my. rd.“
fipymmm M
I. . 1:53.141

#1 . , is... . :r
‘ I. i .. . .
‘fiwhgmwrflfi p . ,

E1 ‘

i

3- $3352
3’35? ::

\
i

3
. EH
. . J“ “.13. a V
.1. 1 nu! . A
at; . e31... .) 3......li5s...f
313.}: .s 9. :5):I.i:.!,..
5‘. I. 04.5! (HUI. ~
, , any»...
F...

{(97}!
\.

{fizmua
s. .
a: s;

a.

. is
3... ...
.II.AJL.5§3§M
t; it .2. ‘0.
3):. , A. 2...:

.: fitufi:
.1 I! It r
3113.51“. .
1.1..{m
.r .r
z

 

 

.14?" W3: . av

was

This is to certify that the

thesis entitled

HIGH DENSITY POLYETHYLENE/WOOD FIBER COMPOSITES :
MEASURING THE EFFECT OF PROCESSING PARAMETERS ON
THEIR PHYSICAL AND MECHANICAL PROPERTIES

presented by

Kajiporn Uerkanarak ‘

has been accepted towards fulfillment
of the requirements for

Master degree in Packaging

Major professor .

Date February 13, 2002

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

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE I DATE DUE

 

OéPfliaQWOUI .,; 03 007;

 

DEC I 2 2005. 1 2 U

 

1:4 in i"

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIFIC/DateDue.p65-p.15

HIGH DENSITY POLYETHYLENE/WOOD FIBER COMPOSITES :
MEASURING THE EFFECT OF PROCESSING PARAMETERS ON
THEIR PHYSICAL AND MECHANICAL PROPERTIES

By

Kajiporn Uerkanarak

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
School of Packaging

2001

ABSTRACT
HIGH DENSITY POLYETHYLENE/WOOD FIBER COMPOSITES :
MEASURING THE EFFECT OF PROCESSING PARAMETERS ON

THEIR PHYSICAL AND MECHANICAL PROPERTIES

By

Kajipom Uerkanarak

High-density polyethylene (HDPE) is number one in the total amount used for
plastic containers and packages. Recycling of HDPE is preferred but it cannot provide the
same properties as pure HDPE does. Lately, the addition of natural fiber reinforcement to
improve overall performance has been rising. This study investigated the effects of
processing parameters (screw speed and position of fiber introduction) on the mechanical
properties of HDPE/wood fiber composites. Tensile properties, impact strength, and
water absorption were evaluated.

It was found that fiber-loading at port 11 provided higher tensile properties than at
port 1. Higher impact strength was obtained when wood fiber was loaded via port I. In the
case of screw speed, the highest tensile properties were obtained when a moderate screw
speed was used (100 RPM). The highest impact strength and water absorption were

obtained when 80- RPM and 120-RPM were used, respectively.

To

My Dearest Parents
Saovanee & Kajonsak Uerkanarak

The Biggest Sister and Brother
Sutinee & T ham’t Uerkanarak

iii

ACKNOWLEDGEMENTS

I would like to express my gratitude to my major professor, Susan Selke, PhD.
(School of Packaging, Michigan State University), and my committee members, Hugh E.
Lockhart, PhD. (School of Packaging, Michigan State University) and Indrek Wichman,
PhD. (Department of Mechanical Engineering, Michigan State University) for their
guidance and academic support.

I would like to thank Dr. Mike Rich from the Composite Materials and Structures
Center for his instructions and frequent help in the use of the extruder and I also wish to
thank Kelby Thayer for his instruction and assistance in use of Instron, and Krittika
Tanprasert for her instruction in use of several machines.

Special thanks go to Chung, Hsin—Yen (Julia) for her assistance in statistical
analysis and statistics software. My appreciation also goes to all my friends who always
support me.

Last, but not least, I would especially like to thank my family for their all

incredible patience and support.

iv

TABLE OF CONTENTS

List of Tables
List of Figures
Chapter 1 Introduction

Chapter 2 Literature Review

2.1 Background in Composite Materials
2.1.1 Matrix
2.1.2 Reinforcement
2.1.3 Interface

2.2 Natural Fiber Reinforced Therrnoplastics

2.3 Compounding

2.4 Prediction of Properties

2.5 Prior Research
Chapter 3 Materials

Chapter 4 Methods

Chapter 5 Results and Discussion

5.1 Tensile Strength

5.2 Yield Strength

5.3 Modulus of Elasticity

5.4 Percent Elongation

5.5 Izod Impact Strength

5.6 Water Absorption

5.7 Discussion
5.7.1 Tensile properties
5.7.2 Impact strength
5.7.3 Water Absorption

vii

viii

10
13
14
16
17
22

28

30

38
40
42
44
46
48

50
51
53

Chapter 6 Conclusions and Recommendations

5. 1

5 .2
Appendix A
Appendix B
References

Conclusions

Recommendations

vi

54
55

56

65

78

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

LIST OF TABLES

Plastic Containers and Packaging by Resin Type
Determination of the Feed Rate of Fiber at Each Screw Speed
of the Extruder

Sample Treatments

Tensile Strength

Yield Strength

Modulus of Elasticity

Percent Elongation

Izod Impact Strength

Percent Increase in Weight due to the Water Absorption
Tensile Strength Data

Yield Strength Data

Elongation Data

Modulus of Elasticity Data

Izod Impact Strength Data

Water Absorption Data

vii

31

32

38

4O

42

44

46

48

57

57

58

58

59

60

Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5
Figure 6.

Figure 7.

LIST OF FIGURES

Port Identification Diagram
Tensile Strength

Yield Strength

Modulus of Elasticity
Percent Elongation

Izod Impact Strength

Water Absorption

viii

33

39

41

43

45

47

49

Chapter 1

INTRODUCTION

In 1998, approximately 220 million tons of municipal solid waste (MSW) was
generated in the United State — an increase of 4 million tons from 1997. Paper and
paperboard products made up the largest component of MSW generated (3 8%) and yard
waste comprised the second largest material component (13%). Glass, metals, plastics,
wood, and food wastes each constituted between 5% and 10% of the total MSW
generated. Rubber, leather, and textiles combined made up about 7% of MSW, while
other miscellaneous wastes made up about 2% of the MSW generated in 1998 [1].

The US. Environmental Protection Agency’s (EPA) integrated waste
management hierarchy includes the following three main techniques:

Source Reduction, also known as “waste prevention”, is the practice of designing,
manufacturing, purchasing, or using materials (i.e. products and packaging) in ways that
reduce the amount or toxicity of trash created. Reusing items is another way to stop waste
at the source because it delays or avoids that item’s entry in the waste collection and
disposal system [2].

Source reduction, including reuse, can help reduce waste disposal and handling
costs, because it avoids the costs of recycling, municipal composting, landfilling, and
combustion. It also conserves resources and reduces pollution, including greenhouse

gases that contribute to global warming.

Disposal includes waste combustion (incineration) and landfilling [3].

Combustion or incineration is used to reduce waste by a controlled burning
process. In addition to minimizing volume, combustors can convert water into steam to
fuel heating systems or generate electricity.

In 1997, about 55% of the MSW generated was disposed in landfills. The number
of municipal solid waste landfills decreased substantially over the last ten years from
about 8,000 in 1988 to 2,314 in 1998 while the average landfill size increased.

Recycling is a series of activities that includes collecting recyclable materials that would
be considered waste, sorting and processing recyclables into raw materials such as fibers,
and manufacturing raw materials into new products [2].

There are three main recycling processes:

Step 1. Collecting and Processing

Collecting recyclables varies from community to community, but there are four
primary methods: curbside, drop-off centers, buy-back centers, and deposit/refund
programs. Then, recyclables are sent to a materials recovery facility to be sorted and
prepared into marketable commodities for manufacturing.

Step 2. Manufacturing

Once cleaned and separated, the recyclables are ready to undergo the second part
of the recycling loop, manufacturing. Recycled materials are used in common household
items such as aluminum, plastic, and glass soft drink containers; and plastic laundry

detergent bottles.

Step 3. Purchasing Recycled Products

Purchasing recycled products completes the recycling loop. By “buying recycled”,
governments, businesses and individual consumers each play an important role in making
the recycling process a success.

During the past few decades, recycling has become popular and successful.
Analysts project that Americans will be recycling at least 83 million tons, or 35% of all
municipal waste, by 2005. In 1998, 27% of MSW was recovered and recycled [1].

While recycling has generally grown, recycling of specific materials has
dramatically increased: 42% of all paper, 35.5% of all plastic soft drink bottles, 59.5% of
all of aluminum beer and soft drink cans, 61% of all steel packaging, 92% of all
automobiles and 64.3% of all major appliances were recycled in 1998 [1].

For the products in MSW, containers and packaging comprised the largest portion
of products generated (33%). Nondurable goods were the second largest portion (27%).
The third category is durable goods, which comprised 16% of total MSW generation [1].

Plastics are a rapidly growing segment of MSW. Plastics are found in durable and
nondurable goods and in containers and packaging, with the latter being the largest
category of plastics in MSW [3]. A

In durable goods, plastics are found in appliances, furniture, casings of lead-acid
batteries, and other products. A wide range of resin types is found in durable goods.

Plastics are found in such nondurable products as disposable diapers, trash bags,
cups, eating utensils, medical devices, etc. The plastic foodservice items are generally
made of clear or foamed polystyrene, while trash bags are made of high-density

polyethylene or low-density polyethylene.

Plastic resins are also used in a variety of container and packaging products such
as polyethylene terephthalate (PET) soft drink bottles and high density polyethylene
(HDPE) bottles for milk and water. Table 1 shows the six types of resins that are most

often used in container and packaging products.

 

 

 

 

 

 

 

Table] Plastic containers and packaging by resin type (%) [3]
PET 14.44
HDPE 37.90
PVC 4.46
LDPE/LLDPE 30.15
PP 10.19
PS 2.23
Other resins 0.63

 

 

 

 

High-density polyethylene (HDPE) is number one in the total amount used.
HDPE is commonly used in packaging containers such as milk, juice and water
containers [4]. These rigid containers are easy to separate and collect, thus making them
become one of the top recycled materials. Recovery of high-density polyethylene milk
and water bottles was estimated at about 31.3% in 1997 [3].

A number of recycling techniques have been developed to obtain well-sorted
resins that can substitute for or blend with virgin resins in many applications [5]. There
have been investigations of utilizing the recycled resins with reinforcements in the form
of composite materials. The addition of up to 40% by weight of fiber reinforcement is

common [6].

The step of compounding is also important to be considered. Compounding is a
process of combining a number of different components into one. These different
components combine to form a new material with properties of its own which are not
necessarily those of its constituents. Therefore, the purpose of this study was to
investigate the effect of processing parameters that are most suitable for the composite
materials of resin and wood fiber, to evaluate the mechanical performance such as tensile
strength and impact strength, as well as the effect of water absorption, of composite
materials, and to compare the mechanical properties of composites when using different
processing parameters.

Natural organic reinforcements such as wood fiber are slowly penetrating the
fiber-reinforced therrnoplastics market, presently dominated by glass fiber and other
mineral reinforcements. This is the result of their low cost, low density, acceptable
specific strength properties, renewable nature, comparative ease of processability,
enhanced energy recovery, and biodegradability [7, 8, 9]. However, they have
disadvantages in that they cannot withstand high temperature because of their low
degradation temperature (200 °C), which restricts the range of plastic materials to be
combined with, to those having low melting temperatures. The second is their water
sorption, which weakens their adhesion to hydrophobic polymer matrices [9].

In this study, high density polyethylene functions as the matrix, and wood fiber
from aspen hardwood is used as the reinforcement. HDPE was selected because of its
high recycling rate. Wood fiber is a choice because it is plentiful, lightweight, non-toxic

and has great strength [10]. It is also low cost, about 40% of the cost of glass fiber [1 1].

The processing parameters were varied, based on the different ports that were used for
fiber loading and the different screw speeds of the extruder when loading the wood fiber
through each port. The variations show how each processing parameter affects the
mechanical performance of the composite materials and which parameter was the most
suitable for the composites. The combination of resin and wood fiber was maintained at
60% resin and 40% wood. To control for variation in the quality of recycled resins, virgin
resin was used [12]. Product performance was evaluated by the mechanical properties of

tensile strength and impact strength and the physical property of water absorption.

