Compression perpendicular-to-grain behaviour of wood

choruspillowUrban and Civil

Nov 29, 2013 (3 years and 6 months ago)

322 views

National
Library
Bibliothèque nationale
du
Canada
Acquisitions
and Acquisitions
et
Bibliographie
Services services
bibliographiques
395
Wellington
Street 395,
rue
Wellington
Ottawa ON
K1A
ON4
OttawaON
K1AON4
Canada
Canada
YOU~
fi&
Vom
nlfërance
Our
file
Narre
réf6nmce
The
author
has
granted
a non-
exclusive licence
allowing
the
National Library of Canada
to
reproduce,
loan,
distribute
or
sell
copies of
this
thesis
in
microform,
paper or
electronic
formats.
The
author
retains
ownership
of the
copyright
in
this
thesis.
Neither
the
thesis
nor
substantial
extracts
fiom
it
may be
printed
or
otherwise
reproduced
without
the
author's
permission.
L'auteur
a
accordé une licence non
exclusive
permettant
a
la
Bibliothèque nationale du Canada de
reproduire, prêter, distribuer ou
vendre des copies de cette thèse sous
la
forme de
microfiche/nlm,
de
reproduction
sur
papier ou
sur
format
électronique.
L'auteur conserve la
propriété
du
droit
d'auteur
qui
protège cette thèse.
Ni
la
thèse
ni
des extraits substantiels
de celle-ci
ne
doivent
être
imprimés
ou autrement reproduits sans
son
autorisation.
ABSTRACT
The major objective of this
study
was to
provide
a
better understanding on
compression
pecpendicular-to-grain
behaviour of wood
and
to examine the
cap
ability
of
mechanicd
models for predicting this behaviour.
A
new
test procedure for testing wood
behaviour
in
compression was developed which enabled the collection of load-
deformation data
and
recordmg
of
cellular
structure images in real
time.
To understand
the relation between cellular structure
defonnation
and
load-deformation behaviour of
wood
in
transverse compression,
specimens
of four species which have
different
celiular
structure
(two
softwoods:
white
spruce
and jack
pine,
two
hardwoods: white ash and
aspen) were subjected to radial compression. Results showed
that,
in
sofhvoods
under
radial compression,
bending
of
celi
walls
occur. Bending deformation was distributed
non-unifonnly
in
a
growth
ring.
The
largest
deformation occurred in the
weakest
zone in
earlywood which led to the
first
collapse
in
this zone. Collapse
then
developed
toward
latewood.
At
the
load
1eveI
used in
this
study, latewood layer
did
not collapse.
Corresponding images recorded
in
real
tirne
and stress-strain curve showed that the
plateau
region
is
a reflection
of
cell
lumen removal in earlywood
and
the upward part
at
the end of stress-strain
curve
is related to
elastic
behaviour of latewood layer.
In
hardwoods, the
mechanisms
were different.
In
ring porous white
ash,
first
collapse
occurred in large vessels located
in
earlywood, but
in
diffuse
porous aspen,
£irst
coilapse
..
11
occurred
in
vessels
aligned
near
each other.
In
tangential compression of
softwood,
it
was
found
that the mechanism of deformation
in
tangential compression was
dominated
by
buckling
of latewood layers.
It
was concluded that
difEerent
mechanical
models should be
deveIoped
for wood
behaviour
in
tangential compression based on
the
mechanism
of
latewood
layer
buckling.
In
this study the
applicability
of some of the mechanical
models
developed
by
Gibson
and
Ashby
(1982)
were
successfùliy
examined
for
softwood
behaviour
in
radial compression.
Ceil
wall
properties
(cell
wall modulus and
yield
point) of
a
fast
grown
white spruce were
calculated
using
these
mechanical
models.
Gross
properties of
a
slow
grown
white spruce
and
jack
pine
were predicted using
calcdated
cell
wall properties
and
cell dimensions of
these
species. To predict
entire
stress-strain
in
radial
compression, a pararneter cailed
compression factor (CF)
was
introduced
in this
study.
CF
relates
the
gross
strain
t o the
number of
collapsed
ceils
and hence the location of
collapse
in
growth ring.
The
pIateau
region
in
stress-strain
cuve
was predicted using the
CF
pararneter
and last
upward
part
of stress-strain curve was predicted
using
calculated latewood modulus. Predicted
stress-
strain
curve
was
verified
by
experirnental
results.
The effects
of
temperature on
ceIl
wall
properties were found through
experiments.
Empirical
models were developed to
describe
these
effects.
The
empirical
models were
successfully
incorporated in the
mechanical
models. The extended models
can
potentially be used to generate stress-strain
curves
of
softwoods
at
any
given
temperature
within
the range studied
in
this
project
based
on
cellular structure
geometry
and
dimensions.
A
thesis
may
bear
only
the
name
of the
author,
but it
required
many
people to
bring
it to
completion.
I
wodd
like to
take
this
opportunity
to
thank
certain persons to whom
1
am
extrernely
gratefùl.
The
author
would
like
to express
his
appreciation to
the
Natural
Sciences and
Engineering Research Council and
Ministry
of
Higher
Education
of
Islamic
Republic of
Iran
for
their
support of this research.
I
would like to
extend
my deepest appreciation to my
supervisor
Dr. Y.H.
Chui,
Professor of wood engineering and
Director
of Wood Science
Technology
Centre of
UNB
for
his
vduable
supervision,
reviewing
my
thesis
and support
during
the conduct of
this
study.
My gratitude
goes
to my
advisory
committee
members: Dr.
M.H.
Schneider Professor of
wood science and technology
at
UNB,
and Dr.
P.
Cooper Associate Professor of wood
science and technology at University of Toronto for their
technical
advices
and reviewing
my
thesis.
1
would
also
like to
thank
Dr. 1.
Smith,
Professor of wood engineering
at
UN3,
Dr.
C.
Dai,
wood composite
scientist
in
F o ~ t e k
Canada
Corp
and Dr. P. Sullivan Associate
Professor
in
composites at
UNB
for
their
advices.
My
thanks
also
goes
to
G.
Daneff
and
A.
Hutton
and
K.
Hunt for reading and
editing
my
thesis.
The
experimentd
section of this
thesis
would not have been conducted without
the
assistance of laboratory technicians.
My
thanks
are to
J.
Brake, D.
McCarthy,
M.
Albright,
D.
Doherty,
J.
Phillips,
G.
Bance,
S.
Bel*
and
D.
Hall for their
help
with
specirnen
preparation, test machine operation and
electronics.
1
would like
to
recognize
the contributions of my parents, my
wife
Marzieh,
my
children
Saeid and Hossein and
other
family
members and
friends
who helped me to
maintain
my
faith
in
hurnanity
and courage me for
life.
It
is
to
these
loving
and caring people
that
1
dedicate this
thesis.
TABLE
OF
CONTENTS
Page
. .
AB
STRACT
........................................................................................................................
-11
ACKNO WLEDGEMENTS
...........................................................................................
iv
.................................................................................................................
LIST
OF
TABLES
x
. .
............................................................................................................
LIST OF
FIGURES
XII
C WT E R
1
.
0
INTRODUCTION
.......................................................................................
1
..........................................................................
1.1
TECHNICAL
BACKGROUND
1
1.2 OBJECTIVES
......................................................................................................
8
..........................................................................
...........
1.3
THESIS
PREVIEW
....
8
CI-LAPTER
2.0
LITERATURE
REVIEW
...................................................................
1 1
...........................................
