⇦Chinese Physics Letters, 2021, Vol. 38, No. 5, Article code 058201 Electrochemical Behavior of Vanadium Carbide in Neutral Aqueous Electrolytes Chaofan Chen (陈超凡)1, Di Pang (庞迪)1, Xiaotong Wang (王晓彤)1, Gang Chen (陈岗)1,2, Fei Du (杜菲)1,2*, and Yu Gao (高宇)1* Affiliations 1Key Laboratory of Physics and Technology for Advanced Batteries (Ministry of Education), College of Physics, Jilin University, Changchun 130012, China 2State Key Laboratory of Superhard Materials, College of Physics, Jilin University, Changchun 130012, China Received 10 January 2021; accepted 5 March 2021; published online 2 May 2021 Supported by the Science & Technology Department of Jilin Province (Grant Nos. 20180101199JC and 20180101204JC), and Jilin Province/Jilin University Co-construction Project-Funds for New Materials (SXGJSF2017-3).
*Corresponding authors. Email: yugao@jlu.edu.cn; dufei@jlu.edu.cn Citation Text: Chen C F, Pang D, Wang X T, Chen G, and Du F et al. 2021 Chin. Phys. Lett. 38 058201    Abstract The V$_{2}$C compound, belonging to the group of two-dimensional transition metal carbonitrides, or MXenes, has demonstrated a promising electrochemical performance in capacitor applications in acidic electrolytes; however, there is evidence to suggest that V$_{2}$C is unstable in an acidic environment. On the other hand, the performance of V$_{2}$C in neutral aqueous electrolytes is still moderate, and has not yet been systematically studied. The charge storage mechanism in a V$_{2}$C electrode, employed in neutral aqueous electrolytes, is investigated via cyclic voltammetry testing and in situ x-ray diffraction (XRD). Good specific capacitances are achieved, specifically 208 F/g in 0.5 M Li$_{2}$SO$_{4}$, 225 F/g in 1 M MgSO$_{4}$, 120 F/g in 1M Na$_{2}$SO$_{4}$, and 104 F/g in 0.5 M K$_{2}$SO$_{4}$. Using in situ XRD, we observe that, during the charge and discharge process, the $c$-lattice parameter shrinks or expands by up to 0.25 Å in MgSO$_{4}$, and 0.29 Å in Li$_{2}$SO$_{4}$ which demonstrates the intercalation/de-intercalation of cations into the $d$-V$_{2}$C layer. DOI:10.1088/0256-307X/38/5/058201 © 2021 Chinese Physics Society Article Text MXene, a family of two-dimensional transition metal carbides/nitrides, synthesized by selectively removing the A element from the precursor phase, M$_{n +1}$AX$_{n}$ (where M is an early transition metal, A stands for any element from the IIIA and IVA groups, X is C and/or N, and $n=1,\, 2$, or 3), has attracted much attention due to their versatile composition, high conductivity and hydrophilic surface.[1,2] There are currently more than thirty known MXene materials, including Ti$_{3}$C$_{2}$,[3] Ti$_{2}$C,[4] Nb$_{4}$C$_{3}$,[5] Nb$_{2}$C,[6] and V$_{2}$C.[7] The particular relevance of this kind of material is its potential application in the field of energy storage;[8–11] as such, the charge storage mechanisms of MXenes in different electrolytes have been studied.[12–16] The majority of studies have focused on Ti$_{3}$C$_{2}$, the most widely investigated member of the MXene group, and a possible mechanism for its intercalation/de-intercalation has been proposed.[3] V$_{2}$C, one of the lightest MXene materials, possessing a unique vanadium metal layer, tends to achieve high specific capacity due to its variable valence (V$^{2+}$, V$^{3+}$, V$^{4+}$, V$^{5+}$) during electrochemical processes. Vanadium carbide was first synthesized in 2013, when it was used as the negative electrode for a lithium ion battery; a capacity of 260 mAh/g was achieved at 1 C, and a high rate performance was reported.[17] Since then, many studies have been devoted to the energy storage applications of V$_{2}$C materials. Dall'Agnese et al.[7] investigated V$_{2}$C as a positive electrode in a sodium ion capacitor, and achieved a specific capacitance of 100 F/g at a scan rate of 0.2 mV/s. Bak et al. studied a V$_{2}$C anode material for Na-ion batteries, reporting on the redox reaction of vanadium as a charge-storage mechanism.[18] Wang et al. examined V$_{2}$C intercalated with cobalt ions, demonstrating its capacity to realize an ultra-long cycle life of more than 15000 cycles, and a high specific Lithium storage capacity of 686.7 mAh/g at 0.1 C, due to the enlarged interlayer space, and the formation of a V–O–Co bond.