Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box-Behnken Design of Experiments

Transcription

1 Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box- Behnken Design of Experiments Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box-Behnken Design of Experiments 1 E.H. Agung, 1 S.M. Sapuan, 2 M.M. Hamdan, 3 H.M.D.K. Zaman, and 4 U. Mustofa 1 Department of Mechanical and Manufacturing Engineering, Universiti Putra Malaysia, UPM Serdang, Selangor, Malaysia 2 Faculty of Engineering, Universiti Pertahanan Nasional Malaysia, Sungai Besi, Kuala Lumpur, Malaysia 3 Radiation Technology, Malaysia Nuclear Agency, Bangi, Malaysia 4 Faculty of Engineering, Universitas Malahayati, Bandar Lampung, Indonesia Received: 8 March 2010, Accepted: 21 December 2010 SUMMARY Mechanical properties of polymer composites are influenced by many factors such as the types of fibres, the types of polymer matrix, the additives used and the adhesion between fibres and polymer matrix. To improve the interfacial adhesion between HIPS matrix and abaca fibres, a study of the optimum use of a coupling agent (MAH) and impact modifier is presented in this paper. Abaca fibre reinforced high impact polystyrene (HIPS) composites were produced with different fibre loadings (30, 40 and 50 wt.%), different compositions of coupling agent, maleic anhydride (MAH) (1, 2 and 3 wt.%) and different compositions of impact-modifier (4, 5 and 6 wt.%). A response surface methodology using Box-Behnken design was used in the design of experiments and analysis of results. Statistical analysis of mechanical properties gave very satisfactory model accuracy, because the coefficient of determinance was for impact strength, for tensile strength, for tensile modulus, for flexural strength, and for flexural modulus. In this study, a loading of abaca fibre of wt.%, maleic anhydride 3 wt.%, and impact modifier 4 wt.% led to optimum individual impact strength. On the other hand, optimum individual tensile strength and tensile modulus were achieved when the loading of abaca fibre was close to wt.%, maleic anhydride 3 wt.% and impact modifier 6 wt.%, but the optimum individual flexural strength and flexural modulus were found when the loading abaca fibre was close to wt.%, maleic anhydride 3 wt.% and impact modifier 4 wt.%. 1. INTRODUCTION Natural fibres have attracted the attention of scientists and technologists because of cost and availability advantages offered over conventional reinforcement materials, and the development of natural fibre composites has been a subject of interest for research. Natural fibres are low cost fibres with low density and high specific properties. The specific properties of some natural fibres are suitable to produce commercial composite products. But, there are some difficulties in dealing with natural Smithers Rapra Technology, 2011 fibres such as natural fibres have large variation in mechanical properties, absorb moisture and have poor thermal characteristics. Therefore, some problems have to be solved for the successful application of natural fibres in composites. One of the issues in natural fibres is how to mix them with composites. The problem of fibrematrix adhesion has been the topic of research interest in the recent years 1. Natural fibres like jute, flax, hemp, coir and sisal proved to be good reinforcement in thermoset and thermoplastic matrices and are being used in automotive applications, construction as well as in packaging industries with few drawbacks 2 6. Abaca fibre, (abaca is a special type of banana which does not bear fruits), is a cellulosic fibre obtained from the pseudo-stem of a banana plant (Musa sepientum) and is produced in the form of a bast fibre 7. In tropical countries, agricultural plants like banana plants are available in abundance. Banana fibre is a waste product of banana cultivation and it can be used without any further investment. The banana fibres can be used for industrial purposes. Abaca is among natural fibres that meet the stringent quality requirements for components used in the exterior of road vehicles, especially regarding resistance Polymers & Polymer Composites, Vol. 19, No. 8,

2 E.H. Agung, S.M. Sapuan, M.M. Hamdan, H.M.D.K. Zaman, and U. Mustofa against stone strike and exposure to damp environments. Pothan et al reported on the dynamic mechanical properties, effect of hybridization and chemical modification of abaca fibre-reinforced polyester composites produced by a compression moulding process. The equilibrium water uptake and diffusion coefficient were found to be minimal for treated fibre composites. Shibata et al. 12,13 investigated the performance of treated and non treated abaca fibre reinforced biodegradable polyester composites using an injection moulding process. Improvement of the mechanical properties of abaca fibre-reinforced composites can be achieved by means of fibre treatment and the use of an appropriate polymer. Bledzki et al. 14 studied the effect of fibre length and compounding processes on mechanical properties of abaca fibrereinforced polypropylene composites. Abaca fibre-reinforced polypropylene composite has remarkable features such as low cost, availability, high specific flexural and tensile strength, good abrasion, acoustic resistance, good resistance to rot and UV rays 15 and can be used in the automotive industry. Statistical methods such as regression analysis can be used to analyze the relationship between the properties of some natural fibre-reinforced polymer composites. Using data collected from a set of experimental trials, regression helps to establish empirically the type of relationship that is present between a variable and its influencing factors. The effects present maximum or minimum conditions of natural fibres reinforced polymer composites. An extended model is needed to observe and to investigate relationship among factors or variables and also to find optimization technique. One of the methods is response surface methodology (RSM) and it is used in this study. Setting up a series of experiments of natural fibre-reinforced polymer composites yields adequate and reliable measurements of the response of interest. This paper discusses the research with theoretical and application study of design of experiment. In the conventional multifactor experiments, optimization is usually carried out by varying a single factor while making all other factors fixed at a specific set of condition. The solving of combined interactions among the factors and finding the optimization are used in the design of experiments with response surface methodology. The definition of the RSM is a collection of mathematical and statistical techniques for designing experiments, building models, evaluating the effects of factors and searching optimum condition of factors for desirable responses 16. The optimization process involves studying the response of the statistically designed combinations, estimating the coefficients by finding a mathematical model that fits best the experimental conditions, predicting the response of the fitted model and checking the adequacy of the model. The most common designs include central composite design (CCD) and Box- Behnken design (BBD) of the principal response surface methodology and they have been widely used in various experiments In the present work, Box-Behnken design is used. The Box-Behnken design requires only three levels of each process factor and only a fraction of all the possible combinations as described on Figure 1. By avoiding the corners of the design research, they allow N experiments to generate a Box- Figure 1. A Box-Behnken design for three factors (n=3) Behnken design for N factors. N must be an integer, 3 or larger. The Box-Behnken design does not contain combinations for which all factors are simultaneously at their highest or lowest level. Therefore, designs are useful in avoiding experiments performed under extreme conditions, for which unsatisfactory results might occur. The Box-Behnken design requires a number of experiments according to N = n 2 + n + c p, where n is the factor number and c p the replicate number of the central point. Such design is formed by combining 2 n factorials with an incomplete block design. Box-Behnken, a spherical and revolving design, has been applied in optimization of chemical and physical processes because of its reasoning approach and excellent outcomes. With this methodology, the empirical model expresses the relationship between independent variables and response variables, and independent variable values that cause the response to become optimum could be determined. 2. EXPERIMENTAL 2.1 Materials The abaca plant is a type of banana plant which is native of the Philippines and it is grown in moderate humidity areas in Indonesia. Abaca fibres used in this study were obtained from Pekalongan, Central Java, Indonesia and were produced by Ridaka Hand Craft. The matrix used for this study was high impact polystyrene (HIPS) Idemitsu 698 Polymers & Polymer Composites, Vol. 19, No. 8, 2011

3 Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box- Behnken Design of Experiments PS HT 50, a product of Petrochemical (M) Sdn. Bhd, Malaysia. It had a density and melt index of 1.04 g/cm 3 and 4.0 g/10 min respectively. Figure 2. The production of abaca fibre reinforced HIPS composites by rolling machine The coupling agent used in this study was maleic anhydride, MAH, (polystyrene-block-poly(ethylene-ranbutylene)-block-polystyrene-graftmaleic anhydride), a product of Sigma Aldrich Inc., Germany. The supplier was Sigma Aldrich Malaysia (M) Sdn. Bhd, Malaysia. The impact modifier was a styrene butadiene styrene (SBS) copolymer rubber (Cyclo resin). The cylindrical granules of cyclo resin were produced by Multiversum, Germany and supplied by PT. Wahana Makmur Kencana, Jakarta - Indonesia. 2.2 Sample Preparation The abaca fibres were dried under sunlight between 27 and 30 C for four days. The dry abaca fibres were cut to 2 3 mm by means of an electronic cutting machine. The matrix, high impact polystyrene (HIPS), maleic anhydride (MAH), cyclo resin and abaca fibres were measured based on the design of experiment and they were classified into three levels (high [+], intermediate [0], and low [-]). The processing of abaca fibre reinforced HIPS composites was accomplished using a rolling machine, as shown in Figure 2. The working temperature of the rolling machine was kept at approximately 200 C. The composites were produced by manually dropping the matrix and fibre into the rolling machine at a very slow rate. The process was continued until all the materials were well mixed. The composites produced were brown in colour following the natural colour of abaca fibres. Sheets of abaca fibrereinforced HIPS composites produced had an average thickness of 1 mm. The composite materials produced were then crushed and pressed to thicknesses of 1, 2 and 3 mm using a hot press. Standard specimens were prepared for the determination of tensile, flexural and impact properties Measurements Tensile Testing Tensile testing was carried out according to ASTM D using a universal testing machine, Instron, model 556 at ambient temperature (27 C). The strain rate was 50 mm/ min with a gauge length of 60 mm. The values reported are the average of three samples tested Flexural Testing Flexural testing was carried out using universal testing machine, Instron, Model 556 at ambient temperature (27 C) according to ASTM D The support span was fixed at 100 mm with a crosshead speed of 3 mm/min. The values reported are the average of three samples tested Impact Testing Notched Izod impact testing was carried out according to ASTM D a using a Toyoseiki pendulum impact testing machine at ambient temperature (27 C). The impact specimens were notched (45 ) to a depth of 2.6 mm. The reported values are the average of three samples tested. 3. DESIGN OF EXPERIMENTS This research used the Box-Behnken design to analyze the mechanical properties of abaca fibre-reinforced HIPS composites, namely impact strength, tensile strength, tensile modulus, flexural strength and flexural modulus. The responses of three factors were to be determined and these three factors were designated as X 1 (abaca fibre), X 2 (maleic anhydride (MAH)) and X 3 (impact modifier (IM)). They were prescribed into three levels, coded as +1, 0, 1 for high, intermediate and low values, respectively as shown in Table 1. Three test variables were coded according to equation (1): ; i = 1,2,3 (1) where x i is the coded value of an independent variable; X i is the actual value of an independent variable; X 0 is the actual value of an independent variable at the centre point; and ΔX is the change value of an independent variable. The Box-Behnken design consists of a set of points lying at the midpoint of each edge and the replicated centre point of the multidimensional cube. All experiments were performed in triplicate and the average values of the mechanical properties were taken as responses. For predicting the optimal point, a second-order polynomial model was fitted to correlate the relationship between the independent variables and responses (mechanical properties). Polymers & Polymer Composites, Vol. 19, No. 8,

4 E.H. Agung, S.M. Sapuan, M.M. Hamdan, H.M.D.K. Zaman, and U. Mustofa Table 1. Levels and code of variables for Box Benhken design For the three factors, the equation is: where y is the predicted response; b 0 is model constant; x 1 ; x 2 and ; x 3 are independent variables; b 1 ; b 2 and b 3 are linear coefficients; b 12 ; b 13 and b 23 are cross-product coefficients; and b 11 ; b 22 and b 33 are the quadratic coefficients. The quality of fit of the polynomial model equation was expressed by the coefficient of determination R 2. (2) 4. Optimization of mechanical properties of abaca fibre-reinforced HIPS composites with desirability function Generally, optimization in industry is needed involving multiple response processes. The desirability function approach is the most widely used method to solve such problems. It is clear that the idea that the quality of a product or processes has multiple quality characteristics, with one of them outside of some desired limits, is completely unacceptable 21. Using equation 3, the method finds the operating conditions x that provide the most desirable response values. For each response Yi(x), a desirability function di(yi) assigns numbers between 0 and 1 to the possible values of Yi, with di(yi) = 0 representing a completely undesirable value of Yi and di(yi) = 1 representing a completely desirable or ideal response value. The individual desirabilities are then combined using the geometric mean, which gives the overall desirability (D) with n denoting the number of responses. For an n responses system, the overall performance of the system is determined by the composite desirability (D) 22. 1/n (3) By using the DOE (Design of Expert) software based on the Box-Behnken design in this research, the concept of composite desirability was applied to find the optimal settings of control factors (loadings of abaca fibres, maleic anhydride and impact modifier). 5. Results and discussion 5.1 Mechanical Properties of Abaca Fibre Reinforced HIPS Composites A three-factor, three-coded level Box-Behnken design was used to determine the responses, i.e. the mechanical properties for different percentages of abaca fibres, maleic anhydride (MAH) and impact modifier (IM). The experiments were carried out based on the design matrix given in Table 2. In order to ensure and decide about the adequacy of the model for Abaca fibre-reinforced HIPS composites, tests for significance of the individual model coefficients and tests for lack-of-fit need to be performed. The ANOVA Tables 3 7, describe the response surface of the quadratic model for mechanical properties of abaca-reinforced HIPS composites. The value of Prob>F in Tables 3-7 for the model is less than 0.05, which indicates that the model is significant. The impact strength of abaca fibre-reinforced HIPS composites, the ratio of abaca fibre (A), two-level interaction of abaca fibre (A 2 ), maleic anhydride (B 2 ) and impact modifier (C 2 ), and also the abaca fibre to maleic anhydride ratios (AB), the abaca fibre to impact modifier ratios (AC), the maleic anhydride to impact modifier ratios (BC) were significant. Other model terms could be said to be insignificant. In a similar manner, the effects of tensile strength, tensile modulus, flexural strength 700 Polymers & Polymer Composites, Vol. 19, No. 8, 2011

5 Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box- Behnken Design of Experiments Table 2. The Box-Behnken design matrix employed for three independent variables along with the observed values and flexural modulus explained in Table 3-7 were based on the each of the values Prob>F. The ANOVA analysis indicated a linear relationship between the main effects of abaca fibre wt.%, maleic anhydride wt.%, impact modifier wt.%, and quadratic relationship with factors of abaca fibre wt.%, maleic anhydride wt.%, impact modifier wt.%. The independent variables and their levels for the Box- Behnken design used in this study are shown in Table 1. Using those relationships, the actual levels of variables for each experiment in the design matrix were calculated, and the results obtained are given in Table 2. The final mathematical equations based on coded factors given in equations (4)-(8) were obtained after the analysis of variance and they gave the levels of weight percentages of abaca fibres, maleic anhydride (MAH) and impact modifier (IM) as determined by Design-expert software. For impact strength, the model equation was: Table 3. ANOVA table (partial sum of squares) for quadratic model (response: impact strength) (4) For tensile strength, the model equation was: (5) For tensile modulus, the model equation was: Table 4. ANOVA table (partial sum of squares) for quadratic model (response: tensile strength) For flexural strength, the model equation was: (6) (7) Finally for flexural modulus, the model equation was: (8) Polymers & Polymer Composites, Vol. 19, No. 8,

6 E.H. Agung, S.M. Sapuan, M.M. Hamdan, H.M.D.K. Zaman, and U. Mustofa Table 5. ANOVA table (partial sum of squares) for quadratic model (response: tensile modulus) This model can be used to predict the mechanical properties of abaca fibre-reinforced HIPS composites within the limits of the experiment. Figures 2-6 show the relationship between the actual (experiments) and predicted values of abaca fibre-reinforced HIPS composites for mechanical properties response. It is seen from Figures 3-7 that the developed models were adequate. The residuals for the predictions of each response are minimum, since the residuals tend to be close to the diagonal line. The R 2 value for impact strength is , tensile strength is , tensile modulus is , flexural strength is and flexural modulus is ; these values are close to 1, which is desirable. All the predicted R 2 values for the mechanical properties of abaca-reinforced HIPS composites are in agreement with the adjusted R 2 (see Tables 3-7). The adequate precision value for the mechanical properties of abaca fibre-reinforced HIPS composites is well above 4. Table 6. ANOVA table (partial sum of squares) for quadratic model (response: flexural strength) Figure 3. Relation between experimental and predicted impact strength (J/m) Table 7. ANOVA table (partial sum of squares) for quadratic model (response: flexural modulus) Figure 4. Relation between experimental and predicted tensile strength (MPa) 702 Polymers & Polymer Composites, Vol. 19, No. 8, 2011

7 Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box- Behnken Design of Experiments Figure 5. Relation between experimental and predicted tensile modulus (GPa) Figure 6. Relation between experimental and predicted flexural strength (MPa) From the results listed in Table 2 and the regression coefficients for the response variables in the experimental design according to equation (3)-(7) (Table 8), the response functions representing mechanical properties were expressed as a function of abaca fibre wt.% (b 1 ), maleic anhydride wt.% (b 2 ) and impact modifier wt.% (b 3 ). The statistical significances of each effect of abaca fibre reinforced HIPS composites are shown in Table 8. Figure 7. Relation between experimental and predicted flexural modulus (GPa) The results are that the values of mechanical properties are strongly affected by the variables selected in the study. This is also reflected in the wide range of values for each coefficient of the terms of equations (4)-(8). The main effects of X 1, X 2, and X 3 represent the average results of changing 1 (one) variable at a time from its low level to its high level. The interaction terms (X 1 X 2, X 1 X 3, X 2 X 3, X 12, Table 8. Regression coefficients of approximate polynomials for response variables in experimental design Polymers & Polymer Composites, Vol. 19, No. 8,

