Investigation of the Manufacturing Orientation Impact on the Mechanical Properties of Composite Fiber-Reinforced Polymer Elements in the Fused Filament Fabrication Process

This article examines the mechanical properties and compatibility of selected composite materials produced with RP technology and the FFF—fused filament fabrication process. The article scales sophisticated modern materials based on PLA—polylactic acid—plastic and its composite variants. The research is carried out on the 3D FFF printer Felix 3.1 with a dual extruder, which works on the “open-source” principle. In this research, elements of the paradigm and methodology of the processing technology for RP were applied; they were implemented according to EN ISO 527 and ISO 2602 standards. The aim of this study was to investigate the impact of 3D-printing strategy on the mechanical properties of 5 types of PLA composites. The results of this research solve the material compatibility problem, primarily through experimental testing of different combinations of filaments in different printing directions. Analysis of the experimental data showed correlations between the choice of printing strategy and mechanical properties, mainly tensile strength of the selected filaments. The research results show the influence of the printing orientation on mechanical properties of 3D printed samples: parts extruded in length orientation showed higher values of tensile strength compared to parts made in width and height. The CarbonPLA material exhibited 10 times higher tensile strength when printed in length compared to samples.


Introduction
The fused deposition modeling (FDM) method is one of the most widespread methods of 3D printing by melting plastic material and was patented in 1989 by Steven Scott Crump. Steven is also behind the designers of Stratasys, one of the leading companies producing FDM printers [1].
In recent years, 3D printing technology has gained significant attention in various industries due to its ability to fabricate complex structures with precision and efficiency. One area of interest is the development of reinforced polymers, where the addition of fillers or fibers enhances the mechanical properties of the printed parts. Among these materials, reinforced polylactic acid (PLA) has emerged as a promising candidate, offering a combination of biocompatibility, ease of processing, and good mechanical performance [2][3][4].
The widespread adoption of 3D printing technology has paved the way for innovative advancements in material manufacturing. Among the various materials utilized in 3D printing, reinforced polylactic acid (PLA) composites have garnered significant interest due to their enhanced mechanical properties and broad applicability. The combination of PLA, a biodegradable and easily processable polymer, with reinforcing agents such as fibers or fillers, offers the potential for producing durable and functional components [5][6][7].
Some studies deal with similar research, such as Hsueh et al. [8]. Their work characterized the polylactic acid (PLA) and polyethylene terephthalate glycol (PETG) materials of FDM under four loading conditions (tension, compression, bending, and thermal deformation) in order to obtain data regarding different printing temperatures and speeds. Jiang et al. [9] published a paper presenting a study of CFF FFF parts produced on desktop 3D printers using commercially available filament. Tensile test samples fabricated with CFF polymer composite and unfilled polymer were printed and then tested following ASTM D3039M. Test bars were printed with FFF bead orientations aligned with the direction of the applied load at 0 • , 45 • , ±45 • , and normal to the loading axis at 90 • . Results for tensile strength and tensile modulus show that CFF coupons in general yield higher tensile modulus at all print orientations and higher tensile strength at 0 • print orientation. Casavola et al. [10] described the mechanical behavior of FDM parts through classical laminate theory and measured the elastic modulus values in the longitudinal and transverse fiber directions. Lee et al. [11] concluded that screen angle, air gap, and layer thickness affect the elastic performance of flexible ABS objects. Also, Anitha et al. [12] studied the effect of layer thickness and showed that the performance increases as the thickness decreases. Ahn et al. [13] conducted several experiments to determine the effects of air gap, bead width, grid orientation, and ABS color on tensile and compressive strength. They determined that the air gap and grid orientation have a significant effect on the tensile strength; on the other hand, the other parameters have a negligible effect. Furthermore, the factors they investigated do not significantly affect the compressive strength. Sood et al. used response surface methodology to analyze the functional relationship between sample strength and several factors, e.g., orientation of the structure, layer thickness, grid angle, and air gap. Their results show that these factors affect bonding and deformation within parts. Saharudin et al.'s [14] work presents the results of mechanical tests of models manufactured with two 3D printing technologies, FDM and CFF. Both technologies use PLA-and PA-based materials reinforced with carbon fibers. The work includes both uniaxial tensile tests of the tested materials and metrological measurements of surfaces produced with two 3D printing technologies. The test results showed a significant influence of the type of technology on the strength of the models built and on the quality of the technological surface layer. After the analysis of the parameters of the primary profile, roughness, and waviness, it can be clearly stated that the quality of the technological surface layer is much better for the models made with the CFF technology compared to the FDM technology. The works of Wysmulski et al. [15] and Rozylo et al. [16] were focused on studying the mechanical properties of 3D printed thin wall samples. They used numerical simulation to predict these mechanical properties. Numerical results were compared with the experimental ones.
Conversely, few articles in the literature deal with the development of predictive models to determine the mechanical properties of FDM parts [17][18][19]. To establish a strong foundation for this study, an extensive review of relevant research articles has been conducted. To fully exploit the potential of reinforced PLA in 3D printing applications, it is crucial to understand the influence of various printing strategies on the resulting mechanical properties. This article aims to provide a comprehensive examination of the 3D printing strategies employed for 5 types of reinforced PLA and their impact on the mechanical properties of the printed components.

