An enhancement in the tensile modulus and bending resistance of polylactic acid/carbon nanotube composite by optimizing FFF process parameters

Abstract

Fused-filament fabrication (FFF) is one of the most common 3D printing methods for thermoplastic polymers and composite materials because it is easy to use and is low-cost. The printed polymer parts for industrial applications require desirable mechanical properties. Therefore, in the present research, the process parameters of fused filament fabrication are optimized to enhance the Young’s modulus and bending resistance of polylactic acid/carbon nanotube (PLA/CNT) composite. For this purpose, the response surface method (RSM) and desirability function technique (DFT) are applied to find the optimal values of the effective parameters of CNT content, printing speed and nozzle temperature. The printed samples were examined by using DSC, TGA and SEM analyses. The results of DSC and TGA analyses indicated that the addition of CNT into PLA enhanced the thermal stability of PLA/CNT composite. It was also observed from the optimization results that the Young’s modulus and bending resistance of PLA/CNT composite improved at CNT content of 2.9 wt%, printing speed of 20 mm/s and nozzle temperature of 210°C.

Keywords

Young’s modulus bending resistance optimization PLA/CNT composite

Introduction

Polylactic acid (PLA) is a biocompatible and biodegradable thermoplastic polymer material that has higher mechanical strength than other thermoplastics, and also has good processability.¹ This polymer is widely used in the packaging industry, food industry and various medical applications such as tissue engineering, drug delivery systems, membranes, implants, scaffolds, and sutures.^2–4 The main advantages of polylactic acid include biocompatibility, biodegradability, easy processing and high mechanical strength compared to other polymers.¹ According to the mentioned properties, PLA as a matrix material has great potential in the development of nanocomposite.⁵ Despite the numerous advantages of PLA, its inherent brittleness and poor thermal stability have restricted its widespread applications.^6,7 Therefore, an additive is needed to ensure the stiffness of the PLA matrix.⁸ One of the excellent additives that simultaneously improves the mechanical properties of polymers is carbon nanotubes (CNTs).^8,9 It is well established that CNTs have good potential for industrial applications because of their unique structure and properties such as excellent mechanical, electrical, and thermal properties.¹⁰ Yang et al.¹¹ reported that the addition of 6 wt% CNT improved the tensile strength and bending resistance of PLA/CNT by about 64.12% and 29.29%, respectively. Zhou et al.⁶ observed that the impact energy, elongation and tensile strength of PLA enhanced by addition of CNTs. They also stated that the thermal stability of PLA/CNTs composite increased than pure PLA. Vidakis et al.⁸ found that the mechanical properties of PLA/CNT composite were significantly improved by incorporation of 1 wt% CNTs. Pinto et al.^9,12 have also found that the mechanical properties of PLA/CNT composite were improved by incorporation of CNTs. Ristić et al.¹³ stated that the addition of CNTs into PLA enhanced both the tensile strength and Young’s modulus of the composite. Lee et al.¹⁴ reported that the Young’s modulus and thermal stability of the ternary PP/PLA/CNT composite improved with the increase of CNT loading. Guo et al.¹⁵ observed that incorporation of CNTs into PLA improved the flammability and thermal properties of the PLA/CNT composite. Rivière et al.¹⁶ reported that with the increase of CNT content up to 5 wt%, the tensile strength of PLA/CNT composite decreased by 26% compared to neat PLA, due to the aggregation of the nanoparticles. Therefore, although the addition of CNTs improves the thermal and mechanical properties of the PLA/CNT composite, but excessive amounts of CNTs lead to the aggregation of the nanoparticles. Therefore, the optimal amount of CNTs should be determined.

In recent years, great attention has been paid to the fabrication of the polymer composites using 3D printers.^17–19 One of the simplest and cheapest techniques for fabrication of polymer-based composites is the fused filament fabrication (FFF) method. In the FFF process, the polymer filaments are placed in a heated nozzle to melt. The nozzle extrudes a semi-solid polymer onto a substrate to create the desired 3D part with deposited layers.²⁰ Usually, the mechanical properties and manufacturing quality of the part produced by the FDM process are poorer than the part produced by conventional manufacturing techniques, which is due to the microstructural anisotropy and the layer-by-layer effect of the forming process.¹¹ Therefore, it is required to improve the mechanical properties and manufacturing quality of the printed parts by optimizing the process parameters. Many attempts have been made to improve the properties and quality of 3D printed parts. The quality of a printed part depends on a number of parameters which control the performance of the printed part.²¹ Anis et al.²² reported that the dimensional quality of 3D printed PLA parts is significantly affected by nozzle temperature and printing speed. Mohammadimehr et al.²³ stated that a reduction in the layer thickness leads to an increase in the dimensional accuracy and surface quality of the PLA samples, while the printing speed leads to an improvement in the mechanical properties. Chacon et al.²⁴ concluded that the increase of layer thickness and the decrease of printing speed improved the mechanical properties of PLA. Ahmed et al.²⁵ observed that the higher tensile strength of CF-PLA/ABS composite was obtained at lower values of printing speed. Magri et al. ²⁶ stated that the tensile properties of both PLA and PLA-CF improved at a nozzle temperature of 230°C. Khosravani et al.²⁷ reported that the higher Young’s modulus and tensile strength are obtained at printing speed of 20 mm/s, however Kuznetsov et al.²⁸ reported that highest tensile strength is obtained at printing speed of 25 mm/s and nozzle temperature of 210°C. In another studies, Altan et al.²⁹ found that the tensile strength of PLA improved at printing speed of 80 mm/s, however Cardoso et al.³⁰ stated that the lower printing speed improved the tensile strength of PLA. Benwood et al.³¹ have also demonstrated that the increase of bed and nozzle temperatures enhanced the impact energy, tensile strength, tensile modulus and bending resistance of PLA. The effect of 3D printing parameters on the mechanical properties of PLA was also examined by Hsueh et al.³² Wach et al.³³ Liparoti et al.³⁴ Vanaei et al.³⁵ Deshwal et al.³⁶ Veeman et al.³⁷ Krishnakumar et al.³⁸ Kumar et al.³⁹ Akhoundi et al.⁴⁰ and Thakur et al.⁴¹

