Fibers and Polymers 2017, Vol.18, No.

11, 2187-2195 ISSN 1229-9197 (print version)

DOI 10.1007/s12221-017-7418-y ISSN 1875-0052 (electronic version)

Vibration Damping of Flax Fibre-reinforced Polypropylene Composites

Md Zillur Rahman*, Krishnan Jayaraman, and Brian Richard Mace
Department of Mechanical Engineering, University of Auckland, Auckland 1142, New Zealand
(Received May 11, 2017; Revised September 6, 2017; Accepted September 11, 2017)

Abstract: This work investigates the effects of fibre content and fibre orientation on the damping of flax fibre-reinforced
polypropylene composites. Laminates of various fibre contents were manufactured by a vacuum bagging process; their
dynamic behaviour were then found from the vibration measurements of beam test specimens using an impulse hammer
technique to frequencies of 1 kHz. The frequency response of a sample was measured and the response at resonance was used
to estimate the natural frequency and loss factor. The single-degree-of-freedom circle-fit method and the Newton’s divided
differences formula were used to estimate the natural frequencies as well as the loss factors. The damping estimates were also
investigated using a “carpet” plot. Experiments were subsequently conducted on a range of samples with different fibre
volume fractions and orientations. The results show significant variations in natural frequencies and loss factors according to
the variations in fibre orientation. Composites containing 45 o, 60 o and 90 o fibre orientation exhibit approximately the same
natural frequencies. Composites with differing fibre orientations exhibit different loss factors for the various modes of
vibration, and the maximum loss factor is obtained for the case of 45 o fibre orientation, with the loss factor generally lying in
the range of 2-7 %. It was found that the loss factor increases with increasing frequency and decreases slightly with increasing
fibre content. These outcomes indicate that flax fibre-reinforced composite could be a commercially viable material for
applications in which noise and vibration are significant issues and where a significant amount of damping is required.
Keywords: Composite materials, Flax-polypropylene composites, Damping, Loss factor, Natural frequency

Introduction into polymeric matrix materials can have multifunctional

capabilities [6] such as vibration control, energy dissipation
In some areas, for example, the automotive industry and heat dissipation along with a high stiffness to weight
(electric/hybrid cars), aerospace and sport, reduction in the ratio.
mass of a structure is becoming increasingly important and To investigate the vibration damping of PFPCs (plant
desirable, as it relates directly to cost, performance and fibre-reinforced polymer composites), experimental dynamic
better fuel efficiency. However, reduction of mass can cause mechanical analysis (DMA) has been extensively used in the
a structure to be more vulnerable to audible noise and past decade [2,3,7-15]. For example, Duc et al. [2,3] measured
vibration. To attenuate the vibrations to a desirable level, the damping of flax, carbon and glass fibre-reinforced epoxy
damping materials may be added to a structure to dissipate composites and showed that flax fibre composites possess a
vibrational energy. Furthermore, increased environmental higher damping than do glass and carbon fibre composites.
consciousness, the reduction of available non-renewable Wielage et al. [7] also found that the loss factors of flax and
sources and the problems inherent to the disposal of waste hemp fibre-reinforced polypropylene (PP) composites are
materials, have all led to a growing interest in the use of greater than those of glass fibre composites. The loss factors
plant fibre based composites for such purposes. increase with the twist angle of the flax yarns and the crimp
One particular advantage of using reinforcing agents such in the flax fabrics; this is due to increasing intra-yarn and
as plant fibres in composites is their multi-scale structure. inter-yarn friction, respectively, as stated in [2,8]. Idicula et
These plant fibres are composed of elementary fibres bonded al. [9] studied the effect of fibre volume fraction on the
by a pectin matrix, where each of the elementary fibres can damping of short banana-sisal hybrid fibre-reinforced polyester
be considered as a composite structure. This structure consists composites and found that composites with a fibre volume
of stiff cellulose microfibrills reinforced in a hemicellulose fraction of 0.40 display the maximum damping. Joseph et al.
and lignin matrix [1]. These fibres are viscoelastic and [10] and Nair et al. [11] reported that the incorporation of
hierarchical in nature which contributes to the dissipation of short sisal fibre reduces the loss factors of the composites
energy [2,3]. Polymers also behave similar to viscoelastic when compared with the polypropylene and polystyrene
materials and show high damping when compared to metallic samples, respectively. Low frequencies results indicate less
materials [4]. Viscoelastic materials have the inherent ability damping of the banana fibre-reinforced polyester composites
to dissipate energy via converting vibration energy into heat compared to measurements at higher frequencies, as described
energy during mechanical deformation [5]. Furthermore, by Pothan et al. [12]. In another study, Guen et al. [13]
interfacial friction between fibre and matrix provides reported that the damping of polyol-treated flax fibre-
additional damping. Therefore, the embedding of plant fibres reinforced epoxy composites is higher than that of non-
treated flax fibre composites.
*Corresponding author: mrah082@aucklanduni.ac.nz In the aforementioned references, except for [1,6], DMA

