- Email: [email protected]
Characterization of residual stress and deformation in additively manufactured ABS polymer and composite specimens
Composites Science and Technology 150 (2017) 102e110
Contents lists available at ScienceDirect
Composites Science and Technology journal homepage: http://www.elsevier.com/locate/compscitech
Characterization of residual stress and deformation in additively manufactured ABS polymer and composite specimens Wei Zhang a, b, c, Amanda S. Wu d, Jessica Sun c, Zhenzhen Quan a, Bohong Gu a, Baozhong Sun a, **, Chase Cotton e, Dirk Heider c, Tsu-Wei Chou b, c, * a
College of Textiles, Donghua University, Shanghai 201620, PR China Department of Mechanical Engineering, University of Delaware, Newark, DE 19716, USA Center for Composite Materials, University of Delaware, Newark, DE 19716, USA d Materials Engineering Division, Lawrence Livermore National Laboratory, CA 94550, USA e Department of Electrical and Computer Engineering, University of Delaware, Newark, DE 19716, USA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 May 2017 Received in revised form 13 July 2017 Accepted 16 July 2017 Available online 17 July 2017
Residual stresses induced in the layer-by-layer fabrication process of additively manufactured parts have significant impact on their mechanical properties and dimensional accuracy. This work aims to characterize the residual stress and deformation in specimens based on unreinforced acrylonitrile-butadienestyrene (ABS), carbon nanotube reinforced ABS and short carbon fiber reinforced ABS. The shrinkage and displacement fields were obtained, respectively, by thermal treatment as well as Digital Image Correlation observation of specimens before and after sectioning. The microstructure and porosity of additively manufactured specimens were also examined using X-ray micro-computed tomography. Specimen shrinkage and porosity content were significantly influenced by the process parameters of raster angle and printing speed, as well as material types. Faster printing speed led to larger porosity and residual stress, as well as higher shrinkage after specimen thermal treatment. Raster angle had a greater influence on specimen shrinkage and porosity as comparing to printing speed. Composite printing wires based on carbon nanotube and short carbon fiber in ABS greatly reduced specimen shrinkage and deformation, while increased the porosity, especially for carbon fiber reinforced ABS specimens. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Residual stress Shrinkage Microstructure Processing parameters Additive manufacturing
1. Introduction Additive manufacturing is defined as a “process of joining materials to make parts from 3D model data, usually layer upon layer, as opposed to subtractive manufacturing and formative manufacturing methodologies” [1] by ISO/ASTM. It refers to a family of fabrication methodologies including fused deposition modeling (FDM) [2], stereo lithography (SLA) [3], selective laser sintering (SLS) and selective laser melting (SLM) [4], inkjet modeling (IJM) [5] and others [6]. In recent years, additive manufacturing has evolved rapidly and been widely used in various manufacturing fields such as aerospace [7], automobile [8], biomedical [9], building [10] and many others [11e13]. This
* Corresponding author. Department of Mechanical Engineering, University of Delaware, Newark, DE 19716, USA. ** Corresponding author. E-mail addresses: [email protected] (B. Sun), [email protected] (T.-W. Chou). http://dx.doi.org/10.1016/j.compscitech.2017.07.017 0266-3538/© 2017 Elsevier Ltd. All rights reserved.
tremendous success could be attributed mainly to its outstanding ability to directly manufacture complex parts without special tools, to greatly reduce material waste, and to significantly reduce the time and cost of manufacturing for novel products and smallquantity productions [14e18]. The effects of an array of processing parameters, such as raster angle, printing speed, layer thickness, etc. on the mechanical properties of additively manufactured parts have been well documented in the literature [19e23]. In order to improve the performance of additively manufactured parts and broaden their applications, researchers have explored various types of printing materials, such as polymers, nanocomposites, fiber reinforced composites, etc. [24e31]. Another important aspect relevant to the mechanical performance of additively manufactured parts is repeated heating and cooling [32,33] resulting from the layer-bylayer building process. Significant residual stresses [34] can thus exist in additively manufactured part. Apart from their detrimental influence on mechanical performance, residual stresses could give rise to part distortion and dimensional inaccuracy [35e38]. Some
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
efforts have been made by researchers in identifying residual stresses in additively manufactured parts. For example, Karalekas and Rapti [39]. used epoxy based photopolymer to study the processing dependence of SLA solidification residual stress using the hole-drilling strain gage method of stress relaxation. Karalekas and Aggelopoulos [40] investigated the shrinkage strains in a SLA cured acrylic photopolymer resin. Kantaros et al. [41] studied the residual strains in ABS parts fabricated by FDM using fiber Bragg grating method. Casavola et al. [42] measured the residual stress in FDM parts made of ABS employing the hole-drilling method combined with electronic speckle pattern interferometry. These research efforts have greatly facilitated the understanding of residual stresses in additively manufactured parts. However, more studies with respect to characterization techniques, printing materials and processing parameters still need to be conducted to minimize the detrimental effects of residual stress. This work aims to characterize the residual stress and deformation in additively manufactured specimens based on unreinforced ABS, carbon nanotube reinforced ABS and carbon fiber reinforced ABS. The shrinkages and displacement fields were obtained by thermal treatment as well as Digital Image Correlation combined with sectioning of specimens, respectively. The effects of two key processing parameters, namely, raster angle and printing speed, as well as reinforcement materials, including carbon nanotube and carbon fiber, on specimen properties have been characterized. The microstructure and porosity of additively manufactured specimens were also examined using X-ray microcomputed tomography.
103
Fig. 1. TGA plot of the as-received CNTABS wire under temperature range of the 25e750 C.
resolution of 0.01 mm. The shrinkage of an additively manufactured specimen in a specific direction was calculated as follows: Shrinkage (%) ¼ ((original dimension (mm.) - dimension after thermal treatment (mm.)) / original dimension (mm.)) 100%
2. Experiment 2.1. Specimen preparation In this work, fused deposition modeling was adopted to fabricate additively manufactured specimens. A total of 27 types of specimens have been fabricated using a QIDI Tech dual-nozzle 3D printer (Ruan Qidi Technology Co., Ltd., Ruan, Zhejiang, China). Three kinds of materials, including pure ABS wire, carbon nanotube reinforced ABS wire (CNTABS) and short carbon fiber reinforced ABS wire (CFABS) (3DTECH, USA) were used. The carbon nanotube content in the wire was measured to be around 8 wt% via thermogravimetric analysis (TGA) in the nitrogen gas, as shown in Fig. 1. The significant weight loss of the samples reflected the thermal decomposition of the ABS matrix. The short carbon fiber content was around 15 wt% [16]. The length distribution of short carbon fibers in the wire was also given in Ref. [16]. The carbon fiber lengths varied from 5 mm to 465 mm, with a number-based average of 71.5 mm. In the fabrication process, three raster angles of 0 , ±45 and 0 /90 as well as three printing speeds of 40 mm/s, 60 mm/s and 80 mm/s were selected to build the specimen in the horizontal plane (X-Y). The raster angle indicates the direction of wire printing with respect to the longitudinal X-axis of the specimens, as shown in Fig. 2. The other processing parameters are shown in Table 1. Two kinds of specimen sizes of around 75 20 3 mm and 40 15 3 mm were fabricated to study the shrinkage and displacement fields, respectively. There are 12e13 layers each specimen. 2.2. Shrinkage measurement In order to measure the shrinkage of additively manufactured specimens, thermal treatment was carried out with the heating temperature of 180 C and heating time of 1 h. The dimensions in length, width and height of specimens before and after thermal treatment were measured by a Vernier caliper with the reading
2.3. Surface displacement field measurement In an effort to characterize specimen residual stress induced by the additive manufacturing process, surface displacement fields were measured by Digital Image Correlation. Digital Image Correlation is a 3D, full-field, non-contact optical technique to measure contour, deformation, vibration and strain on almost any material. In this work, it was employed by digitally comparing images acquired before and after releasing the localized stress by sectioning the specimen. In the measurement, a specimen was painted using a flat white spray paint and a black speckle pattern was manually applied using a flat black spray paint, with a minimum drying time of 24 h at ambient pressure and room temperature in between coatings. After that, the specimen was imaged using a dual camera setup before and after sectioning via lubricated diamond saw. Fig. 3 shows the images of three specimens (ABS, CFABS and CNTABS) before and after painting and sectioning. The displacement fields were calculated using ARAMIS Professional Digital Image Correlation algorithm (GOM International AG) by comparing the acquired images before and after sectioning. The displacement field so obtained is indicative of the surface-level residual stress in the specimen. 2.4. X-ray micro-computed tomography The SkyScan 1172 X-ray micro-CT system (Bruker Corp., Billerica, Massachusetts, U.S.) was used to characterize the microstructure and void distribution of additively manufactured specimens. In order to save scanning time, specimens were cut to the size of 8 20 3 mm. X-ray source voltage and current were 40 KV and 250 mA, respectively. The scanning resolution was around 5.4 mm/ pixel. The image size was set as 2048 1024 pixel. A series of 2D images in X-Z plane were acquired to record the attenuation of X-
104
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 2. Illustration of raster angles for the additively manufactured specimens.
