Mustafa Ünal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mustafa Ünal has a total of 64 co-authors. Here are the top co-authors by number of shared publications:
Raşit Turan
Graduate School of Micro and Nanotechnology Middle East Technical University Ankara Turkey
10 shared publicationsAydın Tankut
Middle East Tech Univ, Ctr Solar Energy Res & Applicat GUNAM, TR-06800 Ankara, Turkey
5 shared publicationsAyşe Merve Genç
Crystal Growth Laboratory Middle East Technical University Ankara Turkey
3 shared publicationsMehmet Can Karaman
Crystal Growth Laboratory Middle East Technical University Ankara Turkey
3 shared publications- Mercouri G. Kanatzidis
Northwestern University
3 shared publications Özden Başar Balbaşı
Crystal Growth Laboratory Middle East Technical University Ankara Turkey
3 shared publications
Latest Publications
- Open Access
- 1 June 2026
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 7 (3) , 030503
The ballad of LLM agents: philosophical reasoning for chemistry
Large language models (LLMs) show remarkable potential for scientific reasoning but often produce unreliable or scientifically unactionable outputs when faced with multi-step logic, domain grounding, and interpretability challenges, especially in complex fields like chemistry and materials science. Here, we introduce a framework of philosophical reasoning agents, inspired by canonical thinkers such as Socrates, Descartes, Kant, and Hume, to guide LLM behavior via structured prompt engineering. These agents embody distinct reasoning paradigms (dialectical inquiry, deductive logic, rule-based judgment, and empirical validation) and are evaluated across multiple chemistry subdomains, physical, analytical, general, inorganic, and organic chemistry, using the ChemBench benchmark. Our agentic prompting approach yields substantial accuracy gains on open-ended numerical chemistry questions, with gains of +11.5 percentage points for GPT-4o with Hume, +4.5 percentage points for GPT-5 with Kant, and +21.8 percentage points for GPT-5.1 with Socrates at the strict 1% error threshold, relative to the corresponding base models. Beyond accuracy, we observe benchmark-level model–agent performance patterns, suggesting that different prompting styles interact differently with each base model. These findings demonstrate that embedding philosophy-of-science principles into multi-agent frameworks can improve and produce interpretable, adaptive, and domain-aligned scientific LLMs.
- 1 June 2026
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 7 (3) , 030503
Open AccessThe ballad of LLM agents: philosophical reasoning for chemistry
Large language models (LLMs) show remarkable potential for scientific reasoning but often produce unreliable or scientifically unactionable outputs when faced with multi-step logic, domain grounding, and interpretability challenges, especially in complex fields like chemistry and materials science. Here, we introduce a framework of philosophical reasoning agents, inspired by canonical thinkers such as Socrates, Descartes, Kant, and Hume, to guide LLM behavior via structured prompt engineering. These agents embody distinct reasoning paradigms (dialectical inquiry, deductive logic, rule-based judgment, and empirical validation) and are evaluated across multiple chemistry subdomains, physical, analytical, general, inorganic, and organic chemistry, using the ChemBench benchmark. Our agentic prompting approach yields substantial accuracy gains on open-ended numerical chemistry questions, with gains of +11.5 percentage points for GPT-4o with Hume, +4.5 percentage points for GPT-5 with Kant, and +21.8 percentage points for GPT-5.1 with Socrates at the strict 1% error threshold, relative to the corresponding base models. Beyond accuracy, we observe benchmark-level model–agent performance patterns, suggesting that different prompting styles interact differently with each base model. These findings demonstrate that embedding philosophy-of-science principles into multi-agent frameworks can improve and produce interpretable, adaptive, and domain-aligned scientific LLMs.
- 29 April 2026
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 18 (18) , 26462-26474
Mitigating Electrochemical Degradation in CsPbBr 3 Gamma Detectors by Organic and Inorganic Encapsulation
CsPbBr3 perovskite semiconductors have emerged as a leading candidate for next-generation radiation detectors because of their exceptional charge transport properties, defect tolerance, and record-breaking sensitivity and energy resolution. Their long-term stability, however, is hindered by electrode-driven electrochemical decomposition, which is accelerated by moisture- and oxygen-assisted ion migration during operation. Here, we investigated organic and inorganic encapsulation strategies as both environmental barriers and means to suppress interfacial degradation pathways. Atomic layer deposition of Al2O3 provided a conformal passivation layer that blocked environmental ingress, suppressed ionic diffusion, reduced leakage current, enhanced energy resolution, and expanded the operational electric-field window beyond 5 kV cm–1. By contrast, organic encapsulants such as paraffin wax and polystyrene slowed moisture diffusion but did not suppress interfacial reactions, with wax extending stability to over 90 days. These results show that ALD-Al2O3 suppresses dominant interfacial degradation pathways, enabling stable, high-field operation and advancing the practical deployment of CsPbBr3 gamma-ray detectors.
