w .
:32:- m... up...”

an

I? .)

m
.125

... v.3.

«MEN "5..
. 4k“
3 wrn

#16918

A)

.z “In.

R.

“.317

x... ...
,

firfii
3- .u. . l.
t: 1!: 2:!

nu Warm ,
.f !.Ju - . o
u“ r .. ”Wm {Whnfifl »
. .4, y I?“
in 144131
.53“

$1 ..
5.. ).
.2.

1515.5:
3.49.6
l

3:... ‘ .
391:}?! t... .
Jon: (I) 33...!
.. It?! Au

3.2:: #99
. .L. A).

 

h. ..;r.r hr.
.81 .{Ii J

5..."...

 

 

.L ‘ .. . . . . .
u : ”5&9... Jam§ngfi . L. 1.35:... H. r IF: V i .l sauna: :L. ,
. - . , V .33..-. u. ; z . . . . $.2......u.h.m?r§§ygfi.

 

 

1, memy
1 O 05 Michigan State
University

This is to certify that the
dissertation entitled

EFFECTS OF COPPER AMINE TREATMENTS ON
MECHANICAL, BIOLOGICAL AND SURFACE/INTERPHASE
PROPERTIES OF POLY(V|NYL CHLORIDE)/WOOD
COMPOSITES

presented by

Haihong Jiang

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Forestry

 

 

W

Major Professor’s Signature
4/5/900f

Date

MSU is an Affirmative Action/Equal Opportunity Institution

-.-.-—.-‘— --.-.-—.--—.—.--.-.-.—-.—.-.-.-.------.-.-.-.-.-.—.--.-.-.--.—.-.-.--.--.-.--—--.-.-

-—.-.- - —

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

DATE DUE

 

1g©lggfi

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 czicficmmou.mms

“——

EFFECTS OF COPPER AMINE TREATMENTS 0N MECHANICAL,
BIOLOGICAL AND SURF ACE/INTERPHASE PROPERTIES OF POLY (VINYL
CHLORIDE)/WOOD COMPOSITES

By

Haihong J iang

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Forestry

2005

ABSTRACT

EFFECTS OF COPPER AMINE TREATMENTS ON MECHANICAL,
BIOLOGICAL AND SURFACE/INTERPHASE PROPERTIES OF POLY (VINYL
CHLORIDE)/WOOD COMPOSITES

By
Haihong Jiang

The copper ethanolamine (CuEA) complex was used as a wood surface modifier and
a coupling agent for wood-PVC composites. Mechanical properties of composites, such
as unnotched impact strength, flexural strength and flexural toughness, were significantly
increased, and fungal decay weight loss was dramatically decreased by wood surface
copper amine treatments. It is evident that copper amine was a very effective coupling
agent and decay inhibitor for PVC/wood flour composites, especially in high wood flour
loading level.

A DSC study showed that the heat capacity differences (ACp) of composites before
and after PVC glass transition were reduced by adding wood particles. A DMA study
revealed that the movements of PVC chain segments during glass transition were limited
and obstructed by the presence of wood molecule chains. This restriction effect became
stronger by increasing wood flour content and by using Cu-treated wood flour. Wood
flour particles acted as “physical cross-linking points” inside the PVC matrix, resulting in
the absence of the rubbery plateau of PVC and higher E’, E” above Tg, and smaller tan
5 peaks.

Enhanced mechanical performances were attributed to the improved wetting

condition between PVC melts and wood surfaces, and the formation of a stronger

interphase strengthened by chemical interactions between Cu-treated wood flour and the
PVC matrix. Contact angles of PVC solution drops on Cu-treated wood surfaces were

decreased dramatically compared to those on the untreated surfaces. Acid-base (polar),
Y“, electron-acceptor (acid) (7+), electron-donor (base) (7.) surface energy components

and the total surface energies increased after wood surface Cu-treatments, indicating a
strong tendency toward acid-base or polar interactions. Improved interphase and
interfacial adhesion were further confirmed by measuring interfacial shear strength

between wood and the PVC matn'x.

To my parents, and my husband

ACKNOWLEDGEMENTS

I would like to express my tremendous and deep thankfulness and appreciation to
Professor D. Pascal Kamdem, my advisor, for patiently guiding and supporting me
throughout these years of study at MSU. He is not only an excellent scientist, and a
diligent leading researcher, but also a warm-hearted educator and instructor, as well as a
parental mentor for me. Without his talented ideas, crucial inspirations, patient guidance
and great encouragements, I could not have finish my Ph.D. research and this
dissertation.

I sincerely and deeply acknowledge all the wonderful support and guidance,
valuable, constructive advice and suggestions from my Ph.D guidance committee,
Professor D. Pascal Kamdem, Professor Lawrence T. Drzal, Professor Melvin R.
Koelling and Dr. Runsheng Yin. I am honored and lucky to have these extraordinary
scientists to be my guidance committees. The chapter of surface and interphase study was
based on knowledge I gained from Dr. Drzal’s class and the excellent research
accomplishments of their Composite Materials and Structures Center. Without Dr.
Drzal’s instructions, we would not have assured the improved decay resistance of copper
amine treated wood-PVC composites. Also, without Dr. Koelling’s suggestions, I would
not have found the reasonable explanation regarding color stability of copper-treated
composites and put forward effective ways for decay protection. Following Dr. Yin’s
advice, I examined the market for WPC and analyzed advantages and disadvantages of
PVC/wood composites.

I also would like to sincerely acknowledge Michigan State University, the USDA

Eastern Hardwood Utilization Program for the funding support, the CertainTeed

Corporation for the Oxyvinyl and the Crompton Corporation for providing raw materials,
the MSU Center for Advanced Microscopy for SEM-EDS analysis, the MSU Composite
Materials and Structures Center for XPS and interfacial shear strength measurements, Dr.
Rui Huang in the MSU Chemistry Department for the assistance in far-IR analysis, as

well as the MSU English Language Center for their great help in grammar corrections.

I also would like to express my tremendous gratitude to Mrs. Kamdem, Mrs. Drzal,
Mrs. Koelling and Mrs. Yin for all their care and support they have given to me

throughout my PhD. research and study at MSU.

Thanks to other students in the group, Pascal Nzokou, Weining Cui, Kyle Wehner,
Joe Pennock, Josh Rawson and Sedric Pankras for all their friendship, assistance and

cooperation.

Finally, thanks to my family: my parents, Guangiian Jiang and Yuzhen Zhang, and
my husband, Shubiao Li, as well as my brother and sister-in-law for always being there

for me.

vi

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES
Chapter 1
Introduction to the Development of Wood and Natural Fiber-Plastic
Composite ......................................................................................... 1-27
1.1 Markets and Manufacturers ............................................................. 2-7
1.2. Formulations and Additives .......................................................... 7-9
1.3 Coupling Agents and Compatibilizers .............................................. 9-13
1.4 Hybrids .............................................................................. 13-15
1.5 Performances and Characteristics ................................................ 15-19
1.6 Background and Objectives ....................................................... 19-20
References ............................................................................... 21-27
Chapter 2
Characterization of Raw Materials and Laboratory Manufacture of PVC/Wood
Flour Composites .............................................................................. 28-36
2.1 Characterization of the Raw Materials .......................................... 28-32
2.1.1 Thermoplastics (PVC) .................................................. 28-29
2.1.2. Wood Flour .............................................................. 29-32
2.2 Preparation of Copper Amine Treating Solutions ................................. 33
2.3 Wood Flour Copper Amine Treatments ............................................. 33
2.4. Compression Molding of PVC/Wood Flour Composites .................... 33-35
2.5 Density Measurement .................................................................. 35
References .................................................................................... 36
Chapter 3
Effects of Copper Amine Treatment on Mechanical Properties of PVC/Wood
Flour Composites .............................................................................. 37-56
3.1 Testing Methods and Equipment ................................................. 37-39
3.1.1 Mechanical Property Testing .......................................... 37-38
3.1.2 Statistical Analysis .......................................................... 38
3.1.3 Scanning Electron Microscope (SEM) and
Energy-dispersive Spectroscopy (EDS) Analysis .................. 38-39
3.2 Results and Discussion ............................................................ 39-53
3.2.1 Impact Properties ........................................................ 39-43
3.2.2 Flexural Properties ...................................................... 43-45
3.2.2.1 Flexural Strength ............................................... 43-44
3.2.2.2 Flexural Modulus ................................................... 44

vii

3.2.2.3 Flexural Toughness ............................................. 44-45

3.2.3 Fracture Surface Analysis .............................................. 45-53
3.3 Conclusions ......................................................................... 53-54
References ............................................................................... 55-56
Chapter 4
DSC and DMA Studies of Poly(Vinyl Chloride)/Wood Flour Composites. . . . . ....57-80
4.1 Principles of DSC and DMA Analysis .......................................... 57-61
4.1.1 DSC (Differential Scanning Calorimetry) ........................... 57-59
4.1.2 DMA (Dynamic Mechanical Analysis) .............................. 60-61
4.2 Applications of DSC and DMA in Wood-Plastic Composites ............... 61-63
4.3 DSC Experimental Materials and Methods ..................................... 64-65
4.3.1 Sample Preparation ......................................................... 64
4.3.2 Experimental Methods .................................................. 64-65
4.4 DMA Experimental Materials and Methods ........................................ 65
4.4.1 Sample Preparation ......................................................... 65
4.4.2 Experimental Methods ...................................................... 65
4.5 Results and Discussion ............................................................ 66-76
4.5.1 DSC Glass Transition Temperature (Tg) ............................ 66-67
4.5.2 DSC Heat Capacity Differences (ACp) ................................. 67-70
4.5.3 DMA Modulus (E’ and E”) ............................................ 70-72
4.5.4 DMA tan6 ................................................................ 73-74
4.5.5 Influence of Wood Copper Treatment ................................ 74-75
4.5.6 Viscosity Measured by DMA .......................................... 75-76
4.6 Conclusions ......................................................................... 76-77
References ............................................................................... 78-80
Chapter 5
Surface and Interphase Characterization of PVC-Copper Amine Treated
Wood Flour Composites .................................................................... 81-107
5.1 Background .......................................................................... 81-86
5.1.1 Fiber-Matrix Interphase Bonding Mechanisms ..................... 81-83
5.1.2 Wood Fiber (Flour)-Plastic Wettabiligy and Interphase ........... 84-85
5.2 Theoretical Calculation of Surface Energy and Surface
Energy Components ................................................................ 86-88
5.3 Contact Angle Measurement ...................................................... 89-90
5.3.1 Wood Cu-Treatment ........................................................ 89
5.3.2 Preparation of PVC solution .............................................. 89
5.3.3 Contact Angle Measurement .......................................... 89-90
5.4 Experimental Design for Interfacial Shear Strength Measurement .......... 90-92
5.4.1 Sample Preparation ......................................................... 90
5.4.2 Interfacial Shear Strength Measurement ............................. 90-92
5.4.3 Statistical Analysis .......................................................... 92
5.5 Results and Discussion ........................................................... 92-102

viii

5.5.1 Wettability Evaluation of PVC to Surface

Modified Wood Surface ................................................ 92-94
5.5.2 Effects of Copper Amine Treatment of Surface Energy
and Surface Energy Components of Wood Surface ................ 95-98
5.5.3 Interfacial Adhesion Estimation by
Interfacial Shear Strength ............................................. 98-102
5.6 Conclusions ...................................................................... 102-103
References ............................................................................ 104-107
Chapter 6
Effect of C0pper Treatment on the Interphase Chemistry of PVC/Wood Flour
Composites ................................................................................... 108-129
6.1 Background .. .......................................................................................... 108-109
6.2 Analytical Techniques .......................................................... 109-1 12
6.2.1 Sample Preparations ....................................................... 109
6.2.2 Fourier Transform Infrared Spectroscopy (FTIR)
and Far IR Analysis .................................................. 109-110
6.2.3 X-ray Photoelectron Spectroscopic (XPS) Analysis ............ 110-112
6.3 Interphase Chemistry of PVC/Copper-Treated Wood
Flour Composites ................................................................ 112-125
6.3.1 Fourier transform infrared spectroscopy (FTIR) and
Far-IR analysis ........................................................ 112-1 18
6.3.2 X-ray Photoelectron Spectroscopic (XPS) Analysis ............ 119-125
6.4 Conclusions .............................................................................. 126
References ............................................................................ 127-129
Chapter 7
Biological Performance and Environmental Durability Evaluation of PVC/Copper-
Treated Wood Flour Composites ........................................................ 130-164
7.1 Introduction ...................................................................... 130-132
7.2 Experimental and Methods ..................................................... 132-138
7.2.1 Fungal Decay ......................................................... 132-135
7.2.1.1 Sample Preparation ....................................... 132-133
7.2.1.2 Accelerated Fungal Decay Tests ........................ 133-135
7.2.2 Outdoor Weathering ................................................. 135-136
7.2.2.1 Sample Preparation ............................................. 135
7.2.2.2 Outdoor Exposure ......................................... 135-136
7.2.3 Artificial Accelerated Weathering .................................. 136-138
7.2.3.1 Sample Preparation ............................................. 136
7.2.3.2 Accelerated Weathering and Color Stability .......... 136-138

7.3 Results and Discussion ......................................................... 138-159
7.3.1 Fungal Decay Resistance of PVC/Copper Amine-Treated

Wood Composites .................................................... 138-152
7.3.2 Effects of Copper Treatments on Natural Weathering Durability of
PVC/Copper Amine-Treated Wood Composites ................ 153-154
7.3.3 Color Stability of PVC/Copper Amine-Treated, Untreated Wood
Composites during Accelerated Weathering ..................... 154-159
7.4 Conclusions ...................................................................... 160-161
References ............................................................................ 162-164
Chapter 8
Conclusions and Recommendations .................................................... 165-168

LIST OF TABLES

Table 1.1. North American PVC-Wood Composite Manufacturers (Schut 1999) ......... 7
Table 1.2. Effects of Coupling Agent PMPPIC on the Mechanical Properties of
PVC/Wood Flour Composites (wood sawdust content was 25 wt%) ........................ 12
Table 1.3. Comparison of mechanical and physical properties for PVC/wood fiber (WF)
composites with wood and other wood composites ............................................ 16
Table 2.1. Physical and mechanical properties of rigid PVCs used in this study ......... 29
Table 2.2. Chemical composition and physical properties of oak ........................... 32
Table 2.3. Elemental composition of red oak ash (%) ........................................ 32

Table 2.4. Compositions of Experimental Samples with Different PVC Types, PVC,
Treated and Untreated Wood Flour Contents, as well as Cu Concentrations in Treated
Wood Flour .......................................................................................... 34

Table 4.1. Compositions of Samples with PVC, Treated and Untreated Wood Flour
Contents, as well as Cu Concentrations in Treated Wood Flour ............................. 64

Table 4.2. Compositions of Experimental Samples with PVC, Treated and Untreated
Wood Flour Contents, as well as Cu Concentrations in Treated Wood Flour ............. 65

Table 4.3. DSC Measured Heat Capacity Difference (ACp) of PVC and PVC/Cu-
Treated, Untreated WF Composites in Comparison with the Modeling Results .......... 70

Table 5.1. Surface Tension Components (mJ/mz) of Testing Solvents at 20 °C .......... 90

Table 5.2. Average Contact Angles and Drop Dimensions of Sessile Drops of PVC
Solution on Untreated and Copper Amine (CuEA)-Treated Oak Surfaces ................. 94

Table 5.3. Average Contact Angles of Testing Solvents on Untreated and treated Wood
Surface .............................................................................................. 95

Table 5.4. Dispersive and Polar Surface Energy Components (mJ/mz) of Treated,
Untreated Red Oak Calculated from Geometric-Mean and Harmonic-Mean Models ....97

Table 5.5. Surface Energy Components (mJ/mz) of Treated, and Untreated Red Oak

Calculated from the Lewis Acid-Base Model ................................................ 98
Table 6.1. Assignment of IR Absorption Bands ............................................. 113
Table 6.2. Assignments of IR Absorption Bands of PVC ................................... 114

xi

Table 6.3. Assignments of far-IR absorption bands of Cu bonded with O, N, and Cl as
observed in copper treated, untreated WF and their composites ........................... 117

Table 6.4. Surface Atomic Compositions (%) of PVC, Wood Flour, and PVC/Cu-treated,
Untreated WF Composites ....................................................................... 120

Table 6.5. The Cls peak deconvolution assignment for PVC, Wood and PVC-Cu-treated
Wood Composites ................................................................................. 123

Table 6.6. The C1, C2 and C3 Peak Area Ratios for PVC, Wood Flour, and PVC/Cu-
treated, Untreated WF Composites ............................................................. 123

Table 7.1. Compositions of Experimental Samples with PVC, CuEA-Treated and
Untreated Wood Flour Contents, as well as Cu Concentrations in Treated Wood Flour
........................................................................................................ 133

Table 7.2. Compositions of Experimental Samples with PVC, CuEA-Treated and
Untreated Wood Flour Contents, as well as Cu Concentrations in Treated Wood
Flour ................................................................................................ 136

Table 7.3. Sample Sizes and Applied Standards for Biological and Weathering
Performance Evaluation .......................................................................... 138

Table 7.4. Weight Loss of PVC, Composites and Red Oak after 3 Months Fungus Decay
Test .................................................................................................. 139

Table 7.5. Moisture Contents of PVC and composites after exposed in the environment
with 60 % and 95% relative humidity .......................................................... 150

xii

LIST OF FIGURES

Figure 1.1. Market demand for natural fiber/plastic composites in North America ....... 3

Figure 1.2. Average market prices of wood fiber/flour and other fillers and
reinforcements ........................................................................................ 4

Figure 1.3. North American market shares for wood/plastic composites with different
polymers (Principia Partners 2003) ............................................................... 5

Figure 1.4. WPC application markets in North America (Clemons 2000) .................. 5
Figure 1.5. Notched impact strength of PVC/wood flour/glass fiber composites with

different glass fiber types, type L glass fiber contents and PVC loading levels.
PW54: PVC: wood + fiber glass = 55:45; PW55: PVC: wood + fiber glass = 50:50;

PW46: PVC: wood + fiber glass = 40:60 ....................................................... 14
Figure 1.6. Color and thickness change of PVC/WF composite and natural wood during

12 month outdoor aging ........................................................................... 18
Figure 2.1. Chemical structure of cellulose .................................................... 29
Figure 2.2. Formulas of the sugar components of polyoses (hemicelluloses) ............. 30

Figure 2.3. Building units of lignin: p-coumaryl alcohol (I), coniferyl alcohol (H),
sinapyl alcohol (111) ................................................................................ 3 1

Figure 2.4. Wood flour particle length distribution ........................................... 32

Figure 3.1. Unnotched impact strength of PVC/Cu-treatedl wood flour composites with:
(a) different PVCl contents and Cu concentrations in treated, untreated wood flour (b)
three PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration
levels ................................................................................................. 40

Figure 3.2. Estimated net notched Izod impact strength of PVC/Cu-treated wood flour
composites with: (a) different PVCl contents and Cu concentrations in WF (b) three PVC
types at the ratio of PVC/wood of 60/40 and at different Cu concentration levels ........ 42

Figure 3.3. Flexural strength of PVC/Cu-treated wood flour composites with:

(a) different PVCl contents and Cu concentrations in treated, untreated wood flour (b)
three PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration
levels ................................................................................................. 46

xiii

Figure 3.4. Flexural modulus of PVC/Cu-treated wood flour composites with:

(a) different PVCl contents and Cu concentrations in treated, untreated wood flour (b)
three PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration
levels ................................................................................................. 47

Figure 3.5. Flexural toughness of PVC/Cu-treated wood flour composites with:

(a) different PVCl contents and Cu concentrations in treated, untreated wood flour (b)
three PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration
levels ................................................................................................. 48

Figure 3.6. ESEM micrographs of the impact fracture surfaces of: (a) sample 264Cu10
showing wood particles pull out (b) sample 264Cu02 showing wood particle fracture ...50

Figure 3.7. EDS surface area scan spectra of: (a) the impact fracture surface of PVC2
alone characterized by the strong Chlorine peak (b) untreated wood flour particles
exhibiting the sharp peaks of Carbon and Oxygen (c) the exposed wood particle surfaces
on the impact fracture surface of sample 264Cu0 showing the weak and small chlorine
peak ((1) the exposed wood particle surfaces on the impact fracture surface of sample
264Cu02 showing a strong and larger chlorine peak .......................................... 52

Figure 4.1. Experimental arrangement of a power compensated DSC ..................... 58

Figure 4.2. Determination of glass transition temperature on a DSC temperature scan
curve .................................................................................................. 59

Figure 4.3. DMA relationship of the complex modulus (E*), elastic modulus (E’), loss
modulus (E”), and tan 8 .................................................................................................... 60

Figure 4.4. DSC diagrams of (1) pure PVC; (2) PW64con-composite made of PVC and
untreated WF at a percentage PVC to wood ratio of 60 to 40; (3) PW64Cu0.2- composite
made of PVC and treated WF containing 0.2% Cu at a percentage PVC to wood ratio of
60 to 40 .............................................................................................. 66

Figure 4.5. Glass transition temperatures of PVC resin and PVC/Cu-treated, untreated
WF composites vs. Cu concentrations in WF ................................................... 67

Figure 4.6. ACp of PVC resin and PVC/Cu-treated, untreated WF composites vs. Cu
concentrations in WF .............................................................................. 68

Figure 4.7. DMA temperature scans of storage modulus (E’) of PVC resin and PVCIWF
composites with increased WF content .......................................................... 71

Figure 4.8. DMA temperature scans of loss modulus (E”) of PVC resin and PVCIWF
composites with increased WF content .......................................................... 72

xiv

Figure 4.9. Tan 8 thermograms of PVC resin and PVCIWF composites with increased
WF content .......................................................................................... 73

Figure 4.10. E’, E” and tan 5 of (a) PW64con: PVC-untreated wood flour composites
with a 40 wt% wood flour content. (b)PW64CuO.2: PVC-Cu treated wood flour
composite made of 40 wt% wood flour which had a 0.2 wt% Cu ........................... 75

Figure 4.11. Loss viscosity of PVC and PVC-untreated wood flour composites with 40,
50 and 60 wt% wood flour content as well as PVC-0.2 wt% Cu containing wood flour
composite with a 40 wt% wood flour content .................................................. 76

Figure 5.1. Adhesion mechanisms (a) molecule chain entanglement following
interdiffusion, (b) electrostatic attraction, (c) cationic groups at the end of molecules
attracted to an anionic surface, leading to molecule orientation on the surface, ((1)
chemical reaction and (e) mechanical keying ................................................... 82

Figure 5.2. Nail-shaped sample prepared for measuring interfacial shear strength
between wood and the PVC matrix .............................................................. 91

Figure 5.3. Wood dowel-pullout test setup ..................................................... 92

Figure 5.4. PVC solution sessile drop shape on (1) untreated oak surfaces; (2) copper
amine-treated oak surfaces ........................................................................ 93

Figure 5.5. Image of the initial contact angle of the PVC solution on untreated oak
surface ................................................................................................ 93

Figure 5.6. Image of the initial contact angle of PVC solution on 0.2 % Cu-untreated
oak surface .......................................................................................... 93

Figure 5.7. Average contact angles and drop dimensions of sessile drops of PVC solution
on untreated and Cu- treated oak surfaces ....................................................... 94

Figure 5.8. Interfacial shear strength between PVC and Cu-treated, untreated wood with
different Cu concentrations in treating solutions ............................................. 100

Figure 5.9. Interfacial shear strength between PVC and Cu-treated, untreated wood with

different Cu concentrations in treating solutions ............................................. 101
Figure 6.1. A core-level electron is ejected by a high energy X-ray photon ............. 111
Figure 6.2. FTIR spectra of PVC, red oak and copper ethanolamine ..................... 114

Figure 6.3. FTIR spectra of untreated red oak, copper amine treated red oak containing
0.2 wt% copper and copper amine treated red oak containing 1.2 wt% copper .......... 116

Figure 6.4. FTIR spectra of composites made of PVC and untreated red oak, copper
treated red oak containing 0.2 and 1.2 wt% copper .......................................... 116

Figure 6.5. Far-IR spectra of copper amine treating solutions with 0.2 and 1.0% Cu, and
their corresponding treated WF, as well as untreated WF ................................... 117

Figure 6.6. Far-IR spectra of (1) treated WF containing 1.2% Cu; (2) PVC/Cu-treated
WF composites made of 0.2 wt% Cu-treated WF; (3) PVC/Cu-treated WF composites
made of 1.2% Cu-treated WF ................................................................... 118

Figure 6.7. XPS survey spectrum of composite made of PVC and Cu-treated WF
containing 0.2 wt% copper ...................................................................... 119

Figure 6.8. Oxygen to carbon ratio, O/C of PVC, treated and untreated oak flour as well
as composites made of 60 wt% PVC and 40 wt% WF (PW64), 50 wt% PVC and50 wt%
WF (PW55) with different Cu concentrations in treated WF ............................... 121

Figure 6.9. Chlorine to carbon ratio, C1/C of PVC, and composites made of 60 wt% PVC
and 40 wt% WF (PW64), 50 wt% PVC and50 wt% WF (PW55) with different Cu
concentrations in treated WF .................................................................... 121

Figure 6.10. C1, C2, and C3 area ratios for of PVC, treated, untreated oak flour and
composites .......................................................................................... 124

Figure 7.1. Weight losses of block samples of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour after the 12 weeks decay test by Pp and IL .139

Figure 7.2. Comparison of composite surface before and after colonized by white-rot
fungus, IL, (1) composite made of 60 wt% untreated wood flour; (2) composite made of
60 wt% of Cu-treated wood flour ............................................................... 140

Figure 7.3. Weight losses of sheet samples of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour after the 12 week decay test by Pp and IL ...142

Figure 7.4. Weight losses of particle samples of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour after the 12 week decay test by Pp and IL ...142

Figure 7.5. Impact strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after decay test by Pp and IL ................... 144

Figure 7.6. Impact strength reductions of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour before and after decay test by Pp and IL . ...144

Figure 7.7. Flexural strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after decay test by Pp and IL ................... 145

Figure 7.8. Flexural strength reductions of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour before and after decay test by Pp and IL ....145

Figure 7.9. Flexural modulus of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after decay test by Pp and IL ................... 146

Figure 7.10. Flexural modulus reductions of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour before and after decay test by Pp and IL ....146

Figure 7.11. Impact strength of PVC, composites made with PVC and untreated, Cu-
treated wood flour after exposed in the environment with different relative humidity (0%
RH means properties measured after oven-dry) .............................................. 148

Figure 7.12. Flexural modulus of PVC, composites made with PVC and untreated, Cu-
treated wood flour after exposed in the environment with different relative humidity (0%
RH means properties measured after oven-dry) .............................................. 149

Figure 7.13. Impact strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after 6 months outdoor weathering ............ 151

Figure 7.14. Flexural strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after 6 months outdoor weathering ............ 151

Figure 7.15. Flexural modulus of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after 6 months outdoor weathering ............ 152

Figure 7.16. Mechanical properties reductions of PVC, oak and composites made of
PVC, CuEA-treated and untreated wood flour with various PVC to wood ratios as well as

different Cu concentrations in wood flour before and after 6 months outdoor weathering
.......................................................................................................... 152

xvii

Figure 7.17. Color changes in Aa (green/red) of PVC, oak and composites made of PVC,
untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to wood
ratios during accelerated weathering ........................................................... 155

Figure 7.18. Color changes in Ab (blue/yellow) of PVC, oak and composites made of
PVC, untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to
wood ratios during accelerated weathering .................................................... 155

Figure 7.19. Color changes in AL (black/white) of PVC, oak and composites made of
PVC, untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to
wood ratios during accelerated weathering .................................................... 156

Figure 7.20. Total color changes in AB of PVC, oak and composites made of PVC,
untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to wood
ratios during accelerated weathering ........................................................... 156

Figure 7.21. Pathway of formation of quinonoid in wood lignin during
photodegradtion ................................................................................... 158

Figure 7.22. Photodegradation of PVC resulting in the formation of HCl ............... 159

xviii

Chapter 1
INTRODUCTION TO THE DEVELOPMENT OF WOOD AND NATURAL
FIBER-PLASTIC COMPOSITES

There are two kinds of definitions about WPC, Wood Polymer Composites and
Wood Plastic Composites. Wood polymer composites are manufactured by impregnating
wood with monomers or prepolymers (Schneider 1994, 1995). Polymerizations can be
initiated by gamma radiation, chemical catalyst or heat reaction to form high molecular
weight polymer and wood composites (Meyer 1965, Wang 1975, Hebeish and Guthire
1981, Witt 1981, Schneider 1994). The main application of this composite is flooring
(Meyer 1977, 1981, 1987).

Wood plastic composites include wood thermosetting composites and wood
thermoplastic composites. Techniques of producing wood thermosetting composites have
been well established since 19003 (Klason 1986, Clemons 2002). PF (phenol
formaldehyde) and UP (urea formaldehyde) are the most common thermosetting resins
used for wood composite products, such as plywood, particleboard, fiberboard and
oriented strand board (OSB) (Kamdem 199la,b,c, 1992a,b, 1993, 1994, Hiziroglu 1995,
Muson 1998).

Wood thermoplastic composites are manufactured by dispersing solid wood
fibers/powder, or wood flour into the molten plastic to form a composite by processing
techniques, such as extrusion, therrnoforrning, and compression or injection molding
(Bledzki 1999). Although wood thermoplastic composites occurred in the United States
since 1980s, it is in the recent years that they have been experiencing a dramatic growth

(Ford 1999, Clemons 2002). WPCs take advantages of low density, low equipment

abrasiveness, relatively low cost, biodegradability from the wood fiber (flour) and the
moisture resistance, dimensional stability from the polymer matrix (Glasser et al. 1999,
Clemons 2002). In addition, the recycling property of plastics has been improved and
various surface optical effects can be obtained by adding different wood species and colored
pigments (Bledzki 19983). It is forecast that the demand for natural fiber/plastic composites
will grow to about 60% per year for building construction products and 50% per year in
automotive applications. Seventy five percent of this demand comes from wood
thermoplastic composites (Kline & Company 2001).

In spite of the advantages, the use of wood in thermoplastics has been plagued by the
thermal stability limitation of the wood and plastics, and the difficulties in obtaining good
filler dispersion and strong interfacial adhesion (Kokta et al. 1986, Bledzki et al. 1998b).
This is because of the natural incompatibility between the hydrophilic, polar wood-fibers
(high surface energy) and hydrophobic, non-polar thermoplastics (low surface energy)
(Kazayawoko et al. 1997a,b, Matuana et al. 1997, 1998a,). Such a phase incompatibility
causes a weak interface between the wood filler and the matrix. Moreover, strong wood-
wood interactions resulting from hydrogen bonding and physical entanglement impair the

dispersion of the fillers in the viscous matrix (Kazayawoko et al. 1999, Lu et al. 2000).

1.1 Markets and Manufacturers

In the last 15 years, natural organic fillers (such as wood-fibers/flour) and other
lignocellulosic fibers (such as hemp, flax, kenaf, jute, sisal, rice hulls, etc.) have slowly
penetrated into North America and global markets and have experienced a dramatic growth

(Clemons 2002, Smith 2001). As shown in Figure 1.1, the prediction of North American

demand for natural fiber/plastic composites was 700 million lb. in 2001, over 875 million
lb. in 2002 and is expected to reach 1.3 billion lb. by 2007 with continued rapid growth
through 2010 (Kline & Company 2001, Clemons 2002, Principia Partners 2003). More
than seventy five percent of that demands is for wood fiber/flour plastic composites
(Principia Partners 2003). The driving force lies on the advantages of WPC products;
they are described as environmental friendly, recyclable, rot and insect resistance and
they need limited routine maintenance (Clemons 2002). In addition, wood fiber/flour is
a relatively low-cost filler for plastics. The average market prices of wood fiber/flour and
other fillers are illustrated in Figure 1.2 Wood fiber/flour is less than one-tenth of the
price of glass fibers and is comparable to other low-cost inorganic fillers, such as talc and

calcium carbonate (Kline & Company 2001).

Ammonlb
14001

 

 

1200*

 

1000

 

800

 

600

 

400*

 

Y\‘\‘\‘Y\\

200 ‘

 

 

           

 

   

2002 2007
CJExpmned

1980 1990 2000 2001

Figure 1.1. Market demand for natural fiber/plastic composites in North America

Compared to wood and treated wood lumber, an additional processing cost is
involved in the manufacture of wood plastic composites, which render their current initial
selling prices higher than that of equivalent solid wood lumber. But the
annualmaintenance cost is considerably limited, so that in the long term, wood plastic

composite products is a better choice (Principia Partners 2003).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Prlce Ilb
(S/) f f #iMO.
0.80- .

D Organic Fillers
0-7°‘/ Inorganlc Flllers :
0.60/ i
0.50 / ,
0.40/ i

./ 7” H
0.30 so 19 $0.22
0.20 -/ ,1
o_1o./ $0.05 - $0.08 f
0.00- ' . , ,
Talc Wood Calelum FlaxIJute Hemp/Kent Glass fibers

fiber/flour carbonate

Figure 1.2. Average market prices of wood fiber/flour and other fillers and
reinforcements

The most commonly used thermoplastics in manufacturing WPCs are polyethylene
(PE), polypropylene (PP), polyvinyl chloride (PVC) and post consumer polymeric
recycled materials (Clemons 2002, Bledzki et al 1998a). The North American market
share of PE-based wood composites is reported to be around 83%, 9% for PVC, 7% for

PP based wood composites in 2002 as illustrated in Figure 1.3 (Principia Partners 2003).

