Materials 2015, 8, 5834-5846; doi:10.3390/ma8095271

OPEN ACCESS

materials ISSN 1996-1944 www.mdpi.com/journal/materials Article

Influence of Layer Thickness and Raster Angle on the Mechanical Properties of 3D-Printed PEEK and a Comparative Mechanical Study between PEEK and ABS Wenzheng Wu 1 , Peng Geng 1 , Guiwei Li 1 , Di Zhao 1 , Haibo Zhang 2 and Ji Zhao 1, * 1

School of Mechanical Science and Engineering, Jilin University, Renmin Street 5988, Changchun 130025, China; E-Mails: [email protected] (W.W.); [email protected] (P.G.); [email protected] (G.L.); [email protected] (D.Z.) 2 Alan G. MacDiarmid Institute, College of Chemistry, Jilin University, Qianjin Street 2699, Changchun 130012, China; E-Mail: [email protected] * Author to whom correspondence should be addressed; E-Mail: [email protected]; Tel.: +86-150-4306-6406. Academic Editor: Reza Montazami Received: 20 May 2015 / Accepted: 19 June 2015 / Published: 1 September 2015

Abstract: Fused deposition modeling (FDM) is a rapidly growing 3D printing technology. However, printing materials are restricted to acrylonitrile butadiene styrene (ABS) or poly (lactic acid) (PLA) in most Fused deposition modeling (FDM) equipment. Here, we report on a new high-performance printing material, polyether-ether-ketone (PEEK), which could surmount these shortcomings. This paper is devoted to studying the influence of layer thickness and raster angle on the mechanical properties of 3D-printed PEEK. Samples with three different layer thicknesses (200, 300 and 400 µm) and raster angles (0˝ , 30˝ and 45˝ ) were built using a polyether-ether-ketone (PEEK) 3D printing system and their tensile, compressive and bending strengths were tested. The optimal mechanical properties of polyether-ether-ketone (PEEK) samples were found at a layer thickness of 300 µm and a raster angle of 0˝ . To evaluate the printing performance of polyether-ether-ketone (PEEK) samples, a comparison was made between the mechanical properties of 3D-printed polyether-ether-ketone (PEEK) and acrylonitrile butadiene styrene (ABS) parts. The results suggest that the average tensile strengths of polyether-ether-ketone (PEEK) parts were 108% higher than those for acrylonitrile butadiene styrene (ABS), and compressive strengths were 114% and bending strengths were 115%. However, the modulus of elasticity for both

Materials 2015, 8

5835

materials was similar. These results indicate that the mechanical properties of 3D-printed polyether-ether-ketone (PEEK) are superior to 3D-printed ABS. Keywords: 3D printing; polyether-ether-ketone (PEEK); raster angle; layer thickness

1. Introduction 3D printing is a new integrated manufacturing technology that involves a variety of disciplines. 3D printing has shown excellent potential to reduce both the cycle time and cost of product development [1]. With the development of 3D printing, a large number of processes have been developed that allow the use of a variety of materials and methods [2,3]. Amongst these technologies, one of the most commonly used is fused deposition modeling (FDM) [4,5], a layer-by-layer additive manufacturing technique, based on computer-aided design (CAD) and computer-aided manufacturing (CAM) [6]. The advantages of this technology are as follows [7]: easy material change, low maintenance costs, supervision-free operation, compact size and low working temperature [8]. However, the main disadvantage of FDM is the narrow range of available materials [9]. Many commercial 3D printers can print only acrylonitrile butadiene styrene (ABS) or poly (lactic acid) (PLA). Most parts fabricated by commercial 3D printers are used as demonstration parts, not as working parts. This limits the use of FDM in industrial applications [10]. Polyether-ether-ketone (PEEK), a semi-crystalline thermoplastic material, is an engineering plastic developed by Imperial Chemical Industries (ICI) in 1977 that has high temperature resistance, superior mechanical strength and outstanding chemical stability. In addition, PEEK is biocompatible and ideally suited for use in biomedical applications. As a consequence of these advantages, PEEK can be used in a wide variety of fields, such as the aerospace, automotive, electronics and medicine industries. The mechanical properties of 3D-printed parts are important indices for evaluating printing quality [11–14]. Regarding this issue, some work has been carried out to determine the effects of production parameters on the mechanical properties of 3D printed parts. Ismail et al. [15] studied the influence of raster angle and orientation on the mechanical properties of ABS printed parts. Tensile and three-point bending tests have been studied. Sood et al. [16] focused on the influence of orientation, layer thickness, raster angle, part raster width and raster-to-raster gap. Tymrak et al. [17,18] described the influence of orientation, layer thickness and raster-to-raster gap on parts that were printed using several commercial 3D printers. The above research has mainly been concerned with ABS materials. This paper reports on the mechanical properties of PEEK samples built by a custom-built 3D printing system. PEEK samples were fabricated for two experiments. The first experiment was used to investigate the influence of printing parameters on mechanical properties. The second experiment was used to compare the mechanical properties of PEEK and ABS FDM parts. 2. Experimental Section Mechanical properties of 3D-printed samples can be influenced by many factors, such as layer thickness, raster angle, build orientations, fill pattern, air gap and model build temperature. Layer thickness is the thickness of the layer deposited by the nozzle. Raster angle is the direction of raster

Materials 2015, 8 Materials 2015, 8

3885 5836

thickness is the thickness of the layer deposited by the nozzle. Raster angle is the direction of raster with shown in Figure 1. Air gapgap is the between two with respect respect to tothe theloading loadingdirection directionofofstress, stress,asas shown in Figure 1. Air is distance the distance between two adjacent deposited filaments in the same layer. Thenumber numberofofcontours contoursisisthe the number number of of filaments adjacent deposited filaments in the same layer. The filaments initially deposited is is thethe width of the filament deposited by the 3D initially deposited along along the the outer outer edge. edge.Bead Beadwidth width width of the filament deposited by the printer nozzle. 3D printer nozzle.

