Effect of fibre loading on mechanical and thermal properties of sisal and kenaf fibre-reinforced injection moulded composites

Abstract

Kenaf and sisal fibres were selected for the development of natural fibre-reinforced thermoplastic composites because they are in abundance in South Africa. These materials can be used in the building industry, where good mechanical and fire resistance properties are required for structural and non-structural applications. High load bearing, moisture and fire can negatively affect the properties of these composites and decrease their performance. The effect of water glass, maleic anhydride-grafted-polypropylene (MAPP) treatment and fibre loadings on mechanical and thermal properties as well as fire resistance of kenaf and sisal fibre-reinforced polypropylene composites were investigated. The fibre-reinforced polypropylene composites were produced by injection and compression moulding processes. A significant improvement of the tensile strength and modulus of kenaf fibre composites was more pronounced when compared to sisal fibre composites. However, the water glass has a slight negative influence on mechanical properties of fibre/polypropylene matrix composites but showed better fire resistance and thermal properties when compared to the untreated fibre/polypropylene composites. Generally, the impact strength of fibre/polypropylene composites decreased with the addition of MAPP. The low impact strength could be the results of weak interfacial bond strength between the fibre and the matrix. The total heat release was reduced by water glass treatment due to the non-flammable polysilicate coating, while the average mass loss rate was slightly reduced by MAPP.

Keywords

Sisal fibre kenaf fibre water glass flammability test

Introduction

Polymer composites using natural plant fibres as reinforcement have attracted a lot of interest in recent years. The main reasons are that these natural fibres can be harvested from renewable resources, possess long aspect ratio for efficient stress transfer and certain aspects of their mechanical properties are comparable to existing inorganic glass fibres.¹ Amongst these plant fibres, sisal, jute, flax, hemp, banana and coconut have attracted particular attention. Early research studies on natural plant fibre-reinforced composites have focused on thermosetting composites and detailed references can be found in two recent review articles.² For fabrication of thermosetting composites, matrix impregnation into the reinforcing fibres is facilitated by the low viscosity of the uncured resin.³ Furthermore, if low cure temperature resin is used, the matrix can be cured without causing thermal degradation to the reinforcing natural fibres. Thermoplastics are alternative candidate matrix materials. They possess a number of obvious advantages over thermoset matrices.⁴ In recent years, there has been a renewed interest in the use of natural fibre as substitute for glass because of the potential advantages of weight saving, lower raw price, recyclable and renewable.⁵ Natural fibres have always found wide application from the time they gained commercial prominence. However, natural fibres exhibit a high degree of moisture absorption, which can be the major problem in many applications.⁶ Because the interface acts as a binder and transfers stress between the matrix and the reinforcing fibre, the quality of a fibre-reinforced composite depends considerably on the fibre/matrix interface. However, studies have indicated that maleic anhydride can serve as effective compatibiliser for natural fibres and polyolefin matrices, resulting in the improvement of the mechanical properties.⁷

Thermal analysis of cellulose derivatives/starch blends, with different sisal short fibre content, was performed by thermogravimetric analysis (TGA)/derivative thermogravimetric analysis (DTGA) under dynamic conditions. Apparent kinetic parameters were determined using a variety of conventional thermogravimetric methods. Two peaks were found: the first is close to 334℃ and the second at 369℃. The apparent activation energy value of the first peak and the maximum temperature value slightly increased. However, the apparent activation energy values for the second peak and the maximum temperature value decreased. The addition of sisal fibres did not produce a significant effect on the thermal degradation of the composites in comparison with the matrix alone.⁸ Natural fibre composites can be a potential candidate for the partial replacement of high-cost glass fibre.⁹ Considerable attention has been given to the investigation sisal and kenaf fibre composites because of their availability in South Africa. The main aim of this study is to manufacture a thermoplastic matrix-based composite with locally available natural fibres (at extreme fibre contents, i.e. 15 wt% and 30 wt%) and matrices which meet some basic requirements (e.g. mechanical properties, fire and moisture resistance) building and construction industries.