Material Filaments Used for Experimental Research
The CarbonFil fiber is based on a unique HDglass blend of PETG. It is reinforced with 20% ultralight and relatively long carbon fibers, resulting in an exceptionally stiff carbon-reinforced fiber. CarbonFil is twice as stiff as HD glass and yet is 10% more impact resistant, a remarkable feature for carbon-reinforced fiber [20].
Metal File Bronze is a PLA-based metal-filled fiber with approximately 80% gravimetric bronze filling. This incredibly high fill of bronze powders allows any FDM 3D printer user to print bronze objects indistinguishable from original bronze castings. Metal-Fil printed objects are very easy to process, making it possible to create stunning bronze objects with various patina effects, including a real bronze look and "cold" bronze metal. It is approximately 300% heavier than PLA (density 3.5 g/cm 3 ). It is easy to process. MetalFil material can be brushed, sanded, polished, waxed, and painted after processing with a patina. It is very easy to print on a direct drive, and the use of Bowden extruders, cold printing, and lack of deformation after cooling are truly unique features for metal-filled filament. It has improved flow behavior and interlayer adhesion [20].
The basic material used in experimental printing is PLA. It is a biodegradable plastic due to its natural origin (corn, sugar cane, or potatoes). By adding other materials to PLA, it is possible to obtain filaments such as wood (particles of wood), plaster (plaster), and BronzeFill (bronze). Conductive PLA is also produced and is designed for small voltages and currents. It is more flexible than classic PLA but at the cost of less adhesion between layers. If the printing temperature is lower than 225 • C, the resulting surface is glossy, but if the temperature is higher than 225-230 • C, the surface will be matte [21][22][23].

Materials Compatibility
The problem of material compatibility is primarily about experimental testing of different combinations. Some materials are difficult and unrealistic to print together because they do not bond together. It requires experimental testing and setting parameters such as extruder or substrate temperatures, print speed, nozzle-to-substrate distance, and more. Individual materials are generally divided into three groups: compatible, experimentally compatible, and incompatible combinations. If we can print materials close to the extruder temperature, those materials will stick together. The subject of our research was to determine which of the investigated composites is the strongest one.