8 E.H. Agung, S.M. Sapuan, M.M. Hamdan, H.M.D.K. Zaman, and U. Mustofa X 22, and X 32 ) show how the mechanical properties of abaca fibre reinforced HIPS composites change when 2 (two) variables are simultaneously changed. The positive coefficients for all 3 (three) independent variables indicate a favourable effect on the mechanical properties. The negative coefficients among 3 independent variables indicate a partitioning of the favourable effect on mechanical properties. Figure 8a. Response surface 3 D plots showing the effect of abaca fibre and maleic anhydride (MAH) on the impact strength (J/m) This is shown in the mechanical properties of abaca fibre-reinforced HIPS composites. The negative coefficients for the main effect indicate that the impact modifier represented an unfavorable effect for abaca fibrereinforced HIPS composites within the range of the research analysed. In this study, only the main effect of maleic anhydride concentration represented favourable results on the impact strength. According to the impact strength, tensile strength and tensile modulus, the main effect of abaca fibre represented favourable resulting. This research indicated that flexural strength and flexural modulus did not represent favourable effects to the result. Figure 8b. Response surface 3 D plots showing the effect of abaca fibre and impact modifier (IM) on the impact strength (J/m) The interactions between 2 variables (X 1 X 2, X 1 X 3, and X 2 X 3 ) indicate unfavourable effects. This is shown in the impact strength, flexural strength, and flexural modulus of abaca fibrereinforced HIPS composites. On the other hand, positive coefficients for the interactions between 2 variables indicate favourable effects on the mechanical properties. Among the 3 independent variables of mechanical properties, the lowest coefficient value is for X 3 (β 3 = and Prob> 0.05), which indicates that this variable is insignificant in the prediction of abaca fibre-reinforced HIPS composites. Figure 8c. Response surface 3 D plots showing the effect of impact modifier (IM) and maleic anhydride (MAH) on the impact strength (J/m) Impact Properties High-impact polystyrene (HIPS) is one of the well-known toughened polymers. The high toughness is given by the rubbery phase. It is noticed that one of the main factors affecting the 704 Polymers & Polymer Composites, Vol. 19, No. 8, 2011

9 Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box- Behnken Design of Experiments impact strength and toughness of HIPS is the rubber-phase particle size and its size distribution When abaca fibres were used as a filler, the impact strength of abaca fibre-reinforced HIPS composites was affected by two variables (maleic anhydride and impact modifier) in combination. Response surface diagrams of the regression equations for impact strength show two opposing effects of the coupling agent (maleic anhydride) and impact modifier when the abaca fibre loading was increased from its experimental minimum to the maximum. In lightly filled composite, a marginal increase in impact strength was observed with an increase in coupling agent of 3 wt.% and impact modifier 4 wt.%. On the other hand, a decrease in impact strength was seen with an increase in the coupling agent of 3 wt.% and impact modifier 6 wt.%. The minimum impact strength was observed for MAH 1 wt.% and IM 4.01 wt.%. To further enhance our understanding of the effect of response variables on abaca fibre-reinforced HIPS composites, the impact strengths were measured and the data were processed by response surface methodology. The effects of the response variables of impact strength are demonstrated by the response surface plots in Figures 8a, 8b, and 8c. The corresponding second-order response was found after ANOVA of the measured value in the regression equation, and the model is presented in equation 1 and listed in Table Tensile Strength and Tensile Modulus The relationship between dependent and independent variables of tensile strength and tensile modulus were studied in the experimental design shown in Table 8, and the effect of the response variables was determined using the model in equations (4)-(8). By 3 D graph plots, the effects of each variable and their interactions on abaca fibre-reinforced HIPS composites at one fixed level of variable (medium level) are shown in Figures 9a and 9b. Figure 9a. Response surface 3D plots showing the effect of abaca fibre and impact modifier (IM) on the tensile strength (MPa) Figure 9b. Response surface 3D plots showing the effect of impact modifier (IM) and maleic anhydride (MAH) on the tensile strength (MPa) Figure 9a shows the tensile strength of abaca fibre-reinforced HIPS composites to be a saddle-shaped curve and indicates two high values (maximum of up to MPa) at abaca loadings close to 45 wt.% and 30 wt.%, the graph of tensile strength starts to decrease and then increase when the loading of abaca fibre is close to 45 wt.% and 30 wt.%. In this condition, the impact modifier loadings are close to 4.5 wt.% and 5.5 wt.%. Figure 9b indicates that the tensile strength of abaca fibre-reinforced HIPS composite passes through a minimum point as a function of impact modifier and maleic Anhydride loadings, while the abaca fiber loading is level (40 wt.%). The case of tensile modulus was studied with regard to the interaction effects (see Table 8), the abaca fibre and impact modifier effects showing the saddle point graph shown in Figure 10a. Figure 10b indicates a minimum value of tensile modulus (reflecting interaction between the impact modifier and Maleic Anhydride) around 1.16 GPa. Upon studying the coefficient values for tensile strength Polymers & Polymer Composites, Vol. 19, No. 8,