Experimental

Materials

The sisal and kenaf fibres used were supplied by the CSIR, Textile Technology Division in Port Elizabeth, South Africa. The homopolymer polypropylene (PP) powder, with a melt flow index of 12 g/10 min (230℃, 2.16 kg) was provided by Plastomark. The compatibiliser used to improve adhesion between the fibre and the matrix was the polypropylene-grafted-maleic anhydride (MAPP) granules, supplied by Eastman Chemical, USA. Water glass (WG) (provided by Protea Chemicals) was used to coat the fibres.

Preparation of sisal/PP and kenaf/PP injection moulded composites

Fibre contents of 15 wt% and 30 wt% sisal and kenaf fibres were weighed and dried at 70℃ in an air-circulating oven for 2 h, in order to reduce the absorbed moisture. A calculated quantity of dried fibres was soaked in WG. The fibres were placed outdoor to be exposed to sunlight for 8 h over two conservative days. Fibres were then dried in an air-circulating oven for 4 h and thereafter granulated to shorter fibre lengths. The WG-coated fibres, MAPP and PP were pre-mixed in a high-speed mixer. The concentrations of WG and MAPP were kept constant at 20% and 5%, respectively. Throughout the experiment, the total composition by weight was maintained to a 1000 g per batch size. Each batch was mixed at an average initial temperature of 25℃ to the maximum of 100℃, for 30–60 min, depending on the fibre content. All the mixtures were kept in the oven at 60℃ to avoid moisture absorption before they were extruded. The fibre/PP composite mixtures are presented in Table 1.

Table 1.

Fibre composites for injection moulding purpose.

Sample code	Composition mixture (wt%)
Sample code	Fibre/WG	PP	MAPP
PP	0/0	100	0
PPM	0/0	95	5
PP15F	15/0	85	0
PP15FM	15/0	80	5
PP15FW	15/3.75	81.25	0
PP15MW	15/3.75	76.25	5
PP30F	30/0	70	0
PP30FM	30/0	65	0
PP30FW	30/7.5	62.5	0
PP30FMW	30/7.5	57.5	5

PP: polypropylene; MAPP: maleic anhydride-grafted-polypropylene; WG: water glass

Extrusion

After the mixing process, the composite materials were compounded with a single-screw extruder. The extruder consists of four different temperature zones, as shown in Table 2. The screw speed was kept at 80 rev/min. As the feeding zone of the extruder was not optimised for the use of cut fibres (sisal and kenaf), the material had to be force-fed with a wooden stick. The collected extrudates were granulated in a standard granulator used in plastics compounding.

Table 2.

Extrusion operating temperatures.

Temperature (℃)
Feed zone	193
Compression zone	190
Metering zone	190
Die	199

Injection moulding

Granulated material was dried in an oven at 70℃ for 2 h before injection moulding. This is to prevent the formation of air bubbles in the specimens which might lower their mechanical properties. The materials compounded were injection moulded into tensile test specimens, using Engel Injection moulding machine. Table 3 shows the injection moulding set parameters for the fibre/PP composites.

Table 3.

Operational parameters set on the injection moulding machine.

Factor, set point temperature (℃)
Barrel 1	210
Barrel 2	210
Barrel 3	210
Melt	185
Mould	23
Pressure zones (bar)
Injection pressure	87
Hold on pressure	65
Back pressure	14
Speed (%)
Injection speed	50
Screw speed	35
Time (s)
Injection time	10
Cooling time	25

Preparation of compression moulded sisal/PP composites

The composite sheets consisted of fibre/WG/matrix/compatibiliser layers where the amount of the compatibiliser and WG was kept at 5% and 20% weight of the total fibre, respectively. The non-woven fibre content was varied, thus 0%, 15% and 30% weight of the matrix. A pure PP matrix was included. A calculated quantity of the polymer/compatibiliser films was stacked between the required numbers of non-woven fibres (untreated or treated) to give the desired fibre content of the composite. The two outer layers of the stack were always of the polymer/compatibiliser or polymer films. The stack consisting of layers of fibre and polymer/compatibiliser was compressed in a rectangular-shaped mould (160 mm × 140 mm × 4 mm) under a pressure of 10 MPa and at a temperature of 190℃ for 10 min. The same pressure was maintained for 15 min during the cooling phase. The materials compressed were moulded as flat test specimens.

