Effectiveness of pista shell filler content and injection molding parameters on characteristics of PLA based green composites

Abstract

This research explores the application of pistachio shell powder (PSP), an agro-waste product, as a bio-filler for polylactic acid (PLA) for the development of sustainable green composites using injection molding. The study utilizes Box-Behnken Design (BBD) to optimize prominent processing parameters, and the properties of the resultant composites are explored through Response Surface Methodology (RSM). The results indicate that the addition of PSP greatly affects the mechanical characteristics of PLA composites, with increasing percentages of fillers contributing to the enhanced tensile and flexural moduli but decreasing elongation and strength. Confirmation of the significant effects of PSP and machining conditions on composite properties is found in ANOVA results, which indicate the suitability of agro-waste fillers to improve the performance of green composites. These results set avenues for applications in biodegradable packaging material and ecologically friendly materials but require further investigation to tackle the batch to batch variability across seasons and regions.

Introduction

In the era of increased environmental awareness and necessity, seeking innovative solutions for waste reduction and low impact living is more vital than ever. With the world population booming now and industrial operations being conducted at an alarming faster pace, the need to deal with the indiscriminate accumulation of domestic waste is now apparent. But within this challenge is an opportunity - the conversion of agro-waste into high value green composites. The need to move toward a more eco-friendly and circular economy has sparked a lot of interest in creating green materials. ¹ People know composite materials for their outstanding physical qualities, but they have depended on man-made reinforcements. Yet, the environmental effects of these materials have led to a change in direction, with a focus on natural fibres/fillers, since they possess some important characteristics to make them good candidate substitute materials in light duty automotive parts.^2,3 This makes them perfect candidates for desired low-cost lightweight structural applications. Consequently, this is why the automotive industry has since turned to these eco-friendly materials for a growing proportion of car components in place of traditional glass fibre composites like door panels; interior panels and even whole body parts. Composites may either use thermoplastic or they can be a mixture of both, thermoplastics and thermosetting polymers, thereby allowing the core function utilizing as much composite material needed to achieve the desired bottom end frankness without running ancillary binders. While thermoplastics (polypropylene, polyethylene, polyester and polylactic acid) have some advantages over the thermosets. ⁴ First, you can recycle thermoplastics as much again at the end of life or some even bio-degrade/compost. In addition, you have a variety of thermoplastics to choose from; each with their own heat resistant applications, strength or other physical qualities. This, of course, allows to tailor composite materials to specific application performance requirements. Researches incorporated natural fibres into polymer matrixes and studying the various parameters, including fibers sequence and fibre orientation, which have an effect on the characteristics of the composites.^5–7 While natural fibres and fillers find applications in composite materials for property improvement, they have different functionalities. Natural fillers are materials used to fill the matrix of composites; normally, they are added to reduce the overall material cost or attain other property improvements/modifications in the final fabricated composites.^8,9

For instance, the addition of silica nanoparticles at different weight percentages: 0 wt%, 3 wt%, 6 wt%, and 9 wt% into sisal/hemp/epoxy composite resulted in decrease void content and improved mechanical properties, and wear resistance. The induced filler particles into the composites replaces the voids by filling up their space thereby enhancing the properties. ¹⁰ In a separate study, the incorporation of mango shell particles into jute/flax/epoxy composites led to significant improvements in both the mechanical properties and wear resistance of the composites. This enhancement demonstrates the effectiveness of filler material as reinforcement in natural fiber-based composites. ¹¹ The study investigated on the impact of introducing discarded groundnut husk filler with varying filler content on the mechanical characteristics of the composite. It was shown that adding up to 40% more filler benefited the mechanical characteristics that including tensile and flexural strength. Furthermore, the composite is more economically feasible and encourages environmental sustainability by reducing waste by utilising discarded components. ¹²

Ramesh et al. (2025) prepared PLA based composites with pomegranate peel filler (5–20 wt%). The 10 wt% filler composite possessed the optimum mechanical performance, with 22% improvement in tensile strength and a 31% rise in tensile modulus. Flexural strength also improved by 28%, and impact strength reached a maximum of 18% above neat PLA. Above 15 wt% filler loading, agglomeration resulted in small losses in strength. Thermal stability was enhanced, and the onset degradation temperature was changed by + 14°C at 10 wt% filler. There was a small increase in water absorption (by ∼6% at 20 wt%), but density was nearly the same, validating lightweight potential for green applications. ¹³ Joseph et al. (2024) employed seashell and fish scale powders (2.5–10 wt%) as fillers in bio-epoxy resin. 5 wt% seashell filler composite showed improvement in tensile strength by 18.6% and flexural strength by 17.8%. Impact strength was enhanced by 24% at 5 wt% filler, while hardness was enhanced by 12%. Thermal analysis revealed the enhancement in glass transition temperature from 91°C to 101°C and thermal stability with a maximum decomposition temperature 22°C above neat resin. Above 7.5 wt% loading, filler agglomeration resulted in decreasing mechanical performance. ¹⁴ Chandran et al. (2024) investigated waste chicken feather filler (2.5, 5, and 7.5 wt%) in a bio-epoxy matrix. The 2.5 wt% composite had the best properties, where tensile strength and flexural strength improved by 15%) and 20%, respectively. Impact resistance also improved remarkably by 27%, and Shore-D hardness improved by 10% against neat epoxy. Thermal analysis indicated enhanced stability, with decomposition onset temperature changed by + 19°C and glass transition temperature from improved by 2.5 wt%. SEM analysis substantiated strong interfacial adhesion at low filler loadings, but at 7.5 wt% voids and agglomeration lowered properties. ¹⁵ Sahayaraj et al. (2021) explored tamarind seed filler in PLA composites with hybridization by sisal fiber. Tensile strength was enhanced by 19% at 15 wt% filler loading, while tensile modulus enhanced by 27%. Flexural strength enhanced by 21%, and impact strength enhanced by 14% with respect to neat PLA. Thermal stability improved, decomposition temperature changed by + 12°C, and water uptake improved moderately (∼7% at 20 wt%). ¹⁶ Altun et al. (2021) produced PLA composites filled with untreated and alkali-treated pistachio shell (PS) filler (5–20 wt%). Optimum results were achieved for 20 wt% treated PS filler. SEM proved that alkali treatment eliminated the impurities on the surface, enhancing the filler–matrix adhesion. Thermogravimetric analysis showed increased stability, with the onset degradation temperature being raised by + 15°C. Density was little altered, proving lightweight promise for packaging and automobile use. ¹⁷ Ramesh et al. (2022) provided a review of the function of organic and inorganic fillers (micro and nano) in polymer composites. In various studies, nano-fillers like nanoclay, CNTs, and graphene enhanced tensile strength by 20%–45%, modulus by 30%–55%, and thermal stability by as much as 40°C over unfilled composites. For example, rice husk powder in HDPE (20–70 wt%) showed improvement in tensile modulus by ∼250%, and fly ash (10 wt%) in epoxy improved flexural strength by ∼18%. The discussion also emphasized that the best filler content is generally 3–10 wt% for nano-fillers and 10–30 wt% for micro-fillers, beyond which agglomeration decreases properties. ⁸

Various agricultural waste items such as egg shells, coconut shell, rice husk, fish bones, fish scales, groundnut and peanut powder, wood sawdust and Samanea saman fillers have been investigated by researchers as fillers into different composites. The results show that these fillers derived from agricultural waste have a positive effect on the material’s attributes, improving strength and stiffness of composites. Additionally, by recycling waste materials and lowering the ecological imprint of composite production, the use of these fillers promotes environmental sustainability.^18,19

