Kang Zhang’s research while affiliated with Xiamen University and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (8)


DOD‐dependent electrochemical performance distinctions between NCM and LRLO. (a) Discharge profiles of NCM half‐cells after 100 cycles at 115 mA/g within the voltage ranges of 3.6–4.3 V (LDOD) and 2.7–4.3 V (HDOD). Insert: Corresponding capacity retention. (b) Discharge profiles of LRLO half‐cells after 100 cycles at 115 mA/g within the voltage ranges of 3.3–4.5 V (LDOD) and 2.7–4.5 V (HDOD). Insert: capacity retention. (c) Discharge profiles of LRLO||Gr pouch‐type full‐cells after 100 cycles at 115 mA/g within the voltage ranges of 3.0–4.43 V (LDOD) and 2.5–4.43 V (HDOD). Insert: Corresponding capacity retention and the pouch‐type full‐cell photograph (Reproduced with permission). (d) The available capacity decay schematic illustrations of NCM and LRLO cathodes under HDOD and LDOD long‐term cycling. Insert: HDOD and LDOD cycling models. (e) Models of ideal cycling step (HDOD) and practical cycling scene (LDOD).
Cationic/anionic charge compensation mechanism of LRLO at different potentials. (a) Evolution of normalized XANES spectra of Ni K‐edge and Mn K‐edge of LRLO during the initial discharge process. Insert: Half‐height changes in XANES at the Ni and Mn K‐edge. (b) O K‐edge sXAS spectra of LRLO during the initial discharge process. Difference spectra are shown inset for clarity. (c) Voltage profiles and corresponding dQ/dV curves in Li half‐cells as the charge window is opened stepwise from 2.5 to 4.5 V. The dQ/dV curves for the 30th cycle within the voltage ranges of 3.3–4.5 V (LDOD) and the 31th cycle within the voltage ranges of 2.7–4.5 V (HDOD) are on the right side.
The charge compensation mechanism of LRLO for HDOD and LDOD cycling. (a) The dQ/dV curves of HDOD (2.7–4.5 V) and LDOD (3.3–4.5 V) cycling from 5th to 50th cycles at a current of 115 mA/g and the 51th cycle at a current of 23 mA/g within the voltage ranges of 2.0–4.5 V (extend cycle). The dQ/dV curves of the 51th cycle discharge state are presented. (b) the O K‐edge absorption spectra and corresponding differential spectra after 50 cycles charge states (4.5 V, green area) and discharge states (yellow area) at HDOD (2.7 V) and LDOD (3.3 V) cycling. (c) Evolution of normalized XANES spectra of Ni K‐edge during the initial discharge process and the 1st and 50th discharge states under HDOD (2.7 V) and LDOD (3.3 V) cycling. (d) The models of reversible redox capacity changes for O and Ni during long cycles of HDOD and LDOD cycling.
Structural evolution of LRLO cathode cycled under different DODs. (a) A schematic model illustrating coordination changes induced by the in‐plane migration of Mn within the TM layer. Green, blue, and red spheres denote Mn‐Mn3, Mn‐Mn4, and Mn‐Mn5 configurations, respectively. The red arrows depict in‐plane migration of TM. (b) The EXAFS spectra of Mn and the intensity variations of Mn‐TM for the 1st and 100th cycles under different DODs. Insert: Mn‐TM coordination and the locally enlarged image of LDOD cycling. (c) High‐resolution RIXS spectra recorded at an excitation energy of 531 eV for LRLO in the OCV, charged (4.5 V, 1st), and discharged states of the same 3.3 V of HDOD and LDOD cycling after 50 cycles. Insert: locally enlarged image of 0.125–2 eV. (d) The formation and evolution schematic illustrations of clusters and the out‐of‐plane migration pathways of TM during different DODs. Colour‐coding in insets is the same as the other structural Figures: green, Li; red, O; blue, TM. The black arrow signifies the electrostatic repulsion between TM at Li sites and TM within the TM layer.
Mechanistic evolution of cathode structure under different DODs cycling. (a) Ex situ sXRD patterns and refined results of HDOD and LDOD cycling after 100 cycles. Insets present the peak intensity ratio of (003)/(104) and the refined values of Li/Ni. Upper left inset: schematic of 003 and 104 crystal face. (b) GITT potential curve and chemical diffusion coefficient [log(DLi+)] of HDOD and LDOD cycling after 10 cycles. (c) TEM images for HDOD and (d) LDOD cycling after 100 cycles; the IFFT images represent surface (R1) and bulk (R2) regions also listed in the Figure.