A review of previous works shows that some researchers have tried to produce PLA/CNT composites by compression molding process.^8–16 Some researchers have also produced PLA samples by FFF method and investigated their mechanical and thermal properties.^22–41 Recently, some researchers have produced PLA/CNT composite by FFF method to investigate its mechanical and thermal properties.^8,11 However, in previous studies, researchers have not investigated the effect of different parameters of FFF process on the mechanical and thermal properties of PLA/CNT composites. In addition, optimizing the mechanical properties of the composite has not been performed in previous studies. Therefore, the primary purpose of the present study is to produce the PLA/CNT composite using the fused filament fabrication process. Moreover, the response surface methodology (RSM) is used to investigate the effect of CNT content, nozzle temperature and printing speed on the mechanical properties of the PLA/CNT composite. The desirability function technique (DFT) is also used to determine the optimum values of the FFF parameters for simultaneous enhancing the Young’s modulus and bending resistance of the PLA/CNT composite. Moreover, the thermal properties and microstructure of printed samples are evaluated to justify the variations of the mechanical properties of the printed samples.

Materials and methods

Materials

The polylactic acid (PLA), used in this study, was an industrial grade (PLA 3052D) in the form of a fine powder supplied from Plastika Kritis S.A (Heraklion, Crete, Greece). The molecular weight and Melt Flow Index of PLA were 116,000 g/mol, and 14 g/10 min (ASTM D1238), respectively. The carbon nanotubes (CNTs), supplied from INP Corporation, have a purity of more than 90%. They have a specific surface area of 250–300 m²/g, an average diameter of 9.5 nm and an average length of 1.5 μm.

Fused filament fabrication

The fused filament fabrication (FFF) process was used in this study to fabricate the PLA/CNT composite. To blend polylactic acid (PLA) with carbon nanotubes (CNTs) at different loadings (0, 2, and 4 wt %), a mechanical mixer was applied. Before the filament extrusion, powder compounds of the PLA/CNT composite were dried by placing in an oven at 80°C. A filament extruder (Filabot EX2, USA) was applied to produce filament with a diameter of 1.5 mm. The extrusion of filament was performed at 205°C with a rotational speed of 3.5 r/min. The optimal values of filament extrusion parameters and melt mixing process were chosen by conducting trial experiments and studying previous works.^1,8 Before conducting each test, the morphology of filaments was checked visually. The fused filament fabrication (FFF) process was implemented by a 3D printer (Model Wanhao D12-230) to print the composite samples. The FFF process was conducted at different values of printing speed and nozzle temperature. The schematic view of the FFF process was also displayed in Figure 1.

Figure 1.

FFF 3D printer used in this study.

Compression molding process

The PLA/CNT composite was also produced by the compression molding process to measure the mechanical properties of the obtained samples. Then, the mechanical properties for compression molded PLA/CNT samples were compared with the mechanical properties for 3D printed PLA/CNT samples. At first, melt compounding process was used to mix 2 wt% CNTs with neat PLA in a co-rotating screw extruder (ZSK 25 P8.2 E WLE) at 200°C for 15 min. Then, the compounded materials were hot-pressed at temperature of 210°C in a Carver Laboratory press (Carver, Inc., Wabash, IN, USA) for 10 min under a pressure of 3.9 MPa to obtain tensile and bending specimens.

Design of experiments

The experiments were designed based on the response surface methodology (RSM) to study the influence of the parameters on the mechanical properties of PLA/CNT composite. At first, various parameters were considered to select the most important parameters. The design space of the input parameters was shown in Table 1. The range of the selected parameters was determined based on the previous research. The parameters that have a greater effect on the Young’s modulus and bending resistance are considered as variables and the less important parameters are kept constant. The constant parameters were determined after conducting preliminary experimental tests. Table 2 indicates the constant parameters of 3D printer process. The values of these constant parameters were chosen in such a way that the maximum mechanical strength is obtained. After determining the significant parameters, the Box-Behnken design was chosen in the RSM method to find a relationship between the independent variables (CNT content, scanning speed and nozzle temperature) and the output responses (tensile modulus and bending resistance). The levels of the selected parameters were determined based on the previous research and conducting the primary experiments. According to RSM method, three levels were considered for each parameter, as shown in Table 3. In the RSM method, 3 center points were selected, and therefore 15 experiments have been implemented, as given in Table 4. The values of Young’s modulus and bending resistance have also measured at different conditions of the process parameters and given in Table 4.

Table 1.

Design space of input parameters.

Parameters	Range	Parameters	Range
CNT content	0–4 wt%	Bed temperature	40–80°C
Air gap	0.05–0.15 mm	Nozzle diameter	0.2–0.5 mm
Layer thickness	0.1–0.5 mm	Nozzle temperature	180–220°C
Printing speed	20–60 mm/s

Table 2.

Constant parameters of FFF process.

Parameters	Nozzle diameter (mm)	Air gap (mm)	Layer thickness (mm)	Bed temperature
Values	0.4	0.05	0.2	80°C

Table 3.

Different levels of the selected parameters.

Parameters	Levels
Parameters	Low (−1)	Middle (0)	High (+1)
CNT content (wt%)	0	2	4
Printing speed (mm/s)	20	40	60
Nozzle temperature (ºC)	180	200	220

Table 4.

Experimental results in design of experiment.

Run	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
CNT content (wt%)	0	4	0	4	0	4	0	4	2	2	2	2	2	2	2
Printing speed (mm/s)	20	20	60	60	40	40	40	40	20	60	20	60	40	40	40
Nozzle temperature (°C)	200	200	200	200	180	180	220	220	180	180	220	220	200	200	200
Young’s modulus (MPa)	2745	3442	2587	2718	2429	2763	2874	3182	3616	3117	3696	3632	3865	3837	3846
Bending resistance (MPa)	43.7	54.2	33.4	41.2	32.6	37.1	32.4	50.5	48.9	34.9	49.7	44.9	42.9	41.7	43.4
Void fraction (%)	7.8	8.1	5.2	5.3	4.8	5	6.8	6.7	6.6	4.6	7.1	6.5	7.2	7.3	7

Mechanical and thermal tests

According to ASTM D-638 standard, the tensile test specimens were printed by the FFF process. Some of the tensile specimens were shown in Figure 2. It can be observed that all the samples broke outside of the gauge length because stress concentration occurred at the fillet radius of the samples, where the geometry changed.^42,43 The tensile test was carried out using a Zwick/Roell-Z100 machine with a crosshead speed of 50 mm/min at room temperature. The bending samples were also printed by FFF process based on ASTM D-790. The tensile and bending tests were also performed on the compression molded samples.

