Experimental investigation on mechanical characterization of 3D printed PLA produced by fused deposition modeling (FDM)

This study aims to systematically experimental investigate the influence of infill-patterns (IPs) on specific mechanical responses of parts fabricated by fused deposition modeling (FDM). A poly-lactic-acid (PLA) feedstock filament is utilized in the manufacturing process. Furthermore, six types of infill-patterns (deposition angle), namely full honeycomb, rectilinear, triangular, fast honeycomb, grid, and wiggle, are designed and printed. In order to determine the mechanical properties of manufactured parts, tensile tests are carried out. The mechanical properties such as extension, stress, elongation, energy, and Young’s modulus are considered as objective functions. As a result, there is a direct correlation between mechanical properties and infill patterns. Thus, it is essential to select the best infill-pattern in terms of their applications, giving sufficient strength without overdoing time and cost. Based on the results, a triangular infill-pattern has a maximum value of ultimate tensile strength and E-module (15.4 and 534 MPa, respectively). On the other hand, the wiggle pattern is more flexible.


Introduction
In recent years, speed up manufacturing processes has been one of the main challenges among academic and industrial centers. In this regard, rapid prototyping (RP) technologies have emerged. The RP is a technique that accelerates the process of manufacturing on different scales [1,2]. The RP is mainly conducted based on two approaches, namely 3D printing or additive manufacturing (AM), which have received increased attention across several disciplines. The AM is a computer control-based system in which different materials such as metallic and plastic parts are built layer by layer [3,4]. To date, various types of AM have been developed and introduced, for instance, selective laser sintering (SLS) [5], selective laser melting (SLM) [6][7][8], and fused deposition modelling (FDM) [9]. Among them, the FDM is the most widely used technique due to its extraordinary properties such as being a clean process, easy to operate, having excellent accuracy and repeatability, and durable manufacturing and dimensionally stabilize. Recently, this method is considered a hot topic among several scholars, which consists of highlighting the FDM technique's properties, analytically and experimentally. According to the literature [10], the main subject in this field is investigating print parameters' effect on mechanical properties and using optimization methods to achieve better properties, however, good researches have been done in other fields such as numerical analysis [11,12]. In fact, the strong impact of printing parameters on the properties of the final piece and the use of new geometric parameters such as infill patterns and raster angle is still the main topic related to the FDM process.
As an initial study, Hutmacher et al [13] investigated the mechanical characteristics of polycaprolactone (PCL) scaffolds fabricated by the FDM. Zein et al [14] modeled a novel scaffold architecture through the FDM was used and selected a honeycomb infill pattern for designing parts. In order to enhance the surface finish of parts manufactured via the FDM, Galantucci et al [15] conducted an experimental study. Kiendl  angles, and they showed that by using and selecting the raster angle, mechanical properties including strength, elongation, toughness, and even failure mechanism could be controlled. Magalhaes et al [17] used the tensile and bending tests in the FDM framework to estimate the stiffness and strength of a sandwich part. Mechanical properties of cellular lattice structures fabricated by the FDM were analyzed numerically by Karamooz et al [18]. The influence of the FDM parameters on the tensile properties of carbon-fiber-reinforced plastic composites has been studied by Ning et al [19]. An experimental investigation was performed by Durão et al [20] to optimize the FDM process parameters. They concluded that print speed and the number of contours play a crucial role in the FDM. Moradi et al [21] aimed to study the influence of the FDM process parameters on manufactured specimens. Besides, they tried to reduce build time and part weight in the FDM process.
Here, a systematic experimental analysis is designed and manufactured FDM parts to define the impact of various types of IPs on mechanical properties. One of the most critical parameters affecting this purpose is the infill-patterns. In fact, increasing mechanical properties of printed parts via the FDM play a crucial role in the field. The infillpattern's shape can significantly affect the mechanical properties, print time, and cost of the obtained part [22]. Therefore, the focus is placed on comparing the mechanical properties of printed samples based on six types of IP. In the current study, comprehensively experimental investigation of IPs on specific mechanical responses of fused deposition modeling (FDM) based on feedstock filament of poly-lactic-acid (PLA) and Six types of infill-patterns (deposition angle), namely full honeycomb, rectilinear, triangular, fast honeycomb, grid, and wiggle are considered.

