Wood–plastic composites from urban waste as
Wood–plastic composites from urban waste as potential building materials
Internacional de Patrimonio y Desarrollo Sustentable, Eje temático:
Patrimonio Natural y Desarrollo de Productos Alternativos,
Campeche, Campeche, Mexico, 3–8 December. Universidad
Autónoma de Campeche, Dirección General de Estudios de Posgrado e Investigación.
Cui, Y., Lee, S., Noruziaan, B., Cheung, M. & Tao, J. (2008) Fabrication
and interfacial modification of wood/recycled plastic composite
materials. Composites: Part A, 39, 655–661.
Eboatu, A.N., Akpuaka, M.U., Ezenweke, L.O. & Afiukwa, J.N. (2003)
Use of some plant wastes as fillers for polypropylene. Journal of
Applied Polymer Science, 90, 1447–1452.
Jacobsen, W.W. (2003) Relleno de fibras lignocelulósicas para composiciones de compuestos termoplásticos, Mexican Patent No. MX
Jiménez, I. (2005) La fuerza de la madera en los plásticos: Una combinación ganadora (The strength of wood in plastics: A winning
combination). Tecnología del Plástico, 20, 24–29.
Lu, J.Z., Wu, Q. & Negulescu, I.I. (2004) Wood-fiber/high-density-polyethylene composites: compounding process. Journal of Applied Polymer Science, 93, 2570–2578.
Maine, F.W. (2004) Woodfiber/Plastics Composites – Introduction and
Overview. Society of Plastics Engineers online presentation. SPE
training and e-learning (2004): <https://members.4spe.org/eLearning/courselist.cfm> [accessed 5 July 2005].
Martínez-Domínguez, O. (2008) Obtención y caracterización de materiales compuestos a base de desechos de madera y polietileno de alta
densidad (Preparation and characterization of composite materials
from wood wastes and high density polyethylene). BSc Thesis.
Instituto Tecnológico de Mérida. Merida, Mexico.
McDonald, A.G., Gallagher, L.W. & Sundar, S.T. (2005) The effect of
wood surface modification and particle size on wood-plastic composite performance. In: Proceedings of the 8th International Conference on Wood fiber-Plastic Composites, Madison, WI, May 23–25,
pp. 163–171. Forest Products Society, 2801 Marshall Ct., Madison,
WI 53705-2295, USA.
Mehta, G., Mohanty, A.K., Thayer, K., Misra, M. & Drzal, L.T. (2005)
Novel biocomposites sheet molding compounds for low cost housing panel applications. Journal of Polymers and the Environment, 13,
Mijangos, D.A. (2004) Comuna aumentará subsidio a recolectoras. Periódico Por Esto!, Sección Ciudad, Calle 60, No. 576 x 73, Centro, C.P.
97000, Mérida, Yucatán, México (2004): <http://www.poresto.net/
index.php?tim=22-10-2004&id=16044> [accessed 22 October 2004].
Ortega-Leyva, M.N. (2008) Compuestos de plástico y madera: qué debe
saber (Composites form plastic and wood: What do you have to
know). Tecnología del Plástico, 23, 23–28.
Pritchard, G. (2004) Two technologies merge: wood plastic composites,
Plastics Additives and Compounding, 6, 18–21.
Puppin, G. (2003) Método de extrusión de perfil compuesto, de resina y
fibra de madera, Mexican Patent No.MX 213391.
Rangaprasad, R. & Vasudeo, Y.B. (2003) Wood plastic composites. Popular Plastics and Packaging, 48, 93–100.
Seethamraju, K.V. & Deaner, M.J. (2001) Resina de ingeniería avanzada y
compuesto de fibra de madera, Mexican Patent No. MX 203754.
Stark, N.M. & Berger, M.J. (1997) Effect of particle size on properties of
wood-flour reinforced polypropylene composites. In: Proceedings of
The Fourth International Conference on Woodfiber-Plastic Composites,
Madison, Wisconsin, May 12–14, 1997, pp. 134–143. Forest Products
Society 2801 Marshall Court, Madison, WI 53705-2295, USA.
Steward, R. (2007) Wood fiber composites: fierce competition drives
advances in equipment, materials and processes. Plastics Engineering,
Stokke, D.D. & Gardner, D.J. (2003) Fundamental aspects of wood as a
component of thermoplastic composites. Journal of Vinyl and Additive Technology, 9, 96–104.
Treffler, B. (2007) La funcionalización de un material natural: El material compuesto alternativo WPC conquista el mundo de la madera y
del plástico (The functionalising of a natural material: The alternative WPC composite conquers the wood and plastic world). Revista
de Plásticos Modernos, 93, 100–102.
R. H. Cruz-Estrada et al.
the problem. Most importantly, both the processor and the
generator should quickly understand and agree on the problem’s source to avoid future occurrences.
