Fracture behaviour of epoxy biocomposite reinforced with short coconut fibres ( Cocos nucifera ) and walnut particles ( Juglans regia L.)

Abstract

Mixed-mode fracture characteristics of epoxy-based biocomposite reinforced with 20 wt% walnut shell particle and 10 wt% coconut fibres are investigated. The biocomposite is fabricated using the squeeze casting method. The positive aspect of hybrid combination of fibre and particle reinforcement is advocated by comparing mode I, mode II and mixed-mode I/II fracture surfaces under a scanning electron microscope. An edge-cracked semicircular arc specimen subjected to symmetric three-point bend (TPB) loading is suggested for fracture toughness testing of biocomposite material. A series of fracture tests are conducted on hybrid biocomposite using the proposed semicircular bend (SCAB) specimen geometry, TPB and four-point bend (FPB) specimens. The average mode I and II fracture toughness obtained from semicircular arc bend (SCAB) specimen are 1.319 MPa and 1.219 MPa $\sqrt{m}$ , respectively. The average values of modes I and II toughness obtained from TPB and FPB specimens are 1.267 and 0.754 MPa $\sqrt{m}$ . The overall ratio of mode I and mode II fracture toughness is found to be 0.762, which is very close to the predicted value of 0.87 by maximum tangential stress criterion. The results obtained from new SCAB geometry are validated with the results obtained from TPB and FPB specimen geometries by statistical significance test. Very good agreement is found between the experimental results obtained from TPB, FPB and the proposed SCAB specimens.

Keywords

Biocomposites fracture toughness scanning electron microscopy biofibre bioparticles

Introduction

In recent decades, natural fibre-reinforced composites are getting much attention in structural application, automobile industries and household applications. Various works on the application of natural fillers and fibres like walnut shell, almond shell, pineapple, sisal, coconut shell and coir, jute, palm, cotton, rice husk, bamboo, wood, and so on as the reinforcements in composites have been reported in the literature.^1,2 A significant number of published studies have been dedicated to mechanical and thermal properties.³ However, because a flaw-free material is extremely difficult to be produced and cracks may be introduced during service, understanding the crack resistance ability is thus essential. Good toughness and crack-stopping capability are particularly important. It has been mentioned that toughness of a brittle thermosetting polymer such as polyester and epoxy can be improved through natural fibre reinforcement.⁴ However, fracture toughness studies of biocomposites have received relatively little attention from the scientific community until now. Due to increasing demand of biocomposites in various applications, it is envisaged that the fracture toughness of biocomposites will play a vital role over the coming years. It is known that biofibres- or bioparticles-reinforced epoxy-based biocomposites are prone to brittle fracture under mechanical loading. Compared with metals, the application of fracture mechanics concepts to polymers and composites is still in the primitive stage. In the polymer field, in spite of some controversies, the fracture toughness tests meant for homogeneous materials are used to determine K_IC and J-integral.^5,6 In these studies, the employed procedures are similar to those applied to metals, as described by ASTM D5045 standards.⁷

Among the published works on biocomposites, only few researchers^{8

–19} have studied the fracture properties such as toughness or critical stress intensity factors of biocomposites. Zamanian et al.¹⁶ have investigated the toughness of epoxy polymers modified with a range of different nanosilica particles using three point bend (TPB) specimen and concluded that addition of nanosilica particles improves the toughness of the epoxy resin to a significant level. Silva et al.¹⁷ characterized short sisal and coconut fibre composites as well as sisal fabric composites using compact tension specimens. It was found that increasing fibre content increased fracture toughness of the composites. At comparative fibre content, sisal fabric composite demonstrated better fracture toughness compared to short sisal fibre composite. The fracture toughness of jute and hemp laminates-reinforced polyester composites was investigated by Hughes et al.⁴ It was found that hemp/polyester composite demonstrated better critical stress intensity factor and energy release rate. At 20% of fibre volume fraction, 313% and 870% improvement in fracture toughness and critical strain energy release rate was achieved for jute/polyester composite, respectively, whereas for hemp/polyester composite, the improvement was 466% and 1740%, respectively. Reis and Ferreira¹⁸ in their work have analysed three different types of natural fibres, namely coir, sugarcane bagasse and banana fibres and concluded that coir and sugarcane bagasse fibres reinforcement improved the fracture toughness by 15.7% and 17.8%, respectively, compared with unreinforced epoxy concrete. However, 22.2% of deterioration in fracture toughness was found for banana pseudostem fibre reinforcement. Fracture energy was found to be improved for all types of natural fibres composites, where 100.8%, 15.9% and 41.1% improvement were achieved for coir, sugarcane bagasse and banana pseudostem fibre, respectively. Wonga et al.¹⁹ have investigated the fracture behaviour of short bamboo fibre-reinforced polyester composites. Results of Wonga et al.¹⁹ have indicated that the fracture toughness of all types of composites is higher compared to neat polyester. Maximum increment of 340% is achieved at 10 mm/50 vol% of fibre reinforcement. Also they have characterized that the toughening mechanisms involved are crack-tip blunting, crack deflection and crack pinning which lead to energy dissipation through matrix plastic deformation, fibre debonding, fibre pull out and fibre damage. Recently, authors²⁰ have studied the mode I and mode II fracture toughness behaviour of coconut fibre and walnut shell particle-reinforced hybrid biocomposite.

