Cycling Stability of a VOx Nanotube Cathode in Mixture of Ethyl

Sep 7, 2016 - From a cost and safety perspective, rechargeable Mg-ion batteries (MIBs) have become promising in recent years as an alternative to conv...
0 downloads 6 Views 5MB Size
Research Article www.acsami.org

Cycling Stability of a VOx Nanotube Cathode in Mixture of Ethyl Acetate and Tetramethylsilane-Based Electrolytes for Rechargeable Mg-Ion Batteries Ju-Sik Kim, Ryoung-Hee Kim, Dong-Jin Yun, Seok-Soo Lee, Seok-Gwang Doo, Dong Young Kim,* and Hyunjin Kim* Samsung Advanced Institute of Technology (SAIT), 130 Samsung-ro, Yeongtong-gu, Suwon-si, Gyeonggi-do 443-803, Korea S Supporting Information *

ABSTRACT: The electrochemical cycling performance of vanadium oxide nanotubes (VOx-NTs) for Mg-ion insertion/ extraction was investigated in acetonitrile (AN) and tetramethylsilane (TMS)-ethyl acetate (EA) electrolytes with Mg(ClO4)2 salt. When cycled in TMS-EA solution, the VOxNT exhibited a higher capacity retention than when cycled in AN solution. The significant degradation of capacity in AN solution resulted from increased charge-transfer resistance caused by the reaction products of the electrolyte during cycling. Mixed TMS-EA solvent systems can increase the cell performance and stability of Mg-electrolytes owing to the higher stability of TMS toward oxidation and the strong Mgcoordination ability of EA. These results indicate that the interfacial stability of the electrolyte during the charging process plays a crucial role in determining the capacity retention of VOx-NT for Mg insertion/extraction. KEYWORDS: vanadium oxide, nanotube, magnesium-ion battery, tetramethylsilane, mixture electrolytes, capacity retention



INTRODUCTION From a cost and safety perspective, rechargeable Mg-ion batteries (MIBs) have become promising in recent years as an alternative to conventional Li-ion batteries (LIBs) for largescale applications, including energy storage systems in smart grids and power sources for electric vehicles.1−3 One of the main limitations of MIB technology is the low performance of the cathode because of the low ionic mobility of divalent Mg ions in the cathode-active materials.4−6 To overcome this limitation, various oxides, selenides, and sulfides have been developed to alter the composition and microstructure of host materials for Mg insertion/extraction.7−12 Among candidate cathode materials, vanadium oxide nanotubes (VOx-NTs) have recently been studied as a promising active material for the cathode owing to its novel structure for the insertion/extraction of Mg ions.13−17 VOx-NTs exhibit high surface area and a layered structure with a large interlayer distance, which allows Mg ions to reversibly move into its interstitial and defect sites. In previous work,17 we investigated VOx-NTs synthesized by the hydrothermal method in the presence of an amine template. As an MIB cathode material, the amine-incorporated VOx-NTs showed an excellent initial capacity of more than 200 mA h g−1 in deaerated acetonitrile (AN) solution containing 0.5 M Mg(ClO4)2, depending on the oxidation state of the vanadium ions and the bonding structure on the surface of VOx.17 However, in spite of its high initial performance, the VOx-NT is expected to exhibit poor capacity retention after long © XXXX American Chemical Society

discharging/charging cycles because of the decomposition of AN at high voltage, even when a large amount of amine was incorporated into VOx (HT-VOx nanotube).17,21−24 On the basis of recent studies that showed that the layer spacing and shape of VOx-NT does not significantly change and that AN could be decomposed at the electrode surface during charging,18−24 it seems reasonable to mainly attribute the capacity fading to interfacial reactions with the electrolyte during charging, rather than to structural changes. As the formation of a passive layer on the cathode surface is generally accompanied by oxidation of the electrolyte, the stability of the electrolyte toward oxidation is expected to play an essential role in determining the capacity retention of VOx-NTs.23,25 In the present work, we propose Mg(ClO4)2-based electrolytes in tetramethylsilane (TMS) mixed solvents. We first investigated the cycling performance of the VOx-NT electrode in TMS-based electrolytes, as TMS has better oxidative stability than AN. Then, ethyl acetate (EA), which strongly interacts with Mg ions, was added to the electrolyte to produce more ionic species. The capacity losses in AN, TMS, and TMS-EA electrolytes obtained from combined experimental and theoretical studies are compared and discussed in terms of interface stability during electrochemical cycles. Received: May 16, 2016 Accepted: September 7, 2016

A

DOI: 10.1021/acsami.6b05808 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 1. (a) High-resolution and (inset) low-resolution SEM images of VOx-NTs. (b) High-resolution and (inset) low-resolution TEM images of VOx-NTs.