Material characterisation

Tensile behaviour

The injection moulded tensile specimens were subjected to tensile testing, using an Instron Model 4302 testing machine according to the ISO R527 standard method. The width and the thickness of each sample were approximately 10 mm and 4 mm, respectively. The cross-head speed was 10 mm/min at a gauge length of 110 mm. The elastic modulus, stress-at-break and strain-at-break were calculated from the stress–strain curve.

Impact behaviour

The injection moulded specimens were subjected to impact testing machine using a Charpy (Ceast 9050) tester. The method used for impact strength testing was according to ISO 179, and the machine was calibrated for a hammer of 15 J at a starting angle of 150°. All test samples were notched, and the energy absorbed was recorded.

Thermal properties

Differential scanning calorimetric (DSC) of composite samples was performed in a DSC Q 2000, Perkin–Elmer thermal analyser, with a nitrogen atmosphere. The composite samples were heated from −60℃ to 200℃, at a rate of 5℃/min. The melt temperature, glass transition temperature and the enthalpy heat of fusion was recorded.

Flammability test method

The flammability analysis was performed only on the sisal compression moulded samples using a cone calorimeter, in accordance with ASTM E1354. The tests were conducted at the CSIR Polymers and Composites fire testing laboratory in Port Elizabeth, South Africa. The compression moulded samples were cut, to the dimensions of 100 mm × 100 mm using a table saw. The composite samples were wrapped round the back and edges with an aluminium foil before placing them on the holder. This was carried out in order to prevent any molten material dripping from the sample on to the load cell. The samples were orientated horizontally and tested at a heat flux of 50 kW/m², which approximates to 772℃. An electrical ignition source was used to ignite flammable volatile gases. The samples were tested in duplicates. The average of the two tests was taken and during tests, the following parameters were determined: time to ignite (TTI), total heat release (THR), mass loss rate (MLR), heat release rate (HRR), peak heat release rate (PHRR), heat of combustion (HC), carbon monoxide (CO) and carbon dioxide (CO₂) yields.

Results and discussion

Mechanical properties

Effect of fibre content on modulus and tensile strength

The tensile modulus and tensile strength of fibre/PP composites are shown in Figures 1 and 2, respectively. Generally, fibre reinforcement resulted in a significant increase in Young’s modulus of kenaf and sisal composites when compared to neat PP (706 MPa). It is clear from Figure 1 that kenaf/PP composites showed better tensile modulus than sisal/PP composites with or without MAPP (or WG). The MAPP-treated 30 wt% kenaf fibre showed better tensile modulus value (10764 MPa) when compared to tensile modulus value (5622.7 MPa) of MAPP-treated 15 wt% kenaf fibre composites. Similar trend was also observed with sisal fibre composites. At 30 wt% fibre loadings, the tensile modulus of kenaf and sisal composites decreased after WG treatment. This could be due to the degradation of fibres at higher concentration that weakened the fibre strengths.

The tensile strength of kenaf and sisal fibre composites showed increasing tendency at 15 wt% fibre loading with the addition of MAPP and then the tensile strength decreased with WG treatment. The addition of MAPP resulted in a strong adhesion between the fibre and the matrix by forming ester bonds.¹⁰ The scanning electron microscope (SEM) micrographs of sisal/PP composites are shown in Figure 3(a) to (c). It is clear from Figure 3(a) that more fibres at 15 wt% sisal/PP/MAPP composite are firmly embedded into polymer matrix layer, even though some gaps between the fibre and matrix were still noticed. The decrease in tensile strength of fibre/PP composites at 30 wt% fibre content with MAPP addition could be due to the excess amount of fibres that lie on top of each other rather than being mixed with the matrix, as shown in Figure 3(b). On the other hand, WG (highly alkaline) is capable of decreasing the fibre strength by breakage of the bond structure and disintegration of the non-cellulosic materials. Figure 3(c) showed that the WG coating on the surface of the fibres was achieved and the roughness on the fibre surface caused poor adhesion, hence fibre–matrix debonding was observed. However, the addition of MAPP- to WG-treated sisal/PP composite improved the tensile strength when compared to either MAPP- or WG-treated composites.

