Nanotube Composite Catalyst

Sep 11, 2013 - Carbon nanofiber/nanotube (CNF/CNT) composite catalysts grown on carbon felt (CF), prepared from a simple way involving the thermal dec...
1 downloads 15 Views 4MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Synergistic Effect of Carbon Nanofiber/Nanotube Composite Catalyst on Carbon Felt Electrode for High-Performance All-Vanadium Redox Flow Battery Minjoon Park,† Yang-jae Jung,† Jungyun Kim,‡ Ho il Lee,‡ and Jeaphil Cho*,† †

School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), 689-798, Ulsan, South Korea ‡ Energy Research Department, Industrial Research Institute, R&D Division, Hyundai Heavy Industries Co., Ltd., 446-912, Yongin, Gyeonggi, South Korea S Supporting Information *

ABSTRACT: Carbon nanofiber/nanotube (CNF/CNT) composite catalysts grown on carbon felt (CF), prepared from a simple way involving the thermal decomposition of acetylene gas over Ni catalysts, are studied as electrode materials in a vanadium redox flow battery. The electrode with the composite catalyst prepared at 700 °C (denoted as CNF/ CNT-700) demonstrates the best electrocatalytic properties toward the V2+/V3+ and VO2+/VO2+ redox couples among the samples prepared at 500, 600, 700, and 800 °C. Moreover, this composite electrode in the full cell exhibits substantially improved discharge capacity and energy efficiency by ∼64% and by ∼25% at 40 mA·cm−2 and 100 mA·cm−2, respectively, compared to untreated CF electrode. This outstanding performance is due to the enhanced surface defect sites of exposed edge plane in CNF and a fast electron transfer rate of in-plane side wall of the CNT. KEYWORDS: Energy storage, redox flow battery, electrode, carbon nanofiber, carbon nanotube, catalyst

T

of the electrodes and alternative electrode materials have been investigated to address these inherent problems.13 To a large extent, there are two categories of modification methods, which include metal-based (noble metal14−19 and metal oxide20,21) and carbon-based materials11,12,22−29 for increasing electrocatalytic activity toward vanadium redox pairs. The precious metal-based materials have shown higher catalytic activity for vanadium ions, but they are not cost-effective or susceptible to hydrogen evolution.15 Therefore, efforts to develop the advanced electrodes having a reasonable cost and stability in acid conditions are concentrated on carbon-based materials.12,30 To date, one-dimensional (1D) carbon nanomaterials have been widely investigated as electrode materials or catalysts for electrochemical activity in the VRFB system to enhance battery performance (summarized in Table S1 of the Supporting Information). The class of 1D carbon nanomaterials can be further subdivided into the class of carbon nanofibers (CNF) and carbon nanotubes (CNT), in which the distinct difference lies in the configuration of the underlying graphene planes created by the alignment of carbon atoms.31 CNF have a

he energy storage system (ESS) enables electricity to be stored efficiently in chemicals and releases it according to demand.1−5 Among the various ESS systems, the vanadium redox flow battery (VRFB) studied by Maria Skyllas-Kazacos in the 1980s has been regarded as an efficient and reliable largescale energy storage device with integration of renewable resources.6 With the external electrolyte liquid reservoir containing electroactive materials, the flow batteries have attractive features like flexible design by controlling two key modular components (i.e., stacks and reactant tanks) compared to conventional secondary batteries where energy-bearing species are stored in electrode structures.7,8 More interestingly, when charging and discharging, VRFB employs the same elements, vanadium, in both electrolytes, and so is less affected by membrane cross contamination resulting in a long cycle life, high energy efficiency, and stability (see Figure S1 of the Supporting Information).9,10 The carbon-based electrodes of VRFB play a key role to provide active sites for vanadium redox reaction just on the surface, where the reaction kinetics and the efficiency of whole cell are determined.11 However, their poor kinetic reversibility and electrochemical activity, especially on the complex positive redox reaction due to the rearrangement of the coordination structures of the VO2+/VO2+ redox couple, limits their extensive application.12 The intensive modification methods © XXXX American Chemical Society