’///f ‘ I

 

Figure 1.3. North American market shares for wood/plastic composites with different
polymers (Principia Partners 2003)

As far as applications are concerned, PE and PVC based WPCs are used to
manufacture mostly building construction products, such as decking, railings, windows
and doorframes. Most PE-based WPCs are used for decking, which contributes about
50% market of the total WPC’s applications, as shown in Figure 1.4. PVC-wood
composites are also used for decking, but they are more often used for windows and
doors, and represent 22% of the market. Automotive applications, which are 14% of the
market share, are dominated by PP-wood composites (Clemons 2000, 2002). Building
construction products are forecast to grow 60% per year while automotive applications

are reported to grow 50% per year from 2000 to 2005 (Kline & Company 2001).

Automotive
14%

 

 

    

Windows/Door
s
22%

Figure 1.4. WPC application markets in North America (Clemons 2000)

Although PE is the dominant plastic used in WPC by far and will remain on its
leading position till 2005, it is also forecast that there will be a dramatic growth for PVC-
wood plastic composites during the next decade (Principia Partners 2003). It is forecast
that the demand for PVC-wood composites will increase about 200% from 2003 to 2010,
130% for PP—wood composites and 40% for PE-wood composites (Principia Partners
2003). PVC-based wood composite is used in building construction application because it
offers acceptable mechanical properties, chemical and water resistance, rot resistance,
stain and paint-ability as well as long lifetime (Patterson 2000). In addition, PVC-wood
composites can be cut, sawed, nailed, screwed, and processed by the conventional
manufacturing equipment (Bledzki et al. 1998a, b).

It is reported that the first commercial WPC product was a PVC/wood composite
flooring tiles, which appeared in the building decoration market around 19503. In 1973, a
Swedish company, Sonesson Plast AB. (MalmO, Sweden), developed a PVC/wood flour
composite product under the registered trade mark of “Sonwood”. Sonwood was used as
a substitute for high grade facing veneer (SW 1419055). In 1993, Andersen Corporation
(Bayport, Minnesota) patented a method to manufacture PVC/wood fiber composite
structural members for windows and doors by using extrusion and injection molding
(Clemons 2000, Schut 1999). CertainTeed Corporation (Valley Forge, PA) started to sell
PVC/wood composite window profiles in 1998. CertainTeed product line for PVC/wood
composite “Boardwalk” decking and railing have received the approval of NES (the
National Evaluation Service) since October 2001 (Schut 1999, CertainTeed Corp. 2001).
Currently, most of the PVC/wood composites are used for window/door profiles,

decking, railing and siding. Interior and marine applications are under development for

PVC/wood composites (Koenig et al. 2002, Malvar 2000). The main PVC/wood

composite manufacturers in North America are listed below in Table 1.1.

Table 1.1. North American PVC-Wood Composite Manufacturers (Schut 1999)

 

 

 

 

 

 

 

 

 

 

 

 

Name Location Product Names Product Applications
Andersen Corporation Bayport, MN Fibrex, Renewal Windows/doors
CertainTeed. Corporation Valley Forge, PA CertaWood Window profiles
CertainTeed Corporation Jackson, MI Boardwalk Decking, railing

Comptrusion Corporation Toronto, Canada ----- Window/door profiles

Crane Plastics Columbus, OH Timber’I‘ech Decking
Forrntech Enterprises Stow, OH ----- Custom profiles
Hoff Forest Products Caldwell, ID Peachtree Windows/doors
Mikron Industries Kent, WA MikronWood Windows, decking
Re-New Wood, Inc. Wagoner, OK Eco-shake Roofing shakes, shingles

Strandex Corporation Madison WI Strandex License, profiles

 

 

 

 

1.2. Formulations and Additives

 

PVC requires the use of several additives to facilitate its process (Nass 1977, Gomez
1984, Titow 1984). Heat stabilizers, processing aids, impact modifiers, lubricants, and
pigments are needed as additives in PVC-wood composites to maintain the intrinsic
properties of the PVC matrix, to facilitate the processability and to enhance the
mechanical properties of the composites (England and Rutherford 1999, Clemons 2002,

Mengeloglu et a1 2000, Matuana 1997).

Lubricants, such as metallic stearates, fatty acids and waxes, are used in PVC-wood
composites to improve wood dispersion, and increase the processing output of extrusion
(Elsevier Advanced Technology 2002). Strukto (2002) reported that their newly
developed lubricant TWP-012, a blend of oleo chemicals and waxes, was specifically
designed for rigid PVC-wood composite extrusion (Elsevier Advanced Technology
2002), and can provide good overall processing characteristics and significantly inhibit
edge tear and melt fracture when combining with suitable impact modifiers (Struktol
2000). The recommended adding amount is about 2 to 8 parts in PVC-wood composites
depending on formulations and processing conditions (Struktol 2002).

Chlorinated polyethylene (CPE), ethylene vinyl acetate (EVA), methacrylate-
butadiene-styrene (MBS) and all-acrylic elastomer (ACR) are major impact modifiers for
rigid PVC (Gomez 1984). Their influences on the mechanical properties of rigid PVC-
wood composites have been investigated (Mengeloglu et a1 2000). The impact strength of
PVC-wood composites can be significantly enhanced by the presence of impact modifiers
and MBS and ACR showed better impact improving effects than CPE (Mengeloglu et a1
2000)

Although plasticizers are crucial for flexible PVC, they can only be used as the
processing aids in rigid PVC in low concentrations due to their deleterious effects on the
mechanical properties of the final products (N ass 1977). It has been reported that the
addition of 15 phr dioctyl phthalate (DOP) resulted in the significantly decreases in
tensile strength, elongation and toughness in PVC/newsprint-fiber composites (Matuana

1997).

Wood is a natural polymer material composed mainly of cellulose, hemi-celluloses,
lignin and extractives, and all of them prone to degrade under environmental ultraviolet
radiation resulting in wood discoloration (Hon and Shiraishi 1991). Due to the presence
of wood, PVC- wood fiber composites experienced a more severely discoloration than
did unfilled PVC when exposed to the UV light (Matuana et a1 2001). The color stability
of PVC-wood composites can be effectively improved by using PVC colorants, such as
titanium dioxide (TiOz), in the composite formulations (Matuana et a1 2001).

Although PVC resin itself is not considered to be susceptible to microbial attack,
antimicrobials might be necessary in formulations for some plasticized or filled PVCs,
especially when the filler is WF (Nass 1977, Elsevier Advanced Technology 2002).
Antimicrobial types and dosages in the composite formulations depend on the wood fiber
type and content, microbe types in the application environment and the environmental
humidity (Elsevier Advanced Technology 2002). It is reported that zinc borates were
useful for preventing rot but not effective for algae growth, whereas isothiazolones, such
as Rohm & Haas’s Vinyzene series biostabilizers and Irgaguard F 3000 antimicrobial
manufactured by Ciba Company are said to be effective against various moulds, fungi

and other microorganisms (Elsevier Advanced Technology 2002).

1.3 Coupling Agents and Compatibilizers

The development of wood-plastic composites is limited by the compatibility issue
arisen between the hydrophilic wood and the hydrophobic plastic matrix (Lu et al. 2000,
Bledzki et al. 1998a, Kokta et al. 1986). The incompatibility prompts a poor interfacial

adhesion between wood and the plastic matrix, and reduces mechanical strengths and

ductility (Oksman and Clemons 1998c, Raj et al. 1992, 1989a,b). Many efforts have been
carried out on the improvement of the interfacial adhesion between wood and polymer
matrix by using coupling agents and compatibilizers (Lu et al. 2000, Maldas et al.
1989a,b, 1988, Oksman and Clemons 1998b,c).

There is an approximately 46% loss on flexural strength and 60% loss in Izod impact
sthength were observed in PVC/wood flour composites compared with the neat rigid PVC
(Patterson 2000, Mengeloglu et al. 2000). For PVC-wood composites, coupling effects of
poly[methy1ene(polyphenyl isocyanate)(PMPPIC), silanes, maleic anhydn'de(MA),
maleated polypropylene(MAPP) and linoleic acid etc. have been examined (Maldas and
Kokta 1989a, Kokta et al. 1990a,b, Maldas et al. 1989c, Mengeloglu et al. 2000, Lu et al.
2002).

Silane is an important coupling agent in the filled plastics industry that used for bonding
the inorganic filler and the polymer matrix (Lu et al. 2000, Bledzki et al. 1998a, Kokta et
al.1990b). Kokta et al. investigated the coupling effects of various types of silanes in
PVC/wood flour composites (Kokta et al.1990b). Generally, silane can be represented as
Y(CH2)nSi(OR)3, where the functional group Y can be amino (Silane A-1100),
methacryloxy (A-174) or epoxy (A-186, A-187) (Kokta et al.1990b, Bledzki et al. 1998).
The alkoxy groups (-OR) can react with the environmental moisture to become silanols
(Kokta et al.1990b, Bledzki et al. 1998a):

Y—(CH2)n—Si(OR)3 + H20 ---> Y— (CH2)n—Si(OH)3
The -OH group side of silanols can connect to the wood surface through hydrogen bonds
formed with the hydroxyl groups of wood. The other side of silanols, the functional groups

Y adhere to PVC matrix by weak Van der Waals forces (Bledzki et al. 1998a) so that strong

10

interfacial adhesion cannot be established between wood and PVC (Maldas et al. 1988).
When Silane, A-1100 was used alone, no improvement in mechanical properties was
observed compared with untreated wood pulp/PVC composites (Kokta et al. 1990b).
Mechanical properties of composites can be increased if adding initiators, such as peroxides
in silane coupling system. This is because the strong chemical bonding can be built between
the functional groups (Y) of silanes and PVC by thermally induced free radical reactions
(Bledzki et al. 1998a, Matuana et al. 1998b). In the presence of lauroyl peroxide, maleic
anhydride and the dispersion solvent, cyclohexanone, the tensile elongation of the resulting
composite increased 15%, on the contrary, 56% decrease was obtained when using
methanol and water as the dispersion solvent alone (Kokta et al. 1990b). Therefore, the
performance of silane coupling agents strongly depended on the type and combination of
dispersion solvents and interfacial reaction initiators (Bledzki et al. 1998a).

Some polymeric coupling agents were widely examined for their effectiveness in the
wood/plastic composites, and they serve more like compatibilizers between the wood phase
and the plastic matrix (Lu et al. 2000, Bledzki et al. 1998a). Poly[methylene(polyphenyl
isocyanate) (PMPPIC) has been proved to be a very useful polymeric coupling agent for
PVC/wood composites (Maldas and Kokta 1989a, Maldas et al. 1989c). The functional
group —N=C=O of isocyanate can build covalent bonds with the -OH group of wood fibers

by chemical reactions to form urethane links:

T O
WN=C=O + HO—-WOOD —>vau~—N-—l‘L—O—WOOD

11

and polymeric chain on the other side of PMPPIC has good compatibility with PVC matrix
(Bledzki et al. 1998a, Maldas et al. 1989c). Table 1.2 indicates that mechanical properties,
such as impact strength, tensile strength, elongation and toughness were significantly

improved by using PMPPIC (Maldas and Kokta 1993).

Table 1.2. Effects of Coupling Agent PMPPIC on the Mechanical Properties of
PVC/Wood Flour Composites (wood sawdust content was 25 wt%)3

 

 

 

 

 

 

 

 

Sawdust (Aspen) Sawdust (Spruce)
Untreated Treated with 8% Untreated Treated with 8%
pme b . 10% PPMIC + 10% PVC
PVC
Unnotched Izod 7.1 9.5 7.2 8.9
Impact Strength (J/m) (1.03) c (1.04) (1.07) (1.22)
Ultimate Tensile 24.2 45.9 23.3 47.6
Strength (MPa) (2.30) (1.17) (2.38) (1.93)
Ultimate Elongation 1.6 2.7 1.5 2.7
(%) (0.14) (0.10) (0.17) (0.25)
Tensile Toughness 0.22 0.66 0.20 0.73
(MPa) (0.03) (0.03) (0.03) (0.03)
Young’s Modulus 2.1 2.4 2.2 2.4
(GPa) (0.19) (0.10) (0.14) (0.24)

 

 

 

 

 

 

a Source: Data from Maldas and Kokta 1993
b Percentages of weight of the wood sawdust
0 Standard deviation of the properties listed in the parentheses

It was reported that about 20% increase in shear strength was obtained when maleated
polypropylene (MAPP) treated wood veneer was used to manufacture PVC/wood laminate
composites in the presence of a benzoyl peroxide initiator (Lu et al. 2002). The performance
of MAPP was less effective in the absence of the reaction initiators (Matuana et al. 1998a,
b).

Guffey and Sabbagh reported their efforts in using chlorinated polyethylene (CPE) as a

compatibilizer for PVC/wood composites system (Guffey and Sabbagh 2002). They

l2

examined CPBs with different molecular weights, crystallinities, and chlorine contents.
Significant improvements in melt strength, elongation to break and surface quality were
found when using CPE with 25% chlorine content, higher molecular weight and low
crystallinity (Guffey and Sabbagh 2002).

Another way to introduce the coupling structure into the wood-PVC system is
through grafting reactions (Bledzki et al. 1998a, Kokta et al. 1990a). Tensile elongations
and breaking energies of PVC/wood flour composites can be improved by H202 initiated
direct grafting reactions which result in the cross-linking between cellulose and wood

fiber (Kokta et al. 1990a).

1.4 Hybrids

The properties of wood-plastic composites can also be enhanced by combining wood
with other fillers to form hybrid reinforcements, by which the advantages of
reinforcements can be maximized and the deficiencies can be minimized (Maldas and
Kokta 1993, Kretsis 1987). The “hybrid effects” of mica, glass fiber and modified wood
pulp have been investigated in PVC-wood composite systems (Maldas and Kokta 1992a,
1993, Takatani et a] 2000, Jiang and Kamdem et al. 2003).

Maldas and Kokta reported that the modulus of PVC/wood composites could be
improved 20 to 30 % by adding mica as the hybrid filler (Maldas and Kokta 1992a). The
mechanical properties of the hybrid composites are strongly influenced by the types of
surface chemical modifications of wood fibers, and the ratio of wood fiber to mica in
hybrid fillers (Maldas and Kokta 1993). Mica combined with isocyanate-coated wood

fibers showed a better hybrid effect on the mechanical properties than combinations of

13

mica with maleic andydreide (MA) or Na-silicate coated wood fibers (Maldas and Kokta
1993).

Glass fiber has been shown to be a good hybrid for PVC/wood composites (Maldas and
Kokta 1992b, c, Jiang and Kamdem et al. 2003). Notched and unnotched impact strength of
PVC/wood flour composites can be strongly enhanced by adding short glass fibers as a
hybrid reinforcement without losing flexural properties (Jiang and Kamdem et al. 2003). An
increase of about 60% in notched impact strength can be achieved by using 5% short glass
fiber with a fiber length of 6.4 mm (with the aspect ratio of about 450) short glass fiber at a
PVC content of 55%, as shown in Figure 1.5 (Jiang and Kamdem et al. 2003). The

significant improvement in impact strength of hybrid composites was attributed to the

70.0

 

§

 

 

 

 

8
o

S
o

 

 

 

 

8
o

 

Notched Impact strength (Jim)
8
O

.3
.0
o

 

 

 

 

 

 

 

 

Figure 1.5. Notched impact strength of PVC/wood flour/glass fiber composites with
different glass fiber types, type IL. glass fiber contents and PVC loading levels

PW54: PVC: wood + fiber glass = 55:45; PW55: PVC: wood + fiber glass = 50:50;
PW46: PVC: wood + fiber glass = 40:60

D 0m E 01.5% 085% m0L2% 0L10% ‘\\\\ 01.20%

14

formation of a three-dimensional network architecture between glass fibers and wood flour
(Jiang and Kamdem et al. 2003).

T akatani et al. reported that excellent fractural strength and water resistance properties
could be achieved in PVC/wood flour composites when replacing some parts of wood flour
by steam-exploded (SE) wood flour (Takatani et al. 2000). Flexural strength of PVC/wood
flour composites was increased 30 to 40 % by using 20 % SE wood and 50 % wood flour
instead of adding 70 % wood flour alone (Takatani et al. 2000). Wood degradation and
decomposition happened during high temperature steam treatments so that the molecular
weight of wood was reduced and its hydrophilic properties could be weakened due to the
dehydration (Takatani et al. 2000, Hon and Shiraishi 1991). It was expectative that the
interphase between PVC and wood flour would be improved by the presence of SE wood

(T akatani et al. 2000).

1.5 Performances and Characteristics

Typical mechanical and physical properties of PVC/WF composites are listed in
Table 1.3 in comparison with natural wood and other wood composites. Flexural strength
and modulus of plastic/WP composites are lower than those of natural wood but
comparable with the properties of wood composites, such as oriented strandboard (OSB)
and medium-density fiberboard (MDF). PVCIWF composite is superior to OSB in
flexural strength and better than that of MDF in both flexural strength and modulus.

Although the specific gravities of plastic/WF composites are higher than that of natural
wood, OSB and MDF, they have much smaller water absorption values (Table 1.3). This

means they are less affected by water related problems, such as fungal growth, thickness

15

swell and shrinkage. Plastic/WP composites generally exhibit better dimensional stability

and fungal resistance than natural wood (Clemons 2002). PVCIWF composites are superior

to HDPE, PP/WF composites with respect to their higher modulus, good creep resistance,

weatherability and flame retardance.

Table 1.3. Comparison of mechanical and physical properties for PVC/wood fiber (WF)
composites with wood and other wood compositesa

 

 

 

 

 

 

 

Properties Woodb Oriented Medium- PVCIWF HDPE/WF PPIWF
and ASTM Oak Pine strandboard densrty Compositesd composites composites
rubm) resinosa) (MDF)

Flexural 96.0 76.0 20.7-27.6 34.5 42.0 19.6 47.9
strength
(MPaXD790)

Flexural 11.3 11.2 4.8-8.2 3.5 5.2 3.8 3.3
modulus
(GPa)(D790)

Specific 0.61 0.46 0.64 0.64-0.80 1.20 1.12 1.05
gravity

(D143)

Water 38.9 70.5 16.5 21.2 1.3 0.7 1.1
absorption

%)(D1037) ‘

 

 

 

 

 

 

 

 

 

a Data sources: FPL, USDA Forest Service, 1999; Xu et a1. 1996; England and Rutherford 1999;
Bledzki et al 1998a.

b Properties at 12 % moisture content.

c . . . .
Medium densrty-exterror adhesrve panels.

d Properties of plastic/WP composites were obtained at a 40 wt% WF loading level.

Mechanical properties of PVCIWF composites are influenced by the WP content and

WP particle size distribution. Within 50 wt% WF loading level, increasing WF content

enlarges modulus but reduces the stress at break (Djidjelli et al. 2002, Jiang and kamdem

2003). PVC can be filled with 40 to 70 wt% WF depending on the ability of the

processing equipment (Brown 2000). WP with a mesh size range from 40 to 80 mesh was

16

 

found to facilitate processing and developing better properties (Patterson 2000). Wood
fiber is different from wood flour in terms of the fiber aspect ratio, the ratio of length to
diameter (l/d). An aspect ratio of 1:1 to 4:1 is for wood flour, whereas 10:1 to 20:1 for
wood fiber. The higher the aspect ratio the larger the modulus, but impact strength may
be reduced (Patterson 2000).

Creep fatigue behavior of PVCIWF composite material is strongly affected by the
applied load, loading time and environmental temperature. It was reported that creep of
PVCIWF composites was greatly increased when the temperature was slightly above
ambient temperature (Sain et al. 2000).

100 and Cho investigated the rheology of PVCIWF composites as well as the effect
of plasticizer and impact modifier (Joo and Cho 1999). They found that, at low shear
rates, shear viscosity of PVC was dramatically increased by the presence of WF, but at
high shear rate region, this effect was almost disappeared because WF oriented along the
PVC flow direction under high shear force. Plasticizer has the ability to decline shear
viscosity of PVCIWF composites, but this trend turned to be compromised by the further
addition of the impact modifier into the plasticized composite system (100 and Cho
1999).

Thermal and dielectric performances of PVCIWF composites were examined by
Djidjelli et al. They reported that the presence of WF had little effect on the glass
transition temperature of PVC, but had tendency to prevent PVC matrix from thermal
degradation (Djidjelli et al. 2002). Perrnittivity and dielectric losses of PVCIWF
composites are higher than those of PVC resin because of introduction of dipoles of

hydroxyl groups from cellulose and hemicelluloses of wood (Djidjelli et al. 2002).

17

Ultraviolet light resistance and weathering dimensional stabilities of PVCIWF
composites are much better than solid wood (Chetanachan et al. 2001). As shown in Fig.
1.6, after 12 month outdoor aging, Color change (AE) of the PVC/WF composite was one
third of solid wood and its thickness change was only one fourth of wood (Chetanachan
et al. 2001). Photo stability of PVCIWF composites can be further improved by adding

rutile titanium dioxide colorant in composite formulations (Matuana et al. 2001).

I L J
T I

Color change (AE)
8 a: 8
«r
+ I—
J-v—r-v-l-fleulerw—Ffi ' fl—arrfififi—H
-b O) (D
Thickness change (%)

888
8

N

 

O 01
L l 1

 

O

1 2 3 4 5 6 7 8 12
Aging time (month)

+ Wood color change + PVCIWF color change
-0— Wood thickness change —A— PVCIWF thickness change

Figure 1.6. Color and thickness change of PVCIWF composite and natural wood during
12 month outdoor aging. (Chetanachan et al. 2001).

Acid-base characteristics of plasticized PVC/newsprint—fiber composites were

investigated by using inverse gas chromatography (10C) technique (Matuana et al,
1998b, 1999). Fiber-matrix specific interaction parameter, ISp was calculated based on the

acid and base constants, K2, and KB of newsprint-fiber and PVC (Matuana et al. 1998b).

The specific interaction parameter was influenced by the type of fiber surface

18

modifications and strongly correlated with the tensile strength of the resulting composite

(Matuana et al. 1998b).

1.6 Background and Objectives

As applications and productions of WPC increase, new issues occur in this field.
Currently, for a standard 0.03 by 0.15 by 3.66 m decking board, the price doubles when
choosing wood-plastic composite boards instead of preservative pressure-treated boards.
Biological and environmental issues which have been bothering wood for a long time,
also affect wood-plastic composite products. Those factors includes fungus decay, mold
growth, sapstain discoloration, migration of tannins, and other colorful extractives to
wood surfaces, stains caused by dropping BBQ sauces, grill oil, weathering discoloration,
and photodegradation (Simonsen et al. 2004, Matuana and Kamdem 2002, 2001,
Pendleton et a1 2002). This finally initiated a nationwide lawsuit against a leading
manufacturer in America regarding the misrepresention of rot resistance, biological and
environmental durability of their products (PRNewswire 2004).

In order to improve biological performance of wood-plastic composites, 3 ~ 4% zinc
borate, DDAC (didecyldimethylammonium chloride), copper-based wood preservatives,
are recommended (Simonsen et al. 2004, Pendleton et a1 2002). Copper-based wood
preservatives are preferred because they are not only good fungicides but also moldicides
(Kamdem and Zhang 2001).

In this study, copper ethanolamine (CuEA) complex was used as a coupling agent for
wood-PVC composites. Coupling agents are characterized by one side of this chemical

reacts with wood filler and another side is connected to polymer matrix. Copper amine is

19

the main component of copper-based wood preservatives. After copper amine
treatments, strong chemical bonding may build up between wood and the copper complex
resulting in a stable copper fixation in wood. On the other hand, amine and hydroxyl
functional groups can react with PVC matrix during PVCIWF composite manufacturing
to form strong connections between copper amine and PVC. As a result, a strong
interfacial adhesion and an improved interphase between wood and PVC are built up,
which leads to better mechanical and physical performance. Additionally, increased
thermal conductivity of copper-treated wood flour facilitated heat diffusion, adequate
flow, and melting of PVC during processing. Wetting between PVC and wood surfaces
was also improved, which modified the dispersion, polar as well as acid, base
characteristics of wood surface energy.

PVC/copper-treated wood flour composites not only have improved mechanical
properties, but also better biological decay (insect, fungus and mold) resistance, especially
when the composites contain 60% or more wood flour. In this study, the influences of
copper treatment on mechanical, thermal properties, and environmental stability were
evaluated. Reasons attributed to the improvement on the mechanical properties of copper
treated PVC/WF composites were investigated by the analysis of surface and interphase

physical and chemical characteristics between copper-treated wood flour and PVC matrix.

20

References

Bledzki, A. K., Reihmane, S., and Gassan, J. 1998a. Thermoplastics reinforced with
wood fillers: A literature review. Polym. Plast. Technol. Eng. 37(4): 451-468.

Bledzki, A. K., Gassan, J ., and Theis, S. 1998b. Wood-filled thermoplastic composites.
Mechanics of Compos. Mater. 34(6):563-568.

Bledzki, A. K., and Sperber, V. E. 1999. Recent developments in wood-filled plastic:
United States, Japan, and Europe. In: Proc. Fifth International Conference on Woodfiber-
Plastic Composites. Forest Prod. Soc., May 26-27, Madison, WI. pp. 187-192.

Burgess, R. H., editor, 1982. Manufacture and Processing of PVC. Applied Science
Publishers Ltd, Barking, England.

CertainTeed Corporation. 2001. CertainTeed Boardwalk Composite Decking and Railing
Receives NES Approval. Company News. October 10, 2001.

Chetanachan, W., Sookkho, D., Sutthitavil, W., Chantasatrasarny, N., Sinserrnsuksakul,
R. 2001. PVC wood: a new look in construction. Journal of Vinyl & Additive
Technology. 7(3):l34-137.

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

Clemons, C. 2000. Woodfiber-plastic composites in the United State-History and current
and future markets. In: 3rd International Wood and Natural Fiber Composites
Symposium. Kassel, Germany. September 19-20, 2000.

Djidjelli, H., Martinez-Vega, J. J ., Farenc, J., and Benachour D. 2002. Effect of wood flour
content on the thermal mechanical and dielectric properties of poly(vinyl chloride).
Macromol. Mater. Eng, 287:611-618.

Elsevier Advanced Technology. 2002. Additive developments aid growth in wood-plastic
composites. SpecialChem: Polymer Additives & Colors. December 11, 2002.

England, A., and Rutherford, R. 1999. Strandex: A wood -thermoplastic composite. In:
Proc. 33“: International Particleboard/Composite Materials Symposium; Washington
State University; Pullman, WA. April 1999. pp. 127-31.

Ford, M. 1999. Research needs of the woodfiber-plastic composites marketplace. In:
Proc. Fifth International Conference on Woodfiber-Plastic Composites. Forest Prod. Soc.,
May 26-27, Madison, WI. pp. 199-201.

Forest Products Laboratory, USDA Forest Service. 1999. Wood Handbook-Wood as an
Engineering Material. Madison, WI. USA.

21

Glasser, W. 0., Taib, R., Jain, R. K., and Kander, R. 1999. Fiber-reinforced
cellulosictherrnoplastic composites. J Appl Polym Sci. 73: 1329-40.

Gomez, 1. L., editor, 1984. Engineering with Rigid PVC: Processability and Applications.
Marcel Dekker, Inc., New York and Basel, New York.

Guffey, V. O. and Sabbagh, A. B. 2002. PVC/wood flour composites compatibilized with
chlorinated polyethylene. SPE ANTEC 2002, San Francisco, CA. USA. May 5-9, 2002.

Hon, D. S., and Shiraishi, N. 1991. Wood and Cellulosic Chemistry. Marcel Dekker, Inc.,

Hebeish, A., and Guthrie, J. T. 1981. The chemistry and technology of cellulosic co-
polymers. Polymer/Properties and Applications 4. Spring-Verlag, NY. 351p.

Hiziroglu, S., and Kamdem, D. P. 1995. Physical and mechanical properties of hardboard
made of black locust furnish. Forest Prod. J. 45(1 1/ 12): 66-70.

Jiang, H., Kamdem, P. D., Bezubic, B. and Ruede, P. 2003. Mechanical Properties of
Poly(Vinyl Chloride)/Wood Flour/Glass Fiber Hybrid Composites. J. Vinyl Additive
Techno.

Joo, Y. L., Cho, M. H. 1999. Utilization of rheology control to develop wood-grain
patterned PVC/wood flour composites. Intern. Polymer Processing XIV 1: 10-20.

Kamdem, D. P., and Zhang, J. 2001. Identification of cupric and cuprous copper in
copper naphthenate-treated wood by X-ray photoelectron spectroscopy. Holzforschung.
55(1): 16-20.

Kamdem, DP, and Sean, T. 1994. The durability of phenolic-bonded particleboards
made of decay-resistant black locust and nondurable aspen, Forest Prod. J. 44(2): 65-68.

Kamdem, D.P., Bose, S. K., and Luner, P. 1993. IGC Characterization of Birch Wood Meal,
Langmuir 9(1 1): 3039-3044.

Kamdem, D. P., and Riedl, B. 1992a. Estimation of the dispersive component of surface
energy of polymer-grafted lignocellulosic fiber with inverse gas chromatography, J.
Adhesion Sci. Technol., 6 (9): 1053-1067.

Kamdem, DP, and Riedl, B.l992b. Inverse gas chromatography of lignocellulosic fibers
coated with a thermosetting polymer: Use of peak maximum and Conder and Young
methods. J. Colloid Interface Sci., 150(2): 507-516.

Kamdem, DP, and Riedl, B. 1991a. Characterization of wood fibers modified by
phenol-forrnaldehyde, Colloid Polym. Sci., 269 (6): 595-603

22

Kamdem, D.P., B. Riedl, S.C. Kaliaguine and A. Adnot.l991b. ESCA Spectroscopy of Poly
(methylmetacrylate) Grafted onto Wood fibers, J. Appl. Polym. Sci. 43: 1901-191 1.

Kamdem, DP. and Riedl, B.l991c. IGC characterization of PMMA grafted onto CTMP
fiber, J. Wood Chem. Technol., 11 ( 1): 57-91.

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

Kazayawoko, M., Balatinecz, J. J., Woodhams, R. T., and Law, S. 1997a. Effects of
wood-fiber surface chemistry on the mechanical properties of wood-fiber-polypropylene
composites. Intern. 1. Polymeric Mater. 37: 237-261.

Kazayawoko, M., and Balatinecz, J. J. 1997b. Adhesion mechanisms in woodfiber-
polypropylene composites. In: Proc. 4th International Conference on Wood fiber-Plastic
Composites, Forest Products Society. Madison, WI, USA, May 12-14. pp. 81-93.

Klason, C., and Kubat, J. 1986. Cellulosic fillers for thermoplastics. Polym. Compos.,
Proc., 28th Microsymp. Macromol., p.153-65. Publisher: De Gruyter, Berlin, Germany.

Kline & Company. 2001. North American demand for plastic fillers and reinforcements,
2000. 6th International Conference on Wood fiber-Plastic Composites, Madison, WI,
USA, May 2001.

Koenig, K.M., and Sypkens, CW. 2002. Wood-plastic composites vie for market share.
Wood & Wood Products. April 2002.

Kokta, B. V., Maldas, D., Daneault, C. and Beland, P. 1990a. Composites of polyvinyl
chloride-wood fiber. Part II: Effect of chemical treatment. Polym. Composites. 11(2): 84-
89.

Kokta, B. V., Maldas, D., Daneault, C. and Beland, P. 1990b. Composites of polyvinyl
chloride-wood fiber. HI: Effect of silane as coupling agent. J. Vinyl Tech. 12(3): 146-
153.

Kokta, B. V., Kamdem, P. D., Beshay, A. D., and Daneault, C. 1986. Use of wood fibers
in thermoplastic composites, III. Polymethylmethacrylate. Polym. Compos. 28:251-67.

Kretsis, G. 1987. A review of the tensile, compressive, flexural and shear properties of
hybrid fiber-reinforced plastics. Compos. 18(1): 13-23.

Lu, J. 2., Wu, Q., and Jr. McNabb, H. S. 2000. Chemical coupling in wood fiber and

polymer composites: A review of coupling agents and treatments. Wood Fiber Sci. 32(1):
88-104.

23

Lu, J. 2., Wu, Q., and Negulescu, I. I. 2002. The influence of maleation on polymer
adsorption and fixation, wood surface wettability, and interfacial bonding strength in
wood-PVC composites. Wood Fiber Sci. 34(3): 434-459.

Maldas, D., and Kokta, B.V. 1993. Performance of hybrid reinforcements in PVC
composites. H: Use of surface-modified mica and different cellulosic materials as
reinforcements, J. Vinyl Technol., 15(1): 38-44.

Maldas, D., and Kokta, B.V. 1989a. Improving adhesion of wood fiber with polystyrene by

the chemical treatment of fiber with a coupling agent and the influence on the mechanical
properties of composites. J. Adhes. Sci. Technol. 3(7): 529-539.

Maldas, D., Kokta, B.V. and Daneault, C. 1989b. Influence of coupling agents and
treatments on the mechanical properties of cellulose fiber-polystyrene composites. J.
Appl. Polym. Sci. 37:751-75.

Maldas, D., Kokta, B.V. and Daneault, C. 1989c. Composites of polyvinyl chloride-wood
fibers: IV. Effect of the natural fibers. J. Vinyl Tech. 11(2):90-99.

Maldas, D., Kokta, B.V., Raj, R. 0., and Daneault, C. 1988. Improvement of the mechanical
properties of sawdust wood fiber-polystyrene composites by chemical treatment. . Polymer
29: 1255-1265.

Maldas, D., and Kokta, B.V. 1992a. Performance of hybrid mica and surface modified wood
fiber reinforced PVC composites. ANTEC’92, 1073-1076.

