Arginine modification of hybrid cobalt/nitrogen Ti₃C₂T_x MXene and its application as a sulfur host for lithium-sulfur batteries

INTRODUCTION

Lightweight and long service rechargeable batteries have been developed to satisfy the increasing demands of industrial needs, military progress and portable life^[1-3]. Furthermore, the development of advanced electrochemical energy conversion and storage systems with a high energy density has been gaining significant attention^[4-7]. Despite the various advantages of existing lithium-ion batteries, their low energy density based on the intercalation mechanism has severely hindered the scale-up fabrication of miniature batteries^[5]. Lithium-sulfur (Li-S) batteries with high theoretical specific capacity (1,675 mAh g^-1) and energy density (2,567 Wh kg^-1) have attracted considerable attention as the next-generation batteries^[8-11]. However, their low sulfur utilization, sluggish redox reaction kinetics, fast capacity fading, and low coulombic efficiency (CE) have hindered their commercialization^[12,13].

To address these issues, several studies have reported the development of advanced sulfur host materials for the Li-S cathodes^[14,15]. For example, carbon materials and conducting material with large specific surface area have been used to inhibit the shuttle effect of polysulfide. In particular, two-dimensional conductive carbon materials have been employed owing to their adjustable morphology and high electrical conductivity^[16-18]. Nevertheless, the interactions are weak between polar lithium polysulfides (LiPSs) and the nonpolar hydrophobic carbon framework play a very limited role in slowing down the shuttle effect or preventing interfacial charge transfer, thus leading to sluggish reaction kinetics^[17,19,20].

The application of Ti₃C₂T_x MXene in electrochemical energy-storage applications has been gaining significant attention owing to their unique properties^[21,22-25], where several studies have employed Ti₃C₂T_x as an electrode substrate for Li-S batteries^[24,26,27] owing to its inherent metallic conductivity (≥ 3.4 × 10³ S m^-1) which can effectively promote a rapid electron transfer. Furthermore, Ti₃C₂T_x MXene has abundant active sites and a variety of surface functional groups which can not only anchor and catalyze the polysulfide lithium through the interaction between the metal and sulfur, but also accelerate the sulfur redox reaction. Despite these advantages, there are still two key challenges facing the optimization of its electrochemical performance^[28,29]: (1) Ti₃C₂T_x nanosheets are easily stacked, which reduces the specific surface area and leads to a decrease in the utilization rate of sulfur; and (2) the surface-active sites cannot easily catalyze the adsorption of large quantities of LiPSs.

Several strategies have focused on increasing the layer spacing and modifying the Ti₃C₂T_x morphology with amino acids^[30,31]. In particular, small molecules that influence Ti₃C₂T_x can weaken the interlayer van der Waals forces and provide a greater interlayer advantage. Unlike large molecules and nanocarbon materials, small molecules do not introduce an insulation phase and avoid the hydrophobicity of the composite materials^[32]. Therefore, there is an urgent need to find appropriate small molecules that can regulate the interlaminar structure and increase the surface area of Ti₃C₂T_x. In addition, the active elements, such as N, Co, Ni, can effectively improve the electrochemical performance of lithium sulfur batteries^[33-35]. Consequently, an efficient method to treat the Ti₃C₂T_x surface termination groups is highly desirable to further improve the electrochemical performance and increase the structural variety of Ti₃C₂T_x based sulfur hosts.

