Multi-Faceted Analysis of Print Parameters’ Impact on Tribological Properties: Exploring the Relationship Between Surface Characteristics and Tribological Performance in Additively Manufactured Polymers and Metals
Date
2024-07-31Type of Degree
PhD DissertationDepartment
Mechanical Engineering
Restriction Status
EMBARGOEDRestriction Type
Auburn University UsersDate Available
07-31-2025Metadata
Show full item recordAbstract
Additive manufacturing (AM) has garnered increasing recognition for its capability to produce components with bespoke mechanical properties. Recent studies underscore that these properties can be meticulously fine-tuned through the strategic control of layer orientation and build structure. This study delves into the influence of print orientation on the tribological properties of 3D-printed polylactic acid (PLA) and Acrylonitrile Butadiene Styrene (ABS). Utilizing fused deposition modeling (FDM), PLA and ABS samples were fabricated in three distinct print orientations. Our tribological analysis reveals that variations in the build direction relative to the sliding direction induce anisotropy in wear characteristics. Intriguingly, samples printed with layers orthogonal to the sliding direction exhibited superior wear properties, while the coefficient of friction remained largely unaffected by the orientation of the print. Furthermore, PLA samples showcased significantly enhanced tribological performance compared to ABS. Variations in sliding speed also affected the wear properties of both materials, suggesting that optimal build orientation can substantially enhance the wear performance of additively manufactured thermoplastics. These findings pave the way for a new paradigm in the design of functionally graded materials. To advance our understanding and enhancement of printed materials, we first sought to determine the most effective measurement technique for capturing surface roughness. Ti-6Al-4V was selected for this purpose due to its as-built metal samples exhibiting the greatest variations, randomness, and extreme features compared to FDM printed samples. Thus, as-built Ti-6Al-4V samples were utilized to evaluate the effectiveness of different surface measuring equipment. Surface characterization of additively manufactured components is intricate due to their complex geometries and the presence of features such as asperities, undercuts, and deep, sharp valleys. Understanding the impact of these surface features on fatigue life is crucial, yet the variability inherent in measurement techniques remains contentious. In this context, the topography of as-built Ti-6Al-4V samples was analyzed using four techniques: contact stylus profilometers, white light interferometers, focus variation microscopy, and X-ray computed tomography. Our qualitative and quantitative analyses identified discrepancies among these measurement methods. Traditional tribological roughness parameters were employed to differentiate these techniques, offering valuable guidance for selecting effective scan parameters and measurement methods. Building on this knowledge, we investigated the effect of different build orientations and normal loads on the wear rate and coefficient of friction of PLA samples. PLA samples, printed in three orientations and tested under varying normal loads (50-100N), demonstrated that those printed with the sliding direction orthogonal to the printed layer achieved the best wear performance. Typically, the wear rate increased with normal load; however, higher infill and print speeds could mitigate this increase, transitioning the wear from abrasive to adhesive. Notably, shape and texture parameters correlated with the coefficient of friction, unlike average surface roughness. Predictive models for friction and wear were developed using machine learning techniques, including Support Vector Machine (SVM), Artificial Neural Network (ANN), Random Forest (RF), and Extreme Gradient Boosting (GBM). Among these, decision tree-based models outperformed others, underscoring the importance of system variables and surface roughness parameters. To further examine the impact of surface roughness on tribological properties, we extended our investigation to the AlSiMg alloy fabricated via laser powder bed fusion. This allowed us to determine whether surface roughness parameters affect metals in a manner similar to polymers. Our findings revealed that samples with varying build orientations exhibited anisotropy that influenced surface conditions and layer orientation, thereby altering wear rates. Machined samples with reduced roughness demonstrated improved wear resistance compared to their as-built counterparts, though the coefficient of friction remained relatively stable. The primary wear mechanism transitioned from abrasion in the initial stages to adhesive wear, attributed to material transfer and oxide layer formation due to sliding-induced heat generation. Delamination and crack propagation were also observed within the wear tracks. Statistical analyses confirmed a robust correlation between wear rate and surface roughness parameters. Shape and texture parameters displayed significant trends with the wear rate. After the initial break-in period, the coefficient of friction stabilized, showing minimal differences regardless of surface condition or build orientation. In summary, this comprehensive study links these four interconnected investigations to enhance the understanding and optimization of tribological performance in additively manufactured materials. By examining the interplay between print orientation, surface roughness, and their effects on wear characteristics, we aim to not only characterize but also improve the functional properties of 3D printed materials through informed design and advanced machine learning techniques.