An Insoluble Amino-Functionalized Hexaazatriphenylene as Stable Organic Cathode in Lithium-Ion Batteries

Abstract

Organic electrode materials have received increasing attention in rechargeable batteries due to their earth abundance and variable structures. However, the practical application of most organic electrode materials is limited by the high solubility in the electrolyte. Herein, an insoluble amino-functionalized hexaazatriphenylene (defined as HATN-[NH₂]₃) in the electrolyte is developed as stable organic cathode material for lithium-ion batteries (LIBs). The resultant HATN-[NH₂]₃ electrode achieves a high reversible capacity of 192.5 mAh g⁻¹ at a current density of 0.05 A g⁻¹. Remarkably, the electrode exhibits almost no capacity fade after 500 cycles at 0.5 A g⁻¹. The high stability can be ascribed to insoluble property caused by hydrogen bonds between HATN-[NH₂]₃ molecules. Moreover, density functional theory calculations suggest that amino functionalization can reduce the band gap of HATN, in favor of improved conductivity and thus enhanced rate performance. This work offers a simple but efficient strategy to develop stable organic electrode materials in LIBs and beyond.

1. Introduction

With the increasing demand of high energy density batteries for electric vehicles and smart grids, high-capacity electrodes (especially for cathode) with high stability are highly desired [1,2]. Apart from transition metal-based cathode materials, organic cathode materials have attracted considerable attention for high energy density batteries because of their high theoretical specific capacity, flexible structural design, high safety, and earth abundance [3,4,5,6,7,8,9]. However, their practical applications in lithium-ion batteries (LIBs) were restricted by the poor cycle stability mainly caused by large solubility in organic electrolytes [10]. A series of strategies, including the polymerization of redox-active organic molecules, the development of inherently insoluble materials, and the recombination of organic molecules with carbon nanomaterials, have been developed to solve the dissolution issues [11,12,13,14,15,16,17,18,19]. Among these strategies, the development of inherently insoluble materials is the most efficient strategy because of its simple synthesis, reduced capacity with a minimum, and various structural design [20,21,22,23,24,25]. For example, Wang et al. reported a small molecule (2,3,7,8-Tetraaminophenazine-1,4,6,9-tetraone) with rich C=O, C=N, and NH₂ groups, which was used as a high-performance organic material for symmetric all-organic batteries. These abundant polar groups benefit to form strong intermolecular interactions, resulting in insolubility and high stability. Though substantial progress has been made, the stability of previously reported organic cathode materials still cannot meet the demand of practical requirements. Therefore, the further development of highly stable cathode materials in LIBs is significantly important.

Hexaazatrinaphthalene (HATN) is an electron-deficient and rigid planar aromatic molecule. As a cathode material, the HATN possesses a high theoretical capacity of 418 mAh g⁻¹ for lithium storage [26,27]. However, the rapid capacity fade limits its application in practical LIBs. The complex of HATN with carbon nanomaterials is an efficient strategy for improving its electrochemical performance [26,28]. For example, the performance of HATN is significantly improved by the construction of HTAN/graphene composite via π–π complex [26]. In addition, the polymerization of HATN molecules also can resolve its dissolution issues [29]. Although these two methods can improve electrochemical performance by decreasing the solubility of HATN, the mass-energy density of the whole electrode largely decreased due to the introduction of large non-active components. Therefore, the simultaneously increasing electrochemical performance but decreasing non-active components in the HATN-based cathode is notably important. Interestingly, the introduction of polar groups on the organic molecules could decrease their solubility, obtaining a stable electrode material [21,30]. For instance, carboxyl group functionalized HATN molecules exhibited strong π–π and H-bond interactions with graphene oxide, suppressing their dissolution in electrolytes and thus enhancing the cycle stability [30]. However, the functionalized groups with high relative molecular mass led to a decreased theoretical capacity for HATN. As reported, amino groups functionalized aromatic ring molecules could largely reduce the solubility in organic solvent due to the formation of hydrogen bonds between adjacent molecules [31,32]. More importantly, the amino groups have small relative molecular mass, resulting in a high theoretical capacity of 374.8 mAh g⁻¹ for amino groups functionalized HATN (HATN-[NH₂]₃). Therefore, amino groups functionalized HATN with low solubility in the organic electrolyte may be a promising high-capacity electrode material for LIBs. To the best of our knowledge, the application of amino groups functionalized HATN as an electrode material has not been reported.