10 E.H. Agung, S.M. Sapuan, M.M. Hamdan, H.M.D.K. Zaman, and U. Mustofa and tensile modulus, the variable X 1 (β 1 = and ) was found to be higher than the variables X 1 and X 3, indicating that it contributes the most in predicting the properties of abaca fibre reinforced HIPS composites. Interactions between X 1 and X 3 provide a more real impact for the yields. It means that abaca fibres were highly influenced by impact modifier. Figure 10a. Response surface 3D plots showing the effect of impact modifier (IM) and abaca fibre on the tensile modulus (GPa) Flexural Strength and Flexural Modulus The statistical analysis of the design shows a high precision of the polynomial model that reflects a high degree of the effects of each variable. The value of correlation (R 2 ) of equation for flexural strength and flexural modulus were found to be and respectively (see Table 8), indicating good fit. Among the dependent variables selected and their interactions, a variable X 2 X 3 for flexural strength had a negative value. This indicates that the variable made a minimal contribution to the interaction of abaca fibre-reinforced HIPS composites. Figures 11a and 11b respectively show the flexural strength and flexural modulus against the loadings of abaca fibres and coupling agent (MAH). It may be observed from the plots that at any given abaca fibre wt.%, the flexural strength and flexural modulus increase with the concentration of the coupling agent (MAH). The graphs for this condition (Figure 11a and 11b) are curvilinear. The positive value of a variable X 1 X 3 (see Table 8), cannot show the interaction effect caused the value of Prob> F more than Figure 10b. Response surface 3D plots showing the effect of impact modifier (IM) and maleic anhydride (MAH) on the tensile modulus GPa Figure 11a. Response surface 3D plots showing the effect of abaca fibres and maleic anhydride (MAH) on the flexural strength (MPa) However, the coupling agent (maleic anhydride) used in this study of the effects on flexural strength and flexural modulus, showed an increase of these properties with the loading of abaca fibre wt.%. The reason for this may have been due to the functionalisation of the coupling agent that improved adhesion between the reinforcing abaca fibres and the matrix. 706 Polymers & Polymer Composites, Vol. 19, No. 8, 2011

11 Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box- Behnken Design of Experiments Figure 11b. Response surface contour plots showing the effect of abaca fibres and maleic anhydride (MAH) on the flexural modulus (GPa) levels of abaca fibre wt.%, maleic anhydride 3 wt.% and minimum level of impact modifier 4 wt.%. The individual response of both the tensile properties (strength and modulus) were predicted with wt.% of loading abaca fibres, 3 wt.% maleic anhydride and a maximum value of impact modifier (6 wt.%). Tensile strength was MPa and tensile modulus 1.52 GPa. From the individual optimised response of flexural properties, the flexural strength was MPa and flexural modulus 4.58 GPa with a composition of abaca fibre wt.%, maleic anhydride 3 wt.% and impact modifier 4 wt.%. 5.2 Optimization of the Experiments Since the interactions were found to be present, the next step was to optimize these interactions so as to obtain maximum mechanical properties and deal with a minimum amount of chemicals added. In order to achieve this, the Design Expert software was used. The optimized properties were based on the models shown in equations (4)-(8). Some solutions for the mechanical properties of abaca fibre-reinforced HIPS composites are shown in Table 9. The performances of the mechanical properties abaca fibre reinforced HIPS composites could not be explained in a single linear relationship. The reason for the increasing desirability factor is because the interface modifiers improved the interfacial adhesion between the fibres and matrix. Therefore, there was better transfer of stress from the matrix to the fibres, leading to the improved tensile and flexural properties 23. According to the several responses evaluated in this experimental design, the optimum points reached individually for each factor did not coincide in all cases. The improved interfacial adhesion was expected to decrease the impact strength. In this study, for the individual response of the optimum impact strength was J/m with optimum To account for this situation, it was necessary to identify a compromise zone where all the experimental responses satisfied the specifications to achieve the proposed aims. In this study, interactions among the abaca fibre, MAH and IM were combined to find the mechanical properties which showed maximum sensitivity. Moreover, to choose the best coordinates of an acceptable compromise, the desirability function was taken into account. The acceptable values of desirability function were those close to one (100%). In this research, mechanical properties (impact strength J/m, tensile strength MPa, tensile modulus 1.31 GPa, flexural strength MPa, Table 9. Optimization experiments of mechanical properties prepared by Box-Behnken design Polymers & Polymer Composites, Vol. 19, No. 8,

