Bismuth Nanoparticle Decorating Graphite Felt as a High-Performance

Feb 11, 2013 - Activated Carbon Fiber Paper Based Electrodes with High Electrocatalytic Activity for Vanadium Flow Batteries with Improved Power Densi...
0 downloads 10 Views 3MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Bismuth Nanoparticle Decorating Graphite Felt as a HighPerformance Electrode for an All-Vanadium Redox Flow Battery Bin Li, Meng Gu, Zimin Nie, Yuyan Shao, Qingtao Luo, Xiaoliang Wei, Xiaolin Li, Jie Xiao, Chongmin Wang, Vincent Sprenkle, and Wei Wang* Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington 99352, United States S Supporting Information *

ABSTRACT: Employing electrolytes containing Bi3+, bismuth nanoparticles are synchronously electrodeposited onto the surface of a graphite felt electrode during operation of an allvanadium redox flow battery (VRFB). The influence of the Bi nanoparticles on the electrochemical performance of the VRFB is thoroughly investigated. It is confirmed that Bi is only present at the negative electrode and facilitates the redox reaction between V(II) and V(III). However, the Bi nanoparticles significantly improve the electrochemical performance of VRFB cells by enhancing the kinetics of the sluggish V(II)/V(III) redox reaction, especially under high power operation. The energy efficiency is increased by 11% at high current density (150 mA·cm−2) owing to faster charge transfer as compared with one without Bi. The results suggest that using Bi nanoparticles in place of noble metals offers great promise as high-performance electrodes for VRFB application. KEYWORDS: Energy storage, redox flow battery, electrode, catalyst, vanadium

T

and overdischarge. A VRFB proposed by Skyllas-Kazacos et al.8 in the 1980s stands out as the most promising and extensively developed flow battery system so far, owing to its advantage of reduced cross-contamination by enlisting the same element, V, in both electrolytes. Compared with other RFB systems, VRFBs have demonstrated several additional advantages, such as excellent electrochemical reversibility and high efficiency. Common challenges for VRFBs include a limitation of the active material concentration, a need for an expensive membrane, and a gas evolution issue. Traditional sulfuric acid-based VRFB technology is significantly hindered by the narrow temperature window of V ion solubility and stability in electrolyte solutions. A recently invented mixed-acid (sulfuric and chlorine-acid) VRFB, however, has achieved a significantly improved energy density and a wider operational temperature window.9 Despite their compelling merits and continuous development over the last 20 years, commercial uptake of VRFBs is still hampered by both technical and economic barriers.10 Among them, high cost is one major factor that has kept the VRFB technology from broader market penetration.11 In this regard, a flow battery capable of operating under high charge/discharge current density is always desired, merely because improved power density can lead to a reduced stack size, achieving a much lower capital cost of the VRFB system. Similar to any

he globally accelerating deployment and integration of renewable energy technologies, such as solar and wind power, underscores a worldwide recognition that reliable, inexpensive, and clean energy is vital for the increasing prosperity and sustainable future of our society.1,2 However, the intermittent nature of renewable energy presents an imperative need to develop large-scale electrical energy storage systems for integrating renewables, because the introduction of more than 20% intermittent energy from renewable sources without the required storage could destabilize the grid with looming threats of voltage increases and frequency fluctuations.3 Among the most promising large-scale energy storage technologies are redox flow batteries (RFBs), which capitalize on reactions between two redox couples to perform the reversible conversion between electrical energy and chemical energy.4−7 With the liquid electrolyte and electroactive materials stored externally, true flow batteries have many advantages over solid-state secondary battery systems, the most prominent of which is the ability to tailor the energy capacity independently from the power output. The decoupling of the power and energy requirements gives considerable design latitude to individually vary the energy capacity or power capability for different energy storage applications. In addition, the role of the electrode in an RFB is merely to provide the electrochemically active surface for the redox reaction to take place. Free from physiochemical changes and mechanical stress, the simplicity of the working mechanism renders long service life to RFBs. Other attractive features of flow batteries include quick response, low self-discharge, and tolerance to overcharge © 2013 American Chemical Society