Table 1 Processing parameters. Parameter
Value
Nozzle diameter Nozzle temperature Print platform temperature Filament feed rate Layer thickness
0.4 mm 230 C 110 C 1.0 mm/s 0.24 mm
Fig. 4. Thermal deformation photos of additively manufactured ABS specimens after thermal treatment at printing speeds of 40, 60 and 80 mm/s (a) 0 , (b) ±45 and (c) 0 / 90 .
Fig. 3. Additively manufactured ABS (top), CFABS (middle) and CNTABS (bottom) specimens shown as-built (left) and after speckle pattern application and sectioning (right).
ray photons in scanning a specimen. A series of cross-section microstructure images were processed by the SkyScan reconstruction program NRecon from these 2D images. The other two cross-section images in X-Y and Y-Z plane were obtained by transformation using software of DataViewer. The porosity of each specimen was obtained by calculating the average of porosities of all slices using the software of CTAn.
3. Results and discussions 3.1. Thermal shrinkage analysis Thermal shrinkage images were obtained for ABS polymer and its composite specimens with three raster angles (0 , ±45 and 0 / 90 ), and three printing speeds (40 mm/s, 60 mm/s and 80 mm/s) after thermal treatment. Fig. 4 shows the comparison of ABS specimens with and without thermal treatment for different raster angles and printing speeds. It can be seen that the heat treated ABS specimens exhibit high shrinkage deformation along the longitudinal direction, especially for raster angle of 0 . This could be attributed to the release of residual stress after thermal treatment.
Furthermore, the shrinkage of specimens increases with increasing printing speed for all raster angles. However, the 0 specimens also exhibit expansion along the transverse direction. In order to clearly identify the thermal deformation of ABS specimens, the shrinkage induced specimen size changes in length, width and height after thermal treatment are shown in Table 2. Positive shrinkage means contraction, while negative value indicates expansion. Three measurements were made along each direction and the average percentage was calculated as the shrinkage. Here, A0, A[0/90] and A[±45] denote the raster angles of 0 , 0 /90 and ±45 , respectively. Also, S40, S60 and S80 denote the printing speeds of 40 mm/s, 60 mm/s and 80 mm/s, respectively. It can be seen that the 0 specimens contract in the length direction and expand in the width and height directions, due to the Poisson's effect in the printed wires. However, for raster angles of 0 /90 and ±45 , the ABS specimens shrink in both length and width directions, while expand in thickness direction. This is due to the orthotropic nature of these two types of specimens. It is also worth noting that the difference of shrinkages between the length and width directions is significant for 0 /90 specimens, and much less for ±45 specimens. This is because the different printing path length of the 0 and 90 layers in the 0 /90 specimens, leading to different temperature changes (as demonstrated in Fig. 9 of reference [43]), and hence, difference in the resulting residual stresses between the length and width directions. Fig. 5 shows the thermal deformation of additively manufactured CNTABS specimens. It is obvious that the heat treated CNTABS specimens contracted along the longitudinal direction for all
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
105
Table 2 Shrinkage induced specimen size changes. Specimen types
Shrinkage (%) Length
Width
Height
ABS-A0-S40 ABS-A0-S60 ABS-A0-S80 ABS-A[0/90]-S40 ABS-A[0/90]-S60 ABS-A[0/90]-S80 ABS-A[±45]-S40 ABS-A[±45]-S60 ABS-A[±45]-S80
28.8 38.74 47.3 10.31 16.54 21.64 6.4 10.77 13.51
13.41 21.99 32.76 8.62 9.15 11.33 8.37 11.23 13.78
23.65 35.64 45.38 20.82 27.83 34.7 10.6 21.41 21.7
CNTABS-A0-S40 CNTABS-A0-S60 CNTABS-A0-S80 CNTABS-A[0/90]-S40 CNTABS-A[0/90]-S60 CNTABS-A[0/90]-S80 CNTABS-A[±45]-S40 CNTABS-A[±45]-S60 CNTABS-A[±45]-S80
10.57 18.18 20.55 5.45 7.99 9.54 4.63 5.92 6.78
0.91 4.87 7.46 9.36 10.7 10.55 8.41 10.35 11.78
13.95 18.75 20.33 18.18 22.82 23.23 19.98 20.07 21.55
CFABS-A0-S40 CFABS-A0-S60 CFABS-A0-S80 CFABS-A[0/90]-S40 CFABS-A[0/90]-S60 CFABS-A[0/90]-S80 CFABS-A[±45]-S40 CFABS-A[±45]-S60 CFABS-A[±45]-S80
3.48 5.43 7.27 2.53 3.9 4.94 1.73 1.86 2.89
0.05 0.48 0.86 4.64 5.22 5.8 4.44 5.91 6.44
1.42 2.47 1.41 4.23 0.57 1.68 0.35 1.05 4.54
Fig. 5. Thermal deformation photos of additively manufactured CNTABS specimens after thermal treatment at printing speeds of 40, 60 and 80 mm/s (a) 0 , (b) ±45 and (c) 0 /90 .
Fig. 6. Thermal deformation photos of additively manufactured CFABS specimens after thermal treatment at printing speeds of 40, 60 and 80 mm/s (a) 0 , (b) ±45 and (c) 0 / 90 .
CNTABS specimens, shrinkage occurs in the thickness of CFABS specimens with raster angles of 0 and 0 /90 . The effects of raster angle and printing speed on the length shrinkages of ABS, CNTABS and CFABS specimens are compared in Fig. 7(a)-(c). It can be clearly seen that shrinkage increases with increasing printing speed for all three kinds of materials at the three raster angles. This implies that the residual stress increases with increasing printing speed. It also can be concluded that the raster angle has a greater influence on length shrinkage compared with printing speed. For all three kinds of materials and three printing speeds, the shrinkages of 0 specimens are larger than those of specimens with other raster angles, particularly for ABS and CNTABS specimens. Furthermore, length shrinkage is highly anisotropic, i.e. orientation dependent. This could be attributed to the effect of flow-induced molecular and reinforcement (CNT, CF) alignment, which greatly constrain the shrinkage of neighboring layers in the 0 /90 and ±45 laminated structures. The ±45 specimens show the minimal shrinkage. Fig. 7(d) presents the length shrinkages of ABS, CNTABS and CFABS specimens at the same printing speed of 60 mm/s. It can be seen that the shrinkages of CFABS specimens are much less than those of ABS specimens at all raster angles, due to the constraining effect of short carbon fibers on specimen contraction. The magnitude of shrinkage of CNTABS specimens lies in between those of ABS and CFABS specimens. Thus, carbon nanotubes are less effective in reducing the shrinkage comparing to short carbon fibers. The influence of volume contents of CNT and CF has not been examined. 3.2. Surface displacement field induced by residual stress
printing speeds and raster angles. Shrinkage induced specimen size change are also shown in Table 2. Overall, the trends of shrinkage deformation of CNTABS specimens are similar to those of pure ABS specimens, and the deformation increases with increasing printing speed. Again, just as ABS specimens, bending also appears in CNTABS specimens. Fig. 6 shows the thermal deformation of additively manufactured CFABS specimens. As indicated in Table 2, the overall dimensional changes in this case are much less than those of ABS and CNTABS specimens, indicating the dominating effect of short carbon fibers in resisting specimen shrinkage. Also, unlike ABS and
Digital Image Correlation measurements in conjunction with specimen sectioning have been employed to identify the surface displacement fields of additively manufactured ABS, CNTABS and CFABS specimens upon releasing the residual stress. Fig. 8 shows the displacement fields on an ABS specimen without heat treatment, which was sectioned into two parts A and B. Referring to the coordinate axes of Fig. 2, the red and blue colors signify positive and negative displacements, respectively, while green color indicates no displacement. In Fig. 8(a), X displacements occur along the edges opposite to the cut surface, and in Fig. 8(b), Y displacements indicate opposite bending deflections in Part A and B of the ABS
106
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 7. Comparison of effects of raster angle and printing speed on length shrinkages for (a) ABS, (b) CNTABS and (c) CFABS, as well as the effects of materials types on length shrinkages (d).