Here, we chose hexaazatrinaphthalene (HATN) as a proof-of-concept model to introduce amino groups via a group modification engineering strategy for boosting the electrochemical performance of organic cathode materials. The amino groups functionalized HATN (HATN-[NH₂]₃) were reported as a stable cathode for LIBs. Owing to the presence of three amino groups, the solubility of HATN-[NH₂]₃ in the organic electrolyte is largely decreased. Meanwhile, the amino functionalization lower band gap of the HATN-[NH₂]₃, resulting in improved intrinsic conductivity. Benefiting from the insoluble property and improved conductivity, the resultant HATN-[NH₂]₃ cathodes exhibited a high initial discharge capacity (192.4 mAh g⁻¹, current density: 0.05 A g⁻¹), long-term cycle stability with a stable specific capacity of 135 mAh g⁻¹, and high-rate capability (105 mA h g⁻¹ at 2 A g⁻¹). In addition, ex-situ powder X-ray diffraction (XRD) and X-ray photoelectron spectra (XPS) results explored the reversible reaction process and found that six lithium ions intercalation and deintercalation reversibly during the discharge/charge process due to a redox couple accessible with Li₆HATN-[NH₂]₃/HATN-[NH₂]₃.

2. Experimental Section

2.1. Materials

O-phenylenediamine (C₆H₈N₂, ≥98%), sodium sulfide hydrate (Na₂S·9H₂O, ≥98%), N-methylpyrrolidone (C₅H₉NO, ≥99.5%), Dimethyl carbonate (DMC, ≥99%) were purchased from Aladdin Reagent Co., Ltd. (Shanghai, China). 4-nitro-1,2-phenylenediamine (C₆H₇N₃O₂, ≥99%) and hexaketocyclohexane octahydrate (HKH, ≥95%) were purchased from Tianjin Xiensi Biochemical Technology Co., Ltd. (Tianjin, China). Poly(1,1-difluoroethylene) (PVDF, average Mw: 900,000) and conductive carbon black (Super P) were purchased from Hefei Kejing Material Technology Co., Ltd. (Hefei, China). The electrolyte 1M Lithium bis(trifluoromethane sulfonyl) imide (LiTFSI) in 1,3-dioxolane (DOL) and dimethoxyethane (DME) (1:1 v/v) was purchased from Suzhou Duoduo Chemical Technology Co., Ltd. (Suzhou, China). All reagents were directly used without any treatment.

2.2. Synthesis of HATN-[NH₂]₃ Molecules

The hexaazatriaphtrine with trinitro groups (HATN-[NO₂]₃) were synthesized by a reaction of 4-nitro-1,2-phenylenediamine and HKH, and then the nitro groups of HATN-[NO₂]₃ were transformed to amino groups, yielding HATN-[NH₂]₃ (C₂₄H₁₅N₉) molecules. The detailed synthetic procedures are illustrated in Supporting Information. For comparison, the HATN (C₂₄H₁₂N₆) molecules without amino groups were also prepared from the reaction of o-phenylenediamine and HKH according to the previously reported procedure [25].