Received: January 18, 2013 Revised: February 4, 2013 Published: February 11, 2013 1330

dx.doi.org/10.1021/nl400223v | Nano Lett. 2013, 13, 1330−1335

Nano Letters

Letter

glovebox, for which glassy carbon, platinum mesh, and Ag/ AgCl are employed as working, counter, and reference electrodes, respectively. The testing was performed from −0.85 to 1.5 V vs Ag/AgCl reference electrode in solutions of 2 M VOSO4 + 5 M HCl with or withoout 0.01 M BiCl3 solutions at a scan rate of 50 mV·s−1. In addition, the morphology and microstructure of modified GFs and the resultant nanoparticles were characterized by a field emission scanning electron microscope (FESEM, JEOL JSM-7600F) and an aberration-corrected scanning transmission electron microscope (STEM) (TEM/STEM, FEI Titan) at 300 kV, respectively. The conductivity of electrolytes with and without the addition of BiCl3 was measured by a benchtop conductivity meter (Orion 3-Star, Thermo Scientific, USA). Figure 1a shows CV curves on glassy carbon electrodes in 2 M VOSO4 + 5 M HCl electrolytes with and without 0.01 M

electrochemical system, increase of current density in a VRFB can increase the polarization, resulting in a declining overall performance. It is therefore of critical importance to improve the rate capability and efficiency of VRFB systems. In a VRFB flow cell, polarization arises from different parts of the system, including mainly kinetic polarization at the electrolyte/electrode interface, ohmic loss of electrolyte, and electrode and concentration polarization from mass transport limitations. As one critical component in a VRFB system, an electrode contributes to the system polarization through not only the ohmic resistance, but also the charge transfer polarization, since the redox reactions between couples of V(II)/V(III) and V(IV)/V(V) take place on the electrode surface. The typical electrode materials for VRFB are graphite felts (GFs) because of their low cost, high stability, high conductivity, corrosion resistance, and high surface area. However, their poor kinetic reversibility and electrochemical activity often limit the VRFB operation to low current density (∼50 mA·cm−2).12,13 An effective way to address this issue is to introduce an electrocatalyst on the surface of the GF to reduce the activation barrier for the redox conversion. Many researchers have attempted to enhance the VRFB performance by depositing conductive noble metals as catalysts, such as Pt, Au, Ir, Pd, and Ru,12,14−17 which however is not practical due to the cost and susceptibility to hydrogen evolution. Recently, low-cost metal oxides such as Mn3O418 and WO319 were reported as catalysts, but their performance is limited by their low conductivity. Moreover, their performance depends heavily on the nano size and uniform distribution of catalysts on GF surfaces, which often involves complex and tedious preparation. In this regard, it is necessary to design advanced catalysts of low cost, high catalytic activity, and high conductivity. Herein, we report a low-cost and highly conductive Bi metal as a novel catalyst to enhance the electrochemical activity of GF electrodes, enabling VRFBs for high current-density operation. A small amount of soluble Bi ions was added to the electrolytes, and the Bi nanoparticles were then simultaneously reduced and electrodeposited on the surface of GFs during the charge of the flow cells, avoiding complex pretreatment procedures. The catalytic effects of Bi on the electrochemical performance of VRFBs are investigated. In this research, original electrolytes were prepared by dissolving 2 M VOSO4 (Aldrich, 99%) in 5 M HCl solutions (Aldrich, 99%).9 A small amount of BiCl3 (Aldrich, 99.5%) was added to the electrolytes with the concentration of Bi3+ varying from 0 to 0.02 at 0.005 M intervals. The corresponding anolytes and catholytes can be prepared by an effective “balance” process.20 The GF electrodes (grade: GFD5EA) are commercially available from SGL Carbon Group, Germany. The setup of flow cells with active electrode areas of 10 cm2 is described in detail in a previous report.21 The single cell was connected to two glass reservoirs containing 50 mL catholytes and 50 mL anolytes, respectively. Both reservoirs were purged with nitrogen gas and then sealed prior to the electrochemical tests to minimize oxidation of V(II) and Bi. Nafion 115 was employed as membranes. Electrolytes were pumped at a flow rate of 20 mL·min−1 (0.33 cm·s−1 face velocity) through a peristaltic pump. The electrochemical performance of the flow battery was carried out using a potentiostat/galvanostat (Arbin Instruments, USA) within the voltage window between 0.8 and 1.6 V under a constant current mode at current densities varying from 50 to 150 mA·cm−2. A cyclic voltammetry (CV) test was conducted in a three-electrode cell in a nitrogen-filled