Fig. 8. Surface displacement distributions on a sectioned ABS specimen without thermal treatment: (a) X-displacement, (b) Y-displacement.
specimen. However, there is essentially no displacement in the thermally treated specimen (Fig. 9). Thus, it can be concluded that fabrication induced residual stress were released by thermal treatment, and the magnitude of deformation in specimens after heat treatment reflects the level of residual stresses. In an effort to compare the deformations among ABS, CNTABS,
and CFABS specimens, the Y displacement distributions on the surfaces of both Part A and B are shown in Fig. 10. In Part A of the specimens, the Y displacements are negative on both ends, while positive in the middle. There is an obvious upward bending deflection. For Part B of the specimens, the Y displacements are positive on both ends, while negative in the middle of the
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 9. Heat treated ABS specimen shows nearly no surface displacement.
specimens, indicating a downward bending deflection. The relative sizes of the positive and negative Y-displacement regions in either Part A or B specimen are the indication of the severity of bending deformation. The bending deflection is the highest in ABS specimen and the lowest in CFABS specimen. The presence of carbon nanotube and carbon fiber in the composite specimens of CNTABS and CFABS improve their flexural modulus and impede the bending deformation. Fig. 11 gives the profiles of Y-displacement at various points on the line along X-direction near the cut surface of ABS, CNTABS and CFABS specimens. In both Part A and B, the largest Y-displacement in ABS gives rise to the highest specimen bending curvature while CFABS shows the smallest Y-displacement and lowest specimen bending curvature. Lastly, it should be mentioned that the opposite direction deflection of Part A and B in a specimen result in expansion in its width as shown in Table 2. 3.3. Characterization of microstructure and porosity The microstructural and porosity characterization of additively manufactured specimens were carried out via X-ray microcomputed tomography. For the effects of printing speed, Fig. 12(a)-(c) show Y-Z plane cross-sections of 0 ABS specimens at three printing speeds of 40, 60 and 80 mm/s. It can be seen that
107
void size increases with increasing printing speed. The porosity of each specimen is the average of porosity values of all sections. The porosities of specimens with printing speeds of 40, 60 and 80 mm/s are, respectively, 0.51%, 0.82% and 3.23%. This finding can be attributed to the different cooling rates of printed wires. Faster printing speed allows less time for the development of interwire adhesion, and, hence, larger pores as well as greater shrinkage when specimens are heat treated. Fig. 12(a’)-(c’) show the X-Y plane cross-sections. Here, the void size and distribution can be seen clearly and the result is consistent with those in Fig. 12(a)-(c). Also, it is worth noting that fast printing speed may lead to the unevenness of the printed wires as shown in Fig. 12(c’). The effects of raster angle on specimen microstructure are demonstrated in Fig. 13(a)-(c). Here, the Y-Z plane cross-sections of ABS specimens are shown for the raster angles of 0 , ±45 and 0 / 90 at printing speed of 60 mm/s. It can be seen that many small voids are evenly distributed in the cross-section of 0 specimen. For raster angle of 0 /90 , numerous large voids exist between the mutually perpendicular printed wires. In comparing to the cases of 0 and 0 /90 , the ±45 specimen shows much less porosity. The porosities of 0 , ±45 and 0 /90 specimens of Fig. 13 are 0.82%, 0.48% and 1.73%, respectively. Fig. 13(a’)-(c’) show the X-Y plane cross-sections for the raster angles of 0 , ±45 and 0 /90 . The pore size and distribution between the printed wires for the X-Y crosssection are consistent with those for the Y-Z cross-section in Fig. 13(a)-(c). Lastly, a comparison among ABS, CNTABS and CFABS specimens is presented. Fig. 14 shows the difference of cross-section microstructure between them. It can be seen that specimen void size increase in the order of ABS, CNTABS and CFABS for both Y-Z and XY cross-sections. Again, these findings confirm that the addition of carbon nanotubes and short carbon fibers constrain the flow of printed wires in the fabrication process, and gives rise to high porosities of 0.48%, 2.39% and 8.54%, respectively, for ABS, CNTABS and CFABS specimens.
Fig. 10. Y-displacement fields of additively manufactured (a) ABS, (b) CNTABS and (c) CFABS specimens.
108
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 11. Profiles of Y-displacement at various points on the line along the X-direction near the cut surface of ABS, CNTABS and CFABS specimens (a) Part A and (b) Part B.
Fig. 12. YeZ plane cross-section microstructures of 0 ABS specimens at printing speeds of (a) 40 mm/s, (b) 60 mm/s and (c) 80 mm/s (a’), (b’) and (c’) are the corresponding X-Y plane cross-section microstructures.
4. Conclusions Major findings of this study are summarized as follows. (1) Faster printing speed leads to larger porosity and residual stress, as well as higher shrinkage after specimen thermal treatment. (2) Specimen shrinkage and porosity content are significantly influenced by the process parameters, raster angle and printing speed, as well as material type. (3) Raster angle has a greater influence on specimen shrinkage and porosity as comparing to printing speed. ABS specimen with raster angle of ±45 show less shrinkage and porosity as comparing to those with raster angle of 0 and 0 /90 .
(4) Composite printing wires based on carbon nanotube and short carbon fiber in ABS greatly reduce specimen shrinkage and deformation especially for CFABS specimens. Void size increases in the order of ABS, CNTABS and CFABS specimens. (5) Digital Image Correlation provides an effective means for measuring specimen surface displacement after sectioning. It provides a unique insight into fabrication induced residual stress. (6) X-ray micro-computed tomography, a nondestructive measurement technique, clearly demonstrates internal microstructure and void size of additively manufactured specimens.
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
109
Fig. 13. YeZ plane cross-section microstructures of ABS specimens at printing speed of 60 mm/s (a) 0 , (b) ±45 and (c) 0 /90 . (a’), (b’) and (c’) are the corresponding X-Y plane cross-section microstructures.
Fig. 14. YeZ plane cross-section microstructures of (a) ABS, (b) CNTABS and (c) CFABS specimens with raster angle of ±45 at printing speed of 60 mm/s (a’), (b’) and (c’) are the corresponding X-Y plane cross-section microstructure.
Acknowledgements Wei Zhang acknowledges the financial support from the China
Scholarship Council (CSC) and the Fundamental Research Funds for the Central Universities of China.
