Thermomechanical fractionation effect on mechanical behaviour of biomaterial based composites

Abstract

The necessity of natural resources protection impose to every country the use of renewable resources based alternatives as annual plants, agricultural and agro-industrial residues for human needs. The aim of this study is to obtain with local biomaterials, efficient flat composites which can be used as substitute of sawlog in furnishing and building works. Two categories of plant material were used: the fibrous material constitutes of core of Hibiscus cannabinus and the binding material from tannin based (Parkia biglobosa and Pithecelobium dulce) and from mucilage based (Bombax costatum and Grewia venusta). The parameters taken into account in the particleboard formulations were the physicochemical treatment of plant materials and the nature of the binders. The interest of this work is in the exploration of new sources of plants likely to be used as binders in the composites sector. Mechanical properties of the panels obtained were measured according to European standards. The interesting results obtained varied with the two types of variables and were in conformity with the American standards.

Full text

Open Archive TOULOUSE Archive Ouverte (OATAO)
OATAO is an open access repository that collects the work of Toulouse researchers and
makes it freely available over the web where possible.
This is an author-deposited version published in : http://oatao.univ-
toulouse.fr/
Eprints ID : 10595
To cite this version : Nenonene, Amen and Sanda, K. and Evon,
Philippe and Rigal, Luc Thermomechanical fractionation effect on
mechanical behaviour of biomaterial based composites. (2007) In: 15th
European Biomass Conference & Exhibition, 07 May 2007 - 11 May
2007 (Berlin, Germany)
Any correspondance concerning this service should be sent to the repository
administrator: staff-oatao@listes-diff.inp-toulouse.fr

THERMOCHEMICAL FRACTIONATION EFFECT ON MECHANICAL BEHAVIOUR OF BIOMATERIAL
BASED COMPOSITES
A.Y. Nenonene, K. Sanda
Unité de Recherche sur les Matériaux et les Agroressources, Ecole Supérieure d’Agronomie, Université de Lomé, B.P. 1515,
Lomé Togo
P. Evon, L. Rigal
Laboratoire de Chimie Agro-Industrielle, UMR 1010, INRA/INP-ENSIACET, 118, Route de Narbonne, 31077 Toulouse
Cedex, France.
Email: amen.nenonene@gmail.com ; komsanda@hotmail.com
; Philippe.Evon@ensiacet.fr
;
Luc.Rigal@ensiacet.fr
ABSTRACT: The necessity of natural resources protection impose to every country the use of renewable resources
based alternatives as annual plants, agricultural and agro-industrial residues for human needs. The aim of this study is
to obtain with local biomaterials, efficient flat composites which can be used as substitute of sawlog in furnishing and
building works. Two categories of plant material were used: the fibrous material constitutes of core of Hibiscus
cannabinus and the binding material from tannin based (Parkia biglobosa and Pithecelobium dulce) and from
mucilage based (Bombax costatum and Grewia venusta). The parameters taken into account in the particleboard
formulations were the physicochemical treatment of plant materials and the nature of the binders. The interest of this
work is in the exploration of new sources of plants likely to be used as binders in the composites sector. Mechanical
properties of the panels obtained were measured according to European standards. The interesting results obtained
varied with the two types of variables and were in conformity with the American standards.
Keywords: Kenaf, Extrusion, Particleboard
1. INTRODUCTION
In the world, deforestation continues at alarming
intervals (13 millions hectares per year between 2000 and
2005). But efforts are made to reduce significantly the
loss of forest surface. However, Africa and South
America remain the continents where the loss of forests is
the widest. According to FAO reports, in 1993 the
surfaces covered by dense forests were of 1400 km2 and
those of clear forests and raised savannah were 11 950
km2. But, in 1970, these areas were not any more larger
than 5 040 km2 (that is an annual rate of deforestation of
180 km2).
A little less than 20 % of annual exploitation of
tropical forests is used for industrial wood (FAO, 1999).
In sub-Saharan Africa, wood constitutes the first source
of energy and building material. It is thus necessary to
turn to alternatives such as the various composites and
panels based on the use of agricultural and agro-industrial
by-products and annual plants in order to reduce human
pressure on natural forest (Gillah et al., 2000, Mohanty et
al., 2002, Mazumder et al., 2005). But, unfortunately
these types of alternatives are lesser-known in developing
countries namely those of Africa.