Chapter 2

LITERATURE REVIEW

2.1 Background in Composite Materials

Demands of materials imposed by today’s advanced technologies have become so
diverse and severe that they often cannot be met by simple single-component materials
acting alone. Therefore, it is necessary to combine several materials into a composite in
which each constituent not only contributes its share, but also combines the action of the
individual properties and provides new performance unattainable by the individaul
constituents [13].

Such composite material systems result in better overall performance and offer the
great advantage of a flexible design; that is one can, in principle, tailor-make the material
as per specifications of an optimum design [14].

When comparing conventional monolithic materials such as aluminum and steel
with composite materials, it is found that composites provide the highest strength and
stiffness as well as fatigue resistance, and they have the lowest thermal expansion. One
important advantage of composites is their light weight compared with the heavy weight
of steel.

The word composite has evolved over a long period. It may be considered as a
system or process of combining two or more reinforcing materials in a matrix binder. On
the other hand, it can be considered to be a new material having characteristics derived

from its processing and from its microstructure [15].

Composites can he basically divided into two forms: (1) composite materials and
(2) composite structures. Composite materials are composed of a reinforcing material,
surrounded by a continuous matrix. They must be capable of arbitrary variation while
composite structures are not capable of arbitrary variation.

Most composite materials have been produced to improve mechanical properties
such as strength, stiffness, toughness, and high-temperature performance. Thus, it is usual
to study the composites that have a common strengthening mechanism. The strengthening
mechanism strongly depends on the geometry of the reinforcement. Therefore, it is quite
convenient to classify composite materials on the basis of the geometry of a reinforcing
material [16].

A common classification of composite materials consists of two types: (1)
particle-reinforced composites and (2) fiber-reinforced composites. The distinguishing
characteristic of a particle is that it is naturally nonfibrous. It may be spherical, cubic,
tetragonal, a platelet, or of other regular or irregular shape, but it is approximately
equiaxial. Particle-reinforced composites are sometimes referred to as particulate
composites.

A fiber is characterized by its length being much greater than its cross-sectional
dimensions. Fibers are small in diameter and will bend easily when pushed axially.
Although they may have outstanding tensile strength, they must be supported to keep
individual fibers from bending and buckling [15]. Fiber-reinforced composites are,

understandably, called fibrous composites [16].

Three basic components in a fiber-reinforced composite are polymer matrices,

reinforcements and interfaces [17].

2.1.1 Matrix

The matrix is the material that gives body, and grips or holds the reinforcements
of the composite together. It usually has lower strength than the reinforcement. The
matrix must be capable of being forced around the reinforcement during some stage in the
composite process. The main functions of the matrix are (i) to protect the fiber from
exposure to the environment as well as against fiber abrasion, (ii) to transfer and
distribute stress loads onto the fiber, and (iii) to separate and keep fibers in the desired
location and orientation [18, 19].

Polymers are generally used for matrices. They are found occurring in nature as
amber, pitch, and resin; and a variety of synthetic polymers are available. Polymers are
selected because they are easily processed and offer good mechanical and dielectric
properties. They can wet the reinforcements as well, resulting in good adhesion. Although
polymers have lower softening points than metals, they are low-density materials. It is
because of the relatively low processing temperatures and production techniques that
many organic reinforcements may be used [15].

All polymers are either thermoplastic or therrnosetting. Probably one of the
earliest distinctions between polymers was based on their reaction to heating and cooling.
Thermoplastics are polymers that can be made to flow when heated and become solid
when cooled. These materials may be softened repeatedly by heat and shaped into useful

products. Most thermoplastic materials, including scrap or damaged pieces, may be

recycled. Continued heating above their melting points will cause them to degrade.
Acrylics, cellulosics, polyarnide, polystyrene, polyethylene, fluoroplastics, polyvinyls,
polycarbonate, and polysulfone are examples of the thermoplastic group. Thermoplastic
materials come in a variety of available forms and are generally fully polymerized.

Thermoplastic polymers in composites have high impact strength and higher
resistance to failure, which provide a better withstanding of matrix microcracking in
composite laminates. However, compared to thermoset matrices, thermoplastic polymers
have developed slowly because of their high melt or solution viscosity, causing difficulty
in incorporation of continuous fibers into the matrix [17].

Thermosetting materials cannot be reshaped or reformed once the material is set
into a final structural framework. The heating and forming process causes them to
undergo a curing reaction. Thermosets will char, burn, or degrade by continued heating,
but they do not remelt. Thermosetting materials are particularly useful in producing many
composite materials because they are available in a variety of forms. Thermosetting
resins in a partially polymerized liquid state may facilitate penetration and wetting of the
other constituents. Members of the thermosetting group include aminos, casein, epoxies,

phenolics, polyesters, silicones, and polyurethanes.

2.1.2 Reinforcement

The reinforcement of therrnoplastics and thermosets by ceramic, metallic or
polymeric fibers is very important. It is easily achieved and the composite formation is
often cheaper than the polymer matrix alone. These reinforced grades marry the strength

and stiffness of such fibers with the good shock resistance of the thermoplastic matrix.

10

The fibers alone are usually very brittle and their strength and stiffness cannot be
fully utilized. The matrix protects these fibers and transfers the load to them. This gives a
material that combines the good properties of the fiber and the matrix, producing an
improvement in the strength, stiffness and creep resistance over those properties for the
matrix alone. These composite materials offer good competition to metals in many
applications in the motor and aerospace industries and in domestic appliances.

The strengthening mechanism depends on the geometry of the reinforcing filler,
which may be one of two types, particulate or fibrous. A particulate filler has no long
dimension, platelets being a noted exception. As a long dimension discourages the
growth of incipient cracks normal to the reinforcement in a brittle matrix, a particle does
not improve the fracture toughness of such a matrix. The exception to this rule is when a
rubberlike substance is dispersed in a brittle matrix. Under these conditions, considerable
toughening occurs, and this method is standard for improving the impact behaviour of
thermoplastics. Typical examples are high-impact polystyrene and ABS.

The particles will also share the load with the matrix, but to a lesser extent than a
fiber. A particulate reinforcer will, therefore, improve stiffness but will not generally
strengthen. Hard particles in a brittle matrix will cause localized stress concentrations in
the matrix, which will reduce the overall impact strength.

Particulate fillers are employed to improve high-temperature performance, reduce
friction, increase wear resistance, improve machinability and reduce shrinkage. In many
cases, particulate fillers are used simply to reduce the cost. Under these conditions, the
additive is a filler, whereas when a considerable improvement in properties of the

composite occurs, the additive is a reinforcement.

11

Fiber reinforcement improves the three main weaknesses of a thermoplastic
matrix: stiffness, strength and creep resistance. The measured strength of most materials
is much less than that predicted by theory because flaws in the form of cracks
perpendicular to the applied load are present in bulk materials. Fibers of non-polymeric
materials have much higher longitudinal strengths in this form because the large flaws are
not generally present in such small cross-sectional areas. However, these small cross-
sectional areas do not permit the use of fibers alone in engineering applications [20].

Composites reinforced with fibers are anisotropic, and properties depend on the
direction of the stress [15]. There are three types of fibrous reinforced composites,
namely particulate, continuous and discontinuous fibers [21].

Particulate composites are made of different sizes and shapes of particles
randomly dispersed in the matrix. Due to the random distribution of particles, these
composites can be considered as quasi-homogeneous on a scale larger than the particle
size. Particulate composites may contain either nonmetallic or metallic particles in a
nonmetallic or metallic matrix. Examples of this type are concrete and glass reinforced
with mica flakes.

Continuous fiber composites are reinforced by long continuous fibers and are the
most efficient for stiffness and strength. They also have greater strength and modulus in
the fiber axis direction and generally lack physical strength in the transverse direction.
The continuous fibers can be all parallel, can be oriented at right angles to each other, or
can be oriented along several directions.

Discontinuous composites consist of short fibers or whiskers in the reinforcing

phase. These short fibers can be either all oriented along one direction or randomly

12

oriented. In a discontinuous fiber composite, the stress along the fiber is not uniform. The
length (l) to diameter (d) ratio of the fiber, commonly called the aspect ratio (l/d),
determines the level of strength that the composite will reach. If the fiber is shorter than
the critical length, the composite will fail at a low strength level [19, 21]. These short
fiber reinforced therrnoplastics are very important in replacing metals in applications
requiring a combination of strength and lightness [20]. The principal fibers in commercial

use are various types of glass and carbon [17].

2.1.3 Interface

The behavior of a composite material is a result of the combined behavior of the

following [14]:
(1) Fiber or the reinforcing element
(2) Matrix
(3) The interface between the fiber and the matrix

To obtain desirable characteristics in the composite material, it is important that
the fibers are not weakened by flaws (surface or internal) and the applied load is
effectively transferred from the matrix to the fibers via the interface. Clearly, the interface
generally has an important bearing in this situation. Specifically, in the case of a fiber
reinforced composite material, the interface, commonly called the interfacial zone,
consists of near surface layers of fiber and matrix and any layer(s) of material existing
between these surfaces [14].

The reason why the interface is very important is that the internal surface area

occupied by the interface is quite extensive. It can easily go as high as 3000 cmz/cm3 in a

13

composite containing a reasonable fiber volume fraction. Thus, it becomes extremely
important to understand what exactly is going on in the interface region of any given
composite system under a given set of conditions.

Wettability of the fiber by the matrix and the type of bonding between the two
components constitute the primary considerations. Additionally, on should determine the
characteristics of the interface and how they are affected by temperature, diffusion,

residual stresses, and so on.

2.2 Natural Fiber Reinforced Thermoplastics

Originally, the purpose of fiber reinforcement was to improve the strength,
stiffness and creep resistance of polymer matrices. It is interesting to analyze how well
these intentions have been realized, to consider any additional, perhaps unexpected
advantages accruing due to reinforcement, and to study the disadvantages that are
involved in reinforcement.

Wood fibers are one of the potential reinforcements. They are naturally plentiful
organic fiber [22]. In 1997, there were about 43.9 percent by weight of wood, paper and
paperboard generated in MSW [3]. Wood fibers also offer strength that is close to that of
the traditional reinforcing materials. The strength and modulus of wood pulp fibers are
comparable with those of glass fibers at the same unit weight. The wood fiber composites
show the same or higher stiffness per weight than the steel, aluminum, glass fiber
composites and talc-contained polyolefins [23]. As a result, cost effectiveness is an
outstanding advantage. The other benefits are low abrasion to machinery, no hazardous

substance generation, low density and the possibility of surface modification [24].

14

However, there are some limitations. The low processing temperature, about 200°C, of
wood fibers makes them unavailable for some polymers that require high melting
temperatures. The water sorption of wood fibers causes a weak interface between the
matrix and wood fiber [9], and leads to biodegradation after repeated exposure [11].

Wood fiber composites can be made by extrusion, compression, or injection
molding to form a variety of products that can be used in packaging, paper products,
building materials, automobile parts, etc. [25].

Composites of thermoplastic polymers and wood fiber yield poor mechanical
properties due to the incompatibility between them. The thermoplastics, especially
polyolefins, are hydrophobic whereas wood fibers are hydrophilic. The difference causes
poor fiber distribution into the matrix and poor interfacial bonds between fiber and
matrix. As a result, the fiber strength cannot contribute to the composite strength as much
as it is supposed to. This phenomenon leads to easy failure of composites in mechanical
testing [9].

According to Chtourou et a1. [9], the interfacial bond between matrix and fiber
plays an important role in the improvement of mechanical properties. It can be
contributed to by five mechanisms: adsorption and wetting, interdiffusion, electrostatic
attraction, chemical links, and mechanical adhesion. In the case of composites consisting
of hydrocarbon polymer matrices and wood fibers, wetting and mechanical adhesion may

be the main influences.

15

2.3 Compounding

Compounding is a process of combining a number of different components into
one. These different components combine to form a new material with properties of its
own which are not necessarily those of its constituents [26].

The extruder is an versatile machine that can be used to form composite materials.
The extrusion process melts and mixes powdered or granular polymers into a continuous
melt. The melt stream is then sent to the die, mold, or accumulator. Screw technology
determines the output, milling rate, die pressure, and type of material that is to be
plasticated. The screw rotates in the extruder barrel. The length (L) and barrel diameter
(D) are expressed as an L/D ratio.

For many operations, multiple-screw extruders are used, for example, in the
compounding and extrusion of many reinforced thermoplastic and thermosetting
composite materials. With twin-screw machines, it is possible to introduce continuous
strands of reinforcement at the hopper. The screw will break the fibers into short lengths.
Adding fibers to a premelted polymer will reduce breakage and extruder wear, while
increasing dispersion and fiber wet out.