RESULTS AND DISCUSSION SEM and transmission electron microscopy (TEM) images of VOx-NTs are shown in Figure 1. Figure 1a shows high and low magnification SEM images of the hydrothermally synthesized VOx-NTs, whereas Figure 1b shows high and low magnification TEM images. The VOx-NTs are observed to be open-ended, multilayered tubular structures, comprising alternating arrangements of VOx(dark and narrow part) and amine (bright and wide part) layers. The detail molecular concentrations of elements in VOx NTs are shown in Table S1. Figures 2a−c show the charge−discharge curves for the VOxNT electrodes in electrolytes with different compositions: AN, TMS, and TMS with 10 vol % EA, respectively. In the TMS electrolyte, a plateau was not observed in the charging curves and the performance of the electrode was significantly poorer than that achieved in the AN electrolyte. Higher viscosities were recorded with TMS (∼44.8 cP) and TMS-EA (23.6 cP) than with AN (0.75 cP). Thus, it is deduced that the high viscosity of TMS causes sluggish Mg ion transport, resulting in a larger overpotential and lower capacity.22 As seen in Figure 2c, the initial discharge capacity was improved to124 mAhg−1, when EA, which has a lower viscosity, was added to TMS. This result indicates that the Mg insertion/extraction performance of the electrode is highly dependent on the viscosity of the electrolyte. Moreover, owing to the relative low binding energy of Mg ions and TMS, fewer ionic species would be produced in the TMS. It should be pointed out that the capacity when using the AN electrolyte drastically decreased after 30 cycles, whereas in the electrolytes containing TMS, the performance was only slightly degraded after 30 cycles. An additional demonstration of the difference in capacity retention with the different electrolytes is given in Figure 3, which shows the discharge capacity and columbic efficiency as a function of discharging/charging cycle. Superior cycling performance and efficiency was obtained for the VOx-NTs in both TMS and in TMS with EA (10%), although the initial capacities in the TMS-based electrolytes were lower than that in the AN-based electrolyte. After the 80th cycle, the capacity retention in AN dropped to 30%, whereas for TMS and TMS with EA(10%), the capacity retentions were estimated to be 70% and 50%, respectively. The capacity retention in AN electrolyte was significantly degraded even when changing the discharging cutoff voltage from −1.5 to −1.3 V (vs Ag/Ag+). This suggests that the degradation in AN electrolyte could not be improved just by lowering utilization of Mg2+.

Figure 2. Charge−discharge curves for VOx-NTs in (a) AN, (b) TMS, and (c) TMS-EA electrolytes with a current density of 60 mA g−1.

Figure 4a and b displays the impedance spectra of the VOxNT electrodes measured in AN and TMS-based electrolytes obtained using an appropriate equivalent circuit. Before cycling, the charge transfer resistance, Rct, had a smaller value in AN than in TMS-based electrolytes, implying that Mg insertion/ extraction into the VOx-NT in AN electrolyte is kinetically favored. This result is consistent with the higher initial capacity B

DOI: 10.1021/acsami.6b05808 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

