1 Introduction

With the spread of portable electronic devices and the boom in the new energy vehicle industry, there is an urgent need to develop energy storage devices with high performance and low cost [1, 2]. Due to the high energy density and good cycle stability, lithium-ion batteries (LIBs) have accounted for most of the consumer battery market. However, the limited resources of lithium ore and its uneven global distribution limit its use in large-scale energy storage applications [3]. Considering the high natural abundance of potassium (2.09 wt%, while 0.0017 wt% for Li) and the similar energy storage mechanism to that of LIBs, potassium-ion batteries (PIBs) have attracted much attention as a promising alternative to LIBs [4,5,6]. However, the larger K+ radius (0.138 nm) leads to slow reaction kinetics and larger volume changes of active material during cycling [7]. Up to now, the development of PIBs is still in its infancy and electrode materials with high capacity and good cycling stability are yet to be developed [8, 9].

Graphite is the most extensively used anode material with excellent electrical conductivity, and large layer spacing (3.4 nm) [10]. Unfortunately, the practical application of graphite anode material remains challenging mainly due to its low capacity. In this regards, alloy-based materials are potential alternatives due to their high capacity. Among the alloy-based materials, Sn4P3, which possesses a high theoretical capacity of 612 mAh·g−1 and synergistic benefits of Sn and P, is deemed as the most promising anode for PIBs [11,12,13]. However, the low electronic conductivity (30.7 S·cm−1) and huge volume changes during charge/discharge process lead to sluggish reaction kinetics and poor cycling stability [12, 14,15,16]. To tackle these limitations, an effective strategy is to combine Sn4P3 with highly conductive skeleton [16,17,18]. For instance, Zhang et al. [12] prepared Sn4P3/C composite via ball-milling technique and studied it as anode for PIBs, which delivers a reversible capacity of 384.8 mAh·g−1 at 50 mA·g−1. However, the Sn4P3-based composites prepared by the ball-milling method are poor in terms of cycle stability [14, 19, 20]. The capacity retention of Sn4P3/C composite is only ~ 6.6% after 120 cycles. Therefore, further efforts are still needed to achieve high capacity, long-term cycle stability as well as high-rate performance.

Ti3C2Tx, as a new member of two-dimensional layered material family, has shown unique advantages in the field of energy storage due to its layered structure, superior electrical conductivity (6500 S·cm−1), and low diffusion barrier [21,22,23,24,25]. Intensive study has been performed on the use of Ti3C2Tx as electrode materials for energy storage batteries, such as hybrid Ti3C2/NiCoP for sodium ion batteries [26], Ti3C2/Si composite for LIBs [27], and MoS2/MXene hybrids for PIBs [28]. Yet, incorporating MXene with high capacity Sn4P3 anode has rarely been tried.

Herein, we prepared a novel sandwich-like structured Sn4P3/Ti3C2Tx composite, which has rarely been reported before, via a solvothermal reaction followed by phosphorization process. Electrochemical study demonstrates that Sn4P3/Ti3C2Tx owns superior cycle stability and rate capability, which benefits from the synergistic effect between Sn4P3 and Ti3C2Tx. In Sn4P3/Ti3C2Tx composite, Ti3C2Tx acts as a conductive matrix to facilitate electron transfer and buffer the volume change of Sn4P3 particles during charge/discharge, while Sn4P3 serves as pillars to prevent the collapse and stacking of Ti3C2Tx sheets.

2 Experimental

Except for Ti3C2Tx powder was purchased from Jilin 11 Technology Co., Ltd, all other materials used in this experiment were purchased from Shanghai Macklin Biochemical Co., Ltd.

