j m a t e r r e s t e c h n o l .

2 0 2 0;9(4):6994–7001

Available online at www.sciencedirect.com

www.jmrt.com.br

Original Article

Curing and thermal properties of co-polymerized

tannin phenol–formaldehyde resin for bonding
wood veneers

Nur Liyana Mohamad Hafiz a , Paridah Md Tahir a,∗ , Lee Seng Hua a,∗ ,
Zurina Zainal Abidin b , Fatimah Athiyah Sabaruddin a , Noryuziah Mohd Yunus c ,
Ummi Hani Abdullah d , Abdul Khalil H.P.S. e
a Institute of Tropical Forestry and Forest Product, Universiti Putra Malaysia (UPM), 43400 UPM Serdang, Selangor, Malaysia
b Chemical Engineering Department, Faculty of Engineering, Universiti Putra Malaysia (UPM), 43400 UPM Serdang, Selangor, Malaysia
c Faculty of Applied Science, Universiti Teknologi Mara Pahang, 26400 Jengka, Pahang, Malaysia
d Faculty of Forestry and Environment, Universiti Putra Malaysia (UPM), 43400 UPM Serdang, Selangor, Malaysia
e School of Industrial Technology, University Sains Malaysia, 11800 Penang, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this study is to assess the curing behaviour and thermal properties of
Received 12 March 2020 phenol–formaldehyde (PF) adhesives with different degree of tannin substitution using sev-
Accepted 10 May 2020 eral thermal analysis techniques. Tannin from Acacia mearnsii bark was copolymerized with
Available online 8 June 2020 PF to form a bonding agent for plywood. The resin mixture was heated at 50 ◦ C prior to
adding 40% w/w tannin solution, followed by a continuous stirring for 2 h. The amounts of
Keywords: tannin substitution were 20%, 30% and 40% (w/w of PF solids) and 100% PF were used as con-
Adhesive curing trol. The study shows that the addition of tannin quickened the gel time of the resin notably,
Glass transition temperature parallel with the increase in viscosity of the resin. The dry shear bond strength of the tannin
Rigidity phenol–formaldehyde (TPF) co-polymer ranged between 1.71 and 2.58 MPa as compared to
Bonding shear 3.41 MPa for PF. At higher addition of tannin, the shear bond strength reduced significantly
predominantly the 40% tannin substitution. Formulation containing 20% tannin was found
comparable to that of neat PF. Thermal test revealed that TPF starts to degrade at a lower
temperature than PF resin. All TPF resins started to cure at about 125 ◦ C and peaked at 160 ◦ C
compared to PF that began to cure at 145 ◦ . The DMA of the resin showed an increase in the
resin stiffness with an addition of tannin. These results imply that even though TPF starts
to degrade at lower temperature, the cured TPF is relatively stronger than pure PF.
© 2020 The Author(s). Published by Elsevier B.V. This is an open access article under the
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

∗
Corresponding authors.
E-mails: parida@upm.edu.my (P.M. Tahir), lee seng@upm.edu.my (L.S. Hua).
https://doi.org/10.1016/j.jmrt.2020.05.029
2238-7854/© 2020 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
 j m a t e r r e s t e c h n o l . 2 0 2 0;9(4):6994–7001 6995