Epoxy-based composites are prone to brittle fracture under mechanical loading. Fractures are often initiated from pre-existing cracks or flaws embedded in the fibre or particle-reinforced composites after the manufacturing process or during the use. These flaws are usually oriented at an arbitrary angle relative to the loading direction. Therefore, the investigation of fracture resistance of epoxy-based hybrid biocomposites under mixed-mode loading is of great importance. The aim of this work is to develop a hybrid biocomposite material composed of epoxy resin and reinforced with walnut shell particles and coconut fibres. Semicircular arc bend specimen (SCAB) geometry along with TPB and four-point bend (FPB) loadings are used to determine the fracture toughness under different modes of loading. The results obtained from new SCAB geometry are validated with the results obtained from TPB and FPB specimen geometries by statistical significance test.

Materials and methods

The raw materials used in this study are walnut shell, coconut fibre, epoxy and hardener. Walnut shells obtained from the nearby market are first cleaned off dirt and impurities and then are chipped using a knife ring flaker. After that, the chipped shells are converted to particle form by Wily mill. The walnut shell particle sizes vary between 1.618 µm and 2.685 µm. Particles are oven dried at 100 ± 5°C for 15–20 min to reach the target moisture content (<2%) before using them as reinforcing material. The oven used in the study with the temperature range of 0–600 ± 1°C and proportional–integral–derivative-controlled specifications was supplied by M/s System Control (Chennai, Tamil Nadu, India). Coconut fibres are extracted from exocarp, washed with distilled water and dried at 100°C for 24 h. The length and diameter of coconut fibres used in the present investigation vary from 1.0 mm to 1.5 mm and from 5.441 µm to 10.673 µm, respectively. Araldite CY 230 epoxy resin is used as an adhesive in the preparation of biocomposite. HY-950 is used as the hardener. CY 230 and HY 951 are supplied by M/s Fine Finish Organics Pvt. Ltd (Chennai, Tamil Nadu, India). Epoxy resins, also known as polyepoxides, are a class of reactive prepolymers and polymers that contain epoxide groups. Reaction of polyepoxides with polyfunctional hardeners HY 951 forms a thermosetting polymer, often with strong mechanical properties. The ultimate tensile strength and modulus of elasticity of the epoxy CY 230 are 44.93 MPa and 1.633 GPa, respectively.²⁰ The biocomposite is fabricated by squeeze casting method. The detailed procedure of casting is described in the works by Singh and co-workers.^21
–23 The mould of size 300 × 250 × 10 mm³ is used for fabrication.

Fracture toughness test

TPB specimen

Figure 1 shows the TPB specimen under symmetrical loading. Mode I fracture toughness K_IC under three-point bending is calculated using the following expression,

K_{I} = \frac{P}{B W^{0 .5}} f (\frac{a}{W}),

where

f (\frac{a}{W}) = 6 {(\frac{a}{W})}^{0.5} \frac{[1.99 - \frac{a}{W} (1 - \frac{a}{W}) (2.25 - 3.93 \frac{a}{W} + 2.7 {(\frac{a}{W})}^{2})]}{(1 + 2 \frac{a}{W}) {(1 - \frac{a}{W})}^{1.5}},

where P, B, W and a are the peak load, specimen thickness, specimen width and crack length, respectively.

Figure 1.

TPB specimen. TPB: three-point bend.

FPB specimen

Single edge-notched FPB specimen shown in Figure 2 is used for mode II fracture experiments. The mode II fracture toughness K_IIC under four-point bending is calculated using the following expression,

K_{I I C} = \frac{Q}{B W^{\frac{1}{2}}} f_{I I} (\frac{a}{W}),

where

Q = P (\frac{L_{1}}{L_{1} + L_{2}} - \frac{L_{3}}{L_{3} + L_{4}}),

where Q and P are the shear and applied forces, respectively.

Figure 2.

FPB specimen. FPB: four-point bend.

The geometry factor is determined from the following relation:

f_{I I} (\frac{a}{W}) = 10.27 {(\frac{a}{W})}^{4} - 16.29 {(\frac{a}{W})}^{3} + 10.01 {(\frac{a}{W})}^{2} + 1.18 (\frac{a}{W})

SCAB specimen

Semi circular bend (SCB) specimen configurations (Figure 3(a) and (b)) are suggested by different test standards for mode I, mode II and mixed-mode I/II fracture toughness test. In order to control the relative combination of modes I and II in the classical SCB specimen, one has to produce specimens with different crack inclination angles. One major drawback for the classical SCB specimen is related to the practical difficulties in producing an angled crack in the specimen, particularly for mode II dominant loading conditions in which the crack angle is relatively large (about 50°). The improved SCB configuration as suggested by Ayatollahi et al.²⁴ also requires a special loading set-up to produce different mode mixity. Special care is needed to adjust the different span lengths from the crack line or load line and may provide measurement error to a significant level. Because of such shortcomings in SCB specimen, a modified arc SCB specimen is proposed in this work to overcome the previous weaknesses. Among the specimen configurations such as classical SCB, modified SCB or Brazilian disc (BD) specimen, SCAB is also suitable for modes I, II and mixed-mode I/II test, and this type of SCAB specimen geometry can save more than 50% of the material as compared to SCB- or BD-type specimens. Figure 3 shows the comparison of three types of specimen geometry. Figure 3(c) shows the geometry and loading conditions for the proposed test configuration called the SCAB. In this test configuration, a semicircular arc specimen of width W that contains an edge crack of length a emanating normal to the inner curve surface of the specimen is loaded at the centre by a TPB fixture. Whilst the specimen is easily manufactured, it does not need complicated loading fixtures for using it in the conventional testing machines. The TPB fixture available with any conventional universal testing machine can be used without any modification or extra attachment. Moreover, the crack is always perpendicular to the tangent at the point of the crack. This simplifies the making of the angled crack. The SCAB specimen can also be used for mixed-mode test just by changing the location of the crack from the load line. Hence, different mode mixities can be obtained in the proposed specimen. The specimen can be used for the simple case of symmetric loading conditions in order to obtain pure mode I fracture toughness for several engineering materials. The finite element (FE) method is employed in this research to determine K_I and K_II. The stress intensity factors K_I and K_II for the SCAB specimen are functions of the crack length a, the specimen width W and crack inclination angle α with respect to the loading direction. This can be written as:

K_{I} = \frac{P_{u}}{2 B W} \sqrt{π a} Y_{I},

K_{I I} = \frac{P_{u}}{2 B W} \sqrt{π a} Y_{I I},

where B is the specimen thickness and Y_I and Y_II are the geometry factors corresponding to modes I and II and function of (a/W, α). P_u is the ultimate load. For calculating Y_I and Y_II, FE models of the SCAB specimen having different crack configurations have been analyzed using the FE code ANSYS (Cecil Township, Pennsylvania, USA). In the models, the following geometry and loading conditions are considered: W = 20 mm, B = 10 mm, P = 500 N and different values of crack lengths. Span distance is kept as 60 mm. The elastic material properties of epoxy as E = 1.633 GPa and ν = 0.35 are considered in the FE models. A total number of 2162 eight-noded plane stress elements are used for each model. The singular elements are considered in the first ring of elements surrounding the crack tip for producing the square root singularity of stress/strain field. The stress intensity factors are directly obtained from the software. Figures 4 and 5 show the values of Y_I and Y_II calculated from several FE analyses performed for different geometry conditions of the SCAB specimen. It is seen from these figures that for the symmetric loading conditions (α = 0°), Y_II equals zero, and thus the specimen is subjected to pure mode I loading. By changing the inclination of the crack with respect to the loading direction, the mode II component also appears in the SCAB specimen. It is seen from the FE results that at α = 58°, the crack completely closes, and failure of the specimen occurs due to shearing effect only. Hence, it is concluded that pure mode II exists at α = 58°. It is seen from Figures 4 and 5 that for α = 58°, the mode I geometry factor Y_I has decreased to a very small value and pure mode II condition has appeared.

Figure 3.

SCB specimen under TPB loading (a) with inclined crack (b) asymmetrical loading (c) SCAB specimen. SCB: semicircular bend; TPB: three-point bend; SCAB: semicircular arc bend.

Figure 4.

Variations of mode I geometry factor Y_I with a/W for different crack inclination angle (α). a: crack length; W: specimen width.

Figure 5.

Variations of mode II geometry factor Y_II with a/W for different crack inclination angle (α). a: crack length; W: specimen width.

Fracture toughness tests are carried out at room temperature in an ADMET servo hydraulic universal test system (Norwood, Massachusetts, USA). For this purpose, single notched TPB, FPB and SCAB specimens (Figures 1, 2 and 3(c)) are employed for mode I, mode II and mixed-mode I/II fracture experiment. Specimens are prepared according to the dimensions shown in Figures 1, 2 and 3(c) from the 10 mm thickness board cast using different reinforcing materials described earlier. The mechanical slit is made at the required position (Figures 1, 2 and (3(c)) by means of a jewellery saw of 0.1 mm thin blade. After obtaining the required size of a mechanical slit, a sharp pre-crack is introduced in the bend specimen by lightly tapping a sharp new razor blade into the tip of the mechanical slit or notch. All the tests are carried out at 0.5 mm/min crosshead speed.

Figure 6 shows the loading set-up for the SCAB specimen geometry. The load–displacement data are recorded during the tests and are shown in Figure 7. All the test samples fractured suddenly from the crack tip and with negligible non-linear deformation show the brittle fracture behaviour of the tested samples. Using the fracture load obtained from each specimen, the stress intensity factors are calculated and the results are presented in Tables 1–3.

Figure 6.

SCAB specimen geometry and loading set-up of walnut shell particle and coconut fibre-reinforced biocomposite. SCAB: semicircular arc bend.

Figure 7.

Load–deflection behaviour of coconut fibre and walnut shell particles-reinforced hybrid biocomposite under (a) three-point bending (SENB (b) four-point bending (c) semicircular arc bending (note: number 1, 2, …, etc mentioned in figures indicate specimen number (replication), alpha indicates crack inclination to the loading direction). SENB: single-edged notch bending.

Table 1.

Fracture load and mode I critical stress intensity factor with their mean and variance obtained from TPB and SCAB α = 0° specimens.

Test no.	Test type	Fracture load (N)	K _IC $(M P a \sqrt{m})$	Mean value $(M P a \sqrt{m})$	Standard deviation	Coefficient of variance (%)
1.	TPB	108.8	0.879	1.267 (TPB)	0.376	29.7
2.	TPB	124.0	1.005
3.	TPB	155.2	1.277
4.	TPB	116.0	0.964
5.	TPB	126.0	1.026
6.	TPB	151.0	1.539
7.	TPB	125.1	2.251
8.	TPB	109.3	1.600
9.	TPB	114.4	1.099
10.	TPB	146.2	1.189
11.	TPB	95.3	0.973
12.	TPB	189.4	1.554
13.	TPB	89.0	0.839
14.	TPB	184.0	1.548
15.	SCAB	328.3	1.133	1.319 (SCAB)	0.300	22.74
16.	SCAB	459.2	1.439
17.	SCAB	290.6	0.961
18.	SCAB	594.3	1.747

TPB: three-point bend; SCAB: semicircular arc bend; K_I: mode I fracture toughness; K_II: mode II fracture toughness; α: inclination angle.

Table 2.

