can occur as a result of subjecting certain types of materials (e.g..
wood products) to higher temperatures. For example, bakeout may lead to excessive drying and
result in stresses that affect both the appearance and structural integrity of some wood products.
7.0 CONCLUSION
This report summarizes information collected from existing sources and a research planning
meeting. The compiled information was used to choose several products for source
characterization studies. Once these emission studies are complete, pollution prevention options
will be identified and evaluated. Interim papers on preliminary research results will be presented
at appropriate conferences or symposiums. A final report featuring the research conducted under
this cooperative agreement between EPA and RTI will be issued upon completion of the research
in 1996. An evaluation of the health and environmental effects was beyond the scope of work
for this project.
53
-------
REFERENCES
AFA (American Fiberboard Association). Palatine, TL. 1994. Correspondence with Louis
Wagner, Director of Technical Services.
APA (American Plywood Association). 1993. Regional Production and Distribution Trends for
Structural Panels and Engineered Wood Product. American Plywood Association,
Tacoma, WA
Baldwin, R.F. 1975. Plywood Manufacturing Practices. Miller Freeman Publications,
San Francisco, CA.
Ball, G.W. 1981. Proceedings of the Fifteenth International Particleboard/Composite Materials
Symposium. Washington State University, Pullman, WA.
Blackman, Ted. 1994. 'Green'Rules Hinder, Aid Engineered-wood Producers. Wood
Technology, (121), (4).
Blomquist, R.F., A.W. Christiansen, R.II. Gillespie, and G.E. Myers. 1981. Adhesive Bonding
of Wood and Other Structural Materials. IJSDA Forest Service, Forest Products
Laboratory. Madison, WT, in cooperation with The University of Wisconsin—Extension,
Madison, WI.
Carliner, Michael. 1994. What's driving lumber prices? Housing Economics. National
Association of Home Builders, Washington, DC. January, pp. 5-9.
Chemical Marketing Reporter. 1990. April 2.
Dunn, Larson B., Jr. 1993. Turning Crops into Plastic...and Cash. Illinois Research.
Spring/Summer edition.
Evergreen Magazine. 1994. Oregon's Forests. 1994 Fact Book. Medford, OR.
Federal Register. 1984. Manufactured Home Construction and Safety Standards:
Supplementary Information, and. Formaldehyde Regulation. Rules and Regulations, Vol.
49, No. 155, 24 CFR Part 3280, Thursday, August 9, 1984.
Frashour, Ron. 1990. Production Variables, Blowline, and Alternative Blending Systems
for Medium Density Fiberboard. In: Proceedings of the NPA Resin and Blending
Seminar. National Particleboard Association, Gaithersburg, MD.
Galbraith, John, Marketing Manager of ICI Polyurethanes Group, West Deptford, NJ, 1995.
Personal Correspondence.
54
-------
GEOMET Technologies, Inc. 1994. Wood Product VOC Emissions Testing, Final report prepared
for the National Particleboard Association, Gaithersburg, MD. GEOMET Report Number
IE-2636.
GTRI (Georgia Tech Research Institute), Atlanta, GA. 1993. Report on Emission Testing for the
National Particleboard Association.
Godish, Thad. 1988. Residential Formaldehyde Contamination: Sources and Levels. Comments
Toxicology 1988, Vol. 2, No. 3, pp. 115-134. Indoor Air Quality Research Laboratory,
Ball State University, Muncie, IN.
Gollob, Lawrence. 1990. Understanding Today's UF Resins. Part A. General
Consideration. In: Proceedings of the NPA Resin and Blending Seminar. National
Particleboard Association, Gaithersburg, MD.
Grantham, J.B., and G.H. Atherton. 1959. Heating Douglas-fir veneer blocks - Does it pay?
Forest Research Laboratory, Corvallis, OR. Bulletin 9.
Graves, Greg. 1990. Catalysts and Scavengers in Particleboard Manufacture. In: Proceedings of
the NPA Resin and Blending Seminar. National Particleboard Association, Gaithersburg,
MD.
Graves, Greg. 1993. Formaldehyde Emission Control Via Resin Technology—North American
Particles. In: Proceedings of the Twenty-Seventh International Particleboard/Composite
Materials Symposium. Washington State University, Pullman, WA.
Grimsrud, D. T., J. T. Brown, W. J. Fisk, and J. R. Girman. 1988. The Indoor Environment of
Commercial Buildings. Prepared for Electric Power Research Institute, Palo Alto, CA, pp.
4-17. July 19.
Groah, William, Technical Director, Hardwood Plywood and Veneer Association, Reston, VA.
1994. Personal Correspondence.
Haygreen, J. G., and J. L. Bowyer. 1989. Forest Products and Wood Science: an Introduction,
2nd Edition. Iowa State University Press. Ames, IA.
I loag, Mike, National Particleboard Association, Gaithersburg, MD. 1993. Personal
communication.
HPVA (Hardwood Plywood and Veneer Association). 1991. Before the EPA: Comments of
the Formaldehyde Institute, National Particleboard Association, Hardwood Plywood
Manufacturers Association on Economic and Technical Implications of Various Regulatory
Options. Hardwood Plywood and Veneer Association, Reston, VA.
55
-------
ICI Personal Communication. 1994. Personal communication between representative of 1CI
Polyurethanes Group, West Deptford, NJ, and Sonji Turner of Research Triangle Institute.
Koch, G.S., F. Klareich, and B. Exstrum. 1987. Adhesives for the Composite Wood Panel
Industry. Noyes Data Corporation, Park Ridge, NJ.
Lehmann, W.F. 1982. Correlations between various formaldehyde tests. In: The Sixteenth
International Particleboard/Composite Materials Symposium Technical Forum.
Washington State University, Pullman, WA.
Lcukroth. R.W. 1991. Carpet Policy Dialogue: Compendium Report. EPA -560/2-91-002
(NTIS PB92-115005) U.S. EPA Office of Toxic Substances, Washington, D.C. September.
Levin, Hal, & Associates. 1989. Building Materials and Indoor Air Quality. Occupational
Medicine: State of the Art Reviews. 4(4): October - December. Hanley & Belfus, Inc.,
Philadelphia, PA.
Lutz, J.F. 1960. Heating veneer bolts to improve quality of Douglas-fir plywood. USDA Forest
Service, Forest Products Laboratory, Madison, WI. Report 2182.
Maloney, T. M. 1977. Modern Particleboard and Dry-Process Fiberboard Manufacturing. Miller
Freeman Publications, Inc., San Francisco, CA.
