Introduction

Contact-electrification (CE) is a ubiquitous charge transfer effect1,2,3, which could render the contact surfaces of two objects being reversely charged after contacts and separations4,5,6. Electrons have been identified as the dominant charge carriers in a majority of cases of CE7,8,9, and the CE-driven electron transfer is able to promote chemical reactions through a process known as contact-electro-catalysis (CEC)10,11. CEC has provided sustainable and effective strategies for the production of reactive oxygen species (ROS)12,13, reduction of metal ions14,15, and synthesis of vital chemicals16,17. In addition, recent studies have also suggested that CEC could be employed for controllable radical polymerization, which highlights the great potential of CEC in facilitating target organic reactions18,19,20. Meanwhile, pristine polymers are the mostly used CEC catalysts due to their high CE performance under ambient conditions21,22,23. However, a variety of significant chemical reactions would be facilitated under high temperatures24,25,26, and pristine polymer-based CEC is less suitable for catalyzing these reactions since the CE ability of polymers would decrease drastically at elevated temperatures27. Thermally stable oxides have been proposed as feasible substitutes for polymers in high-temperature CEC, which could exhibit higher CEC efficiency when the temperature is beyond 70 °C14. However, the intrinsic low CE ability of oxides still renders a relatively sluggish rate of target reactions. Existing investigations have implied that doping with metals28,29, introduction of oxygen vacancies30,31, and assistance with metal-organic frameworks (MOFs)32,33 are three commonly employed methods for improving the catalytic efficiency of oxides. However, these strategies generally rely on delicate synthetic processes that usually involve with high temperatures, and such improvement is often limited to specific oxide substrates. Considering the CE effect is a surface phenomenon that remains largely independent of the bulk property of materials11, we suppose suitable surface engineering may provide a generalized approach for improving the CEC performance of target oxides. Specifically, recent studies on the correlation between surface functional groups and the CE ability have suggested that the density and electron-withdrawing (EW) ability of surface functional groups could significantly affect the overall CE performance34,35. Therefore, we expect the CE ability and corresponding CEC efficiency of oxides could be improved by proper surface modifications, thereby expanding the applicability of CEC across a broader thermal range.

Here, we present a generalized fluorination-based strategy to enable pristine oxides with high CEC performance under both ambient and high-temperature conditions. A polyfluorinated reactant, 1H,1H,2H,2H-perfluorodecyltriethoxysilane (FDTES), is employed for the surface fluorination of target oxides, and the high EW ability of fluorinated functional groups in FDTES is expected to enhance the overall CE abilities36,37. We demonstrate that the CE ability and corresponding CEC efficiency of a variety of oxides (such as SiO2, Al2O3, MgO, and ZrO2) can be improved by this fluorination approach. Taking SiO2 as an example, the quantity of transferred charges upon contact with deionized (DI) water has been improved by 6.12-fold after fluorination, which in turn leads to a substantial enhancement in the yield of ROS produced through CEC. Theoretical calculations reveal that the fluorination could not only reduce the energy barrier for CE-driven electron transfer but also form a hydrogen bond network that further facilitates this process. More importantly, the fluorinated SiO2 (F-SiO2) outperforms pristine fluorinated ethylene propylene (FEP) when the temperature increases to 70 °C. To be specific, the quantity of transferred charges between F-SiO2 and DI water could still reach ~21.9 nC at 70 °C, and the intensity of hydroxyl radical peaks could maintain 90.77 % of that at room temperatures. In contrast, the quantity of transferred charges between FEP and DI water is merely 4.08 nC, and negligible hydroxyl radicals can be obtained under the same conditions. This disparity could be mainly ascribed to the F-SiO2 remains structurally stable even when the temperature reaches 200 °C, and the grafted fluorinated functional group could still effectively facilitate the corresponding electron transfer process. The practicability of F-SiO2 for CEC under high-temperature conditions is first examined by the leaching of cathode materials from spent ternary lithium-ion batteries (LIBs). The leaching efficiencies for Li, Ni, Co, and Mn elements could respectively reach 99.75 %, 99.44 %, 96.77 %, and 94.07 % after ultrasonicating at 70 °C for 300 min. Moreover, by using dimethyl sulfoxide (DMSO) as the solvent, we have investigated the CEC performance of F-SiO2 under further higher temperatures (180 °C). Specifically, the removal ratio of methyl orange in the F-SiO2 group is 6.13 times higher than that in the FEP group under 180 °C, further proving the outperformance of F-SiO2 under high temperatures. Considering the broad feasibility of this fluorination-based enhancement, we envision this strategy could not only provide effective substitutes for polymeric CEC catalysts, but also render a more efficient CEC process, especially under elevated temperatures.

