5.2.1 Single-fibre fragmentation test

The single-fibre fragmentation test (SFFT) method was originally used with metals by Kelly and Tyson¹⁴ who observed a multiple-fibre fracturing phenomenon in a system consisting of a low concentration of brittle tungsten fibres embedded in a copper matrix.² The main assumptions made by Kelly and Tyson were that the matrix yields plastically, the fibre–matrix adhesion is perfect and the shear stress along the fibre is constant and equal to the shear yield strength of the matrix.¹⁵ in addition, a single fragmentation test to measure the interfacial shear strength is available for those composite systems where the ultimate elongation of the matrix is higher than the fibre elongation. The test states that the strain at matrix fracture must be at least three times higher than that of the fibre.¹⁶ The method involves completely embedding a single fibre along the centreline of a relatively large dog-bone shaped specimen of matrix material (Fig. 5.3a–b). The specimen is then pulled in tension uniaxially along the fibre axis. Stress is transferred to the fibre by the matrix. As the strain is increased, the fibre will break. When the fibre first breaks, the matrix restrains the fibre fragments for returning to their unstressed dimension. However, the axial stress in the fibre at a distance from the break increases until it reaches the original stress in the fibre before the first break. Further increase in the specimen strain results in another break in the fibre (Fig. 5.4). This fibre breakage continues with increased strain until the fibre fragments become so small that the matrix can no longer transfer stress over a long enough distance to break the fibre.¹³ At this point, as no more breaks can occur with increased strain, the fragmentation process is saturated and the test is completed.¹⁷

When a matrix is transparent, fibre fracture phenomena can be easily observed through an optical microscope and fragment lengths can be determined. When the matrix used is a semicrystalline thermoplastic, after saturation observation of the fibre fragments by optical microscopy is difficult. In order to properly locate the fibre break points and measure the fragment lengths, the specimens can be melted.¹⁸

Some studies incorrectly assumed that the length of the broken fibre fragments, referred to as the fibre critical length l_c, is an indication of the ability of the polymeric matrix to transfer stress to the fibre.¹³ Kelly and Tyson defined the critical fragment length as the maximum fragment length over which no further fibre fracture occurs.¹⁹ However, the average of the measured fragmentation length, is not equal to l_c ^3,20 because, when the fibre length is just above l_c, the fibre can break into two pieces both having lengths less than l_c. When the fibre length is just below l_c, it can not break any further. Therefore, the measured fragmentation length, should be distributed between and l_c.³

The axial stress acting in a fibre fragment of a single fibre composite specimen varies along the fragment length. The two fibre fragment ends are unloaded: axial stress in the fibre at those points is zero. Moving along the fibre from the end towards the fragment centre, stress increases. This fibre stress build-up occurs owing to stress transfer by the surrounding matrix. For a perfectly plastic matrix ideally adhering to the fibre, the stress build-up in the fibre is rectilinear from the fibre ends to the centre,³ as represented schematically in Fig. 5.5 and 5.6, where are the fibre tensile strengths at different fibre lengths. For the case represented in Fig. 5.5a, the tensile stress that supports the fibre increases linearly from the fibre ends but the highest stress value is not enough to break the fibre. On the other hand, for the case represented in Fig. 5.5b, the fibre length is sufficient to reach the tensile strength value of embedded fibre and the fibre breaks.

The simplest way to relate average fibre length and critical length is the approximation³ shown in equation 5.1,

[5.1]

Several studies^3,9,18,21 assume a distribution function to determine The most commonly used distribution functions are the Gaussian and Weibull distributions.

Based on a force balance in a micromechanical model for a system of elastic fibre embedded in a plastic matrix¹⁷ and assuming that the stress transferred across the interface acts over a length l_c, a force balance can be obtained by setting the total force transferred across the interface equal to the breaking force of the fibre (Fig. 5.6a–b), where is the fibre strength at fibre length l_c and r is the fibre radius. Behind the apparent simplicity of this expression, there is a great deal of complexity because refers to the ultimate strength of very short lengths of the fibre and τ includes a combination of different modes of stress transfer.²² To correctly calculate τ, valid values of , r and l_c must be determined. The fibre tensile strength depends on the distribution of flaws along the length, thus it is strongly influenced by the gauge length. Hence it is incorrect to use the tensile strength values obtained for macroscopic gauge lengths.¹² On the other hand, the value of is too complicated to be experimentally measured because testing the fibre at very short gauge lengths of the order of l_c is impossible in most cases²³ or at least it can entail experimental difficulties.¹² Sometimes this strength value was obtained through a linear extrapolation of experimental data measured in single fibre tensile tests at longer gauge lengths.²⁴ Many studies obtained this strength by extrapolation using the two-parameter Weibull distribution function.^12,25 However, for several reasons, this value of fibre strength obtained may differ from the in situ fibre strength,²³ equation 5.2,

[5.2]

where is the fibre strength at a gauge length of l_o and ρ is the shape parameter of the Weibull distribution function for fibre strength.

The main difficulty in measuring the single-fibre diameter of natural fibres is that it can vary widely between two ends and there can also be differences in diameter among filaments.^26,27 cross-section of natural fibres is polygonal or oval, and somewhat irregularly ellipsoidal.²⁸⁻³⁰ As a result, any determination assuming a round cross-section is expected to give a higher fibre diameter, particularly by an optical method, and thus a lower value of τ ²⁶

As mentioned above, different methods are used to relate average fibre length and critical length. Therefore the calculation of τ depends on the values of _, r and l_c used. Calculation of τ gives an indication of the ability of the matrix to transfer stress to the fibre. Because there is normally no interfacial failure between fibre and matrix, it is incorrect to call τ the interfacial shear strength. Rather, τ is a measure of the stress transfer across the interface, and it should be referred to as the stress transfer or stress transfer coefficient.¹³

Even with the Kelly–Tyson model different approaches can be encountered for calculating the variables in the expression determining the stress transfer coefficient, i.e. the critical length l_c and tensile strength of the fibre at the critical length. Besides, the method assumes a constant shear stress across the interface but ignores the matrix properties, conditions that can hardly be met in real fibre–matrix systems. Therefore, the single-fibre fragmentation test method seems inappropriate for comparison between different systems even on a qualitative basis. However, the main advantage of this method is that it is applicable over a wide range of interfacial bond strengths and it can provide qualitative indications of interfacial bond strength in a given system.²

5.2.2 Fibre pull-out test

Although the fibre pull-out test is popular because of its conceptual simplicity, this test, however, presents some difficulties.¹⁶ For testing, a fibre is embedded in a very thin film or bead of polymer (Fig. 5.7–5.8) and the specimen is gripped in a tensile testing machine. The fibre is pulled out of the polymer and the force required to remove the fibre is measured.13 The adhesive strength is calculated by dividing the measured maximum load by the area of contact of the fibre with the polymer. The area of contact can be measured using optical microscopy. Because of difficulties in measuring the exact embedded surface area for each fibre, it is usually assumed that the fibres are perfectly cylindrical with a smooth surface, which may not be completely valid and it therefore affects the accuracy of calculation.³¹ The small diameter of the fibre requires that the area of interfacial contact be small, otherwise the fibre may break rather than pulling-out, thus altering the data analysis.¹³ Interfacial shear strength values can be obtained according to the Kellys–Tyson equation^7,28,32,33 by using the assumptions that the shear stress at the interface is uniformly distributed along the embedded length,^26,32 the interfacial loading is in shear mode and the friction between debonded fibre and matrix is negligible,³⁴ as shown by equation 5.3:

[5.3]

where F_max is the maximum force recorded by the load cell and d and l are fibre diameter and embedded length, respectively. For the fibre to be pulled out rather than broken, test conditions require that tensile stress in the pulled-out fibre is less than its ultimate tensile strength. Taking into account the Kelly–Tyson equation, the maximum embedded fibre length l_max is given by .As l_max is usually low, it causes experimental difficulties and a large data scatter.²

Figure 5.9 shows a typical pull-out test curve of a flax fibre–polypropylene system. This graph can be separated into three parts corresponding to the different stages involved in pull-out. During the first part of testing, the graph is considered to represent the rectilinear elastic behaviour of the fibre–matrix system and the fibre–matrix interface remains intact. For the second stage, after initiation, debonding occurs by means of crack propagation along the embedded fibre length. The third part occurs after complete debonding has taken place, where the remaining force is the result of frictional interactions between the fibre and the matrix.^26,35

Herrera-Franco et al.³⁶ studying high-density polyethylene (HDPE)–henequen fibre composite noted that pull-out graphs for various surface treatments exhibited the nonlinear behaviour characteristic of a ductile matrix. However, after the load reached its maximum value the graphs showed significant differences. For the native henequen, they observed that the load increased gradually and when it reached a maximum value there was a smooth transition and it began to decrease in a linear fashion until the total embedded length of the fibre was pulled-out. They mentioned that this behaviour agreed well with the behaviour of a poor interphase that resulted because of the incompatibility between the hydrophilic fibre and the hydrophobic matrix. However, they suggested that the load–displacement graph for a fibre treated with an aqueous NaOH solution and then pre-impregnated with dissolved HDPE seemed to be a very strongly bonded interphase because the pull-out force was higher and the pull-out process occurred catastrophically.

