Insights into Homogeneous Bulk Boron Doping at the Tetrahedral Site of NCM811 Cathode Materials: Structure Stabilization by Inductive Effect on TM-O-B Bonds

Abstract

Rechargeable lithium-ion batteries (LIBs) are critical for enabling sustainable energy storage. The capacity of cathode materials is a major limiting factor in the LIB performance, and doping has emerged as an effective strategy for enhancing the electrochemical properties of nickel-rich layered oxides such as NCM811. In this study, boron is homogeneously incorporated into the tetrahedral site of NCM811 through co-precipitation, leading to an inductive effect on transition metal (TM)-O-B bonds that delayed structural collapse and reduced oxygen release. Consequently, these changes culminate in an enhancement of cycling performance, translating to an initial specific capacity of 210 mAh g^-1 and a 95.3% capacity retention after 100 cycles. These interesting findings deepen the understanding of boron doping and shed light on the design of better lithium cathode materials on an applicable scale.

Keywords: B‐doping; cathode material; lithium‐ion battery.

Figure 1

a) XRD Patterns of NCM811_xB, where x = 0, 0.5, 1, 2, 5 at. %, SEM pictures of b₁) NCM811, b₂) NCM811_2B and, b₃) NCM811_5B, b₄) TEM image of NCM811_5B.

Figure 2

Rietveld refinement against the NCM811_5B neutron diffraction data. a) without B, b) with B at the tetrahedral site, c) F_obs‐F_calc value obtained from Rietveld refinement analysis of NCM811_5B neutron diffraction. d) Schematic illustration shows Li and TM octahedral and tetrahedral sites. Difference Fourier maps of (101) plane in NCM811_5B structure, e) without B in the structure, f) with B in tetrahedral site, g) Schematic illustration of (101) plane at d = 1 in NCM811 unit cell, created by VESTA software.^{[ 21 ]}

Figure 3

a) The dQ/dV versus voltage curves were derived from GITT data (see Figure S5, Supporting Information), where voltage points are collected after each rest step, b) Lithium diffusion coefficient (D_Li+) calculated based on GITT data. Rate performance of NCM811_xB versus Li/Li⁺ in the voltage window of c) 3.0–4.4 V, d) 3.0–4.3 V. Cycling stability of NCM811_xB versus Li/Li⁺ in the voltage window of e) 3.0–4.4 V, f) 3.0–4.3 V. Initially, cycling was conducted at a constant current of 9 mA g⁻¹ (0.05C‐rate) for the first two cycles, followed by subsequent cycles of 90 mA g⁻¹ (0.5C‐rate). b–f) include error bars.

Figure 4

NEXAFS data of the pristine NCM811_xB (x = 0, 2, 5 at. %), a) O K edge and b) Ni L edge c) Bonding energy of TM‐O bond is lowered by inductive effect, leading to decreased covalency and a lower chemical potential μ. The electrochemical voltage (V_oc) is determined by the difference in chemical potential between the anode and the lowest unoccupied 3d metal orbital.

Figure 5

a,b) The contour plot displays the in operando XRD data for NCM811 and NCM811_2B at C/20 with a voltage window of 3.0–4 .4V. c–f) The variations in lattice parameters a (= b) and c, isotropic and anisotropic strain values during the initial cycle were determined using Rietveld refinement. The presented values are accompanied by estimated standard deviation. For details about the refinement of isotropic and anisotropic strains, see Supporting Information.

Figure 6

NEXAFS data of cycled samples on O K edge with 0%, 50%, 75%, 80%, and 100% states of charges (SOCs) in partial fluorescence yield. Cycled NCM811 a) at third cycles (cyc.), b) at 103rd cyc. Cycled NCM811_2B c) at 3rd cyc., d) at 103rd cyc.