Solvent polarity-dependent charge polarity recoverable polyaniline-carbon nanotube composites

Xu Dai† ^a, Yanqiu Yao†^b, Yizhuo Wang^a, Jing Wang^b, Tiantian Zhuang^a and Hong Wang*^ab
aState Key Laboratory of Multiphase Flow in Power Engineering & Frontier Institute of Science and Technology, Xi'an Jiaotong University, Xi'an, 710054, China
^bSchool of Energy and Power Engineering, Xi'an Jiaotong University, Xi'an, 710054, China. E-mail: hong.wang@xjtu.edu.cn

First published on 31st July 2025

---

---

Introduction

N-type organic materials are important components in modern organic electronics, including light-emitting diodes (OLEDs), photovoltaics (OPVs), and thermoelectric generators (OTEGs).^1–4 However, their development severely lags behind p-type counterparts due to inherent doping challenges. Most organic semiconductors exhibit oxygen-induced p-doping behavior, where even trace oxygen (<10 ppm) can degrade n-type electrical conductivity and operational stability.⁵ This extreme oxygen sensitivity creates fundamental barriers in both understanding charge-transport mechanisms and engineering air-stable high-performance n-type materials, ultimately constraining advances in organic complementary circuits and energy-harvesting devices.

Recent efforts to mitigate oxygen-induced p-doping in n-type organics have prioritized synthesizing air-stable materials, such as Lu et al.'s coplanar poly(p-phenylenevinylene)(PPV) derivatives, which achieved a power factor (PF) of 1.96 μW m⁻¹ K⁻², and the PFs after doping dropped only 2% in the air for 7 days.⁶ Liu et al. developed an n-type conjugated polymer (n-PT5) blended with poly(ethyleneimine) (PEI) with excellent thermal stability.⁷ The n-type conducting polymer poly(benzodifurandione) (PBFDO) with outstanding electrical conductivity over 2000 S cm⁻¹ was fabricated by Huang et al., showing good stability.⁸ However, the intricate and time-consuming nature of organic synthesis^6,8–11 significantly impedes the rapid development of high-performance n-type thermoelectric materials. Compounding this challenge, the structure–property relationships governing charge transport in organic systems remain insufficiently understood,^12,13 limiting rational design strategies.

Alternative strategies to counter oxygen p-doping focus on environmental control. Prolonged vacuum storage enables oxygen desorption, revealing intrinsic n-type behavior in carbon nanotubes (CNTs) with a stable Seebeck coefficient.^14–16 Similarly, controlled soft plasma processing can eliminate trace oxygen, converting p-type organics into n-type systems with enhanced thermoelectric performance.⁵ However, both approaches face critical limitations: vacuum dependency and plasma-induced metastability lead to rapid property degradation under ambient conditions. These energy-intensive methods lack scalability for practical application, underscoring the urgent need for oxygen-immune materials designs that enable both fundamental polarity studies and air-stable n-type organic electronics.

In this work, we demonstrate a solvent-mediated polarity control strategy that fundamentally suppresses oxygen p-doping in air-stable n-type organic thermoelectrics. While polyaniline/single-walled carbon nanotube (PANI/SWCNT) hybrids are conventionally classified as p-type materials,^17–21 we reveal their dual-polarity nature through dopant-free solvent engineering. Intriguingly, solvent processing enables reversible p-to-n polarity switching in PANI/SWCNT films without chemical dopants. Theoretical modeling revealed a direct correlation between charge polarity and solvent polarity (E_T parameter), achieving a high n-type power factor of 911.3 μW m⁻¹ K⁻² through low-E_T solvent treatment. The solvent-processed n-type films demonstrated exceptional air stability, retaining 73% electrical conductivity and 96% Seebeck coefficient after 4-day ambient exposure. In addition, cyclic polarity reversibility is demonstrated through solvent immersion, highlighting the system's robustness. The solvent-stabilized system shows promise for electricity generation via integrated thermoelectric-water evaporation platforms²² and smart heat management via integrated thermoelectric-heat pipe platforms.²³ This oxygen-immune polarity modulation mechanism establishes a general platform for both probing intrinsic material properties and engineering air-stable n-type organic materials with industrial-scale viability.

