Abstract
The modulus of elasticity is a critical parameter for the performance design and analysis of biofibre-based biocomposite materials. As a result of criteria such as internal heterogeneity, the random distribution of fibres and the success of interfacial adhesion between the fibre and the matrix, it becomes difficult to predict the modulus of elasticity in practical ways. Therefore, one of the aims of this study is to determine the modulus of elasticity of biocomposite material reinforced with discontinuous and random fibres by means of micromechanical models and experimentally. In addition, it is also aimed to reveal which micromechanical model can be used reliably in predicting the modulus of elasticity of both aged and non-aged biocomposite materials due to the relationship between the analytical and experimental results. In order to achieve these objectives, initially, chicken feather fibre/poly (lactic acid) biocomposite specimens having 2, 5 and 10 % chicken feather fibre mass fractions were mixed and manufactured by extruding, and subsequently, tensile test specimens according to the appropriate standard were formed by the injection-moulding method. An agreement between the moduli of elasticity obtained from 6 micromechanical models and experimentally from the slope of the stress–strain curves resulting from tensile tests was determined.
1 Introduction
Throughout history, people have used materials derived from biological raw materials such as leather, silk, wool and cellulose, and are known for the materials and equipment they developed and used. Polymers have an important role in this historical process both in the biological world and especially in industrial economies after the industrial revolution [1, 2]. Thanks to the advancement in chemistry and materials science, new synthetic polymers such as polyethylene, polyamide and polyurethane were introduced worldwide after the industrial revolution. Using these synthetic polymers, materials such as adhesives, rubbers, plastics and fibres, which are frequently encountered in daily life, have been produced [1, 3]. The fact that synthetic polymers are not biodegradable, cannot be obtained from renewable sources and their widespread and frequent use have resulted in several insalubrious environmental conditions [4].
Increased social interest in environmental awareness, unsustainable use of petroleum and petroleum products and new environmental regulations introduced by countries have increased the demand for the use of environmentally friendly products and raw materials [5, 6]. Therefore, today’s focus is on the development and popularisation of environmentally friendly materials such as bio-based polymers and products that can naturally decompose [7, 8]. It is precisely at this point that nature emerges as an inspiration in the development and design of biomaterials. Understanding the structure, function, strength, toughness, ductility and self-healing properties of natural materials can contribute to the development of biopolymers in order to reduce the negative consequences of synthetic polymers and to be an alternative to synthetic polymers [9].
The major reason for the great interest in biodegradable polymers is that they are equipped with some outstanding properties in terms of environmental benefits such as greenhouse gas emissions [4, 10] and renewability of the base material [11, 12]. In this context, poly (lactic acid) (PLA) has been a leader among many biopolymers due to the properties of biopolymers as well as having outstanding physical and mechanical properties [13, 14], thermal plasticity [15], biocompatibility [16] and relatively good availability [17]. PLA, a compostable synthetic polymer, is produced using corn starch-derived monomer feedstock and is a serious substitute for petroleum-based plastics [18, 19]. PLA-based products can be recycled by re-melting and processing the material or by hydrolysis to the basic chemical, lactic acid, after the end of its life [20]. Another possible alternative method for composting PLA is to leave the entire biomass of PLA in the natural life cycle where it degrades to CO2 and water [21, 22]. Extrusion, blow moulding, fibre spinning into various products and injection moulding methods used in the production process of thermoplastic polymers can be applied to PLA for the same purpose [17, 23]. The ease of access to and availability of PLA, a thermoplastic polymer that can be produced from natural sources with a capacity of 140,000 tons per year [24], has paved the way for scientists working in the polymer field to produce new PLA-based biocomposites that can compete with petroleum-based plastics [19, 25].
Owing to its mechanical properties, PLA can be used in a wide range of applications. Examples of the area that does not require high performance include food bags, disposable cutlery and plates, liquid barriers in disposable diapers and slow-release membranes for drug delivery [26]. Furthermore, PLA-based automotive parts have been used in the automotive industry since their CO2 emissions are less than those of petroleum-based automotive parts [17]. In addition to the mentioned fields, it has recently started to be used in fibre production and composite production which can be used in various practical and mechanical fields [27].
Despite its good properties such as easy processability in equipment and good strength that emphasize PLA, its expense and physical properties such as relatively poor impact resistance limit PLA polymer applications [28, 29]. Therefore, one way of eliminating the disadvantages of PLA and making it a cost-effective material, as well as providing better mechanical and thermal properties, is to add reinforcements in the system such as fibres or fillers [30]. Biocomposite materials can be obtained by reinforcing biodegradable polymers such as PLA, cellulose esters, starch polymers and polyhydroxyalkanoates with bio-based fibres [31].
Fibres are generally classified as synthetic or natural fibres. Natural fibres are divided into three types as plant-based, animal-based and mineral-based fibres [6, 8]. Particularly in recent years, in some applications, natural fibres have been chosen instead of man-made fibres [32] due to their low-cost, attainability, low density, sustainability and renewability [33, 34]. There are many examples of plant-based fibres such as kenaf [30], jute [35], banana and bamboo [36, 37]. Examples of mineral-based fibres include amosite, tremolite and crocidolite [38, 39]. Horsehair [40], sheep wool [41] and chicken feather (CF) [42] can be an example for animal-based fibres. The use of natural fibres as reinforcements in composite materials takes researchers’ attention since their specific strength competes with that of synthetic fibres.
In poultry processing plants, a large amount of unwanted waste product is required to be disposed of each year as a CF end product, and it can be of great environmental benefit to consider these waste chicken feathers as a fibre additive in areas such as composite material [43]. Chicken feathers contain 91 % keratin, 8 % water and 1 % lipid [44]. Keratin fibres are non-corrosive, eco-friendly, biodegradable, hydrophobic and renewable, and in addition, they have low density, low cost, high flexibility, good compressibility; hence, these features make them a very good structural additive material in composite materials [42].
In the literature, studies using animal-based fibres as additives are less common than studies using plant-based fibres. One of the studies in which CF was used as reinforcement was carried out by Carrillo et al. [45], and they made the production and characterization of thermoplastic material including PLA, polypropylene and high-density PET as matrix material and whole CF as fibre. The results show that by adding 5–10 % volumetric ratio of whole CF to thermoplastic matrices, the hardness values of composite materials increase. In another study using chicken feather fibre (CFF) as a reinforcement, Baba et al. [46] fabricated CFF/PLA biocomposite material by using PLA as a matrix and CFF as additives and examined the effects of fibre concentration and fibre length on the mechanical properties of CFF/PLA biocomposites. According to the results of the experiment, the modulus of elasticity, compressive strength and hardness values of CFF/PLA biocomposite are higher than pure PLA.
