Eleni Poupaki
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Eleni Poupaki has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Adriaan J. M. Mackus
Department of Applied Physics and Science Education, Eindhoven University of Technology 1 , P.O. Box 513, 5600 MB Eindhoven
4 shared publications - Alex Watkins
University of Warwick
4 shared publications - Bora Karasulu
University of Warwick
4 shared publications - Jennifer D’ Souza
TIB Leibniz Information Centre for Science and Technology, Hannover, Germany
4 shared publications - Sören Auer
TIB Leibniz Information Centre for Science and Technology
4 shared publications Erwin Kessels
Department of Applied Physics and Science Education, Eindhoven University of Technology 2 , Eindhoven 5600 MB
3 shared publications
Latest Publications
- Open Access
- 27 October 2026
- Published by American Chemical Society (ACS) in Chemistry of Materials
The AtomicLimits Database: Towards Artificial Intelligence as a Research Assistant for Atomic Layer Deposition and Etching
The AtomicLimits ALD/ALE database is a community resource that catalogues atomic layer deposition (ALD) and atomic layer etching (ALE) processes reported in the literature. While widely used, it currently captures only a small fraction of the available information, with limited process information per entry, and relies on manual curation. As ALD/ALE process development grows in demand and complexity, with many interdependent parameters and increasingly sophisticated chemistries such as for multicomponent films, decades of experimental and simulation knowledge remain locked across publications, supplementary files, and lab notebooks. Recent advances in artificial intelligence (AI), particularly machine learning (ML) and large language models (LLMs), offer a path to unlock this knowledge for data-driven analysis and process design, but realizing this potential depends on structured, machine-readable data. In this perspective, we propose a framework to evolve AtomicLimits into an AI-ready knowledge infrastructure that integrates literature-extracted, experimental, and simulation data through a community-defined schema, a structured template that specifies which data fields to capture and how to organize them. We outline the AI capabilities this enables, from structured querying to autonomous experimentation, and discuss the coordinated community effort required to realize this vision for the atomic-scale processing field.
- 27 October 2026
- Published by American Chemical Society (ACS) in Chemistry of Materials
Open AccessThe AtomicLimits Database: Towards Artificial Intelligence as a Research Assistant for Atomic Layer Deposition and Etching
The AtomicLimits ALD/ALE database is a community resource that catalogues atomic layer deposition (ALD) and atomic layer etching (ALE) processes reported in the literature. While widely used, it currently captures only a small fraction of the available information, with limited process information per entry, and relies on manual curation. As ALD/ALE process development grows in demand and complexity, with many interdependent parameters and increasingly sophisticated chemistries such as for multicomponent films, decades of experimental and simulation knowledge remain locked across publications, supplementary files, and lab notebooks. Recent advances in artificial intelligence (AI), particularly machine learning (ML) and large language models (LLMs), offer a path to unlock this knowledge for data-driven analysis and process design, but realizing this potential depends on structured, machine-readable data. In this perspective, we propose a framework to evolve AtomicLimits into an AI-ready knowledge infrastructure that integrates literature-extracted, experimental, and simulation data through a community-defined schema, a structured template that specifies which data fields to capture and how to organize them. We outline the AI capabilities this enables, from structured querying to autonomous experimentation, and discuss the coordinated community effort required to realize this vision for the atomic-scale processing field.
- Open Access
- 30 April 2026
- Published by American Vacuum Society in Journal of Vacuum Science & Technology A
- Vol. 44 (3)
Designing an agentic AI workflow for structured information extraction from scientific text: A case study on atomic layer deposition of ZnO and IGZO
The rapid growth of scientific literature in materials sciences and other domains presents a growing challenge where critical process knowledge remains embedded in unstructured text, limiting its reuse for data-driven discovery. Agentic workflows, which integrate large language models (LLMs) with different tools, such as application programming interfaces (APIs), and rule-based functions, have emerged as a promising paradigm to overcome these limitations. In this work, we introduce SciKGExtract, an agentic artificial intelligence (AI) workflow for structured knowledge extraction from scientific publications. The workflow combines LLM-based contextual understanding with deterministic tools for schema validation, data cleaning, and normalization, leveraging authoritative external repositories such as PubChem to ensure accuracy and semantic consistency. Applied to atomic layer deposition (ALD) literature from the AtomicLimits database, SciKGExtract produces a structured dataset capturing experimental parameters and material properties across test cases of zinc oxide (ZnO) and indium-gallium-zinc oxide (IGZO). The extracted data reveal dominant ALD methods and chemistries, variability in reported growth-per-cycle and temperature ranges, and the need for richer metadata to enable reliable cross-study comparisons. These results demonstrate how structured extraction can transform unstructured literature into AI-ready knowledge, accelerating validation, benchmarking, and discovery in emerging materials systems.