Results and discussion

The PANI/SWCNT hybrid films were fabricated through vacuum filtration of composite suspensions, as illustrated in Fig. 1a. In a typical preparation, PANI and SWCNT powders were mixed at various weight ratios and dispersed in dimethyl sulfoxide (DMSO) using a probe-type sonicator for 2 h. The resulting suspension was subsequently vacuum-filtered and thoroughly rinsed with DMSO. After drying under vacuum at 60 °C, the freestanding PANI/SWCNT films were carefully peeled from the filtration membrane for characterization. Specifically, the PANI/SWCNT₁₀ composite containing 10 wt% PANI was selected for detailed analysis.

Raman spectroscopic characterization (Fig. 1b) confirmed the successful integration of both components in the hybrid film. The PANI reference spectrum exhibited characteristic peaks at 1165 cm⁻¹ and 1217 cm⁻¹, corresponding to the C–H bending vibrations of the quinoid/benzenoid ring and the weak C–N stretching.^24,25 SWCNT powder showed its typical disorder-induced mode, D band at 1346 cm⁻¹.^26,27 The PANI/SWCNT₁₀ hybrid film displayed distinct signatures from both constituents, with PANI-specific vibrations at 1160 cm⁻¹ and 1218 cm⁻¹ and SWCNT's D peak at 1346 cm⁻¹, confirming effective composite formation. The enlarged image is shown in Fig. S1. Due to the strong signal of CNT and low content of PANI, the peaks of PANI are relatively weak.

The solvent-mediated modulation of thermoelectric properties in PANI/SWCNT₁₀ hybrid films reveals a distinct polarity-dependent switching behavior. As shown in Fig. 1c, films treated with solvents of varying polarity (DMSO, 1-methyl-2-pyrrolidinone (NMP), dimethylformamide (DMF), ethanol, and H₂O) exhibited reversible carrier-type transitions in air, which also show similar performance in N₂ (Fig. S2), because the solvents have poor oxygen content,^28,29 these solvents act as a protective barrier, avoiding the p-type doping of oxygen and showing the original performance of PANI/SWCNT₁₀ composite. Low-polarity solvents (DMSO, NMP, or DMF; E_T < 45 kcal mol⁻¹)^30–32 induced negative Seebeck coefficients (−45.7 to −57.9 μV K⁻¹), demonstrating stable n-type behavior (Fig. 1d), while high-polarity solvents (ethanol, H₂O; E_T > 45 kcal mol⁻¹)^30–32 restored positive Seebeck coefficients (47.5 to 50.8 μV K⁻¹) characteristic of p-type behavior. This polarity-driven transition contrasts with conventional PANI/SWCNT composites, which typically maintain p-type characteristics under ambient conditions due to oxygen doping effects.^{18,19,33–36} The critical role of solvent polarity is quantitatively shown in Fig. 1d, where the lower polarity solvents (<45 kcal mol⁻¹) promote electron transfer between PANI and CNTs, resulting in n-type properties. This E_T dependence charge polarity result demonstrates solvent polarity as a primary regulator of the electron transfer between PANI and CNTs in the PANI/SWCNT₁₀ hybrid films. Besides, the five solvents could also be categorized into two groups by the presence/absence of protons (large polarity solvents with protons (EtOH, H₂O), small polarity solvents without protons (DMSO, NMP, DMF)). Therefore, to rigorously test the hypothesis, acetonitrile (AN), a high-polarity solvent (E_T = 46 kcal mol⁻¹)³⁰ devoid of protons, was used to immerse the PANI/SWCNT₁₀ films, yielding a positive Seebeck coefficient of 51.0 ± 5 μV K⁻¹, confirming p-type behavior. This result conclusively excludes the presence of protons as the determining factor for p-type switching. When further expanding the solvent polarity range, the oxygen solubility limitation of low-polarity solvents and solvent-SWCNT interactions also affect the Seebeck coefficient, as shown in Fig. S3.

Notably, all solvent-treated films maintained high electrical conductivities of over 600 S cm⁻¹ (Fig. S4), comparable to or exceeding previously reported values for PANI/SWCNT systems.^33,37–39 Film thicknesses were determined through the scanning electron microscopy (SEM) images of the cross-section of the dry films, as shown in Fig. S5.