Received: July 12, 2013 Revised: August 26, 2013

A

dx.doi.org/10.1021/nl402566s | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 1. Upper: Schematic view for the synthesis of CNF/CNT grown on carbon felt surface on Ni nanoparticle seeds via C2H2 gas decomposition above 500 °C. Lower: SEM image of (a) untreated CF, (b) CNF/CNT grown CF surface prepared at 700 °C. HR-TEM images of (c) the CNF/ CNT composite detached from CNF/CNT-700 with magnified image from rectangle in (c), showing carbon nanofiber (CNF) and multiwalled carbon nanotube (CNT) side wall (white line: graphene layer orientation), respectively, and (d) the coexistence structure in CNF/CNT-700 sample.

500 to 800 °C, respectively to obtain CNF/CNT composite, labeled CNF/CNT-T (T = 500, 600, 700, and 800 °C). Figure 1a exhibits the scanning electron microscopy (SEM) image of the untreated CF with the smooth but slightly porous surface, facilitating the metal nanoparticle impregnation sites. Ni catalysts with a size of diameters of 10−80 nm was successfully impregnated on the carbon felt surface (Supporting Information, Figure S2a and b). Energy dispersive X-ray spectroscopy (EDXS) mapping of the CF showed that nickel was uniformly distributed on the CF, loading of about 0.92 wt % (see Supporting Information, Figure S2c). A dense CNF/CNT coverage on the CF was obtained because of well-dispersed Ni catalyst (Figure 1b). In addition, CNF/CNT growth over nickel catalysts is beneficial for good adhesion to carbon felt surface. Although the morphology and diameter of 1D carbon nanostructures is directly linked to the size and nature of catalyst seed, it could be determined indirectly by controlling growth temperature.34 Briefly, at 500 °C, nucleation of the acetylene is slow, and carbon atoms have reached the entire Ni catalyst interface via diffusion and nucleation of CNFs with no hollow core (Figure S3a). At 700 °C, the nucleation starts before the entire metal interface has been saturated with carbon atoms; for this reason, the nucleation begins at the interface among the Ni metal/acetylene gas, which leads to the CNT formation (Figure 1). However, at 800 °C, the graphene wall of

unique morphology in that the graphene layer is tilted against fiber axis, resulting in exposed edge planes toward exterior surface providing active sites for ionic adsorption and chemical bonding directly with the reactants.32 CNT, on the other hand, are comprised of concentric cylinders of a graphene sheet consisting of relatively inert basal planes, which has unique properties, such as good electric conductivity and chemical resistance to acids and bases.33 In these regards, simultaneous usage of both CNT and CNF could provide the best electrocatalytic performance of the cell. To our knowledge, the use of the CNT/CNF composite as electrode catalyst with the advantages of both sides has not been reported in the VRFB system yet. Here, we introduce a CNF/CNT composite catalyst on the carbon felt surface for high performance VRFB. This composite catalyst facilitates the electron and mass transfer kinetics resulting in increased battery efficiency at a high rate by lowering the overpotential for vanadium redox reaction. The CNF/CNT grown on CF electrodes were synthesized in a tube furnace using C2H2 gas at different temperatures ranging from 500 to 800 °C (Figure 1). The CF immersed in nickel nitrate solution was dried at 100 °C for 3 h. The dried CF was calcined under a mixture of hydrogen and Ar gas (10% H2 balanced Ar gas) at 600 °C for 3 h to obtain the reduced Ni catalysts. Finally, CF with reduced Ni was fired at the range of B

dx.doi.org/10.1021/nl402566s | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 2. Ex situ X-ray photoelectron spectroscopy (XPS) analysis of the untreated CF and CNF/CNT-T smaples. (a) High-resolution spectra of C1s a O1s peaks, (b) chemical composition ratio of functional groups from curve fitting of C1s and O1s XPS spectra.