Figure 2.

The printed samples for tensile test.

The differential scanning calorimetry (DSC) analysis was carried out (NETZCH 200 F3, Maia) to determine the crystallization temperature (T_c) and melting temperature (T_m) of printed samples. A nitrogen environment was used to perform DSC analysis with a scan rate of 10°C/min at temperatures of 0 to 200°C. It was performed by heating-cooling-heating cycles. The crystallinity (%) of the samples was calculated by the following equation:

C r y s t a l l i n i t y (%) = (\frac{{∆ H}_{m}}{{∆ H}_{L}}) \times 100 %

(1)

In this equation,

{∆ H}_{L}

indicates the enthalpy for a perfect crystal of PLA (207 J/g) and

{∆ H}_{m}

indicates the measured enthalpy of melting. The thermogravimetric analysis (TGA) was also performed by using a TA instrument 2960 (New Castle, DE, USA). TGA test was conducted in a nitrogen atmosphere from 25 to 600°C by a heating rate of 10°C/min. In this study, it was tried to remove the effect of crystallinity by annealing the printed samples to a constant level of crystallinity. For this purpose, the printed samples maintained at 85°C for 70 min to complete the annealing.

Microstructure analysis

The scanning electron microscopy (VEGATESCAN, Czech Republic) was used to examine the fracture surface of tensile specimens. The surface of specimens was coated with a gold layer to take SEM images in a vacuum atmosphere at a voltage of 20 kV. SEM analysis was also used to detect the dispersion of the CNTs inside the PLA matrix. To examine the quality of the FFF samples, the void fraction of samples was measured at different conditions. First, the cross-section of samples was observed by an optical microscopy (Model Olympus CX23), and then the void fraction of samples was measured by ImageJ software.

Results and discussion

Analysis of thermal properties

The DSC analysis was performed to investigate the thermal properties of PLA/CNT composite. The results of DSC for PLA/CNT composite were presented in Figure 3. From the results of DSC analysis, the thermal properties of PLA/CNT composite were obtained, as given in Table 5. Table 5 indicates that an increase of CNT content from 0 to 4 wt% was followed by an improvement in the melting and crystallization temperatures of the composites. The addition of CNTs into PLA matrix increases the thermal conductivity of the composite, and consequently improves the heat transfer and thermal stability of the PLA/CNT composite, because the thermal conductivity of CNTs is more than PLA.^11,44,45 In addition, DSC results in Table 5 demonstrate that the addition of CNT up to 4 wt% improved the crystallinity percentage of the composite, because CNT nanoparticles acts as nucleation sites for PLA molecules.⁴⁶

Figure 3.

DSC analysis for cooling and heating cycles.

Table 5.

Results of thermal analysis for PLA/CNT composite.

Thermal properties	PLA +0 wt% CNT	PLA +2 wt% CNT	PLA +4 wt% CNT
Crystallization temperature (°C)	115.4	118.6	120.4
Crystallinity (%)	49.6	52.3	54.2
Melting temperature (°C)	169.1	171.5	172.7

The thermogravimetric analysis was also performed to study the thermal behavior of PLA/CNT composite in temperature range of 0 to 600°C, as shown in Figure 4. It can be observed from Figure 4 that the thermal stability of the composite enhanced with the addition of CNTs in the PLA matrix. PLA matrix undergoes thermal degradation, which begins at 298°C with a continuous mass loss up to 541°C. However, the decomposition temperature of the PLA/CNT composite enhanced up to 311 and 338°C after addition of 2 and 4 wt% CNTs in the PLA matrix. The improvement of the decomposition temperature of PLA/CNT composite is owing to the limitation of the mobility of polymer chains by CNT nanoparticles .^47,48 It was also reported by other researchers^6,49 that the decomposition temperature of PLA/CNT composite improved by the increase of CNT content.

Figure 4.

Influence of CNT content on thermogravimetric results.

The thermal stability of the composite was studied by the weight loss temperature at 10, 50 and 90 %, as shown in Table 6. As clear from Table 6, the increase of CNT content resulted in an enhancement in the thermal stability of the composite. The increase of the decomposition temperature of PLA polymer by the addition of CNTs was also reported in.^47–49

Table 6.

The effect of CNT content on Thermogravimetric results of the composite.

Temperature	PLA +0 wt% CNT	PLA +2 wt% CNT	PLA +4 wt% CNT
T₁₀ (°C)	356	379	393
T₅₀ (°C)	418	442	455
T₉₀ (°C)	473	488	501

Analysis of microstructure

The interlayer adhesion in the FFF process is very important, because the failure usually occurs at the interface of deposited layers. Thus, the microstructure and mechanical properties of the printed samples are analyzed to investigate the adhesive between the deposited layers. The microstructure of tensile specimens at different percentages of CNT content was shown in Figure 5. It is obvious from Figure 5 that the nanoparticles was well dispersed in the PLA matrix at CNT content of 2 wt% (Figure 5(b)). A strong interfacial interaction can be attained between the CNTs and PLA matrix under this condition, which leads to an improvement in the mechanical properties of the PLA/CNT composite. However, the increase of CNT content up to 4 wt% was accompanied with the aggregation of the nanoparticles (Figure 5(c)). The aggregation of CNTs results in the poor dispersion of the nanoparticles in the PLA matrix, thereby giving rise to deterioration of the mechanical properties of the composite.

Figure 5.

The influence of CNT content on the fracture surface of composite at: (a) 0, (b) 2, (c) 4 wt%.

Figure 6 presents the microstructure of the PLA/CNT composites printed at different printing speeds. It is apparent from Figure 6(a) that a good bonding was formed between the filament layers at the low level of printing speed (20 mm/s) due to higher heat input. In fact, when the filament is printed at lower speeds, it remains in contact with the heated nozzle for a longer period of time, which led to a rise in the filament temperature. However, an increase in the printing speed from 20 to 30 mm/s increased the repeated thermal cycles at the previous layers in lesser time, which leads to the formation of interlayer cracks²⁵ and consequently weakens the bonding strength (Figure 6(b)). Therefore, the presence of defects was declared in the sample printed at a speed of 30 mm/s. According to Figure 6(c), a rise in the printing speed up to 40 mm/s increased the presence of cracks and cavities in the microstructure of the composite due to the decrease of the cooling time and consequently the reduction of the adhesion between filament layers.