Co-rotating, twin-screw extruders are very commonly used in the plastic industry
for various compounds. They represent the standard for engineering thermoplastics and
polymer alloy compounding. They can also be used in polymers ranging from high
volume polyolefins to high performance liquid crystal polymers to specialty small-lot
operations [26].

The primary advantages of these extruders are their modularity that makes them

very flexible and adaptable to specific process requirements, coupled with their excellent

16

distributive and dispersive mixing. They also have short and narrow residence times,
which can be as short as 5 to 10 seconds for high speed, high torque applications. Such
features allow rapid purging and turnaround times.

The principal disadvantages are that they generate high peak shear rates and are
inefficient as melt pumps. The higher shear rates can be a significant problem at higher
screw speeds.

For the instrumentation control, the most important process parameters are melt
pressure and temperature. They are the best indicators of how well or how poorly an
extruder functions. Other important process parameters are: screw speed, motor load,
barrel temperatures, die temperatures, power draw of the various heaters, cooling rate of

the various cooling units, and vacuum level in vented extrusion [27].

2.4 Prediction of Properties

There are a number of variables that influence the properties of fibrous reinforced
composite materials and structures: (1) interface bond between matrix and fiber, (2)
properties of the fiber, (3) size and shape of the fiber, (4) loading of fiber, (5) processing
technique, and (6) alignment or distribution of the fiber.

The prediction of mechanical properties of aligned, long fibers is difficult to
achieve theoretically and further complications exist in dealing with short fiber
reinforcement. This is due to the possibility of a spectrum of fiber lengths and
orientations caused by processing into the final part.

The prediction of tensile modulus and tensile strength for long fiber reinforcement

will be shown in the following equation [20]:

17

Ec : Em¢m + El¢f (1)

6c : Gm‘l’m + Gl¢f (2)
Where; E = tensile modulus

o = tensile strength

(1) = volume fractions
and subscripts c = composite

f = fiber

m = matrix

Equation (1) represents a simple rule of mixtures, which gives values of E, for
long fiber composites close to values from more rigorous theoretical models and to actual
experimental values. If the values of the Poisson’s ratio, 7, of the fibers and the matrix are
not identical, equation (1) is about 1% in error. For most thermoplastic matrices y is
approximately 0.4.

In the case of continuous fiber reinforcement all the fibers are working at
maximum efficiency, with the average strain in the fiber being equal to that in the matrix.
In aligned short fiber composites this is not so. This is because the fiber restricts the
deformation of the surrounding matrix because it is stiffer than the matrix material. The
load is transferred from the matrix to the fiber via the interfacial shear stresses. Thus, the
calculation of the variation of the shear and tensile stresses along a short elastic fiber in
an elastic matrix are included in the prediction of the tensile modulus of the composite.

The shear stress is greatest at the ends of the fiber and decays to zero somewhere

along it. The tensile stress is zero at each end of the fiber and reaches a maximum at the

18

center. If the fiber is just long enough, the maximum tensile stress reaches the tensile
stress in the matrix. The ratio (L/D)c that occurs under these conditions is called the
critical elastic aspect ratio. For values of L/D less than (L/D)c the tensile stress in the fiber
is always less than that in the matrix, and clearly under these conditions the transfer of
load from the matrix to the fiber is poor and the mechanical properties of the fiber are not
fully utilized.

If L/D > (L/D),, the tensile stress at the interface remains at a maximum over a
greater proportion of the fiber length. Here, the transfer of stress the matrix to the fiber is
very efficient but the average tensile stress in the fiber is always less than that in the
matrix because of the reduced tensile stress at the ends of the fiber. The efficiency of
stress transfer is, therefore, never 100%.

In order to accommodate this change in reinforcement efficiency with fiber
length, the term m, the length correction factor, is considered. Therefore, the tensile

modulus of the composite can be predicted by the following equations [20]:

Ec : nL Ef¢f+ Em¢m (3)

where;

[1 - tanh(,6L / 2)]

 

 

. = 4
m (L / 2) ( )
where L is the fiber length and
V 277G»:
(5)

= [EfAf ln(R /r)]

19

where; Gm = shear modulus of the matrix

r = the radius of the fiber

R = the mean separation of the fibers normal to their
length

Ar = the cross-sectional area of all the fibers in the
composite

It is very important to remember that Ec depends on L/D rather than on L alone.
For the prediction of tensile strength for short fiber reinforcement, the following
equation can be utilized with an additional factor of average tensile stress on the
composite:
0} = Gm¢m + 51¢! (6)
where of is the average fiber stress, which is given by

1

0'f = ‘L‘Of(x)dx (7)

If the tensile stress builds up from the fiber ends in a non-linear way, then the

tensile strength can be calculated by the following equation:

07 = O'foo[1-(1- fl)%] for L > Lc (8)

the tensile stress in a continuous fiber in the same

where; 0'f
matrix under the same loading conditions.
or» = the average stress in the discontinuous fiber within a
distance Lc/2 of either end

L = critical fiber length

C

20

When the fiber length is greater than the critical fiber length, it is assumed that the
fiber failure occurs when of = or». Substituting the equation (8) in equation (6) gives the

following equation:

a. = 041— (1— fl)%]¢m+ am¢m (9)

A comparison between equation (2) and equation (9) shows that discontinuous
fibers provide composites of less strength than continuous fibers. The value of Lc
represents the shortest fiber length that may be broken in a given matrix. Fibers below
this minimum length are not capable of receiving sufficient tensile stress to break them
and composite failure occurs because of failure at the fiber-matrix interface.

The tensile strength of a short fiber reinforced thermoplastic decreases as the
angle between the fiber axis and direction of loading increases. The tensile strength of the
composite in a transverse direction is often less than that of the matrix material owing to
the effect of the fibers.

Impact strength of a composite does not have a simple relationship with adhesion
between filler and polymer, so it is very difficult to predict the impact strength of short
fiber reinforced composites. When brittle fibers are added to brittle matrices, the impact
strength of the composite is often reduced. This is because the matrix is bound by the

fibers and cannot deform to absorb the impact load.

21

2.8 Previous Research

Several researchers have studied the properties of composites of polymer matrices
with reinforcements from natural resources like cellulose fibers, wood fibers, and paper
fibers. Some of those works are summarized as follows.

Yam et a1. [28] investigated the mechanical properties of wood fiber/recycled
HDPE composites in comparison to wood fiber/virgin HDPE. Aspen fiber, a hardwood,
and spruce fiber, a softwood, were used. Recycled HDPE was made from chopped post-
consumer milk bottles. The fiber and polymer were extruded through a co-rotating
interrneshing twin screw extruder at 150°C. The extrudates were compression molded at
150°C and 4.22 MPa for 10 minutes and cooled under pressure for 15 minutes. The
tensile strength and the elastic modulus of composites made from recycled HDPE and
wood fiber were about the same as those of composites made from virgin HDPE. There
was also no significant difference in mechanical properties between the hardwood
composites and the softwood composites.

Kalyankar [29] investigated the mechanical properties of composites made from
recycled HDPE obtained from milk bottles. Aspen fiber, a hardwood, was used. Recycled
HDPE was made from chopped post-consumer milk bottles. Melt mixing of wood fibers
with HDPE in a twin screw extruder was employed and gave a uniform blending.
Mechanical property evaluation showed that the tensile modulus increased with increase
in wood fiber content. The tensile strength, elongation at break and impact strength

decreased. The composite also showed fair stability of dimensions and tensile strength at

22

equilibrium water absorption. Addition of ethylene vinyl acetate (EVA) copolymer as a
bonding agent improved the impact strength of the composite, but not the tensile strength.

Gogoi [25] investigated the effects of fiber pre-treatment, screw configuration of a
twin-screw extruder, and compounding temperature on the mechanical properties of
composites. Granulated HDPE milk bottles were used as the polymer matrix, while aspen
wood fiber was used as the filler. The results showed that tensile strength decreased with
an increase in fiber content. The effect of fiber pre-treatment in terms of tensile and
flexural yield strength showed that acetylated and untreated aspen fibers were better than
heat treated. The mechanical pr0perties of the composite were sensitive to screw
configuration and temperature.

Neiman [3 0] studied the mechanical properties of recycled HDPE and wood fiber
composites. Five additives were used: low density polyethylene (LDPE), stearic acid,
chlorinated polyethylene, maleic anhydride modified polypropylene (MAPP) and
ionomer modified polyethylene. For tensile strength and modulus, only maleic anhydride
modified polypropylene showed potential for improving adhesion between the polymer
matrix and wood fibers. Ionomer modified polyethylene also displayed some positive
results, while the others were determined ineffective for enhancing properties.

Keal [31] conducted research on the effect of dual additive systems on the
mechanical properties of composites of wood fibers and recycled HDPE milk containers.
The additive systems were two of stearic acid, maleic anhydride modified polypropylene
and ionomer modified polyethylene; thus three combinations were used. The use of
additives did improve tensile properties and creep but decreased impact strength. Only

the stearic acid/ionomer additive system did not reduce impact strength. Compared to the

23

effect of single additives, none of the dual additive systems provided significantly better
improvement.

Chtourou et a1. [9] studied composites of recycled polyolefms and wood fibers.
The polyolefins were 95% PE and 5% PP. A mixture of 45% spruce, 45% fir and 10%
poplar produced by chemico-thermomechanical pulp (CTMP) was used as a
reinforcement. The composites were made by injection molding and compression
molding. The impact of fiber concentration, the effect of fiber surface modification by
acetic anhydride and phenol formaldehyde, and the effect of moisture exposure on the
composites were evaluated by tensile properties. The result was that the greater the non-
treated fiber percentage, the higher the Young’s modulus and the strength at yield. More
than 30% fiber could be incorporated into the composites and at 30% fiber content by
weight, Young’s modulus increased 150%. Improvement in tensile strength was observed
in the composites with 10% treated fiber. 10% treated fiber composites also displayed
lower water sorption and higher mechanical properties than 10% non-treated fiber
composites.

Shan Ren and David N.-S. Hon [32] evaluated the effects of components,
processing and additives on the mechanical properties of composites made of newspaper
fiber and polypropylene (PP). The test specimens were made by a mixing and molding
process. The increase in the fiber from O to 10% proportionally reduced the strength.
Then, the strength leveled off at 10 to 50% fiber content and started decreasing again
when the fiber content was greater than 50% as a result of poor matrix-reinforcement
adhesion. In contrast, the modulus of elasticity increased proportionally to the percent of

fiber concentration. The optimum elastic modulus was achieved at the range of 40 to 50%

24

fiber content. The different types of reinforcement, which were commingled newspaper
fiber, TMP and chemical wood pulp, were also tested and showed an improvement of
elastic modulus. The effect of processing temperature revealed that the higher the
temperature, the higher the strength. The optimum temperature was between 190 and 205
°C. The tensile strength also increased proportionally to the addition of additives but
there was no significant effect on modulus of elasticity. From the scanning electron
micrographs, pulled-out fibers at the broken surface after tensile testing were evident. The
presence of stretched fibers confirmed the poor interfacial bonding between matrix and
fiber.

Chotipatoomwan [33] tested the processing and mechanical properties of paper
fiber and high density polyethylene (HDPE) composites. Two kinds of paper fibers,
mixed and deinked paper fibers, were used as fillers in composites. The mechanical and
physical properties were studied by varying the fiber content and using different kinds of
fibers at the same processing conditions. The addition of paper fibers to HDPE caused a
decrease in tensile strength and impact strength, but an increase in tensile modulus. Water
absorption increased due to the addition of paper fibers. Mixed paper fiber composites
had higher strength than deinked paper fiber composites.

Roj anarungtawee [12] studied the effect of mixed resins in different proportions
on the mechanical properties of a plastic/wood fiber composite. Polypropylene (PP) and
high density polyethylene (HDPE) represented the mixed matrices. Aspen wood fiber
served as the reinforcement. The composition of PP and HDPE was varied from 0% to

100% by weight. The mixed resins and wood fiber were compounded in a constant ratio

25

of 60% matrix and 40% reinforcement by a twin screw extruder operating at the PP and
HDPE melting points, 180°C and 150°C, respectively. At 180°C, the maximum ultimate
tensile strength was achieved at 30% PP/70% HDPE. 10% PP/90% HDPE had the
highest modulus of elasticity. At 150°C, only two variations of 0% PP/100% HDPE and
30% PP/70% HDPE were processable. The comparison between two ratios revealed that
the ratio of 0% PP/100% HDPE produced higher tensile strength than the ratio of 30%
PP/70% HDPE did. The modulus did not differ significantly at the two compositions
investigated.

