Amasya University

Artificial intelligence (AI) has become one of the main driving forces of transformation in the financial sector, while simultaneously generating significant implications for environmental sustainability and sustainable development processes. This study analyzes the effects of AI investments in the financial sector on CO2 and total greenhouse gas (GHG) emissions, as well as on Sustainable Development Goal 7 (SDG 7) performance. The analysis is conducted using data from 13 countries with sufficient data availability over the period 2014-2023, compiled from the OECD AI Policy Observatory, the World Bank's World Development Indicators (WDI), and the Sustainable Development Report. The empirical analysis employs Driscoll-Kraay standard errors and the method of moments quantile regression (MMQR) approach. In addition, the robustness of the findings against potential endogeneity is tested using the two-stage least squares (2SLS) method. The results indicate that AI investments in the financial sector have a statistically significant and negative effect on CO2 and GHG emissions, with this effect being more pronounced in countries with higher emission levels. In contrast, although the impact of AI investments on SDG 7 performance remains positive across both models, the findings provide only limited empirical support. Overall, the results suggest that AI investments in the financial sector can serve as an important tool for reducing environmental pressures. Accordingly, it is recommended that policymakers design financial digitalization processes in alignment with environmental objectives and integrate AI-based financial applications with sustainability-oriented strategies.

This study examines the socioeconomic determinants of agricultural CO2 emissions in the 10 highest-emitting agricultural economies from 1992 to 2022. Using the LM bootstrap cointegration method and the method of moments quantile regression (MMQR), the analysis identifies long-run relationships and heterogeneous effects across the emissions distribution. The results indicate that agricultural energy use is the most significant driver of CO2 emissions, with its impact intensifying at higher quantiles, where environmental pressures are more severe. Agricultural value added also increases emissions, showing that productivity gains in these economies remain carbon-intensive. Additionally, agricultural trade openness and rural population size contribute positively to emissions, reflecting structural constraints in agricultural production systems. These findings directly relate to several Sustainable Development Goals (SDGs), particularly SDG 2 (zero hunger) through the need for resilient and low-carbon food systems, SDG 7 (affordable and clean energy) via the transition to renewable energy use in agriculture, SDG 12 (responsible consumption and production) through improved resource efficiency, and SDG 13 (climate action) by highlighting the urgency of reducing agricultural emissions. The quantile results indicate the need for stricter renewable energy mandates in high-emission groups, efficiency-focused technology support in middle groups, and targeted measures to ease trade and population-driven pressures in lower groups. The findings indicate that improving energy efficiency in agricultural production and implementing environmentally aligned trade regulations can deliver measurable reductions in sectoral carbon intensity, directly supporting SDG 13 while maintaining productivity gains consistent with SDG 2 objectives.

A novel approach is introduced for detection of pendimethalin (PEN) - one of the most widely utilized herbicides- based on the synergistic properties of molecularly imprinted polymers (MIPs) and a CuO-Bi2MoO6 (Cu-Bi-Mo) nanocomposite. The methodology combined the specific recognition capabilities of MIPs with the enhanced sensing performance provided by the advanced nanocomposite material. The synthesis of the Cu-Bi-Mo nanocomposite was initiated through the application of the sol-gel method. After the glassy carbon electrode’s modification with the Cu-Bi-Mo nanocomposite, PEN imprinted electrodes were fabricated using cyclic voltammetry (CV) with a dispersion containing 100.0 mM pyrrole (Py) monomer and 25.0 mM PEN molecule. The electrochemical sensor revealed a detection range of 1.0 × 10− 9 M to 1.0 × 10− 8 M PEN and achieved a detection limit (LOD) of 3.30 × 10− 10 M. To demonstrate its practical applicability, the electrochemical sensor was applied to drinking water and orange juice samples, yielding recovery results near 100

PurposeThis study aims to optimize the machining parameters on the average surface roughness (Ra), metal removal rate (MRR), and overcut (OC) values of DIN 1.2365 (H10) steel during the Electrical Discharge Machining (EDM).Design/methodology/approachThe cutting parameters were optimized using multi-response method with Taguchi-based grey relational analysis to determine the optimal machining parameters.FindingsDischarge current was identified as the most influential parameter for both Ra and MRR, while pulse duration emerged as the dominant factor affecting OC. Conversely, ANOVA results revealed that discharge current contributed to 80.62 and 82.27% of Ra and MRR, respectively, while pulse duration accounted for 50.16% of the variation in OC. The experimental study achieved confidence levels of approximately 93%, 92%, and 90% for Ra, MRR, and OC, respectively.Practical implicationsThe study provided a predictive analysis of surface quality in EDM machining of DIN 1.2365 steel, enabling the selection of appropriate parameters. Furthermore, by predicting surface quality, it minimized time and material loss, allowing for cost reductions in production.Originality/valueThis study highlights the use of a multiple optimization method to optimize Ra, MRR, and OC, which are key outputs in EDM machining of DIN 1.2365 steel, commonly used in mold production. The literature generally focuses on similar steels, and it appears that DIN 1.2365 steel, with its superior thermal conductivity, has not received sufficient attention. Furthermore, since studies focusing on a single output in EDM machining of this material and multiple optimization studies are limited, this study makes a significant contribution to the field.

This study investigates the geometric dynamics of electromagnetic wave propagation along optical fibers within a non-Newtonian (multiplicative) differential geometry framework. Modeling the optical fiber as a space curve in multiplicative Euclidean 3-space, we construct a specialized anholonomic coordinate system associated with the multiplicative Frenet frame. Within this setup, we reformulate Maxwell’s equations to derive a novel set of Maxwellian curve evolution equations. We rigorously analyze the polarization evolution of the electric and magnetic fields, establishing explicit relationships between the non-Newtonian Berry phase, Rytov parallel transport, and Fermi-Walker transport laws in both the normal ( ν ) and binormal ( β ) directions. Furthermore, we demonstrate that the derived evolution equations satisfy the Mainardi-Gauss-Codazzi-type compatibility conditions. This geometric formulation reveals a deep intrinsic connection between electromagnetic wave propagation and integrable systems, as exemplified by the emergence of the sine-Gordon equation and the Dini surface geometry in the context of constant negative curvature.