Figure 1. (a) Cyclic voltammograms with glassy carbon as working electrode in solutions of 2 M VOSO4 + 5 M HCl with or without 0.01 M BiCl3 at a scan rate of 50 mV·s−1. (b) Overpotential as a function of current density with pristine glassy carbon and Bi-coated glassy carbon as working electrodes.

BiCl3. It can be clearly seen from Figure 1a that both curves exhibit four main peaks. The corresponding reactions of redox couples are marked in the figure. For the CV curve in the solutions without Bi3+ ions, the oxidation and reduction peak potential separation corresponding to redox couple V(II)/ V(III) was 0.31 V, which was much larger than that of the redox couple V(IV)/V(V) with the value of 0.23 V, suggesting that the sluggish kinetics at the negative side associated with a V(II)/V(III) redox reaction perhaps limit the overall VRFB electrochemical performance. However, upon adding 0.01 M Bi3+ ions into the solutions, the peak potential separation corresponding to redox couple V(II)/V(III) was decreased to 0.22 V, which was comparable with that of V(IV)/V(V). On the contrary, the addition of Bi3+ ions has hardly any effect on the redox reaction of V(IV)/V(V). The standard potential of 1331

dx.doi.org/10.1021/nl400223v | Nano Lett. 2013, 13, 1330−1335

Nano Letters

Letter

Figure 2. (a) FESEM images of GFs modified with Bi nanoparticles at different magnifications after cycling; (b) STEM image of Bi nanoparticles in the anolytes resulting after cycling; (c) and (d) HRTEM and corresponding FFT images of Bi nanoparticles. All the results were obtained from the flow cells using the electrolytes containing 0.01 M Bi3+.

(short duration of 30 s) under different current densities with the same electrolyte mentioned above. The resulting overpotential as a function of charge/discharge current density toward the V(II)/V(III) redox reaction is plotted in Figure 1b. Compared with the pristine carbon electrode, it is clear that the increase of the overpotential with increasing current density was significantly subdued for the Bi-modified carbon electrode in both the charge and the discharge processes, indicating much improved reaction kinetics thanks to the electrocatalytic effect of Bi toward the redox couple of V(II)/V(III). Figure 2a shows the morphology of the GFs decorated with Bi at different magnifications, which was obtained after several cycles in the flow cell using electrolytes containing 0.01 M Bi3+. The nanoparticles were tethered to the surface of the GF and found only on the negative side; they were identified with energy dispersive X-ray spectroscopy (EDX) (see Figure S2) to be Bi metal according to the atomic ratio of Bi/O (≫2/3), which agrees well with the results of CV tests. A trace amount of oxygen was ascribed to the air oxidation of Bi nanoparticles during scanning electron microscope (SEM) sample preparation. In addition, the transmission electron microscopy (TEM) sample was prepared by dropping the anolytes that resulted after cycling onto the regular carbon film grid. The Z-contrast image collected using a high-angle annular dark field detector in a STEM can reflect the average atomic number of a specific region.22,23 Therefore, the brighter the contrast in the Zcontrast image, the higher the average atomic number of this specific region. As shown in Figure 2b, the Z-contrast image clearly resolves the distribution of the heavy Bi element on a carbon film grid. Due to the much larger Z number of Bi (ZBi = 83) as compared to carbon (ZC = 6), the Bi nanoparticles exhibit a much higher contrast. One can see that the Bi