It is necessary to know the behaviour of the polymer-based composite materials selected for use in special applications under the environmental conditions they may be exposed to [47]. Under atmospheric conditions of use the rate of degradation of PLA is slow, and the degradation can last 3–5 years. For this reason, PLA-based composites and materials that are planned to be used in long-term service working process should be subjected to real-time tests, and tests should be performed in real service environments [48]. The behaviour of polymeric materials under service conditions is hard to estimate, due on a large scale to their natural complex morphology. Their morphology, which leads to anisotropic and non-linear behaviour, depends substantially on the thermomechanical history of samples experienced during processing [49, 50]. The long-term ageing process at room temperature has serious effects on a wide range of materials, and it is a natural phenomenon capable of severely affecting the physical properties of the amorphous phase of glassy polymers [51]. PLA is a glassy polymer with an elongation at break of less than 10 % [52].
The ability to predict the macromechanical properties of a composite material using component data is a fundamental tool for the advancement of composite material design. Therefore, the prediction of elastic properties is very crucial for the use of such composites in advanced industries. In order to evaluate the elastic properties of biocomposites, analytical or numerical modelling has been developed based on the knowledge of the individual elastic moduli and the mass or volumetric mixing ratios of the fibres and the matrix material forming the composite material. Characterization of mechanical properties for a randomly distributed and discontinuous short fibre reinforced composite material is carried out by experimental studies. Since the experimental studies require cost, time and effort, some micromechanical models have been developed by researchers in order to estimate the mechanical properties without the need for many tests. Elastic properties of biocomposites formed by reinforcing PLA with short, random and discontinuous CFF were found by Özmen et al. [53] using 6 micromechanical models. Considering the convergence between the elastic moduli obtained from micromechanical models and experimental data, the Nielsen–Chen model which showed the smallest error (1.4 %) [53] was regarded as the most reliable micromechanical model for predicting the elastic modulus.
The effects of long-term ageing process on the mechanical and thermal properties of CFF/PLA biocomposites were investigated by Akderya et al. [54, 55]. Accordingly, the tensile strength, tensile modulus, microhardness values, and thermal stability values of CFF/PLA biocomposites decreased with the long-term ageing effect. The elastic moduli obtained experimentally by Akderya et al. [55] are predicted with this study using micromechanical models, and the most reliable micromechanical model is sought for both aged and non-aged conditions. In this study, six micromechanical models are used to predict elastic moduli of long-term naturally aged CFF/PLA biocomposites formed by reinforcing PLA with discontinuous, random and short CFF which has 2, 5 and 10 % mass ratios. The values obtained from micromechanical models and experimental results are compared. Accordingly, the micromechanical models providing the most reliable and consistent results in predicting the elastic modulus of both aged and non-aged CFF/PLA biocomposites are determined. One of the most important outcomes of this study is that it enables the selection of the most reliable micromechanical model for the prediction of an aged biocomposite elastic modulus reinforced with short, discontinuous and random fibres, and another outcome is to reinforce this reliability by establishing a bridge between both aged and non-aged data. In this way, the modulus of elasticity of the biocomposite material during the design stage and long after its production becomes predictable, which makes significant progress in the design of the biocomposite material.
2 Material preparation
2.1 Preparation of the chicken feathers
The reinforcing component of the biocomposite material, CFF, was supplied from a local company in Manisa/Turkey in the form of waste chicken feathers. In Figure 1, parts of a typical CF are described in detail. As shown in Figure 1, a typical CF consists of three main parts, the central shaft of the feather called rachis, barbs and barbules [54, 55]. The primary structure, rachis, can reach up to 17–18 cm in length, and the barbs are arranged in order and uniformly on the rachis. The secondary structure is of different lengths from the base to the end, as the barbs change their position along the rachis. The lengths of the barbs used in this study vary between 25 and 30 mm. The tertiary section, barbules, vary in length from 0.3 to 0.5 mm and appear to be spines along the barb.
Detailed view of a CF and its parts.
Based on the studies, barbules, together with other fibres, can provide unique structural interaction when it is desired to produce blended yarns from barbs. Barbules can be entangled with different fibres, and this can have a positive effect on the mechanical properties of fibrous composites [56]. It was stated that the barbs are more flexible than rachis, and that they are able to twist and bend together, even if they are in groups [57].
In this study, the barbs, 25–30 mm in length, arranged above the CF were cut manually from the rachis. The barbs were then washed with hot water at 60 °C for cleaning and immersed in the water at room temperature for 24 h after the hot water bath. Finally, the barbs were laid out to dry in a room for 24 h and left in an oven at 60 °C for 6 h to remove the moisture absorbed by the barbs.
2.2 Chicken feather fibre/poly (lactic acid) biocomposites production and specimen preparation
The biopolymer material, the matrix of the biocomposite, was selected as PLA (Nature Works Ingeo Biopolymer 3052D from USA) having a density of 1.24 g cm−3. PLA was mixed with WiseStir HT-50AX mixer (Witeg, Germany) with CFF in mass ratios of 2, 5 and 10 % in order to determine the effect of CFF mass content on mechanical properties. After the mixing process, tensile samples were produced in accordance with ISO 527 [58] standard with the aid of a twin-screw extruder and injection moulds. Tensile test specimen geometries in accordance with ISO 527 and biocomposite specimen with mass ratio of 10 % are shown in Figure 2a and b. More detailed information about the entire production process can be found in the studies of Baba et al. [46] and Özmen et al. [59].
(a) Tensile test specimen geometries in accordance with ISO 527, (b) tensile test specimens of 10 % CFF/PLA biocomposite and pure PLA.
2.3 Long-term ageing conditions
CFF/PLA biocomposite samples with mass ratios of 2, 5 and 10 % CFF were kept in a laboratory with a temperature of 23 ± 2 °C and average humidity of 50 ± 5 % for 5 years (43,800 h) without interrupting and changing the ageing conditions. Laboratory temperature and humidity were provided by an air conditioner, and the samples were positioned so that they are not directly affected by sunlight and airflow. A schematic diagram of the processes followed in this study can be found in the study of Akderya et al. [54].
3 Experimental procedures
3.1 Tensile testing of chicken feather fibre/poly(lactic acid) biocomposites
At least five samples of each parameter were used in the tensile testing of aged and non-aged pure PLA and biocomposite samples having masses of 2, 5 and 10 % CFF. Tensile tests were performed at the room temperature (23 ± 2 °C) and 50 ± 5 % relative humidity on a 100 kN-Shimadzu tensile test machine with a video extensometer equipment at a speed of 1 mm min−1.
3.2 Tensile testing of a single chicken feather fibre
In order to find the mechanical properties of a single CF fibre, the barbs were carefully cut from the rachis. In order to prevent direct contact of the barbs with the grips of the testing device, the barbs were bonded between two composite samples 20 × 20 mm in size as tab materials at the top and bottom (Figure 3), using a small amount of epoxy resin. The samples were then allowed to cure for one day at room temperature. Barbs tensile testing was performed on a 5 kN-Shimadzu tensile testing device at a speed of 1 mm/min and a gauge length of 10 mm [60]. All tests were performed under standard environmental conditions and at 23 ± 2 °C and 50 ± 5 % relative humidity. The test of a sample was discarded if the fibre slipped between the tab materials or if the break did not occur from the gauge length during testing. The tests were performed with at least 5 valid results from each CF. An image taken during the tensile test of a single fibre is shown in Figure 3.