McAlpine, Cindy. 1990. Combination Blending for Particleboard. In: Proceedings of the NPA
Resin and Blending Seminar. National Particleboard Association. Gaithersburg, MD
McCredie, Bill. 1993. The Future of Engineered Wood Panel Availability. Wood & Wood
Products, February, p 60.
Meyer, Beat. 1979. Urea-Formaldehyde Resins. Addison-Wesley Publishing Company, Inc.,
Reading, MA.
Meyer, Beat. 1983. Indoor Air Quality. Department of Chemistry, University of Washington,
Seattle, WA. Addison-Wesley Publishing Company, Inc., Reading, MA.
Meyer, B., B. A. Kottes Andrews, and R. M. Reinhardt. 1986. Formaldehyde Release from Wood
Products. Publication from symposium sponsored by the Division of Cellulose, Paper, and
Textile Chemistry at the 189th Meeting of the American Chemical Society, April 28 - May
3,1985. American Chemical Society, Washington, DC.
Meyer, B., W.S. Johns, and J.K. Woo. 1980. Formaldehyde release from urea-formaldehyde
resins. Forest Prod. J. 30(3), 24.
56
-------
Ivfolhave, Lars. 1982. Indoor Air Pollution Due to Organic Gases and Vapours of Solvents in
Building Materials. Environmental International, 8:117-127.
Moslemi, A.A. 1974. Particleboard, Volume 2: Technology. Southern Illinois University Press,
Carbondale, IL.
NCASI (National Council of the Paper Industry). 1993. A Review of 1991 SARA Section 313
Release Reports Filed by Forest Product Industry Manufacturing Facilities. Technical
Bulletin No. 93-13, ResearchTriangle Park, NC.
NRC (National Research Council). 1981. Formaldehyde and Other Aldehydes. Committee on
Aldehydes, Board on Toxicology and Environmental Health Hazards, Assembly of Life
Sciences, Washington, DC.
NPA (National Particleboard Assocation). 1994a. Buyers and Specifiers Guide to Particleboard
and MDF. National Particleboard Association, Gaithersburg, MD.
NPA. 1994b. 1993 U.S. Annual Shipments and Production. National Particleboard
Association, Gaithersburg, MD.
Nelms, L.H., M.A. Mason, and B.A. Tichenor. 1986. The effects of ventilation rates and product
loading on organic emission rates from particleboard. Refrigerating and Air-Conditioning
Eng. Atlanta, GA. pp. 469-485.
Neretnieks, I., J. Christiansson, L. Romero, L. Dagerhold, and J.W. Yu. 1993. Modeling of
Emission and Re-emission of Volatile Organic Compounds from Building Materials with
Indoor Air Applications. Indoor Air. ISSN 0907-2055. January 21.
Outman, Jim. 1991. Press Vent Formaldehyde Emissions. In: Proceedings of the NPA/APA
Pollution Control Technology Seminars. National Particleboard Association,
Gaithersburg, MD.
Pickrell, J. A., L.C. Griffis, and C.H. Hobbs. 1982. Release of Formaldehyde from Various
Consumer Products. Report by the Lovelace Inhalation Toxicology Research Institute for
the CPSC, Albuquerque, NM.
Pizzi, A. 1983. Wood Adhesives: Chemistry and Technology. Marcel Dekker, Inc. New York,
(NY) and Basel (Switzerland).
Pizzi, A. 1989. Research vs. Industrial Practice with Tannin-based Adhesives. In: Adhesives
from Renewable Resources (Eds. R.W. Hemingway and A.H. Conner). Chapter 19.
American Chemical Society Symposium Series No. 385: pp. 254-270.
57
-------
Pizzi, A., E.P. Von Leyser, J. Valenzuela, and J.G. Clark. 1993. The Chemistry and Development
of Pine Tannin Adhesives for Exterior Particleboard. Holzforschung, Vol. 47. No. 2. pp.
168-173.
Rammon, Rick. 1990. Understanding Today's UF Resins. PartB. Formulating UF Resins
for Different Board Plants. In: Proceedings of the NPA Resin and Blending Seminar.
National Particleboard Association, Gaithersburg, MD.
Roberts, Don G. 1994. Looking through the Branches; Composite Panels: Continuing Scarcity
or Impending Glut? Wood Gundy, Inc., Ottawa, Ontario, Canada, p 5.
Rude, Carl, QO Chemicals, Inc., West Lafayette, IN, 1994. Telephone communication.
Saarela, Kristina, and Maija-Liisa Mattinen. 1993. Variation of Volatile Organic Compounds in a
New Office Building as a Function of Time. In: Proceedings of Indoor Air '93, (2):75-80.
Sellers, T. Jr., and W.A. Bomball. 1990. Forest Products Journal, pp. 40:52.
Sellers, Terry., J. R. McSween, and W. T. Nearn, 1988. Gluing of Eastern Hardwoods:
A Review. U. S. Department of Agriculture, Forest Service, New Orleans, LA.
Sparks, L.E., M.D. Jackson, B.A. Tichenor. 1988. Comparison of EPA test house data with
predictions of an indoor air quality model. In: Proceedings of IAQ '88, Engineering
Solutions to Indoor Air Problems. Atlanta, GA. American Society of Heating,
Refrigerating and Air-Conditioning Engineers, pp. 251-264.
Spedding, D.J., and D.L. Edmondson. 1980. Formaldehyde emission from building materials.
Clear Air (Melbourne), 14(4):35.
Stephanou, A., and A. Pizzi. 1993. Rapid-Curing Lignin-Based Exterior Wood Adhesives: Part
II: Esters Acceleration Mechanism and Application to Panel Products. Holzforschung,
47(6):501-505.
Suchsland, O., and G. E. Woodson. 1987. Fiberboard Manufacturing Practices in the United
States. U.S. Department of Agriculture, Washington, DC.
Sundin, B. 1982. Status of formaldehyde problem. In: Proceedings of the Sixteenth
Particleboard/Composite Materials Symposium. Washington State LTniversity, Pullman,
WA.
Sundin, E. Birger, M. Risholm-Sundman, and K. Edenholm. 1992. Emission of Formaldehyde
and Other Volatile Organic Compounds from Sawdust and Lumber, Different Wood-Based
Panels, and Other Building Materials: A Comparative Study. In: Proceedings of the
58
-------
Twenty-Sixth International Particleboard/Composite Materials Symposium. Washington
State University, Pullman, WA.
Thuagersen, Katharina, L. Guiinarsen, and P. A. Nielsen. 1993. The Effect of Indoor Air Quality
by Water-Damaged Chipboards. In: Proceedings of Indoor Air '93, 2:537-541.
