Improving the Mechanical Properties of Ramie-Polylactic Acid Green Composites by Surface Modication using Single Bath Alkaline and Silane Treatment

This research work emphasises on improving the interfacial adhesion of ramie/ polylactic acid (PLA) composites. For this purpose, ramie fabric was modied using vinyl trimethoxy silane with two different hydrolysing agents, i.e. sodium hydroxide and ammonia. The surface modied ramie fabric was characterised by static water contact angle, elemental dispersive X-ray (EDX) and Fourier transform infrared spectroscopy (FTIR). FTIR and EDX analysis conrmed the presence of silica. The tensile strength of fabric showed a decrease after the silane treatment. The composites were prepared by compression moulding using untreated and treated ramie fabrics with PLA. The treatment improved tensile and impact performance of ramie/PLA composites due to enhanced interfacial adhesion between bre and matrix. Dynamic mechanical analysis (DMA) results revealed that treated ramie/PLA composites have higher storage modulus and lower tangent delta than untreated composites.


Introduction
Use of composites as engineering material has opened up a new horizon of materials science. The widespread use of composites in structural and semi-structural applications has played an instrumental part in shaping the modern developments. However, due to these tremendous advantages, composite consumption has skyrocketed in recent years. According to a survey, municipal solid waste (MSW) generation per capita was of 500 g/day in 2011, which was relatively high in urban India. Enormous waste generation leading to a rise in the greenhouse gas emissions has also been reported, which results in threats like global warming, climate change and poor air quality index [1]. These environmental concerns, although necessary, are limiting the composite utilisation and raising concerns related to sustainable development. The recycling of many composite materials is still hindered by technoeconomic reasons [2]. Thus, renewable and sustainable eco-friendly composites have become the need of the hour. The use of biocomposites in various applications like automotive, packaging and medical eld has become a subject of great interest to the scienti c society [3][4][5]. The use of bio-based polymers with natural bres as reinforcement represents a sustainable solution to recycling and solid waste generation. However, there is a limited number of bio-based polymers used as the matrix in composite materials. Polylactic acid (PLA) is one of the bio-based polymers that ful ls the mechanical, engineering and economic requirements for large scale implementation.
Polylactic acid (PLA), a biodegradable bio-based thermoplastic high-strength, and high-modulus aliphatic polyester is produced from natural renewable sources like corn sugar, sugar cane, potato, etc. [6]. Among the biopolymers, PLA has been reported to be the most successful in developing alternatives and replacing the materials based on non-biodegradable fossil polymers [7]. However, applications of PLA are still limited due to its low exibility. To improve the toughness of PLA, many llers and plasticisers have been used [8,9]. Natural bres are one of the potential candidates in improving the toughness of PLA.
Besides improving the toughness, natural bres offer advantages like environmental friendliness, biodegradable nature and a lower density than the synthetic bres, like glass, carbon etc., commonly used in the composites. Siakeng et al. [10] found that incorporating coir/pineapple bre at 30 wt.% in PLA improved its toughness by 30%. Similarly, Ovlaque et al. [11] reinforced PLA with milkweed oss bres and found that the impact strength of PLA was improved by 12%. Jayamani et al. [5] reported that the impact strength of PLA improved after incorporation of sisal bre to about 6 kJ.m − 2 from 4 kJ.m − 2 .
Aydemira and Gardner [12] used cellulose nano brils (CNFs) to reinforce PLA and polyhydroxybutyrate blend to improve the ductility of PLA. They reported that CNFs improve the mechanical and thermal properties of PLA at 1 wt.% loading.
In general, it has been claimed that natural bres can pave the way for sustainable composites by offering advantages such as weight reduction, low cost, improved mechanical properties and low carbon footprint compared to synthetic bres. However, tensile properties of natural bres are inferior to those of their synthetic counterparts, though the former's speci c tensile strength and stiffness are comparable with glass. For example, ramie bre has higher stiffness than glass, and the former is well accepted as a reinforcement for polymeric composites. Nevertheless, the hydrophilic nature of natural bres results in an inferior interface and thereby causes poor mechanical properties of composites to reinforce hydrophobic matrices [13][14][15][16]. Thus, to improve the interfacial adhesion, chemical or physical modi cations of natural bres are required. Among chemical treatments, alkaline treatment is extensively used for the modi cation of cellulosic bres [17,18]. This treatment exposes cellulose micro brils and promotes the hydroxyl group's ionisation present on the bre surface to the alkoxide groups [19]. True et al.
[18] studied the effect of alkaline and/or silane treatment of the sisal bres on the mechanical properties of sisal/PLA composites and found that the strength of sisal/PLA composite improved signi cantly as compared to that of untreated sisal/PLA composite. Yang et al. [20] used cyclic loading after alkaline treatment to improve ramie/PLA composites' mechanical properties. This study con rmed that the alkaline treatment and cyclic loading act synergistically to enhance ramie/PLA composites' overall mechanical properties. Another well-established way to enhance the interface of natural bres in polymeric matrices is the use of coupling agents. Silane-based coupling agents promote covalent bonding between natural bres and polymeric matrices [21][22][23][24][25][26]. The mechanism of surface modi cation using coupling agents involves the reaction of alkoxysilane group with hydroxy groups present on the bre surface and another active group with a polymer matrix [27]. Song et al. [28] treated hemp bres with silane before manufacturing its composites with PLA. They found that silane treatment helped uniform distribution and better adhesion of the hemp bre into PLA, which resulted in improved mechanical and thermal properties of hemp/PLA composites.
Similarly, Jandas et al. [29] and Li et al. [30] used various silane coupling agents for the surface treatment of natural bres. They found that these composites' mechanical properties improved signi cantly compared to that of untreated natural bre composites. Aphichartsuphapkhajorn et al. [24] used alkaline and silane simultaneously on ax bre followed by composite fabrication with biobased resin furan. The study revealed that the combination of alkali and silane yielded better mechanical properties than that obtained with individual alkali or individual silane treatment.
Ramie is a very strong natural bre and available in Korea, China, Japan, and India's north-eastern parts. With its good mechanical performance, low speci c mass, lustre, absorbance and resistance to bacteria, ramie bre is a promising reinforcement among cellulosic bres [31,32]. The search of extant literature shows that development of green composites using ramie fabric has not been explored much [33][34][35][36].
The present study explores the effect of one bath alkaline/silane treatment on the properties of ramie/PLA composites. In the rst step, ramie fabric was treated with silane and alkali (i.e. sodium hydroxide and ammonium hydroxide), followed by investigating the effect of treatment on the mechanical, morphological properties of ramie fabric. In the second part, PLA composites with untreated and treated ramie fabrics were prepared using compression moulding technique followed by evaluation of mechanical performance and thermomechanical analysis.

