www.advmat.de
COMMUNICATION
www.MaterialsViews.com
All Metal Nitrides Solid-State Asymmetric Supercapacitors Changrong Zhu, Peihua Yang, Dongliang Chao, Xingli Wang, Xiao Zhang, Shi Chen, Beng Kang Tay, Hui Huang, Hua Zhang, Wenjie Mai, and Hong Jin Fan* Supercapacitor is relatively new technology compared to batteries but is regarded a highly promising alternative device for electrochemical energy storage, due to its capability of fast energy delivery (up to tens of seconds) and ultralong cycle life (typically 105 cycles).[1–4] However, the bottleneck of supercapacitor is the relatively low energy density compared to batteries, particularly when carbon materials are utilized as the electric double-layer capacitive electrode.[5–7] Asymmetric supercapacitors (ASCs) can improve the energy density by providing 1 increased voltage of the full devices according to E = CV 2.[8–12] 2 Recently, research on ASCs has been focused mostly on metal oxides because of their high pseudocapacitance through fast reversible redox reactions with ions from electrolyte. However, metal oxides usually suffer from low electrical conductivity and unreliable stability during long cycles. Metal nitrides, in the contrary, show much improved sustainability and superb electrical conductivity (4000−55 500 S cm−1),[13,14] and thus become attractive as SC electrode materials. Among all the reported metal nitrides in energy storage, TiN and Fe2N received particular attention due to their high capacitance (Fe2N as lithium ion battery anode: ≈900 mAh g−1; TiN as supercapacitor cathode: ≈150 F g−1) and electrical conductivity.[15,16] Various types of TiN nanostructures have been employed as supercapacitive electrode such as nanoparticle,[13] nanotube,[17–19] mesoporous microsphere,[20] and nanosheets.[21] However, TiN was found to be easily oxidized in aqueous solutions ⎛ TiN + 2H O → TiO + 1 N + 4H+ + 4e − ,2TiN + 2O → 2TiO + N ⎞ . [ 22 ] 2 2 2 2 2 2 ⎝ ⎠ 2
Therefore, to circumvent the oxidation problem of TiN,
C. Zhu, D. Chao, Dr. S. Chen, Prof. H. J. Fan School of Physical and Mathematical Sciences Nanyang Technological University 637371 Singapore, Singapore E-mail: [email protected] C. Zhu, Dr. H. Huang Singapore Institute of Manufacturing Technology 71 Nanyang Drive, 638075 Singapore, Singapore P. Yang, Prof. W. Mai Department of Physics and Siyuan Laboratory Jinan University Guangzhou 510632, P. R. China X. Wang, Prof. B. K. Tay School of Electrical and Electronic Engineering Nanyang Technological University 50 Nanyang Avenue, 639798 Singapore, Singapore X. Zhang, Prof. H. Zhang School of Materials Science and Engineering Nanyang Technological University 639798, Singapore
DOI: 10.1002/adma.201501838
4566
wileyonlinelibrary.com
researchers combined TiN with more stable materials. They include carbon coating,[22–24] carbon nanotube encapsulation,[25,26] graphene wrapping,[27–30] and forming core-shell structures with TiO2,[31] VN,[32] and MnO2,[33] as well as decoration with polymers such as polypyrrole,[34] polyaniline,[35,36] and PEDOT.[37] Those methods showed high efficiency in enhancing TiN cycling and rate properties. Fe2N has been reported as anode for lithium-ion batteries[38,39] in conjunction with carbonrelated materials. But, the absence of evident charge–discharge voltage plateaus[38] indicates its capacitive property and suitability as the supercapacitor electrode material. Due to the large surface area, high conductivity, and lightweight property, graphene becomes an effective substrate for SCs electrode.[40] Compared to the stacked graphene layers obtained from solution synthesis, vertically aligned graphene nanosheets (GNS) on current collectors (such as carbon fiber) are more advantageous as the later allows a maximum usage of the high surface area. This increased surface area will lead to a high capacitance since a supercapacitor device essentially relies on the available active surface sites for charge absorption (for double-layer capacitor) or redox reactions (for pseudocapacitor). Recently, such GNS substrates have been applied to SnO2 and ZnO-based electrodes to improve LIB performance.[41,42] In this work, we report the fabrication of nanostructured metal nitrides on vertically aligned graphene nanosheets substrate for high-power and ultrastable solid-state ASCs. Atomic layer deposition (ALD) was employed in order to assure the maximum use of the conductive GNS surfaces and have full coverage and tight connection of the active nitride material to the substrate. ALD is proven a powerful tool for nanofabrication and surface engineering of energy materials.[43] We employed solid-state electrolyte to avoid oxidation of metal nitrides.[44–46] The ASCs constructed from TiN@GNSs cathode, Fe2N@ GNSs anode, and poly(vinyl alcohol) (PVA)/LiCl solid electrolyte indeed show an irreversible capacitance of ≈58 F g−1 at 4 A g−1 which is stable up to 20 000 cycles. Moreover, the devices achieve both high volumetric energy density (≈0.55 mWh cm−3) and power density (≈220 mW cm−3) tested at a high current density of 8 A g−1. The fabrication process for both electrodes is shown schematically in Figure 1. Thin films of TiO2 and ZnO with same thicknesses of 20 nm were coated by ALD onto the GNS substrates. To obtain anode precursor, ZnO was first transferred to FeOOH via a solution reaction. Subsequently, the FeOOH@ GNS and TiO2@GNS samples were thermal annealed in a NH3 atmosphere for 1 h at 600 and 800 °C, respectively. The morphology of the samples at different fabrication stages are shown by scanning electron microscopy (SEM) images in Figure 2. The GNSs cover homogeneously the surface of carbon fibers (Figure 2a) with an overall height of ≈800 nm and are less
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, 27, 4566–4571
www.advmat.de www.MaterialsViews.com
Adv. Mater. 2015, 27, 4566–4571
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
COMMUNICATION
amonia annealing were plotted together in Figure 3g. Three typical peaks of graphite were detected in the XRD pattern of bare GNS. After the ALD of TiO2, two weaker peaks of graphene were further weaken and almost disapeared. No obvious peak of TiO2 was observed which may due to the amorphous property of ALD deposited material. As was expected, after high-temperature annealing in amonia atmosphere, three diffraction peaks at 37°, 44°, and 63° appear and correspond well to the face-centered cubic (fcc) TiN (111), (200), and (220), respectively (JC-PDS, #65-0715) which is identical to previous work.[27,48] Same characterization was conducted to the anode material. Figure 4a shows the particle morphology of Fe2N with a diameter ≈10 nm. The Fe2N nanoparticles were detached from the GNS by high-power sonication. TEM mapping of element Fe (blue, Figure 4c) and N (red, Figure 4d) corresponds well to the image profile of Fe2N particles in Figure 4b. Selected area electron diffraction (SAED) pattern (Figure 4e) showed Figure 1. Schematics of the fabrication processes of metal nitride cathode and anode materials. polycrystalline nature of the Fe2N, in which a series of well-defined rings can be assigned to various diffraction planes of (101), (211), and (400) of fcc Fe2N. HRTEM image (Figure 4f) showed the than 5 nm in thickness (Figure 2b). The smooth surface (Figure 2c) of the original ALD TiO2@GNS becomes porous clear lattice fringes with d-spacings of 2.11 and 3.45 Å, which conforms well to the crystal face of (101) and (211). To dem(Figure 2d) after the annealing process, and the porous struconstrate the transformation process of the anode material, the ture is expected to enlarge the specific surface area leading to XRD spectrum of sample at different fabrication stages was increased capacitance. Regarding the anode, after the ZnO@ provided in Figure 4g. Through the comparison, it is clear that GNS (Figure 2e, inset) was dipped in 0.5 M Fe(NO3)3 solupure phase of Fe2N was successfully achieved after the transtion for 2 h, the thin film becomes thinner with a thickness ≈10 nm (Figure 2e), which is FeOOH@GNS according to the formation reaction of amonia annealing. The related diffraction previous study.[47] After the ammonia annealing, the FeOOH peaks of 43°, 56°, 68°, and 75° were indexed to fcc Fe2N latthin film is converted to Fe2N nanoparticles that are dispersed tice of (211), (212), (400), and (213) (JC-PDS, #06-0656), as was reported in previous work.[39,49] uniformly on the surface of GNS (Figure 2f) with an average diameter ≈10 nm. Transmission electron microscopy (TEM) characterizations were performed in order to reveal the detailed nanostructure and phase of the electrode materials. As cathode, TiN shows the pore size varied from 3 to 10 nm (Figure 3a). High-resolution transmission electron microscopy (HRTEM) reveals the lattice of (111) planes (d-spacing 2.45 Å) and (220) ones (d-spacing 1.48 Å) of TiN (Figure 3f), which is in agreement with the fast Fourier transfer (FFT) generated from the same area (Figure 3e). TEM mapping exhibited the consistent element distribution of Ti (Figure 3c) and N (Figure 3d) with the dark field TEM image in Figure 3b. XRD spectrum of bare GNS after oxygen plasma (see the Experimental Section in Figure 2. Morphology characterization of the 3D graphene and the final electrodes. SEM images the Supporting Information), GNS with of a) GNS grown on carbon fiber, b) GNS, c) TiO2@GNS, d) TiN@GNS, and e) FeOOH@GNS 20 nm ALD TiO2, and TiO2@GNS after (inset is the ZnO@GNS), and f) Fe2N@GNS.