The strain-at-break curve of fibre/PP samples during testing is shown in Figure 4. It is shown that the MAPP-treated samples displayed no improvement in the failure strain of the kenaf and sisal fibre composites at 15 wt% fibre loading, while at 30 wt% fibre loading the failure strain of the MAPP-treated sisal fibre composites decreased when compared to the untreated fibres. This behaviour confirmed that an increase in fibre loading caused the material to become stiffer and stronger, hence a significant decrease in strain-at-break was observed. Similar effect was also observed with WG-treated fibre/PP composites.

Effect of fibre content, MAPP and WG on the impact strength

Figure 5 shows the average impact strengths obtained from notched impact tests of fibre/PP composite samples at room temperature. The addition of sisal fibre in PP matrix increased the impact strength of composites while the addition of kenaf fibre in PP decreased the impact strength of composites. Similar trend was also observed with modified fibres (MAPP or WG treatment) where sisal fibre composites showed higher impact strength than kenaf fibre composites. The decrease in impact strength of kenaf/PP composites could be due to the absorbance of total energy dissipated in the composites. When fibre (unmodified or modified) content was increased from 15 wt% to 30 wt%, the impact strength of sisal/PP composites was significantly reduced while the impact strength of kenaf/PP was slightly enhanced with either MAPP or WG treatment. However, the untreated 15 wt% sisal/PP composites showed the highest impact strength value (3.9 KJ/m²) while the MAPP-treated kenaf/PP composites showed the lowest value (0.4 KJ/m²). At 15 wt%, a slight increase in the impact strength (3.01 KJ/m²) of MAPP- and WG-treated sisal/PP composites was observed. This behaviour is as a result of the synergic effect of both MAPP and WG on the interlocking bond in the interphase region of composites. The low impact strength could be the results of the weak interfacial bonding between the fibre and the matrix.¹¹ Generally, sisal/PP composites showed better impact strengths when compared to kenaf/PP composites.

Differential scanning calorimetric analysis

The summary of DSC data is presented in Table 4. Figures 6 and 7 show the melting temperature of kenaf and sisal fibre composites. The DSC scan of neat PP matrix showed a melting peak at about 160℃. It is clear from Table 4 that the addition of either kenaf or sisal fibre to PP caused a slight increased in the melting temperature of composites. However, at 15 wt% fibre content, the WG-treated kenaf/PP and kenaf/PP showed high melting temperature values, 185.9℃ and 167.5℃, respectively, when compared to pure PP. The enhancement in the melting temperature of the 15 wt% WG-treated fibres is due to the increased interaction between the fibre and the matrix. When the fibre was increased from 15 wt% to 30 wt%, the melting temperature showed minimal improvement.

Table 4.

Enthalpy heat of fusion, glass transition and melt temperature of sisal composites.

Sample name	Melt temperature, T_m (℃)	Glass transition temperature, T_g (℃)	Enthalpy, ΔH (J/g)
PP	159.8	−17.0	78.3
PP/MAPP	159.8	−11.0	69.1
15% Sisal/PP	162.5	−11.0	62.6
15% Sisal/PP/MAPP	163.6	−12.0	53.1
15% Sisal/PP/WG	167.5	−11.9	42.0
15% Sisal/PP/MAPP/WG	163.6	−12.0	49.0
30% Sisal/PP	164.1	−13.0	54.5
30% Sisal/PP/MAPP	164.8	−17.0	52.9
30% Sisal/PP/WG	164.1	−22.2	52.8
30% Sisal/PP/MAPP/WG	163.9	−21.7	61.9
15% Kenaf/PP	161.1	−12.5	65.2
15% Kenaf /PP/MAPP	162.6	−15.3	71.6
15% Kenaf /PP/WG	185.9	−9.7	69.0
15% Kenaf/PP/MAPP/WG	163.3	−15.4	82.2
30% Kenaf /PP	162.4	−12.1	66.52
30% Kenaf /PP/MAPP	162.6	−11.9	57.6
30% Kenaf /PP/WG	163.6	−8.6	62.7
30% Kenaf /PP/MAPP/WG	164.8	−24.1	54.3

