Post on 24-Apr-2023
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
1
NOVEL SUSTAINABLE HEMP-BASED COMPOSITES FOR APPLICATION IN THE
BUILDING INDUSTRY: PHYSICAL, THERMAL AND MECHANICAL
CHARACTERIZATION
Enrico Sassoni1,2,*
, Stefania Manzi1,3
, Antonio Motori1,2
, Matteo Montecchi4, Max Canti
4
1Department of Civil, Chemical, Environmental and Materials Engineering (DICAM),
University of Bologna, via Terracini 28, 40131 Bologna, Italy
2Centro Interdipartimentale di Ricerca Industriale (CIRI) Edilizia e Costruzioni
University of Bologna, via Lazzaretto 15/5, 40131 Bologna, Italy
3Centro Interdipartimentale di Ricerca Industriale (CIRI) Meccanica Avanzata e Materiali
University of Bologna, Viale Risorgimento 2, 40136 Bologna, Italy
4CMF TECHNOLOGY S.P.A.
Via Bottegone 73, 41026 Pavullo nel Frignano (Modena), Italy
* Corresponding Author
Tel: +39 051 2090363, Fax: +39 051 2090322, e-mail: enrico.sassoni2@unibo.it
Ph.D.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
2
ABSTRACT
In this study, novel hemp-based composite materials, designed for building application as
components of a multi-layer wall plug for concrete, steel or wood structures, providing both thermal
insulation and physical-mechanical resistance, are presented and characterized. Composite panels
were produced by bonding hemp hurds with a novel hybrid organic-inorganic binder. The panels
were then characterized in terms of physical-microstructural properties (bulk density, water
absorption, swelling in thickness and weight loss after immersion in water, microstructural features
by means of scanning electron microscope and stereo optical microscope observation), thermal
properties (thermal conductivity, reaction to fire) and mechanical properties (compressive strength,
flexural strength, tensile strength, resistance to axial withdrawal of screws). The panels exhibited
promising physical, thermal and mechanical characteristics, generally comparable to those of
commercially available products. In addition, the novel composites have the advantage of a
significantly low environmental impact (thanks to the nature of both the dispersed and the binding
phase) and no negative effects on human health (contrary to many commercial materials,
responsible for emission of formaldehyde). All things considered, the novel composites seem like
very promising materials for application in the building industry. Further tests to assess the
physical-mechanical durability of the panels are currently in progress.
KEYWORDS
Hemp; Environmental sustainability; Green materials; Composite materials; Natural fibres; Thermal
conductivity; Thermal insulation; Durability; Formaldehyde-free materials; Fire reaction
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
3
1. INTRODUCTION
Because of the strong impact of the building industry onto the environment, developing new
materials with a reduced environmental impact has become a very important and urgent issue [1]. A
possible strategy is reducing the use of not renewable resources. For instance, in the case of
structural concrete, recycled aggregates from construction and demolition waste have been
investigated as possible substitutes for natural aggregates [2]. In the case of composite materials
used for building thermal insulation, the possible substitution of inorganic mineral fibers with
natural ones from agricultural resources is one of the studied routes [3, 4].
Indeed, traditional materials used for building thermal insulation, e.g. glass/rock wool,
extruded/expanded polystyrene, polyurethane foam, exhibit significant limitations with regard to
their environmental impact, both in the production phase and in the waste disposal phase. These
materials require petroleum-derived raw materials (e.g., extruded/expanded polystyrene) and/or
very high processing temperatures (e.g., glass/rock wool) and in many cases they cannot be
recycled (e.g., polyurethane foam) [5, 6]. Consequently, in recent years possible alternative
materials with enhanced environmental sustainability have been investigated. Composites
containing natural fibers from agricultural resources (e.g., jute, flax, hemp, cotton, cellulose) have
been extensively studied, as natural fibers exhibit several positive characteristics: low thermal
conductivity, low density, good specific tensile properties [3, 7, 8]. Moreover, natural fibers are
obtained from renewable resources, which makes the environmental impact of natural fiber-based
composites sensibly lower than that of traditional insulating materials [3, 5].
