| Authors | Е.А. Lysenkov1, O.V. Kozlov1, R.V. Dinzhos1 , S.S. Ryzhkov2, F. Korkees3 |
| Affiliations |
1Petro Mohyla Black Sea National University, 54003 Mykolaiv, Ukraine 2Limited Liability Company "International Academy of Marine Sciences, Technologies and Innovations", 54025 Mykolaiv, Ukraine 3Materials Science and Engineering Department, Faculty of Science and Engineering, Swansea University, SA1 8EN Swansea, United Kingdom |
| Е-mail | ealysenkov@ukr.net |
| Issue | Volume 18, Year 2026, Number 4 |
| Dates | Received 15 April 2026; revised manuscript received 20 August 2026; published online 21 August 2026 |
| Citation | Е.А. Lysenkov, O.V. Kozlov, et al., J. Nano- Electron. Phys. 18 No 4, 04022 (2026) |
| DOI | https://doi.org/10.21272/jnep.18(4).04022 |
| PACS Number(s) | 72.80.Tm, 64.60.ah, 72.20. – i, 81.05.Qk |
| Keywords | Polylactic acid (2) , Polymer nanocomposites, Carbon nanotubes (14) , Electrical conductivity (10) , Artificial Intelligence, Fuzzy Logic Models. |
| Annotation |
This study investigates the electrical conductivity of polymer nanocomposites filled with carbon nanotubes, taking into account the influence of both nanotube concentration and diameter. The experimental results reveal a pronounced nonlinear dependence of conductivity on filler content, characterized by a transition from a tunneling-dominated charge transport mechanism at low concentrations to a percolation-controlled regime after the formation of a continuous conductive network. It is shown that the nanotube diameter significantly affects the conductivity over the entire concentration range, particularly in the pre-percolation region, which can be attributed to variations in interparticle contact efficiency and tunneling conditions. A comparative analysis of the experimental data was performed using classical analytical models, including percolation-based approaches and the McLachlan generalized effective medium model. It was found that these models can adequately describe individual conductivity curves; however, their fitting parameters vary substantially with nanotube diameter and often lose physical interpretability. This limitation restricts their applicability for the unified description and prediction of electrical behavior in systems with varying geometrical characteristics of the filler. To overcome these limitations, an intelligent model of electrical conductivity based on fuzzy logic has been developed, which allows simultaneously taking into account the concentration and diameter of nanotubes without the need to specify a rigid analytical dependence. The proposed modeling approach provides high accuracy of reproduction of experimental data and forms a generalized description of the dependence of electrical conductivity on system parameters. The results obtained confirm the effectiveness of the application of artificial intelligence methods for modeling and predicting the properties of complex polymer nanocomposites under conditions of limited experimental data. |
|
List of References |