electrolytes. Figure 5a shows top view of SEM image of asprepared VOx-NT electrode and panels b and c show top view SEM images of the VOx-NT electrodes after 80 cycles in AN and TMS-EA electrolytes. As shown in Figure 5b, the surface of the VOx-NT electrode cycled in the AN electrolyte is covered by a precipitate, indicating that the electrolyte is decomposed. On the other hand, the surface of the VOx-NT electrode cycled in the TMS-EA electrolyte shows clear surface features (shown in Figure 5c), including VOx-NTs and carbon, indicating that no precipitation from decomposition of the electrolyte occurred. Moreover, no significant change in the spectrum was observed from the FT-IR spectroscopy taken after 100th charge/discharge cycle, as shown in Figure 5d, compared with the spectrum taken as-prepared VOx electrode. The comparison of FT-IR spectra are shown in Figure S1. Moreover, the EDX analysis shown in Table 1 reveals that the VOx-NT electrode cycled in the AN electrolyte has a substantially different composition than the electrode cycled in the TMS-EA electrolyte. For the electrode cycled in the AN electrolyte, significantly higher amounts of O, Mg, and Cl were detected when compared with the sample cycled in the TMS-EA electrolyte, which is attributed to precipitation due to electrolyte decomposition. Considering the solvation complex of Mg ions with salt anions and solvent, the (ClO4)−2 anions were probably decomposed by the oxidation of AN solvent during charging. To investigate the change in the oxidation state of vanadium before and after cycling, the V 2p XPS spectrum was obtained for the VOx-NT electrode cycled in the TMS-EA electrolyte, as shown in Figure 6. The V 2p core spectra of both before and after cycling electrodes were divided into three oxidation states: V5+(517.4 eV), V4+(516.2 eV), and V3+(515.3 eV). The V5+:V4+:V3+peak area ratio estimated from the deconvolution of the V 2p XPS peaks were changed from 35:47:16 (before cycling) to 40:36:24 (after cycling), indicating V3+ to V5+ and V4+ and V5+ caused by partial reduction of vanadium in the VOx-NTs.30 In contrast to the XPS results for VOx-NTs cycled in the AN electrolyte in our previous work,17 V3+was observed on the VOx-NT surfaces,even after the 80 cycles, implying the existence of VO6 octahedrons with larger interstitial sites for the insertion/extraction of Mg2+ ions. Given that the VOx-NT electrodes in the TMS-based electrolytes have relatively lower initial activities and show little electrolyte decomposition, the improved capacity retention might be ascribed to the lower utilization of Mg2+ ions and higher interfacial stability.31

Figure 3. Discharge capacity (closed symbol) and columbic efficiency(%) (open symbol) of VOx-NT electrodes measured in different electrolytes: AN, TMS, and TMS-EA.

of the VOx-NT electrode in the AN electrolyte, as shown in Figure 3. For the TMS-based electrolytes, additional capacitance was observed in the high-frequency regime, which might be associated with the slow transport of Mg ions in highviscosity TMS solutions. The charge-transfer impedances in both cases increased after the 80th cycle, while the ohmic resistance, Rs, and Warburg impedance, Zw, were almost unchanged. It should be noted that the increase in nonohmic impedance with cycling was more dramatic for the cell with the AN electrolyte than the cell with TMS-EA. After the 80th cycle, the Rct value significantly increased from 78 to 1750 Ωcm2 for the cell with the AN electrolyte, whereas it increased from 60 to 400 Ωcm2 for the cell with the TMS-EA electrolyte. The impedance data indicates that the capacity fading of the VOx-NTs could be mainly attributed to the degradation of the charge-transfer reaction at the electrode surface, rather than to the transport of Mg ions through the electrode, depending on the kind of electrolyte solution. This result is similar to that result reported earlier with the lamda-MnO2 electrode for the insertion/extraction of Mg ions.23 To further verify that the surface properties were degraded with cycling, we investigated the microstructure and composition of the VOx-NT electrode surfaces using SEM and EDX for electrodes that were cycled in different

Figure 4. Nyquist plots of AC impedance data for the VOx-NT electrodes cycled in (a) AN and (b) TMS-EA electrolytes after the first and 80th cycles. C

DOI: 10.1021/acsami.6b05808 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 5. SEM images of (a) as-prepared the VOx-NT electrodes and after 80 cycles in (b) AN and (c) TMS-EA electrolytes and (d) FT-IR spectra of as-prepared VOx-NT electrode and VOx-NT after 100 cycles in TMS-EA electrolytes.

the M062X functional and the 6-311+G** basis set. Frequency analysis was performed at the same level of theory. Then, the solvent-phase energies were determined based on the selfconsistent reaction field (SCRF) theory, using the conductorlike polarizable continuum medium (CPCM) calculation. The calculated energy values are shown in Figure 7.

