This is to certify that the
thesis entitled

DURABILITY OF WOOD-PLASTIC COMPOSITES
EXPOSED TO FREEZE-THAW CYCLING

presented by

JEANETTE M. PILARSKI

has been accepted towards fulfillment
of the requirements for the

_Master of Scie_nce_ degree in _, Forestrl

 

- I “I a
(a/{VM F 1'. if %\ «Shh/t (i am K

Ma jot?ProfessorTs-Sigﬂnatftre‘ _

M _ L414 245;

Date

 

MSU Is an Affirmative Action/Equal Opportunity Institution

LIBRARIES
MICHIGAN STATE UNIVERSITY
EAST LANSING, MICH 48824—1048

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECAILED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c:/ClRC/DateDue.inddop.15

 

DURABILITY OF WOOD-PLASTIC COMPOSITES
EXPOSED TO FREEZE-THAW CYCLING

By

Jeanette M. Pilarski

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Forestry

2005

ABSTRACT

DURABILITY OF WOOD-PLASTIC COMPOSITES
EXPOSED TO F REEZE-THAW CYCLING

BY
Jeanette M. Pilarski

Two studies were used to assess the durability of wood-plastic composites
(WPCs) exposed to accelerated freeze-thaw cycling. A counter-rotating twin-
screw extruder was used to process various formulations of WPCs, which were
then exposed to accelerated freeze-thaw testing according to ASTM standard D-
6662. This method includes exposure to water submersion, freezing, and thawing
as one complete cycle. The effects of 2 to 15 freeze-thaw cycles were observed
for the density, dimensional stability, ﬂexural strength and stiffness, porosity, and
microscopic appearance of WPCs containing both maple and pine wood ﬂour. Of
the composite formulations tested with varied wood contents, only the lower
wood content WPCs showed resistance to freeze-thaw actions. The density of all
WPCs were unaffected by freeze-thaw actions. The WPCs maintained relatively
stable dimensions after exposure compared to other materials such as solid
wood. The ﬂexural strength and stiffness of all WPCs were signiﬁcantly
decreased by as few as 2 cycles. Scanning electron microscopy conﬁrmed
similarities between water-only and full freeze-thaw cycles. The main cause of
the ﬂexural property losses was attributed to the water portion of this cyclic
process, which led to decreased bonding between the wood and polymer, which
also increased the porosity. The freezing (without water) had a lesser effect on

WPCs compared to the water portion.

Dedicated to my friends and family
Especially Dan, Karana, Bhavesh, Katie and my mom Sue

Without whom I would have never made it this far

iii

ACKNOWLEDGEMENTS

For convincing me that wood-plastics were an exciting and emerging field
I would ﬁrst like to thank Dr. Laurent Matuana, my major professor. He has
taught me several things, of which I will take what I like and leave the rest; I will
learn and go my own way. I have also learned the importance of working smart,
not hard so that I am not just a lab tech and my degree will mean something. I
also need to thank Dr. Daniel Keathley for convincing me to join the Forestry
Department here at MSU. Without the barbeque lunch with his daughter I would
have probably never gone down this path and become what I am today. An
additional thank you to my other committee member Dr. Laura Bix for showing
patience with me through a very stressful process.

The faculty and staff in the Forestry Department have become like my
family during the last frve years I have been in the department. Both Juli and
Carol have always gone above and beyond the call of duty and always brighten
my days. Everyone else here has been wonderful too and it is amusing the
reverence you get as “one of those wood-science people”.

All of my work was supported by the USDA-CSREES Grant-Advanced
Technology Applications to Eastern Hardwood Utilization (Grant No. 2002-
34158-11914). An additional thank you to the companies listed in the text of
chapters 3, 4 and 5 for the kind donation of supplies. Among the companies that
donated supplies are B.P. Solvay Polymers, American Wood Fibers, Struktol,

Eastman Chemicals, and Rohm & Haas.

iv

Lastly, I need to thank all of my friends and family. Karana, Bhavesh, and
Katie have been the best co-workers I could have hoped for. They always cheer
me up and are ready to go get an ice cream cone on a rough day. My mom has
always given me endless support throughout all of my education and life in
general. She has been my best friend and conﬁdant for as long as I can
remember and I owe a lot of what I have accomplished to her sacriﬁce and aid in
all of my pursuits. Dan has also given me unending support that I could not have
done without. I never would have thought when I started work on my Master’s
degree that we would be married and where we are today. You make me smile
each day and never let me give up because you know the true me almost as well
as I do. Thanks to everyone else that I have crossed paths with here at MSU

too, for good or bad it has been a Ieaming experience.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................... , viii
LIST OF FIGURES ............................................................................................... ix
LIST OF ABBREVIATIONS .................................................................................. xi
CHAPTER 1. Introduction ............................................................................. 1
Objectives .................................................................................. 7
References ................................................................................ 8
CHAPTER 2. Background ..................................... 11
Solid wood products ................................................................ 11
Types of wood .............................................................. 1 1
Properties of wood ........................................................ 17
Deterioration of wood .................................................... 20
Preservation of wood .................................................... 22
Summary ...................................................................... 25
Wood-based composites ......................................................... 26
Wood adhesives ........................................................... 26
Veneer products ........................................................... 31
Particle and ﬂake products ........................................... 33
Summary ...................................................................... 36
Wood—plastic composites ........................................................ 37
Development of wood-plastic composites ..................... 37
Raw materials used in wood-plastic composites .......... 39
Polymers .............................................................. 40
Additives .............................................................. 42
Fillers and ﬁbers ................................................... 45
Degradation of wood-plastic composites ...................... 47
Processing of wood-plastic composites ........................ 51
Extrusion .............................................................. 52
Injection molding .................................................. 54
Compression molding .......................................... 55
Testing of wood-plastic composites .............................. 57
Summary ...................................................................... 62
References .............................................................................. 63
CHAPTER 3. Durability of Wood Flour-Plastic Composites Exposed

to Accelerated Freeze-thaw Cycling

Part I. Rigid PVC matrix ........................................................ 68
Abstract ................................................................................... 69
Introduction .............................................................................. 71

vi

CHAPTER 4.

CHAPTER 5.

CHAPTER 6.

Experimental ........................................................................... 75

Materials ....................................................................... 75
Compounding and Extrusion of Composites ................. 76
Freeze-Thaw Cycling .................................................... 77
Modiﬁed Freeze-Thaw Cycling ..................................... 78
Property Testing ........................................................... 79
Statistical Analysis ........................................................ 80
Microscopy ................................................................... 81
Results and Discussion ........................................................... 82
Effect of Wood-Flour Content ....................................... 82
Effect of Number of F reeze-Thaw Cycles ..................... 85
Modiﬁed F reeze-Thaw Cycles ...................................... 88
Scanning Electron Micrographs .................................... 91
Conclusions ............................................................................. 94
References......................................... ..................................... 96

Durability of Wood Flour-Plastic Composites Exposed
to Accelerated Freeze-Thaw Cycling

Part II. High Density Polyethylene Matrix ............................ 99
Abstract ................................................................................. 100
Introduction ............................................................................ 101
Experimental ......................................................................... 103
Materials ..................................................................... 103
Compounding and Extrusion of Composites ............... 103
Freeze-Thaw Cycling .................................................. 104
Property Testing ......................................................... 105
Statistical Analysis ...................................................... 105
Porosimetry ................................................................ 106
Microscopy ................................................................. 107
Results and Discussion ......................................................... 108
Porosimetry ................................................................ 1 10
Scanning Electron Micrographs .................................. 112
Conclusions ........................................................................... 1 14
References ............................................................................ 1 16
Summary of Findings .......................................................... 1 18
Future Work ................................................................ 120
Appendix .............................................................................. 122

vii

Table 2.1.

Table 3.1.
Table 3.2.

Table 3.3.

Table 3.4.

Table 3.5.

Table 3.6.

Table 4.1.

LIST OF TABLES

The relative content of cellulose, hemicellulose, and lignin in
hardwoods and softwoods. ............................................................ 16

Formulations used for rigid PVCIwood-flour composites ............... 75

Density of rigid PVC/wood-ﬂour composites before and after
exposure to 5 full freeze-thaw cycles ............................................. 82

Changes in dimensional stability of rigid PVC/wood-ﬂour
composites after 5 cycles of full freeze-thaw (WFT) exposure ....... 83

F lexural properties of rigid PVCIwood-ﬂour composites before and
after 5 cycles of full Maze-thaw (WFT) exposure .......................... 85

Effect of the number of freeze-thaw cycles on the ﬂexural properties
of maple ﬁlled rigid PVCIwood-flour composites ............................ 87

Effect of the number of freeze-thaw cycles on the ﬂexural properties
of pine ﬁlled n'gid PVCIwood-ﬁour composites ............................... 87

Changes in the physical and mechanical properties in HDPE
composites before and after exposure to freeze-thaw cycling. ....109

viii

Figure 2.1.

Figure 2.2.

Figure 2.3.

Figure 2.4.

Figure 2.5.

Figure 2.6.
Figure 2.7.

Figure 2.8.

Figure 2.9.

Figure 3.1.

Figure 3.2.

Figure 4.1.

LIST OF FIGURES

The three surfaces of wood, transverse, radial, and tangential.
Also an annual growth ring is shown with the
lighter earlywood and darker Iatewood. . ........................................ 12

The interior wood of a tree is known as the heartwood, and the
surrounding lighter colored wood is the sapwood .......................... 13

Fiber tracheids appear like boxes on the transverse side of
softwoods and have small holes, or pits, connecting them. The
large boxes to the left of the image are earlywood, whereas the
smaller boxes on the right are from the latewood .......................... 14

The structure of softwoods (left) is much more simple than that of
hardwoods (right), which contain more types of cells. The large
holes in the hardwood are the vessel elements for conduction ...... 14

The relative difference in sizes between a softwood ﬁber tracheid,
hardwood ﬁber and hardwood vessel element ............................... 15

The structure of polymers commonly used in WPCs ..................... 40

A cross section of an extruder showing all of the components
including the screw down the middle of the barrel ......................... 53

A schematic drawing of ﬂexural testing with variables labeled for
calculating ﬂexural strength (MOR) and stiffness (MOE) ............... 58

The stress-strain curve or load-deﬂection curve that results from
ﬂexural testing of strength and stiffness ......................................... 60

The effect of each freeze-thaw cycle portion on the loss of
ﬂexural properties of rigid PVC ﬁlled with 100 phr wood ﬂour:
(a) maple and (b) pine ................................................................... 89

SEM micrographs of rigid PVC ﬁlled with 100 phr maple wood flour:
(a) unexposed control, (b) sample exposed to ﬁve full freeze-thaw
cycles (water, freezing, and thawing), and (0) sample exposed to
ﬁve water-only cycles (W cycles without freezing) ......................... 93

The effect of freeze-thaw cycling on porosity: (a) maple-HOPE
composites and (b) pine-HDPE composites ................................ 111

ix

Figure 4.2. SEM micrographs of HDPE ﬁlled with maple wood ﬂour: (a)
unexposed control and (b) sample exposed to 15 freeze-thaw
cycles (WFT including water, freezing and thawing). ................... 113

.nt

...
D.
.r.
O d
all
I
t.
l
f
.5.
\.
a .
A I
I
I

.i
or 0. 3,
. . E. .
rt. .4» 7‘1.
’b ll a.
It .1 .r
1" I I. it
. l
.l- l‘ . Ia . ’. .Vw ( t
. . E If
D \I l ‘1‘ 0 .
(v t
. i. .
I l
.I
. “I r. .
o
. r a, no
I .‘
t 1 I .
, l .4
I i J . . not.
0 IO ’ E I l ‘ ..
III ‘ r I I
s l
. r .
..
. . r .
I: . u‘b 9'
J .I4 p.
u. I .i I
..I a r .
.V . . an. .
i.
. .1 .
v E
r . 5.
.l \
. .
a-

‘1‘

’5

CCA
EMC
EPA

HDPE
LDPE
LVL
MDF
MF
MOE
MOR
OSB
PE
PF
phr
PP
PS
PVC
RH
SEM
UF

LIST OF ABBREVIATIONS

Chromated Copper Arsenate (wood preservative)
Equilibrium Moisture Content

(United States) Environmental Protection Agency
Freeze-Thaw only cycle, without water exposure
High-density Polyethylene

Low-density Polyethylene

Laminated Veneer Lumber

Medium Density Fiberboard

Melamine Formaldehyde

Modulus of Elasticity (stiffness)

Modulus of Rupture (strength)

Oriented Strand Board

Polyethylene (high-density or low-density)
Phenol Formaldehyde

Parts per hundred parts of resin

Polypropylene

Polystyrene

Polyvinyl Chloride

Relative Humidity

Scanning Electron Microscopy

Urea Formaldehyde

xi

UV - Ultraviolet (rays or light)

W - Water-only cycle, without exposure to freezing
WFT - Water-Freeze-Thaw cycle, full cycle exposure
WPC - Wood-Plastic Composite

xii

Chapter 1

INTRODUCTION

Wood-plastic composites (WPCs) are relatively new products that have
been developed in part due to health concerns and the phasing out of other
products. WPCs are unique because they can be easily modiﬁed and customized
depending on the requirements of any given application. Some already prevalent
applications of WPCs include decking, docks, landscaping timbers, fencing,
signposts, picnic tables, boardwalks, flooring, railroad ties, playground
equipment, window and door frames, and construction molded atrtomobile
interiors, just to name a few (1-8). The combination of properties from plastics,
wood and other additives can create a composite for seemingly any application
and production has been growing rapidly in the United States (1-3,9-11).

Wood-plastic composites make use of would-be wood wastes as ﬁllers or
reinforcing agents in a polymer matrix. The addition of wood ﬁbers to the neat
polymer enhances the stiffness, lessens the cost compared to mineral reinforcing
agents, and cause less wear on processing equipment because wood is more
ﬂexible (12—19). The plastic portion of the composite is believed to enhance the
resistance to moisture, fungus, bacteria and insects which wood is susceptible to
(1,2). Additives can also be mixed into WPCs, which can further enhance the
properties. Lubricants and coupling agents are the most common additives that

are used to ease processing and enhance mechanical properties.

Wood ﬁbers are naturally attracted to water, whereas plastics are oil-
based and are not miscible with water (20—23). Because of these differences it is
difﬁcult to obtain bonding between the wood and plastic components. Coupling
agents are used to enable and enhance the bonding between the hydrophilic
wood ﬁbers and hydrophobic polymer matrix (20—23). The coupling agent acts by
bonding to the wood and make it appear more plastic-like (21-23). Coupling
agents are most commonly used with polyoleﬁns, such as ethylene or propylene
based polymers (20-23). Coupling agents can be created from several sources,
but commonly involve organic agents, the most common of which are maleated
and acrylic acid-functionalized polyoleﬁns (21-23).

Due to WPC’s desirable combination of attributes, the wood-plastic
composite industry is expected to become a major beneﬁciary of the decision
made by the Environmental Protection Agency (EPA) to stop producing
Chromated copper arsenate (CCA) treated lumber at the end of the year 2003
(24,25). At the end of their service life, CCA treated lumber structures will need to
be replaced by another product with equal properties and superior environmental
preservation. Although CCA is no longer for sale, there are other pressure
treated wood products available. Like CCA, there is concern that these products
may leach chemicals from the wood, posing health hazards to humans and the
surrounding aquatic or terrestrial environment (26). WPCs can yield properties
similar to that of wood and adhesive-bound wood composite products, but with
fawer environmental drawbacks. Adhesive-bound composites, such as
particleboard, medium density ﬁberboard (MDF), or oriented strand board (088)

are frequently bound with formaldehyde-based resins that have concerns about
carcinogenous formaldehyde being emitted from the boards once in use and
even later as a waste (23,27). The health and environmental concerns raised by
pressure treated wood and adhesive-bound wood composites are not present
when considering the use of WPCs. Consequently, wood-plastic composites offer
a healthy alternative for both humans and the environment.

However, many challenges still lie ahead in the thriving wood-plastic
composite market. The emerging status of this product dictates the need for
more and varied research. These relatively new products are increasingly
common in both interior and exterior applications, many of which are listed
above. These applications often involve ground contact or above ground use
where there is a risk of material deterioration. When in ground contact there is a
concern of degradation from biological agents such as fungi and subterranean
termites (4,5,28). Exterior applications also involve exposure to ultraviolet (UV)
rays in sunlight and moisture from precipitation. Wood ﬁbers added into a
polymer matrix have been shown to act as chromophores, since they accelerate
photodegradation compared to neat polymers (29,30). Among the degradative
effects to the WPCs were color changes, increased brittleness, decreased
elasticity, and reduced impact resistance (29). These effects can be as
insigniﬁcant as aesthetic color changes or as drastic as failure of the product that
can lead to injury and subsequent replacement of the degraded material.
Therefore, it can be deduced that WPCs are as susceptible to degradation as

solid wood, but there is much more potential for enhancement in WPCs because
of the ease of modifying formulations.

Some types of exposure have been investigated and evidence has shown
that there is a potential for degradation of wood-plastic composites via fungi,
termites and UV light exposure (4,5,28-30). Other types of exposure, such winter
climate exposure, have yet to be investigated and could result in other forms of
degradation for WPCs. Considerable information is available on the durability of
WPCs exposed to warm climate conditions and improvements can be made in
the future. However, there is no information in the open literature on the freeze-
thaw durability of wood-plastic composites. The durability of WPCs to freeze-
thaw actions is vital because they are currently in exterior use in colder regions
like the northern United States and Canada. The long-term response of wood-
plastic composites to environmental inﬂuences would ideally be evaluated by real
time observations of the materials in natural conditions for several years.
However, this process would be much more extensive, involved, expensive, and
would possess more inherent variation. The control of the accelerated laboratory
tests allow for control of conditions to be more representative of typical conditions
and the data can be obtained in a more timely manner. Additionally, the
laboratory testing allows for comparison between factors that would not be
possible under inherently variable exterior conditions. These reasons
demonstrate that accelerated laboratory freeze-thaw tests are more practical and

will be utilized here.

The results of this study are expected to show that wood-plastic
composites are a durable material that is suitable for exterior applications in cold
northern climates. The composites should perform as well as or better than
current wood products due to the increased fungal and moisture resistance from
the plastic component. Mechanical, physical, and aesthetic properties are
expected to continue to be dependant upon composite composition and
decrease with increasing numbers of freeze-thaw cycles. Properties of concern
for losses include the strength, stiffness, and dimensional swelling. The property
losses will likely manifest themselves from a loss in adhesion between the wood
and plastic that will be observable via losses in density, increased porosity, and
microscopic cracks visible with scanning electron microscopy (SEM). This work
will hopefully lead to future work that can reduce or cause property losses to
plateau over time.

As with other previously investigated agents, there is a potential for both
aesthetic and structural failure of the composites exposed to freeze-thaw
conditions. Since adding wood ﬁbers already increases the brittleness, there is
the concern of further increased brittleness, which often coneleates with
decreased mechanical properties, after exposure to cold and freeze-thaw
conditions (18,19,31-33). Accelerated cyclic freeze-thaw testing will yield
valuable information on the effects of cold climate conditions. This can help in
assessing the possible service life of WPCs used in outdoor applications due to
decreased surface quality and mechanical properties. Freeze-thaw exposure

may result in diminished mechanical properties of these materials that are

already being used in outdoor applications that require the retention of these
properties, such as decking.

This study will enhance the understanding of the durability of WPCs
exposed to accelerated laboratory freeze-thaw cycling. This study will
encompass two common polymers, polyvinyl chloride (PVC) and high density
polyethylene (HDPE). Furthermore, varied wood mesh sizes, wood contents,
maleated polyethylene content in HDPE, and two wood species (hardwood
maple or softwood pine) will be investigated. This will aid in enhancing wood-
plastic composites so that they will have a prolonged service life. The
conclusions from this study will result in enhanced outdoor durability for WPC

building products.

OBJECTIVES

The main goal of this study was to investigate the retention of mechanical

properties of extruded wood-plastic composites exposed to accelerated freeze-

thaw cycling meant to simulate outdoor exposure in northern North America. This

goal was fulﬁlled by addressing the following objectives:

Determine the mechanical (ﬂexural) properties of various compositions of
HDPE and PVC composites before and after exposure to accelerated
laboratory freeze-thaw cycling.

Establish relationships between mechanical properties of composite
samples that have been exposed to varying numbers of accelerated
freeze-thaw cycles.

Examine the causation for mechanical property changes before and after
exposure to freeze-thaw cycling as it relates to portions of exposure.
Evaluate changes in dimensional stability, density and porosity before and
after exposure to freeze-thaw cycling.

Examine the composites before and after exposure to freeze-thaw cycles
for cracking and loss of adhesion on a microscopic level using scanning

electron microscopy, SEM.

REFERENCES

u—L

. C. Clemons (2002) Wood-plastic composites in the United States: the
interfacing of two industries. Forest Pmd. J. 52(6):10-18.

2. J. Patterson (2001) New opportunities with wood-ﬂour-foamed PVC. J. Vinyl
Addit. Technol. 7(3):138-141.

3. Anonymous (2002) Decking sales strong for 1‘t half 2002: Producers continue
to add capacity. Natural & Wood Fiber Composites 1(8):1.

4. M. Mankowski and J.J. Morrell (2000) Patterns of fungal attack in wood-
plastic composites following exposure in a soil block tests. Wood Fiber Sci.
32(3):340-345.

5. RI. Morris and P. Cooper (1998) Recycled plastic/wood composite lumber
attacked by fungi. Forest Prod. J. 48(1):86-88.