Theoretical calculation was performed to understand how the solvents affect the charge polarity in the wetted PANI/SWCNT₁₀ hybrid films. All the density functional theory (DFT) calculations were performed using the CASTEP module, a part of the quantum mechanical software suite Material Studio. Detailed information was provided in the SI.

Band structure and density of states (DOS) analysis reveal solvent polarity-dependent electronic modification in PANI/SWCNT₁₀ hybrid films (Fig. 2a and b). The PANI/SWCNT₁₀ hybrid films are n-type at vacuum conditions where PANI transfers electrons to SWCNT, acting as an n-type dopant. When wetted with solvents of relatively low polarity (E_T < 45 kcal mol⁻¹), such as DMSO, NMP, and DMF, the DOS near the Fermi level exhibits characteristic n-type semiconductor behavior. The Fermi level shifts toward the conduction band minimum (CBM), accompanied by a pronounced electron density of states near the CBM. This electronic configuration suggests abundant free electrons as charge carriers, consistent with the experimentally observed negative Seebeck coefficients indicating n-type doping. In contrast, treatment with high-polarity solvents (E_T > 45 kcal mol⁻¹), such as EtOH and H₂O, induces a p-type electronic structure. The DOS calculations show a significant increase in hole density below the Fermi level, where electrons preferentially occupy higher energy states, leaving abundant holes in the valence band. This aligns with the measured positive Seebeck coefficient, confirming p-type behavior. The polarity-driven doping mechanism involves solvent-PANI interaction: high-polarity solvents (e.g., EtOH, H₂O) act as p-type dopants by modifying PANI's electronic structure. Despite weaker interfacial interactions compared to low-polarity solvents, these polar solvents enable electron withdrawal from SWCNT through dipole-induced charge redistribution. This process generates hole carriers within the hybrid system, facilitating hole-dominated conduction and ultimately stabilizing p-type thermoelectric performance.

Charge density difference calculations further corroborate these conclusions by providing direct visualization of solvent-modulated charge redistribution (Fig. 2c). For the PANI/SWCNT₁₀ hybrid films under vacuum conditions, the electrons accumulate at the PANI/SWCNT interface, as shown in the differential charge density map. When the PANI/SWCNT₁₀ hybrid films are wetted with low-polarity solvents (E_T < 45 kcal mol⁻¹) to induce n-type behavior, the differential charge density maps also reveal electron accumulation at the PANI/SWCNT interface, signifying electron transfer from PANI (acting as an n-type dopant) to SWCNT. This interfacial electron enrichment aligns with the observed n-type conduction dominated by free electrons. Conversely, treatment with high-polarity solvents (E_T > 45 kcal mol⁻¹) to achieve p-type behavior produces distinct charge depletion features. The differential charge density profiles exhibit hole generation near the PANI/SWCNT interface, where PANI acts as a p-type dopant by withdrawing electrons from SWCNT. This localized electron deficiency corresponds to the formation of hole carriers, consistent with the measured p-type thermoelectric response. The polarity-dependent charge redistribution patterns at the heterointerface qualitatively validate the solvent-regulated doping mechanism governing carrier-type inversion in the hybrid system. Bader charge analysis was performed on the PANI/SWCNT interface for all solvent systems, and charge variation (ΔQ) was used to quantitatively confirm the direction of charge transfer (results in Table S1). In low-polarity solvents (DMSO, NMP, and DMF), PANI consistently loses electrons (ΔQ_PANI = +0.047 e⁻ to +0.052 e⁻), while SWCNT gains electrons (ΔQ_SWCNT = −0.045 e⁻ to −0.048 e⁻), further proving electron donation from PANI to SWCNT. In high-polarity solvents, PANI gains electrons (ΔQ_PANI = −0.035 e⁻ to −0.041 e⁻), while SWCNT loses electrons (ΔQ_SWCNT = +0.032 e⁻ to +0.039 e⁻), confirming electron withdrawal from SWCNT.

Therefore, it can be concluded that in the PANI/SWCNT₁₀ hybrid system, solvents with varying E_T values modulate the doping behavior of PANI through distinct interaction mechanisms. Specifically, low-polarity solvents (E_T < 45 kcal mol⁻¹) induce n-type doping by facilitating electron donation from PANI to SWCNT, while high-polarity solvents (E_T > 45 kcal mol⁻¹) trigger p-type doping via electron withdrawal from the CNT framework. This polarity-dependent switching mechanism arises from solvent-regulated interfacial charge transfer processes between PANI and SWCNT.