Maldas, D., Kokta, B.V. 1992b. Performance of hybrid reinforcements in PVC
composites. IV: Use of surface-modified glass fiber and different cellulosic materials as
reinforcements. Intern. J. Polymeric Mater. 17:205-214.

Maldas, D., Kokta, B.V. 1992c. Performance of hybrid reinforcements in PVC
composites. IH: Use of surface-modified glass fiber and wood pulp as reinforcements.
Intern. J. Reinforced Plastics and Composites. 2: 1093-1 102.

Malvar, L.J. 2000. Advanced composites for the navy waterfront infrastructure. Special
Publication ACI SP-l96. American Concrete Institute. llpp.

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

Matuana, L.M., Kamdem, DP, and Zhang, J. 2001. Photoageing and stabilization of rigid
PVC/wood-fiber composites, J. Appl. Polym. Sci. 80: 1943-1950.

24

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

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

Matuana, L.M., Balatinecz, J .J ., Park, C.B., Woodhams, R.T. 1999. Surface characteristics
fo chemically modified newsprint fbers determined by inverse gas chromatography. Wood
Fiber Sci. 31(2) 116-127.

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

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

Meyer, J. A. 1987. Wood polymer composites. Technologist. 1(3): 4-7.

Meyer, J. A. 1982. Industrial use of wood polymer materials: state of the art. Forest Prod.
J. 32(1): 24-29.

Meyer, J. A. 1981. Wood Polymer Materials: State of art. Wood Science. 14(2): 49-54.

Meyer, J. A. 1977. Wood-polymer composites and their industrial applications. In: I. S.
Goldstein (ed.) Wood Technology: Chemical Aspects. ACS Symposium Series 43.
Washington, D. C. 301-325 pp.

Meyer, J. A. 1965. Treatment of wood-polymer systems using heat-catalyst technique.
Forest Prod. J. 15(9): 362-364.

Muson, J. M., and Kamdem, D. P. 1998. Reconstituted particleboards from CCA-treated
red pine utility poles. Forest Prod. J. 48(3): 55-62.

Nass, L. 1., editor, 1977. Encyclopedia of PVC, Marcel Dekker, Inc., New York and
Basel, New York.

Oksman, K., and Lindberg, H. 1998a. Influence of thermoplastics elastomers on adhesion
in polyethylene-wood flour composites. J. Appl. Polym. Sci. 68:1845-1855.

Oksman, K., Lindberg, H., and Holmgren, A. 1998b. The nature and location of SEBS-
MA compatibilizer in polyethylene-wood flour composites. J. Appl. Polym. Sci. 69:201-
209.

25

Oksman, K., and Clemons, C. 1998c. Mechanical properties and morphology of impact
modified polypropylene-wood flour composites. J. Appl. Polym. Sci. 67: 1503-13.

Patterson, J. 2000. New opportunities with wood flour foamed PVC. VINYLTEC 2000,
Rigid PVC in the New Millennium: Innovations, Applications, and Properties.
Philadelphia, PA, USA, October 2000. p. 5-14.

Pendleton, D.E., Hoffard, T.A., Adcock, T., Woodward, B., and Wolcott, M. 2002.
Durability of an extruded HDPE/wood composite. Forest Product J. 52(6):21-27.

Principia Partners and USDA Forest Products Laboratory. 2003. Current and emerging
applications for natural & wood fiber composites. A presentation to: 7m International
Conference on Woodfiber-Plastic Composites (and other natural fibers). May 19-20,
2003, Madison WI.

PRNewswire. 2004. Considering composites? Law Offices of Marc B. Kramer, P.C.

Announces Nationwide Class Action Certified by Court Against Trex Company, Inc. and
ExxonMobil in Case Alleging Product Defects. NEW JERSEY, June 2.

Raj, R. 0., Kokta, B. V., Nizio, J. D. 1992. Studies on mechanical properties of
polyethylene-organic fiber composites. 1. Nut shell flour. J. Appl. Polym. Sci. 45:91-101.

Raj, R. 0., and Kokta, B. V. 1991. Improving the mechanical properties of HDPE-wood
fiber composites with additives/coupling agents. ANTEC. Proc. 49th Annual Technical
Conference. Montreal, Canada, May 5-9, 1991. Society of Plastics Engineers, Brookfield,
CT. pp. 1883-1885.

Raj, R. 0., Kokta, B. V., Maldas, D., and Daneault, C. 1989a. Use of wood fibers in
thermoplastics. VII. The effect of coupling agents in polyethylene-wood fiber
composites. J. Appl. Polym. Sci. 37:1089-1103.

Raj, R. 0., Kokta, B. V., Groleau, 0., and Daneault, C. 1989b. Use of wood fiber as a
filler in polyethylene: studies on mechanical properties. Plast. Rubber Process. Appls. 11:
215-221.

Sain, M. M., Balatinecz, J. and Law, S. 2000. Creep fatigue in engineering wood fiber
and plastic compositions. J. Appl. Polym. Sci. 77: 260-268.

Schnieder, M. H. 1994. Wood polymer composites. Wood Fiber Sci. 26(1): 142-151.

Schnieder, M. H. 1995. New cell wall and cell lumen wood polymer composites. Wood
Sci. Tech, 29(2):121-127.

26

Schut, J. H. 1999. Wood is good for compounding, sheet & profile. Plastic Technology.
March. p.46—52.

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

Smith, P.M. 2001. US. woodfiber-plastic composite decking market. In: Proc. Sixth
International Conference on Woodfiber-Plastic Composites. Forest Prod. Soc., Madison,
WI. p. 13-17.

Struktol Company. 2000. The effects of various additives on the processing and physical
properties of wood-filled PVC. Presentation at: the Wood-Plastic Conference. Baltimore,
MD. December 5, 2000.

Struktol Company. 2002. New engineered products designed specifically for wood-filled
applications. Struktol Knew You “Wood”.

Takatani, M., Kato, 0., Kitayama, T., Okamoto, T. and Tanahashi, M. 2000. Effect of
adding steam-exploded wood flour to thermoplastic polymer/wood composite. J. Wood
Sci. 46:210-214.

Titow, W. V. 1984. PVC Technology. Elsevier applied Science Publishers. London and
New York.

Wang, U. P. 1975. Gamma ray induced low temperature in situ polymerization of vinyl
chloride and copolymerization of vinyl chloride-vinyl acetate in Taiwan-produced woods.
J. Chinese Chem. Soc., 22: 77-89.

Witt, A. E. 1981. Acrylic wood in the United States. Radiat. Phys. Chem. 18(1-2): 67-80.

Woodhams, R. T., Thomas, G. and Rodgers, D. K. 1984. Wood-fiber as reinforcing filler
for polyolefins, Polym. Eng. Sci., 24: 1166-1177.

Xu, W., and Moschler W. W. 1996. A procedure to determine water absorption
distribution in wood composite panels. Wood and Fiber Science.28(3): 286-294.

27

Chapter 2

CHARACTERIZATION OF RAW MATERIALS AND LABORATORY
MANUFACTURE OF PVC/WOOD FLOUR COMPOSITES

2.1 Characterization of the Raw Materials
2.1.1 Thermoplastics (PVC)
PVC is polymerized vinyl chloride through suspension, mass, or emulsion

polymerization methods (Wickson 1993). PVC has the molecular structure below:

‘ECHz— ('33?
Cl

In this study, poly (vinyl chloride) powder (PVCl), is provided and formulated using
a proprietary method by CertainTeed Corporation, Jackson, Michigan. PVCl was used as
supplied.

PVC 190F (PVC2), manufactured by Oxy Vinyls, is formulated with 3% tin
stabilizer, MK-98 (methyltin mercaptide) from Crompton Corp., Hahnville, Louisiana
and 1.5% calcium stearate, extra density grade, manufactured by Crompton Corp.,
Greenwich, Connecticut.

PVC 190F homopolymer resin (PVC3) with an inherent viscosity of 0.73 and the K
value of 58 is manufactured by Oxy Vinyls, LP, Dallas, Texas.

The main physical and mechanical properties of PVC are listed in Table 2.1.

28

Table 2.1. Physical and mechanical properties of rigid PVCs used in this study

 

 

 

 

 

Unnotched Flexural Flexural Flexural Density
impact strength strength modulus toughness (glem’)
(J/m) (MPa) (GPa) (kPa)
PVCl 42.1 36.5 2.68 26.3 1.4
(2.32)* (1.99) (0.09) (1.75)
PVC2 49.8 47.0 3.21 39.1 1.4
(2.69) (2.06) (0.08) (2.45)
PVC3 83.5 88.9 3.13 96.4 1.4
(3.02) (2.68) (0.18) (3.06)

 

 

 

 

 

 

 

* Standard deviations are given in parenthesis.

2.1.2. Wood Flour
Wood is a natural polymer material mainly composed of cellulose, hemicelluloses

(polyoses), lignin and extractives. Cellulose is the major structure component of wood
cell walls. It is characterized as a linear polymer composed of B-D-glucose unit (Figure

2.1) (Fengel and Wegener 1984).

H H2COH
o

   

Figure 2.1. Chemical structure of cellulose

Hemicelluloses have much shorter molecular chains than cellulose. Hemicelluloses
contain various sugar units such as pentoses, hexoses, hexuronic acids and deoxy-hexoses
(Fengel and Wegener 1984, Hon and Shiraishi 1991). Chemical structures of some of

these sugars and uronic acids are listed in Figure 2.2.

29

Pentoses
OH OH
HO
H
B-D-Xylose
H O H
OH
H
a-L—Arabinopyranose

Hexoses Hexuronic acids Deoxy-hexoses

CH20H COOH H
0 OH 0 OH H 0
OH OH OH
HO HO 3
H H OH H
B-D-Glucose B-D-Glucuronic acid a-L-Rhamnose

COOH

CH20H H 0
OH
OH CH
OH OH /<:§5V
HO@ @OH HO
OH OH

B-D-Mannose or-D-Galacturonic acid a-L-Fucose

Figure 2.2. Formulas of the sugar components of polyoses (hemicelluloses)

Lignin is an amorphous polymeric substance, which is mainly located in the middle

lamella and secondary cell walls (Fengel and Wegener 1984, Hon and Shiraishi 1991).

The molecules of lignin consist of aromatic phenylpropane units as shown in Figure 2.3.

Softwood lignin is mainly constructed by guaiacyl units generated from coniferyl alcohol

(II), whereas hardwood lignin contains both gaiacyl and syringyl units, and syringyl unit

is originated from sinapyl alcohols (III). P-coumaryl alcohol (I) is the precursor of p-

hydroxyphenyl units, which build up grass lignin (Hon and Shiraishi 1991).

30

CHZOH CHZOH CH 20H

 

Figure 2.3. Building units of lignin: p-coumaryl alcohol (I), coniferyl alcohol (11),
sinapyl alcohol (111)

Although extractives share only a few percent of wood mass, these low molecular
weight substances have important influences on the properties and processing qualities of
wood. Main wood extractives include aromatic compounds, terpenes, aliphatic acids,
alcohols, tannins and flavonoids (Fengel and Wegener 1984).

Red oak (Quercus rubra) flour (WF) with an average diameter of 0.15 mm and an
average length ranging from 74 to 590 um was used as supplied by CertainTeed
Corporation, Jackson, Michigan. The wood particle length distribution is shown in Figure
2.4. Wood generally contains 50% carbon, 43% oxygen, 6% hydrogen and 1% nitrogen
(Fengel and Wegener 1984). Chemical and elemental compositions as well as basic

physical properties of red oak are listed in Table 2.2, and 2.3 respectively.

31

Frequency (%)
8 8
o o

 

 

0.00 0.50

if T r

1.00

1.50

Wood flour particle length (m)

2.00

Figure 2.4. Wood flour particle length distribution (J iang and Kamdem 2003)

Table 2.2. Chemical composition and physical properties of oak (FPL, USDA Forest

Service 1999)

 

Red oak (Quercus rubra)

 

 

 

 

 

 

 

 

 

 

Cellulose (%) 40.5
Hemicellulose (%) 23.3

Lignin (%) 21.8

Hot-water extract (%) 5 .2

Ash (%) 0.1

Density (g/crfij) 0.65
Green wood moisture content (%) 69-80
Thermal conductivity (Ovendry) (W/mK) 1 0.14

 

Table 2.3. Elemental composition of red oak ash (%) (Misra et al. 1993)

 

 

| P

S K

Ca

M

Al

MnI Cu

 

 

 

I Red oak (Quercus rubra) 1.56

 

1.80 6.08

 

36.58

 

5.20

 

0.68

 

1.49 | 0.07

 

32

 

2.2. Preparation of Copper Amine Treating Solutions

Copper amine treating solutions with different copper concentrations were prepared
by dissolving solid power copper hydroxide into liquid ethanolamine. The molar ratio of
copper to amine is maintained at 1:4. Various target retentions (Table 2.4) of copper were
achieved by treating WF with solutions containing 0.1, 0.2, 0.4, 0.5, 0.6 and 1.0 wt%

elemental copper (Zhang and Kamdem 2000a,b,c).

2.3 Wood Flour Copper Amine Treatments

Treating solutions were used to dip treat WF at a liquid to solid ratio of 1.5:1 at room
temperature and atmospheric pressure. Treated WF was oven dried at 60 °C for 24 hours,
and then at 105 °C for 24 hours to reduce the moisture content to less than 1% before
further processing. High moisture content in WF is proven to interfere blending during
the mold compression of wood plastic composites (Bledzki et al. 1998, Clemons 2002).
The copper contents in treating solutions and in the treated WF were analyzed by using

atomic absorption spectroscopy according to the AWPA standard A1 1-93 (AWPA 2003).

2.4. Compression Molding of PVC/Wood Flour Composites

Untreated and copper amine treated wood flour were mixed with PVC powder for ten
minutes in a 20-liter high-intensity laboratory mixer. Sample compositions are listed in
Table 2.4. About 500 grams of the mixture was poured in an aluminum mould measuring

4 by 280 by 280 mm and the mould placed between two steel platens. A RHM hydraulic

33

Table 2.4. Compositions of Experimental Samples with Different PVC Types, PVC,
Treated and Untreated Wood Flour Contents, as well as Cu Concentrations in Treated
Wood Flour.

 

 

 

 

 

 

Sample ID PVC type PVC Wood flour Cu content in wood flour
(wt %) (wt %) (wt %)

164Cu0 1 60 40 0 (untreated)
164Cu01 1 60 40 0.1

164Cu02 l 60 40 0.2
164Cu05 1 60 40 0.5
164Cu06 1 60 40 0.6
164Cu07 l 60 40 0.7
164Cu12 1 60 40 1.2

155Cu0 l 50 50 0 (untreated)
155Cu01 l 50 50 0.1

155Cu02 l 50 50 0.2

155Cu05 1 50 50 0.5
155Cu06 1 50 50 0.6
155Cu07 1 50 50 0.7
155Cu12 1 50 50 1.2

146Cu0 1 40 60 0 (untreated)
146Cu01 1 40 60 0.1

146Cu02 1 40 60 0.2

146Cu05 1 40 60 0.5
146Cu06 1 40 60 0.6
146Cu07 1 40 60 0.7
146Cul2 1 40 60 1.2

264Cu0 2 60 40 0 (untreated)
264Cu01 2 60 40 0.1
264Cu02 2 60 40 0.2
264Cu05 2 60 40 0.5
264Cu06 2 60 40 0.6
264Cu07 2 60 40 0.7
264Cu12 2 60 40 1.2

364Cu0 3 60 40 0 (untreated)
364Cu01 3 60 40 0.1
364Cu02 3 60 40 0.2
364Cu05 3 60 40 0.5
364Cu06 3 60 40 0.6
364Cu07 3 60 40 0.7
364Cu12 3 6O 40 1.2

 

 

 

 

 

 

 

34

oil heated press with a nominal maximum pressure level up to 6 MPa was used for the
compression molding.

The press platens were maintained at 200 °C. The press cycle consisted of two
phases: the first phase involved the heating of the mould assembly to 200 °C for 8
minutes. After the mould assembly reached the desired temperature, the press was slowly
closed for 2 minutes. The goal of a slow closing of the press was to maintain contact with
the mould assembly as furnish began to melt in order to facilitate the flowing of the
thermoplastic within the mould. This procedure reduces the probability of the formation
of internal air voids in the panel during the release of gases that may cause undesirable
defects.

The second phase is the closure of the press for 5 minutes. After the 15 minutes
press cycle, the caul sheet assembly with the mould containing the molten wood plastic
was removed from the hot press and placed in a cold press for 15 minutes to allow the
composite to harden under pressure. Sample boards were kept in a room condition at

23°C and 65 % relative humidity until further testing.

2.5 Density Measurement
The oven-dried density of samples was measured after 24 hours at 105 °C; the
density values, which ranged from 1.2 to 1.3 g/cm3, were calculated on the base of the

weight divided by the volume and also listed in Table 2.4.

35

References
American Wood-Preservers’ Association (AWPA) Standards. 2003. Granbury, Taxas.

Bledzki, A. K., Reihmane, S., and Gassan, J. 1998a. Thermoplastics reinforced with
wood fillers: A literature review. Polym. -Plast. Technol. Eng. 37(4): 451-468.

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

Fengel, D. and Wgener, G. 1984. Wood, Chemistry, Ultrastructure, Reactions. Walter de
Gruyter, New York.

Forest Products Laboratory, USDA Forest Service. 1999. Wood Handbook-Wood as an
Engineering Material. Madison, WI. USA.

Hon, D. S., and Shiraishi, N. 1991. Wood and Cellulosic Chemistry. Marcel Dekker, Inc.,
Jiang, H., Kamdem, P. D., Bezubic, B. and Ruede, P. 2003. Mechanical Properties of
Poly(Vinyl Chloride)/Wood Flour/Glass Fiber Hybrid Composites. J. Vinyl Additive
Technol. 9(3): 138-145.

Misra, M. K., Ragland, K. W., and Baker, A. J. 1993. Wood ash composition as a
function of furnace temperature. Biomass and Bioenergy. 4(2): 103-116.

Wickson, E. editor. 1993. Handbook of PVC formulating. A Wiley-interscience
Publication, John Wiley & Sons. New York. P. 17.

Zhang, J., and Kamdem, D. P. 2000a. PTIR characterization of copper ethanolamine-
wood interaction for wood preservation. Holzforschung. 54(2): 119-122.

Zhang, J ., and Kamdem, D. P. 2000b. Electron paramagnetic resonance spectroscopic
(EPR) study of COpper amine treated southern pine in wood priservation. Holzforschung.
54(4): 343-348.

Zhang, J ., and Kamdem, D. P. 2000c. X-ray diffraction as an analytical method in wood
priservatives. Holzforschung. 54(1): 27-32.

36

Chapter 3
EFFECTS OF COPPER AMINE TREATMENT ON MECHANICAL
PROPERTIES OF PVC/WOOD FLOUR COMPOSITES
Mechanical property of a material directly reflects its characteristic performance
under external forces and provides important information for its applications. In this
chapter, various mechanical properties (including impact strength and flexural properties)
of PVC/Cu-treated, untreated wood flour composites were measured in order to evaluate
their mechanical performances as well as the effect of wood flour surface copper
modification. The performance of the copper treatment on the WP and then on the
resulted WPC was evaluated by testing the impact and flexural mechanical properties of
the WPC. SEM was also used to examine fractured sample surfaces and compare the

rupture modes of treated and untreated wood fiber/PVC composites.

3.1Testing Methods and Equipment
3.1.1 Mechanical Property Testing

The size of the specimens used for the impact test was 4 by 12.7 by 63.5 mm. Both
Izod unnotched and notched impact strengths of composite samples were measured.
Unnotched impact strength was performed according to ASTM D4812-99 by using a
Tinius Olsen Model 92T Impact Tester.

The Izod notched impact strength was conducted following ASTM D256, Test
Method C with the toss correction. The value measured was called “estimated net

notched izod impact strength”. The notched specimens have notch tip radii of 0.25 mm

37

and were prepared by using the Tinius Olsen Model 899 Automatic Specimen Notcher
designed for plastics. For each sample group, fifteen replicas were tested.

The flexural test specimen size was 4 by 12.7 by 127 mm. The flexural test
properties were conducted according to ASTM D790-99 using the Instron testing
instrument Model 4206. A three-point bending test was performed with a span to
thickness ratio of 16, and the crosshead speed of the Instron Tester was set at 2.2

mm/min. Ten specimens per sample group were tested for flexural properties.

3.1.2 Statistical Analysis

SAS 8.2 system was used for statistical analysis of the effects of Cu-treatment on the
mechanical properties. Two-way ANOVA and generalized lineal model procedure were
used to compare the difference between means and standard deviations of sample groups
at a 95% confidence interval (Jusoh and Kamdem 2001, Kamdem and McIntyre 1998,

Hiziroglu and Kamdem 1995, Kamdem and Scan 1994).

3.1.3 Scanning Electron Microscope (SEM) and Energy-dispersive Spectroscopy
(EDS) Analysis

Fractured surfaces of specimens after the impact test were investigated by using a
JEOL (Japan Electron Optics Laboratories) JSM-6400V SEM with a Lanthanum-
hexaboride (LaBa) emitter and the Noran energy-dispersive spectroscopy (EDS) system.
The accelerating voltage was set at 20 kV for both SEM image and EDS analysis. For
SEM image acquiring, fractured surfaces were pre-coated with gold by using a sputter

coater. The thickness of gold cover layer was around 50 nm. For EDS spectrum and

38

elemental quantitative analysis, the fractured surfaces were pre-coated with a thin layer of
carbon. A minimum of 10 exposed wood flour surfaces on each of the 8 fractured sample
surfaces were examined in order to estimate the average chlorine (from PVC) elemental

amount on the wood flour surface.

3.2 Results and Discussion

Table 2.4 lists the copper retentions of WF after treatment, the compositions of the
panels in terms of percentage by weight of the WF, PVC and types of PVC as well as the
respective density of each type of panels. The density of composite panels varied between

1.21 and 1.33 g/cm3.

3.2.1 Impact Properties

Unnotched impact strength of the boards made with three types of PVC/Cu-treated,
untreated wood flour is represented in Figure 3.1. Statistical analysis of the impact
strength data indicate that copper treatment significantly improved the unnotched impact
strength of PVCl-WF composites at WF loading levels of 40 wt%, 50 and 60 wt%, as
shown in Figure 3.1(a). Treatment with Cu concentrations ranging from 0.2 to 0.6 wt%,
the unnotched impact strength increased by 48.8 % from 33.2 J/m of the untreated sample
to 49.4 J/m for boards made with 0.2 % Cu content treated WF and 60 % PVCl.
Similarly, 35 to 49% enhancement was obtained for panels containing 50 % and 60 %
Cu-treated WF. Unnotched impact strength was reduced by 14% when wood flour

contents were increased from 40-50 % to 60 %.

39

A
2?.

 

70.0 j

 

 

 

Unnotched impact strength (Jlm)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.5 0.6 0.7
Cu concentration in treated wood flour (%)

A
U
V

110.0
100.0

 

 

 

 

Si"!

 

 

70.0 1' l..- :.: .I: ... .
C‘: . :3: ::C " " T

 

 

 

 

ll'1

 

 

Iiii

 

iitiiiil'l
:33

3:13!!!

Unnotched impact strength (Jim)

it!!!

:tLliiJlill:
O I CD ‘
itiiitiiiiiii r-i

iii!!!
I
i

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

d
N

0.1 0.2 0.5 0.6 0.7
Cu concentration in treated wood flour (%)

Figure 3.1. Unnotched impact strength of PVC/Cu-treated wood flour composites with:
(a) different PVCl contents and Cu concentrations in treated, untreated wood flour (b)
three PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration
levels. Legends for Figure 1(a)

I PVCI/WF=60/40 PVCI/WF=50/50 PVCI/WF=40/60

 

 

 

 

 

Legends for Figure 1(b)

 

 

 

 

I PVCl/WF=60/40 H, PVC2/WF=60/40 ;.;.;. PVC3/WF=60I40

 

 

 

 

 

40

Unnotched impact strengths of both PVC2 and PVC3-wood flour composites were
also improved with by using Cu-treated wood flour as shown in Figure 3.1(b). An
increase of 44% of the unnotched impact strength was achieved on boards made with 60
% PVC2 and 40 % Cu-treated WF containing 0.2 % Cu compared with that of
corresponding sample boards made with untreated WF. The same trend was found on
PVC3-wood composites. Unnotched impact strength of boards made with 60 % PVC3
and Cu-treated WF containing 0.2 % Cu was 40% higher than that of the same
composition samples made with untreated WF. All PVC3-wood flour composite have
the larger unnotched impact strengths than those of PVC2-wood flour composites. This
was attributed to the present of the tin thermal stabilizer, which promotes the
embrittlement of the PVC matrix resulting in reductions in some mechanical properties of
the composites (Younan et al. 1983, Nass 1977, Titow 1984).

Figure 3.2 shows the notched impact strength of composite samples after the toss
correction, which is termed as estimated net notched impact strength. For PVCl-wood
flour composites with 40 wt% wood flour, the estimated net notched impact strength was
not affected by the Cu treatment, as shown in Figure 3.2(a). The use of treated WF with
the Cu contents higher than 0.6 wt% decrease the estimated net notched impact strength
at 50 and 60 wt% WF loading levels. The reason of this negative effect on notched
impact strength of composites is not known and will be investigated further. Figure 3.2(b)
shows that the estimated net notched impact strengths of all three types of PVC-wood
flour composites were not influenced by the copper treatment in comparison with

composites made with untreated WP and PVC at the 40 wt% WF content level.

41

 

 

 

_s
.0
0

Estimated net notched
impact strength (Jim)

9'
o

 

 

 

 

 

 

 

 

 

 

 

 

0.0 «

 

0.7 1.2
Cu concentration In treated wood flour (%)

 

 

(b) 20.0

 

15.0

._ _|_4.————_

 

  

   

 

  

 

31113

iztiiiiiiirrtitt

 

  

‘9' U 0 0 i I I
e s s a e e a a a
a a e e e o .

Estimated not notched
impact strength (Jim)
8
0
WW

  

9'
o

 

riirrrriaiiiiiii ri

 

 

 

 

 

 

 

 

r:-
Iiii

 

 

 

 

 

 

 

 

 

 

 

 

.0
o

 

0.0 0.1 02 0.7 12
Cu concentration in treated wood flour (%)

Figure 3.2. Estimated net notched Izod impact strength of PVC/Cu-treated wood flour
composites with: (a) different PVCl contents and Cu concentrations in WF (b) three PVC
types at the ratio of PVC/wood of 60/40 and at different Cu concentration levels.

Legends for Figure 3.2(a)

I PVC 1/WF=60/40

 

PVCl/WF=50/50 PVCl/WF=40/60

 

 

 

 

Legends for Figure 3.2(b)
I PVC1/WF=60/40

 

PVC2/WF=60/4O .;.-.; PVC3/WF=60/40

 

 

 

 

42

Unnotched impact strength of composite materials is known as resulting from the
combination of the crack initiation and the crack propagation energy. Notched impact
strength is mainly attributed to the crack propagation energy because the crack initiation
energy is minimized by the presence of the very sharp notch (Crawford 1998). PVC is a
notch-sensitive material, it is reported that a reduction of about 250% happens on the
notched impact strength when the notch tip radius is reduced from 2.0 mm to 0.5 mm (
Nass 1977, Titow 1984, Crawford 1998). This is why the values of the unnotched impact
strength of the PVC/wood flour composites are higher than the values of the notched
impact strength. The impact test data suggest that the use of WF treated with copper can
effectively increase the crack initiation energy of the composites. However, the WF
copper treatment did not do much contribution to the crack propagation energy. As the
result, the unnotched impact strength rather than the notched impact strength was

improved by the use of copper treated WF in the formulations of PVC/wood composites.

3.2.2. Flexural Properties
3.2.2.1 Flexural Strength

Flexural strength, flexural modulus and flexural toughness of PVC/wood flour
sample composites are illustrated in Figures 3.3 to 3.5. Statistical analysis revealed that
there was significant enhancement of the flexural strength of samples made with PVC]
and Cu-treated WF as presented in Figure 3.3(a). Samples made with 60 wt% PVCl and
40 wt% Cu-treated wood flour exhibit an increase of 36% of the value of flexural
strength, from 27.6 MPa for samples made with untreated WF to 37.4 MPa for those

made with 0.5 % Cu contained WF. Correspondingly, 39 % and 45 % improvement of

43

the flexural strength was observed by the use of 50 wt% and 60 wt% Cu-treated WF with
0.5 % Cu. Sample boards made with 0.5 % Cu contained treated WF shows the higher
flexural strength than those made with 1.2 % Cu contained WF. Figure 3.3(b) shows
flexural strengths of samples made with PVC2, PVC3 and Cu-treated, untreated WF at
the 40 % WF loading level. An increase of 18% was observed with samples made of
PVC2 and Cu-treated WF containing 0.2 % Cu, and a 13% increase was found when

using PVC3 and Cu-treated WF with the same Cu concentration.

3.2.2.2 Flexural Modulus

No significant difference was observed in flexural modulus of the composites made
with PVCl and Cu-treated or untreated WF as illustrated in Figure 3.4(a). A slightly
increase was noticed with composite samples composed of 0.5 % Cu contained WF and
60 wt% or 50 wt% PVCl. The copper treatment of WF also did not significantly
influence the flexural modulus of sample boards made with PVC2, PVC3 and WP
(Figure 3.4(b)). PVCl-WF composite boards show somewhat lower flexural modulus

compared with composite boards made with PVC2 and PVC3.

3.2.2.3 Flexural Toughness

Flexural toughness of PVCIWF composites was obviously improved by the use
copper treatment as shown in Figure 3.5. At the 0.5 wt% Cu concentration level, 40%
improvement in flexural toughness was obtained with PVCl and 40 wt% to 60 wt%
treated WF. No significant statistical difference was observed when the WF proportions

varied from 40 % to 60 % by weight (Figure 35(3)). The lower value of the flexural

toughness was obtained with PVC2 may be attributed to the “antiplasticisation” effect of

the tin stabilizer present in the formulation of PVC2 (Figure 3.5(b)).

3.2.3 Fracture Surface Analysis

Thermodynamically, fiber-matrix interfacial characteristics are crucial to understand
and predict the mechanical behavior of wood/plastic composites (Bledzki et al. 1998,
Matuana et al. 1998, Zhou et al. 2001, Hull and Clyne 1996). These characteristics are
fundamental in understanding the mechanism of fracture initiation, stress transfer and
crack propagation within the composite (Balasuriya et al. 2001, Escamilla et al. 2002,
Ray et al. 2002).