Preparation and Printing of Tested Composite Samples
This experimental research aims to perform tensile tests based on theoretical knowledge and verify and compare the results. Composite printing using two extruders gives us great diversity in terms of its use and arrangement. However, not all variants that may seem practical at first glance are suitable in practice. Therefore, it is necessary to verify the various material designs through 3D printing of samples and analysis of different materials by measuring, for example, the tensile strength and elongation with a tensile test machine. According to the required application, it is possible further to observe dimensional stability, surface treatment, or other properties [5][6][7][8]. This research uses a 3D FFF printer Felix 3.1 with a dual extruder ( Figure 1) for producing test samples. The software for controlling the printer is Repetier-Host V.2.0.5, which then processes the G code according to any slicer. In this case, Slic3r Prusa Edition version 1.35 was used. The CAD software Creo is used for modelling individual parts.
The test samples that were produced on the printer mentioned above correspond to the EN ISO 527 standard [24]. In this case, 30 test specimens were produced from each composite ( Figure 2). According to ISO 2602, accuracy and probability are determined by a 95% confidence interval [25,26]. The test samples that were produced on the printer mentioned above correspond to the EN ISO 527 standard [24]. In this case, 30 test specimens were produced from each composite ( Figure 2). According to ISO 2602, accuracy and probability are determined by a 95% confidence interval [25,26]. In our case, the temperature for the working nozzle of the extruder was set to 220 °C and 60 °C was used for the working plate. Subsequently, the composite test samples were extruded from the extruders layer by layer, alternately from two material filaments. This was continued until the desired shape of the composite sample was created.
Composite test specimens extruded to length by an extruder are formed from 52 layers with a total length of 150 mm and a width of 20 mm. Composite test samples extruded by an extruder in width are made of 10 layers with a total thickness of 4 mm. Test samples produced in height are made of 378 layers with a total height of 150 mm.
As part of this experimental research, 150 test samples were printed. Thirty samples for each tested material were printed, from which 10 composite samples were printed in length, 10 composite samples in width, and 10 composite samples in height.

Experimental Tensile Test of Composite Materials and Its Evaluation According to EN ISO 527 and 2602
When evaluating the results of the static tensile test for the calculation of the tensile strength, measurement results were evaluated, which served to determine the result of the most probable values of the measured quantities and the size of the errors. Part of the evaluation is also the analytical approximation of the acquired dependencies using appropriate mathematical and statistical methods. Following the standard Plastics-Determination of tensile properties EN ISO 527-1, which in the first part describes the general principles, the accuracy of the measured values is determined by calculation using a  The test samples that were produced on the printer mentioned above correspond to the EN ISO 527 standard [24]. In this case, 30 test specimens were produced from each composite ( Figure 2). According to ISO 2602, accuracy and probability are determined by a 95% confidence interval [25,26]. In our case, the temperature for the working nozzle of the extruder was set to 220 °C and 60 °C was used for the working plate. Subsequently, the composite test samples were extruded from the extruders layer by layer, alternately from two material filaments. This was continued until the desired shape of the composite sample was created.
Composite test specimens extruded to length by an extruder are formed from 52 layers with a total length of 150 mm and a width of 20 mm. Composite test samples extruded by an extruder in width are made of 10 layers with a total thickness of 4 mm. Test samples produced in height are made of 378 layers with a total height of 150 mm.
As part of this experimental research, 150 test samples were printed. Thirty samples for each tested material were printed, from which 10 composite samples were printed in length, 10 composite samples in width, and 10 composite samples in height.

Experimental Tensile Test of Composite Materials and Its Evaluation According to EN ISO 527 and 2602
When evaluating the results of the static tensile test for the calculation of the tensile strength, measurement results were evaluated, which served to determine the result of the most probable values of the measured quantities and the size of the errors. Part of the evaluation is also the analytical approximation of the acquired dependencies using appropriate mathematical and statistical methods. Following the standard Plastics-Determination of tensile properties EN ISO 527-1, which in the first part describes the general principles, the accuracy of the measured values is determined by calculation using a In our case, the temperature for the working nozzle of the extruder was set to 220 • C and 60 • C was used for the working plate. Subsequently, the composite test samples were extruded from the extruders layer by layer, alternately from two material filaments. This was continued until the desired shape of the composite sample was created.
Composite test specimens extruded to length by an extruder are formed from 52 layers with a total length of 150 mm and a width of 20 mm. Composite test samples extruded by an extruder in width are made of 10 layers with a total thickness of 4 mm. Test samples produced in height are made of 378 layers with a total height of 150 mm.
As part of this experimental research, 150 test samples were printed. Thirty samples for each tested material were printed, from which 10 composite samples were printed in length, 10 composite samples in width, and 10 composite samples in height.