+2

Depth‐of‐Discharge Dependent Capacity Decay Induced by the Accumulation of Oxidized Lattice Oxygen in Li‐Rich Layered Oxide Cathode
  • Article
  • Publisher preview available

December 2024

·

45 Reads

Kang Zhang

·

Yilong Chen

·

Yuanlong Zhu

·

[...]

·

Shi‐Gang Sun

More and more basic practical application scenarios have been gradually ignored/disregarded, in fundamental research on rechargeable batteries, e.g. assessing cycle life under various depths‐of‐discharge (DODs). Herein, although benefit from the additional energy density introduced by anionic redox, we critically revealed that lithium‐rich layered oxide (LRLO) cathodes present anomalously poor capacity retention at low‐DOD cycling, which is essentially different from typical layered cathodes (e.g. NCM), and pose a formidable impediment to the practical application of LRLO. We systemically demonstrated that DOD‐dependent capacity decay is induced by the anionic redox and accumulation of oxidized lattice oxygen (Oⁿ⁻). Upon low‐DOD cycling, the accumulation of Oⁿ⁻ and the persistent presence of vacancies in the transition metal (TM) layer intensified the in‐plane migration of TM, exacerbating the expansion of vacancy clusters, which further facilitated detrimental out‐of‐plane TM migration. As a result, the aggravated structural degradation of LRLO at low‐DOD impeded reversible Li⁺ intercalation, resulting in rapid capacity decay. Furthermore, prolonged accumulation of Oⁿ⁻ persistently corroded the electrode‐electrolyte interface, especially negative for pouch‐type full‐cells with the shuttle effect. Once the “double‐edged sword” effect of anionic redox being elucidated under practical condition, corresponding modification strategies/routes would become distinct for accelerating the practical application of LRLO.

View access options

Depth‐of‐Discharge Dependent Capacity Decay Induced by the Accumulation of Oxidized Lattice Oxygen in Li‐Rich Layered Oxide Cathode

November 2024

·

25 Reads

Angewandte Chemie

More and more basic practical application scenarios have been gradually ignored/disregarded, in fundamental research on rechargeable batteries, e.g. assessing cycle life under various depths‐of‐discharge (DODs). Herein, although benefit from the additional energy density introduced by anionic redox, we critically revealed that lithium‐rich layered oxide (LRLO) cathodes present anomalously poor capacity retention at low‐DOD cycling, which is essentially different from typical layered cathodes (e.g. NCM), and pose a formidable impediment to the practical application of LRLO. We systemically demonstrated that DOD‐dependent capacity decay is induced by the anionic redox and accumulation of oxidized lattice oxygen (On‐). Upon low‐DOD cycling, the accumulation of On‐ and the persistent presence of vacancies in the transition metal (TM) layer intensified the in‐plane migration of TM, exacerbating the expansion of vacancy clusters, which further facilitated detrimental out‐of‐plane TM migration. As a result, the aggravated structural degradation of LRLO at low‐DOD impeded reversible Li+ intercalation, resulting in rapid capacity decay. Furthermore, prolonged accumulation of On‐ persistently corroded the electrode‐electrolyte interface, especially negative for pouch‐type full‐cells with the shuttle effect. Once the “double‐edged sword” effect of anionic redox being elucidated under practical condition, corresponding modification strategies/routes would become distinct for accelerating the practical application of LRLO.