[19] In addition, CNT@V$_{2}$C was investigated by the same team, in the context of a Zn-ion supercapacitor with 190.2 F/g at 0.5 A/g.[20] The use of few-layer V$_{2}$C was also investigated in an aluminum ion battery, where the reversible intercalation/de-intercalation of aluminum cations was suggested as the mechanism for charge storage.[21] In our previous work, V$_{2}$C was investigated as an electrode material in an aqueous electrolyte (H$_{2}$SO$_{4}$, KOH, and MgSO$_{4}$) for supercapacitors, and impressive capacitance was demonstrated.[22] In this work, we investigate the electrochemical behavior of V$_{2}$C in different neutral aqueous electrolytes (Li$_{2}$SO$_{4}$, Na$_{2}$SO$_{4}$, K$_{2}$SO$_{4}$, and MgSO$_{4}$), and further explore its reaction mechanisms via in situ XRD, taking into account that neutral electrolytes are more environmentally friendly, cheaper and safer, that the $d$-V$_{2}$C material is unstable in acidic electrolytes, and that the dissolution of vanadium may occur during cycling.[22] Experiment. Firstly, 1 g V$_{2}$AlC powder (200 mesh, purchased from Jilin 11 technology Co.) was slowly added into 20 ml HF (49% HF, Aladdin) and stirred for five days at 25℃. The resulting dark green solution was dispersed in DI water, and centrifuged several times at 5000 rpm to obtain a final precipitation at pH $\sim $ 6. The precipitation was then dried at 60℃ in a vacuum oven to produce V$_{2}$C powder. Next, 0.3 g of V$_{2}$C, suspended in 18 mL of deionized H$_{2}$O, was mixed with 2 ml TMAOH (25% TMAOH in water, Aladdin), and stirred for 15 h at room temperature. Residual TMAOH was removed by means of repeated centrifugation at 5000 rpm. The stable delaminated V$_{2}$C colloidal solutions ($d$-V$_{2}$C) were collected after 30 min of sonication and 1 h of centrifugation at 3500 rpm, and a free-standing $d$-V$_{2}$C film was then acquired via vacuum-assisted filtration. Scanning electron microscopy (SEM, JSM-6700 F) was employed to observe the morphology of the resulting V$_{2}$C. X-ray diffraction patterns were collected using a Bruker D8 diffractometer, operated at 40 kV and 40 mA, with Cu $K_{\alpha}$ radiation ($\lambda = 1.5406$ Å). A flexible, binder-free $d$-V$_{2}$C film was used directly as the working electrode, with a counter-electrode comprising over-capacitive activated carbon, containing 5 wt% PTFE, and a reference electrode composed of Ag wire. Here, 0.5 M Li$_{2}$SO$_{4}$, 0.5 M Na$_{2}$SO$_{4}$, 0.5 M K$_{2}$SO$_{4}$, and 1 M MgSO$_{4}$ solutions were used as electrolytes, with cellulose membrane functioning as the separator. Three-electrode Swagelok cells were assembled and tested using a VMP3 potentiostat (Biologic, S.A.). Electrochemical impedance spectroscopy (EIS) measurements were carried out at open circuit potential, with an amplitude of 10 mV, in a frequency range between 100 kHz and 10 mHz. For the in situ XRD test, plastic modified three-electrode cells were assembled with same selection of electrodes as those used in the electrochemical tests above. Cyclic Voltammetry was performed at 0.5 mV/s to control the working electrode potential, using a Vertex-One device, provided by Ivium Technologies B.V. The in situ XRD patterns were recorded in the range of $2\theta =5$–10$^{\circ} $ with steps of 0.02$^{\circ}$. Results and Discussion. Figure 1(a) illustrates the XRD patterns of V$_{2}$AlC, before and after HF etching, and of the $d$-V$_{2}$C film obtained via vacuum filtration. Due to the uneven etching process, the XRD pattern of the etched particles showed two new peaks (at 9.2$^{\circ}$ and 10.9$^{\circ}$), corresponding to the (002) plane of V$_{2}$C MXene, while unreacted V$_{2}$AlC remained dominant. During the delamination process, we found a significant decrease in the (002) peak between V$_{2}$C and the $d$-V$_{2}$C film, corresponding to the $c$-lattice parameter from 16.2 Å and 19.2 Å to 24.3 Å, indicating the intercalation of TMA$^{+}$.[23] An SEM image of the V$_{2}$C sample is shown in Fig. 1(b). The sample exhibits the typical accordion-like structure of MXene, and the cross-section scanning electron microscope image of the $d$-V$_{2}$C film in Fig. 1(c) shows that the film is composed of re-stacked nanosheets.