4567
www.advmat.de
COMMUNICATION
www.MaterialsViews.com
Figure 3. Structure analysis of the cathode material. a) TEM image of TiN@GNS. b–d) TEM elemental mapping of TiN@GNS (scale bar, 50 nm). e) FFT pattern corresponding to the (111) lattice in (f) and (g). f) The HRTEM image of TiN. g) XRD spectra at different fabrication stages.
Both the porous TiN cathode and particle Fe2N anode were homogeneously covered/ distributed on the thin GNS and were assembled to all-metal nitride asymmetric supercapacitor devices. We consider the benefit of such electrode structures to the device performance as follows. First, TiN[13] and Fe2N[39] are high electrical conductive materials, which would be beneficial to improve the ion and electron transport rates and therefore increase the specific capacitance and rate capability. Second, GNS grown on carbon fibers provides a highly conductive and large surface area (467 m2 g−1) scaffold for the active materials. Moreover, this vertically aligned graphene effectively avoids the stacking of individual graphene sheets. By applying ALD deposition, electrode materials can fully cover the entire surface of GNS which ensure maximum usage of the substrate surface. Last, the porous TiN thin films and Fe2N nanoparticles are grafted onto GNS by chemical reaction rather than physical mixing. This assures a tight physical connection of both metal nitride materials with the GNS current collector, and also contributes to a long and stable cycling life as well as a high power density. Before discussing the electrochemical properties of the electrodes, we first consider the possible capacity contribution by
4568
wileyonlinelibrary.com
the GNS substrate both before and after the ammonia annealing. Previous work [50–53] reported that O in C O and OH functional groups on graphene surface is easy to be substituted by NH2 under high-temperature ammonia annealing conditions. Therefore, as control sample, GNS substrates with same size (0.6 × 1 cm2) were annealed both at 600 and 800 °C with other conditions kept the same (see details in the Experimental Section in the Supporting Information, the control experiment). The cyclic voltammetry (CV) curves in the Supporting Information (Figure S1) showed that the capacity contribution of GNS substrates to the electrodes is less than 0.01% (ratios of area enclosed by the CV curves) both before and after ammonia annealing at 600 or 800 °C. Therefore, in our discussion below, we do not consider the capacitance contribution by the substrates. The detailed electrochemical characterization starts with single electrodes in a standard three-electrode testing system with Pt plate as a counter electrode, Ag/AgCl (saturated KCl) as a reference electrode, TiN cathode/Fe2N anode as a working electrode, and 1 M LiCl as an electrolyte. Figure 5a,b shows the CV curves of TiN cathode (0−0.8 V vs Ag/AgCl) and Fe2N anode (−0.8−0 V vs Ag/AgCl) at
Figure 4. Structure analysis of the anode material. a) TEM image of Fe2N@GNS. b–d) TEM elemental mapping of Fe2N particles (scale bar, 50 nm). e) SAED pattern showing (211), (101), and (400) lattice planes of Fe2N. f) HRTEM image of Fe2N. g) XRD spectra at different fabrication stages.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, 27, 4566–4571
www.advmat.de www.MaterialsViews.com
10 mV s-1
0
100 mV s-1
50 mV s 20 mV s-1 10 mV s-1
5 0
Current Density (mA cm-2)
20 mV s-1
10
(c) Fe2N @ GNS
-1
-5
-10
-5
-15 -20
-10 0.2 0.4 0.6 Potential (V vs Ag/AgCl)
-0.8
0.8
-0.6 -0.4 -0.2 Potential (V vs Ag/AgCl)
(e)1.8
180 160 140 120 100 80 60 40 20
TiN Fe2N Device
0
20
40 60 80 Scan speed (mV s-1)
100
6 TiN-Fe N full device 2
100 mV s-1
4 50 mV s-1 20 mV s-1
2
10 mV s-1
0 -2 -4 -6 0.0
0.0
0.4
0.8 1.2 Voltage (V)
(f) 80 Current density (A g-1)
16
0
8
4
60 40 20
2
10 20 30 40 50 60 70 80 90 Time (second)
1.6
Charge Discharge
Full Device
1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
Capacitance (F g-1)
Capacitance (F g-1)
Current Density (mA cm-2)
50 mV s-1
5
0.0
(d)
(b)15
100 mV s-1
Voltage (V)
Current Density (mA cm-2)
TiN @ GNS
COMMUNICATION
(a)10
Ic=Id=4 A g-1
0 0
5000 10000 15000 Number of cycle
20000
Figure 5. Electrochemical properties of individual electrodes and full devices. CV curves of a) TiN@GNS and b) Fe2N@GNS in a standard three-electrode testing system (Ag/AgCl as reference electrode, Pt as counter electrode, and related electrode as working electrode, testing in 1 M LiCl electrolyte) at different scan speeds of 10, 20, 50, and 100 mV s−1. c) CV curves of solid ASC device (TiN@GNS as cathode, Fe2N@GNS as anode, and LiCl-PVA quasi-solid-gel as electrolyte) at different scan speeds of 10, 20, 50, and 100 mV s−1. d) Specific capacitance of both single electrodes and full device at different scan rates calculated from the CV curves from (a–c). e) Charge–discharge curves of an all solid full metal nitride device at different current densities of 2, 4, 8, 16 A g−1. f) Cycling performance of full device at 4 A g−1 in 20 000 cycles with different bending situations. (The mass here is based on total mass of electrode active materials. For the full device, the total mass of TiN and Fe2N is used for calculation.)
