Precision Synthesis and Modulation of Ion Diffusion Interface

Abstract

The development of electrochemical energy storage systems with high-density storage sites is the key and core technology to achieve transformative breakthroughs in battery performance. Herein, we immobilize redox-active naphthalene diamides (TBNDI) within 2D graphdiyne (GDY) to construct carbon-based TBNDI-GDY composite electrodes. This approach creates a hierarchical ion diffusion pathway and enables tunable electronic modulation through a strategic molecular design. The multiredox center electrodes demonstrate prominent kinetic processes, superior interfacial compatibility, and fast desolvation capability. The as-prepared electrodes with pseudocapacitive processes display ultrahigh specific capacity up to 2079 mAh/g at 0.1 A/g, remarkable rate capacity, and ultralong stability for 10,000 cycles even at 5 A/g. Dynamic kinetic tracking and lithium active site visualization confirm that capacity contribution originates from reversible lithium-ion capture as well as Li–C orbital coupling in the sloping voltage regions, nanopore filling, and graphitic region interaction in the subsequent voltage region. Moreover, C=O–N groups with lone pair electron delocalization could modulate the electronic structure and promote reversible redox activities. Our findings highlight that the rational design of the ion diffusion interface from the basic chemical structure can provide giant evolution on the properties of the electrode material for high performance batteries.

Introduction

Rechargeable lithium-ion batteries with high energy density, , brilliant security, and environmental friendliness , have attracted intensive research interest in portable electronic products and various energy storage fields. Given their excellent chemical stability and electrical conductivity, carbon-based materials such as graphite, graphene, and porous carbon are considered to be the most competitive promising candidates. − Moreover, carbon-based materials provide many active sites to fulfill the intercalation, adsorption, or pore-filling of alkali metal ions. , Due to structural defects, specific capacity, rate capability, and incomplete desolvation process, those materials are considered to be bottlenecks for their application as ultra-high-energy electrodes. , Due to the introduction of heteroatoms, the complex porous microstructure and ambiguous electronic structure of these traditional carbon materials hinder the in-depth understanding of the relationship between structure and properties and restrict the breakthrough of energy density. , In order to fabricate high-performance batteries with high energy density, the development of a new carbon-skeleton electrode with a clear structure and composition is the only feasible way. Considering the diversity of emerging carbon materials, it is possible to modulate electronic or pore structure through chemical structure design. Structural design is an effective strategy to meet the demand of brilliant performance electrodes with high capacity, flexible discharge plateaus, and variable microstructure. −

Organic redox-active electrodes have the superiority of reversible multielectron centers, flexible structure, and weaker intermolecular interaction. − Organic redox-active centers such as diamide (C=O–N) groups could decrease the lowest empty molecular orbital (LUMO) and lower the redox potential to realize high energy density or output power density. − Nevertheless, organic electrodes suffer from a severe dissolution problem and inferior conductivity. , It is meaningful to introduce a C=O–N group into the conjugated skeleton of carbon materials to make full use of their structural superiority. Designing a structure with a highly conjugated carbon structure and plentiful C=O–N could enable electrodes with excellent conductive skeleton, multiple active sites, and controllable microstructure with extensive pores. Notably, the excellent interfacial compatibility and good solubility of C=O–N with organic electrolyte are necessary to satisfy fast diffusion and desolvation of metal ions. , This is the key to accelerating reaction dynamics. The emerging carbon allotrope graphdiyne (GDY) with sp and sp² cohybridized carbon have distinctive advantages such as good conductivity, distinctive ion diffusion in plane and out of plane, chemical stability, and tunable pore structure, which enable it to introduce organic groups. − Profiting from those unique features, it is possible to successfully overcome the serious problems that hinder energy density and cycle stability and promote the rapid development of electrode materials.