2.1 WOOD STRUCTURE
AT
MOLECULAR LEVEL 11
.....................
2.2
WOOD STRUCTURE AT ULTRA LEVEL
...,
.........................
13
2.3
WOOD STRUCTURE AT CELLULAR LEVEL
.............................................
14
...
...............................
2.4 WOOD STRUCTURE AT MACROSCOPIC
LEVEL
..,..
16
2.5 WOOD
BEHAVIOUR
DURING
HOT-PRESSING
.........................................
-19
2.6
WOOD
BEHAVIOUR
DURING
TRANSVERSE COMPRESSION
................
23
2.7
EFFECT
OF
TEMPERATURE
AND
MC ON TRANSVERSE
COMPRESSION WOOD
BEHAVIOUR
.........................................................
36
2.8
MODELLING
TRANSVERSE
COMPRESSION BEHAVIOUR
OF WOOD
......................................................................................................
43
.......
...........
2.8.1
Modelling
composite-mat
behaviour
dui ng
hot-pressing
....
44
2.8.2
Modelling
solid
wood
behaviour
during
transverse compression
........
5 2
CHAPTER
3.0 INFLUENCE
OF WOOD CELLULAR STRUCTURE
ON
ITS
BEHAVIOUR
IN
TRANSVERSE COMPRESSION
...................................
60
............................................
.........
3.1
APPARATUS
DEVELOPMENT
...
..........
61
.........................................
3.1.1
Compressing
device
with
a
light
microscope
61
3.1.2
Optical
system
......................................................................................
63
3.1.3
Load
measurement
..........
..
..............................................................
-64
...........
.................................................................
3.2
TEST PROCEDURE
..,.
.....,
66
3
.2.1
.
Radial compression tests
..............................
...
..............
-66
.................
.
3.2.1.1 Radial compression test at
magnification
of
12X
..
67
..........
.
3.2.1.2 Radial compression test at
magnification
of
3 2X
....
...
..
68
3.2.1.3
Radial
compression test at magnification of 160X
..................
69
...................................................
.................
3.2.1.4
Data
analysis
..
70
3.2.2
Tangentid
compression tests
....................................................
70
............................................................................
3.3
SPECIMEN
PREPARATION
71
3.3.1 Species...
...........................................................................................
7
1
3.3
-2
Cutting
of wood
blocks
.......................................................................
-72
. .
.......
.........................*...............................................
3.3.3
Conditionmg
,.
..73
3.3.4
Preparation of specimens for compression tests
.......................titi....ti.....
-74
3.4
RESULTS
AND DISCUSSION
.........................................................................
75
3.4.1
Stress-strain
relationship
of different species in radial compression
......
-75
..................................................
......
3
.4.2
Microscopie
observation
......
-78
.............
.............
3.4.2.1 White
spnice
.,,.
,...
78
...........................................
3.4.2.2 Jack
pine
..................................-a
82
3.4.2.3
Aspen
...................................................................................
86
..............................................................................
3.4.2.4 White
ash
-89
3.4.3 Wood
behaviour
in
tangential
compression
........
..
........................
.....
91
.....................................................
3.4.3.1 White spruce
-92
...............................................................................
3
.4.3.2
Jack
pine
-96
3.5
SUMMARY
AND
CONCLUSIONS
..................................................................
99
CHAPTER
4.0
APPLICATION
OF
CELLULAR
THEORY
IN
PREDICTING
STRESS-STRAIN
CURVE
OF
WOOD
IN
RADIAL
COMPRESSION
.................
102
......................................
4.1
STRESS-STRAIN
AND CELLULAR
STRUCTURE
1 0 3
.*..............
..........*...................................
4-2
LLNEAR
ELASTIC
BEHAVIOUR
...
105
..................................................................................
4.3
PLASTIC
BEHAVIOUR
1
09
4.4 RATIONALES
.................................................................................................
1 12
4.5
PREDICTNG
WOOD
BEHAWOUR
IN
TRANSVERSE
.......................................................
COMPRESSION
........................
....
1
14
4.5.1
Predicting
sofbvood
behaviour
in
radial compression
..........................
115
4.5.2
Determinhg
ce11
wdl
properties
of
white
spruce
.................................
116
.......................................................................................
4.5.3
Verification
-124
4.5.3.1
Predicting
and
experimental
results
of a
slow
grown
white
spruce
properties
in
radial compression
.......................
124
4.5
-3.2
Predicting and experimental results
of
jack
pine
..........................................
properties
in
radial compression 128
vii
4.6
PREDICTING
THE
TOTAL
STRESS-STRAIN
C UR E
IN
RADIAL
COMPRESSION
........
,.
.........................................................................
132
................................................................
4.7
SUMMARY
AND CONCLUSIONS
141
CHAPTER
5.0 INFLUENCE
OF
TEMPERATURE
ON SOFTWOOD
CELL
WALL
AND
ITS
GROSS PROPERTIES
UNDER
RADIAL
COMPRESSION
................
144
...........................................................
5.1
DEVELOPMENT OF
APPARATUS
1 4 5
.............................................................
........
5.2
SPECIMEN
PREPARATION
..,
149
...........................................................................
5.3
EXPERJMENTAL
DESIGN
- 1
50
.................................................................................
5.4
TEST
PROCEDURE
....150
5.5
DATA
ANALYSIS
..........................................................................................
151
.............................................
5.6
RESULTS
AND
DISCUSSION
151
....................................................................
5.6.1
Compression test
results
1
52
5.6.2
Det edat i on
of
ce11
wall
properties
at
different
temperatures
...........
.
158
5.6.3
Predicting
effect
of
temperature
on
ce11
wall
properties
.........
...
......
162
5.6.3.1
Predicting
the effect
of
temperature on
ce11
wall
modulus
.....
163
5.6.3.2
Predicting the effect of
temperature
on
ce11
wall
yield
stress
..
165
5.7
INCORPORATION OF
EFFECT
OF
TEMPERATURE
IN
THE
MECHANICAL
MODELS
.........................................................................
166
5
.7.1
Elastic
behaviour
...............................................................................
1
67
................................................................................
5.7.2
Plastic
behaviour
167
5.8 APPLICATION
OF
MODIFTED
MECHANICAL
MODELS
...........................
168
5.9
SUMn/LARY
AND
CONCLUSIONS
.................................
....
.............
170
C WT E R
6.0
S-Y
AND
RECOMMENDATIONS
..........................
..
.........
1 7 2
SUMMAR.Y
AND
CONCLUSIONS
..............................................................
1 7 2
RECOMMENDATIONS
..............................................................................
178
REFERENCES
...............................................................................................................
1 8 0
APPENDIX
A
................................................................................................................
190
APPENDIX
B
..................................................................................................................
212
APPENDIX
C
.....................
....
.................................................................................
219
LIST
OF
TABLES
Page
Table 2.1 Modulus of
elasticity
and
hardness
of
ce11
wall
......................................................
33
............................
Table 3.1
Experimental
design
for
compression tests at room temperature
72
Table
3.2
Results of radial compression tests on four wood species
......................................
76
Table
4
.
L
Average values of
ce11
dimensions
in
a
farst
grown
white
spruce
..
.........................
119
....
Table 4.2
Results
of radial compression tests on
specimens
of
a
fast
grown
white
spruce
121
Table
4.3
Experimental
gross
modulus
and
calcdated
celi
wall
modulus
of
white spruce
in
radial compression
.......................................
-...
......................................................
-122
Table
4.4
Average
ce11
dimensions of the
fIrst
collapse
region
in
a
fast
grown
.......................
white
spruce
...............