The fractured surfaces of unnotched impact testing boards made of PVC2 and Cu
treated and untreated WF were examined by Scanning electron microscope (SEM)
combining with Energy-dispersive spectroscopy (EDS) analysis. Extensive wood
particle-PVC interfacial debonding can be observed on the fractured surfaces of
composites made of PVC2 and untreated wood flour. It reflects a weak interfacial
bonding between untreated wood flour particles and PVC2 matrix leading to lower
unnotched impact properties (Zhou et al. 2001, Hull and Clyne 1996). The SEM
micrographs of the fractured surface of sample made of PVC2 and Cu-treated wood flour
are exhibited in Figure 3.6. The fractured surfaces of composite samples made of
PVC2/Cu-treated wood were characterized by large scale of wood particle pullout
(Figure 3.6a). Some amount of wood particle rupture(Figure 3.6b), which is known as the

high energy absorbing mechanism, were observed on the unnotched fracture surfaces of

45

A
3
8
O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50.0
E
5 40.0
i.
III (‘1‘
n 1"
5 30-0 - 33$ $11 7.3 H
a 1': ::: ::: :::
E Z: Z: $33 $31
3 20.0 a m III III III ”‘*
3 III ”I III I”
”4 ”I III ”I
u' ”A ”I ”I ”A
10.0 4 $11 $15 $31 $31 —
III. ”I III Ill
[”4 III III Ill
[/11 III III III
[/4 III II II
0.0 4 .. ..
0.0 0.1 0.6 0.7 1.2
Cu concentration in treated wood flour (%)
(b) 100.0
90.0 —-
80.0 -~——-——-
A L 1-
:! 70-0 ; :-: . . . .1; r.
5 -'!‘- :-: .; ;-:- -:. :-:
g 60.0 :3 '. :.: '2' l—
5 . .:. .3. ‘
5 50.0 : _ . .: . .-: :-: —
g e .04 . a «D u:e.d w):-
i 40.0 2 " ...jjt- 336-: Fa.- ~—
3 ' I; .....:-: :0: .1
g 30.0 I -~-- :I- :;~.-; wt- -I- L-
n. 2 r: ,33- :
20.0 ~ Ji: j: ...III -221 3H
.4'. . .....’. ...q,'.i ."
10-O- “ ‘-: 3, ”:3: :::-:j —
00 j T .. a. _ 7
0.0 0.1 0.2 0.5 0.6 0.7 1.2

Cu concentration in treated wood flour (%)

Figure 3.3. Flexural strength of PVC/Cu-treated wood flour composites with: (a)
different PVCl contents and Cu concentrations in treated, untreated wood flour (b) three
PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration levels

Legends for Figure 3.3(a)
I PVCI/WF=60/40

Legends for Figure 3.3(b)

I PVCI/WF=60/40

 

 

 

46

PVC l/WF=50/50

PVC2/WF=60/40

 

 

 

 

 

 

 

 

PVC l/WF=40/60

PVC3/WF=60I40

A
m
v
V
O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6.0
E 5.0
9
3
g 33 i“
ll
5 $3 a
a :3
II
l/
E II i—-
IL //
l/
l/
l/ _
II
II
II
II,
1.2
Cu concentration in treated wood flour (‘16)
(b) 7.0
6.0
a 5.0 — ‘7 T T
9.. " -_ -- 1 :: “
5 4,0 1- .;' 2.3.2: .I Z __
= ‘:::1:~ :::i::- :: °::: .
-i-I- ...": .. ....
E 3.0- ::1:-' .32:- ': 31%
8 {11:3, :.+.:3 “' *1"
3 -.... -1. I“. ..II
2.0- 3...“; .. .; g ; .. .. .H
II. 1...... .. I . .. ....
:1: I- I ~ : :2
1-00 .112} .3? i f : :3 -
"z‘:-: '1 I 3 : 5:
000‘ J I l A V ‘ I
0.0 0.1 0.2 0.5 0.7 1.2

Cu concentration in treated wood flour (%)

Figure 3.4. Flexural modulus of PVC/Cu-treated wood flour composites with: (a)
different PVCl contents and Cu concentrations in treated, untreated wood flour (b) three
PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration levels

 

 

 

 

 

 

 

 

 

 

 

Legends for Figure 3.4(a)
I PVCI/WF=60/40 "'"‘ PVCI/WF=50/50
Legends for Figure 3.4(b)
I PVC l/WF=60/40 PVC2/WF=60/40

 

47

 

 

 

 

 

 

 

 

PVCI/WF=40/60

PVC3/WF=60/40

 

n)
v
8
O

 

40.0

30.0

 

20.0 J

Flexural toughness (kPa)

10.0 ~

 

0.0 -

A
U'
V

110.0

90.0
80.0
70.0

Flexural toughness (kPa)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

,_ T
”I.
335: L
III III.
l/I‘ III.
I” III.
II/ ”/4
II: Ill.
1:, 33$: 3
III III.
”I I”.
ll/ ”/1
{iii 1M
0.0 0.1 0.2 0.7 1.2
Cu concentration in treated wood flour (%)
9:, .. _ _._E
'3' 2'7 r'.- 4'-
. : :-: -:- :-: -l
’ . .'.‘ '.' .'. |
*1 I 2' :::___J
-Ij j .. I 3: W
:: 1:3 3 5:3 .553: __
:3 'EEE' : El: :31 ::: '
..... .. ' t.‘ ..-. T
'2: 3 1‘: :;. :2 'H
0.1 0.2 0.5 0.6 0.7 1.2

 

Cu concentration in treated wood flour (%)

Figure 3.5. Flexural toughness of PVC/Cu-treated wood flour composites with: (a)
different PVCl contents and Cu concentrations in treated, untreated wood flour (b) three
PVC types at the ratio of PVC/wood of 60/40 and at different Cu concentration levels

Legends for Figure 3.5(a)
I PVCl/WF=60/40

Legends for Figure 3.5(b)

I PVC l/WF=60/40

m
VIII/4

AAAAA

 

vvvvv

 

 

 

 

 

 

 

 

 

 

PVCl/WF=50/50

PVC2/WF=60/40

48

 

 

 

 

 

 

 

 

PVCl/WF=40/60

PVC3/WF=60/40

sample made of PVC2 and 0.2 wt% Cu content treated wood flour (264Cu02). This is in
correspondence with its high unnotched impact strength ( Zhou et al. 2001, McHenry and
Stachurski 2003, Song and Hwang 1997).

EDS spectra of PVC2 matrix, single wood flour particles as well as PVC2/Cu-
treated, untreated wood flour composites are given in Figure 3.7. The EDS spectrum of
PVC2 was characterized by a very strong and sharp chlorine peak, a visible carbon peak
and some peaks from oxygen, tin and calcium (Figure 3.7(a)). The EDS spectrum of
untreated wood flour was characterized by a sharp, strong carbon peak and smaller
oxygen peak (Figure 3.7(b)). For Cu-treated wood flour, two extra copper signal peaks
around 0.93 keV (La) and 8.05 keV (Ka) were present (Lide 2003).

Figure 3.7(c) shows the EDS spectrum of exposed wood particle surfaces from
PVC2/untreated wood flour composite sample 264Cu0. Compared with carbon and
oxygen peaks, chlorine (Ka) peak presented at 2.62 keV was very weak. The results of
EDS quantitative analysis indicate that 1.0 — 1.5% chlorine on the wood particle surfaces
of the untreated wood flour composite. The EDS spectrum of wood particle from
composites made of PVC2 and 0.2 wt% Cu contained wood flour is given in Figure 3.7
(d). It is obvious that chlorine peak was much stronger and higher than the chlorine peak
in Figure 3.7(c) in comparison with carbon and oxygen peaks. The average chlorine

content was 4.5 - 5.0% after EDS quantitative analysis.

49

 

Figure 3.6. ESEM micrographs of the impact fracture surfaces of: (a) sample 264Cu10
showing wood particles pull out (b) sample 264Cu02 showing wood particle fracture

50

(a)

(knuns

(b)

Counus

8000-

7000-
6300-
5600-
4900-
4200-
35001
2800.
2100-
1400.

700-

 

0

C
A9 -,

0.000

7000.

6000.
5400.
4800-
4200.
3600+
3000.
2400.
1800.
1200.
600

 

0

U

 

Cl

Sn

 

rsJ

2.000

 

.EE -
4.000 6.000
Eneqn/Uufln

8.000

I] t W'ymfi .A l ‘ “A -...-A .1 A A 4
V V V v '

 

0.000

2.000

4.000 6.000 8.000

Energy (keV)

51

7000. C

6000‘
5400.
4800-
4200-
3600-
3000‘
2400-
1800-

D
1200.Jfi Cl
600
00 . . A“: . . . . .
0.000 2.000 4.000 6.000 8.000
Energy (keV)

Counts

 

 

 

7000-E

60001
5400.
4800.
4200.
3600-
3000-
2400. D
1800.
1200. Ca

600- Cu Sn Cu

C1

Counts

 

 

 

0 ' I 1 *7 1 ' ?
0.000 2.000 4.000 6.000 8.000
Energy (keV)

Figure 3.7. EDS surface area scan spectra of: (a) the impact fracture surface of PVC2
alone characterized by the strong Chlorine peak (b) untreated wood flour particles
exhibiting the sharp peaks of Carbon and Oxygen (c) the exposed wood particle surfaces
on the impact fracture surface of sample 264Cu0 showing the weak and small chlorine
peak (d) the exposed wood particle surfaces on the impact fracture surface of sample
264Cu02 showing a strong and larger chlorine peak

52

The results of EDS analysis suggest that after copper treatment, treated wood
particles can adhere to more PVC matrix, which can not be tore out by the impact force,
than untreated wood flour. This demonstrates that the interfacial adhesion between Cu-
treated wood flour and PVC was stronger than that between untreated wood flour and
PVC matrix implying an improved wood flour-PVC interphase after the copper

treatment.

3.3 Conclusions

Mechanical properties, such as unnotched impact strength, flexural strength and
flexural toughness, were significantly improved by the wood flour copper amine
treatment. The maximum increases on the unnotched impact strength were 48.8 % in
164Cu02, 44% in 264Cu02 and 40% in 364Cu0 comparing with their corresponding
untreated composites. The maximum enhancement of 36% on the flexural strength and
40% improvement on the flexural toughness were obtained in sample 164Cu05. Basically
three types of PVC shared the same improving trend as the increase of Cu concentrations
inside the wood flour. The optimum Cu concentration range for the mechanical properties
was 0.2-0.6 wt% of wood flour. Cu treatment did not make notable contributions to the
improvement on the estimated net notched impact strength and the flexural modulus of
PVC-wood flour composites.

The results of SEM/EDS analysis indicate that PVC, wood particle interfacial
debonding was the main fracture mode dominating the fracture surface of PVC2-
untreated wood flour composites. The fractured surfaces of PVC2-Cu treated wood flour

composites were characterized by the extensive wood particle pullout and wood particle

53

breakage can be observed on the sample with 0.2 wt% Cu in treated wood flour. More
PVC matrix could be found on the exposed Cu treated wood particle surfaces than
untreated wood flour suggesting an improved PVC-wood interfacial adhesion between

PVC and wood flour after Cu-amine treatments.

54

References

American Society for Testing and Materials (ASTM). 2002. Standard test method for
unnotched cantilever beam impact resistance of plastics. ASTM D 4812-99. ASTM,
West Conshohocken, PA.

American Society for Testing and Materials (ASTM). 2002. Standard test methods for
determining the izod pendulum impact resistance of plastic. ASTM D 256. ASTM,
West Conshohocken, PA.

American Society for Testing and Materials (ASTM). 2002. Standard test methods for
flexural properties of unreinforced and reinforced plastics and electrical insulating
materials. ASTM D 790-99. ASTM, West Conshohocken, PA.

Balasuriya P.W., Ye, L., Mai, Y.W., 2001. Mechanical properties of wood flake-
polyethylene composites. Part 1: effects of processing methods and matrix melt flow
behavior. Composites: Part A. 32:619-29.

Bledzki, A.K., Gassan, J., Theis, S. 1998. Wood-filled thermoplastic composites.
Mechanics of Compos Mater. 34(6):563-8.

Crawford, R.J., Plastics Engineering, Third Edition. 1998. Oxford: Butterworth-
Heinemann. p. 148-152.

Escamilla, G.C., Laviada, J.R., Cupul, J.I.C., Mendizabal, E., Puig, J.E., Franco, P.J.H.
2002. Flexural, impact and compressive properties of a rigid-thermoplastic
matrix/cellulose fiber reinforced composites. Composites: Part A. 33:539-49.

Hiziroglu, S., Kamdem, D.P. l995. A mechanical and physical properties of hardboard
made of black locust furnish. Forest Products J. 45(11/12):66-70.

Hull, D., Clyne, T.W., 1996. An Introduction to Composite Materials (second edition),
The press syndicate of the University of Cambridge, Cambridge.

Jusoh, 1., Kamdem, DR 2001. Laboratory evaluation of natural decay resistance and
efficacy of CCA-treated rubberwood. Holzforschung. 55(2): 250-254.

Kamdem, D.P., Sean, ST. 1994. The durability of phenolic bonded particleboard made of
decay-resistant black locust and non-durable aspen, Forest Products J. 44(2):65-68.

Kamdem, D.P., McIntyre, CR. 1998. A chemical investigation of 23 year old copper
dimethyldithiocarbamate (CDDC) treated southern pine. Wood and Fiber Sci. 30(1):64-
71.

Lide, D.R., editor. 2003. CRC Handbook of Chemistry and Physics. The 83rd edition.
Boca Raton, Florida: CRC Press LLC.

55

Matuana, L.M., Woodhams, R.T., Balatinecz, J.J., Park, CB. 1998. Influence of
interfacial interactions on the properties of PVC/cellulosic fiber composites. Polym
Compos. 19(4):446-55.

McHenry, E., Stachurski, Z.H. 2003. Composite materials based on wood and nylon
fiber. Composites: Part A. 34: 171-81.

Nass, L.I., editor. 1977. Encyclopedia of PVC. New York: Marcle Dekker, INC, p. 475.

Ray, D., Sarkar, B.K., Bose, NR. 2002. Impact fatigue behavior of vinylester resin
matrix composites reinforced with alkali treated jute fibers. Composites: Part A. 33:233-
41.

Song, X.M., Hwang, J.Y. 1997. A study of the microscopic characteristics of fracture

surface of MDI-bonded wood fiber/recycled tire rubber composites using scanning
electron microscopy. Wood and Fiber Sci. 29(2): 131-41.

Titow, W.V. 1984. PVC Technology, Fourth Edition. New York: Elsevier Applied
Science Publishers. p. 849-850.

Younan, M.Y.A., Rifai, M.A.E., Mohsen, R., Hennawi, I.M.E. 1983. Effect of stabilizer
type on the mechanical properties of rigid poly(vinyl chloride). J Appl Polym Sci
28:3247-3253.

Zhou, X.F., Wagner, H.D., Nutt, SR. 2001. Interfacial properties of polymer composites
measured by push-out and fragmentation tests. Composites: Part A. 32: 1543-1551.

56

Chapter 4
DSC AND DMA STUDIES OF POLY(VINYL CHLORIDE)/WOOD FLOUR
COMPOSITES

Thermal analysis is the measurement of changes in chemical or physical pr0perties
of a material as a function of temperature. Thermal analysis techniques, such as
Differential Scanning Calorimetry (DSC), and Dynamic Mechanical Analysis (DMA)
have proven to be among the most important and meaningful test methods in the
materials industry and in testing laboratories. Its power lies in understanding materials
behavior during manufacturing processes, in process optimization, reduction of
manufacturing cycle times, failure analysis, while enabling an overall improvement in

material properties and product quality (Gottfried et al. 2004).

4.1 Principles of DSC and DMA Analysis
4.1.1 DSC (Differential Scanning Calorimetry)

Differential Scanning Calorimetry (DSC) has been used for more than two decades.
It measures the temperature and heat flow associated with transitions in materials as a
function of time or temperature in a controlled atmosphere. Such measurements provide
quantitative and qualitative information about physical and chemical changes that involve
endothermic or exothermic processes, or changes in heat capacity. DSC is the most
widely used thermal analysis technique with applicability to polymer and organic
materials. DSC has many advantages including short analysis time, easy sample
preparation, broad temperature range, and excellent quantitative ability.

There are two types of DSC, power compensated DSC and heat flux DSC. In power

compensated DSC, sample and the reference material are heated by separate heaters and

57

temperatures of the sample and reference materials are kept equal. However, in heat flux
DSC, heat flows into both the sample and the reference materials via an electrical disk
and the differential heat flows to the sample and reference are monitored by
thermocouples and recorded by a detector. Figure 4.1 shows the schematic experimental
arrangement of a power compensated DSC in which the amount of energy supplied to or
withdrawn from the sample to maintain zero temperature as well as the differential
between the sample and the reference is recorded and displayed as the ordinate of the

thermal analysis curve.

Sample 5 Reference

Reference

I

Sample

\

\\

\\

 

 

\\
g

Heater Heater

sameIe//////////////////////////// Reference

base base

 

 

 

 

 

 

\\\\\\ .

 

 

 

Figure 4.1. Experimental arrangement of a power compensated DSC

58

DSC has been widely used to study material phase transitions (e.g. solid-solid, solid-
liquid, crystalline to liquid-crystalline and amorphous, liquid-gas), glass transitions
(change of the specific heat), and chemical reactions. Glass transition temperature (T g) is
the temperature at which a reversible change occurs in an amorphous polymer during
heating. Tg is characterized by a sudden transition from hard, glassy, or brittle state to a
flexible or elastomeric state (Wickson 1993). Glass transition temperature (Tg) is a
valuable characteristic parameter of a material and can provide very useful information
regarding the end-use performance of a product (Wickson 1993). Heat capacity (Cp) is
the amount of heat required to increase 1 kg of a material at 1 °C. Cp varies during the
glass transition (Wickson 1993, Billmeyer 1984, Painter 1994). The variation of Cp (ACp)
is defined as the difference in heat capacity before and after glass transition. ACp can be
obtained by DSC, and Tg is determined by the temperature point corresponding to the

half of ACp on the DSC diagram as shown in Figure 4.2.

Heetflawendoup(mW)

 

 

 

20 4o 60 so 100 120 140 160
Tenperature (’0)

Figure 4.2. Determination of glass transition temperature on a DSC temperature scan
curve

59

4.1.2 DMA (Dynamic Mechanical Analysis)

Dynamic Mechanical Analysis (DMA) measures the modulus (stiffness) and
damping (energy dissipation) properties of materials as the materials are deformed under
periodic stress. DMA is particularly useful to evaluate polymers with properties varying
with time, frequency, and temperature due to their viscoelastic nature.

DMA is a sensitive technique for characterizing the mechanical behavior of materials.
Flexural bending or shear deformation is applied to produce strain within a sample. In
DMA, the complex modulus (E*) is separated into two components: elastic modulus (E’)
and loss modulus (E”).

E*=E’+iE” (4-1)
Tan 8 = E”/E’ (4-2)

E’ is a measure of the elasticity of the material and represents the ability of the
material to return or store energy; E” is an imaginary modulus indicating its ability to
lose or dampen energy for friction and internal motions. The ratio of these two effects is
called tan 8, damping. Tan 5, the ratio of E”/E’ is an indicator of how efficiently the
material loses energy to molecular rearrangements and internal friction (Menard 1999).

Figure 4.3 illustrates and shows the significance of E*, E’, E”, and tan 5.

E!

E* E

 

Figure 4.3. DMA relationship of the complex modulus (E*), elastic modulus (E’), loss
modulus (E”), and tan 8.

60

A similar relationship is used to define the complex viscosity, 11* with DMA. 11* is

the tendency of a material to flow. 11’ is its loss viscosity and 11” is the stored viscosity as

expressed in the equation below (Menard 1999).

n*=n’- in” (4-3)

4.2 Applications of DSC and DMA in Wood-Plastic Composites

DSC and DMA have been applied in WPC area to study the influence of the presence
of wood fiber (flour) and compatibilizer as well as wood chemical modification on
thermal behavior of various polymer matrixes (Yap et al. 1991, Kolosick et al. 1992, Kim
et al. 1997, Chuai et al. 2000, Balasuriya et al. 2002, 2003, Liao and Wu 2003, Espert et
al. 2004, Specht et al. 2004). Espert et al. studied thermal behavior of different natural
fiber-filled polypropylene composites by using DSC (Espert et al. 2004). They found the
crystallization and melting temperatures of PP decreased as the fiber content increased.
Reduction in the crystallization temperature of PP was also found on maleated
polypropylene (MAPP) treated wood surfaces (Kolosick et al. 1992). Wood flour acted
as nucleating agents for PP, and B-transition showed that only MAPP treated wood flour
can induce the crystallization of PP (Nunez et al. 2002). Kinetics study found no
differences in the crystal formation nucleated on the wood surface and in the bulk
polymer PP (Harper and Wolcott 2004). But the total degree of crystallinity of PP may be
decreased by the presence of untreated wood flour (Malainine et al. 2004). A similar
wood flour nucleating effect was also found in wood fiber-poly(hydroxybutyrate) (PHB)
and wood fiber-poly(hydroxybutyrate-co-hydroxyvalerate) (PHB/HV) composite systems

(Reinsch and Kelley 1997). Wood fiber increased the crystallization rate of PHB and

61

PHB/HV; however, the ultimate crystallinity was the same in wood fiber reinforced and
unreinforced materials (Dufresne et al. 2003).

DMA studies on MAPP treated wood-PP composites showed that composite
properties decreased for wood flour content higher than 40 wt% (Nunez et al. 2002, Son
et a1. 2003). MAPP had negligible effects on the main transitions of PP while the effect of
fiber content on the intensity and temperature of or-transition was almost proportional to
fiber contents in composites (T ajvidi et al. 2003). Selden et al. reported their DSC results
on UV aged wood fiber-PP composites. The second DSC melting scan of degraded
surface layers indicated a maximum 33 OC decrease in the melting temperature of PP
because of the molecular chain scission and the formation of extraneous groups, such as
carbonyls and hydroperoxides during UV aging (Selden et al. 2004).

For wood-polyethylene (PE) composites, DSC has also been used to evaluate the
crystallinity of HDPE after aging (Sunol and Saurina 2002). Good correlations were
detected between aging time and the degree of crystallinity of PE as well as between
melting enthalpy and wood fiber content. Silane coupling agent treated wood fiber had a
higher loss of crystllinity of PE. A small reduction of the melting peak temperature was
found as the aging time increased (Sunol and Saurina 2002). Increased crystallinity of
PE was observed with increasing wood content up to 50%; the a-transition temperature
and crystallization temperature of PE were shifted considerably in unmodified
composites on DMA curves, and after adding 2 wt% maleic anhydride (MM—modified
polyethylene (MAPE), glass transition temperature shifted to lower temperature range

(Balasuriya et al. 2003).

62

Wang et al. studied boric acid-modified wood fiber-polystyrene (PS) composites by
using DSC and DMA (Wang et al. 1996). They reported that the glass transition
temperature of PS was lowered by the addition of plasticizer, while the presence of boric
acid had little effect.

Djidjelli et al. reported that wood fiber content had little influence on the glass
transition temperature of plasticized PVC that appeared around 55°C on the DSC
therrnograms. They also found the inhibiting effect of wood fiber on PVC thermal
decomposition (Djidjelli et al. 2002). DMA and DSC studies of MAPP-treated wood
veneer-PVC laminated composites showed that storage modulus of composites increased
at low MAPP content and decreased at high MAPP content, but tan6 was independent of
MAPP concentration (Lu et al. 2004). The presence of plasticizer decreased the viscosity
of wood-PVC composites, and acrylic impact modifier increased the viscosity of PVC
while reducing the viscosity of composites (Shah et al. 2004).

Thermal and dynamic mechanical properties of wood plastic composites made of
PVC and copper amine treated, and untreated wood flour were characterized by using
differential scanning calorimetry (DSC) and dynamic mechanical analysis techniques
(DMA). Influence of the addition of wood flour and copper amine treatments on thermal
behavior and glass transition temperature (Tg) of PVC, as well as the heat capacity,
storage and loss modulus, tan5, and viscosity of wood flour-PVC composites were

investigated.

63

4.3 DSC Experimental Materials and Methods
4.3.1 Sample Preparation

Untreated and copper amine-treated wood flour-PVC composites were laboratory
manufactured following procedures described in Chapter 2. Sample compositions and Cu
concentrations in wood flour are listed in Table 4.1. Pure PVC and composite boards then

were ground to 50 mesh (0.3 mm) particles.

Table 4.1. Compositions of Samples with PVC, Treated and Untreated Wood Flour
Contents, as well as Cu Concentrations in Treated Wood Flour

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample ID PVC3 Wood flour Cu content in wood flour
(wt%) (wt%) (wt%)
PW64con 60 40 0 (untreated)
PW64Cu0.1 60 40 0.1
PW64Cu0.2 60 40 0.2
PW64Cu0.5 60 40 0.5
PW64Cu0.6 60 40 0.6
PW64Cu0.7 60 40 0.7
PW64Cul.2 60 40 1.2
4.3.2 Experimental Methods

DSC temperature scans were carried out with a Perkin-Elmer Diamond DSC, and
PYRISTM, Version 4.0 software were used for data acquisition and processing. The
instrument was calibrated with a standard indium sample before analyzing.

About 10 mg of each type of sample was sealed in an aluminum pan and placed in
heating chamber and operated under N2 atmosphere with a pressure of 150 kPa.
Experiment was conducted in 3 steps. In step 1, the samples were heated from 25 °C to
160 °C at a rate of 10 °C /min, and cooled from 160 °C to 25 0C at a rate of 10 °C /min, in
order to eliminate material heat history. Finally they were re-heated from 25 °C to 160 °C

at a rate of 10 °C /min. The therrnogram obtained in the reheating step was used for data

64

 

processing and interpretation. Duplicate samples were conducted for each type of

samples.

4.4 DMA Experimental Materials and Methods
4.4.1 Sample Preparation

Untreated and copper amine-treated wood flour-PVC composites were laboratory
manufactured following procedures described in Chapter 2. Sample compositions and Cu
concentrations in wood flour are listed in Table 4.2. Pure PVC and composite boards

were cut into 1.2 by 3.5 by 15 mm specimens.

Table 4.2. Compositions of Experimental Samples with PVC, Treated and Untreated
Wood Flour Contents, as well as Cu Concentrations in Treated Wood Flour

 

 

 

 

 

 

 

 

 

 

 

 

Sample ID PVC2 Wood flour Cu content in wood flour
(wt%) (wt%) (wt%)
PW64con 60 40 0
PW64Cu0.2 60 40 0.2
PW55con 50 50 0
PW55CuO.2 50 50 0.2
PW46con 40 6O 0
PW46Cu0.2 4O 60 0.2
4.4.2 Experimental Methods

DMA was carried out by a Perkin-Elmer DMA7, and PYRISTM, Version 4.0 software
was used for data acquisition and processing. A 3-point bending—rectangle device testing
system was installed. A 110.0 mN static force and a 100.0 mN dynamic force were
applied on the sample with an oscillating frequency of 1.0 Hz. Testing was conduct under
helium atmosphere with a flow rate of 10.0 ml/min. The sample was heated from 25 °C to

160 °C at a heating rate of 5 °C lmin. Five replicate specimens per type were scanned.

65

 

4.5 Results and Discussion
4.5.1 DSC Glass Transition Temperature (Tg)

PVC generally has a Tg around 80°C to 85°C (T itow 1984). The Tg of cellulose is
between 230 to 250 °C, whereas lignin has a Tg of 90-100 °C at wet condition and 130-
190°C at dry state (Fengel et al. 1984).

Figure 4.4 shows the DSC temperature scan diagrams of pure PVC, composite
sample PW64con and PW64Cu0.2. The glass transition step of PVC was evident in the
DSC curve. After PVC was blended with wood flour, the a-transition step became weak,
especially with Cu-treated wood. Glass transition temperatures of composites and PVC
resin were around 83:1:1°C and illustrated in Figure 4.5. The effect of wood flour content
and Cu concentrations in WP on the glass transition temperatures of PVC was not

considerable in DSC analysis.

 

 

 

 

 

E / i - 3
a. a
a

O

E i
3 '/

.2

ll.

‘5 - 1
g J —

 

[ l l l T P l l l l l I

50 60 70 80 90 100 110 120 130 140 150 160
Temperature (“0)
Figure 4.4. DSC diagrams of (1) pure PVC; (2) PW64con—composite made of PVC and
untreated WF at a percentage PVC to wood ratio of 60 to 40; (3) PW64Cu0.2- composite

made of PVC and treated WF containing 0.2% Cu at a percentage PVC to wood ratio of
60 to 40

66

100.0 ~
95.0 .
90.0 —
85.0
80.0
75.0 .
70.0 -
65.0 .

 

T9 (’6)

 

 

60.0 -—e—~—e. ”2-. . - T - a

PVC 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Cu concentration in wood (%)

Figure 4.5. Glass transition temperatures of PVC resin and PVC/Cu—treated, untreated
WF composites vs. Cu concentrations in WF

4.5.2 DSC Heat Capacity Differences (ACp)

Although wood content and copper treatment had little effect on the PVC glass
transition temperature, the difference of heat capacity (ACp) of these composites were
drastically reduced compared to pure PVC. ACp of PVC, PVC-untreated wood flour
composites, and Cu-treated composites with different Cu contents are given in Figure 4.6.

Heat capacity of solid PVC is 1.2 kJ/kg-K (Nass 1977), almost similar to that of
ovendry wood 1.2 to 1.3 kJ/kg-K (17-27 °C). The Cp of PVC and wood are very high,
almost 3 folds of that of copper, 0.385 kJ/kg-K (Nass 1977). Below glass transition
temperature of PVC, macromolecule chains of PVC are within the frozen state. When

temperature reaches Tg, large segments of main chain of PVC start moving, resulting in

67

the increase of free volume. More heat is needed for heating the material 1 °C so that the
heat capacity (Cp) enlarges during glass transition in PVC polymer (Painter 1994). In
PVC-wood flour composite, when PVC was experiencing its glass transition, the wood
polymer chains were still in frozen state because the temperature was far below glass
transition temperatures of lignin and cellulose. The presence of rigid wood polymer
chains in PVC matrix may block the movement of adjacent PVC chain segments,
preventing the expansion of free volume resulting in low Cp increases compared to those
in pure PVC, and this explains the reduction of ACp. After wood Cu-treatment, a stronger
connection between PVC and wood polymer chains was built up, and as a result, the

movement of PVC chain segments can be further inhibited.

400.0 .
350.0 3
300.0 3
250.0 :
200.0 3
150.0 .
100.0 i
50.0 3

0.0 l

ACP (NJ/9°C)

 

 

 

PVC 0.0 0.1 0.2 0.5 0.6 0.7 1.2

Cu concentrations In wood (%)

Figure 4.6. ACp of PVC resin and PVC/Cu-treated, untreated WF composites vs. Cu
concentrations in WF

ACp of composites made with PVC and Cu-treated, and untreated WF was related to
the ACp of PVC, treated and untreated WF, as well as their weight percentages in the

composite by using the following expressions:

68

ACpPVC/WF = f(ACpPVC’ACpWF’wPVC’a)WF) (4-4)

ACpPVC/Cu-WF = f(ACpPVC’ACpCu—WF’wPVC’wCu-WF) (4-5)
where wPVC , 04W.- , wok,” are weight percentages of PVC, untreated WF, Cu treated WF

respectively. The following two-variable linear model was built to simulate and predict

the heat capacity change of composites made of PVC and Cu-treated, untreated WF:

ACpPVC/WF = wPVC ' ACpPVC + wwr ' ACPWF " [PVCIWF ° wpvc ‘ wwr (4-6)

ACpPVC/Cu-WF = wPVC ' ACpPVC + wCu-WF ' ACpCu-WF _ IPVC/Cu—WF ’ wPVC ' wCu-WF

(4-7)
where IpVC/wp and IpVC/Cu_WF are the probable interaction parameters between

PVC and untreated and Cu-treated WF. [WC/Cu-”- also varies with the copper

concentration, wcu of treated WF as expressed below:

_ 2
IPVC/Cu—WF " awa. + wau + C (4-8)
where a , b , c are constants and can be obtained by the following polynomial equation:

[WWW = —197.9w§u ”80.8%, +114.9 , ,h, R2 of the {mess of

this equation with the curve is 0.9975.

So, a = -197.9, b = 380.8, and c: 114.9.

When 60C“ = 0 (untreated wood flour), IPVC/Cu-wp will become Ipvc/wp , and

[PVCIWF = C (4-9)

69

Since the heat capacities of untreated and treated WF did not change during the glass
transition of PVC, ACpWF = 0 , and ACpCu-WF = 0 , therefore, values of ACp of

composites were calculated, and the results are listed in Table 4.3, which was in

agreement with the experimental values.

Table 4.3. DSC Measured Heat Capacity Difference (ACp) of PVC and PVC/Cu-Treated,
Untreated WF Composites in Comparison with the Modeling Results

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Composite sample ID ACp (mU/g °C) ACp (mJ/g °C)
(Measured) (Predicted)
Pure PVC (190F) 305 305
PVC/untreated WF = 60/40 157 156
PVC/Cu-treated WF = 60/40, Cu% = 0.1 146 147
PVC/Cu-treated WF = 60/40, Cu% = 0.2 132 138
PVC/Cu-treated WF = 60/40, Cu% = 0.5 119 121
PVC/Cu-treated WF = 60/40, Cu% = 0.6 118 118
PVC/Cu-treated WF = 60/40, Cu% = 0.7 118 116
PVC/Cu—treated WF = 60/40, Cu% = 1.2 114 114
PVC/untreated WF = 50/50 127 124
PVC/Cu-treated WF = 50/50, Cu% = 0.5 87 89
PVC/untreated WF = 40/60 96 94
PVC/Cu-treated WF = 40/60, Cu% = 0.5 64 61
4.5.3 DMA Modulus (E’ and E”)

Figure 4.7 shows storage modulus (E’) of PVC resin and PVC/wood flour
composites in DMA temperature scans. It can be seen that composites had higher storage
modulus than pure PVC through the whole temperature scan range. As wood flour
content enlarged from 40 wt% to 50 wt%, storage modulus increased about 30% to 100
MPa, but when wood content reached 60%, storage modulus decreased. This in

agreement with the results reported in literature (Nunez et al. 2002, Son et al. 2003)."

70

After a certain filling limit, PVC was insufficient to cover the large amount of wood
flour, which resulted in poor adhesion and property deterioration of the composites. The
rubbery plateau of PVC after glass transition was dramatically shorten by the presence of
wood flour and almost disappeared when wood content increased to 60 wt%. Composites
can maintain higher storage modulus during a long temperature range above Tg of PVC,
especially for composites containing 50 wt% and 40 wt% wood flour. 'Iheir E’ above Tg
of PVC was even higher than the E’ before Tg of the composite containing 40 wt% wood
flour. Although the composite with 60 wt% wood had a lower E’ before Tg than that of
the composite containing 50 wt% wood flour, its E’ above Tg was higher than that of 50

wt% wood contained composite in a certain temperature range.

Pwsscon

Storage modulus, E' (MPa)

 

10.0. \
0.0 . - . . -2 e e e

 

N.
\~
‘.
~-
_ __
*~~~~_._
—

 

25.0 35.0 .450 55.0 65.0 75.0 85.0 95.0 105.0 115.0 125.0 135.0 145.0
Temperature (°C)

Figure 4.7. DMA temperature scans of storage modulus (E’) of PVC resin and PVCIWF
composites with increased WF content

71

Loss modulus showed similar trends to storage modulus. 'Ihese loss modulus of PVC
and PVC-wood flour composites are illustrated in Figure 4.8. Composites showed larger
E” than PVC, and E” increased as enlarging wood flour content except for composites
with 60 wt% or more wood. Above Tg E” of composites made of 60 wt% of wood higher
than those of composites containing 50 wt% wood until temperature increased to around
140 °C. E’ represents the elastic nature of material, while E” is related to the energy lost
to friction and internal motions reflecting viscous behavior (Menard 1999). The addition
of wood flour led to the increase in both elastic and viscous abilities of composites under

the oscillation load in comparison with PVC.

.81 .3
O O

8
o

 

25.0 .

15.0 .

10.0 -

Loss modulus, E" (MPa)
8

5.0 .

