Introduction

Rechargeable Aqueous Zn-ion batteries (RAZIBs), as one of the multi-valent ion batteries, have garnered widespread attention due to their high safety, low cost, and environmental friendliness. These characteristics make them a highly promising candidate in next-generation efficient rechargeable batteries1,2,3. Given the high theoretical specific capacity (820 mA h g−1 and 5855 mA h cm−3) and relatively low electrochemical potential (−0.76 V vs. Standard Hydrogen Electrode) of zinc anode, developing matchable cathodes with higher operating voltage and accelerated ion diffusion has become a critical challenge, which impedes the practical applications4,5. In the past, a variety of materials have been developed to serve as the high-specific capacity cathodes for RAZIBs, including manganese-based oxides6, Prussian blue analogues7, polymers8, and vanadium-based compounds9,10. However, scarce studies focus on the electrochemical potential of the electrode materials and decipher its determinators, especially the functions of the local chemical coordination and crystal field. Generally, redox ions and their coordination build structural units such as octahedra and tetrahedra to influence the configuration/occupation characteristics of d-orbitals participating into redox reaction11. Therefore, addressing the urgent challenges of manipulating the coordination environment of redox ions and exploring the correlations between the d-orbitals and electrochemical potential is crucial for developing high-energy density RAZIBs.

Hydrated vanadium pentoxide (V2O5·nH2O, VOH) has a bi-layered crystal structure with variable chemical states, offering high specific capacity and regulable structure characteristics12. Strategies have been explored to improve the comprehensive electrochemical performance of VOH, such as the doping of alien cations13, the incorporation of conductive polymers14,15,16,17 and the application of surface coating18. In particular, pre-intercalation, which involves the insertion of foreign guests into the host lattice, has demonstrated considerably effective way of regulating the chemical coordination environment19. With either individual or joint pre-intercalating of metal ions20, non-metal ions/molecules, and organic compounds21, the performance of pre-intercalated cathode could experience significant enhancements due to the expanded interlayer spacing22, shielded electrostatic interaction between the charge carriers and the lattice oxygen23, and enhanced electrical conductivity24. The bulk of these studies concentrate on fast mass transfer to achieve a greater specific capacity and improved rate performance. However, the correlations between the local changeable chemical coordination and the enhanced electrochemical potential in detail are unclear and need to be explored carefully, especially the changes on the electrochemical potentials of VOH.

In this work, we propose that manipulating the coordination environment in VOH alters the V 3 d orbital configurations, thereby modulating the electrochemical potential. Thus, organic cations pre-intercalated VOH are synthesized by a facile hydrothermal reaction with choline and derivatives as the additives to induce the [VO6] octahedra distortion and V 3 d orbitals splitting. The sample with greater distortion achieves a higher operating voltage and rapid ion migration together in RAZIBs. By employing X-ray absorption spectroscopy (XAS) and pair distribution function (PDF), the local coordination environment distortions were analyzed to decipher the relationship between the coordination environment, electronic configurations, and electrochemical performance. This work provides guidance for the design and development of vanadium oxide with high voltage and rapid ion diffusion for high-performance RAZIBs.

