A V2O5.nH2O Nanoflake Electrode for Aqueous Sodium Ion Intercalation Supercapacitors
do not necessarily reflect the views of UKDiss.com.
A V2O5.nH2O Nanoflake electrode for aqueous sodium ion intercalation supercapacitors
Abstract
The high capacitance electrode material V2O5 has been the subject of much research for metal ion energy storage devices due to the variable redox states and multi-electron redox reactions available during the electrochemical process. By expanding the interlayer spacing through hydration to create V2O5.nH2O, the crystalline electrode material is capable of hosting larger ions. This property of the material is beneficial in the creation of an aqueous sodium ion supercapacitor for fast energy storage. In this work, V2O5.nH2O nanoflakes are synthesized through a simple hydrothermal method and assembled directly onto a stainless steel substrate for facile electrode construction. The V2O5.nH2O nanoflake materials shows excellent electrochemical performance in cell testing, delivering a capacity of 180 mA h g-1 in a Na ion cell. These V2O5.nH2O electrodes were used in the construction of an aqueous sodium supercapacitor and demonstrated specific capacitances up to 122.9 F g-1. V2O5.nH2O electrodes were enhanced by the addition of graphene deposited onto the surface of the material and demonstrated capacitances up to 176.5 F g-1. These results are promising in the development of an efficient sodium intercalation electrode for use in an aqueous sodium ion supercapacitor.
Contents
Abstract…………………………………………………………..
Acknowledgements………………………………………………….
1 Introduction……………………………………………………
1.1 Supercapacitors as an energy storage system
1.1.1 Electric double layer supercapacitors………………………….
1.1.2 Pseudocapacitance………………………………………
1.1.3 Electrode and electrolyte materials……………………………
1.2 V2O5/activated carbon sodium ion supercapacitor……………………
1.2.1 Sodium ion supercapacitors………………………………..
1.2.2 Nanostructured V2O5.nH2O electrode
1.2.3 Activated carbon electrode
1.3 Characterization techniques……………………………………
1.3.1 X-ray powder diffraction (XRD)…………………………….
1.3.2 Scanning electron microscopy (SEM)
2 Experimental Details……………………………………………..
2.1 Materials Synthesis of V2O5.nH2O Nanoflakes
2.2 Materials Characterization
2.3 Electrochemical Analysis
3 Results and Discussion…………………………………………….
3.1 Chemical composition and structural analysis of V2O5.nH2O nanoflakes
3.2 Electrochemical performance of supercapacitor devices………………..
3.3 Systematic errors and problematic influences……………………….
4 Conclusions……………………………………………………
5 Bibliography……………………………………………………
This page intentionally left blank.
1 Introduction
In a modern society built upon a heavy dependence on nonrenewable fuels, namely fossil fuels, it has become increasingly important to develop and implement effective strategies to lower the consumption of such costly and environmentally harmful fuels. As a result, in recent times, there has been a growing interest in renewable energy sources and improved energy storage systems (ESSs). In order to fully embrace such renewable systems, amongst others, more efficient ESSs must be created, ones with both high power and high energy density.
For an effective implementation of renewable energy on a global scale, energy storage systems must be able to deal with the intermittent nature of renewable energy sources such as wind or solar power while enabling easy transfer to the grid. Although this has been successfully demonstrated utilizing existing battery technology, it has only been done so at a small scale.1 While renewable energy is chief amongst the applications of better ESSs, more efficient systems for both storage and transfer of energy will always be needed, in consumer electronics or vehicular braking systems, to name just a few. These future needs for higher power capabilities require new devices, such as supercapacitors, to be created.
While significant research has been dedicated to supercapacitor devices, the materials required to create such devices, in particular, lithium, is in high demand and face potential supply shortages. Therefore, the creation of an effective sodium ion supercapacitor, which could potentially store even more energy than a lithium ion supercapacitor, is of the utmost importance.
1.1 Supercapacitors as an energy storage system
Supercapacitors are energy storage devices commonly used in many applications and provide high power capabilities when rapid cycling is required. However, even though supercapacitors can provide hundreds to thousands of times higher power within the same volume of material, they do not have the same amount of charge storage ability as batteries, being usually up to 30 times lower.2 These properties mean that supercapacitors are ideal for cases where very quick bursts of power are required but higher energy capacities as provided by batteries are not necessary. Supercapacitors are also commonly used alongside batteries to create storage devices with separate power and energy properties, improving both and bridging the gap between pure supercapacitors and conventional batteries.
Supercapacitors can naturally be compared to other types of capacitors, namely the electrostatic and electrolytic kinds. In terms of construction, electrostatic capacitors are typically made of two metal electrodes, separated by a dielectric, a non-conducting material, between the two plates. The operating voltage depends upon the strength of the dielectric, and the capacitance of the device, measured in Farads (F), is a ratio of the charge in coulombs on each electrode (Q), to the potential difference between the two plates (V), C = Q/V.3 Electrolytic capacitors are similar in construction but instead have an electrolyte salt directly in contact with the two electrodes. Supercapacitors, take this a further step, also utilizing an electrolyte but pushing the capacitance in terms of surface area (A) and electrode separation (d), C=
ϵ0A/d, to the limit. Supercapacitors have much higher surface areas due to nanomaterials being used to build porous electrodes and a very small separation between the ionic and electronic charges at the electrode-electrolyte surface.4
In comparison with electrolytic capacitors, supercapacitors exhibit lower power output, but can have specific energies several orders of magnitude greater than that of a conventional capacitor.5 Thus, supercapacitor devices are of great interest in research and industry due to the gap they fill between conventional electrolytic capacitors and batteries. While typical electrolytic capacitors have very high power capabilities, reaching well over 10 kW kg-1, and batteries can store great amounts of energy, neither are usually able to supply both. The relative power and energy densities of each type of energy storage device are graphically described in a Ragone plot, with the respective densities represented in each axis and the discharge time of the devices, (E = Pt), shown in diagonal lines. Some different storage technologies are shown in the Ragone plot in Fig. 1. While the power and energy densities of each supercapacitor are important and measureable properties, there are many factors to take into account when creating the best device. Parameters such as cost, safety, and the cycle life of the device must all be taken into account.
A particularly important property of supercapacitors to keep in mind therefore, is that the device exhibits high power capabilities due to its fast discharging, but that it can also be charged in as short a period of time. This means that supercapacitors can be used in applications where energy must be rapidly stored, such as in kinetic energy recovery systems in braking vehicles. Managing the trade-off between power and energy capabilities is the most important part of supercapacitor development, and successfully meeting both requirements can create devices that compare favorably to both capacitors and batteries. Table 1 compares some of the important properties of supercapacitors with capacitors and batteries.
Another great advantage of supercapacitors is their cycle life. Since the energy storage mechanism of supercapacitors do not typically rely on chemical reactions with charge instead stored physically at the electrode-electrolyte surface interface in an electric double layer, the electrodes do not get degraded or corroded by irreversible chemical reactions typical in batteries. This method of charge storage allows supercapacitors to easily exceed the cycle life of batteries.5 Even pseudocapacitive devices, which do rely on chemical reactions, have long cycle lives as the reactions involved are highly reversible. Since there are usually no chemical reactions taking place, there are also no physical changes to the electrode material, such as swelling the is commonplace in battery electrodes.3 While supercapacitors might have extended cycle life due to this charge storage mechanism, this also puts severe limitations on the operating voltage for the device. The voltage window is often very narrow so as to not chemically degrade the electrolyte which functions as the charge carrier in a supercapacitor. Table 2 outlines the differences in the storage mechanisms of batteries and supercapacitors.
