journal of the mechanical behavior of biomedical materials 15 (2012) 141–152

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/jmbbm

Research Paper

Mechanical properties of luffa sponge

Jianhu Shena, Yi Min Xiea,n, Xiaodong Huanga, Shiwei Zhoua, Dong Ruanb
a
Centre for Innovative Structures and Materials, School of Civil, Environmental and Chemical Engineering, RMIT University, GPO Box 2476,
Melbourne 3001, Australia
b
Faculty of Engineering and Industrial Sciences, Swinburne University of Technology, John Street, Hawthorn, VIC 3122, Australia

art i cle i nfo ab st rac t

Article history: The paper presents the first scientific study of the stiffness, strength and energy absorption
Received 3 April 2012 characteristics of the luffa sponge with a view to using it as an alternative sustainable
Received in revised form engineering material for various practical applications. A series of compression tests on
25 June 2012 luffa sponge columns have been carried out. The stress–strain curves show a near constant
Accepted 6 July 2012 plateau stress over a long strain range, which is ideal for energy absorption applications. It
Available online 17 July 2012 is found that the luffa sponge material exhibits remarkable stiffness, strength and energy

Keywords: absorption capacities that are comparable to those of some metallic cellular materials in a

Luffa sponge similar density range. Empirical formulae have been developed for stiffness, strength,

Mechanical properties densification strain and specific energy absorption at the macroscopic level. A comparative

Energy absorption study shows that the luffa sponge material outperforms a variety of traditional engineering

Sustainability materials.
& 2012 Elsevier Ltd. All rights reserved.

1. Introduction to accommodate the natural environment to which they were

Biological materials and structures have distinguished them
from traditional human developed counterparts because of
their unique characteristics (Meyers et al., 2011). The most
attractive one is their long-term sustainability to the natural
environment. The performance of the traditional man-made
materials, notably various metals, metallic alloys, ceramics,
plastics, as well as their composites, significantly surpass the
biological materials (Bonderer et al., 2008; Zhang et al., 2011).
However, most of the man-made materials are not environ-
mentally friendly and little has been concerned with their
sustainability (Zhang et al., 2011). By contrast, those biologi-
cal materials are based on relatively weak base materials
such as minerals and proteins which are easily degradable,
bio-compatible, pollution-free, recyclable and energy-effi-
cient (Zhang et al., 2011). Furthermore, over millions of years
exposed, a large number of biological systems evolve periodic
cells with self-similar hierarchical microstructures. Similar to
turtle’s shell (Damiens et al., 2012), jellyfish mesogloeas (Zhu
et al., 2012), wood (Stanzl-Tschegg et al., 2011; Wegst, 2011),
and E. aspergillum sponge (Mayer, 2011), those hierarchical
architectures are optimised or partially optimised and can
achieve multi-functions with high toughness and efficiency.
As we approach the limit of non-renewable natural resources,
these properties are essential for the long-term sustainability
of our habitat, and is becoming increasingly significant to
human civilisations (Zhang et al., 2011).
Luffa sponge is one of such commercially viable and
environmentally acceptable biological material derived from
fruit of Luffa cylindrica (LC) plant and having recycling cap-
ability and triggered biodegradability (John and Thomas,
2008; Oboh and Aluyor, 2009). It is relatively stable in their

n
Corresponding author. Tel.: þ61 3 9925 3655.
E-mail address: mike.xie@rmit.edu.au (Y. Min Xie).

