Tensile Behavior of PLA and PLA Composite Materials Under Different Printing Parameters

TIBERIU DOBRESCU*, NICOLETA-ELISABETA PASCU, GABRIEL JIGA, IONEL SIMION, VICTOR ADIR, GEORGE ENCIU, DANIELA IOANA TUDOSE University Politehnica of Bucharest, Machine and Production Systems Department, 313 Splaiul Independentei, 060042, Bucharest, Romania University Poltehnica of Bucharest, Engineering Graphics and Industrial Design Department, 313 Splaiul Independentei, 060042, Bucharest, Romania University Politehnica of Bucharest, Strength of Materials Department, 313 Splaiul Independentei, 060042, Bucharest, Romania

PLA has good mechanical properties, thermal plasticity and biocompatibility, is easy to be manufactured. PLA has a high Young modulus and high strength and is not a tough material. PLA resistance can be enhanced by copolymerization [8]. One major disadvantage of PLA is its durability. Consequently, this material is to not being used in automotive industry.
In the production environment of 3D printing products, Polylactic Acid (PLA) is much more widely applicable than ABS, Nylon and various blends of polycarbonate. It is used for 3D printing material with FDM (Fused Deposition Modeling) technology. Since PLA-based materials has a glass transition temperature between (60-70 °C) it is more suitable to be used in FDM.
The FDM principle is based on the following steps: the extrusion material (filament input) is brought into a semi-solid state and forced through a nozzle to form a filament with a narrower diameter than the diameter of the input filament that will rapidly solidify after the extrusion. The input filament will be liquefied in a liquefaction chamber placed before the nozzle. The temperature in the liquefaction chamber is set so that the filament changes its state. The diameter of the filament resulting from the extrusion will remain constant if the movement of the nozzle on the surface is set to a constant speed. Figure 1 shows the extrusion system and its components. The testing of PLA new composite materials is required since they do not behave as expected by datasheets because of their anisotropy and largely unknowns' local material properties after extrusion.
Over the years, the properties of PLA materials have been the subject of extensive researches [9][10][11][18][19][20][21]. The properties of PLA depend on its isomers, processing temperature and cooling time [10,11]. Besides these, the crystallinity rate is an important property of polymers (crystallinity indicates the rate of crystalline regions in the polymer that respects the amorphous content [12]). Crystallinity influences polymerproperties such as hardness, modulus of elasticity, tensile strength, stiffness and melting points.
One of the paper purpose is to highlight the influence of different printing parameters upon one mechanical characteristics such tensile stress, in order to manufacture different consumable components safe in exploitation for short periods of time. Very often, these components are useful in manufacturing, keeping the process up and running during the entire waiting time for the delivery of a new piece of equipment.