different scan rates (10, 20, 50, and 100 mV s−1) separately. Both electrodes show quasirectangular shapes at varied scan rates, demonstrating that the majority capacitance should be ascribed to capacitive behavior.[54] As a comparison, the CV curves of precursor TiO2 and FeOOH (presented in the Supporting Information, Figure S2) show much smaller curve area. The capacitances estimated from the CV curves for the oxides are about ten times lower than the corresponding TiN and Fe2N. CV curves of full device in the range from 0 to 1.6 V at different scan rates are tested in a two-electrode system with TiN as cathode, Fe2N as anode, and PVA/LiCl as electrolyte (Figure 5c). The standard rectangle shape verifies a dominating capacitive behavior.[54] Although there is numerous research work about TiN and Fe2N as electrode material for energy storage system applications, the reaction mechanisms were seldom unraveled. Thus, we proposed a possible explanation for the pseudocapacitance with XPS and Raman data as support (see detailed discussion in the Supporting Information). Gravimetric capacitances at different scan rates were calculated based on Figure 5a–c and plotted in Figure 5d. When tested in 1 M LiCl aqueous solution, the individual TiN and Fe2N electrodes show high capacitance retention ≈80% and ≈77%, respectively, when the scan speed increases from 10 to 100 mV s−1. For the solidstate ASC device, a rather stable capacitance around 60 F g−1 is obtained with less than 2% capacity drop. The improved stabilities of TiN and Fe2N in solid electrolyte should be the reason for the higher rate performance of the device compared to that of single electrodes tested in the aqueous electrolyte. Figure 5e shows the galvanostatic charge–discharge curves of
Adv. Mater. 2015, 27, 4566–4571
the full device between 0 and 1.6 V at four current densities. The linear slopes and triangle shape corroborate the dominating capacitive property of the electrodes in the neutral solid electrolyte, consistent with the CV curves. The single electrodes show similar properties in aqueous electrolyte (Figure S3a,c, Supporting Information). Moreover, both cathode and anode show high stability in 20 000 cycles at 4 A g−1. As for the full device, the cycling property was tested at bending condition from 5000th to 15 000th cycles and then recovered to natural state in last 5000 cycles (Figure 5f). The capacitance profile is nearly flat with a highly reversible capacitance ≈58 F g−1 up to 2000 cycles and the bending does not affect much the capacitance. The Coulombic efficiency within 20 000 cycles is maintained at about 100% (Figure S4a, Supporting Information). More details about the bending test are shown in Figure S4b (Supporting Information). The high capacity retention of the device over 20 000 cycles can be ascribed to the structure stability of the electrode material in the gel electrolyte. To confirm this, the morphology of both cathode and anode of the ASCs after 20 000 cycles was checked with SEM (see Figure S5 in the Supporting Information). It is found that the entire electrode structure is well maintained; both the porous structure of TiN and nanoparticle morphology of Fe2N are well kept although slight aggregation of Fe2N is observed. This confirms the high structure stability of both electrode materials. Nyquist plots of single electrodes and the full device at different cycling stages were tested in aqueous and gel electrolyte, respectively (see Figure S6 in the Supporting Information). Both the single electrode and full device are highly conductive with small
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
4569
www.advmat.de
COMMUNICATION
www.MaterialsViews.com
4570
Fe2N anode (both synthesized on highly conductive graphene nanosheets), and PVA/LiCl neutral electrolyte. The porous structure of TiN and uniform coverage of Fe2N nanoparticles on the GNS substrates are important merits that account for the outstanding cycling stability (≈98% capacity retention in 20 000 cycles) and rate capability (≈99% capacity maintenance when scan rate increases from 10 to 100 mV s−1) of the devices. Five devices are tested and show reproducible results. Typical values of the energy density of Figure 6. a) Cycling performance of five parallel devices at 4 A g−1 within 20 000 cycles. Inset figure shows the percentages of capacity retention. b) Ragone plots of quasi-solid-state TiN≈15.4 Wh kg−1 (≈0.51 mWh cm−3) and power Fe2N ASC in comparison with other PVA-based solid electrolyte SSCs and ASCs. Inset: pink and density of ≈6.4 kW kg−1 (211.4 mW cm−3) are white LEDs in parallel are lit up by two full devices in tandem. obtained. The strategy of combining ALD with ammonia annealing can be extended to other metal nitrides such as SnN, AlN, Zn3N2, and Mg3N2 for electrochemical impedance values (below 5 Ω). After 2000 and 20 000 cycles, the impedance values increase but still acceptelectrochemical energy storage applications. ably low. It is noted that both the nitride electrodes exhibit smaller impedance compared with their oxide source material. The mass loading of TiN and Fe2N is ≈0.5 and ≈0.8 mg cm−2, Supporting Information respectively (see detailed data in Table S1 and the related deterSupporting Information is available from the Wiley Online Library or mination steps in the Supporting Information). To achieve best from the author. performance, the mass ratio between the TiN cathode and Fe2N anode is about 0.69:1 (for details, see the last part in the Supporting Information). To test the repeatability of the methodology, a series of Acknowledgements devices was assembled and their specific capacitances were This research was funded by SERC Public Sector Research Funding measured and compared (Figure 6a). Within 20 000 cycles, all (Grant Number 1121202012), Agency for Science, Technology, and the five devices show very stable and nearly overlapping speResearch (A*STAR), and AcRF Tier 1 (RG104/14). H.Z. thanks the supported from MOE under AcRF Tier 2 (MOE2013-T2-1-034) and cific capacitance ≈58 F g−1. High capacity retention is obtained AcRF Tier 1 (RG 61/12, RGT18/13, and RG5/13) in Singapore. This for all five devices (ranging from 96.8 to 99.8%). The results research was also conducted by NTU-HUJ-BGU Nanomaterials for demonstrate the high reliability and repeatability of the fabricaEnergy and Water Management Programme under the Campus for tion method and the stability of the nanostructured electrode Research Excellence and Technological Enterprise (CREATE), which materials as well. To evaluate the practical application possiwas supported by the National Research Foundation, Prime Minister’s bility of the full device, we calculated the specific gravimetric Office, Singapore. The authors also thank INCUBATION ALLIANCE, Inc. capacitance based on the total mass of the device and plotted for providing the graphene nanosheets/nanoflower (GNS) substrates. the cycling curve in Figure S10 (Supporting Information). Received: April 17, 2015 Due to the light weight of the GNS@carbon fiber substrate Revised: May 20, 2015 −2 −1 (16 mg cm ), the device shows a stable capacitance of 2.5 F g Published online: July 8, 2015 as a whole. For a full device, energy density and power density are two important parameters for the evaluation of the entire device. [1] Z. Liu, J. Xu, D. Chen, G. Shen, Chem. Soc. Rev. 2015, 44, 161. The Ragone plot of the all metal nitride supercapacitor devices [2] G. Wang, L. Zhang, J. Zhang, Chem. Soc. Rev. 2012, 41, 797. was constructed based on the calculated values from the [3] D. Yu, Q. Qian, L. Wei, W. Jiang, K. Goh, J. Wei, J. Zhang, Y. Chen, charge–discharge curves in Figure 5e and compared with preChem. Soc. Rev. 2015, 44, 647. vious work (Figure 6b). The inset figure in Figure 6b shows two [4] Z. Yu, L. Tetard, L. Zhai, J. Thomas, Energy Environ. Sci. 2014, 8, 702. devices in series that light up 11 pink and 10 white LEDs con[5] Q. Qu, S. Yang, X. Feng, Adv. Mater. 2011, 23, 5574. nected in parallel. The calculated volumetric energy density and [6] M. Sevilla, A. B. Fuertes, ACS Nano 2014, 8, 5069. [7] Y. Shi, L. Pan, B. Liu, Y. Wang, Y. Cui, Z. Bao, G. Yu, J. Mater. Chem. power density are 0.61, 0.55, 0.51, 0.37 mWh cm−3 and 52.9, A 2014, 2, 6086. 105.7, 211.4, 422.7 mW cm−3 at 2, 4, 8, 16 A g−1, respectively. [8] T.-W. Lin, C.-S. Dai, K.-C. Hung, Sci. Rep. 2014, 4, 7274. It is noted that the energy density of our nitride-based ASC is [9] F. Wang, S. Xiao, Y. Hou, C. Hu, L. Liu, Y. Wu, RSC Adv. 2013, 3, lower than that of PANI//WOx–MOx ASC;[55] this is ascribed to 13059. the intrinsic high theoretical capacitance of the latter. However, [10] J. Yan, Z. Fan, W. Sun, G. Ning, T. Wei, Q. Zhang, R. Zhang, L. Zhi, the values of energy density and power density are higher than F. Wei, Adv. Funct. Mater. 2012, 22, 2632. other PVA-based solid-state ASC/SSC devices (see more detailed [11] Z. Fan, J. Yan, T. Wei, L. Zhi, G. Ning, T. Li, F. Wei, Adv. Funct. Mater. comparison in Table S2, Supporting Information).[44,45,56] 2011, 21, 2366. In conclusion, metal nitride-based, quasi-solid-state asym[12] P. Tang, Y. Zhao, C. Xu, K. Ni, J. Solid State Electrochem. 2013, 17, metric supercapacitors have been fabricated with TiN cathode, 1701.