2187
2188 Fibers and Polymers 2017, Vol.18, No.11 Md Zillur Rahman et al.

tests were generally conducted at various temperatures with when the modal overlap is less than one. In this approach,
a specific frequency less than or equal to 100 Hz. Etaati et al. the total contribution of all the off-resonant modes is
[16] performed measurements on hemp fibre-reinforced PP assumed to vary very slowly over the narrow frequency
composites with excitation frequencies of up to 200 Hz. The band around a specific resonance. It also provides more
maximum loss factor was found for 30 % by weight hemp robust estimates of loss factor because a number of estimates
fibre composites, and the damping exhibited a slight depen- can be found from responses at different frequencies around
dence on frequency below 30 Hz. However, it is necessary to each resonance and then averaged. In addition, for both
investigate a broader frequency range to characterise the techniques, frequency response is typically measured using a
dynamic behaviour of a material, in particular, the stiffness digital spectral analyser with a given frequency resolution
and loss factor, as audible frequencies range from 20 to which may also lead to inaccuracies. Thus, in this paper,
20,000 Hz. In order to reduce noise and vibration in natural frequencies and loss factors are estimated using the
transport, NFPCs can be considered as potentially useful SDoF circle-fit method [20] and the Newton’s divided
materials for interior application. In vehicles, for example, differences formula [21,22].
frequencies of up to a few kHz are often important, and the One further problem associated with experimental measure-
frequency range from 100-1000 Hz is particularly significant in ments reported in the literature [14-16,18,23,24] is that in
causing fatigue for both driver and passengers [17]. This those studies accelerometers were attached to the specimens.
range also includes much of the energy associated with However, the accelerometer and its lead add mass and
internally and externally radiated noise in transport such as damping to the structure and affect the estimates of natural
cars and trains. Hence, the ability to add damping over a frequencies and loss factors. Therefore, in this work, a non-
wider frequency range is important, and this provides the contacting transducer such as a laser vibrometer is used.
motivation in this paper to measure damping over a wider To date, a number of researches have been carried out on
frequency range of up to 1000 Hz. the vibration damping of PFPCs, but few studies have been
Some studies [14,15,18] have estimated the natural conducted to ascertain the structural damping levels of
frequencies and loss factors of NFPCs. Landro and Lorenzo PFPCs over a broader frequency range. The present work is,
[14,15] examined the dynamic behaviour of plant fibre mat therefore, an attempt to measure the dynamic characteristics
(flax, kenaf and hemp fibres were used in the mat, 50 % by of PFPCs over a broader frequency range. The dynamic
weight) reinforced PP composites over excitation frequencies characterisation concerns various composite beams of different
of up to 4 kHz. An increase in damping with frequency was lengths with three different volume fractions and five different
seen. Kumar et al. [18] reported only the first three natural orientations of flax fibre. The quality of the damping estimates
frequencies and the loss factors associated with short sisal from the “carpet” plots is also investigated.
and banana fibre-reinforced polyester composites for various
fibre lengths (3 mm, 4 mm and 5 mm) and contents (30 %, Experimental
40 % and 50 % by weight). They demonstrated that the
natural frequencies of the composites are not affected Materials
significantly by variations in fibre length. For a constant Polypropylene, a thermoplastic polymer is chosen as a
fibre length (3 mm), two kinds of damping trends were matrix material. Among the thermoplastic polymers (e.g.,
observed with an increase in banana and sisal fibre contents polypropylene, polyethylene and polystyrene), polypropylene
in the case of the first mode: for the former, damping is the most widely used in plant fibre composites due to its
decreases; for the latter, it increases. In those studies (i.e., low density, good mechanical properties, relatively high-
[14,15,18]), the estimation of loss factor was carried out temperature resistance, excellent processability and good
using the half-power bandwidth method, also known as the impact resistance. It is also an attractive choice for vibration
peak-picking method. This method produces an overestimation damping due to its glass transition temperature being below
of damping; this is due to the assumption that only a single room temperature [4]. Polypropylene shows the nature of
mode contributes to the response around each resonance free volume, amorphous and high flexibility of molecules
frequency. However, off-resonant modes can contribute a below the glass transition temperature. Moreover, it undergoes
significant amount to the total response at any resonance. molecular deformation and can dissipate vibration energy
The damping estimation errors can be up to 20 % or more [25]. Polypropylene has a loss factor of 0.06 [26,27] (similar
for a multiple-degree-of-freedom system [19] depending on to that found in this study) which is higher than widely-used
the mode number, modal overlap and loss factor. On the thermosets (e.g., epoxy, loss factor of 0.015 [28,29]). In
other hand, the single-degree-of-freedom (SDoF) circle-fit addition, PP has great potential for composite materials as it
method [20] significantly improves the accuracy of damping can be processed using traditional technologies, for example,
estimation compared to the half-power bandwidth method, extrusion, vacuum bagging, compression and injection
as it considers the frequency range in the close vicinity of a moulding.
natural frequency. This approach is particularly accurate Sheets of PP random copolymer (MOPLEN RP241G)
Vibration Damping of Flax/PP Composites Fibers and Polymers 2017, Vol.18, No.11 2189