Figure 3. Cyclic voltammograms with untreated CF and CNF/CNT-T electrodes prepared at different temperatures (a) V2+/V3+ redox couple and (b) VO2+/VO2+ redox couples in 0.1 M VOSO4 + 2 M H2SO4 electrolyte at the scan rate of 5 mV s−1, respectively, and (c) CNFCNT-700 electrode in 0.1 M VOSO4 + 2 M H2SO4 electrolyte solution at different scan rate. Inset: plot of the anodic peak current (Ipa) versus the square root of the potential scan rate. (d) Plot of peak current density vs square root of scan rate for VO2+/VO2+ redox couples.

CNT could become collapsed due to increasing reaction energy and a too-fast C2H2 decomposition rate (Figure S3C). As a result, we found that 700 °C is the intermediate temperature to obtain the well-developed CNF/CNT composite. The possible reason for the growth of CNF/CNT composite in this study could be the temperature-induced effect, resulting from the

difference of carbon source decomposition and diffusion rate at various temperatures.35,36 As shown in Figure 1b, the CF consisting of an interconnected matrix of CNF/CNT with diameter range from 10 to 80 nm have been successfully prepared. Figure 1c clearly depicts that this composite obtained at 700 °C has a high aspect ratio with the length of a few C

dx.doi.org/10.1021/nl402566s | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Table 1. Electrochemical Propertiesa Obtained from Cyclic Voltammetry Results for Untreated and CNF/CNT-T Electrodes at a Scan Rate of 5 mV s−1 negative half-cell electrode untreated CF CNF/CNT-500 CNF/CNT-600 CNF/CNT-700 CNF/CNT-800 a

Ipa (mA) 46.67 49.61 53.37 32.65

Ipc (mA) −78.17 −72.79 −80.35 −72.79

Vpa (V) −0.18 −0.17 −0.22 −0.22

Vpc (V) −0.88 −0.9 −0.83 −0.9

positive half-cell Ipc/Ipa

ΔE (mV)

Ipa (mA)

Ipc (mA)

Vpa (V)

Vpc (V)

Ipc/Ipa

ΔE (mV)

702.56 725.86 615.07 682.2

59.53 66.00 79.32 87.96 67.40

−32.10 −52.21 −65.29 −74.17 −52.91

1.13 1.07 1.03 1.02 1.04

0.75 0.83 0.88 0.89 0.86

1.85 1.26 1.21 1.19 1.27

375.00 246.61 152.01 136.12 186.58

0.6 0.68 0.66 0.45

pa: anodic current peak, pc: cathodic current peak.

counter, and reference electrodes, respectively. As for the negative electrode (Figure 3a), the anodic and cathodic peaks associated with the V2+/V3+ appear at −0.2 to −0.3 V and −0.8 to −0.9 V (vs Ag/AgCl), respectively. This result demonstrates that the CNF/CNT-T electrode exhibited more pronounced redox peaks than the untreated CF, implying that the poor performance of the V2+/V3+ reduction reaction in negative halfcell was substantially improved by suppressing undesired hydrogen evolution known as a side reaction in the VRFB system. Figure 3b presents the CV curve of the positive electrode reaction corresponding to VO2+/VO2+ redox couples. The onset potential of both redox peaks showed a negative shift to 1.02 V (vs Ag/AgCl) at the anodic reaction corresponding to VO2+ → VO2+. Similarly, in the cathodic reaction, the onset potential associated with VO2+ → VO2+ process exhibited a positive shift to 0.89 V (vs Ag/AgCl). The improved onset potential for both anodic and cathodic processes would be favorable for electron transfer kinetics and beneficial for increasing energy storage efficiency due to lower applied voltage for VRFB, which means the redox reaction of vanadium ions can react more easily on the CNT/CNT-T electrode than untreated one.15 The anodic and cathodic peak current density for positive redox couple reaction in CNF/CNT-700 electrode compared with the untreated one was significantly increased to Ipa = 87.96 and Ipc = −74.17 mA cm−2, suggesting that the electron transfer kinetics for positive redox couple (VO2+/ VO2+) reaction was greatly enhanced on CNF/CNT-700 electrode. Furthermore, the electrochemical impedance spectroscopy (EIS) analysis agrees with the results from CV (see Figure S8c). The catalytic activity of CNF/CNT-700 in terms of redox peak current density and onset potential shows best performance. The reversibility of redox reaction on the electrode surface can be obtained by calculating the ratio of the peak current density (Ipa/Ipc) and the oxidation and reduction peak potential separation (ΔE = Vpa − Vpc).11 Among the samples, the value of the Ipa/Ipc on CNF/CNT-700 at positive electrode was the lowest one indicating that it was close to the value in reversible redox reaction.11 Figure 3c displays the cyclic voltammogram of VO2+/VO2+ redox couples in the CNF/CNT-700 electrode under different scan rate (CV curves of CNF/CNT-T electrode at other temperatures are also shown in Figure S5), in which the value of the Ipa/Ipc was almost constant throughout whole scan rate range due to the increased electron transfer properties at graphene plane of 1D carbon materials. Furthermore, the peak potential separation associated with polarization was decreased to 136 from 370 mV, resulting in enhanced reversibility for vanadium redox couples. All of these electrochemical parameters were summarized at Table 1. Mass transfer properties can be estimated by plotting the peak current density versus the square root of the scan rate