Results

Enhanced CEC-induced ROS yield under room temperature by fluorinated oxides

A polyfluorinated reactant, 1H,1H,2H,2H-perfluorodecyltriethoxysilane (FDTES), is employed for the fluorination of target oxide substrates. Exemplified by SiO2, the FDTES could be modified on its surface through a condensation reaction38, as shown by the schematic illustration in Fig. 1a. The validity of the fluorination process is confirmed by the emergence of F1s peaks in X-ray photoelectron spectra (XPS) of the fluorinated SiO2 (F-SiO2) in Fig. 1b. The survey spectrum of pristine SiO2 and F-SiO2 is available in Supplementary Fig. 1. Fourier transform infrared (FTIR) spectroscopic analysis in Supplementary Fig. 2 has also proved the presence of fluorine-containing functional groups on F-SiO2. According to previous research on the relationship between contact-electrification (CE) ability and surface functional groups34,35,39, the introduced fluorinated functional groups on F-SiO2 are expected to facilitate the process of obtaining electrons from water. We also verified this assumption using a single-electrode mode triboelectric nanogenerator (Supplementary Fig. 3). Results in Fig. 1c indicate that the quantity of transferred charges during contact with deionized (DI) water has been improved from 4.67 nC to 28.58 nC by fluorination. In light of the enhanced CE ability, the performance of F-SiO2 for CEC-induced reactive oxygen species (ROS) generation was examined. Measured waveforms by electron paramagnetic resonance (EPR) spectra in Fig. 1d and e suggested that the yields of hydroxyl and superoxide radicals were both significantly improved. Such consistence between CE ability and CEC-induced ROS generation performance not only confirms the feasibility of FDTES-based enhancement strategy, but also suggests that the CEC should be the dominant mechanism for F-SiO2 to produce ROS. The enhanced hydroxyl radical production could be explained by the high electron affinity of fluorinated functional groups in FDTES34. Thus, the water oxidation reaction (WOR), which produces hydroxyl radicals by grabbing electrons from water molecules, could be significantly facilitated. The formation of superoxide radicals arises from oxygen reduction reaction (ORR), and its improvement could be attributed to that an increasing number of electrons would accumulate on the surface of F-SiO2 during its contact with water. This enables more electrons to be available for subsequent transfer to oxygen molecules for producing superoxide radicals35. In addition, the stability of grafted fluorinated functional groups on F-SiO2 during CEC was also investigated. Negligible variation in the peak intensity and location shift of the F1s peak was observed in XPS spectra of F-SiO2 before and after reaction, as exhibited in Supplementary Fig. 4, implying that the ultrasonication can hardly result in bond cleavages within fluorinated functional groups on F-SiO2 surfaces. FTIR spectroscopic analysis in Supplementary Fig. 5 has further proved the stability of F-SiO2. The fingerprint region in FTIR characterization remains unchanged after reaction, which suggests the high reusability of F-SiO2 for ultrasonication-based CEC.

Fig. 1: Fluorinated SiO2 (F-SiO2) for improved contact-electro-catalysis (CEC).
Fig. 1: Fluorinated SiO2 (F-SiO2) for improved contact-electro-catalysis (CEC).
Full size image

a Schematic illustration of the fluorination process on the SiO2 surface. FDTES refers to 1H,1H,2H,2H-perfluorodecyltriethoxysilane. b F1s spectra in XPS of SiO2 before and after fluorination. c Measured transferred charges when pristine SiO2 and F-SiO2 contacts separately with deionized water. d Electron paramagnetic resonance (EPR) diagrams of hydroxyl radicals produced by employing SiO2 or F-SiO2 for CEC. e EPR profiles of superoxide radicals produced by SiO2 and F-SiO2-based CEC. Source data are provided as a Source Data file.