Experimental Tensile Test of Composite Materials and Its Evaluation According to EN ISO 527 and 2602
When evaluating the results of the static tensile test for the calculation of the tensile strength, measurement results were evaluated, which served to determine the result of the most probable values of the measured quantities and the size of the errors. Part of the evaluation is also the analytical approximation of the acquired dependencies using appropriate mathematical and statistical methods. Following the standard Plastics-Determination of tensile properties EN ISO 527-1, which in the first part describes the general principles, the accuracy of the measured values is determined by calculation using a confidence interval with a probability of 95% according to EN ISO standard no. 2602-including statistical interpretation of test results and mean estimation and named confidence interval. One of the basic standards developed by the ISO/TC 69 technical committee is the EN ISO 2602 standard, which focuses on data's statistical interpretation. The purpose of these standards is to provide a faithful interpretation of the statistical calculations of the recorded test data and thereby create the prerequisites for the correct use of the results obtained for testing and certification in various industrial applications of statistical methods and in the application of other ISO and IEC standards that can use these methods. This standard specifies a precise procedure for obtaining an estimate of the mean and calculating a two-sided and a one-sided confidence interval for the mean using the standard deviation estimate or the sampling interval (Table 1) [26].

Tensile Test Results
The measured and calculated values obtained during the implementation of the experiments are listed for individual composite test samples in the following table (Table 2). Measured force values Fm represent forces at the tensile strength. From the given force values, it is possible to determine the tensile strength limit σM and the proportional elongation at tensile strength εM (Table 2). Test samples after tensile test fracture are shown in Figure 4a. In Figure 4b are CarbonPLA samples after the tensile test.

Tensile Test Results
The measured and calculated values obtained during the implementation of the experiments are listed for individual composite test samples in the following table (Table 2). Measured force values F m represent forces at the tensile strength. From the given force values, it is possible to determine the tensile strength limit σ M and the proportional elongation at tensile strength ε M ( Table 2). Test samples after tensile test fracture are shown in Figure 4a. In Figure 4b are CarbonPLA samples after the tensile test.         Figures 5-10 show the dependence of the displacement on the loading force F, which was evaluated using the data from the testing machine software during the static tensile test.

Documentation of Microstructures of Selected Composite Samples
The samples were observed at the Technical University in Košice in the laboratory of light microscopy and metallography under constant ambient conditions using a Keyence VHX-500 digital microscope (Keyence SV, Mechelen, Belgium). This microscope has a resolution of 18 million pixels. It is shown in Figure 11.

Documentation of Microstructures of Selected Composite Samples
The samples were observed at the Technical University in Košice in the laboratory of light microscopy and metallography under constant ambient conditions using a Keyence VHX-500 digital microscope (Keyence SV, Mechelen, Belgium). This microscope has a resolution of 18 million pixels. It is shown in Figure 11.
When observing the microscopic structure of a sample made of Carbon PLA filament along its length, one can see the failure of the fibers due to the plastic fracture to the narrowing (Figure 12a), as well as the brittle fracture at the point of attachment of the individual layer of the filament. It is also possible to observe the filling of the samples and the interconnectedness of the individual layers when printing the samples in width ( Figure  12b). Figure 11. Keyence VHX-500 digital microscope. Figure 11. Keyence VHX-500 digital microscope.
When observing the microscopic structure of a sample made of Carbon PLA filament along its length, one can see the failure of the fibers due to the plastic fracture to the narrowing (Figure 12a), as well as the brittle fracture at the point of attachment of the individual layer of the filament. It is also possible to observe the filling of the samples and the interconnectedness of the individual layers when printing the samples in width (Figure 12b). Microscopic observation of the fracture, after the tensile strength test was performed on the sample of PLA Carbon + PLA bronze in width, shows the fracture of the printed filaments that occurred during this test. When observing the fracture, the PLA CarbonPLA filament breaks into a taper after stretching, and the PLA Bronze filament has a visibly brittle fracture. From the microscopic photo, it is also clear that the individual layers of filaments are bonded, when alternating layers of Carbon filament and Bronze filament are Microscopic observation of the fracture, after the tensile strength test was performed on the sample of PLA Carbon + PLA bronze in width, shows the fracture of the printed filaments that occurred during this test. When observing the fracture, the PLA CarbonPLA filament breaks into a taper after stretching, and the PLA Bronze filament has a visibly brittle fracture. From the microscopic photo, it is also clear that the individual layers of filaments are bonded, when alternating layers of Carbon filament and Bronze filament are printed (Figure 13a). In Figure 13b it is also possible to observe the chosen route for the filling when printing samples from PLA Carbon + PLA Bronze filament in portrait.
From observation of the PLA Carbon + Bronze sample in width ( Figure 14) under the microscope, it is obvious that during the tensile test this sample suffered a fracture. On the end parts of some of the printed filaments, fracture can be observed until the narrowing caused by pulling of PLA Carbon filament.

