Chinese Chemical Letters  2020, Vol. 31 Issue (4): 1000-1003   PDF    
Nanostructured Ni/Ti3C2Tx MXene hybrid as cathode for lithium-oxygen battery
Caiying Wena, Tianjiao Zhua, Xingyu Lia, Huifeng Lia,*, Xianqiang Huangb, Genban Suna,*     
a Beijing Key Laboratory of Energy Conversion and Storage Materials, College of Chemistry, Beijing Normal University, Beijing 100875, China;
b Shandong Provincial Key Laboratory of Chemical Energy Storage and Novel Cell Technology, School of Chemistry & Chemical Engineering, Liaocheng University, Liaocheng 252059, China
Abstract: Ti3C2 belongs to MXenes family, which is a new two-dimensional material and has been applied in many fields. With simple method of hydrothermal and high temperature calcination, nanostructured Ni/Ti3C2Tx hybrid was synthesized. The stable layer structure of Ti3C2 MXene providing high surface area as well as excellent electronic conductivity are beneficial for deposition and decomposition of discharge product Li2O2. Furthermore, possessing special catalytic activity, Ni nanoparticles with size of about 20 nm could accelerate Li2O2 breaking down. Taking advantage of two kinds of materials, Ni/Ti3C2Tx hybrid as cathode of Li-O2 battery can achieve a maximal specific capacity of 20, 264 mAh/g in 100 mA/g and 10, 699 mAh/g in 500 mA/g at the first cycle. This work confirms that the prepared Ni/Ti3C2Tx hybrid exhibiting better cycling stability points out a new guideline to improve the electrochemical performance of lithium-oxygen batteries.
Keywords: MXene    Nickel    Two-dimensional material    Electronic conductivity    Lithium-oxygen battery    

Among energy storage system, lithium-oxygen battery is recognized as a possible countermeasure for solving energy shortage because of its high theoretical energy density up to 11, 400 Wh/kg, while the poor cyclic property and low kinetic reaction rate of oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) limit its commercial use [1-4].

Many sorts of materials have been utilized as cathodes of Li-O2 battery to improve battery performance such as commodity carbon black, three dimensional structure graphene and nano-porous noncarbon material [5-8]. As a new 2D material, MXene possesses unique electronic and structure properties with electrochemical stability, good electrical conductivity as well as controllable interlayer spacing for diverse intercalants situated at MXene layers, resulting in high surface area [9-11] and it has been extensively used in different energy storage devices such as lithium-ion battery, lithium-sulfur battery and supercapacitor [12-15].

With the popular price, transition metals have been proved to be suitable catalysts for Li-O2 battery, which can reduce charge overpotential, catalyze the decomposition of discharge product and improve cycle performance of lithium-oxygen battery [11, 16-20]. Liu et al. designed 3D ultralight Ni foam with Ru nanoparticles as active electrocatalysts [21] and Oh et al. developed novel mNi-NCNT-MoC-C-embedded microsphere as efficient catalyst [22], both exhibiting excellent Li-O2 battery property. Herein, we report the synthesized Ni/Ti3C2Tx hybrid combining the advantages of MXene and Ni, which makes Ni/Ti3C2Tx an excellent cathode material for Li-O2 battery.