Figure 1. The raster angle of 3D printing. Figure 1. The raster angle of 3D printing. The work thickness (LT) andand raster angle (RA) on work described described herein hereininvestigated investigatedthe theinfluence influenceofoflayer layer thickness (LT) raster angle (RA) the mechanical properties of PEEK samples builtbuilt by a by custom-built 3D printing system. Samples with on the mechanical properties of PEEK samples a custom-built 3D printing system. Samples ˝ ˝ ˝ three rasterraster anglesangles (0 , 30 ) and (200, 300 and 400 µm) 3D with three (0°,and 30°45and 45°)three and layer three thicknesses layer thicknesses (200, 300 and 400were μm)built werebybuilt printing of PEEKofand testedand for tensile, compressive and bending strengths. Thestrengths. printing parameters are by 3D printing PEEK tested for tensile, compressive and bending The printing shown in Table 1. Buildinorientation refersorientation to the inclination of the partinclination in a build platform respect to X, parameters are shown Table 1. Build refers to of part inwith a build platform Y, Z axis (side, flatY, andZ up) ourand study, up along Z axis and up sat along down Z flataxis during with respect to X, axis[18]. (side,Inflat up)samples [18]. In grown our study, samples grown and the building Fillbuilding pattern process. is the motion path ofisthe printing [19]. sat down flatprocess. during the Fill pattern the3D motion pathmachine’s of the 3D liquefier printing head machine’s liquefierthe head [19]. process, During the printing head process, theback liquefier headforming move back and forthstructure formingtoa During printing the liquefier move and forth a rectangular rectangular fill the of between layer. Airtwo gapadjacent is the gap between two layer. adjacent fill the entirestructure internal to portion ofentire layer. internal Air gap portion is the gap rasters on same rasters on same layer. Table 1. The printing parameters of polyether-ether-ketone (PEEK) 3D printing.

Table 1. The printing parameters of polyether-ether-ketone (PEEK) 3D printing.

Control Factors Fixed Factors Control Factors Fixed FactorsValue Factor Level Unit Factor Unit Factor Level Unit Factor Value Unit 200 mm Buildorientation orientation Y-direction (Flat) (Flat) 200 mm Build Y-direction -Layer thickness mm Fillpattern pattern Line 300 mm Fill Line -Layer thickness 300 400 mm Airgap gap mm 400 mm Air 00 mm ˝ 00 Numberof ofcontours contours ° Number 22 - Raster angle Nozzle 30 Nozzle inner inner diameter 0.4 mm Raster angle ˝° 30 0.4 mm 45 ° diameter 45

˝

Materials 2015, 8

5837

Materials 2015, 8

3886

To investigate the relationships between printing factors and mechanical properties, five samples of each geometry were created, with different thicknesses raster angles, using material. To investigate the relationships between layer printing factors andand mechanical properties, fivePEEK samples of each samples geometrywere were built created, withABS different layer to thicknesses angles, using PEEKof material. Identical using material compareand theraster mechanical properties 3D-printed Identical samples were built using ABS material to compare the mechanical properties of 3D-printed PEEK and ABS parts. The geometric models of the tensile, bending samples and compressive PEEK andsimilar ABS parts. The geometric models of the16421-1996, tensile, bending samples and compressive samples samples were to the specifications in GB/T GB/T 9341-2008 and GB/T1041-2008, were similar to the specifications in GB/T 16421-1996, GB/T 9341-2008 and GB/T1041-2008, respectively. Mechanical test sample models conforming to the relevant mechanical test standards were respectively. Mechanical test sample models conforming to the relevant mechanical test standards designed in CATIA V5, and the geometric models were exported as files in stereolithography (STL) were designed in CATIA V5, and the geometric models were exported as files in stereolithography format for import by import the FDM software. The geometric models are shown in Figure 2. 2. (STL) format for by the FDM software. The geometric models are shown in Figure

Figure 2. Standard samples for mechanical tests. (a) Tensile samples; (b) Bending samples; Figure 2. Standard samples for mechanical tests. (a) Tensile samples; (b) Bending samples; (c) Compressive samples. (c) Compressive samples. TM