Tichenor, B.A., and M.A. Mason. 1988. Organic Emissions from Consumer Products and
Building Materials to the Indoor Environment. JAPCA, 38(3), 264-268.
USDC (United States Department of Commerce). 1988. Current Industrial Reports.
Hardwood Plywood. MA24F(88)-1.
U.S. EPA (Environmental Protection Agency). 1994. Formaldehyde (CAS # 50-0-0) Carcinogen
Summary. Integrated Risk Information System (IRIS). Environmental Criteria and
Assessment Office, Cincinnati, OH.
Vaught, C. C. 1989. Evaluation of Emission Control Devices at Waferboard Plants. Publication
No. EPA-450/3-90-002 (NTIS PB90-131442). Control Technology Center, Office of Air
Quality Planning and Standards, U.S. Environmental Protection Agency, Research Triangle
Park, NC.
Wagner, Louis, Director of Technical Services, American Hardboard Association/American
Fiberboard Association, Palatine, IL, 1994. Personal correspondence.
Wellons, J.D., Georgia-Pacific, Atlanta, GA. 1994. Personal correspondence.
White, James T. 1990. Supply, Demand and Cost of Wood Adhesives and Binders - Renewable
and Nonrenewable. Georgia-Pacific Resins, Inc. Decatur, GA.
(WPPF) Wood Panel Products Federation. 1994. Wood Panel Products Technical Information.
Registered Office, 1 Hanworth Road, Feltham, Middlesex TW 135AF.
Wolff, Heinz P. 1996. Product Brochure. Eduard Kusters Maschinenfabrik Gmbk and Co.,
Krefeld, Germany.
Wood Technology. 1994. Panel Review. 121 (4): 14,34.
Youngquist, J.A., B.E. English, II. Spelter, and P. Chow. 1993. Agricultural Fibers in
Composition Panels. In: Proceedings of the Twenty-Seventh International
Particleboard/Composite Materials Symposium. Washington State University, Pullman
WA.
59
-------
Zhao, Lin-wu, B.F. Griggs, Chen-Loung Chen, and J.S. Gratzl. 1994. Utilization of Softwood
Kraft Lignin as Adhesive for the Manufacture of Reconstituted Wood. Journal of Wood
Chemistry and Technology, 14(1): 127-145.
60
-------
Appendix A
Chemistry of Formaldehyde-Based Resins
This appendix presents a detailed discussion of resins used in the manufacture of composite
wood materials. The literature on resins is rather extensive to be reviewed even superficially;
therefore, the reader may wish to review a polymer chemistry handbook for more detailed
information.
A.l Urea-Formaldehyde Resins
Urea-formaldehyde resin was patented in 1920 by Hanns John, Magister of Pharmacy of
Prague (Meyer et al., 1986). The main difference between early urea-formaldehyde resins,
whose mechanisms are depicted below, and modern wood adhesive resins is quality control
during manufacturing and molar ratio of the reagents, urea and formaldehyde (Meyer et al.,
1986).
Urea, H2NCONH2. reacts with formaldehyde to form urea-formaldehyde resins. The four
hydrogen atoms in a urea molecule are potentially reactive. Proper cross-linkage is important
for resin production. Formaldehyde is added, and it reacts with unreacted -NH2 groups to
provide chemical cross-links between polymer chains. When urea and formaldehyde react on
an equimolar basis, the following reactions occur to eventually produce the urea-formaldehyde
resin:
H H
I I
H—C=0 + H2N — C — NH2 —~- HOCH2 — N—C — NH2
Formaldehyde fi B
O 0
Urea Melhylolurea
A monomer is the intermediate product. It polymerizes further to produce a thermoplastic
resin and water.
H H H H
1 > II
HOCH2 — N—C — NH2 + 0=C— H —*• HOCHL-N-C-N-CH90H + H,
I I
O 0
Dimethylolurea
A- 1
-------
Eventually:
~ChL-N~
i
ch2
H H
~N~ CH2—N—C — N~
HOCH
HCHO, urea
0=C
o CH2 + (n-1) H20
O
—-N—CH9—N—C—N
i i
H,C O
2|
— N—CH2~~
n
Some additional formaldehyde is needed to react with a few of the unreacted -NH2 groups and
to provide chemical cross-links between polymer chains. This additional formaldehyde is
significant to the formation of urea-formaldehyde for several reasons. First, an excess of
formaldehyde results in quicker polymerization, which in turn results in lower manufacturing
costs. Second, sufficient formaldehyde is necessary to provide adequate polymerization and
produce satisfactory properties of the desired final product. Finally, excess formaldehyde
results in unreacted formaldehyde in the final product. This excess formaldehyde, also called
"free formaldehyde," slowly diffuses from products and may result in increased formaldehyde
concentrations in indoor air when these products are used indoors (NRC, 1981).
Originally, the formaldehyde-to-urea molar ratio for the urea-formaldehyde resin was 2. This
molar ratio corresponded to the number of chemically reactive groups present in the reagents
and provided for sufficient formaldehyde polymerization of all primary and most secondaiy
amino groups. In the early 1980s, most urea-formaldehyde resins marketed as wood adhesive
resin contained a molar ratio of 1.8 for formaldehyde to urea although proof was available that
lowering the overall molar ratio further reduced the potential for postmanufacture
formaldehyde release (Meyer et al., 1986).
Nonetheless, progress has been made in formulating low-molar ratio resins and in capturing
unreacted methylol groups. Current adhesive resins are manufactured in three or more steps.
The first step still involves the addition of urea to a highly concentrated formaldehyde
solution, frequently with a formaldehyde-to-urea ratio of 4. This reaction produces a mixture
of methylol compounds. Modifications in modern resin production require that urea be added
two or three times. These additions bring the overall molar ratio down to a sufficient level
A - 2
-------
(lower than 1.8) to retain unreacted amino groups. Amino groups behave as scavengers of
formaldehyde that was unreacted or released by hydrolysis of unreacted methylol functions
(Meyer etal., 1986).
A. 2 Phenol-Formaldehyde Resins
Reactions of phenol and formaldehyde produce phenol-formaldehyde resins. Phenol reacts
with both formaldehyde and paraformaldehyde in the presence of an alkali or acid medium to
produce the phenolic resin. The phenol-formaldehyde reactions fall into two groups: (1) the
formation of methylol phenols, called phenol alcohols, and (2) the formation of polynuclear
methylene derivatives. The following is an example of the reaction mechanisms involved in
the former group to form the phenolic resin:
OH
OH
(gpCH2^0)
OH
H+ or OH"
HCHO
>
,CH2OH c6h5oh
>
OH
o¦ Hydroxym ethyiphenol
HCHO. C6H5OH
A - 3
-------
Stages involved in the formation of the phenol-formaldehyde resin presented above are as
follows:
• Phenol and formaldehyde react in an acidic or basic medium to form o- or
/j-hydroxymethylphenol.