micrometers and diameter of a few tens of nanometers. More interestingly, simultaneous formation of CNF and CNT with different graphene sheet orientations is observed in magnified image from rectangles in Figure 1c, in which white lines indicate the different graphene side wall orientation. Other samples, however, exhibit different morphological characteristics (Figure S3). Figure 1d also reveals the copresence of CNF and CNT in as-prepared sample at 700 °C with typical cupstack like carbon nanofiber and the multiwalled carbon nanotube separated by a distance ca. 0.32 nm. Further, the exposed edges of graphitic planes (yellow arrow) were mainly occupied by carbon edge atoms, which might be act as active site for vanadium redox couples. Thus, carbon atoms at the outmost edge in CNF would have a higher reactivity as compared to carbon atoms in untreated CF electrode. Ex situ X-ray photoelectron spectroscopy (XPS) analysis was conducted to elucidate the chemical structures of the resulting samples prepared at different growth temperatures (for full range spectra, see Figure S4). All spectra were calibrated using the C1s peak of carbon present at 284.2 eV. For further analysis, the C1s spectra can be deconvoluted into five peaks at the binding energies of ∼289.8, 288.3, ∼286.4, ∼285.1, and 284.2 eV, corresponding to COO− (C1), CC (C2), C−O (C3), CC(C4), and −COO(C5), respectively (Figure 2a).26 Remarkably, the shape of these five peaks significantly changed with increasing annealing temperature, indicating different amounts of carbon bonding configurations were formed. It is worthy to note that, with the increase of temperature from 500 to 700 °C, the sp2 hybridized carbon peak at ∼284 eV became dominant, implying that more CC bond was incorporated into the carbon network of CNF/CNT-T electrodes. Simultaneously, the contents of sp3 carbon largely decreased possibly due to sufficient energy supplying to form sp2 carbon walls of CNF/CNT by effectively decomposing a carbon source.37 Note that the content of graphitic carbon (sp2) peak is the highest value at 700 °C. Since the oxygen functional group is directly related to the active site of the vanadium ion, the O1s spectra were also deconvoluted into four peaks at binding energies of ∼530.6, ∼531.8, ∼532.9, and ∼534 eV, corresponding to CO(O1), C−OH(O2), C−CO(O3), and carbonate(O4), respectively (Figure 2a).38 The fact that the contents of the CO functional group were gradually increased with the growth temperature (Figure 2b) indicates that the CNF/CNT-T electrodes with edge-rich sp2 hybridized carbon wall might be favorable for vanadium redox couple reaction with high activity. Cyclic voltammetry (CV) was conducted to assess the electrochemical activity of CNF/CNT-T electrode toward the V2+/V3+ and VO2+/VO2+ redox couple in each electrolyte with a three-electrode cell at room temperature, where modified CF electrode, platinum wire, and Ag/AgCl was used as working, D

dx.doi.org/10.1021/nl402566s | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 4. Electrochemical performance of VRFB employing CNF/CNT-T electrodes in flow-type single cell at room temperature: (a) Charge− discharge voltage profiles of untreated and as-prepared electrode at 40 mA·cm−2. (b) Rate performance with increasing the current rate from 40 to 100 mA·cm−2.