To enhance the thermoelectric performance, the PANI/SWCNT films were subjected to cold-compressing at 5 MPa for 18 minutes (Fig. 3a), a process designed to optimize carrier transport pathways through densification.^23,40,41 The cold-compressing protocol is described in the SI. Subsequently, the compressed films (denoted as PANI/SWCNT_{10-compressed}) were immersed in low-polarity solvents (NMP, DMSO, DMF), which were used as received in an air atmosphere. The main purpose of the solvent treatment protocol is to preserve the n-type doping by mitigating oxygen-induced p-doping through controlled solvent polarity.

The cold-compressed PANI/SWCNT films demonstrated a substantial enhancement in electrical conductivity, with a magnitude of improvement. The NMP-treated compressed sample (PANI/SWCNT_{10-compressed}-NMP) exhibited a remarkable 2.3 times increase in the electrical conductivity compared to its non-compressed counterpart (PANI/SWCNT₁₀-NMP), as shown in Fig. 3b. The 2.3-fold enhancement in electrical conductivity (σ) after cold-compression is directly correlated to reduced film porosity and densified morphology, as confirmed by scanning electron microscopy (SEM) analysis (Fig. S6). Similar results have been observed in previous works e.g., ∼7.8 times that of the pristine electrical conductivity for willow catkin-XFS16-CNT_33.3 films after cold-compression,⁴⁰ and 12 times electrical conductivity of original value for CNT films after densification process.⁴¹ This densification enhances carrier mobility through improved inter-nanotube contact and reduced charge transport barriers. The thickness of PANI/SWCNT_{10-compressed}-NMP film was confirmed by a cross-section of the dry film's SEM image, as shown in Fig. S7. The enhancement in electrical conductivity was attributed to the increase of the film's density (ρ), which was typically understood with Maxwell-Eucken's equations as shown below:

	(1)

where σ₀ is the theoretical electrical conductivity of PANI/SWCNT₁₀ with ideal packing density (ρ₀). ρ is the density of the cold-compressed PANI/SWCNT film. β₁ is the constant number determined by the conditions of the pores, which is generally in the range of 1–3 when the shape of the pores is spherical.⁴² According to Maxwell-Eucken's equations, a greater density induces higher electrical conductivity. The cold-compressed PANI/SWCNT films exhibited a greater density and less void compared with the non-compressed films, which is consistent with the previous reports.^43–45 For example, following the density increase of the CNT film from 0.7 to ∼1.7g cm⁻³ after compression, the electrical conductivity dramatically improves from ∼1000 S cm⁻¹ to over 10000 S cm⁻¹.⁴⁴ As demonstrated in prior studies,^23,40,46 cold-compressing significantly reduces porosity (Fig. S6), which is expected to simultaneously enhance σ while suppressing the enhancement of k through phonon scattering at consolidated interfaces, leading to the improvement of the σ/k to increase the ZT value.

Notably, the Seebeck coefficient of PANI/SWCNT films remained essentially invariant under cold-compressing treatment (Fig. 3c). This invariance is attributed to the preservation of SWCNTs' intrinsic chemical environment, a direct consequence of cold-compressing being a purely physical densification process that avoids chemical modification. Furthermore, the isotropic nature of SWCNTs' Seebeck coefficient, as established in prior studies on aligned nanotube systems,^45,47 contributes to the observed stability. Crucially, the synergistic combination of enhanced electrical conductivity (driven by densification) and preserved Seebeck coefficient yielded a 253% improvement in power factor for the compressed NMP-treated composite PANI/SWCNT_{10-compressed}-NMP relative to its uncompressed counterpart (Fig. 3d). The DMSO-treated and DMF-treated PANI/SWCNT₁₀ hybrid films show a similar improvement in thermoelectric performance after compression, as shown in Fig. S8. The underlying mechanism, illustrated in Fig. 3e, involves two complementary effects: (1) structural optimization through cold-compressing, which reduces intertube contact resistance while maintaining SWCNT alignments, and (2) electronic state preservation, where the absence of chemical doping ensures minimal perturbation to the density of states near the Fermi level. The further enhancement in PF could be achieved via tuning the alignment of the CNT in the films to improve the electrical conductivity, as the electrical conductivity of the CNT is anisotropic, and the Seebeck coefficient of the CNT is isotropic.^4,23,45,48