2.3. Material Characterizations

Fourier transform infrared spectroscopy (FT−IR, Nicolet 6700) was collected in the range of 400–4000 cm⁻¹ using a transmission module with 4 cm⁻¹ resolution and 32 scans. Nuclei magnetic resonance (NMR) spectra were carried out on Bruker DPX 400 MHz spectrometers. Solid-state ¹³C CP/MAS NMR spectra were recorded on an Agilent VNMRS-600 spectrometer (Agilent, USA, magnetic field strength 14.1 T) at a resonance frequency of 150.72 MHz for ¹³C using the cross-polarization (CP), magic-angle spinning (MAS), and a high-power ¹H decoupling. The MAS frequency was 10 kHz. Powder X-ray diffraction (XRD) patterns were recorded by AXS D2 with a Cu Kα radiation source (λ = 0.154 nm) at a scan rate of 5° min⁻¹. Ultraviolet-visible (UV/Vis) spectra were measured by using UV 2600. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha) was conducted with 300 W Al Kα radiation. All peaks would be calibrated with C 1s peak binding energy at 284.8 eV for adventitious carbon. The thermogravimetric analysis (TGA) was performed on a Seiko Exstar 6000 instrument over the temperature range of 30 to 800 °C under a nitrogen atmosphere with a heating rate of 5 °C min⁻¹. Scanning electron microscope (SEM, SUPRA 55, ZEISS, Germany) was performed by backscattering electrons in InLens mode at an acceleration voltage of 10 kV. High-resolution transmission electron microscopy (HRTEM) and energy dispersive spectroscopy (EDS) mapping images were collected by TEM (JEOL JEM-F200) at an operation voltage of 200 kV, and the spot size is 0.8 nm.

2.4. Electrochemical Measurements

The cathode electrodes based on HATN-[NH₂]₃, containing 50 wt% of HATN-[NH₂]₃, 40 wt% of carbon black, and 10 wt% of (poly(vinylidene fluoride)) (PVDF), were prepared by casting method. The synthetic procedures are described as follows. Firstly, the active material and carbon black are uniformly mixed in a mortar, and then added to the PVDF solution. Secondly, the mixed solution was homogeneously stirred, and then coated on the carbon-coated aluminum foil by a knife coating method. Finally, the electrodes were vacuum dried at 80 °C for 12 h and used as cathode electrodes for assembling the coin cells. The cathode pellets were 12 mm in diameter, and the average loading of the active material in each pellet was about 0.7−0.9 mg cm⁻².

The electrochemical performances were tested by a 2016-type coin cell with a HATN-based electrode as the cathode, a lithium chip as the counter electrode, a Celgard 2325 film as a separator, and 1 M LiTFSI in DOL/DME (1:1 v/v) as the electrolyte, respectively. All coin cells are assembled in an Ar-filled glovebox (the oxygen and water concentration maintained below 1 ppm). The galvanostatic charge/discharge tests over a voltage range of 1.2–4.0 V were performed on a battery test system (LAND CT2001A, Wuhan, China) at ambient temperature. The capacity of the organic electrode is calculated based on the mass of active materials. Cyclic voltammetry (CV) was tested with a scan speed of 0.2 mV s⁻¹ by an electrochemical workstation (Autolab PGSTAT302N, Switzerland) between 1.2 and 4.0 V. The reversal potential of two HATN-based cathodes is measured to 3.1 V vs. Li/Li⁺ in 0.00244 V step increases. Electrochemical impedance spectroscopy (EIS) was carried out on a Solartron Analytical SI 1287A/SI 1260A potentiostat with a voltage amplitude of 10 mV at a frequency range of 10⁶ Hz to 10⁻² Hz, 10 points per decade. The electrodes of ex-situ XRD and XPS were prepared by the method described above. The fabricated electrodes are assembled into batteries. The cells were disassembled after being charged/discharged to a specific state of charge. The electrodes were rinsed with the solvent (DMC) and dried for characterization. Coulombe efficiency was calculated using Equation (1):

$$CE=\frac{Discharge capacity}{Charge capacity}$$

(1)

