Progress in Crystal Growth and Characterization of Materials 60 (2014) 94e98
Contents lists available at ScienceDirect
Progress in Crystal Growth and Characterization of Materials journal homepage: www.elsevier.com/locate/pcrysgrow
Review
Manufacturing and characterization of porous titanium components Florencia Edith Wiria*, Saeed Maleksaeedi, Zeming He Joining Technology Group, Singapore Institute of Manufacturing Technology, 71 Nanyang Drive, Singapore 638075, Singapore
a b s t r a c t Keywords: 3D printing Titanium Porous Additive manufacturing Powder
A powder-bed 3D printer (3DP) is investigated to fabricate porous titanium components. The titanium material was 3D printed and subsequently post-processed by thermal debinding and sintering. Characterization work was carried out to investigate the effects of sintering temperature on the internal porosity profile and shrinkage of 3D printed titanium components, the effects of different binder content on the overall shape of the pre-designed porous components and the effects of post-processing debinding profiles on the titanium components. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction For implant with load bearing function, it is ideal that the load is shared between the surrounding functional body part and the implant. The stress carried by the implant and the functional body part is directly related to their stiffness [1]. Implants with higher stiffness with that of the surrounding body part may cause issues as the higher stiffness of the implant would take the majority of the loaded stress and thus preventing the functional body part from being loaded properly [2]. For bone implant, this phenomenon is called stress shielding, and could lead to adverse effect of implant loosening in the future [3]. To reduce the high stiffness of metallic implant, it is suggested that implantable devices to be fabricated as porous components. Titanium is the material of choice due to its already favorable
* Corresponding author. Tel.: þ65 6793 8274. E-mail address: fl
[email protected] (F.E. Wiria).
http://dx.doi.org/10.1016/j.pcrysgrow.2014.09.001 0960-8974/© 2014 Elsevier Ltd. All rights reserved.
F.E. Wiria et al. / Progress in Crystal Growth and Characterization of Materials 60 (2014) 94e98
95
mechanical and biocompatible properties [4]. As a result, the porous device's mechanical properties could be tailored to better match the yield strength and elastic modulus of the host bone and therefore avoiding stress-shielding effect associated with a mismatch in bone-implant elastic modulus. Additive manufacturing (AM) technologies, a layer-by-layer manufacturing process which builds 3D objects directly from the objects' computer data models, can aid the fabrication of porous components. Fabrication of porous objects can be done by designing porosity into the objects' computer models. A particular interest is the powder-bed 3-dimensional printer (3DP) to manufacture porous titanium components through an indirect process. 3DP is a relatively low cost AM system, which totals less than US$ 40,000 when combined with a sintering tube furnace [5,6]. This sum is much lower in comparison to direct manufacturing of titanium components using metal-based AM systems, in which the cost starts from US$ 600,000 for powder-bed systems such as selective laser melting [7]. A previous work has shown that the mechanical properties of porous 3D printed titanium can be tailored to be in the range of natural bone [8,9]. This work focuses on the physical characterization of the porous titanium component with respect of porosity and shrinkages profiles, as well as factors affecting the stability of the porous component shape. 2. Methodology CP titanium powder (Grade 2, ASTM F67, size 45 mm, spherical shape, TLS Technik Spezialpulver) was dry mixed with poly(vinyl alcohol) (PVA) (Nippon Gohsei, Gohsenol, NH-18S) in a ball mill mixer (US Stoneware, 764 AVM Jar Mill) at 100 rpm for 10 h. PVA, as the binder, was mixed with concentration of 5, 10, 15 and 30%. Zirconia balls (diameter 20 mm) were used to facilitate the mixing process, with ball to powder ratio set to be 10:1 by weight. CAD files of the components can be created using any drawing software and subsequently converted to .STL file to feed into a 3DP system (3D Systems, ZPrinter 310 Plus). The layer thickness used is 0.1 mm. After the printing process completes, the parts were left overnight in the powder-bed to allow the liquid binder to bond to the powder mixture and produce sufficient strength for handling. The parts were subsequently taken out from the building bed and dried in an oven (Townson Mercer) at 50 C for approximately 1 h to remove the moisture and further strengthen the parts. After depowdering, the parts underwent debinding-cum-sintering process (CM Tube High Temperature Furnace) in Argon environment at the rate of 20 L/min. Sintering temperature was varied at 900, 1000, 1100, 1200, 1250, 1300 and 1350 C. Two profiles with a variety in the debinding temperatures (Fig. 1) were used to investigate the effects of thermal post-processing profiles on the titanium components. For determination of dimensional accuracy and shrinkage a digital calliper with 0.01 mm accuracy. Pore size of the titanium samples were measured by mercury porosimeter (AutoPore IVjMicromeritics). A digital microscopy was used to measure the size of macroscopic pores. Scanning electron microscopy (JEOL) was utilized to examine the pore size.