Reactivity of ROS produced by fluorinated oxides-based CEC

Taking the degradation of methyl orange (MO) aqueous solution as a model reaction, we have investigated the reactivity of ROS produced by fluorinated oxides-based CEC. Exemplified by F-SiO2, the characteristic absorbance peak of MO in UV-Vis spectra diminished with the prolongation of ultrasonication time and approached zero after 120 min, as depicted in Fig. 2a. Ex-situ spectroscopic analyses in Supplementary Fig. 6 indicate that F-SiO2 remains the same after reaction, excluding the possibility of physical adsorption of MO on its surface. Moreover, the correlation between FDTES concentration and MO removal ratio in Fig. 2b suggests that a higher FDTES concentration could lead to a more efficient MO degradation process, and such improvement gets saturation when 5 mM FDTES is employed. This not only verifies the dominant role of FDTES in promoting MO degradation, but also indicates that exposing more binding sites for FDTES on SiO2 may further enhance the CEC performance. The evolution of MO relative concentration in the presence of various scavengers in Fig. 2c indicates that the degradation of MO is mainly driven by •OH and •O2− radicals, and •OH radicals appear as the limiting factor, which could be ascribed to the high electron affinity of fluorinated functional groups on F-SiO2. The generation of •OH and •O2- radicals during F-SiO2-based CEC was further evaluated through the terephthalic acid (THA) and nitro blue tetrazolium (NBT) experiments40,41. Specifically, as exhibited in Fig. 2d, the emission intensity of THA-OH adducts at 425 nm in the F-SiO2 group has been improved by 8.73 times, and the degradation rate of NBT has been enhanced by 6.7-fold compared to the pristine SiO2 group. More importantly, in addition to SiO2, such a fluorination-based CEC enhancement strategy could potentially be extended to a variety of oxides. To unambiguously demonstrate its feasibility, a group of chemically inert oxides has been selected, including Al2O3, MgO, and ZrO2. Figure 2e compared the CE capacities of these oxides before and after the fluorination process, revealing that their CE performances have all been significantly improved after fluorination. Among them, the fluorinated Al2O3 (F-Al2O3) exhibits the highest transferred charges during contact with DI water ( ~ 17.15 nC), followed by fluorinated ZrO2 (F-ZrO2), and fluorinated MgO (F-MgO). The differences in CEC performance among these fluorinated oxides align well with the discrepancies in their CE abilities, as depicted in Fig. 2f. For example, F-Al2O3, which exhibits the highest CE ability (except F-SiO2), is the most effective in MO degradation, as 92.81 % of MO could be removed after ultrasonication for 3 h. Such consistence not only suggests that CEC is the underlying mechanism for MO degradation by these oxides, but also indicates that the selection range of CEC catalysts could be further expanded by this generalized fluorination-based strategy.

Fig. 2: Reactivity of CEC-produced reactive oxygen species (ROS).
Fig. 2: Reactivity of CEC-produced reactive oxygen species (ROS).
Full size image

a UV-Vis spectra of a 5 mg/L methyl orange (MO) aqueous solution during ultrasonication in presence of F-SiO2 powder for 180 min. b Comparison of MO relative concentrations when SiO2 substrates are fluorinated with varying concentrations of FDTES. c Evolution of MO relative concentration in conditions of different radical scavengers. d The production of •OH and •O2− radicals was evaluated by the fluorescence intensity of THA (terephthalic acid)-OH and the absorbance of nitro blue tetrazolium (NBT), respectively. e Measured transferred charges for different pristine or fluorinated oxides during contact with deionized water. f MO removal ratios by using different pristine or fluorinated oxides in CEC. Error bars represent the standard deviation based on three replicate data. Source data are provided as a Source Data file.