Results and discussion

Structure and chemical states

The samples were synthesized through a hydrothermal method and the details are described in the method section. The samples with intercalated choline (C5H13N+OH, Ch) and chlorcholine (C5H13N+Cl, Ch2) cations are denoted as Ch-VOH and Ch2-VOH, respectively. Figure 1a presents the XRD patterns of Ch2-VOH and Ch-VOH, and the strongest peak locates at 6.76° in Ch2-VOH associating to the (001) plane with the lattice spacing of 13.1 Å. In contrast, the (001) plane diffraction peak of Ch-VOH appears at 2θ = 6.8°, indicating a lattice spacing of 12.9 Å. The lattice spacings of (001) plane of both samples exhibit an increase compared to 11.5 Å of pristine VOH (JCPDS 40-1296)25. This observation is further corroborated by the HRTEM images presented in Supplementary Fig. 1. The results from the energy dispersive X-ray spectrometer (EDS) mapping reveal that the chlorine (Cl) signal was exclusively detected in Ch2-VOH, demonstrating a discrepancy in the pre-inserted cations. In the electron paramagnetic resonance (EPR) spectra, both Ch2-VOH and Ch-VOH exhibit symmetric signals at g value = 2.003 (Supplementary Fig. 2a). These signals correspond to unpaired electrons captured at the oxygen vacancies26, providing further evidence of the successful pre-insertion of organic cations. Raman spectra in Supplementary Fig. 2b display four peaks at approximately 990, 690, 406, and 281 cm−1 in both samples. These peaks correspond to the stretching vibration of V = O (990 cm−1) and V2-O (690 cm−1), respectively, as well as the bending vibration of V = O (406 and 281 cm−1)20. Fourier transform infrared (FTIR) spectra in Supplementary Fig. 2c identify that the peaks around 738, 978 and 1005 cm−1 represent the characteristic bands of vanadium oxide framework27. The peak observed at around 738 cm−1 is associated with the stretching vibrations of V-O-V. Meanwhile, the peaks appeared near 1005 and 978 cm−1 are attribute to V = O, which is split into two peaks corresponding to the stretching vibration of V5+ = O and V4+ = O, respectively28. The peak located around 1623 cm−1 in Ch-VOH corresponds to the bending vibration of H–O–H. However, the absence of a distinct bending vibrational peak at the same position in Ch2-VOH indicates that it contains less interlayer water compared to Ch-VOH. In the thermogravimetric analysis - differential scanning calorimetry (TG-DSC) curves (Supplementary Fig. 3), the amounts of pre-inserted cations C5H13N+Cl and C5H13N+OH are nearly identical, approximately 9.0 and 9.1 wt %, respectively. After heating to 350 °C, Ch-VOH exhibits a lower residual mass, corroborating the presence of more interlayer water in Ch-VOH. In addition, a vibrational peak belonging to -OH is observed around 1414 cm−1 only in the Ch-VOH spectrum, further confirming the difference in pre-inserted cations for both samples. Figure 1b, c shows the V K-edge X-ray absorption spectroscopy (XAS) spectra of Ch2-VOH and Ch-VOH. Notably, the pre-edge peak corresponds to 1 s to 3 d electronic transition, demonstrating high sensitivity to the crystal field symmetry as well as to the electron occupied state of the V 3 d orbitals. According to the dipole forbidden principle, the intensity of pre-edge peaks for both samples should be weak or undetected. On the contrary, the pre-edge peak intensities of the samples are strong, indicating the occurrence of orbital hybridization between the O 2p and V 3 d orbitals29. The increase in pre-edge peak intensity from Ch-VOH to Ch2-VOH suggests an increase in the number of 3 d unoccupied orbitals and a greater level of orbital hybridization between O 2p and V 3 d orbitals in Ch2-VOH. Additionally, the findings indicate that the pre-intercalation of C5H13N+Cl induce greater distortion of [VO6] octahedra30. Furthermore, the main absorption edges of Ch-VOH and Ch2-VOH also confirm that V in Ch2-VOH processes a higher oxidation state (Fig. 1c).

Fig. 1: Structure information of Ch2-VOH and Ch-VOH.
figure 1

a XRD patterns. Both samples have the same crystal structure but different spacing distances of (001) crystal plane. b The V K-edge XAS spectra of V foil and V oxides with different oxidation states. The pre-edge peak reflects the octahedra distortion and the 1 s → 3 d electron hopping, the near edge presents the chemical states of V in the different chemical surroundings. c The enlarged image of absorption edge from Fig. b. The change in oxidation state is reflected by the almost 3 eV difference in rising edge energies between V2O3 and V2O5. d V K-edge EXAFS (point) of Ch-VOH and Ch2-VOH and their fitting curves (line) in R-space. The data are K2-weighted without phase-corrected. e O K-edge soft XAS spectra of Ch-VOH and Ch2-VOH. The corresponding wavelet transform diagrams of (f) Ch-VOH and (g) Ch2-VOH. The V-O chemical coordination have changed in the ab plane of the octahedra.