for polypropylene with anti-UV properties via dynamic cur-

1. Introduction ing extrusion. Cardona [21] and Mosiewicki et al. [22] found
that tannin addition slightly increases the flexural properties
The tannin-based adhesive has been used as a partial sub-
of the PF glycerol reinforced composites and woodflour/tannin
stitution for the commercial phenolic resin to minimize
adhesive composites, respectively.
the dependency on petroleum-based resources for decades.
Nonetheless, limited literatures have been reported on the
Due to its chemical structure, condensed tannins have high
curing behaviour and thermal properties of TPF resins partic-
reactivity towards chemicals such as formaldehyde, hence
ularly as wood adhesive. Temperature plays an important role
constitute more than 90% of the total tannin production and
in thermoset based adhesive such as tannin-based adhesive.
tannin-based adhesive in the world [1]. Such wide usage
It is a well-known fact that thermal properties are necessary
is ascribed to the structures of polyphenolic compounds
to determine the resin’s curing behaviour as well as thermal
that contain groups of polyhydroxyflavan-3-ol oligomers and
stability which consequently influences the mechanical prop-
polymers linked by carbon–carbon bonds between flavonol
erties of the adhesive. Due to increasing degree of conversion
sub-units. Many studies have been carried out using different
presented by a hydroxyl group, tannin-based adhesive tends
types of tannin sources [2–4], but the most notable works were
to harden much faster than PF with an increase in temperature
by Pizzi and co-workers [5–7] using wattle and pine tannins
[23]. The rate of hardening depends on the hydroxyl moiety in
for bonding particleboard and plywood. To date, only tannins
the A- and B-ring of the flavonoid structure. It was claimed that
from black wattle (Acacia mearnsii), pine, quebracho, chestnut
higher tannin proportion in tannin-based adhesive provides
and oak have been used commercially [8].
better performance in terms of thermal stability. For instance,
The success in commercializing of tannin could be
the addition of A. deabalta tannin significantly lowers the cur-
attributed to its two important characteristics, its simple
ing temperature of PF [24]. In another study, Vázquez et al.
extraction process uses as well as its resistance against water.
[25] reported that the increasing amount of tannin addition
According to Garcia et al. [9] presence of hydrocolloid gums in
decreases the curing enthalpy for phenolic resin. Moreover,
tannin initiates the increase in viscosity and its coagulation
Chupin et al. [26] found that tannin helps in increasing the
behaviour tends to absorb water molecule rapidly resulting in
thermal stability of resin. Extensive usage of TPF as a matrix
polymerization. Tannin compounds contain enough hydroxyl
for a reinforced composite was also investigated by Barbosa
groups for effective cross-linking with other compounds and
et al. [27] who concluded that the presence of tannin in PF
are a natural and feasible source for coagulant synthesis [10].
resin provides greater adhesion between the fibres and the
Acacia tannin particularly that of A. mearnsii, is known for
matrix. Similar results were reported by Ramires and Frollini
its congruity to phenol-formaldehyde (PF) and it is also the
[28]. They found that TPF composite provides better stiffness
main source of world tannin production. Unlike A. mearnsii, A.
and dispersive energy properties hence suggesting that TPF
mangium tannin has a very high viscosity at a very low solid
superseded PF in terms of thermal stability.
content [11]. The use of polar solvent such as methanol helped
The thermal properties of tannin-based adhesive may be
to increase the yield as well as reduce the viscosity of the
comparable to that of phenolics due to their polypheno-
tannin [12,13] Jahanshaei et al. [14] used methanol to extract
lics structure that is congruent to phenol. Study on curing
Quercus sp. bark and obtained a tannin solution of pH 8 and vis-
behaviour and thermal properties of TPF resins is very cru-
cosity of 400 cP. In another study, El Badawi et al. [2] found that
cial if this resin is to be used in other applications such
A. seyal has rather acidic tannin, i.e. pH 5.4. Other than Aca-
as wood polymer composites, wood coating and laminating
cia, several other tannins such as radiata pine (Pinus radiata),
paper. Such property will also suggest the level of stability of
spruce (Picea abies), maritime pine (Pinus pinaster), a leppo pine
the resin when exposed to high temperature. As these char-
(Pinus halepensis), douglas fir (Abies alba) and eucalyptus (Euca-
acteristics are well-related to the mechanical properties of
lyptus globulus) have been found suitable to be used as adhesive
the adhesive, it is therefore worth to be investigated. There-
in wood industry [15–17]. Other studies have been reported
fore, the aim of this work is to study the curing behaviour of
for different applications which include laminated paper from
TPF adhesives using several thermal analysis techniques to
low viscosity A. mearnsii bark extract [17], di-isocyanates from
determine the energy released during resin curing, the rate
viscous pine tannins as coating material for wood [18].
and extent of chemical conversion, the degradation rate, and
A number of studies have reported the effects of tan-
to estimate the changes in the mechanical properties during
nin addition on the mechanical properties of composite
transformation of the liquid resin into a cross-linked structure.
panels. For instance, Hoong et al. [1] stated that tannin
phenol–formaldehyde (TPF)-bonded plywood exceeds the
requirement for shear strength of EN 314-2, and was compara- 2. Materials and method
ble to that of PF-bonded plywood. Zhou and Du [8] concluded
that tannin resin adhesives should be modified by mixing 2.1. Preparation of resin
them with other synthetic resins or organic or inorganic
modifiers, and by optimizing resin synthesis parameters and The commercial PF adhesive was provided by Malayan Adhe-
hot-pressing conditions to meet the requirements set for the sives and Chemicals Sdn. Bhd., Malaysia. Table 1 gives the
exterior plywood adhesive. Tannin has also been used as physical properties of the resin used in the study. Acacia mearn-
matrix for fibre reinforce composites. Zhu et al. [19] obtained sii tannin powder was obtained from Silvateam, San Michele
high tensile strength flax reinforced tannin-based composite, Mondovì, Italy. Tannin solution with 40% solid content was
whilst Liao et al. [20] used tannin as reinforcing components prepared in a 100 mL beaker. In a reflux condition, a fixed
6996 j m a t e r r e s t e c h n o l . 2 0 2 0;9(4):6994–7001