A working understanding of, and ability to manage, common wood-waste contaminants enables wood-waste processors to address the challenge of meeting end-user quality
specifications better. This basic knowledge also leads to intelligent decisions concerning raw-material sourcing, woodwaste processing, and end-product marketing.
With respect to the plastic waste, similar actions may be
implemented in order to separate and classify them by plastic
type, looking at the identification code if they have one. It
may be useful to classify each type of plastic object by different colours because that would give more chance of having
loads of the same kind. Regarding bottles and containers, the
best option for cleaning operations may be automation.
It should be mentioned that the practices described above
are merely recommendations. Other procedures may be
required. Again, as mention before, how advanced the technology should be, must be the result of rigorous cost–benefit
As a result of this preliminary study, it was possible to prepare WPCs using virgin and recycled HDPE, and wood fibres
originating from the pruning of Merida’s trees, subjecting the
raw materials to minimum previous treatment.
Under impact, the composites performed better when the
lowest amount of wood was used in their preparation. This
may be due to the fact that increasing the wood content in the
composites implies a decrease of the thermoplastic matrix
content, which absorbs the impact energy more efficiently
than the wood. Using R-HDPE for the composite preparation caused a slight decrease in their impact resistance in
comparison with the use of V-HDPE, a fact that is positive
from the economical and environmental points of view. On
the other hand, diminishing the wood particle size resulted in
decay of the composite performance under impact because
the smaller the wood particles were, the bigger the particle
surface area that the polymer matrix had to wet in order to
have strong wood–polymer interphases.
The composite tensile strength decreased as the content of
wood was increased. This happened because the wood particles were acting merely as filler instead of reinforcement. In
this regard, one has to consider that the wood fibres elongation to fracture is lower than that of the polymer matrix at
comparable imposed tensile loads. Therefore, increasing the
wood content in the composites increases also their rigidity,
which makes them less capable of withstanding high levels of
load before they fracture. The effect of wood particle size on
the tensile strength was, as expected, only for composites prepared with V-HDPE. This increased as the amount of smaller
wood particles in them increased because, for a same weight
fraction, a smaller wood particle resulted in larger surfaces
areas that provided better tensile load transfer between the
wood and the polymer matrix. Using R-HDPE for the composite preparation yielded contradictory results as the tensile
strength decreased by increasing the amount of smaller wood
particles in the composites, perhaps because of the fact that
the polymer matrix used for the composite preparation was
actually a polymer blend, whose homogeneity was not guaranteed, contrasting with the known commercial grade HDPE
pure resin. Nevertheless, it was remarkable that the impact
strength of some composites prepared using V-HDPE was
higher than that of the pure V-HDPE.
Flat R-HDPE-based WPCs extrudates obtained at the
laboratory level had reasonably good appearance and looked
well moulded. Also, their Young modulus was higher than that
of the R-HDPE itself. Further experimentation is planned to
optimise the process in order to find adequate conditions to
scale up the process to a pilot level to shape these composites into materials with potential structural applications (e.g.
undulated sheets for roofing).
The authors want to thank to the Mexican Council for Science and Technology and to the Government of the Yucatan
State for the financial support granted to carry out this study
through the projects YUC-2003-C02-021 and YUC-2008-C06107327 (“Fondo Mixto CONACyT-Gobierno del Estado de
Yucatán”). In the same way, gratitude is expressed to Merida’s City Council for the assistance provided.
Bledzki, A.K. & Faruk, O. (2004) Wood fiber reinforced polypropylene
composites: compression and injection molding process. PolymerPlastics Technology and Engineering, 43, 871–888.
Castillo-González, R.E. (2007) M3 de producto de poda de árboles recolectados en el año 2007. E-mail message ([email protected]
gob.mx). Ayuntamiento de Mérida, Dirección de Servicios Públicos
Municipales, Calle 116, No. 277 por Av. Jacinto Canek, Colonia
Bojorquez, C.P. 97203, Mérida, Yucatán, México (20 November 2007).
Correa, C.A., Fonseca, C.N.P, Neves, S., Razzino, C.A. & Hage Jr, E.
(2003) Wood-plastic composites. Polimeros: Ciencia e Tecnologia,
Cruz-Estrada, R.H., Fuentes-Carrillo, P., Martínez-Domínguez, O., Canché-Escamilla, G. & García-Gómez, C. (2006) Obtención de materiales compuestos a base de desechos vegetales y polietileno de alta
densidad (Preparation of composite materials from vegetal wastes
and high density polyethylene). Revista Mexicana de Ingeniería
Química, 5, 29–34.
Cruz-Estrada, R.H., Canché-Escamilla, G., González-Chí, P.I. et al.
(2006a) Utilización de desechos vegetales en la obtención de materiales compuestos termoplásticos. Final Research Report. Fondos
Mixtos CONACyT-Gobierno del Estado de Yucatán, Project No.
Cruz-Estrada, R.H., Canché-Escamilla, G., Herrera-Franco, P.J. et al.