Decoupling the Failure Mechanism of Li‐Rich Layered Oxide Cathode During High‐Temperature Storage in Pouch‐Type Full‐Cell: A Practical Concern on Anionic Redox Reaction

In addressing the global climate crisis, the energy storage performance of Li‐ion batteries (LIBs) under extreme conditions, particularly for high‐energy‐density Li‐rich layered oxide (LRLO) cathode, is of the essence. Despite numerous researches into the mechanisms and optimization of LRLO cathodes under ideal moderate environment, there is a dearth of case studies on their practical/harsh working environments (e.g., pouch‐type full‐cell, high‐temperature storage), which is a critical aspect for the safety and commercial application. In this study, using pouch‐type full‐cells as prototype investigation target, the study finds the cell assembled with LRLO cathode present severer voltage decay than typical NCM layered cathode after high‐temperature storage. Further decoupling elucidates the primary failure mechanism is the over‐activation of lattice oxidized oxygen (aggravate by high‐temperature storage) and subsequent escape of oxidized oxygen species (Oⁿ⁻), which disrupts transition metal (TM) coordination and exacerbates electrolyte decomposition, leading to severe TM dissolution, interfacial film reconstruction, and harmful shuttle effects. These chain behaviors upon high‐temperature storage significantly influence the stability of both electrodes, causing substantial voltage decay and lithium loss, which accelerates full‐cell failure. Although the anionic redox reaction can bring additional energy, but the escape of metastable Oⁿ⁻ species would introduce new concerns in practical cell working conditions.