6. Anonymous (2002) Is fencing the next frontier? WPCs making limited inroads.
Natural 8. Wood Fiber Composites 1(8):3.

7. Anonymous (2002) Products Natural & Wood Fiber Composites 1(10):6.

8. M. Alsentzer (2002) USPL gets RR contract: 10,000 ties for Chicago transit.
Natural & Wood Fiber Composites 1(10):7.

9. J. Patterson (2000) Rigid PVC in the new millennium: Innovations,
applications and properties. Proceedings of Vinylitec 2000, sponsored by
Vinyl Division-SPE, Philadelphia Section. Philadelphia, PA October 11-12,
2000.

10. R. Donnell (2000) Looks like decking. Timber Processing Nov. 2000 : 28-30.

11.P. Mapleston (2002) It’s one hot market for proﬁle extruders. Modem Plastics
June 2002 : 49-52.

12.M. Kazayawoko, J.J. Balatinecz, and L.M. Matuana (1999) Surface
modiﬁcation and adhesion mechanisms in wood ﬁber-polypropylene
composites. J. Mater. Sci. 34:6189-6199.

13.G.E. Myers, I.S. Chahyadi, C.A. Coberly, and 0.8. Ermer (1991) Wood
ﬂour/polypropylene composites: Inﬂuence of maleated polypropylene and
process and composition variables on mechanical properties. Int. J. Polym.
Mater. 15:21-44.

14.S. Takase and N. Shiraishi (1989) Studies of composites from wood and
polypropylenes. II. J. Appl. Polym. Sci. 37(3):645—659.

15.C. Joly, M. Kofrnan, and R. Gauthier (1996) Polypropylene/cellulosic ﬁber
composites: chemical treatment of the cellulose assuming compatibilization
between the two materials. J. Macromol. Sci. - Pure and Appl. Chem.
A33(12):1981-1996.

16. L.M. Matuana, R.T. Woodhams, J.J. Balatinecz, and CB. Park (1998)
Inﬂuence of interfacial interactions on the properties of PVC/cellulosic ﬁber
composites. Polym. Comp. 19(4):446-455.

17.L.M. Matuana, J.J. Balatinecz, and CB. Park (1998) Effect of surface
properties on the adhesion between PVC and wood veneer laminates. Polym.
Eng. Sci. 38(5):765-773.

18. L.M.-Matuana, C.B. Park, and J.J. Balatinecz (1997) The effect of low levels
of plasticizer on the rheological and mechanical properties of polyvinyl
chloride/newsprint-ﬁber composites. J. Vinyl Addit. Technol. 3(4):265-273.

19. F. Mengeloglu, L.M. Matuana, and J.A. King (2000) Effects of impact
modiﬁers on the properties of rigid PVCIwood-ﬁber composites. J. Vinyl Addit.
Technol. 6(3):153—157.

20. R.J. Hernandez, S.E.M. Selke, and JD. Culter (2000) Plastics Packagim:

Prggrties, Processing, Application, and Regulations. Carl Hanser Verlag.
Munich, Germany.

21.J.Z. Lu, 0. Wu, and HS. McNabb Jr. (2000) Chemical coupling in wood ﬁber
and polymer composites: A review of coupling agents and treatments. Wood
Fiber Sci. 32(1):88-104.

22.0. Li and L.M. Matuana (2003) Effectiveness of maleated and acrylic acid-
functionalized polyoleﬁn coupling agents for HDPE-wood ﬂour composites. J.
Therrnoplast. Campos. Mater. 16(6):551-564.

23.T. Sellers Jr. (2001) Wood adhesive innovations and applications in North
America. Forest Prod. J. 51(6):12-22.

24.Anonymous (2002) Arsenic-treated wood to be phased out of residential use.
Chemical and Engineering News Feb. 2002, p. 27.

25.Anonymous (2002) Wood substitutes beneﬁt from EPA move. Plastics in
Canada eNewsletter March 13, 2002.

26. K. Stook, T. Tolaymat, M. Ward, B. Dubey, T. Townsend, H. Solo-Gabriele,
and G. Bitton (2005) Relative leaching and aquatic toxicity of pressure-treated
wood products using batch leaching tests. Environ. Sci. Technol. 39(1):155-
163.

27.Anonymous (2001) Formaldehyde in sawdust from composite wood products.
BioCycle 42(1):23-24.

28.W. Chetanachan, D. Sookkho, W. Sutthitavil, N. Chantasatrasamy, and R.
Sinsennsuksakul (2000) PVC/wood: A new look in construction. Proceedings
of Vinylitec 2000, sponsored by \ﬁnyl Division-SPE, Philadelphia Section.
Philadelphia, PA. October 11-12, 2000.

29. L.M. Matuana and DP. Kamdem (2002) Accelerated ultraviolet weathering of
PVC/wood-ﬂour composites. Polym. Eng. Sci. 42(8):1657-1666.

30. L.M. Matuana, DP. Kamdem, and J. Zhang (2001) Photoaging and
stabilization of rigid PVCIwood-ﬁber composites. J. Appl. Polym. Sci.
80(11):1943-1950.

31.B.D. Park and J.J. Balatinecz (1996) Effects of impact modiﬁcation on the
mechanical properties of wood-ﬁber thermoplastic composites with high
impact polypropylene (HIPP). J. Thermoplast. Comp. Mater. 9:342-364.

32.C.M. Clemons, A.J. Giacomin, and J.A. Koutsky (1997) Dynamic fracture
toughness of polypropylene reinforced with cellulose ﬁber. Polym. Eng. Sci.
37(6):1012-1018.

33.G.A. Cooper and MR. Piggot (1977) Cracking and fracture in composites.
Proceedings of the 4‘” lntemational Conference on Fracture, sponsored by
the lntemational Congress on Fracture (ICF). University of Waterloo, Canada.
June 19-24, 1977.

10

Chapter 2

BACKGROUND

SOLID WOOD PRODUCTS

Wood is one of the most abundant and readily available resources that
persons all over the world have for building materials. However, there are several
factors that need to be addressed when considering the use of solid wood.
Among the basic concepts to understand how to appropriately utilize wood are
the variations in structure between species, effects of atmospheric conditions, the

susceptibility to biological organisms, and preservative chemicals.

Tym of Wood
Wood is deﬁned as a cellular, ligno-cellulosic material (1), but

encompasses many different species, that in actuality are not very similar.
Different species, growing conditions, and defects all result in products that have
vastly different structures and properties, which make every piece of wood
unique. Most commonly, broad delineations are made into hardwoods and
softwoods, but there are numerous differences within these categories as well
(2). Hardwoods are commonly known as deciduous trees that shed their leaves,
whereas softwoods are evergreen trees that have needles, which can be

retained for several years.

11

Wood is an anisotropic material, meaning that it has 3 different properties
depending on the orientation of the wood (1-3). The longitudinal, tangential and
radial surfaces shown if ﬁgure 2.1 all display different components of wood and
therefore have different properties (2,3). Differences are prevalent even within
wood obtained from the same tree. The annual rings visible on the transverse
surface of wood represent changes in wood structure from earlywood to
latewood (ﬁgure 2.1) and the center portion of the tree known as heartwood is
different from the surrounding sapwood (ﬁgure 2.2) (1,2). There are also several
cell types, particularly in hardwoods that wnnot all be distinguished with the
naked eye (2). Changes in grain, knots, decay and other defects can also affect

the appearance and properties of wood.

Transverse
Surface

Annual growth increment
(annual ring)

 

Figure 2.1 The three surfaces of wood, transverse, radial, and tangential. Also an
annual growth ring is shown with the lighter earlywood and darker Iatewood.
Image taken from Haygreen & Bowyer (2).

// SAPWOOIL

\‘

 

Figure 2.2 The interior wood of a tree is known as the heartwood, and the
surrounding lighter colored wood is the sapwood.
Image modiﬁed from Society of Wood Science and Technology.
http://www. swst.org/teachlteach1lstructure1 .ppt#1

The structure of wood is most easily compared to a bunch of straws that
are stacked in bundles, end to end (1). This structure allows for conductance of
water and nutrients within the tree. Softwoods are mostly composed of ﬁber
tracheids, which stack in this manner, that are strength-giving ﬁbers with the
ability to conduct water (1). Figure 2.3 shows the thin walls of softwood ﬁbers and
the numerous connections between the ﬁbers, or pits. Hardwoods do not have as
many as many pits as the softwoods due to their variety of components (1).
Softwoods' mostly homogenous structure, straight-grain and long ﬁbers make it
ideal for uses such as construction beams, poles, pilings, decks, siding, exterior
moldings and premium papers (2). These uses necessitate that it is graded on

strength with appearance as a secondary concern (2).

 

Figure 2.3 Fiber tracheids appear like boxes on the transverse side of
softwoods and have small holes, or pits, connecting them. The large boxes to
the left of the image are earlywood, whereas the smaller boxes on the right
are from the Iatewood. Image taken from Haygreen and Bowyer (2).

 

 

 

Softwood (20X)

Hardwood (20X)

 

 

Figure 2.4 The structure of softwoods (left) is much more simple than that of
hardwoods (right), which contain more types of cells. The large holes in the
hardwood are the vessel elements for conduction.

Image modiﬁed from Society of Wood Science and Technology.
http:/Mww.swst.orglteachlteach1lstructure1 .ppﬂt1

14

 

Hardwoods are a much more diverse group than softwoods, with varied
structures and appearances because they have many more components (1,2).
The variety of cells (as seen in ﬁgure 2.4) in hardwoods means that ﬁbers are
more randomly dispersed and are generally shorter (1). Hardwood ﬁbers are also
not used for as much conduction since there are vessels that form conduits for
this purpose (1). A comparison of the size of the softwood tracheid, hardwood
ﬁber, and hardwood vessel ebments is shown in ﬁgure 2.5.

 

._ W
Softwood Fiber Tracheid
‘- ‘t‘

 

Hardwood Fiber

  

Hardwood Vessel Element

Figure 2.5 The relative difference in sizes between a softwood ﬁber tracheid,
hardwood ﬁber and hardwood vessel element.
Image modiﬁed from Haygreen & Bowyer (2).
Structures within hardwoods that are used to verify species include perforation
plates between these vessels, as well as the location of the vessels in diffuse or
ring-like patterns (2). The variations in hardwoods make them desirable for uses
where their uniqueness can be observed for decorative purposes such as
furniture, cabinets, ﬂooring and paneling. However, hardwoods can be more
dense than softwoods, which makes them more desirable for uses as railroad

ties and less popular species are frequently used for industrial pallets (2). For

15

most applications, though, it is most important for hardwoods to be free from
defects and have clear surfaces.

All wood is composed of three primary constituents. Cellulose,
hemicellulose, and lignin are all present in wood, although to differing extents
depending on species and growing conditions (2). Table 2.1 shows the relative
content of these three components for hardwoods and softwoods. Cellulose is
about half of the weight of any wood and is comprised of glucose units produced
during photosynthesis (2). Cellulose is not a large molecule and can only reach
about 10 pm in length and 8 A in diameter (2). Hemicellulose closely resembles
cellulose, but is comprised of different sugars and has a lower molecular weight,
in part, due to chain branching (2). Lignin is essentially the glue that holds the
wood structure together and is found throughout the wood (2). Less than one
percent of the tree is comprised of extractives (2,4). These extractives give each
tree a unique odor, color, and appearance (2,4). Extractives consist of non-
structural elements within the wood and include elemental compounds or
chemicals that help contribute to natural resistances and tolerances of the wood

(4,6).

Table 2.1 The relative content of cellulose, hemicellulose, and lignin in
hardwoods and softwoods. Table modiﬁed from Haygreen and Bowyer (2).

 

 

 

 

 

 

 

 

Type Cellulose Hemicellulose Lignin
Hardwood 40—44 % 15-35 dry wt.% 18-25 %
Softwood 40 — 44 % 20 — 32 dry wt.% 25 — 35 %

 

16

 

Progrﬂes of Wood

Important properties for consideration with wood include compressive,
tensile, and ﬂexural strength and stiffness, impact strength, durability, shear
strength and toughness (1). However, wood is anisotropic and will behave
differently depending on the direction in which it is loaded for property testing (1-
3). Other materials that provide alternatives to wood such as plastic, metal and
cement are isotropic so the mechanical properties are typically the same in all
three directions (2), which can make them more desirable than solid wood. Since
wood is a naturally occurring material there are several other factors that can
affect these already differing properties. Among the concerns for mechanical
properties in wood are exposure to moisture, temperature, chemicals, and
loading, as well as inherent defects (2,3).

Perhaps one of the greatest concerns when using wood is the exposure to
moisture since properties vary with the atmospheric and wood moisture contents
(1,3). Since wood is a hygroscopic material, it will obtain and release moisture
from its surroundings (1). This is a well known phenomenon with wood and often
results in the shrinking and swelling of its dimensions (2). Changes in the
presence of moisture in wood below about 25% moisture content, also known as
the ﬁber saturation point (FSP), inﬂuences the strength and elastic properties of
wood (2). The lower the moisture content below FSP, the higher these properties
will be since the water is no longer present to obstruct ﬁber bonding (2). Different
species differ in their ability to gain and release moisture because of their

differing structures, so some species may be more desirable where moisture

I7

changes will be present (3). The necessary mechanical and dimensional
properties for wood products in their end use must be related to what moisture it
will be exposed to in order to determine what species best ﬁts those needs.

While indirectly related to the ability to hold moisture, temperature also
affects wood properties. High temperatures decrease mechanical properties and
the presence of a high moisture level makes wood more sensitive to property
loss at high temperatures (2). Generally exposure over 100°C will result in some
destruction of the wood structure (2). Exposure to extreme conditions like steam
can result in permanent irreversible damage to the structure of wood and loss of
the majority of its strength (1). In addition to the concerns for property loss from
moisture changes, the effects from temperature changes can make the use of
wood difﬁcult without proper scnrtiny and observation.

Exposure to alkaline and acidic environments is another weakness of
wood, although it is more resistant to acidity than steel (2). Chemicals are known
to deteriorate the cell walls in wood and result in a loss of strength since the cell
walls provide structural support (2). The insertion of fasteners into wood can also
result in more localized deterioration from the reaction of metal with its
environment (2). Nails, bolts and screws can interact with moisture and wood-
bound extractives to degrade the wood and lose their holding power, but coatings
are used to inhibit this trend (2). The environment to which wood is exposed to
should be carefully monitored to ensure that fasteners are intact and that wood
has not degraded, otherwise a material with superior chemical resistance is

necessary.

18

Applying a continual load or stress to wood over extended periods of time
can result in a loss of properties or fatigue (2). Time is a foe of most materials
because it allows for so many changes to take place. The invasion of
microorganisms, drastic temperature changes, changes in moisture content,
exposure to chemicals, and exposure to other degrading agents are all likely
given enough time (2,3). All or most of these effects will take place during the
service life of wood and necessary actions, such as chemical treatment, need to
be taken to ensure against these effects.

Natural defects in wood can be a result of animals, insects, decay, or
growth history and also can cause mechanical property loss in wood products
(2,3). Factors in the history of a tree’s growth, like broken branches can cause
knots or other defects (2,3). Knots are the most common defect that decreases
wood strength (2). Knots and human error in processing of wood can also result
in abnormal grain patterns, which further reduce properties (2,3). Living creatures
can harm trees in many ways like by inﬂicting wounds and damaging wood by
burrowing or causing pockets of pitch to form to repel insects (3). Decay can also
occur within a tree, primarily from the presence of fungi (2,3). Decay is prohibited
in structural lumber and may be much worse than anticipated due to its ability to
decrease impact strength and toughness (2). Most of these natural defects are
preventable, but require appropriate recognition and treatment to ensure that

wood is a practical material for building uses.

19

Deterioration of Wood

Since wood is part of a natural and organic environment, it is susceptible
to degradation from many types of organisms, some of which were previously
mentioned. The sapwood portion of trees is most commonly used and is
susceptible to deterioration from fungi, insects, bacteria, and marine borers when
exposed outdoors (2,4-6). Some wood species possess an inherent resistance to
these factors from extractives that the tree produces, but these are typically held
in the less desirable heartwood (4,6). Additionally, these naturally resistant trees
are not readily available in most locations and are generally in short supply (5).
Since the susceptible sapwood portion of the tree is used most often, it is
imperative that proper chemical preservation be done to prevent against
unnecessary wood degradation. Selecting a readily treatable species and
selecting proper preservatives can help ensure against untimely failure due to
exposure in biologically adverse environments (2,5). There are also concerns of
non-biological agents that affect wood, particularly heat and ﬁre (2). Other
concerns for wood degradation were previously discussed and include chemical
exposure, metal corrosion, as well as surface weathering and W exposure over
time (2).

The most common cause of wood deterioration is fungi because they can
lead to molds, stains and structural decay (6). Fortunately, fungi require more
than just the food supply from the wood to survive. Fungal growth requires warm,
moist, and oxygen rich environments to survive (2,6). Given proper storage,

seasoning and handling procedures there is no reason that wood should be

20

attacked by fungi (6). However, this is not always the case and wood products
are often left susceptible to fungi. Decay fungi collapse the wood structure and
soft-rot fungi penetrate into the surface of the wood to cause losses in strength
(2). Decay fungi are divided into further categories that attack either the cellulose
and hemicellulose portions of the cell or the lignin that holds the cells together
(2). Regardless of which type, decay initially reduces the ability to withstand
impact, also known as toughness, and then proceeds to reduce ﬂexural strength
properties until there is eventually no strength left in the wood (6).

Wood is also exposed to a wide variety of insects when in outdoor use.
Insects can attack standing trees, cut logs, or lumbar products and include
carpenter ants, termites, beetles, and some species of bees (2,3,6). These
insects cause a variety of holes, burrowing galleries, stains and pitch or gum
pockets (6). Small creatures in seawater can have similar effects on wood
products used in marine applications. Wood used for posts or pilings in marine
exposure are susceptible to attack from shipworrns and pholads that are similar
to mollusks, as well as crustaceans (2,6). These creatures, particularly
shipworrns, attack untreated wood so rapidly that it will not even last a year in
salty or brackish waters (2,6). Since no wood in North America is naturally
resistant to these marine borers, chemical treatment is necessary to prevent their
degradation (2,6). Due to wood’s inherent susceptibility chemical preservation is

essential.

21

Preservation of Wood

Wood preservatives are widely used to prevent against the effects of the
above-mentioned deteriorating organisms. The American Wood Preservers
Association or AWPA, and the EPA regulate the use of wood preservatives since
they are considered pesticides and can be hazardous if improperiy utilized (5).
The chemicals used for treatment are divided into two categories - oilbome and
waterborne preservatives. Oilbome preservatives use an oil to transport the
preservative chemicals into the wood and have no interaction with the wood
structure itself (3). However, waterborne preservatives do interact with the cell
walls in the wood because they are transferred by water (2,3).

Among the most common preservatives used to prevent against fungal
and insect attack are pentachlorophenol, creosote and Chromated copper
arsenate or CCA (2,7). Pentachlorophenol is not usually categorized as either
oilbome or waterborne because it can be used with a variety of volatile carriers
(2,5). Creosote is the most common oilbome preservative and has a high toxicity,
low volatility, easily penetrates and has a long-term record of usage (5). Like
creosote, oilbome preservatives are effective at resisting weathering and decay
they also emit odors, cannot be painted, have a dark brown-black color, are not
clean to handle or touch, and have poor ﬁre performance (5).

There are a variety of common waterborne preservatives, such as CCA,
which involve mixtures of copper, chromium and arsenic that are used when
cleanliness, low odor emissions and paint-ability are of concern (5). However,

waterborne preservatives are water-based and have concerns for leaching,

22

strength reduction, and need to be dried before being put in end-use (2). These
products need to be dried before use in any application because changes in
moisture content affect the dimensional stability of wood, as was previously
described (2). Other pest speciﬁc waterborne preservatives are also becoming
popular, such as the use of boron treatments for resistance against subterranean
termites (5,8). Inevitably, the type of exposure that the wood product will be used
in determines the necessary chemical and degree of retention (2,5).

For example, chemical treatment to resist attack by marine borers is an
involved process compared to the treatment of wood used in terrestrial exterior
applications. Normally a dual treatment of a waterborne preservative like ACA or
CCA, and then treatment with creosote until no more chemical will be absorbed
is necessary to ensure proper marine protection (2,5,6). The efﬁcacy of this
treatment is still not absolute since climactic conditions and marine borer
presence vary greatly around the United States (2,6). These differences mean
that treated pilings will have differences in preservative leaching rates and
subsequent differences in service lifetime (2,6).

Leaching is a concern when using most chemicals for wood preservatives,
due to the inability of wood to retain the chemicals over time. Leaching is
encouraged by the presence of rain and moisture in outdoor exposure and many
studies have focused on the ability of various chemicals to be retained (9,10). In
1984 the EPA released an assessment of the use of preserved wood products
and came to the conclusion that preservative chemicals are effective, although

they can adversely affect human health by direct or indirect contact, and that

23

stricter regulations are needed for handling and treating processes to protect
people and the environment (2). Preservatives may be effective, but there is no
long term solution to maintaining our environmental conditions and using
chemically-treated solid wood products.