The main goal of these studies is to arouse the
interest of the populations and the decision makers for the
fibre and particle boards and to explore the possibilities
of obtaining competitive panels which implementation
will be adapted at the technological level of the under-
developed countries.
But the specific aim of this work is to explore how
thermo-chemical treatment-combined with chemical
attack of the raw matter could affect the mechanical
behaviour of kenaf and natural binders based
particleboards.
2. OBJECTIVES
The scientific questions we attempted to answer
specifically are the following:
• are tannin containing organs from Parkia biglobosa
(Jacq.) Benth. pod and from Pithecelobium dulce
Benth. stem bark and mucilage containing organs
such as calyx of Bombax costatum Pellegr. & Vuillet
and stem bark of Grewia venusta Fresen, enough
adhesive to bind kenaf core particles for the
manufacture of low density particleboards with
bending mechanical properties meeting relevant
standard requirements?
• can twin-screws extrusion of kenaf core with
addition of chemical reagent, improve the
mechanical characteristics of the panels obtained?
3. EXPERIMENTAL METHODS
3.1. Materials
3.1.1 Fibrous material
Particles used for the particleboards manufacturing
were obtained from cultivation experiments of Hibiscus
cannabinus L. (Kenaf), a Malvaceae belonging species
which was carried out from May to August 2005 at the
“Station d’Exprérimentation Agropédagogique” (SEA) of
“Ecole Supérieure d’Agronomie, Université de Lomé” in
Togo.
3.1.2. Tannin containing materials
Two types of plant material have been used as
tannin containing materials in this work. The pod the fruit
of Parkia Biglobosa and the stem bark of Pithecelobium
dulce, both population protected legume species growing

wild in tropical savannah area. These plant organs were
harvested in April 2005 at SEA.
3.1.3. Mucilage containing materials
The calyx of Bombax costatum and the stem bark of
Grewia venusta were the mucilaginous materials used in
this work. The two species grow wild in the Togolese
savannah. The stem bark of G. venusta was collected
from young plants in southern area of Togo in March
2005 and sun dried calyx of B. costatum were from
northern Togo and bought in a local market also in March
2005.
3.2. Methodologies
3.2.1. Preparation of the bioadhesives
The binding vegetable materials were oven dried (70
°C; 48 hours), finely crushed using a RETSCH SM 100
type crusher equipped with a sieve of 2.5 mm mesh, and
resulting powder kept dry at 50 °C in an oven until use.
3.2.2. Particles fractionation
3.2.2.1. Mechanical fractionation
The core of H. cannabinus were oven-dried (70 °C,
48 hours) and milled into particles using the above
crusher equipped with a 5 mm mesh sieve, which gave 67
% of particles and fibres with diameter varying from 0.25
to 1.60 mm, length from 1 to 7 mm for particles and
length ranging from 5 to 19 mm for fibres.
3.2.2.2. Thermo-mechanical and chemical fractionation
Thermo-mechanical fractionation combined with
chemical treatment, responsible of material’s internal
structural modification was carried out in a CLEXTRAL
BC 45 twin screws extruder. Before extrusion process,
the kenaf core was firstly cut out in particles of relatively
large size using an Electra type hammer mill. The coarse
particles of kenaf core were introduced via a dosing
hopper placed at the top of the first element of the
cylinder which delivers the matter inside the sleeve. One
per cent soda solution was continually added into the
sleeve at the place desired using a DKM K20-2/PP/16
pump with a flow regulated to have a liquid / solid ratio
of one (1). The binder consisting of the crude vegetable
organs powder was also added or not added at the place
desired with a flow regulated to have a fibre - binder
proportion of 1-9. The temperature in the die was 60 °C.
Resulting product was oven dried at 70 °C for 48 hours
and kept dry at 50 °C in oven until use.
3.2.3. Elaboration of the particleboards
Two parameters affecting the mechanical properties
of the panels were concerned in formulating the
particleboards: the origin of natural binder (P. biglobosa,
P. dulce, B. costatum or G. venusta) and the method of
particles treatment (mechanical fractionation or
extrusion).