wileyonlinelibrary.com
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, 27, 4566–4571
www.advmat.de www.MaterialsViews.com
Adv. Mater. 2015, 27, 4566–4571
[35] C. Xia, Y. Xie, W. Wang, H. Du, Synth. Met. 2014, 192, 93. [36] C. Xia, Y. Xie, Y. Wang, W. Wang, H. Du, F. Tian, J. Appl. Electrochem. 2013, 43, 1225. [37] Y. Xie, H. Du, C. Xia, Microporous Mesoporous Mater. 2015, 204, 163. [38] M.-S. Balogun, M. Yu, Y. Huang, C. Li, P. Fang, Y. Liu, X. Lu, Y. Tong, Nano Energy 2015, 11, 348. [39] P. Yu, L. Wang, F. Sun, D. Zhao, C. Tian, L. Zhao, X. Liu, J. Wang, H. Fu, Chem. Eur. J. 2015, 21, 3249. [40] X. H. Xia, D. L. Chao, Y. Q. Zhang, Z. X. Shen, H. J. Fan, Nano Today 2014, 9, 785. [41] R. Thomas, G. Mohan Rao, J. Mater. Chem. A 2014, 3, 274. [42] N. Li, S. X. Jin, Q. Y. Liao, C. X. Wang, ACS Appl. Mater. Interfaces 2014, 6, 20590. [43] M. Liu, X. Li, S. K. Karuturi, A. I. Y. Tok, H. J. Fan, Nanoscale 2012, 4, 1522. [44] X. Lu, M. Yu, T. Zhai, G. Wang, S. Xie, T. Liu, C. Liang, Y. Tong, Y. Li, Nano Lett. 2013, 13, 2628. [45] X. Lu, G. Wang, T. Zhai, M. Yu, S. Xie, Y. Ling, C. Liang, Y. Tong, Y. Li, Nano Lett. 2012, 12, 5376. [46] X. Xiao, X. Peng, H. Jin, T. Li, C. Zhang, B. Gao, B. Hu, K. Huo, J. Zhou, Adv. Mater. 2013, 25, 5091. [47] J. Liu, Y. Li, H. Fan, Z. Zhu, J. Jiang, R. Ding, Y. Hu, X. Huang, Chem. Mater. 2010, 22, 212. [48] H. Qi, X. Roy, K. E. Shopsowitz, J. K. H. Hui, M. J. MacLachlan, Angew. Chem. Int. Ed. 2010, 49, 9740. [49] Q. Sun, Z.-W. Fu, Electrochem. Solid-State Lett. 2008, 11, A233. [50] B. Guo, Q. Liu, E. Chen, H. Zhu, L. Fang, J. R. Gong, Nano Lett. 2010, 10, 4975. [51] L. Lai, J. R. Potts, D. Zhan, L. Wang, C. K. Poh, C. Tang, H. Gong, Z. Shen, J. Lin, R. S. Ruoff, Energy Environ. Sci. 2012, 5, 7936. [52] L. Lai, H. Yang, L. Wang, B. K. Teh, J. Zhong, H. Chou, L. Chen, W. Chen, Z. Shen, R. S. Ruoff, J. Lin, ACS Nano 2012, 6, 5941. [53] H. Wang, T. Maiyalagan, X. Wang, ACS Catal. 2012, 2, 781. [54] V. Augustyn, P. Simon, B. Dunn, Energy Environ. Sci. 2014, 7, 1597. [55] X. Xiao, T. Ding, L. Yuan, Y. Shen, Q. Zhong, X. Zhang, Y. Cao, B. Hu, T. Zhai, L. Gong, J. Chen, Y. Tong, J. Zhou, Z. L. Wang, Adv. Energy Mater. 2012, 2, 1328. [56] X. Lu, M. Yu, G. Wang, T. Zhai, S. Xie, Y. Ling, Y. Tong, Y. Li, Adv. Mater. 2013, 25, 267.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
COMMUNICATION
[13] B. Avasarala, P. Haldar, Electrochim. Acta 2010, 55, 9024. [14] I. Milošv, H. H. Strehblow, B. Navinšek, M. Metikoš-Hukovic´, Surf. Interface Anal. 1995, 23, 529. [15] M.-S. Balogun, W. Qiu, W. Wang, P. Fang, X. Lu, Y. Tong, J. Mater. Chem. A 2015, 3, 1364. [16] J. H. Chae, K. C. Ng, G. Z. Chen, Proc. Instit. Mech. Eng. Part A: J. Power Energy 2010, 224, 479. [17] S. A. Al-Thabaiti, R. Hahn, N. Liu, R. Kirchgeorg, S. So, P. Schmuki, S. N. Basahel, S. M. Bawaked, Chem. Commun. 2014, 50, 7960. [18] C. Shang, S. Dong, S. Wang, D. Xiao, P. Han, X. Wang, L. Gu, G. Cui, ACS Nano 2013, 7, 5430. [19] Y. Xiao, G. Zhan, Z. Fu, Z. Pan, C. Xiao, S. Wu, C. Chen, G. Hu, Z. Wei, Electrochim. Acta 2014, 141, 279. [20] J. H. Bang, K. S. Suslick, Adv. Mater. 2009, 21, 3186. [21] T. T. Chen, H. P. Liu, Y. J. Wei, I. C. Chang, M. H. Yang, Y. S. Lin, K. L. Chan, H. T. Chiu, C. Y. Lee, Nanoscale 2014, 6, 5106. [22] X. Lu, T. Liu, T. Zhai, G. Wang, M. Yu, S. Xie, Y. Ling, C. Liang, Y. Tong, Y. Li, Adv. Energy Mater. 2014, 4, 1300994. [23] F. Grote, H. Zhao, Y. Lei, J. Mater. Chem. A 2015, 3, 3465. [24] X. Wang, V. Raju, W. Luo, B. Wang, W. F. Stickle, X. Ji, J. Mater. Chem. A 2014, 2, 2901. [25] A. Achour, J. B. Ducros, R. L. Porto, M. Boujtita, E. Gautron, L. Le Brizoual, M. A. Djouadi, T. Brousse, Nano Energy 2014, 7, 104. [26] Z. L. Wang, Nano Today 2010, 5, 540. [27] Y. Yue, P. Han, X. He, K. Zhang, Z. Liu, C. Zhang, S. Dong, L. Gu, G. Cui, J. Mater. Chem. 2012, 22, 4938. [28] Y. Qiu, K. Yan, S. Yang, L. Jin, H. Deng, W. Li, ACS Nano 2010, 4, 6515. [29] F. Tian, Y. Xie, H. Du, Y. Zhou, C. Xia, W. Wang, RSC Adv. 2014, 4, 41856. [30] Z. Wen, S. Cui, H. Pu, S. Mao, K. Yu, X. Feng, J. Chen, Adv. Mater. 2011, 23, 5445. [31] M.-S. Balogun, C. Li, Y. Zeng, M. Yu, Q. Wu, M. Wu, X. Lu, Y. Tong, J. Power Sources 2014, 272, 946. [32] H. Pang, S. J. Ee, Y. Dong, X. Dong, P. Chen, ChemElectroChem 2014, 1, 1027. [33] Z. Wang, Z. Li, J. Feng, S. Yan, W. Luo, J. Liu, T. Yu, Z. Zou, Phys. Chem. Chem. Phys. 2014, 16, 8521. [34] H. Du, Y. Xie, C. Xia, W. Wang, F. Tian, New J. Chem. 2014, 38, 1284.