110
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
References [1] ISO/ASTM 52900:2015(E), Standard Terminology for Additive ManufacturingGeneral Principles-Terminology, 2015. [2] S.S. Crump, Apparatus and Method for Creating Three-dimensional Objects, Stratasys Incorporated, US, 1992. [3] P.F. Jacobs, Rapid Prototyping & Manufacturing: Fundamentals of Stereolithography, Society of Manufacturing Engineers, 1992. [4] J.-P. Kruth, B. Vandenbroucke, v.J. Vaerenbergh, P. Mercelis, Benchmarking of Different SLS/SLM Processes as Rapid Manufacturing Techniques, 2005. [5] M. Singh, H.M. Haverinen, P. Dhagat, G.E. Jabbour, Inkjet printing-process and its applications, Adv. Mater. 22 (6) (2010) 673e685. [6] J. Gardan, Additive manufacturing technologies: state of the art and trends, Int. J. Prod. Res. 54 (10) (2015) 3118e3132. [7] L.J. Kumar, C.G. Krishnadas Nair, Current Trends of Additive Manufacturing in the Aerospace Industry, 2017, pp. 39e54. [8] S. Curran, P. Chambon, R. Lind, L. Love, R. Wagner, S. Whitted, D. Smith, B. Post, R. Graves, C. Blue, Big area additive manufacturing and hardware-in-the-loop for rapid vehicle powertrain prototyping: A case study on the development of a 3-D-printed Shelby Cobra, SAE Technical Paper, 2016. [9] A.A. Zadpoor, J. Malda, Additive manufacturing of biomaterials, tissues, and organs, Ann. Biomed. Eng. 45 (1) (2017) 1e11. [10] K. Biswas, J. Rose, L. Eikevik, M. Guerguis, P. Enquist, B. Lee, L. Love, J. Green, R. Jackson, Additive manufacturing integrated energydenabling innovative solutions for buildings of the future, J. Sol. Energy Eng. 139 (1) (2016) 015001. [11] A. Ambrosi, M. Pumera, 3D-printing technologies for electrochemical applications, Chem. Soc. Rev. 45 (10) (2016) 2740e2755. [12] N. Guo, M.C. Leu, Additive manufacturing: technology, applications and research needs, Front. Mech. Eng. 8 (3) (2013) 215e243. [13] M.G. Mohammed, R. Kramer, All-printed Flexible and Stretchable Electronics, Advanced Materials, 2017. [14] A.R. Studart, Additive manufacturing of biologically-inspired materials, Chem. Soc. Rev. 45 (2) (2016) 359e376. [15] Z. Quan, A. Wu, M. Keefe, X. Qin, J. Yu, J. Suhr, J.-H. Byun, B.-S. Kim, T.-W. Chou, Additive manufacturing of multi-directional preforms for composites: opportunities and challenges, Mater. Today 18 (9) (2015) 503e512. [16] Z. Quan, Z. Larimore, A. Wu, J. Yu, X. Qin, M. Mirotznik, J. Suhr, J.-H. Byun, Y. Oh, T.-W. Chou, Microstructural design and additive manufacturing and characterization of 3D orthogonal short carbon fiber/acrylonitrile-butadienestyrene preform and composite, Compos. Sci. Technol. 126 (2016) 139e148. [17] X. Wang, M. Jiang, Z. Zhou, J. Gou, D. Hui, 3D printing of polymer matrix composites: a review and prospective, Compos. Part B Eng. 110 (2017) 442e458. [18] Z. Quan, Z. Larimore, X. Qin, J. Yu, M. Mirotznik, J.-H. Byun, Y. Oh, T.-W. Chou, Microstructural characterization of additively manufactured multi-directional preforms and composites via X-ray micro-computed tomography, Compos. Sci. Technol. 131 (2016) 48e60. [19] G.P. Tandon, T.J. Whitney, R. Gerzeski, H. Koerner, J. Baur, Process parameter effects on interlaminar fracture toughness of FDM printed coupons, Mech. Compos. Multi-functional Mater. 7 (2017) 63e71. [20] X. Tian, T. Liu, C. Yang, Q. Wang, D. Li, Interface and performance of 3D printed continuous carbon fiber reinforced PLA composites, Compos. Part A Appl. Sci. Manuf. 88 (2016) 198e205. [21] D.-A. Türk, F. Brenni, M. Zogg, M. Meboldt, Mechanical characterization of 3D printed polymers for fiber reinforced polymers processing, Mater. Des. 118 (2017) 256e265. [22] R.J. Zaldivar, D.B. Witkin, T. McLouth, D.N. Patel, K. Schmitt, J.P. Nokes, Influence of processing and orientation print effects on the mechanical and thermal behavior of 3D-Printed ULTEM® 9085 material, Addit. Manuf. 13 (2017) 71e80.
[23] B. Huang, S. Singamneni, Raster angle mechanics in fused deposition modelling, J. Compos. Mater. 49 (3) (2015) 363e383. [24] T.A. Campbell, O.S. Ivanova, 3D printing of multifunctional nanocomposites, Nano Today 8 (2) (2013) 119e120. [25] B.G. Compton, J.A. Lewis, 3D-printing of lightweight cellular composites, Adv. Mater. 26 (34) (2014) 5930e5935. [26] S.R. Shin, R. Farzad, A. Tamayol, V. Manoharan, P. Mostafalu, Y.S. Zhang, M. Akbari, S.M. Jung, D. Kim, M. Comotto, N. Annabi, F.E. Al-Hazmi, M.R. Dokmeci, A. Khademhosseini, A bioactive carbon nanotube-based ink for printing 2D and 3D flexible electronics, Adv. Mater. 28 (17) (2016) 3280e3289. [27] K. Chizari, M. Arjmand, Z. Liu, U. Sundararaj, D. Therriault, Three-dimensional printing of highly conductive polymer nanocomposites for EMI shielding applications, Mater. Today Commun. 11 (2017) 112e118. [28] Y. Yang, Z. Chen, X. Song, Z. Zhang, J. Zhang, K.K. Shung, Q. Zhou, Y. Chen, Biomimetic anisotropic reinforcement architectures by electrically assisted nanocomposite 3D printing, Adv. Mater. 29 (11) (2017). [29] H.L. Tekinalp, V. Kunc, G.M. Velez-Garcia, C.E. Duty, L.J. Love, A.K. Naskar, C.A. Blue, S. Ozcan, Highly oriented carbon fiberepolymer composites via additive manufacturing, Compos. Sci. Technol. 105 (2014) 144e150. [30] F. Ning, W. Cong, J. Qiu, J. Wei, S. Wang, Additive manufacturing of carbon fiber reinforced thermoplastic composites using fused deposition modeling, Compos. Part B Eng. 80 (2015) 369e378. [31] R. Matsuzaki, M. Ueda, M. Namiki, T.K. Jeong, H. Asahara, K. Horiguchi, T. Nakamura, A. Todoroki, Y. Hirano, Three-dimensional printing of continuous-fiber composites by in-nozzle impregnation, Sci. Rep. 6 (2016) 23058. [32] C. Kousiatza, D. Karalekas, In-situ monitoring of strain and temperature distributions during fused deposition modeling process, Mater. Des. 97 (2016) 400e406. [33] C. Kousiatza, N. Chatzidai, D. Karalekas, Temperature mapping of 3D printed polymer plates: experimental and numerical study, Sensors 17 (3) (2017). [34] P. Mercelis, J.P. Kruth, Residual stresses in selective laser sintering and selective laser melting, Rapid Prototyp. J. 12 (5) (2006) 254e265. [35] K. Chou, Y. Zhang, A parametric study of part distortions in fused deposition modelling using three-dimensional finite element analysis, Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 222 (8) (2008) 959e968. [36] T. Mukherjee, V. Manvatkar, A. De, T. DebRoy, Mitigation of thermal distortion during additive manufacturing, Scr. Mater. 127 (2017) 79e83. [37] W.L. Wang, C.M. Cheah, J.Y.H. Fuh, L. Lu, Influence of process parameters on stereolithography part shrinkage, Mater. Des. 17 (4) (1996) 205e213. [38] M. Kaveh, M. Badrossamay, E. Foroozmehr, A. Hemasian Etefagh, Optimization of the printing parameters affecting dimensional accuracy and internal cavity for HIPS material used in fused deposition modeling processes, J. Mater. Process. Technol. 226 (2015) 280e286. [39] D. Karalekas, D. Rapti, Investigation of the processing dependence of SL solidification residual stresses, Rapid Prototyp. J. 8 (4) (2002) 243e247. [40] D. Karalekas, A. Aggelopoulos, Study of shrinkage strains in a stereolithography cured acrylic photopolymer resin, J. Mater. Process. Technol. 136 (1e3) (2003) 146e150. [41] A. Kantaros, D. Karalekas, Fiber Bragg grating based investigation of residual strains in ABS parts fabricated by fused deposition modeling process, Mater. Des. 50 (2013) 44e50. [42] C. Casavola, A. Cazzato, V. Moramarco, G. Pappalettera, Residual stress measurement in fused deposition modelling parts, Polym. Test. 58 (2017) 249e255. [43] Q. Sun, G.M. Rizvi, C.T. Bellehumeur, P. Gu, Effect of processing conditions on the bonding quality of FDM polymer filaments, Rapid Prototyp. J. 14 (2) (2008) 72e80.
