A Dissertation entitled Additive Manufacturing towards the Realization of Porous and Stiffness-tailored NiTi Implants by Jason M. Walker Submitted to the Graduate Faculty as partial fulfillment of the requirements for the Doctor of Philosophy Degree in Biomedical Engineering ________________________________________ Dr. Mohammad Elahinia, Committee Chair ________________________________________ Dr. Sonny Ariss, Committee Member ________________________________________ Dr.
Sarit Bhaduri, Committee Member ________________________________________ Dr. Christopher Cooper, Committee Member ________________________________________ Dr. Matthew Franchetti, Committee Member ________________________________________ Dr. Komuniecki, Dean College of Graduate Studies The University of Toledo May 2014 Copyright 2014, Jason M.
Walker This document is copyrighted material. Under copyright law, no parts of this document may be reproduced without the expressed permission of the author. An Abstract of Additive Manufacturing towards the Realization of Porous and Stiffness-tailored NiTi Implants by Jason M. Walker Submitted to the Graduate Faculty as partial fulfillment of the requirements for the Doctor of Philosophy Degree in Biomedical Engineering The University of Toledo May 2014 This research is focused on the development of an additive manufacturing process for porous and stiffness-tailored nitinol implants.
Selective laser melting (SLM) is an emerging additive manufacturing (AM) technology, which makes possible the production of 3D parts directly from metallic powders. AM enables production of 3D geometries that are not possible using traditional techniques. Other non-traditional manufacturing methods do not allow for precise control of pore geometry and distribution. With SLM, features such as engineered porosity, hollow parts, curved holes and filigree structures are suddenly realizable.
Of particular interest in this research is the ability to engineer and control porosity. The first stage of this study focused on understanding the processing parameters for manufacturing dense nitinol parts with a Phenix Systems PXM. A parametric analysis of process parameters on the quality and functionality of the SLM nitinol parts was conducted. First, a single track analysis was performed to understand the effect of basic processing parameters on the melting and re-solidification of powder.
Next, a parametric iii analysis of SLM process parameters, including laser power, scan velocity, and hatching spacing, on part density was carried out. Finally, a chemical analysis on impurity pickup during laser processing was performed. Based on the results of these studies, an optimal setup for processing nitinol on the Phenix Systems PXM was determined. Using the optimal parameter setup, SLM nitinol parts were manufactured for thermal, mechanical, and functional analysis.
It was found that the transformation temperatures of SLM nitinol parts were approximately 10 C higher than the powder. This effect is attributed to nickel evaporation during laser processing. Mechanical properties were assessed in compression testing and determined to be very similar to properties of conventionally processed nitinol. The shape memory effect was also demonstrated, indicating that SLM nitinol retains its functional characteristics.
To improve the outcome of long-term metallic implant use, the mechanical properties of implants need to better match those of bone. Nitinol structures with engineered porosity were designed and manufactured by SLM to meet these requirements. It is shown that pore size, shape, and distribution can be customized by CAD and manufactured by SLM. In this study, porous nitinol structures were manufactured with three different porosities (32%, 45%, and 58%) and featured interconnecting pores ranging from 0.0 mm in size.
A reduction in elastic modulus based on porosity is demonstrated. iv Acknowledgements First, I would like to thank my advisor, Dr. Mohammad Elahinia, for the opportunity to work on this great project. I would also like to thank him for his guidance and support in all aspects of my research.
I owe many thanks to my committee members for taking time out of their busy schedules to provide the advice and direction required to complete this research. This work would not be possible without them. I want to thank all of my colleagues and fellow researchers who I had the pleasure of working with in the Dynamic and Smart Systems Laboratory. Finally, I want to thank my family.