Theoretical analysis of the CE-driven electron transfer on fluorinated oxides

In order to elucidate the underlying mechanisms for fluorination-enhanced CEC performance, we employed density functional theory (DFT) simulations to examine the influence from fluorination on the CE-driven electron transfer process at oxide surfaces. Figure 3a, b compared the electronic structures of pristine SiO2 and F-SiO2 in the water environment. The left insets are corresponding optimized structural models, and the details of the simulation are available in Supplementary Note 1. The atomic coordinates of the optimized computational models of SiO2 and F-SiO2 are available in Supplementary Data 1, 2. It can be found that the fluorinated functional groups in F-SiO2 are responsible for accepting transferred electrons, and the highest occupied molecular orbital (HOMO) of water molecules is elevated due to interactions with F-SiO242. Consequently, the energy required for obtaining an electron from water molecules decreased from 5.96 eV to 3.69 eV after fluorination, thereby facilitating the electron transfer process during CE. Calculated energy barriers for transferring electrons between water molecules and other fluorinated oxides were displayed in Supplementary Fig. 7, with F-SiO2 exhibiting the lowest energy barrier. The atomic coordinates of the optimized computational models of other fluorinated oxides are available in Supplementary Data 3–5. Another dominant reason for the outperformance of F-SiO2 should be the formation of hydrogen bond networks as presented in Fig. 3c. The initial and final atomic configurations of ab initio molecular dynamics (AIMD) calculations are exhibited in Supplementary Data 6. Owing to the hydroxyl group that would be produced on SiO2 surfaces by contacting with water43, its synergy with fluorinated groups would give rise to the establishment of hydrogen bond networks, further facilitating the electron transfer process. The role of hydroxyl groups in the formation of hydrogen bond networks was assessed through methoxy group passivation. In contrast to Fig. 3c, Supplementary Fig. 8 suggests that hydrogen bond networks are significantly diminished when hydroxyl groups on the F-SiO2 surface are replaced with methoxy groups. Corresponding initial and final atomic configurations of AIMD calculations can be found in Supplementary Data 7. Additionally, charge density difference simulations were conducted to evaluate the impact of hydrogen bond networks on the CE-driven electron transfer process. The charge density data derived from the equilibrated AIMD configurations are used, and the details are available in Supplementary Note 2. Figure 3d illustrates the simulated spatial distribution of the charge density difference between H2O molecules and methoxy group passivated F-SiO2. A more pronounced extent of charge accumulation and depletion was observed in F-SiO2 with surface hydroxyl groups, as demonstrated by Fig. 3e, implying stronger interactions were developed under this condition and thus promoting the electron transfer process. To be specific, the net number of transferred electrons from water molecules to F-SiO2 could achieve 2.967, which is approximately 2.81 times greater than that under methoxy group passivated conditions. Therefore, the fluorination strategy could effectively enhance the CE abilities of target oxides by decreasing the energy barrier for electron transfer, and the formation of hydrogen bond networks on the F-SiO2 surface could further facilitate the CE-driven electron transfer process, leading to a significantly improved CEC performance.

Fig. 3: Theoretical investigations on the impact from fluorinated functional groups on CE-driven electron transfer process.
Fig. 3: Theoretical investigations on the impact from fluorinated functional groups on CE-driven electron transfer process.
Full size image

Calculated energy barrier for transferring electrons between water molecules and SiO2 (a) as well as F-SiO2 (b). The insets are corresponding optimized structural model. c Simulated distribution of hydrogen bond networks at the adjacent of F-SiO2 and water interface. Simulated spatial distribution of charge density difference between H2O molecules and methoxy group-passivated F-SiO2 (d), as well as F-SiO2 (e). The blue sphere represents for Si atoms in SiO2, and the purple sphere for Si atoms in 1H,1H,2H,2H-perfluorodecyltriethoxysilane (FDTES). The charge density difference is plotted in two parts: one showing regions of electron density accumulation (in yellow) and the other showing regions of electron density depletion (in blue). Source data are provided as a Source Data file.

Superior CEC performance of F-SiO2 under elevated temperatures

In virtue of the high thermal stability of the SiO2 substrate and the outperformance of F-SiO2, we have further explored the feasibility of utilizing F-SiO2 for high-temperature-based CEC. Figure 4a presented the differences in the amount of transferred charges between fluorinated ethylene propylene (FEP) and F-SiO2 when they were separately contacted with DI water under varying thermal conditions. Corresponding waveforms of charge transfer during CE are available in Supplementary Fig. 9. While FEP is the mostly used CEC catalysts due to its superior CE performance, the CE ability of F-SiO2 is only slightly lower than that of FEP at 30 °C. Then, both materials experience a reduction of transferred charges when the temperature is raised to 50 °C, and the extent of decrease for FEP is more obvious, which could be ascribed to its glass transition under elevated temperatures27. As a result, F-SiO2 delivers better CE performance than FEP in this condition. A minor decrease of transferred charges is observed for F-SiO2 when the temperature is further increased to 70 °C. However, FEP can hardly be electrified with water under this situation, as only 4.08 nC of transferred charges were obtained. Corresponding characterizations of CEC performance in Fig. 4b have demonstrated the same variation trend. In terms of the CEC-promoted production of hydroxyl radicals, the yields were comparable in FEP and F-SiO2 groups when the temperature was 30 °C, and the F-SiO2 could still maintain high yields even when the temperature increased to 70 °C. On the contrary, the characteristic quadruplet peaks of DMPO-OH were almost negligible in the FEP group when the temperature was 70 °C, which is in accordance with the observed decline in its CE capability.

