2D WS₂/COOH-Modified MWCNTs for Humidity-Tolerant NO₂ Gas Sensing Applications: Machine Learning Modeling, Density Functional Theory Calculations, and Molecular Dynamics | Sciact - CRIS-система Института катализа

2D WS₂/COOH-Modified MWCNTs for Humidity-Tolerant NO₂ Gas Sensing Applications: Machine Learning Modeling, Density Functional Theory Calculations, and Molecular Dynamics Научная публикация

Журнал

Diamond and Related Materials
ISSN: 0925-9635 , E-ISSN: 1879-0062

Вых. Данные

Год: 2025, Том: 159, Номер: Part B, Номер статьи : 112942, Страниц : 15 DOI: 10.1016/j.diamond.2025.112942

Ключевые слова

Nanocomposite; Carbon nanotubes; Sensor; Nitrogen dioxide; DFT; Molecular simulations

Авторы

Khajavian Mohammad ¹ , Ishchenko Arcady V. ² , Bannov Alexander G. ³

Организации

1	Department of Environmental Engineering, College of Ocean Science and Engineering, National Korea Maritime and Ocean University, 727 Taejong-ro, Yeongdo-gu, Busan, 49112, Republic of Korea
2	Boreskov Institute of Catalysis SB RAS, Novosibirsk, 630090, Russia
3	Novosibirsk State Technical University, Novosibirsk, 630073, Russia

Информация о финансировании (2)

1	Министерство науки и высшего образования Российской Федерации (с 15 мая 2018)	075-12-2021-697
2	Министерство науки и высшего образования Российской Федерации (с 15 мая 2018)	FSUN-2023-0008

Multi-walled carbon nanotubes (MWCNTs) have been extensively utilized in gas-sensing applications due to their high surface area and excellent electrical conductivity. However, their hydrophilic nature makes them susceptible to humidity interference, as water molecule adsorption compromises gas detection accuracy. To address this limitation, a hybrid sensing material comprising carboxyl-functionalized MWCNTs (COOH-MWCNTs) and two-dimensional tungsten disulfide (WS₂) was developed to enhance humidity resistance. The gas-sensing performance of pristine MWCNTs, COOH-MWCNTs, and WS₂/COOH-MWCNTs was systematically evaluated using machine learning (ML modeling, density functional theory (DFT), and molecular dynamics (MD) simulations to investigate their interaction mechanisms with NO₂ molecules. The results revealed significantly stronger interactions between NO₂ and the WS₂/COOH-MWCNT structure, evidenced by more than a threefold increase in sensor response (ΔR/R). SHAP sensitivity analysis, based on random forest ML modeling, showed that the WS₂ layer substantially reduced the impact of humidity-related features on sensor performance. DFT calculations further demonstrated that the NO₂/WS₂/COOH-MWCNT complex exhibited a lower energy gap (Eg = 3.12 eV) compared to NO₂/MWCNT (4.02 eV) and NO₂/COOH-MWCNT (3.58 eV), indicating a more stable interaction between NO₂ and the hybrid surface. The integration of ML predictions with DFT and MD insights confirmed the superior NO₂ sensing capability of WS₂/COOH-MWCNTs.

Khajavian M. , Ishchenko A.V. , Bannov A.G.
2D WS₂/COOH-Modified MWCNTs for Humidity-Tolerant NO₂ Gas Sensing Applications: Machine Learning Modeling, Density Functional Theory Calculations, and Molecular Dynamics
Diamond and Related Materials. 2025. V.159. NPart B. 112942 :1-15. DOI: 10.1016/j.diamond.2025.112942 Scopus

Поступила в редакцию:	8 авг. 2025 г.
Принята к публикации:	6 окт. 2025 г.
Опубликована online:	9 окт. 2025 г.
Опубликована в печати:	1 нояб. 2025 г.

Scopus:

2-s2.0-105019252721

Пока нет цитирований