replacement, and 100% PF resin as control samples denoted

Table 1 – Physical properties of PF resin.
as 20 TPF, 30 TPF, 40 TPF and 100 PF, respectively. The open
Property Value
assembly time was 25 min and the boards were pre-pressed
◦ at room temperature for 5 min. Based on our earlier work [11],
pH at 25 C 8.81
Viscosity in poise at 25 ◦ C 0.22 the consolidated veneers containing TPF were hot-pressed at
Free formaldehyde in % at 25 ◦ C 9.63 140 ◦ C for 10 min, whilst those of PF were cured at 150 ◦ C for
Non-volatile content in % at 150 ◦ C/2 hrs 49.8 10 min. The pressure used was 4 MPa. For every adhesive for-
Water tolerance at 25 ◦ C >4000
mulation, three panels were fabricated. Therefore, a total of
12 plywood panels were produced in this study.
Table 2 – Ingredients used in the preparation of TPF
resin. 2.3.2. Assessment of bond integrity
Ingredient Tannin substitution (%) The bond integrity of the TPF-bonded plywood was examined
based on shear strength and percent wood failure. The ply-
20 30 40
wood was conditioned at 20 ◦ C and 65% RH for 1 week prior
Phenol–formaldehyde at 47.3% 100 100 100 to cutting into test specimens. The plywood shear specimens
solids (g) were tested in dry, wet or cold water soaking and wet/boiled
Tannin powder at 100% solids 9.5 14.2 18.9 conditions using 30 kN Universal Testing Machine (Instron). In
(g)
the wet condition, the specimens were soaked in cold water
Water needed to produce 14.2 21.2 28.4
at 20 ± 3 ◦ C for 24 h, then wiped for excess water and covered
tannin with 40% solids (g)
Total solution (g) 123.7 135.5 147.3 with wet cloth. The boiling test was done by soaking the spec-
imens in boiling water under normal atmospheric conditions
for 72 h, followed by cooling in water at 20 ± 3 ◦ C for 1 h. Wet
amount of PF resin was charged into a 500 mL flask when samples were wiped dry prior to testing. The percentage of
the temperature of the water reaches 50 ◦ C. Once the PF resin wood failure was estimated visually on sheared specimens
reached 50 ◦ C, tannin solution was added followed by a con- to the nearest 5%. Ten samples were used for the bonding
tinuous stirring for 2 h. The formulations were prepared using integrity assessment.
three different tannins to PF ratios namely 20%, 30% and 40%
(w/w of solids of PF resin). Table 2 gives the proportions of each
2.4. Evaluation of thermal properties of TPF
ingredient used to prepare the TPF resin in the study.
2.4.1. Differential scanning calorimetric (DSC)
2.2. Determination of physical properties of adhesive Differential scanning calorimetric was done using the
Research Instrument model Q20 equipment. About 5 g of resin
The physical properties, i.e. pH, gel time, viscosity and solid sample (PF and TPF) was placed on a balance located in the fur-
content, of both TPF and PF resins were determined follow- nace and heat was applied over the temperature range from
ing methods described by Nor Hafizah et al. [11]. The pH of 30 ◦ C to 250 ◦ C at a heating rate of 10 ◦ C/min. The results were
the resin solution was measured by using Mettler Toledo pH analyzed with the Universal Analysis software.
metre. Gelation time was determined by immersing the resin-
filled test tube in a water bath that was maintained at 90 ◦ C
2.4.2. Thermogravimetric analysis (TGA)
and the time taken for the resin to gel was recorded. By using
Thermogravimetric analysis was carried out using Perkin
Brookfield metre, model DV2T with spindle 31, the viscosity
Elmer USA equipment. About 10 g of each cured samples were
of the resin solutions was determined. The solid content of
placed on a balance located in the furnace tube and the heat
the adhesive was calculated based on the oven dry weight of
was applied over the temperature range from 30 ◦ C to 600 ◦ C at
the sample. The sample was dried in an oven at 100 ◦ C for at
a heating rate of 5 ◦ C/min during 90 min in argon gas. Weight
least 4 h. The final weight of the sample was then recorded to
gain (TG curves) and the rate of mass loss (DTG curves) of each
calculate the value of solid content.
formulation was determined.