2.1 Synthesis of Sn4P3/Ti3C2Tx

In a typical preparation, SnCl2·2H2O and Ti3C2Tx powder with different mass ratios of 1:0.5, 1:1 and 1:2 were added into ethanol solution, followed by an ultrasonic treatment process for 1 h. Afterward, 20 ml NaOH solution with concentration of 0.2 mol·L−1 was added into the above mixture and stirred for 0.5 h. The solution was subsequently transferred into Teflon-lined stainless-steel autoclave and reacted at 180 °C for 10 h. The obtained SnO/Ti3C2Tx powder was then washed with deionized water and collected by freeze-drying. To get Sn4P3/Ti3C2Tx, appropriate NaH2PO2·H2O and SnO/Ti3C2Tx powder were annealed in tube furnace in Ar atmosphere with temperature of 280 °C for 0.5 h. Electrochemical results in Fig. S1 show that the Sn4P3/Ti3C2Tx sample with an initial SnCl2·2H2O to Ti3C2Tx mass ratio of 1:1 possesses the optimal performance. Therefore, it was selected for further research. As a control, bare Sn4P3 was prepared in the same process without adding Ti3C2Tx powder.

2.2 Material characterizations

The crystal structure, composition and morphology of the samples were investigated by X-ray diffraction (XRD, Rigaku, Ultima IV) with Cu Kα radiation, X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha), scanning electron microscope (SEM, Zeiss Sigma 300), and transmission electron microscope (TEM, JEOL JEM 2100F), respectively.

2.3 Electrochemical characterizations

The electrode was fabricated by coating the mixed slurry of 80 wt% of the as-prepared Sn4P3/Ti3C2Tx or bare Sn4P3, 10 wt% carbon black, and 10 wt% polyvinylidene fluoride (PVDF) binder in N-methyl pyrrolidone (NMP) onto the copper foil. The electrodes were then vacuum-dried at 85 °C for 12 h and cut into disks with loading of ~ 0.8 mg·cm−2. Electrochemical performance was measured with CR2032-type coin cells, where K foil works as counter/reference electrode and KPF6 in ethylene carbonate/diethyl carbonate works as electrolyte. Galvanostatic charge–discharge (GCD) was tested on a Land CT2001A battery testing system. Cyclic voltammograms (CV) and electrochemical impedance spectroscopy (EIS) tests were carried out at CHI660e electrochemistry workstation. CV was conducted in a voltage range of 0.01–2.00 V. EIS was tested over a frequency range from 10 mHz to 100 kHz by applying an alternating current (AC) signal of 5 mV.

3 Results and discussion

Figure 1 illustrates the synthesis process of Sn4P3/Ti3C2Tx composite. Firstly, SnO nanoparticles are loaded on the surface of Ti3C2Tx by a hydrothermal reaction. Afterward, in situ phosphorization reaction is conducted in a tube furnace and the Sn4P3/Ti3C2Tx composite material is finally obtained.

Fig. 1
figure 1

Schematic illustration of synthesis process of Sn4P3/Ti3C2Tx composite

Full size image

The crystal structures of the materials were investigated by XRD. As shown in Fig. 2, the defined peaks at ~ 6.46° could be assigned to (002) planes of titanium carbide, which agrees well with previous reports [29,30,31]. In the XRD pattern of final product of Sn4P3/Ti3C2Tx, the diffraction peaks locating at ~ 27.7°, 28.8°, 30.3, 31.4°, 44.5°, 45.7° and 56.5° are indexed as (104), (015), (0012), (107), (0114), (110) and (027) planes of rhombohedral Sn4P3 (JCPDS No. 73-1820) respectively [15, 32], while the extra peaks can be attributed to Ti3C2Tx, indicating the successful coupling of Sn4P3 and Ti3C2Tx.