Ricciardi [34] studied the impact of additives on the physical and mechanical
properties of high density polyethylene and paper fiber composites. The additives used
were maleic anhydride modified HDPE (MAHDPE), low molecular weight
polypropylene (Proflow 1000), and low density polyethylene (LDPE) with a high melt
flow index. The fiber loading level remained at approximately 35%. The effects of
MAHDPE were studied at 3%, 6% and 10%. The effects of Proflow 1000 and LDPE
were studied at 5% and 10%, while keeping the MAHDPE at 6%. The additives had no
significant effect on modulus of elasticity, percent elongation, and Izod impact strength.
MAHDPE was found to improve yield strength (at 10%) and tensile strength (at 3%), and
appeared to decrease debonding in water (at 3%, 6% and 10%).

Recently, Thepwiwatj it [3 5] investigated the mechanical properties of composites
of wood fiber and recycled HDPE bottles from household use. Aspen wood fiber was
used as the reinforcement. The composite materials were made by combining the resins

and wood fibers in 5 different ratios; 100% HDPE/0% fiber, 90% HDPE/10% fiber, 80%

26

HDPE/20% fiber, 70% HDPE/30% fiber, and 60% HDPE/40% fiber, and 60% virgin
HDPE resin with 40% fiber. The compounding speed of the extruder was set at 120 RPM
and the temperature of all 6 processing zones was set at 150°C. It was found that an
increase of wood fiber content did not improve the tensile properties of the composites.
With an increase in the fiber concentration, the tensile strength, yield strength and %
elongation decreased, while the modulus and impact strength slightly increased. The fiber
fraction in the composites had a very large effect on water absorption. The more the fiber

content, the higher was the gain in weight.

27

Chapter 3

MATERIALS

High density polyethylene (HDPE) in pellet form was provided by Paxon Polymer
Company under the trade name Paxon®AD 60-007. It was a virgin HDPE homopolymer
having a medium molecular weight distribution. The virgin thermoplastic was able to be
a substitute for the recycled HDPE because the properties of virgin HDPE composites
were the same as those of recycled HDPE composites, according to Yam et al.[28].

HDPE is a milky-white nonpolar, linear thermoplastic. Its density ranges from
0.940 to 0.965 g/cm’. It is one of the most versatile polymers, and is the second most
commonly used plastic in the packaging industry. The molecular chains of HDPE
homopolymer are long and straight with little branching. This close packing produces
HDPE with a crystallinity of 65-90% and contributes to HDPE’s good moisture-barrier
properties, chemical resistance and to its non-transparency. Glass transition and melting
temperatures of HDPE are about -80i10 °C and 138 °C, respectively [12].

Aspen wood fiber was used as the reinforcement. Hardwoods are categorized in
the subdivision angiosperrnae. Hardwood leaves are broad and change color in the fall
season in temperate areas. Aspen is a generally recognized name applied to bigtooth
aspen (Populus grandidentata) and to quaking aspen (P. tremuloids). Aspen wood is
usually straight grained with a fine, uniform texture. It is easily worked. Well-seasoned

aspen lumber does not impart odor or flavor foodstuffs. The wood of aspen is light

28

weight and soft. It is low in strength, moderately stiff, moderately low in resistance to
shock, and has moderately high shrinkage [11].

Aspen hardwood fibers in this experiment were in the form of thermomechanical
pulp (TMP). In this mechanical pulping process, wood chips are fed into a refiner at
about 120°C, which grinds and defibrillates the chips into fibers. There is only a
minimum amount of damage to the lignin or hemicellulose during this pulping process,
so the wood fibers retain nearly all of their lignin and natural waxes, which can help in
better dispersion of wood fiber into the non-polar hydrocarbon polymer matrix [23].

The fibers were conditioned for at least 40 hours at 23 i 2°C and 50 i 5% RH

prior to the testing.

29

Chapter 4

METHODS

ExtrudinLthe Composite MM

Before extruding the composite material, the feed rate to the extruder was
calibrated. It is important to know that the feed rates were impacted by the maximum
amount of fiber that could be pushed down into the port. It was also affected by the screw
speed of the extruder. In this experiment, there were 3 different screw speeds (80, 100
and 120 RPM), so the fiber feed rates were different for each screw speed. Using the

following equation, the percent fiber-loading level was set:

 

X . = 0.4
X + Y
Where: X = Feed rate of wood fiber (g/min)
Y = Feed rate of resin (g/min)
0.4 = The chosen percent fiber loading of the composite

The feed rate of the resin was found experimentally by cutting and weighing the
resin every minute. The feed rate of resin was then substituted into the above equation to

calculate the fiber feed rate.

30

Table 2 Determination of the feed rate of fiber at each screw speed of the extruder

 

 

 

 

 

Screw Speed Measured Resin Feed Average (g/min) Calculated Fiber
(RPM) Rate (g/min) Feed Rate (g/min)
15.31
80 16.40 15.91 10.61
16.02
15.89
100 16.05 15.93 10.62
15.84
16.13
120 15.97 16.29 10.86
16.78

 

 

 

 

A Baker Perkins Model ZSK-30, 30 mm, 26:1 co-rotating twin—screw extruder
(Werner & Pfleiderer Corporation, Ramsey, New Jersey) was employed to compound the
polymers and wood fiber. It consists of three parts, the feed zone, compression zone and
metering zone, functioning differently. The feed zone, attached below the feed hopper,
works as the pathway for the resin pellets to get into the barrel. The compression zone is
where some granules start melting. Then, all become liquid and ready to exit at the die at
a constant rate in the metering zone [6]. The temperature was set at 150°C along the three
zones of the extruder. The set temperature should be a little bit higher than the melting
temperature of HDPE, 130-135°C. Too high a temperature is not desirable because

charring of wood fibers increases. The wood fiber was loaded into the extruder through

31

 

three different ports at three different screw speeds of the extruder (80, 100, and 120
RPM). At the die exit, a continuous stream of well-mixed composite flowed consistently

out and was cut into six inches in length before it solidified at room temperature.

Nine treatments, as shown in Table 3, were performed.

 

 

 

 

 

 

 

 

 

 

 

T_ab_le_3 Sample Treatments
Treatments Screw Speed (RPM) Port Position
1 80 I
2 100 I
3 120 I
4 80 II
5 100 II
6 120 II
7 80 III
8 100 III
9 120 III

 

 

 

32

 

:2: 009

 

Exam—3Q noumomucoE tom

:2: 3.2

 

L

 

  

:1

 

 

 

 

aomqom ._

 

 

 

   

Q
“

 

 

Spoon —_

 

 

 

’
‘

 

55 m:

 

 

 

 

 

A Semi
L
x
] fig 2: ] 55 8s ] 55 m3
= m tom .. _. N tom -_ = fl tom __
I \

mtom “222.45% m SE @632 mg» 525

 

 

 

33

Compression Molding

The extruded material was later compressed into plates using a Carver Laboratory
Press, Model M (Fred S. Carver Inc., Menomonee Falls, Wisconsin). Three pieces of six
inch extrudate were needed for each molding process. Two different sizes of frames were
used for different mechanical tests. The first frame dimensions were 6 x 6 x 0.1 inch, for
tensile test samples. The second frame dimensions were 5 x 5 x 0.125 inch, for both
impact test and water absorption samples. For all samples, the machine direction of the

pieces was noted. The compression molding steps in detail are described in Appendix A.

Specimen Preparation

Once the plates were made, they were cut into suitable pieces for tensile, impact
and water absorption tests. When cutting, the machine direction was noted and kept
consistent for all of the test pieces. The molded sheets were cut into 6 x 0.75 x 0.1 inch
for tensile tests and 2.5 x 0.5 x 0.125 inch for both impact and water absorption tests
using a New Hermes Safety Saw. Then, dumbbell-shape specimens (type I) were made
for tensile test samples by using Tensilkut, Model 10-13 (Tensilkut Engineering Division
Sieburg Industries, Inc., Danbury, Connecticut). The specimens for the impact test were
then notched using a TMI notching cutter. The angle of the notch was 22.5° i 0.5° and
the depth of the notch was 0.1 inch. The specimens were conditioned at 2332 °C and 50¢

5 % RH for at least 40 hours before being tested for tensile strength and impact strength.

34

Tensile Strength Testifl

ASTM 638-99 [36], Standard Test Method for Tensile Properties of Plastics, was
followed. An Instron testing machine, Model SFM-20 (United Calibration
Corp., Huntington Beach, California) was used to perform the tensile strength testing
(Instron operation is in Appendix A). The parameters of the machine were set as follows:
1000 lb load cell, test speed 0.02 inch/minute, extension gage length 2 inches. All test
samples were measured by a digital vemier caliper Digimatic (Mitutoyo Corporation,
Japan).

At least five samples were tested. Tensile strength and modulus of elasticity were
automatically calculated and the curves between load (1b) and % extension were plotted
by a Graphtec XY plotter, type MP 3200. X axis or % extension was set at maximum 3%
and Y axis or load was set at 250 lb maximum load. Statistical analysis of the tensile
strength and elastic modulus was performed using the SPSS program for One Way
ANOVA and Least Significant Difference (LSD). The comparisons between each

composition were analyzed at the 95% confidence level.

Izod Impact Strength Testing

By following ASTM D 256-97 [37], Standard Test Method for Determining the
Izod Pendulum Impact Resistance of Plastics, Izod impact strength was determined. At
least ten samples were tested using a TMI 43-I IZOD impact tester with a 5-lb pendulum
(impact test steps are in Appendix A). They were held as a vertical cantilever beam and

broken with a single swing of the pendulum. When the sample failed, the type of failure

35

was classified following the ASTM standard. The statistical analysis and the comparison

between compositions were analyzed by the method described above.

Water Absorption Testing

ASTM D570-98 [3 8], Standard Test Method for Water Absorption of Plastics,
was followed. The specimens were conditioned by drying in an oven for 24 hours at 503:3
°C, cooled in a desiccator, and immediately weighed to the nearest 0.001 g. Twenty-four
hour Immersion and Long-Term Immersion procedures were employed (both procedures
are in Appendix A). The change in weight was monitored and used to calculate the %
water absorption.

Percent water absorbed was calculated after 24 hours and when equilibrium was
reached, by the formula:

% water absorption = wet weight — conditioned weight x 100
conditioned weight

36

Chapter 5

RESULTS AND DISCUSSION

The mechanical properties of the composite samples were evaluated by the test
methods described in the previous chapter. At least five specimens of each treatment were
tested for tensile properties and water absorption. Twelve specimens of each composition
were tested for Izod impact strength.

As mentioned before, nine treatments were performed. However, the last three
treatments (treatment 7, 8 and 9) did not work as well. This is because the HDPE resins
were still in granular form; not melted, when the wood fiber was loaded into the port.
That resulted in a non-uniform mixture coming out of the die. Thus, the extrudate could
not be molded into uniform plates for testing.

The results of the first six treatments are shown for the following mechanical
properties: tensile strength, yield strength, modulus of elasticity, percent elongation, Izod

impact strength and water absorption.

37

4.1 Tensile Strength

Table 4 presents the tensile strength results for the different six treatments. It was
found that the tensile strength of the composite was higher when fiber was loaded into the
extruder via port 11 than when via port I (Figure 1). It was also found that the 100 RPM
screw speed gave the highest tensile strength of the composite when compared with 80
and 120 RPM, for each fiber-loading port. The one-way analysis of variance method
(AN OVA -— see Appendix B for statistical analysis results) confirmed that there was a
significant difference in the tensile strength between the treatments. However, the result
by the least significant difference method (LSD) at the 95% confidence level showed no
significant difference between treatments 1, 2 and 3, no significant difference between
treatments 3 and 6, and no significant difference between treatments 4 and 5. This might
be affected by the wide range of the tensile strength values of treatments 2, 3 and 6 (see

data in Table 10 in Appendix A).

 

 

 

 

 

 

 

 

 

 

 

 

Lam—e4 Tensile Strength (psi)
Treatment Fiber Loading Screw Speed Average SD
1 80 RPM 2504.333 332.92
2 Port 1 100 RPM 2774.21ab 286.75
3 120 RPM 2563.85ac 278.55
4 80 RPM 3387.78d 257.98
5 Port II 100 RPM 3434.36d 208.31
6 120 RPM 2888.76c 229.79

 

 

 

 

 

Note: Values with the same superscript letter are not significantly different.