The fibre pull-out test is mainly suitable for studying various surface treatments in a given fibre–matrix system of strong fibres and low to medium interfacial bond strength. Comparison between different systems may be misleading. The fibre pull-out test is inadequate for weak fibres because the test gives a significant data scatter owing to the permissible embedded fibre lengths being extremely short.² Pull-out tests entail experimental difficulties like sample clamping and embedded fibre alignment problems, besides the meniscus at the point of a fibre entry into matrix sometimes makes it difficult to determine the exact embedded length of the fibre.¹⁶ As shown in equation [5.3], τ is defined as the ratio of the maximum force measured in a pull-out test to the contact area. This approach is very simplistic and, not surprisingly, gives a large scatter in the experimental data.³⁷ In addition, τ underestimates the interfacial shear strength because the interface fails where the maximum stress is located and the interfacial shear stress is not distributed evenly along the fibre. Moreover, τ decreases as the embedded length increases because of the uneven distribution of shear stress at the interface.³⁸ Interfacial shear stresses calculated from pull-out tests can be used only to estimate the bonding strength and compare the effect of the treatment on the fibre–matrix interfacial bonding strength.³¹

As a result, the apparent IFSS values obtained in various studies can not be compared with each other.^37,39 Pisanova et al.³⁷ proposed to characterize the interfacial bond strength in terms of the local interfacial shear strength. The local IFSS is a semi-empirical parameter that can be determined from experimental data by extrapolating τ to zero embedded length, equation [5.4],

[5.4]

5.2.3 Microdebonding test

The microdebonding test is a modified version of the pull-out test, proposed by Miller et al.⁴⁰ The test was developed for characterising the single-fibre–matrix interface in order to eliminate any meniscus effect.₄₁ In this test a small drop of matrix is deposited onto the fibre at some point. The fibre with the microbead is mounted in a micro-device and then the fibre is pulled out (Fig. 5.10a–b).⁴² The test has been widely used to evaluate τ for both thermoset and thermoplastic matrix composites⁴⁰’⁴¹ because sample preparation is simplified. Although this test is easy to perform, there are several concerns that have to be taken into account such as stress concentrations during specimen loading, non-uniform shear stress distribution along the fibre–matrix interface, geometry of the polymer drop, and the effect of the strain rate. All these factors significantly affect the test results and their scatter. The reliability of the data is dependent on the shape of the drop. Symmetric, round drops are easier to test and analyze than drops with flat surfaces. If the length of drop in contact with the fibre surface exceeds a critical value, the fibre will break before debonding and pull-out.⁴³ Miller et al.⁴⁰ noted a reduced variability in shear debonding results when the drop-sizes were larger, they suggested that the variability occurred because localized surface differences were more likely to be averaged out with a greater contact area.

Results obtained by various methods for characterization of the single fibre–polymer matrix interface adhesion values are not similar. However, several studies⁵’³⁶ observed that when different methods were used for the same system, the results were similar for the different tests.

5.3.1 Introduction

As mentioned above, for measuring interfacial shear strength there are many approaches. However, the experimental τ should be regarded as a mere adhesion parameter and not as the actual stress acting at the fibre–matrix interface.¹⁸ Apart from the intrinsic problems of each method for measuring interfacial shear strength, the use of natural fibres adds another problem associated with them, the highly anisotropic nature of this kind of fibres. The lack of homogeneity in these fibres makes the interpretation of interfacial adhesion test results difficult³⁴ and a wide scattering of results is seen. The fibres are highly anisotropic especially when the diameter is considered because their diameters vary from fibre to fibre.⁴⁴ In addition, several studies^5,45,46 have shown that the cross-section and apparent diameter of natural fibres vary considerably along their length. Some studies²⁸⁻³⁰ observed that the cross-section of natural fibres is not round but somewhat irregularly ellipsoidal and fibres show a hollow multicellular nature. As a result, any determination assuming a round cross-section is expected to give an inaccurate value.²⁸

The situation of single-fibre composites of natural fibre is complex because fibres themselves could be considered composite materials. In addition, if bundles of natural fibres are used instead of a single fibre, two types of interfaces should be considered, one between the fibre bundle and the matrix and the other between the cells.¹¹ Much information is available concerning the lignocellulosic fibre–polymer matrix interfacial adhesion (Table 5.1). An increase in interfacial shear strength for composites based on treated fibres is considered to be evidence of the treatment efficiency. Published results should be treated carefully owing to the assumptions made for data analysis and the lack of homogeneity of natural fibres properties. The obtained results are not immediately representative of the actual composite where performance increase is generally much lower.⁶

5.3.2 Effect of fibre surface modification

The adhesion between two solids is directly related to their surface energies and wetting is a requisite to get a good interfacial adhesion. Moreover, the main disadvantage of natural fibres is their hydrophilic nature, which lowers the compatibility with hydrophobic polymer matrices during composite fabrication.⁴ The effect of different chemical and physical treatments on natural fibre–polymer matrix interfacial shear strength has been investigated^7,8,16,18,26Fibre–matrix adhesion can be the result of combined effects of more than one adhesion mechanism, as mentioned above. Therefore, the study of adhesion of single-fibre composites of natural fibres is complex. Usually lignocellulose fibres after treatments become more hydrophobic indicating that treated fibres can be better wetted out by a hydrophobic matrix. A good adhesion is also desirable to prevent, or inhibit, environmental agents from invading and destroying the interface. For lignocellulosic fibres, the degradation caused by water at the interface is of primary concern because the fibres are highly hygroscopic.²⁷

Fibre surface treatments based on the use of sodium hydroxide, silane and maleic acid anhydride (MA) based copolymer are the most common.^5,7,26,47 Figures 5.11–5.14 show possible mechanisms of adhesion improvement for maleic anhydride polypropylene copolymer (MAPP), silane and sodium hydroxide fibre surface treatments,⁴⁸⁻⁵⁰ respectively. The reaction between cellulose and copolymer can be divided into two main steps, as shown in Fig. 5.12. In the first step, the copolymer is converted into the more reactive anhydride form and sterification of the cellulose fibres takes place in the second step.⁴⁸ As shown in Fig. 5.13, the silane can hydrolyse to some extent to form silanols, thus providing a link to cellulose through their –OH groups by the formation of hydrogen bonds. Mwaikambo et al.⁵⁰ suggested that alkalization of lignocellulosic fibres removed lignin, pectin, waxy substances, and natural oils covering the external surface of the fibre cell wall (Fig. 5.14). This treatment can reveal the fibrils, and gives a rough surface topography to the fibre, as shown by Arbelaiz et al.²⁶ (Figure 5.15a–b). Mwaikambo et al.⁵⁰ also mentioned that alkalization depolymerizes the native cellulose I molecular structure producing short crystallites (Fig. 5.14b).

In the following, a review of obtained interfacial adhesion data for different surface treatments of lignocellulosic fibre-polymer matrix interface adhesion data is presented. Arbelaiz et al.²⁶ studied flax fibres as reinforcement for polypropylene (PP) matrix composites. For improving compatibility between flax fibre bundles and PP matrix, fibre surface treatments with MA, MAPP copolymer, vinyltrimethoxysilane and alkali were carried out. The effect of surface modification on fibre–matrix interfacial shear strength was analyzed. The polar component of the surface energy was lowered by treatment and the treated fibres were better wetted out by the polypropylene matrix than the untreated ones. The apparent interfacial strength, as determined by pull-out tests, for a surface-modified single flax fibre showed that only MAPP and alkali treatment led to improvement of fibre–matrix adhesion increasing the IFSS by around 15 and 8%, respectively. Therefore, besides wetting characteristics, other surface properties should be taken into account for improving the interfacial strength. Arbelaiz et al.²⁶ suggested that for silane and MA treated fibres, the vinyl group, which should interact with the thermoplastic matrix, could not create entanglements, as the compound used for treatment is not long enough. Therefore, stresses could not be transferred from PP chains, through fibre bonded coupling agent, to the fibre. Therefore, to stress transfer, a minimum chain length is necessary.