Figure 1 Ragone plot showing some common energy storage technologies. Each energy storage device is located in unique regions describing their relative power and energy densities. Characteristic charge/discharge times correspond to diagonal lines. The inset window shows energy drops due to internal dissipation and leakage losses for sufficiently high and low power.6
Table 1
Comparison of some properties of capacitors, batteries, and supercapacitors.5
Characteristics | Capacitor | Supercapacitor | Battery |
Specific energy (W h kg-1) | < 0.1 | 1 – 10 | 10 – 100 |
Specific power (W kg-1) | > 10,000 | 500 – 10,000 | < 1000 |
Discharge time | 10-6 to 10-3 | s to min | 0.3 – 3 hr |
Charge time | 10-6 to 10-3 | s to min | 1 – 5 hr |
Coulombic efficiency (%) | About 100 | 85-98 | 70 – 85 |
Cycle-life | Almost infinite | > 500,000 | ~ 1,000 |
Table 2
Comparison of storage mechanisms of batteries and supercapacitors.2
Parameter | Battery | Supercapacitor |
Storage mechanism | Chemical | Physical |
Power limitation | Reaction kinetics, ion transport | Electrolyte conductivity |
Energy storage | High (bulk) | Limited (surface area) |
Charge rate | Limited by kinetics | High, equal to discharge rate |
Limits on cycle life | Mechanical stability, chemical reversibility | Side reactions |
A supercapacitor is made using two electrodes, either identical or different, with a separator between them, which prevents electrical contact between the two electrodes. The separators can be polymer or paper, typically for organic electrolytes, or ceramic or glass, for aqueous electrolytes.4 This separator is ion-permeable and is soaked in the electrolyte and provides a vessel for the electrolyte to be directly in contact with each electrode. Ideally, the separator should be high electrical resistance to prevent overly prevalent leakage current, as well as being as thin as possible to reduce the path electrolyte ions must take to move between electrodes. Depending on whether the two electrodes are the same or different, symmetric of asymmetric supercapacitors, respectively, can be made. Electrolytes are typically either aqueous or organic. The electrolyte breakdown potential is what limits the cells operating voltage window, with organic electrolytes having wider voltage windows than aqueous electrolytes.7
Supercapacitors can be further broken down into three general types, depending on their method of energy storage. Electric double layer capacitors (EDLCs) are based on high specific-surface area, nanoporous materials, with charge stored physically in an electric double layer at the electrode surface. Pseudocapacitors are faradaic supercapacitors, with electrodes typically constructed using conducting polymer or metal oxides. These combine both electrostatic charge storage in an electric double layer as well as faradaic charge storage through fast and highly reversible redox reactions at electrode surfaces, similar to batteries. Hybrid capacitors, also more generally called asymmetric supercapacitors (ASSCs), are constructed with both an electrostatic electrode and a faradaic electrode, combining the properties of both a high power EDLC and a high capacity battery, leading to a middle ground in terms of performance.
The electrochemical performance of all of these devices are formally considered as supercapacitance, with the capacitance, C, depending on the dielectric constant of the electrolyte,
ϵr, the effective thickness of the double layer, d (separation between charges), and the accessible surface, A, as the following equation:
C=ϵrϵ0Ad
where
ϵ0 is the dielectric constant of the vacuum. However, the double layer does not have a conventional dielectric to separate the charges.
The energy, E, stored within a supercapacitor is
E=12CV2
where V is the charged cell voltage. The energy stored in the device is proportional to the capacitance and the cell voltage squared. Thus, when attempting to increase the energy density of the device, it is important to maximise both of these.
The maximum power, Pmax, that a supercapacitor is capable of is proportional to the charged voltage squared and the internal resistance, R, of the full device, as the following equation
Pmax=V24R
The electrical properties of a supercapacitor such as cell voltages and capacitances are dependent on the makeup of the device, with most research attempting to create better electrode materials for use with common electrolytes.
Supercapacitors are typically constructed using two electrodes, which forms an equivalent series circuit of two individual capacitors, C1and C2. The behaviour of each supercapacitor electrode within the device behaves analogously to a conventional capacitor, and the capacitive behaviour of a supercapacitor is equivalent to two conventional capacitors in series, as each electrode has an electric double layer which behaves as a conventional capacitor. Therefore, the total capacitance, Ctotal, is as follows
Ctotal=C1⋅C2C1+C2
Conventional capacitors are typically tested by applying a small AC voltage at a frequency of 100 Hz or 1 kHz. However, the capacitance value of a supercapacitors is heavily dependent upon the frequency of the applied voltage.8 Within the electrode, ions move across different distances inside the pores. The areas at the openings of each pore are easily accessible, leading to an analogously low electrical resistance. However, as the distance the ions move become greater, the higher the equivalent resistance.9 This behaviour can be described by an equivalent series circuit of RC elements, shown in Fig. 2. If the applied voltage has high frequency, thus changing the polarity of the supercapacitor quickly, the total electrode surface area that ions can access within that time is greatly reduced, as the ions cannot reach deeper within the porous material. Thus, due to this strong frequency dependence of the capacitance, the capacitance of a supercapacitor must be measured through a constant current charge/discharge test over a significantly long time.
Figure 2Equivalent series circuit of RC elements of a supercapacitor electrode pores.
As a supercapacitor undergoes charging and discharging, ions in the electrolyte move into and out of the electrode material, with any energy losses during this process measured and considered as the internal resistance of the device. The circuit of RC elements, shown in Fig. 2, represent the pores within the electrode material and show that as the ions travel greater distances, the internal resistance is greater too. When designing an electrode for a supercapacitor, it is important to take into consideration factors affecting the internal resistance of the device as this resistance affects the rate capabilities of the supercapacitor. Ensuring that adequate pore sizes and pathways are available decreases the internal resistance of the electrode material and ensures the supercapacitor can have improved rate performance.
1.1.1 Electric double layer supercapacitors (EDLCs)
An electric double layer is a structure that appears on the surface of an object when it comes into contact with a liquid. The object, when charged, has its charge balanced by an opposite charge forming in the liquid at this interface. Multiple theories model this behaviour, with the three main models illustrated in Fig. 3.
The Helmholtz model, Fig. 3 (a), is the simplest approximation of the spacial distribution of charge at the double layer interface. It is analogous to the classical description of a parallel plate capacitor. The charge on the object is neutralised by oppositely charged ions at a distance d from the object. The Helmholtz theory is the most simplistic, considering the counter charges as a single rigid layer. However, this model neglects any interactions between ions at further distances than this layer and also does not consider the electrolyte concentration.10
Figure 3 Electric double layer models, (a) Helmholtz model, (b) Gouy-Chapman model, and (c) Stern model.7
In reality, the capacitance of a supercapacitor was not constant, but was in fact dependent on the voltage applied to the device and on the concentration of the electrolyte. Thus, the Gouy-Chapman model was developed, shown in Fig. 3 (b), where ionic charges appear in liquid directly around a charged object. Unlike the Helmholtz model, these ions are not rigidly attached, with the first ions at a distance d from the surface.10 This model also introduces a diffuse layer allowing the application of Maxwell-Boltzmann statistics, with the potential decreasing exponentially at increasing distance from the surface interface. However, this model is inadequate for a highly charged double layer,11 as experimentally, the measured thickness of the double layer is greater than the calculated thickness.7
While the Gouy-Chapman model makes significant improvements on the Helmholtz model, it makes the assumption that the ions are point charges and that they are able to approach the surface of the object completely, which is not the case in reality.7 The Stern model is a modified version of the Gouy-Chapman model, combining the Helmholtz model with the Gouy-Chapman model, shown in Fig. 3 (c). It states that the ions in fact have a finite size, and therefore are limited in their ability to approach the surface of the object. Furthermore, instead of the first ions being at a distance d, the Stern model allows for surface-adsorbed ions within a plane at distance d, known as the Stern layer. Ions within the Stern layer are adsorbed to the object, with specifically adsorbed ions forming an inner Helmholtz layer and non-specifically adsorbed ions forming an outer Helmholtz layer.12 Stern’s model therefore suggested both an inner layer (the Helmholtz layer) and an outer layer (the Gouy-Chapman layer).
Although using these three models can provide an adequate understanding of the mechanics of the electrical double layer, they generally are not able to completely explain the real-life behaviour in supercapacitors. The exact mechanisms and methods of ion adsorption in a supercapacitor are extremely complicated and are still not entirely understood.7
However, using these models, we can understand what occurs to ions and electrons in a supercapacitor during cycling. When charging, electrons move from the positive electrode to the negative electrode through the external circuit connected to the supercapacitor. Meanwhile, electrolyte ions move to each electrode, with cations moving to the negative electrode and anions moving to the positive electrode. These ions adsorb to the surface to balance the charge on the solid electrode and forms an electronic double layer. The supercapacitor maintains this state until discharging where the opposite occurs. Electrons move pack through the circuit from the negative to positive electrode and the electrolyte ions de-adsorb from the electrode surfaces.