VRFB flow-type single cell having an electrode area of 5 cm2. All flow cells were fabricated identically with Nafion 117 as a membrane separator, graphite bipolar plate and copper current collector, in which 20 mL of 2 M V(III) + 3 M H2SO4 and 2 M V(IV) + 3 M H2SO4 solution were pumped into the positive and negative electrode (see Figure S9). The operating voltage was fixed at 0.8−1.6 V to avoid side reaction such as O2 and H2 evolution.15 Figure 4a shows the voltage profiles of the untreated and CNF/CNT-700 samples between 0.8 and 1.6 V at a current rate of 40 mA·cm−2 during the first cycle, showing that remarkably decreased overpotential about 109 and 138 mV in both charge and discharge processes, which could be ascribed to the effect of CNF/CNT composite having large active sites for vanadium redox species and high conductivity for electrons transfer resulting in lower charge and higher discharge voltage. For given condition, the cell employing CNF/CNT-700 catalyst exhibited a higher capacity of 24.1 Ah·L−1 at first cycle than that of the untreated one showing a quickly decrease of capacity less than 15 Ah·L−1. The increased capacity around 64% of CNF/CNT-700 sample at the cycling test could be attributed to large active sites and improved mass transfer rate as a result of CNF/CNT coating on the CF surface. To investigate the charge−discharge rate capability and capacity recovery, the cells were charged to 1.6 V at a rate of 40 to 100 mA cm−2 then discharged to 0.8 V at the same rate. As shown in Figure 4b, as increasing the current density, the capacity decreases are observed at all samples, which is usually caused by larger charge/discharge overpotential at a fast current rate.15 The CNF/CNT-700 electrode showed a higher rate performance at 100 mA cm−2 than an untreated one. However, the rate capability of other samples demonstrated was not as good as that grown at 700 °C, which might be caused by small amount of active sites for vanadium ions. A capacity recovery test at a 40 mA·cm−2 rate after 100 mA·cm−2 cycling was investigated for the untreated and CNF/CNT-T electrode. The capacity was gradually decreased at the untreated one due to electrolyte decomposition at the higher charge/discharge overpotential, whereas almost 100% of initial capacity was retained at CNF/CNT-700 sample, demonstrating high resistance to strong acid. These outstanding properties indicated that the CNF/CNT composites were successfully maintained on the carbon felt surface after 30 cycles, allowing for higher discharge rate capability and capacity retention. The Coulomb efficiency (CE), voltage efficiency (VE), and energy efficiency (EE) value of CNF/CNT-700 showed the best

from the Randles−Sevcik equation.15,39 As shown in Figure 3d, the oxidation and reduction peak current density of the positive redox couple on as-prepared electrodes are nearly proportional to the square root of the scan rate, implying that a diffusion process controls the redox reaction behavior of VO2+/VO2+ on the electrode. The slope of as-prepared electrode was larger than that of untreated CF electrode, suggesting a faster mass transfer process on the surface of CNF/CNT-T. Based on the slope of each electrodes, the electrochemical surface area (ECSA) was obtained in the order (m2): CNF/CNT-700 (1.4) > CNF/CNT-600 (1.3) > CNF/CNT-800 (1.1) > CNF/CNT500 (1.0) >untreated CF (0.8). On the other hand, the specific surface area obtained by the BET method is not proportional to the current density. The specific area of CNF/CNT samples follows the order (m2 g−1): CNF/CNT-500 (9.8) > CNF/ CNT-700 (6.8) > CNF/CNT-600 (4.5) > CNF/CNT-800 (4.1), but the untreated CF shows a specific area of ca. 0.4 m2 g−1. Comparing the current density with specific surface area of each sample, it could be considered that the BET surface area has a minimal relationship with the vanadium redox reaction. Accordingly, a key factor affecting on the high mass transfer rate might be ascribed not to the specific surface area but to the electrochemical active surface area. Consequently, in the twophase redox reaction between electrode and electrolyte interface, the larger effective surface area resulting from the CNF/CNT catalyst and its functional groups ensures the higher rate of mass transfer. Moreover, the slope of CNF/ CNT-700 electrode was increased about 1.65 and 2.86 times higher compared with the untreated CF both anodic and cathodic processes, respectively. The relatively higher mass transfer rate of CNF/CNT-700 electrode might be ascribed to its abundant exposed carbon edge atom as active sites for vanadium ion adsorption. From the above results, the higher catalytic activity of the vanadium redox couple reaction on CNF/CNT-T electrode could be attributed to both improved electron transfer kinetics and fast mass transfer rate. The enhanced mass transfer reaction was probably owing to the outer edge sites of CNF/CNT composite. As for electron transfer rate, the lower electron transfer resistance of asprepared electrodes could be partially contributed to the conductivity of the well-developed graphene wall of CNF/ CNT. This phenomenon is believed to be originated from the balance between the electrochemical activity at edge sites of CNF and electrical conductivity of CNT. The electrochemical performance of CNF/CNT-T electrodes was evaluated by the charge−discharge experiments in E