Bi/Bi(III) (0.046 V vs Ag/AgCl) is located between these of V(II)/V(III) and V(IV)/V(V). Bi metal will be electrodeposited on the negative electrode before the reduction of V(III) to V(II), which will then be oxidized to Bi3+ ions prior to the oxidation of V(IV) to V(V). Comparing the CV curves with and without Bi additive, only the redox reaction on the negative electrode side corresponding to the V(II)/V(III) redox reaction was affected, suggesting that Bi metal rather than Bi ions has a catalytic effect toward V(II)/V(III) redox couples and Bi presents in the form of Bi metal on the negative side and Bi ions on the positive side in actual VRFBs. To further investigate the catalytic effect of Bi on the redox couple of V(II)/V(III), CV tests at increasing scan rates from 10 mV·s−1 to 100 mV·s−1 were performed. The test solution contained 2 M V ions (around 50% state of charge (SOC): 50 mol % V2+ + 50 mol % V3+) and 5 M H+. To avoid the direct reduction of Bi3+ into Bi in the electrolytes containing V2+, the Bi metal had to be electrodeposited onto the glassy carbon electrode in advance, which was prepared by a CV scanning from 0 to −0.5 V vs Ag/AgCl in the solutions containing 2 M V3+, 0.01 M Bi3+, and 5 M H+. All of the experiments were carried out in an N2-filled glovebox. Comparing Figures S1a and b, for the Bi-modified glassy carbon electrode, the oxidation and reduction peak potential separation was significantly reduced and remained nearly constant throughout the entire scan rate range. For the Bi-modified glassy carbon electrode, the potential separation at 100 mV·s−1 was much smaller than that of the pristine glassy carbon electrode at 10 mV·s−1, indicating the improvement of redox reversibility toward V(II)/ V(III) ascribed to the Bi catalytic effect. In addition, the influence of the Bi on the overpotential of the V(II)/V(III) redox reaction was measured through shallow charge/discharge 1332

dx.doi.org/10.1021/nl400223v | Nano Lett. 2013, 13, 1330−1335

Nano Letters

Letter

Figure 3. Electrochemical performance of VRFBs employing electrolytes containing different Bi3+ concentrations as a function of charge/discharge current density: (a) charge−discharge curves at 150 mA·cm−2; (b) CE and VE; (c) EE; (d) cycling performance VRFB flow cell with electrolytes containing 0.01 M Bi3+ and without Bi3+ at 50 mA·cm−2.

150 mA·cm−2. Under fixed voltage window, the CE of battery is the ratio between flow cell charge and discharge capacity. The VE is determined by the voltage difference between the charge and discharge processes. The EE value is a derivative of the CE and VE (EE = CE × VE). It can be seen from Figure 3b that, for a given electrolyte composition, CE, with values of around 97%, increases slightly by 0.5−0.8% with increasing current density due to the reduced time of ion crossover through membranes. The addition of Bi has little influence on the improvement of CE. Figure 3b also shows the VE values for flow cells with different Bi3+ concentrations in electrolytes operated under different current densities. A fast charge/ discharge rate can cause large charge/discharge overpotential (see Figure 1b) leading to significant VE decline. At the same charge/discharge rate, as the amount of Bi3+ in the electrolytes increases, the VE values first increase and then decrease, with the highest value for the flow cell with 0.01 M Bi3+ additive in the electrolyte; that VE value reached 80.4%, which is around 12% higher than that without Bi3+ at 150 mA·cm−2. As displayed in Figure 3c, the trend of EE values with Bi3+ concentration and current density is similar to that of VE attributed to minor variation of CE. The flow cell containing 0.01 M Bi3+ in the electrolytes exhibits the highest EE value. Compared with the cell using electrolytes without Bi3+, under the current density of 150 mA·cm−2 the EE value was significantly improved by ∼11%. Figure 3d shows the cycling performance of a VRFB with and without 0.01 M Bi3+ in the electrolytes at the current density of 50 mA·cm−2. The EE values for the cells containing Bi ions were enhanced due to the catalytic effect of Bi nanoparticles. No obvious EE decay was observed over 50 cycles, suggesting excellent stability of the