The capacitance of a supercapacitor depends on the electrode material, specifically on the total surface area available and on the size of pores within the material. The capacitance depends on the amount of ions that can adsorb onto the electrode and if a particular pore is too small, it becomes inaccessible to the ion, and therefore will not contribute to the double layer capacitance.4 As such, not all pores of an electrode material are accessible and therefore the relationship between capacitance and the material specific surface area is not a straightforward linear relation. There are three general sizes of nanopores within an electrode material, namely micropores (< 2 nm), mesopores (2-50 nm), and macropores (> 50 nm). Previous research has suggested that pore sizes below 0.5 nm are not accessible to aqueous electrolyte ions,13,14 and sizes smaller than 1 nm are not accessible to organic electrolyte ions.15
Despite this non-linear relation, the important values regarding capacitance are calculated and published in terms of volume or mass of active material. It is clear that particularly large pores are mostly free space, and as such does not contribute to charge storage, and in fact reduce the volumetric efficient of the material. Likewise, pores which are too small are inaccessible and also decrease the density of the useable electrode material. Therefore, when producing electrodes, it is important to try to create small but electrolyte-accessible pores to maximise useable surface area.16 There is general agreement that micropores should be the primary size present to have the most surface area. The addition of some mesopores is also suggested as the larger pores can provide larger pathways for electrolyte ions to diffuse to the microporous surfaces, improving power capabilities.17
1.1.2 Pseudocapacitance
Pseudocapacitance is an alternative, faradaic method of supercapacitance energy storage. By creating a device that relies on fast and highly reversible redox reactions, it behaves electrically analogous to an EDLC and also maintains good cycle life due to the highly reversible reactions involved. An ideal pseudocapacitor responds identically to an EDLC, with the state of charge of the device changing with the applied voltage, leading to a behaviour that can be formally considered supercapacitance, despite its faradaic nature.
There are many different charge storage mechanisms utilised in the design of pseudocapacitor devices: underpotential deposition, redox reactions of transition metal oxides, intercalation pseudocapacitance,18 and also reversible electrochemical doping and de-doping in conducting polymers.19 In the case of this project, the mechanism of the intercalation of sodium ions is used.
In pseudocapacitors, since faradaic processes are combined with electric double layer charge storage, the specific capacitance of a device is increased and can be as high as 10-100 times that of an EDLC. However, the trade-off is the relatively lower power performance of the pseudocapacitor as the faradaic processes involved are slower.20 Another disadvantage of pseudocapacitors is their potentially poor mechanical stability and low cycle life due to swelling and shrinking of electrodes during charging and discharging, similar to that of batteries.21
1.1.3 Electrode and electrolyte materials
There are many potential electrode materials and electrolytes that can be used in supercapacitor devices. Carbon based materials are the most commonly used as they are relatively low cost and are commonly available. Carbon materials can be further broken down into several different kinds, such as activated carbon, carbide derived carbons, carbon nanotubes, graphene, and mesoporous carbon.7 Activated carbons are most commonly used due to its high surface area and relatively low cost,19 and were used in this project.
Metal oxides are also commonly used as pseudocapacitive materials and possess high specific capacitance and conductivity.4 Metal oxides are versatile and a wide variety of metal oxides are used as electrode materials, such RuO2, IrO2, MnO2, NiO, Co2O3, SnO2, and MoOx.7 V2O5 is also used and is the material that is explored in this project. The most important properties for a metal oxide electrode are wide potential windows and highly reversible redox reactions. Typically, metal oxides also have high conductivity, improving their rate capabilities.22
The electrolyte chosen for the supercapacitor also requires consideration to maximise the voltage window available for the device. The electrolyte concentration also has to be kept high so that during cycling, the electrolyte is not depleted.23 Insufficient electrolyte will lead to a “starvation” effect and reduce the performance of the supercapacitor cell. Typically, concentrations in excess of 0.2 molar are sufficient.3
The most important properties of an electrolyte to be considered are the operating voltage window and the conductivity. These determine the environment in which the device can operate. Wider voltage windows allow for a greater capacity and greater conductivity reduces the effective resistance of the device. The electrolyte should also have low ionic radius to make use of the high surface area of the electrode. Ideally, it should also have low volatility, for safety, and low cost. The electrolyte should also not corrode the electrodes of current collectors, which is dependent on the nature of the electrolyte and on the pH of aqueous electrolytes.
Electrolytes used in supercapacitors are either aqueous or organic. Aqueous electrolytes typically cannot be used for voltages in excess of 1 V, due to the decomposition of water at 1.23 V, whereas organic electrolytes can be used at voltages of 2.7 V or higher.24 However, organic electrolytes typically have much higher specific resistance than aqueous ones, typically with at least 20 times higher resistance,3 leading to reduced power capability. Aqueous electrolytes also typically have much smaller ionic radii than organic electrolytes, as lithium and sodium are most commonly used. This means the pores within the electrode material can be smaller, allowing for higher surface areas.24,25 Furthermore, organic electrolytes are more expensive to produce, due to requiring purification from water, which prevents electrolyte degradation at higher cell voltages. The aqueous electrolyte, NaSO4 is used in this project.
1.2 V2O5/activated carbon sodium ion supercapacitor
The aim of this project was the fabrication of a vanadium pentoxide/activated carbon asymmetric aqueous sodium ion supercapacitor. Energy storage devices utilising sodium ions is a large area of ongoing research, with sodium ion devices being proposed as an alternative to the massively common lithium ion device. This is due to sodium’s relatively lower cost, greater availability, and increased operating safety.26 The sodium intercalation chemistry is very similar to that of lithium, thus making it simpler to develop electrode materials, and indeed, making it possible to even use the same materials for both systems. However, it should be noted that the ionic volume of sodium is about 70% larger than that of lithium, and the insertion of Na+ into the electrode material is much harder that Li+.27 While significant research has been dedicated to sodium ion devices, most have only investigated the use of an organic sodium electrolyte.28–30 The main disadvantage in using aqueous electrolytes is its narrow operating voltage due to the decomposition of water at 1 V. However, the creation of an aqueous electrolyte based device is motivated by higher theoretical capacities, as well as increased safety and decreased cost.
However, this project aims to create a supercapacitor device utilising an aqueous sodium sulfate solution as the electrolyte, alongside well-known electrode materials. By enhancing the well studied V2O5 and developing a V2O5.nH2O nanoflake electrode, a supercapacitor utilising this electrode may be able to effectively intercalate Na ions within an aqueous device. While there are many advantages to an aqueous electrolyte, the ultimate goal would be the creation of a membrane assisted capacitive deionisation (CDI) device. The applications of such a device are immediately obvious, namely its ability to store a non-trivial amount of energy whilst purifying brackish water. While much work has been dedicated to developing CDI devices,31 there are many difficulties to creating an efficient and effective “salt water battery”, and to achieve this, time and effort beyond the scope of this project is required.
1.2.1 Sodium ion supercapacitors
Sodium ion supercapacitors are a type of asymmetric supercapacitor device. These devices combine the properties of both batteries and conventional EDLCs, and can store as much as 5-10 times more energy than conventional EDLCs due to its addition of pseudocapacitance.32 The most common type of asymmetric energy storage device is the lithium ion capacitor. The sodium ion supercapacitor is, in principal, the same, other than its electrolyte is sodium ion based rather than lithium ion based. While EDLCs utilise high surface area carbons to create identical positive and negative electrodes, the typical sodium ion supercapacitor system combines both the high surface area carbon as positive electrode and a compound that supports fast and highly reversible sodium redox reactions, in this case V2O5, as the negative electrode.