dx.doi.org/10.1021/nl402566s | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

utilizing nickel nanoparticles as catalysts. Surprisingly, the asprepared CNF/CNT-700 electrode demonstrated better capacity retention and excellent rate capability up to 100 mA· cm−2 with significantly improved electron transfer and lower polarization. Such results were due to the formation of large edge plane defect of CNF walls and fast electron transfer at basal plane of CNT walls. The CNF/CNT composite on CF electrode is promising for enhancement of the battery efficiency and rate capability in the VRFB system.

performance among the samples under current density range from 40 to 100 mA cm−2 (see Figure S6). To elucidate the mechanism of vanadium redox reaction on the CNF/CNT-T electrodes, the electrocatalytic activities of asprepared samples at different temperatures have been investigated. As depicted in XPS analysis, the increase in carbon content is accompanied by a decline in oxygen content. It is note, however, that the decreased oxygen content with the temperature does not lead to a commensurate drop in redox reaction and cell performance. To confirm the effect of oxygen contents in detail, similar experiments have been carried out with fully oxidized carbon felt (OCF) electrode thermally activated in air at 500 °C for 5 h. The experimental results between OCF and CNF/CNT-700 samples in VRFB system are described in the Supporting Information. As shown in Figure S7, the ratio of oxygen atomic content highly increased to 0.22 compared with untreated one. In addition, the value of O/C content ratio on OCF was almost tripled to CNF/CNT700 electrode, implying that increased oxygen atom could be expected to positively influence on redox reaction. However, the overall cell performance on OCF electrode was not greatly improved (Figure S8), suggesting that the vanadium redox reaction is not dependent on the total amount of incorporated oxygen but more on how the oxygen is incorporated into carbon nanostructures. Indeed, catalytic activities for vanadium redox couple was proportional to the contents of CO and C−OH functional groups, providing active sites for adsorption of active materials. Although these oxygen functional groups were substantially increased in CNF/CNT-800 samples, poor electrochemical performance was observed in the CV and flow cell tests, probably due to collapsed graphene side walls at an increased injection temperature of acetylene (Figure S3). Moreover, the reduced charge/discharge overpotentials at a high current rate and enhanced mass transfer rate on CNF/ CNT coated electrodes having large amounts of edge plane defects were verified by the half-cell and flow cell tests. It is generally believed that these defect sites may serve as the active sites for the various redox species.12 In our sample of CNF/ CNT-700, the ratio of the sp2 hybridized carbon defects was very high, suggesting that it retained large active sites for vanadium redox couples confirmed by the HR-TEM. Furthermore, the polarization during charge/discharge processes was much smaller on CNF/CNT-700 electrode than that on untreated one, allowing for enhanced electrochemical reversibility and reduced onset potential caused by lower kinetic activation energy (Figure 4a). Because of the nature of chemical bonding in graphene plane in CNF and CNT, both materials show different electrochemical properties. The mass transfer reaction for the vanadium ion at the edge plane is faster than at the basal plane, whereas the electron transfer is more favorable at the basal plane. Accordingly, it is reasonable to conclude that balance between the surface defects in the form of exposed edge plane in CNF walls and basal plane surface of CNT walls for fast electron transfer plays a major role in the vanadium redox reaction, showing substantially improved catalytic activity and cell performance. Conclusion. In summary, we investigated the synergistic effect of the carbon nanofiber (CNF)/carbon nanotube (CNT) composite catalysts used as VRFB electrodes. CNF/CNT composites with enhanced electron transfer and well-developed edge plane active sites for vanadium redox couple were synthesized on the carbon felt via a simple acetylene (C2H2) vapor deposition method at different growth temperatures