This paper reports the fabrication of amino acid-regulated Co, N-doped Ti₃C₂T_x (Co-N@Ti₃C₂T_x) substrates bearing multiple polysulfide anchoring sites, which are expected to impart a high polysulfide conversion activity and real capacity to Li-S batteries. The interlayer spacing of Ti₃C₂T_x was extended via an amino acid modification. Arginine, serine, and lysine were employed to obtain Ti₃C₂T_x-Arg, Ti₃C₂T_X-Ser, and Ti₃C₂T_X-Lys, respectively_. The effect of the amino acid chain length on the interlayer spacing of Ti₃C₂T_x was investigated. Finally, the possible application of each of the Co-N@Ti₃C₂T_X-Arg, Co-N@Ti₃C₂T_X-Ser, and Co-N@Ti₃C₂T_X-Lys composites as an S host was assessed. The results revealed that the interlayer spacing of Ti₃C₂T_x increased with an increase in the amino acid chain length. The Co-N@Ti₃C₂T_x-Arg composites exhibited a stronger adsorption for LiPSs via the Co, N active site which suppressed the LiPSs shuttle effect and catalyzed the redox reaction. Consequently, the application of the Co-N@Ti₃C₂T_x-Arg composite as a host of S cathode delivered a capacity of 1,365 mAh g^-1 and exhibited stable cycles with a capacity fading rate of 0.083% per cycle over 200 cycles at 0.2 C.

MATERIALS AND METHODS

RESULTS AND DISCUSSION

Figure 1 illustrates the procedure for the synthesis of the Co-N@Ti₃C₂T_x-Ser, Co-N@Ti₃C₂T_x-Lys, and Co-N@Ti₃C₂T_x-Arg composites as a sulfur host. Ti₃C₂T_x was obtained by selectively etching Ti₃AlC₂ by HCl/LiF to remove the Al atomic layer. Serine, lysine, and arginine (with different chain lengths) were chosen as the interlayer regulators owing to their dual function of being a nitrogen source and widening the Ti₃C₂T_x layer spacing. The nitrogen-source-modified Ti₃C₂T_x was used to build a self-supporting electrode and the self-supporting film was then successfully assembled by vacuum filtration. CoCl₂ was deposited onto the Ti₃C₂T_x film by soaking, and Co and N were doped into the Ti₃C₂T_x film after annealing. As the cathode substrate of Li-S batteries, Co-N@Ti₃C₂T_x-Arg did not only extend the spacing of the nanosheets, but also increased the number of catalytic active sites.

Figure 1. Schematic representation of the synthesis of the Co-N@Ti₃C₂T_x composites.

The SEM results of the Ti₃AlC₂ and etched multilayer Ti₃C₂T_x (Mul-Ti₃C₂T_x) are shown in Figure 2A and B, respectively, while those of Ti₃C₂T_X-Ser, Ti₃C₂T_X-Lys, and Ti₃C₂T_x-Arg, are shown in Figure 2C-E, respectively. The surface of Co-N@Ti₃C₂T_x-Arg [Figure 2F] exhibited some roughness which can be attributed to the capillarity of thermal annealing. Cross sectional SEM images show that the thickness of the Co-N@Ti₃C₂T_x-Arg (31.6 μm) is a few high than that of Co-N@Ti₃C₂T_x-Ser and Co-N@Ti₃C₂T_x-Lys (29.2, 30.8 μm) due to longer chain of arginine [Supplementary Figure 1]. Figure 2G represents the SEM/EDS image of Co-N@Ti₃C₂T_X-Arg. The results [Figure 2H-J] revealed a uniform distribution of N and Co, thus indicating the successful modification of Ti₃C₂T_x by the amino acid and cobalt dichloride. The homogeneous distribution of elements Co throughout Co-N@Ti₃C₂T_x-Arg was visualized by EDS elemental mapping [Supplementary Figure 2]. The EDS spectral images of Co-N@Ti₃C₂T_X-Ser and Co-N@Ti₃C₂T_X-Lys[Supplementary Figures 3 and 4] also confirmed the successful doping of Co and N into the Ti₃C₂T_x composites. Nevertheless, the layer space of the Ser-, Lys-, and Arg-modified Ti₃C₂T_x nanosheets varied, and the crumpling increased with an increase in the amino acid chain length. The XRD pattern of the raw material Ti₃C₂T_x [Supplementary Figure 5] matched with the JCPDS data card. The disappearance of the (104) diffraction peak from the XRD pattern after HCl/LiF etching confirmed that most of the Al in the MAX material was successfully etched.