Among others, hemp (Cannabis sativa) has received a lot of attention because of its good
thermal insulation properties, good mechanical properties, rapid growing (only 3.5 months), high
dry biomass production (4 to 5 times higher than that produced by a forest of the same extension in
one year) and high carbon storage potential [9, 10]. Because of the above described characteristics,
several studies [7, 9-12] have investigated the insulating properties of hemp fibers and hemp hurds
(i.e. the non-fibrous fraction), bonded by either organic or inorganic binders, and several hemp-
based products, in the form of flexible mats and boards, have recently been introduced in the
market. The environmental benefits of hemp-based composites for building thermal insulation, in
place of traditional materials such as rock wool, have been quantitatively assessed by Life Cycle
Assessment (LCA) [10]. Moreover, hemp is currently used to create the so-called "hemp concrete"
(i.e., a mix of hemp particles, lime, water and, in some cases, sand), utilized for wall infilling, floor
and roof insulation and insulating plasters and renders [13]. While hemp concrete is used for its
thermal insulating ability and not for bearing loads, the possible exploitation of hemp fibers tensile
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
4
strength was investigated as well, with the aim of developing hemp fiber-reinforced concretes [14]
and pozzolanic matrix-hemp fiber composites for structural retrofit applications [15].
In this paper, novel hemp-based composite materials for application in the building industry
are presented and the results of a preliminary physical, thermal and mechanical characterization are
reported and discussed. The composites were obtained by bonding hemp hurds with a new hybrid
organic-inorganic binder, based on magnesium oxide and a reactive vegetable protein, in the form
of a flour. Five composite types, with density ranging from about 300 to 1300 kg/m3, were
produced: low density panels were designed for building thermal insulation, while medium and high
density panels were designed as possible substitutes for formaldehyde-bonded wood particle boards,
used in the building and furniture industry. In addition to the cited general uses, the new composites
were specifically developed for creating a multi-layer wall, to be used as a wall plug for concrete,
steel or wood structures: the inner, lower density layers are expected to provide thermal insulation
and the external, higher density layers are expected to provide resistance against physical-
mechanical actions, as well as protection against fire.
Compared to traditional materials used in the building and furniture industry, the new
composites are expected to have:
1) a better environmental impact, both in the production and disposal phases. Indeed, the hemp
hurds used as the dispersed phase and the vegetable protein used for the binding phase are
obtained from renewable resources, while production of magnesium oxide requires lower
temperature than production of other inorganic binders (e.g., lime or cement). As for the end-
of-life disposal, the hemp-based panels are completely recyclable, as they can be ground to
powder and used for production of new panels, by mixing with the binder and some new hemp
hurds.
2) a better impact on human health, as they are free from any toxic organic binder, such as
formaldehyde; the latter, classified as "carcinogenic to humans", can in fact be emitted in the
indoor environment by formaldehyde-bonded building materials, such as mineral wool and
wood particle boards [16, 17]; on the contrary, the newly developed composites will be entirely
formaldehyde-free.
3) a remarkably lower production cost, thanks to the low-cost and renewability of the raw
materials from agricultural resources and to the low processing temperature for magnesium
oxide production. This aspect makes the novel hemp-based panels highly competitive from an
economic point of view and particularly suitable for low-cost buildings construction, especially
in developing countries.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
5
Therefore, considering the promising properties of the new composites in terms of
environmental sustainability, impact on human health and economic competitiveness, this paper
focuses on the physical, thermal and mechanical properties of the new composites.
2. MATERIALS AND METHODS
2.1 Samples
Five composite types were studied, as detailed in Table 1. Low density (LD) and medium density
(MD) composites were produced by using coarse hemp hurds (about 10-30 mm in length and 2-6
mm in diameter), while high density composites (HD-1, HD-2 and HD-3) were produced by using
thin hemp hurds (about 2-5 mm in length and 1-2 mm in diameter). For all the composite types,
hemp hurds were bonded by means of a new hybrid organic-inorganic binder, currently under
patent [18]. The binder can in brief be described as based on magnesium oxide, which reacts when
an aqueous magnesium sulfate solution is added, and a reactive vegetable protein. For composite
production, magnesium oxide was mixed with water, then the vegetable protein, in the form of a
flour, and the magnesium sulfate solution were added. Suitably pre-treated hemp hurds were then
added to the so-prepared binder and then mixed at 80 r.p.m. for 3 minutes. In total, the addition of
the various components and mixing lasted 10 minutes. As reported in Table 1, the binder/hemp
hurds ratio differed as a function of composite density, ranging from 1:1 to 1:5. The fresh
binder/hemp hurds mix was then poured into a 150 × 50 × 5 cm3
wood frame; the frame was then
removed and the fresh composite was pre-heated at 80 °C in a micro-wave unit for 3 minutes, to
allow for uniform heating all through the panel thickness. Finally, the composite was pressed at a
temperature of 80 °C for 3 minutes, the compression ratio (i.e., the ratio between the final and the
initial thickness of the panel) varying as a function of composite density, as specified in Table 1. At
the end of the process, panels having 150 × 50 cm2
dimensions and thickness (t) ranging from 1 to 5
cm, as a function of composite density (Table 1), were obtained. The final appearance of the five
composites is illustrated in Figure 1. To perform the characterization tests described in the
following, specimens with suitable dimensions were obtained from the panels by sawing or, in the
case of fire reaction test, by assembling two panels.