 

 

0.01....-..fi.-...-r.,..jfi
25.0 35.0 45.0 55.0 65.0 75.0 85.0 95.0 105.0 115.0 125.0 135.0 145.0

 

Temperature (°C)

Figure 4.8. DMA temperature scans of loss modulus (E”) of PVC resin and PVCIWF
composites with increased WF content

72

4.5.4 DMA tan5

Tan 5 thermograms of PVC, composites made of PVC, and 40, 50, 60 wt% wood
flour are presented in Figure 4.9. Pure PVC exhibited the largest tan 5 peak with a peak
temperature (PVC Tg by DMA) of 85.8 °C. Tan 5 peaks were drastically weakened by the

presence of wood flour, and disappeared when wood amount increased to 60 wt%. A
slight increase in Tg (from 86.9 to 88.7 °C) was observed when wood content enlarged
from 40 wt% to 50 wt%. Tan 5 reflects the mobility and movement capacity of molecule
chain segments during glass transition. It is clear that the movement of PVC chain
segments during glass transition was significantly limited and obstructed by the presence

of rigid wood molecule chains resulting in much smaller tan 5 peaks in composite

 

0501 PVC
1 -- ----- PW64con
0.50 q ...... PWSScon
.................. pw46con
0.40 .
‘1? 030
8 .

0.20 -

 

 

0.10 ~

 

 

o-m ‘ 1 ' 1 v f ' ' fir 4 1 v 1 1 1 1 1 1 r 1 . 1
25.0 35.0 45.0 55.0 65.0 75.0 85.0 95.0 105.0 115.0 125.0 135.0 145.0

 

Temperature (°C)

Figure 4.9. Tan 5 thermograms of PVC resin and PVCIWF composites with increased
WF content

73

samples. This restriction effect became stronger as increasing wood flour content. Wood
flour particles acted as “physical cross-linking points” inside PVC matrix connecting

PVC macromolecule chains and hindered their movements during glass transition.

4.5.5 Influence of Wood Copper Treatment

E’, E”, and tan 5 of composites made of PVC-Cu treated, and untreated wood flour
with 60 wt% PVC and 40 wt% wood are displayed in Figure 4.10. Wood copper amine
surface modification exhibited a similar effect to increasing wood content on E’, E”, and
tan 5 of the composites. After using Cu-treated wood flour, E’, and E” increased and
PVC rubbery plateau was further weakened indicating the same “chain movement
restriction” effect as increasing wood content in composites. This was also in
correspondence with a smaller tan 5 peak and slightly increased Tg of Cu-treated
composite compared to those of the untreated composite. DMA analysis results of Cu-
treated and untreated composites revealed that stronger adhesion and connection between
PVC and wood polymer chains was built up by wood flour copper amine treatment. This
may be attributed to the formation of stronger “physical cross-linking” or even some
amount of “chemical cross-linking” between PVC and Cu-treated wood molecule chains
(this chemical interaction will be discussed in Chapter 6). As a result, the movability of
PVC chain segments was further lowered by Cu-treated wood flour in the same wood

content level as untreated composites.

74

. 0.32

. 0.31

 

 

t 0.30
, t 0.29

" 8
'. £10.28 8

L

Modulus (MPa)

~ 0.27

~ 0.26

 

'. 0.25

 

 

 

 

o_02.2--..--,-22..,2...s,. 0.24
25.0 35.0 45.0 55.0 65.0 75.0 85.0 95.0 105.0 115.0 125.0 135.0 145.0

Temperature (°C)

Figure 4.10. E’, E” and tan 5 of (a) PW64con: PVC-untreated wood flour composites
with a 40 wt% wood flour content. (b)PW64CuO.2: PVC-Cu treated wood flour
composite made of 40 wt% wood flour which had a 0.2 wt% Cu

4.5.6 Viscosity Measured by DMA

The energy loss portion of viscosity of pure PVC and PVC-treated, and untreated
wood flour composites are presented in Figure 4.11. Composites showed higher
viscosities than pure PVC, and viscosity increased as enlarging wood content in
composites except for the composites containing 60 wt% wood flour. Compared to the
composite containing 50 wt% untreated wood flour, composites made of 40 wt% PVC,
and 60 wt% untreated wood flour exhibited lower viscosity before glass transition of
PVC due to the poor interfacial adhesion, and higher viscosity after glass transition (from
88 to 140 °C) because of the physical cross-linking effect. Composite containing 40 wt%

Cu-treated wood flour demonstrated higher viscosities than corresponding composite

75

.‘J
o

 

 

 

6.0 . PWELchn ___________
“—PW46con ..... ‘
A -....._._'.‘.' ............ "....v-
3 so Pwucuo 2
°- Pwucon
g
‘5
5‘ 3.0
3
PVC
... 2.0 ..............................
> .
1.0.
0.0 - 2 E g f . - a

 

 

25.0 35.0 45.0 55.0 65.0 75.0 85.0 95.0 105.0 115.0 125.0 135.0 145.0

Temperature (°C)

Figure 4.11. Loss viscosity of PVC and PVC-untreated wood flour composites with 40,
50 and 60 wt% wood flour content as well as PVC-0.2 wt% Cu containing wood flour
composite with a 40 wt% wood flour content

made of untreated wood suggesting the improved interfacial adhesion between PVC and

Cu-treated wood flour.

4.6 Conclusions

DSC study showed slight increases in glass transition temperature (Tg) of PVC by
the addition of untreated and Cu-treated wood flour but this effect was not considerable.
Heat capacity differences (ACp) of composites before and after glass transition were
dramatically reduced by the presence of wood flour and wood Cu-treatment. A two-
variable linear model successfully simulated the alteration of ACp in terms of weight
percentages of PVC and wood flour as well as their ACp by introducing a PVC-wood

interaction parameter, Ipvc-cuw]=. The interaction parameter between PVC and copper

76

amine treated wood flour was directly correlated to the copper concentrations in wood
through a polynomial equation. Extreme conditions in this model associated with PVC
alone ((Dpvc=100%) and untreated wood flour (Cu%=0) can also be defined.

Tg of PVC in composites was also determined from the peak temperature of tan 5.
By DMA. Tg slightly shifted to the high temperature range by the presence of wood flour
compared to PVC and by wood flour Cu treatment compared to untreated composites.

It can be seen that composites had increased E’, E”, and viscosity, but had much
smaller tan 5 peaks than unfilled PVC. Enlarging wood flour content or using Cu-treated
wood increased E’, E”, and viscosity, but weakened tan 5.

DMA study revealed that the movement of PVC chain segments during glass
transition was limited and obstructed by the presence of rigid wood molecule chains. This
restriction effect became stronger as increasing wood flour content and by using Cu-
treated wood flour. Wood flour particles acted as “physical cross-linking points” inside
the PVC matrix resulting in the absence of the rubbery plateau and higher E’, E” above
Tg, and smaller tan 5 peaks.

Compared to the composite containing 50 wt% wood flour, when wood content was
enlarged to 60 wt%, this composite exhibited a decrease in E’, E” and viscosity before
glass transition due to the poor interfacial adhesion, but an increase in E’, E”, and
viscosity after glass transition (from 88 to 140 °C) because of the “physical cross-linking”

effect.

77

References

Balasuriya, P. W., Ye, L., Mai, Y.-W.; Wu, J. 2002. Mechanical properties of wood
flake-polyethylene composites. II. Interface modification, Journal of applied Polymer
science, 83(12): 2505-2521.

Balasuriya, P. W., Ye, L., Mai, Y.W. 2003. Morphology and mechanical properties of

reconstituted wood board waste-polyethylene composites, Composite Interfaces, 10(2-3):
319-341.

Billmeyer, F. W. 1984. Textbook of polymer science, Wiley, New York.

Chuai, C., Li, 8., Lin, Y. 2000. Study on compatibility and properties of
polypropylene/grafted wood fiber composites. Zhongguo Suliao. 14(5): 23-28.

Djidjelli, H., Martinez-Vega, J. J., Farenc, J ., and Benachour, D. 2002. Effect of wood flour
content on the thermal mechanical and dielectric properties of poly(vinyl chloride).
Macromol. Mater. Eng., 287: 611-618.

Dufresne, A.; Dupeyre, D., Paillet, M. 2003. Lignocellulosic flour-reinforced
poly(hydroxybutyrate-co-valerate) composites. Journal of Applied Polymer Science,
87(8): 1302-1315.

Espert, A., and Karlsson, S. 2004. Characterization and comparison of the thermal and
mechanical properties of different natural fiber-filled polypropylene composites.
International Conference on Woodfiber-Plastic Composites, 7th, Madison, WI, United
States, May 19-20, 2003, Meeting Date 2003, 293-298.

Fengel, D., and ngener, G. 1984. Wood, Chemistry, Ultrastructure, Reactions. Walter
de Gruyter, New York.

Gottfried, W., Ehrenstein, G., Riedel, T. 2004. Thermal Analysis of Plastics: Theory and
Practice. Hanser Gardner, Cincinnati, Ohio, 396pp.

Harper, D., Wolcott, M. 2004. Interaction between coupling agent and lubricants in
wood-polypropylene composites. Composites, Part A: Applied Science and
Manufacturing, 35A(3): 385-394.

Kim, R. K., Kang, M., Kim, J. P., Kim, Y. H.. Lee, J. 8., Moon, S. H., Yoon, T. H. 1997.
1997. Wood-polymer composites with recycled polyethylene films. International
Conference on Woodfiber-Plastic Composites, 4th, Madison, Wis., May 12-14, 1997,
275-279.

Kolosick, Paul C., Myers, G. E., Koutsky, J. A. 1992. Polypropylene crystallization on
maleated polypropylene-treated wood surfaces: effects on interfacial adhesion in wood-

78

polypropylene composites. Materials Research Society Symposium Proceedings, 266
(Materials interactions relevant to recycling of wood-based products), 137-54.

Lu, J. Z., Wu, Q., Negulescu, I. I. 2004. Surface and interfacial characterization of wood-
PVC composites: Thermal and dynamic mechanical properties. Wood and fiber science,
36(4): 500-510.

Malainine, M.E., Mahrouz, M., Dufresne, A. 2004. Lignocellulosic flour from cladodes
of Opuntia ficus-indica reinforced poly(propylene) composites. Macromolecular
Materials and Engineering, 289(10): 855-863.

Menard,K.P. 1999. Dynamic Mechanical Analysis. CRC Press LLC, Boca Raton,
Florida.

Nass, L. I, editor, 1977. Encyclopedia of PVC, Marcel Dekker, Inc, New York and
Basel, New York.

Nguyen, T., and W. E. Johns. 1978. Polar and dispersion force contributions to the total
surface free energy of wood. Wood Sci. Technol. 12: 63-74.

Nunez, A. J., Kenny, J. M., Reboredo, M. M., Aranguren, M. I., Marcovich, NE. 2002.
Thermal and dynamic mechanical characterization of polypropylene-woodflour
composites. Polymer Engineering and Science, 42(4):?33-742.

Painter, P. C. 1994. Fundamentals of Polymer Science: An Introductory Text. :
Technomic Pub. Co. Lancaster, PA.

Selden, R., Nystroem, B., Langstroem, R. 2004. UV aging of poly(propylene)/wood-
fiber composites. Polymer composites, 25(5): 543-553.

Shah, B. L., Matuana, L. M. 2004. Online measurement of rheological properties of
PVC/wood-flour composites. Journal of Vinyl & Additive Technology, 10(3): 121-128.

Son, J., Gardner, D. J., O'Neill, S., Metaxas, C. 2003. Understanding the viscoelastic
properties of extruded polypropylene wood plastic composites. Journal of Applied
Polymer Science. 89(6): 1638-1644.

Specht, K., Bledzki, A. K. 2004. Influence of fiber treatment on the thermal longtime
behavior of wood and hemp fiber-polypropylene composites. International Conference
on Woodfiber-Plastic Composites, 7th, Madison, WI, United States, May 19-20, 2003,
Meeting Date 2003, 109-117.

Sunol, J. J., Saurina, J. 2002. Thermal analysis of aged HDPE based composites.
Journal of Thermal Analysis and Calorimetry, 70(1): 57-62.

79

Tajvidi, M., Ebrahimi, Gh., Enayati, A. A. 2003. Dynamic mechanical analysis of
compatibilizer effect on mechanical properties of wood flour-polypropylene composites.
Majallah-i Manabi-i Tabi-i Iran, 56(l,2):47-59.

Titow, W. V. 1984. PVC technology. Elsevier Applied Science Publishers. London and
New York.

Wang, Y., Simonsen, J., Pascoal Neto, C., Rocha, J., Rials, T. G., Hart, E. 1996. The
reaction of boric acid with wood in a polystyrene matrix. Journal of Applied Polymer
Science, 62(3): 501-508.

Wickson, E. editor. 1993. Handbook of PVC Formulating. A Wiley-interscience
Publication, John Wiley & Sons. New York. P.17.

Yap, M. G. 8., Que, Y. T., Chia, L. H. L., Chan, H. S. O. 1991. Thermal properties of

tropical wood-polymer composites. Journal of Applied Polymer Science, 43(11):
2057-65.

80

Chapter 5

SURFACE AND INTERPHASE CHARACTERIZATION OF PVC/COPPER
AMINE-TREATED WOOD FLOUR COMPOSITES

5.1 Background

The surface of a solid material is characterized by its surface energy, acid-base
properties and chemical composition. There is a strong correlation between these surface
parameters and the strength performance of the resulting composite (Kamdem et al. 1993,
1992a,b, 1991b,c, Kazayawoko et al. 1997a, b, 1999). The interfacial adhesion of the
wood flour to a PVC matrix is strongly influenced by the surface characteristics of the
wood and the matrix. The surface energy parameters of untreated and Cu-treated wood
are crucial to evaluating the wettability and interfacial adhesion between the wood
surface and PVC matrix. This evaluation will aid in a better understanding and prediction
of the adhesion mechanisms between wood flour and the PVC matrix and in identifying
the major factors that govern the performance characteristics of the PVC/wood flour

composites.

5.1.1 Fiber-Matrix Interphase Bonding Mechanisms

Fiber-matrix adhesion is essential for predicting the physical and mechanical
performance of composites (Drzal et a1, 2001, 1993, l983a,b,c, Madhukar et al. 1992).
Strong adhesion in the fiber-matrix interphase is needed to effectively transfer stress and
distribute load throughout the interphase (Hull and Clyne, 1996, Ulkem and Schreiber

1994). Several adhesion mechanisms have been proposed, including:

81

(b)

(C)

(d)

(e)

 

+ + + +
7
A A A A
/
B/ B B/ B/

.\

2

Figure 5.1. Adhesion mechanisms (2) molecule chain entanglement following
interdiffusion, (b) electrostatic attraction, (c) cationic groups at the end of molecules
attracted to an anionic surface, leading to molecule orientation on the surface, (d)
chemical reaction and (e) mechanical keying

82

adsorption and wetting; molecule entanglement following interdiffusion; electrostatic
attraction; cationic-anionic group attraction; chemical reaction and mechanical
interlocking (Pizzi 1983, Hull and Clyne 1996). Bonding mechanisms are demonstrated
in Figure 5.1.

Adhesion is a thermodynamic process characterized by a level of the adhesion
strength controlled by enthalpy at equilibrium. The first step of adhesion is wetting (Hull
and Clyne, 1996, Drzal 1995, l983c, Ulkem and Schreiber 1994). Satisfactory wetting is
a requisite for the thermodynamic demands of adhesion, and this can only be generated
when the surface energy of the wetting substance (liquid or molten) is lower than that of
the solid substrate (Ulkem and Schreiber 1994, Adamson and Gast 1997).

Many experimental methods have been conducted to evaluate and estimate fiber-
matrix interfacial adhesion. These methods can be categorized into the direct method, the
indirect method and the composite lamina method (Drzal 2001, Ho and Drzal 1996).
Direct methods include single-fiber pull-out, push-out and push-down, single fiber-
fragmentation, embedded fiber compression, micro indentation and microbond test;
Indirect methods include the variable curvature method, slice compression, and voltage
contrast X-ray spectroscopy. Composite lamina methods include: short-beam shear tests,
90° transverse flexural and tensile tests (Drzal 2000, 2001, Ho and Drzal 1996, Herrera-

Franco 1992)

83

5.1.2 Wood Fiber (Flour)-Polymer Interphase

There are two ways to improve wettability of polymer melts to wood fiber surfaces:
decrease surface energy of polymer melts or/and increase surface energy of wood. Most
polymers are hydrophobic, while wood is hydrophilic. The compatibility between wood
and polymer can be improved by reducing hydrophobicity of polymer or hydrophilicity
of wood. Physical and chemical treatments have been used to modify the wood surface to
achieve a better adhesion between wood and the polymer matrix (Mahlberg et al. 1998,
Shi and Drzal 2003, Gardner and Walinder 2001, Jiang and Kamdem 2004, Mohanty et
al. 2002).

Plasma has been reported to be an effective method to increase wood surface energy
(Shi and Drzal 2003, Mahlberg et al. 1998, Yang et al. 1999). The interfacial shear
strength between epoxy and natural fiber pre-treated for 8 minutes in oxygen plasma,
increased over 100% (Shi and Drzal 2003). Plasma irradiation was also found to be
effective in the enhancement in bonding strength between cellulosic fiber and
polyethylene (Yang et al. 1999).

It has been reported that wood surfaces became hydrophobic after incomplete
extraction with organic solvents, such as ethanol-toluene (Maldas and Kamdem 1999,
Kamdem et al. l99la,b,c, 1992a,b, 1993). Increased dispersion component of wood
surface energy, acidity and basicity have been found after wood extraction with
acetone/water, dichloromethane, and ethanol/benzene by IGC (inverse gas
chromatography) studies (Liu et a1. 1998, Simonsen et al. 1997).

Gardner and Walinder reported IGC and contact angle experimental results on wood

surfaces coated with polystyrene-b-poly(styrene-co-acrylic acid) (PSAA), which was an

84

potential compatibilizer for wood-polystyrene composites (Gardner and Walinder 2001,
Gunnells et al. 1994, Walinder and Gardner 2002). Hydrophobicity and surface energy of
wood particle surfaces were increased after wood surface PSAA treatment. Similar
results were also obtained on wood surfaces treated with lignin-styrene graft c0polymer,
and both water angle and binding strength between polystyrene and wood increased
because of this surface modification (Chen et al. 1995).

Silane and maleated-polyethylene (MAPE) and maleated-polypropylene (PAPP) are
popular coupling agents and compatibilizers used in composites. However, after
treatment with silane or MAPE, or MAPP, wood surface energy decreased and became
more hydrophilic (Matuana et al 1998, Kazayawoko et al. 1999, Lu et al. 2005).
Increased adhesion between amino-silane treated wood and PVC was attributed to the
possible acid/base interactions and chemical reactions that occurred in interphase
(Matuana et a1 1998). The mechanical interlocking was reported to be the main reason for
the adhesion enhancement in polypropylene-MAPP wood fiber composites (Kazayawoko
et al. 1999).

Maldas and Kamdem found that after chromated copper arsenate (CCA) preservative
treatment, the hydrophobicity of wood surfaces was significantly increased (Maldas and
Kamdem 1998). The dispersion component (7“) of a wood surface was also increased by
wood CCA-treatment. Composites made of high-density polyethylene and CCA-treated
wood flour exhibited increased mechanical properties resulting from better wetting and
improved interfacial compatibility (Kamdem et al. 2004). Similar mechanical property
improvement had been found when using copper ethanolamine-treated wood flour-PVC

composites discussed in Chapter 3 (Jiang and Kamdem 2004).

85

5.2 Theoretical Calculation of Surface Energy and Surface Energy Components
Contact angle has been widely used to gain information on the surface energy of
solids. This information is valuable to understand and predict the adhesion between a
solid and a liquid.
The relation between contact angle and surface tensions of a liquid and a solid is

described by Young’s equation:

71 0086' = 75 - 7.11. (54)

where 0 is the contact angle, Yr. is the surface tension of liquid, 78 is the surface energy of
the solid and 751. the surface tension of the solid-liquid interface (Adamson and Gast
1997).

In addition, the Young-Dupre equation defines the work of adhesion (WSL), as the
work needed to separate the liquid-solid interface (Adamson and Gast 1997).

WSL = yL(l + cos 6) (5-2)

The surface energy is decomposed into two components, y°s and yPs, in the Young-
Good-Girifalco-Fowkes geometric and harmonic-mean models in order to consider
dipolar or hydrogen bonds interactions. The dispersive surface energy component (Yds)
represents the surface energy due to the dispersive and van der Waals force. The polar
component (7P5) denotes polar interactions.

In the acid-base model, three different surface tension components are considered:

Lifshitz-van der Waal for dispersion) (Yst or yds), electron-acceptor for an acid (365) and

electron-donor for a base (Y's) (Adamson and Gast 1997). The surface energy

86

components of Cu-treated, and untreated wood surfaces were calculated by using the

following equations:

COS 6: [12(1'5) 1’1.) “2 + 20's 7]. P)1/2 }/ YLl— 1 (5-3)

Equation (5-3) is called the Young-Good-Girifalco-Fowkes model, in which surface
energy components of solid were calculated by the geometric-mean method (Girifalco

and Good 1957, Fowkes 1972, Nguyen and Johns 1978).

cosl9=4[{(}’s}’L)/(Ys +7L)+(}’s}’L)/(}’s +71. PM 1 (5-4)

Equation (5-4) calculates dispersive and polar components of solid surface energy by

harmonic-mean method (Wu 1971).

0080 = [{2(}’§‘Wi’fw)”2 + (75701” + (YEYZ)1'2}/71.1-1 (5-5)

Equation (5-5) is the acid-base model for estimating Lifshitz-van der Waal, acid and
base components of the solid surface energy (Van Oss et al. 1988).
The following matrix was used to compute surface energy components from the

models above:

87

 

 

 

 

 

“11 “12 “13 x1 “1
“21 “22 “23 x2 = b2 (5-6)
_“31 “32 “33 J Lx3 _ _b3 2
“11 “12 “13 x1 “1
A: “21 “22 “23 1X: x2 13: b2
_“31 “32 “33 _1 _x3 _ _b3 _

 

 

 

 

 

 

AX :3, then X=A"-B
In this formula, A is the coefficient matrix, such as in equation (5-3), a“ is equal to

(Vaduz, a1; is equal to (Yp L)1/2. X is the variable vector, x1 is equal to (Yd S)1/2’ x2 is equal
to (793)1/2. B is the constant vector, and in the equation (5-3), b; is equal to 11.1

(l+cosO)/2, b2 is equal to 712 (l+cos0)/2. The surface energy of the testing liquid is u.
The objectives of this chapter are to examine the wettability of PVC on Cu-treated
wood surface and to investigate the effect of Cu-treatment on wood surface energy and its
components. In order to explore wood-PVC interphase situations and evaluate the
interfacial adhesion, an experimental method was built up to directly measure interfacial

shear strength between PVC and copper amine-treated wood.

88

5.3 Contact Angle Measurement
5.3.1 Wood Treatment

Red oak specimens measuring 5 by 40 by 80 mm were dip-treated in copper amine
treating solutions with various Cu concentrations (listed in Table 2.1) for 48 hours.
Treated wood surfaces were oven-dried first at 60 °C for 24 hours, and then at 105 °C for

24 hours to reduce the moisture content.

5.3.2 Preparation of PVC solution
The wetting condition between the PVC melts and wood flour was simulated by
evaluating the wettability of wood to PVC cyclohexanone solution through the contact

angle measurements. This was based on the similarity of the PVC and cyclohexanone.

The solubility parameter (5) and surface tension (7) of cyclohexanone are 9.9 calo'slcml'5

and 34.5 dyne/cm at room temperature, which is close to 9.5~10.0 calo'slcml'5 and 33.5

dyne/cm of PVC at 190 °C (Koberstein 1987).
About 50 g of pure PVC was dissolved in 500 ml of cyclohexanone to make a PVC-

cyclohexanone solution of 10 wt%.

5.3.3 Contact Angle Measurement

Sessile drops of about 0.08 ml of the PVC-cyclohexanone solution, deionized water,
diiodomethane, and forrnamide were deposited on treated and non-treated oak surfaces,
and the initial contact angle was measured by using a Contact Angle Goniometer (Rame-
Hart, Inc.). At least 20 droplets of each liquid were measured for each wood sample. The

surface tension and surface tension components of testing solvents are listed in Table 5.1.

89

Table 5.1. Surface Tension Components (mJ/m2) of Testing Solvents at 20 °C (Van Oss

 

 

 

 

 

1994)
Liquid 1° 1°“ (1") 1’ 1' W77—
Diiodomethane (DIM) 50.8 50.8 0 0 O
Forrnamide (FA) 58.0 39.0 2.28 39.6 19.0
Water 72.8 21.8 25.5 25.5 51.0

 

 

 

 

 

 

 

5.4 Experimental Design for Interfacial Shear Strength Measurement
5.4.1 Sample Preparation

Red oak core dowels, 6.3 mm in diameter and 40 mm in length, were dip-treated in
copper amine treating solutions and oven-dried in the same procedure as described in

5.3.1.

5.4.2 Interfacial Shear Strength Measurement

The interfacial shear strength (IFSS) between copper-treated, untreated wood and the
PVC matrix was measured by using the dowel-pullout test method. About 2 g of PVC
powder was put into aluminum pans measuring 8 mm in diameter and 2 mm in height.
Tested wood dowels were placed in the center of the aluminum pan containing PVC,
heated at 200 °C for 3 minutes using a hot plate, then cooled to room temperature to
obtain nail-shaped IFSS samples as shown in Figure 5.2. The so-obtained nail-shaped
sample was then installed properly in an Instron Universal Testing Machine as illustrated
in Figure 5.3 by placing it in a plastic foam holder with the wood dowel embedded
through a 6.8 mm diameter hole on a metal plate. The upward movement of the metal

plate was blocked by the edges of the metal member of testing machine. The tensile test

90

was conducted at a cross-head speed of 2.1 mm/min. During the test, the wood dowel was
gradually pulled out from the PVC matrix. The maximum tensile load corresponded to
the pullout debonding point, and the interfacial shear strength (1:) was calculated using the

following equation:

F

max

.1721 <54)

where Fmax is the maximum tensile load, dis the diameter of wood dowel cross section,

2':

and l the length of wood dowel embedded in PVC. The value of shear strength (1:) allows
us to compare the effect of the copper treatment on the interfacial shear strength between

wood and the PVC matrix.

 

Figure 5.2. Nail-shaped sample prepared for measuring interfacial shear strength
between wood and the PVC matrix

91

 

Figure 5.3. Wood dowel-pullout test setup

5.4.3 Statistical Analysis

SigmaStat software version 2.0 and the one-way ANOVA method were used to
examine the difference between means and standard deviations of contact angles and
IFSS at a 95% confidence interval in order to compare the effects of copper treatments as

well as the copper concentrations.

5.5 Results and Discussion
5.5.1. Wettability Evaluation of PVC to Surface Modified Wood Surface

Figure 5.4 demonstrates the scheme of sessile drop shapes of PVC solution on the
surface of Cu-treated, and untreated red oak surfaces. Figures 5.5 and 5.6 are images of

initial contact angles obtained by Contact Angle Goniometer. A much smaller contact

92

angle was formed on Cu-treated wood surfaces when the PVC solution was used in
combination with a decreased drop height and a larger drop width compared to that on
untreated wood surfaces. The drop was spread out instantly (0° contact angle) when using
cyclohexanone alone without PVC on both Cu-treated and untreated wood surfaces so
that the effect of the solvent was negligible on wood.

Average contact angles and drop dimensions of the PVC solution on Cu-treated, and
untreated wood surfaces are listed in Table 5.2. Contact angles of the PVC solution on
treated oak surfaces are compared in Figure 5.7. The contact angles of PVC solution

drops decreased almost 40% on the Cu-treated oak surface, in the same time the sessile

   

----------------------- r

Figure 5.4. PVC solution sessile drop shape on (1) untreated oak surfaces; (2) copper
amine-treated oak surfaces

    

 

 

Figure 5.5. Image of the initial contact Figure 5.6. Image of the initial contact
angle of PVC solution on untreated oak angle of PVC solution on 0.2 % Cu-
surface untreated oak surface

93

Table 5.2. Average Contact Angles and Drop Dimensions of Sessile Drops of PVC
Solution on Untreated and Copper Amine (CuEA)-Treated Oak Surfaces

 

 

 

 

 

 

Red oak surfaces Contact angle (°) Drop height (mm) Drop width (mm)
Untreated 22.3 (1.1)* 0.42 (0.03) 4.00 (0.24)
CuEA treated (0.2 w% Cu) 12.5 (0.6) 0.31 (0.02) 4.32 (0.22)
CuEA treated (0.4% Cu) 10.2 (1.0) 0.25 (0.02) 4.52 (0.12)
CuEA treated (1.0% Cu) 11.6 (0.7) 0.27 (0.02) 4.51 (0.11)

 

 

 

 

* Standard deviations are given in parentheses.

25.0 .

 

to
9
o

 

 

 

A d
.o s»
O O
A

Contact Angie (°)

.0'
o
L

 

 

 

 

 

 

 

 

 

 

0.0 1 . r .
Untreated 0.2% Cu treated 0.4% Cu treated 1.0% Cu treated

Figure 5.7. Average contact angles and drop dimensions of sessile drops of PVC solution
on untreated and Cu-treated oak surfaces

drop width increased about 10% and the drop height decreased more than 30% in
comparison with those on untreated oak surfaces. These results indicated that the wetting

of PVC on oak surfaces was dramatically improved by the wood surface Cu-treatrnents.

94

 

5.5.2. Effects of Copper Amine Treatments of Surface Energy and Surface Energy
Components of Wood Surfaces

Average contact angles of testing solvents on untreated and Cu-treated oak surface
are given in Table 5.3. The water contact angle increased about 20° after wood surface
Cu-treatrnents indicating the hydrophobic nature of treated wood surfaces. Based on data

in Table 5.3 and the testing liquid surface tension components listed in Table 5.1, total
surface energy, the surface energy dispersion component, Yd, and the polar component VP

of treated and untreated wood surfaces were calculated through Young-Good-Girifalco-
Fowkes geometric-mean model (Equation 5-3), and harmonic-mean model (Equation 5-
4). Results are presented in Table 5.4 computed from two liquid pairs: DIM—FA and

DIM-water. Table 5.5 lists the total surface energy, surface energy Lifshitz-van der Waal

(dispersion)(YLw), electron-acceptor (acid) (7+) and electron-donor (base) (7')

components of treated and untreated wood surfaces computed from the Lifshitz-van der

Waal acid-base model in Equation 5-5.

Table 5.3. Average Contact Angles of Testing Solvents on Untreated and treated Wood

 

 

 

 

 

Surface
Contact angle (°) Diiodomethane (DIM) Formamide (FA) Water
Untreated 26.4 (1.0) 70.3 (2.3) 105.1 (3.8)
CuEA treated (0.2% Cu) 29.9 (2.0) 85.9 (2.9) 123.3 (3.5)
CuEA treated (0.4% Cu) 32.3 (1.9) 87.3 (2.7) 124.6 (3.0)
CuEA treated (1.0% Cu) 32.7 (2.2) 83.5 (2.6) 120.8 (2.9)

 

 

 

 

 

 

0 Standard deviations are given in parentheses.

95

With geometric-mean, harmonic-mean and acid-base models used to study wood
surfaces, negative square roots of surface energy components were generated from model
calculations. Meijer et al. (Meijer et al 2000) encountered this situation and treated the
negative values as their absolute positive values, resulting in reasonable results rather
than omitting them as artifacts (Good 1993). On treated wood surfaces, the polar or
hydrogen bond interaction between water and treated surfaces caused a negative effect on
contact angles (the negative polar item in equations), but the polar components were still
considered to be part of the total surface energy in this study.

It is evident from Table 5.4 that polar component 7" was considerably increased by
wood surface Cu-treatrnents. This was supported by the results calculated from both the
geometric-mean model and the harmonic-mean model, and measured by both liquid
pairs. The polar component 7” of oak surfaces treated with 0.4 wt% copper amine solution
was 10 times higher according to the geometric-mean model, and 3.2 times higher
according to the harmonic-mean model than that of untreated wood. Total surface
energies of 0.2 wt% and 0.4 wt% Cu-treated wood were also increased, but dispersion

surface energy components didn’t show considerable changes after Cu-treatrnents.

96

Table 5.4. Dispersive and Polar Surface Energy Components (mJ/mz) of Treated,
Untreated Red Oak Calculated from Geometric-Mean and Harmonic-Mean Models

 

Geometric-mean Harmonic-mean

 

 

Wood samples Liquid pair 7 Yd Yr? YT yd Yp—
Untreated DIM-FA 46.3 45.7 0.6 47.6 45.8 1.8

 

CuEA treated (0.2% Cu) DIM-FA 50.1 44.3 (-)5.8 51.7 44.5 (-)7.2
CuEA treated (0.4% Cu) DIM-FA 49.4 43.3 (-)6.1 51.0 43.5 (-)7.5
CuEA treated (1.0% Cu) DIM-FA 47.1 43.1 (-)4.0 49.1 43.3 (-)5.8
Untreated DIM-water 46.1 45.7 0.4 47.1 45.8 1.3
CuEA treated (0.2% Cu) DIM-water 48.5 44.3 (-)4.2 51.8 44.5 (-)7.3
CuEA treated (0.4% Cu) DIM-water 47.7 43.3 (-)4.4 51.2 43.5 (-)7.7
CuEA treated (1.0% Cu) DIM-water 46.3 43.1 (-)3.2 49.6 43.3 (-)6.3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Electron-acceptor (acid) (7+), electron-donor (base) (7') as well as acid-base, 7"”

surface components dramatically increased after wood treatments as shown in Figure 5.5.

