Abstract
Natural fibre composites attract industries because of their low density, low cost and the specific mechanical properties they possess in comparison to synthetic fibres. In this work, the randomly oriented sisal fibre–reinforced polypropylene composites are fabricated using extrusion–injection moulding technique. The aim of this study is to experimentally investigate the effect of fibre weight fraction (0%–30% in step of 10%) on vibrational damping and acoustic characteristics. The impulse hammer excitation technique is used to evaluate the free vibrational characteristics, namely, natural frequency and damping. An impedance tube is used in evaluating the acoustic properties, namely, sound absorption coefficient and transmission loss. Experimental results reveal that increase in fibre loading significantly alter the vibrational and acoustic response of the polypropylene composite. Modal analysis shows that incorporation of sisal fibres by 30 wt.% to polypropylene made the natural frequency superior when compared with other compositions. However, damping becomes worse with higher fibre content. In case of acoustic properties, incorporation of fibres at higher fraction enhances the sound absorption coefficient and transmission loss. Experimental results drive the research in development of such new materials system towards the application of vibration and sound diminutions.
Introduction
The natural fibres reinforcement composites are a rising trend even today. Properties such as low density, easy to fabricate, good specific mechanical properties, degradable and low cost made them a coveted material in several industrial sectors. Synthetic fibres are acquired from petrochemical resources. Petroleum crisis and environmental sustainability made significant importance to natural fibre composite. In the meantime, the vegetable fibres have wide range of properties to become engineering materials. 1 India happens to be the second largest producer and consumer of agricultural fibres. Diverse agro-climatic conditions make it possible for the country to produce an array of natural fibres. Natural fibres such as coir, sisal, jute, banana and kenaf are available in abundance in the country. 2
Numerous studies have been performed to investigate the experimental modal analysis on different natural fibre–reinforced thermoset polymer (epoxy or polyester) composites. Vibration characteristics of sisal and cotton polyester, 3 carbon and glass hybrid epoxy, 4 hybrid banana–jute polyester, 5 flax and linen epoxy, 6 Sansevieria and coconut sheath polyester, 7 jute fabric/polyester, 8 glass/epoxy, 9 short banana and coconut sheath polyester 10 and woven glass/carbon epoxy 11 were investigated by various researchers. The parameters taken into consideration for these investigations were fibre content; ordination of fibre; staking sequence; secondary filler reinforcement as nano clay and red mud; weaving architecture of woven fibre; treatment of fibre as untreated, alkaline and silane and end conditions of beam as cantilever, fixed-fixed and simply supported. The improvement in natural frequency of composites observed for increasing weight fraction of fibre and secondary filler reinforcement. The natural frequency was observed to be affected by the changing in the layer sequence and the orientation of the fibres as well. Bennet et al. 7 revealed that the C/S/C stacking sequence has a superior natural frequency as compared with all other sequences in Sansevieria cylindrica (S) and coconut sheath (C)/polyester composites. Senthil Kumar et al. 10 investigated the effect of stacking sequence and reported that coconut/banana/coconut layering pattern showed the higher damping. As an influence of weaving architecture, braided fabric-type weaving pattern has shown higher natural frequencies and loss factors than knitted. Natural frequencies in basalt/epoxy are higher for woven fabric than unidirectional for fixed–fixed boundary conditions compared with other. 9 The effect of fibre treatment declined natural frequency and damping factor. 12 Some studies focusing on vibration damping characterization of different natural fibre–reinforced thermoplastic composites are reported in literature. Etaati et al. 13 compounded the noil hemp/polypropylene (PP) composites with varying fibre contents and a coupling agent to estimate vibration behaviour. At 30 wt.% and in the absence of the coupling agent, the maximum damping ratio is achieved. Chauhan et al. 14 measured damping coefficients decreased by free vibration method of wood flour–filled PP composites. The increase in filler load decreases the damping coefficients, but it is noted that compatalizer has no effect on it. Rahman et al. 15 investigated damping studies of flax fibre–reinforced PP composites from vibration measurements. Laminated beam specimens were manufactured by a vacuum bagging process with various fibre contents (volume fraction of 31%, 40% and 50%) and fibre orientations (45°, 60° and 90°). Fibre orientation has a more prominent impact on the damping than fibre content. Loss factor found was in the range of 2%–7% for flax/PP composite. In another research, 16 he investigated the effect of mode of vibration on damping property. The loss factor bending modes lie in the range of 1.7%–2.2%, whereas for twisting modes, it is 4.8%.
In addition to vibration damping of material, structural engineers have to look of acoustic properties of material to control the noise. Sound absorption coefficient (SAC) and transmission loss (TL) are the properties of look of. Traditionally used materials are polyurethane foams, minerals and glass fibres which are non-biodegradable, hazardous to human health and expensive. As an alternative, researchers have started to use different natural fibres such as ramie, jute, flax, kenaf, sisal, rice straw and oil palm, which are obtained from natural resources. Yang and Li 17 investigated the sound absorption properties of ramie, jute and flax (natural fibres) and glass and carbon (synthetic fibres). It was found that the noise reduction coefficient of ramie, jute and flax is nearly two times higher than that of glass and carbon fibre. Few studies on the acoustic properties of natural fibre–reinforced thermoset and thermoplastic resin composites are also found. Prabhakaran et al. 18 fabricated the flax/epoxy and glass/epoxy composites by using vacuum-assisted resin transfer moulding. The investigation was carried out by using an impedance tube test setup for measurement of SAC. It was found that the maximum SAC for flax/epoxy is of 0.45 at 1000 Hz which is 20%–25% higher than glass/epoxy. SAC of different natural fibre composites such as rice straw/PP, kenaf/PP, oil palm empty fruit bunch (OPEFB)/zein, sisal/polylactic acid (PLA),19–22 ramie/PLA, 23 and polyester fibre/polyethylene 24 was investigated by various researchers. Furthermore, investigation of TL characteristics of olive stone/P, 25 jute/natural rubber latex26,27 and flax/PP 28 was carried out. All the above-mentioned researchers study the effect of different parameters such as frequency, fibre content, fibre treatment and composite thickness and found that SAC and TL improve up to certain level of parameters.
However, comparatively, very few studies have reports on the vibration damping behaviour of natural fibre–reinforced thermoplastic compared with thermoset composites. Also, very limited studies have been reported on the acoustic properties of natural fibre composites. Moreover, vibro-acoustic properties of sisal fibre–reinforced PP composites have not been previously reported. This study focused to investigate vibration damping of sisal/PP composites with the influence of fibre content. The analytically evaluated natural frequencies are reported in order to show a comparison with the measured natural frequencies. In addition, TL properties along with SAC were measured.
Materials and methods
Materials
Sisal fibre (Agave sisalana) produced through decortication method is supplied by Tokyo Engineering Corporation, India. Studied sisal fibre has 66%–72% cellulose, 12% hemicellulose, 10%–14% lignin and 11% moisture content while proceeding manufacturing. The physiomechanical properties of sisal fibre are listed in Table 1. PP of grade H110 MA is provided by Reliance Industries Ltd., India. The properties of PP available in datasheets are density of 0.900 g/cc, melt flow index (MFI) of 11 g/10 min, tensile strength of 36 MPa and elongation at yield of 10%.
Physiomechanical properties of sisal fibre.
Composite fabrication
Neat polymer and composite samples are produced through twin screw extrusion–injection moulding process. Chopped fibres (3–5 mm long) are oven dried at 60°C for 4 h and weighed in batches of 10, 20 and 30 wt.% to the matrix. Weighed fibres are then mixed with PP pellets and compounded using twin screw extruder at a temperature of 165°–210°C. Notations followed in sample preparation are listed in Table 2. The compounded materials are then granulated using a scrap granulator followed by the injection moulding (ASTM E756- and E1050-type specimens). The injection moulding machine (model-OMEGA 80W; Make Ferromatik Milacron) is employed in this study with the moulding temperature in the range of 175°C–195°C and a pressure of 50 bar.
Notations for composite fabrication.
Experimental studies
Experimental modal analysis
Impulse excitation test is conducted on the fabricated samples (200 × 20 × 3 mm) as per ASTM E756 recommendations (shown in Figure 1(a)). This test is used to estimate dynamic properties of composite materials such as the natural frequency and allied damping ratio. A graphic of a free vibration test setup with fixed-free boundary condition is shown in Figure 1(b). The equipment used are impact hammer (Make: PCB Piezoelectric, model 086C03), miniature transducer (Brüel & Kjaer model 352B10) and data acquisition (Brüel & Kjaer with Photon+ software). The process to acquire frequency response function (FRF) curve is used as reported in our earlier work. 29 Equation (1) was used to estimate damping ratio by half-power bandwidth method
where