Table 1. Surface Composition as Determined by EDX Analysis atomic percent (%) electrolyte

V

O

C

Mg

Cl

AN TMS (EA, 10%)

1 1

13.14 1.71

1.6 6.07

15.91 1.78

12.35 0

E(Mg 2 +/Sol) = E0(Mg 2 + + Sol) − E0(Mg 2 +) − E0(Sol)

To understand the differences between the electrolytes consisting of Mg(ClO4)2 in pure AN and in TMS-EA mixtures and the role of the solvent molecules (Sol = AN, EA, and TMS) in these electrolytes, we calculated the interaction energies of Sol with Mg2+, Mg(ClO4)+, and ClO4−using eqs 1−3, respectively, where E0 is the zero point energy (ZPE) corrected value calculated using the Gaussian 09 suite of programs32 with

(1)

E(Mg(ClO4 )+ /Sol) = E0(Mg(ClO4 )+ + Sol) − E0(Mg(ClO4 )+ ) − E0(Sol) (2)

Figure 6. V 2p XPS profile of the VOx-NTs measured (a) before cycling and (b) after 80th cycling in the TMS-EA electrolyte. The V3+, V4+, and V5+ peaks obtained by deconvolution of the V 2p peak are shown. D

DOI: 10.1021/acsami.6b05808 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 7. Gas (black) and solvent (red) phase interaction energies between (a) Mg2+, (b) Mg(ClO4)+, and (c) ClO4− and Sol (AN, EA, and TMS). The ab initio calculations were performed using Gaussian09 at the M062X/6-311+G** level.

E(ClO4 − /Sol) = E0(ClO4 − + Sol) − E0(ClO4 −) − E0(Sol)

(3)

In the AN electrolyte, AN molecules stabilize cationic Mg2+/ Mg(ClO4)+ complexes through Mg−AN coordination, leading to an ionic equilibrium. This process also increases the amount of ClO4− anions available for noncovalent interaction with AN molecules. Figure 7 shows the relatively stronger gas- and solvent-phase interaction energies between Mg cations and AN than those between ClO4− anion and AN. There was no solvent-phase interaction energy between ClO4− anion and AN. In the TMS-EA electrolytes, both of gas- and solvent-phase interaction energies between EA and Mg cations are relatively stronger than those between TMS and Mg cations. There show no solvent-phase interaction between TMS and Mg cations. Such interacting behavior between TMS and Mg cations will shift the position of the chemical equilibrium toward the neutral species, resulting in the low capacity of cells with the TMS electrolyte. This information allows us to draw a schematic view of the TMS-EA electrolytes, as shown in Figure 8. Although the chemical equilibrium cannot be explained by simple energetics, cationic Mg2+/Mg(ClO4)+ complexes would be stabilized mainly by EA molecules through Mg−EA coordination. To compare the redox potentials of AN and TMS-EA electrolytes, the first electron affinities (EA) of Mg-cationic complexes and the first ionization energies (IE) of ClO4− complexes were calculated (Figure 9). The calculations indicate that the TMS-EA electrolyte has a wider electrochemical window than the AN electrolyte. Electrolyte reduction starts by moving one electron from the negative electrode to the lowest unoccupied molecular orbital (LUMO) of Mg-cationic complexes. However, it was difficult to obtain a clear picture of the detailed Mg-deposition and electrolyte-reduction processes in this study. We note that the Mg/Mg2+ potential in organic electrolytes is different from that in aqueous media, which is −2.37 V versus the standard hydrogen electrode, and is determined by Mg-deposition processes involving both anions and solvents.33,34 Thus, here we compared the relative EA values for Mg-cationic complexes in the AN and TMS-EA electrolytes using eqs 4 and 5. The EA values of Mg-complexes in EA are lower than those in AN (Figures 9a and b), indicating that the reduction potentials of the TMS-EA electrolyte is lower than that of the AN electrolyte.

Figure 8. Schematic view of Mg(ClO4)2 in (a) AN and (b) TMS-EA.