...
...-
.........................................................
122
Table
4.5
Experimental gross yield stress and
calcdated
ce11
wall yield stress of white
spruce
in
radial compression
........
....
....
.......................................................................
123
........................................
Table
4.6
Average
ce11
dimensions of
a
slow
gram
white
spruce
125
Table
4.7
Average
ce11
dimensions of
the
first
ce11
collapse
region
in
a
slow grown
white
spruce
............................................................................................................
127
Table
4.8
Predicted and experimental
Es
and
o.
0.f
a
slow
grown
white
spruce
..............
128
Table 4.9 Average
ce11
dimensions
of
jack
pine
..................................................................
-129
Table
4.10
Average
ce11
dimensions
of
the
first
collapse
region
in
jack
pine
.........................
130
........
Table 4.1
1
Predicted and
experimental
Es
and
O
,
of
jack
pine
in
radial compression
132
Table 4.12 Compression factor
(CF)
at
three
positians
on stress-strain
curve
................
..,...
136
Table
4.13
Stress-strain
data at
particular
positions
in
plateau
region
..................................
138
........
Table
4.14
Average
ce11
dimensions in latewood
region
of a slow
grown
white
spruce
138
............................................................
Table 5.1
Specimen
details
and
experirnental
design
140
.........................................................
Table
5.2
Results
of
radhl
compression tests at
30
OC
153
.................................
...................
Table
5.3
Resdts
of
radial
compression tests at
80
OC
....
155
.......................................................
Table 5.4
Resuits
of radial compression tests at 1 10
OC
156
......................................................
Table 5.5 Results of radial compression tests at
140
OC
158
Table 5.6 Average
cell
dimensions of a
fast
grown
white
spmce
.....
......................
.......
159
Table 5.7
Experimental
gr as
Ep
and
o
.
and
calculated
ceil
wall
Ew
and
a
.
....*..*..
..*.............*...............................................
at different
temperatures
..,,.....
160
...........
Table 5.8
Sumrnary
of calculated
modulus
and
yield
stress at different temperatures
162
.......................
Table
5.9
Predicted
gross
modulus and
yield
stress of
jack
pine
at
6%
MC
169
LIST
OF
FIGURES
Page
.......................................................
Figure 1.1
Growth
of
structural panels
in
North
Arnerica
4
Figure
1
-2
S
chematic
stress-strain
curve
under
transverse compression
...................
..
...........
5
Figure
2.1
Schematic
drawing
showing
ce11
wall
components
in
a
microfibril
.........
..
...........
12
...............
.........................................
Figure
2.2
Schernatic
drawing
of
wood ultra
structure
..
13
Figure
2.3
Stress-strain
curves of balsa compressed
in
three
directions
..............................................................................
..
....
(Easterling
et
al.
1982)
..,.
..
28
....................
Figure
2.4
Stress-strain
curves
of
balsa compressed
in
tangential
compression
..
28
Figure
2.5
Stress-strain
curves
of
balsa compressed
in
radial
compression
.........
...
....
......
29
Figure
2.6
Micrographs
corresponding
to different
regions
of radial compression
...............................................................................
stress-strain
(Stefansson.
1995)
54
Figure
3.1
Compression
device
and
stage for
light
microscope
.............................................
62
Figure
3.2
Loading
device
for use
in
SEM
...........................................
..
............................
63
.................................................
Figure
3.3
Schematic
drawing
and
picture
of
optical
system
65
Figure
3.4
Compression test
under
microscope at magnification of
12X
......................
..
.....
68
Figure 3.5 Compression test
under
microscope at
magnification
of
32X
...............................
68
..............................
Figure
3.6
Compression test
under
microscope
at
magnification
of
1
60X
69
..........................................................................................
Figure
3
-7
Specimen
preparation
73
...........................................................................
Figure
3.8
Specimen
with
reference
lines
-74
............................................................
Figure
3
-9
Stress-strain
relationships
in four species
-77
..................................................
Figure 3.10
Elastic
deformation of
ce11
wall
in
white
spruce
79
Figure 3.11
First
cell
collapse
in
white
spruce
.......................................................................
80
Figure 3.12 Collapse of earlywood in white
spruce
..................
...
...................................
81
Figure 3.13
Ce11
collapse at end of
loading
in
white
spruce
...................................................
82
Figure 3.14
EIastic
deformation
in
jack
pine
.........................................................................
83
Figure 3.15
First
ce11
collapse
in
jack
pine
............................................................................
83
Figure 3.16 Development of
ce11
collapse
in
earlywood
ofjack
pine
......................................
84
Figure 3.17 Collapse
of
complete
earlywood
in
jack
pine
.....................................................
85
Figure 3.18
Elastic
defomation
in a
vesse1
element
of
aspen
.................................................
86
Figure 3.19 Initiation of first
ce11
colIapse
in
aspen
...............................................................
87
................................................................
Figure 3.20 Development of
ceU
collapse
in
aspen
88
.................................................................
Figure
3
-2
1
Removing
of
vesse1
cavities
in
aspen
-88
....................................................
Figure 3.22 First
ce11
collapse of large vessels
in
white
ash
89
..............................................................
Figure 3.23
Removal
of
vesse1
cavities
in
white
ash
90
Figure 3.24 Densification of large vessels and
elastic
deformation of fibres in white ash
........
91
Figure
3
-25
Stress-strain
curves
for white spruce
in
radial and
tangentid
compression
..........
93
.............
.............................
Figure 3.26
Tangentid
wall
bending
in
tangentid
compression
..
94
Figure
3
-27
Buckling
of latewood layer in white
spruce
during
tangential
compression
.........
94
Figure 3.28
Collapse
in
earlywood
layers
near
buckled
latewood in white
spruce
..................
95
Figure 3.29 Development of latewood layer
bending
in white spruce
....................................
96
.................
Figure
3.30
Stress-strain
cuves
ofjack
pine
in
radial
and
tangentid
compression
97
Figure 3.3
1
Beginning
of deformation
in
resin
canals
............................................................
98
...............................................................
Figure 3.32
Bending
of latewood layer
in
jack
pine
98
...
Xl l l
.............................................................
Figure
3
-33
Cracks
near
resin
canals
in
jack
pine
-99
Figure 4.1 Schematic
drawing
of cellular structure
....
.............~...~~.~..~~~~.....................~......~..
105
Figure 4.2 Compression
behaviour
of
a
cellular
material
in
X
direction
..............................
106
Figure 4.3
Elastic
behaviour
of cellular
material
under
compression
in
X direction
...........
...
107
Figure 4.4 Projection of deflection
(6
)
in
X direction
..........................
............
...................
108
Figure
4.5
Stress distribution in cross-section
LU
of
ce11
wall at
different
...............................................................................................
stages of bending
1 1 0
Figure 4.6 Intra-ring variation of wall thickness and radial length
in
a fast grown
white spruce
...........................
....
.................................................................
1 9
Figure
4.7
S
tress-strain
relationships
obtained
fkom
radial compression
on
specimens
with
one
growth
ring
of a
fast
grown
white
spruce
.................
..
.........
120
Figure 4.8 Intra-ring variation of wall thickness
and
radial
wall
length of a slow
..............................................................................................
grown
white spruce 125
Figure 4.9
Intra-ring
variation of wall thickness and radial
wall
length
of
jack
pine
...................................................................................................................
130
...........................
Figure 4.10 Schematic
drawing
of a
growth
ring before and
after
loading
134
............