Mom and Dad – You inspire me every day. v Table of Contents Abstract .v Table of Contents. vi List of Tables .x List of Figures .1 Identification of Material Requirements .3 Effect of Surgical Implants on Adaptive Bone Remodeling .4 Biocompatibility of Nitinol .1 Biocompatibility of Porous Nitinol .5 History of Nitinol in Medical Devices .6 Nitinol in Orthopedic Applications .18 3 Shape Memory Alloys .1 History of Nitinol .1 Mechanical Behavior: Shape Memory and Superelasticity .1 History of AM .2 Selective Laser Melting .2 Phenix Systems PXM .3 Existing Research on AM of Nitinol .55 5 Parameter Setup, Part 1: Powder and Single Tracks.2 Quality and Width of Single Tracks .3 Mathematical Prediction of Single Track Width .71 6 Parameter Setup, Part 2: Dense Parts .85 7 Thermal, Mechanical, and Functional Properties of SLM Nitinol .100 8 Designed Porosity in Nitinol Structures .1 Review of Metallic Implant Requirements .2 Computer-aided Design of Porosity.109 9 Conclusions and Recommendations for Future Work .2 Recommendations for Future Work.4 Estimated Market Share and Sales .1 Market Entry and Growth Strategy .3 Sales Tactics and Service .8 Product Development Plan .9 Manufacturing and Operations Plan .11 Overall Project Schedule.12 Critical Risks and Assumptions .143 ix List of Tables 5.1 Transformation temperatures of nitinol powder .2 Chemical impurity contents of nitinol powder .3 Amplitude and exponent of the fitting functions by laser power.1 Impurity contents measured in powder and SLM nitinol part .2 Optimal parameters for SLM processing of nitinol on a Phenix PXM.1 Transformation temperatures of SLM nitinol part and powder .1 Dimensions of unit cells for porous structures and resulting pore formation .2 Elastic modulus for typical implant materials and porous nitinol .1 US 2004 knee implant market share, by company .2 Financial calculations for Ortho3d.142 x List of Figures 3-1 Phase diagram of the transformation temperatures of nitinol .27 3-2 Temperature-induced forward transformation in a zero stress state .28 3-3 Temperature-induced reverse phase transformation in a zero stress state .28 3-4 Twinned martensite is detwinned by an applied load .29 3-5 Martensite remains in detwinned state when the applied load is removed .30 3-6 Upon heating past Af, detwinned martensite transforms into austenite .31 3-8 (a) A crystal lattice and (b) resulting plastic deformation by an ordinary slip .32 3-9 (a) A twinned crystal lattice and (b) resulting pseudoplastic deformation .33 3-10 Stress-strain behavior of nitinol at different operating conditions.37 3-11 Phase diagram of a Ti-Ni alloy .39 3-12 Ms as a function of nickel content in quenched NiTi alloys.40 3-13 Effect of aging temperature on R-phase transformation start temperature .43 4-1 Phenix Systems PXM Selective Laser Melting Machine .49 4-2 Sequence of operations of the SLM process .51 4-3 Visual representation of the Staircase Effect and support structures .52 4-4 Basic scan strategy showing laser trajectories all in the same direction.55 4-5 Alternating x/y scan strategy .55 4-6 Alternating x/y scan strategy with 90 degree rotation per layer .56 xi 5-1 Electrode Induction-melting Gas Atomization (EIGA) .59 5-2 SEM micrograph of the nitinol powder .60 5-3 Transformation temperatures of the nitinol powder .62 5-4 Measured single track width as a function of scan velocity .65 5-5 Micrographs of SLM single track .67 5-6 Amplitude, a, of the fitting function as a function of laser power .69 5-7 Calculated single track width as a function of scan velocity .70 6-1 Cuboid blocks produced by SLM for density analysis .74 6-2 Poor surface quality caused by too high energy input .75 6-3 Relative density of SLM nitinol parts as a function of scan velocity .75 6-4 Relative density of SLM nitinol parts as a function of energy density .76 6-5 SLM nitinol samples mounted and polished for microscopy .77 6-6 Comparison of pore formation in low and high density parts .79 6-7 Optical micrograph of SLM nitinol .80 6-8 Comparison of pore formation in two SLM nitinol parts .81 6-9 Effect of hatch spacing on relative density of SLM nitinol parts .82 6-10 Oxygen content in SLM nitinol parts.85 6-11 Micrograph of SLM nitinol single track with optimal parameter setup .87 6-12 Micrograph of surface of SLM nitinol part with optimal parameter setup .88 7-1 Transformation temperatures of SLM nitinol