A simple synthetic method was applied to get the target Ni/ Ti3C2Tx hybrid. As illustrated in Fig. 1a, the Al atoms layers of Ti3AlC2 were etched by HF, producing accordion-like Ti3C2Tx. Afterwards, Ti3C2Tx was mixed with Ni(NO3)2·6H2O and urea, which were heated at 140 ℃. Ultimately, the obtained Ni/Ti3C2Tx precursor was kept at 400 ℃ in protected gas Ar. The characterizations of scanning electronic microscope (SEM) and highresolution transmission electronic microscope (HRTEM) were employed to analyze the microstructure and morphology of Ni/Ti3C2Tx hybrid. As can be seen in Figs. 1b and c, compared with raw material, the interlayered spacing of Ti3C2Tx is enlarged after HF-etching. The expanded layer structure of MXene, which provides space for nickel nanoparticles to uniformly load on, has been kept after the treatment with hydrothermal and calcination, exhibited in Fig. 1d. Furthermore, Fig. 1e also draws the same conclusion that Ni nanoparticles were evenly dispersed on the surface of Ti3C2Tx. HRTEM image of Ni/Ti3C2Tx (Fig. 1f) shows that the size of nickel nanoparticles is approximately 20 nm, with lattice fringe constants of 0.2037 nm classified as (111) plane for Ni (JCPDS No. 04-0850). The well fabricated Ni/Ti3C2Tx was proved to be good crystallinity by the selected area electron diffraction (SAED) pattern of Ni/Ti3C2Tx (Fig. 1g) which demonstrates the single crystal electron diffraction of Ti3C2Tx and diffraction circles of Ni consistent with (111) plane and (200) plane of the XRD pattern of Ni. The above characterizations confirm the successful preparation of the Ni/Ti3C2Tx hybrid with Ni obtained by pristine Ni2+ material experiencing reduction reaction of Ti3C2Tx. The nanostructured hybrid has special catalytic activity, which contributes to the charge process of lithium-oxygen battery. The energy-dispersive X-ray spectrometer (EDS) is conducted to further elaborate the distribution of Ni on Ti3C2Tx. As shown in Fig. S1b (Supporting information), nickel nanoparticles uniformly grew on Ti3C2Tx. The mapping image of F element shows the distribution of similar concentration as Ni, which exhibits the coexistence of fluorine and nickel atoms, implying that some nickel atoms are bonding with fluorine atoms. The concentrations of F and O are appreciable in the entire MXene sheet due to the terminated functional groups, such as —O, —F and —OH. The EDS mapping certifies that the Ni nanoparticles were well-grown on Ti3C2Tx.

Download:
Fig. 1. (a) The schematic diagram of the synthesized method of Ni/Ti3C2Tx hybrid, SEM images of (b) Ti3AlC2, (c) HF-etching Ti3C2Tx, (d) Ni/Ti3C2Tx hybrid, (e) TEM images of Ni/Ti3C2Tx hybrid, (f) HRTEM image of Ni/Ti3C2Tx. (g) SAED pattern of Ni/Ti3C2Tx.

The X-ray diffraction (XRD) is an important characterization to clarify the effect of HF etching and the composition of final product, which is shown in Fig. 2. The peak at 39.0° which represents Al-layered in Ti3AlC2 disappeared after HF-etching, proving the removal of Al layer and formation of Ti3C2Tx. Ti3C2Tx has a lot of functional groups which are beneficial for the synthesis of Ni/Ti3C2Tx. The peaks at 44.5° and 51.8° correspond to (111) and (200) planes of Ni (JCPDS No. 04-0850), indicating that Ni is in the sample. The peak at 41.8° remains in the sample, proving that the structure of Ti3C2Tx was not destroyed after high-temperature calcination. No other peaks showed in the XRD pattern, demonstrating that Ni/Ti3C2Tx is a pure sample.

Download:
Fig. 2. XRD patterns of Ti3AlC2, HF-etching Ti3C2Tx and final product Ni/Ti3C2Tx.

To get closer understanding of the chemical state of the nanostructured Ni/Ti3C2Tx hybrid, the X-ray photoelectron spectra (XPS) was conducted. As shown in Fig. 3a, the synthesized product was made up of Ni, F, O, Ti and C element. In Fig. 3b, the Ti 2p core level was fitted with three doublets (Ti 2p3/2-Ti 2p1/2) with fixed area ratio equal to 2:1. The peaks locating at 455.52 eV, 456.64 eV, 459 eV of Ti 2p3/2 indicated the presence of Ti-C, Ti2+, Ti-O, separately [23]. Compared to the reported literature [23], these three peaks shifting to high binding energy were possibly attributable to the fact that Ti atoms lose partial electrons when Ti3C2Tx reduces the intermediate product of Ni(OH)2. As shown in Fig. 3c, the Ni 2p spectra of Ni/Ti3C2Tx was well fitted with two spin-orbit doublets related to Ni, Ni-F, and shake-up satellites (denoted as Sat.) [24]. The binding energy of Ni are 852.8 eV in Ni 2p3/2 spin-orbit level, and 870.34 eV in Ni 2p1/2 spin-orbit level, respectively. The peak at 856.8 eV corresponding to Ni-F in Ni 2p3/2 is 0.6 eV lower than the reported data [25], which was ascribed to that the electron of Ni atom was partially transferred to F atom, generating positively charged nickel center [26]. The obtained positive valence of nickel is lower than Ni2+, little different from NiF of NiF2. This result is consistent with the spectra of F 1s (Fig. 3d) whose peak at 684.6 eV of Ni-F is 0.1 eV higher than the reported article [27]. The reason why Ni atom was bonding with F atom rather than O atom is that the bond formation enthalpy of Ni-F is 430 kJ/mol, higher than that of Ni—O (382 kJ/mol), and the dissociation energy of chemical bond of Ni—F (439.7 kJ/mol) is greater than Ni-O (366 kJ/mol), revealing that Ni-F is more stable than Ni-O [28]. In Fig. 3e, the spectra of C 1s could be well deconvoluted into six peaks situated at 282, 282.3, 284.8, 286.3, 288.1 and 292.73 eV which belong to C—Ti—Tx, Ti—C—O, C—C, C—O, O=C—O and C—F bonds, respectively [23, 29]. In Fig. 3f, there exist four deconvolution peaks in the spectra of O 1s, demonstrating the four components of Al2O3, H2O and the bond of C-Ti-(OH)x, C-Ti-Ox [30].