Figure 2. Vacuum bagging technique.

Figure 1. Unidirectional stitched flax fabric.

with a melt flow rate of 1.5 g/10 min determined by ISO

1133 and a thickness of 0.38 mm were used as matrix
material. The polypropylene sheets were produced by Lyondell
Basell Industries and supplied by Field International Ltd.,
Auckland, New Zealand. The properties of PP are shown in Figure 3. A differential scanning calorimetry curve of neat
reference [30]. polypropylene.
Flax is chosen in the present study as a reinforcement
material. Among the various plant fibres (flax, hemp, jute,
ramie and kenaf), flax has high inherent vibration absorption to ensure all the air inside the vacuum bag could be drawn
capability [31,32] in addition to its high specific stiffness into a vacuum port. After sealing the material stack, the air
[33,34]. Flax fibres also offer a combination of low cost, was evacuated from inside the vacuum bag using a vacuum
light weight and high stiffness for applications [33,35,36], pump. The mould was subsequently placed inside the
such as infrastructures, automobiles and consumer products. Elecfurn (FAC 100) oven and heated to a temperature of
High-quality long flax yarns are available in fabric form [37]. 190 oC for 1 hour. After this, the mould was cooled to a room
Unidirectional flax fabric (FlaxPly UD180) with a nominal temperature of around 24 oC. The temperature-pressure
specific weight of 180 g/m2 and density of 1.42 g/cm3 was cycle (cycle 1) was applied according to reference [38]. The
used as reinforcement. A micrograph of the flax fabric is size of the panels was nominally 600 mm×500 mm, with a
shown in Figure 1. The flax fabric (42.5 yarns/cm (warp) target thickness of 3 mm and target fibre volume fractions of
and 3 yarns/cm (weft)) was supplied by Lineo, Meulebeke, 0.31, 0.40 and 0.50. Corresponding to each fibre composition,
Belgium. The weight distribution of flax fabric in the warp beams (450 mm×20 mm) with fibre orientations of 0 o, 30 o,
and weft directions was 95.5 % and 4.5 %, respectively, 45 o, 60 o and 90 o were cut from the panels using an
according to the manufacturer’s datasheet. The role of weft automatic saw. Neat PP samples were also manufactured for
(or fill) yarns is to keep the warp yarns together. The flax comparison with the respective flax/PP samples.
fibres are twisted to some degree within the yarns.
Impulse Hammer Technique
Manufacturing of Composites An impulse hammer technique was used to measure the
Flax fabrics were dried for 24 hours at 70 oC in a vacuum frequency response of flax/PP composite beams (300/350/
dryer (Squaroid duo-vac vacuum oven) to reduce the moisture 400 mm×20 mm×thickness) at an ambient temperature of
content. Vacuum bagging technique (Figure 2) was used to about 21 oC. One end of the composite beam was clamped to
manufacture the composite samples. This technique uses a fixed support and the other end was free to vibrate, as
atmospheric pressure (acts uniformly over the laminate) for illustrated in Figure 4. The excitation was provided by an
holding the laminate in place during the cure cycle. The impact hammer (PCB model: 086E80) with a soft tip at a
consolidation temperature of 190 oC was applied based on distance of 5 mm from the free end of the beam, and the
the differential scanning calorimetry curve of PP, as shown beam response was measured by a laser vibrometer (Polytec
in Figure 3, which was greater than the peak melting model: PDV-100) at a distance of 20 mm from the free end
temperature of 143.58 oC. Dry flax fabrics and PP sheets of the beam. The excitation and response signals were then
were interleaved by a hand lay-up process and placed on an processed by a spectrum analyser which gave the frequency
aluminium plate. Then, a peel ply was used to separate the response functions. Subsequent data processing were performed
breather from the laminate, and the breather was employed in MATLAB [39]. The estimation of natural frequencies and
2190 Fibers and Polymers 2017, Vol.18, No.11 Md Zillur Rahman et al.