• Hydroxymethylphenol reacts with another molecule of phenol.
• Then water is lost, and a compound with two rings joined by a -CH2- link is formed.
Three positions on each phenol molecule are susceptible to attack, which makes the reaction
process difficult to control. Condensations contain many cross-links and produce a
high-molecular-weight polymeric product. The general class of such polymers,
phenol-formaldehyde resins, is the final product. Bakelite is the name of this dark, brittle,
cross-linked polymer. The phenolic resin is one of the oldest synthetic polymers.
Very little free formaldehyde is present in phenolic resins; therefore, emissions are lower than
those of urea-formaldehyde resins. The low free formaldehyde content is due to both the low
formaldehyde-to-phenol molar ratios in resin synthesis and the tendency of nearly all the
formaldehyde to react irreversibly with phenol. The little residual formaldehyde present is
reduced further by reactions that occur when the phenolic resin cures. Under curing
conditions, unreacted formaldehyde continues to react with phenol to form larger
phenol-formaldehyde polymer chains.
A.3 Melamine-Formaldehyde Resins
Melamine, a cyclic trimer of cyanamide, reacts with one to six molecules of formaldehyde,
producing mono- to hexa-methylols. These six methylol compounds are often used to prepare
melamine resins. Melamine and urea chemistries are similar. Boiling the proper
melamine-to-formaldehyde ratio for a short time produces hexa-methylols. The reaction
mechanism for melamine and formaldehyde is as follows:
nh2
N<
+ 6CH2 (OH)2
>N-C
N
C-N<
N
A - 4
-------
Melamine offers six hydrogen atoms to react with formaldehyde to form melamine resins.
Unlike the four potentially reactive hydrogen atoms of urea used to form urea-formaldehyde
resins, this reaction yields better cross-linking, and, thus, better water-resistant adhesives are
produced.
In addition, blends of melamine-formaldehyde resins with cross-linked novalak phenol-
formaldehyde resins are used to produce melamine-formaldehyde phenol-formaldehyde
formulations suitable for exterior exposure. (Novalak phenol-formaldehyde resins are linear in
structure and require heat to cure plus hardener to improve crosslinking.) The hardwood and
laminating industries use urea-formaldehyde/melamine-formaldehyde blends for products that
require improved durability over that of UF and light-colored glue line (Sellers et al., 1988).
A.4 Resorcinol-Formaldehyde Resins
Resorcinol-formaldehyde has been in use since 1943 (Pizzi, 1983). Resorcinol-formaldehyde
resins are synthetic polymer resins based on the condensation reaction of formaldehyde
(derived from methanol) and resorcinol (derived from benzene) (Koch et al., 1987).
The resorcinol-formaldehyde resin is formed by the reaction of resorcinol with deficient
amounts of formaldehyde (less than required for complete cure). Additional formaldehyde is
added at the time of application where curing to a solid state takes place. The following
schematic depicts this reaction:
0] + HCHO
Oj~CH2 [Or-CH:
n>0
A.5 Diphenyl-Methane-Isocyanate (Methylenediphenyl Diisocyanate)
The copolymer methylenediphenyl diisocyanate (PMDI), is a member of the family of
isocyanates that are generally low-viscosity liquids formed through phosgenation of the
reaction products of aniline and formaldehyde. Isocyanates are oil-based products derived
from naphtha as shown in Figure A-l. PMDI is known for its bonding strength. Thus, it has
A - 5
-------
become a prominent adhesive candidate. PMDI has several synonyms, including
diphenyl-methane-isocyanate and methylenediphenyl diisocyanates (MDI). The commercial
product, known as diphenyl-methane-isocyanate, CII2(-CH4)-NC02, is the reaction product of
a urethane bridge:
r-o-co-nh-c6ii4-cii2-c6ii4-nii-co-o-r,
where R is wood cellulose, copolymer, or filler. Not only does PMDI adhere to wood, but it
also adheres to metal surfaces of the presses used in composite wood product fabrication.
Consequently, when PMDI use is not confined to the core of the fabricated composite wood
product, a special release agent is needed to detach the composite wood product from metal
surfaces of the press.
A - 6
-------
NAPHTHA
Cracking
Nitration
Hydrogenation
Condensation
Phosgenation
Refining and chemical
modification
NITROBENZENE
ANILINE
POLYAMINES
MIXTURE
POLYISOCYANATES
MIXTURE
FORMALDEHYDE
HYDROCHLORIC ACID
PHOSGENE
Range of MDI Variants
Figure A-l. Manufacturing route to isocyanate (Ball, 1981).
A - 7
-------
Appendix B
Summary of Indoor Air Emissions Research
Section 2.0 of this report discusses indoor air emissions from finished and unfinished
composite wood. Results from several studies of emissions from composite wood products are
summarized in this appendix.
B.l Melhave Study
In a 1982 Danish study, Molhave determined organic emissions from 42 building materials.
This analysis suggested that building materials may be the primary source of organic
compounds in the indoor environment (Molhave, 1982). In this study, Molhave identified 52
different chemical compounds from the air around 42 building materials. Surface materials,
such as plywood and fiberboard, and materials normally used in interior construction, such as
particleboard, account for live products or 12 percent of the materials studied.
Surface materials were delivered directly from production or factory stock. The producers cut
them in standard A2-format (420 X 594 mm2) and wrapped them in heavy
polyvinylidenechloride foil to prevent contamination. The emissions concentrations were
measured in the air around 0.25 m2 of test materials in a 1-m3 stainless steel chamber supplied
with 0.69 liters fresh air per minute (one air change per 24 hours at 21.1 °C [standard
deviation - 1.7 °C]) and 35 to 40 percent relative humidity. The exhaust from the chamber
was analyzed using combined gas cliromatography/mass spectrometry (GC/MS) for the
qualitative and quantitative analysis. Industrial gas detection tubes were applied for aldehyde,
especially formaldehyde, detection.
The results for the five composite wood products that were evaluated as part of this study are
presented in Table B-l. All five materials were characterized as building materials used in
walls. The first three materials, which are particleboards, were further categorized as
materials used inside the construction, and the plywood and fiberboard were categorized as
surface materials. Formaldehyde emissions were not examined from the first two
particleboard samples listed in the table. Although the remaining three materials in Tabic B-l
were analyzed for aldehyde emissions, no reaction took place in the industrial gas detection
tubes. Consequently, formaldehyde emissions were not detected.