Download:
Fig. 3. XPS survey spectrum of (a) Ni/Ti3C2Tx hybrid, high-resolution XPS spectra of (b) Ti 2p, (c) Ni 2p, (d) F 1s (e) C 1s, (f) O 1s of Ni/Ti3C2Tx hybrid.

The test of N2 adsorption and desorption was done to grasp the specific surface area of Ni/Ti3C2Tx, Ti3C2Tx and Ti3AlC2, presented in Figs. S2a-f (Supporting information). The N2 adsorption-desorption isotherms of these three kinds of materials were in accordance with IV curves of IUPAC. The specific surface area of Ni/Ti3C2Tx (51.282 m2/g) was larger than that of Ti3C2Tx (14.59 m2/g) and Ti3AlC2 (4.726 m2/g), as a result of the growth of Ni nanoparticles enlarging the interlayered spacing of Ti3C2Tx, which is in line with XRD patterns. The density functional theory (DFT) pore size distribution of them (Figs. S2b, d and f) mainly appeared in 2.5~6 nm, manifesting the existence of mesopores. The pore volume of Ni/Ti3C2Tx, Ti3C2Tx and Ti3AlC2 are 0.063 cm3/g, 0.027 cm3/g and 0.007 cm3/g, respectively, implying that Ni/Ti3C2Tx is more suitable for the accommodation of discharge product.

MXene is a new 2D material with excellent electronic conductivity, structure stability, excellent oxidation resistance and potential high surface area which is conducive to oxygen adsorption and discharge product of lithium-oxygen battery to stick. Ni has been proved to lower the charge overpotential of Li-O2 battery. Therefore, Ni/Ti3C2Tx hybrid is predicted to be an excellent Li-O2 battery cathode material. With the aim of exploring the electrochemical working mechanism of Ni/Ti3C2Tx as cathode of Li-O2 battery, the related electrochemical tests were carried out. From the Fig. 4a, we can see that even at high current density of 500 mA/g, Ni/Ti3C2Tx electrode delivers high discharge capacity of 10, 699 mAh/g at first cycle, due to the fact that the high surface area of Ni/Ti3C2Tx provides more space to store discharge product. In the second and third round, the specific capacity still can achieve 8635 mAh/g and 3421 mAh/g. The decrease of specific capacity at different circles can be explained by the difficulty of the discharge product decomposing at high current density, so that a fresh cathode cannot be provided. Compared with Ni/Ti3C2Tx, pure KB has worse specific capacity performance whose discharge specific capacity at first cycle is only 7719 mAh/g, shown in Fig. S3a (Supporting information). As can be seen in Fig. 4b, Ni/Ti3C2Tx cathode can stably cycle for 38 rounds at 100 mA/g with the high limited specific capacity of 500 mAh/g. Improving the current density to 500 mA/g, the battery can cycle for 26 rounds. Even if under the presence of catalysis, discharge product can not fully decompose in the charge process, so that Li2O2, Li2CO3 and LiOH would continuously accumulate on the cathode, which resulted in the cut-off discharge voltage of battery decreasing from 2.75 V to 2.3 V along with the decrease of cycle rounds.

Download:
Fig. 4. Electrochemical performance of KB, Ti3C2Tx and Ni/Ti3C2Tx cathode for lithium-oxygen battery. (a) Discharge and charge curves of full specific capacity of Ni/Ti3C2Tx at 500 mA/g. (b) Discharge termination voltage and cycle rounds of Ni/Ti3C2Tx at 100 mA/g and 500 mA/g. (c) The comparison of KB, Ti3C2Tx and Ni/Ti3C2Tx on the first cycle of discharge specific capacity. (d) The comparison of KB, Ti3C2Tx and Ni/Ti3C2Tx on discharge termination voltage and cycle rounds at 100 mA/g. (e) Nyquist plots of KB, Ti3C2Tx and Ni/Ti3C2Tx. (f) CV curves of KB, Ti3C2Tx and Ni/Ti3C2Tx with sweep rate of 5 mV/s. SEM images of (g) first discharge product and (h) first charge product.