nanoparticles are very small, ranging from 2−50 nm. High resolution TEM (HRTEM) was then used to identify the crystal structure of the Bi nanocrystals. As shown in Figure 2c, the HRTEM image showed ⟨001⟩ zone axis projection of the Bi nanoparticle. An atomic model of the ⟨001⟩ zone-projected Bi metal is overlaid on top of the image. The fast Fourier transform (FFT) of this HRTEM image is shown in Figure 2d, which further confirmed the rhombohedral crystal structure of the Bi nanocrystals. Charge/discharge cycling using a VRFB single flow cell with Nafion 115 as membrane was performed to further understand the catalytic effect of Bi on the electrochemical performance of the VRFB flow cell. For comparison, Figure 3a shows the charge−discharge voltage profiles at the same charge−discharge current density of 150 mA·cm−2 for the pristine electrolyte and electrolyte with Bi ions at 0.01 M concentration. A significantly reduced overpotential marked in the figure in both charge and discharge processes suggests the effect of the addition of Bi ions into the electrolytes and the subsequent reduction and electrodeposition of Bi nanoparticles onto the GF electrode, leading to lower charge voltage and higher discharge voltage. The open-circuit voltage was found to increase from 1.39 to 1.48 V, ascribed to the increasing SOC. The cutoff voltage was fixed at 1.6 V to avoid evolution of H2 and O2. For a given charge/discharge voltage window, the cell employing electrolytes with Bi ions exhibits significantly larger charge and discharge capacities than the one without Bi ions. To improve cell performance, the amounts of Bi ions in the electrolytes were systematically optimized with the concentration of Bi3+ varying from 0 to 0.02 at 0.005 M intervals. The Coulomb efficiency (CE), voltage efficiency (VE), and energy efficiency (EE) were identified under current density variation from 50 to 1333

dx.doi.org/10.1021/nl400223v | Nano Lett. 2013, 13, 1330−1335

Nano Letters

Letter

Figure 4. (a) Specific discharge capacity and (b) discharge energy density of VRFBs as a function of cycle number at different charge/discharge current densities employing the electrolytes containing different Bi3+ concentrations.

cells without Bi3+ in electrolytes. The fuel (V) utilization ratio was increased by more than 30%, which will contribute significantly to cost reduction for VRFB stacks as V electrolyte cost is nearly 40% of the VRFB system cost.11 Furthermore, discharge energy density was determined by both discharge capacity and voltage. As displayed in Figure 4b, the increased discharge voltage and capacity (see Figure 3a) resulted in improved discharge energy density due to the presence of Bi catalysts, 150 mA·cm−2, which is around 2.5 times higher than that without Bi. Based on these experimental results, Bi nanoparticles, acting as a powerful catalyst toward the redox couple of V(II)/V(III), can significantly improve the overall VRFB electrochemical performance. Key parameters of VRFB cell performance, such as VE, EE, utilization of fuels, and discharge energy densities, have been substantially enhanced. Similar experiments have been carried out on the iron− vanadium redox flow battery (IVRFB) recently invented at Pacific Northwest National Laboratory,21,24 which also utilizes redox couples of V(II)/V(III) on the negative half-cell side. The experimental details of the IVRFB system are described in the Supporting Information. Consistent with the results for VRFBs, the CV study conducted in the Fe−V electrolyte shown in Figure S4a indicate that Bi nanoparticles sourced from electrolytes containing 0.01 M Bi3+ ions served as an excellent electrocatalytic material toward the V(II)/V(III) redox couple and barely affected the electrochemical reaction of the Fe(II)/ Fe(III) redox couple at the positive half-cell. Accordingly, as shown in Figure S4b, the EE values were greatly enhanced, by 8%, at the high charge/discharge rate of 100 mA·cm−2. The cell performance of IVRFBs was significantly improved due to the excellent catalytic effect of Bi nanoparticles decorating GFs. In summary, Bi nanoparticles, as a low-cost, conductive, and novel electrocatalyst toward the V(II)/V(III) redox couple was proposed to enhance the electrochemical activity of GF electrodes in VRFB systems. The Bi nanoparticles were synchronously electrodeposited on the negative electrode as catalysts while running the cells using the electrolytes containing Bi ions. No Bi particles were found on the positive half-cell side. Bi nanoparticles deposited on the surfaces of GFs and suspended in the electrolytes facilitate the redox reaction, thus leading to greatly increased VRFB performance in VE, EE, utilization of fuels and discharge energy densities, especially under high current density. These results suggest that Bi nanoparticle-decorating GFs hold great promise as highperformance electrodes for VRFB applications.