(2007) Wood-plastic composites based on recycled urban materials
as an alternative for roofing. In: Abstracts Book of the International
Symposium on Advanced Biomass Science and Technology for Biobased Products, Beijing, China, 23–25 May, p. 98.
Cruz-Estrada, R.H. & García-Gómez C. (2007) El patrimonio natural y
el reciclaje como alternativa viable para el mejoramiento de la vivienda de Mérida, Yucatán. In: Libro de Resúmenes del I Congreso
Wood–plastic composites from urban waste as potential building materials
Fig. 5: Extrudate of blend 50/R-HDPE/60–100, photographed using different zoom lens.
up the process to evaluate the feasibility of forming these
types of composites into materials with potential structural
applications, with the maximum achievable properties to
substitute poor mechanical performance conventional materials. For instance, we have identified a potential application
in the fabrication of undulated sheets for roofing to substitute those made of cardboard, used by the poorest people in
The Yucatan. These types of WPCs can also be used to manufacture wood-like products such as furniture, fencing, railing, skirting boards, decking, marina walkways, window parts
and doors, roofing shakes, shingles, and many other products. It is important, however, to point out that one of the
objectives of this study was to subject the raw materials to the
minimum possible treatments in order to implement, in
future, a low-cost production process. Now, the presence of
impurities is an issue that really matters during the formulation of WPCs, no matter how advanced is the technology to be
applied. In order to obtain a product with the required properties, the raw materials must be as clean as possible. Impurities may cause discontinuous flow during the extrusion of
plastic blends, occasioning rupture of the extrudate emerging
from the die, affecting the mechanical performance of the
resultant product, among many other properties. Therefore,
a reliable, good-quality waste management process should
be implemented in order to minimise impurities. It may be
debatable how advanced the technology should be. The decision is to be taken by the industrialist, and it should be the
result of rigorous cost–benefit analyses, taking into consideration the specific application for which the final products are
intended, among many other issues. Many procedures could be
followed to lower impurities. For instance, visually identifying
common wood waste contaminants could be a good option.
The problem is that the wood waste can be generated
from various sources. This variability usually causes contami-
nation (e.g. presence of dirt, grit, ferrous metals, plastic, glass,
etc.). To compete with virgin wood materials in higher-value
markets successfully, wood-waste processors must minimise
the presence of contaminants in their end-products. A recommended good practice is that operators become aware of the
common wood-waste contaminants that cause problems.
This awareness is one of the first steps in achieving satisfactory quality control: visually inspect all incoming loads of
material to identify contamination.
Processing facilities should designate inspectors who examine incoming loads before unloading them from trucks. Inspectors should verify that no prohibited contaminants are present.
When inspectors find unacceptable material or excessive
contamination, they should either downgrade or reject the
load and document its contaminants to communicate proper
requirements clearly. Inspectors should quickly become alert
to individual generators that deliver certain types of woodwaste containing more contaminants than others. This awareness and subsequent feedback to problem generators can assist
them in controlling contamination. Only when inspectors have
visually inspected loads, can trucks dump the wood waste onto
receiving decks. Then, processors segregate the waste wood in
the storage yard according to the degree of contamination. The
wood waste may be subsequently processed appropriately.
Plant operators should have some mechanism (manual or
automated) for removing these contaminants from the wood
waste. Typically, mechanisms to remove contaminants include
the following: manual picking, magnetic removal, air-density
separation, screening, or chip washing.
Documentation of problem loads is necessary. It is recommended to carry out the following actions. Use a predetermined sampling method, sampling, and storing key portions
until the problem is fully resolved. Immediately, the generator of the load should be contacted to discuss and arbitrate
R. H. Cruz-Estrada et al.
Fig. 4: Effect of wood waste content and wood particle size on the composite performance under tension. Tensile strength of V-HDPE and RHDPE is shown for comparison purposes only.
loads. Therefore, increasing the wood content in the composites increases also the composite rigidity. Consequently,
they are less capable of withstanding higher levels of load
before they fracture. This is the expected behaviour for composites prepared with relatively small wood fibres, but not
with wood flour.
Regarding the effect of wood particle size, the results
were as expected only for the composites prepared with VHDPE. As can be seen from Figure 4, the tensile strength of
the V-HDPE-based composites increased as the amount of
smaller wood particles in them increased. In this respect, one
has to remember that about 90% of the wood particles of
MESH 10-based composites had sizes ranging from < 2.00
to > 0.149 mm, and that only about 10% was < 0.149 mm
(refer to Tables 1 and 3 for mesh openings and wt% distribution of wood). On the other hand, MESH 60–100-based
composites were prepared with a higher amount of smaller
wood particles (i.e. 0.430 mm > sizes > 0.149 mm), whereas
MESH 100-based composites had the smallest wood particles
(i.e. < 0.149 mm). Therefore, the composite tensile strength
increased because for a same weight fraction, a smaller wood
particle resulted in a larger surface area that provided better
load transfer between the wood and the polymer matrix. It is
believed that the interaction at the wood particle–polymer
matrix interphase was stronger for the smaller wood particles
because a more homogeneous dispersion of finer wood in
the matrix occurred with smaller sizes than that for bigger
ones. These results are in good agreement with those observed
by other researches, for instance Cui et al. (2008).