2.3. Determination of bonding strength

2.4.3. Dynamic mechanical analysis (DMA)
Dynamic mechanical analysis measures the complex modu-
2.3.1. Preparation of plywood sample
lus and compliance as a function of temperature, time and
The bond integrity of the TPF co-polymer was assessed by
frequency. The method of testing was adapted from Lei and
glueing of veneers into plywood. Rubberwood (Hevea brasilien-
Frazier [29] by pre-curing the resin at 60 ◦ C.
sis) veneers of 2.6 × 200 × 200 mm3 in size were collected from
Mentakab Veneer Plywood Sdn. Bhd. The moisture content
of the veneers was between 8% and 12%. A predetermined 3. Results and discussion
amount of TPF based on a spread rate of 250 g/m2 was weighed
and spread onto the veneers using single glue line method 3.1. Physical properties of TPF
and assembled into three-layer cross-laminated plywood with
opposing tight-side and loose-side veneers arranged at the The physical properties, i.e. pH value, viscosity, gel time and
glue lines. Four types of plywood were prepared using differ- solid content, of A. mearnsii TPF and PF resins are shown in
ent adhesive formulations, namely 20%, 30%, and 40% tannin Table 3. The pH values of TPFs are almost similar which varied
 j m a t e r r e s t e c h n o l . 2 0 2 0;9(4):6994–7001
Table 3 – Properties of Acacia mearnsii TPF and PF resins.
Resin pH value Viscosity Gel time Solid content
(poise) (mins) (%)
100 PF 8.81 0.22 54 49.80
20 TPF 8.19 4.3 47 45.20
30 TPF 7.41 4.1 40 48.12
40 TPF 7.27 4.95 36 45.15
between 7 and 8 whilst PF has slightly higher pH of 8.81. The
reduction in pH values as the amount of tannin increased sug-
gest that the curing of the TPF would be slower, thus requires
longer time or higher temperature to cure compared to neat
PF. The lowest pH was given by TPF with 40% tannin substi-
tution (40 TPF) which was 7.27, inferring the acidic nature
of tannin. The acidic nature of tannin adhesive of has been
reported by other studies [1,30], where it could be as low as
pH 4.9. This finding suggests that both pressing temperature
and time should be predetermined according to the increase
in tannin content in the phenolic system prior to any bonding
work.
This phenomenon, however, was not backed by the trend in
gel time. For example, the neat PF resin took 54 min to gel but
the 40 TPF required only 36 min whilst both 20 TPF and 30 TPF
took 47 and 40 min respectively. Such behaviour suggests that
pH does not have significant influence over gel time. Instead,
viscosity appears to inversely affect the gel time, where the
higher the viscosity, the shorter is the gel time.
Tannin contains high molecular weight hydrocolloid gums
[5], oligomers and flavonoid [1]. Therefore, the presence of
these compounds makes the tannin solution more viscous
and it is directly proportional to the percentage of tannin
added. As mentioned in Hoong et al. [1], tannin has a possibil-
ity of having up to 11 replicates of oligomers unit that leads to
the more viscous resin.
The addition of tannin appears to have reduced the solid
content of the phenolic resin. As can be seen in Table 3, pure
PF contains relatively higher percentage of solids than the TPF
resins. This is expected as the reaction mechanism of phenol–
and tannin–formaldehyde for chain incorporation is by a con-
densation reaction through the hydroxyl side chains that form
water molecules. Such reaction causes a reduction in molec-
ular weight to the phenolic structure. As phenol has only one
hydroxyl group compared to four or five in tannin, less water
molecules will be produced. The reduction in final solid con-
tent of TPF indicates the reaction level of the tannins as a
result of such a process. Santos et al. [10] attributed this occur-
rence to coagulation of water behaviour carried by tannin that
causes tannin to retain water in the molecules.
3.2. Bonding strength
The bonding strength of tannin–phenolic resin was evaluated
via plywood shear test and the results are given in Table 4. To
evaluate the adhesion of bonded product, both shear strength
and wood failure percentage were analyzed. Ideally, a good
bonding is indicated by high shear strength and high wood
failure percentage [11]. However, shear strength is influenced
by the wood used in the glueing study. For instance, in tropical
countries, rubberwood (density 560–640 kg/m3 ) is used whilst
6997
in temperate countries, beech (density 700–900 kg/m3 ) is used
as substrate for glueing. Higher density wood would normally
give higher shear strength, but low percent wood failure and
vice versa.
In this study, the shear strengths in dry, wet and 72 h
boil condition show similar trend. Increment in tannin con-
tent markedly decreased the shear values. However, all the
shear values met the minimum requirement set by EN314-1
for interior use, i.e. 1.0 MPa. The reduction in shear strength
of TPF-bonded plywood ranged between 16% and 49% as com-
pared to 15–27% for PF-bonded. The 20 TPF showed the best
performance among the TPFs with reduction in shear strength
of only 18% for both wet and boil conditions.
A similar trend can be seen in wood failure percentage
where the failure percentage reduced along with increasing
tannin content. Nevertheless, in dry condition, all the wood
failures of the sheared specimens are ≥90% irrespective of tan-
nin content, suggesting that tannin can effectively bond the
rubberwood veneer and producing excellent bond strength.
According to EN 314-2:1993, wood failure is used to evaluate
the bonding quality for plywood panels with shear strengths
lower than 1.0 MPa. It also specifies that the percentage of
wood failure should be more than 40% for the panels shear
strengths between 0.6 and 1.0 MPa. From this study, it was
found that all the plywood had shear strength of >1.0 MPa,
therefore the wood failure percentage is not required for the
assessment of bond integrity of the plywood. It is notable that
all resin formulations used in this study was able to produce
good bonding properties based on both the shear and wood
failure values. This revealed that the tannin–phenolic resin
was able to penetrate the wood and cure within the wood cells
