Content uploaded by Roger M. Rowell
Author content
All content in this area was uploaded by Roger M. Rowell on Aug 21, 2014
Content may be subject to copyright.
STABILITY
OF
ACETYLATED WOOD
TO
ENVIRONMENTAL CHANGES
Roger
M.
Rowell
Supervisory Research Chemist
Rebecca S. Lichtenberg
Technician
USDA Forest Service
Forest Products Laboratory
One Gifford Pinchot Drive
Madison, WI 53705
-
2398
and
Pia
Larsson
Graduate Student
Chalmers University of Technology
Department of Forest Products and
Chemical Engineering
S
-
412 96
Gothenburg, Sweden
(Received January 1993)
ABSTRACT
Acetylated wood is more dimensionally stable and resistant to biological attack than unacetylated
wood.
In
this study, the stability of acetylated wood was tested under various pH, temperature, and
moisture conditions. Ground acetylated southern yellow pine and aspen flakes were treated with buffer
ranging from 2 to
8
pH, exposed at 24 C,
50
C,
or
75 C for different periods, and tested for acetyl
content. At 24 C, acetylated wood was more stable at pH 6 than pH 2, 4,
or
8.
At
50
C and 75 C,
acetylated wood was more stable at pH 4 than at the other pH values. The half
-
life of acetylated wood
continuously in contact with a buffered liquid at pH 6 and 24 C was approximately 30 years. For
acetylated wood used under normal circumstances, the half
-
life would be expected to be much longer.
Acetylated southern yellow pine and aspen flakes were also (1) kept at 90% relative humidity at
27
C
for
6
years
or
(2) cycled (42
-
day cycle) between 90 and 30% relative humidity for 5 years. The
loss
of acetyl was less than 2% in both the constant and cyclic relative humidity tests. The stability of
acetylated wood suggests that such wood can be used for products exposed to changes in humidity.
Keywords:
Acetylation, stabilization, pH, moisture.
INTRODUCTION
Acetylation of wood is presently a commer-
cial reality in Japan, soon will be a reality in
the United Kingdom and Sweden, and, hope-
1 The Forest Products Laboratory is maintained in co
-
operation with the University of Wisconsin. This article
was written and prepared by U.S. Government employees
on
official time, and it is therefore in the public domain
and not subject to copyright.
Wood and
Fiber
Science, 25(4), 1993, pp. 359-364
fully, will be realized in North America in the
near future. Wood is being acetylated primar-
ily to improve its dimensional stability and,
to a limited extent, to improve its resistance
to attack.
Since the bond formed by the reaction of
acetic anhydride and the wood (cell-wall)poly-
mers is an ester bond, splitting of the acetyl
group has been questioned if the acetylated
product is to be used over the course of many
360
WOOD
AND
FIBER
SCIENCE,
OCTOBER
1993, V. 25(4)
years. Esters of organic compounds are known
to be susceptible to hydrolysis, especially un
-
der alkaline conditions. Tarkow et al. (1950)
immersed acetylated birch veneers in a 9%
so-
lution of sulfuric acid at room temperature for
18 h.
No
loss in antishrink efficiency (ASE)
was found after removal of the acid, and anal
-
ysis of acetyl content showed no loss in acetyl.
In the same acidic solution at 40
C,
deacety
-
lation did occur to some extent, as shown by
a drop in ASE from 75% to 65%. The authors
also subjected acetylated spruce veneers to 10
cycles of humidity changes from
30%
to 97%
relative humidity over a 4
-
month period. After
the 10 cycles, no reduction in ASE occurred.
In a final test, the authors suspended acetylated
birch wood over a saturated solution of sodi
-
um chloride at
80 C
for
3½
days. Again, no
loss
in ASE was observed.
Another source of information about acetyl
stability can be found indirectly in the litera
-
ture on wood aging. Hedges (1990) and Freed
-
land (1990) reviewed the degradation of the
carbohydrate fraction as a function of time and
conditions. The carbohydrate fraction de
-
grades at a much faster rate than does the lignin
fraction, and the hemicelluloses degrade to a
greater extent than does the cellulose. Acetyl
is lost in the hemicellulose fraction but not
completely. Kauri wood from New Zealand,
for example, that was buried for
30,000
years
under conditions that resulted in a
50%
loss
in hemicelluloses and a 60%
loss
in xylose,
showed an 85% loss in acetyl content (Freed
-
land 1990). Similar results have been reported
with other woods under different conditions
for varying periods.
The purpose of our research was to expose
acetylated wood to varying pH, moisture, and
temperature conditions and determine loss of
acetyl. These data can be used to predict the
expected life of an acetylated product under
different use conditions.