2.2 Characterization techniques
2.2.1 Physical-microstructural properties
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
6
2.2.1.1 Bulk density
Bulk density (ρ) was measured on the same samples used for the physical-mechanical tests, as the
average for at least 15 specimens. According to European standard EN 1602 [19], the bulk density
was calculated as the ratio of sample weight (after storing for 28 days in laboratory conditions, θ =
23±2 °C and RH = 50±5%) and sample volume (measured with an accuracy of 0.5%).
2.2.1.2 Water absorption
Water absorption by total immersion (WA) was measured according to European standard EN
12087 [20]. The test was stopped after 24 hours, as this period was considered as sufficiently
significant given the high sorptivity of the panels. For each composite type, 5 specimens (5 × 5 × t
cm3) were used.
2.2.1.3 Swelling in thickness after immersion in water
Swelling in thickness after immersion in water for 24 hours (G) was measured according to
European standard EN 317 [21]. In the absence of a specific standard for measuring this parameter
on thermal insulating products for building applications, the cited standard for particleboards and
fiberboards (to which the studied hemp-based panels can be assimilated) was adopted. For each
composite type, 5 specimens (5 × 5 × t cm3) were used.
2.2.1.4 Weight loss after immersion in water
Weight loss after immersion in water (Δw) was measured by preliminarily drying samples in an
oven at 70 °C for 3 days (to prevent any influence of relative humidity content), then immerging
samples in water for 3 days and then drying again at 70 °C until constant weight. The immersion in
water was prolonged up to 3 days, in view of future durability tests involving this saturation period.
For each composite, 2 specimens (5 × 5 × t cm3) were used.
2.2.1.5 Microscopy observation
The composite microstructure was observed by stereo-optical microscope (SOM, Olympus SZX10)
and scanning electron microscope (SEM, Philips XL20) on the fracture surface of the samples.
2.2.2 Thermal properties
2.2.2.1 Thermal conductivity
Thermal conductivity (λ) was measured for LD and MD composites. In the case of the HD panels,
thermal conductivity is expected to be higher than that of LD and MD types; consequently, in the
multi-layer wall, no significant contribution to thermal insulation from these panels seems
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
7
reasonable (as a matter of fact, HD panels are not designed for providing thermal insulation but are
expected to provide resistance against physical-mechanical stress). Thermal conductivity was
measured according to European standard EN 12667 [22] on 50 × 50 × t cm3 samples, by using the
heat flow meter apparatus.
2.2.2.2 Fire reaction class
Fire reaction was determined for LD and MD composites, assuming that fire reaction of HD panels
will be at least the same as that of MD panels. Measurement was carried out according to European
standard EN 13823 [23] for building products. As required by the cited standard, two boards
measuring 50 × 150 × t cm3 and 100 x 150 × t cm
3 were assembled along the 150 cm edge, so as to
create a right-angle. The so-created specimen was exposed to the flames from a burner placed at the
bottom of the right-corner and its performance was evaluated in terms of heat production, smoke
production and falling flaming droplets and particles [23].
2.2.3 Mechanical properties
2.2.3.1 Compressive strength
Compressive strength (σc,10) corresponding to a strain ε = 10% was determined for LD and MD
composites on 5 specimens (5 × 5 × t cm3) according to European standard EN 826 [24]. The test
was carried out by subjecting samples to axial compressive strength using an Amsler-Wolpert
loading machine (maximum load 10 kN) at a constant displacement rate of 5 mm/min for LD panels
and 3 mm/min for MD panels. When strain reached 10%, i.e. when a displacement of the loading
plates of 5 mm for LD panels (thickness t =50 mm) and 3 mm for MD panels (thickness t =30 mm)
was reached , the corresponding compressive load value was registered. The test was not carried out
in the case of HD panels (thickness t = 10 mm) as the the friction between the loading plates and the
sample would make results scarcely significant.