Herein, we demonstrate a porous electrode (TBNDI-GDY) with explicit chemical structure, diffusion interface, and electronic modulation to satisfy superior capacity and speedy Li⁺ storage. The multiple active site electrodes have some distinctive structural merits including stable structure, superior conductivity, and outstanding 3D diffusion channels, thereby advantageous to realize robust pseudocapacitance kinetics. More importantly, the C=O–N groups with N and O lone pair electron delocalization have contributed to good compatibility with electrolyte and realize effective desolvation capability. Typically, TBNDI-GDY electrodes can deliver an extraordinary reversible capacity (2079 mAh g^–1 at 100 mA g^–1), brilliant rate capability (750 mAh g^–1 at 5 A g^–1) and ultrastable cyclability (0.003% capacity decay single cycle after 10,000 cycles at 5 A/g). Complex kinetic tracking and lithium active site visualization have demonstrated four distinct Li⁺ storage active sites, and a hybrid mechanism involving in lithium-ion capture as well as Li–C orbital coupling, nanopore filling, graphitic region interaction, and reversible C=O–N redox is proposed. Those findings provide new insights into designing multiple-active electrodes with electronic modulation and distinctive lithium-ion storage mechanisms in molecular design for future energy storage.

Experimental Section

Synthesis Procedures for TBNDI-GDY

2,3,6,7-Tetrabromonaphthalene-N,N′-bis­(n-octyl) diimides (TBNDIs) are generally synthesized according to the reported literature. Based on a similar synthesis strategy, we successfully fabricate 2,3,6,7-(trimethylsilyl)­ethynyl-N,N′-bis­(n-octyl) diimides monomer in 95% yield from the starting material TBNDIs.

Cu foils with removed surface oxides are put into a reactor containing pyridine under the protection of argon atmosphere. Then, 30 mg of monomer in 2,3,6,7-(trimethylsilyl)­ethynyl-N,N′-bis­(n-octyl) diimides solution (1 mg/mL) in pyridine is added to the mixture slowly. The mixture is kept at 60 °C for 3 days in dark environments to ensure the complete reaction. The color of the Cu foils turns from yellow to black. The product TBNDI-GDY films are washed with acetone and hot N,N-dimethylformamide in turn and then cleaned thoroughly with 6 M HCl and water, respectively.

Characterization

SEM images are recorded by a field-emission scanning electron microscope (S4800). TEM image, high-resolution image, and electron diffraction dots are collected by a JEM-2100F microscope operating at 200 kV. The thickness and size are characterized by atomic force microscopy (AFM, Ntegra Spectra). Nuclear magnetic resonance (NMR, Bruker Avance 400) is used to record the ¹³C and ¹H spectra. A Thermo Scientific DXRxi (532 nm) instrument equipped with LAND measurements is used to investigate the variation of characteristic peaks during the cyclic process. A traditional two-electrode cell (Beijing Scistar Technology Co., Ltd. (www.bjscistar.com)) is implemented to carry out in situ Raman characterizations. Before conducting experiments, the battery is assembled in an inert atmosphere, which consists of a working, counter, and reference electrode (Li foil) and Celgard membrane. The galvanostatic discharge–charge curves are evaluated with the current density of 0.1 A/g and applied potentials between 0.01 and 3 V. Meanwhile, the corresponding Raman spectra are recorded with a voltage interval of 0.1 V. The FT-IR spectrum is employed to analyze the chemical structure of the products. The chemical composition and elemental valence states are measured by means of a Thermo Scientific ESCALab 250Xi instrument with an Al Kα radiation source.

Electrode Measurement

For evaluating the electrochemistry performance of TBTBNDI-GDY, the CR2032 coin cells are assembled with TBTBNDI-GDY as the working electrode, Li metal foil as the counter and reference electrodes, Celgard membrane as the separator, and 1 M LiPF₆ in ethylene carbonate (EC) and dimethyl carbonate (DEC) with volume ratio of 1:1 supplied by Canrd (Canrd Technology Co., Ltd.) as the electrolyte. The coin cells are assembled in an Ar-filled glovebox in which the content of O₂ and H₂O are both less than 0.01 ppm. Half cells are tested with the voltage range of 0.01∼3 V through LAND equipment. Cyclic voltammetry with various scan rates from 0.1 to 1 mV s^–1are recorded by AutoLAB, and electrochemical impedance spectra (EIS) curves are obtained by AutoLAB with the amplitude of 10 mV in open circuit voltage state.