MATERIALS AND METHODS
Preparation
of
specimens
Southern yellow pine and aspen wood flakes
were acetylated using uncatalyzed acetic an
-
hydride as previously described (Rowell et al.
1986). After final drying, the average analyti
-
cally determined acetyl content (on a weight
basis) was 20.2% for southern yellow pine flakes
and 20.6% for aspen flakes.
pH
and temperature conditions
Part of each flake type was ground to pass
through a 10
-
mesh (2.0
-
mm) screen. Approx
-
imately 14 g of each ground sample was stirred
into a 1
-
liter flask containing 0.5 liter of buffer
solution at pH 2, 4,
6,
or 8. The accuracy of
the buffer solutions was checked with
a
pH
meter at the start of the experiment and each
time a sample was removed for analysis. The
flasks were placed in one of three different tem
-
perature conditions: 24
C
(room temperature)
or
a
50 C
or 75
C
water bath. At various times,
ground wood samples were removed from the
flasks and the acetyl contents determined by
gas chromatography. Duplicate tests were run
and the results averaged.
Humidity conditions
Part of each acetylated flake type was placed
in a constant relative humidity (RH) room at
90%
RH at 27
C.
Samples were removed every
6
months and ground to pass a
1
0
-
mesh screen.
Half of each sample was leached in water for
3
days; acetyl contents were determined for
leached and unleached samples. Triplicate
samples were run and the results averaged.
Another part of each acetylated flake type
was cycled between constant 90% RH, 27
C
and
30%
RH, 27
C
conditions. Each cycle con
-
sisted of 21 days at 90% RH and 21 days at
30%
RH. At the end of each 42
-
day cycle,
samples were removed and ground to pass a
10
-
mesh screen. Half the sample was leached
in water for
3
days. Acetyl contents were de
-
termined for leached and unleached samples.
Triplicate samples were run and the results
averaged.
RESULTS AND DISCUSSION
Because of the limited number of specimens
per individual test, statistical analysis of the
data was not appropriate. The results present
-
Rowell et
al.
-
STABILITY
OF
ACETYLATED
WOOD
TO
ENVIRONMENTAL
CHANGES
361
F
I
G
.
I.
Relationship between acetyl content and time
over a pH range of
2
to
8
at
24
C
for aspen.
ed here should be considered as indicative of
trends that a larger, statistically valid experi
-
ment should confirm.
Assuming that the deacetylation of acety
-
lated wood is dependent on the content of the
acetyl groups, first
-
order reaction kinetics were
used for all calculations. Figures
1
to
3
show
the relationship between acetyl content and
time for aspen over a pH range of 2 to
8
at 24
C,
50
C,
and
75
C,
respectively. Figures 4 to
6 show the same data for pine. These data show
the great difference in stability of the acetyl
group under acidic conditions as compared to
slightly alkaline conditions.
From the slope of each plot in Figs.
1
through
6, it is possible to determine the reaction rate
F
I
G
.
2.
Relationship between acetyl content and time
over a pH range of
2
to
8
at
50
C
for aspen.
F
I
G
.
3.
Relationship between acetyl content and time
over
a
pH range of
2
to
8
at
75
C
for aspen.
constant, k, for each pH and temperature. Fig
-
ure
7
shows the relationship between reaction
rate constant and temperature over a pH range
of 2 to
8
for aspen, and Fig.
8
shows the same
data for pine. Table
1
shows the rate constants
for acetylated aspen and pine at 24
C,
50
C,
and
75
C
at pH values of 2,
4,
6, and
8.
At 24
C,
deacetylation occurred at the slowest rate
at pH 6 and at the fastest rate at pH
8.
At
50
C
and
75
C,
the slowest rate of deacetylation
occurred at pH 4 and the fastest rate at pH
8.
As
expected, raising the temperature at each
pH level greatly increased the rate of deacetyla
-
tion. Also as expected, the acetyl groups were
more stable under acidic conditions than basic.
The rate constants represent the bulk rate of
F
I
G
.
4.
Relationship between acetyl content and time
over a pH range of
2
to
8
at
24
C
for pine.
362
WOOD
AND
FIBER
SCIENCE,
OCTOBER
1993,
V. 25(4)
F
IG
.
5.