Figure 2. SEM images of the (A and B) multilayer Ti₃C₂T_X, (C-E) Ti₃C₂T_X-Ser, and Ti₃C₂T_X-Lys, and Ti₃C₂T_x-Arg, composites; (F-J) SEM image and EDS spectrum mapping: C, Co, and N elemental distribution of the Co-N@Ti₃C₂T_x-Arg composite.

The schematic representation of the amino-acid-modified Ti₃C₂T_x composites with LiF/HCl etching process [Figure 3A] revealed the abundant presence of hydroxyl groups in surface of Ti₃C₂T_x, which can interact with the abundant hydroxyl sites of amino acids to obtain the Co-Ti₃C₂T_x-Ser, Co-Ti₃C₂T_x-Lys, and Co-Ti₃C₂T_x-Arg composites by esterification. The carboxyl groups of the amino acids and the hydroxyl groups on Ti₃C₂T_x were esterified to form longer surface groups on the Ti₃C₂T_X, which is consistent with the morphological changes observed by SEM. Figure 3B shows the XRD patterns of Co-Ti₃C₂T_x-Ser, Co-Ti₃C₂T_x-Lys,Co-Ti₃C₂T_x-Arg, and multilayer MXene (Mul-Ti₃C₂T_x) composites. For Ti₃C₂T_X, the position of the characteristic (002) peak reflected the size of the layer spacing^[38]. The corresponding characteristic angles of the (002) crystal planes of Co-Ti₃C₂T_X-Ser, Co-Ti₃C₂T_X-Lys, and Co-Ti₃C₂T_X-Arg were at 6.47°, 6.3°, and 6.11°, respectively, and the values of the layer spacing obtained by the Bragg equation (2d sin θ = nλ) were 1.37, 1.4, and 1.45 nm, respectively, where d is the distance between the crystal faces, λ is the wavelength of the X-ray (λ = 0.15406 nm),θ is the angle between the corresponding crystal faces and the incident X-ray, n is the diffraction series (generally 1), thus indicating that the layer spacing can be effectively increased by an amino acid modification. The Raman spectra of the modified Co-N@Ti₃C₂T_x were aligned with that of Mul-MXene [Figure 3C]. The spectra revealed that the intensities of Co-N@Ti₃C₂T_X-Ser, Co-N@Ti₃C₂T_X-Lys, and Co-N@Ti₃C₂T_x-Arg gradually increased which can be attributed to the change in the surface roughness of the nanosheets after annealing, which is consistent with the SEM results. A disordered carbon (D band) and in-plane vibrations (G band) characteristic peaks was existed at 1,350 and 1,590 cm^-1[39]. A comparison of the D- and G-peak intensities for Mul-Ti₃C₂T_X and the amino acid-modified Ti₃C₂T_X nanosheets revealed that the D-peak to G-peak intensity ratio (I_D/I_G) of Mul-MXene was higher than those of the amino acid-modified Ti₃C₂T_X, thus indicating that Mul-MXene has more carbon vacancies or larger graphitic regions, which confirmed the doping of nitrogen atoms by replacing some carbon atoms in the Ti₃C₂T_x after pyrolysis. The surface chemical composition of Co-N@Ti₃C₂T_x was further characterized by XPS. In comparison with the XPS spectrum of Mul-Ti₃C₂T_x nanosheets [Supplementary Figure 6], the peaks at 779.8 and 400 eV indicated the presence of N and Co on the