With respect to the effect of using R-HDPE, contradictory
results were obtained in relation to those obtained for the VHDPE-based composites. From Figure 4, it is easy to see
that the tensile strength decreased by increasing the amount
of smaller wood particles in the composites, perhaps because
of the fact that different types of polymer matrixes were used
for their preparation. In other words, the source of the RHDPE used where different kinds of containers collected
from the Separation Plant, manufactured with different
types of unidentified HDPE grades and additives, and subjected to different unknown processing histories. This means
that the polymer matrix used for the composite preparation
was actually a polymer blend, whose homogeneity was not
guaranteed. This may have negatively affected the composite
tensile mechanical performance, contrasting with the results
obtained for the composites prepared with the known commercial grade HDPE pure resin. Somehow, the MESH 60–
100 and MESH 100-based composites, respectively, were
affected most by using R-HDPE in their preparation. Intriguingly, MESH 10-based composites performed slightly better
than their counterparts prepared with V-HDPE.
Nonetheless, given the disappointing results about the use
of R-HDPE, it is remarkable that the tensile strength of
MESH 100-based composites prepared using V-HDPE was
similar to that of V-HDPE. Also, from Figure 4, it can be
noticed that the composites with 50 wt% of wood had higher
levels of tensile strength than that of V-HDPE.
Extruded wood–plastic sheets
Flat extrudates, approximately 1.80 mm thick and 100 mm
wide were continuously obtained. They were cut into sheets,
approximately 40 cm long for exhibition purposes (Figure 5).
It is remarkable that the Young modulus of both the 50/RHDPE/10 and 50/R-HDPE/60–100 composites (1422 ± 372
MPa and 2865 ± 461 MPa, respectively) was higher than that
of the R-HDPE itself (430 ± 40 MPa). Also, to the naked
eye, the extrudates had reasonably good appearance. The
final product seemed to be well moulded and the production
procedure was reproducible. These are encouraging results
that motivate us to carry out further experimentation to optimize the process in order to find the right conditions to scale
Wood–plastic composites from urban waste as potential building materials
the impact resistance of both types of the ‘wood-free HDPE’
employed (i.e. 42 J m–1 and 44 J m–1 average, for V-HDPE
and R-HDPE, respectively) with the composite 50/V-HDPE/
10 (refer to Table 2 for details on the composite constituents), which had the highest resistance of the composites
prepared (~35 J m–1 average). Although the difference is not
really significant, especially for the composites containing
wood wastes, this may be due to the fact that increasing the
wood content in the composites implies a decrease of the thermoplastic matrix content, which absorbs the impact energy
more efficiently than the wood. In this way, the higher the
wood content in the composites, the less efficient they are at
absorbing impact energy, and the highest susceptibility to fracture resulting in lower levels of impact resistance. Also, the
presence of wood-fibre ends within the body of the composites can cause crack initiation leading to failure. The ends
of wood fibres act as notches, which generate considerable
stress concentrations when loaded that could initiate microcracks in the ductile HDPE matrix. Upon loading, these microcracks coalesce to form a main crack. In addition, the interaction between neighbouring fibres appears to constrain matrix
flow, causing embrittlement, which in the end can lower the
impact resistance of the composites.
Regarding the effect of using R-HDPE on the composite
impact resistance, Figure 3 shows that, in general, the resistance slightly tends to diminish in comparison with the use of
V-HDPE for composite preparation. Consequently, using RHDPE instead of V-HDPE makes little difference regarding
the mechanical performance under impact of the composites
experimented with in this work; a fact that turns out to be
beneficial from economical and environmental points of
view. As mentioned before, the decrease in the R-HDPE
based composite impact resistance was relatively low, the
highest drop being around 17% for the composite 60/RHDPE/100 (~20 J m–1 average) with respect to its counterpart 60/V-HDPE/100 (~24 J m–1 average). The decrease in
the impact resistance with respect to the content of wood
waste in the composites is an issue worth considering. For
this particular study, the lowest decrease for the R-HDPE
based composites was observed for the 50/R-HDPE/10 composition (~34 J m–1), representing a reduction of about 23% with
respect to the R-HDPE (average impact resistance = 44 J m–1).
In this respect, we have learned that by decreasing the amount
of wood in the composites, one can actually increase their
impact resistance (Cruz-Estrada et al. 2006a, MartínezDomínguez 2008). However, in doing this, one has to analyse
the cost–benefit relationship, taking into consideration the specific application for which the resultant material is intended.