Results and Discussion

For the purpose of fulfilling the efficient and successive Li ion diffusion, designing electrodes with ordered channel structure, remarkable reactivity, and excellent conductivity is critical. In this work, we immobilize redox-active TBNDI into highly conjugated GDY to synthesize ion diffusion interface tunability as well as electronic modulation TBNDI-GDY electrodes in the molecular level (Scheme ). Combined with uneven charge distribution, excellent conductivity, as well as extensive active sites for GDY, diamide with lone pair electron delocalization is regarded as an electron-withdrawing group to increase accessible active sites, decrease the LUMO as well as the redox potential, and modulate electronic structure to optimize the kinetic process. The synthetic methods of TBNDI-GDY and the ¹H as well as ¹³C nuclear magnetic resonance (NMR) spectra and high-resolution mass spectrometry (MS) of 2,3,6,7-(trimethylsilyl)­ethynyl-naphthalene diimide are shown in Figures S1–S4 combined. These results indicate the successful synthesis of 2,3,6,7-(trimethylsilyl)­ethynyl-naphthalene diimide with a large number of C=O–N units and alkyne bonds.

1. (a) The Inspiration of Molecular Design for TBNDI-GDY Electrodes and (b) the Synthetic Route to Fabricate Naphthalene-Diimide-Units-Substituted GDY.

The synthesis route of TBNDI-GDY is depicted in Figure a. Figure b shows the Cu foil and the black-binder-free TBNDI-GDY film grown on Cu foil. The fabricated TBNDI-GDY film displays pore morphology, as shown in Figure b, which was beneficial to fast ion diffusion and electrolyte penetration. The cross-sectional image of the TBNDI-GDY film in Figures S5 and S6 demonstrated that TBNDI-GDY grew tightly on the surface of Cu. The nitrogen adsorption–desorption curves of TBNDI-GDY (Figure S7) showed that the pore distribution was from 2 to 50 nm and mainly concentrated on 4 nm. The transmission electron microscope (TEM) image distinctly displayed a sheet-like structure consistent with SEM images (Figure d). High-resolution TEM image with different regions in Figures e–j and the corresponding selected area electron diffraction (SAED) pattern in Figure k demonstrated the high crystallization with no noticeable defects as well as a lattice distance of 0.47 nm and hexagonal crystal lattice. The distance of about 2.19 nm suggested that the lattice distance was 0.47 nm, which agreed well with the results of the HRTEM image. The broad peak at 20.53° was assigned to the interlayer distance of 0.432 nm in TBNDI-GDY (Figure S8). Atomic force microscopy showed that such an electrode was flat and integral with the average thickness about 4.38 nm (Figures l and m).

1.

(a) The synthesis route of molecular design for TBNDI-GDY electrodes. (b) Photo of the synthetic 3D binder-free TBNDI-GDY electrodes. (c) SEM images of TBNDI-GDY. (d–j) TEM images and related SAED image (k) of TBNDI-GDY. (l) AFM pattern of designed TBNDI-GDY and (m) the corresponding height.