Experimental part Materials and methods Equipment and software applications
The specimens have been printed on a CREALITY 3D [13] printer equipment. The transfer of the specimen geometric model was performed using ".stl" files exported from AutoCAD version 2019 provided by Autodesk [14]. The software application for generating the G-code in the 3D modeling specimens used was Simplify 3D-version 3.1.0 [15]. Layer deposition is dependent on the extruded filament diameter which is applied in successive planes parallel to each other and perpendicular to the flow direction of the extruded material (Fig. 2). http://www.revmaterialeplastice.ro For the tests, 72 specimens were printed, with 18 samples of each material, taking into account the ASTM D638 standard and using a CREALITY 3D printer. All materials were purchased from different providers (Table1). Concerning the printing methods three different filling densities of the specimen were used: 60%, 80% and 100%. In order to evaluate the mechanical characteristics of the specimen as a function of stacking sequence of the layers deposition, two cross different layer orientations ( 45 O and  60 O ) have been taken into account. All specimens, regardless the material type, were printed at a same extrusion temperature. These temperatures and all tested materials are summarized in Table 2. Other printing parameters such as the layer height, the printing speed were adjusted identically for each material, using the supplier's recommendations to achieve print quality and uniformity for all specimens. The general printing conditions for all four materials require the PLA plastic material to be heated in the extruder at a temperature in the range 160 -210 [ O C] and the heating platform heated to a temperature in the range 0 -60 [ O C]. The extrusion temperature and, implicitly, the extrusion die temperature were set at 205 [ O C]. The technicalcharacteristics of the printing filaments used in the tests are 1.75 [mm] filament diameter, the filament diameter after extrusion being given by the diameter of the extrusion die 0.4 [mm]. The parameters of the printing process were not varied during the experiments: thus, the extrusion speed was set at 60 [mm/s], the way of how the specimens have been created being identical for each specimen.
Since this type of technology requires platform heating, this temperature was set to 55 [ O C]. The temperature of 55 [ O C] was selected by the authors based on the experience gained in previous research studies and it falls in the range recommended by the filament manufacturers. The specimens used for tensile tests have been performed according to ASTM D638. This test method covers the determination of the tensile properties of unreinforced and reinforced plastics in the form of standard dumbbell-shaped test specimens when tested under defined conditions of pretreatment, temperature, humidity and testing machine speed. The geometrical characteristics of such a specimen are depicted in Figure 3.

Materials
In the experimental part, four types of filaments were used for the test specimens manufacturing: PLA (Polylactic acid), PLA with Copper powder, PLA with Aluminum powder and PLA with Graphene. Before printing, it is important for the material to be dry and the print media to be controlled (to have a constant temperature).

PLA material without metallic powder coating
PLA is available on the market in a wide range of variantsand depending on the different colors, they can be opaque or translucent. All these types of PLA heated to 160 [ O C] adhere very well to the Kapton band or to the glass (through a glue). In tests the authors adopted the Kapton tape for bed addition. PLA thermoplastic polyester filament used for 3D printing using the FDM method is provided by Poland's "Devil Design" [22]. The characteristics of PLA filament (with a diameter of 1.75 [mm]) are shown in Table 3 and are specified by the manufacturer.  Table 4.  Table 4.
PLA material with Graphene insertion PLA material with Graphene insertion supplied by FILOALFA and called GRAFYLON is a filament for 3D printing addicted with graphene, to provide excellent mechanical and aesthetic performance. GRAFYLON is suitable for the most popular printers on the market and is completely non-toxic and odorless till high temperatures. This material is a special filament with a very good mechanical strength and thermal conductivity coupled with a highly defined surface finish with priceless printing simplicity. Compared to a traditional PLA, GRAFYLON offers anelastic modulus: + 34%, atensile strength: + 23%, anelongation capacity: + 28% and ideal for printing finished objects [17]. The characteristics of the GRAFYLON filament for 3D Printer with a diameter of 1.75 mm are identical to those in Table 4.. Table 5 summarizes the data generated during experimental printing. Thus, the specimens were rated from 1 to 18 for this type of material and ranked according to the deposition angle (45 O , 60 O ) as well as function of the filling percentage of the specimens (60%, 80% and 100%). Indeed, the scanning angle should vary from 0 o to 90 o , but since at angles below 45 o the specimens failed at low values of tensile force (around 350 N), the authors decided to study the tensile behavior at angles higher than 45 o , considering that future experimental attempts to be performed from 45 o to 90 o from 5 o to 5 o . Although the filament suppliers suggested printing components with a 100% density, the authors intended to highlight how the normal stress varies with decreasing of the filling density percentage.  Figure 4, 60% and 100% warp specimens with thread deposition at ± 60° are much more resistant in tension than 60% and 100% yarn specimens with thread deposition at ± 45°. Differences between fracture strength values for the same filling percentage for 60% and 80% samples but with a different deposition orientation of ±45 o and ±60 o are much lower compared to differences in fracture strength values for the same filling percentage of 100% specimen, but with a different loading orientation of ± 45 O and ± 60 O respectively. In Tables 6-11 are depicted the obtained data of tests performed on PLA samples.  In order to quantify the amount of variation of dispersion of the previous data set the standard deviation was calculated with the formula (  For PLA specimens (Fig. 5) one can conclude that, in case of 45 o angle deposition, the graphic is rather a linear function, in comparison with those with a 60 o angle deposition, where the variation is rather exponential; this is due to the fact that for 45 o the specimen is totally "balanced", with respect to the XY machine coordinate system.