Evaluation of Composite Tested Samples with an Emphasis on Their Production Orientation
The results of the tensile strength measurements were calculated in MPa and recorded in a graphic diagram, and at the same time evaluated according to individual composite materials and the orientation of the produced composite test sample in length, height, and width (Figures 15-17). printed (Figure 13a). In Figure 13b it is also possible to observe the chosen route for the filling when printing samples from PLA Carbon + PLA Bronze filament in portrait.
(a) (b) From observation of the PLA Carbon + Bronze sample in width ( Figure 14) under the microscope, it is obvious that during the tensile test this sample suffered a fracture. On the end parts of some of the printed filaments, fracture can be observed until the narrowing caused by pulling of PLA Carbon filament.

Evaluation of Composite Tested Samples with an Emphasis on Their Production Orientation
The results of the tensile strength measurements were calculated in MPa and recorded in a graphic diagram, and at the same time evaluated according to individual composite materials and the orientation of the produced composite test sample in length, height, and width (Figures 15-17). printed (Figure 13a). In Figure 13b it is also possible to observe the chosen route for the filling when printing samples from PLA Carbon + PLA Bronze filament in portrait. From observation of the PLA Carbon + Bronze sample in width ( Figure 14) under the microscope, it is obvious that during the tensile test this sample suffered a fracture. On the end parts of some of the printed filaments, fracture can be observed until the narrowing caused by pulling of PLA Carbon filament.

Evaluation of Composite Tested Samples with an Emphasis on Their Production Orientation
The results of the tensile strength measurements were calculated in MPa and recorded in a graphic diagram, and at the same time evaluated according to individual composite materials and the orientation of the produced composite test sample in length, height, and width (Figures 15-17). Graphical evaluation of the strength of the composite material PLA Carbon in combination with pure PLA plastic white proves that the samples printed in length show the highest limit of strength which is 25.8 Mpa and thus demonstrate higher strength than the samples printed in height.
Graphical evaluation of the strength of the bronze composite material shows that, in this case, the samples printed in width show the highest strength limit, namely 20.2 MPa, they are stronger compared to the samples printed in height. Based on the results from the graph, it can be concluded that the composite samples produced in height have the lowest strength limit, and the samples produced in length have the highest strength limit. Only in the case of the composite material Bronze is the strength limit of the produced samples higher when printed in width. Graphical evaluation of the strength of the bronze composite material shows that, in this case, the samples printed in width show the highest strength limit, namely 20.2 MPa, they are stronger compared to the samples printed in height. Based on the results from the graph, it can be concluded that the composite samples produced in height have the lowest strength limit, and the samples produced in length have the highest strength limit. Only in the case of the composite material Bronze is the strength limit of the produced samples higher when printed in width. Graphical evaluation of the strength of the bronze composite material shows that, in this case, the samples printed in width show the highest strength limit, namely 20.2 MPa, they are stronger compared to the samples printed in height. Based on the results from the graph, it can be concluded that the composite samples produced in height have the lowest strength limit, and the samples produced in length have the highest strength limit. Only in the case of the composite material Bronze is the strength limit of the produced samples higher when printed in width. MPa.
An evaluation of the comparison of individual composite materials, when printed in height, shows the highest strength in samples of bronze composite material with a strength of 2.7 MPa. A comparison of individual composite materials when printing in landscape shows that with this method of printing, the carbon material in combination with white PLA plastic shows the highest strength limit, namely 21MPa.
As a result of changing the strategy of printing filaments in three directions, the highest tensile strength of the test sample was recorded at 25

