Article Preview
Top1. Introduction
Additive manufacturing (AM), capable of fabricating physical parts directly from a CAD model, offers advantage of building highly complex parts in short span of time which are otherwise very difficult to produce by conventional machining processes (Gibson et al., 2015; Modi, 2018). Being very suitable for mass customization, AM has tremendous potential to be used for medical and biomedical applications, such as fabrication of medical anatomical models, prosthetics, orthotics, customized implants, drug delivery doses, surgical devices & tools and tissue engineering scaffolds (Javaid and Haleem, 2018; Modi and Sanadhya, 2018). Bone tissue engineering (BTE) is an area where synthetic bone scaffolds are being fabricated through additive manufacturing. Stereolithography, fused deposition modeling, selective laser sintering and three-dimensional printing are the processes which have been most explored by the researchers (Leong et al., 2003; Hollister, 2005). Although, bone possesses self-regenerating capability; regeneration is limited merely to small fractures and not for critical ones. The critical sized bone defects need to be filled with a porous scaffold allowing in-growth of blood vessels and bone (Bohner et al., 2005). To match the complex structure and unique properties of human bone, synthetic bone scaffold must possess adequate pore size, interconnected porosity and mechanical strength in addition to biocompatibility and bio resorbability.
Powder bed binder jetting 3D Printing (PBBJ 3DP) process is being widely explored by the researchers nowadays to fabricate the porous bone scaffolds (Sachs et al., 1992; Warnke et al., 2010). Final quality of 3D printed porous scaffold is affected largely by raw material, binding agent, printing parameters and pre & post-processing treatments (Dimitrov et al., 2006). Optimizing process parameters of a 3D printer for given material is one practical solution to reduce the process variability and enhance performance of the printer (Hsu and Lai, 2010). Several engineering tools such as design of experiments (DOE), genetic algorithm, Taguchi method, response surface methodology, artificial neural network, particle swarm optimization, topology optimization, iterative gradient search, ant colony algorithm etc. have been employed by researchers for optimizing manufacturing processes (Bagchi, 1993; Zuo et al., 2006; Raja and Baskar, 2011; Yang et al., 2016; Montgomery, 2017; Ning et al., 2018).
Taguchi design of experiment approach have been explored by several researchers, such as Onuh and Hon (1998), investigated the build parameters of Stereolithography (SLA) for better surface finish. Lakshmi and Arumaikkannu (2014), assessed the influence of process parameters on surface finish in customized bone implant using selective laser sintering (SLS) process. Chhabra and Singh (2012), evaluated surface roughness of ZCast direct metal casting parts. Work on FDM process for enhancement of surface quality, dimensional accuracy, and customization of material properties have also been done by different researchers (Anitha et al., 2001; Sood et al., 2009; Srivastava and Rathee, 2018). However, very limited studies have been found on optimization of 3D printing parameters, especially for calcium sulfate based ceramic composite powder material. Yao and Tseng (2002), optimized the process parameters for ZCorp’s 402 3D printer and achieved a significant reduction in parts build time and powder & glue quantity by 20% for Zp100 and 10% for Zp11 powder. Hsu and Lai (2010) employed Taguchi method and ANOVA analysis to determine the optimal parameter values for improving dimensional accuracy and flexural strength. Vaezi and Chua (2011), studied 3D printing of plaster-based powder Zp102 and found that uniform layer thickness with higher binder saturation results in increased mechanical strength but only at the cost of dimensional accuracy. Suwanprateeb et al. (2012), investigated phase conversion during 3D printing at optimized process parameters found that highest transformation rate results in greater flexural modulus and strength. Rai et al. (2018), performed response surface optimization on calcium sulfate based samples and found that tensile strength decreases with increasing layer thickness. Many other researchers (Li and Cao, 2012; Farzadi et al., 2015; Asadi-Eydivand et al., 2016) concluded that optimizing process parameters significantly improved the mechanical properties of 3D printed parts.