While the Pista shell powder (PSP), used for this study was examined by a very few researchers, the comprehensive study adds up to the knowledge of utilizing this agro waste as a potential bio filler. The lignocellulosic fibres that compose up pistachio shells are widely acknowledged for their exceptional mechanical strength and thermal stability. In nanocomposites, where they improve structural reinforcement and aid in minimising thermal stress, these fibres are crucial. ²⁰ Pistachio shells have been utilised as a reinforcement in thermosetting resins such as epoxy in prior research,^21,22 however it’s also essential that researchers look at the way these could potentially be utilised in bio-based thermoplastics. This approach, particularly emphasises sustainability and environmental benefits in composite manufacture, could advance the development of green composites. Advances in green composites, materials comprising less processing-intensive natural fibres embedded within sustainable matrices signify a crucial development for the field of material science. These composites both solve waste disposal issues and provide an innovative alternative to conventional synthetic materials, utilizing agro-waste. ²³ The development of green composites manufactured from agro-waste (agricultural waste) is a necessary step in the quest for sustainability. This is sustainability intertwined with innovation that tackles both environmental issues and contributes to the circular economy by transforming once waste into high-performance, sustainable materials. This research breaks new ground in the development of agro-waste-based green composites to establish them as a material across different industries dedicated to sustainability. Highlighting the environmental aspects, we not only emphasize the waste valorization potential but also promote moving toward a circular economy. Although research on bio-filler composites has been unfolding, the use of pistachio shell powder as a new, unexplored bio-filler is a novel contribution. In addition, this research overcomes the absence of all-around optimization studies, with a new approach on how various processing parameters, e.g., filler type and injection molding methods, interact to impact the material properties. This two-part innovation developing a novel filler and optimizing processing methods paves the way for next-generation, high-performance green composites that can help industries from automotive to packaging while helping create a greener world.

Materials and Methodology

The materials utilized in the experimental study and the various testing procedures have been outlined briefly in this section.

Resin and Filler

The matrix material utilized in this study is polylactic acid (PLA), a biodegradable thermoplastic resin sourced from renewable resources and supplied by Natur Tec India Pvt. Ltd. PLA has a glass transition temperature of 50°C - 80°C and a melting temperature of 145°C −180°C and. The pista shells employed in this experiment were sourced from a local marketplace. To remove dirt and other impurities, the shells were first soaked in distilled water for about 3 hours and then washed with distilled water. After the shells have completely dried up, the ground. Once ground, the lignocellulosic material was sieved through mesh sizes ranging from 40 to 100 μm to maintain a uniform particle size distribution.. The shell powder was then soaked in 5% w/v NaOH for 3 hours after which they were thoroughly washed with deionized water several times, before being left for sun-drying for 10 days.

XRD Characterization

Pistachio shells are primarily composed of a fibrous structure, essentially a mixture of amorphous and crystalline polymers. Microscopic examination identified that the shells consist of laminar layers of polymer arranged in a wound configuration around small pores in the material. This structure contributes to the overall mechanical strength and durability of the pistachio shells. From Figure 1 (XRD characterization), there is a large peak around 20°–30° showing an increase in intensity sharply, indicating that there is significant diffraction from a particular set of planes in the crystal. The sharp peak indicates a highly ordered crystalline structure. Chemically, pistachio shells consist primarily of cellulose and triglycerides with no notable inorganic compounds. The ratio cellulose and triglycerides and within the shells differs based on how deep the shell is which conforms to functional needs at each given depth. ²⁴ This composition difference affects the properties of the material like its mechanical strength, heat stability and applicability as a filler component in composites. The distinct blend of organic species in pistachio shells identifies them as a rich source of reinforcing composite materials that impart both environmental sustainability and mechanical strength.

Figure 1.

XRD Characterization of the Pista Shell filler.

Design of Experiments with RSM

Response Surface Methodology (RSM) is a set of statistical and mathematical techniques, helpful in simulating and analyzing complex engineering issues. RSM is primarily used to maximize the response surface, which is driven by various process variables and to identify the greatest possible solution through identifying how to configure these variables, which improves the overall performance and efficiency of the process. The main strengths of RSM is the reduction of the number of experimental trials required, and it becomes less complicated for researchers to test more parameters and interactions between them in a better manner. This efficient approach allows for a deeper and more thorough comprehensive understanding of the intricate relationships between the process’s variables while additionally saving time and money. In addition, RSM is used mainly to approximate a set of processes, conduct models for experimentation, validate the effect of different components, build prediction models, and determine optimal conditions to produce desired results. Through systematic reduction of the number of experimental runs required, RSM offers an effective experimentation procedure while maintaining the accuracy and reliability of the outcomes. The experiments were optimized based on a Box-Behnken Design (BBD) model, with four factors at three levels, as indicated in Table 1. The experimental results were analyzed, resulting in the development of a mathematical model representing the correlation between the input machining parameters and the output response variables. The model is a predictive tool that allows a better understanding of the influence of each parameter on the response variables.

Table 1.

Design of Experiments by Box–Behnken design.

Run	Std	Filler percentage (%)	Melting temperature (oC)	Injection pressure (MPa)	Injection speed (mm.s−1)
1	12	30	175	70	50
2	8	20	165	70	50
3	5	20	175	110	40
4	23	20	175	90	50
5	21	20	175	90	50
6	18	30	175	110	50
7	1	10	175	70	50
8	27	20	165	110	50
9	10	30	175	90	40
10	7	20	175	70	60
11	11	10	165	90	50
12	16	20	175	70	40
13	17	10	175	90	40
14	19	10	175	110	50
15	22	20	185	90	60
16	9	10	175	90	60
17	20	30	185	90	50
18	4	30	175	90	60
19	25	20	165	90	60
20	15	20	185	110	50
21	14	20	165	90	40
22	6	20	185	90	40
23	26	20	175	110	60
24	3	10	185	90	50
25	24	20	175	90	50
26	2	30	165	90	50
27	13	20	185	70	50

Fabrication of Composites

The PLA/PSP composite samples were prepared using a 50-ton servo-controlled hydraulic injection molding method. For the generation of optimum processing conditions, the barrel temperature of the injection moulding machine was regulated based on the experimental conditions. The injection moulding process was performed for a period of 2 seconds with a shot weight of 20 g. Both the PLA pellets and the filler material were fed directly into the injection moulding machine hopper during the fabrication of the composites. In order to find the optimal formulation, several tests were performed varying the filler percent by weight (w/w) and the injection parameters. Three samples for each run has been fabricated to ensure the proximity of results.

Tensile Characteristics

Blue Star Universal Testing Machine (UTM) VTES 20, whose load capacity was between 4 and 200 kN was utilized in this research to conduct the tensile properties of the composite specimens that were produced. The test was conducted based on ASTM D638 standard, where the specimens measured 165 mm × 19 mm x 5 mm with a crosshead speed of 2 mm/min. The tensile properties of the composite materials are accurately and efficiently tested through adherence to the set testing methods, allowing for exact comparison and evaluation of their mechanical behaviour. The standard formulas were subsequently applied to evaluate the composite samples’ different tensile properties.

$Tensile\hspace{0.17em}Strength\left(\sigma \right)=\frac{F}{A}$

(1)

$\text{Tensile}\hspace{0.17em}\text{Modulus}=\frac{\sigma }{e}$

(2)

Where, F is load, A is cross-sectional area and $e$ is tensile strain.

Flexural Characteristics

The three point bending apparatus on the UTM was utilized to perform flexural characterisation of the specimens. The test was carried out as per ASTM D790 standard with a cross head speed of 5 mm/min. The size of the specimen is 130 mm × 12.7 mm x 5 mm, following the standard procedure ensures uniform testing conditions and allows for accurate measurement of the flexural characteristics of the composite specimens. The flexural strength and flexural modulus of the composite are calculated by using the following formulas:

$Flexural\hspace{0.17em}Strength=\frac{3FL}{2w{t}^2}$

(3)

$Flexural\hspace{0.17em}Modulus=\frac{m{L}^3}{4w{t}^3}$

(4)

Where, F is load, L is span dimension, m is slope of curve, w is width of sample, and t is sample thickness.