Fig. 1. Thermal post-processing profile with varying debinding temperatures.
96
F.E. Wiria et al. / Progress in Crystal Growth and Characterization of Materials 60 (2014) 94e98
3. Effect of sintering temperature on shrinkage and internal porosity profiles The porous titanium components used in these investigations were fabricated using powder mixture with 30 wt% of PVA. The samples (n ¼ 8) were drawn resembling a cylinder. The shrinkage analysis (Fig. 2) shows that with a sintering temperature of 1100 C or below, the axial (along sample's length) and radial (along sample's diameter) showed no significant variation (p < 0.05) at approximately 17 and 21% reduction, respectively, from the CAD model dimensions. Above the sintering temperature of 1100 C, axial shrinkage increased significantly to approximately 33% while radial shrinkage also increased significantly to the region of 28% from the original dimensions. The samples used in this investigation were sintered with the long axis placed horizontally. This resulted in radial shrinkage being generally higher than the axial shrinkage, which implied that the difference in the radial and axial shrinkage was due to gravity acting on the samples during the sintering process. The increased shrinkage seen at higher sintering temperature is perhaps expected, as there is reduced surface energy and gravity effects becoming more dominant due to the particles undergoing sintering becoming increasingly free to move. The morphology data revealed that the morphology changes between the center, periphery and surface of the samples. The internal porosity ranges from 34 to 40% for all the range of sintering temperatures, with internal porosity in central areas to be slightly higher than the peripheral areas. However, porosity at the surface was found to be in the range of 55e65%, which was higher than that on the central or peripheral areas. Apart from the porosity caused by the removal of binders, it could presumably be due to powder loosening from the surface. The average pore size could be from a combination of the sizes of two types of micro-pore, namely the smaller micro-pores, which are the remnants of the packing gaps between the unsintered titanium particles, and larger macro-pores, which are a result of the pores left by PVA binder. The average surface pore size was found to be relatively larger than the average internal (central and peripheral) pore size. This finding was in agreement with the porosity profile data. 4. Effect of binder content on overall shape of pre-designed porous components Rectangular blocks (16 16 7 mm) were designed with small square channels (2 2 mm) cutting through the sides with 1 mm interval (Fig. 3). Printed parts were sintered at 1300 C for 2 h. This investigation showed that the suitable amount of binder is 5 wt% from the point of distortion and 10 wt % from the point of handling strength to ensure excellent handling capability. According to the feature size of the component and handling strength required, suitable binder amount was used. Generally, a binder amount of 5e10 wt% was sufficient to ensure good particle packing and sintering for components with pre-designed channels.
Fig. 2. Part shrinkage.
F.E. Wiria et al. / Progress in Crystal Growth and Characterization of Materials 60 (2014) 94e98
97
Fig. 3. Titanium cages printed with different PVA content.