Stability remains a critical challenge for n-type materials due to oxygen-induced p-doping degradation. As shown in Fig. 4a, the air stability of PANI/SWCNT₁₀ films was systematically evaluated. The dried n-type PANI/SWCNT₁₀-NMP films exhibited gradual conversion to p-type behavior in ambient air, with their Seebeck coefficient changing from −53.6 to 33.0 μV K⁻¹. In contrast, the wetted PANI/SWCNT_{10-compressed}-NMP films demonstrated remarkable stability, maintaining nearly constant Seebeck coefficient values when exposed to air on the 5th day (Fig. 4b). The superior long-term stability can be attributed to the effective protection provided by the low-polarity NMP solvent, which forms a protective barrier around the n-type composite material (Fig. 4c). The low-polarity solvents act as the oxygen-blocking barrier because they have extremely low solubility of oxygen (O₂) in low-polarity solvents compared to its concentration in ambient air. The molar concentration of O₂ in air is approximately 21 mol%. The solubility of O₂ in common low-polarity solvents is orders of magnitude lower (0.016 mol%∼0.039 mol%).^28,29 This vast difference in concentration creates a significant thermodynamic and kinetic barrier. Therefore, the solvent matrix appears to significantly inhibit oxygen penetration after doping effects, preserving the material's electronic properties under ambient conditions. The conclusion is also identified by the wetted PANI/SWCNT_{10-compressed}-DMSO and PANI/SWCNT_{10-compressed}-DMF films in Fig. S9, the Seebeck coefficients of PANI/SWCNT_{10-compressed} hybrid films are nearly consistent when they are protected by the DMSO and DMF, respectively.

The PANI/SWCNT_{10-compressed} films demonstrated solvent-responsive charge polarity switching behavior. Remarkably, immersion in an NMP/H₂O mixed solvent system induced a controllable transition of the Seebeck coefficient from −59.5 to −37.6 μV K⁻¹ as the H₂O volume fraction increased (Fig. 4d). This polarity modulation stems from high-polarity solvents enabling dipole-induced charge redistribution, which facilitates electron withdrawal from SWCNT networks (Fig. 4e). Systematic solvent engineering allowed reversible switching of the hybrid film's charge transport characteristics between p-type and n-type (Fig. 4f and S10), confirming the material's dual-polarity functionality. Meanwhile, the electrical conductivities and the Seebeck coefficients of compressed PANI/SWCNT₁₀ films immersed in DMSO, DMF, and NMP remained consistent with those of densified films after direct immersion (Fig. 3b, c, S8a, b, d and e), confirming robust stability across low-polarity solvents. The repeated solvent immersion cycles between NMP and H₂O were also conducted (Fig. S11). The composite film retained >80% of its original electrical conductivity after 2.5 full switching cycles (NMP → H₂O → NMP → H₂O → NMP), as shown in Fig. S11a. No obvious degradation was observed in the Seebeck coefficient values throughout the 2.5-cycle switching process (Fig. S11b). The PANI/SWCNT films are supposed to have promising prospects in the field of sensing of NH₃, NO₂, N₂H₄, methanol vapor, metal ions, and so on,^49–53 because the thermoelectric performances of the sensors could be recovered in the immersing low polarity solvent cycle.

Conclusion

We have demonstrated a solvent-mediated polarity control strategy that fundamentally addresses the critical challenge of oxygen-induced p-doping in n-type organic thermoelectric. By exploiting solvent polarity-dependent charge transfer modulation and cold-pressing in PANI/SWCNT hybrids, we achieved reversible p-to-n polarity switching without chemical dopants, yielding a high n-type power factor of 911.3 μW m⁻¹ K⁻² through optimized solvent engineering and densification process. Theoretical insights revealed that low-polarity solvents (E_T < 45 kcal mol⁻¹) stabilize n-type behavior by facilitating electron transfer from PANI to SWCNTs, while high-polarity solvents reverse this polarity via interfacial charge redistribution. The cold-compressed hybrid films exhibited exceptional air stability, retained 73% electrical conductivity and 96% Seebeck coefficient after 4-day ambient exposure, attributed to solvent-derived oxygen-blocking effects. Additionally, the polarity-switching mechanism remains cyclically reversible, enabling dynamic control of carrier transport characteristics. This work shows a universal platform for probing intrinsic material properties and designing air-stable n-type organic electronics, avoiding the limitations of vacuum-dependent or plasma-based approaches. The solvent-mediated polarity control strategy opens new avenues for scalable fabrication of high-performance n-type organic thermoelectric materials and complementary circuits for organic electronics.