1751-6161/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jmbbm.2012.07.004
142 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152
intended lifetime but would biodegrade after disposal in
composting conditions. At the same time, the fruits of LC
have a netting-like fibrous vascular system. When they are
dried, the fibrous network structure serves like an open cell
foam material. It has the potential to be used as an alter-
native material for man-made cellular materials. The impor-
tance of biological materials such as the luffa sponge is
growing as we search for sustainable solutions using new
materials.
Only a limited amount of research has been conducted on
the luffa sponge as a source of bio-fibres and bio-composites
in the last ten years. Those researches indicated that it was a
potential alternative material for packaging (Mazali and
Alves, 2005), water absorption (Bal and Bal, 2004; Demir
et al., 2006), and waste water treatment (Laidani et al., 2001;
Oboh et al., 2011). The luffa fibres were also used as reinfor-
cement fibre for other materials (Boynard and D’Almeida,
2000; Ghali et al., 2009; Laranjeira et al., 2006; Paglicawan
et al., 2005; Tanobe et al., 2005) and cell immobilisation for
biotechnology (Chen and Lin, 2005; Roble et al., 2002; Tavares
et al., 2008; Zampieri et al., 2006). At the same time, sponge
gourd (LC), the origin of luffa sponge material, have not yet
had their potentialities fully explored. With regard to indus-
trial and technological development, the cost of fuel is on the
increase. Oil can be extracted from seeds for industrial use
(Bal and Bal, 2004). The oil extracted from LC is finding
increasing use in the production of biodiesel which is now
gaining wide acceptance because of low CO2 emission and
other considerations (Ajiwe et al., 2005).
However, there is a lack of scientific data on the mechanical
properties of luffa sponge material because up to now its
main practical use is a body scrub in the bathroom. Due to
the lack of experimental proofs, their complex hierarchy
microstructures and other common limitations of biological
materials, potential applications have not been implemented
in practice for luffa sponge material as well as luffa fibres.
To this end, the mechanical properties of luffa sponge
columns were tested and compared with other cellular
materials to check the performance of this light weight
material. Uniaxial quasi-static compressive tests were con-
ducted at a strain rate of 103 s1 by using an Instron
machine to study the mechanical properties of luffa sponge
material. Cylindrical specimens with different relative den-
sity were tested at a room temperature of 251 C and a
humidity of approximately 40%. An energy efficiency method
was adopted to obtain the values of the densification strain
and plateau stress, and thus the energy absorption capacity
per unit volume. The experimental results were also dis-
cussed together with test results published by other research-
ers for other celluar materials.
2. Experiments
2.1. Specimens
The luffa sponge used in our experiments was obtained from
pharmacies in Australia which was sold as a bath sponge. A
brief treatment procedure for manufacturing these bath
sponges from natural luffa fruits was provided by the
supplier. The luffa fruits were harvested after they were fully
mature with their skin turning brown. The dried luffa fruits
were slightly squashed laterally to crack and remove the skin.
Then the two ends of luffa fruits were cut and the seeds were
removed. The original luffa sponges were bleached using
liquid chlorine bleach (4%) for about 1 h to improve their
appearance by making them whiter. After that, they were
soaked in clean water for half an hour and then dried in the
sun. The chemical composition of the luffa sponge depends
on several factors, such as plant origin, weather condition,
soil, pre-treatment, etc. A set of reference values for the
chemical composition of LC foam can be found in a previous
research (Siqueira et al., 2010).
The measured maximum diameter for each specimen of
dry luffa sponges tested was within the range of 55–86 mm.
According to the standard of compressive tests and recent
research (Ashby et al., 2000) on other cellular material, the
specimen size effect is negligible for foams when the dimen-
sion of specimen is sufficiently larger than the cell size, i.e., 7
times of the cell size for metallic foams. Thus the specimen
for the luffa sponge should be large enough to eliminate the
specimen size effect. The available maximum size along
radial direction for luffa sponge is whole section of the luffa
sponge column. For most of the luffa sponge column, the
section area at two ends of the specimen is different, thus
two thin layers of luffa sponge slices were cut at each end.
From a single luffa sponge, several cylindrical test specimens
(approximately 50 mm high) and two measurement slices
(approximately 5 mm thick) from both ends were cut using a
bandsaw cutting machine. After the initial cutting, the speci-
mens were milled further to make the two end surfaces
smooth and parallel. For calculation purpose, the cross
sectional area was taken as the enclosed section (ignoring
any internal voids). Because the cross section for most
specimens was not a perfect circle, the actual cross sectional
area was determined from photographs taken of each mea-
surement slice and processed using image software—
Photoshop. The actual cross sectional area from the two
measurement slices was then averaged. An equivalent dia-
meter was calculated from the average area. This equivalent
diameter was within the range of 42–81 mm. A total of 26
specimens were tested. Fig. 1(a) shows a typical cylindrical
specimen for a compressive test. The volume of the luffa
sponge specimen is the product of this cross sectional area
and the height of the specimen.
2.2. Geometry features and standard variation of density
After the specimens were cut, three different core topologies
were observed from our luffa sponge specimens as shown in
Fig. 2(a). The variation in specimen diameter and cross
sectional area is shown in Fig. 2(b).
The luffa sponge column is composed of luffa fibres. Those
fibres interconnect with each other and form networks with
micro-trusses. The length of those micro-trusses represents
the cell length of the luffa sponge material. A rough mea-
surement indicates that the average length of those micro-
trusses is in the scale of millimetre (1–5 mm). The major
orientation of these luffa fibres in the luffa sponge columns
exhibits a regular pattern. According to different orientation
 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152 143

Core region Inner

surface

Interlayer

Outer
surface

Longitudinal
direction

Circumferential Radial
direction direction

Fig. 1 – A luffa cylindrical specimen for a compressive test: (a) Different regions. (b) Orientation of luffa fibres (left: inner
surface; middle: outer surface; right: cross section). (c) Microstructures of luffa fibres (left: luffa fibre in 1 mm; middle: cross
section of luffa fibre in 101 mm; right: cross section of luffa fibre in 102 mm).

of the luffa fibre, the luffa sponge cylinder can be divided into
four regions, namely, the inner surface, the outer surface,
interlayer and core as shown in Fig. 1(a). On the inner surface,
the thickest luffa fibre grows along longitudinal directions.
While on the outer surface, the thickest luffa fibre grows
along circumferential directions. In the core region, the
strongest luffa fibre is along radial direction. In the interlayer
between inner surface and outer surface, the fibre grows in
all three directions as shown Fig. 1(b).
It can be seen from Fig. 3, the variation of density of the
luffa sponge material is irrelevant to the core topology and
the size of the luffa sponge columns. Luffa sponge is a
material with hierarchical architectures at several length
scales. At each hierarchical level, there is a corresponding
density. We observed three levels, namely, the luffa sponge
column (50 mm), the luffa fibre (1 mm) and the cell wall of
luffa fibre (0.01 mm) as shown in Fig. 1. The density of luffa
sponge column is 25–65 kg/m3. The density for luffa fibre is
350–650 kg/m3. A previous research also showed that
the density of the luffa fibre is approximately 353 kg/m3,
with a Young’s modulus of 1332 MPa (735%) and tensile
strength of 11.1 MPa (785%) (Paglicawan et al., 2005).
And the density for cell wall of luffa fibre is 820–
920 kg/m3 (Siqueira et al., 2010). The microstructures of
cell wall of luffa fibre and their mechanical properties
remained unknown at present. For this reason, the empirical
formulae in this paper will be given on the macroscopic
level.
Similar to other cellular materials, the mechanical proper-
ties of the luffa sponge material are closely related to its
density rather than other geometric features mentioned
above. Thus the specimens were grouped only according to
their densities in order to obtain a reliable empirical formula
to represent their mechanical performance. The effect of
specimen size and core topology on its mechanical properties
was disregarded in the current paper.
144 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152

