Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Suitable Conditions for the Use of Vanadium Nitride as an Electrode for Electrochemical Capacitor

Suitable Conditions for the Use of Vanadium Nitride as an Electrode for Electrochemical Capacitor Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) A1077 Suitable Conditions for the Use of Vanadium Nitride as an Electrode for Electrochemical Capacitor a,b,c a a,b,d,e a,b,∗ Alban Morel, Yann Borjon-Piron, Raul ´ Lucio Porto, Thierry Brousse, c,∗,z and Daniel Belanger ´ Institut des Materiaux ´ Jean Rouxel (IMN), Universite ´ de Nantes, CNRS, 44322 Nantes Cedex 3, France Reseau sur le Stockage Electrochimique de l’Energie (RS2E), FR CNRS 3459, France ´ ´ ´ ` ´ ´ ´ Departement Chimie, UniversiteduQuebec a Montreal, Succursale Centre-Ville, Montreal, Quebec H3C 3P8, Canada ´ ´ ´ ´ ´ ´ Universidad Autonoma de Nuevo Leon, Facultad de Ingenierıa Mecanica y Electrica, San Nicolas de los Garza, 66450 Nuevo Leon, ´ Mexico ´ Universidad Autonoma ´ de Nuevo Leon, ´ Centro de Innovacion, ´ Investigacion ´ y Desarrollo en Ingenier´ ıa y Tecnolog´ ıa, Apodaca, 66600 Nuevo Leon, ´ Mexico ´ Vanadium nitride has displayed many interesting characteristics for its use as a pseudocapacitive electrode in an electrochemical capacitor, such as good electronic conductivity, good thermal stability, high density and high specific capacitance. Thin films of VN were prepared by D.C. reactive magnetron sputtering. The electrochemical stability of the films as well as the influence of dissolved oxygen in 1 M KOH electrolyte were investigated. In order to avoid material as well as electrolyte degradation, it was concluded that vanadium nitride should only be cycled between −0.4 and −1.0 V vs. Hg/HgO. After a 24 hours stabilization period, the prepared −2 VN thin film showed an initial capacitance of 19 mF.cm and a capacity retention of 96% after 10000 cycles. Furthermore, dissolved oxygen in the electrolyte was demonstrated to cause self-discharge up to a potential above −0.4 V vs. Hg/HgO, where VN was shown to be unstable. Additionally, the presence of oxygen was shown to shift the open circuit potential of a VN electrode to about 0 V through self-discharge processes. © The Author(s) 2016. Published by ECS. This is an open access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives 4.0 License (CC BY-NC-ND, http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is not changed in any way and is properly cited. For permission for commercial reuse, please email: [email protected]. [DOI: 10.1149/2.1221606jes] All rights reserved. Manuscript submitted February 11, 2016; revised manuscript received March 14, 2016. Published March 29, 2016. Electrochemical capacitors are currently being developed to com- [−1.2; 0.0] V to [−1.2; −0.3] V vs. Hg/HgO enhanced the capacity plement other energy storage or conversion systems such as batter- retention after 1000 charge/discharge cycles. Further improvements ies and fuel cells. In the past two decades, several electrode mate- were obtained when changing the electrolyte pH from 14 to 12. rials have been investigated. The mostly studied materials include When charging/discharging VN electrodes between −1.2 and −0.4 V 1–8 9–12 13 carbons, conducting polymers as well as pseudocapacitive instead of −1.2 and 0 V vs. Hg/HgO for 2800 cycles in 1 M KOH, 14–18 46 transition metal oxides such as ruthenium dioxide and manganese Porto et al. also found enhancement of the capacity retention from 0 19–24 52 dioxide. In the effort to improve the performance of electrochem- to 70%, respectively. To our knowledge, only Lu et al. studied VN ical capacitors, a widely used approach has been to develop synthetic over a relatively large number of charge/discharge cycles (15000 cy- methods to develop nanomaterials that are believed to lead to increased cles). They have shown that when cycling VN up to 0.1 V vs. Hg/HgO 14,23,25–28 energy density and higher rate capability. in 1 M KOH, most of the capacitance was lost after a few thousands Another approach has been to investigate the charge storage prop- cycles. Interestingly, they also showed that when coating VN with car- erties of novel materials. Accordingly, several research groups consid- bon, the electrode kept up to 88% of its initial capacitance after 15000 29,30 31–37 ered materials such as MXenes and transition metal nitrides. cycles in 1 M KOH. All these studies highlight the need of in-depth In the latter class of compounds, molybdenum nitride was firstly investigations of the suitable conditions for which VN is stable as an 32,33,35,38,39 investigated and in the past decade a great deal of atten- active material for energy storage applications. tion has focused on other nitrides with an intensive focus on vana- Most studies use VN powders prepared via ammonolysis of vari- 37,40–58 40,42,44,45,48,49,53,60 37,58 dium nitride as a consequence of the impressive capacitance of ous vanadium oxides or chloride. As pointed out −1 58 46 1340 F.g reported for nanosized VN particles in 1 M KOH. In- by Porto et al., this gives rise to two main drawbacks: (1) the use deed, VN is an interesting electrode material due to this high reported of different precursors especially oxides may lead to difference in specific capacitance coupled with its close to metallic electronic con- compositions and especially oxygen content in the VN powders and −1 59 −3 ductivity (1.18 S.m ), high density (6.13 g.cm ) and high melting (2) the use of powders imply the use of carbon and polymer addi- point (2619 K). tives in order to prepare composite electrodes, thus preventing the The hypothesis proposed by Choi et al. to explain such a high study of VN intrinsic electrochemical properties. On the other hand, specific capacitance of VN, is that, in addition to electrochemi- thin film deposition techniques such as DC sputtering lead to the cal double layer capacitance, successive fast reversible redox reac- synthesis of VN thin films using only vanadium metal and nitro- tions are taking place, involving surface oxide groups and OH ions gen which can be used as-deposited as electrode material without from the electrolyte. This hypothesis was supported by the observa- the need of binder neither conductive additives. Additionally, tuning 58 41,58 tion of surface oxide via ex-situ XPS and FTIR post cycling the deposition conditions can lead to a wide range of morphologies, measurements. from porous to dense, and to the growth of randomly or preferen- While most studies concentrated on new synthesis of VN with the tially oriented thin films. Furthermore, the use of current collector goal to enhance the specific power density and the specific capaci- is not necessary since VN thin films exhibit high electronic con- tance, which was found in most studies to be limited in the range ductivity, and thus can be used both as active materials and current −1 40,42–46,48–52 of 150 to 300 F.g , only few studies looked at cycling collector. parameters-cycle life relationship. Choi et al. showed that diminish- Herein, we report preparation of VN thin film working electrodes ing the potential window over which the VN electrode is cycled from deposited on soda lime glass by D.C. sputtering and the study of their electrochemical properties by a general procedure aimed at determin- ing the experimental conditions to obtain a stable electrode for use in Electrochemical Society Member. an electrochemical capacitor. E-mail: [email protected] A1078 Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) Experimental Thin film preparation.—VN thin films were prepared by D.C reac- tive sputtering using the AC450 sputtering system from Alliance Con- cept. Prior to deposition, soda lime glass substrates were degreased with successive washing with acetone and ethanol. The deposition −4 chamber was first vacuumed to a pressure of 10 Pa. The substrates 4000 were then cleaned in situ with argon sputter etching (6 Pa, 50 sccm, 100 W RF) for 3 min. In order to ensure good adhesion of the VN thin film to the glass substrate, a titanium layer was first sputtered from a 2 diameter titanium disk of 99.95% purity, in an argon plasma B (1 Pa, 50 sccm, 50 W DC), with a target to substrate distance of 7.5 cm. The VN was then sputtered from a 2 diameter vanadium disk of 99.5% purity, with an applied power to the target of 50 W DC, in a gas 30 40 50 60 70 80 90 mixture of Ar/N (with fixed flows of 30/2.5 sccm respectively) for 2θ (°) Cu a total chamber pressure of 1 Pa. Once again, the target-to-substrate distance was fixed to 7.5 cm. The deposition time was changed to Figure 1. X-ray diffraction pattern of a 60 nm titanium adhesive layer (curve control the film thickness. Unless otherwise stipulated, all prepared A) and of 140 (curve B) and 280 nm (Curve C) thick layer of VN on top of that adhesive layer. Curves D and E present the calculated diffraction pattern for thin films are composed of a 140 nm thick VN layer deposited on top the 280 nm thick VN layer and the difference between the measured and the of a 60 nm thick titanium adhesive layer. calculated pattern obtained from FullProf software, respectively. The (◦)and () symbols indicate diffraction peaks attributed to the VN and the Ti phases, Thin film characterization.—The crystallographic structure of the respectively. as deposited material was investigated by X-ray diffraction with an XRD using PANalytical’s X’Pert Pro diffractometer with Cu Kα ra- diation (λ = 1.5418 Å). Lattice parameter of the VN phase was then sputtering and could be explained by a compressive stress related to calculated from the refinement of the Bragg positions through the 62,63 the deposition conditions. use of FullProf software. The morphological aspects of the thin films Representative SEM images of VN thin films are shown in were observed using FEM-SEM Zeiss Merlin scanning electron mi- Figure 2. The VN film exhibits columnar growth on the glass croscope. substrate (Figure 2a) with an average column diameter of 20 nm (Figure 2b). A similar morphology has been previously observed for Electrochemical characterization.—The electrochemical proper- the same type of deposition process. Moreover, the irregular colum- ties of the thin films were investigated with a Solartron multipoten- nar structure observed in the cross section view (Figure 2a) as well tiostat 1470 coupled to a Solartron frequency analyzer 1255B. All as the observation of some pores in the top view image (Figure 2b) measurements were carried out in a closed three-electrode cell with a suggest the presence of voids in between columns. platinum wire as counter electrode, a Hg/HgO (1 M KOH) electrode as reference electrode and a VN thin film as working electrode and Electrochemical characterization.