Of the 52 different chemical compounds (representing 10 chemical categories), Mcilhave
identified 10 chemical compounds as most frequently found in the air around the 42 building
materials tested. Those compounds are categorized as aromatic compounds and alkanes and
represent a total average concentration of 72.96 fig/m3. Although the author stated the
number of compounds detected and identified, he did not specifically identify the chemicals
emitted from each tested building material. The identification of chemicals emitted from each
material would have been valuable information for this report. Instead, Table B-l lists the
B-l
-------
average concentration of organic vapors emitted by each composite wood material, their
specific emissions rates, and the number of compounds detected and identified.
Table B-l. Results of Analysis of Air Around Materials
Organic
Concentration
(mg/m3)
Emissions
Number of
Type of
Material
Rates
(mg/m2-h)
Compounds
Detected
Identified
Particleboard
1.56
0.120
29
10
Particleboard
1.73
0.130
28
11
Particleboard
3.56
0.140
24
7
Plywood
1.07
0.044
16
0
Wood fiberboard
2.96
0.120
23
7
Excerpted from: Molhave, 1982.
B.2 Levin Study
Using software developed by the U.S. Environmental Protection Agency (RPA), Hal Levin
and Associates, architectural consultants, tabulated model predictions of typical emissions rates
from sources in a 400-m2 office area (Levin et al., 1989). To construct the model, EPA
researchers used the results of chamber, headspace, and test house experiments to construct a
model to predict the total volatile organic compound (TVOC) concentrations in indoor air
(Sparks et al., 1988). Source strengths and ventilation rates were also used in the model to
compute the typical emissions factors and emissions rates for assumed product loadings of
various products.
The predicted emissions rates of four composite wood products and their assumed product
loadings are shown in Table B-2. The estimates for two sources, the second particleboard
sample and the plywood panel, represent predicted formaldehyde (HCHO) emissions only.
The estimates for the two remaining samples, the first particleboard and chipboard, depict
predictions for TVOC emissions. The first particleboard was 2 years old; the other
particleboard and plywood panels were new. Information on the condition of the chipboard
was not available. Table B-3 lists organic compounds emitted from composite wood products.
The tabulated information was extracted from an appendix table in Levin et al. (1989). Source
types are those for which quantitative data on emissions have been obtained from chamber
tests or for which qualitative data are available from many sources (Levin et al., 1989).
Unfortunately, emissions data were not listed in the original table.
B-2
-------
Table B-2. Typical Predicted Emissions Rates for Sources in 400-m2 Office Area
Source"
(Noted Emissions Type)
Emissions Rate
(mg/nr-h)
Assumed Product
Loading
Mass Flow
Rate
Particleboard (TVOC)
Particleboard (HCHO)
Plywood paneling (HCHO)
Chipboard (TVOC)
0.20
300
60
2.00
300
600
1.00
1000
1000
0.13
300
39
Excerpted from: Levin et al., 1989.
a The data do not represent all products of the source type listed. Product-to-product variability can be very high.
Table B-3. Sources of Indoor Organic Compounds
Compounds
Substantiated Sources'*
Formaldehyde
Ethanol
Isopropanol
n-Hexane
2-Methylpentane (isohexane)
Benzene
Benzaldehyde
Ethylbenzene
1,2.4-Trimethylbenzene
n-Propylbenzene limonene
Limoncnc
a-Pinene
Undecane
Plywood, particleboard
Fiberboard
Particleboard
Chipboard
Chipboard
Particleboard
Fiberboard, particleboard
Chipboard
Chipboard
Chipboard
Chipboard
Chipboard
Chipboard
Excerpted from: Levin et al., 1989.
a Selected compounds that have been measured in indoor air and that may have come from material sources.
'* Source types for which quantitative data on emissions have been obtained by chamber tests or for which
qualitative data are available.
B - 3
-------
B.3 Saarela Study
In this study (Saarela and Mattinen, 1993), TVOC concentrations as a ftinction of time were
measured in a new office building with ventilation operating at high efficiency and with
ventilation switched off. The office building ventilation rate was 2.5 air changes per hours
(ACH) from Monday morning to Friday evening. Ventilation was switched off on Saturday.
The new two-floor office building was divided into two parts; eastern and western sections
were separated by a staircase and sanitary rooms. The two floors consisted of open and closed
office spaces, a conference room, and a single, unfurnished space that was not in use at the
time measurements were made for this study. The reader is referred to the original document
(Saarela and Mattinen, 1993) for a more detailed description of the building.
This study was conducted to establish the types and quantities of airborne chemicals in a new
office building and the influence of ventilation on the indoor concentrations of these chemicals.
Air samples were taken on both floors of the new office building in the center of the open
spaces at a height of 1.5 meters. The organic samples were collected in glass tubes containing
150 mg of Tenax TA absorbent. Chemical compounds were identified with a Jeol SX-102
mass spectrometer. Of the identified compounds, the most abundant were chosen for
quantitative followup done with an HP 5890 Series II GC equipped with a flame ionization
detector (FID). The total VOC values were calculated from the FID-detector response in
toluene equivalents. Aldehydes were measured from one site in the building under both
ventilation conditions and were collected in impinger flasks containing dinitrophenylhydrazine
(DNPII) dissolved in 2-M HCL and analyzed with high-performance liquid chromatography
(HPLC).
The measurements taken for an 8-month period in the ventilated building displayed a random
variation of individual organics. Table B-4 lists the most abundant organics selected for
quantitative determination and their maximum concentrations with ventilation on. In this
study, the maximum concentration of formaldehyde reached 0.122 mg/m3 with no ventilation;
however, the formaldehyde concentration was 0.059 mg/m3 at week 8 of the 8-month study
with ventilation. In both ventilated and unventilatcd buildings, aldehydes show an upward
trend in the beginning (Saarela and Mattinen, 1993). In the unventilatcd building, the trend
for most of the identified aldehydes was initially rising concentrations, but concentrations
began to decline after 4 to 8 weeks. In the ventilated building, the aldehydes show an upward
trend on weeks four to eight. The authors concluded that new materials produced high
emissions levels and that, with effective ventilation, the resulting concentrations could be
reduced to levels that would not cause discomfort or adverse health effects.