after the heat is applied to the adhesive system.
3.3. Thermal properties of the tannin phenol
formaldehyde (TPF)
3.3.1. Differential scanning calorimetric (DSC)
DSC has been used extensively for the study of physical
transformations and chemical reactions of materials. The
basic assumption for the application of DSC to the curing of
thermoset polymers is that the measured heat flow (dH/dt)
is proportional to the reaction rate (da/dt). The total heat
detected during a reaction is identical to the heat evolved by
the curing reaction. In other words, DSC measures the amount
of energy absorbed or released by a polymer resin when
it is heated, providing quantitative and qualitative data on
endothermic (heat absorption) and exothermic reaction (heat
evolution) processes. At a maximum temperature of 250 ◦ C,
the curves depict the curing process of phenolics. Fig. 1 illus-
trates DSC curves of PF and TPF of different amount of tannin
substitution. Each type of resin showed a single endothermic
peak having a maximum peak around 100–115 ◦ C. Endother-
mic peak infers the energy needed to cure the resin, where
the higher the energy, the greater is the cross-linking pro-
cess and more heat is being released. Thus, the amounts of
heat released, a typical reaction of endothermic reaction, are
proportional with the number of crosslinks formed [25]. From
Fig. 1, a steep peak can be observed for all tannin–phenolic
resin which suggests the presence of large molecule tannin
6998 j m a t e r r e s t e c h n o l . 2 0 2 0;9(4):6994–7001
Table 4 – Plywood shear strength based on different % of tannin addition on PF.
Panel Dry (MPa) Wood Wet (MPa) % Reduction
failure (%) from dry
100 PF 3.41 100 2.89 15
20 TPF 2.58 100 2.14 18
30 TPF 2.12 90 1.78 16
40 TPF 1.71 90 1.38 19
Fig. 1 – Differential scanning calorimetric curves
comparison between PF and TPF resins.
in the polymerization process. The glass transition tempera-
ture (Tg ) of the resin can also be determined by locating the
heating changes in the curve. Nevertheless, as mentioned by
Matsumae et al. [16], the specific Tg of tannin is difficult to be
determined due to the complex structure of tannin. Hence the
Tg was normally reported as ‘a range of temperature’. From the
curve, increasing tannin ratio in PF decreased the Tg tempera-
ture. The highest Tg was recorded at 108.15 ◦ C for pure PF and
the lowest Tg was recorded by 30 TPF at 103.85 ◦ C.
Addition of tannin increase the heat released due to its
natural large structure and thus creates a larger number of
cross-linking. Besides, this may lead to better mechanical
properties [25]. At around 120–140 ◦ C, the curve in the pres-
ence of tannin presents a minor exothermic peak may be due
to the glass transition and curing phase performed by the
resin. This peak corresponds with the formation of methy-
lene bridges through condensation of methylol groups during
the curing of the resin [25]. In this case, the addition of tannin
causes a gradual decrease in the curing phase of the resin. Pure
PF curing phase marks at around 145 ◦ C followed by 20 TPF
which is at approximately 135 ◦ C. The 40 TPF records around
130 ◦ C of curing phase and finally 30 TPF at roughly 125 ◦ C.
These results supported a finding from Kim and Kim [23] that
stated addition of tannin lowers the curing temperature of
PF.
3.3.2. Thermogravimetric analysis (TGA)
Fig. 2 depicts the comparison of thermogravimetric curves
amongst PF and TPF resins. The thermogravimetric curve for
100PF sample exhibits the first stage of degradation from 50 ◦ C
to around 130 ◦ C. The weight loss shown by pure PF at this
stage was 12.9% with a peak temperature at 103.68 ◦ C as listed
in Table 5. This could be ascribed to the loss of water content
and trace formation of volatile compounds such as CO and
CO2 released by the resin. The second degradation occurs at
Wood 72-h Boil (MPa) % Reduction Wood
failure (%) from dry failure (%)
80 2.48 27 50
70 2.11 18 30
60 1.08 49 50
60 1.09 36 60
Fig. 2 – Thermogravimetric thermogram for PF and TPF
resins.
around 170–250 ◦ C which may be attributed to the degradation
of the adjacent resin chain while further curing process takes
place [31]. Meanwhile, the degradation between 400 ◦ C and
600 ◦ C may be due to the decomposition of rigid segments in
the resin with a release of CO, CO2 , CH4 and carbon formation
[32]. The weight loss in this stage may be due to the break-
down of methylene bridges to form aldehydes and phenols
[33].
The 30 TPF shows a slower degradation curve from 100 ◦ C
to 150 ◦ C as compared to pure PF. Meanwhile, both 20 TPF
and 40 TPF have almost the same degradation rate as that
of 100 PF. The second degradation stage occurs at 500 ◦ C. The
total weight loss of TPF resins decreases along with increas-
ing amount of tannin replacement inferring tannin is more
heat tolerance. At this stage, the degradation was attributed
to the formation of carbon and hydrogen such as CH4 , CO or
CO2 in the presence of C–C cleavage bond [34]. Further weight
loss may be due to further decomposition of tannin to form a
carbon skeleton [32]. At 600 ◦ C, PF with 40% of tannin replace-
ment reports 31.71% residue which is comparable to that of
pure PF sample with residue amounted to 31.15%. The 30 TPF
resin shows the highest amount of residue of 43.93%, implying
it is more thermally stable than the others including pure PF.
3.3.3. Dynamic mechanical analysis (DMA)
Fig. 3 displays the storage modulus (E ) graph for PF and TPF
resins. As observed from Fig. 3, all samples experienced sub-
stantial decrease of E from 40 ◦ C to 60 ◦ C. At around 60 ◦ C, the
E value is constant and it may be due to cross-linking as the
storage modulus stabilize and, subsequently, increases when
it reaches around 80 ◦ C. It should be noted that all the resin was
pre-cured at 60 ◦ C and causes partial glassification of the resin.
Furthermore, heat affects the polymer simultaneously in two
different ways, namely physical effects and chemical effects
[29]. As for physical effects, heat would cause softening and
decrease the stiffness of polymer. Meanwhile, as for chemical
 j m a t e r r e s t e c h n o l . 2 0 2 0;9(4):6994–7001 6999

Table 5 – Temperature degradation point and weight loss of PF and TPF resins.
Sample Initial degradation (◦ C) Initial weight loss (%) Final degradation (◦ C) Final weight loss (%) Total residue (%)

100 PF 103.68 12.90 543.38 8.526 31.15

20 TPF 101.26 12.49 589.07 8.650 33.06
30 TPF 123.05 11.55 581.43 10.52 43.93
40 TPF 110.55 17.76 545.49 6.559 31.71

Table 6 – Value of dynamic mechanical analysis (DMA) for PF and TPF resins.
Sample Storage modulus (E max) Loss modulus (E max) Tan delta peak (◦ C)

(◦ C) (MPa) (◦ C) (MPa)

20 TPF 190.08 2630 170.67 232.6 147.41

30 TPF 196.32 3155 173.81 279.4 156.01
40 TPF 189.03 4321 165.65 357.5 150.10
100 PF 186.51 2839 175.68 354 160.18

Fig. 3 – Storage modulus (E ) of PF and TPF resins. Fig. 5 – Loss modulus (E ) between PF and TPF resins.

Fig. 4 – Tan ı curve for PF and TPF resins.
effects, the heat increases the stiffness of polymer due to the
cross-linking or polymerization during curing reaction under
high temperature. This can be explained that from Fig. 3, the
heating physical effect overcomes the chemical effect during
120 ◦ C to 160 ◦ C. After 160 ◦ C, all samples showed increment of
storage modulus indicate fully crosslinking.
In terms of gelation temperature, the gelation point of the
resin starts when tan ı value forms a peak at around 130 ◦ C.
As shown in Fig. 4, the gel temperature decreased when the
tannin was added and increased from 20% to 40%. The gelling
point starts at 134 ◦ C for 100PF and decreases to around 120 ◦ C
for 40 TPF. The first peak from tan ı found in PF resin can be
set as the gelation point whilst the second peaks for PF resin
refers to glassification.
As observed from Fig. 4, at around 140 ◦ C, the storage mod-
ulus E value increases drastically. Meanwhile, the summary of
dynamic mechanical analysis results is displayed in Table 6.
As mentioned by Kim [35], the active curing reaction can be
decided within the range where the maximum loss modu-
lus E value is shown. Based on the loss modulus E graph
in Fig. 5, a significant peak shifting to the left was observed.
In this case, the addition of tannin decreases the active cur-
ing temperature from 175 ◦ C for 100 PF to 165 ◦ C for 40 TPF.
Additionally, it can be said that the curing for all TPFs starts
at 130 ◦ C and reach its peak of curing temperature at around
160 ◦ C. Unlike tannin–phenolic resins, the curing temperature
of 100 PF only starts at around 145 ◦ C. Therefore, the findings
suggesting that addition of tannin fasten the curing rate of
PF resin. Corresponding to gelation and curing temperature,
addition of tannin shows greater maximum rigidity (E max)
at 200 ◦ C when compared to pure PF. Value of E max for 40
TPF is significantly higher compared to 100PF by 68%. This
value could be attributed to higher degree of crosslinking due
to tannin larger molecular structure.
4. Conclusions
Tannin were extracted from A. mearnsii bark in this study. The
extracted tannin was copolymerized with PF to form a bonding
7000 j m a t e r r e s t e c h n o l . 2 0 2 0;9(4):6994–7001
agent for plywood. Based on the findings obtained, the addi-
tion of tannin into PF prepolymer was found to be workable
at a low temperature (50 ◦ C) under a high-speed shear mix-
ing. In comparison to pure PF, the partial substitution of PF to
tannin resulted in almost comparable bonding strength, faster
curing and a more thermally stable resin. The synthesized TPF
revealed more or less similar physical properties to that of pure
PF. PF with the highest tannin substitution (40 TPF) has the
lowest gel time and solid content but highest viscosity than