Figure 6.

The fracture surface of the composites at printing speeds of (a) 20, (b) 30 and (c) 40 mm/s.

The impact of nozzle temperature on the microstructure of tensile samples was displayed in Figure 7. It can be seen from Figure 7(a) that the adhesion between the filament layers is low at the nozzle temperature of 180°C, because the viscosity of the polymer is high. In the other words, the high viscosity of the filament led to the poor adhesion of the filament layers, which resulted in the formation of cracks and cavities in the microstructure. In fact, when the viscosity of printed filament is high, materials are not fluid enough to be effectively deposited and thus can reduce the adhesion by inhibiting polymer chain interpenetration between adjacent filaments due to the hindering filler.^50,51 Figure 7(b) shows that a rise in the nozzle temperature from 180 to 200°C improved the adhesion between the layers owing to the good viscosity of the printed filament. Therefore, the crack and cavity were not detected on the surface of the sample printed at 200°C. However, it can be observed from Figure 7(c) that the increase of nozzle temperature up to 220°C decreased the adhesion bonding between the filaments, which led to an increase in the existence of cracks in the microstructure. The reduction of adhesion between layers may be because the fluidity of the extruded material is very high, which leads to a decrease in the surface quality of the material and the formation of cracks between the layers.^52–54 Another reason for reducing adhesion between layers may be the thermal deterioration of PLA materials due to the high temperature of the nozzle.⁵² In fact, when the nozzle temperature is too high (220°C), the extruded PLA material is almost liquid or even partially thermally degraded.^52–54

Figure 7.

The fracture surface of the composites at nozzle temperatures of (a) 180, (b) 200 and (c) 220°C.

The quality of the 3D printed samples was analyzed by investigating the void fraction of samples at different values of parameters. Figure 8 indicates the void fraction of samples at different conditions. It can be observed from Figure 8 that the effect of CNT content on the void fraction is small, however the lower void fraction was obtained at 2 wt% CNT. The increase of void fraction at CNT content of 4 wt% may be due to the aggregation of nanoparticles.

Figure 8.

The void fraction of samples at different values of parameters.

Moreover, the quality of printed samples reduced with the increase of printing speed due to the increase of the void fraction. It can be also observed from Figure 8 that the lower void fraction was obtained at nozzle temperature of 200°C. The increase of the void fraction at nozzle temperature of 220°C can be due to the reduction of adhesion between layers, because the fluidity of the extruded material is very high at 220°C.

Analysis of young’s modulus and bending resistance

The responses of Young’s modulus and bending resistance were studied by a statistical technique, namely analysis of variance (ANOVA). The ANOVA technique was used to examine the effect of CNT content, printing speed and nozzle temperature on the Young’s modulus and bending resistance of PLA/CNT composite. Minitab 17 software was used to perform the analysis of variance (ANOVA) for Young’s modulus (YM) and bending resistance (BR), as the results were given in Tables 7 and 8, respectively. A parameter in the ANOVA table is considered significant when its p-value is less than 0.05. Moreover, the greater F-value of a parameter denotes the greater impact of the parameter. It can be understood from Tables 7 and 8 that all parameters of CNT content, printing speed (PS) and nozzle temperature (TE) are significant on the Young’s modulus (YM) and bending resistance (BR). Nevertheless, the F-value of the parameters indicates that the impact of CNT content on the Young’s modulus (YM) is more than the other parameters, while the printing speed (SP) showed the higher impact on the bending resistance (BR). The regression equations (2) and (3) were also extracted from the results of ANOVA for Young’s modulus (YM) and bending resistance (BR).

Y M = - 17097 + 1105 C N T - 29 P S + 196 N T - 209.94 C N T \times C N T - 0.3417 P S \times P S - 0.4935 N T \times N T - 3.535 C N T \times P S + 0.2718 P S \times N T

(2)

B R = - 42.6 - 10.68 C N T - 2.08 P S + 1.253 N T - 0.753 C N T \times C N T + 0.00863 P S \times P S - 0.00377 N T \times N T + 0.0847 C N T \times N T + 0.0058 P S \times N T

(3)

Table 7.

ANOVA results for YM.

Source	DF	Adj SS	Adj MS	F-value	p-value
Model	9	3,609,606	401,067	75.94	0.000
Linear	3	798,310	266,103	50.39	0.000
CNT	1	270,568	270,568	51.23	0.001
PS	1	261,357	261,357	49.49	0.001
NT	1	266,384	266,384	50.44	0.001
Square	3	2,683,897	894,632	169.40	0.000
CNT × CNT	1	2,603,778	2,603,778	493.03	0.000
PS × PS	1	68,958	68,958	13.06	0.015
NT × NT	1	143,864	143,864	27.24	0.003
2-Way interaction	3	127,399	42,466	8.04	0.023
CNT × PS	1	79,965	79,965	15.14	0.012
CNT × NT	1	167	167	0.03	0.866
PS × NT	1	47,267	47,267	8.95	0.030
Error	5	26,406	5281
Lack-of-Fit	3	26,001	8667	42.78	0.023
Pure error	2	405	203
Total	14	3,609,606	401,067	75.94	0.000
R² = 99.27%		R²_Adj = 97.97%		R²_Pred = 88.53%

Table 8.

ANOVA results for BR.

Source	DF	Adj SS	Adj MS	F-value	p-value
Model	9	664.672	73.852	51.98	0.000
Linear	3	502.575	167.525	117.92	0.000
CNT	1	209.234	209.234	147.28	0.000
PS	1	222.268	222.268	156.45	0.000
NT	1	71.073	71.073	50.03	0.001
Square	3	92.702	30.901	21.75	0.003
CNT × CNT	1	33.497	33.497	23.58	0.005
PS × PS	1	43.980	43.980	30.96	0.003
NT × NT	1	8.374	8.374	5.89	0.060
2-Way interaction	3	69.396	23.132	16.28	0.005
CNT × PS	1	1.906	1.906	1.34	0.299
CNT × NT	1	45.928	45.928	32.33	0.002
PS × NT	1	21.562	21.562	15.18	0.011
Error	5	7.103	1.421
Lack-of-Fit	3	5.465	1.822	2.22	0.325
Pure error	2	1.638	0.819
Total	14	671.775
R² = 98.94%		R²_Adj = 97.04%		R²_Pred = 86.43%

In order to verify the adequacy of developed regression models, the variable of R², R²_Adj and R²_Pred were applied. The predictive ability of a regression model increases as the values of R², R²_Adj and R²_Pred approach 100%. As can be seen from Tables 7 and 8, the values of R², R²_Adj and R²_Pred for YM and BR are close to 100%, indicating that the developed models are capable of predicting the YM and BR.