2.2.3.2 Flexural strength
Flexural strength (σf) was measured for all the composite types. For each type, a three points
bending test was carried out according to European standard EN 12089 [25] on 5 specimens (26 ×
12 × t cm3), using an Amsler-Wolpert loading machine (maximum load 5 kN) at a constant
displacement rate of 10 mm/min.
2.2.3.3 Tensile strength perpendicular to the plane
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
8
Tensile strength perpendicular to the plane of the panel (σt) was determined, for all the composite
types, according to European standard EN 1607 [26] on 5 samples (5 × 5 × t cm3). The samples
were glued to 2 steel plates (20 mm thickness, to prevent bending of the plate), by using an epoxy
bi-component resin (Figures 2a-2c). After curing for 24 hours, the so-prepared samples were then
inserted into a steel-made pulling apparatus (Figure 2d), designed according to [27]. This apparatus
is fastened to the loading machine and has a self-aligning attachment to ensure even load
distribution in the sample. The specimen was then subjected to tensile strength test by using an
Amsler-Wolpert loading machine (maximum load 10 kN) at a constant displacement rate of 10
mm/min (Figure 2e). With respect to other methods described in the cited standard, involving direct
attachment of the steel plates to the loading machine, the adopted method has the advantage of
allowing much faster testing. Indeed, as a curing period of 24 hours is needed after gluing the
panels to the steel plates, if the plates were directly attached to the loading machine, no more than
one test per day could be carried out. On the contrary, thanks to the adopted method, all the samples
to be tested were glued to the steel plates on the same day and, after 24 hours, the tensile test was
performed on all the samples, by simply changing the specimen inserted into the purposely
designed pulling apparatus.
2.2.3.4 Resistance to axial withdrawal of screws
As MD and HD panels were designed as possible substitutes for formaldehyde-bonded wood
particle boards, the panel resistance to axial withdrawal of screws was tested. The measurement was
carried out according to European standard EN 320 [28], describing the test method for
particleboards and fiberboards. The test was performed on all the composite types and, for each
type, 5 samples (7.5 × 7.5 × t cm3) were used. For LD and MD types both resistance in the direction
perpendicular to the panel plane (R┴, Figure 3a) and resistance parallel to the panel plane (R
//,
Figure 3b) were determined, while for the HD types only R┴ was measured, because of panel
thickness. For panels with thickness t > 15 mm (i.e., LD and MD) the screwing length is fixed by
the standard (15 mm), whereas for panels with t ≤ 15 mm (i.e. HD) the screwing length corresponds
to panel thickness; therefore, in order to allow for a comparison among results obtained for LD, MD
and HD samples, the specific resistance to axial withdrawal of screws was also calculated as Rs┴ =
R┴/t. After screw insertion into the sample, the test piece was inserted into a purposely designed
system (Figure 3), consisting of a steel plate with a central boring (for restraining the screw) and an
underlying metal jig (for restraining the panel). The steel plate and the metal jig were then inserted
into the same self-aligning pulling apparatus used for measuring the tensile strength perpendicular
to the plane (cf. § 2.2.3.3). The screw and the panel were then pulled apart by applying an axial
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
9
tensile force by using an Amsler-Wolpert loading machine (maximum load 5 kN) at a constant
displacement rate of 10 mm/min, until the maximum tensile load (corresponding to screw
withdrawal) was reached.
3. RESULTS AND DISCUSSION
The results of the physical, thermal and mechanical characterization are reported in Table 2. The
SOM and SEM images of panel microstructure are reported in Figure 4 and Figure 5, respectively.
As expected, the bulk density of the composites progressively increases, from 330 kg/m3
(LD) to 660 kg/m3 (MD) to 1025-1280 kg/m
3 (HD), due to the increase of both the fineness of the
dispersed phase and the compression ratio. The progressive densification of panel microstructure
can be clearly observed in the SOM images (Figure 4), where a progressive reduction in open
porosity can be noticed. As a consequence of the very high porosity, as well as the hydrophilic
behavior of both the hemp hurds and the new binder, the panels exhibit medium-high water
absorption, ranging from 118.4 wt.% (LD) to 10.1 wt.% (HD-3) (Table 2). In the investigated range,
the variation of water absorption as a function of bulk density can be suitably described by a power
function (Figure 6). Consequently to the high water absorption (particularly in the case of the LD
composites), the panels exhibited not negligible swelling in thickness and weight loss after
immersion in water (Table 2), as discussed in detail later in the paper.