2.3. Experimental procedures

Uniaxial compressive tests were conducted at very low strain

rates, namely, 103 s1, to obtain the mechanical properties
under quasi-static loading. An Instron machine was used to
conduct most of these tests as shown in Fig. 4. A Shimaszu
machine was also used for several trail tests. Both machines
were calibrated using the same load-cell to ensure the
consistency of the experiment results. The displacement
and load curves were recorded and the average compressive
force, densification strain and energy absorption can be
worked out using an energy efficiency method. A tempera-
ture and humidity metre was used to monitor the tempera-
ture and humidity around the test specimens during
the tests.
Like other natural sponges such as the sponge Euplectella
aspergillum (Johnsona et al., 2010), the mechanical properties
of luffa sponge are also influenced by the moisture. Thus the
Fig. 2 – Different core topologies and sizes for luffa sponge humidity during the test was monitored. The specimens were
specimens: (a) core topology (three different patterns are placed in the test room for at least 2 h before the compressive
shown, namely two links, three links and four links); (b) test to eliminate the difference in temperature and humidity.
variation of diameter of cross section (from 46 mm to The temperature in the test room was from 201 to 281, but
81 mm). during each test its variation was less than 11. The humidity
was between 30% and 44%, but during each test its variation
65
was less than 1.0%. Photographs were taken during the quasi-
static compressive test at every 30 s which corresponded to a
60 1.5 mm displacement interval for the test velocity used. It
Effect of outer diameter
55 should be noted that for the second specimen, a strong light
was used to illuminate the specimen to get better photo-
50
Density (Kg/m3)

graphs. The consequence of this action was that the tem-

45 perature increased from 21.5 1C to 28.5 1C and the humidity
dropped from 39% to 33% during this test. No obvious
40
variation of stress–strain curve was found for this test as
35 shown in Fig. 5(a).
30

25
3. Experimental results
20
55 60 65 70 75 80 85 90 3.1. Deformation features
Diameter of luffa sponge specimen (mm)
Typical force–displacement curves are given in Fig. 5(a) with
65 64.1 similar density. It shows clearly a fairly constant compressive
Core topology 2
60 Core topology 3 force over a long stroke, which represents an ideal energy
57.4
Core topology 4
55 54.1
50
Density (Kg/m3)

45

40

35 Humidity and Load cell

temperature
30 metre Top platen
28.5
25 27.3

20
0 5 10 15 20 25 30 35 40 45 50 55
Specimen number v0 Luffa specimen

Fig. 3 – Variation of density of luffa cylinders: (a) with Bottom platen

respect to equivalent diameter (no particular trend was
found); (b) with respect to core topology (no particular trend Fig. 4 – Experimental set-up of a quasi-static test of a luffa
was found). sponge column.
 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152 145

Fig. 5 – Deformation features of luffa sponge columns crushed axially: (a) force–displacement curves of specimens of similar
density; (b) micro-truss level rotation and bending of luffa fibre. (Dotted lines represent the original fibre orientation, and the
dash dot lines in the right photograph represent rotated original fibres. It can be seen that the fibre has also bending
deformation.)

absorption feature. Generally, the deformation can be divided
into three regions, namely, a rapid increasing elastic region, a
relative smooth plateau collapse region, and a densification
region with sharp increase of force over displacement as
shown in Fig. 5(a). The deformation process is shown in Fig. 6.
Those photographs confirm that the overall compression
strain of the specimen is axial along the longitudinal (load-
ing) direction rather than folding of the wall of the specimen.
The deformation is uniform in the elastic region. Similar to
metallic foams, the compressive deformation process for
luffa sponge material under quasi-static loading was not
uniform in the plateau collapse region. Localised crushing
band was observed which was similar to the static compres-
sion behaviour of metallic foams described by Bastawros
et al. (2000) and Shen et al. (2010). At a smaller scale, luffa
fibres exhibited axial compression and tension as well as
micro-truss-level distortions and rotations which were simi-
lar to the open-cell metallic foams as shown in Fig. 5(b).
Nominal stress (defined as force over original cross sectional
area) and nominal strain (defined as displacement over
original thickness of the foam specimen) were calculated to
get other mechanical properties.
146 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152

Fig. 6 – Typical deformation process of a luffa sponge material (Density: 52.3 kg/m3, Length: 51.3 mm). The deformation was
uniform until the deformation reach 3.5 mm, then localised crush band occurred at the middle part of the specimen.

3.2. Compressive strength where syf is the yield stress of the base material, rf is the
density of the base material, s0 and r0 are the compressive
Similar to other cellular materials (Gibson et al., 2010), the strength and density of the luffa sponge, A and B are two
compressive strength of the luffa sponge material is taken to constants determined by the topology and failure patterns of
be the initial peak stress if it exists. If there is no such peak cellular material. As mentioned previously, the base solid
stress, the stress at the intersection of two slopes is taken to material of luffa sponge and its mechanical properties
be the compressive strength. The two slopes are the slope for remained unknown at the present due to the lack of scientific
the initial loading and that for the stress plateau. It has been data on the mechanical properties of luffa sponge material.
found by other researchers (Ashby et al., 2000; Lu et al., 1999; The luffa sponge column specimen in our experiment was
McCullough et al., 1999; Miyoshi et al., 1999) that compressive composed of luffa fibre. Our preliminary tensile test results
strength of cellular materials obeys a power law with the on luffa fibres indicated that the density and tensile strength
relative density of luffa fibre varied greatly with different orientations in the
same luffa sponge and with different specimens for the same
!B orientation. A large quantity of tests should be conducted to
s0 r
¼A 0 ð1Þ get an accurate average values for mechanical properties of
syf rf
luffa fibres. Thus the empirical formulae are given in the
 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152 147

Table 1 – Fitting parameters for empirical formulae of strength, densification strain and plateau stress.