Additionally, waterborne preservatives, unlike oilbome preservatives,
interact with the cell walls of the wood and thereby can affect the mechanical
properties (3,5). High temperature and pressures used for treatment, as well as
the presence of high levels of chromium can cause reactions between chemicals
and the woods cell walls, which can lead to decreased mechanical properties
(2). Mechanical properties are all affected differently, but generally ﬂexural
strength, tensile strength perpendicular to the grain, impact strength and energy
absorption related properties are the most affected, with losses ranging from 0%
to 50% under less desirable conditions (3). The main factors that can contribute
to a loss in these mechanical properties are the treatment method, retention,
preservative chemistry, drying conditions before and after treatment, size of
material, quality or grade of material, species, and the use of incising (3). The
use of incising, or puncturing the wood surface to improve treatability and
retention, are of utmost concern for these property losses (3).

As well as the chemical’s toxicity, the method of application and the
degree to which the chemical penetrates and is retained in the wood is essential
to the properties and efﬁcacy of the treatment (2,5). There are a variety of ways
that chemicals can be applied to ensure proper penetration, which will not be

covered here. Obtaining thorough protection of the wood is dependant upon

24

many of the same factors that affect strength retention including the species,
penetrability of heartwood, proportion of heartwood/sapwood, and moisture
content (5). Wood species vary in their ability to resist penetration by chemicals
and the inner heartwood portion of most species is less receptive than the
sapwood (2). Consequently, proper species selection can be helpful in

maintaining strength properties while adding resistance to biological agents.

Summary
The use of solid wood products depends on a variety of factors, most of

which cannot be predicted due to the organic nature of wood. Wood can come
from several types of trees from numerous locations and there will be few
commonalities between them, although they are all susceptible to several factors
in outdoor environments as cut lumber or wood products. Factors like moisture,
temperature changes, fungi, and insects are inevitable in outdoor exposure, yet
there is no absolute way to protect wood. Chemical preservation methods are
commme used but involve the use of harmful chemicals that can cause property
loss and often leach out of the wood after treatment. The use of chemical
treatment is not an exact science because the treatability varies between species
and even between logs of the same species. If processed properly solid wood
can be effectively used in countless applications, however the service life can be
cut short from biological organisms. Wood is affordable and readily available
compared to alternatives like steel or concrete, but the inherent variability and

susceptibility may lead to the need to pursue other alternatives.

25

WOOD-BASED COMPOSITES

Due to the inability to obtain solid wood in desired dimensions and the
production of wood wastes from machining and processing, the development of
wood-based composites began. Trees that were originally used in the United
States had much greater dimensions than it is possible to obtain now due to
rotation lengths and demand (2). Consequently, composite products based on
wood, with some of its appearance and mechanical properties were created.
Wood-based composites use wood in a variety of shapes and sizes to enhance
wood utilization and create products in sizes that were previously not possible.
The creation of wood-based composites has resulted in numerous thennoset

resins being developed, as well as several dimensions of wood being used.

Wood Adhesives

Numerous adhesives have been developed in an attempt to better bind
wood pieces together. These resins vary in the way they function, their origins,
solidiﬁcation method, and prevalence in the marketplace. The majority of resins
are organic polymers, which are simple molecules that have been bound
together by functional groups (11). Although natural resins have been used, the
vast majority of these polymers are now synthetic, mostly due to the enhanced
moisture resistance of synthetics (11-13). Synthetic adhesives can also be
designed to be stronger, more rigid, and more durable than the wood that they

are bonding, whereas natural resins can have inherent variability (11).

26

New binders and binding systems are always being developed, but none
have had a signiﬁcant effect in penetrating the market (12). The majority of
adhesives, 59% in 2001, were synthetic amino resins like urea formaldehyde
(UF), melamine formaldehyde (MF), etc. (12). A close second with 32% of the
2001 total was phenol formaldehyde (PF), whereas only 5% of adhesives used
were isocyanates with the rest of the resins having only a negligible inﬂuence
(12). These three resins; UF, PF, and isocyanates, will be discussed in more
detail later. lsocyanates, UF, PF, and MF are all examples of therrnoset resins,
which undergo irreversible chemical changes and cross-link after reaction
(11,13). Therrnoset resins are known for high strength properties, moisture
resistance, chemical resistance, and rigidity under load (11). There is also a
smaller group of thermoplastic resins that are used in these composites, but they
soften under heat and are less common (11).

All adhesives function by two methods, i.) molecular size remains the
same and ii.) molecular size changes through reaction (1). Phenolic resins are
the most common of the second type, where the molecules increase in size until
they are cured, which offers the greatest interaction with the wood and
processing (1). These reactions can take place in a variety of processing
environments. The most prevalent types of processing with wood adhesives
involve radio-frequency curing, steam injection, and continuous or batch platen
pressing (12,14). Once they are adhered by one of these methods, the resins
then solidify to form a composite product. Solidiﬁcation can occur via heat loss,

solvent loss, chemical reaction, or a combination thereof (1). The one problem

27

with resin solidiﬁcation experienced under all these conditions is shrinkage due to
crystallization or resin solvent loss (1). This is a common problem because most
wood adhesives use water as a carrier because it is cheap, non-toxic, and easily
absorbed by wood (11). As the water is evaporated, dimensional changes within
the composite can lead to undesirable stresses within the product (1).

Resins are also categorized by their uses. Structural, semi-structural and
non-structural applications are all determined by the resins ability to transfer load
within the composite and maintain the integrity (11). Adhesives require different
properties, so a unique resin is often created to ﬁt into one of these categories
depending on the application (11,15). The concern is that non-structural
adhesives with the least resistance to service exposure, lowest rigidity and
weakest bonds also have the most tolerance of wood variation, assembly, and
curing conditions (11). Modiﬁcations are done on resins to try to change this.

Adhesives are modiﬁed to improve their workability, mechanical
properties, shelf life, durability, and cost-effectiveness ratio (11,13). This can
involve the addition of extenders, reinforcing ﬁbers, ﬁllers, catalysts, hardeners or
other resins (11). Although natural resins have been mostly replaced with
synthetic resins, they are often still used as components in these types of resin
blends. Natural adhesives that are used for resin modiﬁcations include lignin,
tannin, caseins, soy, animal blood, and pyrolysis oils (12). Components like lignin
and pyrolysis oils can be used as substitutes for up to 50% of PF resins in some
cases (12). Animal blood is considered “indispensable” as an additive for

industrial and construction plywood made with foamed PF resins (12).

28

Synthetic chemicals used in wood adhesives are toxic to people if in a
high enough concentration and exposed for a long enough time (11). Generally,
the cured state of adhesives is considered safe, but this is not always the case.
Resins like PF efﬁciently consume formaldehyde during reaction and have little
concern of gas emission (11). However, UF is not as efﬁcient as reacting and can
release low concentrations of formaldehyde gas that react with proteins in the
body and are suspected to be a carcinogen (11,15).

Urea formaldehyde, UP is typical of the most common group of resins
used in North America (12,13). A reaction between urea and formaldehyde can
bond up to four formaldehyde molecules to the urea before it is neutralized to
produce a ﬁnal resin content of about 65% resin solids (13). UF is the cheapest
of the common resins, but looses its bond strength in the presence of water, is
brittle once cured, has an acidic nature that makes bonding difﬁcult with oily or
high extractive containing woods, and as previously mentioned can emit
formaldehyde gas (1,11,13,15). Because of its affordability and poor water
resistance, UP is typically used for interior applications like hardwood plywood,
medium density ﬁberboard (MDF) in fumiture or paneling, and particleboard used
in furniture or underlayment (1 ,13,14). As with other resins, UF can be modiﬁed.
Modiﬁcation is typically done to enhance its moisture resistance (13).

There are 2 types of phenol formaldehyde, PF, that are commonly used, a
resale and a novolac (1,13). These resins are made of the same components,
however the resale has less phenol than formaldehyde. The proportions of

phenol and formaldehyde are reversed for the novolac (1,13). The resole only

29

needs heat to solidify and has a very limited storage life, conversely since the
novolac lacks formaldehyde it has a very long shelf life, but needs formaldehyde
added to get hardening (1,13). Due to its superior moisture resistance and
strength retention compared to UF, PF is typically used for exterior applications
like plywood and oriented strand board (083) (13,14). PF is a desirable resin
because it is easy to process, low cost, has low formaldehyde emissions, and is
resistant to water, weather, and high temperature (15).

One of the most common isocyanates is polymeric methylene diphenyl
diisocyanate, PMDI (11,13). PMDI is widely used because of its ability to develop
strong and durable bonds even when the wood has a high moisture content,
while possessing a lower volatility and toxicity compared to other isocyanates
(11-13). The use of PMDI is also encouraged because it is free of formaldehyde
and does not need a solvent or additives (13). Diisocyanates like this are
potentially hazardous and highly reactive, but are considered safe once they are
bonded (11).

There are several drawbacks and hazards with wood adhesives, as there
are with wood preservatives. The toxicity and susceptibility concerns (e.g. UF
with moisture) are being addressed and new resin formulations are constantly
being developed. The variety of resins for wood composites means that there are
an equal number of products that can be produced. The following sections on
veneer and particlel‘flake products are a broad overview and do not cover all of

the uses of adhesive-bound composites. The main wood composites that will be

30

covered, in order of manufacturing quantity in 1998, are plywood, OSB,
particleboard, MDF, hardwood plywood, and laminated veneer lumber (LVL) (12).

Veneer Products

Veneer is produced by rolling a log on a lathe with a special cutter and
shaving off thin layers of wood about 1/16th of an inch thick called plies (2,16).
The veneer can then be used to make either plywood or LVL among other
products, although only plywood and LVL will be covered here. Veneer allows
composite materials to have the same appearance as solid wood, but with larger
dimensions and fewer imperfections in wood quality and appearance.

To make plywood, these plies are stacked perpendicular to each other so
that the grains of the wood is alternating on every other level (2,13,14). These
alternating layers allows for good strength properties, improved dimensional
stability, and help increase the isotropy (2,13,14). Softwood plywood has
experienced competition from OSB and has needed to reduce production costs
and improve products to stay competitive (16). A major problem for plywood is
with the raw materials used, as with solid wood, the appearance and quality of
the available wood is decreasing (13,17). Also, because of the raw material
shaped as plies, plywood is conﬁned to the dimensions of roughly 4 by 8 feet
(17). The properties of the plywood are rated on grades A through D and depend
upon the wood species, material quality, orientation of the piles, type and amount

of adhesive, curing conditions, and pressing conditions (2,13,14).

31

Softwood veneer panels, which are preferred for structural applications
compared to hardwood panels, are almost exclusively bound with synthetic PF
resins (12). Hardwood plywood is generally manufactured for decorative, non-
structural uses and because of its applications is typically bound with UF (12,14).
Not all plywood, hardwood or softwood, is comprised entirely of veneer plies.
Often times a ﬁller of particleboard is used for the core of the panel and the
exterior is faced with veneer plies (2,16). This is often done for a major
competitor of hardwood plywood, where MDF is faced with veneer to give it a
desirable solid wood appearance (13).

Laminated veneer lumber (LVL) is also made with wood veneer plies,
although it is not always a panel like plywood (2,18). In LVL plies are all oriented
in the same direction in an effort to simulate solid wood with enhanced properties
(2,13,16—18). Since LVL is a manufactured product, the length can range up to 80
feet, which solid wood cannot easily obtain (17,18). As with plywood, LVL uses
PF resin, but the veneer plies are generally thicker and range from 1/10th to 1l6"'
of an inch (2,12,18). LVL does cost more than solid wood, but it has already been
dried to uniform moisture content, imperfections from growth like knots are no
longer localized, and it can be made in a continuous process to larger
dimensions than solid wood (2,17,18). Because of this, production has been
growing in North America despite the fact that LVL is mostly used for specialized
applications (12). Since many of these applications are load-bearing, the

adhesive used must be reliable and evenly cured (18). LVL is still competitive in

32

the marketplace because of the reduced availability and increasing price of solid
wood that LVL simulates (18).

Plywood and LVL are not the only products made with veneer, although
they are certainly the most common. Other veneer-based products include wood
l-joists, parallel strand lumber, laminated strand lumber, and Com-Ply® Panels.

lnfonnation about these products is available in other publications (2,16-18).

Particle and Flake Products

Wood waste is commonly used for products such as particleboard, but
wood can be made into several different shapes and sizes. Ideally these wood
pieces will be of uniform size and relatively homogeneous. Factors like wood
species and particle type can greatly affect the ﬁnal product (2). To achieve
homogeneity, wood pieces may be processed at the manufacturing plant using
various milling equipment to control the particle size and geometry (2,14).
Products like particleboard and MDF utilize sawdust, planer shaving and other
mill residues that have been screened for compatibility (13,14,16). Flakes are
used to make OSB and are created from roundwood that has been processed
through a ﬂaking or chipping machine to produce thin wafers of wood about 3
inches long and 0.03 inches thick (2,13,14). The emergence of OSB has been
fueled primarily due to the lack of veneer sized logs because OSB can use
essentially any grade or size of tree (13,14). Although not speciﬁcally a wood
waste, the wood used for OSB is generally under-utilized and this is a value-

added usage of wood (13).

33

A close second to the combined production of plywood and LVL is OSB
(12). OSB has been a main competitor of veneer panel products and is not far
behind veneer panels in production (12,13). The ﬂakes are mechanically oriented
to increase properties and typically bound with PF resin, although there is some
use of PMDI generally in the inner layers (2,12-14). There are 3 to 5 layers in an
OSB panel, which like plywood are placed perpendicular to other layers with the
faces being parallel to each other (2,13). The percent resin content used in OSB
can vary within and between panels, but typically ranges from 2 to 6% depending
on the application, with more resin used for higher durability products (12). The
major drawback of OSB is that it lacks dimensional stability and can experience
irreversible thickness swelling (2,13). Despite this, OSB has become well used in
construction for rooﬁng, walls, and ﬂoor sheathing (14). Although commonly used
for products made from particles, the use of a continuous press system is only
recently becoming popular for OSB production (12,14).

The use of wood residues is more common in particleboard or MDF where
smaller wood particles are desired (2,14). Of all the wood particles used in these
manufacturing processes, only about 1/10th comes from virgin roundwood that is
not a waste (2). UF resins are most commonly used for particleboard and MDF,
which combined made up almost as much of the composite market as OSB in
1998 (12). The use of PF is also sometimes used in particleboard for exterior
uses (2). Over 90% of particleboard and MDF products have overlays or ﬁnish

coatings applied to them for uses in furniture (12,14).

34

Particleboard is a general category that can contain products with different
appearances depending upon the particle’s size and geometry, the density it is
pressed to, and the amount of resin used (2,16). Since particleboard is
commonly used for interior furniture, the surface properties are of concern.
Consequently, particleboard is made of 3 to 5 layers where the face is different
than the core (2,13,14). The core of particleboard is comprised of coarser ﬁbers,
while the surface has smaller particles to increase smoothness and strength
(2,14). The particles are pre-coated with the adhesive and are then laid out into a
mat that is processed in a continuous or platen pressing system (2,14). Although
less common, there are also extrusion pressing systems that are used (13,16).

To make MDF, individual ﬁbers or ﬁber bundles are used with a synthetic
resin, in a dry process similar to that used for hardboard (2,13,16). Compared to
particleboard, MDF is more difﬁcult to process because pressurized reﬁners are
needed to produce the pulp that is used (2). Sugar cane stalks, called bagasse,
have also been successfully used to make MDF (16). Although the pulping to
obtain ﬁbers is an added step, otherwise the processes for making MDF and
particleboard are quite similar. In the United States over 70% of the MDF is used
for furniture and cabinetry (2). Unlike particleboard, MDF is more easily
machined, has smoother more paintable surfaces, and does not need edge
banding to mask its appearance (2,13). Variations in raw wood materials are also
better tolerated in MDF production than particleboard although it cannot be too

drastic because of interactions with the UP resin (16).

35

Summag

The development of wood-based composites bound with thennoset resins
has successfully counteracted some of the difﬁculties with solid wood. Composite
products can create large panels and products from small pieces of wood (2).
Also, high percentages, over 70%, of the raw materials are in the ﬁnal product so
there is much less waste (2). In fact wastes from solid wood processing are often
used in particleboard and MDF manufacturing (2,13,14). Composites may be
more expensive than solid wood, but the strength, uniformity, and dimensional
possibilities make it a valuable substitute (2). Solid wood has several previously
mentioned drawbacks, particularly the decreasing quality of trees. Composite
materials are superior because they can offer the same properties, but the quality
is controlled by resins and manufacturing, which are easily alterable instead of
inherently variable growing conditions (2). The main problems that remain for the
development of thennoset bound composites like plywood, OSB, MDF, and
particleboard are the toxicity and cost of raw materials. There are continually
developing adhesives of several types, but the inertia of the market to
incorporate them and continue with the same old adhesives is inconceivable.
Although certainly more customizable and functional than solid wood, there are

still health concerns that need to be addressed for these resins.

36

WOOD-PLASTIC COMPOSITES

Wood-plastic composites (WPCs) are an up-and-coming ﬁeld of products
that alleviate the susceptibility of solid wood and the toxicity of wood adhesives
and preservatives. There are a variety of mineral and natural ﬁllers/ﬁbers that
have been incorporated into composites around the world (13), but the focus
here will be on wood. Essentially anything can be combined into wood-plastic
composites, so the scope needs to be limited to that pertinent to this work. The
variety of polymers, additives, and ﬁllers used in wood-plastic composites means
that they can be open to several types of degradation. Given time to assess
these hazards and customize formulations, wood-plastic composites have the
potential to become durable materials with fewer concerns than current products.
As with wood-based composites, there are countless combinations of materials
that are categorized as wood-plastic composites. This discussion certainly does

not encompass the diversity of products or current research.

Develogment of Wood-Plastic Commit»
WPCs combine the desirable durability of plastics with the cost

effectiveness and strength of natural ligno—cellulosic ﬁbers as ﬁllers or reinforcing
agents (13,19—21). Similar to the particle composites previously discussed, wood-
plastic composites can use virgin, waste, or recycled components (12.14.22). As
of 1998, recycling was still growing slowly, but materials like polyethylene
terephtalate (PET), high-density polyethylene (HDPE), polypropylene (PP), and

37

low-density polyethylene (LDPE) were available as recycled materials (12). The
natural ﬁbers can be from anything ranging from virgin wood, recycled pallets,
recycled newspaper, or other virgin ligno—cellulosic ﬁbers described later (12,14).
This variety of resources can be used because there is no current standard or
restriction on manufacturing WPCs (13).

Although wood-based composite products are typically more isotropic in
nature than solid wood, plastic products are isotropic with no variation in any
direction (2). Unlike the other wood-based composites, WPCs use thermoplastic
resins, which soften when heated and solidify when cooled (14). As long as the
melting point of the thermoplastic is less than the degrading point of the natural
ﬁbers (about 200°C), the two can successfully be combined (1,12,14,20).
Common therrnoplastics include HDPE, LDPE, PP, polystyrene (PS), and
polyvinyl chloride (PVC) (13,14). During manufacturing these plastics serve to
transport the wood or cellulosic ﬁller, but in the ﬁnal product the ﬁber ﬁller bears
the load to enhance the strength (2).

The combination of the wood and plastic in a composite allows for a new
construction material that was not previously available, with tailor-made
properties (23). Beneﬁts of wood-plastic composites compared to neat plastics
are improved machinability, enhanced thermal stability, and increased stiffness
(13,14). Compared to solid wood, wood-plastic composites have several beneﬁts
such as moisture resistance, less water uptake, reduced thickness swelling, and
insect and fungal resistance (13). These attributes are ascribed to the plastic

matrix thoroughly encapsulating and protecting the wood or natural ﬁbers from

38

moisture that they readily absorb (13). However, the temperature sensitivity of
the plastic portion is often more of a concern than any possible moisture that the
wood may uptake (14). Additionally, these composites possess excellent fastener
(nail, screw, etc.) retention and the ability to be sanded, stained, painted and
ﬁnished for aesthetic purposes, although it is not necessary for proper protection
(13).

Since wood-plastic composites are a relatively new product, they are not
without some problems that need to be solved. Problems that need to be
resolved for WPCs include low impact resistance and increased brittleness
compared to unﬁlled thermoplastics and high density (14,24,25). Currently,
impact resistance and brittleness concerns are being addressed by using
additives, which will be discussed later. The high density and weight of WPCs is
being approached by creating microcellular voids and microstructure, using a
process called foaming (26). This has also inadvertently acted to increase

ductility and impact resistance, but may reduce strength and stiffness (26).

Raw Materials Used in Wood-Plastic Commit»

As the name implies, wood-plastic composites are composed primarily of
wood and plastic. There are, however, several other additives that are used to
enhance processability and properties of the ﬁnal product. The most common
plastics, or polymers, as well as common ﬁllers and additives will be covered
here. Fillers that are used in polymer-based composites besides wood will also

be brieﬂy mentioned to show the diversity and possibilities in this ﬁeld.

39

Polmrs

Polymers are isotropic, meaning that they have the same properties no
matter what the direction of the applied force (2,27). This should hold true for
wood-plastic composites too because they are processed into homogenous
materials (14). As of 2001, the majority of wood-plastic composites produced in
North America, over 80%, were made from polyoleﬁns (23). Polyoleﬁns are a
speciﬁc type of thermoplastic, the most prevalent of which are polyethylene and
polypropylene (27). PVC has also been used in thiswork and will be discussed
as well. A generic structure for PE, PP, and PVC is shown in ﬁgure 2.6.

 

 

‘— substitute polymer
H R for R formed
I l H PE
l l CH3 PP
H H
L_ Cl PVC

 

 

 

 

Figure 2.6 The structure of polymers commonly used in WPCs.