B.4 Meyer Study
The author of this study used data from other sources to compile a table of construction
materials and consumer products (Meyer, 1983). The range of formaldehyde emissions rates
for seven composite wood products, extracted from a list of the 28 construction materials and
consumer products, is provided in Table B-5. Recorded emissions rates of these products
B-4
-------
Table B-4. Concentration Measurements of Organics in a Ventilated Building8
Chemical Compound
Maximum
Concentration (mg/m3)
Toluene
0.121
Formaldehyde
0.059
Octane
0.011
Ethylbenzene
0.007
m+p-Xylene
0.013
Styrene
0.008
o-Xylene
0.011
a-Pinene
0.004
1,2,4-Trimethylbenzene
0.002
Decane
0.003
Undecane
0.004
Dodecane
0.002
Excerpted from: Saarela and Mattinen, 1993.
a Measurements depict the average and maximum concentrations during an 8-month period.
Table B-5. Range of Formaldehyde Emissions Rates for Seven
Composite Wood Products
Material
Emissions Rate
(mg/m2-day)
Reference
Plywood
UF-bonded
1-34
a
Phenolic
0-0.05
b
Hardwood paneling (UF)
1-34
b
Particle board
Standard (IJF)
2-34
b
Low emissions (UF)
0.5-3
b
Phenolic
0-0.001
b
Plywood
0.055
a
Excerpted from: Meyer, 1983.
a Pickrell et al., 1982.
b Meyer, 1979; Meyer et al., 1980; see also Spedding and Edmondson, 1980; Lehmann, 1982; and Sundin. 1982.
B-5
-------
were based on a few samples and on a small number of measurements. Therefore, they represent a
range rather than average emissions rates for the entire class of materials. The literature from
which the author obtained the data is also listed.
B.5 Sundin Study
In this study, researchers measured organic emissions including formaldehyde from several
commonly used building materials (Sundin et al, 1992). The materials evaluated included planed
pine lumber, coated and raw particleboard of different ages, medium-density fiberboard (MDF),
hardboard, plywood, and gypsum board. Dried, unresinated sawdust and shavings were also
included.
An active sampling method was used to collect samples. A special device was fabricated to
extract air samples from the surface of the materials analyzed. This device included a cover
consisting of a desiccator lid attached to a pump by a tube containing the adsorbent inline and a
glass plate for the bottom. Each sample was enclosed in the device, and air was extracted from the
surfaces of the materials and adsorbed on charcoal and Tenax tubes. The adsorbed organic
compounds were analyzed using a GC/MS. The emissions of formaldehyde were evaluated
separately at the Casco Nobel research laboratory in Sundsvall, Sweden. Independent of the
analysis performed at the Casco Nobel Research Laboratory, formaldehyde was also measured
with the InterScan Portable Formaldehyde analyzer 1163.
Table B-6 lists TVOC and formaldehvde concentrations emitted from the tested materials. Table
~
B-7 contains the concentration percentages of aldehydes and terpenes from the TVOC
concentrations. The authors of this study concluded that:
• The identified VOCs emitted from building materials showed good agreement with
the published findings of Nelms et al. (1986).
• The decay process for VOCs emitted from particleboard takes place very rapidly
with a half-life of 3 months or less.
• Particleboard coated with trtelamine film has relatively high VOC emissions
immediately after pressing due to the organic emitters present in the film.
However, measurements after 1 month showed that VOC concentration levels are
similar to other investigated materials.
• Ordinary pine lumber has outstandingly high VOC emissions concentration
(920 ng/m3) among the materials studied. The emissions consists largely of
terpenes. The decay process has not been studied.
B-6
-------
Table B-6. VOC and Formaldehyde Emissions from Various Engineered Woods
VOC
"ormald
ehyde
Type of Material
Total
InterScan"
Chamber15
mg/m2-h
mg/m3
mg/m2-h
mg/m3
mg/m3
Ilardboard, standard 3.2 mm
0.03
0.09
0.03
0.02
0.03
Hardboard, oil-treated 3.2 mm
0.03
0.08
0.03
0.02
0.02
Plywood, UF, 10 mm
0.05
0.15
0.03
0.02
0.05
MDF, UF, 16 mm
0.04
0.11
0.79
0.50
0.97
Lumber, planed, pine 170x25 mm
0.31
0.92
0.03
0.02
—
Particleboard, 2 weeks old, 16 mm
0.15
0.43
0.05
0.03
—
Sawdust & shavings, dried, unresinated
0.06
0.19
0.04
0.02
—
Particleboard, 10 years old, 10 mm
0.04
0.11
0.13
0.08
—
Melamine film-coated PB, 14 mm
0.04
0.11
—
—
—
Melamine PB without film, 14 mm
0.03
0.09
--
—
—
Melamine-UF film-coated PB, 14 mm
0.04
0.16
—
—
—
Melamine-UF PB without film, 14 mm
0.02
0.05
—
-
Excerpted from: Sundin et al., 1992.
a InterScan is a portable formaldehyde analyzer.
b Chamber concentrations were measured at the Casco laboratory.
Table B-7. Percent of Higher Aldehydes (Ilexanal) and Terpenes of Total VOC
Total
VOC
P
art of Tota
tl VOC
Type of Material
Aldehydes
Terpcne
Misc.
Total
/j.g/m2-h
(%)
(%)
(%)
(%)
Hardboard, standard 3.2 mm
90
4
96
0
100
Hardboard, oil-treated 3.2 mm
80
6
94
0
100
Plywood, UF, 10 mm
150
8
25
67
100
MDF, UF, 16 mm
110
8
18
74
100
Lumber, planed, pine
920
1
81
18
100
Particleboard, UF 2 weeks old, 16 mm
430
32
22
46
100
Sawdust and shavings (pine, spruce)
190
27
20
53
100
Particleboard, UF 10 years old, 10 mm
110
6
19
75
100
Excerpted from: Sundin et al., 1992.
B-7
-------
B.6 National Particleboard Association Study
A two-part preliminary study was conducted for the National Particleboard Association (NPA)
to analyze product emissions from particleboard and medium-density fiberboard materials.
The test objectives for the two laboratories participating in this two-part study were different
with subsequent differences in the type of data that were reported. GEOMET Technologies
performed the first analysis and prepared a report on emissions results from chamber analysis.
The objective of the GEOMET study was to determine emissions from composite wood
products from a target list of 26 VOCs (Table B-8). GEOMET evaluated TVOC, individual
volatile organic compounds (IVOC), and formaldehyde emissions from 18 particleboard and
fiberboard products. No data were provided for VOCs that were not on the target list.
Table B-8. NPA Target Compounds
Acetaldehyde
Hexanal
Acetone
Limonene
a-Pinene
m & p-Xylenes
Benzaldehyde
1-Nonanal
Benzene
Octanal
P-Pinene
o-Xylenes
Borncol
p-Cymene
Camphene
Pentanal
Ethylbenzene
Pentane
Heptanal
1-Pentanol
Heptane
Phenol
1-Heptanol
Toluene
2-Heptanone
trans-2-Octanal
Source: GEOMET, 1994.