As a consequence of the cellular structure of hemp hurds (Figure 5) and the high open
porosity of the panels, mainly in the case of LD and MD panels, the new composites exhibit a good
thermal behavior (Table 2). In particular, LD panels have a thermal conductivity λ = 0.078 W/mK,
which is in good agreement with thermal conductivity values reported in the literature for boards of
similar density, obtained from natural resources such as wood wool, wood fibers and wood chips
[29]. Even if the thermal conductivity of LD panels is higher than that of traditional insulating
materials with lower density, in turn LD panels exhibit superior mechanical properties and reaction
to fire.
In terms of reaction to fire (Table 2), LD and, particularly, MD panels exhibited a better
behavior than many commercially available thermal insulating products based on natural fibers.
Indeed, according to classification introduced by European standard EN 13501-1 [30], MD panels
belong to class "B - s2 - d0" and LD panels to class "C - s2 - d0", where "s" indicates smoke
production and "d" presence of flaming droplets/particles. On the contrary, natural fiber-based
composites, without modification, typically belong to classes E and F [7].
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
10
As for mechanical properties, compressive strength increased from 1.15 MPa for the LD
panels to 3.04 MPa for the MD panels (Table 2). The variations in flexural strength, tensile strength
and specific resistance to axial withdrawal of screws as a function of bulk density are illustrated in
Figures 7-9, respectively. As a general rule, panel mechanical properties improved as bulk density
increased. These modifications in panel microstructural and mechanical properties can be achieved
by varying the compression ratio, the binder/hemp hurds ratio and the shape/size of the hurds,
which all act synergically in determining the final features and performance of the panels. In
particular, by comparing panels for which only one variable at a time was varied, the following
considerations can be done:
1) for the same hurds type (coarse) and the same binder/hemp hurds ratio (1:1), by lowering the
compression ratio from 1:1.7 (LD) to 1:3.3 (MD) an increase in bulk density from 330 to 640
kg/m3 was obtained and corresponding increases in compressive strength (from 1.15 to 3.04
MPa) and flexural strength (from 0.90 to 6.83 MPa) were achieved;
2) for the same hurds type (thin) and the same compression ratio (1:4), by lowering the
binder/hemp hurds ratio from 1:1 (HD-1) to 1:1.25 (HD-2) to 1:1.50 (HD-3) a progressive
increase in bulk density (from 1025 to 1280 kg/m3) was obtained and a corresponding increase
in flexural strength (from 10.08 to 17.47 MPa) was achieved.
In the case of the three types of HD panels, results exhibit a higher dispersion, compared to the LD
and MD panels, which can presumably be attributed to the higher difficulty of obtaining materials
with homogeneous distribution of the dispersed phase and binder in the case of the HD panels, as
they were produced by adopting a 1:4 press compression ratio, which represents a limit condition
for the pilot plant used in this study. However, in the case of tensile strength perpendicular to the
panel plane, exhibiting a sensibly higher dispersion than other mechanical properties (Figure 8), the
influence of the testing procedure (particularly gluing of panels to the steel plates) probably plays
an important role as well.
All things considered, the panels exhibit good mechanical properties (particularly in terms of
σf, σt and Rs┴), which make them promising potential substitutes for commercially available
products with similar density, especially in the case of the HD panels. The good mechanical
performance of the composites can be attributed to the good adhesion between the binder and the
hemp hurds. Indeed, as can be observed in the SEM micrographs (Figure 5), the surface of the hemp
hurds appears as well covered with the binder, without significant detachments. The good
correlations found between bulk density and σf (R2 = 0.955), σt (R
2 = 0.959) and Rs
┴ (R
2 = 0.999),
suitably described by power functions, may be useful for predicting mechanical properties of new
composite materials having bulk densities within the studied range.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
11
However, as already mentioned, the panels exhibited not negligible swelling in thickness
and weight loss after immersion in water (Table 2), particularly in the case of the LD composites.