(a) Test specimen and (b) test setup for experimental modal analysis.
Natural frequency of these composites is also calculated using the equation given in the following. 30 Assuming that the composites are quasi-isotropic material
where ωi is the natural frequency (rad/s), E is the flexural modulus (N/m2), A is the cross-sectional area (m2), ρ is the density of composite (kg/m3), I is the moment of inertia (m4), L is the free length of specimen (m), β = 1.875, 4.694 and 7.854 for mode 1, mode 2 and mode 3, respectively, at fixed-free condition.
Acoustic properties
SAC measurement
The SAC of the sisal/PP composites is measured using an impedance tube setup (BSWA Technology Co., Ltd, China) with the two microphones shown in Figure 2. Transfer function method is used to measure SAC as per ASTM E1050. 31 Two microphones are fixed to acquire pressure which is produced by a sound source near the sample. VA-Lab IMP can accurately separate the incident wave from the reflecting wave and calculate the absorption coefficient. The measurement is carried out using impedance tubes with diameters of 100 and 30 mm for measuring frequencies from 63 to 2000 and 500 to 6300 Hz, respectively, at 25°C and 60% relative humidity. In order to eliminate the error due to difference in microphone phase, microphones are interchanged in each measurement. Figure 3 shows that the sisal/PP composite samples adhere with tube internal diameter and thickness.