The electron-donor (base) component (7') of 0.4 wt% of the Cu-treated oak surface was

11 times higher, and the acid component, W, was 9 times higher than with the untreated

surface. This may be associated with the alkalinity of copper treating solution with a pH
higher than 9. Combined acid-base components, Y“, were increased after Cu-treatments,
evidencing the strong acid-base interactions. Small reductions in the van der Waal force
component, 7“, were observed in treated wood compared to those in the untreated
surfaces, but this was negligible within the experimental errors. The total surface energies
of treated wood significantly increased, from 47.4 mJ/m2 of untreated wood to 60.9
mJ/m2 of 0.2 wt% copper amine treated wood. Wood surfaces treated with 1.0 wt%

CuEA have lower surface energy parameters compared to those of the 0.2 and 0.4 wt%

97

 

CuEA treatments but higher than untreated samples. Lowered mechanical properties
were also observed in this sample group (Chapter 3).
Increased surface energies of Cu-treated wood surfaces satisfied the thermodynamic

requirement of wetting, leading to a better wetting condition between PVC and treated

wood surfaces.

Table 5.5. Surface Energy Components (mJ/mz) of Treated, and Untreated Red Oak
Calculated from the Lewis Acid-Base Model.

 

 

 

 

 

Wood sample 7'7 YLW 1” 1' 1/‘5
Untreated 47.4 45.7 0.7 0.9 1.7
CuEA treated (0.2% Cu) 60.9 44.3 6.5 10.6 (-)l6.6
CuEA treated (0.4% Cu) 60.6 43.3 6.8 11.0 (-)l7.3
CuEA treated (1.0% Cu) 55.6 43.1 4.1 9.5 (-)l2.5

 

 

 

 

 

 

 

‘ Liquids: diiodomethane-water-formamide.

5.5.3 Interfacial Adhesion Estimation by Interfacial Shear Strength
The relationship between interfacial shear stress and matrix, fiber bulk properties and
specimen geometry was described and expressed by H. L. Cox’s Shear-Lag Stress

Transfer Model in the following equation (Cox, 1952, Drzal et al, 2001, 1983a,b,c,
Sanadi et al. 1993):

G 1’ 2 sinh [3(5— — x)
r: Ef£m|: ’" :l 2

T (5-8)
25, ln(R/r) 6081155

98

where Hg is the tensile modulus of fiber, 8m the strain in the matrix, Gm the shear modulus

of matrix, R the interfiber spacing, r the radius of the fiber, [3 the scaling factor, L the
length of the fiber fragment, x the radial distance outward, and 1: the interfacial shear
stress at a fixed point.

According to the Shear-Lag Stress Transfer Model, for the same matrix and fixed
fiber geometry, interfacial shear strength depends on the tensile modulus of the fiber and
fiber-matrix interaction or fiber-matrix adhesion. Assuming that copper treatments have
negligible influence on the modulus of wood flour, interfacial shear strength was used to
reflect and characterize fiber-matrix interfacial adhesion.

Interfacial adhesion between PVC and Cu-treated, and untreated wood was
investigated by measuring PVC-wood interfacial shear strength. Results are shown in
Figure 5.8. Statistical analysis revealed that the interfacial shear strength between PVC
and Cu-treated wood is considerably higher than that between PVC and untreated wood.
No significant difference was observed in interfacial shear strength between different Cu
concentrations of treating solutions except when samples were treated with 0.2 wt% and
with 1.0 wt% Cu solution. Interfacial shear strength between PVC and wood increased by
40% after treating oak dowels with 0.2% Cu solution suggesting the formation of strong

interfacial adhesion and an improved PVC-wood interphase.

99

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Interfacial shear strength (M Pa)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I I

Cu-0.1 Cu-0.2 Cu-0.4 Cu-0.5 Cu-0.6 Cu-1.0
Cu concentration in treating solutions (%)

 

 

Untreated

Figure 5.8. Interfacial shear strength between PVC and Cu-treated, untreated wood with

different Cu concentrations in treating solutions

The copper amine solution containing 1.0 wt% copper demonstrated relatively lower
mechanical properties, surface energy and interfacial shear strength than those of
solutions with 0.2 to 0.4 wt% copper. Previous studies revealed that at this level of
treatment, physical adsorption may occur with low quality bonds or interactions between
wood and the treating solution. As shown in Figure 5.9, before 0.4 wt% copper, both
chemical and physical adsorptions occurred, but after this point, only physical absorption

happened on wood. The extra amount of copper can be easily release or leached out from

100

 

 

 

 

900-
, e Sawdust Before Leaching
80'0 : -o—Sewdust After Leaching
A 70.0 - a Cubes Before Leaching
F) 1
E 60.0 q —o—Cubes After Leaching
‘ e
c 500-
o .
3' 40.0 -
g — .
300-
‘8 . a
20.0 “ . .
100-
<- ’- I ' 1:. :1“ 3 4; vlLfi 4‘; fl
0.0 L :0 T ' v I T T I W I H

 

0.0 1 .0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
Cu concentration in treating solutions (%)

Figure 5.9. Interfacial shear strength between PVC and Cu-treated, untreated wood with

different Cu concentrations in treating solutions

treated wood sawdust and cubes. After leaching only about 3.2 kg/m3 copper was

maintained in wood corresponding to the treatment by the 0.4 wt% copper solution. The
presence of much solute may indicate a weak adhesion on interaction sites, yielding
lower properties.

From the results of contact angle measurements and surface energy calculations as
well as the estimation of interfacial adhesion, a good correlation can be built up among
the composite mechanical properties, PVC-wood interfacial adhesion, PVC wettability
and wood surface energy. It is evident that wood surface copper amine modification
increased surface energy and acid-base (polar) surface energy component of wood and

generated interfacial the acid-base interactions. As a result, the wettabilty of PVC on

101

wood surface was improved, a strong interfacial adhesion and interphase were formed,

and mechanical performances of composite were enhanced.

5.6 Conclusions

The wetting condition between PVC melts and wood flour was examined and
simulated by evaluating the wettability of a PVC cyclohexanone solution on Cu-treated
and untreated wood surfaces through contact angle measurements. PVC solution drop
contact angles decreased about 40% on Cu-treated oak surface compared to those on the
untreated surfaces accompanied by the increased sessile drop widths and reduced drop
heights, suggesting that the wetting of PVC on oak surface was improved by the wood
surface Cu-treatments.

Young-Good-Girifalco-Fowkes geometric-mean, harmonic-mean and Lifshitz-van
der Waal acid-base models were applied to study the effect of copper amine modification

on the surface energy and surface energy components of oak surface. Acid-base (polar),

y”, electron-acceptor (acid) (7+), electron-donor (base) (y') surface energy components

were dramatically increased after wood surface Cu-treatments, indicating strong acid-
base and polar interactions. Total surface energies of Cu-treated surfaces increased about
30% compared to that of the untreated surface, which ensured the thermodynamic
requirement of wetting resulting in a better wetting condition between PVC and treated
wood surfaces.

Interfacial adhesion between PVC and Cu-treated, untreated wood was investigated
by measuring PVC-wood interfacial shear strength. Interfacial shear strength between

PVC and wood was significantly increased after wood Cu-treatments. A correlation can

102

be built up among the composite mechanical properties, PVC-wood interfacial adhesion,

PVC wettability and wood surface energy.

103

Reference

Adamson, A.W., and Gast, AP. 1997. Physical Chemistry of Surfaces. A Wiley-
Interscience Publication, John Wiley & Sons, Inc. New York.

Chen, M. J ., Meister, J .J., Gunnells, D.W., Gardner, DJ. 1995. A process for coupling
wood to thermoplastic using graft copolymers. Advanced in Polymer Technology, 14(2):
97-109.

Cox, H. L. 1952. The elasticity and strength of paper and other fibrous materials. Br. J.
Appl. Physics, 3(3): 72-79.

Drzal, L. T., and Herrera-Franco, P. J. 2001. Micro and macro fiber-matrix adhesion
measurement methods in composite materials. In: Proc. 24th Annual Meeting of the
Adhesion Society. pp. 50-52.

Drzal, L. T., and Herrera-Franco, R]. 2002. Measurement methods for fiber-matrix
adhesion in composite materials, Chapter 17 in Volume H, Comprehensive Adhesion
Science: The Mechanics of Adhesion, Rheology of Adhesives and Strength of Adhesive
Bonds, Elsevier Science,

Drzal, L. T., Herrera-Franco, P. and Ho, H. 2000. “Test methods for the measurement of
fiber-matrix adhesion” Chapter 5 in Comprehensive Composite Materials, Elsevier
Science, A. Kelley and C. Zweben, ed.

Drzal, L. T., Opila, R.L., Peppas, N. A., and Schutte, C. eds., 1995. Polymer/Inorganic
Interfaces H, Materials Research Society Symposium Proceedings, v. 385.

Drzal, L. T., and Madhukar, M. 1993. Fiber-matrix adhesion and its relationship to
composite mechanical properties, J. Matr. Sci., 28: 569-610.

Drzal, L. T. 1983a. Composite interphase characterization. SAMPE Journal, 19: 7-13.

Drzal, L. T., Rich, M., and Lloyd, P. l983b. Adhesion of graphite fibers to epoxy
matrices. I. The role of fiber surface treatment, J. Adhesion, 16: 1-30.

Drzal, L. T., Rich, M., Koenig, M., and Lloyd, P. 19830. Adhesion of graphite fibers to
epoxy matrices. II. The effect of fiber finish. J. Adhesion, 16: 133-152.

Gardner, D. J ., Walinder, M. 2001. Surface energy and acid-base properties of wood

particles coated with styrene-acrylic acid copolymers. Abstracts of Papers, 22lst ACS
National Meeting, San Diego, CA, United States, April 1-5, 2001.

104

Herrera-Franco, P.J., Glasser, G., and Drzal, LT. 1992. Interfacial shear strength
measurements in composites using the microindentation technique, American Society for
Composites, Proceedings of the 7th Technical Meeting, State College, PA.

Ho, H., and Drzal, LT. 1996. An evaluation of interfacial mechanical properties of fiber
reinforced composites using the microindentation method”, Composites, Part A, 27A:
961-971.

Hull, D., and Clyne, T.W. 1996.An Introduction to Composite Materials, Second Edition.
Cambridge University Press, Cambridge, UK.

Good, R.J. 1993. In Contact Angle, Wettability and adhesion; Mittalm K.L., Ed., VSP:
Zeist, The Netherlands.

Gunnells, D. W., Gardner, D. J ., and Wolcott, M. P. 1994. Temperature dependence of
wood surface energy. Wood and Fiber Sci., 26(4): 447-55.

Jiang, H., and Kamdem, D. P. 2004 Effects of copper amine treatment on mechanical
properties of PVC/wood flour composites. J. Vinyl Addit. Techn. 10(2): 70-78.

Kamdem, D.P., Jiang, H., Cui, W. Freed, J., and Matuana, L.M. 2004. Properties of wood
plastaic composites made of recycled HDPE and wood flour from CCA-treated wood
removed from service. Composites: Part A. 35:347-355.

Kamdem, D.P., Bose, S. K., and Luner, P. 1993. IGC characterization of birch wood meal,
Langmuir 9(11): 3039-3044.

Kamdem, D. P., and Riedl, B. 1992a. Estimation of the dispersion component of surface
energy of polymer-grafted lignocellulosic fiber with inverse gas chromatography, J.
Adhesion Sci. Technol., 6 (9): 1053-1067.

Kamdem, DP, and Riedl, B. 1992b. Inverse gas chromatography of lignocellulosic fibers
coated with a thermosetting polymer: Use of peak maximum and Conder and Young
methods. J. Colloid Interface Sci., 150(2): 507-516.

Kamdem, DP, and Riedl, B. 19912. Characterization of wood fibers modified by
phenol-formaldehyde, Colloid Polym. Sci., 269 (6): 595-603

Kamdem, D.P., Riedl, B., Kaliaguine, SC, and Adnot.A. 1991b. ESCA spectroscopy of
poly (methylmetacrylate) grafted onto wood fibers, J. Appl. Polym. Sci. 43:1901-1911.

Kamdem, DP. and Riedl, B.l991c. IGC characterization of PMMA grafted onto CTMP
fiber, J. Wood Chem. Technol., 11 (1): 57-91.

105

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

Kazayawoko, M., Balatinecz, J. J., Woodhams, R. T., and Law, S. 1997a. Effects of
wood-fiber surface chemistry on the mechanical properties of wood-fiber-polypropylene
composites. Intern. J. Polymeric Mater. 37: 237-261.

Kazayawoko, M., and Balatinecz, J. J. 1997b. Adhesion mechanisms in woodfiber-
polypropylene composites. In: Proc. 4th International Conference on Wood fiber-Plastic
Composites, Forest Products Society. Madison, WI, USA, May 12-14. pp. 81-93.

Koberstein, J .T. 1987. Encyclopedia of Polymer Science and Engineering, Volume 8,
John Wiley, New York.

Liu, RP, and Rials, TC. 1998. Relationship of wood surface energy to surface
composition. Langmuir. 14:536-541.

Lu, J 2., and Wu, Q. 2005. Surface and interfacial characterization of wood-PVC
composite: Imaging morphology and wetting behavior. Wood and Fiber Science, 37(1):
95-111.

Madhukar, M., and Drzal, L. T. 1992. Fiber-matrix adhesion and its effect on composite
mechanical properties. 1H. Longitudinal compressive properties of graphite/epoxy
composites, J. Comp. Matr., 26: 310-333.

Mahlberg, R., Niemi, H. E.-M., Denes, F., and Rowell, R.M. 1998. Effect of oxygen and
hexamethyldisiloxane plasmas on morphology, wettability and adhesion properties of
polypropylene and lignocellulosics. International Journal of Adhesion and Adhesives,
18(4): 283-297.

Maldas, D.C., Kamdem, DP. 1999. Wettability of extracted southern pine. Forest Prod. J.
49(11/12): 91-93.

Maldas, DC, and Kamdem, DP. 1998. Surface tension and wettability of CCA-treated
red maple. Wood and Fiber Sci. 30(4): 368-373.

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

Meijer, M., Haemers, S., Cobben, W., and Militz, H. 2000. Surface energy

determinations of wood: comparison of methods and wood species. Langmuir. 16:9352-
9359.

106

Mohanty, A.L., Drzal, L. T., and Misra, M. 2002. Engineered natural fiber reinforced
composites: Influence of surface modifications and novel powder impregnation
processing, Journal of Adhesion Science and Technology, 16(8): 999-1015.

Pizzi, A. 1983. Wood Adhesives: Chemistry and Technology. Marcel Dekker, New York.

Sanadi, A. R.; Rowell, R. M.; Young, R. A. 1993. Evaluation of wood-therrnoplastic-
interphase shear strengths. Journal of Materials Science, 28(23): 6347-52.

Shi, G., Drzal, LT. 2003. Effect of surface chemistry on the interfacial adhesion and
mechanical properties of natural fiber reinforced polymer composites. Ph.D Dissertation,
Michigan State University.

Simonsen, J., Hong, Z., and Rials, T.G. 1997. The properties of the wood-polystyrene
interphase determined by inverse gas chromatography. Wood and Fiber Sci., 29(1): 75-
84.

Ulkem, I., and Schreiber, HP. 1994. The role of interactions at interfaces of glass-fiber
reinforced composites. Composite Interfaces, 2(4): 253-263.

Van Oss, C. J. 1994. “Interfacial forces in Aqueous Media, Dekker, New York.

Walinder, ME, and Gardner, DJ. 2002. Surface energy and acid-base characterization
of components used for wood-polymer composites. International Conference on
Woodfiber-Plastic Composites, 6th, Madison, WI, United States, May 15-16, 2001,
Meeting Date 2001: 185-195.

Yang, P; Jinno, K., Nishimoto, A., and Ohsako, Y., Yamauchi, H., and Sasaki, H. 1999.
Reinforcement of cylindrical LVL. 1. Improving the bondability of waste paper-
polyethylene composite core fillers to the cylindrical LVL. Mokuzai Kogyo, 54(9):
416-419.

107

Chapter 6
EFFECTS OF COPPER TREATMENTS ON INTERPHASE CHEMISTRY OF
PVC/WOOD FLOUR COMPOSITES
6.1 Background

From the results of Chapter 5, the surface energy acid-base component of wood
surfaces was significantly increased by wood surface copper amine treatments suggesting
strong acid-base interactions. In this chapter, the interphase chemistry was studied by
using IR and XPS.

Coupling agents are characterized by one side of the chemical reacting with the filler
and another side connecting to a polymer matrix. The adhesion mechanisms of the
various coupling agents and compatibilizers applied in the wood-plastic composite area
have been introduced in Chapter 1.

Copper amine is one of the main components of copper-based wood preservatives,
such as ammoniacal copper quat-type D (ACQ—D), copper dimethyl-dithio-carbamate
(CDDC), and copper azole (AWPA 1998, Kamdem 2001, 19982,b,c, Zhang and Kamdem
2000 a,b,c). A cupric complex is formed when mixing copper hydrate and ethanolamine
that acts as the chelating agent through amino and hydroxyl groups (Bodie et al. 1994).
The deprotonation of hydroxyl groups and the formation of stable chelate rings were
suggested in a high pH solution system (Tauler and Casassas 1986). Wood has a weakly
acidic nature and carboxylic groups and phenolic hydroxyl groups are potentially active
sites for reactions with the copper complex (Zhang and Kamdem 2000d). It was proposed

that, after copper amine treatments, the carboxylic groups in the guaiacyl units in lignin

108

of wood reacted with copper complex to form stable complexes (Ruddick et al. 1992,

2001, Zhang and Kamdem 2000a,b,d).

6.2 Analytical Techniques
6.2.1 Sample Preparations

The pure PVC and composite samples used for the FTIR and XPS analysis were
laboratory manufactured according to the description of Chapter 2 and ground to 50 mesh

(0.3 mm) particles.

6.2.2 Fourier Transform Infrared Spectroscopy (FTIR) and Far IR Analysis

The structure of compounds can be elucidated by the spectral locations of their
infrared absorptions because chemical bonds absorb infrared energy at specific
frequencies (or wavelengths). Absorption of infrared radiation is confined largely to
molecular species that have small energy differences between various vibrational and
rotational states (Skoog et a1 1998).

Middle FI'IR located between 400 to 4000 cm'1 wavenumbers is used for qualitative
and quantitative analysis of the chemical structure and function groups of organic and
inorganic compounds. Far-IR located between 600 to 100 cm'1 is mostly appropriate
when analyzing inorganic and metal coordinate compounds. Infrared spectroscopy (IR) is
among the solid state non-destructive techniques best suited for qualitatively and
quantitatively studying the surface of solids, such as wood (Zhang and Kamdem 2000a,
Craciun and Kamdem 1997a). FI‘IR and Far-IR techniques were used to collect the

spectra of PVC, untreated, Cu-treated wood, and PVC-untreated, Cu-treated composites

109

in order to monitor the chemical structure change which happened before and after the
Cu-treatrnents and after composite processing.

Middle FI‘IR analysis (400 to 4000 cm”) was conducted on a Nicolet Protege 460
spectrometer equipped with a Spectra-Tech diffuse reflectance accessory. Potassium
bromide (KBr) was used to collect the background spectrum. Composite samples were
ground to mesh 50 (0.3 mm) particles and mixed with KBr. Diffuse Reflectance Infrared
Fourier Transform accessory (DRIFT) was used to collect spectra of solid samples at a
resolution of 4cm’l and a total of 64 scans on 400 to 4000 cm”.

A Far-IR analysis (600 to 100 cm") was carried out on a Nicolet 750 Magna-IR
Series II spectrometer equipped with a TGS/PE detector and a silicon beam splitter in 2
cm. A blank mirror was used to collect the background spectrum. Far-IR spectra of the
solid samples were collected at a resolution of 4crn'l and a total of 64 scans in the range

of 600 to 100 cm‘1 wavenumbers.

6.2.3 X-ray Photoelectron Spectroscopic (XPS) Analysis

X-ray photoelectron spectroscopic (XPS) is also called electron spectroscopy for
chemical analysis (ESCA). XPS is a well-established solid-state spectroscopic technique
used to determine the atomic composition and the valence states of elements present on
the surface of solids with a sampling depth of 10 to 50 Angstron (Kamdem and Zhang
2001). XPS provides information on the binding energy, chemical shift of photoelectrons,
and the auger electrons that are used to identify the correspondent atom, its environment,

and its oxidative state (Kamdem and Zhang 2001).

110

The basic principle of an XPS experiment is illustrated in Figure 6.1. An x-ray
photon source directly strikes a sample surface under a vacuum condition. The energy of
the x-ray photon is transferred to a core-level electron of an atom and emits the core-level
electron from this atom (Vickerrnan 1997). The ejected electrons are separated and

collected according to their energy levels by an energy analyzer.

Oxygen atom

  

X-ray photon
hu

0
O1. photoelected electron

Figure 6.1. A core-level electron is ejected by a high energy X-ray photon (Vickerrnan

1997)

The relationship between the electron binding energy and its kinetic energy can be

described by the following equation:

EB =hv—KE (6-1)

where E B is the binding energy of the core-level electron in the atom, hv is the energy

of the X-ray source, and KB is the measured kinetic energy of the emitted electron

111

(Vickerman 1997). The binding energies of the ejected electron are correlated with its
original atomic and molecular environment. The number of the ejected photoelectrons is
proportional to the concentration of the corresponding atoms present in the sample
surface.

In this Chapter, XPS was used to characterize the surface elemental and chemical
compositions of PVC, Cu-treated and untreated wood flour, and PVC/Cu-treated,
untreated wood flour composites samples. The surface concentrations of C, O, N, Cl, Cu,
as well as the ratio of O/C, Cl/C of the above samples were calculated.

A Physical Electronics PHI 5400 ESCA Spectrometer with a standard magnesium x-
ray source located at the MSU Composite Materials and Structures Center was used in
this study. Kinetic energy measurements were made using a hemispherical electrostatic
analyzer with a 150 nm radius working in a constant-pass energy mode. The magnesium
Ka source operated at 15 kV and 20 mA, and a vacuum range of 10'7 to 10 '8 torr was

used. Multipak software was applied for curve fitting and peak area calculations.

6.3 Interphase Chemistry of PVC/Copper-Treated Wood Flour Composites
6.3.1 Fourier transform infrared spectroscopy (FTIR) and Far-IR analysis

Previous studies on the interaction of copper amine (Cu-EA) and wood indicated that
Cu-EA interacted with carboxylic, phenolic hydroxyl and ester functional groups from
lignin to form copper carboxylate and phenolate complexes (Zhang and Kamdem 2000a).
The FTIR absorption assignments in wood and PVC are listed in Table 6.1 and 6.2
respectively (Hon and Shiraishi 1991, Hummel 1971). Figure 6.2 shows the FTIR spectra

of PVC, red oak and copper ethanolamine. Oak was characterized by the O-H stretching

112

at 3408 cm'l, CH2 stretching at 2902 cm", C=O stretching of carboxyl groups at 1740
cm", carbonyl stretching in parasubstituted aryl ketone and conjugated carbonyl at 1660

1, aromatic skeletal vibrations at 1598 cm", C-H deformations, asymmetric and

cm'
symmetric at 1458 and 1369 cm'1 respectively, as well as aromatic C-H out-of-plane
deformation in 910 cm". There was no strong O-H absorbance peak at 3408 cm“1 on the
FTIR spectrum of the PVC. Instead, strong absorbance of CH2 stretching was observed at
2900 cm”. C-H deformations in the presence of chlorine appeared at 1430, 1333 and
1250 cm]. From the FI‘ IR spectrum of copper ethanolamine complex, strong O-H and N-
H absorbance was present around 3256 cm". The absorbance peak observed at 1571 cm'1
was assigned to C-N, and a peak which appeared at 1065 cm'1 was assigned to C-C
stretching (Craciun et al. 1997b). Most of the IR vibrations associated with Cu happened

in the low wavenumber range and were examined by Far-IR.

Table 6.1. Assignment of IR Absorption Bands (Hon and Shiraishi 1991)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Position in cm'I Band assignments

3450-3300 OH stretching (hydrogen bonds)

2851-2967 CH2 stretching
1735 :t 5 C=O stretching in carboxyl groups
1715 :1: 5 =0 stretching in unconjugated ketone and carboxyl groups
1660 t 5 C=O stretching in parasubstituted aryl ketone and conjugated carbonyl
1595 :1: 5 Aromatic skeletal vibrations
1505 :1: 5 Aromatic skeletal vibrations
1460 :1: 5 C-H deformations (asymmetric)
1425 :1: 5 Aromatic skeletal vibrations
1370 :1: 5 C-H deformations (symmetric)
1330 :1: 5 Syringyl ring breathing with C-O stretching
1270 :1: 5 Guaiacyl ring breathing with C-O stretching
1235 :1: 5 Syringyl ring breathing with C-O stretching
1130 :1: 5 Aromatic C-H in-plane deformation, syringyl-type
910 i 5 Aromatic C-H out-of-plane deformation

 

113

 

Table 6.2. Assignments of IR Absorption Bands of PVC (Hummel 1971)

 

 

 

 

 

 

 

 

 

 

 

 

 

Position in cm'1 Band assignments
2850-2970 CH2 stretching
1430 :1: 5 C-H deformations
1330 :1: 5 C-H deformations in —CHCl—
1250 :1: 5 C-H deformations in -CHC1—
1100 :1: 5 CC stretching
960 1 5 CH2 in-plane wag
600-700 C-Cl stretching
8 8%
3 3 § :32
3 "' 8
{‘3
0 0°
9 831’ co
8 '- v-s- g
g P
E 2 §
3 a
< Oak
PVC
ID
8
CuEA E
weo'aéoo’aeeo‘zéoo'zeeoTzdooReoekéooleoofi

Wavenumbe rs (cm")

Figure 6.2. FI‘IR spectra of PVC, red oak and copper ethanolamine

114

The FTIR spectra of the untreated oak, and Cu-treated oak containing 0.2 and 1.2
wt% copper are illustrated in Figure 6.3. No remarkable difference was found on the IR
spectra of the Cu-treated and untreated oak flour except the increase in peak intensity at
1660 cm'1 for the Cu-treated wood compared to the adjacent aromatic skeletal vibrations
at 1598 cm'1 or the aromatic C-H out-of-plane deformation in 910 cm", which exhibited
no change after the Cu-treatments. This peak intensity increased with the copper content
and it was assigned to carbonyl stretching in parasubstituted aryl ketone and conjugated
carbonyl. Similar increases at 1660 cm'1 were observed on the spectra of composites
made of PVC and Cu-treated WF as shown in Figure 6.4, and the peak intensity increased
with the increasing Cu concentration in the treated wood flour. The chemical reactions
between the wood and the copper complex may be related to the modification of the
carbonyl groups of wood. It is also possible that carbonyl groups in lignin may react with
amino groups in ethanolamine to form irnine structure, and C=N stretching shares the
same IR band in 1660 cm”, resulting in an increased peak intensity. Further investigation
needs to be conducted in order to elucidate chemical reactions related to the change of
this peak. Additionally, all the absorption bands of the C-H deformations, asymmetric at
1458cm", symmetric at 1369 cm‘1 as well as the aromatic C-H out-of-plane deformation
at 910 cm" in the composites made with PVC and Cu-treated WF became smaller and
almost negligible when Cu concentration increased to 1.2 wt%. This wasn’t found in the
spectra of Cu-treated wood alone in Figure 6.3; it was probably associated with the

influence of the presence of chlorine in the PVC in the composites (Hummel 1-971).

115

3408

2902
1740
1660
1598

Absorbance

Oak-Cu1.2% '

Oa k-Cu0.2%

Oak

f v T w T v v v y fir *1

fi fi 7 v r

4000 3600 :nmo 2331 2400 2003 1600 1200 800 400

Wave numbers (cm")

Figure 6.3. FTIR spectra of untreated red oak, copper amine treated red oak containing
0.2 wt% copper and copper amine treated red oak containing 1.2 wt% copper

 

;-' §§
i 33
33
8
C
5
0 §
3 W640u1.2% .-
at
l
2:
we40uo. °’
PWMcon
meo'eeeo’eeeo'eeoo‘zeeofieozoWeeon‘oo’eéo'm

Wavenumbers (cm")

Figure 6.4. FTIR spectra of composites made of PVC and untreated red oak, copper
treated red oak containing 0.2 and 1.2 wt% 00pper

116

Table 6.3 lists the assignments of far-IR absorption bands of Cu complexes
coordinated with O, N, and Cl observed in copper treated, untreated WP and their
composites (Ferraro 1971, Polyakov et al. 1990). Figure 6.5 shows far-IR spectra of

copper amine treating solutions and the corresponding treated WF as well as untreated

WF.

Table 6.3. Assignments of far-IR absorption bands of Cu bonded with O, N, and Cl as

observed in copper treated, untreated WP and their composites (Ferraro 1971, Polyakov
et al. 1990)

Position in cm' Band
610-630 Cu-O
510 t 5 Cu-N
470-490 Cu-N

400-420 Cu-N

360 :1: 5 O-Cu-O deformations
342 Cu-Cl

200-230 N-Cu-N deformations

 

555
s: 1% z 3 A3)“
.1110!

JW
CuEAsolutlon-Culifi MM,
M VVJDAN '
Treated WF-Cu1.2%
Treated WF-Cu0.2% VW WW 0’

lhtreated WF

 

Absorbance

 

 

' I V I ' I j t ' U V I

600 500 400 300 200 100
Wavenumber (cm")

Figure 6.5. Far-IR spectra of copper amine treating solutions with 0.2 and 1.0% Cu, and
their corresponding treated WF, as well as untreated WF

117

\
\
‘.
l
\

W“; 0“
‘\

1
J\J\/\N
R342
1A,;
‘1

\«\// 1 342
“W “V
300

400
Wavenumbers (cm )

Absorbance

 

Figure 6.6. Far-IR spectra of (1) treated WF containing 1.2% Cu; (2) PVC/Cu-treated WF
composites made of 0.2 wt% Cu-treated WF; (3) PVC/Cu-treated WF composites made of
1.2% Cu-treated WF

A copper-oxygen vibration in the Cu-EA complex appeared around 610-630 cm",
and O-Cu-O deformations were found at 355 cm". Absorptions at 507 cm", 484 cm",
471 cm'1 and 418 cm'1 were attributed to the Cu-N stretching, and absorption bands
which appeared at 200-230 cm‘1 were assigned to the N-Cu-N deformations.

Far-IR spectra of treated wood flour, composites made of PVC and Cu-treated WF
are illustrated in Figure 6.6. The peak at 342 0m'1from Cu-Cl stretching, was not found
on the spectra of Cu-treated WF, and composite made of PVC and untreated WF. This

result may be related to the formation of Cu-Cl bonding resulting from the interaction

between chlorine in PVC and Cu in Cu-treated wood flour.

118

6.3.2 X-ray Photoelectron Spectroscopic (XPS) Analysis

A XPS survey scan spectrum of the composite surfaces made of PVC and a 0.2 wt%
Cu-treated wood flour is shown in Figure 6.7. Photoelectrons from Cls, N ls, Ols, C12p,
Cu2p were collected. Surface elemental compositions of PVC, treated, and untreated

wood flour, and composites made of PVC and treated, and untreated wood flour are listed

 

in Table 6.3.
4
X10 O4XPS354.spe
5 I I j I I I I I I I
1'2
4.5 — a
4-3 .. -
q 6
3 5 — 9‘ ~
0
a 3

 

3 2.5 - “NE, W ~
«1 wt%
2, W _

MWMM» 1*

 

 

1.5~

0.5 _ h -

o l t l l I 1 I l l L
1100 1000 900 800 700 600 500 400 300 200 100 0
Binding Energy (eV)

 

 

 

 

Figure 6.7. XPS survey spectrum of composite made of PVC and Cu-treated WF
containing 0.2 wt% copper

119

Carbon, oxygen and chlorine were present on the PVC surface. The surface of WF
contained carbon, oxygen and a small amount of nitrogen. An increased content of
nitrogen and copper were detected on the surface of the Cu-treated WF, it was attributed
to the copper and nitrogen picked up through the copper amine treatment. About 12%
chlorine was observed on the PVC/untreated WF composite surfaces. The values of
nitrogen and copper found on the surface of composite made of PVC and Cu-treated WF
were lower than that of the corresponding treated WF. This is because the presence of
PVC (without N and Cu) diluted the elemental concentrations of N and Cu. The

composites showed higher chlorine contents than pure PVC.