High density polyethylene, or HDPE, is produced from polymerizing
ethylene into a linear chain by using a special Ziegler-Nana catalyst (27). This
catalyst can be regulated to control the density and thereby the amount of
crystallinity that is possible (27). The more crystalline the polymer is the higher

density it can be and the mechanical properties (is. stiffness, strength, etc.) will

40

be higher (27). The ability of HDPE to achieve 65-90% crystallinity means that it
is a good moisture barrier, can resist chemicals, and it is also opaque and easily
machinable (27). These properties are primarily affected by the molecular weight
and distribution thereof. As the molecular weight increases the mechanical
properties also increase (27). The only fault with HDPE is that it is susceptible to
stress cracking, so the crystallinity needs to be obstructed or reduced by
processing conditions or adding comonomers (27).

Polypropylene, PP, is in the same polyoleﬁn family as HDPE, but varies in
structure because it has a methyl group (CH3) attached to the main carbon chain
as seen in ﬁgure 2.6. The location of the methyl groups in PP can make an
atactic, isotactic or syndiotactic product (27). Atactic PP has the methyl group
randomly located, isotactic PP has the methyl on alternating carbons, and methyl
groups in pairs next to pairs of hydrogens forms the less common syndiotactic
PP (27). Catalysts are also used for PP to ensure that the product is isotactic,
which allows for increased crystallinity compared to atactic or syndiotactic PP
(27). The increased crystallinity in isotactic PP means that the mechanical
properties are improved compared to when there is less crystallinity (27).
Compared to HDPE, PP is lower density, has a higher melt temperature and
higher stiffness in its neat state (27).

PVC is similar to PP, but has a chlorine atom instead of the methyl group
as seen in ﬁgure 2.6. However, unlike PP it is not possible to achieve full
crystallinity because of the large size of the chlorine group (27). The small

degree of syndiotacticity that is possible in PVC allows for just the slightest

41

amount of crystallinity (27). Thus PVC is mostly amorphous (not crystalline) and
relies upon the strength of the polar chlorine bonds for its mechanical properties
(27,28). The main concern with PVC is its thermal stability because it can
decompose at temperatures as low as 100°C (27,28). To process PVC requires
stabilizers because of this. Plasticizers are also used to manufacture PVC
because it acts as a lubricant between molecules to increase ﬂexibility (27,28).
Other additives are also commonly added to PVC, which combined all serve to

determine the properties of the ﬁnal product (27,28).

Additives

Since the plastic portion of WPCs is oil-based and the wood portion has
an afﬁnity for water, the two do not naturally bond to each other (12,27,29).
Because of this difference in polarity and structure, the wood and plastic
components can be difﬁcult to adhere to each other (1227-31). This can result in
poor properties of WPCs compared to neat polymers, but the properties can be
improved by several additives (27—29). Additives can serve many functions, such
as to improve processability, impact strength, cost effectiveness, ﬂame
resistance, UV resistance, resistance to oxidation, and mechanical or optical
properties and decrease shrinkage (27-34). Several of the additives that are used
for neat polymers and also in WPCs include coupling agents, lubricants, impact
modiﬁers, UV stabilizers, and ﬂame retardants (14,28-34).

Coupling agents are the most common additive used to enhance the

bonding between the hydrophilic natural ﬁbers and hydrophobic plastic molecules

42

(12,14,27,30,31). Typically, coupling agents are used with polyoleﬁns and are a
variation of the polymer’s structure (30,31). For example, HDPE and PP are often
modiﬁed with a maleic anhydride, which make a molecule that is polar on one
end, but non-polar on the other end (30,31,33). This allows for the polar and to
bond to the wood and the remaining non-polar and bonds to the polymer and
creates a bond between the otherwise incompatible wood and plastic portions
(12,30,31,33). The enhanced bonding formed by coupling agents increases the
stiffness, strength, and impact strength (34). This improvement is a result of the
wood ﬁbers breaking and the plastic deforming instead of a loss of bonding
between the ﬁber and matrix that would be observed without the presence of a
coupling agent (34).

Some additives simply function as their names imply. This group includes
lubricants, ﬂame retardants, and W stabilizers. In order to have a desirable
appearance, lubricants are needed to decrease sticking and increase the ﬂow or
processability (27). There are two main types of lubricants, external and internal,
which both aid the processability of the polymer melt and enhance surface
appearance, but do so by different mechanisms (14,27). UV stabilizers and also
antioxidants ensure that damaging ultraviolet rays do not deteriorate the structure
of polymer (27). All polymers are affected differently by UV radiation, but color
changes, decreased ﬂexibility, and severed polymer chains are common (27).

Impact modiﬁers also have an effect similar to what the name implies,
however it encompasses a very broad group. The effect of impact modiﬁers

depends on if they act as crosslinked or uncrosslinked modiﬁers, their

43

compatability with PVC, and the content (2829). Generally crosslinked modiﬁers
yield better results at lower contents and can show more impact strength
improvement (25). Impact modiﬁers also enhance the bonding by creating a
network of elastomers, so they can increase the strength and stiffness (25,27-
29). Despite their name, some work has shown that impact modiﬁers do not
actually increase ductility or reduce the density, but they increase the structure,
resistance, and strength of WPCs (25,29).

As was previously mentioned, PVC requires several additives to ease
processing and enhance its properties that are not used or as common with other
polymers. PVC additives include plasticizers, heat stabilizers, and colorants
among many others (27-29). Plasticizers function to reduce the rigidity of PVC,
which increases the ﬂexibility and processability, while reducing the required
processing temperatures (27-29). The increased processability is a result of
decreased viscosity from the low molecular weight plasticizer (28,29,32). The
vast majority of plasticizers, over 80% are used with PVC because without
plasticizers PVC is essentially useless due to its brittle nature and processing
difﬁculties (27). Heat stabilizers are needed to prevent PVC from degrading
because it is heat sensitive (27-29). While colorants serve no purpose in the
mechanical properties of the composite, they are often used for aesthetic
purposes so that painting and other ﬁnishing techniques are not necessary
(27,32).

Although not truly an additive, occasionally other polymers can be added

to a neat polymer to form a blend. For example, PP often has 1.5 to 7% of

44

ethylene monomers, the building blocks for PE (27). This incorporation of similar
monomers allows PP to retain the majority of its structure but effectively reduces
the crystallinity (27). The reduced crystallinity results in less brittleness, higher
clarity, and lower density because the crystallites are not as organized and

packed together (27).

Fillers and Fibers

As with additives, ﬁllers can serve to reduce the cost, improve some of the
properties of neat polymers, or both (27,32). Fillers fall into 2 groups, inorganic
minerals and organic natural ﬁber ﬁllers (27). Regardless of the type, the ﬁber
ﬁllers carry the structural load in composites, assuming they are properly bound
to the polymer matrix (2,32). They are also available in a variety of shapes and
sizes with names like whiskers, ﬁbers, particles, and ﬂakes (32). Continuous
ﬁbers provide more improvement in mechanical properties, but have a higher
cost associated with them than chopped ﬁbers (27,32). The shorter chopped
ﬁbers or particles are easier to process than long ﬁbers, which tend to break or
agglomerate and not disperse as readily (13,14,29,32). The higher aspect ratio
(length to diameter) of long ﬁbers generally provides better reinforcement
(13,20,27). However, a compatibilizer, intensive mixing, or special feeding
equipment is often needed to help with the dispersion of long ﬁbers (13,20).
Another alternative when using long ﬁbers is to have them woven into a mat and

then form the ﬁnal product (32).

45

Mineral ﬁllers have been the most widely used in the past, although
natural ﬁllers are becoming more prevalent because they are lower cost (13).
Common mineral ﬁllers include calcium carbonate, glass ﬁbers, clay, talc, silica,
carbon ﬁbers, and mica (13,27). There are some problems with mineral ﬁllers
compared to natural ﬁllers, although they can provide better reinforcement
Minerals tend to be denser, and products will be heavy, and manufacturing
equipment wears down from processing (13).

Wood and natural ﬁllers also have beneﬁts over mineral ﬁllers, such as
their low cost, low density, light weight, low abrasiveness, increased ﬂexibility,
and that they are made from renewable resources makes them readily available
and recyclable (13,14). Natural ﬁbers also have a high speciﬁc strength and
stiffness, so for a small piece of ﬁber the strength is still quite high (13).
Common natural ﬁbers that are being used for current research are wood, kenaf,
jute, ﬂax, hemp, soybean hulls, etc. (13,14,22,35). One problem with natural
ﬁbers like these is that the moisture can easily ruin the surface or part quality as
excess moisture evaporates during processing (14,29). This necessitates the use
of ovens to dry the ﬁllers before processing or using machinery equipped with
venting for processing (14,29). The lack of attraction between the wood and ﬁber,
which was previously discussed may also be improved by modifying the ﬁbers
with a chemical treatment (29,33). Wood ﬁbers that were previously used as
CCA treated lumber have been shown to be effectively incorporated into WPCs
with improved ﬂexural strength and stiffness compared to WPCs made with virgin

wood ﬂour (36).

46

Wood ﬂour, particles, or ﬁbers are easily accessible and common ﬁllers for
thermoplastic composites (14). As with particleboard the wood can be a result of
planer shavings, chips, or sawdust (14). They can also be obtained in a variety of
species and sizes (14). Generally softwood species are preferred to hardwoods
to make for easier processing and better mechanical properties (29). The
addition of wood ﬁllers leads to an improvement in the stiffness of the composite
while lowering the abrasiveness on processing equipment compared to mineral
ﬁllers (22.37.38). However, the strength may be reduced and the brittleness
increased as the wood ﬁber content increases (24.25.29). There are an inﬁnite
number of additives and concentrations that make the properties of WPCs easily
customizable to any application, but the sacriﬁce of trade-offs such as these

need to be evaluated.

andation of Wood-Plastic Commit”

As mentioned in Chapter 1, wood-plastic composites are used in a variety
of indoor and outdoor applications. The use of WPCs, particularly in outdoor
applications, raises concerns about their durability and property retention. As with
solid wood, there are concerns for marine, fungal, insect. and W damage. Since
WPCs are a relatively new ﬁeld these problems are just being discovered. There
are a variety of additives that have been developed to help with some of these
problems and more can be developed in the future to prevent against potential

problems as they are discovered.

47

The plastic component of WPCs surrounds the wood, which helps to
reduce the moisture problems and subsequent attacks seen in solid wood (13).
The reduced moisture presence and resistance of plastic means that there is less
likelihood for fungal and insect attack (13). There has been some early work
done on fungus and WPCs that have shown the potential for fungal attack
(39.40). More recent work on commercially available products has shown that
WPCs have a slow moisture uptake, although alter continuous submersion they
can still achieve a moisture content high enough for fungal colonization (41).
Also, when there is incomplete encapsulation of the ﬁbers by the polymer, there
is a probability for fungal attack and thereby discoloration is likely (42). The lack
of proper encapsulation allows for greater moisture uptake and microscopic
cracks which act as entry points for the fungus (42). If the wood is properly
encapsulated, WPCs show a greater weight loss [from fungal consumption of
wood components] at higher wood contents because there is more food available
for the fungi (43).

In order to help prevent against fungal colonization, there has been some
use of new additives. Biocides have been added to several different
combinations of WPCs and tested against both brown- and white- rot fungi (43-
45). Zinc borate is the most commonly used biocide and has shown protection
against a brown-rot fungus at levels as low as 1% for 50% wood content in the
composites (44). Additionally, there has been evidence to show that there is little
concern of biocide leaching at effective loading levels of zinc borate (45). The

concern for fungal attack in WPCs was one issue that was discovered early.

48

Consequently, there has been more time to conduct research and an answer to

this problem seems to be near at hand.

There are still concerns for other types of degradation with WPCs. These
concerns are primarily related to the wood component that was discussed earlier,
although there are some drawbacks of plastics that need to be addressed as
well. Along with fungal attack, there is also concern of attack from subterranean
termites (46). Marine applications also raise concerns for attack by creatures
including marine borers. Plastics like PVC have been used to jacket solid wood
pilings and have been shown to effectively mduce attacks by marine borers (6).
Likely a solid plastic coating would have a similar effect for WPCs. However, this
plastic casing is still susceptible to mechanical damage (6), so it is not the ﬁnal
solution to marine borer resistance because more modiﬁcation needs to be done
on the formulation to ensure mechanical properties are retained.

When in use for exterior applications, as WPCs often are, sunlight and its
UV rays are of concern as well as the moisture from precipitation. Most polymers,
but particularly polystyrene (PS) are susceptible to degradation from W light
(47). The W light is absorbed by the polymer and can result in yellowing and
increased brittleness (47). This is preventable in polymers with the use of
additives. most of which turn the absorbed radiation into heat. so no chemical
change is observed (47). Previous research with WPCs has shown that the
incorporation of wood into the polymer matrix increases the rate of
photodegradation compared to neat polymers (48-50). The effects on WPCs can

range from aesthetic to mechanical failures of the product, but it is unknown as to

49

whether or not the incorporation of plastic additives like benzophenone
derivatives can help reduce the damage from W light WPCs like it does in neat
polymers (47-50). WPCs made with CCA-treated wood have shown increased
photo-protection compared to WPCs with virgin wood ﬂour (37), so perhaps
chemical treatment is a possible solution. It has been established that WPCs
absorb more W light than particleboard because of the addition of the plastic
(36), so further research is wananted.

Other factors that could affect WPCs that have not been extensively
investigated exist as well. Both wood and plastics are susceptible to degradation
in the presence of chemicals (2,47). Exposure to chemicals can cause the wood
structure to collapse and polymers to show local plasticization and cracking
under lower than ordinary loads (2.47). Most polymers used for WPCs (PP, PS,
PE, and PVC) are also susceptible to thennal degradation. Additives are
common and have been previously discussed to prevent against this, particularly
for PVC (27-29,47). Thermal degradation often results in breakage or
degradation of the polymer structure (47). In the extreme case of PVC,
hydrochloric acid can be released as the polymer chain is broken apart
(28.29.47). Further research needs to be done on these and other additive

related issues to determine solutions for degradation from these forces.

50

Processing of Wood-Plastic Commit»
Thermoplastic polymers, and therefore WPCs, can be processed in a

variety of ways depending upon their ﬁnal applications. Parts made from the
previously described polymers (HDPE, PP, and PVC) which are also used for
WPCs are commonly processed by injection molding, extrusion, and
compression molding (27.28.32.47). These three techniques are used to
manufacture the majority of plastic objects (32). Extrusion is a continuous
process, whereas both compression and injection molding are batch processes
(47). Extrusion is preferred for long parts; injection molding is good for complex
shapes and very high volumes, whereas compression molding works well for
large molded shapes (32). There are also other options such as pultrusion,
transfer molding, therrnofonning or calendaring, which will not be discussed here
(14.27.32.47). The choice of method relates strongly to the item being made and
the viscous properties of the polymer composite melt (47). The desired properties
of the ﬁnal product may also be considered because each method produces
differing amounts of ﬁber orientation, which can result in anisotropic products
(32). The same methods of processing that are used for polymers also apply to
WPCs, although an initial compounding step is needed to ensure proper
dispersion of ﬁbers and additives (14). This compounding step may involve dry-
blending, molten-dispersion mixing in a continuous or batch process, or

pelletizing (14).

51

Extrusion

Extrusion is perhaps the most widely used method for plastic processing
and creates products formed into proﬁles, pipes, sheets, or rods (32.47.51). An
extruder has three main parts, the hopper, banal, and die (47). The hopper feeds
pelletized, granulated, or ground materials into the heated banal, although some
extruders are equipped to wed already molten mixtures to the barrel (32.47.51).
The barrel is equipped with either one or two screws depending upon the
processing. A twin-screw extruder is needed when temperature sensitive and
viscous melts are being processed (47). Since these are both attributes of
WPCs, a twin-screw extruder is preferred and has been used for this work.

The screws are custom made for each extruder and contain feeding,
compression and metering sections (32,47). The depth and distance between the
channels in the screw are essential to how the material will ﬂow through the
extruder (51). The initial feeding section is often cooled to ensure that the
polymer does not melt too early and clog the feeding (51). The feeding section
moves materials to the compression section, also known as the transition
section, which begins melting the polymer (47,51). Normally a series of heaters
and coolers are needed to ensure that the temperatures are properly maintained
through this region of the barrel (32,51). The metering section begins to taper in
diameter and combined with shear forces inside the extruder force the melt out of
the extruder and into a shaping die (32,47). The ﬁnished shape is than water-
cooled as it comes out to retain its dimensions and is then cut to length

(32.47.51). Although solid formed components were the focus here, ﬁlms, sheets,

52

 

‘ . I
a
"- " '§
'1 ~
I
. o
\
> ‘ \
v
C
I. 1
1' I 2 I .
'5 \ I ‘ ‘
r In .
'v'Iv
"‘
I - ‘
l
c
‘ ,
v
\I
. - , l >
. T‘s. '
Va
. . ’.r
J .
‘ I
. '- ,__‘._'
‘ t
a,
_ 4
r.‘ ‘ K
., - .\ W. l
’ a . l .
1 c

f
.-> ‘

and coatings can also be created using extruders with different dies (51). An
example of an extruder’s cross section with these sections is shown in ﬁgure 2.7.
The drive provides power to the extruder, thermocouples ensure that the
temperature is properly regulated, and the adapter provides a connection for the
die (27).

feeder
hopper

(4) heaters

 

 

 

 

water feed section compression section metering section
cooﬁng

Figure 2.7 A cross section of an extruder showing all of the components
including the screw down the middle of the banal.
Image slightly modiﬁed from Hernandez, Selke, and Culter (27).
Extrusion may be difﬁcult to set-up, but often requires little monitoring
once in progress, although there are a few concerns for the use of extruders.
When using an extruder it is important to ensure that the feed stock has

components that are all about the same size. otherwise there will be uneven

melting, dispersion and ﬂow (51). The ﬂow of materials through the extruder also

53

needs to be monitored to ensure that there are enough raw materials present to
be fed into the later zones and that materials are not stuck in the threads of the
screw (51). The screws also need to be checked for damage that can occur from
abrasive (mineral) ﬁbers (32). For production of polymer compositions like
WPCs, which can emit gases during production, a vent about % of the way down
the banal is needed to allow gases to escape so that the ﬁnal product will not be
damaged or contain voids (51). The ﬁnal shape of the product mainly depends
upon proper venting and the die. However, shrinkage during cooling occurs from
the presence of crystallinity and needs to be considered when designing a die
and determining cooling treatments (51). This makes amorphous polymers, like
PVC, desirable for the production of shaped extrudates (51 ).

Injection Moldigg

Injection molding involves the use of a screw similar to that in the extruder
to melt the pellets, granules, or ground materials (32,47). As with the extruder,
friction and shear are created inside the barrel and create a uniform mixture of
the molten polymer with the ﬁbers suspended inside (32). A fonrvard movement of
the screw then forces the melt through a gate into a mold under high pressure to
form the desired part or parts depending upon the mold used (47). This
movement helps to create microstructure within the piece as the ﬁbers are
oriented and move in the polymer melt (32). As with extrusion there is also a
concern of shrinkage with crystalline polymers. To prevent shrinkage in injection

molding, a packing state with high pressure is used to counteract any contracting

54

that may otherwise take place (32). The part is then cooled and ejected (32).
Components manufactured by injection molding may be limited to non-load
bearing applications because of poor mechanical properties that result if ﬁber
content, length, alignment, and degradation are not properly considered (32).

One of the main problems with injection molding is that only a few
polymers are commonly used. This is because of the interactions between the
hot polymer melt and the cold mold walls that can lead to “freezing” and impede
proper ﬂow (32). Injection molding does have standard built-in vents to allow for
the displacement of air from in the cavity and the polymer melt (32), but this is
useless if the melt cannot completely ﬁll the mold. The cost of molds can also be
quite high and they need to be able to withstand high temperature and pressure
while retaining the ability to produce needy identical molded parts (32.47).
Anodier small problem is that the molds waste material to ﬂow through runners
and gates to get to the portion of the mold that forms the desired part (47). This
can be alleviated by using hot runners, so the loss of material is signiﬁcantly
decreased. Injection molding does provide short cycle times and the potential for
high volume production, as long as these costs and problems can be properly
considered (32).

Compression Molding

Compression molding is becoming less common, but is the oldest,
cheapest and easiest form of polymer processing (32,47). Although injection

molding is preferred for processing thermoplastics, compression molding is often

55

used for processing therrnosets or when other forms of processing are too costly
(47). Both molded and ﬂat panel parts can be made via compression molding
(32,51). The materials used in compression molding may involve a woven mat
with long ﬁbers, or a nonwoven mat that has been dry-blended and loosely laid
together (14,32). Generally the nonwoven mat is preferred because it allows for a
wider use of ﬁbers, ﬁllers, and additives to be incorporated (14). This mat, also
called a charge, is placed between the two sides of the mold cavity (32). The
mold cavity then closes, but there are two options for what happens next in WPC
production depending upon when the heat is applied. The ﬁrst option involves
consolidation, then heating to melt the thermoplastic, and lastly cooling the press
to solidify and remove the composite (14). The other choice is to heat the
material ﬁrst in an oven or press, and then transfer the hot material to a cool
press where it is consolidated and cooled (14). Generally it is preferred to heat
and cool a consolidated mat in one step like the ﬁrst option (32). Regardless of
the method, the concept of heating the WPC so that the thermoplastic ﬂows
around the ﬁbers is the same. Enough heat needs to be provided so that this ﬂow
happens before the cooling stage is initiated (32). Unlike injection molding, the
orientation of ﬁbers from compression molding is negligible and does not affect
the ﬁnal properties of the panel or molded part (32).