3. Results and Discussion

3.1. Preparation and Characterization of HATN-[NH₂]₃

Figure 1a schematically depicts the synthetic route to two HATN-based molecules. The HATN-[NH₂]₃ was synthesized from the reaction of 4-nitro-1,2-phenylenediamine and HKH, and then reduced the nitro groups to amino groups by sodium sulfide. For comparison, the HATN without amino groups was prepared according to previously reported work [26]. The detailed synthetic procedures were illustrated in Scheme S1 (Supporting Information). Their chemical structures were identified by ¹H NMR, FT−IR, and elemental analysis (Table S1), respectively. ¹H NMR spectrum of HATN-[NH₂]₃ (Figure S1a) shows the characteristic peaks at about 6.55, 7.23, 7.47, and 8.15 ppm, corresponding to the four types of hydrogen (remarked in Figure S1a) in the molecule, and the peak area was consistent with the structural characteristics. The chemical structures of HATN and HATN-[NO₂]₃ were also identified by ¹H NMR spectra (Figures S1b and S2). As shown in Figure S3a, the FT−IR spectrum of HATN-[NO₂]₃ exhibited a main peak at 1624 cm⁻¹ for C=N, and peaks at 1533 and 1346 cm⁻¹ for -NO₂ groups. Compared with (HATN-[NO₂]₃), N−H characteristic peaks at 3423 and 3321 cm⁻¹ were observed in FT−IR spectra of HATN-[NH₂]₃, [33] suggesting that the nitro groups of HATN-[NO₂]₃ were successfully reduced to amino groups. The FT−IR spectrum (Figure S3b) is also consistent with the reported HATN structure [26]. Solid-state ¹³C NMR was used to further determine its structure. Figure S4 shows a peak at 133.4 ppm, indicating the formation of an imine bond in the pyrazine ring. The thermogravimetric results of HATN-[NH₂]₃ show that the mass loss is about 25% below 330 °C. This should be ascribed to the continuous loss of bound water of HATN-[NH₂]₃ and the decomposition of amino groups as the temperature rises. Moreover, the elemental analysis (EA) result showed similar results and indicated the composition of HATN-[NH₂]₃ powders was C₂₄H₁₅N₉·5H₂O (Table S1), where the weight content of bound water and amino groups was about 26.5%. However, compared with the HATN (C₂₄H₁₂N₆) (Table S2), the HATN-[NH₂]₃ showed high thermal stability and the redecomposition temperature was over 550 °C (Figure S7). All the above-mentioned characterizations support the successful synthesis of two HATN-based molecules. The SEM images (Figure S5a,b) and TEM images (Figure S5c) showed a bulk-like morphology for the HATN-[NH₂]₃, and the corresponding energy-dispersive X-ray spectroscopy (EDS) mapping images (Figure S5e,f) show the uniform distribution of C, N elements over the samples. X-ray diffraction (XRD) patterns (Figure S6a) showed three peaks at 7.5, 27.1, and 46.01°, which corresponded to a d-spacing of 11.74, 3.29, and 1.97 Å, respectively. The interfacial distance of 3.29 Å is close to the interplanar distance of π–π stacking and smaller than the typical distance of π–π stacking (3.3–3.4 Å), indicating the short interfacial spacing and strong interfacial intermolecular interactions [34]. However, all the diffraction peaks from HATN-[NH₂]₃ are very broad, indicating the nature of low crystallinity. This is also evidenced by the absence of obvious diffraction rings found in the selection electron diffraction maps (Figure S6b). Therefore, the HATN-[NH₂]₃ is a low crystallinity material.

The amino groups have a significant effect on their physical and chemical properties, particularly electronic structure. To get insight into the effect of amino groups, the energy levels of frontier molecular orbital for two HATN-based molecules are calculated by using the density functional theory (DFT) method, as described in Supporting Information. As shown in Figure 1b, the lowest unoccupied molecular orbital (LUMO) and highest occupied molecular orbital (HOMO) of HATN-[NH₂]₃ are calculated to be −2.20 and −5.51 eV, respectively. The band gap of HATN-[NH₂]₃ (3.31 eV) is obviously lower than that of HATN (3.79 eV), revealing its faster electron transfer rate, as proved by their electrochemical impedance spectra (vide infra). The decreased band gap is ascribed to the strong electron-donating ability of amino groups. The high intrinsic conductivity of organic materials is vital to achieve high-rate capability in LIBs [35,36].