4571
COMMUNICATION
www.MaterialsViews.com
All Metal Nitrides Solid-State Asymmetric Supercapacitors Changrong Zhu, Peihua Yang, Dongliang Chao, Xingli Wang, Xiao Zhang, Shi Chen, Beng Kang Tay, Hui Huang, Hua Zhang, Wenjie Mai, and Hong Jin Fan* Supercapacitor is relatively new technology compared to batteries but is regarded a highly promising alternative device for electrochemical energy storage, due to its capability of fast energy delivery (up to tens of seconds) and ultralong cycle life (typically 105 cycles).[1–4] However, the bottleneck of supercapacitor is the relatively low energy density compared to batteries, particularly when carbon materials are utilized as the electric double-layer capacitive electrode.[5–7] Asymmetric supercapacitors (ASCs) can improve the energy density by providing 1 increased voltage of the full devices according to E = CV 2.[8–12] 2 Recently, research on ASCs has been focused mostly on metal oxides because of their high pseudocapacitance through fast reversible redox reactions with ions from electrolyte. However, metal oxides usually suffer from low electrical conductivity and unreliable stability during long cycles. Metal nitrides, in the contrary, show much improved sustainability and superb electrical conductivity (4000−55 500 S cm−1),[13,14] and thus become attractive as SC electrode materials. Among all the reported metal nitrides in energy storage, TiN and Fe2N received particular attention due to their high capacitance (Fe2N as lithium ion battery anode: ≈900 mAh g−1; TiN as supercapacitor cathode: ≈150 F g−1) and electrical conductivity.[15,16] Various types of TiN nanostructures have been employed as supercapacitive electrode such as nanoparticle,[13] nanotube,[17–19] mesoporous microsphere,[20] and nanosheets.[21] However, TiN was found to be easily oxidized in aqueous solutions ⎛ TiN + 2H O → TiO + 1 N + 4H+ + 4e − ,2TiN + 2O → 2TiO + N ⎞ . [ 22 ] 2 2 2 2 2 2 ⎝ ⎠ 2
Therefore, to circumvent the oxidation problem of TiN,
C. Zhu, D. Chao, Dr. S. Chen, Prof. H. J. Fan School of Physical and Mathematical Sciences Nanyang Technological University 637371 Singapore, Singapore E-mail: [email protected] C. Zhu, Dr. H. Huang Singapore Institute of Manufacturing Technology 71 Nanyang Drive, 638075 Singapore, Singapore P. Yang, Prof. W. Mai Department of Physics and Siyuan Laboratory Jinan University Guangzhou 510632, P. R. China X. Wang, Prof. B. K. Tay School of Electrical and Electronic Engineering Nanyang Technological University 50 Nanyang Avenue, 639798 Singapore, Singapore X. Zhang, Prof. H. Zhang School of Materials Science and Engineering Nanyang Technological University 639798, Singapore
DOI: 10.1002/adma.201501838
4566
wileyonlinelibrary.com
researchers combined TiN with more stable materials. They include carbon coating,[22–24] carbon nanotube encapsulation,[25,26] graphene wrapping,[27–30] and forming core-shell structures with TiO2,[31] VN,[32] and MnO2,[33] as well as decoration with polymers such as polypyrrole,[34] polyaniline,[35,36] and PEDOT.[37] Those methods showed high efficiency in enhancing TiN cycling and rate properties. Fe2N has been reported as anode for lithium-ion batteries[38,39] in conjunction with carbonrelated materials. But, the absence of evident charge–discharge voltage plateaus[38] indicates its capacitive property and suitability as the supercapacitor electrode material. Due to the large surface area, high conductivity, and lightweight property, graphene becomes an effective substrate for SCs electrode.[40] Compared to the stacked graphene layers obtained from solution synthesis, vertically aligned graphene nanosheets (GNS) on current collectors (such as carbon fiber) are more advantageous as the later allows a maximum usage of the high surface area. This increased surface area will lead to a high capacitance since a supercapacitor device essentially relies on the available active surface sites for charge absorption (for double-layer capacitor) or redox reactions (for pseudocapacitor). Recently, such GNS substrates have been applied to SnO2 and ZnO-based electrodes to improve LIB performance.[41,42] In this work, we report the fabrication of nanostructured metal nitrides on vertically aligned graphene nanosheets substrate for high-power and ultrastable solid-state ASCs. Atomic layer deposition (ALD) was employed in order to assure the maximum use of the conductive GNS surfaces and have full coverage and tight connection of the active nitride material to the substrate. ALD is proven a powerful tool for nanofabrication and surface engineering of energy materials.[43] We employed solid-state electrolyte to avoid oxidation of metal nitrides.[44–46] The ASCs constructed from TiN@GNSs cathode, Fe2N@ GNSs anode, and poly(vinyl alcohol) (PVA)/LiCl solid electrolyte indeed show an irreversible capacitance of ≈58 F g−1 at 4 A g−1 which is stable up to 20 000 cycles. Moreover, the devices achieve both high volumetric energy density (≈0.55 mWh cm−3) and power density (≈220 mW cm−3) tested at a high current density of 8 A g−1. The fabrication process for both electrodes is shown schematically in Figure 1. Thin films of TiO2 and ZnO with same thicknesses of 20 nm were coated by ALD onto the GNS substrates. To obtain anode precursor, ZnO was first transferred to FeOOH via a solution reaction. Subsequently, the FeOOH@ GNS and TiO2@GNS samples were thermal annealed in a NH3 atmosphere for 1 h at 600 and 800 °C, respectively. The morphology of the samples at different fabrication stages are shown by scanning electron microscopy (SEM) images in Figure 2. The GNSs cover homogeneously the surface of carbon fibers (Figure 2a) with an overall height of ≈800 nm and are less
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, 27, 4566–4571
www.advmat.de www.MaterialsViews.com
Adv. Mater. 2015, 27, 4566–4571
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
COMMUNICATION
amonia annealing were plotted together in Figure 3g. Three typical peaks of graphite were detected in the XRD pattern of bare GNS. After the ALD of TiO2, two weaker peaks of graphene were further weaken and almost disapeared. No obvious peak of TiO2 was observed which may due to the amorphous property of ALD deposited material. As was expected, after high-temperature annealing in amonia atmosphere, three diffraction peaks at 37°, 44°, and 63° appear and correspond well to the face-centered cubic (fcc) TiN (111), (200), and (220), respectively (JC-PDS, #65-0715) which is identical to previous work.[27,48] Same characterization was conducted to the anode material. Figure 4a shows the particle morphology of Fe2N with a diameter ≈10 nm. The Fe2N nanoparticles were detached from the GNS by high-power sonication. TEM mapping of element Fe (blue, Figure 4c) and N (red, Figure 4d) corresponds well to the image profile of Fe2N particles in Figure 4b. Selected area electron diffraction (SAED) pattern (Figure 4e) showed Figure 1. Schematics of the fabrication processes of metal nitride cathode and anode materials. polycrystalline nature of the Fe2N, in which a series of well-defined rings can be assigned to various diffraction planes of (101), (211), and (400) of fcc Fe2N. HRTEM image (Figure 4f) showed the than 5 nm in thickness (Figure 2b). The smooth surface (Figure 2c) of the original ALD TiO2@GNS becomes porous clear lattice fringes with d-spacings of 2.11 and 3.45 Å, which conforms well to the crystal face of (101) and (211). To dem(Figure 2d) after the annealing process, and the porous struconstrate the transformation process of the anode material, the ture is expected to enlarge the specific surface area leading to XRD spectrum of sample at different fabrication stages was increased capacitance. Regarding the anode, after the ZnO@ provided in Figure 4g. Through the comparison, it is clear that GNS (Figure 2e, inset) was dipped in 0.5 M Fe(NO3)3 solupure phase of Fe2N was successfully achieved after the transtion for 2 h, the thin film becomes thinner with a thickness ≈10 nm (Figure 2e), which is FeOOH@GNS according to the formation reaction of amonia annealing. The related diffraction previous study.[47] After the ammonia annealing, the FeOOH peaks of 43°, 56°, 68°, and 75° were indexed to fcc Fe2N latthin film is converted to Fe2N nanoparticles that are dispersed tice of (211), (212), (400), and (213) (JC-PDS, #06-0656), as was reported in previous work.[39,49] uniformly on the surface of GNS (Figure 2f) with an average diameter ≈10 nm. Transmission electron microscopy (TEM) characterizations were performed in order to reveal the detailed nanostructure and phase of the electrode materials. As cathode, TiN shows the pore size varied from 3 to 10 nm (Figure 3a). High-resolution transmission electron microscopy (HRTEM) reveals the lattice of (111) planes (d-spacing 2.45 Å) and (220) ones (d-spacing 1.48 Å) of TiN (Figure 3f), which is in agreement with the fast Fourier transfer (FFT) generated from the same area (Figure 3e). TEM mapping exhibited the consistent element distribution of Ti (Figure 3c) and N (Figure 3d) with the dark field TEM image in Figure 3b. XRD spectrum of bare GNS after oxygen plasma (see the Experimental Section in Figure 2. Morphology characterization of the 3D graphene and the final electrodes. SEM images the Supporting Information), GNS with of a) GNS grown on carbon fiber, b) GNS, c) TiO2@GNS, d) TiN@GNS, and e) FeOOH@GNS 20 nm ALD TiO2, and TiO2@GNS after (inset is the ZnO@GNS), and f) Fe2N@GNS.
