JIIIIIIHWIIHHUI

—
=
—
—
—
—
=
—
=
=

 

was uulfliflfl'flijflflfilill‘filill’lilljlfl’l‘flifl‘lfill

01 91

 

 

This is to certify that the

thesis entitled

RECONSTITUTED PARTICLEBOARD FRW CCA TREATED RED PINE

presented by

JACOB MARCELLO MUNSON

has been accepted towards fulfillment
of the requirements for

 

 

Masters degree in FORESTRY

BVMW

Major professor

 

Date June 10. 1997

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY

Michigan State
University

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINE return on or before date due.

 

MTE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/” WM“

RECONSTITUTED PARTICLEBOARDS FROM CCA TREATED RED PINE
By

Jacob Marcello Munson

A THESIS
Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Forestry

1997

ABSTRACT
RECONSTITUTED PARTICLEBOARDS FROM CCA TREATED RED PINE
By

Jacob Marcello Munson

Large quantities of pressure treated wood will be coming out of service in the near
future. An attempt to alleviate the amount going to landfill was the focus of this thesis.

Two studies on the recycling of CCA (chromated copper arsenate) treated wood
into wood composites were completed. The first objective was to determine the
feasibility and proper amounts of resin to manufacture recycled composites. The second
objective was to evaluate the effects of the treated wood proportion in the composite on
physical and mechanical properties.

Results from the first study showed that 4 and 8 percent solids were suitable
quantities of resin to manufacture wood composites from CCA treated wood, and there
was no significant biological decay when using CCA treated wood in the boards. From
the second study, it was found that using up to 50 percent treated wood did not
significantly reduce the board physical and mechanical properties.

Overall, the objectives of the studies were met, and it was feasible to manufacture

particleboards utilizing CCA treated wood.

To my late grandfather, Paul C. Munson.

iii

ACKNOWLEDGMENTS

A publication such as this would not be possible without the assistance of my
major professor, Dr. Pascal Kamdem. His patience and understanding while I learned
about wood composites and preservation have been invaluable. Hopefully some day I
will be as professional, ethical, and scientific as he. Dr. Kamdem has not only been a
great mentor, but a fiiend with which I could personally talk to. This relationship is one
that will last a lifetime.

I would also like to thank Miss Renee Essenmacher for her help during pressing
and evaluation of the particleboards. This thesis would have been much more difficult to
produce in the time allotted without her help.

Also, gratitude goes to Georgia-Pacific and Weyerhaueser Corporations for
donating the resins used, Hydrolake Leasing Company for the red pine utility poles, and
Universal Forest Products for use of their equipment.

Finally, I would like to thank my wife, Heidi Munson. Heidi has given me the
insight, determination, and courage to make this thesis possible. She has been absolutely

terrific through both the good times and the tough.

iv

TABLE OF CONTENTS

LIST OF TABLES .
LIST OF FIGURES .
INTRODUCTION

CHAPTER 1
RATIONALE AND SIGNIFICANCE .
Problem Analysis
Study topic
Problem statement
Research .
Hypotheses and assumptions .
Literature Review
Overall Objectives

CHAPTER 2

STUDY I: FEASIBILITY AND RESIN CONTENT DETERMINATION .

Objective
Materials and Methods.
Materials .
Particle Manufacture Methods
Composite Manufacture Methods
Material Testing Methods.
Chemical Analysis and Water Leaching
Biological Performance. . .
Results and Discussion .
Particle Size Analysis .
Particle pH. . .
Chemical Analysis and Water Leaching.
Mechanical and Physical Properties.
Urea formaldehyde particleboards. .
Phenol formaldehyde particleboards.
Decay Resistance. . .
Urea formaldehyde particleboards .

vii

viii

HQQMMNNN

O

11
11
11
ll
12
12
14
15
16
16
16
21
21
23
24
25
27
27

Phenol formaldehyde particleboards

CHAPTER 3
STUDY II: EVALUATION OF PARTICLEBOARDS MADE WITH
DIFFERENT TREATED WOOD PROPORTIONS
Objective
Materials and Methods.
Materials .
Particle Manufacture Methods
Composite Manufacture Methods
Material Testing Methods .
Leaching Test Method...
Determination of CCA Retention 1n Particleboard.
Results and Discussion .
Particle Size Analysis .
Particle pH. .
Mechanical and Physical Properties.
Effect of Resin Content . . .
Effect of CCA-treated Wood Content .
Statistical Analysis .
Determination of CCA Retention 1n Particleboard.
Leaching Test Results. . .
Comparison to ANSI Requirements.
Comparison Between Study I and Study 11
CHAPTER 4
CONCLUSIONS .

CHAPTER 5
RECOMMENDATIONS .

APPENDICES
APPENDIX A: Sample Calculations Worksheet of Particleboard
Production...
APPENDIX B. Leachate Calculations.

LIST OF REFERENCES .

vi

28

3O
3O
30
30
31
31
32
33
33
33
33
34
35
35
42
43
43
45
45
46

48

49

50
51

52

LIST OF TABLES

Table 1 - Chemical retention in particles before and after particleboard
manufacture

Table 2 - CCA leachate content from particleboards after 28 days .

Table 3 - Some ANSI requirements of medium density particleboards
Table 4 - Some mechanical and physical properties of UFR particleboards .
Table 5 - Some mechanical and physical properties of PF R particleboards .

Table 6 - Weight loss in percent of UFR particleboards exposed to fungi using
modified soil block and agar block test methods. . .

Table 7 - Weight loss in percent of PFR particleboards using a
modified soil block test.

Table 8 - Bending, IB, and physical properties of particleboards made with CCA
treated furnish: 4% PF resin . . .

Table 9 - Bending, IB, and physical properties of particleboards made with CCA
treatedfurnish: 8%PFresin . . . . . . . . . . .

Table 10 - Effect of PF resin content on particleboard properties .
Table 11 - Retention of CCA in the particleboards .
Table 12 - Leaching as a percent of the initial total oxides in the particleboards

Table 13 - Resin specifications

vii

22

23

23

26

26

28

29

36

36

37

44

45

50

LIST OF FIGURES

Figure 1 - Sample particleboard cutting pattern .

Figure 2 - Particle length distribution .

Figure 3 - Particle width distribution

Figure 4 - Particle size class distribution via Tyler sieves.

Figure 5 - MOR of particleboards made with untreated and treated particles
Figure 6 - MOE of particleboards made with untreated and treated particles

Figure 7 - IB of particleboards made with untreated and treated particles

viii

15

18

19

20

38

39

40

INTRODUCTION

This thesis sets forth a series of studies on the feasibility to produce wood
composites from CCA treated wood. The first study is done to determine the proper
quantities and what types of adhesives would be suitable to use with CCA treated wood.
The second study uses those resins and resin contents to determine the effect of CCA
treated wood proportions in the particleboards on their mechanical and physical

properties.

CHAPTER 1

RATIONALE AND SIGNIFICANCE

Problem Analysis
Study Topic

The topic of this study is to determine if particleboards can be made from
preservative treated, red pine utility poles. It is an important topic to reduce the amount
of CCA treated wood that goes to land fill each year, and to develop another option for
particleboard furnish. “F urnish” is defined as the material used to produce particleboard.
Research on this topic will advance the field by providing a method from which others
can follow to perfect a product, as well as reduce environmental distress.

The research problem area consists of wood preservation, particleboard
production, new forest products, wood science, recycling, and environmental
contamination.

Due to the increased awareness of large quantities of pressure treated wood
coming out of service, scientists have started to develop methods of reconstituting
pressure treated wood into useful products. In 1995, approximately 579 million cubic
feet of preservative treated wood was produced, of which 424 million cubic feet (73

percent) was treated with chromated copper arsenate (CCA) (American Wood Preservers'

3
Institute, 1996). Looking back on the use of waterborne preservatives, almost a half-

billion cubic feet have been produced yearly since the late 19705. In 1993 it was
estimated that over 353 thousand cubic feet of treated wood were removed fi'om service
(Cooper, 1993).

Wood products for outdoor uses such as decking, shingles, railroad ties, utility
poles, and fence posts are treated with preservatives in order to extend the service life. A
preservative is a chemical compound that is toxic to humans and microorganisms and
hazardous to the environment. The most commonly used wood preservatives are
chromated copper arsenate (CCA), ammoniacal copper zinc arsenate (ACZA), creosote,
and pentachlorophenol (Lehmann, 1969). Waterbome CCA is generally used for the
treatment of structural materials and utility poles. Creosote, a tar-oil type preservative,
and pentachlorophenol are used in railroad ties, timber bridges, and utility poles. Today,
there is a trend toward replacing creosote and pentachlorophenol with CCA because of
their high toxicity(Cooper, 1993).

It is estimated that the service life of CCA treated utility poles and crossties is
approximately 25 years, depending on the geographical location. After 20 years, utility
poles and crossties are usually removed from service. Treated lumber from construction
and demolition sites and cut-offs from treated utility poles and posts also generate treated
wood waste. Therefore, significantly greater volumes of treated wood will be available
due to the increased use of CCA.

