Abstract

This dissertation aims to study the underlying physics of anomalous transport of emulsions in porous media and quantify the role of droplet-pore and droplet-droplet interactions in transport in porous media experimentally. The complexities of this problem arise from the network heterogeneity, correlations between the pores across the medium, interfacial properties of the fluids, and the correlations between the pore scale dynamics and system-level transport. This dissertation presents significant advancements in the study of emulsion dynamics within two-dimensional porous media through the development of innovative image analysis and tracking methods. These advancements include precise detection and tracking of deformable objects, which is particularly challenging in experiments involving large datasets with thousands of objects and millions of positional updates. Utilizing cutting-edge microfluidic techniques, we developed an integrated experimental setup combining an on-demand drop maker and advanced imaging methods. By precisely controlling emulsion size, concentration, and injection rates, we uncovered critical insights into pore-level dynamics and bulk transport properties. Through the application of variational mode decomposition (VMD), we effectively distinguished between uniform and irregular droplet formations, enhancing the accuracy of our measurements. Our findings demonstrate that emulsions predominantly flow through higher velocity pores, often becoming trapped in smaller pores, which reduces porosity and creates preferential pathways. Introducing slight polydispersity in emulsion sizes further improved transport efficiency by revealing additional pathways. We also explored the effects of device scaling, dye contrast adjustments, and interfacial tension on emulsion behavior, leading to refined detection and tracking algorithms. These advancements provide a robust framework for future studies and have significant implications for applications in soil remediation, drug delivery, and enhanced oil recovery. Additionally, the creation of high-quality datasets from our experiments lays the groundwork for leveraging machine learning techniques to further understand and predict the complex nature of emulsion transport in porous media. This interdisciplinary approach offers the potential to bridge experimental observations and theoretical models, enhancing our ability to manage and optimize fluid dynamics in various applications. This research represents a substantial contribution to the field of microfluidics and multiphase flow, offering new methodologies and insights that will propel future innovations and applications. The results of this research provide the necessary research platform to advance the research on transport of deformable particles in porous media. In addition to the research questions, these understandings will impact industrial processes such as filtration in food industry, sorting in pharmaceutical, drug delivery in medical, enhanced oil recovery, and soil remediation in environmental industries.

Publication Date

2024

Document Type

Dissertation

Student Type

Graduate

Degree Name

Imaging Science (Ph.D.)

Department, Program, or Center

Chester F. Carlson Center for Imaging Science

College

College of Science

Advisor

Shima Parsa

Advisor/Committee Member

Jairo Diaz

Advisor/Committee Member

Joseph Hornak

Campus

RIT – Main Campus

Download

Share

COinS