cpl-38-5-058201-fig1.png
Fig. 1. (a) XRD patterns of $V_{2}$AlC, V$_{2}$C, and $d$-V$_{2}$C film after delamination. (b) SEM image of V$_{2}$C after etching. (c) SEM image of filtrated $d$-V$_{2}$C paper.
cpl-38-5-058201-fig2.png
Fig. 2. Cyclic voltammograms of $d$-V$_{2}$C at different scan rates in (a) 0.5 M Li$_{2}$SO$_{4}$, (b) 0.5 M Na$_{2}$SO$_{4}$, (c) 0.5 M K$_{2}$SO$_{4}$, and (d) 1 M MgSO$_{4}$.
The electrochemical behavior of the $d$-V$_{2}$C film was investigated in various neutral electrolytes: 0.5 M Li$_{2}$SO$_{4}$, 0.5 M Na$_{2}$SO$_{4}$, 0.5 M K$_{2}$SO$_{4}$, and 1 M MgSO$_{4}$. Figure 2 shows the cyclic voltammograms. The $C$–$V$ curves obtained in the 0.5 M Na$_{2}$SO$_{4}$ and 0.5 M K$_{2}$SO$_{4}$ electrolytes were nearly ideal—rectangular, and without any visible redox peaks, suggesting the typical EDLC charge storage mechanism.[24,25] However, since the double-layer capacitance is proportional to the specific surface area of electrode, much lower capacitance would be expected with the low SSA of the MXene material if EDLC were the only charge storage mechanism.[3] It is therefore more likely that Na$^{+}$ and K$^{+}$ could access some shallow sites in the $d$-V$_{2}$C film, thereby contributing to the total capacitance. In addition, the shape of the $C$–$V$ curves remained rectangular, and without obvious distortion, with increasing scan rate, which demonstrates an impressive rate performance. With respect to the 0.5 M Li$_{2}$SO$_{4}$ electrolyte, the cyclic voltammogram exhibited an overall rectangular shape, with an obvious peak at around $-0.2$ V versus. Ag, and a reduction peak which was broad and less pronounced. Upon cycling in 1 M MgSO$_{4}$ at 2 mV/s, a pair of broad reduction peaks were evident at around $-0.7$ V versus Ag, which implies the possible electrochemical intercalation of cations.[3]
cpl-38-5-058201-fig3.png
Fig. 3. Electrochemical performance of $d$-V$_{2}$C electrode. (a) Specific capacitance measured from corresponding $C$–$V$ patterns. (b) Galvanostatic charge and discharge profiles at 1 A/g for different electrolytes. (c) Cycle lives measured from galvanostatic charge to discharge at 1 A/g.