Table 6.4. Surface Atomic Compositions (%) of PVC, Wood Flour, and PVC/Cu-treated,

Untreated WF Composites

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample ID C O N C] Cu O/C Cl/C

PVC (190F) 73.4 16.8 0.0 9.8 0.0 0.23 0.13

Oak WF 74.3 25.4 0.3 0.0 0.0 0.34 0.00

Cu-treated WF, 0.2% Cu 75.7 22.3 1.7 0.0 0.3 0.29 0.00
Cu-treated WF, 0.6% Cu 74.7 22.1 2.1 0.0 0.4 0.30 0.00
Cu-treated WF, 1.2% Cu 74.5 22.2 2.7 0.0 0.7 0.30 0.00
PVC/untreated WF = 60/40 71.4 16.5 0.0 12.1 0.0 0.23 O. 17
PVC/Cu-treated WF = 60/40, 0.2% Cu 70.1 15.6 1.5 12.6 0.1 0.22 0.18
PVC/Cu-treated WF = 60/40, 0.6% Cu 70.8 15.9 1.0 12.3 0.1 0.22 0.17
PVClCu-treated WF = 60/40, 1.2% Cu 70.8 16.0 1.1 12.0 0.1 0.23 0.17
PVC/untreated WF = 50/50 72.2 15.3 0.0 12.6 0.0 0.21 0.17
PVC/Cu-treated WF = 50/50, 0.2% Cu 69.8 16.1 1.0 12.9 0.2 0.23 0.18
PVC/Cu-treated WF = 50/50, 0.6% Cu 68.9 16.7 1.1 13.1 0.2 0.24 0.19
PVC/Cu-treated WF = 50/50, 1.2% Cu 71.4 13.7 1.6 13.1 0.3 0.19 0.18

 

 

 

 

 

 

 

 

 

120

 

0.40 —

   
    
       
      

[ICU-0% [Cu-0.2%
0'35 i aCu-O.6% ICU-1.2%
0.30 — 7,
/I,
0.25 - Z,
%<
0 Z
~ 0.20 « Z
0 Z
0.15 - Z,
Z:
0.10 ~ Z
Z;
0.05 « Z
Z
0.00 -

 

  

PVC PW64 PW55 Cu-Oak Oak

Figure 6.8. Oxygen to carbon ratio, O/C of PVC, treated and untreated oak flour as well
as composites made of 60 wt% PVC and 40 wt% WF (PW64), 50 wt% PVC and50 wt%
WF (PW55) with different Cu concentrations in treated WF

I] Cu-O% ECU-0.2%
aCu-0.6% ICU-1.2%

 

   

PVC

Figure 6.9. Chlorine to carbon ratio, Cl/C of PVC, and composites made of 60 wt% PVC
and 40 wt% WF (PW64), 50 wt% PVC and50 wt% WF (PW55) with different Cu
concenuations in treated WF

121

The oxygen to carbon ratio, O/C, and chlorine to carbon ratio, Cl/C, were
demonstrated in Figures 6.8 and 6.9, respectively. Oak flour had the highest oxygen to
carbon ratio. After the wood surface Cu-treatments, the O/C ratio decreased due to the
introduction of more carbon than oxygen in the chemical composition of ethanolamine.
The oxygen in PVC came from the polymerization initiator residues or the formation of
low molecular weight polyperoxides during the polymerization of PVC in the presence of
oxygen (Owen 1984). The composites exhibited a similar O/C ratio to PVC, which was
probably because the wood surface was covered by the PVC. It has been reported that the
addition of wood in the PVC prevents PVC from thermal degradation that is
characterized by dehydrochlorination (Djidjelli et al. 2002, Owen 1984). As a result,
composites can maintain more chlorine in a PVC matrix during processing, and display
higher C1/C ratios than pure PVC board. Further experiment need to be carried out by a
thermogravimetry analysis to confirm this effect.

After peak curve fitting, Cls carbon was deconvoluted to three peaks, as listed in
Table 6.5. The C1 peak corresponds to the carbon bond with another carbon or hydrogen.
C2 was assigned to a carbon bond with nitrogen, hydroxyl, and chlorine or in an ether
structure. The C3 peak represents a carbon-oxygen double bond, C=O structure
(Kamdem and Zhang 2001). The area ratios of deconvoluted peaks to the total Cls peak

areas of the composite samples are shown in Table 6.6 and illustrated in Figure 6.10.

122

Table 6.5. The Cls peak deconvolution assignment for PVC, Wood and PVC-Cu-treated
Wood Composites

 

 

 

 

 

Binding energies (eV) Chemical environment
Cl 284.5-285.0 C-H, C-C
C2 2860-2865 C-N, C-OH, C-O-C, C-Cl
C3 2880-2882, 289.0 C=O, N-C=O, O-C=O

 

 

 

Table 6.6. The C1, C2 and C3 Peak Area Ratios for PVC, Wood Flour, and PVC/Cu-
treated, Untreated WF Composites

 

 

 

 

Sample ID C1 C2 C3
BE (eV) Area (%) BE (eV) Area (%) BE (eV) Area (%)
PVC resin 285.0 63.9 286.1 30.3 288.8 5.8
Oak WF 284.8 58.9 286.5 33.6 288.4 7.5

 

Cu-WF, 0.2% Cu 284.7 68.1 286.3 24.9 288.1 7.0

 

Cu-WF, 0.6% Cu 284.8 69.6 286.3 23.5 288.3 6.9

 

Cu-WF, 1.2% Cu 284.8 70.5 286.5 22.8 288.2 6.7

 

PVCIWF = 60/40 284.5 53.0 286.3 42.0 288.1 5.0

 

PVC/Cu-WF = 60/40, 284.7 46.2 286.2 47.4 288.1 6.4

 

 

0.2% Cu

PVC/Cu-WF = 60/40, 285.0 43.5 286.5 50.0 288.1 6.5
0.6% Cu

PVC/Cu-WF = 60/40, 284.7 48.0 286.4 45.7 288.2 6.3
1.2% Cu

 

PVCIWF = 5050 284.7 44.3 286.3 50.2 288.1 5.6

 

PVC/Cu-WF = 50/50, 284.7 39.1 286.3 54.9 288.1 6.0

 

 

 

 

 

 

 

 

 

0.2% Cu

PVC/Cu-WF = 50/50, 284.9 36.6 286.5 57.1 288.2 6.3
0.6% Cu

PVC/Cu-WF = 50/50, 284.9 48.1 286.4 45.7 288.3 6.2
1.2% Cu

 

123

 

 

The C1 percentage areas, which correspond to the amount of CH and CC structures
increased, whereas C2, which represents the contribution of C-N, C-OH, C-O-C, declined
after the wood Cu-treatrnents. This was associated with the introduction of copper
ethanolamine complex on the wood surface. The ratio of C-C and C-H to C-OH and C-N
is 5 to 2. The composites showed smaller CI but a higher amount of C2 compared to the
wood and PVC. This may be because the presence of CC] from the PVC was counted in
C2, and when using copper amine treated wood, C-N was also included in C2, resulting
in the increased percentage area of C2 and reduction in percentage of C1. The differences
in C3 between PVC and composites, as well as between untreated and treated wood were

negligible.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E101 [2102 1203

Figure 6.10. C1, C2, and C3 area ratios for of PVC, treated, untreated oak flour and
composites

124

It was reported that copper oxides and complexes, such as copper salicylate and
CuO, reacted with the PVC during processing and under visible or UV radiation through
a Friedel-Crafts reaction (Shibata et al. 2000, Wickson 1993, Kwei et al. 1967). This led
to an accelerated degradation, the early embrittlement of the material as well as the
discoloration in outdoor use even in trace amounts (Wickson 1993, Kwei et al. 1967).

Copper and copper containing chemicals were also considered as good flame and
smoke suppressants for PVC. Dehydrochlorination of the PVC was significantly
accelerated by the presence of Cu powder, CuO, CuzO and Cqu at 200 °C, resulting in
the formation of cis-2-pentene and allylic chloride structures and copper additives which
acted as lewis acid catalysts (Wescott et al. 1985, Huang and Stames 1993). Crosslinking
reactions of the PVC at a high temperature was prompted and the rate of segment
pyrolysis was reduced by adding copper additives which were proved to retard smoke
and heat evolution in burning PVC (Stames et al. 2003, Li et al. 1999, Jeng et al. 1994).

PVC can be modified with dimethylglyoxime (DMG) followed by chelating with Cu
(H) to form a PVC-DMG-Cu complex polymer which is considered as a promising
material for flame retardant low-smoke cable sheathing (Moitra et al.1998, Biswas and
Moitra 1989). Cu (11) complexes can react with polymers, such as polyurethane and
polybutadiene rubber to form polymer ionomer-metal complexes in order to improve
electrical conductivity of resulting materials (Kim et al. 1989, Xue et al. 1993).

More research and experimental work need to be conducted in order to obtain stronger
evidence and reveal possible interactions happening in the wood-copper amine complex-

PVC interphase.

125

6.4 Conclusions

An increased intensity of C=O stretching at 1660 cm'1 was observed on th eIR
spectra of the Cu-treated wood and the composites made of PVC and the Cu-treated WP
and the intensity of this peak became stronger as Cu concentrations increased in the
treated wood flour. The results of IR analysis suggested that reactions between copper
amine complex and wood may be related to the change of carbonyl groups of wood.

The XPS analysis showed that carbon, oxygen, nitrogen and chlorine were present on
PVC/Cu-treated wood composite surfaces. The C1 percentage areas increased, whereas
C2 declined after wood copper amine treatments. This was associated with the
introduction of more C-C and CH than C-OH and C-N from the copper ethanolamine
complex to the wood surface. The composites showed smaller C1 but higher C2
compared to the wood and the PVC due to the presence of C-Cl from PVC and C-N from
the copper amine complex. The differences in C3 between the PVC and the composites,
as well as between untreated and treated wood were negligible

Further investigations need to be carried out in order to obtain stronger evidence and

reveal possible interactions happening in wood-copper amine complex-PVC interphase.

126

References

American Wood Preservers’ Association (AWPA). 1998. Book of Standards. Granbury,
Texas.

Biswas, M., and Moitra, S. 1989. Synthesis and some properties of PVC-bound
dimethylglyoxime complexes of cobalt(II), nickel(II), and copper(II). J .Appl. Polym. Sci.
38(7): 1243-52.

Craciun, R., and Kamdem, D. P. 1997a. A XPS and FTIR Applied to the Study of
Waterborne Copper Naphthenate Wood Preservatives. Holzforschung 51(3): 207-213.

Craciun, R., and Kamdem, D. P.,Craig R.M. 1997b. Characterization of CDDC (copper
dimethyl dithiocarbamate) treated wood. Holzforschung 51(6): 519-525.

Djidjelli, H., Martinez-Vega, J. J., Farenc, J ., and Benachour, D. 2002. Effect of wood flour
content on the thermal mechanical and dielectric properties of poly(vinyl chloride).
Macromol. Mater. Eng., 287: 611-618.

Ferraro, J. R. 1971. Low-frequency Vibrations of Inorganic and Coordination
Compounds. Plenum Press. New York.

Hon, D. S., and Shiraishi, N. 1991. Wood and Cellulosic Chemistry. Marcel Dekker, Inc.,

Huang, C.H.O., and Stames, W.H. 1993. Mechanistic studies of copper additives as
smoke suppressants for poly(vinyll chloride). In: Proceeding of Beijing Int. Symp. Flame
Retard, 2" . 168-172.

Jeng, J .P., Terranova, S.A., Bonaplata, E., Goldsmith, K., Williams, D.M., Stames, W.H.
1994. Copper-promoted reductive coupling as a potential means of smoke suppression in
poly(vinyl chloride). Polym. Mater. Sci. Eng. 71:299-300.

Jusoh, I and DR Kamdem. 2000. Penetration and retention of CCA in rubberwood.
Journal of Tropical Forest Products 6(1):77-84

Kamdem, D. P., and Zhang, J. 2001. Identification of cupric and cuprous copper in
copper naphthenate-treated wood by X-ray photoelectron spectroscopy. Holzforschung.
55(1): 16-20.

Kamdem, D. P., and McIntyre, C. R. 1998a. Chemical investigation of 23-year-old
CDDC-treated southern pine. Wood Fiber Sci. 30(1):64-7l.

Kamdem, D. P., Zhang, J., and Freeman, M. H. 1998b. The effect of post-steaming on
copper naphthenate-treated southern pine. Wood Fiber Sci. 30(2):210-217.

127

Kamdem, DR, and McIntyre, C. 1998c. A Chemical investigation of 23-year-old Copper
Dimethyldithiocarbamate (CDDC) Treated southern pine. Wood and Fiber Sci. 30(1):64-
71.

Kamdem, D.P., Fair, R., and Freeman, M. 1996. A Efficacy of Waterbome Emulsion of
Copper Naphthenate as Preservative for Northern Red Oak (Quercus rubra) and Soft
Maple (Acer rubrum), Holz als Roh-und Werkstoff 54: 183-187.

Kamdem, D.P., Gruber, K., and Freeman, M.1995. A Laboratory Evaluation of the Decay
Resistance of Red Oak (Quercus rubra) Pressure Treated with Copper Naphthenate,
Forest Prod. J. 45(9):74-76.

Kim, D.C., Song, H.Y., Back, J.H., and Kang, S.U. 1989. Studies on synthesis of
polyurethane ionomer-metal complex and catalytic properties and electrical conductivity.
Journal of Korean Fiber Society. 26(1):31-39.

Kwei, K.S., and Luongo, JP. 1967. Effect of ultraviolet irradiation on poly(vinyl
chloride) and modified poly(vinyl chloride). Polymer Preprints, 8(1): 588-90.

Li, B., Wang, J .Q., Ding, Y. 1999. Investigation of thermal degradation, flame retardance

and smoke suppression of rigid PVC by using the cone calorimeter. Gaofenzi Cailiao
Kexue Yu Gongcheng. 15(5): 124-127.

Nass, L.I. Editor. 1976. Encyclopedia of PVC. Marcel Dekker. New York.

Moitra, S. Raje, NS, and Shrinet, VP. 1998. Thermal stability characteristics of some
metal-chelate polymers from chemically modified PVC. In: Proceeding of IUPAC
International Symposium on Advances in Polymer Science and Technology. Chennai,
India. Jan 5-9, 1: 477-479.

Owen, ED. 1984. Degradation and Stabilization of PVC. Elsevier Applied Science
Publishers, London and New York.

Polyakov, V. N., Kovalenko, A. L., Zelentsov, V. V., Kryukov, V. V.1990. A
physicochemical study of solid bis-chelate compounds of copper(II) with aminoalcohols.
Russian Journal of Inorganic Chemistry. 35(3):408-411.

Ruddick, J NR. 1992. The fixation chemistry of copper-based wood preservatives. Proc.
Can. Wood Preserv. Assoc., 13:116-137.

Shibata, E.Y., Yamamoto, S., Kasai, NT. 2000. Bwhavior of PCDD/Fs formed during
heat treatment of PVC and copper oxide mixture. Organohalogen Compounds. 46:221-
223.

Sjostrom, E. 1993. Wood Chemistry, Fundamentals and Applications, Second Edition.
Academic Press, Inc. San Diego, USA.

128

Stames, W.H., Pike, R.D., Cole, J.R., Doyal, A.S., Kimlin, E.J., Lee, J .T., Murray, P.J.,
Quinlan, R.A., and Zhang, J. 2003. Cone calorimetric study of copper-promoted smoke

suppression and fire retardance of poly(vinyl chloride). Polym. Degradation and Stability.
82(1): 15-24.

Skoog, D.A., Holler, F. J., and Nieman, T.A. 1998. Principles of Instrumental Analysis,
Fifth Edition. Harcourt Brace & Company. Orlando, Florida.

Tauler, R., and Casassas, E. 1986. The complex formation of Cu(II) with mono- and di-
ehanolamine in aqueous solution. Inorganica Chimica Acta. 114(2): 203-209.

Vickerman, J. C. Editor. 1997. Surface Analysis — The principal Tecnniques. John Wiley
& Sons Ltd.

Wescott, L.D., Stames, W.H., Mujsce, A.M., and Linxwiler, RA. 1985. Mechanistic
studies on the role of copper- and molybdenum-containing species as flame and smoke
suppressants for poly(vinyl chloride). J. Analy. Appl. Pyrolysis. 8: 163-172.

Wickson, E. J. editor. 1993. Handbook of Polyvinyl Chloride Formulating. John Wiley &
Sons, New York.

Xue, H., Bhowmik, P., and Schlick, S. 1993. Direct detection of ionic clustering in
telechelic ionomers by DSC and ESR. Macromolecules. 26(13):3340-3343.

Zhang, J ., and Kamdem, D. P. 2000a. FI'IR characterization of copper ethanolamine-
wood interaction for wood preservation. Holzforschung. 54(2): 119-122.

Zhang, J ., and Kamdem, D. P. 2000b. Electron paramagnetic resonance spectroscopic
(EPR) study of copper amine treated southern pine in wood preservation. Holzforschung.
54(4): 343-348.

Zhang, J ., and Kamdem, D. P. 2000c. X-ray diffraction as an analytical method in wood
priservatives. Holzforschung. 54(1): 27-32.

Zhang, Jun; Kamdem, D. Pascal. 2000d. Interaction of copper-amine with southern pine:
retention and migration. Wood and Fiber Sci. 32(3), 332-339.

129

Chapter 7
BIOLOGICAL PERFORMANCE AND ENVIRONMENTAL DURABILITY
EVALUATION OF PVC/COPPER-TREATED WOOD FLOUR COMPOSITES

In the discussion in the previous chapters, copper amine was found to be an effective
coupling agent for the PVC-wood composite system. Mechanical properties, wettability
as well as interfacial adhesion, were significantly improved by wood surface copper
treatments. Copper amine is a main component of copper-based wood preservatives. In
this chapter, influences of Cu-treatments on biological stability and environmental
durability of PVC-wood flour composites are evaluated. Valuable information can be
generated from these tests about the effectiveness of copper amine acting as a biocide in
wood-plastic composites, especially when the composites contain 60% or more wood

flour.

7.1 Introduction

Wood undergoes biological degradation when subjected to attack by decay fungi,
whereas synthetic plastics are basically fungus free unless additives, such as plasticizers,
lubricants, stabilizers and colorants were used as ingredients (Hon and Shiraishi 1991,
Kamdem 1994a,b, Pendleton et al. 2002). In wood-plastic composites, the plastic matrix
is supposed to protect wood particles against fungal attack by encapsulating them, but
wood in composites may still be susceptible to fungal attack due to the presence of
uncovered wood in composites with high wood contents (Imamura et al. 1998,
Chetanachan et al. 2001, Simonsen et al. 2004).

Fungal decay of HDPE and PP-based wood-plastic was studied. A 10-20% weight

loss was observed in composites with a 50:50 wood to plastic ratio after a 4-month

130

exposure (Manning and Fred 2004). Decreases in the modulus of elasticity and creep
properties were found after only 4 weeks of exposure (Silva et al. 2002). The wood
content in composites was considered to be a key factor related to fungal decay
susceptibility (Simonsen et al. 2004, Pendleton et al. 2002). A scanning electron
microscopy study showed that mycelium grew in the interfacial gaps between wood and
plastic near the sample surface (Pendleton et al. 2002). Zinc borate was successfully
applied to wood-plastic composite systems as a biocide (Simonsen et al. 2004, Pendleton
et a1. 2002, Morrell 2001). It was reported that 2 wt% of zinc borate in a WPC
formulation can effectively prevent fungal decay (Pendleton et al. 2002).

Weathering wood-plastic composites resulted in thermal and photo degradation of
the composites leading to surface color change and a reduction of mechanical properties
(Stark et al. 2004). The environmental stress cracks were a crucial effect in weathering
degradation of wood-HDPE composites, and UV radiation enhanced this effect (Li 2000).
Accelerated weathering caused the formation of a white chalky layer on the surface of
PP/MAPP-natural fiber composites (Rowell et al. 2000). DSC, FTIR and SEM studies
of PP/wood-fiber composites after UV-aging indicated that the PP matrix degraded
during weathering giving rise to PP molecular chain scission and the formation of
carbonyl groups and hydroperoxides. This resulted in chemicrystallization and broad
surface cracking (Selden et al. 2004). Using glass fiber as hybrid reinforcement
significantly increased mechanical properties of PP/natural fiber composites after

weathering (Thwe and Liao 2002).

131

It was reported that wood flour acted as a chromophore substance in the PVC matrix
and accelerated the weathering degradation of PVC and composites, and composites had
larger color change than pure PVC (Matuana and Kamdem 2002, Matuana et al. 2000).

Application effectiveness of copper-based preservatives in fungal decay resistance of
pressure-treated wood as well as their weathering and photo stabilities has been widely
studied (Kamdem et al. 1995, 1996, 2002, Jusoh and Kamdem 2001, Nzokou and
Kamdem 2003, Laks et al. 2002). In this chapter, decay resistance of PVC/CuEA-treated
wood composites is evaluated. The influence of Cu-treatments on mechanical properties
and the color change of wood-PVC composites under artificial weathering are also

investigated.

7 .2 Experimental and Methods
7.2.1 Fungal Decay
7.2.1.1 Sample Preparation

PVC/CuEA-treated, untreated wood flour composite samples were laboratory
manufactured according to the method described in Chapter 2. The sample compositions
are listed in Table 7.1. Three different shapes of specimens were prepared for a fungal
decay test:

Blocks

Composites, PVC and red oak (Quercus rubra) boards were cut into 4 by 12.7 by
63.5 mm specimens for impact strength testing, and 4 by 12.7 by 127 mm specimens for
flexural property testing. The same sample size and shape specimens were also prepared

for outdoor weathering.

132

Table 7.1. Compositions of Experimental Samples with PVC, CuEA-Treated and
Untreated Wood Flour Contents, as well as Cu Concentrations in Treated Wood Flour

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample ID PVC2 Wood flour Cu content in wood flour
wt%) (wt%) (wt%)

PVC 100 0 0

64con 60 40 0 (untreated)
64Cu0.2 60 40 0.2
64Cu1.2 60 40 1.2

46con 40 60 0 (untreated)
46Cu0.2 4O 60 0.2
46Cul.2 40 60 1.2
Red oak 0 100 0
Sheets and particles

Sections of composites, PVC and oak boards were cut into 1 by 4 by 12.7 mm thin
sheets and other boards were ground into 20-30 mesh (0.86-0.52 m) particles for the
fungal decay test in order to increase the possible fungal attack surface area on wood.

Samples were conditioned at 23 °C, and a 65% relative humidity environment to
constant weights (W 1). The moisture contents of the composite samples varied between
1.0% for composite containing 40% wood flour and 3.0% for composites made of 60
wt% wood flour. The block samples were surface cleaned using ethanol. The sheet and
particle samples were sterilized in an autoclave at 110 °C for 30 minutes before being
placed in contact with the test fungus on the feeder strip in the soil containers. Eight

replicates were tested for each sample group.

7 .2.1.2 Accelerated Fungal Decay Tests
The fungal decay test of the WPC was conducted following an adopted AWPA
standard E10, with modifications as described below. A brown-rot fungus, Postia

placenta (Madison 698, ATCC 11538) (Pp) and a white-rot fungus, Irpex lacteus (FF

133

 

Lombard strain, Madison 517, ATCC 11245) (IL) were .used as decay fungi.
Autoclavable food containers with a volume of 600 cm3 were half-filled with forest soil
screened through a US. No. 6 sieve (3.35 mm). A feeder strip, with the size of 3 by 38
byl40 mm, was placed on top of the soil. Pine feeder strips were used for Pp, and aspen
for IL. The loosely capped containers were autoclaved at 110 °C for 40 minutes and
cooled to room temperature. Ten mm square fungus inoculum sections were cut from
near the leading edge of the mycelium in the Petri dish cultures. Sections of the
inoculums were placed in contact with an edge of the feeder strip in the soil container.
The culture containers were closed with the lid loosened, and incubated at the desired
temperature until the feeder strips were covered by mycelium. Sterilized composites, pure
PVC and oak samples were placed on top of the fungus covered feeder strip. The
containers were then incubated in a growth chamber at 26 0C, 70% RH, for 12 weeks.

At the end of the incubation period, the samples were removed from the culture
containers. The particle samples were carefully collected and the mycelium was brushed
off for all samples, and dried at 54 °C for 5 hours. The moisture contents of the composite
samples after incubation varied from 2.6 % to 5.2%. Then all samples were placed in the
conditioning room at 23 °C, 65% RH until the samples reached constant weights (W2)
and the moisture contents of the composite samples came back to 1.0 to 3.0%.

The weight loss before and after the fungal decay test was calculated from the

following equation:

Weight [03.9% = 100%- W2) (7-1)
1

 

134

The unnotched impact strengths of the samples before and after decay were
measured according to ASTM D4812 by using a Tinius Olsen Model 92T Impact Tester.
Flexural properties were tested according to ASTM D790 using the Instron testing
instrument Model 4206. A three-point bending test was performed with a span to
thickness ratio of 16, and the crosshead speed of the Instron Tester set at 2.2 mm/min.
SigmaStat 2.0 software and a one-way ANOVA method were used to compare the
difference between the means and standard deviations of the sample groups at a 95%

confidence interval.

7.2.2 Outdoor Weathering
7.2.2.1 Sample Preparation

Composites, PVC and red oak boards with the same compositions listed in Table 7.1
were cut into 4 by 12.7 by 63.5 mm specimens for impact strength testing, and 4 by 12.7
by 127 mm specimens for flexural modulus testing. Specimens were conditioned at 23 °C,
and a 65% relative humidity environment to constant weights before and after outdoor

weathering. Eight replicates were measured for each sample group.

7 .2.2.2. Outdoor Exposure

The outdoor weathering test was conducted following ASTM D1435 at the Tree
Research Center (TRC) at the Department of Forestry, Michigan State University. For 6
months, started from July, 2004 and ended in winter, January, 2005, the specimens were
exposed horizontally and directly contacted the soil ground instead of being positioned on

horizontal racks. For a total of 182 days of weathering, the average month temperature

135

decreased from 21.3 °C to -5.7 °C, the average relative humidity increased from 53% to
72%, there were 38 days of rainfall, 72 days with the minimum temperature below 0 °C
(freezing day), and 42 days with snow (National Climatic Data Center Data).

Testing of mechanical properties before and after outdoor weathering as well as a

statistical analysis were conducted in the same way described in section 7.2.1.2.

7.2.3 Artificial Accelerated Weathering
7.2.3.1 Sample Preparation

Composites, PVC and red oak boards with compositions listed in Table 7.2 were cut
into 4 by 32 by 76 mm specimens for accelerated weathering. The specimens were
conditioned at 23 °C, and a 65% relative humidity environment to constant weights

before accelerated weathering. Three replicates were tested for each sample group.

Table 7.2. Compositions of Experimental Samples with PVC, CuEA-Treated and
Untreated Wood Flour Contents, as well as Cu Concentrations in Treated Wood Flour

 

 

 

 

 

 

 

 

 

 

Sample ID PVC2 Wood flour Cu content in wood flour
(wt%) (wt%) (wt%)
PVC 100 0 0
PW64-con 60 4O 0 (untreated)
PW64-Cu0.5 60 40 0.5
PW55-con 50 50 0 (untreated)
PW55-Cu0.5 50 50 0.5
PW46-con 40 60 0 (untreated)
PW46-Cu0.5 40 60 0.5
Red oak 0 100 0

 

 

 

 

7.2.3.2 Accelerated Weathering and Color Stability
The color stability of the composite samples was evaluated by estimating the color

change caused by an artificially accelerated weathering schedule as described below.

136

 

The samples were subjected to accelerated weathering by exposure to 340nm Fluorescent
UV lamps in the QUV Accelerated Weathering Tester. The weathering schedule involved
a continuous light irradiation for 2 hours followed with a water spray for 18 minutes
according to ASTM G-154. The average irradiance was 0.85 W/m2 at 340 nm
wavelengths with a chamber temperature of approximately 45°C. The output of the 340
nm lamp was concentrated in the UV region ranging from 400 to 300 nm with the apex at
340 nm. Specimens were mounted on aluminum panels and placed inside the QUV
chamber. The changes on the samples surface were monitored by assessing the
parameters of the sample surface color.

The surface color was determined according to the ISO 2470 Standard using a
Microflash model 200 Reflectometer manufactured by Datacolor International (ISO
2470). The CIELAB system is characterized by three parameters, L’, a. and b'. The L.
axis is defined as lightness, a* and b* are chromaticity coordinates. In the CIELAB
coordinates, +a* is for red, -a* for green, +b* for yellow, -b* for blue, and L* varies from
100 (white) to zero (black). L*, a* and b* color coordinates of each sample before and
after exposure to UV light irradiation were obtained based on a D65 light source as
established by the CIE 1976 (Billmeyer and Saltzman 1981). These values were used to

calculate the color change AE’ as a function of the UV-irradiation period according to

Equations 7-2 to 7-5.

AL = Lf ‘ Li (7-2)
* III *
Ad = af - (1i (7_3)

137

Ab,“ = b; —b* (7_4)

 

an ... ... .. (7-5)
AE =JAL2+Aa 2 +Ab 2

Where, AL‘, Aa' and Ab: are the change between the initial values (L;*, a;* and bi*)

and the final (Lf*, af* and bf*) values. A low AE’ corresponds to a low color change.

No attempt was made to measure the mass loss during the exposure.
A summary of sample sizes for biological performance evaluation as well as applied

standards are listed in Table 7.3.

Table 7.3. Sample Sizes and Applied Standards for Biological and Weathering
Performance Evaluation

 

 

 

 

 

Sample size (mm) Test Standard
4 x 12.7 x 63.5 Impact/Fungus decay ASTM D4812/AWPA E10
4 x 12.7 x 127 Flexural/Fungus decay ASTM D790/AWPA E10
1 x 4 x 12.7 Fungus decay AWPA E10
0.86-0.52 (particles) Fungus decay AWPA E10

 

4 x 12.7 x 63.5 Impact/Outdoor weathering ASTM D4812/ASTM D1453
4 x 12.7 x 127 Flexural/ Outdoor weathering ASTM D790/ ASTM D1453
4 x 32 x 76 Artificial weathering ASTM GlS4/ISO 2470

 

 

 

 

 

 

 

7.3 Results and Discussion
7.3.1 Fungal Decay Resistance of PVC/Copper Amine-Treated Wood Composites

A summary of the weight loss of the block, sheet and particle samples are listed in
Table 7.4. Figure 7.1 shows the weight losses of the block samples of PVC, oak and
composites made of PVC, CuEA-treated and untreated wood flour with various PVC to

wood ratios as well as different Cu concentrations in wood flour after the 12 week fungal

138

Table 7.4. Weight Loss of PVC, Composites and Red Oak after 3 Months Fungus Decay

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Test
Sample ID Block Thin sheet Particle
Pp IL Pp IL Pp IL

PVC 0.17 0.18 0.47 0.45 0.74 0.78
(0.01) (0.01) (0.04) (0.06) (0.07) (0.06)

64con 2.39 2.58 18.56 28.61 30.93 38.10
(0.19) (0.16) (1.63) (2.16) (1.38) (2.63)

64Cu0.2 1.92 1.76 9.82 9.07 25.86 26.27
(0.13) (0.09) (1.12) (0.93) (2.00) (2.77)

64Cu1.2 1.65 1.16 4.32 4.56 20.24 16.68
(0.10) (0.09) (0.63) (0.74) (1.70) (1.70)

46con 3.41 3.49 23.43 35.27 45.04 57.26
(0.19) 10.14) (1.16) (3.06) (4.78) (2.33)

46Cu0.2 2.91 2.77 15.94 17.87 37.04 43.38
(0.14) (0.07) (1.66) (1.07) (2.80) (3.81)

46Cul .2 2.53 1.96 7.91 9.43 30.89 19.62
(0.1 1) (0.1 1) (1.09) (0.78) (2.46) (1.70)

Red oak 51.07 53.99 69.80 71.24 90.49 95.81
(2.1) (4.3) (2.69) (4.56) (5.17) (4.03)

 

 

*Stand deviations are listed in parentheses.

Weight Loss (%)
no on
p o
O O

..L

.0

o
1

 

.0
(3

OFF EIIL

 

 

 

 

 

PVC 64con 64Cu0.2 64Cu12 460m 460u02 4601112 Oak

Figure 7.1. Weight losses of block samples of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour after the 12 weeks decay test by Pp and IL

139

Before Decay After Decay

VF? iii—V" ~ W. — 7" .1'7 "7;”. ('9 .~ 1 . ~-

      
   

->*kw<** .

Figure 7.2. Comparison of composite surface before and after colonized by white-rot
fungus, IL, (1) composite made of 60 wt% untreated wood flour; (2) composite made of
60 wt% of Cu-treated wood flour

decay test. Almost no decay weight loss was found in pure PVC for both Pp and IL,
indicating that PVC was basically immune from brown and white-rot fungi. Untreated
oak had the largest weight losses, over 50% under the attacks from both PP and IL,
whereas block samples of PVC/wood flour composites exhibited much lower weight
losses, less than 4% compared to untreated wood. This suggested that wood was
effectively protected by the plastic matrix after blending and it was hard for the fungus to
penetrate the plastic barrier to reach the wood inside the composite so that fungal decay
was limited to near surface area. Enlarging the wood content in the composites slightly
increased weight loss and increasing the Cu concentration in the wood showed a trend of
reducing weight loss. However, because of low level weight loss, these trends were not
shown clearly. It can be seen that the wood flour particles on composite surfaces became
very white after being decayed by the white-rot fungus, IL. Figure 7.2 compared the

surface before and after the IL attack. The composite surfaces decayed by the brown-rot

140

fungus, Pp, didn’t display much color difference because of the similar wood background
color.