EA(Mg(ClO4 )+ Sol) = E0(Mg(ClO4 )+ + Sol + e−) − E0(Mg(ClO4 )+ + Sol)

In the Mg(ClO4)2-based electrolytes, electrolyte oxidation starts by losing one electron from the highest-occupied molecular orbital (HOMO) of ClO4−, solvent molecules, or their complexes to the positive electrode. Without verifying the Mg/Mg2+ potential of these electrolytes, oxidation can only be discussed relatively by calculating the IE values using eq 6 as shown in Figure 9c. IE(ClO4 −Sol) = E0(ClO4 − + Sol − e−) − E0(ClO4 − + Sol)

(6)

The oxidation of electrolytes usually results in the formation of neutral or cationic radicals, which are immediately stabilized by or undergo further reaction with nearby molecules. Htransport is known to be one of main oxidatively driven intermolecular reactions.35 Thus, H-transport from Sol to

EA(Mg 2 +Sol) = E0(Mg 2 + + Sol + e−) − E0(Mg 2 + + Sol)

(5)

(4) E

DOI: 10.1021/acsami.6b05808 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

using a magnetic stirrer for more than 2 days to obtain a homogeneous sticky solution. The mixture was poured into a Teflon-lined autoclave and heated at 180 °C for 18 h using a microwave (MARS5, CEM Co.). The resulting product was filtered and washed with ethanol several times. The final products were dried in a convection oven at 80 °C for more than 8 h and heated in a vacuum oven at 100 °C for 2 h to remove the residual solvent. Electrochemical Measurements. To make an electrode for electrochemical testing, a VOx-NT electrode was prepared by slurry casting. A composite powder consisting of VOx powder and carbon (Ketjen black) was mixed with polyvinylidene fluoride (PVdF) in Nmethyl pyrrolidone (NMP) solution without any extra conductive carbon. The electrode consists of 60 wt % active material, 20 wt % carbon black, and 20 wt % PVdF. The slurry was then coated on Al foil and finally dried under vacuum at 120 °C for 2 h. After pressing, the dried paste was punched into a disc (1 cm diameter). To investigate the effect of electrolyte solvent on the cyclability of the electrode, charge−discharge curves were measured in 0.5 M Mg(ClO4)2 in AN and TMS solutions using a three-electrode cell with a Ag/AgNO3 (0.1 M AgNO3 in AN or TMS) reference electrode. To exclude side-reactions at the interface between the counter electrode and electrolyte, Pt mesh was used as the counter electrode in the three-electrode cell. Prior to the electrochemical measurements, the electrolyte was deaerated by bubbling with purified N2 gas for 4 h and then dried over molecular sieves. The cells were charged and discharged using a conventional constant-current (CC) protocol with a current density of 60 mA g−1 between 0.4 and −1.5 V vs the reference electrode. To obtain reliable data, five cells were evaluated for each sample. All electrochemical experiments were conducted at 24 °C. The impedance spectra for an open circuit in galvanostatic mode over a frequency range of 0.1 Hz to 10 kHz using an AC perturbation of 10 mV were measured with a frequency response analyzer (Solartron, SI 1255 FRA) in conjunction with a potentiostat (Solartron, SI 1287 ECI). Material Characterization. To identify the morphology of the VOx-NT electrode cycled in AN and TMS electrolytes, scanning electron microscopy (SEM) was performed with a S-4700 FE-SEM (Hitachi). The surface chemical composition of the cycled electrode was characterized using X-ray photoemission spectroscopy (XPS, Quantera II). The spectrometer was equipped with a finely focused Al Kα (hυ = 1486.6 eV) beam source of 100 μm diameter. The residual pressure inside the XPS analysis chamber was 9.3 × 10−7 Pa. To avoid any contamination, the cycled electrodes were dried in the glovebox for 24 h and transferred from an Ar-filled glovebox to the XPS chamber using a vacuum transfer holder without exposing to the air and moisture. The viscosities of electrolytes were measured with a modular compact rheometer (Anton Paar, MCR302) at 25 °C.

Figure 9. First electron affinities (EA) of calculated for (a) Mg2+Sol and (b) Mg(ClO4)+Sol in AN and EA, and (c) first ionization energies (IE) calculated for ClO4−Sol in AN and TMS. (d) The oxidative structures of ClO4−Sol. The ab initio calculations were performed using Gaussian09 at the M062X/6-311+G** level.