- 30 April 2026
- Published by American Vacuum Society in Journal of Vacuum Science & Technology A
- Vol. 44 (3)
Open AccessDesigning an agentic AI workflow for structured information extraction from scientific text: A case study on atomic layer deposition of ZnO and IGZO
The rapid growth of scientific literature in materials sciences and other domains presents a growing challenge where critical process knowledge remains embedded in unstructured text, limiting its reuse for data-driven discovery. Agentic workflows, which integrate large language models (LLMs) with different tools, such as application programming interfaces (APIs), and rule-based functions, have emerged as a promising paradigm to overcome these limitations. In this work, we introduce SciKGExtract, an agentic artificial intelligence (AI) workflow for structured knowledge extraction from scientific publications. The workflow combines LLM-based contextual understanding with deterministic tools for schema validation, data cleaning, and normalization, leveraging authoritative external repositories such as PubChem to ensure accuracy and semantic consistency. Applied to atomic layer deposition (ALD) literature from the AtomicLimits database, SciKGExtract produces a structured dataset capturing experimental parameters and material properties across test cases of zinc oxide (ZnO) and indium-gallium-zinc oxide (IGZO). The extracted data reveal dominant ALD methods and chemistries, variability in reported growth-per-cycle and temperature ranges, and the need for richer metadata to enable reliable cross-study comparisons. These results demonstrate how structured extraction can transform unstructured literature into AI-ready knowledge, accelerating validation, benchmarking, and discovery in emerging materials systems.
- Open Access
- 25 April 2026
- Published by SAGE Publications in Semantic Web: – Interoperability, Usability, Applicability
- Vol. 17 (3)
schema-miner pro: Agentic AI for Ontology Grounding Over LLM-Discovered Scientific Schemas in a Human-in-the-Loop Workflow
Scientific processes are often described in free text, making it difficult to represent and reason over them computationally. We present schema-miner p r o , a human-in-the-loop framework that automatically extracts and grounds structured schemas from scientific literature. Our approach combines large language models for schema extraction with an agent-based system that aligns extracted elements to external ontologies through interpretable, multi-step reasoning. The agent leverages lexical heuristics, semantic similarity, and expert feedback to ensure accurate grounding. We demonstrate the framework on two semiconductor manufacturing workflows—atomic layer deposition and atomic layer etching—mapping process parameters and outputs to the QUDT (Quantities, Units, Dimensions, and Types) ontology. By producing ontology-aligned, semantically precise schemas, schema-miner p r o lays the groundwork for machine-actionable scientific knowledge and automated reasoning across disciplines.
- 25 April 2026
- Published by SAGE Publications in Semantic Web: – Interoperability, Usability, Applicability
- Vol. 17 (3)
Open Accessschema-miner pro: Agentic AI for Ontology Grounding Over LLM-Discovered Scientific Schemas in a Human-in-the-Loop Workflow
Scientific processes are often described in free text, making it difficult to represent and reason over them computationally. We present schema-miner p r o , a human-in-the-loop framework that automatically extracts and grounds structured schemas from scientific literature. Our approach combines large language models for schema extraction with an agent-based system that aligns extracted elements to external ontologies through interpretable, multi-step reasoning. The agent leverages lexical heuristics, semantic similarity, and expert feedback to ensure accurate grounding. We demonstrate the framework on two semiconductor manufacturing workflows—atomic layer deposition and atomic layer etching—mapping process parameters and outputs to the QUDT (Quantities, Units, Dimensions, and Types) ontology. By producing ontology-aligned, semantically precise schemas, schema-miner p r o lays the groundwork for machine-actionable scientific knowledge and automated reasoning across disciplines.
- Open Access
- 16 April 2026
- Published by American Vacuum Society in Journal of Vacuum Science & Technology A
- Vol. 44 (3)
Publishing FAIR and machine-actionable reviews in materials science: The case for symbolic knowledge in neuro-symbolic artificial intelligence
Scientific reviews are central to knowledge integration in materials science, yet their key insights remain locked in narrative text and static PDF tables, limiting reuse by humans and machines alike. This article presents a case study in atomic layer deposition and etching (ALD/E) where we publish review tables as FAIR, machine-actionable comparisons in the Open Research Knowledge Graph (ORKG), turning them into structured, queryable knowledge. Building on this, we contrast symbolic querying over ORKG with large language model-based querying and argue that a curated symbolic layer should remain the backbone of reliable neurosymbolic AI in materials science, with LLMs serving as complementary, symbolically grounded interfaces rather than standalone sources of truth.