Mechanical parameter A(C) error B(D) error Correlation coefficient Standard derivation

Strength 0.38 0.24 0.74 0.095

Plateau stress 0.18 0.11 0.89287 0.0587
Densification strain 0.02172 5.38  104 0.48172 0.02467

Linear fit (y ¼ AþBx) was used for all experimental data. Data for strength and plateau stress were converted to logarithmic scales before fitting.
‘‘A error’’ is the standard error for A.

following format 0.50

Experimental data
B 0.45
s0 ¼ Ar0 ð2Þ Empirical formula, Eq. (2)

Compressive strength (MPa)

0.40
The unknown base material properties are included in the
coefficients A and B. By fitting to Eq. (2) using the quasi-static 0.35
data in our experiment, the parameters A and B are 2.25  103 0.30
and 1.28, respectively if the unit of stress is Pa and that of
0.25
density is kg/m3. The fitting parameters are listed in Table 1.
0.20
For truss material with compression and stretching deforma-
tion pattern, an exponent of 1 is expected for plastic yielding 0.15
and 2 for elastic buckling (Mostafa and Damiano, 2010). For 0.10
open cell foams with bending dominant deformation pattern,
0.05
an exponent of 1.5 is usually found for plastic collapse
strength and 2 for elastic collapse strength (Gibson et al., 0.00
20 25 30 35 40 45 50 55 60 65 70
2010). A lower exponent indicates fewer penalties for
Density (kg/m3)
strength and energy absorption with decreasing density for
designing very light weight cellular materials. The exponent, Fig. 7 – Compressive strength of luffa sponge material. The
1.28 from our test results, compares favourably with that for square symbols represent the experimental data, the error
open-cell foams. But it is larger than that for truss materials. bars show the derivation of experimental data from their
However, for solid material based metallic foams, the expo- empirical values.
nent within the range of 1.5–3 is found in other experimental
work (Gibson et al., 2010). The value from our experiment Young's Modulus
12
suggests that the collapse of the luffa sponge is determined Empirical formula
by the coupling of axial yielding and bending of the luffa
10
fibre. The experimental results and empirical prediction are
Young's moduls (MPa)

shown in Fig. 7. It should be noted that the derivation of luffa

sponge material is large as shown in Fig. 7 and discussed in 8

Section 2.1. The scattering effect may also contribute to the

low exponent. If the bending is the dominant deformation 6
mechanism for luffa fibres, more experimental data will
increase the value of this exponent. On the other hand, if 4
the axial stretching and compressing of luffa fibre is the
dominant deformation mechanism, more experimental data 2
will decrease the value of this exponent.
0
20 25 30 35 40 45 50 55 60
3.3. Elastic modulus Density (kg/m3)

Similar to metallic foams (Ashby et al., 2000), the slope of the
initial loading portion of the curve is lower than that of the
unloading curve. It indicates that there is localised plastic
deformation in the specimen at stress levels well below that
of the compressive strength of the luffa sponge material
reducing the slope of the loading curve. However, the reload-
ing curves after unloading roughly is close to the initial
loading curves. As a result, measurement of Young’s modulus
was made from the slope of the initial loading curve in the
current study. Thus, the slope between 25% and 75% of the
compressive strength is taken as the elastic modulus of luffa
sponge material. Following the same argument for the
compressive strength, for Young’s modulus, the empirical
Fig. 8 – Young’s modulus of luffa sponge material. The
square symbols represent the experimental data, the error
bars show the derivation of experimental data from their
empirical values.
formula is
E ¼ 1:48  106 r0 1:16 ð3Þ
where E is the Young’s modulus of luffa sponge; the unit of
stress is Pa, and that of density is kg m3.
The experimental results and empirical predictions are
shown in Fig. 8. Similarly to the compressive strength, the
exponent, 1.16, is compared favourably with those from
148 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152

Gibson et al. (2010), 1.5. However, for solid material based

1.0
metallic foam, this exponent is within the range of 1–2 found
in other experimental work (Gibson et al., 2010). It should be 0.9
noted that the standard derivation of luffa sponge material is 0.8
fairly large as discussed in Section 2.1. This may also
0.7
contribute to the low exponent.

Stress (MPa)
0.6
0.5
3.4. Unloading and reloading features of luffa sponge
material 0.4
0.3
For any energy absorption devices, the bounce-back beha-
0.2
viour is critical to some of their applications. The bounce-
back behaviour is dominated by the stored elastic strain 0.1
during collapse (Lu and Yu, 2003). This feature can be 0.0
captured by the unloading and reloading curves for a speci- 0.0 0.2 0.4 0.6 0.8 1.0
men at different collapse strains. Typical unload–reload Strain
characteristics at different compressive strains are presented
35
in Fig. 9(a). The relatively high gradient of unloading curves
Density=41.1 kg/m
indicates the stored elastic energy is relatively small for luffa
30
sponge material. The unloading and reloading curves do not

Loading modulus (MPa)

overlap with each other. Due to the steep gradient the 25
dissipated energy in the hysteresis loop is very small com-
pared to the area under the entire stress–strain curve. Thus 20
luffa sponge can be used in those situations where a once-off,
15 Young's Modulus from non-cyclic loading
non-reversible energy absorption is required. The unloading 6.45MPa
curve is not linear and is steeper at the beginning of the
10
unloading. The reloading curve, on the other hand, is almost
linear. This phenomenon indicates the hysteresis of the luffa 5
sponge material at a strain rate of 103 s1. The hysteresis is
more prominent at larger compressive strain. For a given 0
temperature and humidity, it is observed that the unloading 0.0 0.2 0.4 0.6 0.8 1.0
and reloading curves intersect at exact unloading point as Unloading strain
shown in Fig. 9(a). The slope of reloading curve, reloading Fig. 9 – Unloading and reloading features of luffa sponge
modulus, varies with the strain level at the unloading point material: (a) stress–strain curves with 9 unloading points
as shown in Fig. 9(b). The reloading modulus is very similar to and the total unloading strain is about 5% (an enlarged view
the Young’s modulus measured in Section 3.3 at the plateau of unloading features in the elastic deformation region is
collapse region ranging from the first peak stress to the provided); (b) loading modulus in each of the unloading
densification strain. circle versus loading strain at the unloading point.

