Translate

2024年11月27日星期三

Comparative Study of Titanium 32×MXene Coated Carbon Electrodes and Thermally Treated Carbon Electrodes for Vanadium Redox Flow Batteries in RSC Advances


1.Research Background

A major challenge in vanadium redox flow batteries is the competition between the main and side reactions of hydrogen evolution associated with the V(II)/V(III) redox couple at the negative electrode.

2.Research Question

A major challenge in vanadium redox flow batteries (RFBs) is the competition between the main and side reactions of hydrogen evolution associated with the V(II)/V(III) redox reaction at the negative electrode. Vanadium redox flow batteries (VRFBs) are suitable for long-term and fixed energy storage applications. So far, the primary chemical composition of commercialized RFBs is vanadium, and VRFBs are particularly favored due to the four oxidation states of vanadium, which facilitate the use of vanadium-based electrolytes on both sides of the battery. Using the same components in both compartments minimizes the issue of electrolyte contamination. VRFBs are not affected by overcharging or deep discharge, and the highly stable chemical nature of vanadium allows VRFBs to have lifetimes ranging from 15,000 to 20,000 cycles. One of the main challenges in VRFBs is at the negative electrode, where side reactions involving hydrogen evolution occur due to the negative standard potential of the V(II)/V(III) reaction. Appropriate electrode treatments can improve the V(II)/V(III) reaction while suppressing parasitic hydrogen evolution.

The commonly used electrode materials in VRFBs are carbon-based materials, which offer high chemical stability, a wide operating voltage range, and low cost. Carbon felt, paper, and cloth are the most commonly used forms of carbon-based electrodes in VRFBs. Numerous electrode treatment methods have been explored to enhance the electrochemical reaction of vanadium, including thermal treatment, chemical treatment (using strong acids), and electrochemical oxidation. These methods, referred to as intrinsic treatments, involve adding oxygen-containing functional groups to increase the specific surface area of the electrode and the number of active reaction sites, thus improving the hydrophilicity and catalytic properties of the electrode. Another strategy is to coat metal and metal oxide electrocatalysts on the carbon-based electrodes to improve their conductivity and catalytic performance. Similar effects have also been observed by decorating the electrodes with carbon nanomaterials.

To achieve all of these treatment effects, MXene coating can be a more direct approach to enhance the electrocatalytic performance of carbon-based electrodes and improve VRFB performance. MXenes, a family of two-dimensional transition metal carbides and nitrides introduced in 2011, have attracted much attention due to their unique combination of properties, such as high surface area, high electrical conductivity (up to 21,000 S/cm), solution processability (with a water electrostatic potential of 40-60 mV), and rich surface chemistry. As a result, the electrochemical applications of MXenes have significantly grown, and they have been used in various electrochemical energy devices.

Although both compartments of the battery use vanadium electrolytes, the reaction kinetics of the V(II)/V(III) and V(IV)/V(V) redox couples differ. Previous studies have shown that the catalytic activity at the negative electrode highly depends on electrode treatment, with the kinetic rate of V(IV)/V(V) exceeding that of V(II)/V(III). While MXenes have been used in various applications, their use in flow batteries is still relatively new. A. M. Mizrahi et al. pre-treated carbon electrodes with plasma before casting titanium, and found that the optimal MXene coating density was 0.1 mg/cm². M. Jingjing et al. performed thermal treatment at 350°C for 1 hour to enhance the hydrophilicity of the electrode, followed by repeatedly immersing the carbon felt in an MXene dispersion. Their results showed that the diffusion coefficient and reaction rate of the MXene-coated carbon felt were two orders of magnitude higher than that of the original carbon felt. Furthermore, L. Wei et al. achieved an energy efficiency of 81.3% in battery tests, with a current density of 200 mA/cm², by soaking and drying graphite felt dispersed in MXene and Nafion, using Nafion as a binder.

To further simplify the process without the need for any pretreatment or binder, in this work, we developed a simple drop-casting technique to address the hydrophobic nature of the raw carbon paper. The electrodes were first wetted, followed by drop-casting of titanium 3C₂Tx MXene dispersion. Due to the inherent hydrophilicity of titanium, the addition of MXene made the carbon electrode more hydrophilic. Our MXene drop-casting method does not require any complex pretreatment equipment, which can reduce costs and simplify the process for commercial-scale applications. Electrochemical characterization and battery tests show that the MXene-coated electrodes have high electrochemical activity and stability with vanadium electrolytes. In addition, scanning electron microscopy (SEM) imaging and X-ray photoelectron spectroscopy (XPS) analysis were used to investigate the surface characteristics of the electrodes, clearly evaluating the exposed surface area and distribution of the MXene coating.