Contents lists available at ScienceDirect
Composites Science and Technology journal homepage: http://www.elsevier.com/locate/compscitech
Characterization of residual stress and deformation in additively manufactured ABS polymer and composite specimens Wei Zhang a, b, c, Amanda S. Wu d, Jessica Sun c, Zhenzhen Quan a, Bohong Gu a, Baozhong Sun a, **, Chase Cotton e, Dirk Heider c, Tsu-Wei Chou b, c, * a
College of Textiles, Donghua University, Shanghai 201620, PR China Department of Mechanical Engineering, University of Delaware, Newark, DE 19716, USA Center for Composite Materials, University of Delaware, Newark, DE 19716, USA d Materials Engineering Division, Lawrence Livermore National Laboratory, CA 94550, USA e Department of Electrical and Computer Engineering, University of Delaware, Newark, DE 19716, USA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 May 2017 Received in revised form 13 July 2017 Accepted 16 July 2017 Available online 17 July 2017
Residual stresses induced in the layer-by-layer fabrication process of additively manufactured parts have significant impact on their mechanical properties and dimensional accuracy. This work aims to characterize the residual stress and deformation in specimens based on unreinforced acrylonitrile-butadienestyrene (ABS), carbon nanotube reinforced ABS and short carbon fiber reinforced ABS. The shrinkage and displacement fields were obtained, respectively, by thermal treatment as well as Digital Image Correlation observation of specimens before and after sectioning. The microstructure and porosity of additively manufactured specimens were also examined using X-ray micro-computed tomography. Specimen shrinkage and porosity content were significantly influenced by the process parameters of raster angle and printing speed, as well as material types. Faster printing speed led to larger porosity and residual stress, as well as higher shrinkage after specimen thermal treatment. Raster angle had a greater influence on specimen shrinkage and porosity as comparing to printing speed. Composite printing wires based on carbon nanotube and short carbon fiber in ABS greatly reduced specimen shrinkage and deformation, while increased the porosity, especially for carbon fiber reinforced ABS specimens. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Residual stress Shrinkage Microstructure Processing parameters Additive manufacturing
1. Introduction Additive manufacturing is defined as a “process of joining materials to make parts from 3D model data, usually layer upon layer, as opposed to subtractive manufacturing and formative manufacturing methodologies” [1] by ISO/ASTM. It refers to a family of fabrication methodologies including fused deposition modeling (FDM) [2], stereo lithography (SLA) [3], selective laser sintering (SLS) and selective laser melting (SLM) [4], inkjet modeling (IJM) [5] and others [6]. In recent years, additive manufacturing has evolved rapidly and been widely used in various manufacturing fields such as aerospace [7], automobile [8], biomedical [9], building [10] and many others [11e13]. This
* Corresponding author. Department of Mechanical Engineering, University of Delaware, Newark, DE 19716, USA. ** Corresponding author. E-mail addresses: [email protected] (B. Sun), [email protected] (T.-W. Chou). http://dx.doi.org/10.1016/j.compscitech.2017.07.017 0266-3538/© 2017 Elsevier Ltd. All rights reserved.
tremendous success could be attributed mainly to its outstanding ability to directly manufacture complex parts without special tools, to greatly reduce material waste, and to significantly reduce the time and cost of manufacturing for novel products and smallquantity productions [14e18]. The effects of an array of processing parameters, such as raster angle, printing speed, layer thickness, etc. on the mechanical properties of additively manufactured parts have been well documented in the literature [19e23]. In order to improve the performance of additively manufactured parts and broaden their applications, researchers have explored various types of printing materials, such as polymers, nanocomposites, fiber reinforced composites, etc. [24e31]. Another important aspect relevant to the mechanical performance of additively manufactured parts is repeated heating and cooling [32,33] resulting from the layer-bylayer building process. Significant residual stresses [34] can thus exist in additively manufactured part. Apart from their detrimental influence on mechanical performance, residual stresses could give rise to part distortion and dimensional inaccuracy [35e38]. Some
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
efforts have been made by researchers in identifying residual stresses in additively manufactured parts. For example, Karalekas and Rapti [39]. used epoxy based photopolymer to study the processing dependence of SLA solidification residual stress using the hole-drilling strain gage method of stress relaxation. Karalekas and Aggelopoulos [40] investigated the shrinkage strains in a SLA cured acrylic photopolymer resin. Kantaros et al. [41] studied the residual strains in ABS parts fabricated by FDM using fiber Bragg grating method. Casavola et al. [42] measured the residual stress in FDM parts made of ABS employing the hole-drilling method combined with electronic speckle pattern interferometry. These research efforts have greatly facilitated the understanding of residual stresses in additively manufactured parts. However, more studies with respect to characterization techniques, printing materials and processing parameters still need to be conducted to minimize the detrimental effects of residual stress. This work aims to characterize the residual stress and deformation in additively manufactured specimens based on unreinforced ABS, carbon nanotube reinforced ABS and carbon fiber reinforced ABS. The shrinkages and displacement fields were obtained by thermal treatment as well as Digital Image Correlation combined with sectioning of specimens, respectively. The effects of two key processing parameters, namely, raster angle and printing speed, as well as reinforcement materials, including carbon nanotube and carbon fiber, on specimen properties have been characterized. The microstructure and porosity of additively manufactured specimens were also examined using X-ray microcomputed tomography.
103
Fig. 1. TGA plot of the as-received CNTABS wire under temperature range of the 25e750 C.
resolution of 0.01 mm. The shrinkage of an additively manufactured specimen in a specific direction was calculated as follows: Shrinkage (%) ¼ ((original dimension (mm.) - dimension after thermal treatment (mm.)) / original dimension (mm.)) 100%
2. Experiment 2.1. Specimen preparation In this work, fused deposition modeling was adopted to fabricate additively manufactured specimens. A total of 27 types of specimens have been fabricated using a QIDI Tech dual-nozzle 3D printer (Ruan Qidi Technology Co., Ltd., Ruan, Zhejiang, China). Three kinds of materials, including pure ABS wire, carbon nanotube reinforced ABS wire (CNTABS) and short carbon fiber reinforced ABS wire (CFABS) (3DTECH, USA) were used. The carbon nanotube content in the wire was measured to be around 8 wt% via thermogravimetric analysis (TGA) in the nitrogen gas, as shown in Fig. 1. The significant weight loss of the samples reflected the thermal decomposition of the ABS matrix. The short carbon fiber content was around 15 wt% [16]. The length distribution of short carbon fibers in the wire was also given in Ref. [16]. The carbon fiber lengths varied from 5 mm to 465 mm, with a number-based average of 71.5 mm. In the fabrication process, three raster angles of 0 , ±45 and 0 /90 as well as three printing speeds of 40 mm/s, 60 mm/s and 80 mm/s were selected to build the specimen in the horizontal plane (X-Y). The raster angle indicates the direction of wire printing with respect to the longitudinal X-axis of the specimens, as shown in Fig. 2. The other processing parameters are shown in Table 1. Two kinds of specimen sizes of around 75 20 3 mm and 40 15 3 mm were fabricated to study the shrinkage and displacement fields, respectively. There are 12e13 layers each specimen. 2.2. Shrinkage measurement In order to measure the shrinkage of additively manufactured specimens, thermal treatment was carried out with the heating temperature of 180 C and heating time of 1 h. The dimensions in length, width and height of specimens before and after thermal treatment were measured by a Vernier caliper with the reading
2.3. Surface displacement field measurement In an effort to characterize specimen residual stress induced by the additive manufacturing process, surface displacement fields were measured by Digital Image Correlation. Digital Image Correlation is a 3D, full-field, non-contact optical technique to measure contour, deformation, vibration and strain on almost any material. In this work, it was employed by digitally comparing images acquired before and after releasing the localized stress by sectioning the specimen. In the measurement, a specimen was painted using a flat white spray paint and a black speckle pattern was manually applied using a flat black spray paint, with a minimum drying time of 24 h at ambient pressure and room temperature in between coatings. After that, the specimen was imaged using a dual camera setup before and after sectioning via lubricated diamond saw. Fig. 3 shows the images of three specimens (ABS, CFABS and CNTABS) before and after painting and sectioning. The displacement fields were calculated using ARAMIS Professional Digital Image Correlation algorithm (GOM International AG) by comparing the acquired images before and after sectioning. The displacement field so obtained is indicative of the surface-level residual stress in the specimen. 2.4. X-ray micro-computed tomography The SkyScan 1172 X-ray micro-CT system (Bruker Corp., Billerica, Massachusetts, U.S.) was used to characterize the microstructure and void distribution of additively manufactured specimens. In order to save scanning time, specimens were cut to the size of 8 20 3 mm. X-ray source voltage and current were 40 KV and 250 mA, respectively. The scanning resolution was around 5.4 mm/ pixel. The image size was set as 2048 1024 pixel. A series of 2D images in X-Z plane were acquired to record the attenuation of X-
104
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 2. Illustration of raster angles for the additively manufactured specimens.