With respect to the effect of the wood particle size on the
composite impact resistance, Figure 3 shows an increase in the
value as the particle size increased. This occurred because the
smaller the wood particle size, the bigger the particle surface
areas that the polymer matrix had to wet in order to have
strong wood–polymer interphases to withstand impact forces.
Unless this wetting mechanism is guaranteed to be perfect
(together with others that may intervene), then the probabil-
ity of having weaker interphases will be higher, and hence
the lowest levels of impact resistance for the composites prepared using the smallest particles. Stark & Berger (1997)
reported similar behaviour for WPCs prepared with polypropylene and wood particles and fibres from ponderosa pine
(Pinus ponderosa). These researchers explained that, under
impact, the crack propagates at the weaker wood–polymer
interphase as well as through the polymer. Because cracks
travel around the wood particles, the fracture surface area
increases with increasing the particle size. As a result, more
energy is required to fracture the impact specimen with
The results of the tensile strength determination are presented in Figure 4. With respect to the content of wood in
the composites, Figure 4 shows that the tensile strength tends
to decrease as the content of wood in the composite increases.
This occurs for both types of composites (i.e. for those prepared using pure HDPE and for the recycled HDPE-based
composites). The mechanical strength of the composites
decreased because the wood particles were acting merely as
filler instead of acting as reinforcement. It is important to
emphasise that this preliminary study only involved experimentation with wood particles not longer than 2 mm. In this
regard, it is well-known that fibres with relatively high aspect
ratios are used to obtain better mechanical properties so that
the fibres act, in fact, as reinforcement. However, the processing method used for the composite preparation reported in
this work (i.e. extrusion) did not allow the use of very long or
continuous fibres due to agglomeration problems, which could
make difficult the processing because of the generation of
excessive levels of pressure and torque in the system, risking
the well-functioning of the equipment. On the other hand,
the authors are conscious of the fact that higher levels of tensile strength could be achieved by using wood particles with
smaller sizes. This was observed by McDonald et al. (2005) for
HDPE-based composites made with commercial maple wood
flour (particle sizes in the range of 100 to < 400 meshes). They
noticed a slight increase in the composite tensile strength,
increasing the wood content in them. However, in order to do
that, extra milling and screening operations may be needed
to obtain the required amounts to work with, which would
result in a laborious and costly process, resulting in a more
expensive final product.
It is worth mentioning, however, that for the particular
range of wood particle sizes and the two formulations experimented with in this work, the decrease in the tensile strength
with the increase of wood content was not really striking.
However, we have actually observed a more pronounced
decay in composites with contents of MESH 10 wood wastes
ranging from 5–70 wt% (Cruz-Estrada et al. 2006a, Martínez-Domínguez 2008). We explain the decrease in tensile
strength as follows – the wood fibres are more rigid than the
polymer matrix. Their elongation to fracture is lower than
that of the polymer matrix at comparable imposed tensile
R. H. Cruz-Estrada et al.
Fig. 3: Effect of wood waste content and wood particle size on the composite performance under impact. Impact resistance of V-HDPE and RHDPE is shown for comparison purposes only.
Table 3: Weight percentage distribution of wood wastes by mesh number for each sample screened.
is believed that this mainly corresponds to the decomposition
of cellulose with the generation of carbon residues. From
about 400˚C, sample weight loss occurs again in a gradual
fashion, showing certain trend to stabilise. Above 400˚C,
decomposition of the sample’s lignin and carbon residues
occurs. These results suggest that the maximum allowed temperature for processing the wood in order to obtain the composites should not be higher than 250˚C.
%RT = 1.33 for mesh no. 20, and refer to Table 1 for mesh
openings). It was decided to use MESH 60–100 adding up
the particles retained on both meshes no. 60 and 100 in order
to have a higher amount of smaller particles. Note from
Table 3 that the average %RT sum for both meshes is 28.24,
contrasting with the very small amount retained only on
mesh no. 100 (i.e. 8.23%), and the 27.10% retained on mesh
no. 40 consisting of bigger wood particles.
Determination of wood waste particle size distribution
Characterisation of mechanical properties
The percentage by weight of wood waste retained on each
mesh, for each sample screened, is presented in Table 3, which
also includes average values.
These results indicate that the procedure followed to
obtain a higher amount of smaller particles was effective in the
sense that the whole of the milled wood passed through mesh
no. 10. This means that, in theory, all the particles are smaller
than 2 mm as the %RT for this particular mesh was zero.
What is more, less than 1.4% of the milled wood had particle
sizes bigger than 0.85 mm (i.e. notice the average value of
The results of the Izod pendulum impact resistance determination are presented in Figure 3, which depicts the effect of
the wood waste content as well as the wood particle size on
the composites’ performance under impact.
In general, Figure 3 indicates that for these particular two
compositions, the composites perform better under impact
when the lowest amount of wood is used in their preparation. This general trend is observed irrespective of the use
of V-HDPE or R-HDPE. This is easily noticed by comparing
Wood–plastic composites from urban waste as potential building materials
Table 2: Details of the plastic–wood blends prepared.