3.5. Energy efficiency method to determine densification

strain, plateau stress and specific energy absorption Then, the plateau stress spl is calculated as
R ed
sðeÞde
The compressive strength, s0, corresponds to the collapse of spl ¼ 0 ð6Þ
ed
the weakest layer of cells which was crushed first in com-
pression. Hence, this stress is not applicable to representing An example is shown in Fig. 10.
the energy dissipation of foams in compression. It is the The densification strain physically corresponds to the start
plateau stress, spl, which is closely related to the energy point from which the stress starts to rise sharply. From this
dissipation capacity of foams and its value is obtained using point onwards, the luffa sponge can still dissipate energy by
an energy efficiency method (Li et al., 2006). plastic deformation, but its dissipation efficiency will start to
The energy dissipation efficiency, Ed, of the foam is defined decrease. Based on the present experimental data, we can
as (Li et al., 2006) propose that the densification strain is linearly related to the
R ed density of the luffa sponge. Hence
sðeÞde
Ed ðea Þ ¼ 0 ; 0rea r1 ð4Þ ed ¼ CDr0 ð7Þ
sa
By fitting to Eq. (7) using the quasi-static data in our
and the densification strain, ed, is defined as the maximum
experiment, the parameters C and D can be obtained as
value of ei which satisfies the condition of the maximum
0.68 and 1.62  103, respectively. The unit of density is
efficiency
kg/m3. The fitting parameters are listed in Table 1. The
dEd ðea Þ empirical prediction and experimental data are shown in
9ea ¼ ei ¼ 0; 0rei r1 ð5Þ
de Fig. 11.
 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152 149

Strain
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
1.0 0.60
Compressive Stress
0.9 0.54
Energy Efficiency
0.8 0.48
Compressive stress (MPa)
0.7 0.42
Plateau stress

Energy efficiency
0.6 0.31MPa 0.36

0.5 0.30
Compression strength
0.27MPa
0.4 0.24

0.3 0.18
Densification strain
0.2 0.57
0.12

0.1 0.06

0.0 0.00
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Strain

Fig. 10 – Illustration of energy efficiency method.

0.7 0.30
Energy absorption per unit volume (N/mm2)

Experimental data
Empirical formula
0.6
0.25
0.5
Densification strain

0.20
0.4

0.3
0.15

0.2 Experimental data for luffa sponge material

Empirical formula 0.10
0.1 Aluminium foam

0.0 0.05
0.06 0.08 0.10 0.12 0.14 0.16 0.18 20 25 30 35 40 45 50 55 60
Relative density Density (kg/m3)

Fig. 11 – Densification strain for luffa sponge specimens. Fig. 12 – Energy dissipation per unit initial volume versus
relative density. The square symbols represent the
experimental data, the error bars show the derivation of
By fitting to Eq. (2) using the quasi-static data for plateau experimental data from their empirical values.
stress in our experiment, the parameters A and B can be
obtained as 8.65  103 and 1.22, respectively if the unit of
stress is Pa and that of density is kg/m3. The fitting para- the density range in our experiment can be written as
meters are listed in Table 1. They are slight different from
w ¼ 8:65  103 ð0:68r0 1:22 1:62  103 r0 2:22 Þ ð8Þ
these for compressive strength. Similar to the compressive
strength of the luffa sponge, the exponent, 1.22, is compared Fig. 12 shows that the energy dissipation capacity per unit
favourably with those from Gibson et al. (2010), 1.5. However, volume relatively linearly increases with relative density in
for solid material based metallic foam, this figure should be the density range tested.
within the range of 1.5–3 found in other experimental work
(Gibson et al., 2010). Similar to strength of luffa sponge
material, the scattering effect may also contribute to the 4. Comparison with other materials
low exponent.
The energy dissipation capacity of luffa sponge material As a biological material, the experimental data in this paper
can be characterised by the energy absorption per unit initial about luffa sponge material may significantly expand the
volume, w, during the compression process before densifica- density range of previously studied natural cellular materials
tion occurs. This value is equal to the area under the (Gibson et al., 2010) with respect to the strength, as shown in
stress–strain curve. Thus the energy dissipation capacity for Fig. 13. This special feature provides an opportunity to widen
150 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152

1000 Natural cellular materials

such as trabecular bones, palms, and wood 8000

Energy dissipation per unit mass (J/kg)

Aporas® aluminium foam
Luffa sponge Cymate aluminium foam
100 (results from present study) 7000 Polymer foams
Strength (MPa)

Luffa sponge from present study

6000
10
5000
1 4000

3000
0.1
Other biological materials 2000
such as tomato, potato, and apple
0.01 1000
10 100 1000
Density (kg/m3) 0
0 50 100 150 200 250 300
Fig. 13 – Comparison between luffa sponge and other Density (kg/m3)
natural materials (Gibson et al., 2010).
1000
900 Luffa sponge (density: 44 ~ 47 kg/m )
the applications of natural cellular materials in general and
Nickel microlattice (density: 43 kg/m )
the luffa sponge in particular, especially when the light 800