Table 1 Processing parameters. Parameter
Value
Nozzle diameter Nozzle temperature Print platform temperature Filament feed rate Layer thickness
0.4 mm 230 C 110 C 1.0 mm/s 0.24 mm
Fig. 4. Thermal deformation photos of additively manufactured ABS specimens after thermal treatment at printing speeds of 40, 60 and 80 mm/s (a) 0 , (b) ±45 and (c) 0 / 90 .
Fig. 3. Additively manufactured ABS (top), CFABS (middle) and CNTABS (bottom) specimens shown as-built (left) and after speckle pattern application and sectioning (right).
ray photons in scanning a specimen. A series of cross-section microstructure images were processed by the SkyScan reconstruction program NRecon from these 2D images. The other two cross-section images in X-Y and Y-Z plane were obtained by transformation using software of DataViewer. The porosity of each specimen was obtained by calculating the average of porosities of all slices using the software of CTAn.
3. Results and discussions 3.1. Thermal shrinkage analysis Thermal shrinkage images were obtained for ABS polymer and its composite specimens with three raster angles (0 , ±45 and 0 / 90 ), and three printing speeds (40 mm/s, 60 mm/s and 80 mm/s) after thermal treatment. Fig. 4 shows the comparison of ABS specimens with and without thermal treatment for different raster angles and printing speeds. It can be seen that the heat treated ABS specimens exhibit high shrinkage deformation along the longitudinal direction, especially for raster angle of 0 . This could be attributed to the release of residual stress after thermal treatment.
Furthermore, the shrinkage of specimens increases with increasing printing speed for all raster angles. However, the 0 specimens also exhibit expansion along the transverse direction. In order to clearly identify the thermal deformation of ABS specimens, the shrinkage induced specimen size changes in length, width and height after thermal treatment are shown in Table 2. Positive shrinkage means contraction, while negative value indicates expansion. Three measurements were made along each direction and the average percentage was calculated as the shrinkage. Here, A0, A[0/90] and A[±45] denote the raster angles of 0 , 0 /90 and ±45 , respectively. Also, S40, S60 and S80 denote the printing speeds of 40 mm/s, 60 mm/s and 80 mm/s, respectively. It can be seen that the 0 specimens contract in the length direction and expand in the width and height directions, due to the Poisson's effect in the printed wires. However, for raster angles of 0 /90 and ±45 , the ABS specimens shrink in both length and width directions, while expand in thickness direction. This is due to the orthotropic nature of these two types of specimens. It is also worth noting that the difference of shrinkages between the length and width directions is significant for 0 /90 specimens, and much less for ±45 specimens. This is because the different printing path length of the 0 and 90 layers in the 0 /90 specimens, leading to different temperature changes (as demonstrated in Fig. 9 of reference [43]), and hence, difference in the resulting residual stresses between the length and width directions. Fig. 5 shows the thermal deformation of additively manufactured CNTABS specimens. It is obvious that the heat treated CNTABS specimens contracted along the longitudinal direction for all
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
105
Table 2 Shrinkage induced specimen size changes. Specimen types
Shrinkage (%) Length
Width
Height
ABS-A0-S40 ABS-A0-S60 ABS-A0-S80 ABS-A[0/90]-S40 ABS-A[0/90]-S60 ABS-A[0/90]-S80 ABS-A[±45]-S40 ABS-A[±45]-S60 ABS-A[±45]-S80
28.8 38.74 47.3 10.31 16.54 21.64 6.4 10.77 13.51
13.41 21.99 32.76 8.62 9.15 11.33 8.37 11.23 13.78
23.65 35.64 45.38 20.82 27.83 34.7 10.6 21.41 21.7
CNTABS-A0-S40 CNTABS-A0-S60 CNTABS-A0-S80 CNTABS-A[0/90]-S40 CNTABS-A[0/90]-S60 CNTABS-A[0/90]-S80 CNTABS-A[±45]-S40 CNTABS-A[±45]-S60 CNTABS-A[±45]-S80
10.57 18.18 20.55 5.45 7.99 9.54 4.63 5.92 6.78
0.91 4.87 7.46 9.36 10.7 10.55 8.41 10.35 11.78
13.95 18.75 20.33 18.18 22.82 23.23 19.98 20.07 21.55
CFABS-A0-S40 CFABS-A0-S60 CFABS-A0-S80 CFABS-A[0/90]-S40 CFABS-A[0/90]-S60 CFABS-A[0/90]-S80 CFABS-A[±45]-S40 CFABS-A[±45]-S60 CFABS-A[±45]-S80
3.48 5.43 7.27 2.53 3.9 4.94 1.73 1.86 2.89
0.05 0.48 0.86 4.64 5.22 5.8 4.44 5.91 6.44
1.42 2.47 1.41 4.23 0.57 1.68 0.35 1.05 4.54
Fig. 5. Thermal deformation photos of additively manufactured CNTABS specimens after thermal treatment at printing speeds of 40, 60 and 80 mm/s (a) 0 , (b) ±45 and (c) 0 /90 .
Fig. 6. Thermal deformation photos of additively manufactured CFABS specimens after thermal treatment at printing speeds of 40, 60 and 80 mm/s (a) 0 , (b) ±45 and (c) 0 / 90 .
CNTABS specimens, shrinkage occurs in the thickness of CFABS specimens with raster angles of 0 and 0 /90 . The effects of raster angle and printing speed on the length shrinkages of ABS, CNTABS and CFABS specimens are compared in Fig. 7(a)-(c). It can be clearly seen that shrinkage increases with increasing printing speed for all three kinds of materials at the three raster angles. This implies that the residual stress increases with increasing printing speed. It also can be concluded that the raster angle has a greater influence on length shrinkage compared with printing speed. For all three kinds of materials and three printing speeds, the shrinkages of 0 specimens are larger than those of specimens with other raster angles, particularly for ABS and CNTABS specimens. Furthermore, length shrinkage is highly anisotropic, i.e. orientation dependent. This could be attributed to the effect of flow-induced molecular and reinforcement (CNT, CF) alignment, which greatly constrain the shrinkage of neighboring layers in the 0 /90 and ±45 laminated structures. The ±45 specimens show the minimal shrinkage. Fig. 7(d) presents the length shrinkages of ABS, CNTABS and CFABS specimens at the same printing speed of 60 mm/s. It can be seen that the shrinkages of CFABS specimens are much less than those of ABS specimens at all raster angles, due to the constraining effect of short carbon fibers on specimen contraction. The magnitude of shrinkage of CNTABS specimens lies in between those of ABS and CFABS specimens. Thus, carbon nanotubes are less effective in reducing the shrinkage comparing to short carbon fibers. The influence of volume contents of CNT and CF has not been examined. 3.2. Surface displacement field induced by residual stress
printing speeds and raster angles. Shrinkage induced specimen size change are also shown in Table 2. Overall, the trends of shrinkage deformation of CNTABS specimens are similar to those of pure ABS specimens, and the deformation increases with increasing printing speed. Again, just as ABS specimens, bending also appears in CNTABS specimens. Fig. 6 shows the thermal deformation of additively manufactured CFABS specimens. As indicated in Table 2, the overall dimensional changes in this case are much less than those of ABS and CNTABS specimens, indicating the dominating effect of short carbon fibers in resisting specimen shrinkage. Also, unlike ABS and
Digital Image Correlation measurements in conjunction with specimen sectioning have been employed to identify the surface displacement fields of additively manufactured ABS, CNTABS and CFABS specimens upon releasing the residual stress. Fig. 8 shows the displacement fields on an ABS specimen without heat treatment, which was sectioned into two parts A and B. Referring to the coordinate axes of Fig. 2, the red and blue colors signify positive and negative displacements, respectively, while green color indicates no displacement. In Fig. 8(a), X displacements occur along the edges opposite to the cut surface, and in Fig. 8(b), Y displacements indicate opposite bending deflections in Part A and B of the ABS
106
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 7. Comparison of effects of raster angle and printing speed on length shrinkages for (a) ABS, (b) CNTABS and (c) CFABS, as well as the effects of materials types on length shrinkages (d).