The amino groups also play a significant role in suppressing the dissolution of HATN-based electrodes. To carefully investigate its solubility, the HATN-[NH₂]₃ and HATN powders were immersed in the organic electrolytes with different time intervals, and the electrolytes after immersion were characterized by ¹H NMR, FT−IR, and UV-vis measurements. As shown in Figure 2a, the electrolyte containing HATN-[NH₂]₃ exhibited no color change while the electrolyte containing HATN displayed obvious yellow, in line with the color of HATN powder, suggesting that the amino-functionalization can obviously decrease the solubility. Figure S8 shows the color change of electrolytes with increasing immersed time from 1 h to 7 days. Compared with the obvious color change of the electrolytes soaked by the HATN electrode, the electrolytes soaked by HATN-[NH₂]₃ electrodes almost no color change (Figure S8a), illustrating the insoluble characteristic. In addition, the HATN-[NH₂]₃ electrodes showed similar morphology before and after electrolyte soaking (Figure S9a,b), indicating its stable electrode structure. However, the morphology of the HATN electrode largely changed before and after electrolyte soaking (Figure S9c,d). ¹H NMR, FT−IR, and UV-vis spectroscopies were further used to test the solubility properties. UV/vis spectra (Figure 2b) show that the electrolyte soaked by HATN-[NH₂]₃ has no adsorption peaks, further proving its insoluble characteristic. For comparison, the HATN without amino groups exhibits obvious UV/Vis adsorption peaks at the range of 250–430 nm⁻¹ in the electrolyte (Figure 2c), indicating the soluble characteristic of HATN [37]. ¹H NMR spectra of the electrolyte after HATN-[NH₂]₃ soaking (Figure 2e) shows the peaks at 4.88, 3.87, 3.58, and 3.41 ppm, corresponding to characteristic resonances of DME and DOL. The absence of other impurities in the electrolytes at various soaking times further demonstrated that HATN-[NH₂]₃ is insoluble in the electrolyte. FT−IR spectra of the electrolyte after HATN-[NH₂]₃ soaking (Figure 2f) exhibit characteristic peaks of electrolytes, including the −C−H stretching mode of the methylene group of DME and DOL (2941 cm⁻¹), S=O stretching mode (1352 cm⁻¹) and the C−F bond of triflate in LiTFSI (1201 cm⁻¹), and the C−O−C functional group in DOL (982 cm⁻¹), respectively [38]. However, the characteristic peaks of HATN-[NH₂]₃ were not observed in the FT−IR spectra of the electrolyte after HATN-[NH₂]₃ soaking. ¹H NMR, FT−IR, and UV/Vis spectra show similar spectra for the electrolytes soaked by HATN-[NH₂]₃ with different durations and do not appear in the peaks of the HATN-[NH₂]₃, suggesting its insolubility in electrolytes.

The insoluble characteristic of HATN-[NH₂]₃ was ascribed to the amino groups’ functionalization due to the force between adjacent molecules through hydrogen bonds or π-π interaction, leading to decreased solubility [21]. Temperature-dependent in-situ FT−IR measurements were conducted to prove the presence of hydrogen bonds between the HATN-[NH₂]₃ molecules. FT−IR spectra of HATN-[NH₂]₃ given in Figure 2d exhibit a peak at about 3436 cm⁻¹ at 305 K, corresponding to N−H asymmetric stretching of primary amine. As the temperature rises from 305 K to 385 K, the peak at 3436 cm⁻¹ gradually shifted to 3450 cm⁻¹, along with the decrease of peak intensity. The red shift with temperature increase suggests the presence of the hydrogen bonding interaction (N−H┄N) between the HATN-[NH₂]₃ molecules, leading to its insoluble characteristic in electrolytes, which is important to achieve stable electrochemical performances for organic electrodes in LIBs.