Compressive stress (kPa)

weight is a key design requirement.
It is worth noting that human beings have been able to
discover and develop many materials, notably various metals,
metallic alloys, ceramics, plastics, as well as their compo-
sites, with performances significantly surpassing biological
materials (Bonderer et al., 2008; Zhang et al., 2011). When
light weight of the material is considered, various metallic/
ceramic/polymeric foams, honeycombs, and microlattices,
are designed and fabricated using those materials as base
materials. However, most of these man-made materials are
not environmentally friendly and have not been designed
with genuine concerns over their long-term sustainability
(Zhang et al., 2011). An efficient solution is to find an
alternative biological counterpart for those materials in use.
For this purpose, a comparative study has been carried out
below to show the merit of the luffa sponge material besides
its sustainable features.
As a cellular material, the luffa sponge is extremely light as
shown in Fig. 13. At less than 50 kg/m3, its density is about
one fifth of that of a typical aluminium foam (Alporass). For a
similar density (25–65 kg/m3), very few materials currently
exist such as Cymat aluminium foam (Z70 kg/m3), Ni–P
microlattices (Z0.9 kg/m3), silica aerogels (Z1.0 kg/m3), car-
bon nanotube aerogels (Z4.0 kg/m3), and polymer foams
(Z8.0 kg/m3). However, our research results indicate that,
when the luffa column is compressed longitudinally, the
amount of energy it can absorb per unit mass is comparable
to that of the aluminium foam (Idris et al., 2009; Shen et al.,
2010) and most of the polymer foams (Avalle et al., 2001), as
shown in Fig. 14(a). It should be noted that Alporass alumi-
nium foam has superior specific energy absorption than
other metallic foams when their density is lower than 300 kg
m3 (Ashby et al., 2000). The good specific energy absorption
of luffa sponge is attributed partially to its light base material
as well as a higher densification strain. Due to the high
strength-to-weight ratio of cellular materials, luffa sponge
can be used as a good packaging material and an excellent
energy dissipation material. It should be seen that the
average plateau stress of aluminium foam is 4–5 times higher
than the luffa sponge material. To show the efficiency of the
700
600
500
400
300
200
100
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Strain
Fig. 14 – Comparison of energy absorption capacity per unit
mass between luffa sponge and: (a) other man-made foams;
(b) Ni–P microlattice (Schaedler et al., 2011).
microstructures of luffa sponge material, we have also
compared the performance of the luffa sponge with other
cellular materials with similar density mentioned above.
Cymat foam has a strength of 40 kPa at a density of 70 kg m3
(Ashby et al., 2000). Aerogels exhibit a continuous increasing
of stress over strain under compression which is not compar-
able to luffa sponge material. Expended Polystyrene foams
have a strength of 200 kPa at a density of 45 kg m3 (Avalle
et al., 2001). The polyamide reinforced with modified poly-
phenylene and polystyrene foam has a strength of 800 kPa at
a density of 45 kg m3 (Avalle et al., 2001). As an important
example, the comparison with a recently invented metallic
microlattice of a similar density (Schaedler et al., 2011) shows
that the average plateau stress of the luffa sponge (about
350 kPa) is significantly higher than that of the Ni–P micro-
lattice (about 120 kPa), as shown in Fig. 14(b). It should be
noted that the strength decreases with density for any kind of
cellular materials. These comparisons indicate that the luffa
sponge material has better compressive strength than most
of other available cellular material of similar density such as
Cymat aluminium foam, expended Polystyrene foams and
Ni–P microlattices.
 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152
It should be noted that the luffa sponge material is a
cellular material with structural hierarchy as indicated in
Fig. 1. As mentioned previously, the luffa sponge column
specimen is composed of luffa fibres. According to other
research on the microstructures of luffa sponge material
(Boynard and D’Almeida, 2000; Ghali et al., 2009; Laranjeira
et al., 2006; Paglicawan et al., 2005; Tanobe et al., 2005), the
luffa fibres are hollow with many micro-tunnels. Those
microstructures have also been proven by our own investiga-
tions. Similar to other natural cellular material with similar
components, the cell walls of those micro-tunnels may be
fibrous themselves and consist of oriented cellulose nanofi-
brils in a hemicellulose and lignin matrix. Those hierarchical
levels required further investigation. However, a rough esti-
mation can be made for contributions of topology at different
hierarchical levels to the strength, plateau stress and energy
absorption capacities of luffa sponge material using the
available material properties from others research
(Paglicawan et al., 2005). The density is approximately
353 kg/m3 with a Young’s modulus of 1332 MPa (735%) and
tensile strength of 11.1 MPa (785%). The obtained relative
plateau stress versus relative density was compared with
other polymeric materials (Gibson and Ashby, 1982) as shown
in Fig. 15. The experimental data and empirical prediction
exhibit a similar trend with that of other polymeric materials.
It indicates that the relative strength are not superior to
others cellular material if only one level of structural hier-