Fig. 4c presents the contrast of the discharge capacity of KB, Ti3C2Tx and Ni/Ti3C2Tx in various current density at initial round. At 100 mA/g, the discharge specific capacity of Ni/Ti3C2Tx (20, 264 mAh/g) is much higher than that of KB (7719 mAh/g) and Ti3C2Tx (9333 mAh/g). Even at the high current density of 500 mA/g, Ni/Ti3C2Tx with nanosized Ni exerting the strong catalytic effect still remained the high discharge capacity of 10, 699 mAh/g. Only 2767 mAh/g and 3900 mAh/g can be reached by pure KB and Ti3C2Tx at 500 mA/g, respectively. In addition, in terms of cycling stability, the cycle numbers of Ni/Ti3C2Tx cathode are more than twice that of pure KB and Ti3C2Tx, illustrated in Fig. 4d. The betterperformance of Ni/Ti3C2Tx hybrid is derived from the increasing attachment sites for the discharge product provided by Ti3C2Tx with enhanced electronic conductivity, and catalytic activity site stem from nanosized Ni, preventing the electrochemical reaction from terminating.

Fig. S4a (Supporting information) is the discharge and charge curves of first round at 100mA/g and 500mA/g of Ni/Ti3C2Tx. Benefiting from the excellent electronic conductivity of Ti3C2 MXene, the charge overpotential of Ni/Ti3C2Tx hybrid at 100mA/g is 710mV, lower than that of Ni/KB [31] and Ti3C2Tx (1.273V, Fig. S4b in Supporting information), attributed to the interaction of Ni and Ti3C2Tx. At 500mA/g, the charge overpotential increased to 1.14V, perhaps ascribed to the internal electrode resistance [32]. Conversely, the charge overpotiental of pure KB (Fig. S3b in Supporting information) is 1.02V, higher than that of Ni/Ti3C2Tx (710mV), because of Ni/Ti3C2Tx electrode combining one advantage of facilitating decomposition of discharge products Li2O2 and another advantage of the high stable structure of Ti3C2Tx. Moreover, the cycle stability of Ni/Ti3C2Tx cathode is proved by the discharge and charge curves in different rounds at 100mA/g and 500mA/g, exhibited in Figs. S4c and d (Supporting information). The discharge voltage plateau of 100mA/g stabilized at 2.73V at the first 20 rounds and decreased to 2.6V after long-time cycling because of the accumulation of undecomposed discharge product covering active sites of Ni/Ti3C2Tx cathode. However, Ni/ Ti3C2Tx cathode exhibited worse cycling performance at 500mA/g whose discharge voltage plateau stabilized at 2.7V, owing to the lower kinetic reaction rate and difficult decomposition of discharge product.

Electrochemical impedance spectroscopy (EIS) test aims to deeply investigate the electrochemical performance of Ni/Ti3C2Tx hybrid. The Nyquist plots of KB, Ti3C2Tx and Ni/Ti3C2Tx hybrid before cycling can be seen from Fig. 4e. The semicircle at high frequency is associated with the diffusion of Li+ through SEI film on the surface of active material, while the semicircle at medium frequency is related to the charge transfer process and the slope line at low frequency could be utilized to study Warburg diffusion impedance [33, 34]. There are two semicircles of KB and Ti3C2Tx obviouslyappearing in high and medium frequency area. However, the Ni/Ti3C2Tx visibly possesses one semicircle at medium and high frequency. The Nyquist plots could be specifically analyzed by means of simulating the equivalent circuit diagram, illustrated in Fig. 4e. And the fitted datas are detailedly enumerated in Table S2 (Supporting information). It can be seen from the data in Table S2 that the resistanceof the electrolyte(expressedin Rs) of KB islower than that of Ti3C2Tx and Ni/Ti3C2Tx hybrid. The interface SEI resistance (written as Rf) and the charge transfer impedance (represented as Rct) of Ti3C2Tx are 35.48 V and 33.4 V, respectively, less than 40.54 V and 69.22 V of KB, separately, suggesting that Ti3C2Txcan improve electronic conductivity of the cathode. What is more, the valueof thecombination of Rf (2.887 V)and Rct (42.19V) of Ni/Ti3C2Tx are the smallest, confirming that Ni/Ti3C2Tx makes contributions to enhancing catalytic activity and stability of the LiO2 battery cathode [34]. Comparison of the fitting data reveals that Ni/Ti3C2Tx could efficiently heighten electrochemical reaction kinetics of the cathode.

