INTERFACIAL THERMAL AND TRANSPORT LAB (ITTL) : FOCUS AREAS

Overview- To design functional interfaces with tunable characteristics including wettability, wickability, porosity, and roughness, while enhancing thermal, electrical, and mechanical properties. The recent focus is on contemporary 2D materials such as MXene, High Entropy Oxides (HEOs) along with conventionally studied copper/graphene, copper/nickel, and polymer/carbon nanofiber particle/fiber composites involving hierarchical structures for improved multiphase transport. 

Keywords- Additive Manufacturing, Powder Metallurgy, Coatings, Membranes, Wet Chemistry, Design of Experiments

Hypothesis- Compositional and structural heterogeneity at multiple length scales enhances transport properties through improved nucleation, interfacial phenomena, and synergistic effects. 

Methodology- Diverse experimental and computational toolset to enable precise control over material composition, structure, and properties across multiple length scales. Powder metallurgy techniques such as ball milling and sintering to induce porosity in metal particles. Various deposition methods, including electrodeposition, screen printing, drop casting, and spin coating. Formulation of specialized inks, particularly metal organic deposition (MOD) inks and graphene inks, to generate fiber-like morphologies. Magnetic and electric field-assisted alignment techniques to construct hierarchical structures. 

Impact- The first group to demonstrate applications of Metal Organic Deposition (MOD) ink for scaled-up manufacturing for thermal applications. Developed magnetically aligned metal organic deposition ink-based surfaces leading to an improvement of 50% in heat flux and 105% in heat transfer coefficient compared to a plain copper surface.

Representative publication(s):  https://doi.org/10.1016/j.applthermaleng.2022.118473

Overview- To develop fundamental understanding of multiphase transport phenomena, focusing on nucleation, bubble dynamics, and phase change heat transfer in porous media, as well as capillary-driven flows in hierarchical structures. We aim to optimize heat transfer, energy conversion, and material characteristics in thermal management systems, including high-performance heaters with enhanced critical heat flux, heat pipes, and vapor chambers for electronics cooling. Our research extends to water treatment applications, nucleate boiling heat transfer, and enhanced condensation and evaporation processes in thermal systems. 

Keywords- Boiling, Membrane Distillation, Heat and Mass Transfer, Physics-Based Modeling 

Hypothesis- Interfacial characteristics such as roughness, porosity, chemical composition can enhance multiphase transport and heat transfer across solid-solid and solid-fluid interfaces. 

Methodology- Diverse experimental and computational toolset to quantify thermal performance parameters. Advanced prototyping techniques with high-precision data acquisition systems, ensuring result reliability through rigorous uncertainty analysis. Cutting-edge visualization methods enable us to capture complex multiphase phenomena. This experimental approach is complemented by theoretical and physics-based modeling, providing a comprehensive framework for investigating interfacial thermal transport and multiphase flow dynamics across various scales and applications.  

Impact- Record breaking critical heat flux of (CHF) of 289 W/cm² achieved at the lowest recorded wall superheat temperature of 2.2°C for flat surfaces. 

Representative publication(s):  https://doi.org/10.1021/acs.langmuir.4c00902

Overview- To harness advanced AI to create machine learning frameworks that leverage experimental data and physical models to predict material performance and lifespan, facilitating the design of efficient thermal management systems. We also apply Large Language Models (LLMs) to analyze complex survey data and scientific literature, uncovering hidden patterns and generating new hypotheses. 

Keywords- Image segmentation, Large Language Models, Time-Series, Physics-Informed AI 

Hypothesis- Experiment and physics-informed AI algorithms will outperform conventional data-driven approaches in accuracy and interpretability when analyzing multiphase transport phenomena, predicting materials performance, and extracting insights from surveys. 

Methodology- Diverse statistical and machine learning techniques integrating experimental data and physical principles to accelerate discovery and enhance prediction accuracy: a) image segmentation algorithms to analyze experimentally generated multiphase flow images, correlating visual data with transport mechanisms, b) machine learning time series models for the prediction of material life and performance, enabling optimized design and maintenance of thermal management systems, c)  large language models to analyze survey data, uncovering hidden patterns and biases.   

Impact- Our AI-driven image segmentation of 2D SEM images accurately predicts mixed matrix membrane porosity, validated by Brunauer-Emmett-Teller (BET) measurements, enabling rapid, non-destructive characterization of membrane structures, and potentially accelerating the development of high-performance materials for separation and filtration technologies. 

Representative publication(s):  https://doi.org/10.1021/acsomega.4c03024