Karlson et al.⁵¹ studied low-density polyethylene (LDPE) regenerated cellulose system by single fragmentation test. They determined the fibre surface perimeter by the Wilhelmy plate method using a nonpolar liquid that had a lower total surface free energy than regenerated cellulose in order to wet the fibre completely. They found that when regenerated cellulose fibres were mechanically treated, long and numerous twisted fibrils were observed on the fibre surface. The surface area of treated fibres was increased after treatment and, consequently, the interfacial shear strength increased significantly as a result of surface fibrillation. They mentioned that fibrils present on the surface could provide numerous mechanical anchoring sites for the polyethylene matrix.

Adusumalli et al.⁵² determined the interfacial shear strength of different fibre-polypropylene composites. They used regenerated cellulose, ramie and glass fibres. They showed that both cellulose fibres had lower interfacial shear strength than sized glass fibre, presumably because of the different chemical character of cellulose (polar, hydrophilic) and PP (nonpolar, hydrophobic). Suitable surface chemical properties led to good wetting and development of adhesion forces. In spite of regenerated cellulose and ramie being chemically very similar as they consist almost exclusively of cellulose, ramie fibres showed higher interfacial shear strength than regenerated cellulose, presumably because of the higher surface roughness observed for these fibres.

García-Hernández et al.⁴⁴ modified sugar cane bagasse fibres by surface treatments in order to improve fibre adhesion to polystyrene (PS) matrices. The fibres were modified by alkaline treatment, silane coupling agents, fibre coating with PS and grafting of PS on the fibre (with and without crosslinker). According to pull-out tests, the IFSS increased for all treated fibres, as compared with untreated ones. The highest IFSS improvement was obtained for grafting of PS on the fibres in the presence of a crosslinker. The improvement obtained with this treatment was partly attributed to a good interdiffusion between chains of PS in the matrix and the grafted PS with crosslinker. Scanning electronic microscopy (SEM) analysis revealed that after treatment PS aggregations distributed as dots appeared along the fibre surface, resulting in more interlocking sites that contributed to the adhesion.

López Manchado et al.⁵³ investigated the enhancement of interfacial adhesion of PP/ethylene propylene diene monomer (EPDM)-flax fibre composites using MA as compatibilizer. The interface between the fibre and polymer matrix was characterized by pull-out tests. The addition of small amounts of MA-modified matrices significantly increased the interfacial shear strength. This effect was more evident when the matrix was PP modified with MAPP addition. This improvement was attributed to the presence of MA functional groups causing esterification of flax fibres, and thereby increasing the surface energy of the fibres to a value much closer to the surface energy of the matrix. Thus, a better wettability and a higher interfacial adhesion was obtained.

Behzad et al.³¹ modified hemp fibres applying a paper sizing technique using a copolymer of styrene and dimethylaminopropylamine maleimide (SMA) as a surface-modifying agent in order to improve the performance of acrylic matrix composites. The estimated average stress to pull out the treated fibres was 71% higher than that measured for untreated fibres. This improvement was attributed to the presence of the nonpolar phenolic groups in the SMA component increasing the dispersive component of the fibre. Therefore, better spreading and wetting was expected, and, thus, the number of voids and bubbles at the interface, commonly present in all types of composites, would be reduced.

Lopattananon et al.⁵⁴ studied the influence of fibre modification on interfacial adhesion of pineapple leaf fibre–epoxy composites by single-fibre fragmentation tests. They verified that the interfacial shear strength of modified fibres was substantially higher than that of untreated ones. Sodium hydroxide and diglycidyl ether of bisphenol A (DGEBA), and the combination of both, were employed as reagents to modify the fibres. The adhesion between alkali-treated fibre and epoxy matrix increased by mechanical anchoring effects because alkalization made the fibre surface rough. The modification of fibres with epoxy solution resulted in grafting of the epoxy resin molecule at OH sites of the fibre. Therefore, the high interfacial shear strength of epoxy-treated composites might result from higher fibre surface affinity with the epoxy matrix. However, the strongest interfacial adhesion was obtained for the fibres that were treated by alkalization combined with DGEBA deposition. In this instance, the number of grafting sites at OH functional groups of the cellulose fibres increased as a result of both treatments, leading to more reactivity of the fibre with the matrix.

Interfacial evaluation of the various untreated and treated jute and hemp fibres reinforced PP–MAPP matrix composites was investigated by a micromechanical technique combined with acoustic emission (AE) and dynamic contact angle measurements by Park et al.⁵⁵ They modified the fibre surface with alkali and silane treatments and the PP matrix was modified by the addition of different amounts of MAPP. They observed that the IFSS of the natural fibre–MAPP/PP composites significantly increased with increasing content of MAPP in the mixture as well as after treating with alkaline solution and silane coupling agent. They suggested that chemical bridges could be created in the interfacial area between the fibre surface and the matrix for MAPP and silane coupling agent. Moreover, the advancing contact angle of silane-treated fibres was slightly lower than that of untreated fibres because the silane coupling agent blocked the high energy sites. Although the advancing contact angle of alkali treated fibres increased owing to the more hydrophilic fibre surface, the IFSS improvement was the result of weak boundary layers (wax, lignin and pectin) of natural fibres being removed leading to the increase of the surface area.

Morales et al.⁵⁶ used low-energy glow-discharge plasmas to functionalize cellulose fibres. PS films were synthesized by plasma (PPS) on the surface of cellulose fibres to obtain a good compatibility between the fibres and the PS matrix. The average interfacial shear strength was obtained from microdebonding tests for each plasma treatment applied to the fibres. Results showed that the adhesion in the fibre–matrix interface increased upon time for the first 4 min of treatment. However, at longer plasma exposures, fibres might be degraded, thus reducing the adhesion with the matrix. The greatest increase in the interfacial strength after plasma treatment was about 70%. They observed that the surface of the untreated cellulose fibres was smooth and uniform which would make the adhesion to the matrix difficult and produce a very weak interface. On the other hand, micrographs of fibres, exposed to 2 min of periodic glow discharges and 4 min of continuous glow discharges, respectively, showed that small fragments of PS adhered to the fibre at several points, which was an indication that the low-energy glow-discharge plasma modification could create different activation sites that promoted the formation of chemical bonds between the polymeric matrix and the fibres. Consequently fibre–matrix adhesion was improved.

Herrera-Franco et al.³⁶ studied the degree of fibre–matrix adhesion and its effect on the mechanical reinforcement of short henequen fibres and a polyethylene matrix. Henequen fibres were modified by various treatments, including an alkali treatment, a silane coupling agent and a pre-impregnation process with HDPE/xylene solution. The fibre–matrix interface shear strength calculated by both single-fragmentation and pull-out tests, was used as an indicator of the fibre–matrix adhesion improvement. They observed that all surface treatments improved the IFSS. All values obtained by fragmentation tests were higher than pull-out values ones. Although the IFSS values obtained by both methods were different, the lowest and highest values were obtained for the same systems, composites with untreated fibres and fibres treated with NaOH aqueous solution and then a silane coupling agent, respectively. The initial treatment with the aqueous solution of NaOH was thought to remove some lignin and hemicelluloses from the fibre surface, thus increasing the fibre surface area. Such a fibre surface increase resulted in a larger area of contact between the fibres and the matrix. Therefore, the hydroxyl groups on the cellulose fibres could better react with the silane-coupling agent because a larger number of possible reaction sites were available.

Khalil et al.⁴¹ studied the effect of acetylation on the interfacial shear strength of oil palm empty fruit bunch (EFB) and coconut fibre (coir) with various commercial matrices such as epoxy, unsaturated polyester and PS. Pull-out tests showed that composites based on acetylated fibres had a higher interfacial shear strength. They mentioned that IFSS values increased after acetylation because the hydrophobicity of natural fibres increased. In general, coir fibres exhibited higher IFSS values than EFB ones. This was attributed to the higher lignin content of coir.

Li et al.³⁵ determined the interfacial shear strength of untreated and white rot fungi-treated hemp fibre reinforced PP/maleated PP matrix. They observed that the IFSS value of white rot fungi-treated fibre composite was 40% higher than for the untreated fibre one. They suggested that the treatment could remove noncellulosic compounds, thus increasing the surface area and roughness. Therefore, the increase in surface area increased the potential for interaction between hydroxyl sites and the MAPP coupling agent. On the other hand, the higher roughness increased mechanical interlocking between fibres and matrix.