Results and Discussion

In this section, the effect of filler percentage and machining parameters on various output parameters has been discussed. Response Surface Methodology (RSM) model has been used to analyze the effect of input parameters on the output responses based on the experimental data for the same. Furthermore, Analysis of Variance (ANOVA) with 95% confidence interval (CI) was utilized to check the reliability and significance of the model obtained. This statistical model ensures that the model is sufficient and trustworthy in prediction of the result with high confidence in the interaction between variables. ANOVA allows for model validity checks as well as contributions of each factor within the desired confidence level. Fisher’s statistical test, as reflected by the F ratio, the significance probability (P value) and coefficients of determination (R²) and adjusted R², were utilized to test the adequacy and appropriateness of the model. F value and P value represent the factors that have maximum influence on response variables and the statistical significance of process variables, respectively. Taken together, the statistical values provide an overall appraisal of the performance of the model and the ability to accurately depict the interactions between the data.

In the analysis, the F value larger than that of P < .05 indicates that the variables are statistically significant, meaning that the variables have a good positive impact on the response variables. The R² value is the ratio of total variation in the response variable explained by the model. It is the ratio of the regression sum of squares (SSR) to the total sum of squares (SST). The greater the value of R², the higher is the model’s capability to explain variability in the response and, therefore, the better the model’s performance in describing the relationships in the data. The adjusted R² is utilized to examine the model fit and adequacy under adjustment for the number of predictors employed. The backward elimination procedure was utilized in this research to eliminate parameters stepwise one at a time, hence narrowing the fitted quadratic model. This technique increases the reliability of the model by targeting the most influential variables so that only those contributing significantly to the response are maintained. Subsequently, the P values for all parameters were less than 0.05, indicating that both the model and its corresponding terms are statistically significant. This finding validates the model and suggests that the parameters considered have a considerable impact on the response variables, further affirming the robustness of the analysis.

Determination of Input Parameter Effects on Tensile Characteristic

The mathematical equation for the ultimate tensile strength and the above-defined machining parameters is developed through the following expressions. The equations provide a quantitative relationship between the input parameters, for example, filler content, melting temperature, and other machining parameters, and the final ultimate tensile strength of the composite material. By examining these terms, the effect of each parameter on the tensile properties of the material can be more clearly understood, making optimization for better performance easier. Equations (5) and (6) are the coded equation and the true equation, respectively, for describing the relationship between the input parameters and the output response, Tensile Strength.

$Coded\hspace{0.17em}Equation:22.5767-2.6175*A-3.58917*B-2.43167*C-2.33667*D+2.715*AB-2.74*AC+1.2825*AD+2.9025*BC-1.135*BD+2.0825*CD-1.49*A^2+2.095*B^2+1.49625*C^2+0.75125*D^2$

(5)

$Actual\hspace{0.17em}Equation:1057.27-3.82525*A-8.97304*B-3.58121*C-0.192292*D+0.02715*AB-0.0137*AC+0.012825*AD+0.0145125*BC-0.01135*BD+0.0104125*CD-0.0149*A^2+0.02095*B^2+0.00374062*C^2+0.0075125\hspace{0.17em}*\hspace{0.17em}{D}^2$

(6)

The Analysis of Variance (ANOVA) table (Table 2) provides a descriptive view of the effects of factors on the output responses. In the present case, model terms A, B, C, D, and their interactions AB, AC, AD, BC, BD, CD, along with the quadratic terms A², B², and C², are seen to be significant. Among the input parameters, the melting temperature has the most significant influence on the ultimate tensile strength, followed closely by the filler percentage. The coded equation suggests that each individual input parameter negatively affects the tensile strength. However, most of the interaction terms and quadratic terms have a positive influence on the tensile strength, which means that some combinations or levels of these parameters can improve the performance of the material. This pattern suggests that raising the levels of each input parameter results in a decrease in the ultimate tensile strength of the material. These results are shown to reflect that the higher parameters lead to the reduction in the tensile performance of the sample produced. Figure 2 illustrates 3D visualization of interaction between the two most significant parameters and provides detailed insight into the interaction effect on tensile strength. The reduction in tensile strength at a higher content of pistachio shell fillers, which is shown, agrees with previous research.^21,25 This observation predicts that greater filler content might impose conditions, for example, weakened interfacial adhesion or enhanced brittleness, which would reduce tensile strength in the composite. These observations conform to the larger body of work on filler-matrix interaction within composite materials to advocate the value of meticulous optimization of filler content to ensure best mechanical performance.

Table 2.

ANOVA table for tensile strength

Source	Sum of Squares	df	Mean Square	F- value	p- value		Fit Statistics
Model	560.02	14	40.00	57.84	<0.0001	Significant	R2	0.9854
A- filler percentage	82.22	1	82.22	118.88	<0.0001		Adjusted R2	0.9684
B- melting temperature	154.59	1	154.59	223.52	<0.0001		Predicted R2	0.9173
C- pressure	70.96	1	70.96	102.60	<0.0001
D- speed	65.52	1	65.52	94.74	<0.0001
AB	29.48	1	29.48	42.63	<0.0001
AC	30.03	1	30.03	43.42	<0.0001
AD	6.58	1	6.58	9.51	0.0095
BC	33.70	1	33.70	48.72	<0.0001
BD	5.15	1	5.15	7.45	0.0183
CD	17.35	1	17.35	25.08	0.0003
A2	11.84	1	11.84	17.12	0.0014
B2	23.41	1	23.41	33.85	<0.0001
C2	11.94	1	11.94	17.26	0.0013
D2	3.01	1	3.01	4.35	0.0590
Residual	8.30	12	0.6916
Lack of fit	8.07	10	0.8075	7.20	0.1280	Not significant
Pure error	0.2243	2	0.1121
Cor total	568.32	26

Figure 2.

3D interaction graph between Melting Temperature and Filler percentage for Tensile Strength.

The obtained results of the tensile strength are given in Table 3. The 3D surface plot clearly illustrates the effect of filler content (X-axis) and melting point (Y-axis) on the tensile strength (Z-axis) of the material. A colour gradient, ranging from blue (representing low tensile strength) to red (representing high tensile strength), clearly emphasizes variations in tensile strength at various combinations of filler content and melting points. This gradient allows one to gain intuitive insight into how changes in these two most important parameters interplay to influence tensile properties of the material. Peaks on the plot are the best filler percent and melting temperature combinations where the tensile strength is highest.

Table 3.

Tensile strength of the fabricated composites.

Run	Tensile strength
	Sample 1	Sample 2	Sample 3	Average	Std. Deviation (%)
1	23.319	24.452	24.857	24.21	2.69
2	34.789	35.999	36.431	35.74	1.94
3	21.240	22.270	22.638	22.05	2.68
4	21.348	22.419	22.801	22.19	2.77
5	21.809	23.031	23.468	22.77	3.08
6	14.583	15.317	15.579	15.16	2.78
7	23.696	24.808	25.205	24.57	2.60
8	23.602	24.579	24.928	24.37	2.30
9	20.317	21.402	21.790	21.17	2.94
10	21.896	23.122	23.560	22.86	3.08
11	29.934	31.481	32.033	31.15	2.85
12	30.863	31.890	32.256	31.67	1.86
13	27.514	29.048	29.596	28.72	3.07
14	25.757	26.676	27.005	26.48	1.99
15	17.651	18.629	18.979	18.42	3.05
16	19.822	20.646	20.940	20.47	2.31
17	19.464	20.236	20.512	20.070	2.21
18	17.388	18.231	18.532	18.050	2.68
19	27.620	29.097	29.624	28.780	2.95
20	22.175	23.137	23.480	22.930	2.41
21	29.045	30.516	31.041	30.200	2.80
22	23.771	24.547	24.824	24.380	1.83
23	20.760	21.792	22.160	21.570	2.75
24	18.784	19.721	20.056	19.520	2.76
25	22.195	22.927	23.189	22.770	1.85
26	20.233	21.006	21.282	20.840	2.13
27	21.837	22.923	23.311	22.690	2.75

Equations (7) and (8) are the coded equation and actual equation, respectively, for the description of the relationship between the input parameters and the output response, Tensile Modulus. The coded equation reduces the mathematical form by applying coded values for the parameters, whereas the actual equation is based on the actual world values and units, with a direct usage in optimization and analysis of the material properties. Unlike the Ultimate Tensile Strength (UTS), the tensile modulus coded equation indicates that all four input parameters have a positive effect on the tensile modulus. This indicates that as the levels of these parameters increase, the tensile modulus also improves, suggesting enhanced stiffness and resistance to deformation in the fabricated material. Such findings highlight the distinct behaviours of tensile strength and modulus.