To ascertain the structure and chemical composition of the obtained TBNDI-GDY, we utilized some typical characterizations to get the information in detail. The Raman result is shown in Figure a. Two distinct peaks around 1355 and 1585 cm^–1 corresponded to the D band and G band with I _G/I _D about 1.39, demonstrating the TBNDI-GDY with a high degree of graphitization. Specifically, the apparent characteristic peaks at 1966 and 2160 cm^–1 were assigned to the vibrations of conjugated −C≡C–. − Besides, two peaks at 1369 and 2132 cm^–1 confirmed the existence of diimide and alkyne groups in TBNDI-GDY (Figure b). ,− The solid-sate ¹³C spectrum (Figure S9) revealed that the signals at 104.8 and 99.8 ppm were ascribed to the interaction between sp hybrid C and sp² hybrid C, and adjacent C in alkyne units. , As displayed in the X-ray photoelectron spectroscopy (XPS) image, the C peaks (Figure c) could be deconvoluted into four main peaks at 284.5 eV (C=C), 285.0 eV (C≡C), 285.9 eV (C–N), and 288.9 eV (C=O). Combined with the N spectrum in Figure d, we could come to the conclusion that there indeed existed alkyne links and diimide groups in the designed TBNDI-GDY electrode. Meanwhile, the fabricated TBNDI-GDY exhibited an optical energy gap of about 1.06 eV, indicating the intrinsic characteristic of preeminent conductivity in TBNDI-GDY (Figure e). The delocalization of N and O lone pair electrons over the aromatic skeleton could modulate electronic structure, decrease band gap, and enhance redox activity, thereby endowing electrodes with outstanding conductivity. , Besides, the electrode displayed a good compatibility with the electrolyte, of which the contact angle is about 2.5°. The contact angle between the electrode and water is about 81.0° in Figure f. The superhydrophilic surfaces can ensure the infiltration of electrolyte into electrode within momentary time and be conductive to speedy ion diffusion with rapid kinetic process on account of strong interaction between the designed electrode and electrolyte, as depicted in Figure g.

2.

Raman spectrum (a) and FITR image (b) of designed TBNDI-GDY electrodes. (c) C spectrum, (d) N spectrum, and (e) the corresponding UV curve of TBNDI-GDY. (f) Contact angles of electrolyte with TBNDI-GDY. (g) The scheme for good interfacial compatibility.

We assessed the lithium-ion storage performances of binder-free 3D electrodes within the voltage range of 3∼0.01 V. It is noticeable that there were apparent redox behaviors in the first cycle and latter two cycles (Figure a). In the first cycle, binder-free TBNDI-GDY delivered a discharge capacity of 3181.4 mAh g^–1 with a Coulombic efficiency of 65.4%. The irreversible capacity loss is ascribed to the formation of the stable solid electrolyte interphase (SEI). , The charge–discharge curves of the next two cycles were identical, demonstrating the remarkable stability of binder-free electrode. The TBNDI-GDY electrode delivered outstanding stability and almost no capacity decay with a capacity retention of 95.0% at 0.1 A/g in Figure b. Besides, it displayed good rate performance with a reversible capacity of 362.7 mAh g^–1 at 10 A/g and predominant recovery capability when returning to 0.1 A/g (Figure c and Figure S10). In addition, we obtained a cyclic voltammetry (CV) curve to reveal the redox process of TBNDI-GDY between 0.01 and 3 V in the first five cycles (Figure d). In the cathodic process, three reduction peaks located at 1.39, 1.14, and 0.72 V appeared, which was attributed to the Li insertion and lithiation process into TBNDI-GDY. In the anodic process, two distinct voltage plateaus could be observed at around 1.17 and 2.64 V, showing the Li⁺ extraction and delithiation process. In the subsequent cycles, the oxidation peaks and reduction peaks were almost overlap, demonstrating the brilliant reversibility of the Li⁺ lithiation/extraction process. , The results are consistent with the charge–discharge profiles in Figure b.

3.

(a) The charge–discharge curve in the initial three cycles. (b) Cyclic performance at 0.1 A/g. (c) Rate performance. (d) Cyclic voltammetry curve. (e) CV curves at different scan rates from 0.1 to 1 mV/s and relevant linear fitting curve of log­(i)–log­(v) at different oxidation and reduction peaks (f). (g) The fitting pseudocapacitive behavior (purple area) at 1 mV/s and related pseudocapacitive contribution ratio (h). (i) Cyclic stability at 1 A/g. (j) Cycle performance at 5 A/g. (k) The comprehensive performance of TBNDI-GDY with reported literature. (l) The distinctive superiority for outstanding kinetic process.