The question now is what will we do with the large quantities of CCA treated
wood coming out of service in the future? Currently there are four main ways to dispose

of treated wood. The most popular option today is to landfill. If the projections of

4
quantities coming out of service in the near future are correct and considering that wood

is a high volume material, landfills will be overwhelmed. Landfill will become a very
expensive alternative and may also produce environmental problems. In large quantities,
treated wood may leach heavy metals into soils, streams, rivers, and groundwater. The
cost (Stalker, 1993) and the current level of environmental awareness will reduce
landfilling as an attractive option for disposal of treated wood waste (C00per, 1994;
Marer et al., 1992; Webb et al., 1994; Lehmann, 1969)).

Burning treated wood in controlled environment settings could be a viable option.
Recent literature suggests that few to negligible air quality problems were encountered
from burning treated wood at the right temperature and appropriate oxygen rate (Marer et
al., 1992; Pasek, 1995). These results were obtained from laboratory experiments.
Further pilot scale or industrial tests are needed to validate this option. Wood
preservatives absorbed in the wood matrix can be recovered by incineration (Pasek, 1992;
Pasek, 1995), biotechnology, or solvent extraction (Honda et al., 1991).

A third opportunity to dispose of treated wood would be to resaw larger poles,
pilings, timbers, and posts into smaller dimension lumber that could be retreated for
further protection (Felton et al., 1996). This option would greatly reduce the amount of
treated wood going to landfill, but at the same time reduce the amount of decay resistance
in the final product by removing the high-preservative exterior retention zones. This may
also be a good source of raw material for laminated veneer lumber (LVL), gluelam, and
parallam lumber if adhesive problems were overcome.

The feasibility of poles removed from service in Ontario and Québec for sawn

products was investigated (Chow et al., 1984). Significant volumes of high quality

5
roofing products were recovered. However, the reuse of poles and ties is dropping

because of the cost of re-manufacturing and also because incipient decay reduces
mechanical properties needed in sawn lumber.

A final attractive alternative for the discarding of treated wood would be to
convert this high-quantity resource into wood composites. If properly managed, recycled
treated wood can be a good source of fiber for engineered products such as hardboard,
fiberboard, particleboard, oriented strand board (OSB), strandboard, or cement-bonded
boards (Schmidt et al., 1994). Composites made from treated wood waste are also
expected to have an extended service life.

Problem Statement

The problem that is the focus of this study is to reduce the amount of CCA treated
wood that goes to landfill each year.
Research

This research will contribute to our understanding of how compatible CCA treated
wood and urea and phenol formaldehyde resins are. Recent studies have shown a
discrepancy that CCA treated wood was incompatible with these resins. Red pine (Pinus
Resinosa Air.) has not been used in any of these prior studies, and particleboard was not
the target product.

This research will help us understand how the amount of treated fiber affect the
physical properties (such as bending strength and internal bonding strength) of the
particleboard. This can be examined by using five ratios of treated to untreated wood
fiber and examining the physical properties.

Besides physical properties already stated, the thickness swelling and water

6
absorption of the treated particleboard may be less than that of untreated particleboard.

This would benefit the flooring areas around toilets and sinks by making the floor more
water resistant. Water is the main way that fungi travel from place to place and break
down the main components of wood fiber. By using the phenol formaldehyde resin,
which is water resistant, and a treated wood fiber would create a decay resistant product.
This could be proven by the use of an accelerated decay test, where the main types of
fimgi are grown with the particleboard as a food source. The particleboard is weighed
before and after inoculation with fungi to determine the weight loss, or the amount the
fungi consumed. This would help us to learn more about the threshold of CCA
concentration needed to stop decay.

Creating such a product would benefit many user groups. The wood preserving
industry would benefit from such a product by the extension of service life of a treated
product, thus alleviating pressure to find ways to dispose their product. Particleboard
producers would gain an inexpensive, new furnish that would reduce the costs of
harvesting trees and the environmental stresses of over-harvesting. In the same time, less
trees will be harvested and less landfill needed considering that 40 percent of landfill
volume is occupied by wood products. A decay-resistant composite product could be
used for flooring around sinks, toilets, and showers, as well as manufactured housing and
trailer homes. On a global level, Earth would ecologically benefit by decreasing the
heavy metal leaching in high quantities of treated wood around landfills. A reconstituted
board would reduce this problem drastically.

Hypotheses and Assumptions

Assumptions made while conducting this research include: 1. Red pine will have

7
the same characteristics as southern yellow pines (Pinus spp.) or ponderosa pine (Pinus

ponderosa). Other studies were done on southern yellow and ponderosa and showed they
may not behave the same way. 2. It will be feasible that the particles of treated red pine
and untreated red pine will be mixed together in different ratios and still make a standard
particleboard. Hopefully no incompatibility between furnishes will be present. 3. It is
also assumed that the treated poles that will be ground will produce a consistent amount
of CCA oxides. It is most likely that each pole will be treated a little differently, but after
grinding, the amount of CCA present in the fiber will be almost homogeneous
throughout.
Literature Review

Recycling treated wood into composites is not a new concept but very little data
about the feasibility and properties of the final product are available in the literature.

There has been no actual research done on particleboards made with recycled
furnish. Also, little has been done on red pine or CCA treated red pine. At this time,
recycling of forest products has not received the attention that it should. The only
research that has been conducted on making particleboards is based on an objective of
creating a deterioration free product with virgin materials. For example, freshly cut aspen
(Populus tremuloides) was ground into chips, and then the chips were treated and then
pressed into particleboards (Boggio et. al, 1982). Two types of phenolic resins were used
and physical properties of the boards were compared. All of the resin-preservative
combinations exceeded minimum standards set by the American National Standards
Institute (ANSI, 1993) for modulus of rupture and modulus of elasticity (MOR and

MOE) and for internal bond (IB). When compared to untreated control samples, the CCA

8
treated waferboards produced lower strength properties. This study confirmed that CCA

definitely reduces mechanical properties, but not to the extent of jeopardizing the grade of
particleboard.

Hall confirmed this with treated aspen wafers in 1982, except he found that the
MOR dropped by 59% when using CCA treated wood as a furnish. The method of
treatment was different. In this case the wafers were treated with CCA at the same time
of spraying with resin, compared to Boggio's industrial treatment before resin application.
A positive aspect of Hall's work was that the amount of irreversible thickness swell
difference between treated and untreated was under 1%.

A third paper (Schmidt et al., 1983) showed that when boards were made with
aspen wafers and then preservative treated with CCA, there was no loss of strength after
accelerated aging. This backs Boggio's findings that there was little loss in strength, but
was the converse of what Hall found. There were findings that were argumentative
between these three authors. The main question between them was, "is CCA affecting
the bonding between wood and resin?"

To answer this question many studies were done by one scientist (Vick, 1990).
Vick used electron microscopy to evaluate a compatibility problem between CCA and PF
resin. He found that the chemistry of the broken wood surface was mostly made up of
lignocellulose, a primary bonding site for PF resin. This surface had been oxidized with
the CCA chemical, causing the lignocellulose to be covered and weakened. He thought
that this was the main reason why there was so much difficulty getting consistent results.

Several other studies have suggested that mechanical properties of wood based

composites made from CCA treated particles were lower than those of virgin particles

9
(Gertjejansen et al., 1988; Vick, 1980; Vick etal., 1990; Vick et al., 1996). The reduction

in properties was attributed to the surface modification by the preservative treatment or
the interaction between preservatives and resin. All of these studies reporting reduction
of mechanical properties involved methods of first reducing the virgin wood to particles
and then treating the particles with preservatives. The treatment, of course, creates
chemical modification on the surface of the particles which is the interface between
particles and resin during gluing and pressing. The reduction of already treated wood
into particles and reassembling into new forms using adhesives has been done and has
shown positive results.

Research on CCA treated composite production feasibility began again. It was
found that preservative treatment with CCA was not detrimental to adhesive bonding for
different wood species, and even a higher glue line shear strength was reported by
Janowiak in 1992. This means CCA treated wood would be stronger in veneering
applications.

Strength was reported higher again in CCA treated flakeboards by Kumar in 1993.
He found that chemical modification of flakes prior to gluing and pressing into boards
represent a simple, inexpensive method for improving dimensional stability and decay
resistance in CCA treated particleboard. Aspen wafers created boards that showed a
reduced thickness swell by 25-30% with CCA treatment. It also increased MOR by 40%
but did not affect the IB. This would mean that CCA treated aspen, ground and then
glued would be not only feasible, but profitable.

At the same time, a study showed the feasibility of recycling treated hardwoods

into particleboards with UP and PF resin (Suzuki, 1993). He found by doing an exterior

10
durability test, that the PF - CCA treated hardwoods were durable to last 6 years in the

elements.

One other study was done on reconstitution by using previously treated lumber
(Vick et al., 1996). Vick showed that flakeboards made of ring-cut flakes from recycled
CCA treated southern yellow pine (Pinus spp.) lumber and bonded with liquid PF resin
had property values substantially lower than those of untreated. These findings were
based upon flakes of variable size and shape, which caused a difference between
untreated and CCA treated flakes. The particle geometry is known to influence the
bending strength properties of the final product. Our project consists of grinding CCA
treated utility poles into uniform particles and used as raw material for medium density
particleboards.
Overall Objectives

The objective of the first study was to investigate the feasibility of producing
particleboards from CCA treated utility poles using both urea and phenol formaldehyde
as the binders. Then, boards would be made and physical and mechanical properties
would be evaluated.

The objective of the second study was to investigate the effect of increasing the
CCA treated wood proportion in particleboards on some mechanical and physical
properties. Special care was taken to ensure particle size was the same for both virgin
and treated particles. Based on the first study's findings, the optimal resin type and

content were used, boards produced, and properties evaluated.