Fracture load and mode II critical stress intensity factor and their mean and variance obtained from FPB and SCAB α = 60° specimen geometry.

Test no.	Test type	Fracture load (N)	K _IC $(M P a \sqrt{m})$	K _IIC $(M P a \sqrt{m})$	Mean value $(M P a \sqrt{m})$	Standard deviation	Coefficient of variance (%)
1.	FPB	738.7	0	0.359	0.754	0.241	31.97
2.	FPB	1190.0	0	0.576
3.	FPB	1412.0	0	0.774
4.	FPB	1910.0	0	0.985
5.	FPB	1152.0	0	0.747
6.	FPB	1930.0	0	1.081
7.	SCAB	3349.5	0.031	1.219	1.219	0.061	5.02
8.	SCAB	3644.0	0.046	1.144
9.	SCAB	3557.0	0.033	1.294

FPB: four-point bend; SCAB: semicircular arc bend; K_I: mode I fracture toughness; K_II: mode II fracture toughness; α: inclination angle.

Table 3.

The fracture toughness values of biocomposite material for different crack inclination angles obtained from SCAB specimen geometry.

Sl no.	Specimen ID	α (deg)	P_u (N)	K _I $(M P a \sqrt{m})$	K _II $(M P a \sqrt{m})$	Remarks
1.	15-1	15	539.3	1.470	0.092	Mode I/II
2.	15-2	15	386.8	0.932	0.059	Mode I/II
3.	15-3	15	638.5	1.560	0.098	Mode I/II
4.	30-4	30	857.9	1.121	0.176	Mode I/II
5.	30-5	30	842.8	1.125	0.178	Mode I/II
6.	30-6	30	1223.7	1.845	0.297	Mode I/II
7.	45-7	45	2508.0	0.888	0.631	Mode I/II
8.	45-8	45	1879.5	0.653	0.464	Mode I/II
9.	45-9	45	2912.4	0.980	0.674	Mode I/II
10.	45-10	45	2011.9	0.681	0.487	Mode I/II

SCAB: semicircular arc bend; α: inclination angle; P_u: ultimate load; K_I: mode I fracture toughness; K_II: mode II fracture toughness.

Results and discussion

Mode I fracture toughness

In the present investigation, mode I fracture toughness (K_IC) is determined from the TPB specimen and SCAB specimen with α = 0°, and the results are presented in Table 1. Fourteen tests under TPB and four tests under SCAB geometry are conducted. For three-point bending, all fracture toughness tests are conducted for the crack length a = 7.5 mm, width W = 15–15.5 mm and thickness B = 10 mm. The dimensions taken for SCAB geometry are B = 10 mm, W = 20 mm, 2 S = 40 mm and a = 10–10.5 mm. The load displacement curves under three-point bending and SCAB for α = 0°are shown in Figure 7(a) and (b), respectively. It is seen that curves rise in a linear way until their maximum value. An abrupt load drops off after attaining the maximum value describing the fracture behaviour under mode I loading. Under three-point bending, the failure load varies from 89.0 N to 189.4 N whereas under SCAB, the failure load varies from 290.6 N to 594.3 N. Thus, the values of K_IC under the three-point bending are found to be varied between 0.839 MPa $\sqrt{m}$ and 1.60 MPa $\sqrt{m}$ . The values of K_IC under the SCAB loading are found to be varied between 0.961 MPa $\sqrt{m}$ and 1.747 MPa $\sqrt{m}$ . In the three-point bending, the mean, standard deviation and coefficient of variance of K_IC are 1.267 MPa $\sqrt{m}$ , 0.376 MPA $\sqrt{m}$ and 29.7%, respectively. The mean, standard deviation and coefficient of variance of SCAB results are 1.319 MPa $\sqrt{m}$ , 0.30 $M P a \sqrt{m}$ and 22.74%, respectively. The absolute difference between K_IC computed from TPB and SCAB is only 0.05 MPa $\sqrt{m}$ which is about 4%. According to Tada et al.,²⁵ K_I estimation within 5% is sufficient for K_IC computation for any new geometry. As in the present case, it stays below 5% and hence, the proposed SCAB geometry can be used for fracture toughness testing with many advantages over the previous specimen geometry in use. The t test between the mean K_IC of biocomposite determined using SCAB and TPB specimen is found to be 0.414, and the tabulated value at 99% confidence level, 16 degree of freedom is 2.58. This confirms that there is no significant difference between two mean toughness determined using different geometries. Mode I fracture toughness of neat epoxy as reported by Zunjarrao et al.¹⁴ is 1.027 MPa $\sqrt{m}$ . The fracture toughness of the neat epoxy CY 230 is found to be 1.063 MPa $\sqrt{m}$ . Hence, the effective increase in K_IC value due to addition of coconut fibre (10 wt%) and walnut shell power (20 wt%) is about 26% with respect to the work by Zunjarrao et al. and 24% with respect to the results obtained from the TPB test.

Mode II fracture toughness

The mode II fracture toughness (K_IIC) is determined from the FPB specimen and the SCAB specimen with α = 60°. The results are presented in Table 2. In four-point bending, different specimen dimensions of L₁ = L₄ = 20 mm and L₂ = L₃ = 10 mm, crack length a = 4–4.5 mm, thickness B = 10 mm and width W = 10 mm are taken. The geometry parameters of SCAB specimens are a = 10–10.5 mm, W = 20 mm, B = 10 mm and α = 60°. The load displacement curves are shown in Figure 7(b) and (c). Similar nature to mode I is seen in the variation of load displacement. In the four-point bending, the mean, standard deviation and coefficient of variance of K_IIC are 0.754 MPa $\sqrt{m}$ , 0.241 MPa $\sqrt{m}$ and 31.97%, respectively. These values for SCAB geometry are 1.219 MPa $\sqrt{m}$ , 0.061 MPa $\sqrt{m}$ and 5.02%. The overall mean, standard deviation and coefficient of variation are 0.909 MPa $\sqrt{m}$ , 0.297 MPa $\sqrt{m}$ and 32.66%, respectively. The results of t test between the mean K_IIC of biocomposite determined using SCAB and TPB specimen is found to be 2.901, and the tabulated value at 99% confidence level, 7 degree of freedom is 3.0. This confirms that there is no significant difference between the two mean toughness values determined using different geometries.