R-HDPE Wood particle
three heating zones. Rods of approximately 3 mm in diameter were obtained by using a 4-cm long extrusion die of 2 mm
in internal diameter fitted to the extruder. The processing
temperatures were 165˚C in the extruder’s feed zone, 160˚C
in the other two zones, and 155˚C in the extrusion die. The
screws’ speed was set at 50 rpm. The resultant extrudates were
pelleted using a Brabender laboratory pelletiser machine
(type 12-72-000). For comparison purposes, both the V-HDPE
and the R-HDPE were processed at comparable extrusion
conditions to those used in the extrusion of the blends.
Test sample preparation
For the evaluation of mechanical properties, test specimens
were prepared from the composite pellets previously compounded. They were compression moulded into 20 cm × 20
cm flat plaques using a Carver automatic hydraulic press
(model 3891). The moulding was performed at 185˚C and at
a compression force of about 26,690 N (6,000 lbf) for 12.5
min. The thickness of the composite plaques varied depending upon the type of the test specimen required. Afterwards,
the resultant plaques were machined to obtain the specimens
with the dimensions specified in the standard test methods.
The same procedure was followed to obtain test specimens
of the V-HDPE and the R-HDPE from their previously produced pellets.
Fig. 2: The TGA thermogram of the wood wastes.
The tensile strength was determined using an Instron universal testing machine (model 1125), following the procedure
indicated in ASTM D-638 standard test method. Standard
type IV dumbbell-shaped test specimens were experimented
with, at a rate of crosshead motion of 2 mm min–1.
For comparison purposes, the V-HDPE and R-HDPE
specimens were also subjected to all the test methods previously described.
At least five specimens of each type were tested. All the
experiments were performed at room temperature, previously conditioning the specimens in accordance with ASTM
D-618 test method (i.e. relative humidity 50 ± 5%, temperature 23 ± 2°C, for at least 40 h).
Forming of wood–plastic sheets
Pellets (2 kg) of blends 50/R-HDPE/10 and 50/R-HDPE/60–
100 were respectively compounded (refer to Table 2 for
details). They were processed by extrusion using the laboratory twin-screw extruder previously described to produce flat
wood–plastic sheets. The processing temperatures were 165˚C
in the extruder’s feed zone, and 160˚C in the other two zones.
The screws’ speed was set at 50 rpm. A slot die 100 mm wide
(at 155˚C) with a 2 mm maximum allowed lip opening was
Results and discussion
Mechanical performance characterisation
The Izod pendulum impact resistance was determined in
accordance with ASTM D-256 standard test method. The
tests were carried out following method C, and using a Ceast
impact machine (model 6545) with a pendulum hammer
capable of delivering energy of 1 J. The specimens’ dimensions were 12.7 mm thick, 63.5 mm long and 6.35 mm wide.
Each specimen was notched using a Ceast Notchvis instrument type 6816, as indicated in the standard test method.
The results of the thermogravimetric analysis are presented in
Figure 2. For simplicity, only a selected TGA thermogram is
presented. The thermogram shows that the sample weight is
gradually lost up to a temperature of about 250˚C. This small
weight loss is presumably due to the loss of water, absorbed
from the environment, and decomposition of low molecular
weight components. The thermogram also suggests that
between about 250–400˚C, an abrupt weight loss occurred. It
R. H. Cruz-Estrada et al.
Fig. 1: Different views of the wood wastes used as raw material: Dried (A) and milled (B) wood wastes.
This pure HDPE has a melting temperature of 132˚C, a melt
flow rate of 0.03 g min–1 at 190˚C, and a density of 960 kg m–3
at 23˚C. The material is referred to as V-HDPE.
The particle size distribution by mesh number was determined taking the average result of the three screenings. The
following equation was used:
%RT = (WRm/W) × 100
Maleic anhydride grafted high-density polyethylene (Polybond 3009), supplied by Brenntag México, S.A. de C.V. was
used as coupling polymer. Its physical properties are as follow: melt flow rate, 0.5 g min–1 at 190˚C; density at 23°C, 950
kg m–3; and melting point, 127°C. The maleic anhydride level
was 1% by weight (wt%).
where %RT is the percentage by weight of wood retained on
mesh, WRm is the weight of wood retained on the mesh, and
W is the initial total weight of the sieved wood waste. The balance described above was used to weigh the wood retained on
The maximum allowed processing temperature for the wood
wastes was estimated by thermogravimetric analysis (TGA)
within the temperature range 30–700˚C using a Perkin-Elmer
TGA7 thermogravimetric analyzer. Two samples of the material were analyzed at a rate of 10˚C min–1 under a nitrogen
Wood waste–HDPE blends with 50 wt% and 60 wt% content
of wood were prepared using a horizontal mixer with a helical
agitator (Intertécnica Co., model ML-5) and a capacity of 5
litres. The blends were prepared by mixing the raw materials for
5 min. The wood used consisted of particles with different sizes.