The impact of CNT content, printing speed and nozzle temperature on the Young’s modulus (YM) and bending resistance (BR) of the PLA/CNT composite was shown in Figure 9. According to Figure 9, the highest Young’s modulus of the PLA/CNT composite was attained at the CNT content of 2 wt% (Figure 9(a)), but the highest bending resistance of the composite was observed at a CNT content of 4 wt% (Figure 9(b)). The variation of the Young’s modulus and bending resistance of the PLA/CNT composite can be explained based on SEM image in Figure 5. The enhancement of the Young’s modulus at CNT content of 2 wt% is due to the fine distribution of the CNTs and thus the strong interfacial interaction between the nanoparticles and PLA polymer. Nevertheless, the aggregation of the nanoparticles at CNT content of 4 wt% led to a reduction in the Young’s modulus of PLA/CNT composite (Figure 5(c)). However, bending resistance of the composite was improved at CNT content of 4 wt%, because the CNT nanoparticles improved the resistance of the composite against bending force. Similar results were observed in.^6,11,55,56 Figure 9 also indicates that an increase in the printing speed from 20 to 60 mm/s resulted in a reduction in the Young’s modulus and bending resistance of the composite by about 11.5 and 21.8%, respectively. The enhancement of the Young’s modulus and bending resistance at printing speed of 20 mm/s is because of the formation of the good adhesion bonding between filament layers and consequently the reduction of cracks and cavities in the microstructure (Figure 6(a)). However, the increase of printing speed up to 60 mm/s decreased the bonding strength between layers, which resulted in the formation of defects in the microstructure and thus the reduction of the Young’s modulus and bending resistance of PLA/CNT composite (Figure 6(c)). The decrease of the Young’s modulus and bending resistance of PLA at higher printing speed was also reported in.^29,30,57

Figure 9.

The effect of process parameters on (a) Young’s modulus, (b) bending resistance.

It is also obvious from Figure 9 that an increase in the nozzle temperature from 180 to 200°C was associated with an improvement in the Young’s modulus and bending resistance by about 12.2 and 8.9%, respectively. However, an increase of nozzle temperature from 200 to 220°C decreased the Young’s modulus by 1.8% and improved the bending resistance 3.2%. According to SEM image in Figure 7(b), the higher values of the Young’s modulus and bending resistance at nozzle temperature of 200°C are due to the lower viscosity of the filament and consequently the higher adhesion of the filament layers. The strong adhesion of the filament layers at nozzle temperature of 200°C decreases the formation of cracks and cavities (Figure 7(b)), giving rise to an improvement in the Young’s modulus and bending resistance of the composite. In addition, based on SEM image in Figure 7(a), the deterioration of the Young’s modulus and bending resistance at the nozzle temperature of 180°C can be due to the poor adhesion of the filament layers and thus formation of defects in the microstructure. It worth mentioning that when the temperature between deposited layers is insufficient, it can hinder molecular chain fusion, which leads to formation of a weak interlayer bonding.⁵⁸ The obtained results are consistent with those reported in.^{28,33,35,59,60}

The concurrent effect of the parameters on the Young’s modulus and bending resistance was studied by considering the significant interaction in the ANOVA tables. Figure 10(a) and (b) display the interaction effect of CNT × PS and PS × NT on the Young’s modulus, and Figure 10(c) and (d) display the interaction effect of CNT × NT and PS × NT on the bending resistance. From Figure 10(a), it can be observed that when CNT content is 0 wt%, the increase of printing speed has a trivial effect on the Young’s modulus, but when CNT content is 4 wt%, the increase of printing speed decreased the Young’s modulus, because increasing the printing speed reduces the heat input to the polymer and some of the heat is absorbed by CNT nanoparticles. However, the greatest Young’s modulus of the composite was achieved at CNT content of 2 wt% and printing speed of 20 mm/s.

Figure 10.

The interaction effect of parameters on (a, b) Young’s modulus, (c, d) bending resistance.

Figure 10(b) indicates that when printing speed is 20 mm/s, the increase of nozzle temperature slightly improved the Young’s modulus, but when printing speed is 60 mm/s, the increase of nozzle temperature strongly improved the Young’s modulus. However, the higher Young’s modulus of the composite was obtained at printing speed of 20 mm/s and the nozzle temperature of 220°C. According to Figure 10(c), for CNT content of 0 wt%, the bending resistance of the composite slightly reduced with the increase of nozzle temperature, while for CNT content of 4 wt%, the bending resistance of the composite considerably enhanced with the increase of nozzle temperature. The enhancement of bending resistance of the composite at CNT content of 4 wt% and nozzle temperature of 220°C can be due to the fact that some of the heat input was absorbed by CNTs and prevented the thermal degradation of the composite. Figure 10(d) represents that as printing speed is 20 mm/s, the increase of nozzle temperature slightly enhanced the bending resistance, but as printing speed is 60 mm/s, the increase of nozzle temperature considerably enhanced the bending resistance.