Mixed mode (modes I and II) fracture toughness

The mixed-mode fracture tests are conducted on the SCAB specimens with different crack inclination angles, foe example, α = 15°, 30° and 45°. The force–displacement diagrams for α = 0°, 15°, 30° and 60° shown in Figure 7(c) are linear up to fracture. In calculation of mode I and mode II fracture toughness, the peak load in each experiment is considered. This is because the force–displacement diagrams are linear and fulfil all the requirements of ASTM standard. The mixed-mode fracture toughness values are determined using fracture load and geometry factors given in Figures 5 and 6 for SCAB specimen with different values of α. The results are presented in Table 3.

The mean values of K_I and K_II with their statistical properties are shown in Table 4 for biocomposite material. The coefficient of variance is found to vary between 15% and 26% under mixed-mode loading. The higher coefficient of variance is mainly due to inhomogeneity in the material and different failure mechanisms working together in particle and fibre-reinforced hybrid biocomposite. It is seen that as the crack inclination angle increases, the effective stress intensity factor also increases. However, for crack angle α ≥ 45°, the effective stress intensity factor decreases and the value became slightly less than the mode I fracture toughness. In the present investigation, optimum effective stress intensity factor is observed for α = 30°. It is also observed that the load-carrying capacity of mode II specimen increases drastically as compared to the mode I geometry. Even the specimen size in SCAB geometry is same for α = 0° and 60°, the load-bearing capacity with α = 60° is about 8.41 times more than α = 0°. This is because under mode I loading, the crack propagates along the direction of the load line, and hence due to opening of the crack most of the particles ahead of the crack tip are either pulled out or the crack moves around the particle making a curvature path. Under mode II condition, complete crack closure occurs and shear-induced effects become dominant. Under shear loading unlike normal loading, fracture does not take place along the initial crack and thus the chances of particle pull out is less as compared to mode I. During experimental work, it is observed for α = 60° that the closing of crack is started at a load level of 870 N and it closed completely when the load level reached 2000 N, and the specimen finally fractured at a load level of 3557 N. Thus, several mechanisms are working under such loading. Thus due to many toughening mechanisms such as crack bridging, crack curvature and shear-induced effects, more energy is needed and hence the load-bearing capacity under mode II is increased to a significant magnitude as compared to mode I.

Table 4.

Overall mean, variance and coefficient of variance of fracture toughness of 20 wt% walnut shell particle and 10 wt% coconut fibre-reinforced biocomposite.

α (deg)	Mean		Standard deviation		Coefficient of variance
α (deg)	K _I $(M P a \sqrt{m})$	K _II $(M P a \sqrt{m})$	$σ_{K_{I}}$	$σ_{K_{I I}}$	$V_{K_{I}}$ (%)	$V_{K_{I I}}$ (%)
0	1.319	0	0.300	0	22.75	0
15	1.321	0.083	0.277	0.018	20.99	20.66
30	1.364	0.217	0.340	0.057	24.96	26.07
45	0.801	0.564	0.138	0.090	17.21	15.99
60	0.037	1.219	6.65E-03	0.061	17.97	6.173

K_I: mode I fracture toughness; K_II: mode II fracture toughness; α: inclination angle.

Fracture toughness ratio

The ratio of K_IIC over K_IC obtained from TPB and FPB specimens, respectively, varies typically from 0.159 to 1.288. The mean value of the ratio is 0.595 for the investigated biocomposite material. However, from the experimental results of SCAB geometry, the ratio varies from 0.655 to 1.347 with a mean value of 0.924 which is closer to the maximum tangential stress (MTS) predicted value of 0.87. The overall ratio from the results of SCAB, TPB and FPB is 0.762 with a mean K_IC = 1.294 MPa $\sqrt{m}$ and mean K_IIC = 0.987 MPa $\sqrt{m}$ . The theoretical prediction based on the MTS criteria²⁶ suggests the ratio of mode II fracture over mode I fracture as 0.87. However, there are many experimental investigations reporting a ratio of K_IIC/K_IC significantly higher or lower than 0.87. Reviews of different investigations show that this value depends on several factors such as material, specimen geometry, loading conditions and so on. In literature, the ratio of K_IIC/K_IC is found to vary from 0.45 to 2.2. In the present investigation, the mean ratio of K_IIC/K_IC for pure epoxy is found as 0.925. Ayatollahi and Aliha²⁷ have reported that the average value of K_IIC/K_IC for polymethyl methacrylate or Perspex material is 0.542. The theoretical value of fracture toughness ratio based on MTS criteria as 0.87 is obtained under the condition of non-inclusion of T stress or higher order stresses in the MTS criterion. The higher value obtained in the present investigation as compared to theoretical value is due to the non-inclusion of T stress in the theoretical prediction. It has been reported by Ayatollahi et al.²⁴ that the specimen having a negative T stress, the value of K_IIC/K_IC is higher than 0.87. For the case of SCB specimen, where T stress is positive, K_IIC/K_IC assumes a value <0.97. This suggests that SCB specimen gives a conservative value of K_IIC in comparison with other specimens that may have negative T stress associated with the crack-tip stress field. In the present investigation, negative T stress exist for M^e ≥ 0.8 and positive T stress for M^e < 0.8 for SCAB geometry. However, the percentage difference between experimental result and theoretical result is only 6.29% if fracture toughness values obtained from SCAB geometry are considered. This shows that a good agreement exists between experimental result and the MTS criteria based theoretical prediction. The overall percentage difference between the predicted and experimental is 12.4%. Under mixed mode loading condition the ratio between mode II stress intensity factor to mode I stress intensity factor varies from 0.063 to 0.711 for crack inclination angle α = 15° to 45°.