each resonance was analysed. A circle was fitted through the

use of a least-squares error fit to the data when plotted in the
complex plane; the centre and the radius of the circle were
estimated. The Newton’s divided differences formula was
used to determine the location of the natural frequency (ωr)

Figure 4. Schematic diagram of experimental setup for an impulse

hammer technique.

loss factors was made on composite beams of lengths 300,

350 and 400 mm with different fibre volume fractions and
orientations. The results are reported for beams of differing
lengths to provide a range of resonance frequencies. Figure 6. Accelerance magnitude for flax/PP beams of different
fibre orientations and a fibre volume fraction of 0.40.
Frequency Response Functions
The accelerance (acceleration per unit force) was measured
over a frequency range to 1 kHz. Figures 5-7 show examples
of the magnitude of the measured accelerance for flax/PP
samples of various fibre volume fractions and orientations,
and a neat PP sample. There are a number of resonance peaks
for the various modes of vibration in the frequency range, with
the natural frequencies being determined by the stiffness and
length of the beam. The data in the vicinity of these peaks
was then post-processed to determine the natural frequencies
and the loss factors.

Modal Analysis Using the SDoF Circle-fit Method

The extraction of modal parameters such as natural Figure 7. Accelerance magnitude for flax/PP beams of different
frequencies and loss factors from measured frequency fibre orientations and a fibre volume fraction of 0.50.
responses (see Figures 5-7) was performed using the SDoF
circle-fit method [20]. An illustrative example is shown in
Figure 8.
A portion of the data in a narrow frequency range around

Figure 5. Accelerance magnitude for flax/PP beams of different Figure 8. Modal circle for extracting natural frequency and loss
fibre orientations and a fibre volume fraction of 0.31, and a neat factor:* Estimated natural frequency (indicated by an arrow) and o
PP beam. discrete frequency data, angles are in degrees.
Vibration Damping of Flax/PP Composites Fibers and Polymers 2017, Vol.18, No.11 2191

and its value corresponding to the maximum rate of change specimen: 75 mm×12.7 mm×the thickness of the sample)
of phase [21,22]. The loss factor (ηr) was estimated from the are given.
frequency response measurements at frequencies ωa and ωb As the angle between the longitudinal and fibre axes
above and below the natural frequency using equation [20] increases, the drop in natural frequencies is substantial.
as However, the rate of decrease in natural frequencies reduces
2 2
ωa – ωb when the angle between them is greater than 30 o. Increasing
ηr = ---------------------------------------------------
- (1) the angle of fibres from 0 o to 60 o reduces the natural
2⎛ ⎛ θa⎞ ⎛ θb⎞ ⎞ frequency by 63.72 % (from 21.64 Hz to 7.85 Hz) in the
ω r tan ----- – tan -----
⎝ ⎝ 2⎠ ⎝ 2 ⎠⎠
where θa,b are the angles between the radii from the centre of
the circle to the natural frequency and the frequency response
at the chosen frequencies ωa and ωb, respectively. The mean
loss factors were then calculated from 100 estimates by
considering 20 data points, 10 data points below the natural
frequency and 10 data points above the natural frequency.

Results and Discussion

Natural Frequency
The variation of natural frequency with fibre orientation
for different fibre volume fractions are shown in Figures 9-
11. In general, the natural frequencies of the neat PP beam Figure 10. Variation of natural frequency with fibre orientation for
are lower than those of the composite beams, and this is flax/PP beams of a fibre volume fraction of 0.40, and neat PP
expected as the stiffness of the neat PP sample is lower. In beams.
regard to the fibre orientation of 0 o samples, the natural
frequencies increase with higher fibre content. This is due to
the presence of a higher proportion of fibres producing
greater stiffness, and hence the natural frequency is increased.
For the composite orientations of 30 o, 45 o and 60 o, the
natural frequency either decreases by a small amount or is
almost constant with increasing fibre content.
This is consistent with the results obtained from static
three-point bend tests (see Figure 12); the flexural moduli
decreased somewhat for these fibre orientations with
increasing fibre content. The flexural moduli were measured
in accordance with ASTM D760-10 using an Instron 5567
with a 10 kN load cell. The average value and standard error
Figure 11. Variation of natural frequency with fibre orientation for
obtained from at least five specimens (nominal size of each
flax/PP beams of a fibre volume fraction of 0.50, and neat PP
beams. *Natural frequency of bending-twisting coupled modes.