Three different ranges of particle size were experimented with:
Determination of wood waste particle size distribution
Three samples about 380 g each of the milled wood wastes
were weighed in an Ohaus balance (sensitivity, 0.1 g), and
each screened in a Tyler nest of sieves (meshes number 10,
20, 30, 40, 60 and 100) for 15 min, using a W.S. Tyler RO-TAP
sieve shaker (model RX-29). The mesh openings are given in
Table 1: Mesh openings of the sieve column.
1. All the milled wood sieved through mesh no. 10 (i.e. particles of size < 2.00 mm).
2. The milled wood that passed through mesh no. 40, but
was retained on both meshes no. 60 and 100 (i.e. 0.430
mm > particles size > 0.149 mm).
3. All the milled wood that passed through mesh no. 100 (i.e.
particles size < 0.149 mm).
The different ranges of particle sizes described above will be,
respectively, referred to throughout the text as MESH 10,
MESH 60–100, and MESH 100. All the blends had 5 wt% of
the coupling polymer. Table 2 summarizes the details of the
The blends were processed using a Brabender laboratory
conical twin-screw extruder (model CTSE-V/MARK II) with
Wood–plastic composites from urban waste as potential building materials
Rangaprasad & Vasudeo 2003, Seethamraju & Deaner 2001,
Steward 2007, Stokke & Gardner 2003). Additional advantages of WPC products over conventional materials such as
wood are that they have better dimensional stability, require
less maintenance and have better resistance to decay (Maine
2004, Steward 2007, Treffler 2007).
The WPC-products market has mainly flourished in Canada, Europe and the US. Reports from 2005 indicate that the
production there was over 600,000 tonne per year. In contrast, in Latin-American countries (e.g. Mexico), where huge
amounts of wood and plastic wastes are also generated, a significant development of this market has not been observed,
as most of the current work is still at an experimental level
A specific case study is the city of Merida, capital of the
Yucatan State in south-eastern Mexico, where urban wastes
consisting of wood residues and objects made of different
kinds of plastics are produced in quantity. With respect to
the wood wastes, many arise from the periodic pruning of
branches from the different varieties of trees planted in the
city’s square gardens, parks and avenues as part of the cleaning and urbanising programmes established by the City Council (Cruz-Estrada et al. 2006, Cruz-Estrada et al. 2006a).
These wastes are ground into splinters and over 300 m3 are
generated every month (Castillo-González 2007). The wood
splinters are generally used to prepare compost and are randomly piled up in a specially designated disposal site, whose
storage capacity rapidly decreases because there is no strict
control regarding their exact amount and residence time
before they are used for making compost (Cruz-Estrada et al.
2006a, Cruz-Estrada & García-Gómez 2007).
On the other hand, in 2004 it was reported that about
17,800 tonne per month of urban waste (organic and inorganic) was generated in Mérida alone (Mijangos 2004), and
around 200 tonne per month of different types of plastic
objects were reported to be separated from the municipal
solids waste at the city’s Separation Plant (Cruz-Estrada et al.
2006a). Approximately 30% of the total amount of plastic
waste generated (i.e. about 60 tonne per month) corresponded
to different high-density polyethylene (HDPE) items, most of
which were containers for different products (Cruz-Estrada et
al. 2006a, Cruz-Estrada et al. 2007).
This situation calls for action to find an appropriate use
for these residues in the creation of new materials with practical applications. This, at the same time, would also contribute to partially solving the problems related to excessive accumulation of wastes (e.g. creation of habitats for harmful fauna,
and risk of fire during the dry season). Additionally, there is a
clear contribution towards protecting the environment, since
the use of virgin raw materials for manufacturing common
products is diminished. Not less important, all this also helps to
encourage the practice of recycling as a culture in society.
As part of a broad research project involving the study of
the availability of wood and plastic wastes generated in Merida and its surrounding regions to be used in the preparation
of polymer composites with potential applications as struc-
tural materials, we have preliminarily evaluated the tensile
and impact mechanical properties of WPCs obtained from
urban wood wastes and both recycled and virgin HDPE,
respectively. The main aim is to scale up the experimentation
in the laboratory to production processes, which will enable
WPCs design into different products with the maximum
Accordingly, we also report in this paper on the production at the laboratory level of flat wood–plastic extrudates,
with the intention to scale up the process to a pilot level to
evaluate fully the feasibility of using the material for construction. A potential application that we have identified is as roofing to substitute for the poor-quality black asphalt-saturated
corrugated cardboard sheets, used by the poorest people in
the country. This however, will be the subject of further publications.
Materials and methods
Wood waste was collected from one of the City’s dump yards
(‘Centro de Acopio Poniente’), giving them a minimum
previous treatment. These were wood splinters obtained by
pruning and subsequent chipping the different types of trees
planted around the city. The City Council workers transport
these residues to the open air storage facility, where they are
accumulated forming huge piles of rotting material which
eventually is used as compost. For the present work, a pile
was selected, and 9 kg of material was collected from different points at different depths in the pile which was stored for
about 2 months, and consisted of a full mixture of different
species of trees.