Tensile testing of a single CFF.
3.3 Fourier transform infrared spectroscopy measurements
Fourier transform infrared spectroscopy (FTIR) analysis was used to determine the organic components of PLA and CFF chemically and to observe the changes in these components as a result of the ageing process. FTIR spectra were recorded using a Nicolet iS50 FTIR Spectrometer (ThermoFisher Scientific, USA) using ATR (attenuated total reflectance) mode. The spectrum was recorded in the wavenumbers range 4000–650 cm−1.
3.4 Scanning electron microscopy analysis
Scanning electron microscopy (SEM) analysis was used to examine the fracture surface morphology of pure PLA and CFF/PLA biocomposite samples, to investigate fibre/matrix interactions and to obtain a detailed image of CF parts. Due to the lack of support from the same laboratories at the end of the long-term ageing process, micrographs of non-aged and aged samples were examined in different laboratories. An EM-30 SEM (COXEM, USA) device was used to obtain micrographs of the non-aged samples, and a 300 VP SEM (Carl Zeiss, Germany) device was used for aged ones. 15 kV was selected as the acceleration voltage in both devices in accordance with E986 standard.
4 Micromechanical models
In recent years, many construction materials have been developed as a result of engineering studies. Considering the diversity of the production methods of these materials, there is a need for instruments that reliably estimate the behaviour of structures made with these materials [61]. Fibre-reinforced composites have been greatly utilized in engineering due to their high specific moduli and admirable adaptability. Composites containing short fibres are preferred because of their high formability properties, while composites containing long fibres stand out with their superior mechanical properties [62].
The elastoplastic theories for a composite material can be categorized as macro-scale (length scale), meso-scale and micro-scale (micromechanical models). In a macro-model approach, the material subject to the engineering problem is treated as an anisotropic homogeneous laminate. In general, the macro model is used in conjunction with the finite element method, and the required elastoplastic properties of the laminate are obtained by experiment or from a smaller scale model. In the meso-scale model, the mechanical properties of a laminate are derived from the properties of a single layer. This layer is considered to be a homogeneous and orthogonal anisotropic medium, and the mechanical properties of the layer can be acquired by experimental methods or a micromechanical model. A micro-scale model uses the microstructure of the composite and the properties of the composite components when determining the mechanical behaviour of the laminate. Schematic representations of the macro, meso and micro-scale models are shown in Figure 4.
Schematic views of macro, meso and micro scales.
In this study, the Christensen–Waals, Pan, Inverse Rule of Mixtures (IROM), Manera, Nielsen–Chen and Halpin–Tsai micromechanical models were used to estimate the elastic modulus of CFF/PLA biocomposites formed by the dispersion of discontinuous, short and random CF fibres into the matrix material PLA. Although there are many methods for predicting the elastic modulus of composite materials, not all methods are appropriate due to the complex nature of natural fibre reinforced composites.
4.1 The Christensen–Waals model
In the study conducted by Christensen and Waals [63], the theoretical stiffness properties of the randomly oriented fibre system in the isotropic matrix phase were theoretically demonstrated. In this approach, it was aimed to predict the effective isotropic properties of the composite by using the elastic properties of each phase and fibre volumetric concentration ratio. With the obtained equations, the theoretical limits of the composite material subject to the experiments can be determined and the consistency of the theoretical and experimental results can be demonstrated. In addition, this approach provides the opportunity to comment on the effectiveness of bonds, distribution of fibres and other factors that may affect the performance of the composite material. Accordingly, the elastic modulus equation proposed by Christensen and Waals [63] is as follows. In the following equation, the elastic modulus of the composite material is represented by E c, the elastic modulus of the fibres is represented by E f and the modulus of elasticity of the matrix is represented by E m while V f represents the fibre volume fraction.
4.2 The inverse rule of mixtures (IROM) model
This model [64], which has been developed to predict the lower-band modulus of composite materials having continuous and unidirectional fibres, is used in this study to estimate the modulus of elasticity of composite material having discontinuous randomly distributed and short fibres.
4.3 The pan model
Pan [65] developed techniques for 2-D and 3-D situations to estimate the tensile and shear moduli and Poisson’s ratio of randomly dispersed fibrous composite materials. Pan stated the difference of composite with randomly distributed fibres from unidirectional composite. This difference is best expressed by the fibre orientation, in other words, it is revealed by the relationship between V f and A f in the given direction. With the general relationship between V f and A f considering the fibre orientation density function, a model for the determination of elastic properties for the randomly dispersed fibre composite material was developed based on the IROM method. The model equations developed by Pan for the prediction of elastic moduli for 2-D and 3-D cases of a composite material with randomly dispersed fibres are given below.
4.4 The Manera model
Manera [66] stated that most of the models used for the prediction of properties of composite materials are based on the classical laminate analogy which assumes that the material consists of multiple or large numbers of omnidirectional orthotropic plies. The Manera model, combined with classical micromechanical models, tried to reveal composite parameters that are difficult to predict. In this model, based on the classical laminate analogy by Halpin et al. [67], the micromechanical formulation of Puck [68] and the invariants defined by Tsai and Pagano [69] were combined.
The elastic modulus of the composite (E c) can be obtained by laminate analogy using Puck equations and invariant properties of the composite.
The invariants identified by Tsai and Pagano are U 1 and U 4.
It is seen that the following relationship emerges between Equations (6) and (7).
In light of these, the model developed by Manera for predicting the modulus of elasticity of a composite material containing randomly dispersed short fibres is described below.
4.5 The Nielsen–Chen model
In the study conducted by Vannan et al. [70], a micromechanical model integrated into software, which was obtained using models such as Mori and Tanaka method and Nielsen–Chen micromechanical model, predicted the elastic modulus of a composite with discontinuous short fibres. The experimentally obtained modulus of elasticity was compared with the Nielsen–Chen micromechanical model, Shear-lag micromechanical model and the software developed by the author. Vannan and Vizhian [70] expressed the Nielsen–Chen model as follows.
Here, E ∏ is young’s modulus parallel to the axis of load direction and E ⊥ is perpendicular to axis of loading.
4.6 The Halpin–Tsai model
Halpin [71] reported that the success of the lamination approach depends on the assumption of average physical volume and the estimation of stiffness and expansion properties. In the laminated plate theory, a laminate reinforced with short and randomly dispersed fibres is classified as a quasi-isotropic material according to Tsai and Pagano [69]. Accordingly, the following is the micromechanical model for the modulus of elasticity of a composite reinforced with short and random dispersed fibres proposed by Halpin–Tsai. L represents the fibre length, D represents the fibre diameter, ξ represents fibre aspect ratio and η is defined as a function of ξ, E f and E m.
5 Results and discussion
At the stage of production of the CFF/PLA biocomposite, the mixing ratio between the CFF and PLA matrix was realized by mass; however, at the stage of prediction the elastic modulus with the micromechanical models, volumetric ratios were used. Therefore, it is necessary to convert the fibre mass ratio to the fibre volumetric ratio, and this conversion is carried out using the following equation [64].