Materials for the Experiment

Carbon paper (GDL, SiC-type 28 AA) and Nafion 212 (N212) membrane were purchased from a fuel cell store. Vanadium (IV) oxide sulfate hydrate, sulfuric acid, and acetone were purchased from Sigma-Aldrich. Hydrofluoric acid (HF, 48-51% aqueous solution) was obtained from Acros Organics. Lithium chloride (LiCl, 98% purity, Thermo Scientific) and hydrochloric acid (HCl, 12M) were obtained from Fisher Scientific and were used as received. A commercial 1.6 M V(III)/V(IV) electrolyte mixture, H₂SO₄ for battery testing, was purchased from GfE (German Electro Metallurgical Company). All chemicals used in cyclic voltammetry and battery testing were not further purified.

MXene Preparation

To synthesize Ti₃C₂ₓ MXene, the material was first washed with 9M hydrochloric acid (HCl) for 18 hours to remove metallic impurities, then mixed with an etching solution consisting of a 3:1 mixture of 12M HCl, deionized water, and 28.4M hydrofluoric acid (HF) by volume. The mixture was stirred at 400 rpm for 24 hours at 35°C. The etched Ti₃C₂ₓ MXene was then washed with deionized water by repeated centrifugation at 3234 RCF (using about 200 mL of deionized water for 4-5 cycles) until the supernatant reached a pH of 6. For delamination, the etched multilayer Ti₃C₂ₓ MXene precipitate was added to a LiCl solution (typically 50 mL of LiCl solution per gram of starting etched powder). The LiCl and multilayer MXene mixture was then stirred at 400 rpm under a constant argon gas flow at 65°C for 1 hour. The mixture was washed with deionized water and centrifuged at 3234 RCF for 5, 10, 15, and 20 minutes. The final mixture was vortexed for 30 minutes and then centrifuged at 2380 RCF for 30 minutes to ensure that the MXene solution consisted of few-layer to monolayer sheets. The final suspension of Ti₃C₂ₓ MXene was obtained.

Electrode Preparation

The original carbon paper was cut into 5 cm² pieces, which were labeled as untreated carbon paper electrodes. They were then heat-treated in a muffle furnace (Nabertherm L-051H1RN1T5/11/B410) at 500°C in ambient air for 3 hours, with a heating rate of 167°C per hour. The MXene slurry was stored in a freezer at -15°C and was thawed at room temperature before use. Ultrapure water (Milli-Q) was mixed with the MXene slurry to obtain a concentration of 5 mg/ml, and then sonicated for 60 minutes in an ultrasonic cleaning bath (VWRUSC300T).

The untreated carbon paper electrodes, which are hydrophobic, were coated with MXene using a continuous process. The process began by immersing the electrode in an acetone solution to ensure complete wetting. The electrode was then rinsed with water to ensure the complete removal of acetone. Under normal atmospheric conditions, the wetted electrode was immediately coated with a 5 mg/ml MXene dispersion using a micropipette. The electrode was then dried in a vacuum oven (Heraeus D-6450 Hanau) at 100°C for 1 hour. The electrodes were coated on both sides with 0.1 ml, 0.5 ml, and 1 ml of MXene dispersion, achieving coating densities of 0.1 mg/cm², 0.5 mg/cm², and 1.0 mg/cm², respectively. These electrodes were labeled as MX 0.1, MX 0.5, and MX 1.

Electrode Characterization

The surface morphology of the electrodes was analyzed using a Hitachi S-4800 scanning electron microscope (SEM) with an acceleration voltage of 10 kV, a working distance of 9400 μm, and an emission current of 10.1. No sputtering was performed. Surface composition analysis was carried out using X-ray photoelectron spectroscopy (XPS) with monochromatic Al Kα radiation (1486.6 eV) on a Kratos Axis instrument, calibrated with the C 1s peak at 284.5 eV. All XPS data processing was performed using LG4X-V2 software.