Figure 4.1 1 Schematic
drawing
of positions of
strain
analysis
on
stress-strain
curve
135
Figure
4-12
Predicted
entire
stress-strain
curve
in
radial compression for
a
slow
grown
white
spruce
..........................................................................................
139
Figure 4.13 Predicted
and
experïmental
stress-strain
cuves
for a slow
grown
..........................................................................
white spruce
....................
...
-140
........
Figure
5.1
Test set
up
for compression tests
at
elevated
temperatures
....
..........
1 4 6
......................................................
Figure
5.2
A
schematic
drawing
of miniature
hot-press
147
...........................................................................
Figure
5
-3
Heating
systern
................
....
148
Figure
5.4
Stress-straùi
relationships of white spmce
specimens
compressed at
6%
MC
and
30
OC
.......................................................................
152
xiv
Figure
5.5
Stress-strain
relationships of
white
spruce specimens
....................................................................
compressed
at
6%
MC
and 80
OC
154
Figure 5.6
Stress-strain
relationships of
white
spmce
specimens
compressed
at
6%
MC
and
1
10
OC
....................................................................
156
Figure
5
-7
Stress-strain
relationships
of
white spruce
specimens
......................................................................
compressed
at
6%
MC
and
140
OC
157
..........
Figure
5.8
EfTect
of temperature on
ce11
wall
modulus
and
its
regression
fitted
cwve
164
......
Figure 5.9
Effect
of temperature on
ce11
wall
yield
stress and its regression
fitted
curve
166
Figure
5.1 0
Experimentai
and
predicted
stress-strain relationships of
jack
pine
.........
......................
compressed
at
6% MC and
30
OC
and
140
OC
,.,,
.............
,..
170
CHAPTER
1
INTRODUCTION
1.1
TECHNICAL
BACKGROUND
Wood properties
exhibit
wide
variability
due to its
natural
origin.
These
variations
are
in
part
the
result
of
the
growth
conditions
of
wood
brought
about
by
environmental
factors
such
as
clhate,
soil,
water
supply,
and
available
nutrients
(Keith
1961,
Panshin
and
de Zeeuw 1980,
Bodig
aud
Jayne
1982, Dodd 1984, Savidge 1985,
Zhang
and
Chui
1996,
Dutilleul
et
al.
1998
).
Wood
is
assumed
to have
three
mutually
perpendicular
axes
of
symmetry
with respect to its properties: longitudinal, radial, and
tangentiai
with
reference to the
cyiïndrical
bole of a
tree.
Wood
exhibits
different
physical
and
mechanical
behaviours
in
these
three
directions
(Schniewind
1959).
In
addition to
natural
variability
and
anisotropic
properties, wood
is
also
a porous
material,
in
which
the pore
structure
takes
different
forms
according
to the species.
In
hickory,
oak
and ash,
the
pores
are
of
such
magnitude
that they
are
observable to the
naked
eyes,
while
in
other
species,
such
as beech
and
maple,
the pore structure
is
somewhat Iess observable,
Wood is also a heterogenous
material.
Heterogeneiv
appears
at
different
levels.
At
the
rnacro
level,
bot s
and
grain
orientation cause wood to behave
non-uniforrniy,
while
at
the micro
level,
earlywood
and latewood zones,
cryçtalline
and amorphous
sections of
the
ce11
wall,
orientation of microfibrils,
and
dierent
percentage of
chernical
constituents
in
the
ceU
waLi
layers are
ali
sources of heterogeneity.
The aforementioned sources of
heterogeneity
represent
numerous
challenges
to
wood processing
and
utilization.
In
the
put,
some of
these
challenges
could be
conquered by
selective
utilization
of certain
species
and
reliance
on
the
larger
and older
classes
of
trees possessing greater
uniformiq.
It is
now
clear,
however, that socieîy
is
no
longer able to
enjoy
such
luxunes.
Trees are
harvested
at
younger age.
As
a
result,
trees
are
increasingly
characterized
by
their
s md
sizes
and
greater
variability.
Whereas
in
the
past,
the
re-manufacture of wood
has
been
limited
in
scope,
present
utihation
requires
extensive
re-manufacture
to
provide
the types of products needed for
a
modem economy
(Bodig
and
Jayne
1982).
To
combat the
Iack
of
quality
wood
suppiy,
various
silvicultural
practices
such
as
wide-space
planting,
heavy
thinning
and
fertilization
of
stands
have been developed for
the
management
of forests over the last
few
decades
(Zobel
1980).
In
response to
these
changes, trees
are
growing
faster
because
of the
increased
vitality
of their
crown
associated
with
a reduced
cornpetition
for
soil
moisture
and
nutnents
(Bonder
1984).
Since
quaIity
of wood
is
a
resdt
of
a
long process where genetic dispositions of the tree
interact with
environmental
requirements,
manipulation
of
the tree
growth
rate
af3ects
not
only
the
rate of the formation of wood,
but
also
its
technological
and
anatomical
properties
(Bendtson
and
Se&
1986,
Keith
and Chauret 1988).
Ring
width,
wood
densiq,
and fibre
length
are
three
variables
classicaiiy
used to select wood for
solid
wood
and paper
produas
(Dinwoodie
1965,
Zhang
1995,
Dutilleul
et
al.
1998).
Confliaiog
results,
however, were reported and controversy
exists
in
the literature
regarding
the
relationships
among
the
growth
rate, wood
density,
and fibre
length
(Zhang
1995).
In
recent
years,
wood
is
used more as a
high
potential
renewable raw
material
for
secondary
rnanufacturing
produds
such
as
engineered wood products or wood
composites. The engineered wood
product
industry
is
presently
growing
rapidly
Wo n e y
1989,
Guss 1994,
SBA
1997).
Some
popular
engineered wood products,
such
as
Oriented Strand Board (OSB) were
invented
only
20
years ago.
As
shown
in
Table
1.1,
it
is
expected
that
noml
993
to
2003,
the volume of the
production
of
these
products
will
double
in
North Amenca
while
the
volume
of
other
cornpetitive
products
like
plywood decreases
(Guss
1994).
The
reasons
for this rapid
growth
of
OSB
are:
timber
supplies
from
public forests
have been severely
constrained,
the
availability
of
larger
sizes
of structurai wood
&aming
members
has decreased
sharply
and
prices
for conventional lumber and
plywood
have
increased
significantly.
These
issues have made the building
matenals
industry
more dependent upon
reliable
engineering,
which,
in
tuni,
requires
materials
whose
strength
properties are more
accurately
known
within
narrower
limits.
Engineered wood products fit
this
requirement
much
better
than
conventional wood products.
Additionaliy,
the strength values of
commercial
species
have been re-evaiuated,
leading
to a
general
lowering
of
such
properties
fiom
rnany
species and
sizes.
Finally,
residential buildings have become
larger
and
more
complex,
requiring
structurai
members
with
greater
strength
than
dimension
Iumber
can
often
provide
(Guss 1994).
Table 1.1
Growth
of structural panels in
North
America
fiom
1993 to
2003
(Guss
1994)
Wood composite products are
generally
made by a hot-pressing process. Most
Year
Plywood
OSB
composite product properties are
controued
through
a
complex
phenornenon
created by
interaction
effects
of
pressing temperature,
moisture
content of
glued
wood
elements
(mat), type of
binder,
and wood transverse compression stress-strain properties.