—Figure 3 depicts cyclic current collector. For ease of reading, all potentials mentioned in the voltammograms for Ti/VN and Ti electrodes in 1 M KOH without text refer to the Hg/HgO (1 M KOH) electrode potential. The cell was any nitrogen bubbling of the electrolyte. A main redox system cen- designed so that the VN surface area in contact with the electrolyte tered at −0.67 V is observed on top of a capacitive envelope for the was limited to 0.502 cm via the use of a gasket. O or N gas was 2 2 VN electrode. Another set of redox waves with an average potential bubbled in the aqueous 1 M KOH electrolyte for 40 min prior each of −1.00 V for the anodic and cathodic peak potentials is also no- experiment. In the case of experiments with N gas, bubbling was also ticed but only for the VN electrode after 20 cycles. These two redox kept during the measurements. In the case of experiments with O gas, the gas stream was only kept above the solution during measurements. Prior to every EIS measurement, the electrode potential was held for 5 min at the desired potential so that the measurement would be representative of the steady state of the system. Nyquist and phase an- gle vs. frequency plots were then obtained by scanning the frequency from 1 kHz to 10 mHz, using a sinusoidal signal of 10 mV amplitude. Results and Discussion Structural characterization.—XRD patterns of a titanium adhe- sive layer and vanadium nitride thin film of two different thicknesses (140 and 280 nm) deposited on the titanium layer are presented in Fig- ure 1. Comparing the XRD patterns of the samples with and without the vanadium nitride layer enables to distinguish the titanium layer diffraction peaks from those of vanadium nitride. Five diffractions ◦ ◦ ◦ ◦ ◦ peaks observed at 2 value of 38.0 , 44.2 64.3 , 77.2 and 81.3 can be attributed to the vanadium nitride thin film and can be ascribed to the (111), (200), (220), (311) and (222) planes of its cubic crystal system (Fm-3m [225]), respectively. Calculation of the lattice param- eter, based on the peak positions observed on the XRD pattern of the thicker VN film which presents slightly more defined peaks, results in a unit-cell parameter of a = 4.094(6) Å. The cell parameter for the stoichiometric bulk compound being 4.1392 Å (JCPDS file 35-768), this could reveal nitrogen vacancies in the as-deposited vanadium nitride, with a composition of VN . However, a lower lattice pa- 0.84 Figure 2. SEM images of a 280 nm VN/160 nm titanium bilayer thin film rameter has already been reported for VN thin films prepared by D.C. deposited on soda lime glass substrate a) cross section, b) top view. [111] [200] [220] [311] [222] Intensity (arbitrary units) Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) A1079 0,0 0.4 0.2 -0,2 0.0 -0,4 -0.2 -0,6 -0,8 -0.4 -1,0 -0.6 0 10000 20000 30000 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Time / s E vs. Hg/HgO (V) Figure 5. Open-circuit potential decay of a VN thin film following polar- Figure 3. Cyclic voltammograms of a VN thin film (cycle #2 (curve A) and −1 ization at −1.0 and −0.5 V for 5 min. under N (curves A and B) and O cycle #20 (curve B)) and of the adhesive titanium layer (curve C) at 20 mV.s 2 2 (curve C) atmospheres. in non-degassed 1 M KOH electrolyte. systems have already been observed at similar or slightly more positive the presence of oxygen in the electrolyte. Prior to cyclic voltammetry 41,43,58 potentials. Another smaller redox system centered at −0.275 V, experiments, the electrolyte was bubbled with either N or O gas 2 2 41,58 which is not clearly observed here, is occasionally seen. It has for40min.AlsowhenN was replaced by O , the potential of the 2 2 been suggested that these redox systems result from successive tran- electrode was held at −0.5 V in order to avoid any reaction that could +5 +2 58 sitions from V to V of vanadium oxides surface sites. These occur if the OCP would shift to value outside of the −0.5 to −1.0 V hypotheses have been supported by the observation of vanadium ox- interval. Figure 4 depicts the cyclic voltammograms obtained under ide by surface analysis following electrochemical cycling between 0 N and O . Under N , the CV shows a capacitive behavior and redox 2 2 2 41,64 and −1.2 V in 1 M KOH. However, the technical challenge that waves centered at −0.67 V. Anodic and cathodic charges (Q and Q , a c represents an in-situ investigation of such a system makes it difficult respectively) evaluated by integration of the cyclic voltammogram to assign each peak to a specific redox process. Moreover, the loss are very similar with values of 7.7 and 7.8 mC/cm , respectively. This th of electroactivity between the second and 20 cyclic voltammogram relatively good Coulombic efficiency (99%), as well as the absence suggests that the VN thin film is not stable in these conditions. At this of any tail shape at the most positive and negative potential limits moment, it is difficult to explain such a loss during only 20 cycles. of the cyclic voltammogram suggest the absence of any irreversible Figure 3 also shows an irreversible cathodic reaction for potentials processes. Thus, the larger cathodic charge, relative to the anodic more negative than −1.0 V that can be tentatively attributed to the charge, sometimes observed in the literature for VN electrodes can onset of the hydrogen evolution reaction (vide infra). However, by presumably be attributed to the presence of traces of oxygen in the 42,43 decreasing the potential window (between −1.0 and −0.5 V), sta- electrolyte. In contrast, only cathodic currents are observed for the −1 ble cyclic voltammograms can be obtained at 1 mV.s (Figure 4). cyclic voltammogram measured under O . This behavior reveals the These conditions will then be used as starting cycling conditions in irreversible oxygen reduction reaction due to presence of oxygen in order to study the influence of oxygen dissolved in the electrolyte, the solution (Eq. 1). Bubbling oxygen in the electrolyte was performed prior to determine the optimal potential window over which VN elec- only to emphasize the electrochemical response of this reaction on a trodes could be operated. Furthermore, one can notice the negligible VN electrode. However, one should realize that such irreversible reac- contribution of the titanium adhesive layer to the measured currents tion will contribute a cathodic charge (eg. in the presence of traces of (Figure 3,curve C). oxygen) that will counterbalance and hide the effect of any irreversible anodic process, when evaluating the Coulombic efficiency. Influence of oxygen in the electrolyte.—Since metal nitrides − − O + 2H O + 4e  4OH [1] 2 2 are now widely investigated for water splitting (eg. the hydrogen 65 66 The influence of oxygen can also be observed when comparing the evolution and oxygen reduction reactions ), the effect of oxygen evolution of the open-circuit potential (OCP) of the electrode in pres- on the electrochemical behavior of VN was investigated. More specif- ence of N and O (Figure 5). In this experiment, the electrode po- ically, this section aimed to emphasize and demonstrate the effect of 2 2 tential was firstly cycled between −0.5 and −1.0 V and polarized at either −1.0 or −0.5 V prior to monitoring the evolution of the OCP for 8 hours. The measurements made under nitrogen show a slow de- 0,02 cay/recovery of the thin film OCP from −1.00 to −0.76 V and −0.50 to −0.65 V. On the other hand, in the presence of oxygen, a very 0,00 fast decay of potential is observed and reached −0.10 V in less than 25 min and −0.01 V after 8 hours regardless of the initial potential. This drastic self-discharge is another example of the influence of the -0,02 oxygen reduction reaction (Eq. 1) taking place at the electrode until its potential reaches the equilibrium potential of that reaction (E = -0,04 0.298 V vs. Hg/HgO (1 M KOH)), or until an equilibrium is reached between this reduction reaction and an oxidation reaction (e.g., cor- rosion process through oxidation of the material and reduction of -0,06 oxygen). Following an additional 24 hours immersion in presence of oxygen, the degradation of the thin film was visually observed. The -1,0 -0,9 -0,8 -0,7 -0,6 -0,5 fact that the OCP does not reach 0.298 V and that degradation is ob- E vs. Hg/HgO (V) served corroborate the hypothesis of a corrosion process in presence of oxygen. Moreover, a significant corrosion rate has been reported Figure 4. Cyclic voltammograms of a VN thin film under N (curve A) and ◦ 67 −1 O (curve B) atmospheres at 1 mV.s in 1 M KOH. for VN in NaOH at 80 C. -2 -2 I (mA.cm ) I (mA.cm ) E / V vs. Hg/HgO A1080 Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) The cyclic voltammograms and the anodic to cathodic charge ra- 0,1 tio (Qa/Qc) recorded for negative potential limits down to −1.1 V seems to suggest that the hydrogen evolution reaction (HER) is not 0,0 occurring at −1.1 V. Hence, electrolytic decomposition can be pos- 0.97 sibly avoided at VN electrode by arbitrarily restricting the negative -0,1 0.98 0.98 0.97 potential limit to value more positive than −0.95 V. However, de- pending on the kinetics of the HER the data of Figure 6a alone might -0,2 0.94 be insufficient to rule out the possibility that such reaction is tak- -1,2 -1,1 -1,0 -0,9 -0,8 -0,7 -0,6 ing place at this potential. Recently, Wu et al. showed that in situa- a) E vs. Hg/HgO (V) tion where cyclic voltammetric and galvanostatic cycling experiments -0.8 V would suggest that a system has a capacitive behavior, electrochemical -4000 -80 -0.9 V impedance spectroscopy proved to be useful to clearly determine the -1.0 V actual potential range over which true capacitive behavior is kept. -60 -1.1 V Figures 6b and 6c present the Nyquist and phase angle vs. frequency -1.2 V -2000 -40 plots, respectively of electrochemical impedance spectroscopy (EIS) measurements performed at different negative potential limits that -20 were used in the cyclic voltammetric experiments (Figure 6a). When the electrode potential is more positive than −1.0 V a capacitive be- -2 -1 0 1 2 3 02000 havior, demonstrated by an almost vertical line (Figure 6b)and a 10 10 10 10 10 10 phase angle of −80 at lower frequencies (Figure 6c), is observed. b) c) Re(Z) (Ohm) Frequency (Hz) The decrease of the imaginary impedance at 10 mHz when the po- Figure 6. a) Cyclic voltammograms of VN thin film electrode between its tential is changed from −0.8 to −1.0 V indicates a slight increase of open circuit potential and various more negative potential limits. The values the capacitance in agreement with the cyclic voltammetric data (Fig- at each lower potential limit represent the ratio of anodic and cathodic charge ure 6a). However, a drastic change in behavior is observed when the (Qa/Qc). b) Nyquist plots (1 kHz - 10 mHz) and c) Bode plots measured by VN thin film electrode is polarized at potential more negative than electrochemical impedance spectroscopy at these various negative potential −1.0 V. Indeed the Nyquist plot for a potential of −1.1 V presents limits. Measurements made in 1 M KOH under N atmosphere. a large semi-circle characteristic of a charge transfer process (Figure 6b). Moreover, the decrease of the diameter of the semi-circle for a potential of −1.2 V indicates a decrease of the charge transfer resis- It can be noted that the methodology used here can be applied to tance of an electrochemical reaction. The most likely electrochemical the investigation of any new electrode material for electrochemical process is the hydrogen evolution reaction that is taking place at the capacitor. In summary, when investigating a material as an electrode VN/1 M KOH interface. Figure 6c further confirms that the electrode for energy storage applications and when oxygen is present in the is losing its capacitive behavior when the potential is more negative electrolyte, the irreversible oxygen reduction reaction can take place. than −1.0 V. The CV and EIS data allow concluding that a negative In a charge/discharge cycle, the cathodic charge consumed by such potential limit of −1.0 V should be used when cycling a VN thin film irreversible reduction reaction could hide or minimize the presence of electrode in 1 M KOH. an irreversible oxidation reaction. This could result in a capacity loss due to oxidation/dissolution or passivation of the material even though Determination of a suitable upper potential limit.—The proce- the Coulombic efficiency is smaller than 1. Oxygen reduction reaction, dure used to determine the lower potential limit was also adopted to which can also participate to the self-discharge of the electrode should determine the positive potential limit. Figure 7a presents the cyclic be minimized as much as possible for energy storage application. The presence of oxygen can lead to the degradation of the electrode through corrosion process, as it seems to be the case for VN thin 1.59 0.2 film electrodes. Consequently, all further experiments were conducted 1.39 1.20 1.09 under nitrogen bubbling of the electrolyte in a closed cell. 0.84 0.95 0.99 0.98 0.1 Determination of a suitable lower potential limit.—Figure 6a 0.0 presents the different cyclic voltammograms obtained between the -0.1 OCP (−0.65 V) and various more negative potential limit values for the same electrode. The CV is still rectangular in shape down to a neg- -1.0 -0.8 -0.6 -0.4 -0.2 0.0 ative potential limit of −1.1 V, but a large irreversible cathodic current a) E vs. Hg/HgO (V) is measured for potentials more negative than −1.1 V, introducing a -0.7 V slight decrease of the Coulombic efficiency (94%). This current is -0.6 V -4000 -0.5 V most likely attributed to the irreversible hydrogen evolution reaction -80 -0.4 V -3000 (Eq. 2). -60 -0.3 V -0.2 V − − -2000 H O + 2e  H + 2OH [2] 2 2 -40 -0.1 V 0.0 V -1000 -20 Similarly to the oxygen reduction reaction, the charge consumed by this reaction is not recovered upon oxidation and is obviously not 0 useful for energy storage. Again, for the same reason, it could also hide -2 -1 0 1 2 3 0 2000 4000 10 10 10 10 10 10 or minimize the presence of an unwanted irreversible anodic reaction, b) c) Re(Z) (Ohm) Frequency (Hz) when the Coulombic efficiency is considered. Moreover, even if this reaction is known for its slow kinetics on some electrode material, it Figure 7. a) Cyclic voltammograms of VN thin film electrode between −1.0 will lead to self-discharge until its potential reaches the equilibrium V and various more positive potential limits (from −0.7to0V). Thevalues potential of the reaction for this particular electrode/electrolyte couple at each upper potential limit represent the ratio of anodic and cathodic charge (E =−0.931 V vs. Hg/HgO (1 M KOH)). At last, considering the (Qa/Qc). b) Nyquist plots (1 kHz - 10 mHz) and c) Bode plots measured by products of reaction, long term cycling experiments in a closed cell electrochemical impedance spectroscopy at these various negative potential could result in a pressure build up and a change of the electrolyte pH. limits. Measurements made in 1 M KOH under N atmosphere. -2 Im(Z) (Ohm) I (mA.cm ) Theta (°) Im(Z) (Ohm) -2 I (mA.cm ) Theta (°) Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) A1081 voltammograms obtained between −1.0 V and different positive po- tential limits for the same electrode. When cycled up to 0 V, a clear irreversible anodic process can be identified. The increase of the Qa/Qc ratio to value higher than 1 for each potential window where the up- per limit is more positive than −0.3 V is indicative of an irreversible anodic process. Moreover, currents recorded at a specific potential are different from one cycle to another. For example, currents recorded 0,6 0,6 A A i) ii) at −0.2 V for cycles with upper reversal potential of −0.2, −0.1 and 0,3 0,3 0,0 B 0,0 0 V diminish from one cycle to another even though currents recorded -0,3 -0,3 at −0.7 V are quite similar for the same cycles. This observation 5 -0,6 -0,6 suggests a passivation or at least a change in the film surface compo- -1,0 -0,8 -0,6 -0,4 -1,0 -0,8 -0,6 -0,4 E vs. Hg/HgO (V) E vs. Hg/HgO (V) sition. This is further confirmed by the appearance of an irreversible cathodic current between −0.9 and −1.0 V for reversal potential more 0 200 400 600 800 1000 positive than −0.3 V. However, the intensity of current of the redox a) Number of cycles peaks, centered at around 0.67 V depends on the positive potential limit when it is varied between −0.6 and −0.3 V. EIS measurements presented in Figures 7b and 7c show an obvious deviation from a capacitive behavior for potentials more positive than −0.1 V as demonstrated by a decrease of the phase angle at low frequency. This observation suggests the occurrence of an anodic reaction such as an oxidation of the VN surface. The slow decrease of the width of the semi-circle radius observed in Figure 7b when 0,6 the potential is more positive suggests a redox reaction with sluggish iii) -1.0V to -0.5V 0,3 kinetics (large charge transfer resistance). This could however explain 0,0 B -1.0V to -0.4V -0,3 why in most charge/discharge cycling experiments reported in the -1.0V to -0.3V -0,6 literature, a positive potential limit of 0 V is used and that irreversible 20 -1,0 -0,8 -0,6 -0,4 40–43 anodic processes are only observed at low cycling rate. However, E vs Hg/HgO (V) small deviation from the capacitive behavior can already be seen for potential as low as −0.5 V (Figure 7c). It is then difficult to determine 0 200 400 600 800 1000 b) from those EIS measurements which positive potential limit should be Number of cycles chosen in order to ensure long cycle life for the VN thin film electrode. Afterwards, the long-term cycling stability of VN thin films was in- Figure 8. a) Specific capacitance and b) relative capacity retention over 1000 −1 vestigated over different potential windows. Figures 8a and 8b present cyclic voltammetry cycles at 20 mV.s in between −1.0 V and three different −1 upper limits: −0.5 (), −0.4 (●)and −0.3 V (). The insets present the the evolution of capacitance over 1000 cycles carried out at 20 mV.s th first (A) and 1000 (B) cyclic voltammograms obtained for different potential in between −1.0 V and 3 selected positive potential limits (−0.5, −0.4 windows (from left to right, (i- [−1.0; −0.5]; ii- [−1.0; −0.4] and iii- [−1.0; and −0.3 V). A general loss of capacitance was observed over the first −0.3] V). 400 cycles irrespective of the potential window (Figure 8a). When cy- cled up to −0.4 or −0.5 V, the capacitance stabilized after 400 cycles, initial capacitance compared to the former electrode. This observation leading to 88% capacity retention after 1000 cycles. However, when suggests that the capacitance loss observed over the first 500 cycles, cycled up to −0.3 V, Figures 8a and 8b exhibit a continuous capac- th th when the electrode is cycled without any prior stabilization time, is itance loss between the 300 and the 1000 cycle with a capacity probably due to a process taking place when the electrode is in pres- retention of only 72% after 1000 cycles. The linear behavior of this ence of 1 M KOH, whether the electrode is cycled or not. One can capacitance fade suggests that a continuous dissolution or passivation suspect the presence of an unstable surface oxide that contributes to is taking place at the VN electrode every time its potential is more pos- the charge storage process. The dissolution of this surface oxide in the itive than −0.4 V. The general decrease in current density observed on 1 M KOH electrolyte would then lead to a capacitance loss. However, the cyclic voltammograms further corroborates this hypothesis (inset further experiments are needed to validate this hypothesis. In these Figure 8b). Analogously, oxidation and formation of soluble vana- experimental conditions, which are a constant nitrogen bubbling in the dium hydroxides have been suggested to explain fast capacitance loss 45,58 1 M KOH electrolyte, a 24 hours resting period prior to cycling and an when the potential limit extended towards too positive potentials. Cycling a VN electrode by using a positive potential limit of −0.5 or −0.4 V yields the same capacitance retention (Figure 8b). However, −2 thin films cycled between −1.0 and −0.4 V showed a 4 mF.cm capacitance enhancement due to an increase of the contribution of the redox peaks (≈−0.67 V, see insets Figure 8a). From these different observations, potential cycling between −1.0 and −0.4 V appears to be the best conditions to obtain a good stability and high capacitance in a 1 M KOH electrolyte. Cycling stability.