Full size image

The K2-weighted Fourier transforms of the V K-edge extended X-ray absorption fine structure (EXAFS) and the corresponding fitting curves simulated by Artemis31 provide the coordination information of the central atoms (Fig. 1d, e). The fitting on R space of both samples adopt identical structure (Supplementary Fig. 4a) without phase-correction. The slabs of V2O5 bilayers are separated by water molecules and stacked along the c-axis. It has a monoclinic unit cell (space group C2/m) with parameters a = 11.722(3) Å, b = 3.570(3) Å, = 11.520(3) Å, and β = 88.65°. The stacking sequence shows signatures of turbostratic disorder and a structural coherence limited to 50 Å32, and the parameters are detailed in the Supplementary Table 1. The radial structure functions corresponding to the central absorption vanadium atom emphasize that the interlayer spacing increases and the distance from oxygen located at the vertex to vanadium become short after the cation pre-intercalation. The peak at 1.1 Å originates from the single scattering contribution of oxygen along the c-axis (Supplementary Fig. 6). Peak at 1.6 Å corresponds to the cooperative scattering from the four O atoms at the equator within the [VO6] octahedra33. The octahedral crystal field distortion caused by the pre-inserted will lead to a secondary splitting of the 3 d orbitals. Compared to Ch-VOH, there are more shorter V-O along the c-axis in Ch2-VOH (Supplementary Tables 2, 3). This results in a greater crystal field repulsion along the c-axis in Ch2-VOH. In addition, the peak at around 2.8 Å arises from scattering contributions of V-V between adjacent octahedra. Note that the radial distance of V-V in both samples remains unchanged before and after intercalation, suggesting a rigid and stable framework in the whole host structure. The V L-edge and O K-edge soft XAS of Ch-VOH, Ch2-VOH, and VOH are presented in Fig. 1e and Supplementary Fig. 7. The V L-edge spectra (Supplementary Fig. 7a) correspond to dipole selection rules that permit transitions from a defined ground state (2p63dn) to a potential final state (2p53dn+1). The spectra display two peaks around 518 eV and 525 eV, which can be attributed to the 2p3/2 and 2p1/2 energy levels, respectively. The O K-edge spectra represent transitions from the O core level to the unoccupied conduction band that hybrids with O 2p states. Different features in the spectra primarily correspond to bands of metal orbital energy levels. The prominent doublets around 528–537 eV are attributed to V 3 d bands split by the crystal field effect. The doublet splitting in the spectra provides detailed information about V 3 d bands, which can be explained qualitatively in terms of symmetry34. As shown in Supplementary Tables 2, 3, the shortest V-O bond in Ch2-VOH has a higher coordination number compared to Ch-VOH, leading to larger secondary splitting of the t2g orbitals35. Consequently, the dxy orbital of Ch2-VOH has a lower energy level than that of Ch-VOH. The corresponding wavelet transform diagrams in Fig. 1f, g reveal that the strongest peaks stem from the V–O single scattering contribution (1 st coordination sphere), and the other peaks correspond to the V–V (2nd sphere), and V–O (3rd sphere) scattering contribution, respectively.

Furthermore, the synchrotron X-ray atomic pair distribution function (PDF) analysis is performed on these samples. The PDF fits are performed over two different regions, i.e., 1.2 to 11.5 Å for intralayer structure modeling (Fig. 2a–d), and 1.2 to 21.0 Å for interlayer structure modeling (Supplementary Fig. 9), and detailed parameter values are reported in Supplementary Tables 4–6. For r < 11 Å, the PDF can be fitted well by the model. However, for interlayer regions (11 < r < 21.0 Å), the model starts to deviate from the data. Our results are consistent with the literature reports for pristine VOH, indicating that the intra-layer structure of Ch-VOH and Ch2-VOH can be well described, but the interlayer turbostratically disordering exits along the c-axis, and the well-ordered interlayer model can only describe about three layers, as by the refined correlation lengths about 4 nm in the Supplementary Table 7. The local structure distortions in Ch-VOH and Ch2-VOH are confirmed by both the qualitative and quantitative PDF analysis. The PDF data shows that Ch-VOH and Ch2-VOH are different from the pristine VOH, reflecting changes in the local environment and confirming the successful pre-intercalation of organic cations into the samples. The V2O5·nH2O structure model (Supplementary Fig. 4b), features four oxygen atoms at distances ranging from 1.79 to 2.10 Å forming a plane with the central V atom, while oxygen atoms are with ~1.60 Å above and ~2.50 Å below the plane of the V atom. The variation in the lengths of these V-O bonds results in distinguishable splitting in the experimental PDF peak at about ~1.6 Å (Fig. 2e), demonstrating the sufficient real-space resolution of PDF for detecting-structure changes and distortions in the local atomic environment. The PDF peak positions directly reflect the bond length, and the area underneath peaks are related to the atomic coordination environment. Through the structural simulation of PDF, the influence of varying atomic bond lengths on the peaks becomes distinguishable. The first peak at 1.85 Å corresponds to the V-O bonds, where both Ch-VOH and Ch2-VOH exhibit deviations from the pristine sample, indicating octahedral distortions in both samples. Particularly, the deviation of V-O bonds in Ch2-VOH is larger, meaning greater distortion of the octahedra. The second peak at 2.53 Å, representing the nearest O-O bonds, which shifts to low-r region for both pre-intercalated samples. The third peak (3.01 Å) mainly arises from the V-V bonds, with a small contribution from the O-O bonds, and the peak in Ch-VOH exhibits slightly decrease. The fourth peak, located at 3.57 Å, mostly originates from V-V pair distances along a and b axes. These V cations form the motif of the octahedra plane, and the PDF analysis reveals that both the position and the area of this peak remain almost unchanged, indicating the interlayer structure motif or skeleton is intact. Therefore, PDF analysis suggests that the changes of O atoms sites are the primary cause of the structural distortions in Ch-VOH and Ch2-VOH. The comprehensive description of the PDF experiment can be found in the supplementary information (Si).

Fig. 2: Local structure information of VOH, Ch-VOH and Ch2-VOH.
figure 2

Experiment PDF data for pristine VOH, Ch-VOH, and Ch2-VOH samples over the range of: (a) 1.3 ~ 21.0 Å. The experimental X-ray PDF data are fit by the VOH crystal model over the range of 1.2 < r < 11 Å: (b) VOH; (c) Ch-VOH; (d) Ch2-VOH. e The experimental X-ray PDF data at rang of 1.3 ~ 4.2 Å. The changes of O atoms sites are the primary reason for the structural distortions in Ch-VOH and Ch2-VOH.

Full size image

The aforementioned results confirm that the pre-intercalation of organic cations into the interlayer of VOH, as illustrated in Fig. 3a. And the interaction between pre-inserted cations and oxygen induces a distortion of [VO6] octahedra (Fig. 3b) and results in the secondary splitting of 3 d orbitals as shown in Fig. 3c. Under an ideal octahedral coordination field, the first splitting of 3 d orbitals occurs and the five-fold degenerate V 3 d orbitals split into lower-energy three-fold degenerate t2g orbitals and higher-energy two-fold degenerate eg orbitals. Due to the compression along c-axis as verfied by EXAFS and PDF analysis, the crystal field repulsion of Ch2-VOH along the c-axis was increased, leading to a larger secondary splitting energy and lower energy level of the 3 dxy orbital compared to Ch-VOH. With nearly the same amount of cations insertion, Ch2-VOH exhibits a larger lattice spacing, less interlayer water, greater distortion of the [VO6] octahedra.

Fig. 3: Schematic of the structure and orbitals.
figure 3

a Organic cations pre-intercalated hydrated vanadate. The crystal structure keeps the original bi-layered framework and the cations replace partial water and occupy the sites. b The distortion in the [VO6] octahedron. Pre-intercalated cation interacts with the oxygen from [VO6] octahedra and induce the compression along c-axis as verified by XAFS and PDF. c 3 d orbitals splitting of vanadium ion in Ch2-VOH and Ch-VOH. The different interaction and distortion in the [VO6] octahedra induce the different secondary splitting energy, causing different energy levels in orbitals. The lower energy level of 3 dxy orbital offers a higher operating voltage in the redox reaction.

Full size image

X-ray photoelectron spectroscopy (XPS) was used to explore the chemical composition and states of the as-prepared materials with different pre-inserted cations. As shown in Fig. 4a, five peaks in XPS survey spectra correspond to V 1 s, O 1 s, V 2p, N 1 s and C 1 s, respectively. Compared to Ch-VOH, the presence of Cl signal in Ch2-VOH indicates the difference in pre-intercalated cations. In Fig. 4b, the valence band spectra of both samples exhibit a split into two regions. The broad peak centered at around 6 eV is attributed to O 2p, while the narrow and low-intensity region around 1.5 eV belongs to V 3 d36. The occupied V 3d-band width was detected by the X-ray source at approximately 1.5 eV for Ch2-VOH and 1.4 eV for Ch-VOH. Compared to Ch-VOH, a slight shift towards higher energy in the valence band spectrum of V 3 d orbitals in Ch2-VOH is observed. This shift is attributed to the lowering of the energy level of V 3 dxy orbital in Ch2-VOH caused by the greater distortion of the octahedra. The electrons in 3dxy orbital of Ch2-VOH are more stable than those in Ch-VOH and require more energy for extraction. Moreover, the O 2p-band of Ch2-VOH shifts to a lower energy due to the hybridization between V 3d and O 2p37. The high-resolution Cl 2p spectrum of Ch2-VOH splits into two peaks as shown in Fig. 4c. One peak at 199.3 eV corresponds to C-Cl, and the other at 200.8 eV corresponds to Cl 2p1/238. In the V 2p spectra (Fig. 4d), two pairs of peaks appear in the range of 513–527 eV, indicative of V4+ and V5+, respectively. The peak area ratio of V4+ in Ch-VOH closes to that of Ch2-VOH, indicating approximately equal amount of both cations pre-inserted in both samples during the preparation process. In the O 1 s spectra (Fig. 4e), two peaks in the range of 526–534 eV for Ch2-VOH correspond to V-O (529.5 eV) and oxygen vacancies (530.5 eV)39. In addition to the observed peaks of V-O (529.5 eV) and oxygen vacancies (530.4 eV), Ch-VOH exhibits an additional peak at 531.7 eV attributed to interlayer water molecules24, in comparison with Ch2-VOH. These results suggest that C5H13N+Cl cations are more effective than C5H13N+OH cations in terms of excluding interlayer water, even though the pre-insertion amounts of the two cations are nearly identical. This result supports the analysis from TG-DSC. In the N 1 s spectra (Fig. 4f), the signal at around 399.4 and 398.9 eV correspond to C-N bonds from pre-inserted organic cations. Another peak at around 401.5 eV originates from the N-O bonds formed by C5H13N+Cl or C5H13N+OH cations and O in the vanadium oxide layer40. This further verifies that the organic cations of C5H13N+Cl or C5H13N+OH are pre-inserted into the VOH lattice, with the chemical bonds forming between N from the organic cations and O within the [VO6] interlayers of vanadium oxide. Notably, the formation of the N-O bond is beneficial in attenuating the electrostatic attraction between O2− and Zn2+ ions. As described above, the pre-insertion of different organic cations regulates the chemical state and adjusts the energy level of 3 d orbitals.