Fig. 2
figure 2

XRD patterns of Ti3C2Tx, SnO/Ti3C2Tx, and Sn4P3/Ti3C2Tx

Full size image

Morphology of pristine Ti3C2Tx and the product of Sn4P3/Ti3C2Tx was investigated by SEM. Figure 3a shows the SEM image of Ti3C2Tx, in which a typical accordion-like and well-aligned layered structure is clearly observed. The layered structure still maintains even after a long time of hydrothermal reaction and high-temperature phosphating process (Fig. 3b), suggesting the excellent structural stability of Ti3C2Tx. Moreover, Sn4P3 nanoparticles are evenly distributed on the surface of Ti3C2Tx sheets, thus forming a sandwich-like structure (Fig. S2). In this structure, Sn4P3 nanoparticles can act as pillars to prevent the collapse and stacking of Ti3C2Tx sheets, while the Ti3C2Tx sheets could provide a good conductive matrix and unblocked channels for electron transmission and K+ transfer, respectively, thereby achieving synergistic effect. TEM images were collected to reveal the structural and morphological details of Ti3C2Tx and Sn4P3/Ti3C2Tx composite. As shown in Fig. 3c, obvious multilayer structure is observed for matrix. Figure 3d shows high-resolution transmission electron microscope (HRTEM) image and fast Fourier transformation (FFT) pattern of Sn4P3/Ti3C2Tx, clear fringes with lattice spacing of 0.320 and 0.204 nm are observed, corresponding to (104) and (0114) planes of Sn4P3, respectively. HRTEM images in Fig. S3 reveal the expansion of the interlayer spacing from Ti3C2Tx to Sn4P3/Ti3C2Tx, which is mainly caused by the insertion of cations between the layers during synthesis process. Elemental mappings in Fig. 3e suggest that Sn4P3 particles are homogeneously distributed on the matrix of Ti3C2Tx.

Fig. 3
figure 3

SEM images of a Ti3C2Tx and b Sn4P3/Ti3C2Tx; c TEM image of Ti3C2Tx; d HRTEM images and (inset) FFT pattern of Sn4P3/Ti3C2Tx; e TEM image and corresponding elemental mappings of Sn4P3/Ti3C2Tx

Full size image

The chemical states of different elements in Sn4P3/Ti3C2Tx composite were analyzed by XPS. Figure 4a shows the full-scan spectrum of Sn4P3/Ti3C2Tx sample, demonstrating the existence of Ti, C, Sn, P, O and F. Figure 4b–f exhibits the high-resolution XPS spectra of Ti 2p, C 1s, Sn 3d, P 2p and O 1s, respectively. Ti 2p spectrum in Fig. 4b can be deconvoluted into four species of Ti2+, Ti3+, Ti–C and Ti–O, which is consistent with the previous literatures [33, 34]. Figure 4c displays four peaks at ~ 281.5, 284.7, 286.4 and 288.8 eV, which are assigned to the characteristic bonds of C–Ti, C–C, C–O and C=O, respectively. In terms of Sn 3d spectrum in Fig. 4d, the peak at ~ 487.1 eV is resulted from spin-orbital splitting photoelectrons of Sn, which is relevant to the Sn–P bond of Sn4P3 phase. While another peak situated at ~ 485.1 eV corresponds to Sn0 [35, 36]. As displayed in Fig. 4e, P 2p spectrum is divided into three peaks. Two peaks locating at ~ 129.6 and 130.3 eV are related to typical P 2p3/2 and P 2p1/3, corresponding to P–Sn bond of Sn4P3 phase. While the small peak at ~ 133.3 eV is attributed to P–O bond, indicating that the oxidized phosphate species remain on the surface of Sn4P3/Ti3C2Tx. For O 1s spectrum in Fig. 4f, the peaks at ~ 529.4, 531.3, 532.1 and 533.0 eV correspond to O–Ti, O=C, O–C and O–P, respectively. XPS results further illustrate that Sn4P3/Ti3C2Tx composite was synthesized successfully.

Fig. 4
figure 4

XPS spectra of Sn4P3/Ti3C2Tx: a full-scan survey, b Ti 2p, c C 1s, d Sn 3d, e P 2p and f O 1s

Full size image

To evaluate potassium storage properties, the Sn4P3 and Sn4P3/Ti3C2Tx were investigated by using CR2032-type coin cells, in which potassium foils were used as counter/reference electrodes. CV was firstly carried out at a scan rate of 0.1 mV·s−1 with the potential window between 0.01 and 2.00 V. As shown in Fig. S4, a broad peak between 0.01 and 1.00 V is observed in the initial cathodic scan, which is derived from the potassiation reaction to form K-P and K-Sn phases as well as the formation of solid electrolyte interphase (SEI). In the following anodic scan, a weak peak centered at 0.8 V can be observed, corresponding to the de-potassiation process. The second and third CV curves overlap with each other, indicating the good reversibility of Sn4P3/Ti3C2Tx during charge and discharge.