Sound absorption coefficient test system (using MC3522 with built-in power amplify).32

SAC test samples of sisal/PP composites.
TL measurement
TL of sisal/PP composites is measured using the same impedance tube setup (BSWA Technology Co., Ltd, China). The schematic diagram of the experimental setup is shown in Figure 4. TL measurement utilizes a transfer matrix method for the calculation, and data are acquired from four different microphones. As illustrated in Figure 4, two microphones are placed between the source and specimen and remaining two is in opposite side. The impedance tube (diameter 100 mm) which can measure the frequency ranges between 50 and 1600 Hz is utilized. In each measurement, the reading is taken with the extension tube in open-ended and in closed-ended conditions. After finishing the three measurements, the average result is referred.

Setup of transmission loss measurement system (using MC3242, PA50). 32
Results and discussion
Natural frequency
Vibration damping characteristics of sisal/PP is carried out by increasing the weight fraction of fibre. The effect of sisal fibre loading on the natural frequency and damping ratio of composite samples are investigated. The 20-mm-long specimen is rigidly fixed and impacted at three equidistant points by an impact hammer. The measurements taken are force magnitude and the displacement signal in time domain. Then, these time domain data are processed using Photon+ software to get the FRF curve. The coherence value close to 1 ensures the received signal data quality. Figure 5 displays an FRF curve for S10 composites with different hammer points such as H1, H2 and H3. The first three peaks in each curve which are visible from the FRF are three modes of natural frequency. Similar FRF curves obtained for S20 and S30 composites are shown in Figures 6 and 7, respectively.

FRF curve of S10 composite with hammer points H1, H2 and H3.

FRF curve of S20 composite with hammer points H1, H2 and H3.