Li et al.⁵⁷ also investigated interfacial properties of untreated and treated sisal fibres reinforced high-density polyethylene composites. Four fibre surface treatments were applied to fibres in order to improve the interfacial bonding properties between sisal fibre and HDPE matrix. Two types of silanes, 3-aminopropyltriethoxysilane (silane 1) and γ-methacryloxypropyltrimethoxysilane (silane 2), were used as coupling agents to modify the surface of the sisal fibres. On the other hand, permanganate (KMnO₄) and dicumyl peroxide (DCP) were selected as oxidants to treat the fibre surface. The highest IFSS value was obtained after permanganate treatment, the IFSS improvement was 100% from 1.6 MPa (untreated fibre) to 3.2 MPa. The silane 2 also improved the IFSS to 2.9 MPa, whereas the IFSS for sisal fibre treated with silane 1 was similar to that for the untreated fibre composite. They suggested that permanganate could etch the sisal fibre surface and make it rougher, so that mechanical interlocking could be introduced between the sisal fibre and the HDPE matrix which was in accordance with the pull-out curve presented. Comparing the results obtained by silane-treated fibres, it was suggested that because silane 2 had a carbon main chain, it could form van der Waals bonds with the HDPE matrix, which has a similar chemical structure, thus increasing the IFSS. However, as silane 1 did not have a long carbon chain structure, no van der Waals bonding with the HDPE matrix could occur between the matrix and treated sisal fibres. Therefore, the IFSS was almost the same as that for the untreated composites.

Joseph et al.⁵⁸ compared the interfacial shear strength values obtained from single-fibre pull-out tests of composites fabricated using banana fibres and glass fibres with a phenol/formaldehyde (PF) matrix. The results obtained revealed that the adhesion between banana fibres and PF matrix was much higher than that between glass and PF matrix. They suggested that this was the result of the hydrophilic nature of banana fibres, caused by the hydroxyl groups of lignin and cellulose, which could easily form hydrogen bonds with methylol and phenolic hydroxyl groups of the resol resulting in a strong interlocking between banana fibres and resol matrix. However, in glass–PF composites the interfacial shear strength value was very small compared with that of banana fibre composites, suggesting that there was no interaction between fibre and matrix.

Van de Velde et al.⁷ studied the influence of surface treatments and the influence of matrix modification on the apparent interfacial shear strength of flax fibre reinforced polymer matrix. Flax fibre was treated with propyltrimethoxysilane, phenylisocyanate and MA-grafted polypropylene. The matrices used for this study were PP (two types having different molecular weights) and MAPP (two types differing in molecular weight and the content of grafted MA). The first, MAPP1, had a higher molecular weight than the second, MAPP2. However, MAPP1 had a lower grafted MA content than MAPP2 did. The single-fibre pull-out test was used for IFSS measurement, the results of which showed that treatment of flax fibre with propyltrimethoxysilane reduced fibre–matrix adhesion. On the other hand, small improvements were found for the phenylisocyanate-treated flax fibre PP interfacial strength. The presence of the alkyl chain MAPP on the coupling agent could possibly further increase the fibre–matrix adhesion. Comparing the values of systems obtained with different MAPP matrices, the results showed that although the MA content of MAPP1 was lower than for MAPP2, the apparent interfacial shear strength was almost double. It was suggested that an optimum MA content existed, for which the interfacial properties were optimum. Moreover, the higher molecular weight of MAPP1 compared with MAPP2 could also create better entanglements between fibre-bonded MAPP polymer chains and the matrix polymer chains resulting in the improvement in the interfacial shear strength. On the other hand, the difference in interfacial adhesion could also be attributed to the difference in crystallinity between the two MAPP used. Results obtained by differential scanning calorimetry showed that MAPP1 presented higher crystallinity than MAPP2. It was also suggested that crystallites could be formed around the fibre to act like cross-links by tying many molecules together, thus creating a possible positive effect on the interfacial adhesion.

Snijder et al.⁵⁹ studied the coupling efficiency of various MAPP grades with varying MA content grafted per polymeric PP chain and different molecular weights. It was concluded that the molecular weight was a more important factor than the MA content. Felix et al.⁶⁰ studied the mechanical properties of cellulose–PP composites. They used three different MA-based compatibilizing agents: alkylsuccinic anhydride, and two MAPP with the same MA content but different molecular weights. They found that the molecular weight of the compatibilizing agent was crucial for improving the mechanical properties of cellulose-PP composites. The mechanical properties increased with an increased molecular weight of MA-based compatibilizing agent. Similar results were obtained by Arbelaiz et al.⁶¹ for flax–PP composites modified with various MAPP. It was suggested that MAPP with higher molecular weight led to more entanglements between the MAPP chains and the PP matrix and, consequently, improved the mechanical properties of flax–PP composites.

The work done by Eichhorn et al.⁶² involved the modification of the interface between flax fibres and a variety of matrices. The interfacial shear strength was measured by pull-out and single fragmentation tests. As reinforcements, green flax (flax as received from the fields), dew retted (green flax after bacterial treatment on the fields) and a commercial fibre Duralin (flax treated by a method developed by CERES BV, Netherlands, which involves the depolymerization of lignin and hemicelluloses into lower molecular aldehyde and phenolic functionalities followed by a subsequent curing that hydrophobizes the fibre surface) were chosen. The following polymers were used as matrices: unsaturated polyester, epoxy, LDPE, HDPE, isotactic PP and MA-modified PP. Results showed that the greatest average interfacial shear strength was obtained with the Duralin–epoxy system, and the lowest was for green flax and LDPE composite.

On the other hand, more methods such as acetylation and stearic acid sizing have also been used to promote a better interface between dew retted flax and isotactic PP. The aim of both treatments was to treat the hydroxyl groups of the fibres with acid groups and, subsequently, to hydrophobize the fibre surface in order to obtain a better compatibility with PP. SFFT test results showed that acetylation slightly improved the interface for dew retted flax and caused a significant improvement for green flax. Stearic acid sizing caused an improvement in the interface only for low reaction times, whereas for longer reaction times there was a deterioration of the interface.

Joly et al.⁴² used microdebonding tests with ramie–PP composites. The IFSS results confirmed an improvement of 60% when MAPP treated fibres were used. They suggested that the improvement was the result of entanglements or interdiffusion of the PP chains grafted on the cellulosic fibres with the chains of the matrix. When fibres were treated with small alkyl chains (C₈–C₁₈), the IFSS value showed that the length of these alkyl chains had almost no effect on the IFSS. They suggested that the chain length was probably too short to create entanglements with matrix chains and also some degradation of the fibres owing to the swelling medium might mask a positive effect of the compatibilizing agent.

Sydenstricker et al.⁶³ evaluated treated and untreated sisal-reinforced polyester biocomposites. Sisal fibres were treated with NaOH and N-isopropyl acrylamide. The sisal–polyester interface was investigated through pull-out tests in order to find the optimal sisal chemical treatment conditions. Pull-out results for sisal fibre–polyester composites showed that all treatments improved the interfacial adhesion between fibres and matrix. The highest IFSS values were obtained for treatment with 2 wt% NaOH and 2 wt% acrylamide. Moreover, these concentrations also showed the highest fibre tensile strength.

Valadez-Gonzalez et al.⁵ explored the possibility of improving the effective mechanical properties of a HDPE matrix reinforced with henequen fibres by enhancing the fibre–matrix interface physicochemical interactions. Such interactions were characterized using two micromechanical methods, single-fibre fragmentation and pull-out tests. Three different treatments were applied to the henequen fibres: a treatment with an aqueous solution, a treatment first with alkaline solution and then with a silane coupling agent and the last treatment with alkali and then impregnated with a dilute matrix solution. The interfacial shear strength between natural fibres and thermoplastic matrices improved by fibre surface modification. They mentioned that the alkaline treatment increased the surface roughness resulting in a better mechanical interlocking and, moreover, for fibres treated after alkalization with a silane coupling agent and impregnated with a dilute matrix solution, NaOH treatment increased the amount of cellulose exposed on the fibre surface, thus increasing the number of possible reaction sites. The treatment with NaOH removed some lignin and hemicelluloses from the surface of the fibres and, consequently, the hydroxyl groups on the cellulose fibres were able to better interact with the silane-coupling agent because of the availability of a larger number of possible reaction sites. After treating henequen fibres with NaOH and combining silane surface treatment modification with preimpregnation process, the IFSS value was the highest owing to synergistic effect of both treatments.