archy is considered. The prediction formulae for other hier-
archical levels are not available because there is a lack of the
mechanical properties of the base composite materials as
well as its specific composite patterns of those base ingre-
dients. It is worth further research effort to understand and
bio-mimic the luffa sponge material with superior strength
and energy absorption capacities. We also find that the
severely crushed luffa sponge column (up to a nominal final
compressive strain of 95%) is able to recover to its original
length and shape (up to 98%) after it is submerged in water
and then dried. We are currently conducting further investi-
gation on this very interesting phenomenon.
Fig. 15 – Comparison of relative plateau stress between luffa
sponge and other polymeric materials (Gibson and Ashby,
1982).
151
As mentioned in Section 2.1, the luffa sponge used in our
experiments was obtained from pharmacies in Australia
which was sold as a bath sponge. There were two treatments
in manufacturing the bath sponge from natural luffa fruits
which may influence the mechanical properties of luffa
sponge, namely, bleaching and lateral squashing. Similar to
other cellulose based natural fibres such as cotton, bleach
would result in slight loss of mass so as to its mechanical
strength of untreated luffa sponge. The lateral squashing
would result in damage of the original luffa fibres. The
experimental data and empirical formula from our experi-
ment will underestimate the strength of untreated luffa
sponge material.
5. Conclusions
In this study, a series of compressive tests were conducted to
examine the stiffness, strength and energy absorption char-
acteristics of the luffa sponge material under quasi-static
compressive load. The Young’s modulus, compressive
strength, densification strain, plateau stress and energy
absorption capacity of the luffa sponge material have been
obtained for the first time. A set of empirical formulae have
been proposed to predict the mechanical properties of luffa
sponge material on the macroscopic scale. It should be noted
that there is a limitation of our current work, i.e., the luffa
sponge material used in our experiment was bleached and
laterally squashed initially. The effect of such pre-treatments
on the mechanical properties of the luffa sponge is not
available and is under further investigation. From our inves-
tigation, the following conclusions can be drawn.
(1) The stress–strain curves show a near constant plateau
stress over a long strain range, which is ideal for energy
absorption application. It could be used as an alternative
packaging material.
(2) The exponent from our experiments suggests that the
deformation of the luffa sponge is determined by the
coupling of axial compression/tension and bending of
luffa fibres.
(3) It is found that the luffa sponge material exhibits remark-
able stiffness, strength and energy absorption capabilities
that are comparable to those of some metallic cellular
materials such as aluminium foams and Ni–P microlat-
tices. The strength of luffa sponge is better than most of
other available cellular materials in the similar density
range such as expended Polystyrene foams and Ni–P
microlattices. As an ultra-light cellular material, it has
great potential to be used as an environmentally friendly
engineering material.
r e f e r e nc e
Ajiwe, V., Ndukwe, G., Anyadiegwu, I., 2005. Vegetable diesel fuels
from Luffa cylindrica oil, its methylester and ester-diesel
blends. Chemistry Class Journal 2, 1–4.
152 journal of the mechanical behavior of biomedical materials 15 (2012) 141 –152
Ashby, M.F., Evans, A.G., Fleck, N.A., Gibson, L.J., Hutchinson, J.W.,
Wadley, H.N.G., 2000. Metal Foams: A Design Guide. Butter-
worth-Heinemann, Warrendale.
Avalle, M., Belingardi, G., Montanini, R., 2001. Characterization of
polymeric structural foams under compressive impact loading
by means of energy-absorption diagram. International Journal
of Impact Engineering 25, 455–472.
Bal, K.E., Bal, Y., 2004. Gross morphology and absorption capacity
of cell-fibers from the fibrous vascular system of loofah (Luffa
cylindrica). Textile Research Journal 74, 241–247.
Bastawros, A.F., Bart-Simth, H., Evans, A.G., 2000. Experimental
analysis of deformation mechanisms in a closed-cell alumi-
nium alloy foam. Journal of the Mechanics and Physics of
Solids 48, 301–322.
Bonderer, L.J., Studart, A.R., Gauckler, L.J., 2008. Bioinspired
design and assembly of platelet reinforced polymer films.
Science 319, 1069–1073.
Boynard, C.A., D’Almeida, J.R.M., 2000. Morphological character-
ization and mechanical behavior of sponge gourd (Luffa
cylindrica)-polyster composite materials. Polymer-Plastics
Technology and Engineering 39, 489–499.
Chen, J.P., Lin, T.C., 2005. Loofa sponge as a scaffold for culture of
rat hepatocytes. Biotechnology Progress 21, 315–319.
Damiens, R., Rhee, H., Hwang, Y., Park, S.J., Hammi, Y., Lim, H.,
Horstemeyer, M.F., 2012. Compressive behavior of a turtle’s
shell: experiment, modeling, and simulation. Journal of the
Mechanical Behavior of Biomedical Materials 6, 106–112.
Demir, H., Atikler, U., Balköse, D., Tihminlioglu, F., 2006. The effect
of fiber surface treatments on the tensile and water sorption
properties of polypropylene-luffa fiber composites. Compo-
sites Part A 37, 447–456.
Ghali, L., Msahli, S., Zidi, M., Sakli, F., 2009. Effect of pre-treatment
of luffa fibres on the structural properties. Materials Letters
63, 61–63.