Concerning
the last parameter,
in
the
manufacturing
of wood composites, wood
material
1993
21
-2
9.9
as
a
major component
(almost
90 percent of products is
usually
subjected
to
2003
16.7
17.8
compression stress
perpendicular-to-
grain
beyond its
elastic
capacity
at
elevated
temperatures
(Suchesland 1967,
Wolcott
1990,
Steiner
and Dai
1993,
Tabarsa
and Chui
1997).
Conventionai understanding of the
stress-strain
relationship
of
wood
under
transverse compression identifies
three
distinct
regions
(Figure
1
-2).
At
the
b
eginning,
wood
exhibits
elastic
behaviour where load
and
deformation are
hearly
related.
At
the
end
of
the
elastic
behaviour,
ceîl
coilapse
and
inelastic behaviour commences. In
the
inelastic
region
strain
increases rapidly
with
little
or no change in stress. This
region
is
often
called
the plateau.
If
compressive
loading
continues,
dl
cells
wiU
collapse
and
ce11
4
cavities
removed,
causing
the wood to
act
as
a
solid
body which causes stress to
increase
drastically.
This
region
is
called
densification
(Ivanov
1953,
Bodig
1965%
Kennedy
1968)
Figure
1.1
Schematic
stress-strain
cuve
under
transverse
compression.
This is
a
simplification of the
actual
behaviour and
does
not
give
detailed
information
of
wood behaviour in
its
inelastic
region
which
is
affected
by the wood
micro-structure.
This
region
is
very
important
fkom
a
materid
science perspective and
for
the
purpose
of
modelling
of
wood behaviour in transverse compression.
Although,
there
is generaiiy
a
recognized
relationship between
mechanical
properties
and
the
specific
gravity of wood,
this
relationship
does
not hold
for
ali
elastic
properties.
While
the
longitudinal modulus of elasticity varies
with
the
5/4
the power of specific
gravity,
no
relationships
cm
be
found
between
specinc
gravity
and
radial modulus of elasticity
(Bazhenov
el
al.
1953,
Bodig
l96Sa).
Specinc
gravity
cannot
provide
the
desired
information
of wood beyond the
elastic
region
for
modeiiing
purposes
(Tabarsa
and
Chui
1998).
Existing
information or wood
behaviour
during
transverse compression
has
been
obtained
by
applying
test procedures
suggested
by
standards
such
as
America
Society
for
Testing
and
Materials
(ASTM)
D
143 (1997)
which
is
considered inadequate
(Peficane
et
al.
1994).
In
the
ASTM
method,
wood specimen
is
assumed
to behave as
a
homogenous
materiai
where the load and deformation are
uniformly
distributed
over the
specimen
in
the
loading
direction.
Compressive
strength
and
modulus
of
elasticity
are
detennined
during
compression of a wood specimen
up
to
2
percent of its total
compressive
strain.
Transverse compression behaviour
of
wood
is
dependent on
its
anatomical
features
(Bodig
1965%
Tabarsa
and Chui
1
W8),
therefore some species
may
never
reach a
yield
point at
this
level of compression.
The standard
test methods suggest
the
use
of
a
deformation gauge attached to the machine cross
head
or over a gauge
length
for
measuring
deformation.
Since
providing
a
smooth wood d a c e
is
difncdt,
part
of
the
rneasured
deformation is
actually
the
gap between
specimen
and
cross
head
if
cross
head movement
is
measured.
In
addition,
these
rnethods
provide
an
average deformation
data over the
individual
annual
rings. Therefore,
these
methods
usually
lead
to erroneous
results.
As
previously
meationed,
wood
is
a heterogenous and
anisotropic
mat
end.
During
compressive
loading
its behaviour
is
not
only
different
in
its
three
fundamental
directions but aiso
is
different
in
various
segments of
annual
rings
(intra-ring) and
among
individual
rings
(inter-ring)
in the radiai direction.
In
axial
compressive loading,
wood
cells
generaily
fail
dong
a
slip plane.
During
transverse compressive loading, wood is
loaded
either
tangentidy
or
radiaüy.
In
the
tangentid
direction, the load
is
mostly
supported
by
the
stifflatewood
layers.
Conversely,
in
pure
radial compressive loading,
the
extemal
load
is
shared
uniformly
by
ali
individual
cells
located
in
each
annuai
ring
and
most
wood
species
behave
as
an
elasto-plastic
materiai
(Tabarsa
and
Chui
1999).
The
above
perspectives
can
be
conciuded
nom
the
general
knowledge
of
wood
anatomy.
Few researchers
(Ivanov
1953,
Schniewind
1959,
Bodig
1965%
Kunesh
1968,
Keith
1971,
Gibson
and
Ashby
2982,
Dinwoodie
1989,
Stefesson 1995,
Persson
1997)
have tried to apply
these
basic concepts
to
explain
wood behaviour
during
compressive
loading.
Although
research
has
provided
valuable
info&ation
in
the
area
of
compressive
loa-
little
is
hown
cooceming
physical
mechanisms
involved
specincaily
in
the
inelastic
region.
This
study
was
conducted
to
provide
this understanding.
1.2
OBJECTIVES
The
major objectives of
this
study
are:
1.
To
develop a
new
experimental
method for the simultaneous observation of
cellular deformation and recording of transverse compression stress-strain
behaviour.
2.
To
provide
an
advanced
understanding
of
the influence of
anatomical
features
of wood on the stress-strain relationship
in
transverse compression.
3.
To
estimate
cell
wall
properties
(modulus
and
yield
stress)
in
compression
using
experhental
data and micro-mechanical models,
and
the effects of
temperature
on
these
properties.
4.
To
develop
a
method of
predicting
wood
behaviour
during
radiai compression
based
on
ceii
waii
properties and
ceil
waii
dimensions.
An
o v e ~ e w
of
the
chapters
in
this
thesis
is
given
below:
Chapter
1
:
In
this chapter the problem is explained, objectives
outlined
and,
all
chapters are
introduced-
Chapter
2:
Previous
studies are
reviewed
in
this
chqter
under
different
sections:
wood structure,
growth
effect on wood structure and wood density,
wood
composite process, wood behaviour
in
transverse compression,
effect
of
temperature on wood behaviour in transverse
compression,
modelling of wood
behaviour
in
transverse
compression
Chapter
3
:
Apparatus and test procedure developed for this
project
are
described.
Species
and
specirnen
preparation
are
then
explained.
Results
are
discussed
under
two
categories.
First,
dserences
between stress-strain response
of
dserent
species used in
this
study
are discussed,
then
the
rnechanisms
of
cellular
deformation
of
these
species
under
radiai
and
tangentid
compression are
discussed. At the end, the conclusions
drawn
fkom
test
results
are sumrnarised.
Chapter
4:
Some mechanical models developed
by
Gibson
and
Ashby
(1982)
which
are
suitable
for adoption
in
this study
are
discussed.
Then
ce11
wall
properties of
a
fast-
grown
white
spmce
are
determined by
applying
these
models.
The
validity
of
the
determined celi
wall
properties is
examined
with
a
slow-grown
white
spruce and jack
pine.
Methods for
predicting
the plateau and
densification
regions
-
of
stress-
strain
c w e
are
introduced.
Total
stress-strain
c w e of white spruce
is
predicted and compared
with
experirnentai
stress-strain
curve.
Conclusions
drawn
firom
this
chapter are
outlined
at
the
end.
Chapter
5: An
apparatus
developed for
cornpressing
specimens
at
high
temperatures
is
explained.
Spcckcn
preparation,
experimental
design and test procedure are
then
described.
Results
of compression tests at
different
temperatures are
discussed
and
celi
wd
properties
detemiuied
uskg
mechanical
models, described
-
in
Chapter 4.