Luo et al.²⁸ measured the IFSS of henequen fibre–poly(hydroxybutyrate-co-hydroxyvalerate) resin (PHBV) using single-fibre fragmentation and microdebonding tests. The IFSS of henequen fibre-PHBV measured using SFFT and microdebonding tests showed close agreement in spite of the different stresses experienced during each test. They attributed these findings mainly to the mechanical interlocking resulting from the fibre surface roughness, because no hydrogen bonding was possibile for PHBV. However, they suggested that fibre–matrix mechanical interlocking could be limited because of the high viscosity of the resin, which limited its ability to penetrate into and around the cells.

Tripathy et al.¹⁸ studied the interfacial adhesion between four different forms of jute fibres (sliver, bleached, mercerized and untreated) and polyolefinic matrices (LDPE and PP). The fibre–matrix adhesion was estimated by means of single-fibre fragmentation tests. Results indicated a low adhesion between fibres and polyolefins, that confirmed that nonpolar matrices can not develop good adhesion with the polar fibre surfaces. With both thermoplastic matrices, sliver jute fibres and untreated jute fibres showed the highest values of stress transfer ability, which was attributed to the fact that sliver jute and untreated jute fibres have a high lignin content whose aromatic nature might be responsible for somewhat better wettability and interfacial interaction with polyolefins as a result of the higher dispersive forces. On the other hand, the bleached jute fibres and mercerized jute fibres showed the lowest interfacial interaction with polymer matrix. The bleached jute fibres and mercerized jute fibres were chemically treated with acid and acid + base solutions, respectively. As a considerable amount of lignin was removed, the concentration of polar groups at the surface probably increased, and therefore the adhesion to nonpolar polymer was consequently reduced.

Tripathy et al.²⁷ also investigated interfacial strength of untreated jute fibres and fibres treated with three different treatments with an epoxy resin as matrix. They carried out single-fibre composite tests in order to determine the critical fragment length and interfacial strength. The bleached jute fibres showed the highest IFSS value of about 140 MPa. They suggested that the high adhesion for bleached jute fibres might be caused by a greater mechanical anchorage of the epoxy resin on the more regular, clean surface with high microporosity. The possibility of chemisorption of the epoxy resin with hydroxyl groups of the cellulose in the absence of lignin might be another reason for improved adhesion.

Cho et al.⁶⁴ demonstrated that the intensity of electron beam (e-beam) irradiated on henequen fibres strongly influenced the interfacial adhesion between the fibres and a PP matrix. The interfacial shear strength was measured by single-fibre microdebonding tests. Henequen fibre surfaces modified with an appropriate dosage of e-beam, a low e-beam intensity of 10 kGy, resulted in the improvement of the interfacial properties. However, the e-beam irradiation at intensities higher than 10 kGy led to the reduction of the properties investigated. It was suggested that at 200 kGy the interfacial and mechanical properties were increased to the level of the composite with the henequen fibres treated at 10 kGy but the fibre surfaces were likely to be damaged.

Cho et al.⁶⁵ studied natural fibres (jute, kenaf and henequen) reinforced thermoplastic (PLA and PP) and thermosetting (unsaturated polyester) matrix composites. They observed that the interfacial shear strength of untreated kenaf–jute, henequen–PP and henequen–unsaturated polyester green composites was significantly improved by water treatment, particularly with a dynamic ultrasonication method. They suggested that the water treatment removed not only the surface impurities but also some waxy components present on the fibre surfaces. As a result, the treated fibres had rougher surfaces with more crevices than the untreated ones. They concluded that water treatment of natural fibres may be practically favourable with many advantages for improving the properties of composites.

Thamae et al.⁶⁶ studied the influence of the fibre extraction method, alkali and silane treatment on the interface of Agave americana waste–HDPE composites using pull-out tests. They observed that mercerization of agave americana fibres could improve the interfacial shear strength value by about 104% compared with untreated fibre ones. They stated that mercerization removed impurities, exposed hydrophobic lignin, and increased fibre surface area by exposing some ultimate fibres. However, the IFSS measurements reported in this work were very low, which they attributed to the lack of compatibility between natural fibres and HDPE. Silane treatment did not have any influence on the interfacial shear strength between agave americana fibres and HDPE.

Huber et al.⁶⁷ studied natural fibres (cotton, flax and hemp) polymeric matrix adhesion using single-fibre fragmentation tests. As polymeric matrices PP with 2% MA and PLA were used. Although SFFT tests have to be performed with a single fibre, in this study flax and hemp fibre bundles were used. The IFSS measurement for flax–MAPP was about 12 MPa and for hemp–MAPP was higher but the value was not given. Finally, the IFSS for the cotton–MAPP system was very low, about 0.7 MPa, which was attributed to be a consequence of the difference in chemical composition of the cotton fibres. Taking into account that cotton is a fibre made up of only cellulose and hemicellulose compared with the bast fibres hemp and flax, which contain lignin and pectin, these results indicated that fibre composition influences the fibre–matrix adhesion. The SFFT results reported for flax–PLA composites showed a poor adhesion between fibre and matrix because no fragments of flax fibre in the PLA matrix could be generated during the tests. The adhesion between fibre and PLA was not strong enough to transmit forces from the matrix to the fibre. This result was in accordance with SEM micrographs presented.

Joffe et al.⁸ studied flax fibres with vinylester (VE), unsaturated polyester (UP) and epoxy matrices adhesion by SFFT. Fibres were treated with acrylic acid and vinyltrimethoxysilane. They observed that the IFSS of epoxy and VE matrices was somewhat higher than that for UP. The surface treatments used did not lead to significant variation in the IFSS. Hence, they concluded that the adhesion of flax fibres and thermoset matrices did not benefit from the common surface treatments.

Lee et al.⁶⁸ studied the effect of fibre surface treatments on the interfacial properties of henequen–PP composites using single-fibre microdebonding tests. The surfaces of henequen fibres were treated with two different media, tap water and sodium hydroxide, that underwent both soaking and ultrasonic methods. The soaking and ultrasonic treatments with tap water and sodium hydroxide at different concentrations and treatment times significantly influenced the interfacial properties. The greatest improvement in the interfacial shear strength was achieved by ultrasonic alkalization performed with 6 wt% NaOH for 60 min. The IFSS improved by about 138% compared with that of the untreated fibre. The water treatment removed not only the surface impurities but also some waxy components present on the fibre surfaces. As a result, the treated fibres had rougher surfaces with more crevices than the untreated ones.

Liu et al.³³ characterized the interface between cellulosic fibres and PS matrix by microdebonding tests. They found that no consistent relationship existed between IFSS results and the embedded length of the microdrop in the range examined. Moreover, the results showed that the acetylated wood fibres had the highest shear strength. Acetylation also improved the IFSS for cotton, although to a lesser degree. No differences were found in the IFSS with fibre treatments for rayon. The high IFSS between the acetylated wood fibre and the PS matrix was attributed to the improved wetting and spreading of the molten PS on the acetylated wood fibre surfaces.

Lodha et al.³⁰ fabricated composites using ramie fibres and soy protein isolate (SPI). They characterized the interfacial shear strength using microdebonding tests. The average IFSS value was 29.8 MPa. Although the SPI-polymer resin contained several amino acids with polar groups, it also contained amino acids, such as alanine, glycine, leucine and isoleucine, which had nonpolar groups and, therefore, these amino acids could not be hydrogen bonded with cellulose molecules in the fibres.

Czigany et al.⁶⁹ examined the interfacial adhesion of three different natural reinforcing fibres and three different commercial matrix materials using microdebonding tests. The natural fibres used were flax, hemp and sisal. As polymeric matrices, PP, biodegradable MaterBi (a thermoplastic biodegradable polymer made of starch) and PuraSorb (a biodegradable lactide–glycolide copolymer) were used. Based on microdebonding tests, it was revealed that among the examined specimens the strongest adhesion existed in the PuraSorb–sisal combination. Purasorb matrix is highly polar owing to the oxygen atoms present in the polymer chain. They suggested that very strong hydrogen bridges were formed with the –OH groups on the surface of the fibres, resulting in superior fibre–matrix interfacial adhesion.

Zafeiropoulos⁷⁰ studied flax fibre-PP composites and a sensitivity analysis was presented for single-fibre fragmentation tests. The results indicated a large standard deviation for the IFSS measurements, which was partly attributed to the significant standard deviation of the fibre strength as well to the significant standard deviation of the fibre diameter. Traditional stress analysis failed to correctly assess the interface, whilst a statistically based data analysis could overcome the fibre heterogeneity problem. He suggested that it was not possible to assess the effect of the fibre surface treatments at the interface in flax fibre–PP composites.