- 16 April 2026
- Published by American Vacuum Society in Journal of Vacuum Science & Technology A
- Vol. 44 (3)
Open AccessPublishing FAIR and machine-actionable reviews in materials science: The case for symbolic knowledge in neuro-symbolic artificial intelligence
Scientific reviews are central to knowledge integration in materials science, yet their key insights remain locked in narrative text and static PDF tables, limiting reuse by humans and machines alike. This article presents a case study in atomic layer deposition and etching (ALD/E) where we publish review tables as FAIR, machine-actionable comparisons in the Open Research Knowledge Graph (ORKG), turning them into structured, queryable knowledge. Building on this, we contrast symbolic querying over ORKG with large language model-based querying and argue that a curated symbolic layer should remain the backbone of reliable neurosymbolic AI in materials science, with LLMs serving as complementary, symbolically grounded interfaces rather than standalone sources of truth.
- Open Access
A Process Systems Engineering Approach to Model and Optimize Cr6+-Free and Pd-Free Plating on Plastics Technologies
Plating on Plastics (PoP) requires specific surface pre-treatment steps to enable metallization. The conventional PoP industry utilizes hexavalent chromium (toxic, carcinogenic) and palladium (critical raw material) for surface etching and activation, respectively, raising significant health, environmental, and economic concerns. This work is based on a new Cr6+-free and Pd-free PoP technology that uses piranha (H2O2-H2SO4) solutions for surface etching, nickel salts for activation, and NaBH4 for reduction, ultimately forming metallic nucleation sites for downstream electroless plating and electroplating. A comprehensive modeling approach was developed to simulate and predict unit operation performance (reaction kinetics and yields) and material properties (contact angle and adhesion) across processing stages of the new technology. State-of-the-art and data-driven modeling revealed the combinatorial relationships among process performance, the achieved properties and the different settings of process operating conditions. The results also highlighted capabilities for tuning all processes over a range of conditions, reaching desired product specifications (adhesion and thickness). The models were constructed as a Decision Support Tool (DST) serving economic, environmental, safety and Safe and Sustainable by Design (SSbD) objectives. The DST can be used through a user-friendly interface that enables the insertion of user-defined inputs and monitoring of optimization results.
Open AccessA Process Systems Engineering Approach to Model and Optimize Cr6+-Free and Pd-Free Plating on Plastics Technologies
Plating on Plastics (PoP) requires specific surface pre-treatment steps to enable metallization. The conventional PoP industry utilizes hexavalent chromium (toxic, carcinogenic) and palladium (critical raw material) for surface etching and activation, respectively, raising significant health, environmental, and economic concerns. This work is based on a new Cr6+-free and Pd-free PoP technology that uses piranha (H2O2-H2SO4) solutions for surface etching, nickel salts for activation, and NaBH4 for reduction, ultimately forming metallic nucleation sites for downstream electroless plating and electroplating. A comprehensive modeling approach was developed to simulate and predict unit operation performance (reaction kinetics and yields) and material properties (contact angle and adhesion) across processing stages of the new technology. State-of-the-art and data-driven modeling revealed the combinatorial relationships among process performance, the achieved properties and the different settings of process operating conditions. The results also highlighted capabilities for tuning all processes over a range of conditions, reaching desired product specifications (adhesion and thickness). The models were constructed as a Decision Support Tool (DST) serving economic, environmental, safety and Safe and Sustainable by Design (SSbD) objectives. The DST can be used through a user-friendly interface that enables the insertion of user-defined inputs and monitoring of optimization results.
- 10 November 2023
- Published by American Chemical Society (ACS) in ACS Applied Polymer Materials
- Vol. 5 (12) , 10205-10216
Single-Step PEDOT Deposition by oCVD for ITO-Free Deep Blue OLEDs
- 10 November 2023
- Published by American Chemical Society (ACS) in ACS Applied Polymer Materials
- Vol. 5 (12) , 10205-10216
Single-Step PEDOT Deposition by oCVD for ITO-Free Deep Blue OLEDs