Figures 7.3 and 7.4 illustrate the weight loss of the thin sheet and particle samples of
the composites, PVC and untreated oak after exposure to decay fungi: Pp and IL. No
decay happened in pure PVC, but important weight losses were observed in the
composite sheets and particles, extremely high weight losses were also found in the sheet
and particle samples of untreated red oak. The untreated oak sheets lost about 70% of
their weight during the fungal decay test, and weight losses almost reached 100% for
their particle samples. The composite particles had larger decay weight losses than the
sheet samples, which could be attributed to the enlarged wood surface area available for
fungal colonization. For untreated wood-PVC composites, the more wood content, the
higher the weight loss. Particle weight loss under white-rot (IL) attack increased about
50% when the wood content in the composite enlarged from 40% to 60%. Additionally,
1L decay caused higher weight loss than Pp decay, suggesting that the red oak was more
susceptible to white-rot, such as IL decay. The weight loss of PVC/wood flour
composites was dramatically decreased by wood surface copper amine treatments. The IL
decay weight loss of the composite sheets made of 60 wt% PVC and 40 wt% wood flour
decreased about 65% when using Cu-treated wood flour containing 0.2 wt% Cu. Another
50% decrease was achieved by increasing the Cu content in wood flour to 1.2 wt%. A
similar trend was also found in composite made of 60 wt% Cu-treated wood flour. After
using treated wood flour containing 1.2 wt% Cu, the IL decay weight loss of composite
particles with 60 wt% wood was almost the same as that of composite particles consisting

of 40 wt% treated wood flour. This indicated that copper amine was a very effective

141

80.0 T

70.0. DPP IIL

60.0 -

Weight Loss (%)
h 0|
9 P
o o

(.0
.0
O
4_ I

20.0 a

10.0 -

 

 

 

 

 

   

 

 

 

..._- I - 1

PVC 64con 64Cu0.2 64Cu1.2 4600n 460u02 46001.2 Oak

Figure 7.3. Weight losses of sheet samples of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour after the 12 week decay test by Pp and IL

100.0 -

9°91 DPP IIL

   

Weight Loss (%)
8
O

 

 

 

 

 

20.0 ~
10.0 + I
0.0 I -

PVC 64con 64Cu0.2 640u1.2 4600n 460u0.2 460u1.2 Oak

Figure 7.4. Weight losses of particle samples of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour after the 12 week decay test by Pp and IL

142

decay inhibitor for white-rot fungus, IL in PVC/wood flour composites, especially when
composites had high wood flour contents. The decay protection effect increased with an
increased level of copper in the wood, and also with the level of PVC. PVC is good for
moisture control. The less water content, the less decay happens in composites. Copper is
known as a fungicide. The combination of cupric ion Cu2+ with thiol groups in the fungi
body led to denaturation of proteins and enzymes and interference with the activity of the
pyruvate dehydrogenase system (Goodel et al. 2003).

The decay resistance of Cu-treated wood flour against the brown-rot fungus, Pp was
not as good as that against IL. For the particle samples, from Figure 7.4, a decrease in
weight loss of only about 20% was obtained by increasing Cu concentration in wood
flour. At a 60 wt% wood flour loading level, the IL decay weight loss of composite made
of untreated wood was about 25% higher than the Pp weight loss. However, after using
Cu-treated wood flour containing 1.2 wt% Cu, the Pp weight loss was 30% larger than
the IL weight loss. Postia placenta has been reported as tolerant to copper (Clausen and
Green 2003). The decreased effectiveness of Cu-treatments for Pp decay was associated
with this effect, which caused the relatively lower Pp decay resistance of Cu-treated
composites.

Impact strength of composites, PVC, and oak before and after the fungal decay test
are presented in Figure 7.5. Impact strength of untreated oak was 10 folds higher than
those of composites and pure PVC. After decay, the impact strength of oak was reduced
about 50%, but only 15 to 20% reduction was observed with composites and PVC, which
can be seen more clearly in Figure 7.6. For untreated wood-PVC composites, enlarging

wood content in composites resulted in higher impact strength reduction. Impact strength

143

80.0 -

DUneroosed : 800.0
70.0 J 9 pp ,
. - 700.0
A 60.0 q Ill. _ E
g 4 F 600.0 v
V 50.0 - _ 0 §
- 500.
‘ ‘8
g 400 - '
3 ‘ - 400.0 g
g 3°") 1 - 300.0 §
20.0 - - 200.0 g.
. )- 5
10-0 1 - 100.0
0.0 t 0.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PVC 6400f! 640u02 64Cu1.2 4600!! 46000.2 460u12 Oak

Figure 7.5. Impact strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after decay test by Pp and IL

or
.0
o

4

OFF IIL

«5 0|
.0 .o
O O
L

I
4‘1

to
P
0

Impact strength reduction (%)
(to
o
O

.5
o
o

 

 

 

 

P
0

PVC 64

con 64CuO.2 64001.2 460m 46CuO.2 46Cu1.2 Oak

Figure 7.6. Impact strength reductions of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour before and after decay test by Pp and IL

144

me

1000 L D Unexposed H:
I lL

80.0 -

 

40.0 :

Flexrnl strength (MPa)
8
O

20.0 3

 

 

 

 

 

 

 

 

 

 

0.0 ‘

 

PVC 64000 64000.2 64001.2 4600n 46000.2 46001.2 Oak

Figure 7.7. Flexural strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after decay test by Pp and IL

70.0 -
60.0 - Cl PP I IL

50.0 «

N (D A
.o .o s:
O O O

Flexural strength reduction (95)

..s
.0
O

 

 

 

 

p
O
l 1

PVC 64000 640002 640u1.2 4600n 46000.2 46001.2 Oek

Figure 7.8. Flexural strength reductions of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour before and after decay test by Pp and IL

145

173.0 —
12.0 3 DUnexposed
11.0 1 Up?
10.0 i ' "-
9.0 3
8.0 3
7.0 1
6.0 1
5.0 —
4.0 3
3.0 3
2.0 3
1.0 l
0.0 .

Flexural modulus (GPa)

 

 

 

 

 

 

 

 

 

PVC 64con 64000.2 64001.2 46000 46000.2 46001.2 Oak

Figure 7.9. Flexural modulus of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after decay test by Pp and IL

70.0 -
1

60.0 - OPP I IL

50.0 -

40.0 -

30.0 -

20.0 -

Flexural modulus reduction (96)

10.0 1

 

 

 

 

0.0 1
PVC 640011 64000.2 64001.2 4600n 46000.2 46001.2 Oak

Figure 7.10. Flexural modulus reductions of PVC, oak and composites made of PVC,
CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour before and after decay test by Pp and IL

146

 

decay increases in flexural strength were found when using Cu-treated wood flour of
composites was increased at least 40% after using Cu-treated wood flour and this
enhancement can be maintained after fungal attacking (Figure 7.5). Impact strength loss
was reduced by wood Cu-treatments, especially for high wood content composite
samples. At a 60 wt% wood loading level in composite formulation, Cu-treatments
reduced the impact strength by 30% loss after fungal decay.

Flexural strength reduction comparison of composites, PVC, and oak before and after
the fungal decay test are illustrated in Figures 7.7 and 7.8. Larger flexural strength
improvements by the Cu-treatments were observed after fungal decay compared to the
corresponding increases of unexposed samples (Figure 7.7). This suggests that fungal
resistance of composites was improved after wood Cu-treatments. Oak showed very large
flexural strength reductions, about 50% in contrast to only 13% loss in the composite
made of PVC and 40 wt% Cu-treated wood flour at a 0.2 wt% Cu concentration (Figure
7.8). The composite made of 60 wt% untreated wood flour exhibited the highest flexural
strength loss, up to 30%, among composite samples. After wood surface treatments,
flexural strength loss of this composite was controlled within 20%.

Similar trends were also observed in the changes on flexural modulus of PVC/treated,
untreated wood flour composites as presented in Figures 7.9 and 7.10. Wood had much
higher flexural modulus than composites and pure PVC before decay, and after IL decay,
their average flexural modulus dropped about 60% to the same level as those of
composites. Composites had lager flexural modulus than pure PVC, and Cu-treatments

showed little effect on the modulus of composites. Flexural modulus of composites

147

120.0 -
PVC
100.0 1
80.0 -
60.0 -

40.0 -

Impact Strength (Jim)

20.0 J

 

 

0.0 J

 

0% 60%
Relative Humidity (%)

80.0 -
70.0 —
60.0 -
50.0 -
40.0 -
30.0 -
20.0 -
10.0 -

0.0 -

PW64con

Impact Strength (Jim)

 

   

0% 60%
Relative Humidity (%)

120.0 -
PW64000.2
100.0 —
80.0 4
60.0 i

40.0 -

Impact Stength (Jim)

20.0 -

 

 

0.0 '1

 

0% 60%
Relative Humldlty (%)

Figure 7.11. Impact strength of PVC, composites made with PVC and untreated, Cu-
treated wood flour after exposed in the environment with different relative humidity (0%
RH means properties measured after oven-dry)

148

5.0 -
PVC

4.0 -

Flexural Modulus (GPa)

 

 

 

0% 60% 95%
Relative Humidity (%)

Flexural Modulus (GPa)

 

 

Relative Humidity (%)

 

 

7.0 ~
A PW64000.2
a
n.
9
fl
5
5
8
a
8
i
t:
0% 60% 95%
Relative Humidity (96)

Figure 7.12. Flexural modulus of PVC, composites made with PVC and untreated, Cu-
treated wood flour after exposed in the environment with different relative humidity (0%
RH means properties measured after oven-dry)

149

Table 7.5. Moisture Contents of PVC and composites after exposed in the environment

with 60 % and 95% relative humidity

 

 

 

 

 

 

 

 

 

 

 

Relative Humidity (%) 60% 95%
PVC 0 0
PW64con 2.64 (0.21) 3.99 (0.27)
PW64Cu0.2 2.53 (0.23) 3.26 (0.30)
PW55con 3.71 (0.15) 4.99 (0.26)
Pw55CuO.2 3.34 (0.10) 4.63 (0.41)
PW46con 4.32 (0.18) 7.82 (0.41)
Pw46Cu0.2 3.931008) 7.77 (0.31)

 

 

decreased after fungal colonization (Figure 7.9). As the content of untreated wood
increased from 40 to 60 wt%, modulus reductions of composites enlarged from about
20% to over 30% as shown in Figure 7.10. Wood Cu-treatment improved fungal decay
resistance of composites, leading to smaller modulus loss than untreated composite.
Although fungal colonization showed little effect on the weight loss of PVC/wood
flour composites, about 20 to 30% decrease in mechanical properties of composites was
generated. It is well established that moisture content (MC) influences the properties of
wood and wood products (Kamdem et al. 1995, 1994a,b). Impact strength and flexural
modulus of PVC and composites after exposed in the environment with different relative
humidity are shown in Figures 7.11 and 7.12, and the moisture contents of composites in
different environmental relative humidity (RH) are listed in Table 7.5. Statistical analysis
indicated that there was no significant difference between mechanical properties of PVC
and composites after exposing samples in different environmental RH. Influence of
moisture contents in composites on their mechanical properties was negligible so that the
mechanical properties reductions after fungus attack were mainly attributed to the fungal

decay of the materials.

150

 

- DUnex ed - 1000.0
90.0 « pas ~
- lWeathered j 90°-0
8°”: 800.0 :3;
‘g‘ 70.0 . . ..
J 700.0 .18:
g 60.01 ‘ 600.0 3
g 50.0 q . 500.0 g.
C " 4
E 40.0 4 A 4000 .
g 30.0 ~ 300.0 g
20.01 : 200.0 5
10.0 . _ 100.0
0.0 d 0.0

 

 

 

 

 

 

 

 

 

 

 

PVC 64000 640002 64001.2 46000 46000.2 460012 Oak

Figure 7.13. Impact strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after 6 months outdoor weathering

110.0 a
100.0 ‘ CI Unexposed

9°-° ‘ IWeathered
80.0 l

70.0 4
60.0 -
50.0 -
40.0 -
30.0 J
20.0 -
10.0 -

0.0

Flexural strength (MPa)

 

 

 

 

 

 

 

 

 

 

 

 

 

PVC 64con 64000.2 64001.2 46000 46000.2 46001.2 Oak

Figure 7.14. Flexural strength of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after 6 months outdoor weathering

151

13.0 3
12.0 -
11.0 3
10.0 3
9.0 3
8.0 3
7.0 3
6.0 3
5.0 1
4.0 3
3.0
2.0
1.0

Flexural modulus (GPa)

 

0.0 -

4
-r
.4
..
-i

U Unexposed
I Weathered

 

 

 

 

 

 

 

 

PVC 64con 64000.2 64001.2 46000 46000.2 46001.2 Oak

Figure 7.15. Flexural modulus of PVC, oak and composites made of PVC, CuEA-treated
and untreated wood flour with various PVC to wood ratios as well as different Cu
concentrations in wood flour before and after 6 months outdoor weathering

50.0 3

Property reduction (96)

10.0 -

D Impact strength
a Flexural strength
I Flexural modulus

 

 

 

 

 

0.0

 

 

 

 

PVC 6400n 640002 640012 46con 460002 460012 Oak

Figure 7.16. Mechanical properties reductions of PVC, oak and composites made of
PVC, CuEA-treated and untreated wood flour with various PVC to wood ratios as well as
different Cu concentrations in wood flour before and after 6 months outdoor weathering

152

7.3.2 Effects of Copper Amine Treatments on Natural Weathering Durability of
PVC/Copper Amine-Treated Wood Composites

Wood and wood composites in outdoor applications are often subjected to
environmental parameters, such as sunlight, rain, and temperature. Sunlight, rain, and
environmental temperatures, therefore, influence the performances of such products. In
order to evaluate the performance of composites as well as the effects of Cu-treatments
on the environmental stability of composites under the influence of all these factors, an
outdoor weathering test was conducted and the mechanical properties of composites
before and after outdoor weathering were measured.

Impact and flexural strength as well as flexural modulus of PVC/Cu-treated,
untreated wood composites, pure PVC, and red oak are illustrated in Figures 7.13 to 7.15.
Comparison of mechanical property reductions is given in Figure 7.16. Still, solid wood
showed much higher impact, flexural strength, and flexural modulus than composites and
pure PVC. Unlike in the decay test, after 6 months of outdoor weathering, solid wood
exhibited similar property reductions, about 15%, to composites consisting of 40 % wood
flour. Compared to composites made of untreated wood flour, composites consisting of
Cu-treated wood flour had increased impact and flexural strength both before and after
outdoor weathering (Figures 7.13 and 7.14). Although Cu-treatment had little effect on
flexural modulus, an increase of about 25% was found on weathered modulus of
composites made of 60 wt% Cu-treated wood compared to the weathered modulus of
corresponding composites made of untreated wood flour (Figure 7.15). It can be seen
from Figure 7.16 that pure PVC had the smallest mechanical property reductions after
outdoor weathering, which is about 7%. The property reductions of composites made of

40 wt% untreated wood was around 16%, and at least 10% decrease in the property loss

153

was observed when using Cu-treated wood. A more remarkable effect was found in
composites with 60 wt% wood loading level. Property loss after weathering was
dramatically reduced when using Cu-treated wood flour (Figure 7.16), about 45% in
impact strength at the 0.2 wt% Cu concentration level, and 30 to 35% in flexural strength
and modulus. This suggested that Cu-treatment increased the environmental stability of
PVC/wood flour composites, especially for composites with 60 wt% wood. Mechanical
properties lost about 45% for the composite made of 60 wt% untreated wood flour. A
significant amount of surface and interfacial cracks were observed on the cross section of
this composite sample, which may be the result of the “freeze and thaw” effect during
winter. However, there were no noticeable cracks were found on cross sections of other
samples. Weathering of wood is a surface phenomenon. UV light cannot penetrate deep
in wood, and wood lignin may act as a free radical “inhibitor” (Hon and Shiraishi 1991,
Kamdem and Grdlier 2002). In wood-plastic composites, interfacial cracking and bond
separation were generated by swelling/drying of wood particles in “freeze and thaw”
circles. The remarkable weathering stability enhancement of Cu-treated wood at high
wood loading level could be attributed to the increased interfacial adhesion and improved

interphase between wood and PVC matrix after wood surface Cu-treatments.

7.3.3 Color Stability of PVC/Copper Amine-Treated, Untreated Wood Composites
during Accelerated Weathering

One of the advantages of wood-plastic composites is their appearance of natural
wood (J iang and Kamdem 2004). Photodegradation and discoloration occurred when
wood is exposed to UV light (Hon and Shiraishi 1991, Kamdem et al. 2004). This

artificial accelerated weathering test was conducted in order to evaluate the color stability

154

—x— PVC conpound ——+— Oak
—0— PW64-con —s— Rivals-010.5

100‘ -°-A---PW55-con ---A---PW55-0.10.5
-.o_. M4500" --.-- M45005
—- — — -—"x
6.0 3 xx}?!
)1:

 

       
       

 

 

 

‘fix&$*:23'--t.n ""h” ................ 8- ........
s+ ‘._°P..—°. O o... . .
‘+~+-l--+—+\+ . . . 1
‘ T ' T "+ i r .- 4. :
.1601
0 400 800 1200 1600 2000 2400 2800 3200 3600

Duration of the weathering, hours

Figure 7.17. Color changes in Aa (green/red) of PVC, oak and composites made of PVC,
untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to wood

ratios during accelerated weathering

20.0 w. —x— PVC conpound —+— Oak
—0— PW64-con —s— PW64-000.5
16.0 ---A~-° PW55-con

....“ PW55-000.5
—-e—- ems-010.5

‘ —-o—- PW46con
12.0 1 ,x-x'*'x‘*

 

 

 

 

2800 3200 3600

 

1200 16m 2000 2400

o 400 800
Duration of the weathering, hours

Figure 7.18. Color changes in Ab (blue/yellow) of PVC, oak and composites made of
PVC, untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to

wood ratios during accelerated weathering

155

18.0- -"
13.03. :

 

AL
a
o

 

 

 

 

-70

\x-X-x '
.12,0.------AFW55-con ---A---PW55-GJO.5
'X'X --o-- PW46-con --e—- lame-010.5

___. _ __... ’X—X—‘x

 

0 400 800 1200 1000 2000 2400 2800 3200 3600
Duration of the weathering, hours

Figure 7.19. Color changes in AL (black/white) of PVC, oak and composites made of
PVC, untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to
wood ratios during accelerated weathering

o. ..‘
&.01 '.+..+..*..+ *..*-'+..+.._...e*

28.0 4

24.0 <

20.03 :,'

A5

16.0 -

  
 

120 ~ -—x— WCconpomd—r—Oek

—o—PW64-con +rwe4-oros

 

 

8.0T ---A---PW55-con ---A---PW55-000.5

4.0 —-o-- Fw46-con —-e—- PW46-010.5

0.0 . f ' * ' ‘ T ’ ' ' T '
1200 1000 2000 2400 2600 3200 3600

Duration of the weathering, hours

Figure 7.20. Total color changes in AB of PVC, oak and composites made of PVC,
untreated and CuEA-treated wood flour containing 0.2 wt% Cu at various PVC to wood
ratios during accelerated weathering

156

of PVC/untreated, Cu-treated wood flour composites and to investigate effects of wood
surface Cu-treatment on color and photo stabilities of composites.

Color change components, Aa, Ab, AL and total color change AE of PVC/Cu-treated
and untreated wood flour composites, pure PVC, and red oak during 3400 hours artificial
weathering are presented in Figures 7.17 to 7.20. Positive Aa means changing toward red,
negative Aa means changing toward green. It can be seen from Figure 7.15 that during
artificial weathering, the color of pure PVC changed toward red, whereas solid oak
toward green and both of these two trends became stable after 800 hours weathering.
PVC/untreated wood flour composites became greener but in a smaller degree than solid
oak. Aa became stable after 400 hours weathering. Composites made of PVC and Cu-
treated wood showed smallest changes in Aa indicating only a slight change toward green
during weathering. Positive Ab represents changing toward yellow and negative Ab
toward blue. As illustrated in Figure 7.18, oak and composite made of untreated wood
exhibited negative Ab, and PVC and composites consisted of Cu-treated wood showed
positive Ab. Color changes in Ab of composites were not as intense as PVC and solid oak,
and their Ab were between PVC and oak.

Solid oak and composites colors faded light to gray, and PVC became darker during
weathering, as displayed in Figure 7.19. Composites showed lager AL than oak, and AL
increased as increasing wood content in composites. Composites made of Cu-treated
wood flour demonstrated higher AL values than untreated wood composites indicating a
more intensive fading effect. The fading effect of composites happened very quickly

within the first 400 hours of weathering and then became stable.

157

Lignin

 

 

 

OCH3
OH
hv sensitizers
Lignin Lignin Lignin
O
OCH3 OCHa 0011;,
O s
Lignin Lignin
+ OH
OCHa 001-13
0 ' o
Lignin
CH30H ...
O
O
o-quinone

 

Lignin

Lignin OCH" Lignin

HO? L001:“ 00143

“OH

 

Lignin l

 

Lignin
OCH3
3 H o'
OCHa
O
> polymers

 

Figure 7.21. Pathway of formation of quinonoid in wood lignin during photodegradtion

As illustrated in Figure 7.20, the total color change (AE) of composites made of

untreated wood exhibited larger AE than oak during the first 800 hours weathering, after

that AE became stable and lower than the final color change of oak. Enlarging wood

content in composite increased AE. Composites made of 40 wt% Cu-treated wood flour

showed a similar color change pattern to composites containing untreated wood, and had

final AE closer to oak and PVC. When the content of Cu-treated wood flour increased to

158

60 wt%, total color changes of the composite significantly increased. Increasing Cu-
treated wood flour content in PVC/wood flour composite formulation had the tendency to
intensify the discoloration of composites.

Photodegradation of wood is an ultraviolet light component-induced free-radical
reaction. Phenolic hydroxyl group in wood lignin generates free radicals resulting in the
photodissociation and formation of quinonoid structures as shown in Figure 7.21 (Hon
and Shiraishi 1991, Kamdem and Grelier 2002). These free-radical reactions are

responsible for the discoloration and deterioration of wood surface.

 

W CH2— CH— CH2 W __hv9 W (CH=CH)n W + HHCI F

Figure 7.22. Photodegradation of PVC resulting in the formation of HCl

PVC is also susceptible to UV radiation. When PVC was exposed to UV,
discoloration occurred rapidly because of the formation of conjugated polyene sequences,
in the same time, large amounts of hydrogen chloride (HCl) were released (Figure 7.22)
and the polymer chains underwent scissions and crosslinking (Owen 1984). In the
presence of oxygen, hydroperoxide groups were generated during the photodegradation
(Owen l984). Both HCl and peroxide groups are good bleaching agents for wood,

leading to larger AL of composites.

159

7 .4 Conclusions

Block samples of PVC/wood flour composites exhibited much lower weight loss,
compared to untreated wood suggesting that wood was effectively encapsulated and
protected by plastic matrix after blending and fungal decay was only limited in near
surface area. Composite particles and sheets had larger decay weight losses than block
samples, which could be attributed to the enlarged wood surface area available for fungal
colonization.

For untreated wood-PVC composites, the more wood content, the higher weight loss.
Fungal decay weight loss of PVC/wood flour composites was dramatically decreased by
wood surface copper amine treatments. Copper amine was a very effective decay
inhibitor for white-rot fungus, IL in PVC/wood flour composites, especially when
composites had high wood flour contents. Decreased effectiveness of Cu-treatments for
Pp decay was associated with copper tolerance natural of Pp. Decreased mechanical
property losses were also obtained in Cu-treated composites.

Extensive cracking was observed on the cross section of composites containing high
amounts of untreated wood flour leading to larger mechanical property loss, and cracking
was absent on Cu-treated wood-PVC composites. Wood Cu-treatment increased the
environmental stability of PVC/wood flour composites, especially for composites with
high-level wood contents, resulting in lower mechanical property loss. This could be
attributed to the increased interfacial adhesion and improved interphase between wood
and PVC matrix after wood surface Cu-treatments. However, increasing Cu-treated wood
flour content in PVC/wood flour composite formulation had the tendency to intensify the

discoloration of composites. This may be associated with the bleaching effect of

160

 

hydrogen chloride and peroxides, which were generated during the photodegradation of

PVC. This bleaching effect became stronger in the presence of Cu.

161

 

References

American Society for Testing and Materials (ASTM). ASTM G 154 - 98. 1998. In: 1998
Annual Book of ASTM Standards. West Conshohocken, PA.

Billmeyer, F.W., Saltzman, M. 1981. In: Principles of Color Technology. John Viley &
Sons, New York.

Chetanachan, W., Sookkho, D., Sutthitavil, W., Chantasatrasamy, N., Sinserrnsuksakul, R.
2001. PVC wood: a new look in construction. Journal of Vinyl & Additive Technology.
7(3):l34-137.

Clausen, C.A., and Green, F. 2003. Oxalic acid pverproduction by copper-tolerant brown-
rot basidiomycetes on southern yellow pine treated with copper-based preservatives.
International Biodeterioration & Biodegradation. 51:139-144.

Green, F., and Clausen, CA. 2003. Copper tolerance of brown-rot fungi: time course of
oxalic acid production. International Biodeterioration & Biodegradation. 51:145-149.

Goodell, B., Nicholas D. D., and Schultz, T.P. editors. Wood deterioration and
preservation : advances in our changing world. American Chemical Society, Washington,
DC.

Hon, D. S., and Shiraishi, N. 1991. Wood and Cellulosic Chemistry. Marcel Dekker, Inc.,
Imamura, Y., Takahashi, M., Ryu, J.Y., and Kajita, H. 1998. Distribution of polymers in
cell walls and their effect on the decay resistance of wood-plastic composites.

Biocontrol. Sci. 3(2): 109-1 12.

International Organization for Standardization (ISO). 1999. ISO 2470. In: ISO Standards.
Geneva, Switzerland: ISO.

Jiang, H., and Kamdem, DR 2004. Development of poly(vinyl chloride)/wood
composites. A literature review. J.Viny1 Addit. Technol.10(2):59-69.

Jusoh, I., and Kamdem, DP. 2001. Laboratory evaluation of the decay resistance of
rubberwood and CCA type C-treated rubberwood. Holzforschung 55:49-52.

Kamdem, D.P. 1994a. Fungal decay resistance of aspen blocks treated with heartwood
extracts. Forest Products J. 44(1):30-32.

Kamdem, DP, and Sean, S.T. 1994b. The durability of phenolic bonded particleboard

made of decay-resistant black locust and non-durable aspen. Forest Products J. 44(2):65-
68.

162

Kamdem, D.P., Gruber, K., and Freeman, M. 1995. A laboratory evaluation of the decay
resistance of red oak (Quercus rubra) pressure treated with copper naphthenate. Forest
Prod. J. 45(9):74-76.

Kamdem, D.P., Fair, R., and Freeman, M. 1996. A efficacy of waterborne emulsion of
copper naphthenate as preservative for northern red oak (Quercus rubra) and soft maple
(Acer rubrum), Holz als Roh-und Werkstoff 54:183-187.

Kamdem, DP, and Grdlier, S. 2002. Surface roughness and color change of copper
amine and UV absorber-treated red maple (Acer rubrum). Holzforschung. 56:473-478.

Kamdem, D.P., Jiang, H., Cui, W., Freed, J ., and Matuana, L.M. 2004. Properties of
wood plastic composites made of recycled HDPE and wood flour from CCA-treated
wood removed from service. Composites: Part A. 35:347-355.

Laks, P.E., Richter, D.L., Larkin, GM. 2002. Fungal susceptibility of interior
commercial building panels. Forest Products J. 52(5):41-44.

Li, R. 2000. Environmental degradation of wood-HDPE composite. Polymer Degradation
and Stability. 70(2): 135-145.

Manning, M.J., and Ascherl, F. 2004. Borates as fungicides in wood-plastic composites.
International Conference on Woodfiber-Plastic Composites, 7th, Madison, WI, United
States, May 19-20, 2003, Meeting Date 2003: 69-78.

Matuana, L., Kamdem, DR, and Zhang, J. 2001. Photoageing and stabilization of rigid
PVC/wood- fiber composites. J. Appl. Polym. Sci. 80:1943-1950.

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

Morrell, J.J. 2001. Biodeterioration of wood-based composites and its prevention.
Proceedings of the Washington State University International Particleboard/Composite
Materials Symposium, 35mzl7-24.

National Climatic Data Center, Local Climatologically Data, National Weather Service
Forecast Office.

Nzokou, P., and Kamdem, DP. 2003. Fungal decay resistance of non-durable aspen
wood treated with extractives from African padauk (pterocarpus soyauxil). Journal of
Tropical Forest Products. 9(1&2): 125-133.

Owen, ED. 1984. Degradation and Stabilization of PVC. Elsevier Applied Science
Publishers, London and New York.

163

Pendleton, D.E., Hoffard, T.A., Adcock, T., Woodward, B., and Wolcott, M. 2002.
Durability of an extruded HDPE/wood composite. Forest Product J. 52(6):21-27.

Rowell, R.M., Lange, SE, and Jacobson, RE. 2000. Weathering performance of plant-
fiber/thermoplastic composites. Mol. Cryst. and Liq. Cryst., 353:85-94.

Selden, R., Nystrom, B., and Langstrom, R. 2004. UV aging of poly(propylene)/wood-
fiber composites. Polym. Composite. 25(5):543-553.

Silva, A., Freitag, C., Morrell, J .J ., and Gartner, B. 2002. Effect of fungal attack on creep
behavior and strength of wood plastic composites. International Conference on
Woodfiber-Plastic Composites, 6th, Madison, WI, United States, May 15-16, Meeting
Date 2001:73-77.

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

Stark, N.M., and Matuana, L.M. 2004. Surface chemistry and mechanical property
changes of wood-flour/high-density-polyethylene composites after accelerated
weathering]. Appl. Polym. Sci. 94(6):2263-2273.

Thwe, M.M., and Liao, K. 2002. Effects of environmental aging on the mechanical

properties of bamboo-glass fiber reinforced polymer matrix hybrid composites. C
omposites, Part A. 33:43-52.

164

Chapter 8

CONCLUSIONS AND RECOMMENDATIONS

Copper amine complex, a main component of copper based wood preservatives, can
also be used as a coupling agent and wood surface modifier for poly(vinyl
chloride)/wood flour composite system. Mechanical properties, such as unnotched impact
strength, flexural strength and flexural toughness of PVC/wood flour composites were

significantly improved by wood copper amine treatments.

The DSC study showed slight increases in glass transition temperature (Tg) of PVC
by the addition of untreated and Cu-treated wood flour, but this effect was not significant.
Heat capacity differences (ACp) of composites before and after glass transition were
dramatically reduced by the presence of wood flour and wood Cu-treatment.

The DMA study revealed that the movement of PVC chain segments during glass
transition was limited and obstructed by the presence of rigid wood molecule chains. This
restriction effect became stronger as wood flour content increased and by using Cu-
treated wood flour. Wood flour particles acted as “physical cross-linking points” inside
the PVC matrix resulting in the absence of the rubbery plateau and higher E’, E” above
Tg, and smaller tan 8 peaks.

Enhanced mechanical performances resulted from a better wetting condition between
PVC melts and wood flour and the formation of a stronger interphase strengthened by
chemical interactions between Cu-treated wood flour and PVC matrix. Contact angle
measurements showed that PVC solution drop contact angles decreased dramatically on

Cu-treated oak surfaces compared to those on untreated surfaces. Acid-base (polar), y”,

165

 

 

electron-acceptor (acid) (7+), electron-donor (base) (7') surface energy components as

well as the total surface energies dramatically increased after wood surface Cu-treatments,
indicating strong acid-base and polar interactions. This ensured that the thermodynamic
requirement of wetting resulted in a better wetting condition between PVC and treated
wood surfaces. Measuring PVC-wood interfacial shear strength further confirmed an
improved interphase and interfacial adhesion. It is also possible that interfacial
interactions happened between copper amine complex and wood, as well as between
copper complex and PVC matrix during high temperature processing. Further
investigation needs to be conducted in order to obtain more evidences about possible
interfacial reactions.

Fungal decay weight loss of PVC/wood flour composites was dramatically decreased
by wood surface copper amine treatments. Copper amine was a very effective decay
inhibitor in PVC/wood flour composites, especially when composites had high wood
flour contents resulting in smaller mechanical property loss. Extensive cracking was
observed on the cross section of composites containing high amounts of untreated wood
flour, leading to larger mechanical property reductions after outdoor weathering, and
cracking was absent on Cu-treated wood-PVC composites.

However, using copper amine to modify wood surface gives rise to some
inconvenience, such as the color change of composites, increased treatment cost, and
recycling and disposalissues related to the leaching of copper from composites to the

environment.

166

Based on the results of experiments and analyses in this study, the following
recommendations are made:

(1) Copper amine complex is a good coupling agent for PVC/wood composites to
improve interfacial adhesion and mechanical performances as well as biological
stabilities. It is especially effective for enhancing the biological stability and
environmental durability of composite systems with high wood content (> 60 wt%).

(2) The optimum Cu concentration range is 0.2-0.5 wt% of wood flour. Very high Cu
concentrations cause decreases in mechanical properties and affect the color stability of
PVC/wood composites. An increased leaching effect also results in more copper amine
released to the environment.

(3) In composites, where treated wood particles are well encapsulated by the plastic,
copper leaching may not be an important issue, but when a large amount of wood is
added, resulting in insufficient encapsulation and weak interfacial adhesion, effect of
copper leaching should be taken into consideration. In order to evaluate the leachability
of copper from Cu-treated wood-PVC composites, new accelerated leaching methods
should be established based on AWPA E11, Standard Method of Determining the
Leachability of Wood Preservative.

(4) Moisture is a crucial factor in determining the long-term biological stability as
well as the environmental durability of wood-plastic composites. When uncovered wood
particles on the surface of wood-plastic composites are exposed to a humid environment,
they absorb moisture and build a suitable environment for the growth of fungus and mold.
Additionally, wood swells after absorbing water and moisture, shrinks when temperature

lower down below 0 °C, and swells again when temperature increase above ice point.

167

This “freeze-thaw” effect results in interphase separation and wood-plastic matrix
splitting, leading to the formation of interfacial cracks.

Recommended methods to protect exposed wood particles and isolate them from
moisture include using sealant to cover exposed wood composites, coating composite
boards with water-proof paints and applying another plastic layer to the exposed surfaces.

New developments in installation methods, connections and hinges designs, and the
selection of fastener types are also needed to avoid generating more exposed wood

particle surfaces in wood-plastic composites.

168

.0 ‘nl\ ’1‘

Out

    
   

 

IIIIIIIIIIIIIIIIIIIIII

“WWWWMWN

3 1293 02 36 4151