12 E.H. Agung, S.M. Sapuan, M.M. Hamdan, H.M.D.K. Zaman, and U. Mustofa and flexural modulus 4.17 GPa) were predicted for abaca fibre-reinforced HIPS composite that compromised on loading of abaca fibre at wt.%, maleic anhydride 3 wt.% and impact modifier 4 wt.%, this condition resulting in 77.37% desirability. The individual response can be transferred to the application of the material depending on the preferred mechanical properties. The stress transfer between matrix and fibres in composites is not only determined by intrinsic properties of the fibres and matrix, but also by the geometric parameters and fibre arrangement within matrix, such as fibre distribution 24. It can be observed from the SEM micrographs in Figure 12a that relatively excessive amounts of fibres lie against one another rather than being mixed with maximum composition, abaca fibre 50 wt.%, maleic anhydride 3 wt.%, and impact modifier 6 wt.% (Figure 12b) and a relatively small amount of fibres and fractured fibres were observed at minimum composition, abaca fibre 30 wt.%, maleic anhydride 1 wt.%, and impact modifier 4 wt.% (Figure 12b). Due to the addition of MAH and impact modifier, improved fibre matrix adhesion and relatively few fibre fractures were observed for both fibre loadings as demonstrated by SEM micrographs. The addition of MAH and the impact modifier improved the performance of fibre-matrix adhesion. Correlations among the factors were shown in the composition of composites which were influenced by the addition of MAH and impact modifier. 6. Conclusions A Box-Benhken design of experiments was performed based on response surface methodology (RSM) which enabled the determination of the independent or the main effects, the interaction effects and the optimum response for abaca fibre-reinforced Figure 12a. SEM micrographs of abaca fibres-hips composites (abaca fibre 50 wt.%, MAH 3 wt.%, IM 6 wt.%) Figure 12b. SEM micrographs of abaca fibres-hips composites (abaca fibre 30 wt.%, MAH 1 wt.%, IM 4 wt.%) HIPS composites. The addition of MAH and impact modifier improved the tensile and flexural properties of the composites by enhancing the adhesion between the fibres and HIPS. The optimum amount of MAH is 3% and impact modifier is 4% which gave the best individual impact strength with loading of abaca fibre close to wt.%. The optimum individual tensile properties needed abaca fibre wt.%, MAH 3 wt.% and IM 6 wt.%. On the other hand, optimum individual flexural properties were found with abaca fibre wt.%, MAH 3 wt.% and IM 4 wt.%. The mechanical properties where the material compromised on loadings of abaca fibre wt.%, maleic anhydride 3 wt.% and impact modifier 4 wt.%, were found to be: impact strength J/m, tensile strength MPa, tensile modulus 1.31 GPa, flexural strength MPa and flexural modulus 4.17 GPa. The conditions were associated with 77.37% desirability. Acknowledgements The authors are thankful to Universitas Malahayati, Lampung, Indonesia for financial support. Acknowledgements are also due to PT. Tara Plastic Indonesia and Radiation Technology, Malaysia Nuclear Agency, Bangi, Malaysia for providing and testing materials for this study. References 1. Saira T, Munawar A.M. and Shaifullah K., Review. Natural Fiber-Reinforced Polymer Composites. Proc. Pakistan Acad. Sci. 44(2) (2007) Mohanty A.K., Misra M. and Hinrichsen G., Biofibres, 708 Polymers & Polymer Composites, Vol. 19, No. 8, 2011