PP: polypropylene; MAPP: maleic anhydride-grafted-polypropylene; WG: water glass

Table 4 and Figure 8 show the glass transition temperature results of kenaf and sisal fibre composites. The glass transition temperature of PP is −17℃. It is clear from Figure 8 that the glass transition temperature of PP was reduced with the addition of fibres. This behaviour is as a result of fibres restricting rotational movement of polymer chains in a melt after nucleation. However, at 30 wt% fibre content, the addition of MAPP and WG resulted in an increase in the glass transition temperature of about 42% for kenaf composites while the WG-treated sisal fibre composites resulted in an increase in the glass transition temperature of about 31%. Figure 9 shows the enthalpy heat of fusion of kenaf and sisal fibre composites. Generally, the PP matrix showed that with the addition of fibre, the heat of fusion of kenaf and sisal fibre composites with or without treatment was significantly reduced when compared to neat PP (78.3 J/g). However, a slight in increase in heat of fusion value (82.2 J/g) of 15 wt% kenaf/MAPP/WG was observed.

Flammability test results

Table 5 shows the average values of the flammability parameters for sisal/PP composites. Compared to PP30S, the ignite time for MAPP-treated (PP30SM) and WG-treated (PP30SW) composite samples was delayed, and longer time to ignition is a very beneficial outcome where fire/flame retardancy is a factor. The combination of MAPP/WG did not enhance the ignition time. The THR was reduced by WG treatment due to the non-flammable polysilicate coating, while the average MLR was slightly reduced by either MAPP or WG (see Table 5). In the composites with MAPP or WG treatment, there was less PP and this could be the reason for the lower THR. These results suggest that the materials (PP30SM and PP30SW) were not fundamentally affected by the thermal degradation of the PP matrix. The average HRR was slightly reduced by the addition of MAPP, while the HRR was further lowered by the WG treatment. The peak of the HRR, which is of utmost importance as it gives an indication of fire growth, was reduced by the addition of MAPP from 755.0 kW/m² to 523.3 kW/m². This can be explained by the strong adhesion between the sisal and matrix, which increases the thermal stability of the composite and hampers the release of volatile degradation products. This effect was more pronounced with the WG treatment; hence, the PHRR value was reduced between 46% and 49% (see Figure 10). This reduction can be explained by the shielding action of the silicate coating on the sisal fibres.¹⁴

Table 5.

Average values of flammability parameters for MAPP- and/or WG-treated sisal composites.

Parameter	Unit	Material
Parameter	Unit	PP30S	PP30SM	PP30SW	PP30SMW
Time to ignition (TTI)	s	19	31	30.3	29.5
Total heat release (THR)	MJ/m²	131.0	126.0	107.9	105.2
Average mass loss rate (MLR)	g/s.m²	0.031	0.029	0.027	0.028
Average heat release rate (HRR)	kW/m²	116.0	114.2	95.1	98.2
Peak heat release rate (PHRR)	kW/m²	755.0	532.3	382	408
Time to reach PHRR	s	82.5	145	80	85
Average heat of combustion (HC)	MJ/kg	36.8	35.3	35.0	34.9
Average CO yield	kg/kg	0.040	0.045	0.051	0.052
Average CO₂ yield	kg/kg	2.38	2.28	2.21	2.26

The shape of the HRR curves for the WG-treated samples in Figure 10 is typical of samples which show an initial increase in HRR until an effective heat barrier is formed. Once this barrier thickens, there is a steady decrease in HRR. This is in contrast to the HRR curve for PP30S which shows a sharp increase in HRR, indicating that the whole sample is pyrolised at about the same time. The time to reach PHRR was increased by the presence of MAPP. This is in line with the above explanation with respect to sisal/HPP adhesion. On the other hand, the WG treatment did not improve the time elapsed for the PHRR which was practically identical with that of the reference composite (PP30S).

Although, the WG-based coating acted as a heat shield even though it did not affect the volatilisation of the degradation products of PP. The average HC of the system studied was fairly the same, as expected. The average CO yield slightly increased with MAPP and WG treatments. However, the CO₂ yield decreased due to the above-mentioned modifications. This means that the ratio of CO₂ to CO was lower for these samples, suggesting more inefficient combustion and prevention of the conversion process of CO to CO₂.