During discharging, sodium ions intercalate into the bulk of the negative electrode in a pseudocapacitive process, causing a reduction of the host lattice and releasing electrons through the external circuit. Anions adsorb to the surface of the positive electrode, in an electric double layer. During charging, sodium ions deintercalate from the negative electrode, with electrons forced through the external circuit back to the cathode, re-oxidising the host lattice, while anions desorb from the positive electrode. As the adsorption process is non-faradaic, it occurs faster than the sodium ion intercalation at the negative electrode. Therefore, the power density will be determined by the slower, sodium ion exchange process at the negative electrode.32
1.2.2 Nanostructured V2O5.nH2O electrode
Vanadium pentoxide (V2O5) is a promising electrode material for energy storage devices and has been intensively studied in lithium batteries for several decades,33 due to its various coordination numbers and oxidation states. More recently, it has also attracted research as a possible electrode material for hosting larger, sodium ions.27,34 While various forms of nanostructured V2O5 have been created, in particular, hydrated forms of vanadium pentoxide (V2O5.nH2O) have been found to have better chemical stability and reversibly uptake more than two Li+ ions due to the expanded interlayer space from extra inserted water molecules.35 With a layered structure with double V2O5 sheets, V2O5.nH2O contains water molecules in the oxide network (Fig. 4).36,37 While V2O5 is a promising electrode material, its application is generally hindered by the capacity fade with cycling that occurs due to the agglomeration of particles, pore occlusion, degradation of the structure, and low electronic and ion conductivities.27
To solve these issues previous research has investigated complex methods, such as synthesis of nanoarchitectures, coating with conductive materials, or doping with metal elements.38 The capacity retention and rate performance of pure V2O5.nH2O are poor unless it is modified with the addition of nanostructured carbon.39 It has also been found that V2O5 can achieve high performance by constructing a binder-free electrode.27 This electrode has a 3D conductive material (e.g. steel mesh, carbon cloth etc.) as a substrate, while the active material is directly deposited on the substrate to form a nano-array architecture.40 This has many advantages: namely improving the volumetric capacity of the device because of the elimination of binder and conductive agent; avoiding agglomeration of particles, pore occlusion, and degradation of the structure when making and drying the electrode; increasing the electronic conductivity and contact area with electrolyte; and improving the ionic conductivity by shortening the ion diffusion pathway.27
Figure 4 Structure of V2O5.nH2O (polyhedral and ball-stick model) with water molecules shown in green. Characteristic crystal structure distances are shown.37
Despite the suitability of V2O5 as an intercalation electrode in supercapacitors, the poor electronic conductivity and high solubility of V2O5 in aqueous solutions makes it difficult to develop low-cost, aqueous devices with high rate capability and long cycling life. Firstly, by our methods of recrystallizing the vanadium oxide through a hydrothermal method, the issue of solubility is circumvented. This recrystallization process is theorized as follows, with V2O5 that has been dissolved in water and excess hydrogen peroxide forming a solution of diperoxo anions [VO(O2)2]–.27,41 Under high temperature, it decomposes to V2O5.nH2O and O2 on the substrate surface and in bulk solution. Meanwhile, an oxide film and intergranular stress corrosion cracking are generated on the surface of the stainless steel mesh.27 The V2O5.nH2O deposits on the substrate along these cracks to form a solid solution with an oxide film. This growth is no longer soluble and can only be mechanically removed from the substrate.
While the issue of solubility can be simply solved by choosing a suitable synthesis process, the poor conductivity of V2O5 and indeed, the additional resistance of the stainless steel mesh substrate, remains an existing issue. Typically composites of the active material with nanostructured carbons are proposed as a solution.29 However, creation of composite electrode materials require organic binders which are undesirable as they can lower the Na ion diffusion rate and electronic conductivity of the electrode.42 Here we proposed to instead grow graphene onto the surface of our electrode through a simple chemical vapor deposition method. Careful addition of highly conducting graphene to the high storage capacity metal oxide will lead to enhanced Na storage properties.
The V2O5.nH2O electrode was fabricated with the intention to create a hydrated crystalline structure with sufficient interlayer spacing to facilitate Na intercalation. The layered structure of V2O5 and its mixed valence states (V2+, V3+, V4+, V5+) in VO, V2O3, VO2 and V2O5 provides readily accessible redox couples. The most accessible V5+/V4+ redox couple is the most attractive for Na+ ion intercalation. The kinetics and mechanics of sodium intercalation into layered V2O5 are well studied,43 where it can be described by the equation:
V2O5+xe-+xNa+↔NaxV2O5 0<x<2
The intercalation of sodium is accompanied by reduction of the host lattice, as the guest species exists as Na+.44
V2O5 typically exists in the orthorhombic
α-V2O5.45 To understand the intercalation process and structural evolution during the electrochemical process, the energies involved must be considered. Much theoretical work has been committed to investigating this process and has found that different stages are observed during the ion intercalation process.46 This can be explained by considering the competition between interlayer van der Waals interaction and ion-ion Coulomb repulsion. If the energy required to expand the interlayer separation is larger than the Coulomb repulsion between ions, then ions are more favorably intercalated into a single gallery until it reaches a maximum capacity. This can be defined as stage-I. A second stage-II occurs when all galleries are occupied. These schemes are shown in Fig. 5.
Figure 5 Schematic illustration of the stage-I and stage-II arrangements of Na storage in orthorhombic V2O5.46
At the beginning of the electrochemical process, at lower Na concentrations within the lattice, stage-I structures are dominant. As the Na concentration increases during the intercalation process, the lattice expands and leads to stronger repulsive interaction between Na ions. It has been found that as the concentration reach Na0.333V2O5, the repulsive energy between the Na ions becomes greater than the energy required to expand the interlayer separation. In this case, the stage-II structures with Na intercalation in all layers would become prevalent. As Na ions are inserted into the crystal lattice, the structure expands to accommodate the intercalated ions. These changes to the crystal dimension have been well investigated through theoretical methods.46 The proposed nanoflake structure of the V2O5.nH2O electrode materials are theoretically suitable to this intercalation of Na ions due to the adjustable layer spacing. The crystal lattice can expand to allow for ions to be inserted between layers without significant breakdown of structure and as such, it able to store a great amount of charge.27
1.2.3 Activated carbon electrode
Activated carbons are most commonly used due to its high surface area and relatively low cost.19 In terms of this project, the main focus was on the vanadium pentoxide electrode. Furthermore, carbon electrode materials have been exhaustively studied and the properties of the various types of carbon are very well known. As such, it should not be particularly important to discuss the activated carbon electrode. However, its contribution to the device should be noted. Within the asymmetric supercapacitor, the carbon anode is important to prevent any further decreases to the rate capability of the device. As a conventional EDLC electrode, the activated carbon electrode adsorbs the anion during discharging.
As an electrode for our device, activated carbon was mixed with carbon black. Polyvinylidene fluoride (PVDF) was used as a binder and NMP was used as an organic solvent so the aqueous electrolyte would not degrade the electrode. The carbon mixture was coated as a film onto an aluminium foil substrate.
1.3 Characterization techniques
1.3.1 X-ray powder diffraction (XRD)
X-ray powder diffraction (XRD) is an analytical characterization technique used for phase identification of a crystalline material. Crystalline materials act as three-dimensional diffraction gratings for X-ray wavelengths, and as such, causes constructive interference of a monochromatic radiation source. Monochromatic radiation is directed towards a sample and the interaction of the incident rays with the sample produces diffracted rays according to the conditions of Bragg’s Law
nλ=2d sinθ
which relates the wavelength,
λ, to the diffraction angle,
θ, and the lattice spacing, d, in the crystalline sample. These diffracted rays are detected and counted. By scanning through a range of angles, all possible diffraction directions of the crystal lattice can be detected provided the orientation of the powdered material is sufficiently random. Through processing of the diffraction peak angle positions, where significant constructive interference takes place, the lattice spacings can be calculated according to Bragg’s Law. As each crystalline material has a unique set of interlayer spacings, this allows the material to be identified typically by comparison of data with standard reference patterns.47,48
1.3.2 Scanning electron microscopy (SEM)
Scanning electron microscopy (SEM) is an imaging technique using a focused beam of high-energy electrons to generate a variety of signals at the surface of solid samples. Accelerated electrons in a scanning electron microscope carry significant amounts of kinetic energy, which is dissipated as a result of electron-sample interactions when the incident electrons are decelerated in the solid sample. There are several interactions that occur during imaging that give off different signals, including secondary electrons, backscattered electrons, diffracted backscattered electrons, photons, visible light, and heat.49
Secondary and backscattered electrons are the most commonly used for imaging samples. Secondary electrons are emitted by atoms near the surface of a sample as a result of incident electrons excited electrons belonging to these atoms. These excited electrons gain sufficient energy to escape the sample surface and are detected. Backscattered electrons occur when incident electrons, rather than losing energy by exciting electrons within the sample, re-emerge from the surface of the sample. By detecting the signals these electrons produced, the data can be processed to produce an image. Secondary electrons are most valuable for showing morphology and topography of samples and backscattered electrons are most valuable for illustrating contrasts in composition in multiphase samples.50
2 Experimental Details
2.1 Materials Synthesis of V2O5.nH2O Nanoflakes
The anode material was prepared by a facile hydrothermal synthesis, similar to a previously reported method.41 In a typical experiment, 95 mg of commercial V2O5 (99.99% trace metal, Sigma Aldrich) was added to 30 mL of water and 1 mL 30% H2O2 and stirred for 2 hours. The as-prepared solution was then transferred to a Teflon container in a stainless steel autoclave with 10 pieces of 1.6 cm diameter stainless steel mesh discs. The autoclave vessel was then heated at 200°C for 12 hours. After heating, the vessel was removed from the oven and cooled in air for 2 hours. Then, the stainless steel meshes were removed and washed, 3 times with water and 3 times with ethanol, and dried under vacuum at 90°C for 3 hours. The crystallinity of the samples was improved by heat treatment at 300°C.