$Coded\hspace{0.17em}Equation:2170.13+179.064*A+72.5475*B+2.645*C+6.59167*D-9.425*AB-27.31*AC-8.9225*AD+28.085*BC+25.9575*BD-34.71*CD+105.084*A^2-8.49625*B^2-69.7025*C^2-73.485\hspace{0.17em}*\hspace{0.17em}{D}^2$

(7)

$Actual\hspace{0.17em}Equation:-1895.57+9.11742*A+13.2596*B+18.3325*C+46.1225*D-0.09425*AB-0.13655*AC-0.089225*AD+0.140425*BC+0.259575*BD-0.17355*CD+1.05084*A^2-0.0849625*B^2-0.174256*C^2-0.73485*D^2$

(8)

From Table 4, it is evident that input model terms A and B are significant factors affecting the response. Although the interactions between the input parameters were not found to be significant, the quadratic terms A², C² and D² have been identified as significant. This indicates that while the main effects of A and B are crucial, the squared terms of certain parameters also play a notable role in influencing the response. The influence of the significant terms can be interpreted from the coded equation. While the input parameters A and B exhibit a positive effect on the response, the quadratic terms C² and D² demonstrate a negative effect. However, the quadratic term A² shows a positive effect. This suggests that increasing values of A and B enhance the desired output, while certain non-linear effects, as indicated by the quadratic terms, influence the response differently, with A² contributing positively and C² and D² exerting a negative influence. From the ANOVA table, it can also be inferred that among the significant input parameters, filler percentage emerges as the most influential factor. This indicates that variations in the filler content have the greatest impact on the response variable, underscoring its importance in the optimization of the composite young’s modulus.

Table 4.

ANOVA table for tensile modulus.

Source	Sum of Squares	df	Mean Square	F- value	p- value		Fit Statistics
Model	6.309 E+05	14	45,067.71	41.30	<0.0001	Significant	R2	0.9797
A- filler percentage	3.848 E+05	1	3.848 E+05	352.64	<0.0001		Adjusted R2	0.9560
B- melting temperature	63,157.68	1	63,157.68	57.88	<0.0001		Predicted R2	0.8843
C- pressure	83.95	1	83.95	0.0769	0.7862
D- speed	521.40	1	521.40	0.4779	0.5025
AB	355.32	1	355.32	0.3257	0.5788
AC	2983.34	1	2983.34	2.73	0.1241
AD	318.44	1	318.44	0.2919	0.5989
BC	3155.07	1	3155.07	2.89	0.1148
BD	2695.17	1	2695.17	2.47	0.1420
CD	4819.14	1	4819.14	4.42	0.0574
A2	58,893.84	1	58,893.84	53.98	<0.0001
B2	384.99	1	384.99	0.3528	0.5635
C2	25,911.67	1	25,911.67	23.75	0.0004
D2	28,800.24	1	28,800.24	26.40	0.0002
Residual	13,093.33	12	1091.11
Lack of fit	12,835.30	10	1283.53	9.95	0.0947	Not significant
Pure error	258.03	2	129.02
Cor total	6.440 E+05	26

The obtained results of the tensile modulus are given in Table 5. The 3D surface plot (Figure 3) demonstrates the effect of filler percentage (X-axis) and melting temperature (Y-axis) on tensile strength (Z-axis) of the material. The color gradient, transitioning from blue (indicating low tensile strength) to red (indicating high tensile strength), visually emphasizes how the tensile modulus changes in response to varying filler percentages and melting temperatures. This graphical illustration assists in comprehending the interactive effect of these two important parameters on the mechanical properties of the material and can assist in optimizing the tensile modulus of the composite. The increase in tensile modulus with a rise in pistachio shell fillers, as observed in the current study, concurs with the work of other researchers.^11,21,26 This trend indicates that the filler particles are contributing to energy absorption, discouraging crack growth, and thereby improving the young modulus of the materials.

Table 5.

Tensile modulus of the fabricated composites.

Run	Tensile modulus
	Sample 1	Sample 2	Sample 3	Average	Std. Deviation (%)
1	2365.097	2456.398	2502.048	2441.180	2.33
2	1991.900	2062.749	2098.173	2050.940	2.15
3	2008.944	2087.588	2126.910	2074.480	2.36
4	2123.401	2186.941	2218.710	2176.350	1.82
5	2127.497	2186.926	2216.639	2177.020	1.70
6	2290.272	2392.091	2442.999	2375.120	2.67
7	1941.023	2002.220	2032.819	1992.020	1.92
8	1947.634	2025.078	2063.800	2012.170	2.40
9	2328.209	2407.591	2447.282	2394.360	2.07
10	1985.453	2074.562	2119.116	2059.710	2.70
11	1960.312	2046.123	2089.028	2031.820	2.63
12	1951.232	2003.935	2030.286	1995.150	1.65
13	1933.926	2015.136	2055.741	2001.600	2.53
14	1966.327	2048.976	2090.299	2035.200	2.53
15	2154.503	2255.144	2305.464	2238.370	2.80
16	1935.878	2019.209	2060.874	2005.320	2.59
17	2428.503	2506.680	2545.769	2493.650	1.96
18	2302.185	2374.432	2410.555	2362.390	1.91
19	1955.894	2017.462	2048.246	2007.200	1.91
20	2112.036	2178.558	2211.819	2167.470	1.91
21	1930.939	2014.509	2056.293	2000.580	2.60
22	2065.774	2140.350	2177.638	2127.920	2.19
23	1942.030	2011.835	2046.737	2000.200	2.18
24	2145.063	2220.072	2257.577	2207.570	2.12
25	2091.181	2170.189	2209.692	2157.020	2.28
26	2284.017	2369.918	2412.867	2355.600	2.27
27	2016.937	2109.294	2155.472	2093.900	2.75

Figure 3.

3D interaction graph between Melting Temperature and Filler percentage for Tensile Modulus.

Equations (9) and (10) are the coded and true equations, respectively, for the output response, i.e., the percent elongation. These equations yield a mathematical expression between the input parameters and elongation response that enables a quantitative analysis of how various factors affect the material to deform prior to failure. The encoded equation is the simplified version of the model, whereas the real equation represents the parameters of the real world and their respective influences on elongation.

$Coded\hspace{0.17em}Equation:0.883767-0.1394*A-0.30715*B-0.135*C-0.219417*D-0.1732*AB-0.045*AC+0.19*AD+0.0925*BC-0.21075*BD+0.4025*CD+0.087175*A^2+0.31055\hspace{0.17em}*B^2+0.396275\hspace{0.17em}*C^2+0.48215\hspace{0.17em}*D^2$

(9)

$Actual\hspace{0.17em}Equation:117.108+0.17954*A-1.01925*B-0.362136*C-0.354404*D-0.001732*AB-0.000225*AC+0.0019*AD+0.0004625*BC-0.0021075*BD+0.0020125*CD+0.00087175*A^2+0.0031055*B^2+0.000990687*C^2+0.0048215*D^2$

(10)

From equation (9), it can be inferred that all the input parameters have a negative effect on the percentage of elongation of the composite, indicating that increases in these parameters tend to reduce the material’s ability to elongate before failure. However, some of the interactions between the input parameters show a positive effect, while others show a negative effect. Additionally, all the quadratic terms exhibit a positive effect, suggesting that the non-linear relationships between certain parameters can enhance the material’s elongation. For percent of elongation, A, B, C, D, AB, AD, BC, BD, CD, A², B², C², D² are significant model terms indicating that the samples elongation at the point of failure is significantly influenced by all four key input parameters. Among these, the melting temperature is found to be the most significant factor, followed by speed (Table 6). The 3D graph (Figure 4) illustrates that, for both melting temperature and speed, an increase in the levels of these input parameters consistently results in a decrease in the percentage of elongation across each level. This trend suggests that higher melting temperatures and increased speed negatively affect the material’s ability to elongate before failure, which could be attributed to changes in the material’s mechanical properties as these parameters are adjusted.