38

 

 

FimZ

Tensile Strength

 

 

 

4000

._ 3500

m

5"

.g 3000

H

2.”

g 2500

m

g 2000

m

5 1500

[.4

3’»

2 1000

3

< 500

0 f q
1 2 3 4 5 6
Treatments

Treatment 1 = Fiber was loaded via port I and at 80 rpm screw speed
Treatment 2 = Fiber was loaded via port I and at 100 rpm screw speed
Treatment 3 = Fiber was loaded via port I and at 120 rpm screw speed
Treatment 4 = Fiber was loaded via port 11 and at 80 rpm screw speed
Treatment 5 = Fiber was loaded via port II and at 100 rpm screw speed
Treatment 6 = Fiber was loaded via port II and at 120 rpm screw speed

39

 

4.2 Yield Strength

The yield strength results of the test samples are shown in Table 5 and Figure 2.
Similar to the results for tensile strength, the yield strength of the composite was higher
when fiber was loaded into the extruder via port 11 than when loaded via port I, and a
screw speed of 100 RPM gave the highest tensile strength of the composite when
compared with 80 and 120 RPM for each fiber-loading port.

The AN OVA result, again, showed there was a significant difference in the yield
strength between treatments. The LSD results indicated no significant difference between
treatments 1, 2, and 3, between treatments 2 and 6, and between treatments 4 and 5.
However, the group of treatments 1, 2 and 3 had a significant difference in yield strength
from the group of treatments 4 and 5, and treatments 1 and 3 from treatment 6.

The data for yield strength testing and the statistical analysis results can be seen in

Appendices A and B respectively.

 

 

 

 

 

 

 

 

 

 

 

 

M5 Yield Strength (psi)
Treatment Fiber Loading Screw Speed Average SD
1 80 RPM 2441.15a 351.51
2 Port I 100 RPM 2545.16ac 369.08
3 120 RPM 2449.49‘ 307.31
4 80 RPM 3282.55b 298.75
5 Port 11 100 RPM 3417.73b 223.25
6 120 RPM 2871.33‘3 222.68

 

 

 

 

 

Note: Values with the same superscript letter are not significantly different.

40

 

 

Figm 3

Yield Strength

 

 

 

4000

A 3 5 00

'2’.

V 3000

a .

g 2500 - '

m

'3 2000

.2

>‘ 1 500

0

en

E 1000

0 y -:

> 2. ‘

< 500 - y , a

0 . .
1 2 3 4 5 6
Treatments

Treatment 1 = Fiber was loaded via port I and at 80 rpm screw speed
Treatment 2 = Fiber was loaded via port I and at 100 rpm screw speed
Treatment 3 = Fiber was loaded via port I and at 120 rpm screw speed
Treatment 4 = Fiber was loaded via port II and at 80 rpm screw speed
Treatment 5 = Fiber was loaded via port 11 and at 100 rpm screw speed
Treatment 6 = Fiber was loaded via port 11 and at 120 rpm screw speed

41

 

4.3 Modulus of Elasticity

Table 6 and Figure 3 demonstrate the results for modulus of elasticity for the
wood fiber reinforced HDPE composite system.

Unlike tensile strength and yield strength, the ANOVA results showed there was
no significant difference between treatments. Therefore, statistical analysis at the 95%

confidence level by the LSD method was not performed.

Table 6 Modulus of Elasticity (psi)

 

 

 

 

 

 

 

 

 

 

 

 

Treatment Fiber Loading Screw Speed Average SD
1 80 RPM 250,070 19092.13
2 Port I 100 RPM 246,633 13543.50
3 120 RPM 228,685 31164.17
4 80 RPM 268,094 26139.53
5 Port II 100 RPM 287,587 72818.70
6 120 RPM 225,600 35694.02

 

 

 

 

 

42

 

Fi ure4

Modulus of Elasticity

 

 

 

 

A 350,000

I; >

e .

3‘ 300,000 .

3 250,000 .

E

0 200,000

in

E.

:1

lg 150,000

2

“5 100,000

0

en

E 50,000

<1 i .

0 7 : ., ‘
1 2 3 4 5 6
Treatments

Treatment 1 = Fiber was loaded via port I and at 80 rpm screw speed
Treatment 2 = Fiber was loaded via port I and at 100 rpm screw speed
Treatment 3 = Fiber was loaded via port I and at 120 rpm screw speed
Treatment 4 = Fiber was loaded via port 11 and at 80 rpm screw speed
Treatment 5 = Fiber was loaded via port 11 and at 100 rpm screw speed
Treatment 6 = Fiber was loaded via port 11 and at 120 rpm screw speed

43

 

4.4 Percent Elongation

The test results for percent elongation of wood fiber reinforced recycled HDPE

composite systems are presented in Table 7 and Figure 4.

The ANOVA results showed there was a significant difference in the percent

elongation between treatments. The highest percent elongation of the composite was

produced in the process at a screw speed of 80 RPM and fiber loading at port 11.

 

 

 

 

 

 

 

 

 

 

 

 

19M Percent Elongation (%)
Treatment Fiber Loading Screw Speed Average SD
1 80 RPM 1.48“b 1.15
2 Port 1 100 RPM 2.96bc 0.54
3 120 RPM 1.97“” 0.47
4 80 RPM 5.91T 0.60
5 Port 11 100 RPM 3.46c 1.40
6 120 RPM 2.975“ 1.73

 

 

 

 

 

Note: Values with the same superscript letter are not significantly different.
Statistical analysis at the 95% confidence level by the LSD method confirmed that

the percent elongation of the composites was affected by both the fiber loading location

and the screw speed.

44

 

F igge 5 Percent Elongation

 

 

 

 

7
f; 6
e
g 5
a
1‘3
:0 4
.2
E‘. 3
o
a)
e 2
3
< 1
o 7
1 2 3 4 5 6
Treatments
Treatment 1 = Fiber was loaded via port I and at 80 rpm screw speed
Treatment 2 = Fiber was loaded via port I and at 100 rpm screw speed
Treatment 3 = Fiber was loaded via port I and at 120 rpm screw speed
Treatment 4 = Fiber was loaded via port 11 and at 80 rpm screw speed
Treatment 5 = Fiber was loaded via port 11 and at 100 rpm screw speed
Treatment 6 = Fiber was loaded via port II and at 120 rpm screw speed

45

 

4.5 Izod Impact Strength

The results of Izod impact strength testing tabulated in Table 8 and Figure 5, show

that the location of fiber loading had an effect on the impact strength of the test

specimens. The AN OVA results indicated that there was a significant difference between

the treatments. Statistical analysis at the 95% confidence level by the LSD method

confirmed that there was a significant difference between the group of treatments 1, 2,

and 3 and the group of treatments 4, 5, and 6.

However, the different screw speeds at the same location of fiber loading had no

effect on the Izod impact strength of the composite (no significant difference was found

at the 95% confidence level by the LSD method).

 

 

 

 

 

 

 

Table—8 Izod impact strength
Fiber Loading Screw Speed Type of Failure Izod Impact Strength (fi-lb/in)
Average SD
80 RPM Partial and Hinge 1.367a 0.380
Port I 100 RPM Partial, Hinge and 1.083b 0.328
Complete
120 RPM Partial, Hinge and 1.0343 0.129
Complete
80 RPM Partial, Hinge and 0.683“ 0.069
Complete
Port 11 100 RPM Partial, Hinge and 0.699c 0.100
Complete
120 RPM Partial, Hinge and 0577" 0.081
Complete

 

 

 

 

 

 

 

Note: Values with the same superscript letter are not significantly different

46

 

Figm 6

Izod Impact Strength

 

 

Average Izod Impact Strength (ft-lb/in)

 

1 2 3 4 5 6
Treatments

 

Treatment 1

Treatment 2

Treatment 3

Treatment 4

Treatment 5

Treatment 6

= Fiber was loaded via port I and at 80 rpm screw speed
= Fiber was loaded via port I and at 100 rpm screw speed
= Fiber was loaded via port I and at 120 rpm screw speed
= Fiber was loaded via port H and at 80 rpm screw speed
= Fiber was loaded via port II and at 100 rpm screw speed

= Fiber was loaded via port II and at 120 rpm screw speed

47

 

4.6 Water Absorption

Table 9 shows the percent water absorption in terms of % increase in weight for
each treatment. The effect of fiber loading location and screw speed is illustrated in
Figure 6.

As the screw speed increased, the gain in weight due to water absorption
increased. Results from ANOVA (see Appendix B) showed that there was a significant
difference in these six treatments,which means both the fiber loading location and the

screw speed had an effect on water absorption.

 

 

 

 

 

 

 

 

 

 

 

 

m2 Percent increase in weight due to water absorption
Treatment Fiber Loading Screw Speed Average SD
1 80 RPM 0.02343” 0.0065
2 Port I 100 RPM 0.0562c 0.0263
3 120 RPM 0.0823d 0.0132
4 80 RPM 0.0081be 0.0017
5 Port 11 100 RPM 0.0141“c 0.0020
6 120 RPM 0.0138’“ 0.0054

 

 

 

 

 

Note: Values with the same superscript letter are not significantly different.

The LSD method at the 95% confidence level confirmed that there was a

significant difference between treatments 1, 2, and 3, but no difference between

treatments 4, 5, and 6. (Details of the data for water absorption testing and the statistical

analysis are shown in Appendices A and B). That means the fiber loading locations have

more effect on water absorption test than does the screw speed.

48

 

 

Fi ure7

Water Absorption

 

% Change in Weight

0.09 ,
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01

 

2 3 4 5 6
Treatments

 

Treatment 1

Treatment 2

Treatment 3

Treatment 4

Treatment 5

Treatment 6

Fiber was loaded via port I and at 80 rpm screw speed
Fiber was loaded via port I and at 100 rpm screw speed
Fiber was loaded via port I and at 120 rpm screw speed
Fiber was loaded via port II and at 80 rpm screw speed
Fiber was loaded via port H and at 100 rpm screw speed

Fiber was loaded via port II and at 120 rpm screw speed

49

 

4.7 Discussion
There are many factors that influence the properties of composites, such as fiber
type, length of fiber, aspect ratio (length to diameter ratio), fiber alignment, interface,

matrix resin morphology, processing procedure and environmental effects.

4.7.1 Tensile properties

Tensile testing results were used to determine the mechanical strength of the
composites. In this experiment, each test specimen’s tensile strength, yield strength,
modulus of elasticity, and percent elongation were determined. These data are useful for
qualitative characterization and development.

Basically, fiber reinforcement is used to improve the strength of polymer matrix
[3 9]. However, weak adhesion between the polymer and reinforcement can cause
difficulty in development of a composite property.

As the tensile strength and yield strength of the composite are influenced by the
fiber loading location, it indicates that the tensile strength of the composite was lower
when fiber was loaded into the extruder via port I than when loaded via port 11 because
the time for the polymer matrix mixing with the wood fiber was too short, resulting in
poor adhesion at the interface between HDPE and wood fiber or it may be due to poor
adhesion of the fibers in the matrix.

The wide range of tensile properties in the composites, which can be seen in the
high standard deviations, may result from poor fiber dispersion. The high viscosity of the
matrix during the composite fabrication and the polarity of the fiber, which tend to hold

the fibers together, may cause poor distribution of fibers. Instead of spreading out evenly

50

in the polymer, the fibers were more likely to crowd randomly in the matrix, leading to
lack of uniformity of the composite systems. In respect to mixing time, the shorter the
mixing time, the higher was the variation, resulting in lower mechanical strength of the
composite.

The modulus of elasticity or Young’s modulus is simply the slope of the initial
straight portion of the stress-strain curve. According to Bigg [40], the reinforcement
concentration can affect the modulus of the composite. The same mixing ratio of polymer
matrix (60%) and wood fiber (40%) was used in every treatment. It was found that there
was no significant difference in the modulus between treatments.

When the results were compared in the case of screw speed, it was found that the
highest mechanical strength (tensile strength and yield strength) was obtained from the
composite produced at 100 RPM screw speed. This might be because the 80-RPM screw
speed allows too long a mixing time between polymer and wood fiber, resulting in fiber
damage, and the 120-RPM screw speed allows too short a mixing time, resulting in

difficulty mixing the HDPE and wood fiber.