Cyclic Voltammetry (CV)

Cyclic voltammetry (CV) was performed in a three-electrode setup in a beaker. A 40 mL mixture of vanadium (IV) sulfate hydrate and sulfuric acid in water was prepared by dissolving vanadium (IV) sulfate, using 50 mM and 50 mM sulfuric acid solution. A VersaSTAT 4 potentiostat was used for CV measurements. A platinum mesh was used as the working electrode, and an Ag/AgCl reference electrode was employed. A 5 cm² electrode sample was fixed on the working electrode clamp and fully immersed in the solution. The CV scans for the negative side (V(II)/V(III)) used a potential window from 0 V to -0.75 V, while the positive side (V(IV)/V(V)) used a potential window from 0 V to 1.20 V. The scan rates for the negative side (V(II)/V(III)) were 2, 4, 6, 8, and 10 mV/s, and for the positive side (V(IV)/V(V)), the scan rates were 3, 5, 10, 20, 30, and 50 mV/s. When the baseline could not be determined, the peak anodic current density (jpa) was estimated using the semi-empirical method proposed by Nicholson, based on the peak cathodic current density (jpc), the uncorrected peak current relative to the baseline of the cathodic peak (jpa, 0), and the current density at the switching potential (jsp, 0).

(1)The CV data can be further analyzed by calculating the reaction rate and diffusion coefficient. For a single-electron transfer process, when Δp is greater than 200 mV, the process is considered irreversible. Therefore, the relationship between the peak current density and the diffusion coefficient is expressed by Equation (2), which is related to the reaction rate constant (k₀), as described by Randles-Ševčík.

 


The variables used in this study include: jp, which represents the peak current density based on the geometric area (A/cm²), α, the charge transfer coefficient, cb the bulk concentration (mol/cm³), and the variables used in this study are given by Equations (1) and (2). D represents the diffusion coefficient (cm²/s), ε represents the scan rate (s⁻¹), n represents the number of electrons, f represents the Faraday constant (C/mol), E_p represents the peak potential (V), and E₀' represents the formal potential (V). The diffusion coefficient can be determined by plotting jp against the square root of the scan rate and calculating assuming α = 0.5 from the slope. When jp is plotted based on the difference between the peak potential and formal potential, the reaction rate constant can be determined from the intercept.

Performance of a Single VRFB Cell

The VRFB test was conducted using an electrolyte containing 1.6 M V(III) and V(IV) in 2 M sulfuric acid. Electrochemical cells (Fuel Cell Technologies Inc.) were provided with two Poco graphite plates with serpentine flow fields and two 200 μm Viton gaskets. The electrode configuration is presented in Table 1. In this setup, each side of the cell is composed of two square electrodes, each with a geometric area of 5 cm². The electrolyte is injected through a single serpentine graphite flow field into the electrodes, with the active area corresponding to a geometric area of 5 cm². Nafion 212 is used as the proton exchange membrane. Both electrolytes are 13 mL in volume, and the external glass bottle containing the negative electrolyte is purged with nitrogen to prevent oxidation by atmospheric oxygen.

The voltage efficiency (VE), coulombic efficiency (CE), and energy efficiency (EE) of the battery are calculated using Eqn (4)–(6), where Idc and Icc are the discharge current and charge current, respectively, and Id and Vc are the corresponding values for the discharge and charge currents.

 


Results and Discussion

The carbon electrodes coated with Ti₃C₂Tₓ MXene reveal a porous surface structure that effectively holds the fibers together with the binder material. The carbon paper electrodes were coated with Ti₃C₂Tₓ MXene at three different loadings, labeled as MX 0.1, MX 0.5, and MX-1 (see Methods section). As the loading of Ti₃C₂Tₓ MXene increases from 0.1 mg/cm² (MX 0.1) to 1 mg/cm² (MX-1), the pores in the electrode are gradually filled. As shown in Figure 1b, it is evident that MX-0.1 leads to a reduction in the number of exposed surfaces and pores on the carbon paper. With increasing MXene loading (MX-1), the exposed surface area further decreases, and most of the pores are filled with Ti₃C₂Tₓ MXene. At lower MXene loadings (e.g., MX-0.1), some carbon fibers remain exposed (indicated by the red arrows in the figure), while at higher Ti₃C₂Tₓ MXene loadings (1.0 mg/cm²), the fibers are completely covered, as shown in Figure 1c. The large surface area is crucial for reducing activation losses. High hydraulic permeability promotes electrolyte transport and reduces pump losses. When the MXene loading is increased from 0.1 mg/cm² to 1 mg/cm², the exposed carbon fibers covering the electrochemical active sites are covered, and the open pores, which are crucial for the hydraulic permeability of the carbon paper, are mainly covered and sealed. As a result of these observations, we expect that lower MXene loadings (MX-0.1 and MX-0.5) more effectively provide redox catalytic activity while maintaining a high active surface area and electrolyte transport.