Experimental work
Basically, computer-aided design (CAD) models are designed using Solid work software and converted to the stereolithography (STL) formatted file. All of the specimens were printed by Sizan3N FDM 3D printer with PLA filament. The mechanical and physical properties of the PLA selected are listed in table 1. Regarding the software and printing configurations, the toolpath calculation (G-code) was made by simplify3D Slicer Printing parameters.
The design of experiments (DOE) is correlated with preparing an efficient mixture of experiments essential to reach comprehensive investigation, which is statistically meaningful. DOE is a scientific approach that allows researchers to understand the aspects of a process better and determine how inputs (factors) affect outputs (responses). In this study, DOEs can be divided into three main phases: planning, carrying out the tests, and analyzing the results. In this current study, DOEs procedure (Taguchi model) and optimization of primary tests are considered based on previous studies [21,22]. Figure 1 shows the flowchart concept of this study.
• The first stage (Design): Applying computer-aided design (CAD) model and numerical simulation to achieve an optimum production based on chain value.
• The second stage (Manufacturing): Using the STL format file and convert it into the G-code file.
• The third step (Testing): Implementing a non-destructive test (NDT) as an inspection method to achieve a reliable specimen. Figure 2 provides an overview of the experimental works, showing the FDM process's schematic, the IPs used, and finally, the printed samples after the tensile test. In order to ensure the printing quality of the specimens, the temperature of the building platform is controlled at around 60°C, and the filament is heated to 220°C during the printing process. All printing parameters presented in table 2 were considered constant for all samples with different patterns [21].
As mentioned, the main focus is placed on comparing the mechanical properties of printed samples based on six types of IPs is explained in table 3. One of the advantages and necessities of 3D printing is that the parts can be     Full honeycomb A bee panel or honeycomb infill is hexagonal geometry to fine distribution the load and wight based on the nature concept.

Rectilinear
A rectangular pattern is a logical structure in order to have a strong network in all directions.

Triangular
A triangular pattern is a typical structure in everyday life to reach extreme resistance in the direction of the walls.
Fast honeycomb A fast honeycomb is greater than the rectangular pattern but with a longer printing time.

Grid
Grid is a self-explanatory 2D pattern to achieve an acceptable print speed.

Wiggle
This zigzag infill provides sufficient support and increases the rebound force.
hollow in different degrees. Regarding the production chain value, it not only reduces material consumption and cost but also optimizes the final weight and time management. Based on the literature, the most common types of IPs are selected as follows; full honeycomb, rectilinear, triangular, fast honeycomb, grid, and wiggle [23].
In order to define the mechanical properties of the printed materials, a uniaxial tensile test is applied to calculate key factors such as yield, ultimate force, extension, tensile strength, elongation, and absorbed energy for every 6 IPs with a Santam machine. According to figure 3(a), tensile specimens are printed based on ASTM D638 and the tensile test is repeated at least two times for each IPs to make sure the tests' accuracy. Tensile samples and printing procedure are shown in figures 3(b) and (c), respectively. The tensile speed is 0.1 mm/min, and the temperature is controlled at 23°C. The tensile test machine and specimen during and after tensile test are presented in figure 4. Here, the effect of IPs on the mechanical properties is determined based on the fibers' print direction related to the loading direction and adhesion between the layers and fibers. There are two main methods to estimate build time that calculate by software packages, detailed analysis and parametric analysis.  Build time estimations tend to be specific to the system and material being used. In fact, build time is an essential factor in additive manufacturing costs; the details of build time are beyond this study's scope. It is worthwhile noting that the built time has been estimated through the use of a digital timer. As a result, due to the layering process incorporated, there is a direct correlation between the build time and the part height.