4.7.2 Impact strength

The impact strength of such composites depends on the nature of the composite
and the type of impact test [41]. According to Chotipatumwan [33], this test method is
used to determine the resistance to breakage by flexural shock of composites. The notch
on the samples was made to concentrate the stress, minimize the deformation, and direct

the fracture to the part of the specimen behind the notch.

51

For non-brittle polymer matrices, the addition of fiber is generally found to reduce
the impact strength of composite materials because it reduces the surface energy and then
decreases the volume fraction of matrix in the plastic zone.

However, the highest impact strength can be obtained in the case of the poorest
adhesion between the matrix and the wood fiber. This is because the maximum energy
can be dissipated by mechanical fiiction during the pull-out process and by debonding of
the fibers [41].

In this investigation, the highest impact strength was obtained when a fiber-
loading location of port I and 80-RPM screw speed were used. That was not expected.
The 120—RPM screw speed and the same fiber-loading port location should provide the
highest impact strength because the time for mixing was shorter than that occurring at 80
RPM and so provided the poorest adhesion. But the shear rate would be higher and that
could compensate by producing greater mixing force, even though less time.

Fibers also can reduce the impact strength by reducing the elongation to break and
thus may reduce the area under the stress-strain curve. Therefore, the higher percent
elongation of such composites contributes to the higher impact strength.

This does not explain the higher impact strength of the composites produced when
the fiber-loading location of port I was used, compared to the composites produced when
the fiber-loading location of port II was used. The higher percent elongation of
composites produced when the fiber-loading location of port 11 was used should provide
higher impact strength than the lower percent elongation of composites produced when

the fiber-loading location of port I was used.

52

4.7.3 Water absorption

This test was used to determine the relative rate of water absorption, and had
two main functions. First, it is a guide to the proportion of water absorbed by a material
and consequently, in those cases where the relationships between moisture and electrical
or mechanical properties, dimensions, or appearance have been determined, as a guide to
the effects of exposure to water or humid conditions on such properties. Secondly, it is
used as a control test on the uniformity of a product.

Results showed that a higher amount of water was absorbed when fiber loading
port I was used than when fiber loading port 11 was used. As the structure of the wood
fiber is hydrophilic, it does not adhere very well to the hydrophobic structure of the
polymer matrix, especially when the mixing time is too short. After processing of the
composites, the thermal shrinkage of the matrix results in a gap surrounding the fiber.
Then the composite materials can take up a large amount of water because the hydroxyl
groups in the fiber structure interact with the surrounding water. This causes swelling of
the fibers which can fill the gap between the fibers and the polymer matrix. Then,
mechanical properties might be decreased.

Composite materials, when the fiber was loaded via port I, had more gaps
between the polymer matrix and wood fiber than when loaded via port 11. This resulted in
higher water absorption in the composites when fiber was loaded via port I than when
loaded via port 11.

When comparing the effect of screw speeds on water absorption, it seems that

there was not much difference between treatments.

53

Chapter 6

CONCLUSIONS AND RECOMMENDATIONS

6.1 Conclusions

In the investigation of the fiber loading location, it was found that tensile strength,
yield strength, modulus of elasticity and percent of elongation were higher when fiber
was loaded into the extruder via port 11 than via port I. This is because of the longer
mixing time of the HDPE and wood fiber composites. But higher impact strength and
water absorption were obtained when fiber was loaded into the extruder via port I than via
port II because of the poorer adhesion between the matrix and the wood fiber.

In the case of the screw speed, the lOO-rpm screw speed provided the highest
tensile properties except for percent elongation. The highest impact strength was obtained
when the 80-rpm screw speed was used. The highest percent water absorption was found
when the 120-rpm screw speed was used.

Overall, most properties of composites were improved when the fiber loading

location was port 11 and the 100—rpm screw speed was used.

54

6.2 Recommendations for future research

As the incompatibility of the fibers and the matrix, including poor dispersion
of the fibers in the matrix, causes adhesion between HDPE and wood fiber to be
relatively poor, so it produces poor result for the mechanical and physical properties.

Therefore, the recommendation for further research is to improve the interface
adhesion by adding coupling agents or surface modification such as maleic anhydride
modified HDPE (MAHDPE), Proflow 1000, etc.

Moreover, the fiber might be treated to obtain a better distribution of fiber in the
matrix. This method may be a way of improving the mechanical properties of the

composites.

55

Appendix A

56

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M Tensile Strength Data (psi)
Sample Port I Port 11
80 RPM 100 RPM 120 RPM 80 RPM 100 RPM 120 RPM
1 2819.84 3092.98 2208.75 3449.22 3347.29 2572.29
2 2175.80 2392.15 2925.79 3319.24 3626.63 2935.93
3 2543.31 2565.81 2375.07 3228.10 3150.40 3198.42
4 2833.41 2932.23 2631.65 3801.59 3396.45 2946.46
5 2149.29 2887.90 2677.98 3140.77 3651.05 2790.71
Average 2504.33 2774.21 2563.85 3387.78 3434.36 2888.76
SD 332.92 286.75 278.55 257.98 208.31 229.79
Maximum 2833.41 3092.98 2925.79 3801.59 3651.05 3198.42
Minimum 2149.29 2392.15 2208.75 3140.77 3150.40 2572.29
Iat;l_e_l_1_ Yield Strength Data (psi)
Sample Port I Port II
80 RPM 100 RPM 120 RPM 80 RPM 100 RPM 120 RPM
1 2816.49 3067.36 2084.67 3039.51 3342.37 2571.04
2 2175.80 2310.88 2792.36 3309.29 3608.03 2876.93
3 2234.32 2116.93 2167.77 3161.25 3095.90 3183.38
4 2832.70 2725.00 2630.02 3787.15 3393.71 2935.52
5 2146.43 2505.64 2572.61 3115.55 3648.65 2789.80
Average 2441.15 2545.16 2449.49 3282.55 3417.73 2871.33
SD 351 .51 369.08 307.31 298.75 223.25 222.68
Maximum 2832.70 3067.36 2792.36 3787.15 3648.65 3183.38
Minimum 2146.43 2116.93 2084.67 3039.51 3095.90 2571.04

 

 

 

 

 

 

 

57

 

 

Table 12

Modulus of Elasticity Data (psi)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Port I Port II
80 RPM 100 RPM 120 RPM 80 RPM 100 RPM 120 RPM
1 271,026 243 ,315 216,709 262,920 264,343 202,785
2 270,506 229,353 207,456 262,920 416,369 178,581
3 239,331 240,246 195,872 310,612 236,657 265,288
4 231,897 263 ,224 265,873 265,558 260,707 227,725
5 237,593 257,027 257,513 239,088 259,861 253,623
Average 250,070 246,633 228,685 268,094 287,587 225,600
SD 19092.13 13543.50 31164.17 26139.53 72818.70 35694.02
Maximum 271,026 263,224 265,873 310,612 416,369 265,288
Minimum 231,897 229,353 195,872 239,088 236,657 178,581
1% Elongation Data (%)
Sample Port I Port 11
80 RPM 100 RPM 120 RPM 80 RPM 100 RPM 120 RPM
1 1.82 3.52 1.43 5.75 2.34 1.86
2 0.66 3.29 2.73 6.56 2.61 2.39
3 1.14 2.54 1.84 5.02 2.59 5.98
4 3.31 2.24 1.95 6.35 4.19 2.77
5 0.48 3.20 1.88 5.86 5.59 1.83
Average 1.48 2.96 1.97 5.91 3.46 2.97
SD 1.15 0.54 0.47 0.60 1.40 1.73
Maximum 3.31 3.52 2.73 6.56 5.59 5.98
Minimum 0.48 2.24 1.43 5.02 2.34 1.83

 

 

 

 

 

 

 

58

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:35 83 $3 $3 $2 32 mm

:3 a: $3 3.3 $3 $2 can}.

92 u was a 83 o R: a ass a 83 a 2

Rec 0 23 m ”N; a mm? o 33 a £2 a :

mas o 53 m as a Geo m 82 a E: m 2

5s a 966 3 fine a 2 2 o 83 a 8a.. a a

~65 o 56 m 266 a was a :2 u was a m

and 0 Rec 0 an; a on: a E: a a: a s

as o 83 m 56 a 82 m 3.2 o 22 a a

mas o ES a 83 a on: a £3 a was a m

Sec o :2 m 38 a a: m 32 m a: a a

22 u 33 o Geo m an: a Rad : 82 a m

:2 o «as u was a 49.? a 33 m a: a N

ass : $3 0 see a S3 0 22 a 32 a _

3375 came Eéé 223 3345 same 3575 225 ESE 22E @375 823
fiwaobm E fiwcobm mo fiwcobm mo 530ch we 3:85 we fiwqobm mo
cofi 09¢. wed 09¢. cod 09¢. vofi on»... pod 09¢. pod 09C.
2.2 OS 2.2 2: 2.2 ow 2mm 2: 2,2 2: 2.2 ow
a :8 :5 23%

@375 sac swag sass Be flamed.

59

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1311M Water Absorption Test
Sample Port I Port 11
80 RPM 100 RPM 120 RPM 80 RPM 100 RPM 120 RPM

1 0.0308 0.1137 0.0740 0.0057 0.0129 0.0191
2 0.0213 0.0297 0.0987 0.0078 0.0164 0.0067
3 0.0280 0.0392 0.0662 0.0097 0.0155 0.0115
4 0.0246 0.0369 0.0919 0.0089 0.0147 0.0166
5 0.0306 0.0647 0.0628 0.0105 0.0157 0.0053
6 0.0224 0.0588 0.0853 0.0078 0.0140 0.0167
7 0.0127 0.0593 0.0859 0.0060 0.0140 0.0161
8 0.0169 0.0476 0.0938 0.0085 0.0099 0.0185
Average 0.0234 0.0562 0.0823 0.0081 0.0141 0.0138
SD 0.0065 0.0263 0.0132 0.0017 0.0020 0.0054
Maximum 0.0308 0.1137 0.0987 0.0105 0.0164 0.0185
Minimum 0.0127 0.0297 0.0628 0.0057 0.0099 0.0053

 

 

 

 

 

 

 

60

 

Compression Molding Process

 

l.

2.

Set both platens (top and bottom) to the desired temperature (150 °C)

Assemble the frame and chrome platens in the mirror-like structure of Mylar sheets
and the metal platens

Wait for 15 minutes to heat the system up to 150°C

Place the assembly on the bottom platen

Close the hydraulic chamber by turning the knob at the base clockwise

Use pump handle to apply pressure up to 30,000 psi

Hold for 5 minutes (if pressure drops, apply more pressure using handle)

. Turn off the heat by setting temperature to 50°C

Turn on cooling water slowly
When the temperature reaches 50°C (around 5 minutes), turn off cooling water,
release the pressure by turning the knob counterclockwise.
Disassemble the assembly
Clean up: - set lower platen temperature at 200°C
- place the frame on the lower platen for about 5 minutes

- use scraper to scrape out resin

61

Izod Impact Strength Test

1.

2.

10.

11.

12.

13.

Press left soft key and hold while turning power switch on. Release soft key.

Press the key labeled “ID”, 1

Enter “Test ID”, “Data”, and “Sample ID” by pressing a soft key under corresponding
choices. After each selection press “enter” key to store the information in computer

memory.

. Bring the 5 1b. pendulum to its latched position.

Press pendulum key. Enter pendulum weight. Press “Enter”.

. Release pendulum without specimen. Press “enter”.

Select type of test (Izod) using soft key below test type as viewed on LCD display.
Enter the average sample width in “mil”. This is “width” at the notch—what we
would normally call the thickness. Press “enter”.

Enter temperature of testing. Press “enter”.

Press “enter” again. You will be asked to enter the specimen width. Measure
thickness at notched position with vemier caliper. Enter in mils, then press “enter”.
Place specimen in the clamp; use the jig for positioning the notch to align in the
center, with the notch facing the pendulum. Remove the jig.

Release the pendulum. Catch it after the beep, before it hits the sample again.
Identify the type of breakage as a complete break, hinge break, partial break or non-
break.

Repeat steps 9-12 until all specimens are finished.

62

14. To print out your data, turn on the printer. Press the key “print”. Press the right soft
key under “report”. The display will change. Press the right soft key again under the

display “print”. The printer will start to generate the report.

63

flter Absorption Test

1.