Empirical
models
are
developed
to represent the effect
of
temperature
on
ceil
wall
properties.
These
empïrical
models
are
incorporated
within
the
mechanical
models for
predicàog
wood behaviour
under
radial
compression at
any
temperature
in
the range used
in
these
experiments.
Modified
mechanical
models
are
applied
on
jack
pine
to
predict
its
gross
properties at
different
t ernperatures. Total stress-strain
curves
are
predicted for
two
extreme
temperature
levels
and
compared
with
experimental
resuits.
Chapter
6:
The
thesis
is
summarized
and
conclusions
outlined.
Some
cornrnents
and
recornmendations for
fùture
work
are
given.
Since
a major
part
of
this
study
deds
with
the
correlation
between
the
anatomical
structure
and
its
transverse compression behaviour, it
is
important to gain
an
understanding
of
the
basics
of the
wood
structure
prior
to
reviewing
related
studies.
2.1
WOOD
STRUCTURE
AT MOLECULAR
LEVEL
A woody
ce11
regardless of its type
and
function
may
be considered as
a
basic unit
of wood
matenal
which
is
composed
of
millions of individual
cells.
A
woody
ce11
has
a
hollow (lumen) enveloped by
walls
and
appears
in hexagonal,
rectangular
and
sometimes
round shapes.
It
has been
conventionaliy
accepted that regardless
of
the wood species
and
types
of
woody
cells,
the
ceU
wail
is
basically
made of
three
main
natural
polyrners:
cellulose, hemicellulo se
and
Ligni~.
These
polymers appear as a fibre-composite
which
is
cailed
microfibrils
in
celi
wail.
Microfibrils
are
essentially
a
core
of
crystalline
cellulose
encased
in
a
shell of hemicellulose
and
lignin
(Figure
2.1).
Although
the chemical
structures
of
cellulose
are
assumed
to
be the
same
in
aii
wood
species, the
chemical
structures of hemicellulose
and
lignin
are
different
among
species.
On
the other
hand,
the
also
among
trees
of
a
M y,
diierent
parts
of
a single
tree,
difFerent
segments of
an
annual
Nig,
difrent
kinds
of
woody
cens,
and
even
in
merent
hyas
of
a single
tell
w d
(Sjosuom
1981,
RoweU
1983).
These
three
polymers
dictate
most
of
the
physical
and
mechanical
behaviours
of
wood.
--
I
7-
-dn-
0-
I
9!
7-
CIL-
F-
mon
I
-1
A
or
1 1
.rp--'.rc
0s
Figure.
2.1
Schematic
drawing
showing
celi
w
l
wall
chernical
elements
in
a
microfibrfi.
The
other components
like
extractives
and
ash
appear
in
wood
in
low
quantities
and have
an
insignificant
effect
on
mechanical
properties
of wood.
2.2
WOOD
STRUCTURE
AT
ULTRA
LEVEL
The
cell
wail
is composed of
difFerent
layers.
On
the
outside,
the space between
neighbouring
cells
is
called
the
middle
lamella
which
is
made
up of
mainly
lignin
and
acts
as
a
binder
between
cells.
The
Grst
layer
of the
cell
wall
is
called
the
primary
wall and the
second one
is
the secondary
wall
(Figure
2.2).
The
primary
wall
is
composed
of
a
loosely
packed
and
randomly
formed
network
ofmicrofibrils
and
it
has
a
high
lignin
and
Figure.
2.2
Schematic
drawing
of
wood ultra structure
adopted
&om
Dinwoodie
(1989).
low cellulose content. The thickest
layer
of
the
cell
wall is
the
secondary
wall
which
consists
of
diree
layers:
S 1,
S2,
and
S3.
The
microfibriUar
angle
(MFA)
varies
among
these
layers. The
MFA
in
the
S
1
and
S3
layers lies
between
50'
to
70".
Since
S2 layer
accounts
for
approxhately
70%
of
the
thickness
of
the
cell
wall,
the
orientation
of
the
cellulose
MFA
in
this
layer
has the
strongest
inauence
on
mechanical
properties
of wood
(Sahiberg
1997).
The
mean
MFA
in
the
S2
layer
of
the
secondary
waii
has
been used
as
a
key
parameter
in
understanding
the
mechanical
behaviour
of
whole
wood
(Meyland
and
Probine
1969).
MFA
of
S2
layers
of pi ms
radiata
was
measured
using
X-ray diffraction
by
Cave
(1997)
and
reported to be
20".
MFA
varies
not
only
among
cell
wall
layers
but
also in the
annual
ring
dong
the
radial
direction.
In
one
study
conducted
by
Pyrkjeeide
(1990)
on
Norway
spruce,
it was reported that
the
MFA
of
earlywood
was
40-60"
and
that
of
latewood was
5
"-20
O.
2.3
WOOD
STRUCTURE
AT
CELLULAR LEVEL
During
post
cambial
enlargement,
active
cell
division
ceases
and immature
cells
dSerentiate
into
the
various
kinds
of
mature
cells
that
are
characteristics
of
the
species
(Panshin
and
de
Zeeuw
1980).
Longitudinal tracheids
and
ray
parenchyma
are
two
types
of
cell
which are found
in
al l
coniferous
wood. Longitudinal tracheids which
constitute
over
90
percent of the volume
of
soffwoods
are
relatively
long
(3
-0-5
.O
mm).
The
earlywood
longitudinal
tracheids
are
hexagonal
in
cross section,
while
the latewood
tracheids
are
more
likely
to be
rectangular
in
shape.
The
change
in
the
cross-sectional
shape of tracheids
is
associated
wiîh
seasonal
growth.
The
largest
radial
diameter,
with a
minimal
wall
thickness for
the
species,
occurs
a short distance
fiom
the
beginning
of
the
growth
incrernent
(shortly
d e r
growth
starts
in
the
s p ~ g ).
Later
in
the growth
season
the
radial
tracheid
diameter
decreases and the
narrowest
tracheids are
formed
at
the
outer
margin
of
the growth increment
just
before the
end
of
the
growth
season.
The
tangential
diameter
of tracheids
remains
quite
constant
within
a growth
ring
and
does
not
Vary
much
by
the
age
of the tree. The
walls
of
the
longitudinal tracheids
are
commonly
marked
with
pits
(Desch
and
Dinwoodie
1981).
These
pits across the radial
face of
an
earlywood
tracheid
are
always
numerous
and
maximum
in
diameter
while
pits
on
the
radial face
of
the latewood
are
less numerous and
smaller
in
diameter.
The
bordered pits on
the
tangential face of longitudinal tracheids are always
smaller
than
those
on
the
radial walls.
Parenchyma
are
thin
walled,
short
cells
which
have
simple
pits.
Some parenchyma
celis
run
perpendicdar
to
the
longitudinal direction
wbich
are
cailed
ray parenchyma.
These
elements
transfer
the
sap
£iom
outer
part
(sapwood) to
inoer
part
of
a
tree.
The
pits
in
a
cross field between
the
ray parenchyma and longitudinal tracheids
Vary
between
species and
are
used for
softwood
identification.
In
hardwoods, fibres, vessels, and
parenchyma appear
in
all
species. Fibres
are
long
and
narrow
cells
with
closed ends
which
provide
mechanicd
support
and vessels
are
long
tubelike
structure which
are
conductive
elements
(Panshin
and
de
Zeeuw
1980).
Among
these,
tracheids
in
soffwoods
and fibres
in
hardwoods
dominate
mechanical
properties
of
wood.