3.2.3.1. Preparation of the particleboard mat
In case of their use, the binders were incorporated at
the rate of 10 % of fibre dry matter. For all types of
panel, the H. cannabinus particles were mixed with the
binder as appropriate according to the desired
formulation of the panel and on the basis of 150 g initial
dry matter for each particleboard. A total of 40 g of water
were used for humidification.
For the extrusion fractionation based type of panels,
the binders were mixed to the particles during the
extrusion process. This homogenous mixture was also
mixed with 50 g of water.
In both cases, the mixture was done manually for 2
min then mechanically for 10 min using a PERRIER 721
mixer.
3.2.3.2. Thermo-pressing of the panels
The mat was introduced in an aluminium mould (27
cm x 27 cm x 5 cm) electrically preheated at 180 °C. The
mould was then closed and placed between two heating
plates of a manual hydraulic thermopress of CARVER
type (maximum pressure: 11 tons.m
-2
) and the pressure
was gradually applied as follows: 5 tons.m
-2
for 60
seconds and then 10 tons.m
-2
for 4 minutes. The panel
was then removed from the mould and weighted after
cooling with ambient air.
3.2.4. Mechanical testing of the particleboards
Six specimens of 150 mm x 50 mm and 6 specimens
of 50 mm x 50 cm were cut from each particleboard
according to the NF-EN 326-1 (1993) standard and
conditioned at 20 °C , 65 % relative humidity for 14
days before testing. The 150 x 50 specimens were used
for bending testing according to NF-EN 310 (1993)
standard and the 50 x 50 specimens for internal bond
strength testing on in accordance with NF-EN 319,
(1993) standard requirement. For both mechanical tests, a
JFC type H5KT testing machine was used.
4. RESULTS AND DISCUSSION
Panel average density was 435.62 ± 21.85 kg.m
-3
.
This type panel could be classified as low density
particleboards (NF-EN 309, 2005) intended for a general
use in dry medium. It can be used for insulation, ceiling
and for wall coating. These panels density can be
compared to those obtained by Sellers et al., 1993. These
authors reported panel density ranging from 300 to 500
kg.m
-3
for kenaf core particleboards manufactured
respectively with Urea Formaldehyde (UF), Phenol
Formaldehyde (PF) and Polymeric Diphenylmethane
Diisocyanate (PMDI) resins.
4.1. Effect of the source of the natural binder on the
panels mechanical properties
Comparing the binders, the panels can be classified
in 3 categories (panels without binders added, panels
containing mucilaginous binder and panels with tannic
binder). Whatever the method of fractionation used, the
modulus of bending elasticity (MOE) was more
important for the panels containing a tannic binders than
those containing mucilaginous binders whereas the latter
showed a similar effect as adhesiveness panels (table 1).
The tannic binders showed the same prevalence on
the bending strength (MOR) and on the internal bond
strength (IB). But, for these two parameters,
mucilaginous binders induced an improvement
comparing with the results obtained on Binderless panels

Table I: Influence of the nature of the natural binder on
the mechanical properties of the panel
Mechanical
properties MOE (MPa)
MOR
(MPa) IB (MPa)
Treatment N-
Ext. Ext. N-
Ext. Ext. N-
Ext. Ext.
Binderless
Panel 344 1854 1.72 10.12 0.11 0.28
Tannin
based
Panels 433
2598
2.59
17.28
0.15
0.78
Mucilage
based
panels 353
1885
2.54
10.61
0.10
0.47
MOE: Modulus of bending elasticity; MOR: Bending
strength; IB: Internal bond strength; N-Ext.: Non
extruded matter; Ext. Extruded matter.
namely on the IB of matter extruded ones. Several
authors reported that tannin based binder improves
particleboard mechanical properties (Sowunmi et al.,
1996, Garnier et al., 2001, Bisanda et al., 2002, Li et al.,
2004) relating to the improving effect of tannin based
binder (containing hardener or not) on the particleboards
mechanical characteristics. The mucilaginous binders did
not seem to influence the panels bending properties
compared to that of panels without binder.