■

ASSOCIATED CONTENT

S Supporting Information *

Schematic illustration of VRFB, reference table; SEM images, XRD analysis, EDXS patterns, and electrochemical cell performance data. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS

■

REFERENCES

This research was supported by MSIP (Ministry of Science, ICT&Future Planning), Korea, under the C-ITRC (Convergence Information Technology Research Center) support program (NIPA-2013-H0301-13-1009) supervised by the NIPA (National IT Industry Promotion Agency).

(1) Dunn, B.; Kamath, H.; Tarascon, J. M. Science 2011, 334 (6058), 928−35. (2) Halls, J. E.; Hawthornthwaite, A.; Hepworth, R. J.; Roberts, N. A.; Wright, K. J.; Zhou, Y.; Haswell, S. J.; Haywood, S. K.; Kelly, S. M.; Lawrence, N. S.; Wadhawan, J. D. Energy Environ. Sci. 2013, 6 (3), 1026. (3) Leung, P.; Li, X.; Ponce de León, C.; Berlouis, L.; Low, C. T. J.; Walsh, F. C. RSC Adv. 2012, 2 (27), 10125. (4) Ding, C.; Zhang, H.; Li, X.; Liu, T.; Xing, F. J. Phys. Chem. Lett. 2013, 4 (8), 1281−1294. (5) Yang, Z.; Zhang, J.; Kintner-Meyer, M. C.; Lu, X.; Choi, D.; Lemmon, J. P.; Liu, J. Chem. Rev. 2011, 111 (5), 3577−613. (6) Skyllas-Kazacos, M. J. Electrochem. Soc. 1986, 133 (5), 1057. (7) Wang, W.; Kim, S.; Chen, B.; Nie, Z.; Zhang, J.; Xia, G.-G.; Li, L.; Yang, Z. Energy Environ. Sci. 2011, 4 (10), 4068. (8) 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. (9) Zaffou, R.; Li, W. N.; Perry, M. L. Vanadium Redox Flow Batteries for Electrical Energy Storage: Challenges and Opportunities. In Polymers for Energy Storage and Delivery: Polyelectrolytes for Batteries and Fuel Cells; ACS Symposium Series, Vol. 1096; American Chemical Society: Washington, DC, 2012; pp 107−127. (10) Shin, S.-H.; Yun, S.-H.; Moon, S.-H. RSC Adv. 2013, 3 (24), 9095−9116. (11) Han, P.; Yue, Y.; Liu, Z.; Xu, W.; Zhang, L.; Xu, H.; Dong, S.; Cui, G. Energy Environ. Sci. 2011, 4 (11), 4710. (12) Jin, J.; Fu, X.; Liu, Q.; Liu, Y.; Wei, Z.; Niu, K.; Zhang, J. ACS Nano 2013, 7 (6), 4764−4773. (13) Li, W.; Liu, J.; Yan, C. Carbon 2013, 55, 313−320. (14) Tseng, T. M.; Huang, R. H.; Huang, C. Y.; Hsueh, K. L.; Shieu, F. S. J. Electrochem. Soc. 2013, 160 (4), A690−A696.