In Equation (16), M f and ρ f represent the mass ratio and density of the fibre, respectively. In addition, M c represents the mass ratio of the biocomposite, and ρ m indicates the density of the matrix. Volumetric fraction equivalents of the amount of reinforcement CFF in the content of biocomposite based on Equation (16) are given in Table 1.
% mass and volumetric fractions of CFF.
| % mass fractions (w) of CFF | % volumetric fractions (V f ) of CFF |
|---|---|
| 0 | 0 |
| 2 | 2.76 |
| 5 | 6.83 |
| 10 | 13.4 |
5.1 The elastic modulus of chicken feather fibre
In the microscopic examination, it was found that the cross-sectional diameter of the barb decreases from the part to the end of the barb, and thus the cross-sectional area decreases. The microscopic measurement of the diameter of a single CFF is given in Figure 5. The variation of the cross-sectional area of a CF is given schematically in Figure 6. Accordingly, the cross-sectional area of the fibre at lengths 0, x and L is A 0, A x and A L , respectively.
Microscopic examination of an aged CFF diameter.
Schematic view of the cross-sectional area of a CFF with irregular diameter with and without force applied in the tensile direction.
A stress–strain graph of the tensile test results obtained from long-term aged CFF is given in Figure 7. While the elastic modulus of the aged CFF is determined from the slope of the linear section of this graph, the elastic modulus of non-aged CFF is taken from the study of Özmen et al. [46, 53].
Stress–strain graph of the long-term aged single CFF.
The following equations are used to calculate the effective area in the stress calculation [60]. Assuming that the fibre diameter changes linearly along the length of the fibre, A x becomes;
The effective fibre cross-sectional area, A e, is expressed using the following equation:
Local strain (ε x ) and stress (σ x ) are expressed in the equations followed. E represents the modulus of elasticity and F represents the applied force.
When Equation (20) used in Equation (19), ε x becomes,
The following equations are used to calculate the change in fibre length (l) along the fibre.
When we replace Equation (18) in Equation (22), we have
If Equation (23) is rearranged to obtain E,
5.2 Elastic modulus of poly(lactic acid)
The elastic modulus of PLA, which is selected as the matrix of aged [55] and non-aged [46, 53] biocomposite materials, is experimentally determined with a tensile device to achieve at least 5 valid results in accordance with ISO 527 standard. E m, which is the elastic constant used to predict the elastic modulus of composite from micromechanical analysis, is obtained from the stress–strain graphs from the studies conducted by Akderya et al. [55] and Özmen et al. [46]. The values obtained from the slope of the linear part of the stress–strain graph are given in Table 2.
5.3 Elastic moduli of chicken feather fibre/poly(lactic acid) biocomposites with micromechanical models
In this section of this study, results found using the micromechanical models are compared with obtained experimental data for both aged and non-aged [46, 53] CFF/PLA biocomposites, and the convergencies between them are illustrated in Figure 8. In addition to this, the elastic modulus of both aged and non-aged [46, 53] CFF and pure PLA, which are used to calculate elastic modulus of CFF/PLA biocomposites by micromechanical models, are tabulated in Table 2.
Bar charts of CFF/PLA biocomposite elastic moduli determined by micromechanical models and experiments.
Elastic moduli values of non-aged CFF/PLA biocomposites, CFF and pure PLA and micromechanical models used in this paper are taken from the study conducted by Özmen and Baba [46, 53]. Elastic moduli of long-term aged ones are obtained both experimentally and using micromechanical models applied to non-aged ones. It is aimed to determine the micromechanical model(s) that can predict the values closest to the experimental results for both the aged and the non-aged samples.
In the Christensen–Waals model as shown in Figure 8a, the error ratio increases as the CFF volume fraction increases in non-aged and aged samples. Furthermore, the error ratio is higher in aged samples than in non-aged ones. For CFF/PLA biocomposite with the highest mass fractions of CFF, the difference between the experimental and the micromechanical result is 14.13 % for the non-aged sample, while this difference is 47.99 % for the aged one.
When the micromechanical calculations are made with Pan 2D and Pan 3D models, it is realized that the modulus of elasticity values obtained by the 2D model can make closer approximations to the experimental values compared to the 3D model for both aged and non-aged samples as shown in Figure 8c and d. However, in both Pan models, accuracy decreases with increasing CFF volume fraction between experimental results and results obtained with the micromechanical approach. This difference is particularly noticeable for aged samples. For aged 2, 5 and 10 % CFF/PLA biocomposites, there is a 1.68, 7.41 and 25.13 % difference between Pan’s 2D micromechanical model approach and the experimental results. This difference is 6.39, 17.35 and 38.87 % for 2, 5 and 10 % CFF/PLA, respectively, when calculated with Pan’s 3D model.
Among all the micromechanical models used in this study, the Manera model (Figure 8e) has the highest errors from the experimental data for the elastic moduli of both aged and non-aged CFF/PLA biocomposites. It is observed that the error of the elastic moduli obtained with this model increases with increasing CFF content for all specimens. The highest error is observed in 10 % CFF/PLA samples, which is 16.74 % for the non-aged one and 51.06 % for the aged one.
Among the micromechanical models, IROM (Figure 8b), Nielsen–Chen (Figure 8f), and Halpin–Tsai (Figure 8g) models are the most reliable models that generate the results closest to the experimentally obtained elastic moduli for both aged and non-aged specimens. The error percentages from experimental data exhibited by the Nielsen–Chen micromechanical model for non-aged 2, 5 and 10 % CFF/PLA are 1.36, 0.53 and 0.40 %, respectively, and in addition, for the aged ones these rates are 5.53, 2.26 and 5.24 %, respectively. Based on the IROM, Nielsen–Chen and Halpin–Tsai models, as the volumetric concentration of CFF increases, the convergence between the experimental results and the values obtained by the micromechanical approach increases, which can be expressed as the opposite behaviour compared to other models. Halpin–Tsai and IROM models show the same error values, which are 7.06, 5.93 and 1.74 % for aged 2, 5 and 10 % CFF/PLA biocomposites, respectively. In Halpin–Tsai model, the inclusion of ξ and η in the calculation makes a difference to the IROM model in the 6th digit in the decimal place.
The reason why the Nielsen–Chen, IROM and Halpin–Tsai micromechanical model approaches provide good agreement results with the experimentally available modulus of elasticity for both aged and non-aged samples is the fact that these models successfully include significant critical factors in the calculation of the elastic modulus. Random distribution of fibres, random orientation of fibres, irregular shape of fibres, interphase zone between fibre and matrix are some of the critical factors expressed by Hassanzadeh-Aghdam et al. [72]. In addition, the non-isotropic nature of the CFF can be considered among these criteria.
The Nielsen–Chen, Halpin–Tsai and IROM models exhibit good agreement with the experimental results of elastic moduli. Therefore, these models can provide accurate results under non-aged and aged conditions. According to these models, in aged conditions, the highest error of 7.06 % elicited by Halpin–Tsai model for 2 % CFF/PLA samples.