Twenty-Four Hour Immersion

The conditioned specimens shall be placed in a container of distilled water
maintained at a temperature of 23 i 1°C (73.4 3: 18°F), and shall rest on edge and
be entirely immersed. At the end of 24, +1/2, -0 h, the specimens shall be
removed from the water one at a time, all surface water wiped off with a dry cloth,
and weighed to the nearest 0.001 g immediately.

Long-Term Immersion

To determine the total water absorbed when substantially saturated, the
conditioned specimens shall be tested as described in Twenty-F our Hour
Immersion except that at the end of 24 h they shall be removed from the water,
wiped free of surface moisture with a dry cloth, weigh to the nearest 0.001 g
irrnnediately, and then replaced in the water. The weighings shall be repeated at
the end of the first week and every two weeks thereafter until the increase in
weight per two-week period, as shown by three consecutive weighings, averages
less than 1% of the total increase in weight or 5 mg, whichever is greater; the

specimen shall then be considered substantially saturated.

64

APPENDIX B

65

One-Way Analysis of Variance (ANOVA)

 

 

 

 

TENSILE STRENGTH
Source Sum of Squaresl df Mean Square Sig.
Between Groups 4025312622 5 805062.524 11.145 .000
Within Group 1733620006 24 72234.167
Total 5758932628 29

 

 

 

 

 

 

 

66

 

Post Hoc Tests

 

 

 

 

 

 

 

 

Multiple Comparisons
Dependent Variable: TENSILE STRENGTH
LSD
Treatment Treatment Mean Std. Error Sig. 95% Confidence Interval
(I) (J) Difference
(I-J)
Lower Bound Upper Bound
1 2 -269.8840 169.981 .125 -620.7083 80.9403
3 -59.5180 169.981 .729 -410.3423 291.3063
4 —883.4540’ 169.981 .000 -1234.2783 -532.6297
5 -930.0340’ 169.981 .000 -1280.8583 -579.2097
6 -384.4320' 169.981 .033 -735.2563 -33.6077
2 1 269.8840 169.981 .125 -80.9403 620.7083
3 210.3660 169.981 .228 -140.4583 561.1903
4 -6135700‘ 169.981 .001 -964.3943 462.7451
5 -660.1500' 169.981 .001 -1010.9743 809.3251
6 -1 14.5480' 169.981 .507 465.3723 236.2763
3 1 59.5180 169.981 .729 -291.3063 410.3423
2 -210.3660 169.981 .228 -561.1903 140.4583
4 -823.9360' 169.981 .000 -1 174.7603 -473.1117
5 -870.5160' 169.981 .000 -1221.3403 -5 19.6917
6 -324.9140 169.981 .068 -675.7383 25.9103l
4 1 883.4540' 169.981 .000 532.6297 1234.2783
2 613.5700' 169.981 .001 262.7457 964.3943!
3 823.9360' 169.981 .000 473.1117 1 174.7603
5 -46.5800 169.981 .786 -397.4043 304.2443]
6 499.0220‘ 169.981 .007 148.1977 849.8463
5 1 930.0340' 169.981 .000 579.2097 1280.8583
2 660.1500' 169.981 .001 309.3257 1010.9743
3 870.5160' 169.981 .000 519.6917 1221.3403
4 46.5800 169.981 .786 -304.2443 397.4043l
6 545.6020. 169.981 .004 194.7777 896.4263
6 1 384.4320. 169.981 .033 33.607 735.2563
2 1 14.5480 169.981 .507 -236.2763 465.3723
3 324.9140 169.981 .068 -25.9103 675.7383
4 -499.0220' 169.981 .007 -849.8463 -148.1977
5 -545.6020’ 169.981 .004 -896.4263 -1 94.7 777

 

 

 

 

 

 

 

 

* The mean difference is significant at the .05 level.

67

One-Way Analysis of Variance (ANOVA)

 

 

 

 

YIELD STRENGTH
Sum of Squares df Mean Square F Sig.
Between Groupsl 4644715690 5 928943.138 10.266 .000
Within Group 2171599296 24 90483.304
Total 6816314986 29

 

 

 

 

 

 

 

 

68

Post Hoc Tests

Multiple Comparisons

Dependent Variable: YIELD STRENGTH

 

 

 

 

 

 

 

 

 

 

 

 

 

LS_D
Treatment Treatment Mean Std. Error Sig. 95% Confidence Interval
(I) (J) Difference
(I-J)
Lower Bound Upper Bound
l 2 -104.0140 190.245 .590 -496.6613 288.6333
3 -8.33 80 190.245 .965 400.9853 384.3093
4 -841.4020' 190.245 .000 -1234.0493 -448.754
5 -976.5840' 190.245 .000 -1369.2313 -583.936
6 -430.1860' 190.245 .033 -822.8333 -37.538
2 1 104.0140 190.245 .590 -288.6333 496.6613
3 95.6760 190.245 .620 -296.9713' 488.3233
4 -737.3880' 190.245 .001 -1 130.03 53 -344.7407
5 -872.5700° 190.245 .000 -1265.2173 479.9223
6 -326.1720 190.245 .099 -718.8193 66.4753
3 1 8.3380 190.245 .965 -384.3093 400.9853
2 -95.6760 190.245 .620 488.3233 296.9713
4 -833.0640' 190.245 .000 -1225.71 13 -440.4167
5 -968.2460° 190.245 .000 -1360.8933 675.5983
6 -421.8480' 190.245 .036 -814.4953 29.2003
4 1 841 .4020' 190.245 .000 448.7547 1234.0493
2 737 .3880' 190.245 .001 344.7407 1130,0353
3 83.0640 190.245 .000 440.4167 1225.71 13
5 -135.1820 190.245 .484 627.8293“ 257.4653
6 41 12160 190.245 .041 18.5687 803.8633
5 1 976.5840' 190.245 .000 583.936 1369.2313
2 872.5700' 190.245 .000 479.9227 1265.2173
3 968.2460' 190.245 .000 575.5987 1360.8933
4 135.1820 190.245 .484 -257 .4653 527.8293
6 546.3980 190.245 .008 153.7507 939.0453
6 1 430.1860' 190.245 .033 37.5387 822.8333
2 326.1720 190.245 .099 -66.4753 718.8193
3 421 .8480' 190.245 .036 29.2007 814.4953
4 -411.2160° 190.245 .041 -803.8633 -18.5687
5 -546.3980° 190.245 .008 -939.0453 -153.7507

 

 

 

 

* The mean difference is significant at the .05 level.

69

One-Way Analysis of Variance (AN OVA)

 

 

 

 

MODULUS OF ELASTICITY
Sum of Squares] df Mean Square Sig.
Between Groups 13990490985 .200 5 2798098197040 1.91 .129
Within Group 35092021182000 24 1462167549250
Total 49082512167200 29

 

 

 

 

 

 

 

 

70

Post Hoc Tests

Multiple Comparisons

Dependent Variable: MODULUS OF ELASTICITY

 

 

 

 

 

 

 

 

 

 

 

 

LSD
Treatment Treatment Mean Std. Error Sig. 95% Confidence Interval
(I) (J) Difference (I-J)
Lower Bound Upper Bound
1 2 3437,6000 24184.024 .888 -46475.7725 53350.9725
3 21386.0000 24184.024 .385 -28527.3725 71299.3725
4 -l 8149.0000 24184.024 .460 -68062.3725 31764.3725
5 -37516.8000 24184.024 .134 -87430.1725 12396.5725
6 24470.2000 24184.024 .322 -25443.1725 74383 .5725
2 1 -3437.6000 24184.024 .888 -53350.9725 46475.7725
3 17948.4000 24184.024 .465 -31964.9725 67861.7725
4 -21586.6000 24184.024 .381 -71499.9725 28326.7725
5 -40954.4000 24184.024 .103 -90867.7725 8958.9725
6 21032.6000 24184.024 .393 -28880.7725 70945.9725
3 1 -21386.0000 24184.024 .385 -71299.3725 28527.3 725
2 -17948.4000 24184.024 .465 -67861.7725 31964.9725
4 -39535.0000 24184.024 .115 -89448.3725 10378.3725
5 -58902.8000' 24184024 .023 -108816.1725 -8989.4275
6 3084.2000 24184.024 .900 -46829.1725 52997.5725
4 1 18149.0000 24184024 .460 -31764.3725 68062.3 725
2 21586.6000 24184.024 .381 -28326.7725 71499.9725
3 39535.0000 24184.024 .115 -10378.3725 89448.3725
5 -19367.8000 24184.024 .431 -69281.1725 30545.5725
6 42619.2000 24184024 .091 -7294. 1725 92532.5725
5 1 37516.8000 24184.024 .134 -12396.5725 87430.1725
2 40954.4000 24184.024 . 103 -8958.9725 90867.7725
3 589028000 24184.024 .023 8989.4275 108816.1725
4 19367 .8000 24184.024 .431 -30545.5725 69281.1725
6 619870000 24184.024 .017 12073.6275 111900.3725
6 1 -24470.2000 24184.024 .322 -743 83 .5725 25443.1725
2 -21032.6000 24184.024 .393 -70945.9725 28880.7725
3 -3 084.2000 24184.024 .900 -52997.5725 46829.1725
4 -42619.2000 24184.024 .091 -92532.5725 7294.1725
5 -61987.0000' 24184024 .017 -1 l 1900.3725 -12073.6275

 

 

 

* The mean difference is significant at the .05 level.

71

 

One-Way Analysis of Variance (ANOVA)

 

 

 

 

PERCENT ELONGATION
Sum of Squaresl dfl Mean Square F Sig.
Between Groupsi 59.780 5 11.956 10.058 .000
Within Groups 28.530 24 1.189
Total 88.309 29

 

 

 

 

 

 

 

 

72

Post Hoc Tests

Multiple Comparisons

Dependent Variable: PERCENT ELONGATION

 

 

 

 

 

 

 

 

 

 

 

 

 

L_S_D
Treatment Treatment Mean Std. Error Sig. 95% Confidence Interval
(I) (J) Difference
(I-J)
Lower Bound Upper Bound
1 2 -l .4760T .690 .043 -2.8992 -5.2813E-02
3 -.4840 .690 .489 -1.9072 .9392
4 -4.4260' .690 .000 -5.8492 -3 .002
5 -1.9820' .690 .008 -3 .4052 -.558
6 -1.4840' .690 .042 -2.9072 -6.0813E-0
2 1 1 .4760' .690 .043 5.281E-02 2.899
3 0.9920 .690 .163 -.4312 2.415
4 -2.9500' .690 .000 -4.3732 -1.526
5 -0.5060 .690 .470 -1.9292 .917
6 -8.0000E-03 .690 .991 -1.4312 1.415
3 1 0.4840 .690 .489 -.9392 1.907
2 -0.9920 .690 . 163 -2.4152 .431
4 -3.9420° .690 .000 -5.3652 -2.518
5 -1.4980’ .690 .040 -2.9212 -7.4813E-02
6 -1 .0000 .690 .160 -2.4232 .4232
4 1 4.4261? .690 .000 3 .0028 5.8492
2 2.9500' .690 .000 1.5268 4.3732
3 3.9420' .690 .000 2.5188 5.3652
5 2.4440' .690 .002 1 .0208 3.8672
6 2.9420' .690 .000 1.5188 4.3652
5 1 1.9820' .690 .008 .5588 3.4052
2 0.5060 .690 .470 -.9172 1.9292
3 1.4980' .690 .040 7.481E-02 2.9212
4 -2.4440' .690 .002 -3 .8672 -1.0208
6 0.4980 .690 .477 -.9252 1.9212
6 1 1 .4840’ .690 .042 6.081E-02 2.9072
2 8.000E-03 .690 .991 -1.4152 1.4312
3 1.0000 .690 .160 -.4232 2.4232
4 -2.9420' .690 .000 -4.3652 -1.5188
5 -0.4980 .690 .477 -1.9212 .9252

 

 

 

 

* The mean difference is significant at the .05 level.