As for swelling (G), the highest values were obtained for HD-1 panels (G = 10.6%), while
significantly lower values were obtained for LD and MD panels (3.9% and 4.7%, respectively) and
HD-2 and HD-3 panels (4.9% and 3.6%, respectively). This behavior, not corresponding to a trend
directly related to panel density, may be explained considering that, in the case of LD and MD
panels, more voids are available for internally accommodating panel swelling, while, in the case of
HD-2 and HD-3 panels, their higher tensile strength is able to effectively contrast swelling. As for
the weight loss after immersion in water, it progressively decreased from 16.3 wt.% (LD) to 6.9
wt.% (MD) to 3.0-3.7 wt.% (HD). The weight loss can presumably be ascribed to some dissolution
of the binder, which, in turn, is responsible for the detachment of hemp hurds. Consistently, weight
loss decreases as water absorption decreases (i.e., as bulk density increases) and as the binder/hemp
hurds ratio decreases.
Even if total immersion in water represents a very extreme condition that thermal insulating
panels very improbably would experience in service (as they are generally protected from direct
exposition to water, included possible rising damp from the ground), the evaluation of the physical-
mechanical durability of the new panels, in terms of resistance to water-related weathering
processes (e.g., wetting-drying cycles and freezing-thawing cycles) certainly needs to be
specifically investigated. Such work is currently in progress.
4. CONCLUSIONS
In this study, novel composite materials, designed for creating a multi-layer wall with thermal
insulation properties, to be used as a wall plug for concrete or steel structures, were obtained by
bonding hemp hurds with a novel hybrid organic-inorganic binder.
The panels exhibited promising physical, thermal and mechanical properties, generally
comparable to those of commercially available products. Indeed, the LD panels (ρ = 330 kg/m3),
designed as thermal insulators to be used in the internal layer of the wall, exhibited a good thermal
conductivity (λ = 0.078 W/mK), a quite good reaction to fire (class "C - s2 - d0", according to EN
13501-1 [30]) and relatively good mechanical properties (σc,10 = 1.15 MPa, σf = 0.90 MPa, σt = 0.18
MPa, Rs┴ = 2.9 N).
The MD panels (ρ = 640 kg/m3) exhibited a higher thermal conductivity (λ = 0.138 W/mK),
but still a very good reaction to fire (class "B - s2 - d0", according to EN 13501-1 [30]) and good
mechanical properties.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
12
The HD panels (ρ = 1025-1280 kg/m3), designed as possible substitutes for high density
formaldehyde-bonded wood particle boards and intended as the external layers of the multi-layer
wall, providing resistance to physical-mechanical actions, exhibited mechanical properties that
make them promising substitutes for commercially available products with similar density (σf =
10.08-17.47 MPa, σt = 0.82-1.88 MPa, Rs┴ = 81.9-147.2 N, depending on panel density). The
reaction to fire of HD panels was not directly tested but was assumed to be at least equal to that of
the MD panels.
Considering the above mentioned results, as well as the advantages in terms of
environmental sustainability and impact on human health, the new composites seem like very
promising new materials for the building sector. In addition, the new composites are very
competitive also from the economic point of view, thanks to the low cost of the vegetable resources
used for the binder and the dispersed phase, as well as the relatively low cost of the magnesium
oxide used for the binder.
Considering the not negligible swelling in thickness and weight loss after immersion in
water, further studies aimed at investigating the physical-mechanical durability of the panels, in
terms of resistance to wetting-drying and freezing-thawing cycles, are currently in progress.
ACKNOWLEDGEMENTS
The financial support of Emilia-Romagna Region (POR-FESR funds) is gratefully acknowledged.
Prof. Andrea Saccani (DICAM, University of Bologna) is gratefully acknowledged for
collaboration on the SEM observation.