CHAPTER 2

STUDY 1: FEASIBILITY AND RESIN CONTENT DETERMINATION

Objective

This study had two objectives. The first objective was to investigate the feasibility
of producing particleboards from CCA treated utility poles using urea and phenol
formaldehyde resins as bonding agents. The second objective was to evaluate physical
and mechanical properties of those boards.
Materials and Methods
Materials

Untreated and CCA treated red pine (Pinus resinosa Ait.) poles were obtained
from Hydrolake Leasing Service in McBain, Michigan. All poles were kiln-dried to 30
percent moisture content (MC) prior to treatment. Half the poles were then pressure
treated with a 2 percent total oxide solution of CCA-C for 6 hours with a modified full-
cell method. The treatment included an hour of initial vacuum at 91 kPa (27 inches) of
mercury followed by 4 hours of pressure at 1.03 MPa (150 psi), and a final vacuum of 1
hour. The target retention of total oxides in the poles was 9.6 kg/m3(0.60 pcf ). The

treated poles were then air-dried to 19 percent MC.

11

1 2
Particle Manufacture Methods

Poles were chipped with an Morbark Eager Beaver Chipper, and chips reduced
into particles with a laboratory hammermill. Particles from untreated red pine were
named virgin furnish and particles from CCA treated red pine named treated furnish.

Attention was paid to particle size to avoid an imbalance which could create panel
warping and strength reduction. Particles were sifted by size with a vibrating inclined
screen, and only particles passing through 10 but held by 16 mesh screens were selected
for particleboard production. A screen analysis was performed on a sample of both
particle types with an electric shaker for 5 minutes using screen 8, 10, 16, and 30 mesh.
The thickness, width, and length of 1000 screened particles were also measured by light
microscope. Screened particles were air dried to 5 i 2 percent MC and used to
manufacture particleboards.

The pH of both furnishes was determined with a pH meter. In a beaker, 10 g of
screened particles were mixed in 100 ml of distilled water for 30 minutes using a
sonicated bath and the pH determined.

Composite Maflfiwmre Methods

Commercial liquid urea formaldehyde resin (UFR) and phenol formaldehyde resin
(PFR) provided by Georgia Pacific, each containing around 50 d: 1 percent solids, were
used in this study. Screened and dried particles were sprayed with 2.5, 4, and 6 percent
resin solids content based on oven-dry weight of the furnish. A sample board calculation
is shown in Appendix A.

Several studies reported reduced mechanical properties of wood composites made

of CCA treated wood (Vicks, 1990; Boggio and Gertjejansen, 1982) due to the

13
interference of the preservatives with the adhesion of hot or cold press adhesives. The

weak adhesion is also explained by the reduction of active hydroxyl or carboxylic groups
on the wood surface. The pH, surface energy, and surface tension of wood modified after
preservative treatment may also contribute to the weak adhesion of treated wood. Several
options have been proposed to improve adhesion of CCA treated wood. The most
promising alternatives are: the improvement of adhesive formulation to increase the
mechanical interlocking by a deep penetration of adhesive in wood (Vick, 1990), the
treatment of wood with a surface modifier such as sodium hydroxide (Vick, 1980) which
increases wood surface reactivity with adhesives, or the use of a relatively high amount of
resin (Moslemi, 1974). Different amounts of resin were used to produce boards in order
to evaluate their effect on the properties of reconstituted boards.

Particles were sprayed with resin in a laboratory rotary drum blender for 5
minutes and mats were hand-formed in a 40.6 cm square (16" square) frame. The MC of
the particles in the mat before pressing was 8 i 2 percent. A Berthelsen thermo-oil
heated hydraulic press was used to press the mat between two steel platens to a nominal
thickness of 10 mm (0.375"). The time interval from the application of resin to pressing
was kept constant at 5 minutes in all trials. The press time was 6 minutes, the pressure
800 psi (8.28 MPa), the press temperature 190°C (325°F), and the closing time 13
seconds.

Low density areas on each board were removed by trimming one inch on all
edges. Trimmed boards were kept at 65 percent relative humidity (RH) and 20°C (68°F )

for at least 40 days before testing or until they reached their equilibrium moisture content

14
(EMC). The EMC and density of boards were 5i] percent and 750450 kg/m3(47:1:3 pcf ),

respectively.

Boards with untreated and CCA treated red pine were manufactured containing
2.5, 4, or 6 percent by weight UFR or PFR. Ten boards of each type were made, for a
total of 120 boards. The manufacture sequence of boards was designed such that all the
replicates of each type were made consecutively, due to some problems associated with
the cleaning of the rotary drum blender.
Material Testing Methods

Samples were cut from each board to conform with American Society for Testing
Materials (ASTM) Standard number D103 7-95 guidelines to obtain: 2 specimens for
bending, 3 for thickness swelling (TS) and water absorption (WA), 6 for internal bond
(1B), and five 2.5 by 2.5 cm (1" x 1") strips for soil block (See Figure 1). Samples were
stored in a room conditioned at 65 i 1 percent RH and temperature of 20 :t 3° C (68 i 6°
F) until tested. ASTM Standard 1037-95 tests for static bending and IB were all
conducted using an Instron testing machine. TS and WA were conducted by ASTM

Standard 1037 as well.

15

 

 

 

 

 

 

 

 

 

113
"NS

113

BENDING
113
BENDING

113 T13

113

113 m

 

 

 

 

 

 

 

 

Figure 1 - Sample particleboard cutting pattern

Chemical AnjaLvsis and Water Leaching

Three samples measuring 10.2 by 10.2 by 1 cm (4" x 4" x 0.375") with a
combined surface area of 720 cm2 (111 inz) and a total of 200 g (0.45 lb) oven dry
weight were completely immersed in 500 ml distilled water for one month. About 15 m1
of water was sampled every 48 hours and analyzed for copper, chromium and arsenic
content using atomic absorption spectroscopy (AAS). Leached and unleached solid
samples were acid digested and their metal content determined. The detection limit of the

AAS used was less than 100 ppb for copper and chromium, and about 10 ppm for arsenic

16
(AWPA, 1996). All samples with less than 10 ppm arsenic were spiked with a 20 ppm

known standard, this allows us to measure 1 ppm arsenic concentration.
Biological Performance

A modified AWPA protocol (E10-9l) laboratory soil block test was conducted
described below ( Kamdem, 1995). Brown fungi Gloeophylleum trabeum (Pers. Ex Fr.)
Murr.(Madison 617 ATCC 11539) and Poria placenta (Fr.) Cooke(Madison 698, ATCC
11538), and white rot fungi Trametes versicolor (L. Ex FR.) Pilat (R-105), Irpex lacteus
Fries (PP-105915), and Pleurotus ostreatus(.1acq. Ex Fr.) Kummer(ATCC 32237) were
used. Specimens used for the bioefficacy measured 2.5 by 2.5 by 1 cm (1" x 1" x 0.375").
Culture boxes were incubated until the aspen feeder strip was covered by fungus.
Specimens were placed on the fungus covered feeder strip and kept for 16 weeks at 90
percent RH and 25°C (77° F). After 16 weeks, specimens were removed from culture
boxes, scraped clean to remove superficial mycelium and reconditioned at 65 percent RH
and 20°C (68° F) until they reach their EMC and their weight stabilized. Weight loss after
exposure to pure culture of a test fungus was used as the index of decay.
Results and Discussion
Particle Size Analysis

Particle length, width, and size distribution are summarized in Figures 2,3, and 4,
respectively. About 80 :1: 5 percent of the thousand particles analyzed by light
microscopy had an average length of 3.6 i 1.0 mm (0.14 i 0.04") and an average width
of 1.5 i 0.3 mm (0.06 i 0.01"). The average slendemess ratio was 2.64: 1, which is
defined as the ratio of length to the diameter of the particle. From the Tyler sieve

analysis, Figure 4 depicts particle size ranged from 0.5 to 2.3 mm (0.02 to 0.09"). This

17
shows the same distribution for both treated and untreated furnish by weight. Eighty-

seven percent of all particles were between 1.52 and 1.78 cm (0.06 and 0.07") in size.
This data is in agreement with microscopy measurements based on distribution and
standard deviation. Any differences between microscopy and Tyler sieve data could be
explained by the many angles at which particles could pass through the Tyler sieves (0 -

180 degrees).

   

 

l---TA .wrgmh

Qt

T—l

 

Ii1IiIllllllmilIllIl1111111IIIIIIIIIIIIi1IllllllllilIIIIl111mlIll11111"!11111111111111

    

107

 

[llIlllllllllllllllllllllllllIll1111111111llllllll[Illlll111111lllllllllllIlllllllllllll|ll|!illllllllllllIllllllllllllllllllllllll

L Particle length (mm)

 

Figure 2 - Particle length distribution

 

 

        
   

Ill

lllllllll?