Ti₃C₂T_X, respectively^[24]. The C 1s spectra [Figure 3D, Supplementary Figures 7 and 8] revealed the presence of five peaks at approximately 281.8, 284.6, 285.2, 286.1, and 288.5 eV corresponding to the C-Ti, C-C, C-N, C-O, and O=C-O bonds, respectively. The disappearance of the C-Ti-O bond in the modified Ti₃C₂T_x compared to that of Mul-Ti₃C₂T_x confirmed the reaction of amino acids with the-OH functional group on the Ti₃C₂T_x surface. The additional C-N bond confirmed the interaction between the three amino acids and Ti₃C₂T_X. The O 1s spectra [Figure 3E, Supplementary Figures 7 and 8] revealed the presence of three peaks at approximately 529.2, 530.7, and 532.2 eV attributed to the presence of Co-O, O-C=O, and oxygen vacancies, respectively. The Co 2p spectra [Figure 3F, Supplementary Figures 7 and 8] were decomposed into two distinct peaks, Co 2p3/2 (780.2 eV) and Co 2p1/2 (795.7 eV), thus confirming the successful modification of the cobaltous oxide (CoO) nanoparticles in the three different pleated MXene nanosheets. The high-resolution N 1 s spectra [Figure 3G, Supplementary Figures 7 and 8] revealed the presence of pyridinic-N (397.5 eV), pyrrolic-N (398.3 eV), and quaternary-N (400.6 eV) in the Co-N@Ti₃C₂T_X-Ser, Co-N@Ti₃C₂T_X-Lys, and Co-N@Ti₃C₂T_x-Arg composites, thus confirming that N was doped into Ti₃C₂T_x rather than being present as a residue or impurity. The Co-N@Ti₃C₂T_X-Ser composite was composed of 48.65% Ti-N bond, 16.2% pyrrolic N, and 35.14% pyridinic N, while the Co-N@Ti₃C₂T_X-Lys composite was composed of 24.24% Ti-N bond, 30.30% pyrrolic N, and 45.45% pyridinic N. The Co-N@Ti₃C₂T_x-Arg composite was composed of 11.11% Ti-N bond, 59.26% pyrrolic N, and 11.11% pyridinic N^[40]. The proportion of the Ti-N bonds decreased with an increase in the amino acid chain length, which can be attributed to a mount of NH₃ being produced during metal doping, which reduced the number of generated Ti-N bonds. Binding energy and affinity of polar titanium with elemental sulfur (S8) and polar polysulfide (Li₂S_n, 4 ≤ n ≤ 8) can be effectively enhanced by strong Lewis acid-base interactions with pyridine N and pyrrole N^[41]. The Ti 2p spectra [Figure 3H, Supplementary Figures 7 and 8] revealed three Ti 2p1/2/ Ti 2p3/2 double bonds: Ti-O (465.7/458.7 eV, 465.6/459.6 eV, 464.2/458.1 eV), Ti-N (462.6/456.4 eV, 462.6/456.7 eV, 463.8/457.6 eV), and Ti-C (461.2/455.4 eV, 461.5/455.5 eV, 461.2/454.5 eV), respectively. Compared to the pure Ti₃C₂T_X, the Ti-O bonding in the Co-N@Ti₃C₂T_X-Ser, Co-N@Ti₃C₂T_X-Lys, and Co-N@Ti₃C₂T_x-Arg composites increased, which can effectively anchor polysulfides. The Ti-O bond content initially increased and then decreased, indicating the Co-N@Ti₃C₂T_X-Ser,Co-N@Ti₃C₂T_X-Lys and Co-N@Ti₃C₂T_x-Arg have different degrees of oxidation during synthesis.