REFERENCES
[1] Franzoni E., Materials selection for green buildings: which tools for engineers and architects?, Procedia
Engineering 21 (2011) 883-890
[2] Manzi S., Mazzotti C., Bignozzi M.C., Short and long-term behavior of structural concrete with recycled concrete
aggregate, Cement & Concrete Composites, 37 (2013) 312–318
[3] Korjenic A., Petrànek V., Zach J., Hroudovà J., Development and performance evaluation of natural thermal-
insulation materials composed of renewable resources, Energy and Buildings 43 (2011) 2518-2523
[4] La Mantia F.P., Morreale M., Green Composites: A brief review, Composites: Part A 42 (2011) 579-588
[5] Papadopoulos A.M., State of the art in the thermal insulation materials and aims for future developments, Energy
and Buildings 37 (2005) 77-86
[6] Jelle B.P., Traditional, state-of-the-art and future thermal building insulation materials and solutions - Properties,
requirements and possibilities, Energy and Buildings 43 (2011) 2549-2563
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
13
[7] Zach J, Hroudovà J., Brožovskỳ, Krejza Z., Gailius A., Development of thermal insulating materials on natural
base for thermal insulation systems, Procedia Engineering 57 (2013) 1288-1294
[8] Ku H., Wang H., Pattarachaiyakoop N., Trada M., A review on the tensile properties of natural fiber reinforced
polymer composites, Composites: Part B 42 (2011) 856-873
[9] Motori A., Manzi S., Montecchi M, Canti M., A preliminary study of the physical and mechanical properties of
sustainable hemp fibers based composite materials for building insulated walls, Proceedings of ECCM15 - 15th
European Conference on Composite Materials, Venice (Italy) 24-28 June 2012, p. 1-8
[10] Zampori L., Dotelli G., Vernelli V., Life Cycle Assessment of hemp cultivation and use of hemp-based thermal
insulator materials in building, Environmental Science & Technology 47 (2013) 7413-7420
[11] Balčiūnas G., Vėjelis S., Vaitkus S., Kairytė A., Physical properties and structure of composite materials made by
using hemp hurds and different binding materials, Procedia Engineering 57 (2013) 159-166
[12] Manzi S., Sassoni E., Motori A., Montecchi M., Canti M., New composite panels with hemp hurds
for sustainable buildings, Environmental Engineering and Management Journal 12, S11, Supplement (2013) 31-34
[13] Arnaud L., Gourlay E., Experimental study of parameters influencing mechanical properties of hemp concretes,
Construction and Building Materials 28 (2012) 50-56
[14] Awwad E., Mabsout M., Hamad B., Farran, M. T., Khatib H., Studies on fiber-reinforced concrete using industrial
hemp fibers, Construction and Building Materials 35 (2012) 710-717
[15] Asprone D., Durante M., Prota A., Manfredi G., Potential of structural pozzolanic matrix-hemp fiber grid
composites, Construction and Building Materials 25 (2011) 2867-2874
[16] Gunshera J., Mentese S., Salthammer T., Andersen J.R., Impact of building materials on indoor formaldehyde levels:
Effect of ceiling tiles, mineral fiber insulation and gypsum board, Building and Environment 64 (2013) 38-145
[17] Costa N. A., Pereira J., Ferra J., Cruz P., Martins J., Magalhães, Mendes A., Carvalho L.H., Scavengers for
achieving zero formaldehyde emission of wood-based panels, Wood Science Technology 47 (2013) 1261-1272
[18] International Patent Application WO2013/061182 (filed on 02/05/2013)
[19] EN 1602, Thermal insulating products for building applications - Determination of the apparent density, 2013
[20] EN 12087, Thermal insulating products for building applications - Determination of long term water absorption
by immersion, 2013
[21] EN 317, Particleboards and fiberboards - Determination of swelling in thickness after immersion in water, 1994
[22] EN 12667, Thermal performance of building materials and products - Determination of thermal resistance by
means of guarded hot plate and heat flow meter methods - Products of high and medium thermal resistance, 2001
[23] EN 13823, Reaction to fire tests for building products - Building products excluding floorings exposed to the
thermal attack by a single burning item, 2010
[24] EN 826, Thermal insulating products for building applications - Determination of compression behavior, 1998
[25] EN 12089, Thermal insulating products for building applications - Determination of bending behavior, 1999
[26] EN 1607, Thermal insulating products for building applications - Determination of tensile strength perpendicular
to faces, 2013
[27] Franzoni E. et al, New external thermal insulating composite boards for the building envelope retrofitting, in press
[28] EN 320, Particleboards and fiberboards - Determination of resistance to axial withdrawal of screws, 2011
[29] Domìnguez-Muñoz F., Anderson B., Cejudo-Lòpez J., Carrillo-Andrès A., Uncertainty in the thermal conductivity
of insulation materials, Energy and Buildings 42 (2010) 2168
[30] EN 13501-1, Fire classification of construction products and building elements - Part 1: Classification using data
from reaction to fire tests, 2009
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
14
Table 1. Nominal characteristics of the five composite types and corresponding technological parameters.
Sample Label Nominal density
[kg/m3]
Nominal thickness
[mm]
Binder/hemp hurds
ratio [wt./wt.]
Compression
ratio
Low density LD 300 50 1:1 1:1.7
Medium density MD 600 30 1:1 1:3.3
High density-1 HD-1 1100 10 1:1 1:4
High density-2 HD-2 1200 10 1:1.25 1:4
High density-3 HD-3 1300 10 1:1.50 1:4
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
15
Table 2. Results of physical, thermal and mechanical characterization of the composite panels (ρ = bulk density; WA =
water absorption after 24 hours; G = swelling in thickness after immersion in water for 24 hours; Δw = weight loss after
immersion in water for 3 days; λ = thermal conductivity; σc,10 = compression strength corresponding to a strain ε = 10%;
σf = flexural strength; σt = tensile strength perpendicular to the plane; R┴ = resistance to axial withdrawal of screws
perpendicular to the plane; Rs┴ = specific resistance to axial withdrawal of screws perpendicular to the plane; R
// =
resistance to axial withdrawal of screws parallel to the plane). Values are averages for 5 samples, except ρ and Δw
values that are averages for 15 and 2 values, respectively; standard deviations in brackets.
LD MD HD-1 HD-2 HD-3
t = 50 mm t = 30 mm t = 10 mm t = 10 mm t = 10 mm
Physical properties
ρ [kg/m3] 330 (± 31) 640 (± 43) 1025 (± 44) 1210 (±47) 1280 (± 125)
WA [wt.%] 118.4 (± 18.6) 39.3 (± 6.3) 28.4 (± 6.4) 21.2 (± 7.2) 10.1 (± 6.5)
G [%] 3.9 (± 0.9) 4.7 (± 0.5) 10.6 (± 1.2) 4.9 (± 2.0) 3.6 (± 1.7)
Δw [wt.%] 16.3 (± 0.0) 6.9 (± 0.2) 3.7 (± 0.1) 3.9 (± 0.5) 3.0 (± 0.3)
Thermal properties
λ [W/mK] 0.078
0.138
-
-
-
Fire reaction class C - s2 -d0 B - s2 -d0 -
-
-
Mechanical properties
σc,10 [MPa] 1.15 (± 0.24) 3.04 (± 0.14) -
-
-
σf [MPa] 0.90 (± 0.19) 6.83 (± 0.48) 10.08 (± 1.17) 16.41 (± 1.50) 17.47 (± 3.46)
σt [MPa] 0.18 (± 0.05) 0.49 (± 0.03) 0.82 (± 0.27) 1.79 (± 0.78) 1.88 (± 0.82)
R┴ [N] 146.5 (± 40.4) 687.4 (± 85.5) 960.4 (± 91.3) 1214.0 (± 175.6) 1792.2 (± 277.7)
Rs┴ [N/mm] 2.9 (± 0.8) 21.5 (± 3.1) 81.9 (± 9.9) 145.6 (±12.4) 147.2 (± 21.1)
R// [N] 147.0 (± 42.9) 795.9 (± 80.9) -
-
-
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
16
Figure 1. Appearance of the five composite types.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
17
Figure 2. Tensile strength test: preparation of LD (a), MD (b) and HD (c) samples by gluing to steel-plates; insertion
into the purposely designed apparatus (d) and exemplification of sample failure (e).
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
18
Figure 3. Resistance to axial withdrawal of screws test: exemplification of R┴
(a) and R//
(b) determination for a MD
panel and R┴
determination for a HD panel (c).
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
19
Figure 4. SOM images of the microstructure of LD (a), MD (b) and HD-2 (c) panels (the length of the marker is 1 mm).
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
20
Figure 5. SEM images of the microstructure of LD (a, b), MD (c, d) and HD-2 (e, f) panels. For each panel type, the
first image has a magnification of 250× (150× in the case of LD) and the second image has a magnification of 500×.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
21
Figure 6. Water absorption against bulk density of the five composite types.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
22
Figure 7. Flexural strength against bulk density of the five composite types.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
23
Figure 8. Tensile strength against bulk density of the five composite types.
Sassoni E., Manzi S., Motori A., Montecchi M., Canti M., Novel sustainable hemp-based composites for
application in the building industry: physical, thermal and mechanical characterization, Energy and Buildings 77 (2014) 219-226, DOI:10.1016/j.enbuild.2014.03.033
24
Figure 9. Specific resistance to axial withdrawal of screws against bulk density of the five composite types.