The modulus of elasticity predicted by the micromechanical models and those achieved experimentally are schematized in Figure 9. Overall, Christensen–Waals, Pan (2D), Pan (3D) and Manera provide reliable results for biocomposites reinforced only with low percentage (2 %) CFF, while IROM, Nielsen–Chen and Halpin–Tsai models provide more reliable results as the percentage CFF in the biocomposite increases. However, among all these models, the Nielsen–Chen model provides the most reliable results for aged and non-aged CFF/PLA biocomposites with different percentages (2, 5 and 10 %) of CFF.
Elastic modulus obtained with micromechanical models for (a) non-aged, and (b) aged CFF/PLA biocomposites.
The differences between the results of micromechanical models and experiments can be observed as a result of nonhomogeneous dispersion and distinct physical properties of CFF and the presence of the CFF/PLA interface. The micromechanical models are based on multiple basic assumptions. One of these basic assumptions is that composite phases are considered to be homogeneous. There may be differences in the physical properties of natural fibres due to their inhomogeneity. It is known that natural fibres are not homogeneous even among themselves, since they do not have the same physical properties [73]. Another assumption is that there is an excellent bond between the matrix and the fibre [74]. After the manufacturing process, an excellent bond at reinforcement–matrix interfaces is rarely encountered. The production of an excellent composite material capable of meeting these assumptions is rarely encountered in real life [74, 75]. Thus, the experimental data may differ from the values obtained by micro-model approaches. Therefore, there is a need for the development of new micromechanical models or to add coefficients to existing micromechanical models. The new or revised models need to include statistical information on the actual distribution of the reinforcement within the composite, take into account that the bond between the matrix–reinforcement may be affected at the end of processes such as production or ageing. In addition, phases that cause a significant change in the elastic modulus of the composite need to result in fluctuations in the elastic modulus values obtained from the models.
The decreases in properties of polymer-based materials are influenced strongly by factors such as thermal, environmental and chemical ageing, mechanical stress, environmental exposure throughout their service life [76]. These effects trigger the decrease of the free volume which is also closely related to the decline in mechanical properties such as reduction in damping, brittleness and shrinkage. The free volume affects the mobility of most polymer chains and has an effect on mechanical properties [51]. The effect of the long-term ageing process on mechanical properties is that the elastic modulus of CFF/PLA specimens having the same CFF mass fraction is lower compared to that of non-aged ones, and this can be readily seen by examining the experimental results in Figure 9. Ageing has a significant effect on the physical properties of the amorphous phase in glassy or partially glassy polymeric materials. The effect of the ageing process has occurred around the glass transition temperature of the material, and it becomes noticeable by a decline in enthalpy and entropy, the observation of shrinkage in a specific volume and a decline in the mobility of molecules which can be related to mechanical properties of the material [51].
The degradation process which can be triggered by stress, heat, radiation, oxygen or chemical environments, means partial decomposition of the polymer. The degradation of polymeric materials occurs particularly during the processing and during daily use. The rate of degradation depends on the chemical structure and crystallinity of the polymeric material, defects in its structure and the nature of the environment in which the polymer is used or located [47, 76]. The phenomenon of degradation constitutes some factors that cause irreversible changes in the properties of the material, and some of these factors are crosslinking, thermal oxidation and chain scission. The decrease in mechanical properties of CFF and pure PLA is due to macromolecular chain scissions by exposure to long-term ageing [76].
The FTIR spectra are used to evaluate the effects of the long-term ageing process on the structure of CFF, PLA and CFF/PLA biocomposites in Figure 10. The infrared spectra of aged and non-aged chicken feather fibres are shown in Figure 10a. The absorption band at 3280 cm−1 represents the N–H (amine) group of an amino acid from CFF. The absorption band at 2923 cm−1 represents the C–H stretching vibrations, and the absorption band at 1647 cm−1 represents C=C stretching vibration. The band at 1537 cm−1 indicates C=C bending. The vibration absorption band at 1235 cm−1 represents C–O (carboxylic acid) resulting from an amino acid of the CFF [45, 77]. When the spectra of the aged and non-aged CFF are evaluated together, the band intervals of the characteristic data that are distinct do not change; however, the intensity of the bands weakens occasionally with the effect of the ageing process.
FTIR spectra of (a) aged and non-aged CFF, (b) aged and non-aged PLA, and (c) aged CFF, PLA and CFF/PLA biocomposites.
According to Figure 10b, the absorption band of PLA at 1764 cm−1 represents C=O stretching vibrations. Absorption bands at 1188 and 1086 cm−1 represent C–H stretching vibration of PLA. These characteristic data are distinct bands belonging to the ester group [78, 79]. In particular, by comparing the spectra of aged and non-aged PLA (Figure 10b), it is seen that the characteristic bands of PLA do not change with the long-term ageing effect, whereas the intensity of the bands weakens under the effect of the ageing process. Thus, it can be seen that the vibration absorption bands of PLA at the mentioned points (1086, 1188 and 1764 cm−1) are weakened.
The FTIR spectra of aged CFF, PLA and CFF/PLA biocomposites are given in Figure 10c. Accordingly, it is seen from Figure 10c that the characteristic bands of the spectrum of 2 and 10 % CFF/PLA biocomposites are weaker (lower transmittance values) than that of 5 % CFF/PLA. From this point of view, the increase in the difference between the elastic moduli provided by the Nielsen–Chen model, which is determined the most reliable model in this study, and the data obtained experimentally for 2 and 10 % aged CFF/PLA samples, reveals that there is a concordance between FTIR results and the study.
The disappearing or decreasing band indicates that the chain is completely broken or the macromolecular backbone is divided into short chain structures, respectively [78]. The chain scission mechanism generated by the ageing process can be understood from the decrement in the bands of the aged and the non-aged sample. In addition, the aged PLA shows the same characteristic absorption bands as the non-aged PLA. This is an indication that no new bond or strong chemical interaction occur in PLA with the effect of the ageing process [80]. From the point of mechanical properties of the composite material, it can be correlated that the decrease in mechanical properties is due to the deterioration of the bonds causing fibre matrix interactions.
5.4 Morphological analyses
In this study on prediction of modulus of elasticity by using micromechanical models, it is seen that especially three models come to the fore. The Nielsen–Chen, Halpin–Tsai and IROM models show superior success in predicting the modulus of elasticity of non-aged samples and maintain this success for aged ones. It is thought that the slight increase in the error in the prediction of the modulus of elasticity of aged samples may be due to the parameters which micromechanical approach models do not consider such as quality of adhesion between fibre and matrix. In order to determine these parameters, SEM micrographs of the fracture surfaces of non-aged and aged 2 % CFF/PLA bio-composite samples obtained after tensile tests are presented in Figure 11 for evaluation. The area highlighted by the red frame indicates the gaps between the fibre and the matrix, the blue ones indicate microvoids and the orange one indicates cavities.
SEM micrographs of (a) non-aged, (b) aged 2 % CFF/PLA biocomposite.