I

lllllllllllllllllllfl

19

l

l Virgin
Recycled ll

   

 

 

 

  

_ ,m EE::2:__m____§:=__=22E

 

:

 

:______=___:_:_:_:

 

 

    
   

   

 

:5 g: M

  

.______2::=3_____.__.___________.__z_.___,____‘_=_______=______w______._______.._._:§._:._:=WEiW75EgE:_gEE_=____=_____:__:_

 

E: =__=_a_________=_=_=__=___w_____=_______________=____________________:__=________=___________E;__v_=_______________a:,___:__.=_____==_3.5..Eis:_.__=_____§__=_

 
      

 

___=__.__s

 

 

 

. W .EA___=1._______A___________==_=_é_§%§___£___:§__E
5 4 4 3 3 2 2 1 1
3.; 55:60..”—

 

 

2=2:.__,_____._________m_________________:__:______22.52%:E:

                

Particle width (mm)

 

Figure 3 - Particle width distribution

20

 

 

 

 

 

.h
0
l

 

 

 

 

 

 

 

 

Frequency (%)

N
O
l

10

OJ
0

 

 

 

111

1

1111111111111

—l
%l
E
E
E
%
E
E
E
E
E
g
E
E
E
E:
g
E
=
E
E
g
E
E

111111111111

111511511111

lllllllli

.—_—-
=
-—w—

 

 

 

 

1.0 1.2 1.4 1.7

.0
01

 

IVirgin

g Recycled ~

 

l

 

 

 

 

 

_‘-—'—_—_'
.____
r:
a
E
E
=
.___
.__.
E—
5
5‘5
E
E
E
E
=
=
E
=
'—.—"“—.
E
E
E
__.
E;
E.
E
._t—_—“——__:
.=__.._:
E
E
=2
E
2
.__._ .
E
E
E
:2
2:
52
E
g
:1
=
g
=
:7-

 

 

2.0 2.4

Particle size (mm)

j

 

 

 

Figure 4 - Particle size class distribution via Tyler sieves

 

 

 

 

21
Particle pH

The particle pH was found to be slightly different for virgin versus treated V
particles. Virgin particles had a pH of 4.9 compared to 5.1 for the CCA—treated furnish.
However, pH obtained through this method is not representative of the pH of the wood
surface since the pH meter evaluates the concentration of H+ liberated in the water
medium. Therefore the pH could be influenced by the solvent. Thomason and Pasek
have shown that pH of southern yellow pine tested in water is different than tested in
acetone (17). Knowing the low pH of CCA (pH 5 2), the similarity in pH of CCA treated
and virgin red pine suggest that wood may behave as a buffer, or the retention of CCA in
the wood analyzed was low.

Chemical Mlvsis and Water Leaching

The chemical retention in the particles from utility poles before and after board
manufacturing are listed in Table 1. Data in Table 1 indicate a negligible or undetectable
loss of copper, chromium, or arsenic during the board manufacture processing. The total
oxide retention of particles from treated poles was 7.95 kg/m3 (0.50 pet). The retention
was lower than the target retention of CCA treated red pine poles (0.60 pcf) but higher
than the retention of ground contact commodities (0.40 pot). A statistical analysis
showed no significant difference at the 5% level between the chemical content of
particles before and after the particleboard fabrication. Depletion of copper, chromium,
or arsenic during the manufacture of particleboard from CCA treated utility poles of red
pine is negligible and insignificant at a 5% level, although the reduction of poles into

particles involved high mechanical and thermal activities during the reduction of poles

22
into particles with the harnmerrnill and a high pressure of about 8.28 Mpa (1200 psi)

level and 190°C temperature is used for particleboard formation.

Table 1 - Chemical retention in particles before and after particleboard manufacture

 

 

 

 

 

CCA Retention in Particles Retention in Boards
Chemical
Kg/m3 (pct) Kg/m3 (pct)
CuO 1.41 (0.09) 1.23 (0.08)
CrO3 3.68 (0.23) 3.20 (0.20)
A520, 2.84 (0.18) 3.04 (0.19)
Total Oxide 7.93 (0.50) 7.47 (0.47)

 

 

 

 

 

Table 2 lists the amount of heavy metal leached from particleboards made of
virgin (V) and recycled (R) CCA treated furnish after a one month immersion in distilled
water. The levels of copper and chromium released during the soaking period vary with
the type of resin, the amount of resin, and the furnish used. Boards made of CCA treated
furnish, evidently leached more Cu, Cr, and As than those made of virgin furnish. Boards
made with high resin content (6%) leached less CCA components than those made of low
resin content (2.5%). UFR boards leached more CCA components than PFR boards

because PFR is water resistant.

23
Table 2 - CCA leachate content from particleboards after 28 days

 

 

 

 

 

Furnish Type Resin Type Resin [Cu] [Cr] [As]
Content
% ppb" ppb ppm
Virgin UF 2.5 60 200 0.28
4 50 1 80 --
6 40 120 0.04
PF 2.5 30 -- --
4 300 -- --
Reconstituted UF 2.5 3200 2800 12.9
4 2500 2400 17.7
6 1430 1400 10.5
PF 2.5 1000 800 25
4 700 600 14
6 700 700 14

 

 

 

 

 

 

 

 

Mechanical and Physical Progerties

Tables 3 lists the ANSI A208.1-1993 requirements of mat formed medium density
boards for MOE, MOR, and IB. The MOE, MOR, IB, TS, and WA values of boards

made of UP R and PF R are summarized in Tables 4 and 5, respectively.

Table 3 - Some ANSI requirements of medium density particleboards

 

 

 

 

 

 

 

MOE MOR IB
Grades kpsi MPa psi MPa psi kPa
M-l 250.2 1725 1595 1 1 58 400
M-S 275.7 1900 1813 12.5 58 400
M-2 or PBU 326.3 2250 2103 14.5 65 400
M-3 or D2 398.9 2750 2393 16.5 80 551.6
D-3 449.6 3100 2828 19.5 80 551.6

 

 

 

 

 

 

 

 

 

24
The average density of reconstituted boards or boards made of virgin firmish was

750i50 Kg/m3 (46.8i3.1 pcf) which corresponds to medium density. Their equilibrium
moisture content (EMC) prior to testing was 5 i 1 percent.

VUFR represent boards made of virgin (V) furnish using urea formaldehyde resin
(UF R). RUFR is reconstituted (R) board made of CCA treated utility poles using urea
formaldehyde resin. VPF R defines board made of virgin furnish with phenol
formaldehyde resin (PF R). RPFR is reconstituted boards from CCA treated poles with
phenol formaldehyde resin.
Urea formaldehyde particleboards

VUFR at 4% resin content met the ANSI requirements of MOE, MOR, and IB for
grades M-l for non-structural underlayment, M-S, M-2, or PBU for underlayment
flooring products and M-3 or D2 for home decking materials. At 4% resin content,
RUFR did not meet the grades M-3 or D2 MOE requirement, but satisfied the conditions
when 6% resin content was used. Resin content could be used to increase the bending
strength of reconstituted composites as illustrated in Table 4. The statistical significance
of the effect of the resin content and the firrnish on the MOE, MOR, and IB is also shown
in Table 4. The MOE of VUFR boards with 2.5, 4, or 6% resin content were similar but
the MOR of 6% resin content boards is higher than that of 4 or 2.5% resin content
boards. The MOE of RUFR was more sensitive and increased with the resin content.
MOE of 6% resin content of RUFR was similar to the MOE of VUFR at the same resin
level, suggesting a probable maximum plateau for the resin contribution. At 6% resin
content, the MOR of RUF R was similar to that of VUFR at 4%. At the low resin content

(2.5 to 4%), the MOR of RUFR were lower than that of VUFR which was in agreement

25
with literature data suggesting the reduction of properties of composites made of CCA

treated wood (Vick, 1980; Archer et al., 1993). The MOR was more sensitive than the
MOE and the IB. 7

A two-way analysis of variance (ANOVA) test conducted at a 95% level of
confidence indicated that 1B of RUFR was higher than that VUFR and also that the effect
of resin content from 2.5 to 6 percent was insignificant (Table 4) on 1B of furnish used.
The reason for the higher IB values was not understood. Archer et a1. (Archer, 1993)
reported higher IB value of boards made of CCA treated wafers in contradiction with
Schmidt (Schmidt, 1991, Boggio and Gertjejansen, 1982). Water absorption (WA) and
thickness swelling (TS) of VUF R were reduced considerably and in agreement with the
hydrophobicity created by CCA treatment. For RUFR at 4% resin content, WA was only
81% compared to 133% for VUFR. Water absorption and thickness swelling also
decreased with the increase of resin content for reconstituted wood.
Phenol formaldehyde particleboards

The properties of boards made of phenol formaldehyde resin are included in Table
5. At 2.5% resin content VPFR boards met the M-1 and M-S requirements for MOE,
MOR, and IB. VPFR can be used for commercial non structural underlayment. At 4%
resin content, MOE, MOR and 1B of VPFR was increased and M—2 or PBU conditions

(Table 3) satisfied for home decking products.