The experimental samples and ABS plus -P430. High-performance PEEK The experimental samples were were mademade from from PEEKPEEK and ABS plusTM -P430. High-performance PEEK material material was obtained from the Jilin University Special Plastic Engineering Research Co. Ltd. was obtained from the Jilin University Special Plastic Engineering Research Co. Ltd. (Changchun, China), with (Changchun, China), with a melting point of 334 °C, a glass transition temperature of 143 °C and a melting point of 334 ˝ C, a glass transition temperature of 143 ˝ C and a melt index of 44 g/10 min. a melt index of 44 g/10 min. PEEK samples were then built with a custom-built 3D printing system. In PEEK samples were then built with a custom-built 3D printing system. In this printing machine, a stage this printing machine, a stage platform moves in the X- and Y- axes, and a head including a nozzle tip platform movesheater in themoves X- and Y- axes, and awhile headfusing including a nozzlematerial. tip and ABS nozzle heater moves and nozzle along the Z-axis, and depositing is inexpensive, hasalong the Z-axis, while fusing and depositing ABS inexpensive, high mechanical strength, and is widelymaterial. used in the 3D is printing industry has [20].high ABSmechanical plusTM-P430strength, material and TM is widely used Inc., in theEden 3D Prairie, printingMN, industry ABSto plus -P430 material (Stratasys Inc., Eden 3D Prairie, (Stratasys USA) [20]. was used make ABS samples. ABS is ideal for building with uPrint 3D Printers (Stratasys Prairie, USA). ABS samples with MN, parts USA)when wascombined used to make ABSSE samples. ABS is idealInc., for Eden building 3DMN, parts when combined with uPrint SE 3D Printers usingPrairie, default building parameters obtain optimal mechanical uPrintwere SE built 3D Printers (Stratasys Inc., Eden MN, USA). ABS to samples were built with uPrint properties. The SR-30 Soluble Support (synthetic thermoplastic polymer) was used as the support material. SE 3D Printers using default building parameters to obtain optimal mechanical properties. The SR-30 The mechanical properties of PEEK and ABS materials are shown in Table 2. Testing of mechanical Soluble Support (synthetic thermoplastic polymer) was used as the support material. The mechanical properties of the raw materials were performed on samples manufactured by injection molding. properties of PEEK and ABS materials are shown in Table 2. Testing of mechanical properties of the raw materials Table were performed on properties samples manufactured by injection molding. 2. Mechanical of PEEK and acrylonitrile butadiene styrene (ABS). Factor Tensile Strength Elastic Limit Factor Compressive Strength Tensile Strength Compressive Modulus Bending Strength Elastic Limit Bending Modulus Compressive Strength

PEEK 100.0 MPa 72.0 MPa PEEK 118.0 MPa 100.0 MPa 3.8 GPa 163.0 72.0 MPa MPa 4.0 GPa 118.0 MPa 3.8 GPa 163.0 MPa 4.0 GPa

ABS 37.0 MPa 31.0 MPa ABS 37.0 MPa 37.0 MPa 2.3 GPa 53.0 MPa 31.0 MPa 2.2 GPa 37.0 MPa 2.3 GPa 53.0 MPa 2.2 GPa

Table 2. Mechanical properties of PEEK and acrylonitrile butadiene styrene (ABS).

Compressive Modulus Bending Strength Bending Modulus

Materials 2015, 8

5838

The mechanical tests were carried out on an autograph universal materials testing machine, equipped with a 50-kN load cell. To minimize experimental error, five samples were built and tested under identical conditions for each mechanical test and the mean values were taken as the results. The samples were performed according to the GB/T 16421-1996 Standard and the samples stretched at a strain rate of 1 mm/min. The three-point bending tests were performed according to the GB/T 9341-2008 Standard with the loading rate of 2 mm/min, the span was 64 mm and the head of loading device was composed of steel cylinders 10 mm in diameter. The compressive test was performed according to the GB/T1041-2008 Standard at a loading rate of 2 mm/min. 3. Results and Discussion 3.1. Influence of Layer Thickness and Raster Angle on Mechanical Properties of PEEK Table 3 shows the practical mean values of mechanical tests for the 3D-printed PEEK samples. From Table 3, it can be noted that the samples built with a 300-µm layer thickness had the greatest strengths in all mechanical tests. The strength of samples with a 400-µm layer thickness decreased significantly. We can also see that samples built with raster angles of 0˝ /90˝ had the greatest mechanical strengths. Although the layer thickness had a great influence on tensile strength, it had little influence on bending and compressive strengths. Because the compressive sample was cylindrical, there was no effect of the raster angle. Table 3. Mechanical properties in different layer thickness and different raster angles. Factors

Layer Thickness (µm)

Raster Angle (˝ )

200 300 400 ˝ 0 /90˝ 30˝ /´60˝ 45˝ /´45˝

Tensile strength (MPa)

Bending strength (MPa)

Compressive strength (MPa)

40.1 56.6 32.4 56.6 41.8 43.3

52.1 56.1 48.7 56.1 48.5 43.2

53.6 60.9 54.1 -

As indicated in Table 3, the tensile strengths of the PEEK samples were significantly affected by the layer thickness and raster angle. In samples printed at raster angles of 0˝ /90˝ , the filaments were oriented parallel to the load direction, producing the strongest samples. Similarly, in samples printed in other orientations, there was a finite angle between the printed microstructural elements and the load direction. The filaments were thus subjected to both tensile and shear stresses, leading to failure at low tensile strength. During compressive tests, the samples undergo pressure aligned in the axial direction and the adjacent layers bear the shear stress because of the additive build up, which can cause the single layers to slide along one another until the sample finally breaks. The compressive strength of printed samples is significantly lower than that of injected materials because the printed samples are subject to flaws, comprising both extensive pores that are initially squeezed out and weak interlayer bonding.