part and nitinol powder.90 7-2 Stress-strain curve for SLM and conventional nitinol in the martensitic state .93 7-3 Fracture of SLM nitinol samples at 45º with respect to the loading direction.95 7-4 Stress-strain curves of shape memory cycling to 300 MPa .97 xii 7-5 Stress-strain curves of shape memory cycling to 900 MPa .98 7-6 Cumulative irreversible strain in SLM nitinol during cyclic testing .103 8-2 Porous SLM nitinol structures .104 8-3 Micrographs of porous SLM nitinol structures .105 8-4 Stress-strain curves to 12% strain for porous SLM nitinol .106 8-5 Stress-strain curves to 2.5% strain for porous SLM nitinol .109 10-1 (a) Typical total knee replacement and (b) porous nitinol structure .123 10-2 Estimated number of total knee replacements performed in the US .126 10-3 Projected primary total knee arthoplasties in the US through 2030 .127 10-4 Projected total knee revisions in the US through 2030 [206] .127 10-5 Break-even analysis for Ortho3d .141 xiii Chapter 1 Introduction The aim of this research is to investigate the use of laser-based additive manufacturing to produce porous and stiffness-tailored nitinol implants. To improve the outcome of long-term metallic implant use, the mechanical properties of implant materials need to better match those of bone.
Orthopedic implants are used to restore and maintain proper biomechanical function in the body, yet current technologies utilize materials that possess drastically different mechanical properties than the biological materials they replace. Furthermore, these implants are introduced into highly unique environments (i. different patients), yet are mass designed with little regard for the individuals’ needs. Most metallic implants are made out of 316 stainless steel, titanium (Ti-6Al-4V), or cobalt-chrome.
Each of these materials can be 5 – 15 times stiffer than the surrounding bone tissue. During long-term fixation, this severely disrupts the physiomechanical processes which regulate bone remodeling. As a result, surrounding bone tissue degrades which leads to implant loosening and, ultimately, catastrophic failure. 1 To meet this unmet need, nitinol is identified as a more suitable material for use in long-term metallic implants.
Nitinol features high strength, low stiffness, high recoverable strain, and good biocompatibility. Additionally, an additive manufacturing processing called selective laser melting (SLM) is identified as a better processing route than traditional methods. SLM circumvents current issues with manufacturing nitinol components (i. conventional machining of nitinol is incredibly difficult) and opens up possibilities for patient-specific implant design.
During the SLM process, a laser beam is used to manufacture a solid part by selectively binding powder particles through localized heating. The part is produced additively by manufacturing one cross-sectional layer at a time, which is digitally created by slicing a CAD model. Direct-from-CAD manufacturing allows for the design of structures with engineered porosity. In this way, the stiffness of porous structures can be modulated through the customization of pore size, shape, and distribution.
In summary, the primary objective of this research is to manufacture nitinol components with engineered porosity by selective laser melting to achieve desired mechanical properties.1 Approach SLM is a layer-based manufacturing process which utilizes direct melting of parts from a metallic powder-bed using a high-power fiber laser. The process features a considerable number of parameters which require optimization including powder characteristics, laser parameters, and layering technique. Even the smallest compositional 2 variances and all types of microstructural defects can strongly affect the integrity and functionality of nitinol parts. Considerable work has been done to optimize the SLM process for materials such as titanium and stainless steel, but it is far from being established as a processing method for nitinol.
To this end, research is conducted on an SLM machine (Phenix Systems PXM) to develop and optimize the manufacturing method for dense nitinol components. A variety of analytical methods were used to evaluate and characterize fabricated AM nitinol components. These included microscopy, density analyses, differential scanning calorimetry (DSC), chemical analyses, and mechanical testing.