Gibson, L.J., Ashby, M.F., 1982. The mechanics of three-dimen-
sional cellular materials. Proceedings of the Royal Society of
London Series A 382, 43–59.
Gibson, L.J., Ashby, M.F., Harley, B.A., 2010. Cellular Materials in
Nature and Medicine. Cambridge University Press, Cambridge.
Idris, M.I., Vodenitcharova, T., Hoffman, M., 2009. Mechanical
behaviour and energy absorption of closed-cell aluminium
foam panels in uniaxial compression. Materials Science and
Engineering A 517, 37–45.
John, M.J., Thomas, S., 2008. Biofibres and biocomposites. Carbo-
hydrate Polymers 71, 343–364.
Johnsona, M., Walterb, S.L., Flinnc, B.D., Mayerc, G., 2010. Influ-
ence of moisture on the mechanical behavior of a natural
composite. Acta Biomaterialia 6, 2181–2188.
Laidani, Y., Hanini, S., Henini, G., 2001. Use of fiber luffa cylindrica
for waters traitement charged in copper. Study of the possi-
bility of its regeneration by desorption chemical. Energy
Procedia 6, 381–388.
Laranjeira, E., Carvalho, L.H.d., Silva, S.M.d.L., D’Almeida, J.R.M.,
2006. Influence of fiber orientation on the mechanical proper-
ties of polyester/jute composites. Journal of Reinforced Plas-
tics and Composites 25, 1269–1278.
Li, Q.M., Magkiriadis, I., Harrigan, J.J., 2006. Compressive strain at
the onset of the densification of cellular solids. Journal of
Cellular Plastics 42, 371–392.
Lu, G., Lu, G.Q., Xiao, Z.M., 1999. Mechanical properties of porous
materials. Journal of Porous Materials 6, 359–368.
Lu, G., Yu, T.X., 2003. Energy Absorption of Structures and
Materials. Woodhead Publishing Ltd, Cambridge.
Mayer, G., 2011. New toughening concepts for ceramic compo-
sites from rigid natural materials. Journal of the Mechanical
Behavior of Biomedical Materials 4, 670–681.
Mazali, I.O., Alves, O.L., 2005. Morphosynthesis: high fidelity
inorganic replica of the fibrous network of loofa sponge (Luffa
cylindrica). Anais da Academia Brasileira de Ciências 77, 25–31.
McCullough, K.Y.G., Fleck, N.A., Ashby, M.F., 1999. Uniaxial
stress–strain behaviour of aluminium alloy foams. Acta Bio-
materialia 47, 2323–2330.
Meyers, M.A., Chen, P.-Y., Lopez, M.I., Seki, Y., Lin, A.Y.M., 2011.
Biological materials: a materials science approach. Journal of
the Mechanical Behavior of Biomedical Materials 4, 626–657.
Miyoshi, T., Itoh, M., Mukai, T., Kanahashi, H., Kohzu, H., Tanabe,
S., Higashi, K., 1999. Enhancement of energy absorption in a
closed-cell aluminum by the modification of cellular struc-
tures. Scripta Materialia 41, 1055–1060.
Mostafa, S.A.E., Damiano, P., 2010. Multiscale structural design of
columns made of regular octet-truss lattice material. Interna-
tional Journal of Solids and Structures 47, 1764–1774.
Oboh, I.O., Aluyor, E.O., 2009. Luffa cylindrica—an emerging cash
crop. African Journal of Agricultural Research 4, 684–688.
Oboh, I.O., Aluyor, E.O., Audu, T.O.K., 2011. Application of luffa
cylindrica in natural form as biosorbent to removal of divalent
metals from aqueous solutions—kinetic and equilibrium
study. In: Einschlag, F.S.G. (Ed.), Waste Water—Treatment
and Reutilization. InTech, pp. 195–212.
Paglicawan, M.A., Cabillon, M.S., Cerbito, R.P., Santos, E.O., 2005.
Loofah fiber as reinforcement material for composite. Philip-
pine Journal of Science 134, 113–120.
Roble, N., Ogbonna, J., Tanaka, H., 2002. A novel circulating loop
bioreactor with cells immobilized in loofa (Luffa cylindrica)
sponge for the bioconversion of raw cassava starch to ethanol.
Applied Microbiology and Biotechnology 60, 671–678.
Schaedler, T.A., Jacobsen, A.J., Torrents, A., Sorensen, A.E., Lian, J.,
Greer, J.R., Valdevit, L., Carter, W.B., 2011. Ultralight metallic
microlattices. Science 334, 962–965.
Shen, J., Lu, G., Ruan, D., 2010. Compressive behaviour of closed-
cell aluminium foams at high strain rates. Composites Part B-
Engineering 41, 678–685.
Siqueira, G., Bras, J., Dufresne, A., 2010. Luffa cylindrica as a
lignocellulosic source of fiber, microfibrillated cellulose, and
cellulose nanocrystals. BioResources 5, 727–740.
Stanzl-Tschegg, S.E., Keunecke, D., Tschegg, E.K., 2011. Fracture
tolerance of reaction wood (yew and spruce wood in the TR
crack propagation system). Journal of the Mechanical Beha-
vior of Biomedical Materials 4, 688–698.
Tanobe, V.O.A., Sydenstricker, T.H.D., Munaro, M., Amico, S.C.,
2005. A comprehensive characterization of chemically treated
Brazilian sponge-gourds (Luffa cylindrica). Polymer Testing 24,
474–482.
Tavares, J., Israel, N., Rui, O., Wilton, S., Valderi, D., 2008. Nitrifica-
tion in a submerged attached growth bioreactor using Luffa
cylindrica as solid substrate. African Journal of Biotechnology
7, 2702–2706.
Wegst, U.G.K., 2011. Bending efficiency through property gradi-
ents in bamboo, palm, and wood-based composites. Journal of
the Mechanical Behavior of Biomedical Materials 4, 744–755.
Zampieri, A., Mabande, G.T.P., Selvam, T., Schwieger, W., Rudolph,
A., Hermann, R., Sieber, H., Greil, P., 2006. Biotemplating of
Luffa cylindrica sponges to self-supporting hierarchical zeolite
macrostructures for bio-inspired structured catalytic reactors.
Materials Science and Engineering C 26, 130–135.
Zhang, Z., Zhang, Y., Gao, H., 2011. On optimal hierarchy of load-
bearing biological materials. Proceedings of the Royal Society
B 278, 519–525.
Zhu, J., Wang, X., He, C., Wang, H., 2012. Mechanical properties,
anisotropic swelling behaviours and structures of jellyfish
mesogloea. Journal of the Mechanical Behavior of Biomedical
Materials 6, 63–73.