The desirability function technique was employed to concurrently enhance the Young’s modulus and bending resistance of the PLA/CNT composite. According to this technique, the output responses of Young’s modulus and bending resistance are first transformed to the individual desirability function (d). The geometric mean of the individual desirability functions is then attained to estimate the combined desirability function (D). Eventually, the optimal condition of the parameters is identified by maximizing the combined desirability functions. The desirability function can have a value between zero and one. The number one is the ideal value of the desirability function. Different objectives can be considered to calculate the individual desirability function for each response: 1- higher is better, 2- lower is better, 3- a certain value is better.^20,61,62 The objective of this paper is to obtain higher values of the Young’s modulus and bending resistance of the PLA/CNT composite, therefore for estimating the individual desirability function, the first objective is considered. For the objective of “higher is better” the following equation can be calculated to obtain the individual desirability function (d_i):

d_{i} = {\begin{cases} 0 x < L \\ {(\frac{x - O}{O - L})}^{t} L \leq x \leq O \\ 1 x > O \end{cases}

(4)

where, O denotes the objective value of a response (x), L denotes the lower value of a response (x), and t-value represents the weighting factor of a response (x). By calculating the geometric mean of individual desirability functions (d_i) for Young’s modulus and bending resistance, the combined desirability function (D) is also obtained as follows:

D = \sqrt[2]{d_{1} (Y o u n g^{'} s m o d u l u s) \times d_{2} (B e n d i n g r e s i s t a n c e)}

(5)

In Eq. (5), d₁(Young’s modulus) represents the individual desirability function for Young’s modulus, d₂(Bending resistance) represents the individual desirability function for bending resistance, and D represents the combined desirability function.^20,61,62 In the present work, the individual and combined desirability functions were calculated by Minitab®17 software. Figure 11 displays the optimization plot obtained by desirability function technique. The optimization plot in Figure 11 indicates that a high combined desirability has been obtained (D = 0.99), indicating that the estimated values of the Young’s modulus and bending resistance are close to their maximum value (objective value). In addition, Figure 11 indicates that the Young’s modulus and bending resistance of the composite can be improved at CNT content of 2.9 wt%, printing speed of 20 mm/s and nozzle temperature of 210°C. It was also predicted that the Young’s modulus and bending resistance of the PLA/CNT composite improved up to 3855 MPa and 54.1 MPa at the optimal condition of the parameters. It should be explained that in optimal conditions, the CNT content is estimated to be 2.9 wt%, because more than this amount, the aggregation of nanoparticles occurs. Moreover, the printing speed of 20 mm/s was chosen in optimal conditions because increasing the printing speed reduces the temperature of the filament and thus reduces the interfacial bonding between the layers. The nozzle temperature of 210°C was also chosen in optimal conditions, because above this temperature, the fluidity of the filament layers increases too much and it causes a decrease in the quality surface of the filament and the formation of cracks.

Figure 11.

The results of optimization of parameters by desirability function method.

Several verification experiments were performed at the optimal condition of parameters to validate the accuracy of the optimization results. The results of the prediction and verification experiments were shown in Table 9. It can be concluded from the results in Table 9 that the verification experiments are in close agreement with the prediction results.

Table 9.

The values of verification tests and prediction results at the optimal condition.

Responses	Prediction results	Verification tests	Error (%)
Young’s modulus (MPa)	3855	3929	1.9
Bending resistance (MPa)	54.1	52.33	3.2

The microstructure of the printed samples has been analyzed in the optimal conditions of the parameters to be compared with other samples. Figure 12 shows the microstructure of the printed sample at the optimal conditions. As can be seen from Figure 12(a), despite the increase of CNT content up to 3 wt%, the nanoparticles are well distributed in the PLA matrix and there is no effect of aggregation. Figure 12(b) shows that at the optimal conditions of the parameters, the interfacial bonding between the filament layers increased and the presence of cracks and voids decreased.

Figure 12.

The microstructure of the printed sample at optimal conditions with magnifications of (a) 20,000 and (b) 100.

The bending resistance and Young’s modulus for the compression molded PLA/CNT samples were compared with those obtained for 3D printed PLA/CNT samples at optimal conditions, as given in Table 10. It can be observed from Table 10 that the Young’s modulus of 3D printed PLA/CNT sample improved compared to the compression molded PLA/CNT sample, while the bending resistance of 3D printed PLA/CNT sample reduced compared to the compression molded PLA/CNT sample.

Table 10.

Comparison of mechanical properties of compression molded PLA/CNT samples with 3D printed PLA/CNT at the optimal condition.

Responses	Compression molding	3dD printing
Young’s modulus (MPa)	3591	3929
Bending resistance (MPa)	96.4	52.33

Conclusion

In the present study, the Young’s modulus and bending resistance of the PLA/CNT composite improved concurrently by using the response surface method and desirability function technique. The obtained results are summarized as follows:

1- The DSC results showed that an increase in CNT content up to 4 wt% improved the melting and crystallization temperatures of the composites. Moreover, the thermogravimetric analysis indicated that the thermal stability of the composite enhanced with the addition of CNTs in the PLA matrix.

2- The highest Young’s modulus of the composite was attained at a CNT content of 2 wt% owing to good dispersion of the nanoparticles in the PLA matrix, while the Young’s modulus of the composite was maximized at a CNT content of 4 wt% owing to the absorption of some energy by nanoparticles.

3- A rise of printing speed from 20 to 60 mm/s decreased the Young’s modulus and bending resistance of the composite due to the formation of a poor adhesion bonding between filament layers and consequently the formation of cracks and cavities in the microstructure.

4- The increase of nozzle temperature from 180 to 220°C improved the Young’s modulus and bending resistance of the composite, because of the good adhesion of filament layers.

5- The results of optimization demonstrated that the Young’s modulus and bending resistance of the composite can be enhanced simultaneously at a CNT content of 2.9 wt%, printing speed of 20 mm/s and nozzle temperature of 210°C. Under this optimal condition, the Young’s modulus and bending resistance were predicted to be 3855 MPa and 54.1 MPa, respectively.

Footnotes

Declaration of conflicting interests

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

Funding

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

ORCID iDs

Hatam Hardani

Mahmoud Afshari

Data Availability Statement

The authors confirm that the data supporting the findings of this study are available within the article.

References

Petousis

Vidakis

Mountakis

, et al. Multifunctional material extrusion 3D-printed antibacterial polylactic acid (PLA) with binary inclusions: the effect of cuprous oxide and cellulose nanofibers. Fibers 2022; 10: 52.

Divakara Shetty

Shetty

. Investigation of mechanical properties and applications of polylactic acids—a review. Mater Res Express 2019; 6: 112002.

Singhvi

Zinjarde

Gokhale

. Polylactic acid: synthesis and biomedical applications. J Appl Microbiol 2019; 127: 1612–1626.

Hamad

Kaseem

Yang

, et al. Properties and medical applications of polylactic acid: a review. Express Polym Lett 2015; 9: 435–455.