Materials Materials 2015, 2015, 888 Materials 2015,

5839 3888 3888

Weak Weak interlayer interlayer bonding bonding is is responsible responsible for for aaa decrease decrease in in bending bending strength, strength, which which can can cause cause welded welded Weak interlayer bonding is responsible for decrease in bending strength, which can cause welded layers to to delaminate. delaminate. During the top top of of the the previous previous layer During printing, printing, each each new new layer layer will will overlay overlay the before layers layer before before material solidification from the the melt melt occurs, which leads to shrinkage in the the previous layer. The material solidification solidification from from the melt occurs, occurs, which which leads leads to to shrinkage shrinkage in in the previous previous layer. layer. The The residual residual material residual stresses in in the the printed printed samples samples resulting resulting from from the the volume volume shrinkage shrinkage were were examined examined to to determine determine the the stresses in the printed samples resulting from the volume shrinkage stresses were examined to determine the weak interlayer interlayer bonding. bonding. As As the layer thickness increases, the accuracy of the overall finished sample sample weak interlayer As the the layer layer thickness thickness increases, increases, the the accuracy accuracy of of the the overall overall finished finished weak bonding. sample geometry decreases decreases and and the the outline outline more more readily readily displays displays the the staircase staircase phenomenon. phenomenon. geometry decreases geometry and the outline more readily displays the staircase phenomenon.

3.2. Comparison Comparison of of ABS ABS and and PEEK PEEK Tensile Tensile Strengths Strengths 3.2. Comparison of ABS and PEEK Tensile 3.2. Strengths To evaluate evaluate the the performance performance of of printed printed PEEK PEEK samples, samples, we we built built samples samples from from PEEK PEEK and and ABS ABS To evaluate the performance using optimal optimal parameters. parameters. Figure Figure shows ABS ABS and and PEEK PEEK samples samples after after the the tensile tensile test. test. Engineering Engineering Figure 33 shows using the tensile test. stress-strain curves curves for for PEEK PEEK and and ABS ABS samples samples are are shown shown in in Figure Figure 4. 4. stress-strain shown in Figure 4.

Figure styrene (ABS); (b) Figure 3.3. 3. Fractured Fractured tensile tensilesamples. samples. (a) (a)acrylonitrile acrylonitrilebutadiene butadiene styrene (ABS); Figure Fractured tensile samples. (a) acrylonitrile butadiene styrene (ABS); polyether-ether-ketone (PEEK). (b) polyether-ether-ketone polyether-ether-ketone (PEEK). (b) (PEEK).

Figure Tensile stress-strain stress-strain curves curves of of acrylonitrile acrylonitrile butadiene butadiene styrene styrene (ABS) Figure 4. 4. Tensile Tensile (ABS) and and Figure 4. stress-strain curves of acrylonitrile butadiene styrene (ABS) and polyether-ether-ketone (PEEK). polyether-ether-ketone (PEEK). (PEEK). polyether-ether-ketone

The ABS ABS curve curve includes includes regions regions of of elastic elastic and and plastic plastic deformation, deformation, accompanied accompanied by by necking necking The ABS curve accompanied The includes regions of elastic and plastic deformation, by necking deformation. The The stress-strain behavior under tensile tensile stress stress was was initially initially nonlinear. nonlinear. After After the the sample sample deformation. The stress-strain stress-strain behavior behavior under under tensile stress After the deformation. was initially nonlinear. sample

Materials 2015, 8 Materials 2015, 8

5840 3889

reached peak stress, the resisting stress was almost constant with the increase in strain through shape. It reached peak stress, the resisting stress was almost constant with the increase in strain through shape. can be concluded that the fracture of ABS occurs mainly via damage to the raster. As the tensile stress It can be concluded that the fracture of ABS occurs mainly via damage to the raster. As the tensile increases, the failure will begin at the weakest raster and next weakest raster will break, in sequence, stress increases, the failure will begin at the weakest raster and next weakest raster will break, until total failure of the sample. When the stress reaches a certain constant value, a long propagation in sequence, until total failure of the sample. When the stress reaches a certain constant value, a long process occurs in theoccurs neck.inCraze is the mainis plastic deformation mechanism of ABS, of with a great propagation process the neck. Craze the main plastic deformation mechanism ABS, with number of crazes appearing perpendicular to the direction of the tensile loading. Crazes initiate, widen, a great number of crazes appearing perpendicular to the direction of the tensile loading. Crazes initiate, then suffer of the raster as tension If crazes extend to both endstoofboth the sample, widen, thenbreakdown suffer breakdown of the raster asincreases. tension increases. If crazes extend ends of the the sample fails with insignificant necking deformation, because molecular chains initially in an unoriented sample, the sample fails with insignificant necking deformation, because molecular chains initially in state transformstate to a transform more highly state of necking. transformation process causesprocess strain an unoriented to aoriented more highly oriented stateThe of necking. The transformation hardening, and ensures the crazes extending both ends, is similar causes strain hardening, anduniform ensures expansion the uniformofexpansion of crazestoextending to which both ends, which to is metal deformation hardening caused by uniform deformation. similar to metal deformation hardening caused by uniform deformation. The curve clearly clearly differs differs from from the the ABS ABS curve. curve. As Asthe theload loadincreased, increased,PEEK PEEKfirst firstyielded yieldedatata The PEEK PEEK curve amaximum maximumstress, stress,then thennecking neckingdeformation deformationappeared appearedatat the the tensile tensile fracture fracture surface surface and and the the samples samples broke after reaching the maximum stress, when the strain reached 87%. In PEEK samples, there waswas no broke after reaching the maximum stress, when the strain reached 87%. In PEEK samples, there obvious necking deformation at the tensile fracture surface, and and no obvious neckneck propagation untiluntil the no obvious necking deformation at the tensile fracture surface, no obvious propagation fracture of samples afterafter yielding. the fracture of samples yielding. Figure 5 shows the average Figure 5 shows the average tensile tensile property property values values for for both both ABS ABS and and PEEK. PEEK. The The elastic elastic limit limit from from five tensile experiments with PEEK was 50.8 MPa; the tensile strength of PEEK was 56.6 MPa, while five tensile experiments with PEEK was 50.8 MPa; the tensile strength of PEEK was 56.6 MPa, while the the elastic elastic limit limit of of ABS ABS was was 22.9 22.9 MPa, MPa, and and the the tensile tensile strength strength of of ABS ABS was was 27.1 27.1 MPa. MPa. The The values values of of these 122% andand 108% higher thanthan for ABS. As shown in the in figure, the tensile these properties propertiesfor forPEEK PEEKwere were 122% 108% higher for ABS. As shown the figure, the properties of 3D-printed ABS samples were lower than injection-molded ABS by 26.2% for the elastic tensile properties of 3D-printed ABS samples were lower than injection-molded ABS by 26.2% for the limit 26.8% the tensile Likewise, the tensile of 3D-printed PEEK samples elasticand limit and for 26.8% for thestrength. tensile strength. Likewise, theproperties tensile properties of 3D-printed PEEK were lower thanlower injection-molded PEEK by 29.4% elasticfor limit 43.4% for the samples were than injection-molded PEEKforbythe29.4% theand elastic limit andtensile 43.4%strength. for the However, the lower strength 3D-printed samples compared with injection-molded samples can be tensile strength. However, theoflower strength of 3D-printed samples compared with injection-molded mainly the gaps betweentofilaments the extensive poresand within for samplesattributed can be to mainly attributed the gapsand between filaments the filaments, extensive especially pores within PEEK samples because the poor printing quality filaments, especially forofPEEK samples because of obtained. the poor printing quality obtained.