Characterization of Na2O‐based high capacity presodiation agent. a) Voltage profile for (Na2O + NiO) composite cathode at 10 mA g⁻¹ with a 4.5 V upper cutoff potential; the inset shows the pure Na2O cathode capacity (<15 mAh g⁻¹) at cutoff 4.5 V. b) Schematic illustration of the preparation procedure of NNO composite presodiation agent by high‐energy ball milling. c) sXRD pattern and d) NPD pattern of NNO presodiation agent and the results of fitting via Rietveld refinement. e) The density of states (DOS) of Na2O (top) and Ni–Na2O (bottom). f) Gibbs free energy diagrams during the pure Na2O and Ni–Na2O decomposition process.
The structural and local covalent environment evolution during NNO decomposition. a) Ex situ sXRD patterns and the results of fitting via Rietveld refinement of NNO presodiation agent in different voltages. The sXRD peak intensities of Ni–Na2O presodiation agent at 2θ = 14.0°, 19.8°, 23.3°, 24.3°, and 28.3° (Ni–Na2O), 2θ = 16.4°, 18.9°, and 26.9° (NiO), respectively. b–d) TEM images of charged NNO presodiation agent cathode at pristine, charge 4.3 V, and discharge 1.5 V. Insets present the corresponding SEAD patterns. The red and yellow marker lines represent the NiO and Ni–Na2O phases, respectively, and the white ones represent the amorphous phase. e) Ni K‐edge X‐ray absorption near‐edge structure (XANES) spectra at different voltage states. Upper left inset: enlarged pre‐edge of Ni K‐edge XANES spectra and a schematic illustration of Ni–O tetrahedral (tetra.) configuration (Ni in Na2O) and octahedral (octa.) configuration (Ni in NiO). Inset below right: edge positions of Ni K‐edge at different voltage states. f) Ni K‐edge EXAFS spectra (weighted by k³) of pristine and charged 4.5 V NNO presodiation agent. The Ni molar percentage pie chart of Ni–Na2O (green region) and NiO (purple region) in NNO presodiation agent. The fitted Ni–O coordination numbers (C.N.) are shown in the inset.
Analysis of oxygen behavior upon Na2O oxidation in NNO. a) The top panel shows the galvanostatic charge curve for the initial charging process of the NNO presodiation agent cathode at a current density of 50 mA g⁻¹. Five points are labeled in the charge curve: OCV, 3.25 V,3.45 V, 3.65 V, and 4.5 V, respectively. b) The middle panel shows OEMS results of corresponding time‐resolved evolution rates for O2 and CO2 during initial charging. c) TMS result: amounts of O2 and CO2 collected from the NNO presodiation agent plates with different specific voltages. d) O K‐edge XANES spectra of the NNO presodiation agent cathode and standard samples at TEY modes. The peak at 533 eV represents oxygen in the antifluorite structure of Na2O; the peak at 531.4 eV represents σ* (O─O) peroxide species; and the peak at 531.2 eV represents the hybridized state of O 2p and Ni 3d orbitals in NiO. e) The schematic diagram illustrating the dynamics of electrochemical transfer oxidation process during the charging of Ni–Na2O. The electronic structure of Na2O exhibits a charge transfer electronic ground state in the reduced phase (left). The O (2p) lone pairs denote |O2p. The splitting of the O 2p narrow band into distinct σ, π, π*, and σ* states is illustrated by the red (ΔσO─O) and green (ΔπO─O) arrows, with respect to the conversion of O─O dimer species (bottom horizontal axis).
Electrochemical performance after presodiation. a) Initial charge profiles of Na3V2(PO4)3 (NVP) and Na2/3Ni2/3Mn1/3O2 (NNMO) with n wt % (n = 0%, 5%, and 10%) NNO presodiation agent cathodes at 10 mA g⁻¹ in the range from 2.5 to 4.3 V and from 2.0 to 4.15 V, respectively. b) Cycling performance of NVP without or with NNO presodiation agent cathodes at 50 mA g⁻¹ in the range from 2.5 to 4.3 V. Inset: the bar chart corresponds to the charge and discharge capacity of the initial cycle at 10 mA g⁻¹. c) Galvanostatic charge/discharge curves of HC||NVP without or with 10 wt% NNO presodiation agent coin‐type full‐cell at 10 mA g⁻¹ (1 st) and 50 mA g⁻¹ (5–50 th) in the range from 1.0 to 4.2 V. d) Cycling performance of HC||NVP without or with 10 wt% NNO presodiation agent coin‐type full‐cell in the range from 1.0 to 4.2 V. e) Galvanostatic charge/discharge curves of HC||NNMO without or with 10 wt% NNO presodiation agent coin‐type full‐cell at 10 mA g⁻¹ (1st) and 50 mA g⁻¹ (5–50 th) in the range from 0.5 to 4.0 V. f) Cycling performance of HC||NNMO without or with 10 wt% NNO presodiation agent coin‐type full‐cell in the range from 0.5 to 4.0 V.
The impact of NNO on the electrode–electrolyte interface and associated perspective on presodiation agent. a) TOF‐SIMS characterization of pure NVP and NVP‐NNO cycled cathode electrodes after 150 cycles. The normalized depth profiles of the interface and bulk fragments illustrate the structure of CEI. b) 3D renderings of selected secondary ion fragments of different CEI. The sputtered volume is 100 µm (length) × 100 µm (width) × 150 nm (height). c) TEM images of NVP‐NNP after 200 cycles. The IFFT results for the surface and bulk regions are also listed in the figure. The light green‐dashed ground state region is identified as NVP structure, white is defined as carbon layer region, and yellow is assessed as a CEI architecture. d) Scheme of CEI formation and changes on NVP and NVP‐NNO cathodes summarized from characterization data in FEC‐containing electrolytes. e) Perspectives for cathode presodiation agent.
Achieving High‐Capacity Cathode Presodiation Agent Via Triggering Anionic Oxidation Activity in Sodium Oxide