Figure 3(a) summarizes the specific capacitance obtained at different scan rates. A high specific capacitance of 225 F/g was obtained at 2 mV/s in 1 M MgSO$_{4}$; when the scan rate was increased to 100 mV/s, the capacitance decreased to a moderate value of 130 F/g. In 0.5 M Li$_{2}$SO$_{4}$, a better rate performance was demonstrated, with capacitance of 208 F/g and 160 F/g at 2 mV/s and 100 mV/s, respectively. The higher capacitance in Li$_{2}$SO$_{4}$ and MgSO$_{4}$ may be attributed to charge transfer between the intercalated cations and MXene, in addition to surface adsorption capacitance; charge redistribution occurs via orbital coupling between the surface termination and ions, leading to depletion of the electrostatic potential difference. Figure 3(b) shows the GCD curves of the $d$-V$_{2}$C electrode at a current density of 1 A/g. There was a plateau at around $-0.3$ V in 0.5 M Li$_{2}$SO$_{4}$, consistent with the $C$–$V$ test. In addition, linear GCD curves were obtained in 0.5 M Na$_{2}$SO$_{4}$ and 0.5 M K$_{2}$SO$_{4}$ without an obvious IR drop. The cycle life of the $d$-V$_{2}$C, tested by galvanostatic charge-discharge at a high current density of 1 A/g, is given in Fig. 3(c). The capacitance values obtained in 0.5 M Na$_{2}$SO$_{4}$ and 0.5 M K$_{2}$SO$_{4}$ were 96 F/g and 83 F/g, respectively, and remained stable at around 97% and 98% after 1000 cycles. The capacitance slightly decreased with the number of cycles in Li$_{2}$SO$_{4}$ and MgSO$_{4}$, and capacitance retention levels after 1000 cycles were 75% and 87.5%, respectively. To explore the electrochemical kinetic of different ions, the pseudocapacitive contribution of MXene in MgSO$_{4}$ and Li$_{2}$SO$_{4}$ is illustrated in Fig. S1 in the Supplemental Material; the capacitive contribution is increased, with a maximum value of 92.1% in Li$_{2}$SO$_{4}$, and 87.3% in MgSO$_{4}$ at 100 mV/s with an increase in the scan rate. Moreover, the capacitive contribution was found to be 69.7% at 10 mV/s in Li$_{2}$SO$_{4}$, higher than the 59.3% found in MgSO$_{4,}$ consistent with its reasonable rate performance. Electrochemical impedance spectroscopy (EIS) was also performed to investigate the internal resistance and ion-diffusion process. As shown in Fig. S2, there is no visible semicircle in any electrolyte, which corresponds to low charge transfer resistance ($R_{\rm ct}$). In 0.5 M K$_{2}$SO$_{4}$, the equivalent high-frequency resistance (ESR) determined by the intersection of the high-frequency arc and the real axis is lower than for the other electrolytes, indicating that the inherent resistance in the cell is lower. Figure S3 shows the XRD patterns of a $d$-V$_{2}$C electrode immersed in 0.5 M Li$_{2}$SO$_{4}$ and 1 M MgSO$_{4}$ electrolytes. When immersed in the electrolytes, shift of peaks occurred, which may be attributed to the spontaneous intercalation of cations.[26] The in situ x-ray diffraction was used to further verify the intercalation of Li$^{+}$ and Mg$^{2+}$ during cycling. Figures 4(a) and 4(b) are contour maps of the in situ XRD patterns collected in the range $2\theta = 5^{\circ}$–$10^{\circ}$ during cyclic voltammetry in Li$_{2}$SO$_{4}$ and MgSO$_{4}$. A clear shift of (002) was visible during cycling in both electrolytes, confirming the intercalation/de-intercalation phenomenon of Li$^{+}$ and Mg$^{2+}$ into $d$-V$_{2}$C. Figures 4(c) and 4(d) show the change in the $c$-lattice parameter, depending on the potential during the first three cycles. A slight shrinkage of the $c$ value was observed with increasing voltage, which could be attributable to the electrostatic attraction interaction between positively charged solvated cations and the negatively charged MXene surface. For example, the first cycle in Li$_{2}$SO$_{4}$ showed that the $c$-lattice parameter decreased continuously to 24.61 Å from 24.32 Å, from $-0.1$ to $-1.3$ V, versus Ag, then expanded continuously to 24.58 Å when the potential returned to $-0.1$ V. The $c$-lattice parameter therefore changed by as much as 0.29 Å during this single cycle; the irreversible change in the $c$ value after each cycle could be explained by incomplete extraction of Li$^{+}$, and shrinkage of the interlayer space may lead to the observed decrease in capacitance. The $c$-lattice parameter changed by up to 0.25 Å in MgSO$_{4}$ during a single cycle, and was reversible during cycling.