Result and discussion
At first, the observed sample's quality was checked and ensured the expected one and matched with the set parameters are presented in tables 2 and 3. Based on observation and special shape check method, there are no voids or surface cracks over the printed layer, and surface topography of all IPs specimen has clarified the achievement of good enough printing. Due to the varieties of available Additive Manufacturing (AM) technologies, one of the most critical points in cost-effective strategy is building time. However, for FDM, other factors are also involved, such as supports, fill density, contour and raster width, etc. In this section, full factorial mechanical properties and time conditions are shown in table 4. According to the results, the printing time of all produced samples with different IPs is in the range of 72 to 75 min and there is not much difference between them. Of course, the fast and full honeycomb have less and triangular IP has more printing time, respectively. The mechanical properties are then compared in three different sections (tensile strength, elongation and absorbed energy) for the maximum and the fracture points. The triangular and grid IPs are demonstrated the maximum and minimum tensile strength, respectively. On the other hand, wiggle has the highest elongation and absorbed energy values.
Firstly, tensile strength results and Young's modulus for every six IPs are compared with each other and are summarized in figures 5-9. From these figures, it is seen that the amount of tensile force, strength, elongation, absorbed energy and Young's modulus experienced by printed samples has a significant dependency on infillpattern. Generally, according to figure 5, the strength properties of the triangular and grid patterns are the highest and lowest, respectively. The grid and triangular are the most used pattern [24]. Grid pattern has the most significant advantage over other patterns is its simplicity and less time.
Moreover, triangular is a two-dimensional grid made of triangles, while a vertical force is applied to the surface of the object, this pattern shows high resistance and strength. This pattern is also suitable for thin rectangular pieces because there is little contact between these objects' walls. According to figure 5, the geometry of triangular resembles a truss with a continuous structural element. This structure leads to a homogenous distribution of stress in FDM samples. Because the distance between joint points in the grid pattern is long and a less infill density and much-trapped air, the grid infill-pattern resists the lowest value of the stress and force [25]. It is to be noted that for other patterns, this phenomenology is analogous [22,26]. Here, the trapped air in different types of patterns is calculated and applied in design section ( figure 6).
Regarding Young's modulus, every six IPs are compared in figure 7. Young's modulus and strength variations are almost identical, with the highest and lowest values being for triangular and full honeycomb, respectively. Of course, except for the triangular pattern, which is much higher than the others are, rectilinear and wiggle are close to each other (580.4 and 571.2 Mpa), and the last three patterns (full and fast honeycomb and grid) are not much different. Of course, the ratio of the maximum Young's modulus to the minimum is equal to 1.25, which indicates the relatively strong effect of the pattern on the elastic properties of the printed specimens. Figure 8 reveals the elongation values for maximum elongation at UTS and total elongation at fracture point for each internal pattern. It could be concluded that wiggle pattern possesses the most considerable elongation and formability due to its flexible structure. Besides, the filaments in wiggle patterns are integrated and aligned, leading to a higher value of deformation. In this regard, the wiggle infill looks like wavy or zigzag lines, and aligning the filament strands with the applied force increases the ductility [14,25]. In fact, in the same direction, changing the string's geometry from the wavy to the smooth with by applying force, the elongation and formability increases [27,28]. This pattern is primarily used for flexible filaments such as softer nylon or other such rubbery filaments, where parts are designed for exceptional circumstances (twist, compress and rebound force). Moreover, this infill can still provide support to perpendicular walls.
Flexible productions in the field of engineering design and manufacturing refer to methods that can adapt when external changes occur. Some infill patterns lend themselves more to flexibility than others. The wiggle pattern resembles waves oriented in a single direction. Although every pattern has different advantages and disadvantages, the honeycomb pattern is the most practical and standard method in the 3D printing technique. In fact, the honeycomb pattern is an inspiring nature concept, which mainly in the construction of actual honeycombs. In this pattern, high infill promotes increased strength with diverse setting infill at a lower setting will also increase in a more flexible product. The honeycomb pattern strikes the perfect balance of strength and material use. Varying combinations of infill percentage and pattern can influence strength, material usage, and print times [29]. Overall, in order to improve the mechanical performance of a 3D printed part, it comes at the expense of print quality, speed, and affordability [30].
Energy absorption is another critical factor that is investigated in this study. In automotive structures, it is necessary to minimize the occurrence of damage caused by crash loads. Automobile manufacturers have applied various methods and systems to dissipate collision forces and decrease them by absorbing and redistributing them, which finally reduces their effects on the vehicle's occupants. The structures considered to have good energy absorption behavior used in the majority of transportation vehicles are the thin-walled (tubular) structures. Absorbed energy is defined as the surface below the load-displacement curve, which could be used to evaluate the toughness and the unanimity gained by the structures, especially in the automotive constructions during the frontal crash. The toughness study can also give us more accurate and useful information about the material's behavior because it includes both the parameters of strength and ductility. Therefore, the wiggle IP with lower strength has the highest absorbed energy toughness. According to figure 9, the fast honeycomb pattern and grid have the maximum and minimum absorbed energy, respectively.

Conclusion
This paper investigates different types of infill-patterns (IPs) to accomplish fused deposition modeling (FDM) parts with respect to improving mechanical characteristics and the effect of IPs on the mechanical properties. The infill-patterns prepared were full honeycomb, rectilinear, triangular, fast honeycomb, grid, and wiggle. In addition, tensile strength, elongation, absorbed energy, and Young's modulus was defined as targets. An empirical setup was implemented, and tensile tests were conducted to calculate mechanical responses. This investigation concludes that: In the triangular infill, due to the distance between joint points in the grid pattern is short. Subsequently, a less much-trapped air, the grid infill-pattern resists the highest value of the strength and Young's modulus and the maximum values of ultimate tensile strength were achieved for the triangular pattern. It is to be noted that for other patterns, this phenomenology is analogous.
Based on results, it could be concluded that wiggle pattern possesses the largest ductility and elongation due to its flexible structure. Overall, in order to improve the mechanical performance of a 3D printed part, it comes at the expense of print quality, speed and affordability. The best responses in elongation were obtained by wiggle pattern.
In terms of the absorbed energy as toughness evaluation, the wiggle pattern and fast honeycomb had the maximum and minimum absorbed energy, respectively.