Results and discussions
Moreover, in case of 60% angle deposition it is easily to be seen that as the percentage of filling increases, the graphic will present a steeper variation, concluding that it is advisable to have specimens with high fill percentage when high values of stresses are requested.
For low interior fill percentage, the infill angle does not affect the stress behavior. For values between approximately 65% and 85% filling percentage the specimens with an infill angle of ± 45 o have a better behavior regarding the tensile stress than the other ones. For specimens with an interior filling percentage over 85% it is suitable to make deposition with an infill angle of 60%.
In Table 13 are synthesized data resulted during 3D printing process with PLA specimens with Copper powder insertion. The specimens were ranking from 19 to 36. In Tables 14-19 are depicted the obtained data of tests performed on PLA sampleswith Copper powder insertion.  The tensile strength for the test specimens  19 ÷ 36 performed on PLA specimens with Copper powder insertion (Fig. 6) shows a significant variation from 20.13 [MPa] to 30.14 [MPa]. As it can be seen in Figure 6, the 60%, 80% and 100% fill patterns with thread deposition at ± 45° are more resistant in tension than those with a wire deposition at ± 60°. The differences between the fracture strength values for the same sample filling rate but with a different orientation of ± 45 o and ± 60 o deposition were much higher for the filling percentages of the 60% and 80% specimens. Differences between fracture strength values for the same 100% fill rate of the specimen but with a different orientation of ± 45 o and ± 60 o deposition are insignificant. http://www.revmaterialeplastice.ro In case of PLA specimens with copper powder insertion ( Figure 7) for thread deposition at 45 o compared to the deposition at 60 o the fracture strength is higher for all breakage rates. For PLA specimens with Copper powder insertion with thread deposition at 60° on the graphic depicted in Figure 7, a decrease in tensile strength in the vicinity of 80% filling percentage of the specimen can be observed. These values are the lowest averages of the obtained ones within that series of tests.The tensile strength for the test specimens  37÷54 performed on the PLA specimens with aluminum powder insert (Figure 8) shows a small variation compared to those presented above from 19.65 [MPa] to 30.95 [MPa]. As can be seen in Figure 8, specimens for all filling types with thread deposition at ± 45° are more resistant in tension than specimens with thread deposition at ± 60°. As in previous cases, the differences between the fracture strength values for the same percentage of the specimen filling but with a different orientation of ± 45 o and ± 60 o deposition were much higher for the fill percentages of the 60% and 80% specimens. The differences between the fracture strength values for the same percent filling, i.e. 100% of the specimen, but with a different orientation of ± 45 o and ± 60 o deposition, are insignificant.In Table 20 are synthesized data resulted during 3D printing process with PLA-Aluminum powder type of material. The specimens were ranking from 37 to 54.  In Tables 21-26 are depicted the obtained data of tests performed on PLA-Aluminum powder samples.   In case of PLA specimens with Aluminum powder insertion (Fig. 9) for thread deposition at  45 o compared to the deposition at  60 o the fracture strength is higher for all breakage rates. For PLA specimens with an Aluminum powder insertion with thread deposition at  45 o and  60 o , it can easily be seen from the graph represented in Figure 9, the curves of the fracture strength values for these specimens, according to the percentage of filling, have the same allure.