26

._0>o_ 00003300 $3 a 3 62.156 bwcwomfiwfi “0: 8a .832 08$ 2: E 330:8 £808 £08300 55$»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

son as. :2 exam 3a: $8 £0 3% 02 32: 3.2 0
£3 $3 as? s; Q mmm was $22 ~42 3:: a: a
so: $3 as? s; 3me $2 3 82 as 38 mg nm 89%
can: :8 5m 5m Ema $8 33$ 2.: 30 5 5w 4
as: sea as: :8 308m R8 00 mm: 3: A33 8m 2 52>
s as s s as. 82 an 82 «a an: .x. 8...:
008000
$~-<3 EL; 5m. 3 E. we .02 «02 a 58m geese
$30922th Man .«0 85.5003 .3330 00m 3253008 080m - m 2an
ass s3. s8 :2 3 2m Neon 3:08 :2 0.38 82 e
as; as? as? $8 303». $8 Eons ”.2 3:2 2: a
$3 at as? $3 358 ME: 32: a: 302 0:: 3 88¢
:2: $2. saw :3. 32% RR 33mm ”.8 3;: E 0
same «A: as x? 33m 82. SENS S: SE: 3 a
as”: some $82 :3 30% 20m 3 «SN 0: 30 t: E 2 53$
.x. .x. s. so a: £2 an 82 E s: as 8...:
008000
5N4? 09.4.3 new. me E. E m0: m0: a seam Ease

 

 

 

 

 

 

 

 

 

$383250 ME: 00 8000080 33930 use E20388 080m - v 2an

 

27
RPF R boards with 2.5% resin content met the M-1 grade requirement but fail to

satisfy the M-S requirement. RPFR at 4% PFR satisfies the M-S conditions but not the
M-2 compared to VPFR at 4%. Even at 6% resin content, RPFR did not meet the
conditions for M-2 grades. Reconstituted wood composites with 4 to 6% resin content
can be used for floor underlayment but not for home decking. The low WA and TS
(Table 5) of reconstituted products are some of the advantages that can be exploited from
using indicated CCA treated utility poles as furnish. Meanwhile, more research needs to
be done to understand the low mechanical properties of reconstituted wood compared to
those from virgin fiimish. IB of RPFR was higher than that of VPFR and also increased
with the resin content. The high IB of RPFR was not predictable based on available
information, more work need to be done to explain the IB behavior.
Decay resistance
Urea formaldehyde particleboards

OSB made of CCA treated wafers has been reported to impart adequate decay and
termite resistance under laboratory and field conditions (Archer et al., 1993). Decay tests
were conducted with boards made of urea and phenol formaldehyde resin. Weight losses
measured from an agar and also from soil block test are reported in Table 6. Weight loss
of RUFR boards were generally low or insignificant compared to that of VUF R boards.
The weight loss of UFR boards decreased with the increase of resin content confirming
some bioactivity of the resin. Another advantage of reconstituted wood from CCA

treated poles will be their durability when exposed to fungi.

28

Table 6 - Weight loss in percent of UP R particleboards exposed to fungi using modified

soil block and agar block test methods

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Test Particles Reconstituted Virgin
Type
Resin 2.5 4 6 2.5 4 6
Content (%)
Agar Control 1(2) 2( 1) 1(2) 2( 1) 1(2) 1 (2)
Block
P. placenta 1 (2) 1 (2) 1(2) 29(5) 28(5) 13(7)
G. trabeum 2(1) 1 (2) 2( 1) 37(12) 20(10) 7(3)
P. ostreatus 1(2) 2(1) 1(2) 11(14) 25(5) 30(10)
T. versicolor 1(2) 2(1) 1(2) 22(12) 20(5) 5(2)
Soil Control 1 (2) 2( 1) 1 (2) 39(1) 2(2) 2( 1)
Block
P. placenta 1 (2) 1 (2) l (2) 20(5) 15(5) 12(8)
G. trabeum 2( 1) l (2) 1 (2) 30(7) 21(3) 15(6)
P. ostreatus 2(1) 1 (2) 2(1) 42(6) 27(10) 42(5)
T. versicolor 1 (2) 1 (2) 1 (2) 20(7) 18(3) 2(2)

 

Numbers in parentheses represent standard deviation

Phenol formaldehyde particleboards

A soil block decay test was performed on boards made with 2.5 and 4 percent

PF R and results are shown in Table 7. Little or no decay occurred using PFR at any

content on boards made with CCA treated fumish. Virgin boards did, however, show

weight loss due to decay.

 

29
Table 7 - Weight loss of PFR particleboards using a modified soil block test

 

 

 

Furnish used virgin reconstituted
Resin content 2.5% 4% 2.5% 4%
Control 0(0) 0(0) 0(0) 0(0)

G. trabeum 24(27) 0(0) 0(0) 0(0)
P. ostreatus 0(0) 0(0) 0(0) 0(0)
P. placenta 44(22) 18(28) 0(0) 0(0)
T. versicolor 10(24) 0(0) 0(0) 0(0)
Standard deviations are in parentheses

 

 

 

 

 

 

 

CHAPTER 3

STUDY II: EVALUATION OF PARTICLEBOARDS MADE WITH DIFFERENT
TREATED WOOD PROPORTIONS

Objective

The objective of this study was to investigate the effect of increasing the CCA
treated wood proportion in particleboards on some mechanical and physical properties.
Special care was taken to ensure particle size was the same for both virgin and treated
particles. In a previous study, acceptable bending strength was attained with a liquid
phenol formaldehyde(PF) at 4 and 8 percent. The same resin levels were used in this
study.
Materials and Methods
Materials

Untreated and CCA treated red pine (Pinus resinosa Ait.) poles were obtained
from Hydrolake Leasing Service in McBain, Michigan. All poles were kiln-dried to 30
percent moisture content (MC) prior to treatment. Half of the poles were then pressure
treated with a 2 percent total oxide solution of CCA-C for 6 hours with a modified full-
cell method. The treatment included an hour of initial vacuum at 91 kPa (27 inches) of

mercury followed by 4 hours of pressure at 1.03 MPa (150 psi), and a final vacuum of 1

30

3 1
hour. The target retention of total oxides in the poles was 9.6 kg/m3(0.60 pcf ). The
treated poles were then air-dried to 19 percent MC.
Particle Manufacture Methods

Poles were chipped with an Morbark Eager Beaver Chipper, and chips reduced
into particles with a laboratory hammermill. Particles from untreated red pine were
virgin firmish and particles from CCA treated red pine named treated furnish.

Special care was taken for particle size to avoid imbalance which could create
panel warping and strength reduction. Particles were sifted by size with a vibrating
inclined screen, and only particles passing through 10 but held by 16 mesh screens were
selected for particleboard production. A screen analysis was performed on a sample of
both particle types with an electric shaker for 5 minutes using screen 8, 10, 16 and 30
mesh. The thickness, width, and length of 1000 screened particles were also measured by
light microscope. Screened particles were air dried to 53:2 percent MC and used to
manufacture particleboards.

The pH of both furnishes was determined by using a pH meter. In a beaker, 10 g
of screened particles were mixed in 100 ml of distilled water for 30 minutes using a
sonicated bath and the pH determined.

Composite Manufacture Methods

 

Treated and virgin particles were mixed at five proportions by weight, namely
100, 75, 50, 25, and 0 percent treated wood content. These mixtures were then sprayed
with GP© 107C38 RESI-STRAN Oriented Strand Board phenol-formaldehyde resin (PF)

containing 55 percent resin solids. Ten replicates were manufactured at each proportion

32
of treated wood content with 8 percent resin solids, while 6 particleboards were
manufactured at each proportion of treated wood content with 4 percent resin solids.
Overall a total of 80 boards were fabricated.

Particle mixtures were sprayed with resin in a laboratory rotary drum blender for
5 minutes and mats were hand-formed in a 40.6 cm square (16" square) frame. The MC
of the particles in the mat before pressing was 84:2 percent. A Berthelsen thermo-oil
heated hydraulic press was used to press the mat between two steel platens down to a
nominal thickness of 10 mm (0.3 75" ). The time interval from the application of resin to
pressing was kept constant at 5 minutes in all trials. The press time was 6 minutes, the
pressure 800 psi (8.28 MPa ), the press temperature 190°C (325°F), and the closing time
13 seconds.

Low density areas on each board were removed by trimming one inch on all
edges. Trimmed boards were kept at 65 percent relative humidity (RH) and 20°C (68°F )
for at least 40 days before testing or until they reached their equilibrium moisture content
(EMC). The EMC and density of boards were 5d:1 percent and 750:1:50 kg/m3(47&:3 pcf ),
respectively.

Material Testing Methods

Samples were cut from each board to conform with American Society for Testing
Materials (ASTM) Standard number D1037-95 guidelines to obtain: 2 specimens for
bending, 3 for thickness swelling (TS) and water absorption (WA), and 6 for internal
bond (IB) (2) (See Figure 1). Samples were stored in a room conditioned at 65 :1: 1

percent RH and temperature of 20i3°C(68i 6°F) until tested. ASTM Standard 103 7-95

33

tests for static bending and IB were all conducted using an Instron testing machine (2).
TS and WA were conducted by ASTM Standard 1037 as well.
Leaching Test Method

A modified American Wood Preservers' Association (AWPA) (3) leaching test
was also performed on each board type as follows: two liters of water were added to
twelve 7.62 cm by 7.62 cm by 1 cm (3" by 3" by 0.375") samples in a plastic container
with a cover to reduce water evaporation. Aliquots of the leachate were taken every three
to five days for 28 days and analyzed for chromium, copper, and arsenic content.
Determination of CCA Retention in Particleboard

An ASOMA X-ray Fluorescence Analyzer, Model 100, provided by Universal
Forest Products in Grand Rapids, MI, was used to determine the concentration of
chrome, copper, and arsenic in the finished particleboard product at each level of
treatment (0, 25, 50, 75, and 100 percent CCA-treated wood). Samples of 5 grams from 3
random boards at each treatment level were taken, ground, and oven-dried at 100 i 3° C
(212 i 6° F) for 1 hour. The samples were then analyzed at a density of 750 :1: 50 kg/m3
(47 :1: 3 pct) on the ASOMA.
Results and Discussion
Particle Size Anialysis

Since the same method of particle manufacture was used as in the first study, the
results of analysis were the same.