July 2024

·

86 Reads

·

5 Citations

Compensating for the irreversible loss of limited active sodium (Na) is crucial for enhancing the energy density of practical sodium‐ion batteries (SIBs) full‐cell, especially when employing hard carbon anode with initially lower coulombic efficiency. Introducing sacrificial cathode presodiation agents, particularly those that own potential anionic oxidation activity with a high theoretical capacity, can provide additional sodium sources for compensating Na loss. Herein, Ni atoms are precisely implanted at the Na sites within Na2O framework, obtaining a (Na0.89Ni0.05□0.06)2O (Ni–Na2O) presodiation agent. The synergistic interaction between Na vacancies and Ni catalyst effectively tunes the band structure, forming moderate Ni–O covalent bonds, activating the oxidation activity of oxygen anion, reducing the decomposition overpotential to 2.8 V (vs Na/Na⁺), and achieving a high presodiation capacity of 710 mAh/g≈Na2O (Na2O decomposition rate >80%). Incorporating currently‐modified presodiation agent with Na3V2(PO4)3 and Na2/3Ni2/3Mn1/3O2 cathodes, the energy density of corresponding Na‐ion full‐cells presents an essential improvement of 23.9% and 19.3%, respectively. Further, not limited to Ni–Na2O, the structure–function relationship between the anionic oxidation mechanism and electrode–electrolyte interface fabrication is revealed as a paradigm for the development of sacrificial cathode presodiation agent.



[a] Typical initial galvanostatic charge/discharge curves of LRLO (2.0–4.8 V) cathode. Current density: 10 mA/g. [b] Potential/Time‐dependent in situ SERS spectra of LRLO recorded during first charging. The black traces present the Raman signal collected at open circuit voltage (OCV) for comparing. The newly‐produced Raman peaks have been assigned to superoxo, peroxo and Li2CO3 species, respectively. [c] ¹⁹F and ¹H NMR spectra of electrolytes after 30 cycles (extracted by D2O solvent, benzene as internal standard for fair normalization) of LRLO in different electrolytes, asterisks represented products unknown. [d] Possible reaction Schemes of HFE under nucleophilic attack on LRLO consistent with experimental and calculated results displayed. (e Schematic illustration of HFE intrinsic mechanism on LRLO interface for CEI fabrication.
[a] EIS‐GITT voltage profiles and corresponding logarithm of DLi+ of LRLO in EE and EEF electrolytes during first charge. [b] Picked Nyquist plots at 5th, 10th and 15th charging points from Figure 2 (a). [c,d] The variation of RCEI and Rct of LRLO during first charging in EE and EEF electrolytes. [e,f] Nyquist plots for LRLO in EE and EEF at different cycles. Corresponding fit results for the electrochemical impedance RCEI and Rct at the 3rd, 10th and 50th cycles. [g,h] TEM images of LRLO after 100 cycles using EE and EEF electrolytes. The IFFT results for the surface and bulk regions are also listed in Figure. The orange‐dashed ground state region is identified as a spinel structure, blue is defined as a distorted region, and red is assessed as a CEI morphology.
[a] Charge and discharge curve of LRLO by CCCV protocol. [b] The comparison of overall polarization ΔVcha‐dis (Vcha minus Vdis) on LRLO during cycling in different electrolytes. [c] The cycling performance (discharge capacity) of LRLO in EE and EEF electrolytes. [d] The variation of average voltage during charging and discharging along with cycling of LRLO in EE and EEF electrolytes. [e] The attribution of polarization. [f] The comparison of Rv and Lv changes during cycling in EE and EEF electrolytes. [g] The ratio of the charging capacity of the CV procedure to the total capacity of the current cycle at different cycle numbers in EE and EEF electrolytes. [h] The ratio of the irreversible capacity to the total capacity of the current cycle at different cycle numbers in EE and EEF electrolytes. [i] Floating current is obtained by holding 10 h at 4.8 V and the variation of floating current during cycling in EE and EEF electrolytes.
[a] O K‐edge sXAS spectra of discharged LRLO CEI with etching, fabricated by different electrolytes, including EEand EEF. [b] Characterization of the components of the CEI layer on LRLO after 50 cycles in different electrolytes via XPS F1s spectrums. [c] The concentration variation of elements C, O, F and P on LRLO in EE‐ and EEF‐CEI after different cycles with and without etching. [d] TOF‐SIMS characterization of the cycled cathode electrodes in EE and EEF after 100 cycles. The normalized depth profiles of the interface and bulk fragments illustrate the structure of the CEI. [e] 3D renderings of selected secondary ion fragments of different CEI. The sputtered volume is 100 μm (length) ×100 μm (width)×150 nm (height). [f] The CEI construction mechanism of HFE.
[a,b] Comparison of SEM morphologies on individual LRLO particles and electrode cracks after 300 cycles in EE and EEF. [c,d] The cross‐section images of electrodes after cycling in different electrolytes, local region is enlarged for comparing by‐products deposition from electrolyte decomposition. [e] Raman spectra of different LRLO samples after 100 cycles in EE and EEF compared with the pristine electrode (dark plots). [f] The comparison of structural rearrangement of LRLO cycled in EE and EEF via Raman intensity ratio of Ilayer/Ispinel and XRD intensity ratio of I(003)/I(104).
Manipulated Fluoro‐Ether Derived Nucleophilic Decomposition Products for Mitigating Polarization‐Induced Capacity Loss in Li‐Rich Layered Cathode

