|
Unraveling the rheology-printability relationship of 3D printable biomaterials for food application |
|---|---|
| รหัสดีโอไอ | |
| Title | Unraveling the rheology-printability relationship of 3D printable biomaterials for food application |
| Creator | Theo Claude Roland Outrequin |
| Contributor | Paiboon Sreearunothai, Advisor |
| Publisher | Thammasat University |
| Publication Year | 2567 |
| Keyword | 3D food printing, Rheology, Biopolymers, Additive manufacturing, Food proteins, Artificial intelligence, Machine learning, Smart processing |
| Abstract | The attraction toward 3D food printing is growing significantly since the development of extrusion-based 3D printing technology. Biopolymers are frequently incorporated into the food ink formulation to modify the rheological properties and make the formulation processable, significantly impacting the ability of food products to be printed. The 3D food printing process comprises four stages: (a) the ink preparation, (b) the ink extrusion, (c) the ink deposition, and finally (d) the formation of the 3D structure. From a recent literature review, it has been demonstrated that the extrudability of an ink is controlled by its yield stress and rheological flow parameters, which may be obtained from the Power Law or the Herschel-Bulkley model. These parameters directly influence the minimal applied pressure required to initiate the ink’s flow. Moreover, an acceptable range of viscoelastic properties resulting in stable 3D printed structure formation has been drawn. The combination of these rheological parameters within the biopolymers-based ink formulation remains an important challenge. Printing and rheological parameters possess a strong influence on the printed object. The control over the ink deposition step is also important to obtain printed objects with dimensions closely aligned to the intended dimensions. The first part of this study aims to understand the impact of the rheological properties on the fidelity of the printed object presented by the printed line width. Having a printed line width close to the one prepared in the model is essential to obtain a correct shape fidelity as the extrusion-based 3D printing technology corresponds to the deposition of material layer by layer. In this work, the factors affecting material spreading in 3D printing, focusing on pectin inks, are investigated. These factors include rheological properties, such as flow properties and complex modulus, which significantly influence the spreading ratio, with higher concentrations reducing spreading but complicating extrusion. The impact of printing parameters, such as nozzle diameter, pressure, and speed, on the spreading ratio were also analyzed. Balancing these factors is crucial for achieving target dimensions, considering the trade-off between reduced spreading and potential extrusion challenges. Machine learning regressions are applied to establish a robust framework comprising all the previous parameters for precise ink deposition control, advancing biomaterial 3D printing, and fostering innovation in fabricating functional, dimensionally accurate objects. The Extra Tree regression model with four features (pressure, speed, G_0^*, and n) demonstrated that the rheological properties accounted for 92% of the weight in the spreading variation, whereas the printing parameters contributed to 8%.After studying the dimensional changes in 2D lines, a deeper focus on the parameters yielding a correct formation of a 3D structure is made. Mung bean protein isolate was added to the pectin solution to form an ink formulation with enhanced rheological properties. Moreover, the addition of protein powder promoted the development of ink with improved nutritional benefits to fabricate high-protein 3D-printed snacks. The dimensional accuracy of the printed shape is improved with the increasing protein content, as it yields stronger viscoelastic properties. Machine learning classification was utilized as a quality control tool to assess the printability of a formulation. This approach classified print samples as either 'within range' or 'out of range' based on their features and deviations, with an acceptable error range set at 10% deviation from the target height. Additionally, the impact of baking on the shape of the printed object was examined, revealing that formulations with higher biopolymer content were more effective at preserving the shape of the 3D-printed baked snack.Lastly, the combination of 3D food printing and Artificial Intelligence (AI) is discussed as food additive manufacturing technology has advanced rapidly to help promote the creation of diverse and customized food products. However, scaling up this process remains challenging, thus limiting its use in the food industry. Integrating AI technology into the printing process presents transformative opportunities for the food industry, especially by enhancing personalized nutrition and production efficiency. AI-driven food ink development can enable the precise formulation of inks tailored to specific dietary needs, improving both nutritional value and user satisfaction while optimizing functional properties for efficient processability. AI-driven selection of printing parameters based on the knowledge of food ink rheological properties could accelerate research and development, thus fostering innovation. Additionally, real-time monitoring and automated adjustments during printing can enhance process efficiency and ease the final quality control. Overall, the adoption of AI combined with automation in 3D food printing is foreseen to revolutionize the food industry, offering scalability to the printing process to better accommodate evolving consumer demands. |