Casalini

Rossi

Castrovinci

, et al. A perspective on polylactic acid-based polymers use for nanoparticles synthesis and applications. Front Bioeng Biotechnol 2019; 7: 259.

Zhou

Lei

Yang

, et al. Preparation and characterization of polylactic acid (PLA) carbon nanotube nanocomposites. Polym Test 2018; 68: 34–38.

Rosli

Ahmad

Anuar

, et al. Mechanical and thermal properties of natural rubber-modified poly(lactic acid) compatibilized with telechelic liquid natural rubber. Polym Test 2016; 54: 196–202.

Vidakis

Petousis

Kourinou

, et al. Additive manufacturing of multifunctional polylactic acid (PLA)—multiwalled carbon nanotubes (MWCNTs) nanocomposites. Nanocomposites 2021; 7: 184–199.

Pinto

Ramos

Alves

, et al. Comparative failure analysis of PLA, PLA/GNP and PLA/CNT-COOH biodegradable nanocomposites thin films. Procedia Eng 2015; 114: 635–642.

10.

Doagou-Rad

Islam

Jensen

Influence of processing conditions on the mechanical behavior of MWCNT reinforced thermoplastic nanocomposites. Procedia CIRP 2017; 66: 131–136.

11.

Yang

Zhou

, et al. Effects of carbon nanotube on the thermal, mechanical, and electrical properties of PLA/CNT printed parts in the FDM process. Synth Met 2019; 253: 122–130.

12.

Pinto

Ramos

Alves

ASF

, et al. Dispersion and failure analysis of PLA, PLA/GNP and PLA/CNT-COOH biodegradable nanocomposites by SEM and DIC inspection. Eng Fail Anal 2017; 71: 63–71.

13.

Ristić

Miletić

Vukić

, et al. Characterization of electrospun poly(lactide) composites containing multiwalled carbon nanotubes. J Thermoplast Compos Mater 2021; 34(5): 695–706.

14.

Lee

J-S

Hwang

G-H

Kwon

, et al. Influences of carbon nanotube on structures and properties of compatibilized polylactide/polypropylene blend-based ternary nanocomposites. J Thermoplast Compos Mater 2023; 36(7): 2815–2835.

15.

Guo

Xin

Zhai

, et al. Flammability and thermal properties of modified carbon nanotubes in poly(lactic acid). J Thermoplast Compos Mater 2019; 32(8): 1107–1122.

16.

Rivière

Nypelö

Obersriebnig

, et al. Unmodified multi-wall carbon nanotubes in polylactic acid for electrically conductive injection-moulded composites. J Thermoplast Compos Mater 2017; 30(12): 1615–1638.

17.

Ivanov

Kotsilkova

Xia

, et al. PLA/Graphene/MWCNT composites with improved electrical and thermal properties suitable for FDM 3D printing applications. Appl Sci 2019; 9: 1209.

18.

Dou

Wang

, et al. Three-dimensional-printed carbon nanotube/polylactic acid composite for efficient electromagnetic interference shielding. Polymers 2023; 15: 3080.

19.

Batakliev

Georgiev

Angelov

, et al. Synergistic effect of graphene nanoplatelets and multiwall carbon nanotubes incorporated in PLA matrix: nanoindentation of composites with improved mechanical properties. J Mater Eng Perform 2021; 30: 3822–3830.

20.

Taher

Afshari

Houmani

, et al. Simultaneous enhancement of the impact strength and tensile modulus of PP/EPDM/TiO2 nanocomposite fabricated by fused filament fabrication. Colloid Polym Sci 2024; 302: 393–407.

21.

Goh

Yap

Tan

HKJ

, et al. Process–structure–properties in polymer additive manufacturing via material extrusion: a review. Crit Rev Solid State Mater Sci 2020; 45(2): 113–133.

22.

Anis

Ansari

Kamil

. Effect of print speed and extrusion temperature on properties of 3D printed PLA using fused deposition modeling process. Mater Today Proc 2021; 45: 5462–5468.

23.

Farazin

Mohammadimehr

. Effect of different parameters on the tensile properties of printed Polylactic acid samples by FDM: experimental design tested with MDs simulation. Int J Adv Manuf Technol 2022; 118: 103–118.

24.

Chacon

Caminero

Garcia-Plaza

, et al. Additive manufacturing of PLA structures using fused deposition modelling: effect of process parameters on mechanical properties and their optimal selection. Mater Des 2017; 124: 143–157.

25.

Ahmed

Hussain

Al-Ghamdi

, et al. Mechanical properties of an additive manufactured CF-PLA/ABS hybrid composite sheet. J Thermoplast Compos Mater 2021; 34(11): 1577–1596.

26.

Magri

El Mabrouk

Vaudreuil

, et al. Mechanical properties of CF-reinforced PLA parts manufactured by fused deposition modeling. J Thermoplast Compos Mater 2021; 34(5): 581–595.

27.

Khosravani

Reinicke

. Effects of raster layup and printing speed on strength of 3D-printed structural components. In: Proceedings of the 1st European-Structural-Integrity-Society (ESIS) Virtual European Conference on Fracture (ECF). Funchal, Madeira, Portugal: Virtual, 2020, pp. 720–725.

28.

Kuznetsov

Solonin

Tavitov

, et al. Increasing strength of FFF three-dimensional printed parts by influencing on temperature-related parameters of the process. Rapid Prototyp J 2020; 26: 107–121.

29.

Altan

Eryildiz

Gumus

, et al. Effects of process parameters on the quality of PLA products fabricated by fused deposition modeling (FDM): surface roughness and tensile strength. Mater Test 2018; 60: 471–477.

30.

Cardoso

PHM

Teixeira

Calado

VMD

, et al. Mechanical and dimensional performance of poly(lactic acid) 3D-printed parts using thin plate spline interpolation. J Appl Polym Sci 2020; 137: 49171.

31.

Benwood

Anstey

Andrzejewski

, et al. Improving the impact strength and heat resistance of 3D printed models: structure, property, and processing correlationships during fused deposition modeling (FDM) of poly(lactic acid). ACS Omega 2018; 3: 4400–4411.

32.

Hsueh

Lai

Wang

, et al. Effect of printing parameters on the thermal and mechanical properties of 3D-printed PLA and PETG using fused deposition modeling. Polymers 2021; 13: 1758.

33.