Compression molding has been partially phased out due to injection
molding because injection molding can produce similar parts with more
automation (32). Compression molding does still have the advantage that there

are fewer stress points than in injection molding because of the lesser amount of

56

deformation (32). Also, compression molding allows longer ﬁbers to stay intact
since they do not have to ﬂow through the gate and higher ﬁber contents can be
used because ﬂow is not as much of a concern (32). Compression molding also
yields parts with very low dimensional tolerance requirements and has more

repeatability over time even under fast production times (32).

T»ting of Wood-Plastic COMET”!

The phrase ‘mechanical properties’ is often used when referring to the
strength, stiffness, impact strength, etc. of materials. They are also called
strength properties or by the name of the individual property of concern. These
properties are essentially the same for all materials that are used for the same
applications regardless of whether they are wood, plastic, a composite thereof, or
a different material like steel, concrete, etc. A thorough discussion of common
and less common properties was found for wood, but it can be extended to
WPCs since they are used for the same applications and need the same
properties

In the course of this text the phrase mechanical properties was often used
to refer to the ﬂexural strength and stiffness. The ﬂexural strength and stiffness
are used to reﬂect the maximum load that a material can retain under a bending
procedure (3). The bending involves both compressive forces on the top of the
sample and tension forces on the bottom of the sample (2). Figure 2.8 shows a
sketch of how ﬂexural testing is conducted with a force, F, applied to the middle

of a sample set across a span denoted by the supports. One test can yield both

57

the ﬂexural strength (MOR) and stiffness (MOE), both of which are calculated in
the untis of psi or MPa. Since WPCs are most commonly used for decks, their
ability to resist loading in this manner is of utmost concern. Consequently,
ﬂexural testing was the primary test done throughout this work to determine

mechanical property retention.

   

d (depth)

 

Figure 2.8 A schematic drawing of ﬂexural testing with variables labeled for
calculating ﬂexural strength (MOR) and stiffness (MOE).
Image was modiﬁed from the Society of Wood Science and Technology.
http:llwww.swst.org/teach/teach2/properties2.ppt#1

The result of a ﬂexural test is a graph similar to that shown in ﬁgure 2.9.
The x-axis is the strain or deﬂection and the y-axis is stress or load. Either way
the same graph results. If the graph is stress vs. strain, then the ﬂexural stiffness
(MOE) can be calculated as the slope of the line, the stress divided by the strain
up to the proportional limit (2). The proportional limit, shown in ﬁgure 2.9 is where

the linear portion of the stress-strain curve ends and it becomes curvilinear (2).

58

The MOE can also be calculated by using equation 2.1. The variables are
illustrated in ﬁgure 2.8, where F represents the force applied, D is the deﬂection

caused by the load at midspan, L is the span, and I is the moment of inertia (2).

FL3 1 _ bd3
481D 12

MOE =

 

 

(Equation 2.1)

The MOR is calculated using the same variables as the MOE. The
calculation for MOR is shown in equation 2.2. The force, F, this time is the
maximum load to cause the sample to break, and L, b, and d are the same
dimensions as before (2). The dimensions L. b, and d are used in these
equations because the samples being tested are uniform rectangles with the
same cross sectional sizes (2). If the samples had a different shape, then
different equations would be needed to calculate the MOE and MOR (2). Flexural
testing also assumes that the sample is supported only by the span supports and

that the sample does not slip from these supports during testing (2).

1.5FL

MOR = b d2 (Equation 2.2)

 

59

  
 

  

\

proportional
limit

Stress (or load)

 

 

Strain (or deﬂection)

Figure 2.9 The stress-strain curve or load-deﬂection curve that results
from ﬂexural testing of strength and stiffness.
Image was modiﬁed from the Society of Wood Science and Technology.
http://www.swst.orglteach/teach2/properties2.ppt#1
The tensile properties are also used by themselves to determine the ability
of a material to stretch in a vertical direction in each of the anisotropic directions
(3). Compressive forces are often checked by themselves too. Compressive
forces are applied in all anisotropic directions to determine the ability of each
direction to resist compression (3). Isotropic materials like plastics, and
presumably WPCs, are only tested in one direction for these tests.
Other common properties to check are the shear strength, impact strength
and hardness (3). Shear strength is the ability to resist slippage within the sample
(3). The impact strength is tested by dropping a weight onto a sample to

determine how much energy can be absorbed by the material (3). Hardness is

the resistance of a material to have an indentation form in it (3). During the test a

60

ball is forced into the sample to measure me penetration (3). Contrary to some
people’s beliefs, hardness is not related to the title hardwood or softwood (3).

Other less common properties that can also be of concern for more
speciﬁc applications include torsion, toughness, creep, fatigue, and fracture
toughness, among many others (3). Torsion is the resistance to twisting and can
be measured as a strength or stress (3). The energy required to cause failure
from a sudden centrally-located force is the toughness (3). Creep relates to the
ability of a material to hold a load over time and the amount that the sample
deforms (3). Fatigue is similar to this, although cyclic loads are applied instead of
a constant weight (3). Fracture toughness is a resistance of the material to form
ﬂaws which can perpetuate into tetal failures (3).

ASTM lntemational has established testing methods for all of these
properties and more for most common building materials. These standard testing
methods relate to mechanical, physical, and aesthetic properties that may
degrade over the service life of a product. Because of the variability of products
such as wood to change their properties under different climactic conditions (e.g.
moisture contents) these standards provide a common means to test and
compare materials.

The research presented here followed ASTM standards as closely as
possible, since WPCs are relatively new and have very few established
standards. The ﬂexural properties were tested using ASTM D6109, which is the
standard test method for ﬂexural properties of unreinforced and reinforced plastic

lumber. Three-point bending tests were conducted and the strength and stiffness

61

of the WPCs were determined in the manner previously mentioned. Dimensional
stability, a physical property has been given little attention, but was tested
according to ASTM D6341, as discussed in ASTM D6662. Another physical
property, density, was also tested, although no standard test method was found.
Freeze—thaw cycling was conducted in accordance with ASTM 06662. the
standard speciﬁcation for polyoleﬁn-based plastic lumber decking boards. Decks
have become a popular use of WPCs and this was the most closely related
standard, although the sample dimensions needed to be modiﬁed.

Summag
Wood—plastic composites offer numerous advantages over solid wood,

neat polymers, and wood-based composites bound with thennoset resins.
Among the primary beneﬁts is the option of several polymers, ﬁllers, and
additives to customize the ﬁnal product so that it meets the requirements of end-
use applications. The wood and polymer components, as well as their
composites are still susceptible to various forces. Some of these forces, like
fungi, have been researched and new additives are being developed for
incorporation into WPCs. This work seeks to investigate another previously
uninvestigated effect of freeze-thaw cycling, which may also affect (extruded)
WPCs. This evaluation of effects will also be statistically modeled in an effort to

determine an optimum formulation for degradation resistance.

62

REFERENCES

. A.A. Marra (1992) Technolpgy of Wood Bondipg: Principles in Practice.
VanNostrand Reinhold. New York.

. J.G. Haygreen and J.L. Bowyer (1996) Forest Products and Wood Science;
An Introduction, 3rd edition. Iowa State University Press. Ames, IA.

. D.W. Green, J.E. Vlﬁnandy, and DE. Kretschmann (1999) Mechanical
Properties of Wood. Chapter 4 in Forest Products Laboratory ed. Wood
Handbook - Wood as an Epgineeripg Material. Gen. Tech. Rep. FPL-GTR-
113. US. Department of Agriculture, Forest Service, Forest Products
Laboratory. Madison, WI.

. W. Simpson and A. TenWolde (1999) Physical Properties and Moisture
Relations of Wood. Chapter 3 in Forest Products Laboratory ed. Wood
Handbook — Wood as an Epgineeripg Material. Gen. Tech. Rep. FPL-GTR-
113. US. Department of Agriculture, Forest Service. Forest Products
Laboratory. Madison, WI.

. R.E. lbach (1999) Wood Preservation. Chapter 14 in Forest Products

Laboratory ed. Wood Handbook - Wood as an Epgineering Material. Gen.
Tech. Rep. FPL-GTR-113. US. Department of Agriculture. Forest Service,

Forest Products Laboratory. Madison, WI.

. T.L. Highley (1999) Biodeterioration of Wood. Chapter 13 in Forest Products

Laboratory ed. Wood Handbook - Wood as an Engineeripg Material. Gen.
Tech. Rep. FPL-GTR-113. US. Department of Agriculture, Forest Service,

Forest Products Laboratory. Madison. WI.

. W.J. Catallo, T.E. Shupe, and RR. Gambrell (2004) Hydrothermal treatment
of CCA- and penta- treated wood. Wood Fiber Sci 36(2):152-160.

. BM. Ahmed, J.R.J. French, and P. Vinden (2004) Evaluation of borate
formulations as wood preservatives to control subterranean termites in
Australia. Holzforschung 58(4):446-454.

. J.L. Taylor and PA. Cooper (2003) Leaching of CCA from lumber exposed to
natural rain aboveground. Forest Prod. J. 53(9):81-86.

10. K. Stock, T. Tolaymat, M. Ward, 8. Dubey, T. Townsend, H. Solo-Gabriele,

and G. Bitton (2005) Relative leaching and aquatic toxicity of pressure-treated
wood products using batch leaching tests. Environ. Sci. Technol. 39(1):155-
163.

63

11.C.B. Vick (1999) Adhesive Bonding of Wood Materials. Chapter 9 in Forest
Products Laboratory ed. Wood Handbook - Wood as an Epgineering
Material. Gen. Tech. Rep. FPL-GTR-113. US. Department of Agriculture,
Forest Service, Forest Products Laboratory. Madison, WI.

12.T. Sellers Jr. (2001) Wood adhesive innovations and applications in North
America. Forest Prod. J. 51(6):12-22.

13. L.M. Matuana and PA Heiden (2004) Wood Composites. In the

Emma of Pplmipr Science and Technolpgy. John Wiley & Sons, Inc.
New York.

14.J.A. Youngun (1999) Wood-based Composites and Panel Products.
Chapter 10 in Forest Products Laboratory ed. Wood Handbook — Wood as an
Engineeripg Material. Gen. Tech. Rep. FPL—GTR—113. US. Department of
Agriculture, Forest Service, Forest Products Laboratory. Madison, WI.

15.B.D. Park, B. Riedl, E.W. Hsu, and J. Shields (1998) Effects of weight
average molecular mass of phenol-formaldehyde adhesives on medium
density ﬁberboard performance. Holz. Roh. Werkst. 56:155-161.

16.T.M. Maloney (1996) The family of wood composite materials. Forest Prod. J.
46(2): 1 9-26.

17.L.M. Guss (1995) Engineered wood products: the future is bright. Forest
Prod. J. 45(7/8):17-24.

18. DA Pease ed. (1994) Panels: Products, Applications and Production Trends.
2"cl edition. Miller Fmeman Publications, Inc. San Francisco, CA.

19. L.M. Matuana, R.T. Woodhams, J.J. Balatinecz, and CB. Park (1998)
Inﬂuence of interfacial interactions on the properties of PVC/cellulosic ﬁber
composites. Polym. Comp. 19(4):446-455.

20.C. Clemons (2002) Wood-plastic composites in the United States: the
interfacing of two industries. Forest Prod. J. 52(6):10-18.

21.R.M. Rowell, S.E. Lange, and RE. Jacobson (2000) Weathering performance
of plant-ﬁberlthen'noplastic composites. Mol. Cryst. Liq. Cryst. 353:85—94.

22.T. Sellers, G.D. Miller, and M. Katabian (2000) Recycled thermoplastics
reinforced with renewable lignocellulosic materials. Forest Prod. J. 50(5):24-
28.

23.J. Patterson (2001) New opportunities with wood-ﬂour-foamed PVC. J. Vinyl
Addit. Technol. 7(3):138-141.

24. L.M. Matuana, C.B. Park, and J.J. Balatinecz (1997) The effect of low level
plasticizer on the rheological and mechanical properties of polyvinyl
chloride/newsprint-ﬁber composites. J. Vinyl Addit. Technol. 3(4):265-273.

25. F. Mengeloglu, L.M. Matuana, and J.A. King (2000) Effects of impact
modiﬁers on the properties of rigid PVC/wood-ﬁber composites. J. Vinyl Addit.
Technol. 6(3): 153-157.

26. L.M. Matuana, C.B. Park, and J.J. Balatinecz (1998) Cell morphology and
property relationships of microcellular foamed PVC/wood-ﬁber composites.
Polym. Eng. Sci. 38(1 1): 1 862-1 872.

27.R.J. Hernandez, S.E.M. Selke, and JD. Culter (2000) Plastics Packaging:
Promrties, Processipg, Application, and Rgulations. Cari Hanser Verlag.

Munich, Germany.

28.W.V. Titow (1984) PVC Technolpgy. 4th edition. Elsevier Applied Science
Publishers Ltd. New York.

29.J. Zazyczny and L.M. Matuana (in press) Chapter 7: Fillers and Reinforcing
Agents in DE. \Mlkes, J.W. Summers, and CA Daniels, eds. PVC
Handbook. Hanser Publications. Munich, Germany.

30.0. Li and L.M. Matuana (2003) Effectiveness of maleated and acrylic acid-
functionalized polyoleﬁn coupling agents for HDPE-wood ﬂour composites. J.
Thermoplast. Compos. Mater. 16(6):551-564.

31.J.Z. Lu, Q. Wu. and HS. McNabb Jr. (2000) Chemical coupling in wood ﬁber
and polymer composites: A review of coupling agents and treatments. Wood
Fiber Sci. 32(1):88-104.

32.S.G. Advani and EM. Sozer (2003) Process modeling in commsites
manufacturing. Marcel Dekker, Inc. New York.

33.M. Kazayawoko, J.J. Balatinecz and L.M. Matuana (1999) Surface
modiﬁcation and adhesion mechanisms in wood ﬁber-polypropylene
composites. J. Mater. Sci. 34:6189-6199.

34.V.N. Hristov, S.T. Vasileva, M. Krumova, R. Lach, and G.H. Michler (2004)
Deformation mechanisms and mechanical properties of modiﬁed
polypropylene/wood ﬁber composites. Polym. Comp. 25(5):521-526.

35.J.L. Julson, G. Subbarao, D.D. Stokke, H.H. Gieselman, and K.

Mathukumarappan (2004) Mechanical properties of biorenewable ﬁber/plastic
composites. J. Appl. Polym. Sci. 93(5):2484-2493.

65

36. DP. Kamdem, H.H. Jiang, W.N. Cui, J. Freed, and L.M. Matuana (2004)
Properties of wood plastic composites made of recycled HDPE and wood
ﬂour from CCA-treated wood removed from service. Compos. Part A - Appl.
Sci. 35(3):347-355.

37.G.E. Meyers, I.S. Chahyadi, C.A. Coberly, and 0.8. Enner (1991) Wood ﬂour/
polypropylene composites: inﬂuence of maleated polypropylene and process
and composition variables on mechanical properties. Int. J. Polym. Mater.
1521-44.

38.8. Takase and N. Shiraishi (1989) Studies of composites from wood and
polypropylenes. II. J. Appl. Polym. Sci. 37(3):645-659.

39.M. Mankowski and J.J. Morrell (2000) Patterns of fungal attack in wood-
plastic composites following exposure in soil block tests. Wood Fiber Sci.
32(3):340-345.

40. PI. Morris and P. Cooper (1998) Recycled plastic/wood composite lumber
attacked by fungi. Forest Prod J. 48(1):86-88.

41 .W.H. Wang and J.J. Morrell (2004) Water sorption characteristic of two wood-
plastic composites. Forest Prod. J. 54(12):209-212.

42.B. Dawson-Andoh, L.M. Matuana, and J. Harrison (2004) Mold susceptibility
of rigid PVC/wood-ﬂour composites. J. Vinyl Addit. Technol.10(4):179486.

43.J. Simonsen, C.M. Freitag, A. Silva, and J.J. Morrell (2004) Wood/plastic
ratio: Effect on performance of borate biocides against a brown rot fungus.
Holzforschung 58(2):205-208.

44.8. Verhey, P. Laks, and D. Richter (2001) Laboratory decay resistance of
woodﬁberl‘thermoplastic composites. Forest Prod. J. 51(9):44-49.

45. DE. Pendleton, T.A. Hoffard, T. Adcock, B. Woodward, and MP. Wolcott
(2002) Durability of an extruded HDPE/wood composite. Forest Prod. J.
52(6):21-27.

46.W. Chetanachan, D. Sookkho, W. Sutthitavil, N. Chantasatrasamy and R.
Sinsermsuksakul (2000) PVC/wood: A new look in construction. Proceedings
of Vinyltec 2000, sponsored by Vinyl Division-SPE, Philadelphia Section.
Philadelphia, PA. October 11-12, 2000.

47.J.R. Fried (1995) Poﬂer Science and Technolpgy. Prentice Hall. New
Jersey.

66

48. L.M. Matuana and DP. Kamdem (2002) Accelerated ultraviolet weathering of
PVCIwood-ﬂour composites. Polym. Eng. Sci. 42(8):1657-1666.

49. NM. Stark and L.M. Matuana (2003) Ultraviolet weathering of photostabilized
wood-ﬂour-ﬁlled high-density polyethylene composites. J. Appl. Polym. Sci.
90(10):2609—2617.

50. L.M. Matuana, D.P. Kamdem, and J. Zhang (2001) Photoaging and
stabilization of rigid PVCIwood-ﬁber composites. J. Appl. Polym. Sci.
80(11):1943-1950.

51.P.N. Richardson (@1974) Introduction to Extrusion. Eighth printing, Society of

Plastics Engineers (SPE) processing series. Society of Plastics Engineers,
Inc. Brookﬁeld, CT.

67

Chapter 3

DURABILITY OF WOOD FLOUR-PLASTIC COMPOSITES EXPOSED TO
ACCELERATED FREEZE-THAW CYCLING
PART I. RIGID PVC MATRIX

This chapter is slightly modiﬁed from Journal of Vinyl & Additive Technology that

was published in January 2005. 11 (1) : 1-8. It is co-authored by J.M. Pilarski1
and L.M. Matuana‘.

‘ Department of Forestry, Michigan State University, East Lansing, MI, 48824

68

ABSTRACT

This study examined the effects of accelerated freeze-thaw actions on the
durability of wood ﬁber-plastic composites. Rigid PVC formulations ﬁlled with
various concentrations of wood ﬂour (both pine and maple) were processed in a
counter-rotating twin-screw extruder and exposed to cyclic freeze-thaw actions
according to ASTM standard D6662. Freeze-thaw cycling was also modiﬁed by
omitting portions of the test (either the water or freezing) to verify whether or not
moisture was the primary cause for property loss. The durability of exposed
samples was assessed in terms ﬂexural properties, density and dimensional
stability. Scanning electron micrographs of unexposed and freeze-thaw exposed
samples were taken to qualitatively evaluate the interfacial adhesion between the
wood ﬂour and PVC matrix.

The experimental results indicate that the density was not affected by
freeze-thaw cycling. The dimensional stability was also relatively unaffected,
although greater wood ﬂour content exhibited greater dimensional change. The
loss in stiffness of the composites was statistically signiﬁcant after only 2 freeze-
thaw cycles, regardless of both the wood species and content. Conversely, the
strength of the composites was not signiﬁcantly affected by 5 freeze-thaw cycles
at lower wood ﬂour contents (50 and 75 phr). The deleterious effects of the
freeze-thaw actions on the strength of the composites become apparent at higher
wood ﬂour content (100 phr) after only 2 freeze-thaw cycles for maple ﬂour and 5

freeze-thaw cycles for pine ﬂour. The property loss was attributed primarily to the

69

water portion of the cycling which appears to have led to the decreased

interfacial adhesion between the wood ﬂour and rigid PVC matrix.

70

INTRODUCTION

Recently, wood ﬁber-plastic composites (WPCs) have been gaining
popularity in a variety of applications because WPCs combine the desirable
durability of plastics with the cost effectiveness of wood ﬁbers as a ﬁller or
reinforcing agent (1-5). The addition of wood ﬁbers leads to an improvement in
the stiffness of the composite while lowering the abrasiveness on processing
equipment (6-8) compared to mineral ﬁllers. Because of these attributes, WPCs
are being used in a variety of innovative applications, such as decking, docks,
landscaping timbers, fencing, playground equipment, window and door frames,
etc. (5,9-12). Among these, building applications are the largest and fastest
growing market for WPCs (13). One of the most prevalent uses is as decking and
railings, which have shown strong sales and some manufacturing companies
have doubled production from 2001 to 2002 (10). This production has been
forecasted to double again by the year 2005 (13). One reason for the popularity
of WPCs is the need to replace Chromated copper arsenate (CCA) pressure-

treated solid wood in several outdoor applications (14,15).

This increased use of WPCs by the construction industry has resulted in
concern about their durability in outdoor environments. The products may be in
ground contact or in aboveground exterior use where there is a risk of material
deterioration. When in ground contact WPCs are subject to degradation from

biological agents such as fungi and subterranean termites (11,12,16). Exposure

71

to the ultraviolet (UV) rays in sunlight and moisture have also been shown to

cause degradation in aboveground exterior environments (17,18).

Various climactic conditions cause millions of dollars of material damage
every year and high costs may be involved in replacing these damaged products.
Therefore, when a new material is developed, it is important to determine how
durable the material will be in a variety of environmental conditions.
Consequently, in colder regions where the freeze-thaw action is prevalent.
freeze-thaw durability may be of signiﬁcant importance in determining the service
life of WPCs, but has not yet been assessed. Similar studies on freeze-thaw
exposure of composites made with HDPE and 50% maple wood ﬂour have been
shown to have signiﬁcant losses in both ﬂexural strength and stiffness (19). The
PVC matrix has not previously been assessed to see how it’s performance under

freeze-thaw conditions compares to other matrices.