Figure 9. Variation of natural frequency with fibre orientation for

flax/PP beams of a fibre volume fraction of 0.31, and neat PP Figure 12. Flexural modulus for flax/PP samples of different fibre
beams. volume fractions and orientations, and a neat PP sample.
2192 Fibers and Polymers 2017, Vol.18, No.11 Md Zillur Rahman et al.

Figure 13. Accelerance magnitude for different flax/PP beam Figure 14. Variation of loss factor with frequency for flax/PP
lengths of a fibre volume fraction of 0.50. beams of different fibre orientations and a fibre volume fraction of
0.31, and neat PP beams.

case of the first mode of a sample with a fibre volume

fraction of 0.31. Incorporation of fibres (Vf = 0.50) into PP
increases the natural frequency of the 0 o fibre oriented
samples from approximately 7.59 Hz to approximately
26.95 Hz. The highest natural frequency occurs for 0 o fibre
orientation because the fibres are stiffer in tension than the
matrix. Kumar et al. [18] also reported that the more rigid
specimens have higher natural frequencies. The lowest
natural frequency occurs for 90 o fibre orientation in the
instance of two distinct fibre volume fractions of 0.31 and
0.40; in contrast, in the case of 0.50 fibre volume fraction,
the maximum natural frequency occurs at 0 o fibre orientation
and the minimum at 60 o, although the variation between
Figure 15. Variation of loss factor with frequency for flax/PP
60 oand 90 o is negligible. beams of different fibre orientations and a fibre volume fraction of
The magnitude of the peak value of the response at 0.40, and neat PP beams.
resonance decreases with increasing beam length, as
illustrated in Figure 13. The modal density increases with
increasing length. A bending-twisting mode (see Figure 7)
appears for beam lengths of 300 mm, 350 mm, 400 mm, and
a fibre volume fraction of 0.50 and orientation of 30 o. It
originates from the coupling between bending and twisting
when fibre orientation is different from 0 o or 90 o.

Loss Factor
Effect of Fibre Content
Figures 14-16 show the variation of mean loss factor with
frequency of flax/PP composite beams for fibre volume
fractions of 0.31, 0.40 and 0.50, respectively, and neat PP
beams. Linear fits are also shown for each set of data. Figure 16. Variation of loss factor with frequency for flax/PP
The loss factor of neat PP is approximately 0.06 at low beams of different fibre orientations and a fibre volume fraction of
frequencies. This value matches with the value reported by a 0.50, and neat PP beams. *Loss factor of bending-twisting coupled
number of authors [26,27]. As the frequency increases, the modes.
loss factor of neat PP increases, rising to 0.090 at approximately
500 Hz. despite a decrease in the amount of the viscoelastic polymer.
With increased fibre contents of 40 % and 50 % by In general, as fibre content increases, the loss factor decreases.
volume, flax/PP samples exhibit approximately the same However, no significant difference in the loss factors was
loss factors (loss factors lie predominantly in the range of 4- observed with the addition of an increase in the amount of
6 % for the fibre orientations of 30 o, 45 o, 60 o and 90 o, and flax fibre. This is probably due to the existence of more
2-2.5 % for the fibre orientation of 0 o) as 31 % by volume, fibres offering more fibre/matrix interfaces. This results in
Vibration Damping of Flax/PP Composites Fibers and Polymers 2017, Vol.18, No.11 2193