Figure 5. Comparison on tensile property between acrylonitrile butadiene styrene (ABS) Figure 5. Comparison on tensile property between acrylonitrile butadiene styrene (ABS) and polyether-ether-ketone (PEEK). and polyether-ether-ketone (PEEK).

Materials 2015, 8 Materials 2015, 8

5841 3890

Figure Figure 66 shows shows scanning scanning electron electron microscopy microscopy (SEM) (SEM) images images of of fracture fracture cross-sections cross-sections of of PEEK PEEK and and ABS along the longitudinal direction. These images of the fracture surface show that failure was caused ABS along the longitudinal direction. These images of the fracture surface show that failure was by differing reasons. Although individual had melted we cantogether, still distinguish caused by differing reasons. ABS Although ABSrasters individual rasterstogether, had melted we canevery still raster in the images, and the fracture of ABS was mainly caused by damage to the rasters pulling and distinguish every raster in the images, and the fracture of ABS was mainly caused by damage to the rupturing. As the force increased, the force unit areathe would the filaments tensile limit. rasters pulling andload rupturing. As the load force per increased, forcereach per unit area would reach the In the printed samples the fracture would begin approximately at the weakest filament, and the fracture filaments tensile limit. In the printed samples the fracture would begin approximately at the weakest would propagate the samples The result is that the stress continues to increase andthe thestress next filament, and theuntil fracture would failed. propagate until the samples failed. The result is that weakest will fail.and Bythe observing the failure ofBy PEEK samples, can see that there were continuesraster to increase next weakest rastersurfaces will fail. observing thewe failure surfaces of PEEK no obvious and thethere samples to have melted intosamples a block.appeared This was samples, werasters can see that wereappeared no obvious rasters and the to most have likely meltedcaused into a by a combination of the high extruder temperature and filaments that created significant thermal bonding block. This was most likely caused by a combination of the high extruder temperature and filaments between bothsignificant raster andthermal layers, causing fusion. that created bondinggreater between both raster and layers, causing greater fusion.

Figure 6. SEM images of fracture cross sections of polyether-ether-ketone (PEEK) Figure 6. SEM images of fracture cross sections of polyether-ether-ketone (PEEK) and acrylonitrile butadiene styrene (ABS). (a) ABS, 25ˆ; (b) ABS, 70ˆ; (c) PEEK, 25ˆ; and acrylonitrile butadiene styrene (ABS). (a) ABS, 25×; (b) ABS, 70×; (c) PEEK, 25×; (d) PEEK, 70ˆ. (d) PEEK, 70×. 3.3. Comparison of ABS and PEEK Compressive Properties Properties Figure 7 shows shows ABS ABS and and PEEK PEEK samples samples after after compressive compressive tests. tests. Figure 8 shows compressive engineering stress-straincurves curvesfor forABS ABSand andPEEK. PEEK.As Asthe thefigure figureillustrates, illustrates,thethe compressive strength engineering stress-strain compressive strength of of ABS is obvious, it is difficult to confirm unique compressive for PEEK. The ABS is obvious, whilewhile it is difficult to confirm a uniqueacompressive strength forstrength PEEK. The stress-strain stress-strain curves ABS are very similar to the tensile for ABS. The stress-stain curve of curves for ABS arefor very similar to the tensile curves for curves ABS. The stress-stain curve of PEEK is PEEK initiallyand linear the increases stress increases withdevelopment the development of significant deformation. With initiallyis linear the and stress with the of significant deformation. With an an increase compressivestress, stress,thethecurve curvebecomes becomes nonlinear nonlinear and and inelastic. inelastic. The The significant significant initial increase in in compressive initial

Materials 2015, 2015, 88 Materials

5842 3891 3891

deformation appears appearstoto toresult resultfrom from extensive pores becoming squeezed out. Ifcould we could could reduce the deformation extensive pores becoming squeezed out. out. If weIf reducereduce the pores appears result from extensive pores becoming squeezed we the pores within within the printed printed part, the compressive compressive strength and printing printing quality of 3D-printed 3D-printed PEEK parts within the printed part, the compressive strengthstrength and printing quality of 3D-printed PEEK parts could be pores the part, the and quality of PEEK parts could be substantially improved. substantially improved.improved. could be substantially

Figure 7. Fractured compressive compressive samples. (a) (a) acrylonitrile acrylonitrile butadiene butadiene styrene (ABS); 7. Fractured Fractured Figure 7. compressive samples. (a) acrylonitrile butadiene styrene (ABS); (b) polyether-ether-ketone (PEEK). polyether-ether-ketone (PEEK). (b) polyether-ether-ketone (PEEK).

Figure 8. Compressive stress-strain curves of acrylonitrile butadiene styrene (ABS) and Figure 8. 8. Compressive Compressive stress-strain stress-strain curves curves of of acrylonitrile acrylonitrile butadiene butadiene styrene styrene (ABS) (ABS) and and Figure polyether-ether-ketone (PEEK). polyether-ether-ketone (PEEK). (PEEK). polyether-ether-ketone The compressive strength and and compressive compressive modulus modulus of of five five Figure 9. 9. compressive strength five experiments experiments are The experiments are shown shown in in Figure According to to Figure Figure 9, 9, the strength of of PEEK PEEK was 60.9 MPa and the strength the compressive compressive strength was 60.9 MPa and the compressive compressive strength According of ABS ABS was was 28.4 Thus, the the value value for for PEEK PEEK was was 114% 114% higher while the the MPa. Thus, Thus, the value for PEEK was 114% higher than than that that for for ABS, ABS, while of 28.4 MPa. MPa. compressive both. As As the the figure shows, the 3D-printed samples failed at 76.7% compressive modulus moduluswas wassimilar similarforfor for both. As the figure shows, the 3D-printed 3D-printed samples failed at compressive modulus was similar both. figure shows, the samples failed at 76.7% of the the value of the the injection-molded injection-molded ABS.the Likewise, the injection-molded injection-molded PEEK failedand at of the value of value the injection-molded ABS. Likewise, injection-molded PEEK failed at 118 MPa, 76.7% of of ABS. Likewise, the PEEK failed at 1183D-printed MPa, and and the the 3D-printed samples failed at 76.7% of of the the injection-molded injection-molded PEEK. indicate The test testthat results the samples failed at 76.7% failed of the at injection-molded PEEK. The test results the 118 MPa, 3D-printed samples 76.7% PEEK. The results indicate that that strength the compressive compressive strength and modulus modulus of of ABS injection-molded ABS samples wereand higher by compressive and modulus of injection-molded samples wereABS higher by 76.9% 35.3% indicate the strength and injection-molded samples were higher by 76.9% and of 35.3% than those those of of the samples. 3D-printed ABS samples. samples. Similarly,PEEK the 3D-printed 3D-printed PEEKatsamples samples than those the 3D-printed ABS Similarly, the 3D-printed samples failed 51.6% 76.9% and 35.3% than the 3D-printed ABS Similarly, the PEEK failed at 51.6% 51.6% of of the the injection-molded injection-molded PEEK’s compressive strength,PEEK and 3D-printed 3D-printed PEEK samples of the at injection-molded PEEK’s compressive strength, and 3D-printed samples had 79.1% lower failed PEEK’s compressive strength, and PEEK samples had 79.1% 79.1% lower lower compressive modulus than that that of of PEEK. injection-molded PEEK. PEEK. compressive modulus than thatmodulus of injection-molded had compressive than injection-molded

Materials 2015, 8 Materials 2015, 8

5843 3892

Figure 9. Comparison on compressive property between acrylonitrile butadiene styrene Figure 9. Comparison on compressive property between acrylonitrile butadiene styrene (ABS) and polyether-ether-ketone (PEEK). (ABS) and polyether-ether-ketone (PEEK). 3.4. Comparison Comparison of of ABS ABS and and PEEK PEEK Bending Bending Properties Properties 3.4. Figure Figure 10 10 shows shows ABS ABS and and PEEK PEEK samples samples after after three-point three-point bending bending tests. tests. Figure Figure 11 11 shows shows bending bending engineering ABS and and PEEK samples. According to the GB/T9341-2008 bending engineering stress-strain stress-straincurves curvesforfor ABS PEEK samples. According to the GB/T9341-2008 standard,standard, the bending set as the value thatvalue causes deformation. The bending bending the strength bending is strength is set as the that3.5% causes 3.5% deformation. The strength bending and bending modulus of five experiments are shown in Figure 12. The bending strength of PEEK strength and bending modulus of five experiments are shown in Figure 12. The bending strengthwas of 56.2 MPa, than higher that ofthan ABSthat (48.6 MPa).(48.6 TheMPa). bending PEEK of wasPEEK 1.6 GPa, PEEK was 15% 56.2 higher MPa, 15% of ABS Themodulus bending of modulus was which waswhich very was closevery to that main the for poor property is the is weak 1.6 GPa, closeoftoABS. that ofThe ABS. Thereason main for reason theflexural poor flexural property the interlayer bonding. The 3D-printed ABS samples had bending strengthstrength and bending modulus modulus reduced weak interlayer bonding. The 3D-printed ABS samples had bending and bending by up toby 8.2% 20.8%, compared compared with thosewith of injection-molded ABS. The ABS. weak reduced up and to 8.2% andrespectively, 20.8%, respectively, those of injection-molded interlayer greatly influenced 3D-printed PEEK sample properties, because the bending strength The weakbonding interlayer bonding greatly influenced 3D-printed PEEK sample properties, because the and bending modulus were reduced by up to 65.5% respectively. bending strength and bending modulus were reducedand by58.9%, up to 65.5% and 58.9%, respectively.