B-8
-------
Georgia Tech Research Institute (GTRI) performed the second analysis of a subset of
the materials evaluated by GEOMET. The primary objective of the GTRI work was to
determine if a-pinene and other terpenes were being emitted from composite wood products
that were not detected in the GEOMET study. The GTRI evaluation reported all VOCs
emitted from the composite wood products detected in the chamber air regardless of the source
of the chemical. A major concern of the NPA with the GTRI data was the identification of
unexpected VOCs detected during emissions tests. These unexpected VOCs included freons,
chloroform, 1,1,1-trichloroethane, and benzene (Table B-9). According to GTRI, these
unexpected VOCs resulted from contamination of the composite wood samples from sample
containers. For simplicity, individual analyses comprising the NPA's two-part study are
presented separately below.
Table B-9. Sample Container Contaminants
Acetone
Longifolene
Benzene
Pentanes
Bromod ichloromcthane
Thujone
Chloroform
Thujopsene
Geraniol
Toluene
Isoprene
T richlorofluoromethane
Source: Personal communication, Mike Iloag, National Particleboard Association, November 16, 1993.
B.6.1 Part I-GEOMET
GEOMET specimens consisted of both finished and unfinished particleboard and MDF.
Table B-10 summarizes these product test specimens. Test specimens, selected and prepared
for evaluation by NPA, were delivered to GEOMET. Then, each specimen, with an area
consisting of 0.0105 m2 per side and a loading factor of 0.41 m2/m\ was conditioned for a
5-day period in the 0.052-m1 chamber. Following the conditioning period, separate air
samples were collected at 24 hours for TVOC, at 24¥z hours for IVOC, and at 25 hours for
formaldehyde. This sample pattern was chosen because samples could not be taken
simultaneously due to the low chamber flow rate (-0.87 L/min'). TVOC and IVOC air
samples, which were collected in sorbent tubes, were analyzed with a GC. IVOC analysis
included identification and quantitation of 26 target compounds. Formaldehyde concentrations
for each specimen were determined using the Chromatropic Acid method (NIOSII 3500) and
are listed in Table B-l 1. The detection limit for each of the specimens was 13.9 /zg/m3, except
for Specimen 1A for which the detection limit was 41.7 fig/m3.
B - 9
-------
Table B-10. Test Specimens
Specimen ID
Substrate
Unfinished PB 1A
Southern Yellow Pine
Unfinished PB IB
Southern Yellow Pine
Unfinished PB 2A
Western Fir
Unfinished PB 2B
Western Fir
Unfinished PB 3A
Hardwood
Unfinished PB 3B
Hardwood
Unfinished MDF 4A
Medium-Density Fiberboard
Unfinished MDF 4B
Medium-Density Fiberboard
Finished PB 5A
Southern Yellow Pine
Finished PB 5B
Southern Yellow Pine
Finished PB 6A
Southern Yellow Pine
Finished PB 6B
Southern Yellow Pine
Finished PB 7A
Southern Yellow Pine
Finished PB 7B
Southern Yellow Pine
Finished PB 8A-1
Southern Yellow Pine
Finished PB 8A-2
Southern Yellow Pine
Finished MDF 9A
Medium-Density Fiberboard
Finished MDF 9B
Medium-Density Fiberboard
Source: GEOMET, 1994.
Specimens 1A through 4B were composed of various wood substrates without a finished
surface. Specimens 5A through 9B were covered with various finishes using four different
adhesives. Specimens 1A and IB and Specimens 5A through 8B were all composed of
southern yellow pine (SYP). A direct comparison was made between the TVOC concentration
levels of the unfinished SYP substrate, Specimens 1A and IB, to the finished SYP substrates,
Specimens 5A through 8B. Ail of the finished specimens had lower TVOC concentration
levels than the unfinished specimens, with the exception of the 120-hour analysis for Specimen
8A-2. Concentrations of the identified targeted organic compounds for the nine specimen
pairs are shown graphically for specimens analyzed at 24 and 120 hours in Figure B-l. The
reader is referred to the individual study for any additional information (GEOMET, 1994).
B - 10
-------
3000
¦ 24-hours
~ 120-hours
2500
2000
1500
1000
500
0
-------
Table B-ll. Formaldehyde Emissions Data from Engineered Wood Products
Specimen
Concentration, /*g/m3
24 Hours
120 Hours
Unfinished PB (1A)
603
421
Unfinished PB (IB)
123
123
Unfinished PB (2A)
99
132
Unfinished PB (2B)
809
375
Unfinished PB (3A)
120
95
Unfinished PB (3B)
246
153
Unfinished MDF (4A)
183
175
Unfinished MDF (4B)
110
129
Finished PB (5A)
116
54
Finished PB (5B)
127
79
Finished PB (6A)
65
62
Finished PB (6B)
9
26
Finished PB (7A)
364
408
Finished PB (7B)
719
371
Finished PB (8A-1)
14
28
Finished PB (8A-2)
21
21
Finished MDF (9A)
140
153
Finished MDF (9B)
246
89
Compiled from: GEOMET, 1994.
B.6.2 Part II - GTRI
GTRI also prepared a report on emissions results from chamber analyses. GTRI evaluated
TVOC emissions from 10 product specimens consisting of eight urea-formaldehyde bonded
particleboards and two medium-density fiberboards. Compounds were identified and
quantitated by comparisons to authentic standards, and basic emissions factors were calculated
using procedures outlined in the standard method given in the Carpet Policy Dialogue Report
(Leukroth, 1991). This method is a generally accepted method used to measure chemical
emissions from dry products. Like carpet, the composite wood products tested had flat, dry
surfaces and were allowed to offgas in a chamber.
Various organic compounds were identified. The primary organic compounds emitted were a
variety of terpenes. The analyzed substrates were urea-formaldehyde-bonded particleboards,
B - 12
-------
except for two MDF specimens. Although not specifically stated by the author, one can
assume that Specimen 6, an interior-grade MDF, was bonded with a urea-formaldehyde resin,
which is typically used to bond composite wood products designed for interior use. Using the
same logic, one can make a similar assumption that exterior-grade MDF, Specimen 7, was
bonded with a phenol-formaldehyde resin, which is moisture-resistant and more appropriate
for exterior use. Specimens 1 through 3 were analyzed for total and specified VOCs at 24
hours only. Specimen 10 emitted the highest levels of formaldehyde and TVOC. At 120
hours, the organic emissions for Specimens 9 and 10 were still increasing; however, the
formaldehyde emissions for Specimen 10 were constant at 24 and 120 hours. The
formaldehyde and acetaldehyde levels for Specimen 8 continued to increase at 120 hours while
the TVOC levels decreased. Excluding Specimen 6, the acetaldehyde emissions for all the
samples decreased significantly after 24 hours. Table B-12 lists formaldehyde emissions data
from this study, both concentration and emissions rates, for Specimens 4 through 10 taken at
24 and 120 hours for seven of the boards in the GTRI analysis. Specimens 1 through 3 were
not analyzed for formaldehyde. For additional information, the reader is referred to the
original document (GTRI, 1993).