F

dx.doi.org/10.1021/nl402566s | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

(15) Li, B.; Gu, M.; Nie, Z.; Shao, Y.; Luo, Q.; Wei, X.; Li, X.; Xiao, J.; Wang, C.; Sprenkle, V.; Wang, W. Nano Lett. 2013, 13 (3), 1330− 1335. (16) Tsai, H. M.; Yang, S. J.; Ma, C. C. M.; Xie, X. F. Electrochim. Acta 2012, 77, 232−236. (17) Huang, R. H.; Sun, C. H.; Tseng, T. m.; Chao, W. k.; Hsueh, K. L.; Shieu, F. S. J. Electrochem. Soc. 2012, 159 (10), A1579−A1586. (18) González, Z.; Sánchez, A.; Blanco, C.; Granda, M.; Menéndez, R.; Santamaría, R. Electrochem. Commun. 2011, 13 (12), 1379−1382. (19) Wang, W. H.; Wang, X. D. Electrochim. Acta 2007, 52 (24), 6755−6762. (20) Kim, K. J.; Park, M. S.; Kim, J. H.; Hwang, U.; Lee, N. J.; Jeong, G.; Kim, Y. J. Chem. Commun. (Cambridge) 2012, 48 (44), 5455−7. (21) Yao, C. A.; Zhang, H. M.; Liu, T.; Li, X. F.; Liu, Z. H. J. Power Sources 2012, 218, 455−461. (22) Li, W.; Liu, J.; Yan, C. J. Solid State Electrochem. 2013, in press. (23) Flox, C.; Skoumal, M.; Rubio-Garcia, J.; Andreu, T.; Morante, J. R. Appl. Energy 2013, 109, 344−351. (24) Li, W. Y.; Liu, J. G.; Yan, C. W. Electrochim. Acta 2012, 79, 102− 108. (25) González, Z.; Vizireanu, S.; Dinescu, G.; Blanco, C.; Santamaría, R. Nano Energy 2012, 1, 833−839. (26) Li, W.; Liu, J.; Yan, C. Carbon 2011, 49 (11), 3463−3470. (27) Han, P.; Wang, H.; Liu, Z.; Chen, X.; Ma, W.; Yao, J.; Zhu, Y.; Cui, G. Carbon 2011, 49 (2), 693−700. (28) Yue, L.; Li, W. S.; Sun, F. Q.; Zhao, L. Z.; Xing, L. D. Carbon 2010, 48 (11), 3079−3090. (29) Wei, G. J.; Jia, C. K.; Liu, J. G.; Yan, C. W. J. Power Sources 2012, 220, 185−192. (30) Wang, S.; Zhao, X.; Cochell, T.; Manthiram, A. J. Phys. Chem. Lett. 2012, 2164−2167. (31) Iijima, S.; Ichihashi, T. Nature 1993, 363 (6430), 603−605. (32) Huang, J.; Liu, Y.; You, T. Anal. Methods 2010, 2 (3), 202. (33) Zhu, J.; Holmen, A.; Chen, D. ChemCatChem 2013, 5 (2), 378− 401. (34) Martínez-Latorre, L.; Armenise, S.; Garcia-Bordejé, E. Carbon 2010, 48 (7), 2047−2056. (35) De Jong, K. P.; Geus, J. W. Catal. Rev. 2000, 42 (4), 481−510. (36) Chen, X.-W.; Su, D. S.; Hamid, S. B. A.; Schlögl, R. Carbon 2007, 45 (4), 895−898. (37) Vollebregt, S.; Ishihara, R.; Tichelaar, F. D.; Hou, Y.; Beenakker, C. I. M. Carbon 2012, 50 (10), 3542−3554. (38) Li, W.; Liu, J.; Yan, C. Electrochim. Acta 2011, 56 (14), 5290− 5294. (39) Flox, C.; Rubio-Garcia, J.; Nafria, R.; Zamani, R.; Skoumal, M.; Andreu, T.; Arbiol, J.; Cabot, A.; Morante, J. R. Carbon 2012, 50 (6), 2372−2374.

G

dx.doi.org/10.1021/nl402566s | Nano Lett. XXXX, XXX, XXX−XXX