Table 6.

ANOVA table for percent of elongation.

Source	Sum of Squares	df	Mean Square	F- value	p- value		Fit Statistics
Model	5.07	14	0.3622	62.62	<0.0001	Significant	R 2	0.9865
A- filler percentage	0.2332	1	0.2332	40.32	<0.0001		Adjusted R 2	0.9707
B- melting temperature	1.13	1	1.13	195.73	<0.0001		Predicted R 2	0.9229
C- pressure	0.2187	1	0.2187	37.81	<0.0001
D- speed	0.5777	1	0.5777	99.88	<0.0001
AB	0.1200	1	0.1200	20.75	0.0007
AC	0.0081	1	0.0081	1.40	0.2596
AD	0.1444	1	0.1444	24.97	0.0003
BC	0.0342	1	0.0342	5.92	0.0316
BD	0.1777	1	0.1777	30.72	0.0001
CD	0.6480	1	0.6480	112.04	<0.0001
A2	0.0405	1	0.0405	7.01	0.0213
B2	0.5144	1	0.5144	88.93	<0.0001
C2	0.8375	1	0.8375	144.80	<0.0001
D2	1.24	1	1.24	214.35	<0.0001
Residual	0.0694	12	0.0058
Lack of fit	0.0684	10	0.0068	14.07	0.0681	Not significant
Pure error	0.0010	2	0.0005
Cor total	5.14	26

Figure 4.

3D interaction graph between Melting Temperature and Speed for Percentage of Elongation (Tensile).

The obtained results of the percentage of elongation are given in Table 7. The effect of the filler percentage on the material’s percentage of elongation can be observed from Figure 5. As the filler percentage increases, the percentage elongation of the matrix shows a consistent declining trend with each increase in the input parameter levels. This implies that increased filler content lowers the material’s elongation ability before failure, probably because of the increased stiffness and lower ductility brought by the fillers. This can be attributed to the fact that the addition of filler, with suitable shape and size, to the matrix leads to a decrease in the ductility of the matrix while also increasing its strength.^11,25 The fillers contribute to increased strength but come at the expense of reduced elongation or ductility. By reinforcing the composite, the fillers enhance its overall strength, yet they limit its ability to deform before failure, thereby reducing the material’s elongation properties. This is a common phenomenon when incorporating fillers into composite materials, as the material becomes stiffer and more brittle.

Table 7.

Percentage of elongation of the fabricated composites.

Run	Tensile percentage of elongation
	Sample 1	Sample 2	Sample 3	Average	Std. Deviation (%)
1	1.274	1.318	1.340	1.310	2.09
2	2.112	2.182	2.217	2.170	2.01
3	1.385	1.428	1.449	1.420	1.87
4	0.830	0.865	0.882	0.859	2.52
5	0.872	0.902	0.917	0.897	2.09
6	1.021	1.056	1.074	1.050	2.10
7	1.498	1.549	1.574	1.540	2.05
8	1.613	1.682	1.716	1.670	2.57
9	1.379	1.441	1.472	1.430	2.70
10	1.264	1.308	1.330	1.300	2.11
11	1.464	1.508	1.529	1.500	1.81
12	2.489	2.563	2.600	2.550	1.81
13	2.006	2.071	2.104	2.060	1.98
14	1.423	1.468	1.490	1.460	1.91
15	0.881	0.913	0.929	0.908	2.20
16	1.109	1.159	1.184	1.150	2.71
17	0.694	0.722	0.736	0.718	2.43
18	1.243	1.288	1.310	1.280	2.18
19	1.987	2.063	2.101	2.050	2.31
20	1.220	1.257	1.275	1.250	1.83
21	1.913	1.982	2.016	1.970	2.17
22	1.629	1.679	1.704	1.670	1.87
23	1.741	1.788	1.812	1.780	1.66
24	1.291	1.338	1.362	1.330	2.22
25	0.875	0.901	0.915	0.897	1.85
26	1.535	1.590	1.617	1.580	2.16
27	1.345	1.388	1.409	1.380	1.93

Figure 5.

Effect of Filler Percentage on Percentage of Elongation (Tensile).

Determination of Input Parameter Effects on Flexural Characteristic

Equations (11) and (12) represent the coded and actual equations, respectively, for the flexural strength of the composite material.

$Coded\hspace{0.17em}Equation:68.72-10.32*A-5.74833*B-14.48*C-7.22167*D+24.6375*AB-3.4025*AC+1.98*AD+5.1525*BC-5.305*BD-3.045*CD-1.84208*A^2+7.66292*B^2+7.64792*C^2-4.08458*D^2$

(11)

$Actual\hspace{0.17em}Equation:3407.93-42.8697*A-31.9887*B-7.5725*C+13.6204*D+0.246375*AB-0.0170125*AC+0.0198*AD+0.0257625*BC-0.05305*BD-0.015225*CD-0.0184208*A^2+0.0766292*B^2+0.0191198\hspace{0.17em}*\hspace{0.17em}{C}^2-0.0408458*D^2$

(12)

The Analysis of Variance (ANOVA) table (Table 8) provides a comprehensive examination of the various factors that contribute to the variation in the output responses. In this case A, B, C, D, AB, AC, AD, BC, BD, CD, A², B², C² are significant model terms. Among the input parameters, pressure is found to exert the most substantial impact on the ultimate tensile strength, followed by filler percentage. The equation indicates that each individual input parameter exerts a negative influence on tensile strength, with the exception of speed. Additionally, while some interaction and quadratic terms demonstrate a positive effect on flexural strength, others show a detrimental effect. This suggests that specific combinations or levels of these parameters may enhance the material’s performance, while others may hinder it. This trend suggests that an increase in the level of the input parameter, speed, results in an enhancement of the material’s flexural strength. In contrast, an increase in the filler percentage leads to a reduction in flexural strength. Figure 6 presents a three-dimensional visualization of the interaction between the two most influential parameters, providing a detailed representation of how their combined effects influence flexural strength. The observed reduction in flexural strength as the content of fillers increases, aligns with findings reported in previous studies.^26,27 From the 3D diagram, it can be observed as filler percentage and speed increases, the flexural strength consistently decreases with each additional level of filler as discussed earlier.

Table 8.

ANOVA table for flexural strength.

Source	Sum of squares	df	Mean square	F- value	p- value		Fit Statistics
Model	8543.18	14	610.23	113.00	<0.0001	Significant	R2	0.9925
A-filler percentage	1278.03	1	1278.03	236.67	<0.0001		Adjusted R2	0.9837
B-melting temperature	396.52	1	396.52	73.43	<0.0001		Predicted R2	0.9570
C-pressure	2516.04	1	2516.04	465.93	<0.0001
D-speed	625.83	1	625.83	115.89	<0.0001
AB	2428.03	1	2428.03	449.63	<0.0001
AC	46.31	1	46.31	8.58	0.0126
AD	15.68	1	15.68	2.90	0.1141
BC	106.19	1	106.19	19.67	0.0008
BD	112.57	1	112.57	20.85	0.0006
CD	37.09	1	37.09	6.87	0.0224
A2	18.10	1	18.10	3.35	0.0921
B2	313.17	1	313.17	58.00	<0.0001
C2	311.95	1	311.95	57.77	<0.0001
D2	88.98	1	88.98	16.48	0.0016
Residual	64.80	12	5.40
Lack of fit	63.93	10	6.39	14.76	0.0651	Not significant
Pure error	0.8664	2	0.4332
Cor total	8607.98	26

Figure 6.