4.2. Effect of the fractionation methods on the panels
mechanical properties
In first global analysis, results obtained showed that
the extrusion of the basic materials of the particleboards
elaborated in this work improved in a largely significant
way the mechanical properties of the panels.
4.2.1. Fractionation effect on the panel modulus of
bending elasticity
The treatment of the matter (fibre and binder plant
organs) by extrusion in the presence of a solution of soda
and at 60 °C and under pressure constraints, increased the
modulus of bending elasticity (fig. 1) from 4 to 7 times in
comparison with the values obtained on the panels made
with kenaf core treated by single mechanical crushing.
0
500
1 000
1 500
2 000
2 500
3 000
3 500
M O E ( M Pa)
Non
binder P. dulce P.
biglobosa B.
costatum G.
venusta
BINDER
Extruded
Non Extruded
Figure 1: Effect of the fractionation process on the panel
modulus of bending elasticity
For the panels based on single mechanical ground
material, none average MOE values obtained (344; 463;
403; 382 and 323 MPa respectively for binderless, P.
dulce, P. biglobosa, B. costatum and G. venusta based
panels) had been able to reach the required value by
American standard ANSI, 208.1, 1999 (55O MPa) for
this parameter. This situation can be explained by a low
effectiveness of the mucilaginous material taken in a
rough state. The use of these plants extracts would allow
better appreciation of real adhesive effect of pectins on
the panel's mechanical behaviour.
Conversely panels obtained with the extruded
matter had largely exceeded (1,854; 2 041; 3,155; 1,439;
2,331 MPa respectively) this standard requirement. The
best MOE value was reached with the panel containing
the extruded mixture of kenaf core particles and P.
biglobosa fruit pod. It clearly appeared that the extrusion
treatment coupling with soda attack of the particles
improved highly the panel’s modulus of bending
elasticity.
These results could be compared to that of Webber
et al., (2000) with kenaf-MDI particleboard (1731 MPa).
But, with lower density panels, Sellers et al., (1995) got
MOE value lesser than ours (125 MPa) with conventional
particleboard resins (UF, PF, PMDI). As for Xu et al.,
(2003), MOE value (2,500 MPa) obtained with
binderless kenaf particleboards using steam injection
pressing process was higher than ours (1 854 MPa) on
panels without binder made with extruded raw matter.
4.2.2. Fractionation effect on the panels bending strength
The effect of extrusion is also very remarkable on
the bending strength (Fig. 2) and the increase in this
parameter was about 5 times the values measured on the
panels with non extruded matter. Indeed, the MOR values
obtained on the non extruded raw matter panels were
very low and ranging from 1.72 to 3.17 MPa. As for the
extruded raw matter panels, the MOR values varied from
8.07 for binderless panels to 23.02 MPa for the panel
containing tannin crude organ from Parkia biglobosa.
The average value of bending strength of extruded raw
matter panels was 13.18 MPa. Except for the panels with
P. biglobosa, all the values of MOR obtained with the
other binders on the panels containing non-extruded
matter are lower than that required by the standard (3
MPa). All the value of the extruded raw matter panels
exceeded largely the ANSI 208.1 required value. These
results namely extruded matter panels ones were closed
to those obtained Webber et al., (2000) (7.1 – 19.3 MPa)
and Xu et al., (2003) (12.6 MPa).
0
5
10
15
20
25
M OR (MP a)
Non
binder P. dulce P.
biglobosa B.
costatum G. venusta
BINDER
Extruded Non Extruded
Figure 2: Effect of the fractionation process on the panel
bending strength
4.2.3. Fractionation effect on the panels internal bond
strength
The internal bond strength of the panels also
reflected the prevalence of the extrusion on the treatment
by single mechanical crushing (Fig. 3). However, the
ratio is less significant than that raised with the first two

parameters. But, P. dulce gave the best IB characteristic
and close to the value induced by P. biglobosa. The
panels with tannic adhesives gave better IB value than
those with mucilaginous binder.
0
0,2
0,4
0,6
0,8
Inte rnal B ond ( MP a)
Non binder P. dulce P.
biglobosa B.
costatum G. venusta
BINDER
Extruded Non Extruded
Figure 3: Effect of the fractionation process on the panel
internal bond strength
The results obtained for IB varied from 0.12 to 0.28
MPa with 0.15 MPa as average value for the non
extruded matter based panels and from 0.11 to 0.74 MPa
with 0.52 MPa as average value for the extruded ones.