Comparing the fracture surfaces of non-aged and aged 2 % CFF/PLA biocomposite samples, it is noticed that the aged has more cavities, microvoids and gaps between the fibre and the matrix. As an effect of the ageing process, the cavity and microvoids in the aged sample increased, whereas the interfacial bonding success between fibre and matrix, one of the most significant factors affecting the mechanical properties of biocomposite samples [81], decreased. The success of the bond between the fibre and the polymeric matrix in a composite material is important as it affects the physical and mechanical properties of the composite material. The ability to bond fibre and matrix regularly across the interface is a combination of several parameters, some of which are interface strength, interface layer, interface thickness, and surface energy of the fibres. Any parameter that deteriorates the interface quality detaches the results of the micromechanical analysis from their actual values [82, 83].
6 Conclusions
In this study, the elastic moduli of long-term aged CFF/PLA biocomposite samples were found by using micromechanical models, and the reliability of the micromechanical models was investigated by comparing the obtained results with experimental results. The following comments can be made in accordance with the micromechanical approach and experimental results:
The Christensen–Waals micromechanical model showed close to experimental results in predicting the modulus of elasticity of composite materials containing 2 % fibre, while it shows significant errors in composite materials containing more than 2 % fibre. The convergence decreases as fibre content increases; the highest error is observed in 10 % CFF/PLA biocomposites.
Pan’s 2D and 3D micromechanical models have failed to predict the modulus of elasticity of aged CFF/PLA biocomposites and the accuracy decreased as the CFF content of aged samples increased.
Among the micromechanical models used, the most serious unsuccessful prediction attempts were made with the Manera model.
Nielsen–Chen, Halpin–Tsai and IROM micromechanical models have developed in-situ and reliable predictions of elastic moduli of both aged and non-aged biocomposite samples. Halpin–Tsai and IROM models have made accurate predictions of elastic moduli of 10 % CFF/PLA non-aged biocomposites, in particular.
Based on the results presented here, it has been found that some micromechanical models have the ability to reliably predict the critical parameters such as elastic moduli of biocomposite materials at the design stage. Moreover, estimations and insight into the long-term mechanical performance of biocomposites can be obtained using these micromechanical models that provide reliable predictions. However, it is clear that reliability decreases when it comes to the ageing effect. The main reason for this is the degradation of the fibre–matrix interface adhesion quality due to the ageing effect. Most of the developed micromechanical models attempt to reveal composite elastic properties, mainly based on the properties of the fibre and matrix. Models considering fibre–matrix interaction are limited, and the added coefficients for this purpose are often insufficient to reflect reality. For this reason, it is necessary to add coefficients to include the interface interaction in micromechanical models.
-
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
-
Research funding: None declared.
-
Conflict of interest statement: The authors declare no conflicts of interest regarding this article.
References
1. Abhilash, M., Thomas, D. In Biopolymer Composites in Electronics; Sadasivuni, K. K., Cabibihan, J. J., Ponnamma, D., Al-Maadeed, M. A. S. A., Kim, J., Eds. Elsevier: Amsterdam, Netherlands, 2017; pp. 405–435.10.1016/B978-0-12-809261-3.00015-2Search in Google Scholar
2. Rangappa, S. M., Siengchin, S., Parameswaranpillai, J., Jawaid, M., Ozbakkaloglu, T. Polym. Compos. 2022, 43, 645–691. https://doi.org/10.1002/pc.26413.Search in Google Scholar
3. Girijappa, Y. G. T., Rangappa, S. M., Parameswaranpillai, J., Siengchin, S. Front. Mater. 2019, 6, 1–14. https://doi.org/10.3389/fmats.2019.00226.Search in Google Scholar
4. Vinod, A., Sanjay, M. R., Suchart, S., Jyotishkumar, P. J. Clean. Prod. 2020, 258, 120978. https://doi.org/10.1016/J.JCLEPRO.2020.120978.Search in Google Scholar
5. Mohammed, L., Ansari, M. N. M., Pua, G., Jawaid, M., Islam, M. S. Int. J. Polym. Sci. 2015, 2015, 1–15. https://doi.org/10.1155/2015/243947.Search in Google Scholar
6. Sanjay, M. R., Madhu, P., Jawaid, M., Senthamaraikannan, P., Senthil, S., Pradeep, S. J. Clean. Prod. 2018, 172, 566–581. https://doi.org/10.1016/J.JCLEPRO.2017.10.101.Search in Google Scholar
7. Yu, L., Dean, K., Li, L. Prog. Polym. Sci. 2006, 31, 576–602. https://doi.org/10.1016/j.progpolymsci.2006.03.002.Search in Google Scholar
8. Sanjay, M. R., Siengchin, S., Parameswaranpillai, J., Jawaid, M., Pruncu, C. I., Khan, A. Carbohydr. Polym. 2019, 207, 108–121. https://doi.org/10.1016/J.CARBPOL.2018.11.083.Search in Google Scholar
9. Bassas-Galia, M., Follonier, S., Pusnik, M., Zinn, M. In Bioresorbable Polymers for Biomedical Applications; Perale, G., Hilborn, J., Eds. Woodhead Publishing: Cambridge, UK, 2017; pp. 31–64.10.1016/B978-0-08-100262-9.00002-1Search in Google Scholar
10. Kamarudin, S. H., Abdullah, L. C., Aung, M. M., Ratnam, C. T., Jusoh Talib, E. R. Mater. Res. Express 2018, 5, 085314. https://doi.org/10.1088/2053-1591/aabb89.Search in Google Scholar
11. Huda, M. S., Drzal, L. T., Mohanty, A. K., Misra, M. Compos. Sci. Technol. 2008, 68, 424–432. https://doi.org/10.1016/j.compscitech.2007.06.022.Search in Google Scholar
12. Chakoli, A. N., Sui, J., Amirian, M., Cai, W. J. Polym. Res. 2011, 18, 1249–1259. https://doi.org/10.1007/s10965-010-9527-9.Search in Google Scholar
13. Siakeng, R., Jawaid, M., Ariffin, H., Sapuan, S. M., Asim, M., Saba, N. Polym. Compos. 2019, 40, 446–463. https://doi.org/10.1002/pc.24747.Search in Google Scholar
14. Madhavan Nampoothiri, K., Nair, N. R., John, R. P. Bioresour. Technol. 2010, 101, 8493–8501. https://doi.org/10.1016/j.biortech.2010.05.092.Search in Google Scholar PubMed
15. Chun, K. S., Husseinsyah, S., Osman, H. J. Polym. Res. 2012, 19, 1–8. https://doi.org/10.1007/s10965-012-9859-8.Search in Google Scholar
16. Yokohara, T., Yamaguchi, M. Eur. Polym. J. 2008, 44, 677–685. https://doi.org/10.1016/j.eurpolymj.2008.01.008.Search in Google Scholar
17. Pang, X., Zhuang, X., Tang, Z., Chen, X. Biotechnol. J. 2010, 5, 1125–1136. https://doi.org/10.1002/biot.201000135.Search in Google Scholar PubMed
18. Petinakis, E., Yu, L., Simon, G., Dean, K. In Fiber Reinforced Polymers – The Technology Applied for Concrete Repair; Masuelli, M. A., Ed. IntechOpen: London, 2013.Search in Google Scholar
19. Mukherjee, T., Kao, N. J. Polym. Environ. 2011, 19, 714–725. https://doi.org/10.1007/s10924-011-0320-6.Search in Google Scholar
20. Kale, G., Kijchavengkul, T., Auras, R., Rubino, M., Selke, S. E., Singh, S. P. Macromol. Biosci. 2007, 7, 255–277. https://doi.org/10.1002/mabi.200600168.Search in Google Scholar PubMed
21. Iovino, R., Zullo, R., Rao, M. A., Cassar, L., Gianfreda, L. Polym. Degrad. Stab. 2008, 93, 147–157. https://doi.org/10.1016/j.polymdegradstab.2007.10.011.Search in Google Scholar
22. Xie, L., Xu, H., Wang, Z. P., Li, X. J., Bin Chen, J., Zhang, Z. J., Yin, H. M., Zhong, G. J., Lei, J., Li, Z. M. J. Polym. Res. 2014, 21, 1–15. https://doi.org/10.1007/s10965-014-0357-z.Search in Google Scholar
23. Gupta, B., Revagade, N., Hilborn, J. Prog. Polym. Sci. 2007, 32, 455–482. https://doi.org/10.1016/j.progpolymsci.2007.01.005.Search in Google Scholar
24. Shen, L., Worrell, E., Patel, M. Biofuels, Bioprod. Biorefining 2010, 4, 25–40. https://doi.org/10.1002/bbb.189.Search in Google Scholar
25. Wu, C. S., Tsou, C. H. J. Polym. Res. 2019, 26, 1–9. https://doi.org/10.1007/s10965-019-1710-z.Search in Google Scholar
26. Psomiadou, E., Arvanitoyannis, I., Biliaderis, C. G., Ogawa, H., Kawasaki, N. Carbohydr. Polym. 1997, 33, 227–242. https://doi.org/10.1016/S0144-8617(97)00032-5.Search in Google Scholar
27. Murariu, M., Bonnaud, L., Yoann, P., Fontaine, G., Bourbigot, S., Dubois, P. Polym. Degrad. Stab. 2010, 95, 374–381. https://doi.org/10.1016/J.POLYMDEGRADSTAB.2009.11.032.Search in Google Scholar
28. Cohn, D., Hotovely Salomon, A. Biomaterials 2005, 26, 2297–2305. https://doi.org/10.1016/j.biomaterials.2004.07.052.Search in Google Scholar
29. Hassan, M. M., Koyama, K. J. Polym. Res. 2020, 27, 1–11. https://doi.org/10.1007/s10965-019-1991-2.Search in Google Scholar
30. Ochi, S. Mech. Mater. 2008, 40, 446–452. https://doi.org/10.1016/j.mechmat.2007.10.006.Search in Google Scholar
31. Mohanty, A. K., Misra, M., Hinrichsen, G. Macromol. Mater. Eng. 2000, 276–277, 1–24. https://doi.org/10.1002/(SICI)1439-2054(20000301)276:1<1::AID-MAME1>3.0.CO;2-W.10.1002/(SICI)1439-2054(20000301)276:1<1::AID-MAME1>3.0.CO;2-WSearch in Google Scholar
32. Ramamoorthy, S. K., Skrifvars, M., Persson, A. Polym. Rev. 2015, 55, 107–162. https://doi.org/10.1080/15583724.2014.971124.Search in Google Scholar
33. Nunna, S., Chandra, P. R., Shrivastava, S., Jalan, A. K. J. Reinforc. Plast. Compos. 2012, 31, 759–769. https://doi.org/10.1177/0731684412444325.Search in Google Scholar
34. Ku, H., Wang, H., Pattarachaiyakoop, N., Trada, M. Compos. B Eng. 2011, 42, 856–873. https://doi.org/10.1016/j.compositesb.2011.01.010.Search in Google Scholar
35. Jiang, N., Yu, T., Li, Y., Pirzada, T. J., Marrow, T. J. Compos. Sci. Technol. 2019, 173, 15–23. https://doi.org/10.1016/J.COMPSCITECH.2019.01.018.Search in Google Scholar
36. Li, W., He, X., Zuo, Y., Wang, S., Wu, Y. Int. J. Biol. Macromol. 2019, 141, 325–332. https://doi.org/10.1016/J.IJBIOMAC.2019.09.005.Search in Google Scholar
37. Shih, Y. F., Huang, C. C. J. Polym. Res. 2011, 18, 2335–2340. https://doi.org/10.1007/s10965-011-9646-y.Search in Google Scholar
38. IARC Monogr. Eval. Carcinog. Risks to Humans 100 C 2011, 219–309.Search in Google Scholar
39. Bloise, A., Kusiorowski, R., Gualtieri, A. F. Minerals 2018, 8, 1–13. https://doi.org/10.3390/min8070274.Search in Google Scholar
40. Bolormaa, B., Drean, J. Y., Enkhtuya, D. J. Nat. Fibers 2007, 4, 1–11. https://doi.org/10.1080/15440470801893281.Search in Google Scholar
41. Bharath, K. N., Pasha, M., Nizamuddin, B. A. J. Ind. Textil. 2016, 45, 730–751. https://doi.org/10.1177/1528083714540698.Search in Google Scholar
42. Bansal, G., Singh, V. K., Gope, P. C., Gupta, T. J. Graph. Era Univ. 2017, 5, 16–24.Search in Google Scholar
43. Bartels, T. J. Exp. Zool. B Mol. Dev. Evol. 2003, 298, 91–108. https://doi.org/10.1002/jez.b.28.Search in Google Scholar PubMed
44. Belarmino, D. D., Ladchumananandasivam, R., Belarmino, L. D., Pimentel, J. R. d. M., da Rocha, B. G., Galvão, A. O., de Andrade, S. M. B. Mater. Sci. Appl. 2012, 03, 887–893. https://doi.org/10.4236/msa.2012.312129.Search in Google Scholar
45. Carrillo, F., Rahhali, A., Cañavate, J., Colom, X. J. Reinforc. Plast. Compos. 2013, 32, 1419–1429. https://doi.org/10.1177/0731684413500546.Search in Google Scholar
46. Baba, B. O., Özmen, U. Polym. Compos. 2017, 38, 837–845. https://doi.org/10.1002/pc.23644.Search in Google Scholar