FRF curve of S30 composite with hammer points H1, H2 and H3.
First three modes are bending, twist and second bending. The influence of fibre loading on natural frequency (Mode 1) is shown in Figure 8. For S10, S20 and S30, the natural frequencies obtained are 20.92, 21.77 and 22.58 Hz, respectively. The improvement in natural frequency with increasing fibre loading is almost 4% because of higher bending stiffness of composite at increased fibre loading. From Table 3, it is witnessed that all three modes of natural frequencies (mode 1, mode 2 and mode 3) increased with the rise in loading of fibre. In Mode 1, mode 2 and mode 3, natural frequencies of 10 wt.% are 20.92, 147.17 and 426.00 Hz, respectively. Hence, it can be concluded that higher natural frequencies are exhibited in higher mode.

Effect of fibre loading on the natural frequency (Mode1).
Effect of fibre loading on the natural frequency (Modes 1, 2 and 3).
Damping ratio
The damping ratio (mode 1) with different fibre loadings is shown in Figure 9. The rise in weight fractions has a negative significance in damping property. The behaviour of sisal fibre is elastic, whereas PP matrix is viscoelastic in nature. The rise in fibre content changed the behaviour from viscoelastic to elastic, which caused decrease in damping ratio. Mode 2 and mode 3 damping ratio of sisal/PP composites are tabulated in Table 4 showing the similar trends as those of mode 1. For S10 sisal/PP, damping ratio decreased with the increase in the mode of vibration. It is consistent with an increase in natural frequency at higher mode which reduces the damping ratio. Such behaviour is also reported by Kumar et al. 10 for banana/polyester and sisal/polyester composites and Etaati et al. 13 for hemp/PP composites.

Influence of fibre loading on damping ratio (Mode 1).
Influence of fibre loading on damping ratio (Modes 1, 2 and 3).
A comparison of experimental and analytical natural frequencies
In Table 5, the analytically calculated natural frequencies for mode 1, mode 2 and mode 3 of sisal/PP at different fibre loadings are reported. These natural frequencies are calculated by assuming composites as quasi-isotropic. Bending modulus and density inputs for this are obtained by conducting three-point bending and density test as per ASTM D790 and D792. A similar trend can be observed with a slight deviation in analytical natural frequency compared with experimental one due to quasi-isotropic assumption.
Comparison of experimental and analytical natural frequency of sisal/PP composites (Modes 1, 2 and 3).
SACs
Figure 10 shows the results of the SACs of sisal/PP composites at the frequencies ranging from 500 to 6000 Hz. Significant variation are noted in SAC due to the increase in fibre loading. Increase in fibre loading directly enhanced the SACs of the composites.

Fibre loading effects on the sound absorption coefficients.
For S10 composites, the maximum SAC observed is 0.16 at 2000 Hz, whereas for S20, SAC is 0.23 at 4000 Hz. This indicates that around 43.75%, enhancement in SAC is noted with the increase in fibre loading from 10 to 20 wt.%. Increase in tortuosity due to the volumetric increase of fibres in PP matrix at higher fibre volume support the increase in SAC. In the meantime, after a critical weight percentage, such enhancement is not recorded due to the messy structure at very high fibre loading. Increase in fibre loading from 20% to 30 wt.% reduces the SAC to 0.19 at a frequency of 2500 Hz. The reduction of SAC for S30 indicates higher fibre content in a small volume which creates an over compact structure. This structure therefore now has collapsed fibres which tend to reflect sound waves rather than absorbing them. 33 A similar result is also obtained by Jayamani et al. 34 for sisal/PLA composites. Authors found that the SAC for Pure PLA with 10, 20, 25 and 30 wt.% fraction of sisal fibre are 0.06, 0.07, 0.075, 0.08, 0.085. The SAC of composites with 30 wt.% fraction of fibre is larger than other compositions of composites. Jiang et al. 24 fabricated the seven-hole hollow polyester fibre (SHPF)–reinforced chlorinated polyethylene (CPE) composites to investigate the effect of fibre content (0, 5, 10, 15 and 20 wt.%) and composite thickness (1, 2 and 3 mm) on sound absorption property. Maximum SAC of 0.42 is observed for 20 wt.% fibre content. Yang et al. 35 experimented with rice straw–wood particle reinforced urea–formaldehyde (UF) and showed improvement in SAC with the increase in fibre content.
SAC of composites is influenced by frequency. As the frequency increases, the SAC of composite also increases up to certain frequency and then starts to decrease. This behaviour of the composite shows that sisal fibres have the sound reflecting characteristics at certain frequencies and absorbing at other frequencies. Thus, such composition of composites can be utilized in acoustic application of specific application range. Similar results are also reported by Sambu et al. 36 for 50-mm-thick Arenga pinnata/natural rubber composites. Authors have found two peaks of SAC at frequencies of 1200 and 3500 Hz. The acoustic studies of natural fibre–reinforced composites show that noise reduction capability of vegetable fibre is good at higher frequencies and can be applied in fields such as aeronautical engineering.
TL analysis
Figure 11 shows the variation of transmission with respect to frequency traces from the 100-mm-diameter impedance tube of 3-mm-thick samples. The result shows two peaks in the curve for all compositions which attribute a maximum TL. The first peak value obtained for neat PP and S10, S20 and S30 composites are 17.05 and 17.92, 20.57 and 23.35 dB, respectively, and second peak values obtained are 17.16, 17.80, 20.15 and 22.37 dB. The frequency range is 100–200 Hz for first peak and 1250 Hz for second peak. The improvement in TL of 10, 20 and 30 wt.% composites over PP is of 3.73%, 13.20% and 11.01%, respectively, which indicates a positive effect of sisal fibre incorporation. The trough point in the curve is seen at 400 Hz with TL of 8–10 dB. This point shows the minimum possible TL throughout the frequency range where the noise level is maximum. The improvement in TL is because of increasing weight fraction of fibre which makes the composite structure more compact while also increasing the density of the composites. This indicates that the sound waves travelled for longer time in the sample thickness. Figure 12 shows a plot of the average TL values over the frequency range of 80–1600 Hz against the composition of studied materials. It can be seen that this relationship is approximately linear. Similar results for improvement in TL are also observed by Fatima and Mohanty26,27 for jute fibre and felt-reinforced natural rubber latex composites. In another study, Naghmouchi et al. 25 find TL of olive stone–filled PP composites of 4 mm thickness with different filler percentages and also confirmed the improvement in TL with respect to the increase in the weight fraction of filler. Mean TL of 12-mm-thick commercial gypsum material 25 is 27 dB, while for a fibre glass 27 of 88.9 mm thickness, the TL is 52 dB; whereas sisal/PP has a thickness of 3 mm and gives a TL of 23.35 dB. It indicated that the TL results of current studies are comparable with commercial sound proofing material.

Variation of TL of sisal/PP composites with respect to frequency.

Average TL for all compositions of sisal/PP composites.
Conclusion
In this study, sisal/PP composite specimens are successfully prepared by extrusion followed by injection moulding at different fibre loading. Following conclusions have been drawn on the basis of the results attained:
Natural frequency of the PP composite increases as a function of fibre loading.
The highest damping ratio of 0.4460 is gained for S10 composite in a first bending mode of vibration.
In contrast, increase in fibre loading made PP composite so rigid and damping properties weaker.
In all modes of vibration, experimental natural frequency shows good agreement with analytically calculated natural frequency.
Maximum SAC obtained is 0.23 at 4000 Hz for S20 composition and maximum TL 23.35 dB for S30 composition of sisal/PP composites.
Average TL of 19.58 dB was observed for S30 composition which is 40.05% more than that of pure PP.
As a whole, the studied sisal/PP composite has a great potential to work as acoustic proof member in automotive and aerospace applications.
Footnotes
Acknowledgements
The authors would like to be grateful to Centre for Bio-Polymer Science & Technology (CBPST), Kochi, Kerala, for providing extrusion and injection moulding facility. Also thanks to Dr Muhammad Khusairy Bin Bakri, Swinburne University of Technology, Sarawak Campus, Malaysia for his assistance in this research.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.