—To further investigate the cycling stability, the VN thin film was tested for 10000 cycles between −1.0 and −0.4 V in 1 M KOH. Figure 9 shows that cycling in these conditions allowed the electrode to keep 74% of its initial capacitance. How- ever, a significant loss (13%) of capacitance takes place during the 0 2500 5000 7500 10000 first 500 cycles. This loss can either be due to either a degradation Cycle number process that occurred upon the first 500 cycles, or a process that took place once the VN thin film was in presence of the electrolyte. In Figure 9. Specific capacitance over 10000 cyclic voltammetry cycles at −1 order to discriminate between these two hypotheses, the capacitance 20 mV.s in 1 M KOH between −1.0 and −0.4 V, immediately after be- evolution of an electrode cycled after being kept in the electrolyte for ing exposed to the electrolyte (curve A) and after a stabilization time of 24 h 24 hours is also presented in Figure 9. This electrode shows a smaller (curve B), prior to cycling. -2 -2 Specific capacitance (mF.cm ) Capacity retention (%) Specific capacitance (mF.cm ) -2 -2 i (mA.cm ) i (mA.cm ) -2 i (mA.cm ) A1082 Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) optimized potential window ([−1.0; −0,4] V), an impressive capacity 21. T. Brousse, M. Toupin, R. Dugas, L. Athouel, O. Crosnier, and D. Belanger, ´ J. Electrochem. Soc., 153, A2171 (2006). retention of 96% was achieved over 10000 cycles. 22. D. Belanger, ´ T. Brousse, and J. W. Long, Electrochem. Soc. interface, 49 (2008). 23. S. Devaraj and N. Munichandraiah, J. Phys. Chem. C, 112, 4406 (2008). 24. O. Ghodbane, J.-L. Pascal, and F. Favier, ACS Appl. Mater. Interfaces, 1, 1130 (2009). Conclusions 25. J. P. Zheng, P. J. Cygan, and T. R. Jow, J. Electrochem. Soc., 142, 2699 (1995). 26. K. E. Swider-Lyons, C. T. Love, and D. R. Rolison, J. Electrochem. Soc., 152, C158 This study has clearly demonstrated that VN is active for the oxy- (2005). gen reduction reaction and the presence of oxygen in the electrolyte 27. K.-H. Chang and C.-C. Hu, J. Electrochem. Soc., 151, A958 (2004). should be avoided as well as the negative potential excursion should 28. W. Sugimoto, H. Iwata, Y. Yasunaga, Y. Murakami, and Y. Takasu, Angew. Chem. Int. Ed. Engl., 42, 4092 (2003). be limited in such a way to avoid the hydrogen evolution reaction. 29. M. R. Lukatskaya, O. Mashtalir, C. E. Ren, Y. Dall’Agnese, P. Rozier, P. L. Taberna, Moreover, in order to maintain a long cycle life, a VN thin film elec- M. Naguib, P. Simon, M. W. Barsoum, and Y. Gogotsi, Science, 341, 1502 trode should not be cycled to potential more positive than −0.4 V. As (2013). a result, cycling a VN thin film electrode between −1.0 and −0.4 V 30. Y. Dall’Agnese, M. R. Lukatskaya, K. M. Cook, P.-L. Taberna, Y. Gogotsi, and in 1 M KOH electrolyte after a prior 24 hours stabilization period re- P. Simon, Electrochem. Commun., 48, 118 (2014). −2 31. M. Wixom, L. Owens, J. Parker, J. Lee, I. Song, and L. Thompson, Electrochem. Soc. sulted in an initial capacitance of 19 mF.cm with a capacity retention Proc., 96-25, 63 (1997). of 96% after 10000 cycles. It is worth noting that such performance 32. L. Owens, L. T. Thompson, and M. R. Wixom, High surface area mesoporous desigel combined with the simple and reproducible PVD synthesis of VN thin materials and methods for their fabrication. US Patent 5837630, 1998. films make them attractive for electrochemical microsupercapacitor 33. C. Z. Deng and K. C. Tsai, Electrochem. Soc. Proc., 96-25, 75 (1997). 46,47,69,70 34. D. A. Evans and J. R. Miller Capacitor including a cathode having a nitride coating. applications. US patent 5754394, 1998. The VN pseudocapacitive behavior is thought to be due to sur- 35. L. T. Thompson, M. R. Wixom, and J. M. Parker, High surface area nitride, carbide face oxide redox processes. This has been further supported by sur- and boride electrodes and methods of fabrication thereof. US Patent 5680292, 1997. face characterization following cycling experiment between 0.0 and 36. D. Choi and P. N. Kumta, 2298 (2006). 41,58 37. D. Choi and P. N. Kumta, Electrochem. Solid-State Lett., 8, A418 (2005). −1.2 V. However, the fact that the VN was found to be unstable in 38. S. L. Roberson, D. Finello, and R. F. Davis, J. Appl. Electrochem., 29, 75 (1999). 1 M KOH for potential more positive than −0.4 V demonstrates that 39. T.-C. Liu, W. G. Pell, and B. E. Conway, J. Electrochem. Soc., 145, 1882 (1998). further investigation is needed to confirm that the observed surface 40. X. Zhou, H. Chen, D. Shu, C. He, and J. Nan, J. Phys. Chem. Solids, 70, 495 (2009). oxide was not just the result of some oxidation/degradation process. 41. D. Shu, C. Lv, F. Cheng, C. He, K. Yang, J. Nan, and L. Long, Int. J. Electrochem. Sci., 8, 1209 (2013). Additionally, in this study, the presence of oxygen was shown to 42. A. M. Glushenkov, D. Hulicova-jurcakova, D. Llewellyn, G. Q. Lu, and Y. Chen, shift the open circuit potential of a VN electrode to about 0 V by Chem. Mater., 22, 914 (2010). self-discharge. Consequently, in order to make sure that ex-situ char- 43. L. Zhang, C. M. B. Holt, E. J. Luber, B. C. Olsen, H. Wang, M. Danaie, X. Cui, acterization measurements are representative of the real state of the X. W. Tan, V. Lui, W. P. Kalisvaart, and D. Mitlin, J. Phys. Chem. C, 115, 24381 (2011). material in the electrolyte, characterization of the surface composition 44. P. Pande, P. G. Rasmussen, and L. T. Thompson, J. Power Sources, 207, 212 (2012). after cycling should be performed without exposing the VN electrode 45. R. L. Porto, R. Frappier, J. B. Ducros, C. Aucher, H. Mosqueda, S. Chenu, to ambient air. B. Chavillon, F. Tessier, F. Chevire, ´ and T. Brousse, Electrochim. Acta, 82, 257 (2012). 46. R. Lucio-Porto, S. Bouhtiyya, J. F. Pierson, A. Morel, F. Capon, P. Boulet, and Acknowledgments T. Brousse, Electrochim. Acta, 141, 203 (2014). 47. E. Eustache, R. Frappier, R. L. Porto, S. Bouhtiyya, J.-F. Pierson, and T. Brousse, The financial supports of the Natural Science and Engineering Electrochem. Commun., 28, 104 (2013). 48. C. M. Ghimbeu, E. Raymundo-Pinero, ˜ P. Fioux, F. Beguin, ´ and C. Vix-Guterl, J. Research Council of Canada (NSERC) as well as the French Ministry Mater. Chem., 21, 13268 (2011). of Education, Research and Technologie are gratefully acknowledged. 49. S. Dong, X. Chen, L. Gu, X. Zhou, H. Wang, Z. Liu, P. Han, J. Yao, L. Wang, G. Cui, Ecole Polytechnique de Montreal ´ (cm ), NanoQAM and CQMF are and L. Chen, Mater. Res. Bull., 46, 835 (2011). also acknowledged. 50. F. Cheng, C. He, D. Shu, H. Chen, J. Zhang, S. Tang, and D. E. Finlow, Mater. Chem. Phys., 131, 268 (2011). 51. P. J. Hanumantha, M. K. Datta, K. S. Kadakia, D. H. Hong, S. J. Chung, M. C. Tam, References J. A. Poston, A. Manivannan, and P. N. Kumta, J. Electrochem. Soc., 160, A2195 (2013). 1. A. G. Pandolfo and A. F. Hollenkamp, J. Power Sources, 157, 11 (2006). 52. X. Lu, T. Liu, T. Zhai, G. Wang, M. Yu, S. Xie, Y. Ling, C. Liang, Y. Tong, and Y. Li, 2. J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon, and P. L. Taberna, Science, Adv. Energy Mater., 4, 1300994 (2013). 313, 1760 (2006). 53. Z.-H. Gao, H. Zhang, G.-P. Cao, M.-F. Han, and Y.-S. Yang, Electrochim. Acta, 87, 3. C. Largeot, C. Portet, J. Chmiola, P. Taberna, Y. Gogotsi, and P. Simon, J. Am. Chem. 375 (2013). Soc., 130, 2730 (2008). 54. R. Wang, X. Yan, J. Lang, Z. Zheng, and P. Zhang, J. Mater. Chem. A, 2, 12724 4. S. A. Al-Muhtaseb and J. A. Ritter, Adv. Mater., 15, 101 (2003). (2014). 5. M. Sevilla, S. Alvarez, T. A. Centeno, A. B. Fuertes, and F. Stoeckli, Electrochim. 55. Y. Su and I. Zhitomirsky, J. Power Sources, 267, 235 (2014). Acta, 52, 3207 (2007). 56. Y. Yang, R. Kirchgeorg, R. Hahn, and P. Schmuki, Electrochem. Commun., 43,31 6. D. N. Futaba, K. Hata, T. Yamada, T. Hiraoka, Y. Hayamizu, Y. Kakudate, O. Tanaike, (2014). H. Hatori, M. Yumura, and S. Iijima, Nat. Mater., 5, 987 (2006). 57. M. Balogun, W. Qiu, W. Wang, P. Fang, X. Lu, and Y. Tong, J. Mater. Chem. A, 3, 7. S. Talapatra, S. Kar, S. K. Pal, R. Vajtai, L. Ci, P. Victor, M. M. Shaijumon, S. Kaur, 1364 (2015). O. Nalamasu, and P. M. Ajayan, Nat. Nanotechnol., 1, 112 (2006). 58. D. Choi, G. E. Blomgren, and P. N. Kumta, Adv. Mater., 18, 1178 (2006). 8. Y. Gogotsi, A. Nikitin, H. Ye, W. Zhou, J. E. Fischer, B. Yi, H. C. Foley, and 59. S. T. Oyama, The chemistry of transition metal carbides and nitrides; S. T. Oyama, M. W. Barsoum, Nat. Mater., 2, 591 (2003). Ed.; Chapman & Hall, 1996. 9. D. Belanger, ´ X. Ren, J. Davey, F. Uribe, and S. Gottesfeld, J. Electrochem. Soc., 147, 60. P. Krawiec, P. L. De Cola, R. Glaser, ¨ J. Weitkamp, C. Weidenthaler, and S. Kaskel, 2923 (2000). Adv. Mater., 18, 505 (2006). 10. C. K. Chiang, C. R. Fincher, Y. W. Park, A. J. Heeger, H. Shirakawa, E. J. Louis, 61. T. Onozuka, J. Appl. Cryst. 11, 132 (1978). S. C. Gau, and A. G. MacDiarmid, Phys. Rev. Lett. Lett., 39, 1098 (1977). 62. J. C. Caicedo, G. Zambrano, W. Aperador, L. Escobar-alarcon, and E. Camps, Appl. 11. M. Mastragostino, C. Arbizzani, R. Paraventi, and A. Zanelli, J. Electrochem. Soc., Surf. Sci., 258, 312 (2011). 147, 407 (2000). 63. K. Ait Aissa, A. Achour, J. Camus, L. Le Brizoual, P. Y. Jouan, and M. A. Djouadi, 12. A. Rudge, J. Davey, I. Raistrick, S. Gottesfeld, and J. P. Ferraris, J. Power Sources, Thin Solid Films, 550, 264 (2014). 47, 89 (1994). 64. B. D. Choi, G. E. Blomgren, and P. N. Kumta, 1178 (2006). 13. T. Brousse, D. Belanger, and J. W. Long, J. Electrochem. Soc., 162, A5185 (2015). 65. W.-F. Chen, J. T. Muckerman, and E. Fujita, Chem. Commun., 49, 8896 (2013). 14. B. E. Conway, Electrochemical Supercapacitors: Scientific Fundamentals and Tech- 66. C. Pozo-Gonzalo, O. Kartachova, A. A. J. Torriero, P. C. Howlett, A. M. Glushenkov, nological Applications; Kluwer, 1999. D. M. Fabijanic, Y. Chen, S. Poissonnet, and M. Forsyth, Electrochim. Acta, 103, 151 15. S. Trasatti and G. Buzzanca, J. Electroanal. Chem., 29, A1 (1971). (2013). 16. R. Fu, Z. Ma, and J. P. Zheng, J. Phys. Chem. B, 106, 3592 (2002). 67. M. Yang, A. J. Allen, M. T. Nguyen, W. T. Ralston, M. J. MacLeod, and F. J. DiSalvo, 17. W. Dmowski, T. Egami, K. E. Swider-lyons, C. T. L. Love, and D. R. Rolison, J. J. Solid State Chem., 205, 49 (2013). Phys. Chem. B, 106, 12677 (2002). 68. T.-H. Wu, Y.-H. Chu, C.-C. Hu, and L. J. Hardwick, Electrochem. Commun., 27,81 18. T. Lui, W. G. Pell, and B. E. Conway, Electrochim. Acta, 42, 3541 (1997). (2013). 19. H. Y. Lee and J. B. Goodenough, J. Solid State Chem., 148, 81 (1999). 69. M. Beidaghi and Y. Gogotsi, Energy Environ. Sci., 7, 867 (2014). 20. M. Toupin, T. Brousse, and D. Belanger, Chem. Mater., 16, 3184 (2004). 70. M. Brunet and D. Pech, Techniques de l’ingenieur ´ 2012, Re210. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of the Electrochemical Society IOP Publishing

Suitable Conditions for the Use of Vanadium Nitride as an Electrode for Electrochemical Capacitor

Download PDF
6 pages

Loading next page...