Fig. 4: Comparison of CEC performance between pristine polymers and fluorinated oxides under elevated temperatures.
Fig. 4: Comparison of CEC performance between pristine polymers and fluorinated oxides under elevated temperatures.
Full size image

a Evolution of transferred charges per cycle for fluorinated ethylene propylene (FEP) and F-SiO2 during their contacts with deionized water under different temperatures. b Corresponding electron paramagnetic resonance (EPR) diagrams for FEP and F-SiO2 during CEC under different temperatures. c Comparison of contact-electrification (CE) abilities among various conditions of FEP membrane. d In-situ FTIR spectroscopy of F-SiO2 when the temperature varies from 30 to 200 °C. e Simulated distribution of hydrogen bond networks at the adjacent of F-SiO2 and water interface when the temperature is increased to 350 K. Error bars represent the standard deviation based on three replicate data. Source data are provided as a Source Data file.

The diminution of CE performance at elevated temperatures could typically be explained by the structural changes and thermionic emission of electrons44. In order to clarify the dominant influencing mechanism, we have compared the CE performance of pristine and quenched FEP membranes to decouple these two factors. The quenched FEP refers to a FEP film that is kept at 70 °C for 1 h followed by cooling down to room temperature (RT). Obtained results in Fig. 4c indicate that the quantity of transferred charges in the quenched FEP group is only 9.24 nC, even the measurement was performed at RT, which is 69.62% less than that of the pristine FEP group under the identical condition. Such a loss should be mainly ascribed to the structural change, as the thermionic emission of electrons usually occurs at significantly higher temperatures. This result indicates that the structural alteration during heating is the principal reason for the decreased CE abilities, thus highlighting the significance of utilizing thermally stable catalysts for enhancing the CEC efficiency at high temperatures. The thermal stability of F-SiO2 was proved by the in-situ FTIR spectroscopy as depicted in Fig. 4d. The fingerprint region of F-SiO2 remains unchanged even as the temperature rises from 30 to 200 °C, indicating that the fluorinated functional groups are thermally stable and can maintain stable binding to the SiO2 substrates. Corresponding thermogravimetry (TG) results in Supplementary Fig. 10 have also confirmed the thermal stability of F-SiO2, as only 0.61% weight loss was observed when the temperature increased to 200 °C. Moreover, theoretical simulations in Fig. 4e indicate that the hydrogen bond network still exists when the temperature increases to 350 K, and Supplementary Fig. 11 has also verified the stability of the hydrogen bond network by introducing Na+ and Cl− ions as additives. Corresponding initial and final atomic configurations of AIMD calculations can be found in Supplementary Data 8, 9. Simulated result in Supplementary Fig. 11a implies that the overall distribution of hydrogen bond networks remains nearly the same after introducing ions, except for a tiny diminution of hydrogen bonds in the vicinity of Na+ and Cl-. The stability of hydrogen bond networks under various conditions was further confirmed through analyzing the intensity and distribution of hydrogen bonds along the Z-axis of F-SiO2/water interface, as shown in Supplementary Fig. 11b. Although the introduction of Na+ and Cl- ions would affect the shape of peaks to some extent, neither obvious shift of the position nor variation of the intensity of peaks was observed under all the conditions investigated, which indicates the hydrogen bond network can stably exist at F-SiO2/water interface. Hence, the fluorinated functional groups that maintain stable binding to SiO2 substrate should play a dominant role in enabling a broad temperature window, and the reduced energy barrier, as well as the hydrogen bond networks, are two main reasons for the enhanced CEC performance of F-SiO2.