Table B-12. Formaldehyde Emissions Data from Particleboard and MDF
Specimen
Product Line
Concentration
(/ig/m3)
Emissions Rate
(pg/mMi)
4
11/16" Industrial
27.29a
20.54"
127.52a
95.98b
5
5/8" Underlayment
50.26
41.17
237.08
194.20
6
3/4" Interior (MDF)
9.66
13.75
44.72
63.66
7
3/4" Exterior (MDF)
3.10
3.29
14.49
15.37
8
5/8" Red Oakprint
18.78
19.60
88.58
92.45
9
5/8" White Oakprint
c
7.92
c
38.08
10
5/8" Particleboard
59.21
59.00
279.29
278.30
Excerpted from: GTRI, 1993.
a Sample taken at 24 hours for all top numbers in column.
L' Sample taken at 120 hours for all bottom numbers in column.
f No data for sample.
B - 13
-------
Emissions of VOCs from Specimens 4 through 8 were lower at the 120-hour evaluation than they
were at the 24-hour evaluation. However, VOC emissions from Specimens 9 and 10 increased at
120 hours. Figure B-2 graphically depicts TVOC emissions rates from each specimen. The
reader should note that, due to contamination concerns, compounds identified as sample
container contaminates (Table B-9) have been deleted from the TVOC emissions totals depicted
in Figure B-2.
B.7 Thegersen Study
A Scandinavian study (Thogersen et al., 1993), showed that water-damaged chipboards result in
substandard air quality due to increased formaldehyde emissions and microbiological growth.
Nine chipboard specimens of identical dimensions (200 x 800 x 13 mnr) with untreated edges
were placed in chambers, three specimens per chamber. The first specimen set was
used as a reference specimen and was kept at constant standard conditions (21 ± 1 °C, 30 ± 15
percent RH, 15=1 m/s air velocity, and 0.80 ± 0.05 L/s air supply) established by the
researchers for the duration of the experiment . The second specimen set was submerged in a
water bath for 24 hours and then placed in a special climate chamber without air change,
maintaining 80 percent humidity for 6 days. Following simulated water damage and
conditioning, specimen set 2 was placed in its chamber. Specimen set 3 was subjected to the
same simulated water-damage process as specimen set 2. However, following the simulated
water-damaging process, specimen set 3 was placed into its chamber for bakeout at a temperature
(35 ± 1 °C) much higher than the reference temperature (percent RH, air velocity, and air supply
were the same as the reference specimen).
Study investigators reported that, for the reference specimen, Specimen 1, the formaldehyde
concentration decayed exponentially until it flattened to a steady state after 11 days. At this
steady-state, the chamber concentration of formaldehyde for the reference specimen was about
0.01 mg/m3. Investigators stated that, during the decay phase, chemical substances diffused to
the surface and evaporated; after emissions were stabilized, subsequent decomposition
determined the emissions rate. The water-damaged specimens, both Specimens 2 and 3, which
were wetted with water, had high emissions rates immediately following return to their
respective chambers. The researchers assumed that the high emissions rates for water-damaged
Specimens 2 and 3 were due to:
• Increased diffusion speed produced by higher water content
• Organic compounds transported into the air along with water evaporation
• Glue dissolution promoted by higher water content.
In spite of these high emissions rates, concentration levels for Specimens 2 and 3 decreased more
rapidly than the concentration levels of the reference specimen. The rapid concentration decay
was attributed to the lower total initial amount of formaldehyde present in the specimens after
water damage. Researchers believed that some formaldehyde had been released during the
simulated water-damage treatment. They also postulated that water-damage treatment caused
B - 14
-------
35000
30000
25000
20000
15000
10000
5000
1
124-hours
~ 120-hours
5 6
Specimen
8
10
Figure B-2. TVOC emissions rates (GTRI, 1993).
-------
considerable washout, which rarely occurs during a more gradual moistening, as occurs by
diffusion or dripping in a real building. The bakeout specimen, Specimen 3, produced increased
emissions during its bakeout period, but emissions soon steadied at a low rate. Researchers
concluded that bakeout conditions should be maintained for several days to achieve a significant
formaldehyde reduction to promote sufficient lasting effects.
When emissions were fairly steady from t - 400 h to t = 500 h, the mean chamber concentration
of formaldehyde for the water-damaged Specimen 2 was about 0.0475 mg/m3. The mean
formaldehyde concentration for Specimen 3, the water-damage specimen that underwent
bakeout, was about 0.02 mg/m3. Comparing the formaldehyde concentrations for all three
specimens at steady-state, the water-damaged specimen that did not experience bakeout had a
formaldehyde concentration five times higher than the reference specimen. The water-damaged
specimen that did experience bakeout appeared to have a formaldehyde concentration level twice
that of the reference specimen (0.02 mg/m3). Investigators stated that the small concentration
difference between the reference specimen (0.01 mg/m3) and the bakeout specimen (0.02 mg/m3)
could not be interpreted because both concentrations were close to the detection limit of the
measuring instrument (0.06 mg/m3).
The authors discussed only formaldehyde concentrations in this paper; however, they did state
that TVOC concentrations conformed to curves similar to those describing formaldehyde
concentration. Researchers concluded that water-damaged chipboards result in increased
emissions of formaldehyde and other organic compounds, even when the chipboards dried, and
that water-damaged chipboards result in less acceptable air quality when the chipboards are wet.
Microbiological samples were also collected and investigated as part of this study.
Microbiological samples were collected by pressing agar plates to the chipboard specimens, and
samples were then cultivated in a laboratory. The microbiological evaluation revealed
significant growth of micro fungi after water damage. Ten species were identified, with the
penicillium species dominating. Spores began growing when humidity was introduced, and the
specimen contained spores even after drying. Following bakeout, spores were able to grow. The
researchers added that, in a building, these spores can float through the air and adhere to material
surfaces where they can grow under favorable conditions. Once established, these
microorganisms are very difficult to eliminate. Total removal of infected materials is the only
remedy when this type of growth occurs.
B - 16
------- |