In order to find out electrochemical catalytic effect of Ni/Ti3C2Tx in lithium-oxygen battery, cyclic voltammetry (CV) tests are implemented in Fig. 4f. The onset reduction potential and peak current as well as the oxidation peak current of Ni/Ti3C2Tx are higher than those of pure KB and Ti3C2Tx, respectively, suggesting that Ni/Ti3C2Tx contributes to higher ORR activity and better OER catalytic activity. Two apparent OER peaks at lower potential and higher potential for Ni/Ti3C2Tx can be seen in Fig. 4f, which are associated with the deintercalation of outer part of Li2O2 and the oxidation of most Li2O2 [35, 36]. The morphology of discharge product Li2O2 is like thin slice [37], shownin Fig. 4g. Aftercharging, flaky Li2O2 broke down, leading to no thin sheet in Fig. 4h.

To sum up, Ni/Ti3C2Tx hybrid was synthesized through hydrothermal along with calcination. Layer structure of MXene with Ni distributing uniformly on the interlayer and surface of Ti3C2Tx, was not destroyed afterkeeping at high temperature in the protected gas Ar. The Ni/Ti3C2Tx hybrid serving as cathode catalyst for Li-O2 battery delivers high discharge specific capacity (20, 264mAh/g at 100mA/g and 10, 699mAh/g at 500mA/g) and stable cycle performance maintaining 38 rounds at 100mA/g. The high surface area and excellent electronic conductivity of MXene and outstanding catalytic activity of Ni make Ni/Ti3C2Tx a suitable cathode catalyst for lithium-oxygen battery. This work provides a promising strategy to design MXene-based nanomaterials applied in Li-O2 battery cathode.

Acknowledgment

This work was supported by the National Natural Science Foundations of China (Nos. 21871028, 21471020 and 21771024).

Appendix A. Supplementary data

Supplementary material related to this article can be found, inthe online version, at doi:https://doi.org/10.1016/j.cclet.2019.09.028.