To understand the kinetic process of the TBNDI-GDY electrode, CV at different scan rates were performed to investigate the pseudocapacitive contribution and diffusion coefficients of Li⁺ in Figures e–h and Figures S11–S13. Figure e shows that the locations of oxidation and reduction peaks shift to higher and lower voltage with the increase of scan rate due to enhanced polarization, respectively. Besides, we carried out a galvanostatic intermittent titration technique (GITT) method to further understand excellent specific capacity and rate capability. As seen in Figure S12a, the lithium-ion diffusion coefficients for the TBNDI-GDY electrode were 4.40 × 10^–10 ∼ 1.25 × 10^–8 cm² s^–1 for discharge and 7.80 × 10^–9 ∼ 1.07 × 10^–7 cm² s^–1 for charge. In addition, the TBNDI-GDY electrode manifested weakened voltage polarization and ohmic polarization in Figure S12b and charge transfer resistance and solution resistance in the high frequency region (Figure S13). Furthermore, we could get the values of parameter b to evaluate the storage mechanism of Li⁺ via the equation i = av ^b . A diffusion kinetic process is dominant when the value of b approaches 0.5 while a capacitive behavior is the main storage process when the b value is close to 1. As displayed in Figure f, the b parameters were 0.865, 0.832, 0.789, and 0.746, referring to different anodic and cathodic peaks, implying the capacitive lithium-ion storage mechanism for a 3D binder-free electrode. We also analyzed quantitatively the capacitive contribution through the equation i = k ₁ v + k ₂ v ^1/2 in Figures g and h and Figure S14 (k ₁ v and k ₂ v ^1/2 represent the capacitive and diffusion dominant process, respectively , ). The proportion of the pseudocapacitive process was estimated to be 70% in total charge storage, which was advantageous to the Li⁺ diffusion. Accompanied by an enlarged scan rate from 0.1 to 1 mV/s, the capacitive contribution gradually increased from 44% to 70%, suggesting an extraordinary Li⁺ kinetic process and an outstanding rate capability for TBNDI-GDY. More importantly, the binder-free TBNDI-GDY electrode demonstrated an excellent capacity retention of 94.0% with ultralong cycle about 2000 cycles at 1 A/g and a superior coulomb efficiency of 99.9% (Figure i). The electrodes also displayed an exceptional energy density of 759 Wh/kg under 1 A/g (Figure S15). Notably, the capacity retention of TBNDI-GDY could reach 70% even at 5 A/g after 10,000 cycles, as displayed in Figure j. Due to the architecture of the binder-free electrode and outstanding kinetic process, the electrochemistry performance of TBNDI-GDY exceeded the advanced anodes in previous literature such as MoS₂/Ti₃C₂, Fe₂O₃/GO, BNO, and so on (Figure k). − Noticeably, such superior long-term and rate capability originate from speedy electron transportation with moderate electronic structure, 3D ion diffusion, plenty of active sites, and good conductivity with conjugated structure (Figure l).

Raman spectra are recorded to reveal the solvation structure evolution in graphite and TBNDI-GDY. The characteristic peaks located at 738.1, 896.7, and 969.3 cm^–1 attributed to the PF₆ ^– breathing mode, the breathing mode in the EC ring, and the O–CH₃/O–C₂H₅ stretching mode of DMC in electrolyte are shown in Figure a. It is clear that the modes attributed to EC and DMC exhibited obvious red shift for TBNDI-GDY in contrast to hardly any shift for graphite. In contrast, the characteristic peaks for PF₆ ^– display obvious blue shifts, indicating the enhanced strong interaction between PF₆ ^– and Li⁺. These results confirm diminished Li-ion solvation on account of strong interaction of Li-ions with alkyne bonds of TBNDI-GDY. As shown in Figure b, TBNDI-GDY has strong interaction with Li⁺ due to the existence of alkyne bonds, thereby diminishing the coordination interaction between Li⁺ and solvent molecule to expedite the desolvent process. We could observe that C 1s peaks show a positive shift in the initial stage and a negative shift in the latter stage during the discharge process (Figure c). This result demonstrates that incomplete charge transfer could induce reversible Li–C formation in the sloping regions and effective charge transfer to sp²-C on account to electron occupation of the π* antibonding band.

4.

(a) Raman spectra of the electrolyte and electrolyte wetted graphite or TBNDI-GDY. (b) The speculated desolvation process for lithium-ion storage in TBNDI-GDY. The evolutions of binding energy of C (c) and Li (d) with different state of discharge (blue lines) or charge (red lines) etching for 300 s. (e) The peak position of Li and C with different depths. The discharge–charge curves with constant current in the first cycle and in situ Raman spectroscopy test during discharge (f) and charge (g). (h) The lithium-ion storage mechanism for TBNDI-GDY.