73

One-Way Analysis of Variance (ANOVA)

 

 

 

 

 

IMPACT STRENGTH
Sum of Squares] df Mean Square F Sig.
Between GroupsL 5.521 5 1.104 22.611 .000
Within Groupsl 3.174 65 4.883E-02
Total 8.695 70

 

 

 

 

 

 

 

 

74

Post Hoc Tests

Multiple Comparisons

Dependent Variable: IMPACT STRENGTH

 

 

 

 

 

 

 

 

 

 

 

fl
Treatment Treatment Mean Std. Error Sig. 95% Confidence Interval
(I) (J) Difference
(I-J)
Lower Bound Upper Bound
l 2 .2834' .090 .003 .1032 .463
3 .3334' .092 .001 .1492 .5177
4 .6842' .090 .000 .5040 .8643
5 .6681' .090 .000 .4879 .8483
6 .7903' .090 .000 .6101 .97041
2 1 -.2834' .090 .003 -.4636 -.1032
3 5.003E-02 .092 .589 -.1342 .2343
4 .4008' .090 .000 .2206 .5809
5 .3847' .090 .000 .2045 .5648
6 .5068' .090 .000 .3267 .6870
3 1 -.3334' .092 .001 -.5177 -.1492
2 -5.0030E-02 .092 .589 -.2343 .1342
4 .3507' .092 .000 .1665 .5349
5 3346‘ .092 .001 .150 .5189
6 .4568' .092 .000 .2723 .6410
4 1 -.6842’ .090 .000 -.8643 -.5040
2 -.4008' .090 .000 -.580 -.2206
3 -.3507° .092 .000 -.534 -.1665
5 -1.6083E-02 .090 .859 -.1963 .1641
6 . 1061 .090 .244 -7.4089E-02 .2863]
5 1 -.6681‘ .090 .000 -.8483 -.487
2 -.3847° .090 .000 -.5648 -.2045
3 -.3346° .092 .001 -.5189 -.150
4 1.608E-02 .090 .859 -.1641 .1963
6 .1222 .090 .180 -5.8006E-02 .3023
6 1 -.7903' .090 .000 -.970 -.6101
2 «5068’ .090 .000 -.687 -.3267
3 -.4568' .092 .000 -.641 -.2726
4 -. 1061 .090 .244 -.2863 7.409E-02
5 -. 1222 .090 .180 -.3023 5.801E-02

 

 

 

 

 

 

 

* The mean difference is significant at the .05 level.

75

One-Way Analysis of Variance (ANOVA)

 

 

 

 

WATER ABSORPTION
Sum of Square df Mean Square F Sig.
Between GroupsI 3.527E-02 5 7.053E-03 44.911 .000
Within Groups 6.596E-03 42 1.570E-04
Total 4.186E-02 47

 

 

 

 

 

 

 

 

76

Post Hoc Tests

Multiple Comparisons

Dependent Variable: WATER ABSORPTION

 

 

 

 

 

 

 

 

 

LS_D
Treatment Treatment Mean Std. Error Sig. 95% Confidence Interval
(I) (J) Difference
(I-J)
Lower Bound Upper Bound
1 2 -3 .2825E-02' .006 .000 -4.5470E-02 -2.0180E-02
3 -5.8912E-02’ .006 .000 -7.1558E-02 -4.6267E-02
4 1.530E-02' .006 .019 2.655E-03 2.795E-02
5 9.275E-03 .006 . 146 -3 .3 70 1 E-03 2.192E-02
6 9.600E-03 .006 .133 -3.0451E-03 2.225E-0
2 1 3 .282E-02’ .006 .000 2.018E-02 4.547E-02
3 -2.6088E-02° .006 .000 -3 .8733E-02 -1 .3442E-02
4 4.812E-02' .006 .000 3.548E-02 6.077E-02
5 4.210E-02' .006 .000 2.945E-02 5.475E-02
6 4.243E—02' .006 .000 2.978E-02 5.507E-02
3 1 5.891E-02' .006 .000 4.627E-02 7. 1 56E-02
2 2.609E-02' .006 .000 l .344E-02 3 .873 E-02
4 7.421 E-02' .006 .000 6. 1 57E-02 8.686E-02
5 6.819E-02' .006 .000 5.554E-02 8.083E-02
6 6.851E—02' .006 .000 5.587E—02 8.1 16E-02
4 1 -1 .5300E-02' .006 .019 -2.7945E-02 -2.6549E-03
2 -4.8125E-02' .006 .000 -6.0770E-02 -3 .5480E—02
3 -7.4213E-02° .006 .000 -8.6858E—02 -6.1 567E-02
5 -6.0250E-03 .006 .342 -1 .8670E-02 6.620E-03
6 -5 .7000E-03 .006 .368 -1 .8345 E-02 6.945E-03I
5 1 -9.2750E—03 .006 .146 -2.1920E-02 3.370E-03
2 -4.2100E-02' .006 .000 -5 .4745E-02 -2.9455E-0
3 -6.8188E—02' .006 .000 -8.0833E-02 -5 .5542E-02
4 6.025E—03 .006 .342 -6.6201E-03 1 .867E-02
6 3.250E-04 .006 .959 -1 .2320E—02 1 .297E-02
6 1 -9.6000E-03 .006 .133 -2.2245E-02 3.045E-03
2 -4.2425E-02° .006 .000 -5 .5070E-02 -2.9780E-02
3 -6.8512E-02' .006 .000 -8.1 158E-02 -5 .5867E-02
4 5 .700E-03 .006 .368 -6.945 1E-03 1 .83 5E-02
5 -3 .2500E-04 .006 .959 -1 .2970E-02 1 .232E-02

 

 

 

 

 

 

 

* The mean difference is significant at the .05 level.

77

10.

11.

12.

REFERENCES

Municipal Solid Waste Generation, Recycling and Disposal in The United States:
Facts and Figures for 1998. [Online] Available. http://www.epa/gov/osw,
April 2000.

Source Reduction and Reuse. [Online] Available
http://www.epa.gov/epaoswer/non-hw/muncpl/sources.htrnl , January 2001.

Characterization of Municipal Solid Waste in the United States. [Online]
Available

http://www.epa.gov/epaoswer/non-hw/muncpl/msw98.htrnl , January 2001.

Kimbell D., Recycling in Americapa Reference Hgdbook, ABC-CLIO Inc.,
California, 254 pp. (1992).

Bisio, L.A., and M. Xanthos (ed), How to Manage Plastic Waste: Technology and
Market Opportunitites. Hanser, New York, 253 pp. (1995).

Birley, A.W., B. Haworth and J. Batchelor, Physics of Llastics: Processing,
Properties and Materials Engineering, Oxford University Press, New York (1992).

George, J ., S.S. Bhagawen, N. Prahakaran, and S. Thomas, “Short Pineapple-
Leaf-Fiber-Reinforced Low-Density Polyethylene Composites”, Journal of
Applied Polymer Science. Vol. 57, pp. 843-854 (1995).

Kamani, R., “Kenaf-Reinforced Polypropylene Composites”, MS. Thesis,
Michigan State University, East Lansing, MI, 1996.

Chtourou, H., B. Riedl, and A. Ait-Kadi, “Reinforcement of Recycled Polyolefins
with Wood Fibers”, Journal of Reinforced Plastics and Composites, Vol. 11, pp.
372-394 (1992).

Marcin, Thomas C., “Technological Assessment of Composite Combinations of
Wood and Nonwood Materials”, Proceedings of the Society of America_n_
Foresters Natiorgl Convention. Bethesda, MD, pp. 559-560 (1991).

Babyak, Richard J ., Wood Reinforced Plastic, Appliance Manufacturer March, 84
pp. (1 993).

Rojanarungtawee, S., “Composite of Wood Fiber and Mixed Recycled
Thermoplastics”, MS. Thesis, Michigan State University,East Lansing, MI, 1998.

78

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

Dietz, Albert G.H., Compoaite Materials. American Society for Testing and
Materials, Philadelphia, PA, 55 pp. (1965).

Chawla, K.K., Composite Materials. Springer-Verlag, New York Inc., Germany,
292 pp. (1987).

Richardson, T., Composites: A Design Guide, Industrial Press Inc., New York,
(1987)

Agarwal, B.D., and L.J. Broutman, Agni/sis a_n_d Performance of Fiber
Composites. John Wiley & Sons, Inc., United States, 449 pp. (1990).

Malllick, P.K., F iber-Reinforced Composites: Materials. Manufacturing, and
Design, Marcel Dekker, Inc., New York, 469 pp. (1988).

Gibson, R.F., Principles of Composite Material Mechanics, McGraw-Hill, Inc.,
New York, 425 pp. (1994).

Simpson, R.J., “Composite Materials from Recycled Multi-Layer Polypropylene
Bottles and Wood Fibers”, MS. Thesis, Michigan State University, 1991.

Clegg, D.W., and AA. Coolyer (ed), Mechanical Properties of Reinforced
Thermoplastics, Elsevier Applied Science Publishers, New York, 326 pp. (1986).

Daniel, M., and O. Ishai, Engineering Mechanics of Composite Materials. Oxford
University Press, New York, 395 pp. (1994).

Pieper, R, “Wood Waste Alchemy”, Biocycle, 40 pp. (1993).

Woodhams, R.T., G. Thomas, and BK. Rodgers, “Wood Fibers as Reinforcing
Fillers for Polyolefins”, Polymer Engineering and Science Vol. 24, pp.1166-1171
(1984)

Beshay, A.D., V.K. Bohuslav, and Claude Daneault, “Use of Wood Fibers in
Thermoplastic Composites II: Polyethylene”, Polvmer Composites. Vol. 6, No. 4,
262 pp. (1985).

Gogoi, B.K., “Processing-Morphology-Property Relationships for Compounding
Wood Fibers with Recycled HDPE Using a Twin-Screw Extruder”, MS. Thesis,
Michigan State University, East Lansing, MI, 1989.

Wildi, R. and Maier, C., Understanding Compounding, Hanser/Gardner
Publications, Inc., Cincinnati, 189 pp. (1998).

79

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Rauwendall, C., UndersgndingExmsion. Hanser/Gardner Publication, Inc., pp.
23-24 (1998).

Yam, IK., V. Kalyankar, S. Selke, and C. Lai, “Mechanical Properties of Wood
Fiber/Recycled HDPE Composites” Society of figfic Engineers Technical
Conference ANTEC’88. pp. 1809-1911, (1988)

Kalyankar, V., “Mechanical Characteristics of Composites Made from Recycled
HDPE Obtained from Milk Bottles”, MS. Thesis, Michigan State University,
East Lansing, MI, 1989.

Nieman, K.A., “Mechainical Property Enhancement of Recycled HDPE and
Wood Fiber Composites Due to the Inclusion of Additives”, MS. Thesis,
Michigan State University, East Lansing, MI, 1989.

Keal, M.D., “The Effect of Dual Additives Systems on the Mechanical Properties
of Aspen Fiber/Recycled HDPE Composites”, MS. Thesis, Michigan State
University, East Lansing, MI, 1990.

Ren, S., and David N.-S. Hon, “Newspaper F iber—Reinforced Polypropylene
Composite”, Journal of Reinforced Plastics and Composites. Vol. 12, No. 12, pp.
1311-1322, (1993).

Chotipatoomwan, T., “Processing and Mechanical Property Testing of Paper
Fiber and HDPE Composites”, MS. Thesis, Michigan State University, East
Lansing, MI, 1998.

Ricciardi, J .J ., “High Density Polyethylene/Paper Fiber Composites: Measuring
the Impact of Additives on Their Physical and Mechanical Properties”, MS.
Thesis, Michigan State University, East Lansing, MI, 1999.

Thepwiwatj it, N., “Composite of Wood Fiber and Recycled HDPE Bottles from
Household Use”, MS. Thesis, Michigan State University, East Lansing, MI,
2000.

ASTM D63 8-99, “Standard Test Method for Tensile Properties of Plastics”,
Annual Book of ASTM Standards, Philadelphia, P.A., Vol. 8.01, pp. 46-54
(2000)

ASTM D256-97, “Standard Test Methods for Determining the Izod Pendulum

Impact Resistance of Plastics”, Annual Book of ASTM Standards, Philadelphia,
P.A., Vol. 8.01, pp. 1-12 (2000).

80

38.

39.

40.

41.

ASTM D570-98, “Standard Test Method for Water Absorption of Plastics”,
Annual Book of ASTM Standards, Philadelphia, P.A., Vol. 8.01, pp. 32-34
(2000)

Katz, SH, and J .V. Milewski (ed), Handbook of Fillers for Plastics. Van
Nostrand Reinhold Co., New York, 467 pp. (1987).

Bigg, D.M., “Mechanical Properties of Particulate Filled Polymers” Polymer
Composites. Vol. 8, No. 2, pp. 115 (1987).

Lawrence E.N., Mechanical Properties of Polymers and Composites. Volume 2,
Marcel Dekker, Inc., New York, 556 pp. (1974).

81

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

llllllllllll[llllllllllllll