V2O5.nH2O/graphene electrodes were prepared by a simple chemical vapor deposition (CVD) technique. V2O5.nH2O nanoflake electrodes were prepared as above by the facile hydrothermal method and the as-prepared electrodes were then placed in a CVD tube furnace. The samples were first calcined under argon at 650°C for 30 minutes and then graphene was grown under an acetylene flow for 10 minutes. Finally, the samples were cooled to room temperature under argon.
It is worth noting that precursor materials were of relatively low cost and that all methods used during the synthetic process were efficient without involving toxic or environmentally harmful compounds.
2.2 Materials Characterization
V2O5.nH2O nanoflake active material was separated from the stainless steel substrate by sonication in ethanol. Typically, the powder solid active material was kept in suspension in ethanol at room temperature until dry samples were required, at which time the material was dried at 90°C under vacuum for three hours. The as synthesized V2O5.nH2O stainless steel electrodes were also used for materials characterization.
X-ray powder diffraction (Rigaku Smartlab) with a CuK
αradiation source (scan rate 2° min-1, scan range 5°-85°) was used to characterize the structure of the sample. The morphologies and structures were examined and imaged by using a field-emission scanning electron microscope (FEI Quanta 400).
2.3 Electrochemical Analysis
As-prepared heat-treated V2O5.nH2O nanoflakes on stainless steel mesh were used directly as working electrodes in a three-electrode system, with a platinum wire counter electrode and a silver/silver chloride (Ag/AgCl) reference electrode. 1M Na2SO4 was used as the electrolyte in this system. Electrochemical measurements were also made using both a CR2032-type coin-cell system and a Swagelok system. The active material on the electrodes varied from approximately 0.9 to 2.5 mg cm-2.
Three-electrode systems were used to analyze electrolyte stability and electrode material stability in an aqueous sodium electrolyte system. Coin-cell and Swagelok systems were used to analyze capacity and rate capabilities of a full cell system.
For the Na-ion cell used to measure galvanostatic charge/discharge capacities,
Na metal was used as the anode, while 1M NaClO4 in 1:1 ethylene carbonate and dimethyl carbonate was used as electrolyte and quartz microfiber filters (Whatman) as the separator. This cell was assembled in an argon glove box (H2O and O2 < 0.1 ppm, Mbraun, Unilab). Both coin cells and Swagelok cells analyzing an aqueous sodium system were assembled in a lab environment, using an activated carbon cathode (80 wt% activated carbon, 10 wt% carbon black, 10 wt% PVDF, with NMP as solvent), quartz microfiber filters (Whatman) as separator and 1M Na2So4 as electrolyte. The mass loading of the activated carbon electrode was approximately 3 to 4 times that of the V2O5.nH2O active material used.
Cyclic voltammetry tests for both half-cell and full cell systems were carried out on an Autolab PGSTAT302N system at room temperature. Galvanostatic charge/discharge tests were carried out on an Arbin BT-2043 battery tester at room temperature.
3 Results and Discussion
3.1 Chemical composition and structural analysis of V2O5.nH2O nanoflakes
The chemical composition and structure of the V2O5 nanoflakes on stainless steel substrate were characterized by X-ray diffraction. The XRD patterns are shown in Fig. 6. It can be seen in Fig. 6 (a) that the as-prepared sample presents a growth of the (001) plane primarily. Calcination is well known as an effective method of removing extra intermolecular water and improving the crystallinity of electrode materials. The sample that underwent heat treatment at 300°C showed improved crystallinity, exhibiting more distinct peaks in Fig. 6 (b). Both XRD patterns are indicative of a crystalline orthorhombic V2O5, with the heat treated sample showing improved crystallinity, and are as expected.51 According to the XRD results, the value of n in V2O5.nH2O is 1.6 for the as-prepared sample. These XRD patterns indicate the relative purity of the sample produced as V2O5.nH2O and indicates the crystal structure of the sample. The crystallinity is particularly important as it is the property that directly allows for Naion intercalation into the electrode.
Figure 6 a) XRD pattern of as-prepared sample b) XRD pattern of sample heat-treated at 300°C.
The morphology of the V2O5.nH2O electrodes can be seen in the scanning electron microscopy (SEM) images shown in Fig. 7. Before the reaction, the stainless steel substrate has a smooth surface. After the hydrothermal process, the surface of the stainless steel is almost completely covered with small nanoflakes. A full flake is approximately 5-10 µm in height and tens of nanometres in width. These nanoflakes are almost completely vertical to the stainless steel substrate and forms continuous nanoflake arrays across the surface.
The samples demonstrate near uniform growth with only small areas of agglomeration or non-uniform growth. The most likely cause of a “broken” area, where the stainless steel substrate is not completely covered by nanoflakes, is in the physical handling of the samples prior to imaging, rather than any faults during synthesis. This indicates that the hydrothermal process was successful in growing nanoflakes directly onto the surface of the stainless steel substrate, with excellent mass loading of the active material, and that the as-prepared sample retained structural stability and could be used directly as an electrode in a sodium ion supercapacitor.
Figure 7 SEM images of samples at various magnifications.
3.2 Electrochemical performance of supercapacitor devices
All electrochemical testing was performed using 300°C heat-treated samples. The electrochemical performance of the V2O5.nH2O Na ion coin cells were tested through galvanostatic charge/discharge cycling at a current density of 100 mA g-1 and are shown in Fig. 8. The mass of active material loading on this electrode was 1.83 ± 0.1 mg. The electrode delivers an initial discharge capacity of 180.0 ± 1.5 mA h g-1 and demonstrates an excellent capacity retention rate of 90.5% after 50 cycles. The theoretical capacity was calculated using the Faraday’s Law equation
Qtheoretical=nF3600mw
where n is the number of the charge carrier, F is the Faraday constant, and mw is the molecular weight of the electrode material. The theoretical capacity was calculated to be 147 mA h g-1. The actual capacity is higher than the calculated theoretical capacity due to the possibility of intercalating more than one Na ion for every V2O5 molecule. In reality, it is more likely that two Na ions are intercalated for every V2O5 molecule to form Na2V2O5. Assuming two sodium ion intercalations, the theoretical capacity then becomes 294 mA h g-1. However, not all the active material volume may be accessible and indeed, there are not necessarily two Na ions intercalated for each V2O5.
Figure 8 Specific capacity of V2O5.nH2O Na ion coin cell at a current density of 100 mA g-1, over 50 cycles.
These results are very encouraging when considering the design of an aqueous sodium ion supercapacitor utilising this material. Specific capacities of up to 319 mA h g-1 have been reported for nanostructured V2O5 electrodes.52 However, the nanoflake structure presented here is being used within an aqueous supercapacitor. As such, the versatility and potential of such an application outweighs the need of excessively high specific capacity. A specific capacity of 180 mA h g-1 combined with a capacity retention rate of over 90% after 50 cycles is an excellent starting point to further develop this material. In particular, the exceptional capacity retention rate suggests that the unique nanoflake morphology is suitable for the insertion of Na ions without degradation of the electrode material. It is clear that the nanoflake morphology is able to accommodate large volume changes repeatedly and is able to continue cycling without significant breakdown of the V2O5.nH2O crystal structure. The cycling stability and capacity retention is related to the current density applied,43 and given more time, higher current density tests, up to 1000 mA g-1, should be performed to investigate the high current capabilities of the supercapacitor device. At higher current densities, lower capacities are expected, but structural stability of the electrode material under a high rate may be confirmed.
Figure 9CV curves of potentiostatic charge/discharge testing of V2O5.nH2O electrode in a half-cell system.