Torres et al.⁷¹ studied the interfacial properties of sisal-reinforced polyethylene composites by means of single-fibre fragmentation tests. Fibre treatment with stearic acid increased the interfacial shear strength by 23% with respect to untreated fibres. The improvements in IFSS found for the treated specimens were consistent with observations from SEM micrographs. They suggested that the variability of the results was relatively high as indicated by a high standard deviation (21%), which might be attributed to experimental errors (opacity of the matrix), as well as the inherent characteristics of the fibres and the fibre–matrix interface.

Wazzan⁷² modified date palm (Phoenix dactylifera L.) fibre surface using different treatments in order to improve their adhesion to polyester matrices. Three fibre surface treatments were used, an aqueous alkaline solution, silane coupling agent, and a combination of both treatments. He investigated the improvement in adhesion through single-fibre pull-out tests. IFFS values increased for all treated fibres compared with nontreated fibres. In particular, the combination of alkaline and silane coupling agents resulted in the best adhesion improvement to the polyester matrix. The alkali treatment gave up to 18% increase in the shear strength. Wazzan mentioned that this improvement was because of the removal of pectins. A higher shear strength was obtained when using the fibres treated with the silane coupling agent. However, the best effect was reached with a combination of both treatments, which increased the average value of fibre–matrix shear strength by more than 40% with respect to the composite made with untreated fibres. The IFSS varied with the fibre embedded for all date palm fibre–polyester systems. The apparent IFSS increased as the embedded length decreased.

Park et al.⁷³ evaluated the IFSS values of various ramie and kenaf fibre–epoxy composites using the combination of microdebonding test and nondestructive acoustic emission. They observed that the IFSS was higher for ramie fibre than for kenaf fibres. Kenaf fibres contain a higher proportion of hemicelluloses, lignin, and waxes than ramie fibres and these compounds are responsible for poor adhesion between the fibre surface and epoxy matrix because of their different inherent properties.

Park et al.⁵⁵ studied jute and hemp fibres and noticed that the IFSS of a given natural fibre–matrix system increased when the critical embedded area decreased. The critical embedded area was used instead of the conventional critical embedded length because of the different diameter for both fibres. The critical embedded area was obtained by the intersection of two linear regression lines: one is the fibre pull-out linear regression line, whereas the other is the fibre fracture linear regression line. From the experimental plots, they observed that the critical embedded area decreased with an increase in MAPP content in the PP–MAPP matrix material or after treating the natural fibres with alkaline solution and silane coupling agent.

5.3.3 Effect of processing parameters

The presence of moisture has a negative effect on the performance of lignocellulosic fibre based composites.⁷⁴ Water molecules absorbed in natural fibres can reduce the intermolecular hydrogen bonding between cellulose molecules in the fibre and establish intermolecular hydrogen bonding between cellulose molecules in the fibre and water molecules, thereby reducing the interfacial adhesion between the fibre and the matrix and, as a result, decreasing the mechanical properties of the composites.⁶¹

Chen et al.⁷⁵ fabricated bamboo strips–vinyl ester resin composites to investigate the influence of moisture on interfacial shear strength by pull-out tests. The objective was to clarify how prefabrication exposure to moisture affects the matrix hardening process and to establish the relative importance between prefabrication moisture exposure and post-fabrication moisture exposure. The relative humidity (RH) in composite manufacture had a severe impact on the IFSS of the resulting composites. The IFSS achieved at normal room conditions (20 °C, 60% RH) was only a half of what was achieved in the dry condition. Composites produced at high RH (80 and 90%) had a negligible interfacial strength. They suggested that the high moisture content of henequen fibres would evaporate during processing, resulting in micro-voids around the fibres, which would thus reduce the interface area and decrease the IFSS and, consequently, the composite properties.^75,76 The IFSS decreased almost linearly with the increase in bamboo moisture content from 0 to about 12%. The rate of decrease suddenly accelerated after this point. A further increase of 2% in bamboo moisture content caused a reduction of more than 80% in the IFSS. Reported optical photographs showed that the specimen produced at dry condition showed no visible voids at the interface between bamboo fibres and the matrix whereas the sample produced at 80% RH exhibited thin void lines along the fibrous region and wide void spaces along the ground matrix regions, which was in accordance with the IFSS.

Several studies⁷⁷⁻⁷⁹ reported that the presence of lignocellulosic fibres could lead to the development of transcrystallinity (TC). Some studies ^32,60,78,79 were conducted with natural fibre–polymer composites to observe the effect of a transcrystalline region on composite interfacial properties. The results reported are fairly contradictory regarding the effect of a transcrystalline interphase on the composite interfacial properties. Zafeiropoulos et al.⁷⁸ used four different types of flax fibres, green flax, dew-retted flax, Duralin flax, and stearic acid-treated flax with two different iPP matrices. The effect of various processing conditions such as the crystallization temperature, time and cooling rates on TC was investigated and the effect of the TC layer upon the mechanical properties of the interface was studied using single-fibre fragmentation tests. They concluded that the presence of TC in the flax–PP system had a profound effect on the interface characteristics. The interface was strongly enhanced, as shown by fragmentation tests. However, there was no difference for different thickness or crystallization temperature and the cooling rate from the melt to the crystallization temperature did not affect the morphology of the transcrystalline layer.

Eichhorn et al.⁶² investigated the effect of cooling rates upon the interface of flax-isotactic PP composites by single-fibre fragmentation tests. Two different flax fibres were used: dew retted flax and green flax. The results showed that there was a significant improvement in the IFSS when TC was present, as shown by the sharp decrease of the critical length. They concluded that TC improved the interfacial stress transfer and the thickness of the transcrystalline layer did not affect the interface. Felix and Gatenholm⁷⁹ employed a single-fibre fragmentation test to evaluate the effect of a transcrystalline interphase morphology on the shear stress transfer in cotton fibre–PP system. They found that the presence of a trancrystalline interphase improved the shear transfer up to ~ 100% depending on the thickness of the transcrystalline layer. They suggested that slow cooling favours the kinetics of the approach of PP molecules and, hence, interfacial adsorption, which yields an ordered transcrystalline PP interphase having a high density of secondary bonds with the cellulose surface. They concluded that a careful control of thermal conditions during processing may be sufficient for obtaining satisfactory interfacial and thus composite properties.

However, Garkhail et al.⁸⁰ studied the effect of transcrystallization on the interfacial shear strength in PP–flax fibre composites by fibre pull-out tests. They found that the IFSS of the PP–flax system was slightly decreased in the presence of a transcrystalline interphase.

George et al.³² studied the effect of TC on the interfacial shear strength of flax fibre-reinforced polypropylene by pull-out tests. They observed that with increasing transcrystallinity thickness the value of IFSS decreased. However, the differences observed were very small.

Finally, some mention should be made of the effect of the type of matrix chosen for composite fabrication. Khalil et al.⁴¹ observed that thermosetting polymer showed a higher IFSS than thermoplastic matrices. They suggested that the thermosetting systems had a higher wettability than thermoplastic ones owing to the fact that thermosetting plastics are cured ‘in situ’ from relatively low molecular weight compounds while mixing with fibres, followed by the polymerization reaction. On the other hand, thermoplastics are mixed in the molten state where the wettability is limited. Another reason for the higher IFSS of thermosets compared with thermoplastics could be the chemical components of natural fibres that could be more reactive or compatible with thermoset matrices.

1. Netravali, A.N., Henstenburg, R.B., Phoenix, S.L., Schwartz, P. Interfacial shear strength studies using the single-filament-composite test. I: Experiments on graphite fibres in epoxy. Polymer Composites. 1989; 10(4):226–241.

2. Narkis, M., Chen, E.J.H. Review of methods for characterization of interfacial fibre–matrix interactions. Polymer Composites. 1988; 9(4):245–251.

3. Wimolkiatisak, A.S., Bell, J.P. Interfacial shear strength and failure modes of interphase-modified graphite-epoxy composites. Polymer Composites. 1989; 10(3):162–172.

4. Arbelaiz, A., Fernandez, B., Cantero, G., Llano-Ponte, R., Valea, A., Mondragon, I. Mechanical properties of flax fibre/polypropylene composites. Influence of fibre/matrix modification and glass fibre hybridization. Composites, Part A: Applied Science and Manufacturing. 2005; 36A(12):1637–1644.