Figure 5a shows the selected charge/discharge profiles at current density of 50 mA·g−1 between 0.01 and 2.00 V (vs. K/K+). In the first cycle, Sn4P3/Ti3C2Tx electrode delivers a discharge and charge capacity of 586.6 and 340.9 mAh·g−1, respectively, corresponding to an initial Coulombic efficiency of ~ 58%. The irreversible loss of the initial capacity is mainly resulted from the formation of SEI films on the surface of the electrode material [37]. In the subsequent cycles, Sn4P3/Ti3C2Tx gives stable capacity of ~ 304.8 mAh·g−1 and the Coulombic efficiency reaches up to > 98%. Compared with Sn4P3/Ti3C2Tx composite electrode, the capacity of Sn4P3 electrode is only ~ 228 mAh·g−1 under the same current density.

Fig. 5
figure 5

a Selected galvanostatic charge/discharge profiles of Sn4P3/Ti3C2Tx at current density of 50 mA·g−1; b rate capability and c cycle performance of Sn4P3 and Sn4P3/Ti3C2Tx electrodes at 500 mA·g−1; d Nyquist plots of Sn4P3 and Sn4P3/Ti3C2Tx electrodes before and after 50 cycles; e long-term cycling performance of Sn4P3/Ti3C2Tx electrode at 1 A·g−1 (working voltage of 0.01–2.00 V)

Full size image

Rate capability of pure Sn4P3 and Sn4P3/Ti3C2Tx composite was measured at various current densities ranging from 0.05 to 2.00 A·g−1. As shown in Fig. 5b, Sn4P3/Ti3C2Tx composite delivers reversible capacities of 304.8, 270.7, 220.3, 167.4, 120.8 and 60.7 mAh·g −1 at current densities of 0.05, 0.10, 0.20, 0.50, 1.00 and 2.00 A·g−1, respectively, which is obviously superior to that of pure Sn4P3. In addition, most of the capacity is recovered (~ 292.5 mAh·g −1) when the current density returns to 0.05 A·g−1, demonstrating that the electrode is able to accommodate the massive current changes in application. It is worth mentioning that the rate capability of Sn4P3/Ti3C2Tx presented in our work surpasses Sn-based anodes reported previously (Fig. S5) [38,39,40], which is mainly due to that the accordion-like skeleton with good conductivity allows easy penetration of electrolyte and transfer of electrons.

Cycling performance of the bare Sn4P3 and Sn4P3/Ti3C2Tx was compared at 0.5 A·g−1 in Fig. 5c. Sn4P3/Ti3C2Tx electrode delivers a charge capacity of 173.1 mAh·g−1 for the first cycle, which is significantly higher than 64.5 mAh·g−1 obtained from Sn4P3 electrode. Moreover, the Sn4P3/Ti3C2Tx electrode retains a high capacity of 162.6 mAh·g−1 with capacity retention of ~ 94% after 50 cycles. By contrast, the capacity of bare Sn4P3 electrode decays quickly with cycling and only 48.2 mAh·g−1 is remained at 50th cycle. Nyquist plots were collected and exhibited in Fig. 5d. It can be clearly seen that the charge transfer resistance of Sn4P3/Ti3C2Tx decreases with cycling, due to the activation of electrode [14]. Whereas the Sn4P3 electrode shows larger impedance after 50 cycles, which is attributed to the pulverization and exfoliation of Sn4P3 particles (details will be presented by ex-situ SEM shown in Fig. 6). The smaller resistance well accounts for the excellent cycling stability and rate capability of Sn4P3/Ti3C2Tx composite. Long-term cycle performance of Sn4P3/Ti3C2Tx was further evaluated at a current density of 1.0 A·g−1. As displayed in Fig. 5e, Sn4P3/Ti3C2Tx electrode still delivers a specific capacity of 103.2 mAh·g−1 even after 300 cycles, which is ~ 81% of its initial reversible capacity.