Figure 3. (A) Schematic illustration of the Ti₃C₂T_x modified by the amino acids. (B) XRD patterns of the Co-Ti₃C₂T_x-Ser, Co-Ti₃C₂T_x-Lys, Co-Ti₃C₂T_x-Arg, and Mul-Ti₃C₂T_x composites. (C) Raman spectra of the Co-N@Ti₃C₂T_x-Ser, Co-N@Ti₃C₂T_x-Lys, Co-N@Ti₃C₂T_x-Arg, and Mul-Ti₃C₂T_x composites from 0 to 3,250 cm^-1. (D) C 1s XPS spectrum of the crumpled Co-N@Ti₃C₂T_x-Arg composite. (E) O 1s spectrum of the Co-N@Ti₃C₂T_x-Arg composite. (F) Co 2p spectrum of the Co-N@Ti₃C₂T_x-Arg composite. (G) N 1s spectrum of the Co-N@Ti₃C₂T_x-Arg composite. (H) Ti 2p spectrum of the Co-N@Ti₃C₂T_x-Arg composite.

The possibility to employ each of the obtained composites as an S cathode was then investigated [Figure 4]. Free-standing Co-N@Ti₃C₂T_x-Arg films were obtained by vacuum filtration and exhibited excellent flexibility [Supplementary Figure 9]. A representation of the application of Co-N@Ti₃C₂T_X-Arg/S electrodes in Li-S batteries is shown in Figure 4A, while the rate capabilities of Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys/S and Co-N@Ti₃C₂T_X-Arg/S electrodes are shown in Figure 4B. Specifically, the Co-N@Ti₃C₂T_X-Arg/S electrode exhibited the highest rate performance with initial discharge capacities of 1,365.0, 1,246.4, 1,305.3, 1,020.9, and 856.5 mAh g^-1 at current rates in the range of 0.1 to 2 C. Even at 0.2 C, a high specific capacity of 1,359.9 mAh g^-1 was maintained, thus indicating an outstanding electrochemical reversibility. The Co-N@Ti₃C₂T_X-Arg/S electrode demonstrated a better performance than the Mul-Ti₃C₂T_x/S, Co-N@Ti₃C₂T_X-Ser/S and Co-N@Ti₃C₂T_X-Lys/S electrodes at different rates, thus indicating the significance of the strong adsorption ability of LiPSs, which facilitated the efficient capture of the dissolved LiPSs for subsequent conversions and effectively suppressed the shuttle effect. The rate capability of Mul-Ti₃C₂T_x composite cathode in Supplementary Figure 10 shows lower discharge capacity than Co-N@Ti₃C₂T_x-Arg, indicating Co-N@Ti₃C₂T_x-Arg can effectively promote LiPSs conversion. Supplementary Figure 11 shows the discharge/charge curves of the Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys/S, and Co-N@Ti₃C₂T_X-Arg/S composite electrodes at 0.2 C, the Co-N@Ti₃C₂T_X-Arg/S composite electrodes display two stable voltage platforms. The long-term cycling performances of the batteries at different rates were then investigated. The cycling performance of Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys/S, and Co-N@Ti₃C₂T_X-Arg/S was tested at 0.2 C (1.0 C = 1,675 mAh g^-1, based on sulfur). The results [Figure 4C] revealed that the Co-N@Ti₃C₂T_X-Arg/S composite sulfur electrodes showed a high initial specific capacity of 1,314.3 mAh g^-1 at 0.2 C, Furthermore, a high specific (1,098 mAh g^-1) was maintained (corresponding to a capacity retention rate of 84.1%) and the corresponding decay rate per cycle was 0.083% after 200 cycles, while the Co-N@Ti₃C₂T_X-Ser/S and Co-N@Ti₃C₂T_X-Lys/S electrodes delivered low initial capacities of 539.5 and 200 mAh g^-1, respectively. And the areal capacity Co-N@Ti₃C₂T_x-Ser/S, Co-N@Ti₃C₂T_x-Lys/S, and Co-N@Ti₃C₂T_x-Arg/S are shown in Supplementary Figure 12. The CE of the Co-N@Ti₃C₂T_X-Arg/S electrodes [Figure 4D], was stabilized at a high value of 99.6%, whereas those of the Co-N@Ti₃C₂T_X-_Ser/S and Co-N@Ti₃C₂T_X-Lys/S electrodes fluctuated. The reason is that there are more active sites on Ti₃C₂T_x surface, which can effectively inhibit the dissolution of LiPSs into the electrolyte and reduce the shuttle phenomenon, this improves the utilization of S8. To further compare the electrochemical performance of Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys/S, and Co-N@Ti₃C₂T_X-Arg/S electrodes, the EIS experiments were conducted in the frequency range of 100 kHz to 10 mHz [Figure 4E], all of them show the high-frequency region and the low-frequency region, where R_Ω is the electrolyte impedance and Rct is the charge transfer resistance. The charge transfer resistance was 15.7, 19.8 and 36.3 Ω for Co-N@Ti₃C₂T_X-_Ser/S, Co-N@Ti₃C₂T_X-Lys/S, and Co-N@Ti₃C₂T_x-Arg/S composite cathode, respectively. The results revealed that Co-N@Ti₃C₂T_x-Arg/S composite cathode has highest charge transfer resistance, which due to that long chain amino acids increased layer spacing and reduced interlayer connectivity. However, discharge capacity is affected by many factors, including the number of active sites, lamellar morphology, etc, large layer spacing exposes more catalytic sites, effectively catalyzing and inhibiting the shuttle effect of polysulfide. Therefore, Co-N@Ti₃C₂T_X-Arg/S electrode exhibits higher discharge capacity. These findings reflect the enhanced lithium ion transport and low resistance of electrolytes of the Co-N@Ti₃C₂T_X-Lys/S electrodes during an electrochemical operation. Figure 4F shows the decay rate of the Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys/S, and Co-N@Ti₃C₂T_X-Arg/S electrodes, where the Co-N@Ti₃C₂T_X-Arg/S electrode exhibited an excellent performance with a high-capacity retention of 83.4%, However, the other two composites exhibited retentions of 54.4% and 91.0% (the initial capacity is only 200 mAh g^-1). Compare previous articles of MXene-based as host materials, the performance of Co-N@Ti₃C₂T_x-Arg/S electrodes have significant improvement [Supplementary Table 1]. The excellent electrochemical performance obtained with the Co-N@Ti₃C₂T_X-Arg/S composite sulfur electrodes is mainly attributed to the synergistic effect of the strong adsorption and excellent catalytic activity achieved by the increased active site for the sulfur electrodes.