Moisture sorption during simulated warm climate accelerated weathering
tests has been shown to adversely affect the properties of other wood-polymer
composites. Stark (20) examined the effects of moisture on ﬂexural properties of
40% wood ﬂour ﬁlled polypropylene composites and reported that composites
exposed to a water bath for 2000 hours at 267°C experienced 39% loss in
modulus and 22% loss in strength. Tensile properties and notched Izod impact
strength also signiﬁcantly decreased for composites soaked in the water bath
(20). HDPE ﬁlled with 65% wood lost approximately 40% in tensile modulus after

saturation with water (21). However, when only 30% wood ﬁller was used with

72

HDPE matrix, a loss of approximately 25% in ﬂexural modulus was observed

after exposure to boiling water for 50 hours (22).

Clemons and lbach (23) preconditioned 50% pine ﬂour ﬁlled HDPE
composites for fungal testing and observed that more moisture was absorbed
during ambient temperature water soak than during cyclic boiling tests. However,
the cyclic boiling tests did more damage to the samples. Composite samples
preconditioned for 5 cycles of ambient temperature water and then 65% relative
humidity (RH) at 27°C showed a 23.5% decrease in ﬂexural strength (23).

Rapid hourly freeze-thaw changes from -10°C to 15°C following a 2 week
water soak showed a decrease in tensile strength for 50% straw-PE composites
(24). This loss in tensile strength was attributed to either previous water or
fungus exposure and not to the freezing (24). Karbhari et al. (25) observed
prevalent microcracking in E-glass vinylester composites that were cycled
between -10°C and 225°C. There was also evidence of ﬁber-matrix debonding
and salt water had a more prevalent effect than deionized water, however no

causation for either observation was given (25).

The processing method was also shown to have an effect on moisture
absorption of WPC products. Extruded composite samples absorbed more
moisture than compression or injection molded samples (23). Several authors
have also reported an increase in water uptake as the ﬁber loading increased in
WPCs, regardless of ﬁber and matrix types (8,20,26,27). The increased ﬁber.
content allows for more ﬁbers to undergo ﬁber wicking, so the moisture uptake is

continually increasing with time (8). Different lignocellulosic ﬁllers have been

73

shown to lead to varying degrees of water absorption, likely due to the varied
adhesion between the ﬁber and the polymer matrix (27). Composites made with
wastepaper had greater water absorption than those manufactured with wood
ﬂour or cellulose ﬁbers (27). For aspen wood-polypropylene composites, a
maleated polypropylene compatibilizer has been shown to reduce swelling and

moisture uptake (8).

Although the durability of WPCs exposed to biological organisms and
ultraviolet light has been extensively investigated (8,12,17,18,20—24), little
information is available in the open literature on the freeze-thaw durability of
these composites. This study was aimed to assess the effects of accelerated
freeze-thaw cycling on the density, dimensional stability, and ﬂexural strength
and stiffness of rigid PVC ﬁlled with various concentrations of pine or maple
wood ﬂour. Particular emphasis was placed on understanding the effects of the
number of freeze-thaw cycles and moisture absorption on the properties of the

composites.

74

EXPERIMENTAL

Materials

The 0.425 mm (40 mesh) maple and pine wood ﬂours were utilized as
hardwood and softwood species, respectively. The wood ﬂour was oven-dried at
105°C for approximately 48 hours before processing to remove moisture. The
PVC matrix and other additives used in the manufacture of the composites are

listed in table 3.1.

Table 3.1 Formulations used for rigid PVCIwood-ﬂour composites.

 

 

Ingredients Concentration (phr)

PVC (K-value = 66) (Oxyvinyls) 100
Tin stabilizer (PlastiStab 2808) (OMG Americas) 2

Calcium stearate (Synpro) 1.5
Parafﬁn wax (Gulf Wax) 2
Processing aid (Paraloid K-120)‘ 2
Processing aid (Paraloid K-175)‘ 2

Impact modiﬁer (K-334)‘ 10

40 mesh maple or piniwbrgod) ﬂour (American Wood 50, 75, 100
r rs

 

1Supplied by Rohm 8 Haas Co.

75

Comppunding and Extrusion of Commit»

A 10 liter high intensity mixer (Papenmeier, Type TGAHK20) was used for
room temperature dry blending of the PVC matrix, dried wood ﬂour (pine or
maple) and other additives at the concentration levels listed in table 3.1. Once
mixed, the compounded materials were fed into a 32 mm conical counter-rotating
twin-screw extruder (C.W. Brabender Instruments Inc.) with a length to diameter
ratio of 13:1. This twin screw extruder was equipped with an unpressurized vent
to allow any residual moisture to escape and was powered by a 7.5 hp Intelli-
Torque Plasti-Corder Torque Rheometer”. The rotational speed of the screws
was maintained at 50 rpm throughout processing. However, the extruder’s

temperature proﬁles were varied as follows.

The ﬁrst portion of this study examined the effect of wood ﬂour content
(50—100 phr) on the physical and mechanical properties of the composites
exposed to 5 freeze-thaw cycles. For this study, the temperature proﬁle from the
hopper to the horizontal die was set at 190/175/170/180 °C. The rectangular die
created samples with a nominal width and depth of 2.54 cm (1") and 0.95 cm
(318'), respectively.

For the second part of this study, the wood content was maintained at 100
phr and the number of freeze-thaw cycles was varied at either 0 (control), 2, 5, 8
or 12. Although excellent quality samples were previously produced with 50 and
75 phr wood ﬂour in the ﬁrst part of this study, the increased melt viscosity with
100 phr wood ﬂour made processing at the lower temperature proﬁle difﬁcult and

the appearance of the ﬁnal sample suffered accordingly (poor surface quality,

76

rough and tearing edges). Consequently, a different temperature proﬁle was
selected for the second part of this study with varied numbers of freeze-thaw
cycles. The processing temperature proﬁle from the hopper to the die was set at
190/185l180/180°C. The same die discussed above was used.

Freeze-Thaw Cycling

Freeze-thaw cycling was done in accordance with a modiﬁed ASTM
06662-01 (28), the Standard for Polyoleﬁn-Based Plastic Lumber Decking
Boards. One complete freeze-thaw cycle consisted of three parts: (i) a water
soak until equilibrium moisture content (EMC), (ii) exposure to freezing for 24
hours, and (iii) thawing for 24 hours. The water submersion part of the cycle was
conducted in ambient room temperature water (approximately 21°C) in a
workshop environment. Each sample was submerged for 24 hour intervals until
the weight gain was less than 1%, which implies that the sample has reached its
EMC. The freezing was conducted in a GE 0.38 cubic meter (13.4 cu ft) chest
freezer controlled to —27°C :t 2°C. Lastly the samples were thawed in a walk-in

conditioning room at 23°C :I: 2°C and 50% 1: 4% RH.

77

Modiﬁed Freeze-Thaw Cycling
Our preliminary study showed that rigid PVCIwood-ﬂour composites

exposed to freezethaw cycling lose some of their mechanical properties (29).
This loss in mechanical properties is thought to be largely due to the effect of
moisture since moisture changes can lead to a degradation of interfacial
properties in composites (25). To verify whether or not moisture is the primary
cause of the observed property loss, freeze-thaw cycling was modiﬁed by

omitting portions of the test (either the water or freezing part) as follows:

(i) The Freeze-thaw only (FT) cycle had the freezing and thawing
portions of the exposure. The water portion was omitted to ﬁnd
out if freezing had a signiﬁcant effect on the mechanical property

loss.

(ii) The Water only (W) cycle had the water submersion and
conditioning at 23°C 1: 2°C and 50% :I: 4% RH portions. The
omission of the freezing portion would show if exposure to water

caused a signiﬁcant amount of the mechanical property loss.

This part of the study used composites with 100 phr wood ﬂour
manufactured in the second part of the study (high temperature proﬁle) and the

cycle length for both the FT and W cycles was kept constant at 5 cycles.

78

Frogs! Testing

The three-point ﬂexural tests were carried out in the walk-in conditioning
room at 23°C 1 2°C and 50% a: 4% RH on an lnstron 4206 (with Series IX
software) testing machine. The crosshead rate was 4.5 mmlmin in conformance
with ASTM standard 06109-97 (30), the Standard Test Methods for Flexural
Properties of Unreinforced and Reinforced Plastic Lumber. Unless otherwise
mentioned, at least 8 replicates were tested to obtain an average value for each
formulation. Data were collected on modulus of rupture (MOR or ﬂexural
strength) and modulus of elasticity (MOE or ﬂexural stiffness). Per the standard
(30), all unexpowd and freeze-thaw exposed samples were stored in the walk-in
conditioning room described above for at least 48 hours prior to property testing.

Dimensional stability was measured in conformance with ASTM standard
D6341 as mentioned in 06662-01 (28) using digital calipers for measurements of
thickness and width before and after exposures for possible thickness swelling
and width changes. Five replicates were tested to obtain an average value for

each formulation.

The density proﬁle across the width of the sample was determined using
an automated and nondestructive X-ray Density Proﬁler (Model QDP-01X
equipped with Version NT 1.14 software) following the approach described by the
manufacturer (Quintek Measurement Systems, Inc., Oak Ridge, Tennessee,

1999). Five replicates were tested for each condition shown.

79

Statistical Analﬁls

To determine the effect of freeze-thaw cycling on ﬂexural properties of the
composites with various wood ﬂour contents, a two-sample t-test was carried out
with an a signiﬁcance value of 0.05, comparing the exposed and unexposed
data. Comparisons were only performed within species type and between the
unexposed control and samples exposed for 5 freeze-thaw cycles of each wood
content addition level (50, 75 and 100 phr). All statistical analyses were
performed using Design Expert software (Version 6) from Stat-Ease, Inc.
Minnesota.

The statistical analysis for the effect of the number of freeze-thaw cycles
on the properties of the composites was also performed, using the same
software. A two-sample t-test with a = 0.05 was used to compare each cycle
exposure to the unexposed control. Comparisons were only performed within
species type and between unexposed and exposed samples. Ten replicates
were tested to obtain an average value for each exposure cycle.

Ten replicates were also used for the modiﬁed freeze-thaw cycles, water
only (W) and freeze-thaw only (FT), which were compared to the 5 full freeze-
thaw cycles with all of the components (water-freeze-thaw). Samples exposed to
both 5 W and 5 FT cycles were also compared to the unexposed controls using

the same two-sample t-test (a. = 0.05).

80

 

 

Microscopy

Scanning electron microscope (SEM) images of unexposed and exposed
samples (both full and modiﬁed freeze-thaw cycles) were taken to verify whether
or not moisture changes affected the interfacial adhesion in the composites.
Images were taken of the broken surface after ﬂexural testing. Since the most
detrimental effect of freeze-thaw cycling on ﬂexural properties was seen in rigid
PVC ﬁlled with 100 phr maple ﬂour, representative images were taken only from
these composites. These samples were oven dried at 105°C for 24 hours, to
remove moisture, and were than gold coated, for conductance, immediately
before insertion into the SEM. Gold coating was done using a deposition of 20
mA for 2 runs of 4 minutes each. A JEOL JSM-6400 SEM was used with
accompanying analySlS software for image acquisition. The acquired images
had a resolution of 2048 x 1536 pixels with a pixel dwell time of 50 us. The
accelerating voltage was set to 15 W, with a condenser lens of 11 and a working

distance of 39 mm.

81

RESULTS AND DISCUSSION

Effect of Wood-Flour Content

Freeze-thaw actions did not appear to have a substantial effect on the

density of rigid PVCIwood-ﬂour composites. The test results show that the

densities were similar for both the unexposed and full freeze-thaw cycle exposed

samples, irrespective of wood ﬂour content and species (table 3.2).

Table 3.2 Density of rigid PVC/wood-ﬂour composites before and after
exposure to 5 full freeze-thaw cycles.

 

DenSitY (9/0113)

 

Samples 50 phr 75 phr 100 phr

 

Control 5 cycles Control 5 cycles Control 5 cycles

 

Composites with
1.15 1.14 1.21 1.23 1.08 1.10
maple ﬂour
Composites with
_ 1.13 1.15 1.21 1.21 1.12 1.15
pine ﬂour

 

Table 3.3 summarizes the changes in dimensional stability of rigid

PVCIwood-ﬂour composites after exposure to 5 complete freeze-thaw cycles.

The overall dimensional stability was relatively unaffected by the freeze-thaw

cycling, regardless of wood species. Even after 5 freeze-thaw cycles, the width

and thickness changed less than 1.3% and 2.2%, respectively, for the most

82

severely affected composition of rigid PVC ﬁlled with 100 phr maple ﬂour (table
3.3). It was also observed that both the thickness and width changes due to
freeze-thaw cycling increased with wood ﬂour content. Increasing the wood ﬂour
content in the PVC matrix allows more water absorption and potentially
diminished encapsulation of the wood ﬂour by the matrix, thus leading to greater

dimensional changes.

Table 3.3 Changes in dimensional stability of rigid PVC/wood-ﬂour composites
after 5 cycles of full freeze-thaw (WFT) exposure.

 

Percent change in Composites with maple ﬂour Composites with pine ﬂour

 

 

 

properties from the
control 50 phr 75 phr 100 phr 50 phr 75 phr 100 phr
Thickness swell (%) 0.73 1.28 2.17 0.55 0.97 1.72
Width change (%) 0.19 0.32 1.32 0.17 0.44 0.85

 

The percent loss in MOR and MOE of rigid PVC/wood-ﬂour composites
ﬁlled with both maple and pine species exposed to 5 full freeze-thaw cycles were
compared to the unexposed control and the results for the various wood ﬂour
addition levels are summarized in table 3.4. Rigid PVCIwood-ﬂour composites
with lower wood ﬂour contents (50 and 75 phr) of either wood species retained
their strength after exposure to 5 freeze-thaw cycles. The loss in the MOR of the
composites exposed to 5 freeze-thaw cycles was statistically signiﬁcant only for

100 phr of wood ﬂour content for both wood species (table 3.4). This trend was

83

 

expected because the more wood present in the sample, the greater the uptake
of moisture and, therefore, a more dramatic decrease in mechanical properties
(8,20,26,27). Unlike the ﬂexural strength, 5 freeze-thaw cycles caused a
signiﬁcant loss in the MOE for all composites, regardless of wood species and
content (table 3.4). The stiffness of rigid PVCIwood-ﬂour composites with 100 phr
maple and pine decreased by 34% and 30%. respectively, after exposure to 5 full
freeze-thaw cycles. Although a statistical comparison between the maple and
pine was not performed, the results listed in table 3.4 indicate that the
composites with maple wood ﬂour were more detrimentally affected by 5 freeze-

thaw cycles than the pine wood ﬂour composites.

As previously mentioned, the degraded mechanical properties of rigid
PVC-wood ﬂour composites after accelerated freeze-thaw cycling is thought to
be largely due to a degradation of interfacial adhesion in the composites as a
result of exposure to moisture. During freeze-thaw exposure, the samples cycled
through environments of soaking in a water bath until equilibrium moisture
content, freezing at -27°C for 24 hours and thawing at 23°C and 50% RH for
another 24 hours. The majority of this cycling is changing moisture contents
which have been shown to adversely affect the properties of other wood—plastic
composites (20.23-25.29). Moreover, visual observations show a distinctive
change in appearance and lightening in color of exposed samples. Most of the
samples had developed tearing and rough edges following exposure to 5 freeze-
thaw cycles compared to the control. These rough edges are likely a result of

moisture changes and ﬁber-matrix debonding (25).

84

Table 3.4 Flexural properties of rigid PVC/wood-ﬂour composites before and

after 5 cycles of full freeze-thaw (WFT) exposure.

 

 

 

 

Properties Composites with maple ﬂour Composites with pine ﬂour
50 phr 75 phr 100 phr 50 phr 75 phr 100 phr
MOR of 36.61 46.40 31.55 40.30 43.57 37.60
unexposed
samples (MPa)
MOE of 2.04 3.38 2.54 2.35 3.33 3.34
unexposed
samples (GPa)
LOSS In MOR NS NS S NS NS S
(%), after WFT 4.22 5.91 15.19 2.22 8.49 12.53
exposure
Loss in MOE S S S S S S
(%), afterWFT 16.8 22.13 33.93 9.96 18.11 29.79
exposure

 

1
NS means that ﬁre change is not statistically signiﬁcant whereas S implies that the difference is
statistically signiﬁcant at values of ‘Prob>|t|" less than 0.05.

Effect of Number of Freeze-Thaw Cycles

As mentioned, processing conditions had a prevalent effect on the surface

quality and appearance of the composite samples made with 100 phr wood ﬂour.

Composite samples used in the second portion of this study were processed

using a higher temperature proﬁle (190/185/180/180°C) due the increased melt

viscosity of the blend with 100 phr wood ﬂour. Consequently, the processing was

much easier and the samples had more thorough edge adhesion than those

previously manufactured with the lower temperature proﬁle (190/175/170/180°C).

These samples were exposed to various numbers of freeze-thaw cycles (0, 2, 5,

85

8 or 12) and the measured ﬂexural properties are listed in tables 3.5 and 3.6 for
maple and pine, respectively. Both ﬂexural strength and stiffness signiﬁcantly

decreased with increasing numbers of freeze-thaw cycles.

For 100 phr maple ﬂour ﬁlled rigid PVC samples, the loss in ﬂexural
properties (both MOR and MOE) of the exposed samples was signiﬁcant when
statistically compared to the control even after 2 freeze-thaw cycles. Composites
with 100 phr pine ﬂour also showed a signiﬁcant decrease in strength and
stiffness after freeze-thaw exposures, although slightly less pronounced than
maple composites. After exposure to 2 full freeze-thaw cycles the loss in ﬂexural
strength was only 3.5%, which is not statistically signiﬁcant when compared to
the unexposed control. However, a signiﬁcant loss in strength occurred after 5
full freeze-thaw cycles and all cycles thereafter. Unlike for the strength, the loss
in ﬂexural stiffness of pine ﬂour ﬁlled rigid PVC was signiﬁcant after only 2 full
freeze-thaw cycles and leveled-off after about 8 freeze-thaw cycles.
Consequently, it is assumed that composite with pine ﬂour are not likely to loose
much more than 20% of their stiffness. This signiﬁcant loss in ﬂexural properties
of rigid PVC/wood-ﬂour composites may be a cause for concern with these
materials being used in decking and other constructionlsemi-structural

applications.

86

Table 3.5 Effect of the number of freeze-thaw cycles on the ﬂexural properties
of maple ﬁlled rigid PVCIwood-ﬂour composites.

 

(a) Composites with maple ﬂour1

 

Number of Freeze-Thaw Cycles

 

Properties
0 2 5 8 12

 

MOR(MPa) 33.81 31.49S 29.763 29.738 28.663

MOE (GPa) 2.71 2.228 2.058 1.95S 1.90S

 

lNS means that the change compared to the contrOI is not statistically
signiﬁcant whereas S implies that the difference is statistically signiﬁcant at
values of “Prob>|t|” less than 0.05.

Table 3.6 Effect of the number of freeze-thaw cycles on the ﬂexural properties
of pine ﬁlled rigid PVC/wood-ﬂour composites.

 

(b) Composites with pine ﬂour1

Properties

 

Number of Freeze-Thaw Cycles

 

O 2 5 8 12

 

MOR(MPa) 38.93 37.57NS 36.713 35.98S 35.39S

MOE (GPa) 3.37 2.948 2.748 2.688 2.71S

 

lNS means that the change compared to the control is not statistically
signiﬁcant whereas S implies that the difference is statistically signiﬁcant at
values of “Prob>|t|” less than 0.05.

87

Modiﬁed Freeze-Thaw Cycles

To determine the cause of the above mentioned loss in mechanical
properties in rigid PVC/wood-ﬂour composites, the full freeze-thaw cycle was
broken down into its components. Either the water or freezing portion of the full
freeze-thaw cycle was omitted and the number of cycles was held constant at

ﬁve.

Figure 3.1 shows the effect of the freeze-thaw cycle portions on the
ﬂexural properties of PVC ﬁlled with 100 phr maple ﬂour. The freezing portion (FT
only cycle) had very little effect on the ﬂexural properties of the composites and
was not statistically different from the control. Five FT only cycles actually
showed a slight increase in MOR compared to the unexposed control. This
insigniﬁcant increase in MOR leads us to believe that the freezing portion of the
full freeze-thaw cycle does not signiﬁcantly affect the ﬂexural properties of the
composites with maple ﬂour. However, the water portion of the cycle (exposed to
5 W cycles) had a signiﬁcant impact upon the ﬂexural properties of the
composites compared to the unexposed control. Flexural properties of the
composites exposed to 5 full freeze-thaw cycles (water, freezing and thawing or
WFT) were not statistically different from those exposed to water only cycles
(with no freezing). Consequently, the loss in both MOE and MOR of the
composites ﬁlled with 100 phr maple ﬂour could be attributed to the water portion

of the full freeze-thaw cycle process.

88

A
N
V

 

   

 

 

 

 

 

 

 

 

I: 25 EMOR
°- 2
g 0 IZIMOE
.E 15
8 10
a s a
§ 0
3 -5
a.
-10
WFT FT W
Cycles
bl
( 30
E‘
a
2
O.
.E
3
2
E
0
2
O
n.
WFT FT W
Cycles

Figure 3.1 The effect of each freeze-thaw cycle portion on the loss of ﬂexural
properties of rigid PVC ﬁlled with 100 phr wood ﬂour:
(a) maple and (b) pine.