Fig. 8. Surface displacement distributions on a sectioned ABS specimen without thermal treatment: (a) X-displacement, (b) Y-displacement.
specimen. However, there is essentially no displacement in the thermally treated specimen (Fig. 9). Thus, it can be concluded that fabrication induced residual stress were released by thermal treatment, and the magnitude of deformation in specimens after heat treatment reflects the level of residual stresses. In an effort to compare the deformations among ABS, CNTABS,
and CFABS specimens, the Y displacement distributions on the surfaces of both Part A and B are shown in Fig. 10. In Part A of the specimens, the Y displacements are negative on both ends, while positive in the middle. There is an obvious upward bending deflection. For Part B of the specimens, the Y displacements are positive on both ends, while negative in the middle of the
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 9. Heat treated ABS specimen shows nearly no surface displacement.
specimens, indicating a downward bending deflection. The relative sizes of the positive and negative Y-displacement regions in either Part A or B specimen are the indication of the severity of bending deformation. The bending deflection is the highest in ABS specimen and the lowest in CFABS specimen. The presence of carbon nanotube and carbon fiber in the composite specimens of CNTABS and CFABS improve their flexural modulus and impede the bending deformation. Fig. 11 gives the profiles of Y-displacement at various points on the line along X-direction near the cut surface of ABS, CNTABS and CFABS specimens. In both Part A and B, the largest Y-displacement in ABS gives rise to the highest specimen bending curvature while CFABS shows the smallest Y-displacement and lowest specimen bending curvature. Lastly, it should be mentioned that the opposite direction deflection of Part A and B in a specimen result in expansion in its width as shown in Table 2. 3.3. Characterization of microstructure and porosity The microstructural and porosity characterization of additively manufactured specimens were carried out via X-ray microcomputed tomography. For the effects of printing speed, Fig. 12(a)-(c) show Y-Z plane cross-sections of 0 ABS specimens at three printing speeds of 40, 60 and 80 mm/s. It can be seen that
107
void size increases with increasing printing speed. The porosity of each specimen is the average of porosity values of all sections. The porosities of specimens with printing speeds of 40, 60 and 80 mm/s are, respectively, 0.51%, 0.82% and 3.23%. This finding can be attributed to the different cooling rates of printed wires. Faster printing speed allows less time for the development of interwire adhesion, and, hence, larger pores as well as greater shrinkage when specimens are heat treated. Fig. 12(a’)-(c’) show the X-Y plane cross-sections. Here, the void size and distribution can be seen clearly and the result is consistent with those in Fig. 12(a)-(c). Also, it is worth noting that fast printing speed may lead to the unevenness of the printed wires as shown in Fig. 12(c’). The effects of raster angle on specimen microstructure are demonstrated in Fig. 13(a)-(c). Here, the Y-Z plane cross-sections of ABS specimens are shown for the raster angles of 0 , ±45 and 0 / 90 at printing speed of 60 mm/s. It can be seen that many small voids are evenly distributed in the cross-section of 0 specimen. For raster angle of 0 /90 , numerous large voids exist between the mutually perpendicular printed wires. In comparing to the cases of 0 and 0 /90 , the ±45 specimen shows much less porosity. The porosities of 0 , ±45 and 0 /90 specimens of Fig. 13 are 0.82%, 0.48% and 1.73%, respectively. Fig. 13(a’)-(c’) show the X-Y plane cross-sections for the raster angles of 0 , ±45 and 0 /90 . The pore size and distribution between the printed wires for the X-Y crosssection are consistent with those for the Y-Z cross-section in Fig. 13(a)-(c). Lastly, a comparison among ABS, CNTABS and CFABS specimens is presented. Fig. 14 shows the difference of cross-section microstructure between them. It can be seen that specimen void size increase in the order of ABS, CNTABS and CFABS for both Y-Z and XY cross-sections. Again, these findings confirm that the addition of carbon nanotubes and short carbon fibers constrain the flow of printed wires in the fabrication process, and gives rise to high porosities of 0.48%, 2.39% and 8.54%, respectively, for ABS, CNTABS and CFABS specimens.
Fig. 10. Y-displacement fields of additively manufactured (a) ABS, (b) CNTABS and (c) CFABS specimens.
108
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
Fig. 11. Profiles of Y-displacement at various points on the line along the X-direction near the cut surface of ABS, CNTABS and CFABS specimens (a) Part A and (b) Part B.
Fig. 12. YeZ plane cross-section microstructures of 0 ABS specimens at printing speeds of (a) 40 mm/s, (b) 60 mm/s and (c) 80 mm/s (a’), (b’) and (c’) are the corresponding X-Y plane cross-section microstructures.
4. Conclusions Major findings of this study are summarized as follows. (1) Faster printing speed leads to larger porosity and residual stress, as well as higher shrinkage after specimen thermal treatment. (2) Specimen shrinkage and porosity content are significantly influenced by the process parameters, raster angle and printing speed, as well as material type. (3) Raster angle has a greater influence on specimen shrinkage and porosity as comparing to printing speed. ABS specimen with raster angle of ±45 show less shrinkage and porosity as comparing to those with raster angle of 0 and 0 /90 .
(4) Composite printing wires based on carbon nanotube and short carbon fiber in ABS greatly reduce specimen shrinkage and deformation especially for CFABS specimens. Void size increases in the order of ABS, CNTABS and CFABS specimens. (5) Digital Image Correlation provides an effective means for measuring specimen surface displacement after sectioning. It provides a unique insight into fabrication induced residual stress. (6) X-ray micro-computed tomography, a nondestructive measurement technique, clearly demonstrates internal microstructure and void size of additively manufactured specimens.
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
109
Fig. 13. YeZ plane cross-section microstructures of ABS specimens at printing speed of 60 mm/s (a) 0 , (b) ±45 and (c) 0 /90 . (a’), (b’) and (c’) are the corresponding X-Y plane cross-section microstructures.
Fig. 14. YeZ plane cross-section microstructures of (a) ABS, (b) CNTABS and (c) CFABS specimens with raster angle of ±45 at printing speed of 60 mm/s (a’), (b’) and (c’) are the corresponding X-Y plane cross-section microstructure.
Acknowledgements Wei Zhang acknowledges the financial support from the China
Scholarship Council (CSC) and the Fundamental Research Funds for the Central Universities of China.
