×

Changes in local thermal regime due to AI data centres: A remote sensing analysis

Professor: Prof. Eswar R

Description:

Large AI data centres have been setup or under development at different parts of the globe.There are media reports that these large AI data centres produce heat and noise affecting the neighbourhoods in which they are setup. But are they really changing the thermal patterns in the locality? is an open question. The aim of this project is to analyse thermal images from remote sensing satellites. Students will identify the large AI data centres across the globe and analyze thermal images to see if there are any changes in the thermal footprint. The students will be able to learn remote sensing image analysis and as a part of the literature review, can learn about the energy and water demands of the AI data centres.

Pre-requisites:

• Introduction to remote sensing and python programming

×

Steel Slag–Derived Circular CaO Sorbents for Low-Cost Industrial CO2 Capture via Closed-Loop Organic Acid Extraction and Calcium Looping

Professor: Prof. Arnab Dutta

Description:

This project is a research-driven initiative focused on converting steelmaking waste, specifically steel slag, into calcium-based solid sorbents for industrial CO2 capture. Conventional amine-based capture systems remain expensive and difficult to scale in heavy industry due to energy-intensive regeneration, solvent degradation, corrosion, and unfavorable operating economics. In parallel, integrated steel plants generate large volumes of calcium-rich slag that is often underutilized or diverted into low-value applications. The proposed work targets this dual challenge by developing a circular pathway that valorizes slag into reusable CaO-based sorbents compatible with high-temperature capture processes such as calcium looping.

Objectives:

1. Develop an indirect carbonation route to selectively extract calcium from steel slag using mild organic acids under moderate conditions.


2. Regenerate the leaching medium to enable a closed-loop, low-waste process with minimal reagent make-up.


3. Convert calcium-rich intermediates into porous, reactive CaO-based sorbents via controlled thermal activation.


4. Evaluate sorbent performance under cyclic carbonation: calcination conditions, benchmarking capacity retention, kinetics, attrition, and stability.


5. Assess process integration potential at steel plants, prioritizing low-cost inputs, operational simplicity, and scalable unit operations.

Approach:

The work investigates controlled acid leaching to separate calcium from complex silicate/ferrite matrices, followed by recovery of calcium precursors and regeneration of the organic acid. Thermal activation will be optimized to engineer pore structure and reactivity. Cyclic testing will establish capture performance and durability, providing evidence for pilot-relevant feasibility.

Deliverables

• A characterized dataset of slag composition, calcium extraction efficiency, and impurity management strategies.
• A validated closed-loop leaching and solvent regeneration protocol with mass and energy balances.
• Prototype CaO-based sorbents with defined porosity, reactivity, and cyclic stability metrics.
• Cyclic carbonation–calcination performance reports (≥50–100 cycles target), including attrition and deactivation analysis.
• A preliminary techno-economic and integration blueprint for on-site deployment at steel plants, identifying key scale-up risks and pathways to pilot demonstration.

Pre-requisites:

• NA

×

Machine Learning–Driven Strategies for Sustainable CO₂ Utilization and Transformation

Professor: Prof. Debabrata Maiti

Description:

The rapid increase in atmospheric CO₂ levels is one of the most pressing challenges of climate change and sustainability. While carbon capture technologies are advancing, an equally important frontier lies in sustainable CO₂ utilization, where captured CO₂ is transformed into value-added chemicals, fuels, or materials. This project aims to explore how machine learning (ML) can accelerate the discovery and optimization of sustainable CO₂ transformation pathways.

The primary objective of this project is to introduce undergraduate students to data-driven approaches for sustainability-oriented chemical transformation, focusing on CO₂ conversion reactions. Students will work with curated datasets from literature and open chemical databases that include reaction conditions, catalysts, and yields for CO₂ reduction or functionalization processes. Using these datasets, the project will involve developing ML models to identify patterns that govern reaction efficiency, selectivity, and sustainability metrics such as energy input and atom economy.

The project will begin with an overview of CO₂ utilization chemistry and sustainability considerations, followed by hands-on training in data preprocessing, feature extraction, and model building using standard ML techniques (e.g., regression and classification models). Students will then apply these models to analyze existing reaction datasets and generate insights into factors influencing successful CO₂ transformation. Where feasible, simple model-based optimization or hypothesis generation will be explored to propose potentially improved reaction conditions or catalyst features.

Deliverables:

1. a clean, well-documented dataset related to CO₂ transformation reactions,
2. trained ML models with performance evaluation and interpretation,
3. visualizations highlighting key trends relevant to sustainable chemistry, and
4. a short technical report and presentation summarizing findings and sustainability implications.

Pre-requisites:

• Basic knowledge of undergraduate-level chemistry (physical/organic chemistry fundamentals)
• Introductory familiarity with Python programming
• Willingness to learn basic concepts of machine learning and data analysis
• Prior ML knowledge is helpful for the project

×

Integrating climate module in an integrated assessment model

Professor: Prof. Yogendra Shastri

Description:

An integrated assessment model (IAM), known as the Generalized Global Sustainability Model (GGSM) has been developed by us in the past and used to study various sustainability related aspect such as the food-energy-water nexus (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0266554) and circular economy (https://www.sciencedirect.com/science/article/pii/S0921344919303660). It is a model that integrates macroeconomics and technology related aspects in a food-web model. The model needs to be extended to include climate modules, so that climate change related effects can be studied through the integrated model. The goal of this project is to add climate related components, especially those specific to the water resources (such as precipitation). The work will require historical data analysis and identifying simplified functional forms and model connections. The work is completely computational in nature and highly interdisciplinary. Coding in Python along with some data analysis will be required.

Pre-requisite:

Prior experience in computer programming, preferably in a procedural language. Based knowledge of mass balance and mathematical modeling. Students from chemical, civil, CESE, and CTARA may particularly be appropriate.