Particle length, width, and size distribution are summarized in figures 2, 3, and 4,

respectively. About 80 2t 5 percent of the thousand particles analyzed by light

34
microscopy had an average length of 3.6 i 1.0 mm (0.14 d: 0.04") and an average width
of 1.5 :1: 0.3 mm (0.06 :1: 0.01 "). The average slendemess ratio was 2.64: 1, which is
defined as the ratio of length to the diameter of the particle. From the Tyler sieve
analysis, Figure 4 depicts particle size ranged from 0.5 to 2.3 mm (0.02 to 0.09"). This
shows the same distribution for both treated and untreated furnish by weight. Eighty-
seven percent of all particles were between 1.52 and 1.78 cm (0.06 and 0.07 ") in size.
This data is in agreement with microscopy measurements based on distribution and
standard deviation. Any differences between microscopy and Tyler sieve data could be
explained by the many angles at which particles could pass through the Tyler sieves (0 -
180 degrees).
Particle pH

Since the same particles were used in Study II as in Study 1, the particle pH was
the same. The results are summarized again here.

The particle pH was found to be slightly different for virgin versus treated
particles. Virgin particles had a pH of 4.9 compared to 5.1 for the CCA-treated fumish.
However, pH obtained through this method is not representative of the pH of the wood
surface since the pH meter evaluates the concentration of H+ liberated in the water
medium. Therefore the pH could be influenced by the solvent. Thomason and Pasek
(1997) have shown that pH of southern yellow pine tested in water is different than tested
in acetone. Knowing the low pH of CCA (pH .2. 2), the similarity in pH of CCA-treated
and virgin red pine suggest that wood may behave as a buffer, or the retention of CCA in

the wood analyzed was low.

35

Mechanical and Physical Prgerties
Effect of resin content

The MOR, MOE, and 1B of reconstituted particleboards made with 4 and 8
percent solid resin content with increasing amounts of CCA treated particles are shown in

Tables 8, 9 and 10, and Figures 5, 6, and 7.

36

_0>0_ 000000000 $3 0 00 000000.00 0000050me 000 000 0000— 0800 05 3 0032.00 3002

 

 

 

 

 

 

 

:30 5.4m 30$ :30 303% 0.88 30 088 2;: .<.z.<.z 2:
£30 0.9% :08 :08 3:.on $08 30 same 4.802 .<.z.<.z we
0.0.3 :03 :08 s02 30:; 4.88 30.090 282 .<.z.<.z om
$0.6 some. 5.8 :02 34.0% in: 3082 EMS: .<.z.<.z mm
5.3 $0.2. 0.0.3 :30 30.0% ”.22 39:8 033: .<.z.<.z o

as so as as EM £2 E s: E 0%. 0..

EN :N

-55 -55 mg: mp mm - E 002 002 m: 8005 Base

 

 

 

 

 

 

 

 

 

 

500.. 00 o\°w Emma—.50 000000 <00 505 0008 0000000000 .00 000000000 03930 000 .E .mE000m - 0 030%

 

 

 

 

 

 

 

:30 $4.05 :3; :02 3038 20: 3008: 0.83 @002 28 2:
$30 0.0.? 0.0.3 so? 32: 0.82 300.82 380 33: oi: me
:30 5.? seem so: 3080 0.3.2 33:: was: 34.0: 202 om
so? am: 5.3.. :0: 00:02 082 30%: 008$ 33:: So: 2
.003 some $0? snmm 30.02 0.32 3028 4.302 Emma is: o

as .x. as as as. £2 E s: E 0%. so

0% 5 EN

-55 -55 - mp mm- 00 002 002 a £250 305

 

 

 

 

 

 

 

 

 

E80 00 $0 ”0005.0 000000 <00 505 0008 000000200000 00 00000000 000930 000 .m: .w5000m - w 053

37

030> 5.0080 n 0 000:3
2: .. E \ E - MET 005% .x.

 

 

 

 

 

 

 

 

 

 

KEN- o\o~ m- gmm- o\°m m- $0 e\omv - cg

$3.. $07 $3. fibm- $3 $00 - mm

o\cmm.. $5M- $3. $3. fa: o\om0 - cm

KEN- o\om m- c\eov. foam- Xbo— $0m - mm

30m- o\ow m- o\oo m- $31 o\cmo_ §mm - o
3503 Nix? vméH N-mH mOE M002 9

 

00000080 00000200000 so E00000 £000 00 00 80mm - 3 030B

 

38

 

20000 0770004,0_40 00,,0.00 004,440.,4__4,Wj
18000 1
16000 1
14000 3,
12000 3

10000

C!)
O
O
O

Modulus of Rupture (kPa)
8
8

 

0 25 50 75 3 100
CCA-Treated Furnish Content (%)

 

Figure 5 - MOR of particleboards made with untreated and treated particles

__J

 

39

 

 

Modulus of Elasticity (MPa)

 

CCA-Treated Furnish Content (%)

 

 

 

Figure 6 - MOE of particleboards made with untreated and treated particles

‘ 1400 2 2 , 7
% -4%F$
Resin \
l 1200 ,,
‘10
10
151000,
is
‘5
i:
2 800‘
‘4.-
1w
|0
: 1
‘o 600.
10
HE
\E
i3 4%»
15 p
l l 1
‘ 200» i ~

 

1W

0 25
CCA-Treated Furnish Content (%)

liigfl wggmigmifirev¥xAfli_V4wwfi ,

Figure 7 — IB of particleboards made with untreated and treated particles

41

Results show that 8 percent PF board property values were substantially higher
than those made with 4 percent PF. As expected, increasing the resin content increased
the properties of the board in agreement with the literature (Boggio et al., 1982,
Gertjejansen et al., 1988; Vick, 1980; Vick et al., 1990; Vick et al., 1996). The increase
of PF resin content from 4 to 8 percent, resulted in a 33 percent increase in MOR for
boards made of virgin fumish, and an increase of 45 i 5 percent for other boards
containing treated furnish.

The MOE of boards containing 0 or 25 percent treated furnish increased 100
percent when 8 percent PF was used compared to 4 percent PF. The MOE of boards
containing 50 or 75 percent treated wood increased by an average of 78 percent by
doubling the resin content. When using 100 percent treated wood the MOE did not
change when applying 8 percent instead of 4 percent PF resin (Table 10). The MOE ws
reduced either because of a low resin content or the particle aspect ratio was not large
enough.

The 18 values for 4 percent PF boards are shown in Figure 6. The IB strength for
boards made of 8 percent PF were over the limit of the load cell available.

The only differences in TS or WA found in this study were between boards made
of 4 or 8 percent PF resin. Boards made with 8 percent PF swelled 37 percent less than
those made with 4 percent PF after 2 hours. After 24 hours, TS decreased by 33 percent
by using 8 instead of 4 percent PF. Also, 8 percent boards absorbed 32 percent less water
after 2 hours compared to 4 percent boards. After 24 hours, boards made with 8 percent

PF absorbed 29 percent less water than boards made with 4 percent PF. This can be

42
explained by PF being an exterior resin used for water-resistant applications. As
expected, increasing resin content decreases thickness swelling and water absorption.
Effect of C CA-treated wood content

MOR, MOE, and IB are reduced when a greater proportion of treated particles
were used. These findings are consistent with the literature (Boggio et al., 1982,
Gertjejansen et al., 1988; Vick, 1980; Vick et al., 1990; Vick et al., 1996). The reduction
in mechanical properties has been explained by the incompatibility of PF resin bonds
with CCA treated wood (Boggio et al., 1982, Gertjejansen et al., 1988; Vick, 1980; Vick
et al., 1990; Vick et al., 1996). It has been suggested that insoluble chromium, copper,
and arsenic solids present on the cell walls may reduce the formation of bonds between
adhesives and wood. Recently it has been attributed to the reduction of wood cell
physical and mechanical properties due to loss of some cell wall components during CCA
treatment (22). Winandy et a1. (1997) reported that CCA treatment reduces the extractive
content, and up to 20 percent of the hemicellulose in wood.

The use of 50 percent CCA-treated wood content in the manufacture of
particleboard did not affect the MOR, MOE, and IB significantly. At 75 percent treated
wood content, the MOR was significantly different than the MOR at O, 25, and 50 percent
treated wood content. The IB and MOE were reduced significantly only for samples
containing 100 percent CCA—treated wood.

Dimensional stability (Tables 8 and 9) was not affected by the proportion of

CCA-treated wood in the particleboard. Table 10 gives the effect of PF resin content on

43

the change in TS and WA. The TS and WA for boards made with 4 or 8 percent PF were
not affected significantly by the treated wood content.
Statistical An_alvsis

A statistical analysis of mechanical and physical properties was executed. A
Tukey two-way AN OVA was used to make simultaneous pairwise comparisons between
dependent data sets using two independent variables. The treated wood content and resin
content were used as the independent variables. Data was analyzed to test if the
distributions were normal and if the variances were equal in order to compare the data
with significance. All tests were compared with 95 percent confidence.
Determination of CCA Retention in Particleboard

The results of ASOMA analysis of particleboard CCA concentrations are shown
in Table 4. As expected, the actual quantities of chromic oxide, copper oxide, and arsenic
pentoxide gradually increase with nominal treated wood concentration. The oxide
concentrations in the boards made with 8 percent PF resin were insignificantly different
with oxide concentrations in boards made with 4 percent PF resin. The retentions of
CCA in kg/m3 in the boards were higher than that of the poles because the density of the
particleboard was about twice that of red pine. Red pine wood has a density of 385 kg/m3

(24 pct) while the board densities were 750 kg/m3 (47 pct).