Fig. 9 shows the electrochemical performance of the V2O5.nH2O electrode in a three-electrode system using 1M Na2SO4 as electrolyte. The potentiostatic cyclic voltammetry curves of V2O5 were performed at a scan rate of 0.1 mV s-1 in the range from 0.3 to 0.8 V vs. Ag/AgCl and did not reveal a significant increase of current at the upper voltage limit, indicating there was no electrolyte decomposition. This is as expected as water decomposition only occurs above 1 V in this system. The CV curves show well defined, reversible redox peaks, corresponding to the reversible intercalation and deintercalation of Na ions into the V2O5.nH2O bulk electrode material.53 The peak occurring at 0.09 V is due to the initial insertion of Na ions into V2O5.nH2O and the conversion from V2O5.nH2O into Na0.333V2O5.nH2O. The following peak at -0.04 V is due to the further insertion into the host lattice to further convert to Na0.666V2O5.nH2O. Finally, the peak at -0.25 V is due to the complete insertion of Na ions in the intercalation process, converting entirely to NaV2O5.nH2O and fully intercalating one Na ion to each molecule of V2O5.46,54 The peaks between 0.2 V and 0.4 V during the charge cycle indicate the reversible nature of these intercalation events as the Na ions are removed from the host lattice.
The CV was performed 5 times and each scan is shown in Fig. 9. As can be seen, each individual scan shows that the electrochemical performance was approximately the same, suggesting that a supercapacitor device using this electrode would have good cycling stability and capacity retention. The peaks occur at the same voltage in each cycle indicating that the intercalation/deintercalation process is truly reversible and that the same V5+/V4+ redox couple is being accessed and the redox event is occurring in each individual charge/discharge cycle. However, the main disadvantage of V2O5 is still prevalent. These scans were performed at a 0.1 mV s-1 scan rate, which is decidedly slow. An increased scan rate may not yield as distinct redox peaks due to the slow diffusion rate of sodium.46
A representative cycle of the CV scan was chosen to be used to calculate the sodium intercalation capacitance of the V2O5.nH2O electrode, shown in Fig. 9. The mass of active material on the electrode was measured to be 1.65 ± 0.1 mg. The CV curve was plotted and the capacitance C was calculated through a purpose-built Python script using the equation
C= ∫ViVfI dV2mν(Vf-Vi)
where the integral implies the area enclosed within the CV curve, m is the mass of the electrode active material, v is the scan rate, and Vfand Viare the upper and lower values of the potential window respectively. The specific capacitance was calculated to be 616.8 ± 5.7 F g-1. These results are useful in validating the mechanism of sodium ion intercalation into the V2O5.nH2O electrode material and confirming the stability of both the electrode and the electrolyte in an aqueous sodium environment.
The electrochemical performance of an V2O5.nH2O Na ion full cell aqueous sodium supercapacitor is shown in Fig. 10. The activated carbon electrode was used as an anode and 1M Na2SO4 as electrolyte. The sample was tested through potentiostatic cyclic voltammetry at several different scan rates in a voltage range of 0 V to 1V. The CV curves of the V2O5.nH2O supercapacitor exhibit the roughly rectangular shape expected of a supercapacitor CV curve. The CV curves do not indicate any significant unexpected current changes, meaning no part of the cell had decomposed. The total mass of active material of the V2O5.nH2O electrode and activated carbon electrode material was 7.06 ± 0.1 mg. The device’s specific capacitance was calculated to be 122.9 ± 1.7 mF g-1 at a scan rate of 0.1 mV s-1. As the scan rates are increased though, the capacitance of the device falls. The nature of the slow redox reactions and slow ion diffusion during Na ion insertion into the V2O5.nH2O lattice limits the rate capability of the device. Furthermore, despite excellent mass loading of electrode material on the substrate, some of the electrode volume may be inaccessible to the Na ions during diffusion. Pore openings with sizes that are too small may not allow for diffusion of the bulkier Na ion into the material and will limit the capacitance and cause greater internal resistance, also limiting the rate capability. The poor conductivity of V2O5 can also hinder device performance at higher scan rates. The full set of specific capacitance values for each scan rate can be found in the comparison Table 2.
Figure 10CV curves of V2O5.nH2O/activated carbon aqueous sodium full cell supercapacitor device.
It should also be noted that there was no significant increase in current at the upper limit of the voltage window, indicating that there was no electrolyte breakdown. Despite several cycles, the CV curve maintained the approximate shape expected of a supercapacitor device, even at higher scan rates, without any noticeable change in electrochemical behaviour evident from the CV. For V2O5 based sodium ion supercapacitors, specific capacities of up to 450 F g-1 have been reported.55,56 Our obtained values of 122.9 ± 1.7 F g-1 at a scan rate of 0.1 mV s-1 are unfortunately not as high as could be expected from such an electrode material. However, these reported values are within an organic sodium electrolyte device and the use of such a material in an aqueous sodium system has not been sufficiently investigated so to the best of our knowledge, no comparable results for such a system could be found. For context, an organic electrolyte based device should suffice. As organic electrolytes are capable of having a much larger voltage window, these are naturally expected to have a significantly higher specific capacitance. For such a device, the specific capacitance could be as much as 4 or 5 times greater than that for an aqueous device utilising the same electrode materials.3,4 As such, our obtained values appear to be comparable when considering such factors.
However, as mentioned, the main drawback of using a V2O5 electrode material and such a method of preparing it as an electrode is both the poor conductivity of V2O5 and of its stainless steel substrate. Despite obtaining a reasonably acceptable specific capacitance for a full cell supercapacitor device, it can be expected that addressing these issues would greatly improve the rate capability of the device. In an attempt to improve the V2O5.nH2O electrode, carbon was deposited onto as-prepared samples through a simple chemical vapour deposition method. SEM images of the samples (Fig. 11) with carbon deposited show significant addition of carbon onto the surface of the V2O5.nH2O nanoflakes.
Figure 11SEM images of samples with carbon deposition, marked in red.
Figure 12CV curves of graphene enhanced V2O5.nH2O/activated carbon aqueous sodium full cell supercapacitor device.
The deposited carbon is believed to be graphene layers and ideally is deposited onto the electrodes in low amounts so as to not make any pores within the V2O5.nH2O inaccessible. The combined redox capacitance from the V2O5.nH2O and the electric double layer charge storage contribution from the graphene in the electrode material is expected to give a higher capacitance.20 As well as this, the addition of carbon is also expected to increase the conductivity of the electrode, improving the rate capability. The electrochemical performance of the graphene enhanced V2O5.nH2O aqueous sodium supercapacitor is shown in Fig. 12.
The total active material mass of the graphene enhanced V2O5.nH2O electrode and the activated carbon electrode was 8.05 ± 0.1 mg and the specific capacity was calculated to be 176.5 ± 1.9 F g-1. This presents a significant increase from the heat-treated, as-prepared V2O5.nH2O electrode, likely due to the contribution of electric double layer charge storage due to the addition of carbon to the electrode material. However, as can be seen in the full set of specific capacitances in Table 2, the only improvement in specific capacitance was at the lowest scan rate of 0.1 mV s-1.
Table 2
Full set of CV results of V2O5.nH2O supercapacitors
Specific capacitance, as-prepared
mA h g-1 |
Specific capacitance, graphene enhanced
mA h g-1 |
Scan Rate mV s-1 |
122.9 | 176.5 | 0.1 |
51.4 | 49.0 | 1 |
18.1 | 11.9 | 10 |
7.9 | 5.4 | 50 |
5.5 | 2.7 | 100 |
Despite the improvement in specific capacitance, the rate capability of the supercapacitor was not in fact improved. The proposed cause is the overloading of carbon onto the surface of the electrode. If too great an amount of carbon is deposited onto the surface of the V2O5.nH2O nanoflakes may cover a large surface area, blocking access to the bulk of the material. This may obscure the larger pore sizes which facilitate quick diffusion of Na ions through the bulk material, removing many major pathways and preventing a similar level of diffusion as in the non-enhanced material. This effect would not be noticeable at low scan rates but becomes pronounced at higher scan rates. The improved conductivity of the electrode is not enough to overcome the higher effective resistance due to the inability of the Na ions to speedily diffuse through the material and insert into the V2O5.nH2O crystal lattice. Furthermore, the heating of the electrode material to temperatures of 650°C would remove a significant amount of water from the V2O5.nH2O structure, reducing the interlayer spacing within the crystal lattice. This would lead to reduce the overall capacitance and the rate performance of the electrode as Na ions would not be inserted into the host lattice as readily.