WFT cycle represents the full freeze-thaw cycle (water soak-freeze-thaw).
W cycle includes only the water soaking and conditioning at 23°C :I: 2°C and
50% :t 4% RH portions (without freezing).

FT cycle includes freezing and thawing only (without water soak).

The cycle length for WFI', W, and FT cycles was kept constant at 5 cycles.

89

Rigid PVC ﬁlled with 100 phr of pine ﬂour showed a slightly different trend
than that observed with maple ﬂour (ﬁgure 3.1). Compared to the unexposed
control, both the ﬂexural strength and stiffness of the composites ﬁlled with pine
ﬂour increased after exposure to 5 freeze-thaw only (FT) cycles. However, this
increase in ﬂexural properties was not statistically signiﬁcant compared to the
control, meaning that ﬂexural property loss cannot be attributed to the freezing
portion of the cycle. The loss in MOE of the composites exposed to 5 water only
(W) cycles was statistically the same as that of 5 full freeze-thaw (WFT) cycles.
For the loss in MOR, however, 5 water only cycles produced signiﬁcantly
different results compared to 5 full freeze-thaw (WFT) cycles. Nevertheless, the
signiﬁcance level for the MOR values was 0.0483, which is very close to the
necessary 0.050 to be statistically the same. Since the freezing part of the cycle
is known to have no effect, the ﬂexural property loss is attributed to the water
submersion part of the full freeze-thaw cycle for 100 phr pine ﬁlled rigid PVC as

well.

For 100 phr (about 46 wt.%) pine ﬂour ﬁlled rigid PVC samples, a
decrease of approximately 3.5% in ﬂexural strength was observed in this study
with exposure to 5 water only cycles. This loss in MOR was considerably less
than that reported by Clemons and lbach (23) for 50% pine ﬂour ﬁlled HDPE
matrix. A decrease of approximately 24.5% in ﬂexural strength was observed
after 5 cycles of ambient temperature water soaking followed by 65% RH at 27°C
conditioning (23). This indicates that PVC is likely a superior matrix for resistance

towater only cycling compared to HDPE matrix.

90

Scanning Electron Mlcggraphs

Representative SEM images of the broken surface after ﬂexural testing
(ﬁgure 3.2) were taken to qualitatively determine the effects of the freeze-thaw
cycling (full and modiﬁed cycles) on the interfacial adhesion between the wood
ﬂour and PVC matrix. As mentioned previously the 100 phr maple ﬁlled rigid PVC
was used for imaging since the most prevalent impact on properties was seen in

these samples.

For the unexposed control samples, the SEM image (ﬁgure 3.2a) shows
that there was more breakage of the polymer matrix and even ﬁbers are tearing
in half. Although there were some gaps or crevices, these were probably due to
manufacturing or the stresses from ﬂexural testing, however the matrix and ﬁbers

were still well adhered to each other.

A considerable number of holes and cavities that have been formed from
ﬁbers pulling out of the matrix are clearly seen for the composite sample exposed
to 5 full freeze-thaw cycles with water, freezing and thawing (ﬁgure 3.2b). There
are also intact ﬁbers that are indicative of holes in the other half of the broken
sample where the ﬁbers used to be bound. These holes and intact ﬁbers are
evidence of loss in bonding between the ﬁbers and the matrix for the composites
exposed to 5 full freeze-thaw cycles. This result is consistent with ﬁndings from
vinylester/E-glass composites where microcracking and ﬁber-matrix debonding

were observed after exposure to water (25).

The SEM image of the composite sample exposed to 5 water only (W)

cycles (ﬁgure 3.20) bears a striking resemblance to that of 5 full freeze-thaw

91

(WFT) cycles (ﬁgure 3.2b). As seen with 5 full freezethaw cycles (ﬁgure 3.2b).
there are a lot of holes where ﬁbers used to be and some ﬁbers that have left
holes on the other half of the broken sample, suggesting that the water only and
full freeze-thaw cycles are essentially the same. This striking resemblance further
conﬁrms that the majority of the ﬂexural property loss is due to the water portion

of the freeze-thaw cycling process.

92

.8530: Songs 8.96 >>v $.96 2:988; mac 9 oowoaxo
oEEmw 8v new .3539: cam .mENoot .585 mo_o>o BmcsroNooc :3 was 9 nowocxo anmm 5
__o.Eoo oomoaxocs Amy usoc coo; came Ea cor 53> 8:: o>n_ Em: co 23392:. 2mm mm 239“.

. , ‘56“ .i .
t . .21;
.. . v

a

 

CONCLUSIONS

This study examined the effects of accelerated freeze-thaw cycling on the
durability of rigid PVC/wood-ﬂour composites. Particular emphasis was placed on
investigating the effects of wood ﬂour content, number of freeze-thaw cycles,
freeze-thaw cycle portions and water absorption on the properties of the
composites. The following conclusions can be drawn from the experimental

results.

Cyclic freeze-thaw actions did not affect the density of rigid PVC/wood-
ﬂour composites at any wood ﬂour content for either the maple or pine ﬁller. Both
the thickness swell and width changes due to freeze-thaw cycling increased with
wood ﬂour content. However, the greatest increase was still very small compared

to other materials such as solid wood.

The stiffness of the composites was signiﬁcantly affected by 5 full freeze-
thaw cycles, regardless of both the wood species and content. Conversely, the
strength of the composites was not signiﬁcantly affected by the freeze-thaw
actions at lower wood ﬂour content (50 and 75 phr). However, the deleterious
effects of freeze-thaw cycling on the strength of the composites became
apparent at the highest wood ﬂour content (100 phr) due to the increased wood
content and therefore increased moisture uptake, which possibly caused
decreased encapsulation of the wood ﬂour by the matrix. The 100 phr maple

samples did not appear as well processed as the other samples, as was

94

evidenced by the lower than expected density. This manifested itself in lower

performance after exposure in the freeze-thaw test.

The loss in ﬂexural strength of the remade composites ﬁlled with 100 phr
wood ﬂour was statistically signiﬁcant after 2 freeze-thaw cycles, regardless of
wood species. However, maple ﬂour ﬁlled rigid PVC composites showed a
signiﬁcant loss in ﬂexural strength after 2 full freeze-thaw cycles, whereas it took
5 full freeze-thaw cycles for the composites with 100 phr pine ﬂour to show a

signiﬁcant decrease in ﬂexural strength.

The water only portion of the cycling process had the greatest impact
upon the ﬂexural properties and is likely attributable for the majority of the
property loss due to the moisture effects on the adhesion between the matrix and
wood ﬂour. The freeze-thaw only portion of the cycling did not signiﬁcantly affect

the ﬂexural properties of the composites.

Moreover, SEM images exhibited similarities between water only (W)
cycles and full freeze-thaw (WFT) cycles. Both of these exposures showed a loss
of bonding between the wood ﬂour and PVC matrix, which resulted in cavities
and intact ﬁbers protruding from the matrix. However, the unexposed control

composite exhibited matrix and ﬁber breakage.

Overall, rigid PVC/wood-ﬂour composites have been shown to have been
signiﬁcantly impacted by cyclic water-freeze-thaw actions. With uses in outdoor
exposure, such as decking, the durability of WPCs is a concern. Other types of
WPCs used in outdoor exposures should be studied.

95

REFERENCES

. L.M. Matuana, C.B. Park, and J.J. Balatinecz (1997) The effect of low level
plasticizer on the rheological and mechanical properties of polyvinyl
chloride/newsprint-ﬁber composites. J. Vinyl Addit. Technol. 3(4):265—273.

. L.M. Matuana, J.J. Balatinecz, and CB. Park (1998) Effect of surface
properties on the adhesion between PVC and wood veneer laminates. Polym.
Eng. Sci. 38(5):765-773.

. L.M. Matuana, R.T. Woodhams, J.J. Balatinecz, and CB. Park (1998)
Inﬂuence of interfacial interactions on the properties of PVC/cellulosic ﬁber
composites. Polym. Comp. 19(4):446-455.

. F. Mengeloglu, L.M. Matuana, and J.A. King (2000) Effects of impact
modiﬁers on the properties of rigid PVC/wood-ﬁber composites. J. Vinyl Addit.
Technol. 6(3):153-157.

. C. Clemons (2002) Wood-plastic composites in the United States: the
interfacing of two industries. Forest Prod. J. 52(6):10-18.

. G.E. Meyers, I.S. Chahyadi, C.A. Coberly, and US. Ermer (1991) Wood ﬂour/
polypropylene composites: inﬂuence of maleated polypropylene and process
and composition variables on mechanical properties. Int. J.Polym. Mater.
15:21-44.

. S. Takase and N. Shiraishi (1989) Studies of composites from wood and
polypropylenes. II. J. Appl. Polym. Sci. 37(3):645-659.

. RM. Rowell, S.E. Lange, and RE. Jacobson (2000) Weathering performance
of plant-ﬁberlthennoplastic composites. Mol. Cryst. Liq. Cryst. 353:85-94.

. J. Patterson (2001) New opportunities with wood-ﬂour-foamed PVC. J. Vinyl
Addit. Technol. 7(3):138-141.

10.Anonymous (2002) Decking sales strong for 1“ half 2002: Producers continue

to add capacity. Natural & Wood Fiber Composites. 1(8):1.

11.M. Mankowski and J.J. Morrell (2000) Patterns of fungal attack in wood-

plastic composites following exposure in soil block tests. Wood Fiber Sci.
32(3):340-345.

12.P.I. Morris and P. Cooper (1998) Recycled plastic/wood composite lumber

attacked by fungi. Forest Prod J. 48(1):86-88.

96

13. P. Mapleston (2002) It’s one hot market for proﬁle extruders. Modem Plastics
June 2002: 49-52.

14.Anonymous (2002) Arsenic-treated wood to be phased out of residential use.
Chemical and Engineering News. Feb. 2002: 27.

15.Anonymous (2002) Wood substitutes beneﬁt from EPA move. Plastics in
Canada eNewsletter. March 13, 2002.

16.W. Chetanachan, D. Sookkho, W. Sutthitavil, N. Chantasatrasamy and R.
Sinserrnsuksakul (2000) PVC/wood: A new look in construction. Proceedings
of Vinyltec 2000, sponsored by Vinyl Division-SPE, Philadelphia Section.
Philadelphia, PA. October 1 1-12, 2000.

17.L.M. Matuana, D.P. Kamdem, and J. Zhang (2001) Photoaging and
stabilization of rigid PVC/wood-ﬁber composites. J. Appl. Polym Sci.
80(11):1943-1950.

18. L.M. Matuana and DP. Kamdem (2002) Accelerated ultraviolet weathering of
PVC/wood-ﬁber composites. Polym. Eng. Sci. 42(8):1657-1666.

19.JM. Pilarski and L.M. Matuana (2005) Durability of wood ﬂour-plastic
composites exposed to accelerated freeze-thaw cycling. Part II. High density
polyethylene matrix. SUBMITTED to J. Appl. Polym. Sci.

20. NM. Stark (2001) Inﬂuence of moisture absorption on mechanical properties
of wood ﬂour-polypropylene composites. J. Thermoplast. Comp. Mater.
14(5):421-432.

21 . S.V. Rangaraj and L.V. Smith (2000) Effects of moisture on the durability of a
wood/thennoplastic composite. J. Thermoplast. Comp. Mater. 13(3):140-161.

22.J.J. Balatinecz and 8D. Park (1997) The effects of temperature and moisture
exposure on the properties of wood-ﬁber thermoplastic composites. J.
Thermoplast. Comp. Mater.10(9):476-487.

23.C.M. Clemons and RE. lbach (2004) Effects of processing method and
moisture history on laboratory fungal resistance of wood-HDPE composites.
Forest Prod. J. 54(4):50-57.

24.D.A. Johnson, D.A. Johnson, J.L. Urich, RM. Rowell, R. Jacobson and DP.
Cauﬁeld (1999) Weathering characteristics of ﬁber-polymer composites.
Proceedings of the 5‘” lnt’l Conference on Woodﬁber-Plastic Composites.
Sponsored by the USDA Forest Service. Madison, WI May 26-27, 1999. pp.
203-209.

97

25.V.M. Karbhari, J. Rivera, and J. Zhang (2002) Low-temperature hygrotherrnal
degradation of ambient cured E-glassNinylester composites. J. Appl. Polym.
Sci. 86(9):2255-2260.

26. KC. Manikandan Nair and S. Thomas (2003) Effect of ageing on the
mechanical properties of short sisal ﬁbre reinforced polystyrene composites.
J. Thermoplast. Comp. Mater. 16(3):249-271.

27M. Tajvidi and G. Ebrahimi (2003) Water uptake and mechanical
characteristics of natural ﬁller-polypropylene composites. J. Appl. Polym. Sci.
88(4):941-946.

28.American Society of Testing and Materials (2001) ASTM 06662-01 Standard
speciﬁcation for polyoleﬁn-based plastic lumber decking boards. Philadelphia,
PA.

29.L.M. Matuana and J.M. Maurice (2003) Effect of freeze-thaw cycles on the
properties of rigid PVC/wood-ﬂour composite lumber. Proceedings of Vinyltec
2003, sponsored by Vinyl Division-SPE, Ohio Firelands Section. Huron, OH.
October 27-29, 2003.

30.American Society of Testing and Materials (1997) ASTM 06109-97 Standard

test methods for ﬂexural properties of unreinforced and reinforced plastic
lumber. Philadelphia, PA.

98

Chapter 4

DURABILITY OF WOOD FLOUR-PLASTIC COMPOSITES EXPOSED TO
ACCELERATED FREEZE-THAW CYCLING
PART II. HIGH DENSITY POLYETHYLENE MATRIX

This chapter is slightly modiﬁed from Jouma/ of Applied Polymer Science. In

submission as of February 2005. It is co-authored by J.M. Pilarski1 and L.M.
Matuana‘.

‘ Department of Forestry, Michigan State University, East Lansing, MI, 48824

99

ABSTRACT

This study examined the durability of extruded HDPE/wood-ﬂour
composites exposed to 15 accelerated cycles of water submersion, freezing and
thawing according to ASTM standard D6662. The durability of both maple and
pine composites was assessed by testing the ﬂexural properties and density.
Mercury intrusion porosimetry and scanning electron microscopy were also used
to evaluate the interfacial adhesion between the matrix. and wood ﬂour before
and after exposure to accelerated freeze-thaw cycling. Freeze-thaw actions had
no apparent effect on the density of the composites alter exposure, regardle’ﬁs of
the wood species. However, these actions led to moisture uptake, which
decreased the interfacial adhesion and increased the pore size and quantity in

the composites, which resulted in a signiﬁcant loss in ﬂexural properties.

100

INTRODUCTION

Wood-plastic composites (WPCs) have become prevalent in many
building applications partially due to the need to replace pressure-treated solid
lumber (1-3). Several applications of WPCs require outdoor exposure or ground
contact, where the material is susceptible to various weakening agents, such as
fungi and insects (4-6). The weather itself may also cause the product to
deteriorate, which is evident in work done on moisture and ultraviolet radiation
(7,8). However, the effect of exposure to cold winter conditions is not well-
documented. Although WPCs are typically seen as combining the best properties
of wood and plastic, their combination makes a brittle product, which may
become more brittle by cold exposure and further decrease the mechanical
properties (9-11). Since these products are used in building applications the

retention of strength is essential.

Our preliminary study showed that rigid PVC/wood-ﬂour composites
exposed to freeze-thaw cycling lose some of their mechanical properties (12).
The in-depth study conducted with PVC ﬁlled with maple and pine wood ﬂour
showed signiﬁcant losses in both ﬂexural strength and stiffness after exposure to
various numbers of freeze-thaw cycles (13). Additionally. this study showed the
development of cracking and de-bonding inside of the composite, which was
attributed to the absorption of water during weathering (13). Although PVC was
previously examined, the majority of the WPCs in North America, over 70%, are

manufactured with polyethylene thermoplastics (14). This study was aimed to

101

 

 

further investigate the effects of accelerated freeze-thaw cycling on the density
and ﬂexural properties of HDPE composites ﬁlled with 50% maple or pine wood
ﬂour after exposure to extreme conditions. Since the previous studies examined
the effect of varying numbers of freeze-thaw cycles, this work focused on the

extreme case of 15 freeze-thaw cycle exposures.

102

 

 

EXPERIMENTAL

Materials

Due to the popularity of polyethylene resins in the wood-plastic composite
market (14), HDPE, designated as FORTIFLEXTM B53-35H-FLK from B.P.
Solvay Polymers (Houston, TX), was used as the polymeric matrix. The polymer
was in ﬂake form with a melt ﬂow index (MFI) of 0.49 g/10min and density of 0.9
g/cm3. The 0.425mm (40 mesh) maple and pine wood ﬂours were obtained from
American Wood Fibers (Schoﬁeld, WI) to compare hardwood and softwood
species. The wood ﬂour was oven-dried at 105°C for approximately 48 hours
before processing to remove moisture. The TPW104 lubricant from Struktol"
Company (Stow, OH) was used to reduce friction between components and
equipment, which eased processing. This lubricant is a blend of an aliphatic

carboxylic acid salts and mono and diamides.

Comppunding and Extrusion of Commit»

A 10 liter high intensity mixer (Papenmeier, Type TGAHK20) was used for
room temperature dry blending of the HDPE matrix, dried wood ﬂour and
lubricant. The formulation of the composites was maintained at 50% wood ﬂour
(maple or pine), 44% HDPE and 6% lubricant by total weight of the composites.
The compounded materials were fed into a 32 mm conical counter-rotating twin-

screw extruder (C.W. Brabender Instruments Inc.) with a length to diameter ratio

103

of 13:1. This twin screw extruder was equipped with an unpressurized vent to
allow a residual moisture to escape and was powered by a 7.5 hp Intelli-Torque

Plasti-Corder Torque Rheometer”.

The heating proﬁle, from the hopper to the horizontal die was 190, 175.
170, 165°C with a rotational speed of the screws maintained at 40 rpm
throughout processing. The rectangular die created samples with a nominal

width and thickness of 2.54 cm (1") and 0.95 cm (3/8"), respectively.

Freeze-Thaw Cycling

Freeze-thaw cycling was done in accordance with a modiﬁed ASTM
06662-01 (15), the Standard for Polyoleﬁn-Based Plastic Lumber Decking
Boards. One complete freeze-thaw cycle consisted of three parts: (i) a water
soak until equilibrium moisture content (EMC), (ii) exposure to freezing for 24
hours, and (iii) thawing for 24 hours. The number of freeze-thaw cycles was
maintained at 15 cycles to examine extreme conditions since intermediate cycles

have already been examined (13).

The water submersion part of the cycle was conducted in room
temperature water (approximately 21°C). Samples were weighed every 24 hours
until the weight gain was less than 1%, which implied that the sample had
reached its EMC. A General Electric 0.38 cubic meter (13.4 ft3) chest freezer

controlled to -27°C :I: 2°C was used for the freezing portion of the cycle. Lastly,

104

the samples were thawed in a walk-in conditioning room at 23°C 1: 2°C and 50%

:I: 4% relative humidity (RH).

Propppty T»ting

The three—point ﬂexural tests were carried out in the walk-in conditioning
room using an lnstron 4206 testing machine (with Series IX software) according
to ASTM standard D6109-97 (16), the Standard Test Methods for Flexural
Properties of Unreinforced and Reinforced Plastic Lumber. Fifteen replicates
were tested and data was collected on modulus of rupture (MOR or ﬂexural
strength) and modulus of elasticity (MOE or ﬂexural stiffness).

The density proﬁle across the thickness of the sample was determined
using an automated and nondestructive X-ray Density Proﬁler (Model GDP-01X
equipped with Version NT 1.14 software) following the approach described by the
manufacturer (Quintek Measurement Systems, Inc., Oak Ridge, Tennessee,

1999). Five replicates were tested for each formulation.

Statistical Analysis

Design-Expert v.6 software from the Stat-Ease Corp. (Minneapolis, MN)
was used to perform statistical analyses to determine the effect of freeze-thaw
cycling on properties of the composites. A two-sample t-test and Duncan's
multiple range tests were employed to determine the statistical differences

among the variables investigated at a 95% signiﬁcance level. Comparisons were

105

 

done between exposed and unexposed samples as well as between wood

species for density and ﬂexural properties.

Porosimet_ry

Porosity testing was conducted by the Micromeritics Instrument
Corporation to quantify the porosity in the samples before and after freeze-thaw
exposures. Samples had a nominal size of 2.54 cm (1") thick by 0.950m (3/8”)
wide from the die on the extruder. The sample was cut from near the middle of
the sample after ﬂexural testing. The broken end was out off and then a 2.54 cm
(1") length was out. One sample each of the unexposed (control) and 15 freeze-
thaw exposures were tested for composites with both maple and pine wood ﬂour.
An AutoPore IV 9250 was used for testing, which has a maximum pressure
capability of 414 MPa (60,000 psi). Mercury was used for all pore intrusion since

it cannot move through the samples by capillary action.

Penetrometers were used to hold the individual samples when they were
exposed to mercury intrusion and had a nominal volume of 6 mL. Initially the
penetrometers were placed in low pressure chambers to force mercury into the
larger macropores, then they were individually moved into high pressure

chambers for mesopore testing and the majority of the data collection.