Conflicts of interest

The authors declare no competing interests.

Data availability

The data that support the findings of this study are available from the corresponding author, [H. W.], upon reasonable request.

Enlarged Raman spectra, Seebeck coefficients of samples in N₂, Seebeck coefficients of samples wetted with hexane and toluene, electrical conductivity of samples after immersing in various solvents, thicknesses and surface morphology of sample, thermoelectric performance and air stability of sample wetted with DMSO and DMF, and the electrical conductivity and Seebeck coefficients of the samples in the reversible p-to-n polarity switching in the cycle of various solvents and the cycle of NMP and H₂O. See DOI: https://doi.org/10.1039/d5ta04654h.

Acknowledgements

H. W. acknowledges financial support from the National Natural Science Foundation of China (Grant No. 52276014 and 52488201), National Defense Basic Scientific Research Project (JCKYS2024DC04), Shaanxi Provincial Natural Science Basic Research Program – Key Projects – Frontier Exploration Project Demand Traction Category (2025JC-QYXQ-024), Shaanxi Provincial Natural Science Basic Research Program- General Project – Youth Project (Category C) (2025JC-YBQN-476), and the Fundamental Research Funds for the Central Universities, and the World-Class Universities (Disciplines). This work was also supported by the HPC Platform and the Instrument Analysis Center, Xi'an Jiaotong University.