 
/lp/iop-publishing/suitable-conditions-for-the-use-of-vanadium-nitride-as-an-electrode-9myd2WipRV

References (65)

Copyright
Copyright © The Author(s) 2016. Published by ECS.
ISSN
0013-4651
eISSN
1945-7111
DOI
10.1149/2.1221606jes
Publisher site
See Article on Publisher Site

Abstract

Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) A1077 Suitable Conditions for the Use of Vanadium Nitride as an Electrode for Electrochemical Capacitor a,b,c a a,b,d,e a,b,∗ Alban Morel, Yann Borjon-Piron, Raul ´ Lucio Porto, Thierry Brousse, c,∗,z and Daniel Belanger ´ Institut des Materiaux ´ Jean Rouxel (IMN), Universite ´ de Nantes, CNRS, 44322 Nantes Cedex 3, France Reseau sur le Stockage Electrochimique de l’Energie (RS2E), FR CNRS 3459, France ´ ´ ´ ` ´ ´ ´ Departement Chimie, UniversiteduQuebec a Montreal, Succursale Centre-Ville, Montreal, Quebec H3C 3P8, Canada ´ ´ ´ ´ ´ ´ Universidad Autonoma de Nuevo Leon, Facultad de Ingenierıa Mecanica y Electrica, San Nicolas de los Garza, 66450 Nuevo Leon, ´ Mexico ´ Universidad Autonoma ´ de Nuevo Leon, ´ Centro de Innovacion, ´ Investigacion ´ y Desarrollo en Ingenier´ ıa y Tecnolog´ ıa, Apodaca, 66600 Nuevo Leon, ´ Mexico ´ Vanadium nitride has displayed many interesting characteristics for its use as a pseudocapacitive electrode in an electrochemical capacitor, such as good electronic conductivity, good thermal stability, high density and high specific capacitance. Thin films of VN were prepared by D.C. reactive magnetron sputtering. The electrochemical stability of the films as well as the influence of dissolved oxygen in 1 M KOH electrolyte were investigated. In order to avoid material as well as electrolyte degradation, it was concluded that vanadium nitride should only be cycled between −0.4 and −1.0 V vs. Hg/HgO. After a 24 hours stabilization period, the prepared −2 VN thin film showed an initial capacitance of 19 mF.cm and a capacity retention of 96% after 10000 cycles. Furthermore, dissolved oxygen in the electrolyte was demonstrated to cause self-discharge up to a potential above −0.4 V vs. Hg/HgO, where VN was shown to be unstable. Additionally, the presence of oxygen was shown to shift the open circuit potential of a VN electrode to about 0 V through self-discharge processes. © The Author(s) 2016. Published by ECS. This is an open access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives 4.0 License (CC BY-NC-ND, http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is not changed in any way and is properly cited. For permission for commercial reuse, please email: [email protected]. [DOI: 10.1149/2.1221606jes] All rights reserved. Manuscript submitted February 11, 2016; revised manuscript received March 14, 2016. Published March 29, 2016. Electrochemical capacitors are currently being developed to com- [−1.2; 0.0] V to [−1.2; −0.3] V vs. Hg/HgO enhanced the capacity plement other energy storage or conversion systems such as batter- retention after 1000 charge/discharge cycles. Further improvements ies and fuel cells. In the past two decades, several electrode mate- were obtained when changing the electrolyte pH from 14 to 12. rials have been investigated. The mostly studied materials include When charging/discharging VN electrodes between −1.2 and −0.4 V 1–8 9–12 13 carbons, conducting polymers as well as pseudocapacitive instead of −1.2 and 0 V vs. Hg/HgO for 2800 cycles in 1 M KOH, 14–18 46 transition metal oxides such as ruthenium dioxide and manganese Porto et al. also found enhancement of the capacity retention from 0 19–24 52 dioxide. In the effort to improve the performance of electrochem- to 70%, respectively. To our knowledge, only Lu et al. studied VN ical capacitors, a widely used approach has been to develop synthetic over a relatively large number of charge/discharge cycles (15000 cy- methods to develop nanomaterials that are believed to lead to increased cles). They have shown that when cycling VN up to 0.1 V vs. Hg/HgO 14,23,25–28 energy density and higher rate capability. in 1 M KOH, most of the capacitance was lost after a few thousands Another approach has been to investigate the charge storage prop- cycles. Interestingly, they also showed that when coating VN with car- erties of novel materials. Accordingly, several research groups consid- bon, the electrode kept up to 88% of its initial capacitance after 15000 29,30 31–37 ered materials such as MXenes and transition metal nitrides. cycles in 1 M KOH. All these studies highlight the need of in-depth In the latter class of compounds, molybdenum nitride was firstly investigations of the suitable conditions for which VN is stable as an 32,33,35,38,39 investigated and in the past decade a great deal of atten- active material for energy storage applications. tion has focused on other nitrides with an intensive focus on vana- Most studies use VN powders prepared via ammonolysis of vari- 37,40–58 40,42,44,45,48,49,53,60 37,58 dium nitride as a consequence of the impressive capacitance of ous vanadium oxides or chloride. As pointed out −1 58 46 1340 F.g reported for nanosized VN particles in 1 M KOH. In- by Porto et al., this gives rise to two main drawbacks: (1) the use deed, VN is an interesting electrode material due to this high reported of different precursors especially oxides may lead to difference in specific capacitance coupled with its close to metallic electronic con- compositions and especially oxygen content in the VN powders and −1 59 −3 ductivity (1.18 S.m ), high density (6.13 g.cm ) and high melting (2) the use of powders imply the use of carbon and polymer addi- point (2619 K). tives in order to prepare composite electrodes, thus preventing the The hypothesis proposed by Choi et al. to explain such a high study of VN intrinsic electrochemical properties. On the other hand, specific capacitance of VN, is that, in addition to electrochemi- thin film deposition techniques such as DC sputtering lead to the cal double layer capacitance, successive fast reversible redox reac- synthesis of VN thin films using only vanadium metal and nitro- tions are taking place, involving surface oxide groups and OH ions gen which can be used as-deposited as electrode material without from the electrolyte. This hypothesis was supported by the observa- the need of binder neither conductive additives. Additionally, tuning 58 41,58 tion of surface oxide via ex-situ XPS and FTIR post cycling the deposition conditions can lead to a wide range of morphologies, measurements. from porous to dense, and to the growth of randomly or preferen- While most studies concentrated on new synthesis of VN with the tially oriented thin films. Furthermore, the use of current collector goal to enhance the specific power density and the specific capaci- is not necessary since VN thin films exhibit high electronic con- tance, which was found in most studies to be limited in the range ductivity, and thus can be used both as active materials and current −1 40,42–46,48–52 of 150 to 300 F.g , only few studies looked at cycling collector. parameters-cycle life relationship. Choi et al. showed that diminish- Herein, we report preparation of VN thin film working electrodes ing the potential window over which the VN electrode is cycled from deposited on soda lime glass by D.C. sputtering and the study of their electrochemical properties by a general procedure aimed at determin- ing the experimental conditions to obtain a stable electrode for use in Electrochemical Society Member. an electrochemical capacitor. E-mail: [email protected] A1078 Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) Experimental Thin film preparation.—VN thin films were prepared by D.C reac- tive sputtering using the AC450 sputtering system from Alliance Con- cept. Prior to deposition, soda lime glass substrates were degreased with successive washing with acetone and ethanol. The deposition −4 chamber was first vacuumed to a pressure of 10 Pa. The substrates 4000 were then cleaned in situ with argon sputter etching (6 Pa, 50 sccm, 100 W RF) for 3 min. In order to ensure good adhesion of the VN thin film to the glass substrate, a titanium layer was first sputtered from a 2 diameter titanium disk of 99.95% purity, in an argon plasma B (1 Pa, 50 sccm, 50 W DC), with a target to substrate distance of 7.5 cm. The VN was then sputtered from a 2 diameter vanadium disk of 99.5% purity, with an applied power to the target of 50 W DC, in a gas 30 40 50 60 70 80 90 mixture of Ar/N (with fixed flows of 30/2.5 sccm respectively) for 2θ (°) Cu a total chamber pressure of 1 Pa. Once again, the target-to-substrate distance was fixed to 7.5 cm. The deposition time was changed to Figure 1. X-ray diffraction pattern of a 60 nm titanium adhesive layer (curve control the film thickness. Unless otherwise stipulated, all prepared A) and of 140 (curve B) and 280 nm (Curve C) thick layer of VN on top of that adhesive layer. Curves D and E present the calculated diffraction pattern for thin films are composed of a 140 nm thick VN layer deposited on top the 280 nm thick VN layer and the difference between the measured and the of a 60 nm thick titanium adhesive layer. calculated pattern obtained from FullProf software, respectively. The (◦)and () symbols indicate diffraction peaks attributed to the VN and the Ti phases, Thin film characterization.—The crystallographic structure of the respectively. as deposited material was investigated by X-ray diffraction with an XRD using PANalytical’s X’Pert Pro diffractometer with Cu Kα ra- diation (λ = 1.5418 Å). Lattice parameter of the VN phase was then sputtering and could be explained by a compressive stress related to calculated from the refinement of the Bragg positions through the 62,63 the deposition conditions. use of FullProf software. The morphological aspects of the thin films Representative SEM images of VN thin films are shown in were observed using FEM-SEM Zeiss Merlin scanning electron mi- Figure 2. The VN film exhibits columnar growth on the glass croscope. substrate (Figure 2a) with an average column diameter of 20 nm (Figure 2b). A similar morphology has been previously observed for Electrochemical characterization.—The electrochemical proper- the same type of deposition process. Moreover, the irregular colum- ties of the thin films were investigated with a Solartron multipoten- nar structure observed in the cross section view (Figure 2a) as well tiostat 1470 coupled to a Solartron frequency analyzer 1255B. All as the observation of some pores in the top view image (Figure 2b) measurements were carried out in a closed three-electrode cell with a suggest the presence of voids in between columns. platinum wire as counter electrode, a Hg/HgO (1 M KOH) electrode as reference electrode and a VN thin film as working electrode and Electrochemical characterization.