Practicability of F-SiO2 in recycling spent cathodes of LIBs

The feasibility of F-SiO2 in contact-electro-catalysis (CEC) at elevated temperatures was demonstrated through the leaching of cathode materials from spent ternary lithium-ion batteries (LIBs). Figure 5a presents a flowchart of the entire recycling process, with a detailed description provided in the Methods section. Various dielectric powders, including F-SiO2, SiO2, and FEP, were individually mixed with spent cathode materials as well as citric acid, and the ultrasonication-based CEC-leaching procedure was performed at 70 °C for 300 min. A blank control group was also performed under the same conditions, except that the citric acid was the only leaching agent. The CEC-produced ROS is supposed to promote the reduction of valuable metals in cathodes from poorly soluble high-valence state to soluble low-valence state14. Aliquots were sampled at specific time intervals and subjected to the UV-Vis spectroscopy and inductively coupled plasma optical emission spectrometry (ICP-OES) analysis. The UV-Vis spectral profiles in Fig. 5b reveal that the highest absorbances of these ions were achieved in the F-SiO2 group, followed by SiO2, FEP, and control groups. The evolution of metal concentrations in different groups during the CEC-leaching process were characterized by the ICP-OES analysis, and obtained results in Supplementary Fig. 12 suggest that the leaching efficiency of Li, Ni, Co, and Mn all increased with prolonged ultrasonication time. The final leaching efficiencies for each group are compared in Fig. 5c, which indicates that the leaching efficiencies of target elements in the F-SiO2 group are all the highest, with 99.75% of Li, 99.44% of Ni, 96.77% of Co, and 94.07% of Mn. In addition to developing catalysts with higher CEC performance, we expect the overall leaching efficiency could also be enhanced by optimizing the pretreatment procedure of spent LIBs. For example, polyvinylidene difluoride (PVDF) and Al foil are generally regarded as impurities during the pretreatment of cathode materials. On the one hand, the CE effect also exists at the PVDF-water interface34,35, and PVDF has been employed to degrade MO aqueous solution via CEC10. On the other hand, recent studies have suggested that the presence of reductive impurity elements (e.g., Cu, Al, Fe) could significantly enhance the dissolution of active materials45,46. Therefore, we suppose that some impurities during the pretreatment may be employed to facilitate the subsequent CEC-leaching process, which is expected to render a more cost-effective and efficient recycling process.

Fig. 5: Application of CEC in regeneration of cathode materials from spent lithium-ion batteries.
Fig. 5: Application of CEC in regeneration of cathode materials from spent lithium-ion batteries.
Full size image

a Schematics for the regeneration process of NCM ternary cathodes via CEC-leaching. b UV-Vis spectra of leachate obtained by using different dielectric powders for CEC-leaching. c Corresponding leaching efficiencies of different metal elements in each group. d Comparison of XRD spectra between commercialized and regenerated NCM622 ternary cathode powders. Source data are provided as a Source Data file. Created with Adobe Illustrator.

In virtue of the stability of F-SiO2, we have investigated the reusability of F-SiO2 and found that negligible diminution in the leaching efficiency was observed after recycling F-SiO2 for 3 times (Supplementary Fig. 13). The regeneration of the 622 ternary cathode was performed after the leaching process. The leaching solution was first filtered to remove dielectric powders before introducing nickel acetate, cobalt acetate and manganese acetate with a final metal concentration ratio reaching 6:2:2. After magnetic stirring at 80 °C for 12 h, the suspension was transferred to a muffle furnace and kept at 900 °C for 6 h to obtain the recycled cathode materials. The X-ray diffraction (XRD) pattern of the recycled NCM622 ternary cathode powder in Fig. 5d closely resembles that of the commercialized NCM622 cathode, indicating the high purity of the obtained cathode materials47. The excellent reusability of F-SiO2 and reduced leaching time due to its superior CEC performance at elevated temperatures suggest the advantage of employing fluorinated oxides for CEC at high temperatures. Moreover, we have investigated the CEC performance of F-SiO2 under further higher temperatures (180 °C) by using dimethyl sulfoxide (DMSO) as the solvent. DMSO is selected because it is a commonly used high-boiling-point organic solvent, and previous studies have proved that DMSO can be employed for CEC degradation of organic pollutants48. The degradation of MO (dissolved in DMSO) was employed as a model reaction, and the highest decoloration ratio was obtained in the F-SiO2 group (86.44%), as depicted in Supplementary Fig. 14. While the removal ratio of MO was nearly the same in the blank control (11.35%) and FEP groups (14.11%). The outperformance of F-SiO2 under 180 °C further verifies the feasibility of broadening the temperature window for CEC applications by this fluorination-based strategy.