13 Optimization of the Mechanical Properties of Abaca Fibre-Reinforced High Impact Polystyrene (HIPS) Composites Using Box- Behnken Design of Experiments biodegradable polymers and biocomposites. An overview. Macromolecular Materials and Engineering, 1 24 (2000) Mohanty A.K., Misra M. and Drzal L.T., Sustainable bio-composites from renewable resources: Opportunity and challenges in the green materials world. Journal of Polymers and the Environment, 10 (2002) Maguro A., Vegetable fibress in automotive interior components. Die Angewandte Makromolekulare Chemie, 272 (1999) Gross R.A., and Karla B., Biodegradable polymers for the environment. Science, 297 (2002) Puglia D., Biagiotti J. and Kenny L.M., A review on natural fibres-based compositespart II: Application of natural reinforcements in composite materials for automotive industry. Journal of Natural Fibres, 1 (2005) Joseph S., Sreekala M.S., Oommen Z., Koshy P. and Thomas S., A comparison of the mechanical properties of phenol formaldehyde composites reinforced with banana and glass fibres. Composites Science and Technology, 62 (2002) Pothan L.A., Oommen Z. and Thomas S., Dynamic mechanical analysis of banana fibres reinforced polyester composites. Composites Science and Technology, 63 (2003) Pothan L.A. and Thomas S., Polarity parameters and dynamic mechanical behaviour of chemically modified banana fibres reinforced polyester composites. Composites Science and Technology, 63 (2003) Pothan L.A., Neelakantan N.R., Rao B. and Thomas S., Stress relaxation behavior of banana fibresreinforced polyester composites. Journal of Reinforced Plastics and Composites, 23 (2004) Pothan L.A., and Thomas S., Effect of hybridization and chemical modification on the water-absorption behavior of banana fibres-reinforced polyester composites. Journal of Applied Polymer Science, 91 (2004) Shibata M., Takachiyo K-I., Ozawa K., Yosomiya R., and Takeishi H., Biodegradable polyester composites reinforced with short abaca fibres. Journal of Applied Polymer Science, 85 (2002) Shibata M., Ozawa K., Teramoto N., Yosomiya R., and Takeishi H., Biocomposites made from short abaca fibres and biodegradable polyesters. Macromolecular Materials and Engineering, 288 (2003) Bledzki A.K., Faruk O., and Mamun A.A., Abaca fibres reinforced PP Composites: Influence of fibres length and compounding processes on the mechanical properties. Polimery, 53 (2008) Proemper E., New automotive interior parts from natural fibres materials. 7th International AVK- TV Conference, Baden-Baden, Germany pb-8 (2004). 16. Box G.E.P., Hunter W.G. and Hunter J.S., Statistics for Experimenters. (John Wiley & Sons, Inc) New York, pp (1978). 17. Dean A.M. and Voss D.T., Design and Analysis of Experiments. (Springer-Verlag, Inc, New York), pp (1999). 18. Oscar S.C., Gerardo S.C., Jose L.P.H., Mariano G.R. and Ernesto F.T., Effect of substrate composition on the mycelia growth of Pleurotus ostreatus: an analysis by mixture and response surface methodologies. Process Biochemistry, 35 (1999) Qiu L.X. and Chen L.M., Application of second-order regression model in optimization of pharmacy techniques. Chinese Journal of Health Statistics, 16 (1999) Muthukumar M., Mohan D. and Rajendran M., Optimization of mix proportions of mineral aggregates using Box-Behnken design of experiments. Cement and Concrete Composites, 25 (2003) Ryan Thomas P., Modern Experimental Design, John Wiley & Sons, Inc., Hoboken, New Jersey (2007). 22. Anderson M.J., Whitcomb P.J., RSM Simplified Optimizing Processes Using Response Surface Methods for Design of Experiments, Productivity, Inc., New York NY (2005). 23. Shah B.L., Matuana L.M. and Heiden P.A. Novel Coupling Agents for PVC-wood-flour Composites, Journal of Vinyl & Additive Technology, 11(4) (2005) Pan N., Theoretical determination of the optimal fibres volume fraction and fibres-matrix property compatibility of short fibres composites. Polymer Composites, 14 (1993) Polymers & Polymer Composites, Vol. 19, No. 8,

14 E.H. Agung, S.M. Sapuan, M.M. Hamdan, H.M.D.K. Zaman, and U. Mustofa 710 Polymers & Polymer Composites, Vol. 19, No. 8, 2011