—Figure 3 depicts cyclic current collector. For ease of reading, all potentials mentioned in the voltammograms for Ti/VN and Ti electrodes in 1 M KOH without text refer to the Hg/HgO (1 M KOH) electrode potential. The cell was any nitrogen bubbling of the electrolyte. A main redox system cen- designed so that the VN surface area in contact with the electrolyte tered at −0.67 V is observed on top of a capacitive envelope for the was limited to 0.502 cm via the use of a gasket. O or N gas was 2 2 VN electrode. Another set of redox waves with an average potential bubbled in the aqueous 1 M KOH electrolyte for 40 min prior each of −1.00 V for the anodic and cathodic peak potentials is also no- experiment. In the case of experiments with N gas, bubbling was also ticed but only for the VN electrode after 20 cycles. These two redox kept during the measurements. In the case of experiments with O gas, the gas stream was only kept above the solution during measurements. Prior to every EIS measurement, the electrode potential was held for 5 min at the desired potential so that the measurement would be representative of the steady state of the system. Nyquist and phase an- gle vs. frequency plots were then obtained by scanning the frequency from 1 kHz to 10 mHz, using a sinusoidal signal of 10 mV amplitude. Results and Discussion Structural characterization.—XRD patterns of a titanium adhe- sive layer and vanadium nitride thin film of two different thicknesses (140 and 280 nm) deposited on the titanium layer are presented in Fig- ure 1. Comparing the XRD patterns of the samples with and without the vanadium nitride layer enables to distinguish the titanium layer diffraction peaks from those of vanadium nitride. Five diffractions ◦ ◦ ◦ ◦ ◦ peaks observed at 2 value of 38.0 , 44.2 64.3 , 77.2 and 81.3 can be attributed to the vanadium nitride thin film and can be ascribed to the (111), (200), (220), (311) and (222) planes of its cubic crystal system (Fm-3m [225]), respectively. Calculation of the lattice param- eter, based on the peak positions observed on the XRD pattern of the thicker VN film which presents slightly more defined peaks, results in a unit-cell parameter of a = 4.094(6) Å. The cell parameter for the stoichiometric bulk compound being 4.1392 Å (JCPDS file 35-768), this could reveal nitrogen vacancies in the as-deposited vanadium nitride, with a composition of VN . However, a lower lattice pa- 0.84 Figure 2. SEM images of a 280 nm VN/160 nm titanium bilayer thin film rameter has already been reported for VN thin films prepared by D.C. deposited on soda lime glass substrate a) cross section, b) top view. [111] [200] [220] [311] [222] Intensity (arbitrary units) Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) A1079 0,0 0.4 0.2 -0,2 0.0 -0,4 -0.2 -0,6 -0,8 -0.4 -1,0 -0.6 0 10000 20000 30000 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Time / s E vs. Hg/HgO (V) Figure 5. Open-circuit potential decay of a VN thin film following polar- Figure 3. Cyclic voltammograms of a VN thin film (cycle #2 (curve A) and −1 ization at −1.0 and −0.5 V for 5 min. under N (curves A and B) and O cycle #20 (curve B)) and of the adhesive titanium layer (curve C) at 20 mV.s 2 2 (curve C) atmospheres. in non-degassed 1 M KOH electrolyte. systems have already been observed at similar or slightly more positive the presence of oxygen in the electrolyte. Prior to cyclic voltammetry 41,43,58 potentials. Another smaller redox system centered at −0.275 V, experiments, the electrolyte was bubbled with either N or O gas 2 2 41,58 which is not clearly observed here, is occasionally seen. It has for40min.AlsowhenN was replaced by O , the potential of the 2 2 been suggested that these redox systems result from successive tran- electrode was held at −0.5 V in order to avoid any reaction that could +5 +2 58 sitions from V to V of vanadium oxides surface sites. These occur if the OCP would shift to value outside of the −0.5 to −1.0 V hypotheses have been supported by the observation of vanadium ox- interval. Figure 4 depicts the cyclic voltammograms obtained under ide by surface analysis following electrochemical cycling between 0 N and O . Under N , the CV shows a capacitive behavior and redox 2 2 2 41,64 and −1.2 V in 1 M KOH. However, the technical challenge that waves centered at −0.67 V. Anodic and cathodic charges (Q and Q , a c represents an in-situ investigation of such a system makes it difficult respectively) evaluated by integration of the cyclic voltammogram to assign each peak to a specific redox process. Moreover, the loss are very similar with values of 7.7 and 7.8 mC/cm , respectively. This th of electroactivity between the second and 20 cyclic voltammogram relatively good Coulombic efficiency (99%), as well as the absence suggests that the VN thin film is not stable in these conditions. At this of any tail shape at the most positive and negative potential limits moment, it is difficult to explain such a loss during only 20 cycles. of the cyclic voltammogram suggest the absence of any irreversible Figure 3 also shows an irreversible cathodic reaction for potentials processes. Thus, the larger cathodic charge, relative to the anodic more negative than −1.0 V that can be tentatively attributed to the charge, sometimes observed in the literature for VN electrodes can onset of the hydrogen evolution reaction (vide infra). However, by presumably be attributed to the presence of traces of oxygen in the 42,43 decreasing the potential window (between −1.0 and −0.5 V), sta- electrolyte. In contrast, only cathodic currents are observed for the −1 ble cyclic voltammograms can be obtained at 1 mV.s (Figure 4). cyclic voltammogram measured under O . This behavior reveals the These conditions will then be used as starting cycling conditions in irreversible oxygen reduction reaction due to presence of oxygen in order to study the influence of oxygen dissolved in the electrolyte, the solution (Eq. 1). Bubbling oxygen in the electrolyte was performed prior to determine the optimal potential window over which VN elec- only to emphasize the electrochemical response of this reaction on a trodes could be operated. Furthermore, one can notice the negligible VN electrode. However, one should realize that such irreversible reac- contribution of the titanium adhesive layer to the measured currents tion will contribute a cathodic charge (eg. in the presence of traces of (Figure 3,curve C). oxygen) that will counterbalance and hide the effect of any irreversible anodic process, when evaluating the Coulombic efficiency. Influence of oxygen in the electrolyte.—Since metal nitrides − − O + 2H O + 4e  4OH [1] 2 2 are now widely investigated for water splitting (eg. the hydrogen 65 66 The influence of oxygen can also be observed when comparing the evolution and oxygen reduction reactions ), the effect of oxygen evolution of the open-circuit potential (OCP) of the electrode in pres- on the electrochemical behavior of VN was investigated. More specif- ence of N and O (Figure 5). In this experiment, the electrode po- ically, this section aimed to emphasize and demonstrate the effect of 2 2 tential was firstly cycled between −0.5 and −1.0 V and polarized at either −1.0 or −0.5 V prior to monitoring the evolution of the OCP for 8 hours. The measurements made under nitrogen show a slow de- 0,02 cay/recovery of the thin film OCP from −1.00 to −0.76 V and −0.50 to −0.65 V. On the other hand, in the presence of oxygen, a very 0,00 fast decay of potential is observed and reached −0.10 V in less than 25 min and −0.01 V after 8 hours regardless of the initial potential. This drastic self-discharge is another example of the influence of the -0,02 oxygen reduction reaction (Eq. 1) taking place at the electrode until its potential reaches the equilibrium potential of that reaction (E = -0,04 0.298 V vs. Hg/HgO (1 M KOH)), or until an equilibrium is reached between this reduction reaction and an oxidation reaction (e.g., cor- rosion process through oxidation of the material and reduction of -0,06 oxygen). Following an additional 24 hours immersion in presence of oxygen, the degradation of the thin film was visually observed. The -1,0 -0,9 -0,8 -0,7 -0,6 -0,5 fact that the OCP does not reach 0.298 V and that degradation is ob- E vs. Hg/HgO (V) served corroborate the hypothesis of a corrosion process in presence of oxygen. Moreover, a significant corrosion rate has been reported Figure 4. Cyclic voltammograms of a VN thin film under N (curve A) and ◦ 67 −1 O (curve B) atmospheres at 1 mV.s in 1 M KOH. for VN in NaOH at 80 C. -2 -2 I (mA.cm ) I (mA.cm ) E / V vs. Hg/HgO A1080 Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) The cyclic voltammograms and the anodic to cathodic charge ra- 0,1 tio (Qa/Qc) recorded for negative potential limits down to −1.1 V seems to suggest that the hydrogen evolution reaction (HER) is not 0,0 occurring at −1.1 V. Hence, electrolytic decomposition can be pos- 0.97 sibly avoided at VN electrode by arbitrarily restricting the negative -0,1 0.98 0.98 0.97 potential limit to value more positive than −0.95 V. However, de- pending on the kinetics of the HER the data of Figure 6a alone might -0,2 0.94 be insufficient to rule out the possibility that such reaction is tak- -1,2 -1,1 -1,0 -0,9 -0,8 -0,7 -0,6 ing place at this potential. Recently, Wu et al. showed that in situa- a) E vs. Hg/HgO (V) tion where cyclic voltammetric and galvanostatic cycling experiments -0.8 V would suggest that a system has a capacitive behavior, electrochemical -4000 -80 -0.9 V impedance spectroscopy proved to be useful to clearly determine the -1.0 V actual potential range over which true capacitive behavior is kept. -60 -1.1 V Figures 6b and 6c present the Nyquist and phase angle vs. frequency -1.2 V -2000 -40 plots, respectively of electrochemical impedance spectroscopy (EIS) measurements performed at different negative potential limits that -20 were used in the cyclic voltammetric experiments (Figure 6a). When the electrode potential is more positive than −1.0 V a capacitive be- -2 -1 0 1 2 3 02000 havior, demonstrated by an almost vertical line (Figure 6b)and a 10 10 10 10 10 10 phase angle of −80 at lower frequencies (Figure 6c), is observed. b) c) Re(Z) (Ohm) Frequency (Hz) The decrease of the imaginary impedance at 10 mHz when the po- Figure 