pure PF. The TPF was proved to be suitable in plywood fabrica-
tion. Although the shear strength of the TPF-bonded plywood
was lower than that of the PF-bonded plywood, it surpassed
the minimum requirement stipulated in the EN314 standard.
PF with 30% tannin (30 TPF) substitution has the lowest cur-
ing temperature of around 125 ◦ C compared to 145 ◦ C in PF. In
addition, TPF with 30% tannin substitution also has the high-
est thermal stability where the residue at 600 ◦ C was recorded
as 43.93%. Therefore, it can be concluded that PF with 30% tan-
nin substitution (30 TPF) performed relatively better than the
pure PF, 20 TPF and 40 TPF in terms of having the fastest curing
rate, slowest degradation rate, strongest maximum rigidity as
well as high shear bond strength.
Acknowledgements
This study was financially supported by the Ministry of Higher
Education Malaysia (grant HiCoE 6369109) and Aerospace
Malaysia Innovation Centre (grant 6300142).
references
[1] Hoong YB, Paridah MT, Loh YF, Jalaluddin H, Chuah LA. A
new source of natural adhesive: Acacia mangium bark
extracts co-polymerized with phenol–formaldehyde (PF) for
bonding Mempisang (Annonaceae spp.) veneers. Int J
Adhesion Adhesives 2011;31(3):164–7.
[2] Elbadawi M, Osman Z, Paridah T, Nasroun T, Kantiner W.
Mechanical and physical properties of particleboards made
from Ailanthus wood and UF resin fortified by Acacias
tannins blend. J Mater Environ Sci 2015;6(4):1016–21.
[3] Gonultas O. Properties of pine bark tannin-based adhesive
produced with various hardeners. BioResources
2018;13(4):9066–78.
[4] Moubarik A, Pizzi A, Allal A, Charrier F, Charrier B.
Cornstarch and tannin in phenol–formaldehyde resins for
plywood production. Indus Crops Prod 2009;30(2):188–93.
[5] Pizzi A. Tannins: prospectives and actual industrial
applications. Biomolecules 2019;9(8):344.
[6] Pizzi A, Mittal KL, editors. Handbook of adhesive technology.
Boca Raton: CRC Press; 2017.
[7] Pizzi A. Recent developments in eco-efficient bio-based
adhesives for wood bonding: opportunities and issues. J
Adhesion Sci Technol 2006;20(8):829–46.
[8] Zhou X, Du G. Applications of tannin resin adhesives in the
wood industry. In: Aires A, editor. Tannins-structural
properties, biological properties and current knowledge.
London: IntechOpen; 2020.
[9] Garcia DE, Glasser WG, Pizzi A, Lacoste C, Laborie MP.
Polyphenolic resins prepared with maritime pine bark
tannin and bulky-aldehydes. Indus Crops Prod 2014;62:84–93.
[10] Santos AF, Luz LA, Napoleão TH, Paiva PM, Coelho LC.
Coagulation, flocculation, agglutination and
hemagglutination: similar properties? Adv Chem Res
2014;20:51–70.
[11] Nor Hafizah AW, Paridah MT, Nor Yuziah MY, Zaidon A, Choo
ACY, Nor Azowa I. Influence of resin molecular weight on
curing and thermal degradation of plywood made from
phenolic prepreg palm veneers. J Adhesion 2014;90:210–29.
[12] Dang XC, Nguyen XH, Dinh HD, Le Thi MT, Nguyen VT,
Hoang TH, et al. Tannins: extraction from plants. In: Aires A,
editor. Tannins-structural properties, biological properties
and current knowledge. London: IntechOpen; 2020.
[13] Altemimi A, Lakhssassi N, Baharlouei A, Watson DG,
Lightfoot DA. Phytochemicals: extraction, isolation, and
identification of bioactive compounds from plant extracts.
Plants 2017;6(4):42.
[14] Jahanshaei S, Tabarsa T, Asghari J. Eco-friendly
tannin–phenol formaldehyde resin for producing wood
composites. Pigment Resin Technol 2012;41(5):296–301.
[15] Bertaud F, Tapin-Lingua S, Pizzi AN, Navarrete PA, Petit-Conil
MI. Development of green adhesives for fibreboard
manufacturing, using tannins and lignin from pulp mill
residues. Cell Chem Technol 2012;46(7-8):449–55.