4567
www.advmat.de
COMMUNICATION
www.MaterialsViews.com
Figure 3. Structure analysis of the cathode material. a) TEM image of TiN@GNS. b–d) TEM elemental mapping of TiN@GNS (scale bar, 50 nm). e) FFT pattern corresponding to the (111) lattice in (f) and (g). f) The HRTEM image of TiN. g) XRD spectra at different fabrication stages.
Both the porous TiN cathode and particle Fe2N anode were homogeneously covered/ distributed on the thin GNS and were assembled to all-metal nitride asymmetric supercapacitor devices. We consider the benefit of such electrode structures to the device performance as follows. First, TiN[13] and Fe2N[39] are high electrical conductive materials, which would be beneficial to improve the ion and electron transport rates and therefore increase the specific capacitance and rate capability. Second, GNS grown on carbon fibers provides a highly conductive and large surface area (467 m2 g−1) scaffold for the active materials. Moreover, this vertically aligned graphene effectively avoids the stacking of individual graphene sheets. By applying ALD deposition, electrode materials can fully cover the entire surface of GNS which ensure maximum usage of the substrate surface. Last, the porous TiN thin films and Fe2N nanoparticles are grafted onto GNS by chemical reaction rather than physical mixing. This assures a tight physical connection of both metal nitride materials with the GNS current collector, and also contributes to a long and stable cycling life as well as a high power density. Before discussing the electrochemical properties of the electrodes, we first consider the possible capacity contribution by
4568
wileyonlinelibrary.com
the GNS substrate both before and after the ammonia annealing. Previous work [50–53] reported that O in C O and OH functional groups on graphene surface is easy to be substituted by NH2 under high-temperature ammonia annealing conditions. Therefore, as control sample, GNS substrates with same size (0.6 × 1 cm2) were annealed both at 600 and 800 °C with other conditions kept the same (see details in the Experimental Section in the Supporting Information, the control experiment). The cyclic voltammetry (CV) curves in the Supporting Information (Figure S1) showed that the capacity contribution of GNS substrates to the electrodes is less than 0.01% (ratios of area enclosed by the CV curves) both before and after ammonia annealing at 600 or 800 °C. Therefore, in our discussion below, we do not consider the capacitance contribution by the substrates. The detailed electrochemical characterization starts with single electrodes in a standard three-electrode testing system with Pt plate as a counter electrode, Ag/AgCl (saturated KCl) as a reference electrode, TiN cathode/Fe2N anode as a working electrode, and 1 M LiCl as an electrolyte. Figure 5a,b shows the CV curves of TiN cathode (0−0.8 V vs Ag/AgCl) and Fe2N anode (−0.8−0 V vs Ag/AgCl) at
Figure 4. Structure analysis of the anode material. a) TEM image of Fe2N@GNS. b–d) TEM elemental mapping of Fe2N particles (scale bar, 50 nm). e) SAED pattern showing (211), (101), and (400) lattice planes of Fe2N. f) HRTEM image of Fe2N. g) XRD spectra at different fabrication stages.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, 27, 4566–4571
www.advmat.de www.MaterialsViews.com
10 mV s-1
0
100 mV s-1
50 mV s 20 mV s-1 10 mV s-1
5 0
Current Density (mA cm-2)
20 mV s-1
10
(c) Fe2N @ GNS
-1
-5
-10
-5
-15 -20
-10 0.2 0.4 0.6 Potential (V vs Ag/AgCl)
-0.8
0.8
-0.6 -0.4 -0.2 Potential (V vs Ag/AgCl)
(e)1.8
180 160 140 120 100 80 60 40 20
TiN Fe2N Device
0
20
40 60 80 Scan speed (mV s-1)
100
6 TiN-Fe N full device 2
100 mV s-1
4 50 mV s-1 20 mV s-1
2
10 mV s-1
0 -2 -4 -6 0.0
0.0
0.4
0.8 1.2 Voltage (V)
(f) 80 Current density (A g-1)
16
0
8
4
60 40 20
2
10 20 30 40 50 60 70 80 90 Time (second)
1.6
Charge Discharge
Full Device
1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
Capacitance (F g-1)
Capacitance (F g-1)
Current Density (mA cm-2)
50 mV s-1
5
0.0
(d)
(b)15
100 mV s-1
Voltage (V)
Current Density (mA cm-2)
TiN @ GNS
COMMUNICATION
(a)10
Ic=Id=4 A g-1
0 0
5000 10000 15000 Number of cycle
20000
Figure 5. Electrochemical properties of individual electrodes and full devices. CV curves of a) TiN@GNS and b) Fe2N@GNS in a standard three-electrode testing system (Ag/AgCl as reference electrode, Pt as counter electrode, and related electrode as working electrode, testing in 1 M LiCl electrolyte) at different scan speeds of 10, 20, 50, and 100 mV s−1. c) CV curves of solid ASC device (TiN@GNS as cathode, Fe2N@GNS as anode, and LiCl-PVA quasi-solid-gel as electrolyte) at different scan speeds of 10, 20, 50, and 100 mV s−1. d) Specific capacitance of both single electrodes and full device at different scan rates calculated from the CV curves from (a–c). e) Charge–discharge curves of an all solid full metal nitride device at different current densities of 2, 4, 8, 16 A g−1. f) Cycling performance of full device at 4 A g−1 in 20 000 cycles with different bending situations. (The mass here is based on total mass of electrode active materials. For the full device, the total mass of TiN and Fe2N is used for calculation.)