47. Kotsilkova, R., Angelova, P., Batakliev, T., Angelov, V., Di Maio, R., Silvestre, C. Coatings 2019, 9, 1–14. https://doi.org/10.3390/coatings9060355.Search in Google Scholar
48. Rasselet, D., Ruellan, A., Guinault, A., Miquelard-Garnier, G., Sollogoub, C., Fayolle, B. Eur. Polym. J. 2014, 50, 109–116. https://doi.org/10.1016/j.eurpolymj.2013.10.011.Search in Google Scholar
49. Araújo, M. C., Martins, J. P., Mirkhalaf, S. M., Lanceros-Mendez, S., Pires, F. M. A., Simoes, R. Appl. Surf. Sci. 2014, 306, 37–46. https://doi.org/10.1016/j.apsusc.2014.03.072.Search in Google Scholar
50. Mirkhalaf, S. M., Fagerström, M. Mech. Time-Dependent Mater. 2019, 25, 119–131. https://doi.org/10.1007/s11043-019-09429-w.Search in Google Scholar
51. Acioli-Moura, R., Sun, X. S. Polym. Eng. Sci. 2008, 48, 829–836. https://doi.org/10.1002/pen.21019.Search in Google Scholar
52. Sin, L. T., Rahmat, A. R., Rahman, W. A. W. A. Polylactic Acid; Elsevier: Oxford, UK, 2013, 177–2019. https://doi.org/10.1016/B978-1-4377-4459-0.00005-6.Search in Google Scholar
53. Özmen, U., Baba, B. O. Mater. Tehnol. 2016, 50, 141–146. https://doi.org/10.17222/mit.2014.182.Search in Google Scholar
54. Akderya, T., Özmen, U., Baba, B. O. J. Polym. Res. 2020, 27, 162. https://doi.org/10.1007/s10965-020-02132-2.Search in Google Scholar
55. Akderya, T., Özmen, U., Baba, B. O. Fibers Polym. 2021, 22, 2602–2611. https://doi.org/10.1007/s12221-021-0304-7.Search in Google Scholar
56. Reddy, N., Yang, Y. J. Polym. Environ. 2007, 15, 81–87. https://doi.org/10.1007/s10924-007-0054-7.Search in Google Scholar
57. Butler, M., Johnson, A. S. J. Exp. Biol. 2004, 207, 285–293. https://doi.org/10.1242/jeb.00746.Search in Google Scholar PubMed
58. ISO 527-1:2019 Plastics – Determination of tensile properties, https://www.iso.org/standard/75824.html (Accessed Jul 18, 2022).Search in Google Scholar
59. Özmen, U., Baba, B. O. J. Therm. Anal. Calorim. 2017, 129, 347–355. https://doi.org/10.1007/s10973-017-6188-5.Search in Google Scholar
60. Zhan, M., Wool, R. P. Polym. Compos. 2011, 32, 937–944. https://doi.org/10.1002/pc.21112.Search in Google Scholar
61. Malekmohammadi, S., Tressou, B., Nadot-Martin, C., Ellyin, F., Vaziri, R. J. Compos. Mater. 2014, 48, 1857–1874. https://doi.org/10.1177/0021998313490977.Search in Google Scholar
62. Wang, Y., Huang, Z. M. Materials 2018, 11, 1–15. https://doi.org/10.3390/ma11101919.Search in Google Scholar PubMed PubMed Central
63. Christensen, R. M., Waals, F. M. J. Compos. Mater. 1972, 6, 518–535. https://doi.org/10.1177/002199837200600307.Search in Google Scholar
64. Kaw, A. K. Mechanics of Composite Materials, 2nd ed.; CRC Press: Boca Raton, Florida, USA, 2005.Search in Google Scholar
65. Pan, N. Sci. Eng. Compos. Mater. 1996, 5, 63–72.Search in Google Scholar
66. Manera, M. J. Compos. Mater. 1977, 11, 235–247. https://doi.org/10.1177/002199837701100208.Search in Google Scholar
67. Halpin, J. C., Jerine, K., Whitney, J. M. J. Compos. Mater. 1971, 5, 36–49. https://doi.org/10.1177/002199837100500104.Search in Google Scholar
68. Puck, A. Kunststoffe, 1967, 57, 965–973.Search in Google Scholar
69. Tsai, S. W., Pagano, N. J. In Composite Materials Workshop; Tsai, S. W., Halpin, J. C., Pagano, N. J., Eds. Technomic Publishing Co., Lancaster, PA, 1968, 233–252.Search in Google Scholar
70. Vannan, E., Vizhian, P. J. Miner. Mater. Char. Eng. 2014, 02, 61–69. https://doi.org/10.4236/jmmce.2014.21010.Search in Google Scholar
71. Halpin, J. C. J. Compos. Mater. 1969, 3, 732–734. https://doi.org/10.1177/002199836900300419.Search in Google Scholar
72. Hassanzadeh-Aghdam, M. K., Ansari, R., Darvizeh, A. Mech. Adv. Mater. Struct. 2018, 1–15. https://doi.org/10.1080/15376494.2018.1472336.Search in Google Scholar
73. Wallenberger, F. T., Weston, N. E. Natural Fibers, Plastics and Composites; Springer: US, 2004.10.1007/978-1-4419-9050-1Search in Google Scholar
74. Sendeckyj, G. P. Mechanics of Composite Materials; Elsevier: Academic Press, Cambridge, Massachusetts, USA, 1974.Search in Google Scholar
75. Broutman, L. J. J. Adhes. 1970, 2, 147–160. https://doi.org/10.1080/0021846708544590.Search in Google Scholar
76. Moraczewski, K., Stepczyńska, M., Malinowski, R., Karasiewicz, T., Jagodziński, B., Rytlewski, P. Polymers 2019, 11, 1–20. https://doi.org/10.3390/polym11040575.Search in Google Scholar PubMed PubMed Central
77. Salehuddin, S. M. F., Wahit, M. U., Kadir, M. R. A., Sulaiman, E., Abu Kasim, N. H. Malaysian J. Anal. Sci. 2014, 18, 368–375.Search in Google Scholar
78. Bao, Y. J. Thermoplast. Compos. Mater. 2018, 31, 1341–1351. https://doi.org/10.1177/0892705717738290.Search in Google Scholar
79. Oliveira, M., Santos, E., Araújo, A., Fechine, G. J. M., Machado, A. V., Botelho, G. Polym. Test. 2016, 51, 109–116. https://doi.org/10.1016/j.polymertesting.2016.03.005.Search in Google Scholar
80. Chieng, B. W., Ibrahim, N. A., Wan Yunus, W. M. Z., Hussein, M. Z. Adv. Mater. Res.; Trans Tech Publications Ltd; Bäch SZ, Switzerland, 2014; pp. 136–139.Search in Google Scholar
81. Gan, Y. X. Int. J. Mol. Sci. 2009, 10, 5115–5134. https://doi.org/10.3390/ijms10125115.Search in Google Scholar PubMed PubMed Central
82. Aliotta, L., Gigante, V., Coltelli, M. B., Cinelli, P., Lazzeri, A. Int. J. Mol. Sci. 2019, 20, 1–14. https://doi.org/10.3390/ijms20040960.Search in Google Scholar PubMed PubMed Central
83. Lazzeri, A., Phuong, V. T. Compos. Sci. Technol. 2014, 93, 106–113. https://doi.org/10.1016/j.compscitech.2014.01.002.Search in Google Scholar
© 2022 Walter de Gruyter GmbH, Berlin/Boston