electrocatalytic effect rendered by Bi nanoparticles over repetitive cycling. As previously discussed, the resistance of electrolytes contributes to the overall polarization and therefore affects the energy efficiency of flow cells. In our case, the conductivities of the electrolytes with different concentrations of Bi3+ ions were measured to be in a narrow range of 297−310 S/m. Thus, the change of electrolyte resistance is not considered to play a major role in the improvement of electrochemical performance. As discussed regarding the CV test (Figure 1), it is electrodeposited Bi particles that render a catalytic effect at the negative half-cell promoting the V(II)/V(III) redox reaction. Accordingly, it was inferred that the Bi catalyst greatly accelerates the redox reaction, especially at high charge/ discharge rates, presumably by lowering the kinetic activation energy. It is believed that the Bi nanoparticles were first electrochemically reduced and deposited onto the surface of the graphite fibers within the GF electrode as evidenced in Figure 2a. However, under the flowing electrolytes, it is very likely that some of the electrodeposited Bi particles were dislodged and as a result gave rise to the suspended Bi nanoparticles in the electrolytes confirmed in Figure 2b. The suspended Bi nanoparticles can also come into effect as catalysts when in contact with the GF electrode while flowing through. However, the increase of Bi3+ concentration in electrolytes may cause more Bi nanoparticles to be electrodeposited and agglomerated into larger particles on the surfaces of GFs, resulting in a higher tendency of Bi to be displaced from the surfaces of GFs. As shown in Figure S3, few Bi particles were observed on the surfaces of GFs after cycling when the concentration of Bi3+ ions was increased to 0.02 M. As a result, the optimal amount of Bi3+ in the electrolytes (0.01 M) was obtained. As is shown in Figure 1b, a fast charge or discharge rate can elevate the overpotential, which can result in the reduction of charge/discharge capacity due to a reduced useful voltage window. Figure 4a shows clearly that discharge capacity decreases with increasing discharge rate for electrolytes with different Bi3+ concentrations. However, the overpotential can be significantly reduced by adding appropriate Bi nanoparticles onto GFs as catalysts, which affords faster charge transfer. The flow cells with optimum Bi nanoparticles as catalysts sourced from the electrolytes containing 0.01 M Bi3+ ions exhibit the highest specific capacity, which reaches up to 64% of theoretical capacity (26.8 Ah·L−1) at 150 mA·cm−2, showing a large improvement compared with 30% of theoretical capacity for the 1334

dx.doi.org/10.1021/nl400223v | Nano Lett. 2013, 13, 1330−1335

Nano Letters

■

Letter

(17) Huang, R. H.; Sun, C. H.; Tseng, T.; Chao, W.; Hsueh, K. L.; Shieu, F. S. J. Electrochem. Soc. 2012, 159 (10), A1579−A1586. (18) Kim, K. J.; Park, M. S.; Kim, J. H.; Hwang, U.; Lee, N. J.; Jeong, G.; Kim, Y. J. Chem. Commun. 2012, 48, 5455−5457. (19) Yao, C.; Zhang, H.; Liu, T.; Li, X.; Liu, Z. J. Power Sources 2012, 218, 455−461. (20) Li, B.; Li, L.; Wang, W.; Nie, Z.; Chen, B.; Wei, X.; Luo, Q.; Yang, Z.; Sprenkle, V. J. Power Sources 2013, 229 (0), 1−5. (21) Wang, W.; Nie, Z. M.; Chen, B. W.; Chen, F.; Luo, Q. T.; Wei, X. L.; Xia, G. G.; Skyllas-Kazacos, M.; Li, L. Y.; Yang, Z. G. Adv. Energy Mater. 2012, 2 (4), 487−493. (22) Gu, M.; Song, C.; Yang, F.; Arenholz, E.; Browning, N. D.; Takamura, Y. J. Appl. Phys. 2012, 111 (8), 084906−084906−6. (23) Browning, N.; Chisholm, M.; Pennycook, S. Nature 1993, 366 (6451), 143−146. (24) Wang, W.; Kim, S.; Chen, B.; Nie, Z.; Zhang, J.; Xia, G.-G.; Li, L.; Yang, Z. Energy Environ. Sci. 2011, 4 (4), 4068−4073.