Fig. 4: Chemical states of elements and valence bands of both samples.
figure 4

a XPS survey spectra. The signal from element Cl verifies the difference of intercalants. b Valence band spectra. The occupied V 3d band in Ch2-VOH presents a slight shift toward high binding energy because of the lower energy level of dxy after the secondary splitting. c High resolution spectrum of Cl 2p collected from Ch2-VOH. It confirms the C-Cl connection in the molecular chain. d V 2p spectra. Organic intercalants introduce V4+ in both lattices because of charge balance, confirming the successful pre-intercalation. e O 1 s spectra. The detected water signal suggests a more competitive expulsion from C5H13N+Cl on lattice water compared to C5H13N+OH. f N 1 s spectra. The presence of C-N bonds from the skeletons of organic cations and the formation of N-O bonds suggest the strong chemical interaction between the intercalants and the lattice oxygen.

Full size image

Reaction kinetics and electrochemical performance

The electrochemical performance was assessed using a zinc plate as the anode and a 3 M Zn(OTF)2 (OTF− = CF3SO3−) aqueous solution as the electrolyte. The galvanostatic charge/discharge (GCD) profiles recorded at a specific current of 1 A g−1 are shown in Fig. 5a, the clearly visible charge/discharge platforms at 1.1 ~ 0.8 V and 0.8 ~ 0.3 V are observed in the Ch2-VOH curves, relating to the insertion/extraction of active Zn2+. Importantly, the mid-point voltage of Ch2-VOH is significantly higher than that of Ch-VOH due to the difference in the energy levels of the 3 d orbitals for the two samples. As discussed above, the [VO6] octahedra in Ch2-VOH exhibit greater distortion than those in Ch-VOH, resulting in a larger splitting energy in the V t2g orbitals of Ch2-VOH (Fig. 3c). Consequently, the 3dxy orbital of Ch2-VOH demonstrates a stronger electron affinity than that of Ch-VOH, leading to a higher voltage platform compared to Ch-VOH. At the specific currents of 1, 2, 4, and 8 A g−1 (Supplementary Fig. 10), the discharging voltage platforms of Ch2-VOH are higher than those of Ch-VOH, confirming the superior voltage outputs in Ch2-VOH. The CV contour plots of both samples are shown in Fig. 5b, c. The cathodic peaks at ~0.93 V (peak 4) and ~0.45 V (peak 3) correspond to the reduction of V5+ to V4+ and V4+ to V3+, respectively. The anodic peaks at ~0.73 V (peak 1) and ~1.16 V (peak 2) during the anodic sweep originate from the transition from V3+ to V4+ and V4+ to V5+ in the charging process. Ch2-VOH cathode shows a higher current response compared with Ch-VOH, especially in the region of V4+/V5+ transition41. Meanwhile, with the decrease in scan rates, the current response of the Ch2-VOH cathode remains evident, indicating excellent Zn2+ diffusion kinetics. Moreover, Ch2-VOH exhibits a higher response current peak compared to Ch-VOH, indicating a higher specific capacity. In addition, the open circuit voltage (OCV) of Ch2-VOH is higher than that of Ch-VOH (Fig. 5d) due to the lower energy of V 3dxy orbital of Ch2-VOH than that of Ch-VOH. Galvanostatic intermittent titration technique (GITT) was performed to estimate the Zn2+ diffusion coefficients of both samples. The Zn2+ diffusion coefficients (DZn2+) of Ch2-VOH and Ch-VOH, range from 10−8 to 10−10 cm2 s−1 in Fig. 5e, are superior to those of previously reported vanadium-based cathodes in Zn2+ storage42,43. During the discharge process, the DZn2+ of Ch2-VOH is measured within a range of 5.0 × 10−9 ~ 6.3 × 10−10 cm2 s−1, while Ch-VOH exhibits a relatively low diffusion coefficient of 3.1 × 10−9 ~ 4.0 × 10−10 cm2 s−1. Similarly, in the charging process, Ch2-VOH displays a significantly higher DZn2+ value in comparison to Ch-VOH. It is further verified that the distorted structure of Ch2-VOH caused by the pre-inserted C5H13N+Cl enables rapid migration of Zn2+and improves the reaction kinetics. Figure 5f illustrates the rate performance of Ch2-VOH and Ch-VOH, where the specific current is gradually increased from 0.5 to 8 A g−1 and subsequently reverted to 0.5 A g−1. At a specific current of 8 A g−1, Ch2-VOH retains 60.3% of its initial capacity measured at specific current of 0.5 A g−1, whereas Ch2-VOH is only 57.8%. Moreover, the discharge capacity of Ch2-VOH recovers to 404 mA h g−1 when the specific current returns to 0.5 mA g−1, suggesting its excellent rate capability. The energy efficiency of Ch2-VOH is evidently higher than that of Ch-VOH (Fig. 5g). The enhanced efficiency is attributed to the higher average voltage and rapid ion diffusion of Ch2-VOH acquired by the modulation of local chemical coordination. Figure 5h shows the cycling performance at a current density of 4 A g−1. The maximum capacity of Ch2-VOH and Ch-VOH batteries reach 285.6 mA h g−1 and 285.9 mA h g−1, respectively, after the period of activation caused by electrolyte penetration. After 1000 cycles, the capacity of the Ch2-VOH battery shows almost no attenuation, maintaining an impressive 98.8% of its maximum capacity. In contrast, the capacity of the Ch-VOH battery decays to 241 mA h g−1, retaining 85.6% of its maximum capacity. In comparison to other reported cathode materials19,44,45, Ch2-VOH exhibits excellent cycling life and Zn2+ diffusion kinetics. Overall, Ch2-VOH cathode presents higher open circuit and operating voltages, and preferable electrochemical reversibility as well as better cycling stability, which are attributed to the distorted [VO6] octahedra and open structure generated by C5H13N+Cl cations pre-intercalation.