5. Valadez-Gonzalez, A., Cervantes-Uc, J.M., Olayo, R., Herrera-Franco, P.J. Effect of fibre surface treatment on the fibre–matrix bond strength of natural fibre reinforced composites. Composites, Part B: Engineering. 1999; 30B(3):309–320.

6. Gauthier, R., Joly, C., Coupas, A.C., Gauthier, H., Escoubes, M. Interfaces in polyolefin/cellulosic fibre composites: chemical coupling, morphology, correlation with adhesion and aging in moisture. Polymer Composites. 1998; 19(3):287–300.

7. Van de Velde, K., Kiekens, P. Influence of fibre surface characteristics on the flax/polypropylene interface. Journal of Thermoplastic Composite Materials. 2001; 14(3):244–260.

8. Joffe, R., Andersons, J., Wallstroem, L. Interfacial shear strength of flax fiber/thermoset polymers estimated by fibre fragmentation tests. Journal of Materials Science. 2005; 40(9/10):2721–2722.

9. Rao, V., Drzal, L.T. The dependence of interfacial shear strength on matrix and interphase properties. Polymer Composites. 1991; 12(1):48–56.

10. Felix, J., Enhancing interactions between cellulose fibres and synthetic polymers PhD thesis. Chalmers University of Technology, Goteborg, 1993.

11. Scarponi, C., Pizzinelli, C.S. Interface and mechanical properties of natural fibres reinforced composites: a review. International Journal of Materials & Product Technology. 2009; 36(1–4):278–303.

12. Pisanova, E.V., Zhandarov, S.F., Dovgyalo, V.A. Interfacial adhesion and failure modes in single filament thermoplastic composites. Polymer Composites. 1994; 15(2):147–155.

13. Lee, S.M., International encyclopedia of composites Volume 1. VCH publishers, New York, 1989.

14. Kelly, A., Tyson, W.R. Tensile properties of fibre-reinforced metals. Copper/tungsten and copper/molybdenum. Journal of the Mechanics and Physics of Solids. 1965; 13(6):329–350.

15. Bian, X.S., Kenny, J.M., Ambrosio, L., Nicolais, L. Development and application of the single fibre technique (SFT) for the evaluation and analysis of fibre/matrix adhesion in polymeric composites. Chimica Oggi. 1991; 9(6):44–49.

16. Netravali, A.N., Li, Z.F., Sachse, W., Wu, H.F. Determination of fibre matrix interfacial shear strength by an acoustic emission technique. Journal of Materials Science. 1991; 26(24):6631–6638.

17. Wu, H.F., Biresaw, G., Laemmle, J.T. Effect of surfactant treatments on interfacial adhesion in single graphite/epoxy composites. Polymer Composites. 1991; 12(4):281–288.

18. Tripathy, S.S., Levita, G., Di Landro, L. Interfacial adhesion in jute polyolefin composites. Polymer Composites. 2001; 22(6):815–822.

19. Li, Z.F., Grubb, D.T., Phoenix, S.L. Fibre interactions in the multi-fibre composite fragmentation test. Composites Science and Technology. 1995; 54(3):251–266.

20. Ohsawa, T., Nakayama, A., Miwa, M., Hasegawa, A. Temperature dependence of critical fibre length for glass fibre-reinforced thermosetting resins. Journal of Applied Polymer Science. 1978; 22(11):3203–3212.

21. Drzal, L.T. Adhesion of graphite fibres to epoxy matrices. III. The effect of hygrothermal exposure. Journal of Adhesion. 1985; 18(1):49–72.

22. Favre, J.P., Jacques, D. Stress transfer by shear in carbon fibre model composites. Part 1. Results of single-fibre fragmentation tests with thermosetting resins. Journal of Materials Science. 1990; 25(2B):1373–1380.

23. Goda, K., Park, J.M., Netravali, A.N. A new theory to obtain Weibull fibre strength parameters from a single-fibre composite test. Journal of Materials Science. 1995; 30(10):2722–2728.

24. Zhao, F.M., Okabe, T., Takeda, N. The estimation of statistical fibre strength by fragmentation tests of single-fibre composites. Composites Science and Technology. 2000; 60(10):1965–1974.

25. Weibull, W. A statistical theory of the strength of materials. Handlingar Stocklholm: Ingenioersvetenskapsakademien; 1939.

26. Arbelaiz, A., Cantero, G., Fernandez, B., Ganan, P., Kenny, J.M., Mondragon, I. Flax fibre surface modifications: Effects on fibre physico mechanical and flax/polypropylene interface properties. Polymer Composites. 2005; 26(3):324–332.

27. Tripathy, S.S., Di Landro, L., Fontanelli, D., Marchetti, A., Levita, G. Mechanical properties of jute fibres and interface strength with an epoxy resin. Journal of Applied Polymer Science. 2000; 75(13):1585–1596.

28. Luo, S., Netravali, A.N. Characterization of henequen fibres and the henequen fibre/poly(hydroxybutyrate-co-hydroxyvalerate) interface. Journal of Adhesion Science and Technology. 2001; 15(4):423–437.

29. Prasad, B.M., Sain, M.M. Mechanical properties of thermally treated hemp fibres in inert atmosphere for potential composite reinforcement. Materials Research Innovations. 2003; 7(4):231–238.

30. Lodha, P., Netravali, A.N. Characterization of interfacial and mechanical properties of green composites with soy protein isolate and ramie fibre. Journal of Materials Science. 2002; 37(17):3657–3665.

31. Behzad, T., Sain, M. Surface and interface characterization of untreated and SMA imide-treated hemp fibre/acrylic composites. Polymer Composites. 2009; 30(6):681–690.

32. George, J., Garkhail, S.K., Wieland, B., Peijs, T., A study on the transcrystallisation behaviour of flax fibre reinforced polypropylene composites and effect on mechanical properties Sao Pedro, Brazil. Proceedings from the third international symposium on natural polymers and composites, and the workshop on progress in production and processing of cellulosic fibres and natural polymers. 2000:408–413.

33. Liu, F.P., Wolcott, M.P., Gardner, D.J., Rials, T.G. Characterization of the interface between cellulosic fibres and a thermoplastic matrix. Composite Interfaces. 1994; 2(6):419–432.

34. Stamboulis, A., Baillie, C., Schulz, E. Interfacial characterisation of flax fibre-thermoplastic polymer composites by the pull-out test. Angewandte Makromolekulare Chemie. 1999; 272:117–120.

35. Li, Y., Pickering, K.L., Farrell, R.L. Determination of interfacial shear strength of white rot fungi treated hemp fibre reinforced polypropylene. Composites Science and Technology. 2009; 69(7–8):1165–1171.

36. Herrera-Franco, P.J., Valadez-Gonzalez, A. A study of the mechanical properties of short natural-fibre reinforced composites. Composites, Part B: Engineering. 2005; 36B(8):597–608.

37. Pisanova, E., Zhandarov, S., Mader, E. How can adhesion be determined from micromechanical tests? Composites, Part A: Applied Science and Manufacturing. 2001; 32A(3–4):425–434.

38. Zhang, C., Qiu, Y. Modified shear lag model for fibres and fillers with irregular cross-sectional shapes. Journal of Adhesion Science and Technology. 2003; 17(3):397–408.

39. Pitkethly, M.J., Favre, J.P., Gaur, U., Jakubowski, J., Mudrich, S.F., Caldwell, D.L., Drzal, L.T., Nardin, M., Wagner, H.D., Di Landro, L., Hampe, A., Armistead, J.P., Desaeger, M., Verpoest, I. A round-robin program on interfacial test methods. Composites Science and Technology. 1993; 48(1–4):205–214.

40. Miller, B., Muri, P., Rebenfeld, L. A microbond method for determination of the shear strength of a fibre/resin interface. Composites Science and Technology. 1987; 28(1):17–32.

41. Khalil, H.P.S.A., Ismail, H., Rozman, H.D., Ahmad, M.N. The effect of acetylation on interfacial shear strength between plant fibres and various matrices. European Polymer Journal. 2001; 37(5):1037–1045.

42. Joly, C., Gauthier, R., Chabert, B. Physical chemistry of the interface in polypropylene/cellulosic-fibre composites. Composites Science and Technology. 1996; 56(7):761–765.

43. George, J., Sreekala, M.S., Thomas, S. A review on interface modification and characterization of natural fibre reinforced plastic composites. Polymer Engineering and Science. 2001; 41(9):1471–1485.

44. Garcia-Hernandez, E., Licea-Claverie, A., Zizumbo, A., Alvarez-Castillo, A., Herrera-Franco, P.J. Improvement of the interfacial compatibility between sugar cane bagasse fibres and polystyrene for composites. Polymer Composites. 2004; 25(2):134–145.