With the addition of graphene, the specific capacitance achieves a large increase. Furthermore, there is also the consideration of improving the cycling stability of the V2O5.nH2O electrode. The CV curves in Fig. 12 demonstrate the typical rectangular behaviour of a supercapacitor. However, when compared to the previous, non-enhanced sample CV curves in Fig. 10, the enhanced electrode CV curves are distinctly smoother and maintain a more consistent behaviour as the scan rate is increased. This implies an increased structural stability under higher rates and can improve the cycle life of the device. However, the addition of graphene is primarily to improve the rate capability of the V2O5.nH2O electrode, which was not achieved. It has been proposed that more precise tuning and control of the carbon growth onto the surface of the material may allow for enough carbon to be deposited so as the improve the rate capability through increased electrical conductivity, but not so much as to prevent the normal diffusion of Na ions through the bulk material.57
These data show good potential of V2O5.nH2O as a suitable electrode material for the construction of an aqueous sodium supercapacitor. A coin cell using this electrode material exhibited promising specific capacities and good cycling stability. CV curves of both as-prepared heat-treated V2O5.nH2O electrodes as well as graphene enhanced electrodes exhibit excellent specific capacitance values. The main issue remains the deficient rate capability of such an electrode, due to the poor conductivity of V2O5. Many improvements could be made to build upon the work that has been conducted throughout this project, such as optimising the nanostructured carbon deposition process. However, simply obtaining more data and repeating some of these analyses would help achieve a better understanding of how an aqueous sodium supercapacitor may be created utilising this electrode. Optimising the mass loading of V2O5.nH2O on the substrate during the synthesis process and ensuring consistent growth is also an area that could be improved. Ideally, a full cell supercapacitor would have been created and analysed through galvanostatic charge/discharge cycling to obtain a true capacity for a complete supercapacitor device. This kind of testing would also investigate the feasibility of creating a commercial device using this technology. Due to the nature of the type of testing required to obtain meaningful data to investigate these parameters, significantly more time and dedication would be required to improve upon the results of this research. Unfortunately, the charging and discharging of a coin cell for 100 cycles at varying rates over several weeks to obtain more cycling stability data simply was not an available luxury. Indeed, the ultimate goal of creating a “salt water battery” would require work far beyond the scope of this project. However, through these results, the successful creation of an aqueous sodium ion supercapacitor has been demonstrated. The apparent next step would be in attempting to create a similar device using NaCl as electrolyte, rather than Na2SO4. If such a device was manufactured successfully, the possibility to obtaining clean water from salt water while storing significant energy could become a reality.
3.3 Systematic errors and problematic influences
At this point, it is worth considering some systematic errors and problematic influences on the work and measurement conducted during this project.
Firstly, the measuring of the mass of the electrodes was the largest cause of potential errors in the final data. While the stainless steel substrate for all electrodes were cut using the same machine and to the same dimensions, it is not out of the question that certain meshes were of different mass due to deformations in the mesh. This of course leads to potentially erroneous measurements of the mass of the synthesised V2O5.nH2O electrodes. To prevent this becoming a large issue, the mass of each stainless steel mesh substrate was accurately measured prior to the hydrothermal process to ensure consistent mass throughout. Once each electrode had been prepared, the mass of each electrode was also measured and the active material mass calculated through subtraction, which leads to potential propagation of any uncertainties from previous measurements. Furthermore, these could cause inaccuracies in the calculation of the capacities or current densities of measured cells. Accordingly, any calculated values are presented with the necessary uncertainties.
The methods used in electrochemical testing of the samples are standard procedures commonly used to test supercapacitor devices. However, these methods also come with some systematic errors that are difficult to eliminate. Although both the V2O5.nH2O cathode and activated carbon anode were fabricated to have the same cross-sectional area, during assembly of coin cells and Swagelok-type cells, it cannot be ensured that the device maintained the same overlapping area. This may therefore cause imbalances in current density across the device on a particular electrode, leading to possible errors in the charge/discharge measurements. Furthermore, due to the nature of the Swagelok-type cells used for full cell testing, there were additional contributions to the resistance of the supercapacitor that would potentially affect the rate capability of the device. This additional resistance would be problematic to accurately measure and to take into account when calculating the actual specific capacitance of the supercapacitor device.
While the materials used to make the activated carbon electrode is standard and incorporated in many devices,5 the focus of this work was on the V2O5.nH2O cathode. Therefore, very little consideration was given to the activated carbon anode and given the time constraints of this project, there was no opportunity to investigate whether or not the anode was inhibiting the performance of the supercapacitor. If the specific capacity of the carbon anode was in fact lower than that of the V2O5.nH2O electrode, then, due to the operating principles of a supercapacitor, the specific capacitance of the device would in fact be lower than the maximum that could be obtained, as not all the active mass of the V2O5.nH2O is hosting Na+, even if all accessible area of the activated carbon electrode is adsorbing the anion. While this issue was addressed by simply increasing the mass loading on the activated carbon electrode to 3 or 4 times that of the V2O5.nH2O, because of time constraints, whether or not this attempt to bypass the problem was successful could not be investigated.
4 Conclusions
In summary, a V2O5.nH2O nanoflake electrode was prepared by a hydrothermal method, depositing the material directly onto a stainless steel substrate for simple supercapacitor electrode construction. The material was heat-treated at 300°C to dehydrate and improve the crystallinity of the electrode material. Cell utilizing this electrode and an aqueous sodium electrolyte were fabricated and cell testing of this material showed good electrochemical performance. The unique nanoflake morphology makes excellent contact with the electrolyte to shorten the diffusion length for the charge carrying Na ions and the crystal structure of the V2O5.nH2O material allows for smooth intercalation of Na ions. This method of synthesizing electrodes and the promising electrochemical performance of the material in aqueous sodium supercapacitors is expected to be helpful in future designs of an aqueous sodium “salt water battery”.
5 Bibliography
1. Nair, N. K. C. & Garimella, N. Battery energy storage systems: Assessment for small-scale renewable energy integration. Energy Build. 42, 2124–2130 (2010).
2. Miller, J. R., Simon, P. & Patrice, S. Electrochemical Capacitors for Energy Management. Science (80-. ). 321, 651–652 (2008).
3. Kötz, R. & Carlen, M. Principles and applications of electrochemical capacitors. Electrochim. Acta 45, 2483–2498 (2000).
4. Sharma, P. & Bhatti, T. S. A review on electrochemical double-layer capacitors. Energy Convers. Manag. 51, 2901–2912 (2010).
5. Pandolfo, A. G. & Hollenkamp, A. F. Carbon properties and their role in supercapacitors. J. Power Sources 157, 11–27 (2006).
6. Christen, T. & Carlen, M. W. Theory of ragone plots. Journal of Power Sources 91, 210–216 (2000).
7. González, A., Goikolea, E., Barrena, J. A. & Mysyk, R. Review on supercapacitors: Technologies and materials. Renew. Sustain. Energy Rev. 58, 1189–1206 (2016).
8. Conway, B. E. Transition from ‘Supercapacitor’ to ‘Battery’ Behavior in Electrochemical Energy Storage. J. Electrochem. Soc. 138, 1539 (1991).
9. Frackowiak, E. Carbon materials for supercapacitor application. Phys. Chem. Chem. Phys. 9, 1774 (2007).
10. Endo, M., Takeda, T., Kim, Y. J., Koshiba, K. & Ishii, K. High Power Electric Double Layer Capacitor ( EDLC ’ s ); from Operating Principle to Pore Size Control in Advanced Activated Carbons. Carbon Sci. 1, 117–128 (2001).
11. Bagotsky, V. Fundamentals of Electrochemistry. (John Wiley & Sons, Inc, 2005).
12. Zhang, L. L. & Zhao, X. S. Carbon-based materials as supercapacitor electrodes. Chem. Soc. Rev. 38, 2520 (2009).
13. Shi, H. Activated carbons and double layer capacitance. Electrochim. Acta 41, 1633–1639 (1996).
14. Qu, D. Studies of the activated carbons used in double-layer supercapacitors. J. Power Sources 109, 403–411 (2002).
15. Kim, Y. J. et al. Correlation between the pore and solvated ion size on capacitance uptake of PVDC-based carbons. Carbon N. Y. 42, 1491–1500 (2004).
16. Chmiola, A. J. et al. 249 Anomalous Capacitance Less Than Increase in Carbon at Pore Sizes. 313, 1760–1763 (2015).
17. Vix-Guterl, C. et al. Electrochemical energy storage in ordered porous carbon materials. Carbon N. Y. 43, 1293–1302 (2005).