During the charge process, the binding energy of C 1s shifted in the opposite direction, indicating a highly reversible lithium-ion storage mechanism in TBNDI-GDY. Moreover, the Li 1s peak (Figure d) showed an opposite shift in contrast to Li 1s during the discharge and charge process, respectively. This result confirmed the strong interaction, cation−π effect, and reversible Li–C orbital coupling between alkyne and Li⁺. In addition, we could observe the nearly identical tendency in different etching depths of Li and C elements in Figure e. We conducted in situ Raman meaning to detect the phase transition and atomic environment evolution during the discharge and charge process to unravel the storage mechanism and storage reversibility. As displayed in Figures f and g and Figure S16, the location, shape, and intensity of the D band, G band, and alkyne bond would change gradually during the lithiation and delithiation process. Obviously, we could observe that the alkyne peak at 2098 cm^–1 gradually became weaker and eventually vanished at 0.4 V with the continuous strength of the discharge states. This is attributed to the high electronegativity of sp-C that would induce incomplete charge transfer and form Li–C orbital coupling accompanied by the transformation of alkyne bonds and alkene feature bonds. Simultaneously, the G band displayed an indistinct red shift and diminished intensity when the voltage approached 0.1 V. This phenomenon could be attributed to the formation of LiC _x with the continuous insertion of lithium ions into sp² hybrid C regions and negative charge transfer to hexagonal atom benzene rings in sp²-C regions. ,− In general, negative charge transfer would occupy the π* antibonding band to induce electron–phonon coupling, weakening as well as lengthening sp²-C as well as bringing about massive of Li–C generation, which inevitably induced the red shift of the G band. , The continuous lithium-ion interaction at high lithiation state eventually would lead to the diminishment of the G band. Besides, the lower intensity and distinct red shift of the D band could be ascribed to the introduction of Li⁺ occupying diacetylenic triangular–like pore centers and edge planes breathing motion in sp²-C regions for TBNDI-GDY. We could observe the complete disappearance of the D band as well as the G band at the finial discharge states. During the lithiation process, we could observe clearly the return to origin states for the D band as well as the G band and reversible recovery of alkyne bonds with high reversibility. In the delithiation process, reversible phenomenon with recovery of the D band and G band could be observed, demonstrating lithium-ion filling in massive amounts of pores composed of highly conjugated sp-C and inserting into the interlayer distance in the sp²-C region are removed and extracted. In the subsequent delithiation states, we could distinguish the reappearance of characteristic peaks at 2098 cm^–1 with enhanced intensity, confirming reversible structural stability. As depicted in Figure h, lithium ions would be stored in alkyne active sites with a reversible lithium-ion capture as well as Li–C orbital coupling and then insert into the interlayer distance in sp²-C regions as well as fill into the centers of diacetylenic triangular–like pores.