cpl-38-5-058201-fig4.png
Fig. 4. In situ XRD patterns of $d$-V$_{2}$C during electrochemical cycles in (a) 0.5 M Li$_{2}$SO$_{4}$, and (b) 1 M MgSO$_{4}$. Change of the $c$-lattice parameter with potential during the first 3 cycles in (c) 0.5 M Li$_{2}$SO$_{4}$, and (d) 1 M MgSO$_{4}$.
In general, the multi-scale computation method is a useful tool helpful to reveal the electrochemical storage principle of an energy-storage material,[27] MXene is no exception. Mu et al. combined DFT calculations and an in situ XRD technique to investigate the charge-storage mechanism of MXene in H$_{2}$SO$_{4}$, and confirmed that the $c$-lattice of Ti$_{3}$C$_{2}$ can change with H$^+$ intercalation, due to the steric effect, together with electrostatic force.[15] By combining first-principles calculations and the implicit solvation model and investigating the electronic states during charge process, Ando et al. proposed the interaction between partially intercalated dehydrated cations and MXene layers.[28] We can deduce that the charge-storage mechanism of V$_{2}$C may be closely related to the size of cations; cations with smaller ion radii, such as Mg$^{2+}$ and Li$^{+}$, could intercalate into the interlayer space, and interact with the surface V atom and termination groups, as illustrated in Fig. 5, while larger ions such as Na$^{+}$ and K$^{+}$ could only access a few shallow sites, thereby contributing to superior rate and cycling performance.
cpl-38-5-058201-fig5.png
Fig. 5. Schematic illustration of V$_{2}$C during the discharge process. (yellow, V; sky blue, C; red, O; white, H; interlayer species: cations with different ion radius).
In summary, we have investigated the charge storage mechanism in V$_{2}$C electrode in various neutral aqueous electrolyte via cyclic voltammetry and in situ XRD, and good electrochemical performance is achieved. Redox peaks are visible in the $C$–$V$ patterns in MgSO$_{4}$ and Li$_{2}$SO$_{4}$, and a slight shrinkage in the $c$ value of 0.25 Å in MgSO$_{4}$, and 0.29 Å in Li$_{2}$SO$_{4}$ is observed with increasing voltage by means of in situ XRD. This definitively demonstrates the presence of cation intercalation into the $d$-V$_{2}$C layer. The superior cycle life of MgSO$_{4}$ can also be explained by the reversible change in the $c$ value during cycling, while the lower capacitance retention of Li$_{2}$SO$_{4}$ is attributed to the shrinkage of interlayer space during cycling. In Na$_{2}$SO$_{4}$ and K$_{2}$SO$_{4}$, nearly rectangular $C$–$V$ patterns with great cycling performance are demonstrated, in spite of their modest capacitance. In conclusion, the practical application of V$_{2}$C materials in aqueous electrolytes can be realized in a neutral environment, and careful selection of electrolytes with different cations should be considered for different demands: larger cations are suitable for energy storage devices where a long cycle life is a major requirement, while smaller cations are able to fulfill a requirement for high capacitance. The authors particularly wish to express their appreciation to Professor Yohan Dall'Agnese for great advice and support to this paper. References Two-Dimensional Transition Metal Carbides25th Anniversary Article: MXenes: A New Family of Two-Dimensional MaterialsCation Intercalation and High Volumetric Capacitance of Two-Dimensional Titanium CarbideA Non-Aqueous Asymmetric Cell with a Ti 2 C-Based