Materials And Methods
Polylactic acid pellets (4043D) with 6% D-lactide and 94% L-lactide, were purchased from NatureWorks Co. Ltd. Coimbatore, India. Sodium hydroxide and ammonia solutions (25 wt. % NH 3 ) were procured Merck Chemicals, India. Vinyl trimethoxy silane, used as a coupling agent, was purchased from TCI Chemicals, India. The plain-woven ramie fabric used in this study, having fabric sett of 16 cm − 1 in both warp and weft directions, was woven on a CCI sample weaving loom. The ramie yarn's linear density was 80 tex (g km − 1 ), and the areal density of the woven fabric was 285 g m − 2 . Figure 1 shows SEM images of ramie yarn and fabric.

Ramie fabric treatment
For the hydrophobic treatment of ramie fabric, two different types of alkali, i.e. sodium hydroxide (NaOH) and ammonia were used as hydrolysing agents for silane. Before modi cation, ramie fabric was washed with ethanol for 30 min and dried at 60 ℃ for 30 min. The silane concentration used for the treatment of ramie fabric was 10% on weight of fabric with material to liquor ratio of 1:30. A solution containing vinyl trimethoxy silane in ethanol/water solvent mixture (80/20 v/v) was stirred at 40 ℃ for 30 min, followed by sodium hydroxide ammonia with stirring for next 30 min. In case of treatment with sodium hydroxide, 1M solution was prepared in deionised water, and 1 ml of this solution was added in the 100 ml of solution containing silane.
On the other hand, for hydrolysis with ammonia, 10 ml of as received ammonia was used in 100 ml silane solution in ethanol/water. The fabric samples were dipped into this solution for 12 h, followed by drying at 80 ℃ for 1.5 h. The schematic of silane treatment is shown in Fig. 2. In this study, neat ramie fabric, ramie fabric treated with silane hydrolysed with ammonia and sodium hydroxide are coded as NR, HRSA and HRSS, respectively.