Wach

Wolszczak

Adamus-Wlodarczyk

. Enhancement of mechanical properties of FDM-PLA parts via thermal annealing. Macromol Mater Eng 2018; 303: 9.

34.

Liparoti

Sofia

Romano

, et al. Fused filament deposition of PLA: the role of interlayer adhesion in the mechanical performances. Polymers 2021; 13: 399.

35.

Vanaei

Shirinbayan

Vanaei

, et al. Multi-scale damage analysis and fatigue behavior of PLA manufactured by fused deposition modeling (FDM). Rapid Prototyp J 2021; 27: 371–378.

36.

Deshwal

Kumar

Chhabra

. Exercising hybrid statistical tools GA-RSM, GA-ANN and GA-ANFIS to optimize FDM process parameters for tensile strength improvement. CIRP J Manuf Sci Technol 2020; 31: 189–199.

37.

Veeman

Palaniyappan

. Process optimisation on the compressive strength property for the 3D printing of PLA/almond shell composite. J Thermoplast Compos Mater 2023; 36(6): 2435–2458.

38.

. In vitro degradation analysis and mechanical characterization of PLA-CF composites prepared by fused filament fabrication technique for bio-medical applications. J Thermoplast Compos Mater 2023. DOI: 10.1177/08927057231219928.

39.

Kumar

Singh

, et al. On investigations of thermal conductivity, circumferential compressive strength, and surface characterization of 3D-printed hybrid blended magnetostrictive PLA composite. J Thermoplast Compos Mater 2022; 35(5): 631–650.

40.

Akhoundi

Nabipour

Kordi

, et al. Calculating printing speed in order to correctly print PLA/continuous glass fiber composites via fused filament fabrication 3D printer. J Thermoplast Compos Mater 2023; 36(1): 162–181.

41.

Thakur

Singh

Kumar

. On investigation of flexural, morphological, and shape memory properties for 3D printed PLA-PVDF NFs-PLA interleave composite structures for smart gripper applications. J Thermoplast Compos Mater 2024; 37(4): 1407–1428.

42.

Tymrak

Kreiger

Pearce

. Mechanical properties of components fabricated with open-source 3-D printers under realistic environmental conditions. Mater Des 2014; 58: 242–246.

43.

Gordelier

Thies

Turner

, et al. Optimising the FDM additive manufacturing process to achieve maximum tensile strength: a state-ofthe-art review. Rapid Prototyp J 2019; 25: 953–971.

44.

Mora

Verma

Kumar

. Electrical conductivity of CNT/polymer composites: 3D printing, measurements and modeling. Compos B Eng 2020; 183: 107600.

45.

Pietrzak

Raniszewski

Szymanski

. Multiwalled carbon nanotubes polylactide composites for electrical engineering—fabrication and electrical properties. Electronics 2022; 11: 3180.

46.

Niu

Wang

, et al. Enhanced nucleation rate of polylactide in composites assisted by surface acid oxidized carbon nanotubes of different aspect ratios. ACS Appl Mater Interfaces 2011; 3: 3744–3753.

47.

Banerjee

Ray

. An overview of the recent advances in polylactide-based sustainable nanocomposites. Polymer Engineering Sci 2021; 61: 617–649.

48.

Yang

Zhang

, et al. Selective localization of carbon nanotubes and its effect on the structure and properties of polymer blends. Prog Polym Sci 2021; 123: 101471.

49.

Park

Lee

Kim

. Isothermal crystallization behavior and mechanical properties of polylactide/carbon nanotube nanocomposites. Compos Appl Sci Manuf 2013; 46: 11–18.

50.

Ginoux

Vroman

Alix

. Influence of fused filament fabrication parameters on tensile properties of polylactide/layered silicate nanocomposite using response surface methodology. J Appl Polym Sci 2021; 138: 50174.

51.

Das

Gilmer

Biria

, et al. Importance of polymer rheology on material extrusion additive manufacturing: correlating process physics to print properties. ACS Appl Polym Mater 2021; 3: 1218–1249.

52.

Wang

Deng

, et al. Effects of fused deposition modeling process parameters on tensile, dynamic mechanical properties of 3D printed polylactic acid materials. Polym Test 2020; 86: 106483.

53.

Benali

Habibi

, et al. Stereocomplexed PLA nanocomposites: from in situ polymerization to materials properties. Eur Polym J 2014; 54: 138–150. DOI: 10.1016/j.eurpolymj.2014.03.004.

54.

Sin

Rahmat

Rahman

WAWA

. Thermal properties of poly (lactic acid). Polylactic acid: PLA biopolymer technology and applications. Book Series: PDL Handbook Series 2013; 10: 109–141.

55.

Urquijo

Aranburu

Dagréou

, et al. CNT-induced morphology and its effect on properties in PLA/PBAT-based nanocomposites. Eur Polym J 2017; 93: 545–555.

56.

Wang

Mei

Wang

, et al. Improved fracture toughness and ductility of PLA composites by incorporating a small amount of surface-modified helical carbon nanotubes. Compos B Eng 2019; 162: 54–61.

57.

Jaya Christiyan

Chandrasekhar

Venkateswarlu

. Flexural properties of PLA components under various test condition manufactured by 3D printer. J. Inst. Eng. (India) Ser. C 2018; 99: 363–367.

58.

Bakhtiari

Nikzad

Tolouei-Rad

. Influence of three-dimensional printing parameters on compressive properties and surface smoothness of polylactic acid specimens. Polymers 2023; 15: 3827.

59.

Behzadnasab

Yousefi

Ebrahimibagha

, et al. Effects of processing conditions on mechanical properties of PLA printed parts. Rapid Prototyp J 2019; 26: 381–389.

60.

Luzanin

Movrin

Stathopoulos

, et al. Impact of processing parameters on tensile strength, in-process crystallinity and mesostructure in FDM-fabricated PLA specimens. Rapid Prototyp J 2019; 25: 1398–1410.

61.

Taher

Hardani

Bakhshi

, et al. Enhancing the tensile properties of PA6/CNT nanocomposite in selective laser sintering process. Polym Compos 2023; 44: 1290–1304.

62.

Samadi

Ayaz

Afshari

, et al. An investigation on the friction stir welding of PP/TiO2 nanocomposites for improving the tensile strength and hardness of the weld joint. Colloid Polym Sci 2023; 301: 465–480.