Paper on Deciphering Oxygen Evolution Reaction Activity: A QM/ML Approach with Single Atom Catalysts
Prof. Ranjith Thapa in collaboration with two of his research scholars, Mr E. S. Erakulan Mr Sourav Ghosh and has come up with a groundbreaking research that has resulted in the publication of a scholarly paper titled, Specific Descriptor for Oxygen Evolution Reaction Activity on Single Atom Catalysts Using QM/ML.
Abstract of the paper
Descriptors are properties or parameters of a material that is used to explain any catalytic activity both computationally and experimentally. Such descriptors aid in designing the material’s property to obtain efficient catalyst. For transition metals, d-band center is a well-known descriptor that shows Sabatier type relation for several catalytic reactions. However, it fails to explain the activity when considering same metal active site with varying local environment. To address this, density functional theory was used for single atom catalysts (SACs) embedded on armchair and zigzag graphene nanoribbons (AGNR and ZGNR). By varying the anchoring nitrogen atoms’ orientation and considering pristine and doped cases, 432 active sites were used to test the oxygen evolution reaction (OER) activity. It was observed that S and SO2 dopant helps in reducing the overpotential on Co-SAC (h = 0.28 V). Along with the d-band center, a total of 105 possible descriptors were individually tested and failed to correlate with OER activity. Further, PCA was employed to narrow down unique descriptors and machine learning algorithms (MLR, RR, SVR, RFR, BRR, LASSO, KNR and XGR) were trained on the two obtained descriptors. Among the models, SVR and RFR model showed highest performance with R2 = 0.89 and 0.88 on test data. This work shows the necessity of a multi-descriptor approach to explain OER catalytic activity on SAC and the approach would help in identifying similar descriptors for other catalytic reactions as well.
Social Implications:
Computational studies have proven to be a vital tool to predict new materials and also assess the behaviour towards various catalytic reactions. They also identify the innate properties of the material which drives the catalytic activity. It helps in designing the material with required property to improve the catalytic activity. Descriptors are such computationally obtained properties/parameters of a material that has a meaningful relation with any catalytic property of a chemical reaction. d-band center, given by Hammer and Norskov in 1995, explained the binding strength of oxygen atom on pure transition metals. The d-band center shows Sabatier type relation with chemical activity and has been widely used to explain the catalytic activity of several reactions since its formulation. The adsorbate state after interaction with delocalized s-states of the metal atom is almost constant while that resulting from d-states interaction, is split into bonding and antibonding states. Hence the s-states were not considered. It is well known that, when the dimensions of a system are lowered the states become narrow and localized. In such systems, the d-band center does not explain the catalytic activity well and it is an open research problem.
Future Projects:
Density functional theory with machine learning approach could further be used and improved on similar SACs from which a predictive model equation could be constructed. Also, the proposed models are open to exploration on other catalytic reactions as well.
The authors thank SRM University-AP and National Super Computing mission for providing the computational facility.
- Published in Departmental News, News, Physics News, Research News
Enhancing Visual Saliency in Group Photographs: A Novel Approach for Improved Security and Healthcare Applications
Dr Ravi Kant Kumar, an Assistant Professor at the Department of Computer Science and Engineering, and his research scholar, Ms Gayatri Dhara, have come up with a patent titled “A System and Method for Enhancement Of Visual Saliency Of Intended Face In Group Photography.” The patent, with Application Number 202441040020, employs pathbreaking technology to enhance security and healthcare applications, with real-time face recognition and remote diagnostics.
Abstract:
Visual saliency is a way of figuring out which parts of a scene draw our attention the most. When looking at a crowd or a group of faces, our eyes naturally focus more on certain faces than others. This happens because some faces have dominant features that stand out more. For faces that don’t naturally catch our attention, there is a need to make them more noticeable. This new method and system are designed to do just that. The system calculates scores based on various factors like skin tone, colour, contrast, position, and other visual details. These scores help identify which face needs enhancement, making it more prominent in a group of faces. The primary advantage of this invention is its potential to improve user experience in various applications, such as photo editing, social media, security systems, and more. By giving users, the control to select and enhance a specific face, it allows for a more personalised and targeted approach to face recognition and enhancement. This could be particularly beneficial in scenarios where the user wants to highlight a specific individual in a group photo or in a crowd. Overall, this invention represents a significant advancement in the field of face recognition and image enhancement, offering a novel and user-centric approach to visual saliency. It opens up new possibilities for user interaction and control in image editing and face recognition technology.