Fig. 5: Electrochemical properties Ch2-VOH and Ch-VOH cathodes in aqueous zinc ion batteries.
figure 5

a The GCD curve at a specific current of 1 A g−1. CV contour of (b) Ch2-VOH and (c) Ch-VOH. The larger area and stronger intensity at the same scan rate reflect Ch2-VOH with better reaction kinetics. d Open circuit voltage. The higher midpoint voltage and OCV of Ch2-VOH verify the impacts of local coordination manipulation. e Zn2+ diffusion coefficients. The Zn2+ diffusion coefficients of Ch2-VOH are higher than that of Ch-VOH in the charging and discharging process under a specific current of 50 m A g−1. f Rate performance. The enhanced capacities of Ch2-VOH at high current densities are offered by the fast ion diffusion and charge transfer. g Energy efficiency at the different current densities. The smaller polarization from fast mass transfer enables the higher energy efficiencies. h Cycling performance at specific current of 4 A g−1. The crystal structure of Ch2-VOH become more robust in comparison with Ch-VOH.

Full size image

In summary, this work establishes a model linking the local chemical coordination and electron orbital occupation with electrochemical performance on the basic of crystal field theory. V2O5·nH2O pre-intercalated with chlorcholine cation with a greater local distortion achieves excellent Zn2+ diffusion kinetics and elevated operating voltage as well as enhanced cycling stability in comparison with pre-intercalated choline cation. These enhancements attribute to the induced distortion in octahedral coordination field of vanadium. The t2g orbitals of Ch2-VOH exhibit higher secondary splitting energy and the 3dxy orbital possesses a lower energy level. Thereby, it elevates the open circuit voltage from 1.55 V to 1.68 V, and the mid-point voltage presents an increase of 50 mV. In addition, the strong interaction between the organic cation and oxygen from the [VO6] octahedra stabilize the bi-layered structure of hydrated vanadate, achieving an impressive capacity retention of 98.8% at specific current of 4 A g−1 over 1000 cycles. This work provides insightful guidance for the voltage regulation in the cathode materials of rechargeable batteries.