References
[1]
G. Girishkumar, B. McCloskey, A.C. Luntz, S. Swanson, W. Wilcke, J. Phy. Chem. Lett. 1 (2010) 2193-2203. DOI:10.1021/jz1005384
[2]
L. Grande, E. Paillard, J. Hassoun, et al., Adv. Mater. 27 (2015) 784-800. DOI:10.1002/adma.201403064
[3]
J. Lu, L. Li, J.B. Park, et al., Chem. Rev. 114 (2014) 5611-5640. DOI:10.1021/cr400573b
[4]
L. Wang, Y. Zhang, Z. Liu, L. Guo, Z. Peng, Green Energy Environ. 2 (2017) 186-203. DOI:10.1016/j.gee.2017.06.004
[5]
C. Liu, R. Younesi, C.W. Tai, et al., J. Mater. Chem. A Mater. Energy Sustain. 4 (2016) 9767-9773. DOI:10.1039/C5TA10690G
[6]
M.M.O. Thotiyl, S.A. Freunberger, Z. Peng, P.G. Bruce, J. Am. Chem. Soc. 135 (2013) 494-500. DOI:10.1021/ja310258x
[7]
M.M.O. Thotiyl, S.A. Freunberger, Z. Peng, etal., Nat.Mater. 12 (2013) 1049-1055.
[8]
C. Wang, Z. Xie, Z. Zhou, Appl. Mater. 7 (2019) 040701. DOI:10.1063/1.5091444
[9]
N.K. Chaudhari, H. Jin, B. Kim, et al., J. Mater. Chem. A Mater. Energy Sustain. 5 (2017) 24564-24579. DOI:10.1039/C7TA09094C
[10]
M. Naguib, V.N. Mochalin, M.W. Barsoum, Y. Gogotsi, Adv.Mater. 26 (2014) 992-1005. DOI:10.1002/adma.201304138
[11]
V.M.H. Ng, H. Huang, K. Zhou, et al., J. Mater. Chem. A Mater. Energy Sustain. 5 (2017) 3039-3068. DOI:10.1039/C6TA06772G
[12]
Y.T. Liu, P. Zhang, N. Sun, et al., Mater 30 (2018) 1707334.
[13]
L. Yu, L. Hu, B. Anasori, et al., ACS Energy Lett. 3 (2018) 1597-1603. DOI:10.1021/acsenergylett.8b00718
[14]
Q. Zhao, Q. Zhu, J. Miao, P. Zhang, B. Xu, Nanoscale 11 (2019) 8442-8448. DOI:10.1039/C8NR09653H
[15]
X. Zhang, Z. Zhang, Z. Zhou, J. Energy Chem. 27 (2018) 73-85. DOI:10.1016/j.jechem.2017.08.004
[16]
D. Ji, S. Peng, D. Safanama, et al., Chem. Mater. 29 (2017) 1665-1675. DOI:10.1021/acs.chemmater.6b05056
[17]
M.W. Yuan, R. Wang, W.B. Fu, etal., ACS Appl. Mat. Interfaces 11 (2019) 11403-11413. DOI:10.1021/acsami.8b21808
[18]
M. Yuan, C. Nan, Y. Yang, et al., ACS Omega 2 (2017) 4269-4277. DOI:10.1021/acsomega.7b00497
[19]
M. Yuan, L. Lin, Y. Yang, et al., Nanotechnology 28 (2017) 185401. DOI:10.1088/1361-6528/aa66bc
[20]
M. Yuan, Y. Yang, C. Nan, et al., Appl. Surf. Sci. 444 (2018) 312-319. DOI:10.1016/j.apsusc.2018.02.267
[21]
Z. Liu, N. Feng, Z. Shen, et al., ChemSusChem 10 (2017) 2714-2719. DOI:10.1002/cssc.201700567
[22]
Y.J. Oh, J.H. Kim, S.K. Park, et al., Chem. Eng. J. 351 (2018) 886-896. DOI:10.1016/j.cej.2018.06.166
[23]
J. Yan, C.E. Ren, K. Maleski, et al., Adv. Funct. Mate 27 (2017) 1701264. DOI:10.1002/adfm.201701264
[24]
A.P. Grosvenor, M.C. Biesinger, R.S.C. Smart, N.S. McIntyre, Surf. Sci. 600 (2006) 1771-1779. DOI:10.1016/j.susc.2006.01.041
[25]
M. Jin, G. Zhang, F. Yu, et al., Phys. Chem. Chem. Phys. 15 (2013) 1601-1605. DOI:10.1039/C2CP43357E
[26]
H. Zhang, L. Yu, T. Chen, W. Zhou, X.W. Lou, Adv. Funct. Mater 28 (2018) 1807086. DOI:10.1002/adfm.201807086
[27]
H. Han, J. Woo, Y.R. Hong, Y.C. Chung, S. Mhin, ACS Appl. Energy Mater. 2 (2019) 3999-4007.
[28]
Y.M. Lee, K.M. Nam, E.H. Hwang, et al., J. Phys. Chem. C 118 (2014) 10631-10639.
[29]
A. Rozmyslowska-Wojciechowska, E. Karwowska, S. Pozniak, et al., RSC Adv. 9 (2019) 4092-4105. DOI:10.1039/C8RA07633B
[30]
J. Halim, K.M. Cook, M. Naguib, et al., Appl. Surf. Sci. 362 (2016) 406-417. DOI:10.1016/j.apsusc.2015.11.089
[31]
C. Wang, Y. Zhao, J. Liu, et al., Chem. Commun. 52 (2016) 11772-11774. DOI:10.1039/C6CC05349A
[32]
Z. Hou, J. Long, C. Shu, et al., J. Alloys. Compd. 798 (2019) 560-567. DOI:10.1016/j.jallcom.2019.05.131
[33]
X. Yang, L. Zhang, F. Zhang, Y. Huang, Y. Chen, ACS Nano 8 (2014) 5208-5215. DOI:10.1021/nn501284q
[34]
K. Zhang, Y. Xu, Y. Lu, et al., J. Mater. Chem. A Mater. Energy Sustain. 4 (2016) 6404-6410. DOI:10.1039/C6TA01118G
[35]
T. Zhu, X. Li, Y. Zhang, et al., J. Electroanal. Chem. 823 (2018) 73-79. DOI:10.1016/j.jelechem.2018.05.032
[36]
X. Li, T. Zhu, C. Wen, et al., Electrochim. Acta 317 (2019) 367-374. DOI:10.1016/j.electacta.2019.05.054
[37]
Z. Sun, M. Yuan, L. Lin, et al., ACS Appl. Energy Mater. 2 (2019) 4144-4150. DOI:10.1021/acsaem.9b00328