3D interaction graph between Filler Percentage and Pressure for Flexural Strength.

The obtained results of the flexural strength are given in Table 9. The 3D surface plot demonstrates the effect of filler percentage and pressure on the flexural strength (FS) of the material. As both filler percentage and pressure increase, the flexural strength generally shows a downward trend, as illustrated by the declining slope and the color gradient shifting from yellow (indicating high FS) to blue (indicating low FS). The plot identifies an optimal region for achieving maximum flexural strength, emphasizing that strategic adjustments to these two variables are crucial for optimizing the material’s performance. Among the input parameters, pressure is found to be the most significant parameter followed by filler percentage.

Table 9.

Flexural strength of the fabricated composites.

Run	Flexural strength
	Sample 1	Sample 2	Sample 3	Average	Std. Deviation (%)
1	77.333	80.414	81.955	79.900	2.40
2	105.529	109.167	110.986	108.560	2.09
3	65.293	67.402	68.456	67.050	1.96
4	67.322	69.912	71.207	69.480	2.32
5	66.135	68.782	70.105	68.340	2.42
6	46.112	47.382	48.018	47.170	1.68
7	91.443	95.772	97.936	95.050	2.84
8	67.425	69.544	70.603	69.190	1.91
9	57.758	59.573	60.481	59.270	1.91
10	82.427	85.323	86.771	84.840	2.13
11	111.267	115.327	117.357	114.650	2.21
12	90.494	93.322	94.736	92.850	1.90
13	79.604	82.240	83.558	81.800	2.01
14	73.401	76.436	77.954	75.930	2.49
15	51.094	53.466	54.652	53.070	2.79
16	59.798	61.421	62.232	61.150	1.65
17	82.456	85.437	86.928	84.940	2.19
18	44.931	46.862	47.828	46.540	2.59
19	76.074	79.538	81.270	78.960	2.74
20	66.919	68.937	69.945	68.600	1.83
21	78.764	81.316	82.592	80.890	1.97
22	74.161	76.632	77.868	76.220	2.02
23	45.148	47.203	48.230	46.860	2.73
24	55.620	57.349	58.213	57.060	1.89
25	65.961	68.816	70.244	68.340	2.61
26	42.501	44.276	45.164	43.980	2.52
27	84.359	87.961	89.762	87.360	2.57

From the ANOVA table (Table 10), it is evident that the factors A, B, C, D, as well as the interaction terms AB, AC, AD, BC, BD, CD, and the quadratic terms A², B², C², D², all demonstrate statistical significance. Upon further examination of the coded equation, it becomes apparent that each of the individual input parameters A, B, C, and D, exert a positive influence on the tensile modulus of the composite material. In addition, most of the interaction and quadratic terms also exhibit positive effects. However, it is noteworthy that the interaction term AD stands out, as it is the only factor that shows a negative impact on the tensile modulus. This suggests that while the main effects and the majority of interactions and quadratic terms contribute positively to the mechanical properties of the composite, the specific interaction between factors A and D may have a detrimental influence. This trend, when observed, states that a rise in the values of the input parameters, that is, Pressure and Filler Percentage, will result in the enhancement of the material’s flexural modulus. That is, as these values are increased, there is also an increase in the material’s resistance to bending, which accentuates their contribution to the improvement of the material’s mechanical property. To better illustrate such a relationship, Figure 7 offers a three-dimensional representation illustrating the interaction of these two most significant parameters. Such a graphical presentation provides a clear understanding of how the coupled effects of Pressure and Filler Percentage together influence the variations of the flexural modulus, thus facilitating a clearer interpretation of their coupled effect on the performance of the material. Equations (13) and (14) are the actual and coded equations, respectively, for the composite material’s flexural modulus.

$Coded\hspace{0.17em}Equation:1.52064+0.137735*A+0.0493598*B+0.141221*C+0.049788*D+0.12746*AB+0.0763335*AC-0.1126*AD+0.0464808*BC+0.236871*BD+0.152449*CD+0.0411985*A^2+0.0956013*B^2+0.156528*C^2+0.0548798*D^2$

(13)

$Actual\hspace{0.17em}Equation:65.32-0.20381*A-0.494512*B-0.149793*C-0.510506*D+0.0012746*AB+0.000381667*AC-0.001126*AD+0.000232404*BC+0.00236871*BD+0.000762243*CD+0.000411985*A^2+0.000956013*B^2+0.000391321*C^2+0.000548798*D^2$

(14)

Table 10.

ANOVA table for flexural modulus.

Source	Sum of Squares	df	Mean Square	F- value	p- value		Fit Statistics
Model	1.14	14	0.0812	170.69	<0.0001	Significant	R2	0.9950
A- filler percentage	0.2277	1	0.2277	478.82	<0.0001		Adjusted R2	0.9892
B- melting temperature	0.0292	1	0.0292	61.49	<0.0001		Predicted R2	0.9714
C- pressure	0.2393	1	0.2393	503.37	<0.0001
D- speed	0.0297	1	0.0297	62.56	<0.0001
AB	0.0650	1	0.0650	136.68	<0.0001
AC	0.0233	1	0.0233	49.02	<0.0001
AD	0.0507	1	0.0507	106.67	<0.0001
BC	0.0086	1	0.0086	18.18	0.0011
BD	0.2244	1	0.2244	472.05	<0.0001
CD	0.0930	1	0.0930	195.53	<0.0001
A2	0.0091	1	0.0091	19.04	0.0009
B2	0.0487	1	0.0487	102.52	<0.0001
C2	0.1307	1	0.1307	274.84	<0.0001
D2	0.0161	1	0.0161	33.79	<0.0001
Residual	0.0057	12	0.0005
Lack of fit	0.0056	10	0.0006	18.29	0.0529	Not significant
Pure error	0.0001	2	0.0000
Cor total	1.14	26

Figure 7.

3D interaction graph between Filler Percentage and Pressure for Flexural Modulus.

The obtained results of the flexural modulus are given in Table 11. The three-dimensional surface plot correctly demonstrates the combined effect of both pressure and filler percentage on material flexural modulus (FM). When the level of filler percentage and pressure both rise, there is a rising trend in flexural modulus that is graphically demonstrated by the surface’s inclining slope, along with a color gradient progression from blue (which represents a lower flexural modulus) towards yellow (representing a higher flexural modulus). This graphical representation emphasizes the positive relationship between these two variables and the performance of the material. The plot also detects a certain area where the flexural modulus is at its highest value, emphasizing the need to optimize the levels of pressure and filler percentage with great care. This observation trend of flexural modulus increase with the filler content is in accordance with earlier works. Such behavior is due to the enhancement in mechanical properties by reason of addition of fillers towards increasing stiffness and rigidity to the composite system. Addition of filler contributes to the enhanced stress-carrying capacity of the matrix, ultimately yielding higher flexural modulus values.^28,29

Table 11.

Flexural modulus of the fabricated composites.