Zn
species with broad
rays,
the
mechanical
properties
are
aiso
affected
by the rays
(Youngs
1957,
Schniewind
1959, Kennedy
1968).
2.4
WOOD STRUCTURE
AT
MACROSCOPIC
LEVEL
At
this
level, wood structure
is
discussed
at
the
scale
of
the
annual
growth
rings.
Trees
grown
in
a
region
with
distinct
seasonal
climates,
in
which
a
growing
penod
alternates
with
a resting state, the
growth
rings
appear
in
cross section. The width
of
the
growth
ring,
or the number of rings
in
one
centimetre
in
the radial
direction,
is a measure
of the
growth
rate
(Desch
and
Dinwoodie
198
1).
In
softwoods
and
Mg
porous hardwoods, variation
in
ring
width is associated
with
the variations in
the
proportion of latewood
and
earlywood.
h
d i s e
porous
hardwoods, identical pores
are
distiibuted
unifonnly
ail
over the cross section.
Earlywood
and
latewood
regions
are
not
as
distinct
as
sofhvood
or ring porous
hardwoods.
In
aU
species,
extremely narrow and extremely broad
Nigs
are
an
indication
of
exceptionaiiy
weak
wood.
Probably, there is
an
optimum
rate of growth for the
production of
the
strongest wood, but
the
rate
dZFers
with
species. For
softwoods
this
optimum is about 7
to
20
rings per
2.5
cm
while
in
ring
porous hardwoods this optimum
is
6
to
10 rings
per
2.5cm.
@esch
and
Dinwoodie
198
1).
In
the
past,
annual
ring
studies
have
been
conducted
mostly by
dendrochronologists.
ln
recent decades, however,
many
investigators
have
tried
to
correlate some
wood
properties
such
as
density
to
wood
anatorny
at
the
annual
ring
level.
Variation
in
specific
graMty
of
white spmce wood
was
observed
in
relation to the rate of
growth and distance
fkom
the
pith
by
Keith
(1961).
It
was
concluded that over
40
percent
of
the
observed variations
in
specific
gravity
could
be related to the variation
in
Nig
width
for
a
specimen
inside
the
6oL
ring.
The
inter-tree (between
trees)
and
intra-tree
(in one tree) ring width
and
wood
density variations were
studied
by
Zhang
and
Jiang
(1998).
Specimens
used in that
study
were
sampled
from
Meen
years
old
black spruce
trees
grown
in
New Brunswick. It
was
suggested that the wood
density
characteristics
are
under
stronger genetic
control
than
ring
width
charactenstics.
Wood density characteristics have a
remarkable
variance
component due to
the
family.
This, to
a
Iess
extent,
applies
to
ring
width.
The
intra-tree
variation of
vaitous
wood
characteristics
is
considerably
larger
than
inter-tree variation in
black spmce. Compared to ring width
charactençtics
(except latewood width and
latewood percent),
intra-ring
wood density charactenstics (except lat ewood density)
show
a
smaller
inter-tree variation but
a
larger intra-tree variation. The latewood
width
and latewood percent show the
srnailest
inter-tree variation and
the
largest
intra-tree
variation.
Cambial
age
explains
much
more variation
in
most
&a-ring
wood density
characteristics
than
the
ring
width.
Wood
density
of
growth
rings
in
black spruce is
dependent
more
on cambial
age
than
growth rate. This investigation
was
continued by
Zhang
(1998)
and
he
found that
with
an
increasing
cambial age both
ring
width and
ring
density
tend to
exhibit
a constant tree to
tree
variation,
whereas
most
other
intra-ring
wood
charactenstics show
a
smaller tree-
to-
tree variation. Correlation between
ring
density
and
most
intra-ring
characteristics
(except
eariywood
density) tend to be
weaker
with
increasing
cambial
age.
White
spmce
specific
gravity
and
tracheid
lengths
were
measured
fiom
increment
cores
of
10
trees each
f?om
four
selected
white
spruce
stands in
Alberta by Taylor
et
al.
(1
982).
The average
specinc
gravity
found
(0.3
3 8)
was
sùnilar
to that found for
other
regions
in
North
America.
Although
average
values
are
quite
unifonn
throughout
the
species range,
there
were
rather
large
specinc
gravi@-
differences
in
the
wood of
individual
trees.
The
results
of regression
analysis
on
values
for
individual
height showed
that
specitic
gravity
increases
nom
the base to the top
of
the
stem.
The relationship
between
specinc
gram
and
radial tree growth
was
quite
complex for
the
trees
studied.
At
breast height,
specific
gravity
decreases
from
the
pith
outward
reaching
a
minimum
at
10
to
25
years
of age
then
rebounding
to
a
higher
specific
gravity at
30
to
40
years
of
age.
In
subsequent growth increments,
specific
gravity
remained
relatively constant
or
increased
only
slightly.
At
the
higher
stem
sarnpling,
the
initiai
decrease
through
the first
10
to
15
years
was
followed
by a
relatively
constant or
decreasing
specific
gravity.
Tracheid length varied
consistently
f?om
tree to
tree and
fiom
height to height
within
the
trees. Tracheid length
increased
with
increasing
rings
f?om
the
pith
outward.
For
comparable growth increments, tracheids
were
slightly
shorter
at
the
breast -height
sampling
point
than
at
samphg
points
higher
in
the
stem.
It
can
be
inferred
that
annual
ring
characteristics have
been
targeted recently
in
research
studies
and
this
type
of investigation is
developing
rapidly.
In
the
near
future
more information
on
annual
ring
behaviour
WU
be
needed.
These
studies
indicated
that
most
variation
of
wood
quality
occurs
dong
radial
direction
both
at
macroscopic
(among
annual
rings
eom
pith
toward
bark)
and
microscopie
Ievel
( i
one
growth
ring
fiom
earlywood
to latewood).
The
intention
of
this
study
is
to
provide
a
new
understanding
of
wood behaviour
in
radial compression
by
focushg
on
the
deformation
within
one
annual
ring.
2.5
WOOD
BEHAVIOUR
DURING
HOT-PRESSING
In
mdact ur i ng
wood composites, wood
elements
in
the
form
of
particles,
wafers, or
strands
are
mixed
with
a
proper
glue
(urea,
phenol,
or
melamine
formaidehyde)
and
formed
into
a
low
density
mattress (mat).
The
mat is
then
hot-pressed
for a
few
minutes and a
board
is
produced. The
properties
of
the
end
product
are
partk
controlled
by
this
process.
Numerous investigations
have
been
done
to address the
relationship
between processing parameters
and
end-product
performance.
In
order
to
show
the importance of
wood behaviour
during
hot-pressing, a
few
articles
that
discuss
wood behaviour
during
hot
pressing
are
reviewed
in
this section.
The
structure
of
a
particle-mat
(formed
wood-glue
system)
and
its
behaviour
during
hot pressing was studied
by
Suchesland
(1967).
He
indicated
that
the
finction
of
the pressing operation
during
manufacturing
of
particleboard
is to develop an
adhesive
bond
between
individual
particles. A
fundamental
variable
affecting
board properties
is
the
degree
of
defonnation
of
particles
during
hot-pressing. The densification
of
the mat
under
pressure
is
significantly
affected
by
press pressure,
temperature
and
mat
moisture
content. Wood elements are
subjected
to compression
perpendicular
to
the
grain
during
hot-pressing. The
deformation
of wood
under
the compression stresses
depends
on the
intra-mat
environment
(temperature
and
humidity).