January 2024

·

72 Reads

·

15 Citations

Electrolyte engineering is a fascinating choice to improve the performance of Li‐rich layered oxide cathodes (LRLO) for high‐energy lithium‐ion batteries. However, many existing electrolyte designs and adjustment principles tend to overlook the unique challenges posed by LRLO, particularly the nucleophilic attack. Here, we introduce an electrolyte modification by locally replacing carbonate solvents in traditional electrolytes with a fluoro‐ether. By benefit of the decomposition of fluoro‐ether under nucleophilic O‐related attacks, which delivers an excellent passivation layer with LiF and polymers, possessing rigidity and flexibility on the LRLO surface. More importantly, the fluoro‐ether acts as “sutures”, ensuring the integrity and stability of both interfacial and bulk structures, which contributed to suppressing severe polarization and enhancing the cycling capacity retention from 39 % to 78 % after 300 cycles for the 4.8 V‐class LRLO. This key electrolyte strategy with comprehensive analysis, provides new insights into addressing nucleophilic challenge for high‐energy anionic redox related cathode systems.


Manipulated Fluoro‐Ether Derived Nucleophilic Decomposition Products for Mitigating Polarization‐Induced Capacity Loss in Li‐Rich Layered Cathode

December 2023

·

23 Reads

Angewandte Chemie

Electrolyte engineering is a fascinating choice to improve the performance of Li‐rich layered oxide cathodes (LRLO) for high‐energy lithium‐ion batteries. However, many existing electrolyte designs and adjustment principles tend to overlook the unique challenges posed by LRLO, particularly the nucleophilic attack. Here, we introduce an electrolyte modification by locally replacing carbonate solvents in traditional electrolytes with a fluoro‐ether. By benefit of the decomposition of fluoro‐ether under nucleophilic O‐related attacks, which delivers an excellent passivation layer with LiF and polymers, possessing rigidity and flexibility on the LRLO surface. More importantly, the fluoro‐ether acts as "sutures", ensuring the integrity and stability of both interfacial and bulk structures, which contributed to suppressing severe polarization and enhancing the cycling capacity retention from 39% to 78% after 300 cycles for the 4.8 V‐class LRLO. This key electrolyte strategy with comprehensive analysis, provides new insights into addressing nucleophilic challenge for high‐energy anionic redox related cathode systems.

Citations (1)


... As displayed in Figure 5i, Figure S43 and Figure S44, the multi-component nature of SEI is demonstrated by the observed amorphous and crystalized region with thickness of around 120 nm. A mixture of crystalline lattice fringes of ZnO (100), Zn 5 -(OH) 8 Cl 2 · H 2 O (101) are identified in the SEI, which may possess high interfacial energy to suppress the detrimental side reactions [45] (Figure 5j-5l). Compared to HAE, the poly-inorganic composite aggregates and larger crystal lattice area can be detected with numerous non-uniform protrusions in Aqua, which may irreversibly block ion transport and cause accumulation of by products in the stripping/plating processes. ...

Reference:

Evolution of Frustrated Coordination in Eutectic Electrolyte Driven by Ligand Asymmetry toward High‐Performance Zinc Batteries
Gradient Interphase Engineering Enabled by Anionic Redox for High-Voltage and Long-Life Li-Ion Batteries
  • Citing Article
  • February 2024

Journal of the American Chemical Society