an increase in the number of energy dissipation sites. This is

also supported by the statement of Chauhan et al. [40] that
increasing filler content can contribute to the creation of
large number of interfacial areas which increases vibrational
energy dissipation. A similar observation was reported by
Liang and Tjong [41]. Moreover, this is due to the internal
structure of the fibre which induces high internal friction
[23]; in particular, this occurs between the cellulose and
hemicellulose in each wall and the friction between the cell
walls [2,3,8]. The addition of plant fibres also contributes to
energy dissipation, even at ambient temperature, and results
in the significant damping of the composites over a wide
Figure 17. Variation of loss factor with fibre orientation for flax/
range of frequency [14]. In addition, high damping can be
PP beams of a fibre volume fraction of 0.31, and neat PP beams.
attributed to the presence of voids (e.g., the intrinsic porosity
of plant fibres) and the viscoelastic nature of matrix and
reinforcement.
The loss factor of composites is, however, lower than that
of neat PP because of the lower amount of viscoelastic PP.
This coincides with the trend reported by Chandra et al. [42]
that high content of resin leads to higher damping due its
viscoelastic nature. The composite samples are stiffer than
the neat PP resulting in more stored energy for the same
deformation. The higher stored energy leads to lower
damping in the composites as compared to the neat PP. The
stiffness is strongly dependent on the presence of fibres,
while the loss factor is dominated by the matrix; this results
in decreased loss factors due to the incorporation of flax Figure 18. Variation of loss factor with fibre orientation for flax/
fibres into PP. A similar observation was also reported by PP beams of a fibre volume fraction of 0.40, and neat PP beams.
other authors [3,16,40,42,43]. However, interfacial damping
ameliorates this. When compared to the base PP samples, at
a fibre volume fraction of 0.31, the flexural stiffness of the
composite material increases by about 270 % (from 1.07
GPa to 3.96 GPa, as shown in Figure 12), whereas the loss
factor decreases by about 30 % (from 0.076 to 0.054) in the
case of 30 o fibre oriented samples.
Effect of Fibre Orientation
Figures 17-19 illustrate the variation of mean loss factor
with fibre orientation of flax/PP composite beams for fibre
volume fractions of 0.31, 0.40 and 0.50, respectively, and
neat PP beams. Notably, these variations of loss factor with
fibre orientation are also due to the effect of different natural Figure 19. Variation of loss factor with fibre orientation for flax/
frequencies. However, in these particular cases, the effect of PP beams of a fibre volume fraction of 0.50, and neat PP beams.
the different natural frequencies is not mentioned to show *Loss factor of bending-twisting coupled modes.
the trend of loss factor with fibre orientation only.
The effect of reinforcement orientation on the damping of fibre volume fractions, and the damping is found to lie in the
composites is noticeable. The 0 o fibre oriented samples range of 4-7 % for this fibre orientation. A similar trend was
demonstrate sharper resonance peaks (see Figures 5-7), observed by Gao et al. [44] and Adams and Maheri [45] in
while 30 o, 45 o, 60 o and 90 o fibre oriented samples exhibit the cases of carbon fibre-reinforced epoxy, and carbon and
broader peaks with the exception of the first two peaks. This glass fibre-reinforced plastic composites, respectively. This
indicates that the damping is greater for such orientations. is because the total energy is dominated by the in-plane
The loss factor increases with increasing fibre orientation shear strain energy [46]; the in-plane shear strain energy is
up to 45 o, and it can then be seen to decrease slightly as the maximised at this fibre orientation in flax/PP composites.
fibre orientation increases to above 45 o. Maximum damping The maximum loss factor of 0.071 is observed for a 300 mm
is obtained for the fibre orientation of 45 o in the case of all beam (Vf =0.31 and θ =45 o) at approximately 500 Hz. The
2194 Fibers and Polymers 2017, Vol.18, No.11 Md Zillur Rahman et al.

average loss factor for the 45 o fibre oriented samples show a

significant increase in loss factor of around 150 % (from
0.021 to 0.052) when compared to the 0 o fibre oriented
samples. At very low frequency, fibre orientations other than
0 o fibre orientation exhibit almost the same damping. The
damping is minimal and the stiffness is at its maximum at
0 o, i.e., in the fibre direction. This is reasonable, since the
fibres play a dominant role at 0 o fibre orientation and fibres
exhibit less damping and a much higher stiffness than the
matrix material. As the orientation angle increases, the
general trend is for the damping to increase (up to a certain Figure 20. Damping “carpet” plot (beam length: 300 mm, V =0.50,
f
angle) and the stiffness to decrease. The bending-twisting θ =0 o and mode 1).
modes exhibit a loss factor of 0.032 on average; this is less
than other modes of the same sample.
Damping increases when frequency increases for a given
fibre volume fraction. A maximum increase of loss factor of
60 % (0.042 to 0.067) is found for a fibre volume fraction of
0.31 and orientation of 60 o. The increase is in the range of
25-60 % over the frequency range to 1 kHz for all fibre
volume fractions and orientations. The increase in damping
with frequency may come from fibres and/or fibre/matrix
interactions.
The fibre orientation (30 o, 45 o, 60 o and 90 o) of the
sample affects the loss factor more than the fibre content
(31 %, 40 % and 50 % by volume). Overall, the loss factor Figure 21. Damping “carpet” plot (beam length: 300 mm, V =0.50,
f