Therefore,
the deformation behaviour
of
any
individual
particles at a
given
location
in
the mat
wilI
be
controlled
by
changes
in
temperature
and moisture content.
D u ~ g
consolidation of a mat considerable
intemal
stresses
are
accumdated
because
of changes of temperature
and
moisture
content
which
e e c t
final
product
properties.
W e
particleboard
might
be stable
as
it
leaves
the
press,
the excessive densification of parts of
the
mat represents a
potential
for considerable
recovery upon
an
increase
in
moisture
content. The
vertical
density distribution has a
marked
eEect
on the
mechanicd
properties of the board. The density
of
a board
depends
not
only
on
the
degree of
densincation
of the mat but
also
on the density of the
particles.
The degree of contact
depends
on
stress-strain
behaviour of
wood
elements.
The
interna1
structure of fibreboard
and
compressive deformation of the individual
fibres
were
studied
by
Mataki
(1972).
The
cross section of the
fibre
ïncludmg
the
cell
wd,
c d
lumen,
and
intercellular
space
was
measured.
During
manufacturing,
the cross-
section
of
the
individual
fibres is
changed
h m
a
near-circular
into
a
two-lobed
and
fïnally
to a flat ribbon
shape.
It
was
pointed out that the ratio of
fibre
dimension
in
the
fibreboard is dependent on the
original
dimensions of the fibre
especially
with
the
ratio of
the fibre
waii
thickness
to fibre radius.
During
the application
of
pressure to a mat, the deformation
of
the
fibre bundles
differs
fundamentaily
from
the single fibre.
The
compression behaviour of
boards
composed
of
fibre bundles
is
difFerent
from
boards composed
of
single fibres. The
fibres
of
a
bundle
mutually
retain
compressive
deformation
so that the
original
structure
of the
fibres
is
more evident
and
even
the
vessels
are
not completely closed. By
Uicreasing
the
pressure, fibre
bundle
is
deformed
permanently
which
means
ceil
wds
t h s t
laterally
so
that
the
adjacent
fibres
collapse.
W~thin
the
fibreboard, the
fibre
bundle
is
responsible
for
local
irregularities
in the
deformation
of
the
fibres.
Since
the contact pressure between
the
fibre bundles
is
irregular,
fibres
near
the voids are
partially
compressed,
while
fibres
located at points of
fibre
bundies
contact
&er
more
extensive deformation.
In
contrast,
boards composed
primarily
of
single fibres are
characterised
by
a
more
uniforni
internal
structure.
With
the
application
of pressure
up
to
1.3
MPa
to
the
fibre
mat,
large voids between fibres decrease
in
size
and
nurnber
and
the occupation
ratio
of the
fibre cell
wall
(cell
wall
volume to total volume of the board ratio) increases.
However,
since
only
a
moderate
amount
of compressive deformation of the
fibres
occurs,
the
occupation
ratio of
the lumen
remah
relatively constant.
By
increasing
pressure to
2.3
MPa
most
of cell
lumen
is
closed and the void volume between fibres
is
reduced
which
leads
to an increase
in
the occupation ratio of
the
fibre
cell
wall.
The
net
effect
is,
of
course, reflected
in
a
rapid
increase
in
bonding
area
between
fibres.
At
pressures in
excess of
2.75
Mea
the fibres deform
sufliciently
and the compressive deformation of
the
fibre reaches
an
upper
lirnit.
The compression behaviour of wood
during
the
production of
plywood
was
investigated by
Welonse
et
a1.(1983).
Based
on
the
observations of
their
study,
the
loss
in
the
thickness
ofplywood
panels
compressed
in
the
hot-press
may
be
a
result
of
uniform
fibre
compression at
the
panel surface
in
the
eariywood
zones
of
growth rings.
A
zone of
highly
compressed earlywood fibres
was
a
nomal
feature
at
gluelines
of
the plywood
panels
regardiess
of the
amount
of compression.
Such
compression is
essential
if the
two
irregular
surfaces
are
to
conform
to one
another.
However,
when
thick-walled,
latewood
fibres were adjacent to the
glueline,
Little
or
no compression
occurred
even when they
were heated and
plasticized.
When
the panel
compression
was
less
than
5
percent,
the
glueline
was
the major
zone
of compression.
When
the panel compression was
greater
than
7
percent,
additional
bands of
thin-wailed
fibres were compressed in
earlywood
zones of
growth
rings.
The
compressed fibres may have been near the glueline.
Such
compression
fdure
would occur only
whea
the compressive
strength
ofthose
fibres
has
been exceeded and
the
wood
has
been weakened
fiom
overioading.
The
behaviour
of a
waferimat
during
hot pressing was
studied
by
Smith
(1982).
It
was assumed that a
randomly
fonned
wafer-mat
is formed
of colurnns
with
smali
cross-
sectional
area
Wafen
are
distributed
randody
during
fomiing
which
means
that the
number
of wafers sit
in
each
column
would not be
equal
with
neighbouring
columns. As
pressure
is
applied
against
çuch
a
mat
surface,
these
colurnns
containing
uneven wafers
behave
diierently
because
they
have
different
heights. That is,
the
columns with more
wafer
(tdler
columns) support the
bulk
of
the
initial
load.
Aithough
the load
is
applied
evenly
over
the
entire
mat surface, the stress becomes concentrated over
small
portion of
the
area
(taller
columns).
When the stress concentration
increases
above
the proportional
limt
of individuai wood wafers a
collapse
of the wood cellular
structure
occurs. The
column
containing
such
wafers
can
then
no
longer
support the
entire
load
so
the
stress
is
partially
shifted
to some adjacent columns.
In
areas
of the lowest
density,
the stress
concentration
may
never
develop
sufficientiy
to
cause
permanent deformation.
This
pattern
of
shifang
stress
concentrations
continues
until
the
board
reaches
the
ha1
thickness
and
average
density.
AU
of
the
above
mentioned documents
indicate
that
the wood element
stress-
strain
behaviour
during
hot-pressing
plays
an
important
role
in
mo-g
end-product
properîies.
A
better
understanding
of
this
behaviour
will
enable
us
to
make
good
decisions
for
advancing
t dt i onal
products
or
inventing
more
efficient
new
products.
2.6
WOOD
BEHAVIOUR
UNDER
TRANSVERSE
COMPRESSION
The
stress-strain
relationship
of wood
in
compression
across
the
pain
differ
s
from
that
dong
the
grain.
These
relations, especially
inelastic
behaviour, are
conspicuously
affected
by the
anatomical
characteristics
of wood. S
O
fMroods
and
hardwoods
behave
differently.
Numerous
investigations
have been
conducted
on
wood
behaviour
in
compression.
Since
this
study
is
intended
to
explain
wood
behaviour
in
compression
perpendicular
to
grain,
some
articles
discussing
this
behaviour
are
reviewed below
,
Bazhenov
et
aL(1953)
conducted
a
study
on
the
compression
behaviour
of
different species
under
compression
perpendicular-to-grain.
Specimens
with
dimensions
20~20x30
mm
were
prepared
Born
different
species and
conditioned
to
a
moisture
content of 10
percent.
One
group
of
specimens
was
subjected
to
radial
compression
and
the
other
group
to
tangential
compression.
Bazhenov
et
al.
(1
953)
22
inferred
that
the
expression
"proportional
limit"
of
load-deformation
curve
represents
a
purely
graphical
and
conditional
value.
They
divided
the
cume
into
"straight
linen
and
"
curvilinear"
elements.
Their
tem
"plastic