Scanning electron microscope (SEM)

The reinforcement by incorporation of particles or fibres has been articulated as the result of strong interfacial bonding between the matrix material and the reinforcing elements either in the fibre form or in the particulate or hybrid form. Typically, the degree of bonding is also influenced by the dispersion of particles in the polymer matrix. A homogeneous dispersion will increase the effective particle polymer interfacial area and leads to the strong interaction. In addition, good dispersion also reduces large and dense particle agglomeration, which tends to act as defects, constituting stress concentrations under applied loads, which facilitate the triggering of micro-cracks and their progression to material failure.²⁰ Several toughening mechanisms of particle–fibre-reinforced composites have been reported in the literature. Some of the toughening mechanisms in epoxy matrices are crack path deflection,^28,29 plastic deformation,^30
–32 crack front pinning,^32,33 shear-induced mechanism³⁴ and particle debonding and subsequent void growth.^35,36

The micrographs of different magnifications obtained from SEM analysis are presented in Figure 8 to study the toughening mechanism of 20 wt% walnut shell particles and 10 wt% short coconut fibre-reinforced epoxy composites. Figure 8(a) shows the SEM micrographs of failed specimen under mode I loading. Three distinct zones such as machined cut, crack extended by a razor blade and crack propagation zone are clearly visible in the figure. The smooth surface at the middle of Figure 8(a) indicates the condition of the crack tip. Approximately 265.882 µm is extended from the saw cut notch by a new razor blade. This length slightly varies from specimen to specimen. The growth direction in Figure 8(a) is from right to the left. Figure 8(b) shows the dispersion of the walnut particles in the matrix material. In Figure 8(b), it is clear that the walnut shell particles are well dispersed and there is no such dense particle agglomeration. It is reported by many researchers^20,37,38 that small particle of uniform size and shape has the ability to enhance both the stiffness and ductility through its dispersion in the matrix. Therefore, more energy can be absorbed when the walnut shell particle is added to the matrix in addition to the coconut fibre. The increase in the fracture toughness due to the addition of walnut particles may be due to the particle obstruction in the propagation of micro-cracks by inhibiting their growth. Under loading, deformation tends to concentrate at the initiative micro-cracks instead of being distributed evenly throughout the material. Therefore, the initiation of more micro-cracks is delayed due to the continuing energy dissipation at the deformation zones. When the strain energy around the existing cracks grow greatly, de-bonding occurs near the interface. Hence, the weakening effect of particles prevails under loading and accounts for the reduction in ductility. Figure 8(c) shows the adhesion of the walnut particles on the coconut fibre. The diameter of the fibre shown in Figure 8(c) is 145.48 µm. The contribution of the short-fibre reinforcement to the total interfacial area plays a significant role on the toughness properties.²⁰ This is due to the large surface-to-volume ratio of short-fibre reinforcement. Increase of interfacial area due to the short fibre is detrimental to crack resistance because crack tends to propagate mainly into the short-fibre/matrix interfaces. It has been reported that the main reason for fracture in discontinuously reinforced systems is the formation and coalescence of voids arising from reinforcement fracture ahead of crack tips.^39,40 Considering the effects of the above factor, walnut shell particles are found to be more useful in combination of short coconut fibres in enhancing the crack resistance of composite. Thus, short-fibre and particle-reinforced hybrid composites could provide technical advantages over conventional discontinuously reinforced composites. The combination of short fibre and particle enables better control of damage tolerance properties as well as greater control of the size and extent of reinforcement. In the present investigation, enhancement in the fracture toughness is seen because of many mechanisms acting simultaneously.

Figure 8.

SEM images of fracture surfaces (a) saw cut, crack tip and fracture zone, (b) dispersion of walnut shell particles, (c) adhesion of walnut shell particles on coconut fibre, (d) mode I (TPB geometry), (e) mode II (FPB geometry), (f) mixed mode M^e = 0.960 and (g) mixed mode M^e = 0.899. SEM: scanning electron microscopic; TPB: three-point bend; FPB: four-point bend.

The fracture processes taking place on or in the vicinity of the fracture surface are analysed using SEM. Figure 8(d) and (e) shows that the fracture surface just near the crack tip of the mode I and mode II specimen failed under the three-point bending and the four-point bending, respectively. It is reported that the pull out of the particle or fibre from the matrix material indicates a bridging mechanism in particle–fibre-reinforced composites.³⁶ Figure 8(d) reveals that failure occurs due to pull out of the particles and fibres without noticeable plastic deformation. The plastic deformation could not be noticed because of the presence of hard walnut particles in the matrix material. The material is in front of the crack tip in the presence of hard particles of walnut shell, the crack propagates through the matrix material leaving almost negligible plastic deformation. Johnsen et al.³⁵ in their study have reported that particle de-bonding is one of the major toughening mechanisms for nanoparticle–epoxy composites. Figure 8(e) shows the fracture surface of the FPB specimen failed under pure mode II condition. In Figure 8(e), exposed fibres can be observed in large numbers and length compared to Figure 8(d), where most of the fibres are broken without sliding, which denotes strong interfacial adhesion characteristics. Hence, under mode II loading, the failure of the biocomposite is mostly because of fracture of the fibres. However, very few pull out failures of the fibres and particles are also seen in Figure 8(e). From these observations, it can be said that under pure mode II loading the failure is mainly due to shearing and the growth takes place in a curvature manner. This restricts the de-bonding of the fibre and the crack propagates due to splitting of the fibres into two or more pieces.