Preparation of ramie/PLA composites
For manufacturing of ramie/PLA composites, the ramie fabric was placed between two neat PLA lms followed by compression moulding. For lm preparation, PLA pellets were dried at 60 ℃ for 12 h in a vacuum oven. A pre-weighted amount of PLA pellets was then placed inside the mould, followed by placement of the mould between two solid stainless-steel plates. The thickness of the stainless-steel mould was 0.2 mm. After that, the stainless-steel plates were placed inside compression moulding machine (LabTech LP-S-30) for preheating at 180°C for 2 minutes followed by pressing at 80 bar pressure for 3 minutes and drying in an ambient atmosphere. The thickness of all the lms was 200 ± 20 µm. Figure 3 shows the schematic of the manufacturing process of PLA lms.
For manufacturing of ramie/PLA composites, the ramie fabric was placed between two neat PLA lms. The samples were then preheated for 10 s at 170°C, followed by pressing at an optimum pressure of 10 bar for 2 min. Use of higher pressure than this resulted in shearing of warp and weft yarns. The bre volume fraction was kept at 40 ± 2 % for all ramie/PLA composites. Table 1 shows the description of the sample codes used for composites in this study. Surface morphology of ramie/PLA composites, before and after silane treatment, was examined by Zeiss EVO 50 scanning electron microscopy (SEM) at 10 kV operating voltage. SEM also analysed the crosssections of fractured composite samples after tensile testing. All the samples were dried in a vacuum oven and coated with 5 nm thick gold coating before SEM characterisation.

Elemental dispersive X-ray (EDX)
Elemental dispersive X-ray (EDX) was used to identify elemental composition of neat and silane treated ramie fabrics. Zeiss EVO 50 scanning electron microscope operating at 10 kV was used for this study.

Contact angle Measurement
To con rm the hydrophobicity of ramie fabric, static contact angle measurement was performed on KRUSS drop shape analyser (DSA100). The static water contact angle was measured by placing a droplet of distilled water (2 µL) onto the fabric surface using a syringe. The angle was measured at ve different positions of fabric, and the average value was reported.

Fourier transform infrared spectroscopy (FTIR)
Chemical modi cation of ramie fabric by silane was con rmed by FTIR analysis. The spectra were collected from 4000 to 400 cm − 1 wavenumber using a Nicolet 20SXB FTIR spectrometer (Thermo Fisher Scienti c Inc., U.S.A.) with 64 scans for each sample in attenuated total re ectance (ATR) mode. To eliminate the effect of moisture, samples were dried at 40 ºC before testing.

Tensile testing
The tensile testing of ramie fabric was done as per ASTM 5035 whereas ramie-PLA composites were tested as per ASTM D3039 using a universal tensile tester (Tinius Olsen H5KS). Samples with a gauge length of 100 mm and 25 mm wide were tested at a crosshead speed of 10 mm/min. Five specimens were tested for each sample of the fabric, and four samples were tested for composites, and then the average was calculated.