Figure 10. Fractured bending bending samples. samples. (a) (a) acrylonitrile acrylonitrile butadiene butadiene styrene styrene (ABS); Figure 10. Fractured (ABS); (b) polyether-ether-ketone (PEEK). (b) polyether-ether-ketone (PEEK).

Materials 2015, 8

5844

Materials 2015, 8 Materials 2015, 8

3893 3893

Figure 11. 11. Bending of acrylonitrile acrylonitrilebutadiene butadiene styrene (ABS) Figure Bendingstress-strain stress-strain curves curves of styrene (ABS) and and Figure 11. Bending stress-strain curves of acrylonitrile butadiene styrene (ABS) and polyether-ether-ketone (PEEK). polyether-ether-ketone (PEEK). polyether-ether-ketone (PEEK).

Figure 12. Comparison on bending property acrylonitrile butadiene Figure 12. Comparison on bending property between between acrylonitrile butadiene styrene styrene (ABS) (ABS) and polyether-ether-ketone (PEEK). and polyether-ether-ketone (PEEK). Figure 12. Comparison on bending property between acrylonitrile butadiene styrene (ABS) and polyether-ether-ketone (PEEK). 4. Conclusions

4. Conclusions

The aim of this paper was to investigate the effects of raster angle and layer thickness on 4. Conclusions

properties samples. The experiments rasterthickness angle and on layer The aimmechanical of this paper was ofto3D-printed investigate the effects of raster confirmed angle andthat layer mechanical The aim of this paper was to investigate the effects of raster angle and layer thickness on thickness both have a marked effect on tensile, compressive and three-point bending properties. properties The of 3D-printed samples. The experiments confirmed that raster angle and layer thickness both optimal mechanical properties of PEEK were in samplesconfirmed with a 300-μm thickness and layer mechanical properties of 3D-printed samples. Thefound experiments that layer raster angle and have athickness marked effect on compressive three-point bending optimal mechanical a raster angle of tensile, 0°/90°. both have a marked effect on and tensile, compressive andproperties. three-point The bending properties. ˝ study also compared mechanical properties of PEEK andthickness ABS sample parts. Comparing properties of Our PEEK were found inthe samples with a 300-µm and a raster angletheof 0and /90˝ . The optimal mechanical properties of PEEK were found inlayer samples with a 300-μm layer thickness mechanical properties of ABS and PEEK samples made by 3D printing, it can be concluded that the Our study also of compared a raster angle 0°/90°. the mechanical properties of PEEK and ABS sample parts. Comparing the properties of each part are decreased through 3D-printing compared with those of the raw materials. Our study also compared properties of PEEK ABS sample Comparing that the the mechanical properties of ABS the andmechanical PEEK samples made by 3Dand printing, it canparts. be concluded In this study, the tensile strength of 3D-printed PEEK was about 56 MPa, which is equivalent to that of mechanical properties ABS andstudies, PEEK samples madeproperties bycompared 3D printing, it can be concluded that the properties of each part parts. areofdecreased through 3D-printing with those of themay raw materials. nylon injection In future the mechanical of 3D-printed PEEK parts be of each partstrength are decreased through 3D-printing those of the materials. In thisproperties study, the tensile of 3D-printed PEEK wascompared about 56with MPa, which is raw equivalent to that In this study, the tensile strength of 3D-printed PEEK was about 56 MPa, which is equivalent to that of of nylon injection parts. In future studies, the mechanical properties of 3D-printed PEEK parts may nylon injection parts. In future studies, the mechanical properties of 3D-printed PEEK parts may be be improved by increasing the control accuracy and hardware precision of the 3D-printing system. In this study, the mechanical properties of 3D-printed PEEK samples (tensile, compressive and three-point bending) were higher than those of ABS samples printed by commercial 3D printers. Specifically, the tensile, compression and bending strengths of PEEK samples were higher than those of ABS samples