106

Microscopy

Scanning electron microscope (SEM) images of unexposed and exposed
samples were taken to verify whether or not freeze-thaw actions affected the
interfacial adhesion in the composites. Images were taken of the broken surface
after ﬂexural testing of the composites with maple ﬂour since the most signiﬁcant
ﬂexural loss was observed with these samples. The samples were oven dried at
105°C for 24 hours, to remove moisture, and were then gold coated, for
conductance, immediately before insertion into the SEM. Gold coating was done
using a deposition of 20 mA for 2 runs of 4 minutes each. A JEOL JSM-6400
SEM was used with accompanying analySlS software for image acquisition. The
accelerating voltage was set to 15 W, with a condenser lens of 11 and a working

distance of 39 mm.

107

RESULTS AND DISCUSSION

Only the extreme case of 15 full freeze-thaw cycles was examined in this
study. Results from a previous study with rigid PVC/wood-ﬂour composites
indicated a general trend of increasing property loss with increasing numbers of
freeze-thaw cycles (13). However, there was a plateau effect after a threshold of
about 5 cyclic exposures were completed. Consequently, these HDPE/wood-ﬂour
composites were exposed to 10 more cycles than those with PVC for a total of 15
cycles to simulate the most extreme effects possible so that the maximum

property loss was certain to be observed.

Changes in density and ﬂexural properties in HDPE/wood-ﬂour
composites before and after exposure to freeze-thaw cycling are summarized in
table 4.1. The observed loss in ﬂexural properties was considerably larger for the
composites with maple ﬂour than those with pine ﬂour. Composites made with
maple wood ﬂour lost about 21% of their strength and almost 50% of their
stiffness. Whereas. composites with pine ﬂour only lost about 5% in strength and
near 38% of their stiffness. These losses are all signiﬁcant compared to their
unexposed counterparts (control samples). The greater ﬂexural property loss in
maple ﬂour ﬁlled HDPE composites compared to those with pine was likely a
result of their lower density and greater porosity (discussed in the following
section). This favored water uptake and decreased interfacial bonding after

freeze-thaw exposure in maple composites.

108

The density within either species composite was unaffected by freeze-
thaw cycling exposure (table 4.1). However, the composites with pine ﬂour were
higher density and therefore less porous than those with maple ﬂour. This made
pine samples less sensitive to the damage caused by freeze-thaw exposure.
Higher density WPCs absorb less moisture (17), which is the primary cause of
mechanical property loss. As expected, composites with pine wood ﬂour reached
the equilibrium moisture content faster and subsequently gained less moisture
per cycle (3.1 wt.% vs. 4.6 wt.% for maple) than those with maple ﬂour.
Consequently, the higher density pine composites were more likely to retain most
of their ﬂexural properties, particularly their bending strength. Greater moisture
uptake observed in maple composites may be indicative of decreased bonding

between the matrix and ﬁller. which resulted in the greater ﬂexural property loss.

Table 4.1 Changes in the physical and mechanical properties in HDPE
composites before and after exposure to freeze-thaw cycling.

 

 

 

Flexural and Cornposlt» wlth maple flour1 Composites with pine flour1
Physical
Properties 0 (control) 15 cycles 0 (control) 15 cycles

 

MOR (MPa) 20.73 A a: 2.26 16.31 3 1: 1.40 24.68 C s 1.04 23.47 D 1 0.51

MOE (GPa) 1.63 A s 0.33 0.83 B 1 0.11 2.21 c a 1.04 1.38 D 1 0.07

 

Density (910m3) 0.89 A s 0.03 0.90 A s 0.02 0.99 B 1: 0.003 1.01 B s 0.02

 

TMeans with the same letter are not signiﬁcantly different based on Duncan’s grouping.

109

Porosimegy

Porosimetry was conducted for quantitative evaluation of the porosity in
the samples before and after 15 freeze-thaw exposures. Figure 4.1 illustrates
graphs of pore size (diameter) class versus incremental intrusion (pore volume)
for composites with both wood species. A distinct difference between the
exposed and unexposed samples is clearly seen in the range between about 1
pm and 6 pm. As the incremental intrusion increased, there were more pores of
that particular diameter. In other terms, exposure to freeze-thaw cycles increased
the incremental intrusion, implying that more pores were created in this range,
regardless of wood species. This general increase in porosity was likely due to a
decrease in bonding between the matrix and ﬁbers after exposure. Moreover, this
loss in bonding resulted in a small amount of debris being present in the water
after the submersion portion of the cycle, which could be the material that was
previously in the newly created porous spaces. The increase in the size and

number of pores correlates with the loss in mechanical properties.

The denser composites with pine ﬂour had fewer and smaller sized pores
compared to the maple composites. The higher density and lower overall porosity
allowed pine composites to resist moisture uptake and better retain interfacial
bonding. This was expected since the decrease in ﬂexural properties was not as

drastic for the pine samples as for the maple.

110

 

0.0080

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Maple-control

5i - - - -Maple-15 full cycles

5 0.0060— —-

C

.9

VI

3 0.0040 4- ,

.5

1'9

5

5 0.0020

2

O

.5

0.0000 . , - .
0 1 2 3 4 5 6 7
Pore slze(pm)
0.0080

Pine-control

51 - - - -Pine-15 full cycles

E 0.0060 ~-———— —— > ”T

C

.2

m

5 0.0040 -_,‘_________,__2 2-. _ _,- _ —— --—--

C

E

5

5 0.0020 - T‘ — 7

o . ' ‘

r: . ‘
— \
0.0000 . .

o 1 2 3 4 5 6 7

Pore size (pm)

Figure 4.1 The effect of freeze-thaw cycling on porosity: (a) maple-HDPE
composites and (b) pine-HDPE composites.

111

Scanning Electron Micmraphs

Scanning Electron Microscopy (SEM) was used to evaluate the interfacial
adhesion between the matrix and ﬁbers in order to determine to the main cause
for property loss for the exposed composites. In previous work SEM images of
rigid PVC/wood-ﬂour composites showed that freeze-thaw cycling reduces the

bonding between the matrix and ﬁbers, thus reducing the ﬂexural properties (13).

The HDPE matrix behaved similarly to PVC and SEM images shown in
Figure 4.2 support the ﬁnding of our porosity investigation (Figure 4.1). The
unexposed sample showed a considerable amount of ﬁber and matrix breakage
with very little intact material on the surface (Figure 4.2a) as a result of good
adhesion between the maple wood ﬁbers and HDPE matrix. Conversely,
exposure to 15 freeze-thaw cycles decreased bonding between the matrix and
ﬁbers as clearly shown in Figure 4.2b where intact ﬁbers as well as a cavity
where an intact ﬁber was pulled out from the matrix are seen. This loss in
bonding supports the porosity results and is consistent with previous work on the
freeze-thaw cycling of rigid PVC/wood-ﬂour composites (13) and E-glass with

vinyl ester (1 8).

112

 

Figure 4.2 SEM micrographs of HDPE ﬁlled with maple wood ﬂour: (a)
unexposed control and (b) sample exposed to 15 freeze-thaw
cycles (WFT including water, freezing and thawing).

113

CONCLUSIONS

The results of this study indicated that accelerated freeze-thaw cycling has
a signiﬁcant impact upon the durability of HDPE composites ﬁlled with 50% of
either maple or pine wood ﬂour. The HDPE/maple-ﬂour composites showed the
greatest losses of 49% in MOE and 21% in MOR after 15 accelerated freeze-
thaw cycles. This property loss may cause problems if this material is used in
construction applications. The pine-HDPE composites, on the other hand,
performed much better for MOR with a loss of only 5%, but had a loss of 37% for
the MOE.

The cycling actions had no apparent effect upon the density of either the
maple or pine ﬂour ﬁlled HDPE composites. However, a lower density was
observed for maple samples compared to the pine, which resulted in increased

water uptake and property loss for the composites with maple ﬂour.

Regardless of wood species, more and larger pores were created in the
composites exposed to freeze-thaw actions. particularly in the range between
about 1 and 6 um. However, the composites with maple ﬂour had a higher
porosity than those with pine ﬂour for both the control and exposed composites,
likely because of their lower density. This increased porosity was partially the

cause for the lower ﬁexural properties of the composites with maple ﬂour.

SEM images conﬁrmed that composite samples exposed to 15 freeze-

thaw cycles had decreased bonding between the matrix and ﬁbers as expected

114

from porosity tests. This led to ﬁbers extending out of the matrix and leaving
cavities in the other side of the break surface after exposure. The unexposed
samples (control) clearly showed the matrix and ﬁbers breaking, so the cycling

must have diminished the interfacial bonding.

115

REFERENCES

. Anonymous (2002) Decking sales strong for 1" half 2002: Producers continue
to add capacity. Natural & Wood Fiber Composites. 1(8):1.

. P. Mapleston (2002) It’s one hot market for proﬁle extruders. Modern Plastics
June 2002: 49-52.

. Anonymous (2002) Arsenic-treated wood to be phased out of residential use.
Chemical and Engineering News. Feb. 2002: 27.

. M. Mankowski and J.J. Morrell (2000) Patterns of fungal attack in wood-
plastic composites following exposure in soil block tests. Wood Fiber Sci.
32(3):340-345.

. P.l. Morris and P. Cooper (1998) Recycled plastic/wood composite lumber
attacked by fungi. Forest Prod J. 48(1):86-88.

. W. Chetanachan, D. Sookkho, W. Sutthitavil, N. Chantasatrasamy and R.
Sinsemisuksakul (2000) PVC/wood: A new look in construction. Proceedings
of Vinyltec 2000, sponsored by Vinyl Division-SPE, Philadelphia Section.
Philadelphia, PA. October 11-12, 2000.

. NM. Stark and L.M. Matuana (2003) Ultraviolet weathering of photostabilized
wood-ﬂour-ﬁlled high-density polyethylene composites. J. Appl. Polym. Sci.
90(1 0):2609—261 7.

. L.M. Matuana and DP. Kamdem (2002) Accelerated ultraviolet weathering of
PVC/wood-ﬁber composites. Polym. Eng. Sci. 42(8): 1 657-1 666.

. L.M. Matuana, C.B. Park, and J.J. Balatinecz (1997) The effect of low levels
plasticizer on the rheological and mechanical properties of polyvinyl
chloridelnewsprint-ﬁber composites. J. Vinyl Addit. Technol. 3(4):265-273.

10.F. Mengeloglu, L.M. Matuana, and J.A. King (2000) Effects of impact

modiﬁers on the properties of rigid PVC/wood-ﬁber composites. J. Vinyl Addit.
Technol. 6(3):153-157.

11.G.A. Cooper and MR. Piggot (1977) Cracking and fracture in composites.

Proceedings of the 4'” lnt’l Conference on Fracture, sponsored by the
lntemational Congress on Fracture (ICF). University of Waterloo, Canada.
June 19-24, 1977.

116

12.L.M. Matuana and J.M. Maurice (2003) Effect of freeze-thaw cycles on the
properties of rigid PVC/wood-ﬂour composite lumber. Proceedings of Vinyltec
2003, sponsored by Vinyl Division-SPE, Ohio Firelands Section. Huron, OH.
October 27-29, 2003.

13.JM. Pilarski and L.M. Matuana (2005) Durability of wood-ﬂour plastic
composites exposed to accelerated freeze-thaw cycling. Part I. Rigid PVC
matrix. J. Vinyl Addit. Technol. 11(1):1-8.

14.J. Patterson (2001) New opportunities with wood-ﬂour-foamed PVC. J. Vinyl
Addit. Technol. 7(3):138-141.

15.American Society of Testing and Materials (2001) ASTM D6662-01 Standard
speciﬁcation for polyoleﬁn-based plastic lumber decking boards. Philadelphia,
PA.

16.American Society of Testing and Materials (1997) ASTM D6109-97 Standard
test methods for ﬂexural properties of unreinforced and reinforced plastic
lumber. Philadelphia, PA.

17.C.M. Clemons and RE. lbach (2004) Effects of processing method and
moisture history on laboratory fungal resistance of wood-HDPE composites.
Forest Prod. J. 54(4):50-57.

18.VM. Karbhari. J. Rivera, and J. Zhang (2002) Low-temperature hygrothennal

degradation of ambient cured E-glass/Vinylester composites. J. Appl. Polym.
Sci. 86(9):2255-2260.

117

Chapter 5

SUMMARY OF FINDINGS

The use of accelerated laboratory testing has shown that extruded wood-
plastic composites are all susceptible to freeze-thaw actions, although to varying
degrees. Composite compositions were varied with different wood contents,
wood species, plastic types, and additives. Regardless of the formulation,
signiﬁcant property loss was observed for all composite formulations. Some
formulations took slightly longer to reach this level of property loss, but property
loss is inevitable for all WPCs studied. Signiﬁcant effects were seen after as few
as 2 freeze-thaw cycles for some composites. Generally, the stiffness was
affected to a greater extent than the strength. However, there has not been any
conclusive link as to what has caused the MOE to be affected more than the
MOR.

Work with both PVC and HDPE composites have shown that the density
of the composites remain stable throughout the freeze-thaw exposure. The
dimensional stability is also relatively well intact as was shown for PVC. There
were some dimensional increases, likely from the exposure to water, but these
were relatively low compared to other materials, such as solid wood or other
wood-based composites. The water portion has also been attributed as the main
cause of ﬂexural property loss. This was only tested for PVC composites with
maple and pine wood ﬂour, but likely extends to HDPE composites because the

composites have behaved similarly in other ways.

118

Scanning electron microscopy (SEM) has shown that the exposure to
freeze-thaw cycling increases microscopic cracking. PVC composites ﬁlled with
100 phr (~46 wt.%) maple wood ﬂour and HDPE composites with 50% maple
wood ﬂour were examined since they displayed greater ﬂexural property losses
compared the pine ﬁlled composites. The full cycle exposure and water-only
exposure appeared the same in SEM images for both polymers, which further
conﬁrms that the water portion is attributable for the majority of ﬂexural property
losses. The control samples showed thorough adhesion between the wood and
polymer, however, after exposure it appears that the water had penetrated the
samples and caused a loss of bonding between the wood and polymer.

Porosity testing on HDPE composites further conﬁrmed that there was a
loss of bonding between the wood ﬂour and polymer matrix in freeze-thaw
exposed composites. Porosity testing on both maple and pine ﬁlled composites
exhibited more and larger pores forming after freeze-thaw exposure. This
increased porosity is likely a result of decreased bonding, which in turn
decreases the ﬂexural properties. The ﬂexural properties decrease from
increased porosity because the polymer matrix can no longer transfer the applied
load to the wood component, which provides the structural strength.

Generally, the maple ﬁlled composite were more susceptible to freeze-
thaw cycling than the pine ﬁlled composites for both HDPE and PVC composites.
This could be a result of the wood ﬁbers behaving differently during processing.
Chemical variations between the wood species could result in different abilities to

form bonds with the polymers. The 50% maple-HDPE composites and the

119

100 phr maple-PVC composites both had lower densities than their pine
counterparts. Lower density materials are well known to have lower mechanical
(ﬂexural) properties, so this could be a reason for their poorer performance. The
lower density also likely leaves more points of entry from moisture to be
absorbed into the WPC.

The interaction between the different wood ﬂour species and polymer
matrices are not currently understood. HDPE composite with 50% maple wood
ﬂour lost 21% of their strength and 49% of their stiffness after 15 freeze-thaw
. cycles. PVC composites with 100phr of maple wood ﬂour lost 15% of their
strength and 30% of their stiffness after exposure to 12 full freeze-thaw cycles. A
plateau trend was observed for PVC where the majority of the property loss was
observed after 5 freeze-thaw cycles, so these values are likely representative of

most of the property loss that will occur in PVC composites.

Future Work

The effect of exposure to freeze-thaw cycling had not been previously
evaluated for WPCs before this study. The results have clearly shown that WPCs
are susceptible to freeze-thaw cycling and are likely susceptible to other
environments as well. Consequently, exposures to all types of environments
need to be assessed for wood-plastic composites.

Maple and pine wood ﬁbers exhibited different physical and mechanical
properties when used in the exact same formulation. Consequently, an analysis

of how wood species affects bonding within the composite would be beneﬁcial.

120

Perhaps there is a chemical difference between species that could enhance
bonding between the wood and plastic. Chemical modiﬁcation has been proven
to be effective at enhancing bonding, but maybe a wood species could be found
that would improve the properties without further modiﬁcation.

WPCs are also currently made with varying wood contents. This work had
a limited range of wood contents because of the need to gain a basic
understanding of freeze-thaw effects. The testing of freeze-thaw durability for
HDPE composites with wood contents other than 50% and PVC composites with
more than 100 phr (~46 wt.%) wood contents could be beneﬁcial. The lower
wood contents for PVC have been shown to have improved freeze-thaw
resistance and might be a solution to the detrimental effects found in HDPE
composites. Additionally, the PVC composites have shown improved freeze-thaw
resistance compared to HDPE composites, but this may be different at higher
wood contents and if additives were used in the HDPE composites. Coupling
agents are often used with HDPE, and future should include this additive.

Further optimization of WPC formulations is deﬁnitely needed to decisively
prove what can enhance freeze-thaw resistance. There are a variety of other
polymers, wood species, wood sizes, and countless additives that can be
incorporated into WPCs. Perhaps there is some phenomenon that occurs during
the processing of some of these other components that could enhance the
freeze-thaw resistance. This study only provided introductory information that
WPCs are susceptible to freeze-thaw effects and there is a lot of work yet to be

done on determining a formulation that is resistant to these freeze-thaw actions.

121

APPENDIX

This appendix contains additional scanning electron microscopy (SEM)
images that were taken of 100 phr (~46%) PVC and 50% HDPE composites ﬁlled
with both maple and pine wood ﬂour. SEM images were displayed in the text for
maple ﬁlled composites because they showed a greater loss in ﬂexural strength
and stiffness than pine ﬁlled composites. The greater ﬂexural property loss meant
that the maple composites’ SEM images more cleariy displayed the effects of
freeze-thaw cycling. These images include controls, full (water-freezethaw) cycle
exposures, water-only exposures (with no freezing), and freeze-thaw only
exposures (with no water submersion). These are abbreviated as control, FT
only, W only, and WFT. The SEM images will be grouped by composite and
exposure type and in order of increasing magniﬁcations (25x and 50x).

The HDPE composites’ SEM images are shown ﬁrst. These images
include the effects of freeze-thaw cycle portions, which were previously only
investigated for PVC composites. There are also additional control and 15 WFT
images, as well as SEM images of pine that were not used in the journal article.

Next are the PVC images. The ﬁrst two sets of SEM images are for PVC
composites made as part of the study to examine the effects of wood content.
These images were taken to show similarities between the two different groups
of PVC samples that were made for the work in chapter 3. The last set of images
show the effects of the freeze-thaw cycling and freeze-thaw cycle portions after 5

exposures on PVC-wood ﬂour composites, for both maple and pine composites.

122

Control maple-HDPE composites at 25x resolution.

 

123

Control maple-HDPE composite at 50x resolution.

:00 INT“

.1

 

15 FT only maple-HDPE composite at 25x.

 

124

15 FT only maple-HDPE composite at 25x resolution (cont'd).

 

125

15 FT only maple-HDPE composite at 50x resolution shared few similarities.

50C rim

 

126

15 W only maple-HDPE composites at 25x resolution.

 

127

15 W only maple-HDPE composites at 50x resolution.

 

128

15 WFT exposures at 25x for a maple-HDPE composite.

 

129

15 WFT resolutions at 50x for maple-HDPE composites (cont’d).

 

130

Control for HDPE-pine composites at 50x resolution.

 

131

HDPE-pine composites exposed to 15 WFT cycles at 25x.

 

132

15 WFT exposed HDPE-pine composite at 50x. Fibers left their imprint, but were
no longer bonded to the polymer matrix.

 

 

133

Control 100 phr maple PVC composite from varied wood content study at 25x.

‘ 500 pm ‘

 

134

Control 100 phr maple PVC composite from varied wood content study at 50x

(cont'd). The strand on the right of the top image shows a torn apart ﬁber.

500 pm“

 

135

Control 100 phr pine PVC composite from the varied wood content study at 25x.

l mrn

 

136

Control 100 phr pine PVC composite from the varied wood content study at 50x.

 

A 230x magniﬁcation image of the wood in a control 100 phr pine-PVC composite.

4: lOO pm

 

137

The rest of the SEM images are from the cycling effect study.

Control 100 phr maple-PVC composites at 50x.

 

GOO lll‘li

138

100 phr maple-PVC composites exposed to 5 FT only cycles at 25x resolution.

 

139

50x resolution images of 100 phr maple-PVC composites after 5 Fl' only cycles.

As with their HDPE counterparts there were few similarities in these images.

 

140

5 W only cycles of 100 phr maple-PVC composites at 25x.

 

141

5 W only cycles of 100 phr maple-PVC composites at 50x resolution.

 

".600 um

142

A 100 phr maple-PVC composite exposed to 5 WFI' cycles at 25x.

 

A 370x image of a degraded maple wood ﬁber in a 100 phr maple-PVC
corn osite exposed to 5 W

a

 

143

100 phr maple-PVC composites exposed to 5 WFT cycles at 50x resolution.

 

 

A control 100 phr pine-PVC composite at 25x resolution.

 

145

A 100 phr pine-PVC composite after 5 FT only cycles at 25x.

 

146

100 phr pine-PVC composites after 5 FT only cycles at 50x (cont'd).

500 pm

 

147

A 100 phr pine-PVC composite at 15x resolution after 5 W only cycles.

 

148

A 100 phr pine-PVC composite after 5 W only cycles at 50x resolution.

 

149

100 phr pine-PVC composites after 5 WFT cycles at 50x resolution.

503 lll‘ll

 

150

    
 

lliliiiiiiiiiiitrill