different scan rates (10, 20, 50, and 100 mV s−1) separately. Both electrodes show quasirectangular shapes at varied scan rates, demonstrating that the majority capacitance should be ascribed to capacitive behavior.[54] As a comparison, the CV curves of precursor TiO2 and FeOOH (presented in the Supporting Information, Figure S2) show much smaller curve area. The capacitances estimated from the CV curves for the oxides are about ten times lower than the corresponding TiN and Fe2N. CV curves of full device in the range from 0 to 1.6 V at different scan rates are tested in a two-electrode system with TiN as cathode, Fe2N as anode, and PVA/LiCl as electrolyte (Figure 5c). The standard rectangle shape verifies a dominating capacitive behavior.[54] Although there is numerous research work about TiN and Fe2N as electrode material for energy storage system applications, the reaction mechanisms were seldom unraveled. Thus, we proposed a possible explanation for the pseudocapacitance with XPS and Raman data as support (see detailed discussion in the Supporting Information). Gravimetric capacitances at different scan rates were calculated based on Figure 5a–c and plotted in Figure 5d. When tested in 1 M LiCl aqueous solution, the individual TiN and Fe2N electrodes show high capacitance retention ≈80% and ≈77%, respectively, when the scan speed increases from 10 to 100 mV s−1. For the solidstate ASC device, a rather stable capacitance around 60 F g−1 is obtained with less than 2% capacity drop. The improved stabilities of TiN and Fe2N in solid electrolyte should be the reason for the higher rate performance of the device compared to that of single electrodes tested in the aqueous electrolyte. Figure 5e shows the galvanostatic charge–discharge curves of
Adv. Mater. 2015, 27, 4566–4571
the full device between 0 and 1.6 V at four current densities. The linear slopes and triangle shape corroborate the dominating capacitive property of the electrodes in the neutral solid electrolyte, consistent with the CV curves. The single electrodes show similar properties in aqueous electrolyte (Figure S3a,c, Supporting Information). Moreover, both cathode and anode show high stability in 20 000 cycles at 4 A g−1. As for the full device, the cycling property was tested at bending condition from 5000th to 15 000th cycles and then recovered to natural state in last 5000 cycles (Figure 5f). The capacitance profile is nearly flat with a highly reversible capacitance ≈58 F g−1 up to 2000 cycles and the bending does not affect much the capacitance. The Coulombic efficiency within 20 000 cycles is maintained at about 100% (Figure S4a, Supporting Information). More details about the bending test are shown in Figure S4b (Supporting Information). The high capacity retention of the device over 20 000 cycles can be ascribed to the structure stability of the electrode material in the gel electrolyte. To confirm this, the morphology of both cathode and anode of the ASCs after 20 000 cycles was checked with SEM (see Figure S5 in the Supporting Information). It is found that the entire electrode structure is well maintained; both the porous structure of TiN and nanoparticle morphology of Fe2N are well kept although slight aggregation of Fe2N is observed. This confirms the high structure stability of both electrode materials. Nyquist plots of single electrodes and the full device at different cycling stages were tested in aqueous and gel electrolyte, respectively (see Figure S6 in the Supporting Information). Both the single electrode and full device are highly conductive with small
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
4569
www.advmat.de
COMMUNICATION
www.MaterialsViews.com
4570
Fe2N anode (both synthesized on highly conductive graphene nanosheets), and PVA/LiCl neutral electrolyte. The porous structure of TiN and uniform coverage of Fe2N nanoparticles on the GNS substrates are important merits that account for the outstanding cycling stability (≈98% capacity retention in 20 000 cycles) and rate capability (≈99% capacity maintenance when scan rate increases from 10 to 100 mV s−1) of the devices. Five devices are tested and show reproducible results. Typical values of the energy density of Figure 6. a) Cycling performance of five parallel devices at 4 A g−1 within 20 000 cycles. Inset figure shows the percentages of capacity retention. b) Ragone plots of quasi-solid-state TiN≈15.4 Wh kg−1 (≈0.51 mWh cm−3) and power Fe2N ASC in comparison with other PVA-based solid electrolyte SSCs and ASCs. Inset: pink and density of ≈6.4 kW kg−1 (211.4 mW cm−3) are white LEDs in parallel are lit up by two full devices in tandem. obtained. The strategy of combining ALD with ammonia annealing can be extended to other metal nitrides such as SnN, AlN, Zn3N2, and Mg3N2 for electrochemical impedance values (below 5 Ω). After 2000 and 20 000 cycles, the impedance values increase but still acceptelectrochemical energy storage applications. ably low. It is noted that both the nitride electrodes exhibit smaller impedance compared with their oxide source material. The mass loading of TiN and Fe2N is ≈0.5 and ≈0.8 mg cm−2, Supporting Information respectively (see detailed data in Table S1 and the related deterSupporting Information is available from the Wiley Online Library or mination steps in the Supporting Information). To achieve best from the author. performance, the mass ratio between the TiN cathode and Fe2N anode is about 0.69:1 (for details, see the last part in the Supporting Information). To test the repeatability of the methodology, a series of Acknowledgements devices was assembled and their specific capacitances were This research was funded by SERC Public Sector Research Funding measured and compared (Figure 6a). Within 20 000 cycles, all (Grant Number 1121202012), Agency for Science, Technology, and the five devices show very stable and nearly overlapping speResearch (A*STAR), and AcRF Tier 1 (RG104/14). H.Z. thanks the supported from MOE under AcRF Tier 2 (MOE2013-T2-1-034) and cific capacitance ≈58 F g−1. High capacity retention is obtained AcRF Tier 1 (RG 61/12, RGT18/13, and RG5/13) in Singapore. This for all five devices (ranging from 96.8 to 99.8%). The results research was also conducted by NTU-HUJ-BGU Nanomaterials for demonstrate the high reliability and repeatability of the fabricaEnergy and Water Management Programme under the Campus for tion method and the stability of the nanostructured electrode Research Excellence and Technological Enterprise (CREATE), which materials as well. To evaluate the practical application possiwas supported by the National Research Foundation, Prime Minister’s bility of the full device, we calculated the specific gravimetric Office, Singapore. The authors also thank INCUBATION ALLIANCE, Inc. capacitance based on the total mass of the device and plotted for providing the graphene nanosheets/nanoflower (GNS) substrates. the cycling curve in Figure S10 (Supporting Information). Received: April 17, 2015 Due to the light weight of the GNS@carbon fiber substrate Revised: May 20, 2015 −2 −1 (16 mg cm ), the device shows a stable capacitance of 2.5 F g Published online: July 8, 2015 as a whole. For a full device, energy density and power density are two important parameters for the evaluation of the entire device. [1] Z. Liu, J. Xu, D. Chen, G. Shen, Chem. Soc. Rev. 2015, 44, 161. The Ragone plot of the all metal nitride supercapacitor devices [2] G. Wang, L. Zhang, J. Zhang, Chem. Soc. Rev. 2012, 41, 797. was constructed based on the calculated values from the [3] D. Yu, Q. Qian, L. Wei, W. Jiang, K. Goh, J. Wei, J. Zhang, Y. Chen, charge–discharge curves in Figure 5e and compared with preChem. Soc. Rev. 2015, 44, 647. vious work (Figure 6b). The inset figure in Figure 6b shows two [4] Z. Yu, L. Tetard, L. Zhai, J. Thomas, Energy Environ. Sci. 2014, 8, 702. devices in series that light up 11 pink and 10 white LEDs con[5] Q. Qu, S. Yang, X. Feng, Adv. Mater. 2011, 23, 5574. nected in parallel. The calculated volumetric energy density and [6] M. Sevilla, A. B. Fuertes, ACS Nano 2014, 8, 5069. [7] Y. Shi, L. Pan, B. Liu, Y. Wang, Y. Cui, Z. Bao, G. Yu, J. Mater. Chem. power density are 0.61, 0.55, 0.51, 0.37 mWh cm−3 and 52.9, A 2014, 2, 6086. 105.7, 211.4, 422.7 mW cm−3 at 2, 4, 8, 16 A g−1, respectively. [8] T.-W. Lin, C.-S. Dai, K.-C. Hung, Sci. Rep. 2014, 4, 7274. It is noted that the energy density of our nitride-based ASC is [9] F. Wang, S. Xiao, Y. Hou, C. Hu, L. Liu, Y. Wu, RSC Adv. 2013, 3, lower than that of PANI//WOx–MOx ASC;[55] this is ascribed to 13059. the intrinsic high theoretical capacitance of the latter. However, [10] J. Yan, Z. Fan, W. Sun, G. Ning, T. Wei, Q. Zhang, R. Zhang, L. Zhi, the values of energy density and power density are higher than F. Wei, Adv. Funct. Mater. 2012, 22, 2632. other PVA-based solid-state ASC/SSC devices (see more detailed [11] Z. Fan, J. Yan, T. Wei, L. Zhi, G. Ning, T. Li, F. Wei, Adv. Funct. Mater. comparison in Table S2, Supporting Information).[44,45,56] 2011, 21, 2366. In conclusion, metal nitride-based, quasi-solid-state asym[12] P. Tang, Y. Zhao, C. Xu, K. Ni, J. Solid State Electrochem. 2013, 17, metric supercapacitors have been fabricated with TiN cathode, 1701.