[16] Matsumae T, Horito M, Kurushima N, Yazaki Y. Development
of bark-based adhesives for plywood: utilization of flavonoid
compounds from bark and wood. II. J Wood Sci 2019;65(1):9.
[17] Abdullah UH, Pizzi A, Rode K, Delmotte L, Zhou X, Mansouri
HR. Mimosa tannin resins for impregnated paper overlays.
Eur J Wood Wood Prod 2013;71(2):153–62.
[18] Garcia DE, Glasser WG, Pizzi A, Paczkowski SP, Laborie MP.
Modificationof condensed tannins: from polyphenol
chemistry to materials engineering. N J Chem 2016;40:
36–49.
[19] Zhu J, Abhyankar H, Nassiopoulos E, Njuguna J. Tannin-based
flax fibre reinforced composites for structural applications in
vehicles. IOP Conf Ser: Mater Sci Eng 2012;40(1):
012030.
[20] Liao J, Brosse N, Pizzi A, Hoppe S. Dynamically cross-linked
tannin as a reinforcement of polypropylene and UV
protection properties. Polymers 2019;11(1):102.
[21] Cardona F. Effect of tannin on flexural properties of phenol
formaldehyde glycerol reinforced composites: prelieminary
results. J Reinf Plast Compos 2010;29(15):3543–53.
[22] Mosiewicki M, Aranguren MI, Borrajo J. Thermal and
mechanical properties of woodflour/tannin adhesive
composites. J Appl Polym Sci 2004;91(5):3074–82.
[23] Kim S, Kim HJ. Curing behavior and viscoelastic properties of
pine and wattle tannin-based adhesives studied by dynamic
mechanical thermal analysis and FT-IR-ATR spectroscopy. J
Adhesion Sci Technol 2003;17(10):1369–83.
[24] Lisperguer J, Saravia Y, Vergara E. Structure and thermal
behavior of tannins from Acacia dealbata bark and their
reactivity toward formaldehyde. J Chilean Chem Soc
2016;61(4):3188–90.
[25] Vázquez G, López-Suevos F, González-Alvarez J, Antorrena G.
Curing process of phenol–urea–formaldehyde–tannin (PUFT)
adhesives. J Thermal Anal Calorimet 2005;82(1):143–9.
[26] Chupin L, Charrier B, Pizzi A, Perdomo A, Charrier-El
Bouhtoury F. Study of thermal durability properties of
tannin–lignosulfonate adhesives. J Thermal Anal Calorimet
2015;119(3):1577–85.
[27] Barbosa V Jr, Ramires EC, Razera IA, Frollini E. Biobased
composites from tannin–phenolic polymers reinforced with
coir fibers. Indus Crops Prod 2010;32(3):305–12.
[28] Ramires EC, Frollini E. Tannin–phenolic resins: synthesis,
characterization, and application as matrix in biobased
composites reinforced with sisal fibers. Compos Part B: Eng
2012;43(7):2851–60.
[29] Lei H, Frazier CE. A dynamic mechanical analysis method for
predicting the curing behavior of phenol–formaldehyde resin
adhesive. J Adhesion Sci Technol 2015;29(10):981–90.
 j m a t e r r e s t e c h n o l . 2 0 2 0;9(4):6994–7001
[30] Carvalho AG, Zanuncio AJ, Mori FA, Mendes RF, da Silva MG,
Mendes LM. Tannin adhesive from Stryphnodendron
adstringens (Mart.) Coville in plywood panels. BioResources
2014;9(2):2659–70.
[31] Zhang W, Yang X, Li C, Liang M, Lu C, Deng Y.
Mechanochemical activation of cellulose and its
thermoplastic polyvinyl alcohol ecocomposites with
enhanced physicochemical properties. Carbohyd Polym
2011;83(1):257–63.
[32] Konai N, Raidandi D, Pizzi A, Girods P, Lagel MC, Kple M.
Thermogravimetric analysis of anningre tannin resin.
Maderas. Ciencia y tecnología 2016;18(2):245–52.
7001
[33] Papadopoulou E, Chrissafis K. Thermal study of
phenol–formaldehyde resin modified with cashew nut shell
liquid. Thermochim Acta 2011;512(1–2):105–9.
[34] Khan MA, Ashraf SM. Studies on thermal characterization of
lignin. J Thermal Anal Calorimet 2007;89(3):993.
[35] Kim S. Environment-friendly adhesives for surface bonding
of wood-based flooring using natural tannin to reduce
formaldehyde and TVOC emission. Bioresour Technol
2009;100(2):744–8.