3.2. Electrochemical Performance

The electrochemical performance of the HATN-based cathode was evaluated over a voltage range of 1.2−4.0 V at room temperature. The discharge/charge curves of HATN-[NH₂]₃ at the first three cycles were exhibited in Figure 3a. HATN-[NH₂]₃ has an initial capacity of up to 192.4 mAh g⁻¹ with a high coulombic efficiency (CE) of 90% at 0.05 A g⁻¹. Meanwhile, the capacity loss in the following cycles may be on account of the formation of solid electrolyte interphase (SEI) films and the activated process [39,40]. Figure 3b shows that the typical cyclic voltammograms (CV) curves of HATN-[NH₂]₃ electrodes at the initial three cycles, displaying three distinct couples of redox peaks at 2.66/2.63, 2.09/2.08 and 1.79/1.74 V vs. Li⁺/Li. Notably, the CV curves of the initial three cycles are almost completely overlapped, indicating the highly reversible and stable electrochemical behavior of the HATN-[NH₂]₃ electrodes. This is also proved by the above charge-discharge curves. CV curve of HATN (Figure S10) also showed three distinct couples of redox peaks at 1.51/1.31, 2.47/2.31, and 2.49/2.79 V vs. Li/Li⁺. These results indicate that the two structures have similar multi-electron lithium storage behavior [26]. As expected, the HATN-[NH₂]₃ electrodes exhibited a more stable cycle performance than that of the HATN electrode at 0.05 A g⁻¹ (Figure 3c). After 60 cycles, its capacity remained stable and maintained at 185 mAh g⁻¹ (96% retention), while the capacity of HATN decreased to 123 mAh g⁻¹ (52% retention), although stable first three cycles (Figure S11). It should be noted that the carbon black additive exhibited a low capacity of ~16.7 mAh g⁻¹ at a current density of 0.05 A g⁻¹ (Figure S12). Additionally, the HATN-[NH₂]₃ electrodes achieved a high CE of 98% with the increase of cycle number and kept stable. However, the HATN showed unstable CE at different cycles due to its unstable electrode structure in the electrolyte. More importantly, the HATN-[NH₂]₃ electrodes also exhibited highly stable cyclability at a high current density of 0.5 A g⁻¹ (Figure 3d). After 500 cycles, the discharge capacity of HATN-[NH₂]₃ was maintained at 135 mAh g⁻¹ with an average CE of 99.5 %, and there was no capacity decay during the entire cycle, which outperforms most reported HATN-based positive electrodes (Table S3). As a comparison, the capacity of HATN electrode was only 50 mAh g⁻¹ after 500 cycles, and the capacity retention rate was only 24%. The poor cycle stability of HATN is ascribed to it being highly soluble in the electrolyte, which leads to a serious loss of active materials from the electrodes during the cycling process, as shown in Figure S13. However, the electrode morphology of HATN-[NH₂]₃ after 10 cycles is similar to the pristine morphology (Figure S14). The electrode stability of HATN-[NH₂]₃ was also confirmed by almost no color change of the disassembly separator before and after charging and discharging 10 cycles (Figure S15). The high structural stability of HATN-[NH₂]₃ in the electrolyte during the discharge/charge process is extremely important for long-term stability.

Figure 3e shows the rate capabilities of two HATN-based electrodes. The capacities of HATN-[NH₂]₃ electrodes are 148, 133, 126, 115, and 105 mAh g⁻¹ at various current densities of 0.1, 0.3, 0.5, 1, and 2 A g⁻¹, respectively. The rate performance of the HATN-[NH₂]₃ electrodes is also obviously better than that of the HATN electrode. The rate performance was mainly affected by charge and solid-state ion transport resistances of electrode materials. The EIS was performed to study the charge and ion transport properties, and the test data were fitted by an equivalent circuit diagram (Figure 3f,g) [41,42]. The fitting data showed that the charge transfer resistance (Rct) of both cathodes increased after cycling. Notably, the HATN-[NH₂]₃ cathodes always showed smaller charge transfer resistance than the HATN cathodes (Figure 3h), indicating its quicker charge transfer ability. The Warburg impedance is expressed by the Warburg coefficient (σ), which is the slope of the Z’∼ ω^−1/2 line obtained from the Z’ ∼ ω^−1/2 plot (ω = 2πf) in Figure S16 [43]. The σ of HATN-[NH₂]₃ after 10 cycles is calculated to be 52.66 Ω s^−1/2, which is the same as the σ of the pristine HATN-[NH₂]₃ (58.68 Ω s^−1/2). However, the σ of HATN after 10 cycles is calculated to be 2335.97 Ω s^−1/2, which is obviously higher than that of the fresh HATN (754.81 Ω s^−1/2). This indicates that the HATN-[NH₂]₃ cathodes have better lithium ion diffusion and stability than HATN during cycling, which is attributed to the good chemical stability of HATN-[NH₂]₃. Benefiting from low Rct and higher stable structure in the electrolyte, HATN-[NH₂]₃ electrodes achieved long-term cycle life and excellent rate capacity, indicating the key role of amino groups in boosting the electrochemical performance for HATN-based cathode. In view of the above excellent electrochemical performance, the HATN-[NH₂]₃ is a promising candidate as cathode material for commercial LIBs due to its cost-effective and mature preparation of organic molecules, simple electrode processing procedures with combinability of current industrial technology for battery manufacturing.

