In Situ Chemical Modulation of Graphitization Degree of Carbon Fibers and Its Potassium Storage Mechanism

Abstract Graphite is considered to be the most auspicious anode candidate for potassium ion batteries. However, the inferior rate performances and cycling stability restrict its practical applications. Few studies have investigated the modulating the graphitization degree of graphitic materials. Herein, a nitrogen‐doped carbon‐coated carbon fiber composite with tunable graphitization (CNF@NC) through etching growth, in‐situ oxidative polymerization, and subsequent carbonization process is reported. The prepared CNF@NC with abundant electrochemical active sites and a rapid K+/electron transfer pathway, can effectively shorten the K+ transfer distance and promote the rapid insertion/removal of K+. Amorphous domains and short‐range curved graphite layers can provide ample mitigation spaces for K+ storage, alleviating the volume expansion of the highly graphitized CNF during repeated K+ insertion/de‐intercalation. As expected, the CNF@NC‐5 electrode presents a high initial coulombic efficiency (ICE) of 69.3%, an unprecedented reversible volumetric capacity of 510.2 mA h cm−3 at 0.1 A g−1 after 100 cycles with the mass‐capacity of 294.9 mA h g−1. The K+ storage mechanism and reaction kinetic analysis are studied by combining in‐situ analysis and first‐principles calculation. It manifests that the K+ storage mechanism in CNF@NC‐5 is an adsorption‐insertion‐insertion mechanism (i.e., the “1+2” model). The solid electrolyte interphase (SEI) film forming is also detected.

The Rs, Rf, Rct, and RW signified the internal resistance, the electrolyte/SEI interface resistance, the charge-transfer resistance, and the Warburg resistance, respectively. [1]e diffusion coefficient of K + could be calculated by the following formula: In this formula, DK was the diffusion coefficient of K + , R was the gas constant (8.314J mol -1 K -1 ), T was the test temperature of battery (298 K), F was the Faraday constant (96500 C mol -1 ), A was the surface area of electrode (1.13 cm -2 ), n was the number of electrons involved in the insertion reaction (n = 1), C was the K + concentration in electrode (C = 1.0×10 -3 mol cm -1 ) and σ' was the Warburg coefficient equaled to the slopes of the Z'(Ω) and ω -1/2 (ω = 2πf) lines in the low-frequency region. [2]

Figure S4 .
Figure S4.The ratio of mesoporous and microporous for Brunauer-Emmett-Teller and pore volume of CNF and CNF@NC-5.

Figure S8 .
Figure S8.The charge-discharge curves for selected cycles of CNF electrode at 0.1 A g -1 .

Figure S19 .
Figure S19.The CNF@NC-5 electrode after 2000 cycles: a) low-magnification and b) high-magnification SEM images; c) the corresponding EDS mapping images.

Figure S20 .
Figure S20.An equivalent circuit for EIS analysis.

Figure S21 .
Figure S21.EIS measurement at various temperatures of a) CNF@NC-5 electrode, and b) CNF electrode.c) The calculation of activation energy of CNF and CNF@NC-5 electrodes.

Figure S22 .
Figure S22.Determination of ΔEt and ΔEs from the measured GITT profiles.

Figure S23 .
Figure S23.CNF@NC-5//AC PIHCs.a) Diagram of PIHCs.b) CV profiles of AC in half cells between 2.0 and 4.0 V vs K + /K at 0.5 mV s -1 .c) The 2nd, 3rd, and 4th chargedischarge curves at 0.1 A g -1 of AC in half cells.d) Potassiation and depotassiation capacity and coulombic efficiency at 0.1 A g -1 of AC in half cells.e) Long-term cycle performance of CNF@NC-5//AC PIHCs at 1.0 A g -1 over 3000 cycles.

Table S1 .
The volumetric specific capacity of reported carbon materials in KIBs.

Table S2 .
Electrochemical performance of reported carbon materials in KIBs.