5. Effect of debinding profiles on shrinkage Rectangular blocks (n ¼ 12) with dimension 18 cm 18 cm 6 mm were used to investigate the shrinkage profiles during the debinding and sintering stages of the thermal post-processing. Higher shrinkage was observed in debinding stage for samples in profile A as compared to profile B. Samples in profile A underwent debinding at higher temperatures than actual PVA decomposition temperatures, which had peak degradation at 250 C according to thermal gravimetric analysis. Hence the samples had undergone rapid PVA decomposition and transited into sintering stage earlier, with the titanium atoms collapsing into the void spaces and caused the higher shrinkage values in the debinding stage. 6. Conclusion There are many factors that contribute to the reproducibility and stability of porous 3D printed titanium components. In this paper, several factors were investigated, namely the effects of sintering temperature, debinding profiles and binder content of the powder mixture on the overall shape, shrinkage and internal porosity profiles. Increase in sintering temperature caused increased shrinkage. Generally, the shrinkage was higher in the Z-direction, with respect to the placement of the part during sintering process. This is because gravity may have caused the titanium particles to move downwards when the particles become more increasingly free to move. The porosity profile showed that the surface porosity was higher than that on the central or peripheral areas. Both the surface and internal porosity profiles showed that the smaller particle size ranges gave a larger peak as compared to the larger particle size ranges. The smaller size ranges was likely to be attributed to the sintering micro-pores while the larger size ranges was likely to be attributed to the porogen micro-pores. Binder amount of 5e10 wt% was relatively sufficient to ensure good particle packing, handling stability and sintering for components with pre-designed channels. References [1] S. Ramakrishna, J. Mayer, E. Wintermantel, K.W. Leong, Biomedical applications of polymer-composite materials: a review, Compos. Sci. Technol. 61 (2001) 1189e1224. [2] L.L. Hench, Bioceramics: from concept to clinic, J. Am. Ceram. Soc. 74 (1991) 1487e1510. [3] L.L. Hench, Bioceramics, J. Am. Ceram. Soc. 81 (1998) 1705e1728. [4] G. Lütjering, J.C. Williams, Titanium, Springer-Verlag, Berlin Heidelberg, 2003. [5] PR Newswire, Z Corp. introduces two value-priced 3D printers starting at just $14,900, in: http://www.prnewswire.com/ news-releases/z-corp-introduces-two-value-priced-3d-printers-starting-at-just-14900-99011014.html. [6] MTI Corporation 1600 C and atmosphere tube furnace (80 mm OD), in: http://www.mtixtl.com/1600vacuumtubefurnacegsl-1600x-80.aspx. [7] T. Wohlers, Wohlers Report 2011: Additive Manufacturing and 3D Printing State of the Industry e Annual Worldwide Progress Report, Wohlers Associates: Colorado (2011). [8] F.E. Wiria, J.M.S. Yong, P.N. Lim, F.C.W. Goh, J.F. Yeo, T. Cao, Printing of titanium implant prototype, Mater. Des. 31 (2010) S101eS105. [9] S. Maleksaeedi, J.K. Wang, A. El-Hajje, L. Harb, V. Guneta, Z. He, F.E. Wiria, C. Choong, A.J. Ruys, Toward 3D printed bioactive titanium scaffolds with bimodal pore size distribution for bone ingrowth, Procedia CIRP 5 (2013) 158e163.
98
F.E. Wiria et al. / Progress in Crystal Growth and Characterization of Materials 60 (2014) 94e98
Dr. Florencia Edith Wiriais a Research Scientist at Singapore Institute of Manufacturing Technology of Agency for Science, Technology and Research. She obtained her Ph.D.from the School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore. Her entire research career revolves around material processing and development for 3D additive manufacturing systems (such as powder-bed inkjet 3D printing, stereolithography, fused deposition modeling and selective laser melting) and fabrication of porous materials for biomedical applications. Her curiosity has led her to work on research projects involving polymers and ceramics material, as well as titanium for low modulus applications. She also enjoys interaction with her students as a relaxing way to impart her knowledge and experience.
Dr. Saeed Maleksaeediobtained his Ph.D.in 2009 from School of Materials Science and Engineering in Shiraz University in Iran. He has been working in Singapore Institute of Manufacturing Technology as a Research Scientist since 2010. His research interests lie in the field of powder processing of materials for structural, biomedical and energy applications. He has been engaged in 3D additive manufacturing R&D for the last several years for fabrication of ceramic and metallic components especially for biomedical and dental applications. The processes and materials he has focused on are inkjet 3D printing of titanium, stereolithography of ceramic and metallic components, indirect 3D additive manufacturing using sacrificial casting and selective laser melting of titanium alloys. He has published several papers in prestigious journals and filed a few patents on manufacturing processes.
Dr. Zeming Heis a Research Scientist at Singapore Institute of Manufacturing Technology of Agency for Science, Technology and Research. His research interest is in materials processing using 3D additive manufacturing techniques, such as inkjet 3D printing and electron beam melting. Dr. He obtained his Ph.D.degree in 2001 from School of Materials Science and Engineering, Nanyang Technological University in Singapore. He has worked at different institutions in China, Singapore, Germany,The Netherlands and Denmark on developing structural and functional materials for defensive, biomedical and energy applications. He has accumulated vast experience in ceramic materials development using powder processing technologies, published nearly 70 scientific and technical papers and served as referees for 7 prestigious journals, such as Apply Physics Letters. In 2005, Dr. He was awarded “European Commission Marie Curie Fellowship” for developing nano-structured thermoelectric materials with high figure of merit for energy conversion.