110
W. Zhang et al. / Composites Science and Technology 150 (2017) 102e110
References [1] ISO/ASTM 52900:2015(E), Standard Terminology for Additive ManufacturingGeneral Principles-Terminology, 2015. [2] S.S. Crump, Apparatus and Method for Creating Three-dimensional Objects, Stratasys Incorporated, US, 1992. [3] P.F. Jacobs, Rapid Prototyping & Manufacturing: Fundamentals of Stereolithography, Society of Manufacturing Engineers, 1992. [4] J.-P. Kruth, B. Vandenbroucke, v.J. Vaerenbergh, P. Mercelis, Benchmarking of Different SLS/SLM Processes as Rapid Manufacturing Techniques, 2005. [5] M. Singh, H.M. Haverinen, P. Dhagat, G.E. Jabbour, Inkjet printing-process and its applications, Adv. Mater. 22 (6) (2010) 673e685. [6] J. Gardan, Additive manufacturing technologies: state of the art and trends, Int. J. Prod. Res. 54 (10) (2015) 3118e3132. [7] L.J. Kumar, C.G. Krishnadas Nair, Current Trends of Additive Manufacturing in the Aerospace Industry, 2017, pp. 39e54. [8] S. Curran, P. Chambon, R. Lind, L. Love, R. Wagner, S. Whitted, D. Smith, B. Post, R. Graves, C. Blue, Big area additive manufacturing and hardware-in-the-loop for rapid vehicle powertrain prototyping: A case study on the development of a 3-D-printed Shelby Cobra, SAE Technical Paper, 2016. [9] A.A. Zadpoor, J. Malda, Additive manufacturing of biomaterials, tissues, and organs, Ann. Biomed. Eng. 45 (1) (2017) 1e11. [10] K. Biswas, J. Rose, L. Eikevik, M. Guerguis, P. Enquist, B. Lee, L. Love, J. Green, R. Jackson, Additive manufacturing integrated energydenabling innovative solutions for buildings of the future, J. Sol. Energy Eng. 139 (1) (2016) 015001. [11] A. Ambrosi, M. Pumera, 3D-printing technologies for electrochemical applications, Chem. Soc. Rev. 45 (10) (2016) 2740e2755. [12] N. Guo, M.C. Leu, Additive manufacturing: technology, applications and research needs, Front. Mech. Eng. 8 (3) (2013) 215e243. [13] M.G. Mohammed, R. Kramer, All-printed Flexible and Stretchable Electronics, Advanced Materials, 2017. [14] A.R. Studart, Additive manufacturing of biologically-inspired materials, Chem. Soc. Rev. 45 (2) (2016) 359e376. [15] Z. Quan, A. Wu, M. Keefe, X. Qin, J. Yu, J. Suhr, J.-H. Byun, B.-S. Kim, T.-W. Chou, Additive manufacturing of multi-directional preforms for composites: opportunities and challenges, Mater. Today 18 (9) (2015) 503e512. [16] Z. Quan, Z. Larimore, A. Wu, J. Yu, X. Qin, M. Mirotznik, J. Suhr, J.-H. Byun, Y. Oh, T.-W. Chou, Microstructural design and additive manufacturing and characterization of 3D orthogonal short carbon fiber/acrylonitrile-butadienestyrene preform and composite, Compos. Sci. Technol. 126 (2016) 139e148. [17] X. Wang, M. Jiang, Z. Zhou, J. Gou, D. Hui, 3D printing of polymer matrix composites: a review and prospective, Compos. Part B Eng. 110 (2017) 442e458. [18] Z. Quan, Z. Larimore, X. Qin, J. Yu, M. Mirotznik, J.-H. Byun, Y. Oh, T.-W. Chou, Microstructural characterization of additively manufactured multi-directional preforms and composites via X-ray micro-computed tomography, Compos. Sci. Technol. 131 (2016) 48e60. [19] G.P. Tandon, T.J. Whitney, R. Gerzeski, H. Koerner, J. Baur, Process parameter effects on interlaminar fracture toughness of FDM printed coupons, Mech. Compos. Multi-functional Mater. 7 (2017) 63e71. [20] X. Tian, T. Liu, C. Yang, Q. Wang, D. Li, Interface and performance of 3D printed continuous carbon fiber reinforced PLA composites, Compos. Part A Appl. Sci. Manuf. 88 (2016) 198e205. [21] D.-A. Türk, F. Brenni, M. Zogg, M. Meboldt, Mechanical characterization of 3D printed polymers for fiber reinforced polymers processing, Mater. Des. 118 (2017) 256e265. [22] R.J. Zaldivar, D.B. Witkin, T. McLouth, D.N. Patel, K. Schmitt, J.P. Nokes, Influence of processing and orientation print effects on the mechanical and thermal behavior of 3D-Printed ULTEM® 9085 material, Addit. Manuf. 13 (2017) 71e80.
[23] B. Huang, S. Singamneni, Raster angle mechanics in fused deposition modelling, J. Compos. Mater. 49 (3) (2015) 363e383. [24] T.A. Campbell, O.S. Ivanova, 3D printing of multifunctional nanocomposites, Nano Today 8 (2) (2013) 119e120. [25] B.G. Compton, J.A. Lewis, 3D-printing of lightweight cellular composites, Adv. Mater. 26 (34) (2014) 5930e5935. [26] S.R. Shin, R. Farzad, A. Tamayol, V. Manoharan, P. Mostafalu, Y.S. Zhang, M. Akbari, S.M. Jung, D. Kim, M. Comotto, N. Annabi, F.E. Al-Hazmi, M.R. Dokmeci, A. Khademhosseini, A bioactive carbon nanotube-based ink for printing 2D and 3D flexible electronics, Adv. Mater. 28 (17) (2016) 3280e3289. [27] K. Chizari, M. Arjmand, Z. Liu, U. Sundararaj, D. Therriault, Three-dimensional printing of highly conductive polymer nanocomposites for EMI shielding applications, Mater. Today Commun. 11 (2017) 112e118. [28] Y. Yang, Z. Chen, X. Song, Z. Zhang, J. Zhang, K.K. Shung, Q. Zhou, Y. Chen, Biomimetic anisotropic reinforcement architectures by electrically assisted nanocomposite 3D printing, Adv. Mater. 29 (11) (2017). [29] H.L. Tekinalp, V. Kunc, G.M. Velez-Garcia, C.E. Duty, L.J. Love, A.K. Naskar, C.A. Blue, S. Ozcan, Highly oriented carbon fiberepolymer composites via additive manufacturing, Compos. Sci. Technol. 105 (2014) 144e150. [30] F. Ning, W. Cong, J. Qiu, J. Wei, S. Wang, Additive manufacturing of carbon fiber reinforced thermoplastic composites using fused deposition modeling, Compos. Part B Eng. 80 (2015) 369e378. [31] R. Matsuzaki, M. Ueda, M. Namiki, T.K. Jeong, H. Asahara, K. Horiguchi, T. Nakamura, A. Todoroki, Y. Hirano, Three-dimensional printing of continuous-fiber composites by in-nozzle impregnation, Sci. Rep. 6 (2016) 23058. [32] C. Kousiatza, D. Karalekas, In-situ monitoring of strain and temperature distributions during fused deposition modeling process, Mater. Des. 97 (2016) 400e406. [33] C. Kousiatza, N. Chatzidai, D. Karalekas, Temperature mapping of 3D printed polymer plates: experimental and numerical study, Sensors 17 (3) (2017). [34] P. Mercelis, J.P. Kruth, Residual stresses in selective laser sintering and selective laser melting, Rapid Prototyp. J. 12 (5) (2006) 254e265. [35] K. Chou, Y. Zhang, A parametric study of part distortions in fused deposition modelling using three-dimensional finite element analysis, Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 222 (8) (2008) 959e968. [36] T. Mukherjee, V. Manvatkar, A. De, T. DebRoy, Mitigation of thermal distortion during additive manufacturing, Scr. Mater. 127 (2017) 79e83. [37] W.L. Wang, C.M. Cheah, J.Y.H. Fuh, L. Lu, Influence of process parameters on stereolithography part shrinkage, Mater. Des. 17 (4) (1996) 205e213. [38] M. Kaveh, M. Badrossamay, E. Foroozmehr, A. Hemasian Etefagh, Optimization of the printing parameters affecting dimensional accuracy and internal cavity for HIPS material used in fused deposition modeling processes, J. Mater. Process. Technol. 226 (2015) 280e286. [39] D. Karalekas, D. Rapti, Investigation of the processing dependence of SL solidification residual stresses, Rapid Prototyp. J. 8 (4) (2002) 243e247. [40] D. Karalekas, A. Aggelopoulos, Study of shrinkage strains in a stereolithography cured acrylic photopolymer resin, J. Mater. Process. Technol. 136 (1e3) (2003) 146e150. [41] A. Kantaros, D. Karalekas, Fiber Bragg grating based investigation of residual strains in ABS parts fabricated by fused deposition modeling process, Mater. Des. 50 (2013) 44e50. [42] C. Casavola, A. Cazzato, V. Moramarco, G. Pappalettera, Residual stress measurement in fused deposition modelling parts, Polym. Test. 58 (2017) 249e255. [43] Q. Sun, G.M. Rizvi, C.T. Bellehumeur, P. Gu, Effect of processing conditions on the bonding quality of FDM polymer filaments, Rapid Prototyp. J. 14 (2) (2008) 72e80.