Characterisation of silane modi ed ramie fabric
After the silane treatment of ramie fabric with sodium hydroxide and ammonia as a catalyst, the samples were tested for hydrophobicity by measuring the contact angle using drop shape analyser. Before and after silane treatment, the surface morphology of ramie fabric was analysed by scanning electron micrographs shown in Fig. 4.
As shown in Fig. 4, neat ramie bres have a clean and smooth surface. After the silane treatment, a uniform coating of silane, with small silica particles, is visible on the bre surface. The surface of treated bres also displays small cracks and increased roughness. Figure 5 shows the static water contact angle (CA) of the silane treated ramie fabrics. The contact angles of 128 and 118 indicate that the silane treated fabrics have become hydrophobic.
This can be attributed to the con guration of silane molecules at the bres' surface, which is supposed to bend and orient its non-polar head (-CH 2 ) towards the surface. Simultaneously, its -OH group form strong hydrogen bonding with the hydroxyl groups of cellulose. The vinyl group (CH = CH) on the surface of ramie fabric is con rmed by FTIR peak present at 1600 cm − 1 .
The FTIR spectra of neat and silane treated ramie fabrics are shown in Fig. 6. The wavenumber region from 500 to 4000 cm − 1 was studied to con rm any chemical linkage between the silane and ramie bre.
For untreated ramie, the spectra at 2900 cm − 1 belong to the vibrations of C-H stretching from -CH 2 group of cellulose and hemicellulose [25,37]. Similarly, the peaks ranging from 3000 cm − 1 to 3500 cm − 1 show hydroxyl group presence in cellulose. After silane treatment of ramie bre, a new absorption band appears at 760 cm − 1 , representing the vibrations of the Si-O-Cellulose and Si-C bonds linked to the hydroxyl groups of the bre surface [38,39]. It is also noticeable from the FTIR that the peak intensity decreases between bands from 1200 cm − 1 to 900 cm − 1 after treatment with silane using sodium hydroxide as a catalyst during the treatment. This suggests that sodium hydroxide had also removed non-cellulosic content present on the surface of the ramie bres. A similar nding has also been reported by other researchers [40] To further con rm the presence of silane, elemental analysis of neat and treated ramie fabrics was conducted. Figure 7 illustrates the EDX spectra of neat and treated ramie fabrics. Table 5 shows the concentration of carbon, oxygen, silica and sodium in ramie fabric before and after silane treatment. Presence of silica is established in silane treated samples (HRSA and HRSS). Besides, ramie fabric treated with silane and sodium hydroxide also shows the presence of sodium. Neat ramie fabric (NR), ramie fabric treated with silane hydrolyse using ammonia (HRSA) and sodium hydroxide (HRSS)

Tensile properties of ramie fabrics
Tensile strength and elongation at break of ramie fabrics before and after silane treatment is shown in Fig. 8. Neat ramie fabric shows tensile strength of 5.8 cN tex − 1 and elongation at break of 20.5%. However, silane treated ramie fabrics display decreased tensile strength with a slight increase in elongation at break. This change is more prominent in case of sodium hydroxide, and silane treated ramie fabric. The decrease in tensile strength is probably due to ramie's additional deligni cation resulting in itching of bre surface. The presence of OH − and Na + ions in the silane solution causes slight removal of pectin hemicelluloses, waxes and lignin, which has a detrimental effect on ramie fabric [41].
One-way ANOVA was used to determine whether there is a signi cant difference in the means of tensile strength of ramie fabric before and after silane treatment. Besides, Tukey's test, which is a single-step multiple comparison statistical test, was used to determine whether the tensile strength of ramie fabrics after silane treatment with sodium hydroxide or ammonia differ signi cantly from each other. Tables 3  and 4 show the analysis of variance (ANOVA) and Tukey's test.
Neat ramie fabric (NR), ramie fabric treated with silane hydrolyse using ammonia (HRSA) and sodium hydroxide (HRSS) It is evident from the ANOVA results that the loss in tensile strength is signi cant after silane treatment as the P-value (0.0029) is lower than 0.05. On the other hand, Tukey's test reveals that silane treatment with ammonia does not signi cantly affect the tensile properties of ramie fabric. In contrast, it is signi cant in case of silane treatment with sodium hydroxide. Table 5 shows the ANOVA results for the tensile strength of the three ramie/PLA composites. Form the Pvalue (6.0571×10 − 5 ); it can be inferred that ramie/PLA composites' strength improves signi cantly after silane treatment of ramie fabric.