In Table 28 are synthesized data resulted during 3D printing process with PLA-Graphene powder insertions type of material. The specimens were ranking from 55 to 72. In Tables 28-33 are depicted the obtained data of tests performed on PLA with Graphene samples.  Tensile strength for the test specimens  55 ÷ 72 performed on Graphene powder insertions ( Figure 10) shows a significant variation from 20.13 [MPa] to 39.39 [MPa]. As it can be seen in Figure 10, the specimens with 60% respectively 80% infill with a thread deposition at ± 45° are more resistant in tension than those with a wire deposition at ± 60°. The differences between the fracture strength values for the same sample fill rate but with a different orientation of ± 45 o and ± 60 o deposition were much higher for the fill percentages of the 60% and 80% specimens. In the same time, from the same graph shown in Figure 10 it can be seen that for this combination of test material for 100% the fracture strength is higher for a deposition at ± 60 o than deposition to ± 45 o , differences in fracture strength being insignificant   In the case of PLA + Graphene specimens (Fig. 11) for thread deposition at  45 o compared to the deposition at  60 o the fracture strength is higher only for the percentages close to the 60% filling degree. As the specimens' filling percentage increases, the thread deposition at  45 o and 60 o can easily be seen from the graph depicted in Figure 11 From the chart shown in Figure 12 (in the case of interior fill percentage -60%), it can be noticed that the PLA + Copper powder filaments have a close fracture strength as average values with those of PLA + Graphene specimens and very close to the average values of PLA + Aluminum powder specimens. PLA specimens do not align with these values. The graph also shows that for this filling percentage of the specimen, the samples obtained by thread deposition at  45 o have a higher breaking strength than the ones obtained by depositing the thread at  60 o in case of powder-coated specimens. For PLA specimens, the samples obtained by thread deposition at  60 o have a fracture strength higher than the specimens obtained by a thread deposition at  45 o .  rom the chart depicted in Figure 13 (in the case of Interior fill percentage -80%) it can be seen that the PLA + Copper powder filaments have a low fracture strength compared to the others. The graph also shows that for this filling percentage of the specimen, the samples obtained by depositing the thread at 45 o have a higher fracture strength than the samples obtained by depositing the thread at 60 o for all specimens. Also from the same graph we can rank the materials according to fracture strength as follows: PLA specimens have the highest tensile strength, the PLA + Graphene and PLA + Aluminum powder specimens are close and follow the PLA and PLA + Copper powder samples have the lowest breaking strength. From the graph depicted in Figure 14 (in case of internal filling percentage of 100%) it can be seen that the PLA + Copper powder filaments have a close fracture strength regarding the average values to those with PLA + Aluminum powder.
PLA specimens do not align with these values, these ones presenting the highest fracture strength for both types of thread deposition.
The graph shows also that for this filling percentage of the specimen, the samples obtained by thread deposition at  60 o have a higher fracture strength than those obtained by depositing the thread at  45 o in case of PLA and PLA + Graphene specimens.
For PLA + Copper and PLA + Aluminum metal powder specimens, the samples obtained by thread deposition at  60 o have a fracture strength close to that of the specimens obtained by thread deposition at  45 o and for both materials the deposition at  45 o is more advantageous in terms of tensile strength.

Materials
Interior fill percentage -80% Interior fill percentage -80% and Infill angle -±45°I nterior fill percentage -80% and Infill angle -±60° During the printing process some difficulties arose regarding the adhesion of the materials to the printing platform. Thus, the version directly printed on the platform glass through an adhesive was not adopted because for this printer type (with a 300 x 300 [mm] high print bed) the specimens did not adhere to the printing plate, their construction being sometimes deformed and during the printing process, the test specimens were unevenly contracted in the clamping area on the folding bed. Finally, it was adopted the specimens' fixation using the Kapton tape which was changed to each print in order to ensure the same printing conditions for the test specimen. Also from an initial test phase in the data collected from the slicing program of the geometric models Tables 5, 13, 20 and 27 it can be noticed that there is a difference between the test time of the samples provided by the slicing software and the actual time of making the specimens.