ASSOCIATED CONTENT

S Supporting Information *

Detailed experimental information for the IVRFB system as well as the CV scans at different scan rates, EDX data, and SEM images of GFs with different Bi3+ concentrations. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 1-509-372-4097. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors would like to acknowledge financial support from the U.S. Department of Energy’s (DOE’s) Office of Electricity Delivery and Energy Reliability (OE) (under Contract No. 57558). We also are grateful for enlightening discussions with Dr. Imre Gyuk of the DOE-OE Grid Storage Program. The S/ TEM work was conducted in the William R. Wiley Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by DOE’s Office of Biological and Environmental Research and located at PNNL. Pacific Northwest National Laboratory is a multiprogram national laboratory operated by Battelle for DOE under Contract DEAC05-76RL01830.

■

REFERENCES

(1) Yang, Z.; Zhang, J.; Kintner-Meyer, M. C. W.; Lu, X.; Choi, D.; Lemmon, J. P.; Liu, J. Chem. Rev. 2011, 111 (5), 3577−3613. (2) Dunn, B.; Kamath, H.; Tarascon, J.-M. Science 2011, 334 (6058), 928−935. (3) Steve Olson, R. W. F. The National Academies Summit on America’s Energy Future: Summary of a Meeting, Washington, D.C, March 13−14, 2008. (4) Wang, W.; Luo, Q.; Li, B.; Wei, X.; Li, L.; Yang, Z. Adv. Funct. Mater. 2013, 23, 970−986. (5) Leung, P.; Li, X.; Ponce de Leon, C.; Berlouis, L.; Low, C. T. J.; Walsh, F. C. RSC Adv. 2012, 2 (27), 10125−10156. (6) Weber, A.; Mench, M.; Meyers, J.; Ross, P.; Gostick, J.; Liu, Q. J. Appl. Electrochem. 2011, 41 (10), 1137−1164. (7) Skyllas-Kazacos, M.; Chakrabarti, M. H.; Hajimolana, S. A.; Mjalli, F. S.; Saleem, M. J. Electrochem. Soc. 2011, 158 (8), R55−R79. (8) Skyllas-Kazacos, M.; Rychick, M.; Robins, R. All-vanadium redox battery. U.S. Patent 4786567, 1988. (9) Li, L.; Kim, S.; Wang, W.; Vijayakumar, M.; Nie, Z.; Chen, B.; Zhang, J.; Xia, G.; Hu, J.; Graff, G.; Liu, J.; Yang, Z. Adv. Energy Mater. 2011, 1 (3), 394−400. (10) Kear, G.; Shah, A. A.; Walsh, F. C. Int. J. Energy Res. 2011, 36, 1105−1120. (11) Kamath, H.; Rajagopalan, S.; Zwillenberg, M. Vanadium Redox Flow Batteries: An In-Depth Analysis; Electric Power Research Institue: Palo Alto, CA, 2007. (12) Sun, B. T.; Skyllas-Kazacos, M. Electrochim. Acta 1991, 36 (3− 4), 513−517. (13) Sun, B.; Skyllas-Kazacos, M. Electrochim. Acta 1992, 37 (13), 2459−2465. (14) Flox, C.; Rubio-Garcia, J.; Nafria, R.; Zamani, R.; Skoumal, M.; Andreu, T.; Arbiol, J.; Cabot, A.; Ramon Morante, J. Carbon 2012, 50 (6), 2372−2374. (15) Tsai, H.-M.; Yang, S.-J.; Ma, C.-C. M.; Xie, X. Electrochim. Acta 2012, 77, 232−236. (16) Wang, W. H.; Wang, X. D. Electrochim. Acta 2007, 52 (24), 6755−6762. 1335

dx.doi.org/10.1021/nl400223v | Nano Lett. 2013, 13, 1330−1335