IIT Bombay develops predictive models for sintering-induced shrinkage

Researchers at the Indian Institute of Technology (IIT) Bombay have made significant strides in the field of additive manufacturing by developing predictive models that can accurately calculate sintering-induced shrinkage and deformation in 3D printed components. This groundbreaking work, led by Professor Gurminder Singh from the Department of Mechanical Engineering, has been documented in two studies focusing on ceramics and copper materials.

Understanding Sintering-Induced Shrinkage

Sintering is a critical process in additive manufacturing, particularly for materials like ceramics and metals. During this process, parts undergo heat treatment, which often leads to shrinkage and significant dimensional changes. This shrinkage can affect the final geometry of the printed components, making it essential to predict these changes accurately before manufacturing begins.

Research Overview

The research team aimed to address the challenges associated with shrinkage by developing models that allow for the prediction of final part geometry at the design stage. This approach eliminates the need for iterative physical testing, which can be time-consuming and resource-intensive.

Ceramics Study

In the ceramics study, researchers Pranith Kumar Reddy Puchakayla and Professor Prasanna Gandhi collaborated with Professor Singh to investigate 3 mol% yttria-stabilised zirconia (3-YSZ). They modeled the material as a viscous fluid during the heating process. Key findings from this study include:

• The team measured density and viscosity changes across temperature intervals.
• Simulations were validated against three geometries: a cylinder, an I-section, and a branched “pine tree” structure.
• The model predicted final dimensions with an error range of just 0.8% to 2.03%.

Professor Singh emphasized the importance of viscosity in determining how easily the material flows during printing, how well the layers stack, and the internal stress stored in the printed part. Additionally, relative density plays a crucial role before sintering, indicating the amount of solid material present and the pore volume that exists. Lower density before sintering correlates with higher shrinkage and deformation.

Copper Study

The copper study, conducted by researcher Sri Bharani Ghantasala and Professor Singh, involved the development of a hybrid model for copper materials. Key aspects of this study include:

• The use of an artificial neural network (ANN) trained on experimental and computer-generated data from eight sintering experiments.
• Seven input variables were analyzed to determine their influence on the final part geometry.
• SHAP (SHapley Additive exPlanations) analysis identified process time and heating rates as the dominant factors affecting final geometry.
• The hybrid model achieved a 98% success rate in matching experimental results.

This innovative approach outperformed traditional computer models, particularly when predicting the final shape of complex overhanging geometries. Professor Singh noted that this transition from trial-and-error sintering to predictive, model-based manufacturing represents a significant paradigm shift in advanced additive manufacturing science.

Implications for the Future

The development of these predictive models has far-reaching implications for the field of additive manufacturing. By enabling designers to account for shrinkage and deformation during the design phase, manufacturers can create parts that meet precise specifications without the need for extensive physical testing. This not only saves time and resources but also enhances the overall efficiency of the manufacturing process.

Potential for Smart CAD Tools

Professor Singh envisions a future where smart CAD (Computer-Aided Design) tools can automatically apply predicted shrinkage fields to designs, generating pre-compensated geometries directly. This advancement could streamline the design process and further improve the accuracy of 3D printed components.

Conclusion

The research conducted by IIT Bombay represents a significant advancement in the understanding and prediction of sintering-induced shrinkage in additive manufacturing. By utilizing physics-based and hybrid AI approaches, the team has developed models that can accurately predict the final dimensions of printed parts, paving the way for more efficient and effective manufacturing processes. As the field continues to evolve, these innovations will play a crucial role in shaping the future of additive manufacturing.

Note: The findings discussed in this article are based on research published by IIT Bombay and reflect the current state of knowledge as of October 2023.