45. Barkakaty, B.C. Some structural aspects of sisal fibres. Journal of Applied Polymer Science. 1976; 20(11):2921–2940.

46. Mukherjee, P.S., Satyanarayana, K.G. Structure and properties of some vegetable fibres. Part 1. Sisal fibre. Journal of Materials Science. 1984; 19(12):3925–3934.

47. Joly, C., Kofman, M., Gauthier, R. Polypropylene/cellulosic fibre composites: chemical treatment of the cellulose assuming compatibilization between the two materials. Journal of Macromolecular Science Pure and Applied Chemistry. 1996; A33(12):1981–1996.

48. Felix, J.M., Gatenholm, P. The nature of adhesion in composites of modified cellulose fibres and polypropylene. Journal of Applied Polymer Science. 1991; 42(3):609–620.

49. Bledzki, A.K., Gassan, J. Composites reinforced with cellulose based fibres. Progress in Polymer Science. 1999; 24(2):221–274.

50. Mwaikambo, L.Y., Ansell, M.P. Chemical modification of hemp, sisal, jute, and kapok fibres by alkalization. Journal of Applied Polymer Science. 2002; 84(12):2222–2234.

51. Karlsson, J.O., Blachot, J.F., Peguy, A., Gatenholm, P. Improvement of adhesion between polyethylene and regenerated cellulose fibres by surface fibrillation. Polymer Composites. 1996; 17(2):300–304.

52. Adusumali, R.B., Reifferscheid, M., Weber, H., Roeder, T., Sixta, H., Gindl, W. Mechanical properties of regenerated cellulose fibres for composites. Macromolecular Symposia. 2006; 244:119–125.

53. Lopez Manchado, M.A., Arroyo, M., Biagiotti, J., Kenny, J.M. Enhancement of mechanical properties and interfacial adhesion of PP/EPDM/flax fibre composites using maleic anhydride as a compatibilizer. Journal of Applied Polymer Science. 2003; 90(8):2170–2178.

54. Lopattananon, N., Payae, Y., Seadan, M. Influence of fibre modification on interfacial adhesion and mechanical properties of pineapple leaf fiber-epoxy composites. Journal of Applied Polymer Science. 2008; 110(1):433–443.

55. Park, J.M., Quang, S.T., Hwang, B.S., DeVries, K.L. Interfacial evaluation of modified jute and hemp fibers/polypropylene (PP)-maleic anhydride polypropylene copolymers (PP-MAPP) composites using micromechanical technique and nondestructive acoustic emission. Composites Science and Technology. 2006; 66(15):2686–2699.

56. Morales, J., Olayo, M.G., Cruz, G.J., Herrera-Franco, P., Olayo, R. Plasma modification of cellulose fibers for composite materials. Journal of Applied Polymer Science. 2006; 101(6):3821–3828.

57. Li, Y., Hu, C., Yu, Y. Interfacial studies of sisal fiber reinforced high density polyethylene (HDPE) composites. Composites, Part A: Applied Science and Manufacturing. 2008; 39A(4):570–578.

58. Joseph, S., Sreekala, M.S., Oommen, Z., Koshy, P., Thomas, S. A comparison of the mechanical properties of phenol-formaldehyde composites reinforced with banana fibres and glass fibres. Composites Science and Technology. 2002; 62(14):1857–1868.

59. Snijder, M.H.B., Bos, H.L. Reinforcement of polypropylene by annual plant fibers: optimisation of the coupling agent efficiency. Composite Interfaces. 2000; 7(2):69–79.

60. Felix, J.M., Gatenholm, P. The effect of compatibilizing agents on the interfacial strength in cellulose-polypropylene composites. Polymeric Materials Science and Engineering. 1991; 64(3):123–124.

61. Arbelaiz, A., Fernandez, B., Ramos, J.A., Retegi, A., Llano-Ponte, R., Mondragon, I. Mechanical properties of short flax fibre bundle/polypropylene composites: influence of matrix/fibre modification, fibre content, water uptake and recycling. Composites Science and Technology. 2005; 65(10):1582–1592.

62. Eichhorn, S.J., Baillie, C.A., Zafeiropoulos, N., Mwaikambo, L.Y., Ansell, M.P., Dufresne, A., Entwistle, K.M., Herrera-Franco, P.J., Escamilla, G.C., Groom, L., Hughes, M., Hill, C., Rials, T.G., Wild, P.M. Current international research into cellulosic fibers and composites. Journal of Materials Science. 2001; 36(9):2107–2131.

63. Sydenstricker, T.H.D., Mochnaz, S., Amico, C. Pull-out and other evaluations in sisal-reinforced polyester biocomposites. Polymer Testing. 2003; 22(4):375–380.

64. Cho, D., Lee, H.S., Han, S.O., Drzal, L.T. Effects of e-beam treatment on the interfacial and mechanical properties of henequen/polypropylene composites. Advanced Composite Materials. 2007; 16(4):315–334.

65. Cho, D., Seo, J.M., Lee, H.S., Cho, C.W., Han, S.O., Park, W.H. Property improvement of natural fiber-reinforced green composites by water treatment. Advanced Composite Materials. 2007; 16(4):299–314.

66. Thamae, T., Baillie, C. Influence of fibre extraction method, alkali and silane treatment on the interface of Agave americana waste HDPE composites as possible roof ceilings in Lesotho. Composite Interfaces. 2007; 14(7–9):821–836.

67. Huber, T., Muessig, J. Fibre matrix adhesion of natural fibres cotton, flax and hemp in polymeric matrices analyzed with the single fibre fragmentation test. Composite Interfaces. 2008; 15(2–3):335–349.

68. Lee, H.S., Cho, D., Han, S.O. Effect of natural fiber surface treatments on the interfacial and mechanical properties of henequen/polypropylene biocomposites. Macromolecular Research. 2008; 16(5):411–417.

69. Czigany, T., Morlin, B., Mezey, Z. Interfacial adhesion in fully and partially biodegradable polymer composites examined with microdroplet test and acoustic emission. Composite Interfaces. 2007; 14(7–9):869–878.

70. Zafeiropoulos, N.E. On the use of single fiber composites testing to characterize the interface in natural fiber composites. Composite Interfaces. 2007; 14(7–9):807–820.

71. Torres, F.G., Cubillas, M.L. Study of the interfacial properties of natural fibre reinforced polyethylene. Polymer Testing. 2005; 24(6):694–698.

72. Wazzan, A. The effect of surface treatment on the strength and adhesion characteristics of Phoenix dactylifera-L (date palm) fibers. International Journal of Polymeric Materials. 2006; 55(7):485–499.

73. Park, J.M., Son, T.Q., Jung, J.G., Hwang, B.S. Interfacial evaluation of single ramie and kenaf fiber/epoxy resin composites using micromechanical test and nondestructive acoustic emission. Composite Interfaces. 2006; 13(2–3):105–129.

74. Joseph, P.V., Rabello, M.S., Mattoso, L.H.C., Joseph, K., Thomas, S. Environmental effects on the degradation behaviour of sisal fibre reinforced polypropylene composites. Composites Science and Technology. 2002; 62(10–11):1357–1372.

75. Chen, H., Miao, M., Ding, X. Influence of moisture absorption on the interfacial strength of bamboo/vinyl ester composites. Composites, Part A: Applied Science and Manufacturing. 2009; 40(12):2013–2019.

76. Stark, N. Influence of moisture absorption on mechanical properties of wood flour-polypropylene composites. Journal of Thermoplastic Composite Materials. 2001; 14(5):421–432.

77. Arbelaiz, A., Fernandez, B., Ramos, J.A., Mondragon, I. Thermal and crystallization studies of short flax fiber reinforced polypropylene matrix composites: effect of treatments. Thermochimica Acta. 2006; 440(2):111–121.

78. Zafeiropoulos, N.E., Baillie, C.A., Matthews, F.L. A study of transcrystallinity and its effect on the interface in flax fibre reinforced composite materials. Composites, Part A: Applied Science and Manufacturing. 2001; 32A(3–4):525–543.

79. Felix, J.M., Gatenholm, P. Effect of transcrystalline morphology on interfacial adhesion in cellulose/polypropylene composites. Journal of Materials Science. 1994; 29(11):3043–3049.

80. Garkhail, S., Wieland, B., George, J., Soykeabkaew, N., Peijs, T. Transcrystallisation in PP/flax composites and its effect on interfacial and mechanical properties. Journal of Materials Science. 2009; 44(2):510–519.