Methods

Synthesis of samples

All chemicals were used as received without further purification steps.

To prepare the samples, 2 mmol of V2O5 (99.6 + %, supplier: Sigma-Aldrich) was dissolved in 50 mL of deionized water (DI H2O) with 2 mL of H2O2 (30%, supplier: Aladdin). Separately, 1 mmol of C5H13Cl2N or C5H14ClNO (supplier: Aladdin) was dissolved in 30 mL of DI water. Two solutions were mixed and filled into a 100 mL of Teflon lined stainless steel autoclave (internal diameter: 42 mm, height: 72 mm), then heated and maintained at 120 °C for 6 h.

Afterward, the precipitates were obtained by centrifugation, followed by washing with water and ethanol three times. The collected precipitates were dried overnight in an electric oven at 70 °C, resulting in greenish powders. The resulting products were further dried at 120 °C in a vacuum oven for 2 h.

Preparation of cathode host and battery assembly

For electrode preparation, the active material, conductive carbon, and polyvinylidene fluoride (PVDF) binder were mixed with a weight ratio of 7:2:1 in N-methyl-2-pyrrolidone (NMP) solvent to produce a slurry. This slurry was coated onto a titanium foil serving as the current collector. The resulting electrodes were subsequently dried in a vacuum oven at 60 °C overnight. The active material was loaded onto the electrodes at a density ranging from 1 to 2 mg/cm2. To prepare coin batteries (2032 type with 316 stainless steel), Zn metal disc was utilized as the anode with a diameter of 14 mm, and the batteries were filled with 80 μL of 3 M Zn(Otf)2 aqueous solution as the electrolyte. Separator between electrodes was a glass fiber filter of Whatman, Grade GF/A.

Characterizations

The crystal structure of the samples was identified by Cu Kα radiation X-ray diffraction (XRD, SmartLab, 3 kW). The bonding properties of the samples were analyzed by Fourier transform infrared (FT-IR, Bruker ALPHA) absorption spectroscope and Raman spectroscope (Thermo Scientific DXR 2xi Micro-Raman). Transmission electron microscopy images were obtained by transmission electron microscope (TEM, JEOL, 3200fS) with an accelerating voltage of 300 kV. X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi) were used to collect the surface chemical states of the samples. The experimental PDF data are carried out at BL08W at Super Photon ring−8 GeV (SPring-8) in Japan. The soft-XAS of the V L-edge and O K-edge was performed by National Synchrotron Radiation Research Center (NSRRC) in Taiwan at BL20A1 beamline. V K-edge X-ray absorption spectra were collected using an easyXAFS300+ benchtop instrument (easyXAFS, LLC), using a Ge (211) ball-bending crystal analyzer and an Ag anode X-ray tube. Spectra were deadtime corrected and energy were calibrated using a standard V metal foil. Thermogravimetry - differential scanning calorimetry (TG-DSC) analysis was conducted by Jupiter Netzsch, 449F3 with a temperature range of 40–700 °C in a flowing of Air. Electron Paramagnetic Resonance (EPR) was performed by a Bruker A300 spectrometer.

Electrochemical measurements

All electrochemical measurements were conducted at 25 °C and at the voltage range of 0.2–1.6 V. The redox reactions of the cathodes were detected by cyclic voltammetry (CV). In addition, the galvanostatic intermittent titration technique (GITT) and galvanostatic charge and discharge assessments were used to analyze reaction kinetics and cycling stability on a Neware tester (MIHW-200-160CH-B). GITT tests were carried out at a current density of 50 milliamperes per gram and involved a charge/discharge duration of 10 min for each step.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.