of flax/PP composite samples is in the range of 2-7 % for all θ =0 o and mode 2).
studied fibre volume fractions and orientations. It can be
inferred from the above discussion that plant fibres possess a and 50 % by volume, respectively. In relation to this deviation,
significant vibration damping capacity. some possible measurement errors include measurement
noise, clamping pressure, air damping and non-uniformity in
Diagnosis of the Quality of the Damping Estimates the laminate (voids, variations in thickness and improper
The loss factor estimated using equation (1) depends on bonding).
the chosen frequencies ωa and ωb in addition to the natural
frequency (ωr). Different choices of ωa and ωb give different Conclusion
estimates of loss factor. The variation in the individual
estimates of loss factor for many combinations of the The effects of the fibre content and fibre orientation on the
selected frequencies can be seen from the damping “carpet” damping of NFPCs were estimated from vibration measure-
plots, with two examples shown in Figure 20 and Figure 21. ments. The SDoF circle-fit method and the Newton’s
The plots are presented for the first and second modes of a divided differences formula provide accurate estimates of
beam (beam length=300 mm, Vf =0.50 and θ =0 o) using all natural frequencies and loss factors. Changes in fibre content
combinations of the 20 selected frequencies below and and orientation produce different natural frequencies and
above the natural frequency. This gives rise to 100 estimates loss factors for the same geometry and boundary condition.
of loss factor from which are obtained the mean and This offers an additional freedom for the design of a composite
variance of the estimates. For a perfect case, the surface laminate-allowing for the prospect of alternating the fibre
ought to be smooth, flat and level. The surfaces of the plots orientation to affect the structure’s stiffness and damping.
are almost horizontal, smooth and flat, apart from some tilt This knowledge also makes the use of these materials attractive
and unevenness. The coefficients of variation of the 100 as it is feasible to increase damping without increasing mass
estimates are 1.76 % (where mean loss factor of 0.017 and or changing geometry.
standard deviation of 0.0003) and 1.58 % (where mean loss Of all the parameters, fibre orientation has the most
factor of 0.019 and standard deviation of 0.0003) for mode 1 significant impact on the damping. At each fibre volume
and 2, respectively. These are small and indicate that the fraction, the loss factor increases up to 4-7 % for fibre
estimates are of good quality. On the whole, the coefficients orientation up to 45 o and then decreases slightly. The loss
of variation are in the range of 1.04-9.92 %, 1.40-10.86 % factor generally lies in the range of 2-7 %, irrespective of
and 0.43-11.11 % in the case of fibre content of 31 %, 40 % fibre volume fraction and fibre orientation. The increase in
Vibration Damping of Flax/PP Composites Fibers and Polymers 2017, Vol.18, No.11 2195