Discussion
Previously published research on the process of 3d printing of reinforced PLA materials mainly focused on the investigation of material and process parameters and their impact on the mechanical properties and surface parameters [27][28][29]. Evaluation of mechanical properties, mainly tensile strength, of 5 types of reinforced PLA materials was the subject of the presented work. The research also focused on the printing strategy and its impact on the deformation behavior. Other works [15,16,30] were also aimed at predicting mechanical properties, deformation behavior of different types of reinforced 3D printed material using numerical simulation The work of Guessasma et al. [31] involved experiments and numerical simulations of carbon reinforced PLA material with a focus on the infill parameters. They studied the impact of infill rate on the tensile properties.
The experimental results in this work showed the importance of printing strategy on the mechanical properties and deformation behavior of printed samples. The values of tensile strength σ M and the proportional elongation at tensile strength ε M are shown in Table 1. These results suggest that printing strategy has a significant impact on the tensile strength and deformation behavior of tested materials. In addition, the types of filament, in our case, PLA-based filaments with carbon or bronze reinforcement, were also studied, and their impact on mechanical properties was evaluated using uniaxial tensile tests. Moreover, stress-strain behavior of 5 tested filaments was compared for samples made using three types of printing strategy in Figures 5-10. The printing strategy has a significant effect on the results of the tensile tests. The printing strategy that produced tensile test samples in height showed lowest values of tensile strength and elongation at break. Thus, based on making these samples, this printing strategy is unsuitable for the production of parts that will experience tensile load. Higher values of tensile strength were measured for samples produced in width compared to samples produced using printing strategy "height". These values were up to 10 times higher compared to samples produced in height. The highest values of tensile strength were measured when samples were printed in length for the majority of cases. Only Bronze PLA material that was printed in width had higher values of tensile strength compared to samples printed in length. A comparison of all types of filaments produced in length is shown in Figure 10, which shows that Carbon PLA material showed much higher values of elongation compared to other materials.

Conclusions
Considering the possibility of more massive application in practice of even cheaper rapid prototyping devices to produce final products, it is necessary to conduct research on the possibility of increasing the mechanical properties of products produced on cheaper and more easily available devices for rapid prototyping, as well as devices using fused deposition modeling technology. The results of this paper can help with FFF-RP methods, which offer us many variants of how individual models produce and apply elements of the paradigm and methodology of process technologies for DP and 3D printing. The results from experiments were used by a Czech car manufacturer to optimize the production of 3D printed prototypes of certain car parts and production jigs.
The experimental tests showed that by using carbon composite as a material it is possible to increase the tensile strength of the PLA plastic used, and thus, the mentioned technology can be used in practice even for more demanding applications. The highest strength was achieved with samples made of CarbonPLA + PLA white and Bronze + PLA white composites in the length direction. Strength values for all investigated materials dropped significantly when testing samples were printed perpendicular to the applied force. For the samples produced in this way, the strength reached about 10% of the strength of the samples printed in the direction of the stress force. Experiments have proven that with a suitable printing method we can create prototype production algorithms that correspond to the required explicit application in the production of a model or component. Based on the analyses performed and the results achieved, it can be concluded that:

•
The use of a suitably chosen composite material as a reinforcing substance can influence and change selected mechanical properties of products produced by the fused deposition modeling method in rapid prototyping technologies compared to the use of PLA plastic material, which is the most often used material in the mentioned technology.

•
In the printing strategy, part orientation is an important parameter that must be considered when printing parts made of reinforced PLA material. Printing strategy should be also based on the type of load that the part can experience in practical use.