SEM analysis of Tensile fractured samples
SEM images of the broken tensile samples of neat and silane treated ramie/ PLA composites are shown in Fig. 10. Neat ramie/PLA composite shows ramie bres' debonding from matrix, suggesting a poor interfacial adhesion between bres and PLA. Also, ramie yarns in weft direction are not su ciently adhered to PLA matrix, and their debonding from matrix can be identi ed. Orue et al.
[18] also reported for sisal/PLA composites that the bres are debonding from matrix results from poor interfacial properties.
In silane treated ramie/PLA composites, the ramie bres seem intact with the PLA matrix in both cases, i.e. with sodium hydroxide and ammonia. The bre breakage is more prominent than the bre pull-out, unlike the neat ramie/PLA composites. Oushabi et al. [39] also reported that the alkali and silane treatment of date palm bre improves the interfacial strength of palm bres and epoxy composites. This improvement in the impact strength of neat ramie/PLA composite can be attributed to higher elongation at break of the ramie fabric (20.5%). Also, ramie bres which act as reinforcement help in better stress transfer during the impact loading and the presence of the ramie fabric hinder the crack propagation after crack initiation thus resulting in improvement in impact performance [32,42]. Similar improvements in impact strength after reinforcing polymer matrices with high elongation bres have been reported by other researchers also [46,47]. After silane treatment with ammonia, the impact strength of ramie/PLA composites was improved by 13% (from 194.5 J m − 1 to 220.5 J m − 1 ). However, composite containing ramie fabric having silane treatment with sodium hydroxide shows only 3% improvement in impact strength. This additional improvement in the impact strength, after silane treatment, can be attributed to the improved adhesion between hydrophobic bre and matrix [30,47].

Dynamic mechanical analysis of ramie/PLA composites
The storage modulus and damping factor (tan δ) of ramie/PLA composites, evaluated in the three-point bending mode, is shown in Fig. 12. It can be seen that the storage moduli of silane treated ramie fabric reinforced composites are higher than that of neat ramie/PLA composite. This implies better elastic behaviour of composites reinforced with silane treated fabrics. Over the range of temperatures under investigation, the silane and sodium hydroxide treated ramie fabric reinforced composite shows the highest storage modulus. Change in the loss factor (tan δ) of ramie/PLA composites as a function of temperature is shown in Fig. 12b. The composite having neat ramie fabric as reinforcement exhibits a higher tan δ value as compared to those of composites reinforced with silane treated fabrics. The higher value of tan δ in neat composite can be attributed to e cient energy dissipation due to higher friction values at the poor interface between neat ramie and PLA matrix. Since the tan δ peak value is related to bre matrix adhesion, lower tan δ peak values of silane treated composites, correspond to better adhesion and compatibility between silane treated ramie and PLA matrix [16].

Conclusions
In this study, the mechanical and thermo-mechanical performance of ramie/PLA biocomposites was explored as a function of bre surface treatment. Static water contact angle con rmed that the fabric surface had been modi ed to hydrophobic from being hydrophilic after treatment with silane. FTIR and EDX of the treated fabric further con rmed the surface modi cation of the fabric. Though the tensile strength of fabric declined after silane treatment, especially with sodium hydroxide, the tensile strength of composites reinforced with silane treated fabrics was 10-12% higher than that of composite reinforced with neat fabric. Also, DMA analysis showed that silane ramie/PLA composites have higher storage modulus at the onset of transition temperature. The results imply that treatment of silane improves the bre-matrix interface and thereby the stress transfer from matrix to bre enhances; as a result, tensile, impact and dynamic mechanical properties improve.

Declarations
Funding This research did not receive any speci c grant from funding agencies in the public, commercial, or notfor-pro t sectors.

Con ict of Interests
The authors declare that they have no con ict of interest.   Tensile properties of PLA and ramie/PLA composites PLA (NP), ramie fabric treated with silane hydrolyse using ammonia (HRSA_Com) and sodium hydroxide (HRSS_Com) Figure 10 SEM images of tensile broken ramie/PLA composites Neat ramie (NP_Com), ramie fabric treated with silane hydrolyse using ammonia (HRSA_Com) and sodium hydroxide (HRSS_Com) Page 20/20