Relationship between acetyl content and time
over a pH range
of
2
to
8
at
50
C
for
pine.
deacetylation of wood
-
that is, under the con
-
ditions used in these experiments, it is not pos
-
sible to see the difference between the rate of
deacetylation of the lignin fraction separately
from the rate for the hemicelluloses. Because
isolated lignin has been shown to acetylate
faster than the hemicelluloses,
it
might be ex
-
pected that their rates of deacetylation would
be different (Rowell et al. in press). It is also
interesting that the rate of deacetylation was
essentially the same for the hardwood and soft
-
wood used in the study reported here. It might
be expected that since they differ both in the
content and type of hydroxyl groups in both
lignin and hemicelluloses, the rate of deacety
-
lation might be different.
F
IG
.
7.
Relationship between reaction rate constant
(k)
and temperature (l/T) over a pH range of
2
to
8
for aspen.
Using the reaction rate constant, the half
-
life for the acetyl groups can be determined.
When half the acetyl content is hydrolyzed,
dimensional stability and biological resistance
are reduced and the half
-
life data can be used
as an indication of how long acetylated wood
could be expected to perform in a given pH
and temperature environment. Table
1
shows
the half
-
life for aspen and pine at each pH and
temperature used.
The half
-
life data suggest that wood acety
-
lated to
20
weight percent gain would deacety
-
late to 10 weight percent gain in approximately
30
years at room temperature at pH
6.
The
dimensional stability or
ASE
of wood acety
-
lated to about 10% is about
50%
compared to
F
IG
. 6. Relationship between acetyl content and time F
IG
.
8.
Relationship between reaction rate constant
(k)
over a pH range
of
2
to
8
at
75
C
for
pine. and temperature
(1/T)
over a pH range
of
2 to
8
for pine.
Rowell
et al.
-
STABILITY
OF
ACETYLATED
WOOD
TO
ENVIRONMENTAL CHANGES
363
T
ABLE
1.
Rate constant and half
-
life
of
acetylated aspen
and pine at different
pH
and temperature conditions.
85%
for wood acetylated to 20 weight percent
gain.
From the slope of the plots in Figs. 7 and
8,
it is possible to calculate the activation energy
for the deacetylation at each pH. Table 2 shows
the activation energy for aspen and pine at
each pH level used in these experiments. The
activation energy
is
about
90
kJ/mole at pH
6,
65
kJ/mole at pH 4,
60
kJ/mole at pH 2,
and
55
kJ/mole at pH
8.
Table
3
shows the acetyl content for aspen
and pine flakes kept at
90%
RH and 27
C
for
6
years. Within experimental error limits, very
little acetyl content was lost during this time.
So
little acetyl was lost that rate constants, half
-
life estimates, and activation energy calcula
-
tions could not be determined. It is interesting
to note that whereas untreated flakes darkened
as a result of attack by microorganisms in the
T
ABLE
2.
Activation energyfor deacetylation
of
aspen and
pine over temperature range
of
24
C
to
75
C.
90%
RH room within 2 weeks, the acetylated
flakes were still their original color after
6
years.
Table 4 shows the acetyl content for aspen
and pine flakes cycled between
90%
and
30%
RH at
27
C.
Since each cycle represents 42
days, a total of
41
cycles represents about 4.7
years. As with the flakes kept at constant RH,
very little loss of acetyl was observed after the
cyclic RH conditions.
The acetyl values in Tables
3
and 4 for
leached and unleached samples are essentially
the same, showing that no soluble acetate
-
con
-
taining compounds were formed during the
experiments.
CONCLUSIONS
Results from these experiments show that
acetylated wood is much more stable under
slightly acid conditions as opposed to slightly
alkaline conditions. Acetylated wood should
be stable for a very long time under normal
air temperatures and humidity. In addition,
the stability of acetylated wood under condi
-
tions of high or cyclic humidity suggests that
such wood can be used for products exposed
to changes in humidity.
The data can be used to predict the stability
of acetylated wood or other biobased fiber
composites at any pH and temperature com
-
bination to estimate the life expectancy of the
product in its service environment. However,
it must be remembered that data for this study
were collected using either finely ground pow
-
der or small fiber. As a result, a very large
364
WOOD
AND
FIBER SCIENCE, OCTOBER
1993,
V. 25(4)
T
ABLE
3.
Acetyl content
of
aspen and pine
flakes
after exposure to
90%
relative humidity at
27
C.
T
ABLE
4.
Acetyl content
of
aspen
and pine
flakes
after
REFERENCES
relative humidity cycles.
surface area came into contact with the various
pH and temperature environments. We expect
that these data represent the fastest possible
degradation rates and that composites made
from acetylated fiber would degrade at
a
much
slower rate since there would be much less
surface contact. This is known to be the case
in archaeological wood, in which acetyl groups
have been very stable over thousands of years
and little
loss
of acetyl has been observed.
Printed
on
recycled
paper