wileyonlinelibrary.com
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, 27, 4566–4571
www.advmat.de www.MaterialsViews.com
Adv. Mater. 2015, 27, 4566–4571
[35] C. Xia, Y. Xie, W. Wang, H. Du, Synth. Met. 2014, 192, 93. [36] C. Xia, Y. Xie, Y. Wang, W. Wang, H. Du, F. Tian, J. Appl. Electrochem. 2013, 43, 1225. [37] Y. Xie, H. Du, C. Xia, Microporous Mesoporous Mater. 2015, 204, 163. [38] M.-S. Balogun, M. Yu, Y. Huang, C. Li, P. Fang, Y. Liu, X. Lu, Y. Tong, Nano Energy 2015, 11, 348. [39] P. Yu, L. Wang, F. Sun, D. Zhao, C. Tian, L. Zhao, X. Liu, J. Wang, H. Fu, Chem. Eur. J. 2015, 21, 3249. [40] X. H. Xia, D. L. Chao, Y. Q. Zhang, Z. X. Shen, H. J. Fan, Nano Today 2014, 9, 785. [41] R. Thomas, G. Mohan Rao, J. Mater. Chem. A 2014, 3, 274. [42] N. Li, S. X. Jin, Q. Y. Liao, C. X. Wang, ACS Appl. Mater. Interfaces 2014, 6, 20590. [43] M. Liu, X. Li, S. K. Karuturi, A. I. Y. Tok, H. J. Fan, Nanoscale 2012, 4, 1522. [44] X. Lu, M. Yu, T. Zhai, G. Wang, S. Xie, T. Liu, C. Liang, Y. Tong, Y. Li, Nano Lett. 2013, 13, 2628. [45] X. Lu, G. Wang, T. Zhai, M. Yu, S. Xie, Y. Ling, C. Liang, Y. Tong, Y. Li, Nano Lett. 2012, 12, 5376. [46] X. Xiao, X. Peng, H. Jin, T. Li, C. Zhang, B. Gao, B. Hu, K. Huo, J. Zhou, Adv. Mater. 2013, 25, 5091. [47] J. Liu, Y. Li, H. Fan, Z. Zhu, J. Jiang, R. Ding, Y. Hu, X. Huang, Chem. Mater. 2010, 22, 212. [48] H. Qi, X. Roy, K. E. Shopsowitz, J. K. H. Hui, M. J. MacLachlan, Angew. Chem. Int. Ed. 2010, 49, 9740. [49] Q. Sun, Z.-W. Fu, Electrochem. Solid-State Lett. 2008, 11, A233. [50] B. Guo, Q. Liu, E. Chen, H. Zhu, L. Fang, J. R. Gong, Nano Lett. 2010, 10, 4975. [51] L. Lai, J. R. Potts, D. Zhan, L. Wang, C. K. Poh, C. Tang, H. Gong, Z. Shen, J. Lin, R. S. Ruoff, Energy Environ. Sci. 2012, 5, 7936. [52] L. Lai, H. Yang, L. Wang, B. K. Teh, J. Zhong, H. Chou, L. Chen, W. Chen, Z. Shen, R. S. Ruoff, J. Lin, ACS Nano 2012, 6, 5941. [53] H. Wang, T. Maiyalagan, X. Wang, ACS Catal. 2012, 2, 781. [54] V. Augustyn, P. Simon, B. Dunn, Energy Environ. Sci. 2014, 7, 1597. [55] X. Xiao, T. Ding, L. Yuan, Y. Shen, Q. Zhong, X. Zhang, Y. Cao, B. Hu, T. Zhai, L. Gong, J. Chen, Y. Tong, J. Zhou, Z. L. Wang, Adv. Energy Mater. 2012, 2, 1328. [56] X. Lu, M. Yu, G. Wang, T. Zhai, S. Xie, Y. Ling, Y. Tong, Y. Li, Adv. Mater. 2013, 25, 267.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
COMMUNICATION
[13] B. Avasarala, P. Haldar, Electrochim. Acta 2010, 55, 9024. [14] I. Milošv, H. H. Strehblow, B. Navinšek, M. Metikoš-Hukovic´, Surf. Interface Anal. 1995, 23, 529. [15] M.-S. Balogun, W. Qiu, W. Wang, P. Fang, X. Lu, Y. Tong, J. Mater. Chem. A 2015, 3, 1364. [16] J. H. Chae, K. C. Ng, G. Z. Chen, Proc. Instit. Mech. Eng. Part A: J. Power Energy 2010, 224, 479. [17] S. A. Al-Thabaiti, R. Hahn, N. Liu, R. Kirchgeorg, S. So, P. Schmuki, S. N. Basahel, S. M. Bawaked, Chem. Commun. 2014, 50, 7960. [18] C. Shang, S. Dong, S. Wang, D. Xiao, P. Han, X. Wang, L. Gu, G. Cui, ACS Nano 2013, 7, 5430. [19] Y. Xiao, G. Zhan, Z. Fu, Z. Pan, C. Xiao, S. Wu, C. Chen, G. Hu, Z. Wei, Electrochim. Acta 2014, 141, 279. [20] J. H. Bang, K. S. Suslick, Adv. Mater. 2009, 21, 3186. [21] T. T. Chen, H. P. Liu, Y. J. Wei, I. C. Chang, M. H. Yang, Y. S. Lin, K. L. Chan, H. T. Chiu, C. Y. Lee, Nanoscale 2014, 6, 5106. [22] X. Lu, T. Liu, T. Zhai, G. Wang, M. Yu, S. Xie, Y. Ling, C. Liang, Y. Tong, Y. Li, Adv. Energy Mater. 2014, 4, 1300994. [23] F. Grote, H. Zhao, Y. Lei, J. Mater. Chem. A 2015, 3, 3465. [24] X. Wang, V. Raju, W. Luo, B. Wang, W. F. Stickle, X. Ji, J. Mater. Chem. A 2014, 2, 2901. [25] A. Achour, J. B. Ducros, R. L. Porto, M. Boujtita, E. Gautron, L. Le Brizoual, M. A. Djouadi, T. Brousse, Nano Energy 2014, 7, 104. [26] Z. L. Wang, Nano Today 2010, 5, 540. [27] Y. Yue, P. Han, X. He, K. Zhang, Z. Liu, C. Zhang, S. Dong, L. Gu, G. Cui, J. Mater. Chem. 2012, 22, 4938. [28] Y. Qiu, K. Yan, S. Yang, L. Jin, H. Deng, W. Li, ACS Nano 2010, 4, 6515. [29] F. Tian, Y. Xie, H. Du, Y. Zhou, C. Xia, W. Wang, RSC Adv. 2014, 4, 41856. [30] Z. Wen, S. Cui, H. Pu, S. Mao, K. Yu, X. Feng, J. Chen, Adv. Mater. 2011, 23, 5445. [31] M.-S. Balogun, C. Li, Y. Zeng, M. Yu, Q. Wu, M. Wu, X. Lu, Y. Tong, J. Power Sources 2014, 272, 946. [32] H. Pang, S. J. Ee, Y. Dong, X. Dong, P. Chen, ChemElectroChem 2014, 1, 1027. [33] Z. Wang, Z. Li, J. Feng, S. Yan, W. Luo, J. Liu, T. Yu, Z. Zou, Phys. Chem. Chem. Phys. 2014, 16, 8521. [34] H. Du, Y. Xie, C. Xia, W. Wang, F. Tian, New J. Chem. 2014, 38, 1284.
4571