Figure 8(f) and (g) shows the SEM micrographs of fracture surface of the biocomposite broken under mixed-mode loading conditions. Figure 8(f) and (g) presents the fracture surface of specimen for α = 15° and 30°, respectively. The random orientation of fibres and breaking of fibre are clearly visible in Figure 8(f). Figure 8(f) reveals hybrid failure mechanism of fibre cracking and fibre pull out. Some parabolic features are also present in Figure 8(f). This type of pattern is seen in mixed mode (α = 15°) along with some matrix cracking and particle pull out. Under mixed-mode loading condition, the crack initiates at an angle θ₀, does not grow in a straight line and the crack path profile is curved. The curved path of crack growth can be due to the fact that the MTS direction changes as the crack advances. Thus, at each step of crack propagation, the crack plane rotates and grows in a new direction which has the MTS. The presence of stiff walnut shell particles in each step of crack plane rotation can provide a high resistance against crack propagation, which results in a higher fracture energy. A curved fracture trajectory comes across more particles during its extension resulting in an increased reinforcement efficiency of these particles. One of the most important fracture mechanisms in the particulate composite is crack deflection.⁴¹ Ayatollahi et al.^36,42 have already mentioned that such crack deflection and crack deviation mechanisms resulting to a curvilinear crack path in their studies. In Figure 8(d), it is observed that the initial crack plane and the crack initiation planes are different. This indicates about the increased shear effect under mode II dominant loading. Figure 8(e) and (f) shows that the fracture surfaces are rougher than that of α = 0°. The increasing roughness indicates that the intermolecular forces may be less and because of reinforcing effect the failure load increases. Similar conclusions are made by Wetzel et al.⁴¹ Besides the above observations, some features like the waviness and curved shape of fillers in the matrix are also seen. These natures of the particles make the pull out process more difficult and result in higher fracture resistance.

Conclusions

The use of coconut fibres and walnut shell particles as reinforcing materials in epoxy resin are made and the mode I, mode II and mixed-mode I/II fracture toughness are investigated. Hybridization of fibre with particle improves the fracture toughness of the biocomposite material. The overall mean values of mode I and mode II fracture toughness of biocomposite are 1.279 $M P a \sqrt{m}$ and 0.909 $M P a \sqrt{m}$ , respectively. An increment of about 24% is achieved under mode I loading. Increase in the mode I fracture toughness is observed under mixed-mode loading also and the highest mode I fracture toughness is observed at mode mixity 0.9 (crack inclination angle α = 30°) which is about 6.7% higher than the mode I fracture toughness under mode I loading. However, mode II fractures toughness increases with increase of the crack inclination angle. The ratio of K_IIC over K_IC varies typically from 0.159 to 1.347 with mean value of 0.762 which is closer to the MTS criterion predicted value of 0.87. The experimental results show that a combination of different toughening mechanism works under mixed-mode loading condition and hence increase in the toughness value is obtained. From the present observations, it can be concluded that the toughening mechanisms involved in the particle–fibre-reinforced biocomposite are crack deflection, crack bridging and shear-induced effects which lead to energy dissipation through matrix plastic deformation, fibre debonding, fibre pull out and fibre splitting or fracture. Thus short-fibre/particle hybrid composites could provide mechanical advantages over conventional discontinuously reinforced composites. The combination of short fibre and particle enables better control of damage-tolerance properties.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Maya

Sabu

. Review: biofibres and biocomposites. Carbohydr Polym 2008; 71: 343–364.

Faruk

Bledzki

Finkb

. Biocomposites reinforced with natural fibers: 2000–2010. Prog Polym Sci 2012; 37(11): 1552–1596.

Mohanty

Misra

Hinrichsen

. Biofibers biodegradable polymers and biocomposites: an overview. Macromol Mater Eng 2000; 276/277(1): 1–24.

Hughes

Hill

CAS

Hague

JRB

. The fracture toughness of bast fibre reinforced polyester composites. Part 1: evaluation and analysis. J Mater Sci 2002; 37(21): 4669–4676.

Wong

Mai

. Fracture resistance and microstructures of unreinforced and fiber reinforced PA6,6/PP/SEBS-g-MA. In: Proceedings of the 56th annual technical conference of the society of plastics engineers, Atlanta, USA, 26–30 April, 1998, pp. 3113–3117. Society of Plastic Engineers, ISSN-ISBN:1-56676-669-9.

Atodaria

Putatunda

Mallick

. A fatigue crack growth model for random fiber composites. J Compos Mater 1997; 31(18): 2645–2659.

ASTM D 5045-96. Standard test method for plane-strain fracture and strain energy release rate of plastic materials, 1996.

Han

. Short-fiber/particle hybrid reinforcement: effects on fracture toughness and fatigue crack growth of metal matrix composites. Compos Sci Technol 2007; 67: 1719–1726.

Laffan

Pinho

Robinson

. Translaminar fracture toughness testing of composites: a review. Polym Test 2012; 31: 481–489.

10.