All of the panels showed an internal bond strength value,
higher than the American standard requirement (0.1
MPa). The internal bond strength values of non extruded
matter panels were similar to that of Sellers et al., (1995):
0.20 MPa) on low density panels made with kenaf core
and conventional particleboard resin (UF, PF and PMDI).
The panels made with extruded raw matter gave IB value
closed to those reported by several authors (Webber et
al., 2000; Xu et al., 2003).
CONCLUSION
It can be concluded that whatever the nature of the
binder, high-performance particleboard can be
manufactured with kenaf core by using extrusion
treatment combined with soda attack of the raw matter.
The tannic binders conferred to the panels more
interesting mechanical properties than the mucilaginous
ones. In the manufacturing of the particleboards, the
tannic binders can be valid alternatives to the synthetic
binders often more expensive and especially more
prejudicial with environmental and human health. But the
tannic material’s effect has to be improved by using pure
material such as extract or chemical modification of the
matter or adding a hardener.
The mucilaginous binders did not seem to influence
some mechanical properties of the panels namely bending
parameters.
REFERENCES
[1] ANSI A208.1, Particleboard. American National
Standard Institute. National Particleboard
Association. Gaithersburg. MD. (1999) 9p.
[2] E. T. N. Bisanda, W. O. Ogola, J. V. Tesha,
Characterisation of tannin resin blends for
particleboard applications. Cement and Concrete
Composites, (2003). 25 (6), 593-598.
[3] FAO, Situation des forêts dans le monde. Rome,
(1999). 154 p.
[4] S. Garnier, Z. Huang, A. Pizzi, Dry I.B. forescasting
of commercial tannin adhesives-bonded
particleboard by MTA bending. Holz als Roh- und
Werkstoff (2001) 59, 46.
[5] P.R.. Gillah, M.A. Irle, S.A. Amartey, Development
and production of laboratory scale novel MDF
panels from composite and nonwoven mattresses of
sisal and wood fibre mixtures. Holz als Roh- und
Werkstoff (2000) 58: 324 - 330.
[6] K. Li, X. Geng, J. Simonsen, J. Karchesy, Novel
wood adhesives from condensed tannins and
polyethylenimine. International Journal of Adhesion
& Adhesives (2004). 24: 327–333
[7] B.B. Mazumder, A. Nakgawa-izumi, K. Kuroda, Y.
Ohtani, K. Sameshima, Evaluation of havesting
times on kénaf bast lignin by pyrolysis-gas
chromatography. Industrial Crops and Products
(2005) 21: 17-24.
[8] A. K. Mohanty, M. Misra, L. T. Drzal1, Sustainable
Bio-Composites from Renewable Resources:
Opportunities and Challenges in the Green Materials
World. J. Polym. Environ. (2002) 10 (1/2): 19 - 26.
[9] T.J. Sellers, G.D. Miller, M.J. Fuller, Kenaf core as a
board raw material. Forest Product Journal (1993)
43 (7/8): 69-71
[10] T. J. Sellers, G. D., Miller, M. J. Fuller,
Lignocellulosic-Based Composites Made of Core
from Kenaf, an Annual Agricultural Crop. Abstarct,
IUFRO XX World Congress - Caring for the Forest:
Research in a Changing World, 6-12 August 1995,
Tampere, Finland. Improving Our Understanding of
Processes and Products. (1995) Abstract.
[11] S. Sowunmi, R.0. Ebewele, A.H. Conner, H. River,
Fortified Mangrove Tannin-Based Plywood
Adhesive. J. Appl. Polym. Sci. (1996) 62: 577 - 584
[12] C.L. Webber, R.E. Bledsoe, V.K. Bledsoe, C.D.
Ray, J.L. Blalock, Industrial Grade Particleboard.
Kenaf Association International Conference
Proceedings. 2000. (2000) P. 38. (Abstract)
[13] J. Xu, G. Han, E.D. Wong, S. Kawai, Development
of binderless particleboard from kenaf core using
steam-injection pressing. J. Wood Sci. (2003) 49 (4)
327-332.