Figure 4. (A) Schematic illustration of the fabrication of the Co-N@Ti₃C₂T_X-Arg/S electrodes for Li-S batteries. (B) Rate performances of the Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys /S, and Co-N@Ti₃C₂T_X-Arg/S electrodes. (C) Long-term cycling of the Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys/S, and Co-N@Ti₃C₂T_X-Arg/S electrodes. (D) CEs of the Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys /S and Co-N@Ti₃C₂T_X-Arg/S electrodes at different rates. (E) Nyquist plots of the Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys /S and Co-N@Ti₃C₂T_X-Arg/S electrodes, and the inset shows the equivalent circuit and fitted results. (F) Decay rates of the Co-N@Ti₃C₂T_X-Ser/S, Co-N@Ti₃C₂T_X-Lys/S and Co-N@Ti₃C₂T_X-Arg/S electrodes.

CONCLUSIONS

This study reported the synthesis of Co, N doped Ti₃C₂T_x MXene via thermal annealing of arginine-modified and CoCl₂-deposited Ti₃C₂T_x. The long molecular chain of arginine effectively increased the layer spacing of the Ti₃C₂T_x nanosheets. Once applied to Li-S batteries, the Co-N@Ti₃C₂T_x-Arg composite sulfur electrodes exhibited an electrochemical performance which was higher than those of Co-N@Ti₃C₂T_X-Ser and Co-N@Ti₃C₂T_X-Lys composite sulfur electrodes since the exposed active site effectively promoted the chemisorption of LiPSs. The Co-N@Ti₃C₂T_x-Arg composite sulfur electrode exhibited an initial discharge specific capacity of 1,314.3 mAh g^-1 and a reversible specific capacity that was maintained at 1,096 mAh g^-1 at 0.2 C even after 200 cycles. The results of this study revealed that combining amino acids with the highly conductive Ti₃C₂T_x can enable the development of promising small-molecule-modified ultrathin 2D materials which can be employed in high-performance Li-S batteries.