References

1. L. Deng, Y. Liu, Y. Zhang, S. Wang and P. Gao, Adv. Funct. Mater., 2023, 33, 2210770 CrossRef CAS .
2. W. Zhao, J. Ding, Y. Zou, C.-a. Di and D. Zhu, Chem. Soc. Rev., 2020, 49, 7210–7228 RSC .
3. H. Wang and C. Yu, Joule, 2019, 3, 53–80 CrossRef CAS .
4. X. Hao, J. Wang and H. Wang, Chem. Soc. Rev., 2025, 54, 1957–1985 RSC .
5. Y. Huang, K. Wu, Y. Sun, Y. Hu, Z. Wang, L. Yuan, S. Wang, D. Ji, X. Zhang, H. Dong, Z. Gong, Z. Li, X. Weng, R. Huang, Y. Cui, X. Chen, L. Li and W. Hu, Nat. Commun., 2024, 15, 626 CrossRef CAS .
6. Y. Lu, Z. D. Yu, R. Z. Zhang, Z. F. Yao, H. Y. You, L. Jiang, H. I. Un, B. W. Dong, M. Xiong, J. Y. Wang and J. Pei, Angew. Chem., Int. Ed., 2019, 58, 11390–11394 CrossRef CAS PubMed .
7. X. Fan, J. Liu, X. Duan, H. Li, S. Deng, Y. Kuang, J. Li, C. Lin, B. Meng, J. Hu, S. Wang, J. Liu and L. Wang, Adv. Sci., 2024, 11, 2401952 CrossRef CAS .
8. H. Tang, Y. Liang, C. Liu, Z. Hu, Y. Deng, H. Guo, Z. Yu, A. Song, H. Zhao, D. Zhao, Y. Zhang, X. Guo, J. Pei, Y. Ma, Y. Cao and F. Huang, Nature, 2022, 611, 271–277 CrossRef CAS .
9. J. Han, H. Fan, Q. Zhang, Q. Hu, T. P. Russell and H. E. Katz, Adv. Funct. Mater., 2020, 31, 2005901 CrossRef .
10. X. Zhao, D. Madan, Y. Cheng, J. Zhou, H. Li, S. M. Thon, A. E. Bragg, M. E. DeCoster, P. E. Hopkins and H. E. Katz, Adv. Mater., 2017, 29, 1606928 CrossRef .
11. D. Yuan, Y. Guo, Y. Zeng, Q. Fan, J. Wang, Y. Yi and X. Zhu, Angew. Chem., Int. Ed., 2019, 58, 4958–4962 CrossRef CAS .
12. X. Fan, J. Liu, X. Duan, H. Li, S. Deng, Y. Kuang, J. Li, C. Lin, B. Meng, J. Hu, S. Wang, J. Liu and L. Wang, Adv. Sci., 2024, 11, 2401952 CrossRef CAS .
13. K. Feng, H. Guo, J. Wang, Y. Shi, Z. Wu, M. Su, X. Zhang, J. H. Son, H. Y. Woo and X. Guo, J. Am. Chem. Soc., 2021, 143, 1539–1552 CrossRef CAS .
14. S.-H. Jhi, S. G. Louie and M. L. Cohen, Phys. Rev. Lett., 2000, 85, 1710–1713 CrossRef CAS .
15. P. G. Collins, K. Bradley, M. Ishigami and A. Zettl, Science, 2000, 287, 1801–1804 CrossRef CAS .
16. K. Bradley, S.-H. Jhi, P. G. Collins, J. Hone, M. L. Cohen, S. G. Louie and A. Zettl, Phys. Rev. Lett., 2000, 85, 4361–4364 CrossRef CAS PubMed .
17. C. Meng, C. Liu and S. Fan, Adv. Mater., 2010, 22, 535–539 CrossRef CAS .
18. Q. Yao, Q. Wang, L. Wang and L. Chen, Energy Environ. Sci., 2014, 7, 3801–3807 RSC .
19. H. Li, S. Liu, P. Li, D. Yuan, X. Zhou, J. Sun, X. Lu and C. He, Carbon, 2018, 136, 292–298 CrossRef CAS .
20. J. Chen, L. Wang, X. Gui, Z. Lin, X. Ke, F. Hao, Y. Li, Y. Jiang, Y. Wu, X. Shi and L. Chen, Carbon, 2017, 114, 1–7 CrossRef CAS .
21. J. Huang, X. Liu and Y. Du, J. Materiomics, 2024, 10, 173–178 CrossRef .
22. Z. Gong, A. Suwardi and J. Cao, Adv. Funct. Mater., 2025, 2423371,  DOI:10.1002/adfm.202423371 .
23. H. Wang, K. Li, X. Hao, J. Pan, T. Zhuang, X. Dai, J. Wang, B. Chen and D. Chong, Nat. Commun., 2024, 15, 5617 CrossRef CAS PubMed .
24. Q. Yao, L. Chen, W. Zhang, S. Liufu and X. Chen, ACS Nano, 2010, 4, 2445–2451 CrossRef CAS .
25. S. Tao, B. Hong and Z. Kerong, Spectrochim. Acta, Part A, 2007, 66, 1364–1368 CrossRef .
26. P. Zhang, J. Fan, Y. Wang, Y. Dang, S. Heumann and Y. Ding, Carbon, 2024, 222, 118998 CrossRef CAS .
27. Y. Zhou, Q. Wei, M. Zhang, H. Nakajima, T. Okazaki, T. Yamada and K. Hata, ACS Appl. Mater. Interfaces, 2024, 16, 4199–4211 CrossRef CAS .