Silva

Spinelli

Bose Filho

. Fracture toughness of natural fibers/castor oil polyurethane composites. Compos Sci Tech 2006; 66: 1328–1335.

11.

Zhao

Botsis

. Experimental and numerical studies in model composites, part I: experimental results. Int J Fract 1996; 82: 153–174.

12.

Choi

Takahashi

. Toughness and microscopic fracture mechanisms of unfilled and short-glass-fiber-filled poly(cyano arylether). J Mater Sci 1996; 31(3): 731–740.

13.

Botsis

Beldica

. Strength characteristics and fatigue crack growth in a composite with long aligned fibers. Int J Fract 1995; 69(1): 27–50.

14.

Zunjarrao

Singh

. Characterization of the fracture behavior of epoxy reinforced with nanometer and micrometer sized aluminum particles. Compos Sci Technol 2006; 66: 2296–2305.

15.

Kim

Park

Lee

. Fracture toughness of the nano-particle reinforced epoxy composite. Compos Struct 2008; 86: 69–77.

16.

Zamanian

Mortezaei

Salehnia

. Fracture toughness of epoxy polymer modified with nanosilica particles: particle size effect. Eng Fract Mech 2013; 97: 193–206.

17.

Silva

Spinelli

Bose Filho

. Fracture toughness of natural fibers/castor oil polyurethane composites. Compos Sci Technol 2006; 66(10): 1328–1335.

18.

Reis

JML

Ferreira

AJM

. A contribution to the study of the fracture energy of polymer concrete and fibre reinforced polymer concrete. Polym Test 2004; 23(4): 437–440.

19.

Wonga

Zahi

Low

. Fracture characterisation of short bamboo fibre reinforced polyester composites. Mater Des 2010; 31: 4147–4154.

20.

Rao

Gope

. Fracture toughness of walnut particles (Juglans regia L.) and coconut fiber-reinforced hybrid biocomposite. Polym Compos. Epub ahead of print 18 February 2014. doi: 10.1002/pc.22926.

21.

Singh

Gope

. Silica-styrene-butadiene rubber filled hybrid composites: experimental characterization and modeling. J Reinforc Plast Compos 2010; 29(16): 2450–2468.

22.

Gope

Singh

. Effect of filler addition and strain rate on the compressive strength of silica- styrene butadiene rubber filled epoxy composites. J Poly Eng Sci 2011; 51(6): 1130–1136.

23.

Singh

Gope

Chauhan

. Mechanical behavior of banana fiber based hybrid biocomposites. J Mater Environ Sci 2012; 3(1): 185–194.

24.

Ayatollahi

Aliha

MRM

Saghafi

. An improved semi-circular bend specimen for investigating mixed mode brittle fracture. Eng Fract Mech 2011; 78: 110–123.

25.

Tada

Paris

Irwin

. The stress analysis of cracks handbook. 3rd ed. New York: ASME, 2000.

26.

Williams

Ewing

. Fracture under complex stress: the angled crack problem. Int J Fract 1972; 8: 441–446.

27.

Ayatollahi

Aliha

MRM

. On determination of mode II fracture toughness using semi-circular bend specimen. Int J Solids Struct 2006; 43: 5217–5227.

28.

Faber

Evans

. Crack deflection processes: I. Theory. Acta Metall 1983; 31(4): 565–576.

29.

Faber

Evans

. Crack deflection processes: II. Experiment. Acta Metall 1983; 31(4): 577–584.

30.

Lee

Yee

. Inorganic particle toughening II: toughening mechanisms of glass bead filled epoxies. Polymer 2001; 42: 589–597.

31.

Kawaguchi

Pearson

. The moisture effect on the fatigue crack growth of glass particle and fiber reinforced epoxies with strong and weak bonding conditions. Part 2. A microscopic study on toughening mechanism. Comp Sci Technol 2004; 64: 1991–2007.

32.

Kinloch

Maxwell

Young

. Micromechanisms of crack propagation in hybrid-particulate composites. J Mater Sci Lett 1985; 4: 1276–1279.

33.

Spanoudakis

Young

. Crack propagation in a glass particle-filled epoxy resin. Part 1 effect of particle volume fraction and size. J Mater Sci 1984; 19: 473–486.

34.

Bhattacharjee

Knott

. Effect of mixed mode I and II loading on the fracture surface of polymethyl methacrylate (PMMA). Int J Fract 1995; 72: 359–381.

35.

Johnsen

Kinloch

Mohammed

. Toughening mechanisms of nanoparticle-modified epoxy polymers. Polymer 2007; 48(2): 530–541.

36.

Ayatollahi

Shadlou

Shokrieh

. Mixed mode brittle fracture in epoxy/multi-walled carbon nanotube nanocomposites. Eng Fract Mech 2011; 78: 2620–2632.

37.

Argon

Cohen

. Toughenability of polymers. Polymer 2003; 44: 6013–6032.

38.

Komarneni

. Nanocomposites. J Mater Chem 1992; 2(12): 1219–1230.

39.

Le Roy

Embury

Edwards

. A model of ductile failure based on the nucleation and growth of voids. Acta Metall 1981; 29(8): 1509–1522.

40.

Hunt

Jr Osman

Lewandowski

. Micro- and macrostructural factors in DRA fracture resistance. JOM-US 1993; 45: 30–35.

41.

Wetzel

Rosso

Haupert

. Epoxy nanocomposites–fracture and toughening mechanisms. Eng Fract Mech 2006; 73: 2375–2398.

42.

Shadlou

Alishahi

Ayatollahi

. Fracture behavior of epoxy nanocomposites reinforced with different carbon nano-reinforcements. Comps Struct 2013; 95: 577–581.