Chapple and Anandjiwala¹² and Helwig and Paukszta¹³ reported that the incorporation of natural fibres in polyolefins is usually accompanied by an enhanced smoke development. The data on the smoke development and release show the following ranking: PP30SM >PP30S >PP30SW ∼PP30SMW. According to this ranking, WG-treated composite samples have better beneficial effect of producing less smoke.

Conclusions

The fibre content (sisal or kenaf) has a great influence on mechanical properties. It was observed that in all fibre/matrix composites, increasing the amount of fibres resulted in increases in tensile strength and Young’s modulus. The impact strength of sisal fibre composites is higher than that of kenaf fibre composites. Generally, the impact strength of fibre/PP with MAPP decreased but increased with the addition of WG. The high impact strength could be the results of the strong interfacial bond strength between the fibre and the PP matrix. A decrease in strain-at-break was observed with an increase in the fibre content. This is due to the close fibre orientation to the surface of the sample. WG increased the melting temperature of kenaf and sisal composites at a given fibre content. WG slightly enhanced the glass transition temperature of kenaf (with MAPP) and sisal composites but reduced the heat of fusion at a given composites.

The addition of WG to the fibre/PP matrix composites showed a positive influence on the flammability properties. Therefore, WG seems to be a promising route to overcome the general problems of high flammability with natural fibre-reinforced thermoplastic composites. The WG has a slight negative influence on mechanical properties of fibre/PP matrix composites. It was found that the WG-coated fibre/PP composites showed excellent thermal properties when compared to the untreated fibre/PP composites. The THR was reduced by WG treatment due to the non-flammable polysilicate coating, while the average MLR was slightly reduced by either MAPP or WG. In the composites with MAPP or WG treatment, there was less PP and this could be the reason for lower THR.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Joseph

Rabello

Mattoso

LHC

. Environmental effects on the degradation behaviour of sisal fibre reinforced polypropylene composites. Compos Sci Technol 2002; 62: 1357–1372.

Zhangb

Ronga

. Self-reinforced melt processable composites of sisal. Compos Sci Technol 2003; 63: 177–186.

Bledzki

Gassan

. Composites reinforced with cellulose based fibres. Prog in Polym Sci 1993; 24: 1083–1093.

Funga

Xinga

Lia

RKY

. An investigation on the processing of sisal fibre reinforced polypropylene composites. Compos Sci Technol 2003; 63: 1255–1258.

Mai

. Sisal fibre and its composites: a review of recent developments. Compos Sci Technol 2000; 60: 2037–2055.

Espert

Vilaplana

Karlsson

. Comparison of water absorption in natural cellulosic fibres from wood and one-year crops in polypropylene composites and its influence on their mechanical properties. Compos Part A: Appl Sci Manuf 2004; 35: 1267–1276.

Khoathane MC. The processing properties of natural fibre reinforced higher α-olefin based thermoplastis. Masters Thesis, Tshwane University of Technology, Pretoria, RSA, 2005.

Alvarez

Zquez

. Thermal degradation of cellulose derivatives/starch blends and sisal fibre biocomposites. Polym Degrad Stab 2004; 84: 13–21.

Patel

. Natural fibres composites: green and ecofriendly. FRP Today 2010; 34: 21–25.

10.

Bledzki

Mamun

Faruk

. Abaca fibre reinforced PP composites and comparison with jute and flax fibre PP composites. Express Polymers Letters 2007; 11: 755–762.

11.

Torres

Cubillas

. Study of the interfacial properties of natural fibre reinforced polyethylene. Polym Test 2005; 24: 694–698.

12.

Chapple

Anandjiwala

. Flammability of natural fiber-reinforced composites and strategies for fire retardancy: a review. J Thermoplast Compos Mater 2010; 23: 871–893.

13.

Helwig

Paukszta

. Flammability of composites based on polypropylene and flax fibers. Mol Cryst Liq Cryst 2000; 354: 373–380.

14.

Medina

Schledjewski

. Water glass as hydrophobic and flame retardant additive for natural fibre reinforced composites. J Nanostruct Polym Nanocompos 2009; 5: 107–114.