The wood residues were initially dried at 70˚C for 24 h in
a home-made convection oven in order to eliminate the
excess of moisture according to the procedure indicated in
ASTM D-4442 standard test method. Afterwards, the material
was milled with a Pagani granulating machine (model 1520)
fitted with a screen plate drilled with holes of 4 mm in diameter. In order to obtain smaller particles, the milled material
was milled again, but using this time a screen with holes of 1
mm in diameter. Figure 1 shows different views of the wood
wastes used as raw material.
The matrix material consisted of different high-density polyethylene (HDPE) bottles and containers provided by Merida’s Separation Plant. They were manually separated and
washed. Then, they were cut into pieces about 10 cm long,
and washed again. The caps, the stickers and the bottles with
non-water soluble residues were discarded. Afterwards, the
clean material was ground using the granulating equipment
previously described. A screen plate with holes of 4 mm in
diameter was used in this occasion. This material will be
referred to as R-HDPE.
For comparison purposes, composites using virgin HDPE
(grade 6003 from Petroquímica Morelos) were also prepared.
Los Angeles, London, New Delhi
© The Author(s), 2010. Reprints and permissions:
Waste Management & Research
2010: 28: 838–847
A preliminary study on the preparation of
wood–plastic composites from urban wastes
generated in Merida, Mexico with potential
applications as building materials
Ricardo H. Cruz-Estrada, Gustavo E. Martínez-Tapia, Gonzalo Canché-Escamilla,
Pedro I. González-Chí, Cesar Martín-Barrera, Santiago Duarte-Aranda,
Javier Guillén-Mallette, Carlos V. Cupul-Manzano, Osvaldo Martínez-Domínguez
Centro de Investigación Científica de Yucatán, Unidad de Materiales, Mérida, Yucatán, México
Facultad de Arquitectura, Universidad Autónoma de Yucatán, Mérida, Yucatán, México
A preliminary study on the use of wood and plastic wastes generated in Merida, Mexico to assess their potential for the development of building materials is reported. Composites based on recycled, high-density polyethylene (R-HDPE) loaded with
wood particles were prepared. The R-HDPE was collected from Merida’s Separation Plant, where it was sorted from other residues, either organic or inorganic. Composites based on virgin, high-density polyethylene (V-HDPE) were also prepared to
assess the effect of the R-HDPE on the composite’s mechanical properties. The wood came from the trims of different varieties
of the city’s trees that are periodically pruned as part of the cleaning and urbanising programmes implemented by the City Council. A batch of this material was selected at random to incorporate into both the R-HDPE and V-HDPE. Different wood particle
sizes were experimented with to obtain extruded composites with contents of 50% and 60% by weight of wood that were characterized under tension and impact. Flat wood–plastic extrudates with reasonable good appearance were also produced at the
laboratory level as a first step to find an adequate route to scale-up the process to a pilot level to evaluate the feasibility of producing alternative building materials.
Keywords: wood–plastic composites, urban-waste recycling, building materials, extrusion processing, mechanical performance
During the last decade, there has been an increase in the use
of materials made of thermoplastic resins loaded with fillers
rich in lignocellulosic fibres due to the added benefits that
they provide to the materials in terms of lightness, low cost,
mechanical resistance, better resistance to moisture, attack
by insects and micro-organisms, etc. (Jacobsen 2003, OrtegaLeyva 2008, Puppin 2003, Rangaprasad & Vasudeo 2003,
Seethamraju & Deaner 2001, Steward 2007, Stokke & Gardner 2003, Treffler 2007). A few examples of such fillers are
sisal, flax, hemp and kenaf fibres. In addition, the use of wood
fibres/flours from oak, maple, spruce, pine and fir trees has
grown considerably. Normally, the most successful cases are
the applications of materials based on wood fibres and high
density polyethylene or polypropylene (Bledzki & Faruk
2004, Eboatu et al. 2003, Lu et al. 2004, Maine 2004, Steward
2007, Treffler 2007). These materials, the so-called wood–
plastic composites (WPCs), are normally extruded or injection moulded into different products with specific transverse
sections, designs and geometries to satisfy the growing needs
of different markets, namely building materials, automotive,
marine, basic infrastructure, etc. (Correa et al. 2003, Jacobsen 2003, Maine 2004, Mehta et al. 2005, Pritchard 2004,
Corresponding author: Ricardo H. Cruz-Estrada, Centro de Investigación Científica de Yucatán, Unidad de Materiales, Calle 43, No. 130, Colonia
Chuburná de Hidalgo, C.P. 97200 Mérida, Yucatán, México.
E-mail: [email protected]
Received 25 June 2009; accepted in revised form 2 September 2009
Figures 1, 5 appear in color online: http://wmr.sagepub.com