To gain a deeper understanding of the lithium-ion storage mechanism, we monitored the composition transformation with different states of discharge and charge for the TBNDI-GDY electrode by ex situ FT-IR spectra. As demonstrated in Figures a and c, the strong adsorption peaks at 1664 and 1366 cm^–1 belonged to the typical C=O and C–N stretching varication in pristine TBNDI-GDY materials. , Apparently, the C=O characteristic peak shifted gradually to 1645 cm^–1 with the voltage altering from 3.0 to 0.01 V owing to the lithiation process. Meanwhile, the characteristic peak at 1140 cm^–1 (C–O–Li) appeared and strengthened, indicating the transformation from C=O to C–O–Li during the lithiation reaction. In addition, the functional group C–N–Li infrared peak at 1501 cm^–1 emerged while the C–N became weakened at 1.5 V and ultimately disappeared at 1.1 V during the discharge process. During the charge process, the C=O peak returned to the pristine location at around 1664 cm^–1 with the weaking of C–O–Li (Figures b and d). At the same time, the characteristic peak of C–N–Li vanished with the recovery of C–N, revealing the highly reversible Li⁺ uptake and removal in TBNDI-GDY electrodes. The 3D Nyquist plots are shown in Figures e, f, and S17 during the cyclic process. The semicircles in high-frequency regions correspond to charge transfer resistance (R _ct) and solution resistance (R _s). Apparently, the values of R _s and R _ct gradually increased, and visible boundaries of them could be seen during the lithiation stage. The values of R _ct and R _s could return to the original state after delithiation, suggesting good reversibility. In situ XPS with Ar⁺ sputtering at different depths were implemented to reveal the structure evolution of TBNDI-GDY during cycling. As can be seen, the intensities of C=O and C–O in the O spectrum were significantly reduced and increased during the lithiation process, demonstrating the transformation from C=O to C–O (Figure g). It was noteworthy that the C=O peaks with different depths did not disappear in the end of discharge, indicating the incomplete lithiation owing to strong repulsive force (Figure S18). During the delithiation stage, the C–O peak almost disappears completely and the C=O peak recovered (Figure h). We could observe the relative amount of evolution of C–O and C=O in Figure i in the first cycle. Furthermore, we could observe that the N spectrum displayed a positive shift during discharge and shifted to negative binding energy during charge with etching times of 0, 150, and 300 s (Figures S19–S21). Therefore, the storage mechanism of TBNDI-GDY is involved in complete redox reversible diimide groups. ,, In summary, we speculated that the lithium-ion storage mechanisms were the existence of nanopore filling, graphitic region interaction, and reversible lithium-ion capture, as well as Li–C orbital coupling and redox-reversible diimide groups.

5.

Ex situ FT-IR spectrum for TBNDI-GDY with different discharge (a) or charge (b) states for the TBNDI-GDY electrode and the corresponding color mappings (c and d). The typically 3D Nyquist plots of TBNDI-GDY during the lithiation (e) and delithiation process (f). The O spectrum during the discharge (g) and charge process (h) and the corresponding relative content during the cycle (i). (j) Schematics of the proposed redox mechanism.

Conclusions

In summary, we construct precisely multiple redox-active TBNDI-GDY electrodes with tunable electronic structure and an ion diffusion interface to regard as high-capacity electrodes in molecular engineering. The electrodes display the intrinsic characteristics of outstanding electronic conductivity, 3D ion diffusion channels, favorable interfacial compatibility, and a speedy desolvation process. Therefore, TBNDI-GDY electrodes deliver ultrahigh specific capacity (2079 mAh/g at 0.1 A/g), excellent rate performance, and extraordinary long-term cycle under 5 A/g on account of dominant capacitive process and remarkable lithium-ion diffusion. Moreover, compressive characterizations reveal the complex redox kinetics, visualize lithium active sites, and propose a novel hybrid mechanism involving reversible lithium-ion capture as well as Li–C orbital coupling, graphitic region interaction, nanopore filling, and redox-reversible diimide groups. The TBNDI-GDY electrode with electronic structure modulation, diffusion interface modification, and coordination environment optimization in the molecular design not only provides a new avenue for devising and constructing high energy density electrodes but also reveals profound understanding of the Li⁺ storage mechanism.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/prechem.5c00030.

• Experimental section for fabrications, characterizations, and electrochemistry measurements and more details on TEM, XPS, GITT, Raman, EIS, and corresponding data analysis (PDF)

Y.L. conceived and designed the research and reviewed and edited the manuscript. J.G. synthesized the electrode materials and carried out a series of characterization to analyze the results. J.X. gave some useful suggestions for synthesizing the electrode. C.H. assisted with the data analysis and organized and wrote the paper. X.L., X.F., W.L., and Z.Z. gave useful help.

This work was supported by the Fundamental Research Center of Nano Toxicology Chemistry, the National Natural Science Foundation of China (22388101), the National Key Research and Development Project of China (2022YFA1204500, 2022YFA1204501, 2022YFA1204503, 2018YFA0703501), the Key Program of the Chinese Academy of Sciences (XDPB13), and the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0520200).

The authors declare no competing financial interest.