6. a) Cyclic voltammograms of VN thin film electrode between its tential is changed from −0.8 to −1.0 V indicates a slight increase of open circuit potential and various more negative potential limits. The values the capacitance in agreement with the cyclic voltammetric data (Fig- at each lower potential limit represent the ratio of anodic and cathodic charge ure 6a). However, a drastic change in behavior is observed when the (Qa/Qc). b) Nyquist plots (1 kHz - 10 mHz) and c) Bode plots measured by VN thin film electrode is polarized at potential more negative than electrochemical impedance spectroscopy at these various negative potential −1.0 V. Indeed the Nyquist plot for a potential of −1.1 V presents limits. Measurements made in 1 M KOH under N atmosphere. a large semi-circle characteristic of a charge transfer process (Figure 6b). Moreover, the decrease of the diameter of the semi-circle for a potential of −1.2 V indicates a decrease of the charge transfer resis- It can be noted that the methodology used here can be applied to tance of an electrochemical reaction. The most likely electrochemical the investigation of any new electrode material for electrochemical process is the hydrogen evolution reaction that is taking place at the capacitor. In summary, when investigating a material as an electrode VN/1 M KOH interface. Figure 6c further confirms that the electrode for energy storage applications and when oxygen is present in the is losing its capacitive behavior when the potential is more negative electrolyte, the irreversible oxygen reduction reaction can take place. than −1.0 V. The CV and EIS data allow concluding that a negative In a charge/discharge cycle, the cathodic charge consumed by such potential limit of −1.0 V should be used when cycling a VN thin film irreversible reduction reaction could hide or minimize the presence of electrode in 1 M KOH. an irreversible oxidation reaction. This could result in a capacity loss due to oxidation/dissolution or passivation of the material even though Determination of a suitable upper potential limit.—The proce- the Coulombic efficiency is smaller than 1. Oxygen reduction reaction, dure used to determine the lower potential limit was also adopted to which can also participate to the self-discharge of the electrode should determine the positive potential limit. Figure 7a presents the cyclic be minimized as much as possible for energy storage application. The presence of oxygen can lead to the degradation of the electrode through corrosion process, as it seems to be the case for VN thin 1.59 0.2 film electrodes. Consequently, all further experiments were conducted 1.39 1.20 1.09 under nitrogen bubbling of the electrolyte in a closed cell. 0.84 0.95 0.99 0.98 0.1 Determination of a suitable lower potential limit.—Figure 6a 0.0 presents the different cyclic voltammograms obtained between the -0.1 OCP (−0.65 V) and various more negative potential limit values for the same electrode. The CV is still rectangular in shape down to a neg- -1.0 -0.8 -0.6 -0.4 -0.2 0.0 ative potential limit of −1.1 V, but a large irreversible cathodic current a) E vs. Hg/HgO (V) is measured for potentials more negative than −1.1 V, introducing a -0.7 V slight decrease of the Coulombic efficiency (94%). This current is -0.6 V -4000 -0.5 V most likely attributed to the irreversible hydrogen evolution reaction -80 -0.4 V -3000 (Eq. 2). -60 -0.3 V -0.2 V − − -2000 H O + 2e  H + 2OH [2] 2 2 -40 -0.1 V 0.0 V -1000 -20 Similarly to the oxygen reduction reaction, the charge consumed by this reaction is not recovered upon oxidation and is obviously not 0 useful for energy storage. Again, for the same reason, it could also hide -2 -1 0 1 2 3 0 2000 4000 10 10 10 10 10 10 or minimize the presence of an unwanted irreversible anodic reaction, b) c) Re(Z) (Ohm) Frequency (Hz) when the Coulombic efficiency is considered. Moreover, even if this reaction is known for its slow kinetics on some electrode material, it Figure 7. a) Cyclic voltammograms of VN thin film electrode between −1.0 will lead to self-discharge until its potential reaches the equilibrium V and various more positive potential limits (from −0.7to0V). Thevalues potential of the reaction for this particular electrode/electrolyte couple at each upper potential limit represent the ratio of anodic and cathodic charge (E =−0.931 V vs. Hg/HgO (1 M KOH)). At last, considering the (Qa/Qc). b) Nyquist plots (1 kHz - 10 mHz) and c) Bode plots measured by products of reaction, long term cycling experiments in a closed cell electrochemical impedance spectroscopy at these various negative potential could result in a pressure build up and a change of the electrolyte pH. limits. Measurements made in 1 M KOH under N atmosphere. -2 Im(Z) (Ohm) I (mA.cm ) Theta (°) Im(Z) (Ohm) -2 I (mA.cm ) Theta (°) Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) A1081 voltammograms obtained between −1.0 V and different positive po- tential limits for the same electrode. When cycled up to 0 V, a clear irreversible anodic process can be identified. The increase of the Qa/Qc ratio to value higher than 1 for each potential window where the up- per limit is more positive than −0.3 V is indicative of an irreversible anodic process. Moreover, currents recorded at a specific potential are different from one cycle to another. For example, currents recorded 0,6 0,6 A A i) ii) at −0.2 V for cycles with upper reversal potential of −0.2, −0.1 and 0,3 0,3 0,0 B 0,0 0 V diminish from one cycle to another even though currents recorded -0,3 -0,3 at −0.7 V are quite similar for the same cycles. This observation 5 -0,6 -0,6 suggests a passivation or at least a change in the film surface compo- -1,0 -0,8 -0,6 -0,4 -1,0 -0,8 -0,6 -0,4 E vs. Hg/HgO (V) E vs. Hg/HgO (V) sition. This is further confirmed by the appearance of an irreversible cathodic current between −0.9 and −1.0 V for reversal potential more 0 200 400 600 800 1000 positive than −0.3 V. However, the intensity of current of the redox a) Number of cycles peaks, centered at around 0.67 V depends on the positive potential limit when it is varied between −0.6 and −0.3 V. EIS measurements presented in Figures 7b and 7c show an obvious deviation from a capacitive behavior for potentials more positive than −0.1 V as demonstrated by a decrease of the phase angle at low frequency. This observation suggests the occurrence of an anodic reaction such as an oxidation of the VN surface. The slow decrease of the width of the semi-circle radius observed in Figure 7b when 0,6 the potential is more positive suggests a redox reaction with sluggish iii) -1.0V to -0.5V 0,3 kinetics (large charge transfer resistance). This could however explain 0,0 B -1.0V to -0.4V -0,3 why in most charge/discharge cycling experiments reported in the -1.0V to -0.3V -0,6 literature, a positive potential limit of 0 V is used and that irreversible 20 -1,0 -0,8 -0,6 -0,4 40–43 anodic processes are only observed at low cycling rate. However, E vs Hg/HgO (V) small deviation from the capacitive behavior can already be seen for potential as low as −0.5 V (Figure 7c). It is then difficult to determine 0 200 400 600 800 1000 b) from those EIS measurements which positive potential limit should be Number of cycles chosen in order to ensure long cycle life for the VN thin film electrode. Afterwards, the long-term cycling stability of VN thin films was in- Figure 8. a) Specific capacitance and b) relative capacity retention over 1000 −1 vestigated over different potential windows. Figures 8a and 8b present cyclic voltammetry cycles at 20 mV.s in between −1.0 V and three different −1 upper limits: −0.5 (), −0.4 (●)and −0.3 V (). The insets present the the evolution of capacitance over 1000 cycles carried out at 20 mV.s th first (A) and 1000 (B) cyclic voltammograms obtained for different potential in between −1.0 V and 3 selected positive potential limits (−0.5, −0.4 windows (from left to right, (i- [−1.0; −0.5]; ii- [−1.0; −0.4] and iii- [−1.0; and −0.3 V). A general loss of capacitance was observed over the first −0.3] V). 400 cycles irrespective of the potential window (Figure 8a). When cy- cled up to −0.4 or −0.5 V, the capacitance stabilized after 400 cycles, initial capacitance compared to the former electrode. This observation leading to 88% capacity retention after 1000 cycles. However, when suggests that the capacitance loss observed over the first 500 cycles, cycled up to −0.3 V, Figures 8a and 8b exhibit a continuous capac- th th when the electrode is cycled without any prior stabilization time, is itance loss between the 300 and the 1000 cycle with a capacity probably due to a process taking place when the electrode is in pres- retention of only 72% after 1000 cycles. The linear behavior of this ence of 1 M KOH, whether the electrode is cycled or not. One can capacitance fade suggests that a continuous dissolution or passivation suspect the presence of an unstable surface oxide that contributes to is taking place at the VN electrode every time its potential is more pos- the charge storage process. The dissolution of this surface oxide in the itive than −0.4 V. The general decrease in current density observed on 1 M KOH electrolyte would then lead to a capacitance loss. However, the cyclic voltammograms further corroborates this hypothesis (inset further experiments are needed to validate this hypothesis. In these Figure 8b). Analogously, oxidation and formation of soluble vana- experimental conditions, which are a constant nitrogen bubbling in the dium hydroxides have been suggested to explain fast capacitance loss 45,58 1 M KOH electrolyte, a 24 hours resting period prior to cycling and an when the potential limit extended towards too positive potentials. Cycling a VN electrode by using a positive potential limit of −0.5 or −0.4 V yields the same capacitance retention (Figure 8b). However, −2 thin films cycled between −1.0 and −0.4 V showed a 4 mF.cm capacitance enhancement due to an increase of the contribution of the redox peaks (≈−0.67 V, see insets Figure 8a). From these different observations, potential cycling between −1.0 and −0.4 V appears to be the best conditions to obtain a good stability and high capacitance in a 1 M KOH electrolyte. Cycling stability.