Run	Flexural modulus
	Sample 1	Sample 2	Sample 3	Average	Std. Deviation (%)
1	1.552	1.623	1.658	1.611	2.74
2	1.572	1.637	1.669	1.626	2.48
3	1.627	1.685	1.715	1.676	2.18
4	1.477	1.527	1.553	1.519	2.08
5	1.480	1.525	1.547	1.517	1.84
6	1.989	2.081	2.128	2.066	2.79
7	1.477	1.529	1.555	1.520	2.13
8	1.736	1.812	1.850	1.799	2.63
9	1.800	1.853	1.879	1.844	1.78
10	1.442	1.498	1.526	1.489	2.35
11	1.571	1.628	1.656	1.618	2.19
12	1.639	1.714	1.751	1.701	2.74
13	1.243	1.302	1.332	1.293	2.86
14	1.626	1.679	1.705	1.670	1.97
15	1.961	2.017	2.045	2.008	1.74
16	1.578	1.623	1.646	1.616	1.75
17	1.907	1.967	1.996	1.957	1.89
18	1.658	1.728	1.764	1.717	2.56
19	1.401	1.448	1.471	1.440	2.02
20	1.966	2.023	2.051	2.013	1.76
21	1.737	1.819	1.860	1.805	2.83
22	1.379	1.435	1.462	1.425	2.43
23	1.999	2.087	2.131	2.073	2.65
24	1.397	1.454	1.483	1.445	2.47
25	1.478	1.537	1.567	1.527	2.42
26	1.580	1.628	1.653	1.620	1.87
27	1.594	1.666	1.703	1.654	2.74

When flexural testing is considered, percentage elongation at break depends upon four major factors and their interaction with the quadratic terms of each factor to a great extent. A closer inspection of the inherent mathematical equation will show that every input parameter depends inversely upon the elongation at break of the composite material. In particular, the higher the levels of these parameters, the lower is the elongation at break. This reveals that all of the parameters tested make a negative contribution to the stretching capacity of the material up to the point of breaking and consequently reduce the ductility. These parameters, where their level is increased at the expense of the material, generally making it more rigid or brittle, and thereby reducing the elongation potential of the material. Moreover, the interaction terms (which describe how two or more factors together influence the elongation), mostly non-significant and quadratic terms (which account for non-linear effects of individual factors), these higher-order terms allow for a more nuanced understanding of how the parameters work both independently and in combination to affect the material’s mechanical performance. It is also noteworthy that, as indicated in equation (15), the majority of the interaction terms, though predominantly non-significant, and quadratic terms exhibit a positive influence on the composite’s characteristics. The only exceptions to this trend are the interactions between factors AC and CD, both of which were found to be non-significant in terms of their effect on the material’s performance. Equations (15) and (16) outline the mathematical relationships describing the elongation percentage of the composite material. In equation (15), the coded equation offers a standard means of determining the effects of any input factors, making it easy to contrast and condense in the experimental analysis. On the other hand, equation (16), which is the working equation, provides real values for these normalized interactions and their practical relationships.

$Coded\hspace{0.17em}Equation:2.6-0.61*A-0.316667*B-0.5825*C-0.524167*D+0.4375*AB-0.0025*AC+0.095*AD+0.1075*BC+0.025*BD-0.0475*CD+0.47*A^2+0.425*B^2+0.56125*C^2+0.27875*D^2$

(15)

$Actual\hspace{0.17em}Equation:190.798-1.061*A-1.66754*B-0.363625*C-0.372542*D+0.004375*AB–0.000013*AC+0.00095*AD+0.0005375*BC+0.00025*BD-0.0002375*CD+0.0047*A^2+0.00425*B^2+0.00140312*C^2+0.0027875*D^2$

(16)

The Analysis of Variance (ANOVA) table (Table 12) provides a descriptive view of the effects of factors on the output responses. The obtained results of the percent of elongation for flexural testing are given in Table 13. Additionally, Figure 8 provides a three-dimensional visualization of the interaction between the two most influential parameters, namely Filler Percentage and Pressure, offering a comprehensive illustration of how their joint effects contribute to the material’s flexural strength. This 3D surface plot shows the relationship between filler percentage (X-axis), pressure (Y-axis), and percentage elongation at fracture, PE (%) (Z-axis). As the filler percentage increases and the pressure increases, the percentage elongation at fracture generally decreases, as indicated by the slope from red (high value of elongation to blue value of elongation). This suggests that higher filler percentages and high pressure reduce the material’s ability to stretch before breaking. The plot highlights how varying these parameters can impact the material’s flexibility or ductility, with lower elongation at higher filler and lower temperature values.

Table 12.

ANOVA table for percent elongation.

Source	Sum of Squares	df	Mean Square	F- value	p- value		Fit Statistics
Model	16.24	14	1.16	81.50	<0.0001	Significant	R2	0.9896
A- filler percentage	4.47	1	4.47	313.79	<0.0001		Adjusted R2	0.9775
B- melting temperature	1.20	1	1.20	84.56	<0.0001		Predicted R2	0.9409
C- pressure	4.07	1	4.07	286.14	<0.0001
D- speed	3.30	1	3.30	231.70	<0.0001
AB	0.7656	1	0.7656	53.80	<0.0001
AC	0.0000	1	0.0000	0.0018	0.9673
AD	0.0361	1	0.0361	2.54	0.1372
BC	0.0462	1	0.0462	3.25	0.0966
BD	0.0025	1	0.0025	0.1757	0.6825
CD	0.0090	1	0.0090	0.6342	0.4413
A2	1.18	1	1.18	82.79	<0.0001
B2	0.9633	1	0.9633	67.70	<0.0001
C2	1.68	1	1.68	118.06	<0.0001
D2	0.4144	1	0.4144	29.12	0.0002
Residual	0.1708	12	0.0142
Lack of fit	0.1670	10	0.0167	8.79	0.1064	Not significant
Pure error	0.0038	2	0.0019
Cor total	16.41	26

Table 13.

Percentage of elongation of the fabricated composites.

Run	Flexural percentage of elongation
	Sample 1	Sample 2	Sample 3	Average	Std. Deviation (%)
1	3.361	3.516	3.594	3.490	2.77
2	4.582	4.724	4.795	4.700	1.88
3	3.339	3.461	3.522	3.440	2.21
4	2.499	2.597	2.646	2.580	2.37
5	2.490	2.587	2.635	2.570	2.35
6	2.418	2.505	2.548	2.490	2.17
7	4.630	4.811	4.901	4.780	2.36
8	3.003	3.144	3.214	3.120	2.81
9	3.222	3.316	3.363	3.300	1.78
10	3.405	3.532	3.595	3.510	2.25
11	4.634	4.834	4.934	4.800	2.60
12	4.314	4.514	4.614	4.480	2.78
13	4.467	4.651	4.743	4.620	2.48
14	3.707	3.807	3.857	3.790	1.65
15	2.447	2.547	2.597	2.530	2.46
16	3.099	3.245	3.318	3.220	2.83
17	2.957	3.057	3.108	3.040	2.06
18	2.216	2.293	2.332	2.280	2.11
19	3.143	3.236	3.283	3.220	1.81
20	2.605	2.720	2.777	2.700	2.65
21	4.044	4.160	4.217	4.140	1.74
22	3.224	3.376	3.451	3.350	2.82
23	2.215	2.294	2.333	2.280	2.15
24	3.275	3.426	3.501	3.400	2.76
25	2.560	2.669	2.723	2.650	2.56
26	2.586	2.711	2.774	2.690	2.90
27	3.741	3.872	3.938	3.850	2.13

Figure 8.

3D interaction graph between Filler Percentage and Pressure for Percentage of Elongation (Flexural).

Surface Morphology

The SEM analysis emphasizes the importance of filler properties, dispersion quality, and filler-matrix adhesion in designing composites with optimal mechanical performance. To understand the mechanical performance and failure mechanisms of a composite material, detailed microscopic analysis of its structure is essential. In this research work, scanning electron microscopy was employed to study the fractured surfaces of composite specimens, which helps gain understanding regarding their microstructure and fracture behavior.

Key Aspects from SEM Analysis.

• Morphological Characteristics of the Filler

As shown in Figure 9(a), and (b), the crystallinity of the filler is of critical importance. Such crystalline features would certainly carry a certain distinctive property, capable of influencing the mechanical performance of the composite significantly. From structural features like these, it is implied that the filler is one of the contributors toward reinforcing the composite by giving strength and stability.