18. Pell, B. E. C. Æ. W. G. Double-layer and pseudocapacitance types of electrochemical capacitors and their applications to the development of hybrid devices. 637–644 (2003). doi:10.1007/s10008-003-0395-7
19. Simon, P. & Burke, A. Nanostructured Carbons: Double-Layer Capacitance and More.
20. Chuang, C., Huang, C., Teng, H. & Ting, J. Effects of Carbon Nanotube Grafting on the Performance of Electric Double Layer Capacitors. 6476–6482 (2010). doi:10.1021/ef101208x
21. Zhu, Z., Wang, G., Sun, M., Li, X. & Li, C. Electrochimica Acta Fabrication and electrochemical characterization of polyaniline nanorods modified with sulfonated carbon nanotubes for supercapacitor applications. Electrochim. Acta 56, 1366–1372 (2011).
22. Lee, H., Suk, M., Hoi, I., Do, J. & Lee, Y. RuOx / polypyrrole nanocomposite electrode for electrochemical capacitors. Synth. Met. 160, 1055–1059 (2010).
23. Metallization, E., Organic, O. F., Using, P., With, P. & Anions, Z. 161 electroless metallization of organic polymers using the polymer as a redox reagent: reaction of polyimide with zintl anions. 102, 161–172 (1983).
24. Soc, C., Wang, G. & Zhang, J. Chem Soc Rev CRITICAL REVIEW A review of electrode materials for electrochemical supercapacitors. 797–828 (2012). doi:10.1039/c1cs15060j
25. Ellenbogen, J. C. Supercapacitors : A Brief Overview. (2006).
26. Kühnel, R.-S., Reber, D. & Battaglia, C. A High-Voltage Aqueous Electrolyte for Sodium-Ion Batteries. ACS Energy Lett. 2005–2006 (2017). doi:10.1021/acsenergylett.7b00623
27. Wang, H. et al. Open-Structured V2O5·nH2O Nanoflakes as Highly Reversible Cathode Material for Monovalent and Multivalent Intercalation Batteries. Adv. Energy Mater. 7, 1–8 (2017).
28. Pasta, M. et al. Manganese–cobalt hexacyanoferrate cathodes for sodium-ion batteries. J. Mater. Chem. A 4, 4211–4223 (2016).
29. Lu, K. et al. An advanced aqueous sodium-ion supercapacitor with a manganous hexacyanoferrate cathode and a Fe 3 O 4 /rGO anode. J. Mater. Chem. A 3, 16013–16019 (2015).
30. Ellis, B. L. & Nazar, L. F. Sodium and sodium-ion energy storage batteries. Curr. Opin. Solid State Mater. Sci. 16, 168–177 (2012).
31. Porada, S., Zhao, R., Van Der Wal, A., Presser, V. & Biesheuvel, P. M. Review on the science and technology of water desalination by capacitive deionization. Prog. Mater. Sci. 58, 1388–1442 (2013).
32. Sivakkumar, S. R. & Pandolfo, A. G. Electrochimica Acta Evaluation of lithium-ion capacitors assembled with pre-lithiated graphite anode and activated carbon cathode. Electrochim. Acta 65, 280–287 (2012).
33. Wu, C. & Xie, Y. Promising vanadium oxide and hydroxide nanostructures : from energy storage to energy saving. 1191–1206 (2010). doi:10.1039/c0ee00026d
34. Links, D. A. ChemComm The rechargeable aluminum-ion battery. 12610–12612 (2011). doi:10.1039/c1cc15779e
35. Gu, S. et al. Con fi rming reversible Al 3 þ storage mechanism through intercalation of Al 3 þ into V 2 O 5 nanowires in a rechargeable aluminum battery. Energy Storage Mater. 6, 9–17 (2017).
36. Takahashi, K., Wang, Y. & Cao, G. Ni – V 2 O 5 ‚ n H 2 O Core – Shell Nanocable Arrays for Enhanced Electrochemical Intercalation. 48–51 (2005). doi:10.1021/jp044772j
37. Petkov, V. et al. Structure of V 2 O 5 · nH 2 O Xerogel Solved by the Atomic Pair Distribution Function Technique. 10157–10162 (2002). doi:10.1021/ja026143y
38. Liu, D., Liu, Y., Candelaria, S. L. & Cao, G. Atomic layer deposition of Al 2 O 3 on V 2 O 5 xerogel film for enhanced lithium-ion intercalation stability. 30, 1–6 (2012).
39. Liu, Q. et al. electrochemical performance for Li-ion batteries. Nat. Commun. 6, 1–10 (2015).
40. Pan, G. X., Xia, X. H., Cao, F., Chen, J. & Zhang, Y. J. Carbon cloth supported vanadium pentaoxide nanoflake arrays as high-performance cathodes for lithium ion batteries. Electrochim. Acta 149, 349–354 (2014).
41. Zhai, T. et al. Centimeter-long v2O5 nanowires: From synthesis to field-emission, electrochemical, electrical transport, and photoconductive properties. Adv. Mater. 22, 2547–2552 (2010).
42. Perera, S. D. et al. Vanadium oxide nanowire-carbon nanotube binder-free flexible electrodes for supercapacitors. Adv. Energy Mater. 1, 936–945 (2011).
43. Tepavcevic, S. et al. Nanostructured bilayered vanadium oxide electrodes for rechargeable sodium-ion batteries. ACS Nano 6, 530–538 (2012).
44. Braithwaite, J. S., Catlow, C. R. A., Gale, J. D. & Harding, J. H. Lithium Intercalation into Vanadium Pentoxide: a Theoretical Study. Chem. Mater. 11, 1990–1998 (1999).
45. Shklover, V. & Haibach, T. Crystal Structure of the Product of Mg 2 ؉ Insertion into V 2 O 5 Single Crystals. J. Solid State Chem. 323, 317–323 (1996).
46. Wang, D., Liu, H., Elliott, J. D., Liu, L.-M. & Lau, W.-M. Robust vanadium pentoxide electrodes for sodium and calcium ion batteries: thermodynamic and diffusion mechanical insights. J. Mater. Chem. A 4, 12516–12525 (2016).
47. Morgan, D. J. Modern Powder Diffraction. Clay Miner. 25, 544–545 (1990).
48. Dinnebier, R. E. & Billinge, S. J. L. Powder diffraction: Theory and practice. Theory and practice (2008). doi:10.1039/9781847558237
49. JEOL. Scanning Electron Microscope A To Z. Serv. Adv. Technol. 32 (2006). doi:10.1017/S1431927605507797
50. Seiler, H. Secondary electron emission in the scanning electron microscope. J. Appl. Phys. 54, (1983).
51. Pan, A. Q., Wu, H. Bin, Zhang, L. & Lou, X. W. (David). Uniform V2O5 nanosheet-assembled hollow microflowers with excellent lithium storage properties. Energy Environ. Sci. 6, 1476 (2013).
52. Yue, Y. & Liang, H. Micro- and Nano-Structured Vanadium Pentoxide (V2O5) for Electrodes of Lithium-Ion Batteries. Adv. Energy Mater. 7, 1–32 (2017).
53. Wu, J. et al. A Scalable Free-Standing V2O5/CNT Film Electrode for Supercapacitors with a Wide Operation Voltage (1.6 V) in an Aqueous Electrolyte. Adv. Funct. Mater. 26, 6114–6120 (2016).
54. Chao, D. et al. A V2O5/conductive-polymer core/shell nanobelt array on three-dimensional graphite foam: A high-rate, ultrastable, and freestanding cathode for lithium-ion batteries. Adv. Mater. 26, 5794–5800 (2014).
55. Zhu, J. et al. Building 3D structures of vanadium pentoxide nanosheets and application as electrodes in supercapacitors. Nano Lett. 13, 5408–5413 (2013).
56. Khoo, E., Wang, J., Ma, J. & Lee, P. S. Electrochemical energy storage in a β-Na0.33V2O5 nanobelt network and its application for supercapacitors. J. Mater. Chem. 20, 8368 (2010).
57. Qin, M. et al. Facile synthesis of multiwalled carbon nanotube–V2O5 nanocomposites as cathode materials for Li-ion batteries. J. Solid State Electrochem. 18, 2841–2846 (2014).
This page intentionally left blank.