—To further investigate the cycling stability, the VN thin film was tested for 10000 cycles between −1.0 and −0.4 V in 1 M KOH. Figure 9 shows that cycling in these conditions allowed the electrode to keep 74% of its initial capacitance. How- ever, a significant loss (13%) of capacitance takes place during the 0 2500 5000 7500 10000 first 500 cycles. This loss can either be due to either a degradation Cycle number process that occurred upon the first 500 cycles, or a process that took place once the VN thin film was in presence of the electrolyte. In Figure 9. Specific capacitance over 10000 cyclic voltammetry cycles at −1 order to discriminate between these two hypotheses, the capacitance 20 mV.s in 1 M KOH between −1.0 and −0.4 V, immediately after be- evolution of an electrode cycled after being kept in the electrolyte for ing exposed to the electrolyte (curve A) and after a stabilization time of 24 h 24 hours is also presented in Figure 9. This electrode shows a smaller (curve B), prior to cycling. -2 -2 Specific capacitance (mF.cm ) Capacity retention (%) Specific capacitance (mF.cm ) -2 -2 i (mA.cm ) i (mA.cm ) -2 i (mA.cm ) A1082 Journal of The Electrochemical Society, 163 (6) A1077-A1082 (2016) optimized potential window ([−1.0; −0,4] V), an impressive capacity 21. T. Brousse, M. Toupin, R. Dugas, L. Athouel, O. Crosnier, and D. Belanger, ´ J. Electrochem. Soc., 153, A2171 (2006). retention of 96% was achieved over 10000 cycles. 22. D. Belanger, ´ T. Brousse, and J. W. Long, Electrochem. Soc. interface, 49 (2008). 23. S. Devaraj and N. Munichandraiah, J. Phys. Chem. C, 112, 4406 (2008). 24. O. Ghodbane, J.-L. Pascal, and F. Favier, ACS Appl. Mater. Interfaces, 1, 1130 (2009). Conclusions 25. J. P. Zheng, P. J. Cygan, and T. R. Jow, J. Electrochem. Soc., 142, 2699 (1995). 26. K. E. Swider-Lyons, C. T. Love, and D. R. Rolison, J. Electrochem. Soc., 152, C158 This study has clearly demonstrated that VN is active for the oxy- (2005). gen reduction reaction and the presence of oxygen in the electrolyte 27. K.-H. Chang and C.-C. Hu, J. Electrochem. Soc., 151, A958 (2004). should be avoided as well as the negative potential excursion should 28. W. Sugimoto, H. Iwata, Y. Yasunaga, Y. Murakami, and Y. Takasu, Angew. Chem. Int. Ed. Engl., 42, 4092 (2003). be limited in such a way to avoid the hydrogen evolution reaction. 29. M. R. Lukatskaya, O. Mashtalir, C. E. Ren, Y. Dall’Agnese, P. Rozier, P. L. Taberna, Moreover, in order to maintain a long cycle life, a VN thin film elec- M. Naguib, P. Simon, M. W. Barsoum, and Y. Gogotsi, Science, 341, 1502 trode should not be cycled to potential more positive than −0.4 V. As (2013). a result, cycling a VN thin film electrode between −1.0 and −0.4 V 30. Y. Dall’Agnese, M. R. Lukatskaya, K. M. Cook, P.-L. Taberna, Y. Gogotsi, and in 1 M KOH electrolyte after a prior 24 hours stabilization period re- P. Simon, Electrochem. Commun., 48, 118 (2014). −2 31. M. Wixom, L. Owens, J. Parker, J. Lee, I. Song, and L. Thompson, Electrochem. Soc. sulted in an initial capacitance of 19 mF.cm with a capacity retention Proc., 96-25, 63 (1997). of 96% after 10000 cycles. It is worth noting that such performance 32. L. Owens, L. T. Thompson, and M. R. Wixom, High surface area mesoporous desigel combined with the simple and reproducible PVD synthesis of VN thin materials and methods for their fabrication. US Patent 5837630, 1998. films make them attractive for electrochemical microsupercapacitor 33. C. Z. Deng and K. C. Tsai, Electrochem. Soc. Proc., 96-25, 75 (1997). 46,47,69,70 34. D. A. Evans and J. R. Miller Capacitor including a cathode having a nitride coating. applications. US patent 5754394, 1998. The VN pseudocapacitive behavior is thought to be due to sur- 35. L. T. Thompson, M. R. Wixom, and J. M. Parker, High surface area nitride, carbide face oxide redox processes. This has been further supported by sur- and boride electrodes and methods of fabrication thereof. US Patent 5680292, 1997. face characterization following cycling experiment between 0.0 and 36. D. Choi and P. N. Kumta, 2298 (2006). 41,58 37. D. Choi and P. N. Kumta, Electrochem. Solid-State Lett., 8, A418 (2005). −1.2 V. However, the fact that the VN was found to be unstable in 38. S. L. Roberson, D. Finello, and R. F. Davis, J. Appl. Electrochem., 29, 75 (1999). 1 M KOH for potential more positive than −0.4 V demonstrates that 39. T.-C. Liu, W. G. Pell, and B. E. Conway, J. Electrochem. Soc., 145, 1882 (1998). further investigation is needed to confirm that the observed surface 40. X. Zhou, H. Chen, D. Shu, C. He, and J. Nan, J. Phys. Chem. Solids, 70, 495 (2009). oxide was not just the result of some oxidation/degradation process. 41. D. Shu, C. Lv, F. Cheng, C. He, K. Yang, J. Nan, and L. Long, Int. J. Electrochem. Sci., 8, 1209 (2013). Additionally, in this study, the presence of oxygen was shown to 42. A. M. Glushenkov, D. Hulicova-jurcakova, D. Llewellyn, G. Q. Lu, and Y. Chen, shift the open circuit potential of a VN electrode to about 0 V by Chem. Mater., 22, 914 (2010). self-discharge. Consequently, in order to make sure that ex-situ char- 43. L. Zhang, C. M. B. Holt, E. J. Luber, B. C. Olsen, H. Wang, M. Danaie, X. Cui, acterization measurements are representative of the real state of the X. W. Tan, V. Lui, W. P. Kalisvaart, and D. Mitlin, J. Phys. Chem. C, 115, 24381 (2011). material in the electrolyte, characterization of the surface composition 44. P. Pande, P. G. Rasmussen, and L. T. Thompson, J. Power Sources, 207, 212 (2012). after cycling should be performed without exposing the VN electrode 45. R. L. Porto, R. Frappier, J. B. Ducros, C. Aucher, H. Mosqueda, S. Chenu, to ambient air. B. Chavillon, F. Tessier, F. Chevire, ´ and T. Brousse, Electrochim. Acta, 82, 257 (2012). 46. R. Lucio-Porto, S. Bouhtiyya, J. F. Pierson, A. Morel, F. Capon, P. Boulet, and Acknowledgments T. Brousse, Electrochim. Acta, 141, 203 (2014). 47. E. Eustache, R. Frappier, R. L. Porto, S. Bouhtiyya, J.-F. Pierson, and T. Brousse, The financial supports of the Natural Science and Engineering Electrochem. Commun., 28, 104 (2013). 48. C. M. Ghimbeu, E. Raymundo-Pinero, ˜ P. Fioux, F. Beguin, ´ and C. Vix-Guterl, J. Research Council of Canada (NSERC) as well as the French Ministry Mater. Chem., 21, 13268 (2011). of Education, Research and Technologie are gratefully acknowledged. 49. S. Dong, X. Chen, L. Gu, X. Zhou, H. Wang, Z. Liu, P. Han, J. Yao, L. Wang, G. Cui, Ecole Polytechnique de Montreal ´ (cm ), NanoQAM and CQMF are and L. Chen, Mater. Res. Bull., 46, 835 (2011). also acknowledged. 50. F. Cheng, C. He, D. Shu, H. Chen, J. Zhang, S. Tang, and D. E. Finlow, Mater. Chem. Phys., 131, 268 (2011). 51. P. J. Hanumantha, M. K. Datta, K. S. Kadakia, D. H. Hong, S. J. Chung, M. C. Tam, References J. A. Poston, A. Manivannan, and P. N. Kumta, J. Electrochem. Soc., 160, A2195 (2013). 1. A. G. Pandolfo and A. F. Hollenkamp, J. Power Sources, 157, 11 (2006). 52. X. Lu, T. Liu, T. Zhai, G. Wang, M. Yu, S. Xie, Y. Ling, C. Liang, Y. Tong, and Y. Li, 2. J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon, and P. L. Taberna, Science, Adv. Energy Mater., 4, 1300994 (2013). 313, 1760 (2006). 53. Z.-H. Gao, H. Zhang, G.-P. Cao, M.-F. Han, and Y.-S. Yang, Electrochim. Acta, 87, 3. C. Largeot, C. Portet, J. Chmiola, P. Taberna, Y. Gogotsi, and P. Simon, J. Am. Chem. 375 (2013). Soc., 130, 2730 (2008). 54. R. Wang, X. Yan, J. Lang, Z. Zheng, and P. Zhang, J. Mater. Chem. A, 2, 12724 4. S. A. Al-Muhtaseb and J. A. Ritter, Adv. Mater., 15, 101 (2003). (2014). 5. M. Sevilla, S. Alvarez, T. A. Centeno, A. B. Fuertes, and F. Stoeckli, Electrochim. 55. Y. Su and I. Zhitomirsky, J. Power Sources, 267, 235 (2014). Acta, 52, 3207 (2007). 56. Y. Yang, R. Kirchgeorg, R. Hahn, and P. Schmuki, Electrochem. Commun., 43,31 6. D. N. Futaba, K. Hata, T. Yamada, T. Hiraoka, Y. Hayamizu, Y. Kakudate, O. Tanaike, (2014). H. Hatori, M. Yumura, and S. Iijima, Nat. Mater., 5, 987 (2006). 57. M. Balogun, W. Qiu, W. Wang, P. Fang, X. Lu, and Y. Tong, J. Mater. Chem. A, 3, 7. S. Talapatra, S. Kar, S. K. Pal, R. Vajtai, L. Ci, P. Victor, M. M. Shaijumon, S. Kaur, 1364 (2015). O. Nalamasu, and P. M. Ajayan, Nat. Nanotechnol., 1, 112 (2006). 58. D. Choi, G. E. Blomgren, and P. N. Kumta, Adv. Mater., 18, 1178 (2006). 8. Y. Gogotsi, A. Nikitin, H. Ye, W. Zhou, J. E. Fischer, B. Yi, H. C. Foley, and 59. S. T. Oyama, The chemistry of transition metal carbides and nitrides; S. T. Oyama, M. W. Barsoum, Nat. Mater., 2, 591 (2003). Ed.; Chapman & Hall, 1996. 9. D. Belanger, ´ X. Ren, J. Davey, F. Uribe, and S. Gottesfeld, J. Electrochem. Soc., 147, 60. P. Krawiec, P. L. De Cola, R. Glaser, ¨ J. Weitkamp, C. Weidenthaler, and S. Kaskel, 2923 (2000). Adv. Mater., 18, 505 (2006). 10. C. K. Chiang, C. R. Fincher, Y. W. Park, A. J. Heeger, H. Shirakawa, E. J. Louis, 61. T. Onozuka, J. Appl. Cryst. 11, 132 (1978). S. C. Gau, and A. G. MacDiarmid, Phys. Rev. Lett. Lett., 39, 1098 (1977). 62. J. C. Caicedo, G. Zambrano, W. Aperador, L. Escobar-alarcon, and E. Camps, Appl. 11. M. Mastragostino, C. Arbizzani, R. Paraventi, and A. Zanelli, J. Electrochem. Soc., Surf. Sci., 258, 312 (2011). 147, 407 (2000). 63. K. Ait Aissa, A. Achour, J. Camus, L. Le Brizoual, P. Y. Jouan, and M. A. Djouadi, 12. A. Rudge, J. Davey, I. Raistrick, S. Gottesfeld, and J. P. Ferraris, J. Power Sources, Thin Solid Films, 550, 264 (2014). 47, 89 (1994). 64. B. D. Choi, G. E. Blomgren, and P. N. Kumta, 1178 (2006). 13. T. Brousse, D. Belanger, and J. W. Long, J. Electrochem. Soc., 162, A5185 (2015). 65. W.-F. Chen, J. T. Muckerman, and E. Fujita, Chem. Commun., 49, 8896 (2013). 14. B. E. Conway, Electrochemical Supercapacitors: Scientific Fundamentals and Tech- 66. C. Pozo-Gonzalo, O. Kartachova, A. A. J. Torriero, P. C. Howlett, A. M. Glushenkov, nological Applications; Kluwer, 1999. D. M. Fabijanic, Y. Chen, S. Poissonnet, and M. Forsyth, Electrochim. Acta, 103, 151 15. S. Trasatti and G. Buzzanca, J. Electroanal. Chem., 29, A1 (1971). (2013). 16. R. Fu, Z. Ma, and J. P. Zheng, J. Phys. Chem. B, 106, 3592 (2002). 67. M. Yang, A. J. Allen, M. T. Nguyen, W. T. Ralston, M. J. MacLeod, and F. J. DiSalvo, 17. W. Dmowski, T. Egami, K. E. Swider-lyons, C. T. L. Love, and D. R. Rolison, J. J. Solid State Chem., 205, 49 (2013). Phys. Chem. B, 106, 12677 (2002). 68. T.-H. Wu, Y.-H. Chu, C.-C. Hu, and L. J. Hardwick, Electrochem. Commun., 27,81 18. T. Lui, W. G. Pell, and B. E. Conway, Electrochim. Acta, 42, 3541 (1997). (2013). 19. H. Y. Lee and J. B. Goodenough, J. Solid State Chem., 148, 81 (1999). 69. M. Beidaghi and Y. Gogotsi, Energy Environ. Sci., 7, 867 (2014). 20. M. Toupin, T. Brousse, and D. Belanger, Chem. Mater., 16, 3184 (2004). 70. M. Brunet and D. Pech, Techniques de l’ingenieur ´ 2012, Re210.

Journal

Journal of the Electrochemical Society – IOP Publishing

Published: Mar 29, 2016

There are no references for this article.