28. J. H. Dymond, J. Phys. Chem., 1967, 71, 1829–1831 CrossRef CAS .
29. T. Sato, Y. Hamada, M. Sumikawa, S. Araki and H. Yamamoto, Ind. Eng. Chem. Res., 2014, 53, 19331–19337 CrossRef CAS .
30. U. Mayer, V. Gutmann and W. Gerger, Monatsh. Chem., 1975, 106, 1235–1257 CrossRef CAS .
31. V. Gutmann, Coord. Chem. Rev., 1976, 18, 225–255 CrossRef CAS .
32. Y. Nonoguchi, M. Nakano, T. Murayama, H. Hagino, S. Hama, K. Miyazaki, R. Matsubara, M. Nakamura and T. Kawai, Adv. Funct. Mater., 2016, 26, 3021–3028 CrossRef CAS .
33. H. Li, Y. Liang, Y. Liu, S. Liu, P. Li and C. He, Compos. Sci. Technol., 2021, 210, 108797 CrossRef CAS .
34. H. Li, Y. Liu, P. Li, S. Liu, F. Du and C. He, ACS Appl. Mater. Interfaces, 2021, 13, 6650–6658 CrossRef CAS PubMed .
35. H. Li, Y. Liu, S. Liu, P. Li, C. Zhang and C. He, Compos. Part Appl. Sci. Manuf., 2023, 166, 107386 CrossRef CAS .
36. S. Yin, W. Lu, X. Wu, Q. Luo, E. Wang and C.-Y. Guo, ACS Appl. Mater. Interfaces, 2021, 13, 3930–3936 CrossRef CAS .
37. C. Cho, B. Stevens, J.-H. Hsu, R. Bureau, D. A. Hagen, O. Regev, C. Yu and J. C. Grunlan, Adv. Mater., 2015, 27, 2996–3001 CrossRef CAS .
38. L. Feng, R. Wu, C. Liu, J. Lan, Y.-H. Lin and X. Yang, ACS Appl. Energy Mater., 2021, 4, 4081–4089 CrossRef CAS .
39. S. Wang, F. Liu, C. Gao, T. Wan, L. Wang, L. Wang and L. Wang, Chem. Eng. J., 2019, 370, 322–329 CrossRef CAS .
40. Y. Wang, K. Li, J. Wang, X. Dai, X. Sun, D. Chong, J. Yan, L. Zhang and H. Wang, J. Mater. Chem. A, 2022, 10, 25740–25751 RSC .
41. Y.-M. Liu, X.-L. Shi, T. Wu, H. Wu, Y. Mao, T. Cao, D.-Z. Wang, W.-D. Liu, M. Li, Q. Liu and Z.-G. Chen, Nat. Commun., 2024, 15, 3426 CrossRef CAS PubMed .
42. R. Goodall, L. Weber and A. Mortensen, J. Appl. Phys., 2006, 100, 044912 CrossRef .
43. Y. Wang, X. Dai, J. Pan, J. Wang, X. Sun, K. Li and H. Wang, J. Mater. Chem. A, 2024, 12, 18948–18957 RSC .
44. K. Li, X. Sun, Y. Wang, J. Wang, X. Dai, Y. Yao, B. Chen, D. Chong, J. Yan and H. Wang, Small, 2023, 19, 2304266 CrossRef CAS .
45. X. Sun, Y. Wang, K. Li, J. Wang, X. Dai, D. Chong, J. Yan and H. Wang, Adv. Funct. Mater., 2022, 32, 2203080 CrossRef CAS .
46. X. Dai, Y. Wang, X. Sun, K. Li, J. Pan, J. Wang, T. Zhuang, D. Chong, J. Yan and H. Wang, Adv. Mater., 2024, 36, 2312570 CrossRef CAS .
47. K. Fukuhara, Y. Ichinose, H. Nishidome, Y. Yomogida, F. Katsutani, N. Komatsu, W. Gao, J. Kono and K. Yanagi, Appl. Phys. Lett., 2018, 113, 243105 CrossRef .
48. Y. Wang, Q. Li, J. Wang, Z. Li, K. Li, X. Dai, J. Pan and H. Wang, Nano Energy, 2022, 93, 106804 CrossRef CAS .
49. L. Xue, W. Wang, Y. Guo, G. Liu and P. Wan, Sens. Actuators, B, 2017, 244, 47–53 CrossRef CAS .
50. M. H. Mostafa, E. S. Ali and M. S. A. Darwish, Mater. Chem. Phys., 2022, 291, 126699 CrossRef CAS .
51. A. Bora, K. Mohan, D. Pegu, C. B. Gohain and S. K. Dolui, Sens. Actuators, B, 2017, 253, 977–986 CrossRef CAS .
52. N. Ansari, M. Y. Lone, Shumaila, J. Ali, M. Zulfequar, M. Husain, S. S. Islam and S. Husain, J. Appl. Phys., 2020, 127, 044902 CrossRef CAS .
53. M. Ding, Y. Tang, P. Gou, M. J. Reber and A. Star, Adv. Mater., 2011, 23, 536–540 CrossRef CAS PubMed .

---

Footnote

† These two authors contribute equally to the paper.

This journal is © The Royal Society of Chemistry 2025