3.3. Lithium Storage Mechanism

The possible lithium storage mechanism of HATN-[NH₂]₃ was illustrated in Figure 4a. The three Li ions firstly store in HATN-[NH₂]₃ via Li–N bond, and then another three Li ions store again [26]. The intercalation/deintercalation electrochemical process can be reversibly performed. Figure 4b shows the initial galvanostatic discharge and charge curves of HATN-[NH₂]₃ electrodes at diverse discharge and recharge states at 0.05 A g⁻¹. Ex-situ XRD patterns of the HATN-[NH₂]₃ electrodes at diverse discharged and charged states (corresponding to the marked points in Figure 4b) were measured to explore the reversible electrochemical redox process (Figure 4c). The main diffraction peaks for the HATN-[NH₂]₃ electrodes did not change significantly under different charging and discharging states, which indicated that the structure of the sample remained stable during the lithium-ion intercalation. In addition, the strongest peak at about 27.1° that corresponded to the layer spacing moved to a lower angle with the insertion of more lithium ions and returned back to a higher angle with the extraction of lithium ions, indicating that the interfacial spacing increased after the storage of Li ions. During the charging and discharging process, the Li₂CO₃ peaks (as an important component of SEI) appeared at different charging and discharging states, suggesting the formation of stable SEI. These results indicated the reversible insertion/extraction of Li ions and the stable structure of HATN-[NH₂]₃ during the charging and discharging process [44,45].

To further inquire into the reaction mechanism, ex-situ XPS was used to test the HATN-[NH₂]₃ samples with diverse discharge and recharge states. Figure 4d shows that the high-resolution N 1s spectrum of the as-prepared electrode could be divided into two peaks at 399.38 and 400.78 eV, corresponding to conjugated (sp²) –N= and non-conjugated (sp³) –NH₂ groups, respectively. During the discharging process (from point A to C), the peaks of the –N= weakened gradually when discharged to 1.5 V. Contrastively, a new peak emerged at 399.08 eV ascribed to a newly-formed Li–N single bond (399.08 eV) when discharged to 1.2 V, implying that reduction occurred at N atoms in the pyrazine ring and the formation of a Li–N single bond. During the charging process, the peak intensity of the C=N double bond (399.38 eV) gradually enhanced and recovered to its pristine state, and simultaneously the peak of the Li-N single bond gradually reduced [46,47,48]. When charging to 4.0 V, the Li–N bond disappeared, indicating that the lithium ions were basically released, and the charging and discharging process was a reversible process.

4. Conclusions

In summary, a novel insoluble hexaazatriphenylene-based organic cathode was developed for high-performance LIBs. Due to the amino group functionalization, the HATN-[NH₂]₃ resolves the problem of the high solubility of HANT-based electrodes in non-aqueous electrolytes. Benefiting from poor solubility in organic electrolytes and increased conductivity, the HATN-[NH₂]₃ cathodes showed an initial discharge capacity of 192.4 mAh g⁻¹ with a high initial CE of 90% at 0.05 A g⁻¹. Furthermore, the cathode also exhibited high cycle stability without capacity decay at 0.5 A g⁻¹ for 500 cycles and outstanding rate capability (105 mA h g⁻¹ at 2 A g⁻¹). Ex XRD and XPS measurements confirmed that the discharge and charge of the HATN-[NH₂]₃ electrode is a reversible process. This work provides a simple and facile strategy to resolve the dissolution issue of organic electrodes via amino group functionalization for the stable operation of rechargeable batteries.