damping is in the range of 25-60 % over the frequency range 20. D. J. Ewins, “Modal Testing: Theory, Practice and
of 1 kHz. The damping “carpet” plots indicate the quality of Application”, pp.310-318, Research Studies Press Ltd.,
the estimates of loss factor. The coefficient of variation of Baldock, Hertfordshire, England, 2000.
the loss factor estimated at each resonance is in the range of 21. J. Silva and N. Maia, J. Sound Vib., 124, 13 (1988).
0.43-11.11 %. 22. N. Maia and J. Silva, “Theoretical and Experimental
Given this, it can be concluded that the high damping Modal Analysis”, pp.219-227, Research Studies Press
capability of flax fibre-reinforced composite materials with Ltd., Baldock, Hertfordshire, England, 1997.
low weight make these materials appropriate for applications in 23. M. Assarar, W. Zouari, H. Sabhi, R. Ayad, and J.-M.
cases where low weight and high damping are sought. Berthelot, Compos. Struct., 132, 148 (2015).
24. K. Cheour, M. Assarar, D. Scida, R. Ayad, and X.-L.
Acknowledgements Gong, Compos. Struct., 152, 259 (2016).
25. V. Geethamma, R. Asaletha, N. Kalarikkal, and S. Thomas,
The first author acknowledges the financial support Resonance, 19, 821 (2014).
provided by the Department of Mechanical Engineering, 26. I. C. Finegan, G. G. Tibbetts, and R. F. Gibson, Compos.
University of Auckland, New Zealand. Sci. Technol., 63, 1629 (2003).
27. C. Subramanian, S. B. Deshpande, and S. Senthilvelan,
References Adv. Compos. Mater., 20, 319 (2011).
28. R. Chandra, S. P. Singh, and K. Gupta, J. Compos. Tech.
1. Y. Li, Y. Luo, and S. Han, J. Biobased Mater. Bioenergy, 4, Res., 25, 1 (2003).
164 (2010). 29. R. Chandra, S. P. Singh, and K. Gupta, J. Sound Vib., 262,
2. F. Duc, P. E. Bourban, and J. A. E. Manson, Compos. Sci. 475 (2003).
Technol., 102, 94 (2014). 30. M. Z. Rahman, K. Jayaraman, and B. R. Mace, Polym.
3. F. Duc, P. E. Bourban, C. J. G. Plummer, and J. A. E. Compos., doi:10.1002/pc.24486 (2017).
Manson, Compos. Pt. A-Appl. Sci. Manuf., 64, 115 (2014). 31. “Flaxtape Technical Guide”, Available at http://www.
4. D. D. L. Chung, J. Mater. Sci., 36, 5733 (2001). lineo.eu/#!products accessed 02 February 2015.
5. D. I. Jones, “Handbook of Viscoelastic Vibration Damping”, 32. L. Pil, F. Bensadoun, J. Pariset, and I. Verpoest, Compos.
p.7, John Wiley & Sons Ltd., Chichester, West Sussex, Pt. A-Appl. Sci. Manuf., 83, 193 (2016).
England, 2001. 33. L. Yan, N. Chouw, and K. Jayaraman, Compos. Pt. B-Eng.,
6. R. F. Gibson, Compos. Struct., 92, 2793 (2010). 56, 296 (2014).
7. B. Wielage, T. Lampke, H. Utschick, and F. Soergel, J. 34. A. Mohanty, M. Misra, and G. Hinrichsen, Macromol.
Mater. Process. Technol., 139, 140 (2003). Mater. Eng., 276, 1 (2000).
8. F. Duc, P. E. Bourban, and J. A. E. Manson, J. Reinf. Plast. 35. K. L. Pickering, M. G. A. Efendy, and T. M. Le, Compos.
Compos., 33, 1625 (2014). Pt. A-Appl. Sci. Manuf., 83, 98 (2015).
9. M. Idicula, S. K. Malhotra, K. Joseph, and S. Thomas, 36. D. B. Dittenber and H. V. GangaRao, Compos. Pt. A-Appl.
Compos. Sci. Technol., 65, 1077 (2005). Sci. Manuf., 43, 1419 (2012).
10. P. Joseph, G. Mathew, K. Joseph, G. Groeninckx, and S. 37. “LINEO-advanced Flax”, Available at http://www.lineo.eu/
Thomas, Compos. Pt. A-Appl. Sci. Manuf., 34, 275 (2003). accessed 05 October 2016.
11. K. Nair, S. Thomas, and G. Groeninckx, Compos. Sci. 38. S. Kazmi, R. Das, and K. Jayaraman, J. Mater. Process.
Technol., 61, 2519 (2001). Technol., 214, 2375 (2014).
12. L. A. Pothan, Z. Oommen, and S. Thomas, Compos. Sci. 39. MATLAB R2013b. The MathWorks Inc., Natick,
Technol., 63, 283 (2003). Massachusetts, 2013.
13. M. J. L. Guen, R. H. Newman, A. Fernyhough, and M. P. 40. S. Chauhan, A. Karmarkar, and P. Aggarwal, J. Appl.
Staiger, Compos. Pt. A-Appl. Sci. Manuf., 67, 37 (2014). Polym. Sci., 114, 2421 (2009).
14. L. D. Landro and W. Lorenzi, J. Biobased Mater. Bioenergy, 41. J. Liang, R. Li, and S. Tjong, J. Thermoplast. Compos.
3, 238 (2009). Mater., 13, 12 (2000).
15. L. D. Landro and W. Lorenzi, Macromol. Symp., 286, 145 42. R. Chandra, S. P. Singh, and K. Gupta, Compos. Struct.,
(2009). 46, 41 (1999).
16. A. Etaati, S. A. Mehdiza-deh, H. Wang, and S. Pather, J. 43. K. Joseph, S. Thomas, and C. Pavithran, J. Reinf. Plast.
Reinf. Plast. Compos., 33, 330 (2013). Compos., 12, 139 (1993).
17. T. Dias and R. Monaragala, Meas. Sci. Technol., 17, 2499 44. Y. Gao, Y. Li, Y. Hong, H. Zhang, and X. He, Polym.
(2006). Polym. Compos., 19, 119 (2011).
18. K. S. Kumar, I. Siva, P. Jeyaraj, J. T. W. Jappes, S. C. 45. R. Adams and M. Maheri, J. Alloys Compd., 355, 126
Amico, and N. Rajini, Mater. Des., 56, 379 (2014). (2003).
19. J.-T. Wang, F. Jin, and C.-H. Zhang, Soil Dyn. Earthq. 46. S. J. Hwang and R. F. Gibson, J. Compos. Mater., 26, 2585
Eng., 39, 138 (2012). (1992).