Materials 2015, 8

5845

by 108%, 114% and 115%, respectively, while little significant difference was found between the compressive and flexural modulus of PEEK and ABS. Experimental studies and comparative analyses were carried out to study the factors affecting PEEK 3D print forming quality, in the hope of providing reference conditions under which to print PEEK. However, we recognize that there is much work left to do in this area. Further research is needed to reduce pore formation during the printing process and to improve interlayer bonding. PEEK has favorable properties, including excellent mechanical properties in both static and dynamic tests and high chemical resistance [21,22]. It is believed that PEEK may be a significant and promising material for industrial applications of 3D-printed components. Acknowledgments This research is supported by National Natural Science Foundation of China (No. 51205163), Specialized Research Fund for the Doctoral Program of Higher Education of China “SRFDP” (No. 20120061120030), Young Research Fund of Jilin Province Science and Technology Development Plan (No. 20140520124JH) and Young Teachers and Students to Interdisciplinary Cultivating Project of Jilin University (No. JCKY-QKJC28). Author Contributions Wenzheng Wu designed the experiment and wrote the final manuscript. Peng Geng Guiwei Li and Di Zhao wrote the first draft and prepared the experiment. Haibo Zhang carried out the experimental study. Ji Zhao supervised the research, design of the study and carried out the conception. All authors contributed to the analysis for results and conclusions and revised the paper. Conflicts of Interest The authors declare no conflict of interest. References 1. Ahn, S.; Montero, M.; Odell, D.; Roundy, S.; Wright, P.K. Anisotropic material properties of fused depositionmodeling ABS. Rapid Prototyp. J. 2003, 8, 248–257. [CrossRef] 2. Sun, Q.; Rizvi, G.M.; Bellehumeur, C.T.; Gu, P. Effect of processing conditions on the bonding quality of FDM polymer filaments. Rapid Prototyp. J. 2008, 14, 72–80. [CrossRef] 3. Upcraft, S.; Fletcher, R. The rapid prototyping technologies. Rapid Prototyp. J. 2003, 23, 318–330. [CrossRef] 4. Martínez, J.; Diéguez, J.L.; Ares, E.; Pereira, A.; Hernández, P.; Pérez, J.A. Comparative between FEM models for FDM parts and their approach to a real mechanical behavior. Procedia Eng. 2013, 63, 878–884. [CrossRef] 5. Es-Said, O.S.; Foyos, J.; Noorani, R.; Mendelson, M.; Marloth, R.; Pregger, B.A. Effect of layer orientation on mechanical properties of rapid prototyped samples. Mater. Manuf. Process. 2000, 15, 107–122. [CrossRef] 6. Mota, C.; Puppi, D.; Dinucci, D.; Errico, C.; Bártolo, P.; Chiellini, F. Dual-Scale polymeric constructs as scaffolds for tissue engineering. Materials 2011, 4, 527–542. [CrossRef]

Materials 2015, 8

5846

7. Anitha, R.; Arunachalam, S.; Radhakrishnan, P. Critical parameters influencing the quality of prototypes in fused deposition modeling. J. Mater. Process. Technol. 2001, 118, 385–388. [CrossRef] 8. Bernarand, A.; Fischer, A. New trends in rapid product development. CIRP Ann. Manuf. Technol. 2002, 51, 635–652. [CrossRef] 9. Thrimurthulu, K.; Pandey, P.M.; Reddy, N.V. Optimum part deposition orientation in fused deposition modeling. Int. J. Mach. Tools Manuf. 2004, 44, 585–594. [CrossRef] 10. Wu, G.; Langrana, N.A.; Sadanji, R.; Danforth, S. Solid freeform fabrication of metal components using fused deposition of metals. Mater. Des. 2002, 23, 97–105. [CrossRef] 11. Nikzad, M.; Masood, S.H.; Sbarski, I. Thermo-mechanical properties of a highly filled polymeric composites for Fused Deposition Modeling. Mater. Des. 2011, 32, 3448–3456. [CrossRef] 12. Lee, J.; Huang, A. Fatigue analysis of FDM materials. Rapid Prototyp. J. 2013, 19, 291–299. [CrossRef] 13. Croccolo, D.; de Agostinis, M.; Olmi, G. Experimental characterization and analytical modelling of the mechanical behaviour of fused deposition processed parts made of ABS-M30. Comput. Mater. Sci. 2013, 79, 506–518. [CrossRef] 14. Vigliotti, A.; Pasini, D. Stiffness and strength of tridimensional periodic lattices. Comput. Methods Appl. Mech. Eng. 2012, 229, 27–43. [CrossRef] 15. Durgun, I.; Ertan, R. Experimental investigation of FDM process for improvement of mechanical properties and production cost. Rapid Prototyp. J. 2014, 20, 228–235. [CrossRef] 16. Sood, A.K.; Ohdar, R.K.; Mahapatra, S.S. Experimental investigation and empirical modeling of FDM process for compressive strength improvement. J. Adv. Res. 2012, 32, 81–90. [CrossRef] 17. Tymrak, B.M.; Kreiger, M.; Pearce, J.M. Mechanical properties of components fabricated with open-source 3-D printers under realistic environmental conditions. Mater. Des. 2014, 58, 242–246. [CrossRef] 18. Sood, A.K.; Ohdar, R.K.; Mahapatra, S.S. Parametric appraisal of mechanical property of fused deposition modeling processed parts. Mater. Des. 2010, 31, 287–295. [CrossRef] 19. Agarwala, M.K.; Jamalabad, V.R.; Langrana, N.A.; Safari, A.; Whalen, P.J.; Danforth, S.C. Structural quality of parts processed by fused deposition. Rapid Prototyp. J. 1996, 8, 248–257. [CrossRef] 20. Kantaros, A.; Karalekas, A. Fiber Bragg grating based investigation of residual strains in ABS parts fabricated by fused deposition modeling process. Mater. Des. 2013, 50, 44–50. [CrossRef] 21. Bigg, D.M. Mechanical, thermal, and electrical properties of metal fiber-filled polymer composites. Polym. Eng. Sci. 1979, 19, 1188–1192. [CrossRef] 22. Bigg, D.M. Mechanical properties of particulate filled polymers. Polym. Compos. 1987, 8, 115–122. [CrossRef] © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).