ClO4−, as shown in Figure 9d, is considered in eq 6. Calculations show that IE(ClO4−TMS), which models the oxidation potential of the Mg(ClO4)2-TMS-EA electrolyte, is slightly higher than IE(ClO4−AN), which models the oxidation potential of the Mg(ClO4)2-AN electrolyte. In addition to the wider electrochemical window of the TMS-EA electrolyte compared with that of the AN electrolyte, the low dipole moment (∼0 D) of TMS can hinder the access of TMS to the surface of the VOx-NT electrode during charging conditions. In contrast, the dipole moment of AN is ∼4 D. This is consistent with the literature that experimentally shows the TMS-based electrolyte has high voltage stability above 5.0 V(Ag/Ag+).36



CONCLUSIONS Improved cycling retention of VOx-NT for Mg insertion/ extraction was achieved using electrolytes consisting of Mg(ClO4)2 in TMS-EA mixed solvents. This system has higher electrochemical stability than the AN system. On the basis of the high charge-transfer resistance and the presence of a precipitate on the surface after cycling in the AN electrolyte, it is suggested that the degradation of the electrochemical performance of VOx-NTs is most likely due to interfacial reaction products formed by decomposition of the electrolyte during charging. We believe that these combined theoretical and experimental studies open the possibility of demonstrating Mg electrolytes with electrochemical stability and high capacity for VOx-NT electrodes.





ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b05808.



EXPERIMENTAL SECTION

Molecular concentrations of elements in VOx NTs and FT-IR spectra of as-prepared VOx-NT electrode and VOx-NT after 20 cycles in ACN electrolytes (PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Tel: +82-10-6367-7594. Fax: +82-31-8061-1319.

Synthesis of VOx Nanotubes. The VOx-NTs used in this study were prepared using a microwave-assisted hydrothermal method that has been described elsewhere.17,26−29 We have used the same synthesis method with HT-VOx nanotube in the previous paper.17 Vanadium pentoxide (0.91 g) and octadecylamine (1.35 g) were dissolved in 5 mL of ethanol and 25 mL of water. The precursor solution was mixed

Notes

The authors declare no competing financial interest. F

DOI: 10.1021/acsami.6b05808 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces



Molecular Storage of Mg Ions with Vanadium Oxide Nanoclusters. Adv. Funct. Mater. 2016, 26 (20), 3446−3453. (21) Jiao, L.; Yuan, H.; Wang, Y.; Cao, J.; Wang, Y. Mg Intercalation Properties into Open-Ended Vanadium Oxide Nanotubes. Electrochem. Commun. 2005, 7, 431−436. (22) Jiao, L.; Yuan, H.; Si, Y.; Wang, Y.; Cao, J.; Gao, X.; Zhao, M.; Zhou, X.; Wang, Y. Electrochemical Insertion of Magnesium in OpenEnded Vanadium Oxide Nanotubes. J. Power Sources 2006, 156, 673− 676. (23) Rasul, S.; Suzuki, S.; Yamaguchi, S.; Miyayama, M. High Capacity Positive Electrodes for Secondary Mg-Ion Batteries. Electrochim. Acta 2012, 82, 243−249. (24) Krumeich, F.; Muhr, H.-J.; Niederberger, M.; Bieri, F.; Schnyder, B.; Nesper, R. Morphology and RopochemicalReactions of Novel Vanadium Oxide Nanotubes. J. Am. Chem. Soc. 1999, 121, 8324−8331. (25) Li, J.-M.; Chang, K.-H.; Wu, T.-H.; Hu, C.-C. MicrowaveAssisted Hydrothermal Synthesis of Vanadium Oxides for Li-Ion Supercapacitors: The Influences of Li-Ion Doping and Crystallinity on the Capacitive Performances. J. Power Sources 2013, 224, 59−65. (26) Livage, J. Hydrothermal Synthesis of Nanostructured Vanadium Oxides. Materials 2010, 3, 4175−4195. (27) Shao, N.; Sun, X.-G.; Dai, S.; Jiang, D.-e. Electrochemical Windows of Sulfone-Based Electrolytes for High-Voltage Li-Ion Batteries. J. Phys. Chem. B 2011, 115, 12120−12125. (28) Kim, H.; Kim, R.-H.; Lee, S.-S.; Kim, Y.; Kim, D. Y.; Park, K. Effects of Ni Doping on the Initial Electrochemical Performance of Vanadium Oxide Nanotubes for Na-Ion Batteries. ACS Appl. Mater. Interfaces 2014, 6, 11692−11697. (29) Kim, H.; Kim, D. Y.; Kim, Y.; Lee, S.-S.; Park, K. Na Insertion Mechanisms in Vanadium Oxide Nanotubes for Na-Ion Batteries. ACS Appl. Mater. Interfaces 2015, 7, 1477−1485. (30) Xu, K.; Angell, C. A. Sulfone-Based Electrolytes for Lithium-Ion Batteries. J. Electrochem. Soc. 2002, 149, A920−A926. (31) Corr, S. A.; Grossman, M.; Furman, J. D.; Melot, B. C.; Cheetham, A. K.; Heier, K. R.; Seshadri, R. Controlled Reduction of Vanadium Oxide Nanoscrolls: Crystal Structure, Morphology, and Electrical Properties. Chem. Mater. 2008, 20, 6396−1076. (32) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision D.01; Gaussian, Inc.: Wallingford, CT, 2009. (33) Kim, D. Y.; Lim, Y.; Roy, B.; Ryu, Y.-G.; Lee, S.-S. Operating Mechanisms of Electrolytes in Magnesium Ion Batteries: Chemical Equilibrium, Magnesium Deposition, and Electrolyte Oxidation. Phys. Chem. Chem. Phys. 2014, 16, 25789−25798. (34) Roy, B.; Kim, D. Y.; Lim, Y.; Lee, S.-S.; Son, Y.-H.; Doo, S.-G. Cyclic Silicon−Nitrogen−Silicon Core Derived Silylamido−Magnesium Compounds for Magnesium−Battery Electrolytes with Improved Oxidation Stability. J. Power Sources 2015, 297, 551−555. (35) Kim, D. Y.; Park, M. S.; Lim, Y.; Kang, Y.-S.; Park, J.-H.; Doo, S.-G. Computational Comparison of Oxidation Stability: Solvent/Salt Monomers vsSolvent−Solvent/Salt Pairs. J. Power Sources 2015, 288, 393−400. (36) Abouimrane, A.; Belharouak, I.; Amine, K. Sulfone-Based Electrolytes for High-Voltage Li-ion Batteries. Electrochem. Commun. 2009, 11, 1073−1076.