44

02000000 030 so 000m

 

 

 

 

 

 

 

 

 

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00.0 00.00 00.0 00.0 00.0 00.0 00.0 v: 00.0 0.0
00.0 00.0 00.0 00.0 00.0 0.1 00.0 00.0 00.0 0.0 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 5.4 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 2.0 00.0 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0
000 000000 000 ”00000 000 ”0000 000 “8000 000 ”8000 .0
30000000 cos—00:00:00
000.0 no? 000 ”06 00832 .0582
00008920000 “.0. $0 0o 00200000050 .0390. I
00.0 00.00 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 04.0 00.0 00
00.0 00.0 3.0 00.0 00.0 0: 00.0 00.0 00.0 00.4 00
20 E0 00.0 00.0 00.0 00.0 00.0 00.0 2.0 04.0 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 0
.00 0500 000 "800 000 0500 000 00000 000 0800 00
000300000 cozmbcmocou
0000 “000. 000 600 00832 .0582
000.080.2000 “E 000 0o 0020000000000 .0390.

 

 

 

 

 

0000000000000 000 E <00 00 0000000000080 008000 - 2 0300.

45
Leaching Test Results
Leaching test results are shown in Table 5. Chromium, copper, and arsenic in the
leachate were analyzed over a 28 day period for 4 percent PF boards and over a 14 day
period for 8 percent PF boards. The corresponding metal oxides were calculated
assuming that copper was present as copper oxide (CuO), chromium as chromium
trioxide (Cr03), and arsenic as arsenic pentoxide (A320,) (See Appendix B). These
leaching results are expressed in percent of initial CCA oxides present in the
particleboard.

Table 12 - Leaching as a percent of the initial total oxides in the particleboards

 

4% PF after 14 days 8% PF after 14 days 4% PF after 28 days

 

 

 

Nominal CrO3 CuO A3205 CrO3 CuO A8205 CrO3 CuO A3205
Concentration (%) (%) (%) (%) (%) (%) (%) (%) (%)
O 0 O O O O O O O O
25 O O 2 0 O 1 O O 3
50 O O 2 0 0 1 O 0 2
75 O 0 1 O O 1 O O 2
100 O O 1 O O 1 O 0 2

 

 

 

 

 

 

 

 

 

 

 

As shown in Table 12, negligible amounts of CCA were leached from the
Particleboards after 14 and 28 days. No CrO3 and CuO were leached at all, while 1 to 3
Percent of the arsenic pentoxide was leached. These values were within the error of the
E=quipment used.

Q. omparison to ANSI Requirements

Finally, in comparison with ANSI standard A208.1-1993 for medium density

particleboards, 4 percent PF boards with 0 and 25 percent treated wood satisfied the

requirements for an M-l grade for IE and MOR. These boards failed MOE requirements.

46

All other 4 percent PF boards passed an M-l grade for IB. Boards made with 8 percent
PF and O, 25, and 50 percent treated furnish passed requirements for an M-2 grade, while
75 and 100 percent treated firmish passed an M-S grade. Both M-2 and M-S are grades
used for particleboard underlayment or subflooring. These results show that a maximum
of 50 percent CCA treated particles can be incorporated in particleboard if an 8 percent
PF resin content is used for the manufacture of underlayment or subflooring.
Comparison Between Study L and Studv II

Properties of boards made in Study I were not similar to those produced in Study
II with the same type and amount of resin and the same proportion of CCA treated wood.
The discrepancy cannot be explained and this may represent a major problem of this
study. However, several speculative explanations can be given with regard to the
variability.

The most probable reasons could be the control of manufacturing parameters.
The press closing times used in this study were shorter (9 to 15 seconds) than the
commercial practice consisting of a minimum of 30 to 45 seconds (Suchsland, 1986;
Chow, 1984) necessary to reduce density variation within a board. A short closing time
is known to create variable MOE, MOR, and IB within a board. An appropriate close
time would have reduced the density variability, as well as made the MOE, MOR, and the
IB more consistent.

The second oversight was using two different batches of furnish of CCA treated
red pine. Each study was performed with a limited amount of CCA treated fiber, and

therefore, a second batch was required to complete the study. This was a limited amount

47

because boards were made at first with 8 percent instead of 4 percent PF in the second
study due to board calculation error. This may have changed property values
significantly between Studies I and II.

Finally, the third overlooked possibility was using industrial-commercial resins.
We didn't want to vary or change the glue type, but Georgia-Pacific had been using one
formulation when the first study was initiated, and a second resin when the second study
began. The objective of the study was to determine feasibility without specific concerns
on glue, and whether or not we could satisfy the requirements for particleboard. It was
assumed that the properties would not significantly change due to the resin formulation.

This may not have been true.

CHAPTER 4

CONCLUSIONS

From the results of the first study, reconstituted particleboards can be made from
red pine utility poles treated with CCA at 0.60 pcf using UFR or PFR. A level of 4 to 6
percent resin content was needed to produce boards with bending properties comparable
to those made of virgin furnish. With 4 percent or more resin content ANSI requirements
for medium density mat formed particleboards for underlayment were satisfied. Water
absorption capacity, thickness swelling, and biological durability were improved for
reconstituted boards. The amount of metal leached from the boards was negligible.

For the second study, particleboards were manufactured containing 0, 25, 50, 75,
and 100 percent CCA treated red pine utility poles and bonded with 4 or 8 percent liquid
PF resin. From the results, no significant difference was found between boards made of
50 percent treated wood and untreated particleboard in mechanical and physical
properties, but to meet ANSI requirements a higher resin content should be utilized. The
MOE, MOR, and IB strength were reduced significantly when the board contained 75
percent CCA treated particles at both 4 and 8 percent solids resin content. There was no
significant change in dimensional stability due to the amount of treated furnish used, but

there was a difference of up to 35 percent when using a higher resin content.

48

CHAPTER 5

RECOMMENDATIONS

This study showed that reconstituting freshly treated wood to make particleboards
was feasible with 50 percent treated wood content and 8 percent PF resin with short close
times. Longer close times should be used in future studies to reduce density variation.

In the future, it may be feasible to manufacture wood composites with a mixture
of recycled wood materials and freshly treated wood or virgin materials, but not using
100 percent recycled wood materials. Further studies on the actual recycling of decks
into wood composites is imperative to understand the feasibility of it and to reduce the
amounts of CCA treated wood that will go to landfill each year.

Additionally, an analysis of the costs and benefits of this type of recycling is
necessary. Understanding the impact on particleboard producers is important to the
implementation of such a production.

Finally, as shown in this study, the leaching of reconstituted particleboards from
CCA treated red pine is negligible. Further investigations of the leaching of CCA from

reconstituted particleboards need to be investigated further to confirm these results.

49

APPENDICES

APPENDIX A

APPENDIX A

Sample Calculation Worksheet of Particleboard Production

Final (target) board dimensions: 15" x 15" x 0.375"

VolumeOD of target board: 15" * 15" * 0.375" = 84.375 in3

VolumeOD of target board conversion: 84.375 in3 * (2.54 cm/in)3=1383 cm3
Target board densityOD: 0.75 g/cm3 (Allowable range: 0.70-0.80 g/cm3)
Target board MassOD: 0.75 g/cm3 * 1383 cm3 = 1038 g

Resin solids (dry): 4% 1038 g * 0.04 = 41.52 g

Particles (dry): 96% 1038 g * 0.96 = 996.48 g

Table 13: Resin specifications

 

I Resin type PF UP I

Solid content 53% 55%
Specific gravity 1.219 1.270

 

 

Add moisture content of 3.5%

Mass of particles with water: 996.48 g / (l-MC) = 1033 g
Mass of liquid resin: 41.52 g / 0.55 = 75.49 g

Liquid volume of resin: 75.49 g / 1.270 = 59 ml

Mat weight: 1033 g + 75.49 g = 1108 g

50

APPENDIX B

 

APPENDIX B

Leachate Calculations
First, convert metal in leachate to total mg of oxide leached.

Example:
XpartsCr x Cr03 _ XmgCrOz
millionHzO Cr LHzO

 

 

 

 

 

 

W x LH20 = XmgCrOs
LH 20
Second, determine how many milligrams of metal oxide were in the particleboard to
begin with.
Example:

Determine total milligrams of wood:

12 samples with 0.75 g/cm3 density,

and volume of 10.16 cm x10.16 cm x 1 cm = 103.2 cm3

Total volume: 12 x 103.2 em3 = 1238.7 total cm3

Total mass: 1238.7 cm3 x 0.75 g/cm3 = 929 g of particleboard, or 929,000 mg

Then find milligrams of metal oxide in that wood:

R = Retention of CrO3 in percent

R x 929,000 mg = CmgCrO3

where C = mg of metal oxide in particleboard

Finally, make a percentage that leached of the total C CA oxide initially present in the

board

 

XmgCr03Leached
C mgCr03initial

x 100 = %Cr03Leached

 

 

 

 

51

LIST OF REFERENCES

American National Standards. 1993. Particleboard (ANSI). ANSI A208.1 - 1993.
National Particleboard Association 18928 Premiere Court, Gaithersburg, Md.

American Society for Testing and Materials. 1996. Standard Methods of evaluating the
properties of wood-base fiber and particle panel materials. ASTM Design.
D1037-78. ASTM, Philadelphia, PA.

American Wood Preservers' Association. 1997. 1996 Book of Standards. Woodstock,
Maryland.

American Wood Preservers' Institute. 1996. The 1995 Wood Preserving Industry
Production Statistical Report. AWPI; Vienna, Virginia.

Archer, K. J ., D. A. Fowlie, and A. F. Preston. 1993. Test methods and treatments in
development of protected composite products. In: Protection of wood-based
composite products. Forest Prod. Soc., Madison, WI, pp. 38-43.

Boggio K. And R. Gertjejansen. 1982. Influence of ACA and CCA waterborne
preservatives on the properties of aspen waferboard. Forest Prod. J. 32(3):22-26.

Chow, P. and J .J . J anowiak. 1984. The internal bond and shear strength of hardwood

veneered particleboard composites. Wood and Fiber Sci. 18(1): 99-106.

52

53
Cooper, P. A. 1993. The potential for reuse of treated wood poles removed from service.

Proc. 2nd Inter. Symp. on Wood Preservation. Cannes, Mandelieu, France. DOC
IRG/WP 93 50001: 252 264.

Cooper, P. A. 1994. Disposal of treated wood removed from service: the issues. In:
Environmental considerations in the manufacture, use, and disposal of
preservative-treated wood, Forest Products Society; Madison, Wisconsin, pp. 85
90.

Ellis, S. C., Ruddick, N. R., and Steiner, P. R. 1993. A feasibility study of composites
produced from telephone directory paper, plastics and other adhesives. Forest
Prod. J. 43(7/8): 23-26.

English, B. 1992. Meeting society’s challenge: value-added products from recycled
materials. Pacific Rim Bio-Based Composites Symposium, 9-13 November,
Rotorua, NZ.

Falk, R. 1994. Housing products from recycled wood waste. Proceedings of Pacific
Timber Engineering Conference, July 11-15, Gold Coast, Australia.

Felton, C. C. and R. C. DeGroot. 1996. The recycling potential of preservative-treated
wood. Forest Prod. J. 46(7/8): 37-46.

Gatchell, C.J., B.G. Heebink, and RV. Hefty. 1966. Component variables on properties
of particleboard for exterior use. Forest Prod. J. 16(4): 46-57.

Gertjejansen, R., J. Haygreen, M. Hyvarinen, and D. French. 1972. Physical properties
of phenolic bonded wafer-type particleboard from mixtures of aspen, paper birch,

and tamarack. Forest Prod. J. 23(6): 24-28.

54
Gertjejansen, R.O., E. L. Schmidt, and D. C. Ritter. 1988. Assessment of preservative-

treated aspen waferboard after 5 years of field exposure. Forest Prod. J. 39(4): 15-
19.

Hall, H.J., R.O. Gertjejansen, E.E. Schmidt, C.G. Carll, and RC. DeGroot. 1982.
Preservatives treatment effects on mechanical and thickness swelling properties of
aspen waferboard. Forest Prod. J. 32(1 1/ 12): 19-26.

Hashim, R., R. J. Murphy, D.J. Dickinson, and J .M. Dinwoodie. 1994. The mechanical
properties of boards treated with vapor boron. Forest Prod. J. 44(10): 73-79.

Haygreen, J .G. and J .L. Bowyer. 1989. Forest Products and Wood Science: An
Introduction. Iowa State University Press. Ames, Iowa.

Honda A., Kanjo Y., Kimoto A., Koshii K., and S. Kashiwazaki. 1991. Recovery of
copper, chromium, and arsenic compounds from waste preservative-treated wood.
Kyoto, Japan. DOC IRG/WP 91-3651.

Ince, P.J. and D. Zhang 1994. Impacts of recycling technology on North American fiber

' supply and competitiveness. Proceedings of 3rd International Symposium, Sept.

13-14. Seattle, WA.

Kamdem, DP. and S. Hiziroglu. 1994. Some of the mechanical and physical properties of
hardboard made of black locust, Forest Prod. J. 45(11): 23-26.

Kamdem, DP. and ST. Trek. 1994. The durability of phenolic-bonded particleboards
made of decay-resistant black locust and nondurable aspen. Forest Prod. J. 44(2):

65-68.

55
Kumar, S. and J .J . Morrell. 1993. Improved composites fi'om chemically modified

particles. In: Protection of Wood-Based Composite Products. Forest Prod. Soc.:
33-37.

Laufenberg, TL. and S. Abubakr. 1994. Expanding research horizons: USDA forest
service initiative for developing recycled paper technology. Proceedings of 3d
International Symposium, Sept. 13-14. Seattle, WA.

Laufenburg, T., R. Horn, T. Wegner and S. Bean. 1992. Improvements in recycling
wood and wood-fiber products. 1992 Yearbook of Agriculture, 264-275. USDA.
Washington, DC.

Lehmann, W.F. 1969. Resin efficiency in particleboard as influenced by density,
atomization, and resin content. Forest Prod. J. 20(11): 48-59.

Marer, P. J. and M. Grimes. 1992. Wood Preservation. University of California statewide
integrated pest management project. Division of Agriculture and Natural Res.
Pub. 3335.

Michell, A. J. 1980. Composites of commercial wood pulp fibres and cement Pinus
radiata and Eucalyptus species. APPITA 33(6): 461-463.

Moslemi, A. A. 1974. Particleboard: Vol. 2. Southern Illinois University Press.
Carbondale, IL.

Minor, J .L., C.T. Scott and RH. Atalia. 1993. Restoring bonding strength to recycled
fibers. 1993 Recycling Symposium. University of Wisconsin.

Pasek, E. A. 1995. Air emissions from burning CCA treated wood. In: Wood preservation
in the ‘905 and beyond. Forest Products Society conference proceedings, pp 42-

48. September, 1994. Savannah, Georgia.

56
Pasek, E. A. and C. R. McIntire. 1992. Burning of CCA treated wood with minimal

metal loss. AWPI Workshop, ibid.

Rowell, RM. 1992. Opportunities for lignocellulosic materials and composites.
Proceedings of symposium; 1990 August 26-31. Washington, DC. American
Chemical Society symposium series 476.

Rowell, RM. 1992. Property Enhancement of Wood Composites. Composite
applications- The role of matrix, fiber, and interface. New York: VCH Publishers,
Inc. Chapter 14.

Rowell, R.M., et. a1. 1992. Opportunities for composites from recycled wastewood-
based resources: problem analysis and research plan. Forest Prod. J. 43(1) 55-63.

Schmidt, E.L., H.J. Hall, R.O. Getjejansen, C.D. Carl] and RC. DeGroot. 1983.
Biodeterioration and strength reductions in preservative treated aspen waferboard.
Forest Prod. J. 33(11/12): 45-53.

Schmidt, R., R. Marsh, J .J . Balatinecz, and RA. Cooper. 1994. Increased wood-cement
compatibility of chromate-treated wood. Forest Prod. J. 44(7/8): 44-46.

Schwartz, H.G. and M.H. Simatupang. 1984. Suitability of beechwood for the
manufacture of cement bonded wood composites. Holz als Roh - und Werstoff.
Berlin, W. Ger.: Springer. 42(7).

Stalker I. N. 1993. Disposal of treated wood after service. Proceeding CWPA:159-174.

Suchsland, O. and Woodson, G. E. 1986. F iberboard manufacturing practices in the
United States, United States Department of Agriculture, Agriculture Handbook

no.640.

57
Suzuki, Kentaro. 1993. Field evaluation of composite products, six years in service.

Protection of Wood-Based Composite Products, Forest Prod. Soc.: 44-48.

Vick, C. B. 1980. Structural bonding of CCA treated wood for foundation systems. Forest
Prod. J. 30(9):25-32.

Vick, C.B., R.C. DeGroot and J. Youngquist. 1990. Compatibility of nonacidic
waterborne peresevatives with phenol-formaldehyde adhesive. Forest Prod. J.
40(2): 16-22.

Vick, C.B., R.L. Geimer, and J .E. Wood, Jr. 1996. F lakeboards from recycled CCA
treated southern pine lumber. Forest Prod. J. 46(1 1/ 12): 89-91.

Webb, D. A. and D. E. Davis. 1994. Spent treated wood products: alternatives and their
reuse or recyclability. In : Environmental considerations in the manufacture, use,
and disposal of preservative-treated wood, Forest Products Society, Madison,
Wisconsin.

Winandy, J. E. and J .J . Morrell. 1997. Effects of waterborne preservatives treatment on
mechanical properties. Am. Wood. Pres. Assn. Preprints. Technical Forum,
Doubletree Hotel; Pittsburgh, PA.

Winandy, J. E. and K.A. McDonald. 1993. Material selection and preservative
treatments for outdoor wood structures. Wood Design Focus 4(3): 8-13.

Youngquist, J .A., B.W. English and RM. Rowell. 1993. Alternative uses for waste-
paper in wood-based composite products. Proceedings: 1993 Recycling
symposium; Feb. 28-March 4. New Orleans, LA. Atlanta, GA: TAPPI Press:

359-365.

"I11111111111111111111111“