Discussion

A generalized fluorination-based surface modification strategy was developed for improving the contact-electro-catalysis (CEC) performance of target oxides. Our results suggested that this strategy is applicable to a variety of oxides, such as SiO2, Al2O3, MgO, and ZrO2. Exemplified by pristine SiO2 and F-SiO2, we demonstrated that the yield of reactive oxygen species (ROS) could be significantly promoted by using F-SiO2 for CEC, and the extent of enhancement was positively related to the concentration of utilized FDTES. Theoretical simulations suggested that the fluorinated functional group could facilitate the CE-driven electron transfer by decreasing the energy barrier, and its synergy with hydroxyl on the F-SiO2 surface could form a hydrogen bond network, rendering a further expedited electron transfer process. More importantly, owing to the thermal stability of F-SiO2 and the facilitated electron transfer enabled by fluorinated functional groups, the F-SiO2 could exhibit higher CEC efficiency than polymers could do at elevated temperatures. This was exemplified by the CEC-leaching of cathode materials from spent ternary LIBs at 70 °C, where leaching efficiencies for all elements exceeded 90% within 300 min. Besides, by using dimethyl sulfoxide (DMSO) as the solvent, we have investigated the CEC performance of F-SiO2 under further higher temperatures (180 °C). To be specific, the removal ratio of methyl orange in the F-SiO2 group is 6.13 times higher than that in the FEP group under 180 °C, further suggesting the outperformance of F-SiO2 under elevated temperatures. Considering this fluorination approach is feasible for a variety of oxides, we expect this strategy could not only expand the selection range of effective CEC catalysts, but also enable the application of CEC for a series of significant reactions that usually occur at high temperatures.

Methods

Instrumentation and chemicals

Materials were obtained from commercial suppliers and were not subject to further modifications unless otherwise noted. Methyl orange [C14H14N3NaO3S, Macklin, 98%], p-benzoquinone [C6H4O2, Macklin 99.5%], ter-butanol [C4H10O, Macklin, 99.9%], SiO2 [2 μm, Macklin, 99.99%], Al2O3 [5 ~ 6 μm, Macklin, 99.99%], MgO [100 nm, Macklin, 99.99%], ZrO2 [2 μm, Macklin, 99.99%], FDTES [C16F17H19O3Si, Macklin, 96%], terephthalic acid (THA) [C8H6O4, Macklin, 99%], nitro blue tetrazolium (NBT) [C40H30N10O6·2Cl, Macklin, 98%], and 5,5-Dimethyl-1-pyrroline N-Oxide [Dojindo, C6H11NO] were purchased from the corresponding suppliers. Spent cathode sheets were purchased from a local battery recycling company. All reactions were conducted using ultrasonic baths with 120 W and 40 kHz (Yumeng, YM-020T).

Sample preparation

The oxide powder was first rinsed with ethanol and ultrapure water separately and then dried in an oven at 40 degrees overnight. In the meantime, FDTES was added into absolute isooctane to reach a final concentration of 5 mM. Afterwards, 5 g oxide powder was stirred with 200 mL as-prepared FDTES solution for 24 h for surface fluorination. The powder after fluorination was rinsed by isooctane and then isopropanol before dried overnight.

A 5-ppm aqueous solution of methyl orange was prepared by dissolving 5 mg of methyl orange (C14H14N3NaO3S) in 1 L of ultrapure water, followed by magnetic stirring for 1 hour.

For the degradation of methyl orange at 180 °C, 5 mg methyl orange was dissolved in 1 L dimethyl sulfoxide (DMSO) solution, and 50 mL of the as-prepared solution, as well as 20 mg F-SiO2, were transferred together into a jacketed beaker that was connected to a thermostat. The temperature during ultrasonication was controlled at 180 ± 1 °C.

20 mg of oxide powder were added into 50 mL as-prepared methyl orange aqueous and then magnetic stirred at 1000 rpm for 48 h before ultrasonication. The temperature during ultrasonication was regulated. Aliquots were sampled at 0, 5, 10, 15, 30, 60, 90, 120, 180 mins intervals during ultrasonication for the subsequent UV-Vis measurements.