Fig. 6
figure 6

SEM images of a initial and b 50 charge/discharge cycled Sn4P3 electrodes; c initial and d 50 charge/discharge cycled Sn4P3/Ti3C2Tx electrodes (insets being digital photos of each electrode); Schematic illustration of potassiation process and de-potassiation of e Sn4P3 and f Sn4P3/Ti3C2Tx

Full size image

To further explore the reason for the largely improved potassium storage performance of Sn4P3/Ti3C2Tx composite, morphology of the electrodes before and after cycling was studied by SEM. As shown in Fig. 6a, c, both Sn4P3 and Sn4P3/Ti3C2Tx electrodes present flat surface before cycling. While after 50 cycles, cracks are clearly observed in Sn4P3 electrode (Fig. 6b). In sharp contrast, the surface of Sn4P3/Ti3C2Tx electrode remains flat and intact, except for some fibers from glass fiber separator (Fig. 6d). These results reveal that the introduction of Ti3C2Tx could effectively alleviate the stress induced by volume change during charge/discharge, thus enabling the integrity of electrode and fast transfer of electrons, which contribute greatly to the electrochemical performance (Fig. 6e, f).

Reaction kinetics of Sn4P3/Ti3C2Tx electrode was analyzed by CV. As shown in Fig. 7a, the intensity of peaks increases with the scanning rate increasing from 0.2 to 1.0 mV·s−1. However, the CV curves always maintain a similar shape, which means that Sn4P3/Ti3C2Tx electrode owns good response ability under fast scanning rate. Diffusion and capacitive contribution are qualitatively determined by the relationship between current (i) and scan rate (v) according to the following equations:

$$i = a{ }v^{b}$$
(1)
$$\lg{i} = b{ }\lg{v} + \lg{a}$$
(2)

where \(a\) and \(b\) are adjustable constants, and the value of b can be confirmed by the slope of lg\(i\)-lg\(v\) curves. When the value of \(b\) is close to 0.5 or 1.0, the reaction process is diffusion control or capacitance control, respectively. As depicted in Fig. 7b, the calculated b values of the anodic and cathodic peaks are 0.86 and 0.75 respectively, which implies that the electrochemical reaction process is controlled by both diffusion and capacitance behavior. The ratios of diffusion and capacitive contribution are obtained based on the following equation:

$${i _{\left( V \right)}} = {k_{1} v} + {k_{2} v}^{1/2}$$
(3)

where \({k}_{1}v\) and \({k}_{2}{v}^{1/2}\) correspond to the contributions of the capacitive effect and diffusion-controlled process, respectively. As illustrated in Figs. 7c, d and S6, the ratio of capacitive contribution gradually increases with the increase of scanning rate. This demonstrates that the electrochemical reaction process of the Sn4P3/Ti3C2Tx electrode is mainly controlled by the capacitive behavior at high rate, which is beneficial for rate performance.

Fig. 7
figure 7

a CV curves of Sn4P3/Ti3C2Tx electrode at various scan rates; b lgi-lgv plots obtained from corresponding CV curves; c capacitive charge storage contribution at 1 mV·s−1; d separation of contributions from capacitive- and diffusion-controlled capacities at various scan rates

Full size image

4 Conclusion

In summary, sandwich-like structured Sn4P3/Ti3C2Tx composite was synthesized via solvothermal reaction along with low-temperature phosphating process. The electrochemical properties of Sn4P3/Ti3C2Tx composite is evaluated as anode for PIBs. The introduction of the highly conductive Ti3C2Tx matrix not only provides channels for fast electron transfer, but also alleviates the volume change of Sn4P3 upon charge/discharge. Moreover, the loading of Sn4P3 nanoparticles serves as pillars to prevent Ti3C2Tx sheets from collapse or stack. Owing to the synergistic effect between the two components, Sn4P3/Ti3C2Tx exhibits significantly improved electrochemical performance than Sn4P3, which can be ranked as a high-performance anode material for PIBs.