REFERENCES

(1) Etacheri, V.; Marom, R.; Elazari, R.; Salitra, G.; Aurbach, D. Challenges in the Development of Advanced Li-Ion Batteries: A Review. Energy Environ. Sci. 2011, 4, 3243−3262. (2) Dunn, B.; Kamath, H.; Tarascon, J.-M. Electrical Energy Storage for the Grid: A Battery of Choices. Science 2011, 334, 928−935. (3) Yang, Z.; Zhang, J.; Kintner-Meyer, C. W.; Lu, X.; Choi, D.; Lemmon, J. P.; Liu, J. Electrochemical Energy Storage for Green Grid. Chem. Rev. 2011, 111, 3577−3613. (4) Aurbach, D.; Lu, Z.; Schechter, A.; Gofer, Y.; Gizbar, H.; Turgeman, R.; Cohen, Y.; Moshkovich, M.; Levi, E. Prototype Systems for Rechargeable Magnesium Batteries. Nature 2000, 407, 724−727. (5) Yoo, H. D.; Shterenberg, I.; Gofer, Y.; Gershinsky, G.; Pour, N.; Aurbach, D. Mg Rechargeable Batteries: An On-Going Challenge. Energy Environ. Sci. 2013, 6, 2265−2279. (6) Shterenberg, I.; Salama, M.; Gofer, Y.; Levi, E.; Aurbach, D. The Challenge of Developing Rechargeable Magnesium Batteries. MRS Bull. 2014, 39, 453−460. (7) Liu, B.; Luo, T.; Mu, G.; Wang, X.; Chen, D.; Shen, G. Rechargeable Mg-Ion Batteries Based on WSe2Nanowire Cathodes. ACS Nano 2013, 7, 8051−8058. (8) Liu, Y.; Jiao, L.; Wu, Q.; Zhao, Y.; Cao, K.; Liu, H.; Wang, Y.; Yuan, H. Synthesis of rGO-Supported Layered MoS2 for HighPerformance Rechargeable Mg Batteries. Nanoscale 2013, 5, 9562− 9567. (9) Chernova, N. A.; Roppolo, M.; Dillon, A. C.; Whittingham, M. S. Layered Vanadium and Molybdenum Oxides: Batteries and Electrochromics. J. Mater. Chem. 2009, 19, 2526−2552. (10) Imamura, D.; Miyayama, M.; Hibino, M.; Kudo, T. Mg Intercalation Properties into V2O5Gel/Carbon Composites UnderHigh-Rate Condition. J. Electrochem. Soc. 2003, 150, A753−A758. (11) Nam, K. W.; Kim, S.; Lee, S.; Salama, M.; Shterenberg, I.; Gofer, Y.; Kim, J.-S.; Yang, E.; Park, C. S.; Kim, J.-S.; Lee, S.-S.; Chang, W.-S.; Doo, S.-G.; Jo, Y. N.; Jung, Y.; Aurbach, D.; Choi, J. W. The High Performance of Crystal Water Containing Manganese Birnessite Cathodes for Magnesium Batteries. Nano Lett. 2015, 15, 4071−4079. (12) He, D.; Wu, D.; Gao, J.; Wu, X.; Zeng, X.; Ding, W. Flower-Like CoS with Nanostructures as a New Cathode-Active Material for Rechargeable Magnesium Batteries. J. Power Sources 2015, 294, 643− 649. (13) Cui, C.-j.; Wu, G.-m.; Shen, J.; Zhou, B.; Zhang, Z.-h.; Yang, H.y.; She, S.-f. Synthesis and Electrochemical Performance of Lithium Vanadium Oxide Nanotubes as Cathodes for Rechargeable LithiumIon Batteries. Electrochim. Acta 2010, 55, 2536−2541. (14) Kim, J.-S.; Chang, W.-S.; Kim, R.-H.; Kim, D.-Y.; Han, D.-W.; Lee, K.-H.; Lee, S.-S.; Doo, S.-G. High-Capacity Nanostructured Manganese Dioxide Cathode for Rechargeable Magnesium Ion Batteries. J. Power Sources 2015, 273, 210−215. (15) Krtil, P.; Kavan, L.; Novák, P. Oxidation of Acetonitrile-Based Electrolyte Solutions atHigh Potentials: An In Situ Fourier Transform Infrared Spectroscopy Study. J. Electrochem. Soc. 1993, 140, 3390− 3395. (16) Pereira-Ramos, J. P.; Messina, R.; Perichon, J. Electrochemical Formation of a Magnesium Vanadium Bronze MgxV2O5 in SulfoneBased Electrolytes at 150°C. J. Electroanal. Chem. Interfacial Electrochem. 1987, 218, 241−249. (17) Kim, R.-H.; Kim, J.-S.; Kim, H.-J.; Chang, W.-S.; Han, D.-W.; Lee, S.-S.; Doo, S.-G. Highly Reduced VOx Nanotube Cathode Materials with Ultra-High Capacity for Magnesium Ion Batteries. J. Mater. Chem. A 2014, 2, 20636−20641. (18) Novák, P.; Desilvestro, J. Electrochemical Insertion of Magnesium in Metal Oxides and Sulfides from Aprotic Electrolytes. J. Electrochem. Soc. 1993, 140 (1), 140−144. (19) An, Q.; Li, Y.; Deog Yoo, H.; Chen, S.; Ru, Q.; Mai, L.; Yao, Y. GrapheneDecorated Vanadium Oxide Nanowire Aerogel for LongCycle-Life Magnesium Battery Cathodes. Nano Energy 2015, 18, 265− 272. (20) Cheng, Y.; Shao, Y.; Raju, V.; Ji, X.; Mehdi, B. L.; Han, K. S.; Engelhard, M. H.; Li, G.; Browning, N. D.; Mueller, K. T.; Liu, J. G

DOI: 10.1021/acsami.6b05808 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX