Better Wettability of Pandanus Utilis Alkaline Treated Fibres Tead to Higher Tensile Strengths of Composites.

Composite materials made with synthetic bres like E-glass, Kevlar or carbon have helped to provide a wide array of products to society with specic engineering properties. However, these materials have a high carbon footprint as well as being non-biodegradable. The use of natural bre, as a substitution to these man-made bres, has been studied and encouraging results are being obtained. In this study, the use of ‘Pandanus utilis’ bre as a reinforcing agent in plastic was investigated with the aim of exploring specic properties such as the tensile strength of the bre, its wettability and the effect of bre length after treating the bre with two different NaOH solutions. Results have shown that better reinforcement was obtained for the composites (11.10 ± 2.53MPa) with bres subjected to a more aggressive treatment (2.5%NaOH for 2h) compared to the composite made with bres having maximum tensile strength (168 ± 12MPa at 0.5% NaOH for 14h), due to a better hydrophilicity of the alkaline treated bre (87.37° internal angle). Within the range of short chopped bre length tested (6 to 15 mm), it was shown that there was a general decrease in the tensile strength of the composite.


Introduction
In this era where the world is facing a major problem of plastic pollution, people are converging their ideas towards the use of natural bre in different composite applications. For the past few decades there has been an increasing use of natural plant bres in the manufacture of polymer composites, mostly in the automotive sector. This is due to the relatively high speci c strength, bio-degradability property, lower cost of production, and low abrasive properties.
For Small Island Developing States (SIDS), such as the Mauritius(part of Mascarene islands), the demand for glass-reinforced products (GRP) such as letter box, surfboard, kayaks, utility boxes, wind turbine blades, swimming pool furniture and water tanks are increasing year by year showing the signi cant use of glass bres. However being a SIDS without much raw material resources, Mauritius has to import glass bres, which involves high transportation cost and leading to a high carbon footprint. Glass bre tends to cause additional machining costof 2.65€/cm 3 since the tools required in the manufacturing process of GRP need to be replaced frequently given the abrasive nature of glass bre. [1] In order to meet the Sustainable Development Goals (SDGs), SIDS have to provide new opportunities for business development for their citizens in the sectors that are present locally such as agriculture, shing and tourism. The creation of a bre micro-industry would entail the creation of new job opportunities particularly in rural areas. This would help to alleviate poverty and minimise social inequalities. For the past 50 years, Mauritius' economy has been based on the sugarcane industry (60 % of the export earnings in 1979, [2]) due to the preferential tariff and guarantee market in Europe Union (EU), causing the maximum development of the sugarcane plantation across the island. However in October 2017, EU has abolished all sugar quotas. Therefore, the local sugar planters are facing several di culties in sustaining sugar cane plantation particularly with the rising labour cost, and it has been reported that there are more than 9,000 hectares of abandoned sugar cane elds [3]. Some other farmers have started to shift towards pineapple and banana cultivation in order to generate an income while being more resilient to climate change. Thus, the creation of the micro industry for the production and extraction of the bre from agro waste will not only prevent the farmers from losing their jobs but also allows the development on the agricultural sector.
Furthermore, many of the SIDS possess a rich ora of endemic plants, and for example the Mascarenes islands are a hotspot for the conservation of several endemic plants. As a mean to support the conservation effort, the leaves of many endemic plants such as Pandanus species, palm species are a goodsource o bres for the micro industry.
Natural bre has proved to gain more strength under alkaline treatment as a reinforcing agent in biocomposites at proper concentration and soaking time. Alkaline treatment, which meets both time and cost constraints, allows natural bre to be exploited e ciently and economically. Treated data palms bre (DPF) with 5%w.t NaOH has an optimum tensile strength of 460MPa. [4] The optimum treatment for banana bre was found to be 11g/L NaOH for 150 min at 90°C. [5] After being soaked for 10min at 55°C at 5%w.t NaOH, treated ax bre has an optimum tensile strength of 611 MPa. [6] However, the main idea after treating the natural bre is to use the latter as a reinforcing element in the bio composite. But the optimum alkaline treatment for producing the strongest bre may well differs from the treatment which will produce the stronger bio composite. This is because there are other factors, particularly the bre to matrix adhesion which tends to play an important role in in uencing the composite mechanical strength. The optimum treatment for producing a strong single Kenaf bre was found to be 4%w.t NaOH for a period of 30min at 60°C but the author did not mention the speci c alkaline treatment which leads to an increase in strength of 11.84% of the composites [7]. Taha et al (2007) did optimize the strength of date palm bre (DPF) for its application in polymeric composites, mentioning that alkaline treatment does also modi es the surface of the bre. [8] On the other hand, Rizal et al (2018) observed the relationship of the soaking time of alkaline treatment on Typha bre with its wettability properties and as a result improving the tensile strength of the biocomposites. [9] The strength of Alfa reinforced polyester increases as the concentration of NaOH increases for the same soaking time. [10] Few authors did consider the impact of optimizing the alkaline treatment in terms of the interaction between the base matrix and the treated bre. Their main concern was to improve factors like the length or weight ratio of the bre. Hemp bre propylene composites proved to be stronger (47.2 MPa) with a weight ratio of 40% and a length of 1-3cm compared to bre having length of 10cm. [11] At a weight ratio of 30% compare to 0%, 10% and 20%, sisal reinforced polypropylene has the highest tensile strength. [12] Ijuk bre at a length of 50mm had a relatively higher tensile strength compare to a bre length of 10mm according to Santhiarsa (2016). [13] Research has also been conducted on the Pandanaceae species and their application in biocomposite. An optimum tensile strength of 17 MPa was obtained using 'Pandanus Fascicularis' in polyester composites while the potential application of 'Pandanus utilis' in epoxy composites as a substitution to glass bre was discussed by Deesoruth et al (2014) with a biocomposite having a compressive strength of 97.9 MPa for weight ratio of 10% of the optimum treated bre (5% w.t NaOH for 45min at 75°C). [14,15] In this study, an investigation was performed in order to evaluate the effect of the alkaline treatment on the wettability of 'Pandanus utilis' bre and the impact of this property on the tensile strength of the reinforced plastic composite produced by compression moulding. Furthermore, in the same process, the relationship of the tensile strength of the reinforced biocomposite with the bre chopped length used for the biocomposite was studied and compared to the composite being produced industrially by a local company.

Methodology Extraction
The extraction of the bre was done using a Phoenix Decorticator, and all the 'Pandanus utilis' leaves were taken from the same tree to minimize the effect of external factors such as geographical location, temperature, the soil quality. The bre were then cleaned and dried in an oven for 24h at 60 C.

Alkaline Treatment
The bres were treated under two set of conditions; a rst set with 0.5%wt NaOH for 14h (SFOT) as being the optimum alkaline condition to obtain the maximum bre tensile strength of 160 MPa as per the published work Ra dison et al (2018). [16] Given that the objective is to improve the wettability of the bre to the matrix, an exposure to a higher concentration of alkaline treatment would lead to the de brillation of the bre and increased its roughness. Thus, a second set of conditions, 2.5%wt NaOH for 2h (SFST) was conducted, which, according to Ra dison et al (2018), should result in a decrease of 40% in the tensile strength (120MPa). [16] A bre to solution ratio of 1 to 30 was used and the bres were completely immersed in the respective solution. After the appropriate soaking time, the bres were neutralized using distilled water and dried in an oven for 24h at 60 C.

Manufacturing of the composites
The bre composite was fabricated based on the method in place at one of the leading manufacturer of composite in Mauritius.The local manufacturer uses 6 mm chopped glass bresto produce GRP for different commodity products by compression moulding. Thus the baseline for 'Pandanus utilis' bre length was set at 6mm, and the effect of the two alkaline treatments on the 6 mm chopped bres were evaluated by testing the composite tensile strength.
Subsequently, the effect of the alkaline (SFST) treated chopped bre length of 6mm, 9mm, 12mm and 15mm on the tensile strength of the resulting composite was investigated. A bre volume ratio of 15% was used, as per the local manufacturer's practice.The bres were added slowly to the polyester resin during the mixing process and the resulting dough was mixed using a motorised Z-bladder for achieving homogeneity. The dough was then placed in a compression moulding machine (available at the local company used for producing GRP products) and a load of 1000kg/cm 2 at a temperature of 150 C for 1 min to produce rectangular plate of 315mm x 250mm x 4mm. The composite plates were then allowed to cool at room temperature.

Tensile Testing
The tensile test of the 'Pandanus Utilis' bre was done according to the ASTM C-1557 with a load cell of 10kgf at a speed of 3.5mm/min (for failure of bre to occur within 30 s) on a Testometric M500-50 AT. The gauge length used was 25.4mm and the overall bre length was 30mm. 20 single bre specimens were tested and their individual cross-sectional area was measured at the broken point (during the tensile test), along each bre using ImageJ software with a USB connected microscope.
For the tensile testing of the bre composite, the dumbbell shape was cut from the compression moulding composite plate using a CNC milling machine according to the Type 1 of the ASTM D-638. The tensile test was conducted on the above tensile machine with a load cell of 5000 kgf and a speed of 0.35mm/min with a sample size of 9 specimens. The average cross sectional dimensions of each of the 9 composite samples for each condition were measured using a Vernier caliper.

FTIR
FTIR tests were carried out on the untreated and mercerized bres using a Bruker single bounce ATR-FTIR spectrometer, equipped with its OPUS software. Calibration was carried out before each measurement. All spectra were recorded in the range from 4000 to 500cm -1 .

Wettability test
Contact angle test was done at the Center for Biomedical and Biomaterials Research (CBBR), Mauritius. The specimens were prepared on a glass slide and the inter-bre gaps were minimized thus increasing the accuracy of thetesting method. The tests were carried out on Kruss Drop Shape Analyzer-DSA 100 using a water droplet of 2µL at 25 C with 5 samples being taken on each specimen and Young Laplace Fitting method was used.  (2017) have recorded an increase of 76% in the tensile strength of the date palm bre after the latter was treated with its optimum alkaline treatment. [4] Jute bre has an increase of 59% in its modulus after being treated with an alkaline treatment of 1% w.t NaOH. [17] The second alkaline treatment (2.5% w.t NaOH for 2 h -SFST) has yielded a bre tensile strength of 117.9 ± 12.3 MPa, about 30 % lower as compared to the optimal NaOH treatment (SFOT). This shows that a higher concentration of the alkaline treatment as compared to the optimum condition produced a treated bre with a lower tensile strength. The same observation was made by Pickering et al (2007) with hemp bre. [11] With an increase of 5%w.t in its NaOH concentration in the alkaline treatment, a decrease of 45

Results And Discussion
MPa was recorded in the hemp bre, although the soaking time was reduced by 30min.

Wettability test
The average internal contact angle obtained for the SFOTsample was 95.23° ± 3.49° and for the SFST specimen was 87.37° ± 4.97°, showing that the more aggressive alkaline treatment does affect the hydrophobicity of the bre. The SFST condition allows the bre to reach the hydrophilic state as compared to a hydrophobic state obtained with the SFOT condition. The same observation was made with Typha bre by Rizal et al (2018) where the internal contact angle of the water droplet increases with an increase in the soaking time (from 2 hours to 8 hours) at a constant alkaline concentration of 5% w.t NaOH. [9] In order to further understand the difference in the effect of the chemical treatments on the properties of the bresthe detection of speci c chemical functional groups by the FTIR technique is presented in the next section.

FTIR results
The decrease in the peak at 1726 cm -1 wavelength shows the signi cant removal of non-cellulosic component such as the lignin and hemicelluloses. This change has also been observed in Alfa bre when treated with 5%NaOH. [10] Roy et al (2012) observed the same phenomena as the concentration of the alkaline solution is increased when treating jute bre. [18] The absorbance peak at 1735 cm -1 is reduced in Typha bre after being treated with sodium hydroxide showing the successful removal of the carbonyl group present in hemicelluloses. [9] As these chemical groups are removed from the natural ber after treatment, the treated bre has a higher ratio of cellulose present in the bre therefore having higher tensile strength as compared to untreated bre. The SFST treated bre showed a smaller peak around bres, thereby removing more of the hemicellulose and leading to higher de brillation. This would then lead to an increased in the surface roughness of the bres, which would yield a better wettability property.
Thus in the present study, it can be expected that the composite manufactured with SFST treated bres would have a higher tensile strength as compared to composite with SFOT bres. This will be discussed in the next section.
Tensile strength of the brebased composite A rst comparison of the tensile strength of the biocomposite and the glass reinforced plastic (GRP) produced using the industrial procedure ( bre chopped at 6mm length) present at the industry was made. Since the density of the 'Pandanus utilis' bre is half the glass bre, industrial volume ratio was used instead of the mass ratio for adequate mixing of the natural bre and the polyester resin. [19] The tensile strength of the SFOT bre polyester composite was 8.32 ± 1.30 MPa, and 11.10 ± 2.53 MPa for SFST bre polyester composite. Based on a t-test, the calculated t value is 2.953 whereas the critical t value is 2.306, which con rms that there is a difference between the two mean values. These results of tensile strength con rm the results of the contact angle measurements where SFST bres had a smaller internal contact angle as compared to the SFOT bres. Thus the SFST bres have a better wettability property with the polymer resin leading to better mechanical strength.
On the other hand the 6 mm chopped bre glass polyester composite bre has a tensile strength of 33.10 ± 2.48 MPa, which is 198 % higher than that of the SFST. However, the SFST composite does not represent an optimized condition for highest interfacial shear strength between the bre and the polyester matrix. Furthermore the effective impact of the bre length and bre ratio in the matrix has also not yet been optimized in this study.
It is observed that although the SFST bres have 40% lower tensile strength (118 MPa) as compared to the SFOT bres (168 MPa); the SFST bre composite has a tensile strength which is 33.7 % higher than SFOT based bre composite.This implies that the adhesion of the treated bre with the base polyester matrix plays a critical role in transmitting the load from the matrix to the bres. The higher adhesion of the SFST bre to the matrix is explained by the lower internal contact angle as compared to the SFOT bre. The same observation was made by Rizal et al (2018) with Typha bre, where the maximum tensile strength of the Typha bre reinforced composite was recorded with an alkaline treatment of 5%w.t NaOH for 4h which had acontact angle lower than 87.5°. [9] This clearly showed that the hydrophilicity state of the treated bre do affect the nal strength of the biocomposite.
The above ndings of the present study are further supported by the research work on jute bres; by combining the observation made by Roy et al (2012) and those of Lakshmanan et al (2016). Roy et al (2012) observed that the optimum treatment for producing a strong single Jute ber (610MPa) is 0.5% wt NaOH for 24h at ambient temperature, while Lakshmanan et al (2016) reported that the optimum treatment for producing a strong biocomposite (48.3MPa), with based matrix unsaturated polyester resin, using treated Jute bre is 1%w.t NaOH for 60min at 30°C. [17,18] It is thus observed, that a jute bre treated with a stronger alkaline solution, resulting in a lower tensile strength (25.7 %) but with better contact angle, will produce a stronger bre based composite.
Effect of the bre length on the tensile strength of the composite The tensile strength of the respective brecomposite produced using the different bre length is shown in the table below. As the length of the bre is increased, the tensile strength of the biocomposite decreases to a minimum value of 3.23 ± 1.37 MPa at 15mm chopped length. Although Deesoruth et al (2014) have determined that the critical length of 'Pandanus utilis' bre is L c = 2.64mm, the present ndings show that the tested lengths L between 2Lc (6 mm) and 6Lc (15 mm) of the treated bre lead to a reduction in the tensile strength of the composite. The same observation was made by Yang et al (2019). [15,20] Fig 3 Tensile strength of the different composites produced using different ber length Takagi and Ichihara (2004) observed bre pull-out phenomenon when NaOH treated bamboo bres of 8 mm length or less was used in a resin, whereas bre fracture occurred when the bre length was 15 mm or more. These authors concluded that the bre critical length was about 12 mm whereas theoretically they calculated the critical length as being 30 mm. However, these authors did not perform any NaOH process optimization to determine the best bre to resin adhesion. [21] Iqbal et al (2017) had also investigated the effect of jute bre length (10-30 mm) on the tensile strength of a polystyrene untreated bre composite. According to the authors calculation the critical length (L c ) was 0.24 mm, and the minimum bre length to be considered as a continuous bre (15L c ) was 3.6 mm.
They have reported no signi cant improvement in the tensile strength with bre length of 10 mm but an increase of about 18 % at bre length of 30 mm as compared to the pure unreinforced polymer. It can be observed that these authors have used bre lengths much greater than the minimum continuous untreated bre, and that a signi cant improvement in the TS of the composite was only achieved at very long length, typically 125L c . [22] Bisariaet al (2015) investigated the effect of untreated jute bres (5-20 mm) on the mechanical properties of the bre epoxy composite. [23] These authors have reported a general decrease in the tensile strength of the resulting composite for bre length between 5 and 20 mm as compared to that of the unreinforced polymer. These results are to some extent along the same line as those of Iqbal et al (2017) particularly for bre length of 20 mm or less. Both Bisaria et al (2015) and Iqbal et al (2017) have used the hand layup technique to fabricate the composite, but each with a different procedure, and also with a different polymer. [22,23] From the ndings of Takagi and Ichihara (2004), Iqbal et al (2017), Bisaria et al (2015), and the results of the present study, it is noted that there are several factors which would impact on the resulting tensile strength of the composite, namely, the optimized interfacial shear strength of the bre to polymer, the bre to resin ratio, the distribution of the bres in the polymer (which would affect the void spaces in between the bres network), presence of air bubbles, and the method of fabrication of the composite (which relates to the effective pressure applied to bind the bre to the polymer). [21][22][23] Alkaline treatment with NaOH will tend to modify the surface roughness of the bre thereby increasing the area of contact with the polymer. But the NaOH has to be optimized for maximum bre to matrix interfacial shear strength rather than maximum tensile strength of the bre. Furthermore the shape and size of the bres also play an important role in ensuring a proper adhesion to the matrix.

Conclusion
This study has shown that the ultimate goal in producing the strongest natural bre based composite is to improve the interaction of the bre with the base matrix and to achieve this goal, several factors need to be considered and optimised such as the hydrophilicity (appropriate treatment of the bre), chopped length, bre to resin ratio and fabrication among others. In this study it has been shown that the treated bre with the highest tensile strength does not produce the stronger composite. The crucial role of the hydrophilicity of the natural treated bre has been shown, and although not being optimised, has shown to have better effect on the resultant strength of the biocomposite being produced. Fibre lengths L between 2Lc and 6 Lc have shown to produced weaker bio composite in this study leading to the fact that ber length has an impact on the tensile strength of the bre. It should be noted that the range of bre lengths was chosen based on the current practice at the local composite manufacturing company, and also on the methodology of the mixing process for the dough preparation prior to the compression moulding process.
Finally, it would seem that using much longer bre length than the critical length with the optimum NaOH treatment for maximum bre to matrix interfacial strength, appropriate pressure for bre wetting to the matrix, which would also remove air bubbles, effective bre distribution in the matrix could potentially improve the tensile strength of the natural bre composite. work was carried out to compare the optimum alkaline treatments yielding highest bre tensile strength with a slightly more aggressive alkaline treatment; and the tensile strength, FTIR analysis and contact angle measurement have shown better wettability of the bres to the polyester matrix. A composite with a higher mechanical strength was thus obtained. The study has also shown that chopped short natural bres within the close range of the commercial chopped bre glass length do not lead to an improvement in the mechanical strength, but rather the opposite effect.