• Filler Dispersion and Compatibility

Figure 9.

SEM Micrograph of the Composite Specimens.

Figure 9(c)-(f), show a good dispersion of the filler in the composite matrix. Equal dispersion of filler suggests an adequate blending, which is paramount to achieving homogeneity in mechanical properties of the bulk. The proper dispersion reduces weak-link formations, thus improving the reliability of the composite under mechanical stresses. Powder pullouts did form, but there were also regions where the filler permanently adhered to the binder. These are zones where it can be inferred that the filler did bundle into the matrix, indicating good interfacial bonding, thus imparting resistant features to the composite when subjected to force.

• Microstructural Behavior Under Load

The pull-outs visible in the micrographs give pertinent insights into the behavior of composite materials under different applied stresses. They show that stress distribution and bonding strength at the filler-matrix interface are critical factors influencing failure. The strong adhesion regions increase the composite’s strength against load application, while localized detachment indicates potential failure points.

Optimization of Process Parameters

In the present section, the experimentation for the optimization of the process parameters is continued using the Response Surface Optimization (RSO) technique. The principle RSO behind this is to derive individual desirability values for the respective response parameters and weigh them according to the objective importance. Then, these desirability values are aggregated into a composite desirability function representing the overall desirability of a multi-response system.

Steps in Optimization by RSM.

• Calculation of desirability values for each output response so as to measure how close each is to its respective target value.

• Weighting of the responses to prioritize objectives in the optimization problem.

• Combination of individual desirability values into a single composite function that indicates the performance of the system overall.

The output responses are thus optimized by varying the input parameters concerned in such a manner that the optimal combination (based upon the defined criterion) can be used. The optimal input parameters acquired with this method for maximize criteria (Criterion 1) for every output parameter are: Filler percentage: 30, Melting temperature: 185°C, Pressure: 70, Speed: 40.684, with a desirability of 0.777.

However, from the experimental investigation, it was observed that the addition of filler produced different output response patterns, while certain characteristics of the composite increased, others exhibited a declining trend. Consequently, an approach was explored to optimize the output results according to the behaviour of filler is also checked, by defining maximize and minimize criterion for different output responses. For the (Criterion 2) Criteria of TS (Min), TM (Max), PE (Min), FS (Min), FM (Max), and PE (Min) the obtained input parameters are: Filler percentage: 30, Melting temperature: 178.619°C, Pressure: 104.238, Speed: 53.342, with a desirability of 0.920. And, For the (Criterion 3) Criteria of TS (Max), TM (Min), PE (Max), FS (Max), FM (Min), and PE (Max) the obtained input parameters are: Filler percentage: 17, Melting temperature: 165°C, Pressure: 70, Speed: 60, with a desirability of 0.904. While the above criterion were given based on the observed effect of filler percentage on the characteristic of the composite (taken into consideration for confirmation analysis), different criterion produces different sets of optimal input parameters with different desirability levels.

Confirmation Analysis

The confirmation analysis, was conducted to assess the accuracy and reliability of the RSM model. ³⁰ The confirmation analysis indicates the validation of the RSM model based on the predicted values and the experimental results is within the limit. The test conditions employed for the above analysis were chosen from obtained input parameters for various criterion of output parameters. In order to verify the performance of the model, the actual experimental data were compared with the predicted values from the RSM model. The maximum percent error for each experiment setup was determined using equation (17) to estimate the predictive ability of the model. This calculation yields a value of deviation from predicted to experimental, and this tells us how accurate the model is. Table 14 illustrates the results of the analysis that this presents, with the efficacy of the RSM model being such that it effectively predicts the result within reasonable margins of error. Therefore, the RSM model is a consistent and proven model that has been developed and validated and proved to have high accuracy in predictions. For the sake of the table, A stands for the Filler Percentage, B for the Melting Temperature, C for the Pressure, and D for the Speed. This notation simplifies the interpretation of the input parameters.

$\%\hspace{0.17em}Error=\left|\frac{Exprimental\hspace{0.17em}value-Predicted\hspace{0.17em}value}{Exprimental\hspace{0.17em}value}\right|\times 100$

(17)

Table 14.

Confirmation analysis table for RSM.

Criteria	A	B	C	D	% error for TS	% error for TM	% error for PE	% error for FS	% error for FM	% error for PE
Criterion 1	30	185	70	40	−4.775	2.301	4.621	−2.220	3.014	−4.464
Criterion 2	30	175	100	50	−2.447	1.795	3.820	3.734	−1.010	2.086
Criterion 3	17	165	70	60	3.842	−2.385	2.934	−1.832	3.363	0.462

Conclusions

The present research study dwelled into the utilization of Pista Shell, an agro waste, a hard compound which is of no use, but rich in crystallinity, as a bio filler into PLA to produce Sustainable Green Composites through Injection Molding Technique. The fabrication process employed the Box-Behnken Design to optimize input parameters, while the properties of the resulting green composites were further refined using Response Surface Methodology (RSM). The following insights can be drawn from the experimental study.

• ANOVA results also show the addition of pista shell has a significant effect on the materials characteristics alongside the machining parameters indicating that the addition of pista shell filler (an agro waste) can give a value addition to the composites, and also that the machining input parameters have a significant impact on the composite’s characteristics, with both quadratic and interaction terms showing a notable effect on output parameters.

• The Incorporation of pista shell filler significantly affected the properties of the composite and hence can be considered a potential filler material to synthesize green composites. Composites with high percentage of filler has shown a high modulus characteristics for both tensile as well as flexural characterization, whereas the strength of the composite showed a reverse trend.

• The maximum tensile strength of 35.74 MPa was achieved using a 20% filler content, while the minimum, 15.16 MPa, was realized at 30% filler. For tensile modulus, the highest value of 2493.65 GPa was realized for the 30% filler composite, while the lowest of 1992.02 GPa was realized by the 10% filler composite. In addition, the elongation percentage showed a definite fall with an increase in filler percentage, indicating a high reverse relationship between filler content and material flexibility. Such values are looked into from the filler percentage point of view, since the input machining parameters also play an influential role in the final output property.

• In flexural properties, the maximum flexural strength of 114.65 MPa was recorded at 10% filler and the minimum, 43.98 MPa, at 30% filler. In flexural modulus, the maximum of 2.065 GPa and 2.072 GPa were recorded at 30% and 20% filler, respectively, which were affected by the changing machining input parameters. On the other hand, the lowest value of 1.29 GPa was recorded at 10% filler. Similar to tensile strength, the percentage elongation trend was also the same, reducing with increasing filler content. These are indicative of the influence of the filler percentage, with machining input parameters also exerting a significant effect in defining the end product properties.

• The optimal input parameters acquired through the multi parameter optimization for output varied on the effect of input parameters on the output responses. The results of the confirmation tests aligned closely with the predicted values from RSM, demonstrating the model’s accuracy and reliability.

Limitations of the Study

Batch to Batch Variability

When natural fillers are employed in PLA composites, seasonal fluctuations of plant growth can influence the filler properties. Natural fillers collected at various points of the year may have varying important properties including moisture content, fiber thickness, and level of lignification. For example, increased moisture content can decrease compatibility with the PLA matrix, resulting in less durable composites. To address seasonal variations, three batches of natural fillers were taken and three batches of composite samples were fabricated maintaining a constant injection molding parameters to find out the batch to batch variability. While the process parameters used were, 10% Filler Concentration, Melting Temperature of 165 (^oC), Injection Pressure of 70 (MPa) and Injection Speed of 70 (mm.s⁻¹). The results obtained from showed a standard deviation in the range of 7%–12% across various properties from batch to batch fillers. While in the present study, the batch to batch variability has been tested for three samples across three different seasons from the same supplier, the optimization becomes more challenging when the supplier changes, as the properties of natural fillers changes from region to region.