The powders after reactions were separated from the solution using centrifugation at 6948 g for 30 min. The filtered powders were then dried in an oven at 40 °C overnight before analysis.

A terephthalic acid solution was prepared by dissolving 332.4 mg of p-phthalic acid and 760 mg of sodium phosphate tribasic dodecahydrate in ultrapure water. The nitro blue tetrazolium solution was prepared by dissolving 5 mg of nitro blue tetrazolium in 1 L of ultrapure water, followed by magnetic stirring for 1 h.

FTIR samples were prepared by grinding 10 mg of oxide powder with 100 mg of KBr and pressing the mixture into a pellet.

Samples for EPR analysis were prepared by mixing 25 ml of ultrapure water or dimethyl sulfoxide (DMSO) and 10 mg of oxide powders. Approximately 0.25 mL of DMPO was added to the solution prior to ultrasonication.

Sample characterization

The absorbance of the aliquots was measured using a Cary 3500 UV-Visible Multicell spectrophotometer, with a scanning wavelength range of 250–650 nm. The samples were placed in a Hellma Analytics QS high-precision cell with a light path of 10 mm.

X-ray photoelectron spectroscopy (XPS) measurements were performed using a Thermo Fisher Scientific K-Alpha. The Al kα ray source (hv = 1486.6 eV) was operated in a vacuum of 1 × 10−9 mBar, with an operating voltage of 15 kV and a filament current of 10 mA. The pass energy was set at 30 eV.

The Raman spectroscopy analysis was conducted on a LabRam HR evolution (HORIBA, SAS France), using a range from 300 to 1400 cm−1.

FTIR analyses were conducted on a Bruker Vertex 80 v on a range from 400 to 3000 cm−1.

The emission spectra of THA-OH were recorded on an Edinburgh Instruments FLS 980, using λexcitation = 225 nm and λemission = 425 nm.

Electron paramagnetic resonance was recorded on a Bruker EMX plus-9.5/12/P/L. The measurements were performed in X-band (9.812056 GHz) with amplitude modulation of 1 G, microwave power of 20 mW, amplitude modulation frequency of 100 kHz, conversion time of 26.70 ms, and time constant of 0.01 ms.

Thermogravimetric (TG) analysis was conducted using the SDT Q600 synchronous thermal analyzer from TA Instruments, USA. The temperature range is from room temperature to 200 °C with a heating rate of 10 °C/ min. The weighing resolution is 0.1 μg, and the DSC resolution is less than 1 μW.

In-situ FTIR analysis was conducted on a Frontier FT-IR Spectrometer (PerkinElmer) equipped with a demountable gas cell. The sample was heated to the target temperature and maintained for 15 min prior to measurement.

Pretreatment of spent LIBs

The discharged cathode sheets were obtained from a local battery-recycling company (Beijing, China). These electrode sheets were first cut into small pieces (4 cm × 4 cm) and then placed in a crucible and heated in a muffle furnace at 600 °C for 60 min to remove cross-linking agents (PVDF and others). The ternary cathode material was then peeled off from the aluminum film, followed by grounding into powder using a mortar and pestle for subsequent leaching experiments.

CEC-leaching of target metals

For metal leaching experiments, 20 mg of catalysts and 400 mg of the spent cathode materials were added to 40 ml of 0.2 M aqueous citric acid solution. The as-prepared solution was magnetically stirred for 1 h and then placed in an ultrasonic bath (40 kHz, 120 W) for 5 h at 70 °C, with a sampling interval of 1 h. After ultrasonication, the aliquots were centrifuged at 10004 g for 3 min. Metal concentrations in the filtrate were measured by inductively coupled plasma optical emission spectrometry (ICP-OES).

Regeneration of cathode materials

Based on the metal leaching efficiency obtained by ICP-OES, nickel acetate, cobalt acetate and manganese acetate were added to the leaching solution to achieve a final ratio of 6:2:2 in the amount of substance of the corresponding metals. The resulting solution was magnetically stirred (1000 rpm/min) at 80 °C for 12 h. Subsequently, the precipitated metal compounds were placed in a crucible and placed in a muffle furnace and heated to 900 °C and kept for 6 h. The resulting electrode material was subjected to XRD characterization.