- 29 April 2026
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 18 (18) , 26462-26474
Mitigating Electrochemical Degradation in CsPbBr 3 Gamma Detectors by Organic and Inorganic Encapsulation
CsPbBr3 perovskite semiconductors have emerged as a leading candidate for next-generation radiation detectors because of their exceptional charge transport properties, defect tolerance, and record-breaking sensitivity and energy resolution. Their long-term stability, however, is hindered by electrode-driven electrochemical decomposition, which is accelerated by moisture- and oxygen-assisted ion migration during operation. Here, we investigated organic and inorganic encapsulation strategies as both environmental barriers and means to suppress interfacial degradation pathways. Atomic layer deposition of Al2O3 provided a conformal passivation layer that blocked environmental ingress, suppressed ionic diffusion, reduced leakage current, enhanced energy resolution, and expanded the operational electric-field window beyond 5 kV cm–1. By contrast, organic encapsulants such as paraffin wax and polystyrene slowed moisture diffusion but did not suppress interfacial reactions, with wax extending stability to over 90 days. These results show that ALD-Al2O3 suppresses dominant interfacial degradation pathways, enabling stable, high-field operation and advancing the practical deployment of CsPbBr3 gamma-ray detectors.
- Open Access
- 24 December 2025
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 6 (4) , 045073
Towards philosophical reasoning with agentic LLMs: Socratic method for scientific assistance
As large language models (LLMs) become central tools in science, improving their reasoning capabilities is critical for meaningful and trustworthy applications. We introduce a Socratic agent for scientific reasoning, implemented through a structured system prompt that guides LLMs via classical principles of inquiry. Unlike typical prompt engineering or retrieval-based methods, our approach leverages definition, analogy, hypothesis elimination, and other Socratic techniques to generate more coherent, critical, and domain-aware responses. We evaluate the agent across diverse scientific domains and benchmark it on the abstraction and reasoning corpus challenge dataset, achieving 97.15% under a fixed prompting protocol and without fine-tuning or external tools. Expert evaluation shows improved reasoning depth, clarity, and adaptability over conventional LLM outputs, suggesting that structured prompting rooted in philosophical reasoning can improve the scientific utility of language models.
- 24 December 2025
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 6 (4) , 045073
Open AccessTowards philosophical reasoning with agentic LLMs: Socratic method for scientific assistance
As large language models (LLMs) become central tools in science, improving their reasoning capabilities is critical for meaningful and trustworthy applications. We introduce a Socratic agent for scientific reasoning, implemented through a structured system prompt that guides LLMs via classical principles of inquiry. Unlike typical prompt engineering or retrieval-based methods, our approach leverages definition, analogy, hypothesis elimination, and other Socratic techniques to generate more coherent, critical, and domain-aware responses. We evaluate the agent across diverse scientific domains and benchmark it on the abstraction and reasoning corpus challenge dataset, achieving 97.15% under a fixed prompting protocol and without fine-tuning or external tools. Expert evaluation shows improved reasoning depth, clarity, and adaptability over conventional LLM outputs, suggesting that structured prompting rooted in philosophical reasoning can improve the scientific utility of language models.
- 23 October 2025
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 7 (21) , 9838-9849
Assessment of Blocking Contacts in Suppressing Polarization Effects in CsPbBr 3 Perovskite Detectors
- 23 October 2025
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 7 (21) , 9838-9849
Assessment of Blocking Contacts in Suppressing Polarization Effects in CsPbBr 3 Perovskite Detectors
- 12 June 2025
- Published by American Chemical Society (ACS) in ACS Photonics
- Vol. 12 (7) , 3574-3580
High-Performance Electron Transport in CsPbBr3 Perovskite Crystals for Room-Temperature Radiation Detection
- 12 June 2025
- Published by American Chemical Society (ACS) in ACS Photonics
- Vol. 12 (7) , 3574-3580
High-Performance Electron Transport in CsPbBr3 Perovskite Crystals for Room-Temperature Radiation Detection
- 4 June 2024
- Published by Wiley in Surface and Interface Analysis
- Vol. 56 (10) , 713-719
Surface and interface chemistry of bromine–methanol‐etched Cd 0.9 Zn 0.1 Te crystals
Chemical polishing or chemo‐mechanical polishing is crucial as a last step of surface preparation to remove the damaged layer and contaminants from the surface of CdZnTe crystals. The bromine–methanol solution is widely used for this purpose. However, bromine–methanol solution enriches the surface with Te and results in poor performance of CdZnTe crystals. In this study, the effect of the chemical polishing with 5% bromine–methanol solution on the surface and at the interface is investigated and it is demonstrated that etching duration strongly influences surface stoichiometry and interface contaminants. The evolution of the surface topography with etching and chemical changes are presented. It is shown that after 90 s etching/polishing, subsurface damage is removed and Te enrichment is minimum. Moreover, interface layer thickness is the smallest for 90 s etching duration. It is presented that further increase in the etching duration disturbs the surface stoichiometry and interface depth. It also calculated that 90 s of etching shows low interface barrier and symmetrical current–voltage curve.