Two-Dimensional Negative ElectrodeLi-ion uptake and increase in interlayer spacing of Nb4C3 MXeneAmine-Assisted Delamination of Nb 2 C MXene for Li-Ion Energy Storage DevicesTwo-Dimensional Vanadium Carbide (MXene) as Positive Electrode for Sodium-Ion CapacitorsMXene: a promising transition metal carbide anode for lithium-ion batteriesConductive two-dimensional titanium carbide ‘clay’ with high volumetric capacitancePseudocapacitance of MXene nanosheets for high-power sodium-ion hybrid capacitorsUltra-high-rate pseudocapacitive energy storage in two-dimensional transition metal carbidesSolving the Capacitive Paradox of 2D MXene using Electrochemical Quartz-Crystal Admittance and In Situ Electronic Conductance MeasurementsProbing the Mechanism of High Capacitance in 2D Titanium Carbide Using In Situ X-Ray Absorption SpectroscopyElectrochemical and in-situ X-ray diffraction studies of Ti 3 C 2 T x MXene in ionic liquid electrolyteRevealing the Pseudo‐Intercalation Charge Storage Mechanism of MXenes in Acidic ElectrolyteInfluences from solvents on charge storage in titanium carbide MXenesNew Two-Dimensional Niobium and Vanadium Carbides as Promising Materials for Li-Ion BatteriesNa-Ion Intercalation and Charge Storage Mechanism in 2D Vanadium CarbideAtomic Cobalt Covalently Engineered Interlayers for Superior Lithium-Ion StorageDelaminating Vanadium Carbides for Zinc‐Ion Storage: Hydrate Precipitation and H + /Zn 2+ Co‐Action MechanismTwo-Dimensional Vanadium Carbide (MXene) as a High-Capacity Cathode Material for Rechargeable Aluminum BatteriesTwo-dimensional vanadium carbide (V2C) MXene as electrode for supercapacitors with aqueous electrolytesIntercalation and delamination of layered carbides and carbonitridesMaterials for electrochemical capacitorsA review of electrode materials for electrochemical supercapacitorsIon-Exchange and Cation Solvation Reactions in Ti 3 C 2 MXeneMulti-scale computation methods: Their applications in lithium-ion battery research and developmentCapacitive versus Pseudocapacitive Storage in MXene
[1] Naguib M et al 2012 ACS Nano 6 1322
[2] Naguib M et al 2014 Adv. Mater. 26 992
[3] Lukatskaya M R et al 2013 Science 341 1502
[4] Come J et al 2012 J. Electrochem. Soc. 159 A1368
[5] Zhao S S et al 2017 Energy Storage Mater. 8 42
[6] Mashtalir O et al 2015 Adv. Mater. 27 3501
[7] Dall'Agnese Y et al 2015 J. Phys. Chem. Lett. 6 2305
[8] Naguib M et al 2012 Electrochem. Commun. 16 61
[9] Ghidiu M et al 2014 Nature 516 78
[10] Wang X et al 2015 Nat. Commun. 6 6544
[11] Lukatskaya M R et al 2017 Nat. Energy 2 17105
[12] Levi M D et al 2015 Adv. Energy Mater. 5 1400815
[13] Lukatskaya M R et al 2015 Adv. Energy Mater. 5 1500589
[14] Lin Z F et al 2016 Electrochem. Commun. 72 50
[15] Mu X P et al 2019 Adv. Funct. Mater. 29 1902953
[16] Wang X H et al 2019 Nat. Energy 4 241
[17] Naguib M et al 2013 J. Am. Chem. Soc. 135 15966
[18] Bak S M et al 2017 Adv. Energy Mater. 7 1700959
[19] Wang C D et al 2018 Adv. Mater. 30 1802525
[20] Wang C D et al 2019 Small Methods 3 1900495
[21] VahidMohammadi A et al 2017 ACS Nano 11 11135
[22] Shan Q M et al 2018 Electrochem. Commun. 96 103
[23] Mashtalir O et al 2013 Nat. Commun. 4 1716
[24] Simon P and Gogotsi Y 2008 Nat. Mater. 7 845
[25] Wang G P et al 2012 Chem. Soc. Rev. 41 797
[26] Ghidiu M et al 2016 Chem. Mater. 28 3507
[27] Shi S Q et al 2016 Chin. Phys. B 25 018212
[28] Ando Y et al 2020 Adv. Funct. Mater. 30 2000820