Practical and Social Implications
The practical implementation of this research lies in its ability to identify and enhance faces within a group or crowd that do not naturally draw attention. This innovative method and system address this issue by calculating saliency scores based on factors such as skin tone, colour, contrast, position, and other visual details. These scores are then used to identify faces that need enhancement to become more prominent in a group. The system’s ability to enhance specific faces has significant practical applications in several fields.
Photo Editing: Users can easily enhance specific individuals in group photos, ensuring that everyone stands out as desired. This is particularly useful for personal photos, event photography, and professional photo editing.
Social media: Enhanced face recognition and saliency can improve user experience by allowing users to highlight specific people in their posts, making photos more engaging and personalised.
Security Systems: In surveillance and security applications, the ability to enhance less prominent faces can improve the accuracy of face recognition systems, aiding in the identification of individuals in crowded or low-visibility conditions.
Collaborations:
SRM University-AP,
Dr Ravi Kant Kumar,
Mrs Gayathri Dhara.
Future Research Plans:
Future plans for this visual saliency-based face enhancement system include refining algorithms for greater accuracy and efficiency, and integrating with popular photo editing software and social media platforms for seamless user experience. The technology will be expanded into security and healthcare applications, enhancing real-time face recognition and remote diagnostics. Emphasis will be placed on reducing biases, ensuring privacy protection, and enabling user customisation. Collaborations with academic institutions will drive further research, while commercialisation efforts will focus on launching products globally.
- Published in CSE NEWS, Departmental News, News, Research, Research News
Dr Negi’s Research Exploration of the Sugeno Exponential Function and Its Multidisciplinary Applications
Dr Shekhar Singh Negi from the Department of Mathematics has published a research paper titled “A note on Sugeno exponential function with respect to distortion.” Dr Negi’s research investigates the Sugeno exponential function. This research develops new mathematical tools and rules to work with a different way of measuring things, which can be useful in various fields like economics, biology, or any area where traditional measurements don’t quite fit the problem at hand.
Abstract:
This study explores the Sugeno exponential function, which is the solution to a first order differential equation with respect to nonadditive measures, specifically distorted Lebesgue measures. We define k-distorted semigroup property of the Sugeno exponential function, introduce a new addition operation on a set of distortion functions, and discuss some related results. Furthermore, m-Bernoulli inequality, a more general inequality than the well-known Bernoulli inequality on the real line, is established for the Sugeno exponential function. Additionally, the above concept is extended to a system of differential equations with respect to the distorted Lebesgue measure which gives rise to the study of a matrix m-exponential function.
Finally, we present an appropriate m-distorted logarithm function and describe its behaviour when applied to various functions, such as the sum, product, quotient, etc., while maintaining basic algebraic structures. The results are illustrated throughout the paper with a variety of examples.
Collaborations:
Prof. Vicenc Torra, Professor at the Department of Computing Science at Umea University. His area of research include artificial intelligence, data privacy, approximate reasoning, and decision making.
Future Research Plans:
To explore the aforementioned derivative and investigate results with applications in real life.
- Published in Departmental News, Math News, News, Research News
Research Paper on ESG Scores and Their Effect on Polluting Companies After COVID-19
The Department of Commerce, under the Paari School of Business, is proud to present the research publication of Dr Lakshamana Rao Ayyangari, Guest Faculty Dr Sankar Rao, and Research Scholar Mr Akhil Pasupuleti. Their research paper, titled “Assessing the impact of ESG scores on market performance in polluting companies: a post-COVID-19 analysis,” is featured in the Q2 journal “Discover Sustainability.” Here is an interesting abstract of their research.
Abstract:
The study aims to unravel the impact of Environmental Social Governance (ESG) scores on the firm’s market performance of polluting companies. Moreover, the study also finds out the moderating effect of green initiatives. The study’s population consisted of 67 companies that were chosen from the list of polluting companies given by the Central Pollution Control Board of India for the post-COVID-19 timeframe of 2020–2023. The results indicate that the performance of ESG will improve the financial performance of the company.
Practical Implementation:
The analysis showed that companies with higher ESG scores generally perform better in the market. This means that firms that are more responsible in terms of environmental, social, and governance practices tend to do well financially. However, the study found that green initiatives did not have a significant impact on this relationship.
These findings are important for company managers and stakeholders. Understanding the connection between ESG practices and market performance can help managers create strategies to improve their ESG scores, ultimately boosting their financial performance.
Future Research Plans:
i) Focus on the R&D investment and sustainability.
ii) Studying the relationship between green finance and sustainability
iii) Exploring the relationship of CSR in sustainability
- Published in Commerce Current Happenings, Departmental News, News, Research News