- 4 June 2024
- Published by Wiley in Surface and Interface Analysis
- Vol. 56 (10) , 713-719
Surface and interface chemistry of bromine–methanol‐etched Cd 0.9 Zn 0.1 Te crystals
Chemical polishing or chemo‐mechanical polishing is crucial as a last step of surface preparation to remove the damaged layer and contaminants from the surface of CdZnTe crystals. The bromine–methanol solution is widely used for this purpose. However, bromine–methanol solution enriches the surface with Te and results in poor performance of CdZnTe crystals. In this study, the effect of the chemical polishing with 5% bromine–methanol solution on the surface and at the interface is investigated and it is demonstrated that etching duration strongly influences surface stoichiometry and interface contaminants. The evolution of the surface topography with etching and chemical changes are presented. It is shown that after 90 s etching/polishing, subsurface damage is removed and Te enrichment is minimum. Moreover, interface layer thickness is the smallest for 90 s etching duration. It is presented that further increase in the etching duration disturbs the surface stoichiometry and interface depth. It also calculated that 90 s of etching shows low interface barrier and symmetrical current–voltage curve.
- 2 May 2023
- Published by Elsevier in Journal of Crystal Growth
- Vol. 615
Production of detector grade CdZnTe crystal with VGF furnace by analyzing segregation of Zn and In
- 2 May 2023
- Published by Elsevier in Journal of Crystal Growth
- Vol. 615
Production of detector grade CdZnTe crystal with VGF furnace by analyzing segregation of Zn and In
- 7 May 2022
- Published by Springer Nature in Journal of Electronic Materials
- Vol. 51 (9) , 4675-4680
Production of High-Performance CdZnTe Crystals Grown by THM for Radiation Detection Applications
- 7 May 2022
- Published by Springer Nature in Journal of Electronic Materials
- Vol. 51 (9) , 4675-4680
Production of High-Performance CdZnTe Crystals Grown by THM for Radiation Detection Applications
- 15 December 2020
- Published by American Physical Society (APS) in Physical Review A
- Vol. 102 (6) , 063521
Implementation of generalized Harvey-Shack theory in light scattering from rough surfaces
We present a discrete implementation of generalized Harvey-Shack scalar scattering theory to calculate angular intensity distributions from height profiles of select randomly textured surfaces proposed for use in solar cells and covering a wide range of surface characteristics. We compare these calculations to high-resolution angular intensity distribution measurements. These comparisons suggest that the pupil function does benefit from an additional correction factor for rough surfaces containing lateral feature sizes on the order of the wavelength, which can be attributed to effective medium effects. Moreover, secondary interactions within the surface topography are shown to be a mechanism that partly redistributes scattered power, affecting angular intensity distribution results. These mechanisms emerge as the two main limitations of the Harvey-Shack scalar scattering theory in the far field, which nonetheless produces remarkable quantitative predictions.
- 15 December 2020
- Published by American Physical Society (APS) in Physical Review A
- Vol. 102 (6) , 063521
Implementation of generalized Harvey-Shack theory in light scattering from rough surfaces
We present a discrete implementation of generalized Harvey-Shack scalar scattering theory to calculate angular intensity distributions from height profiles of select randomly textured surfaces proposed for use in solar cells and covering a wide range of surface characteristics. We compare these calculations to high-resolution angular intensity distribution measurements. These comparisons suggest that the pupil function does benefit from an additional correction factor for rough surfaces containing lateral feature sizes on the order of the wavelength, which can be attributed to effective medium effects. Moreover, secondary interactions within the surface topography are shown to be a mechanism that partly redistributes scattered power, affecting angular intensity distribution results. These mechanisms emerge as the two main limitations of the Harvey-Shack scalar scattering theory in the far field, which nonetheless produces remarkable quantitative predictions.