 


Scanning electron microscope images at 500 μm and 50 μm scales (insets). (a) Untreated carbon paper, (b) untreated carbon paper coated with MXene at a coating density of 0.1 mg/cm², and (c) untreated carbon paper coated with MXene at a coating density of 1.0 mg/cm². The red arrows in the insets of (a) and (b) point to the exposed carbon fibers, while (c) shows complete coverage of the carbon fibers by MXene.

XPS analysis of the treated carbon paper reveals additional surface functionalities and elements introduced by the thermal treatment and MXene coating. For calibration purposes, the C 1s peak at 284.5 eV was used, taking the high conductivity of the carbon paper as a reference point. The curve-fitting results for carbon (C), oxygen (O), titanium (Ti), fluorine (F), and chlorine (Cl) ions are shown in the figures. In Figure 2a, the overall survey spectrum displays significant peaks corresponding to carbon (C1s) at 284.5 eV, oxygen (O1s) at 532.7 eV, titanium (Ti 2p) at 454.7 eV, fluorine (F1s) at 684.8 eV, and chlorine (Cl 2p) at 199.0 eV, attributed to the introduction of Ti₃C₂Tₓ MXene. In Figure 2b, the C1s spectrum shows a distinct peak originating from C at 284.5 eV. As we move to higher binding energies at 286.0 eV, 288.4 eV, and 291.3 eV, these can be attributed to C-O, CO, and π-π* vibration features. Furthermore, in the MX-0.1 and MX-0.5 electrodes, the presence of titanium carbide peaks at 282.0 eV was observed when MXene was combined. From Figure 2a, after thermal treatment, the chemical adsorption of oxygen on the carbon paper surface remains unchanged, with the same atomic percentage (13%) as untreated carbon paper, indicating equivalent and uniform composition. Deconvolution of the C1s and O1s spectra revealed the presence of two oxygen-containing functional groups, particularly C-O at 532.9 eV and O at 531.3 eV. The percentage of carbon in the O1s curve fitting results shows a significant difference between untreated carbon paper (3.5%) and thermally treated carbon paper (28%). This suggests a preference for the formation of C-O groups during thermal treatment, while the atomic percentage of oxygen remains constant for the O-containing groups during the process.

 


Component peak fitting of the prepared carbon paper electrodes. (a) Wide XPS survey spectra of untreated carbon paper, thermally treated carbon paper, MX-0.1, and MX-0.5. (b) C1s, (c) O1s, (d) Ti2p, (e) F1s, (f) Cl2p.

The introduction of Ti₃C₂Tₓ MXene on the MX-0.1 and MX-0.5 electrodes led to the presence of two species in the O1s spectrum: CO (531.5 eV) and Ti-O (529.8 eV), resulting in an oxygen atomic percentage of 10% for both MX-0.1 and MX-0.5. Previous studies confirm that the Ti 2p spectrum reveals multiple peaks corresponding to different oxidation states, with the peak at 454.7 eV corresponding to titanium carbide (Ti-C), followed by Ti²⁺ (455.7 eV) from titanium oxide, Ti³⁺ (456.6 eV) from titanium fluoride, and Ti⁴⁺ (458.3 eV) from TiO₂. Since the MXene coating was performed at room temperature and the MXene-coated carbon paper was not treated, we do not expect MXene to have undergone significant oxidation. However, it is essential to quantify any potential oxidation of MXene during storage, as oxidation could reduce the high electrical conductivity of MXene. In the Ti 2p spectrum (Figures 2c and d), small amounts of Ti were confirmed to be present, indicating the termination groups, with peaks for Ti-F at 685.0 eV and Ti-Cl at 199.1 eV, similar to previous reports. For MX-0.1 and MX-0.5, the atomic percentages of oxygen, fluorine, and chlorine were recorded as 10%, 3%, and 1%, respectively. Notably, MX-0.5 showed a higher titanium atomic percentage (31%) compared to MX-0.1 (25%). Compared to the thermal treatment process, the MXene coating method offers unique advantages due to its inherent hydrophilicity, attributed to the termination groups. By adding MXene as a coating on the original carbon paper using the drip-casting process proposed in this work, the hydrophilicity was improved without requiring additional pre-treatment.

he cyclic voltammetry (CV) in a three-electrode system was used to investigate the electrochemical performance of the electrodes. The ratio of the anodic peak current to the cathodic peak current (j_pa/j_pc) and the peak-to-peak separation (ΔE_p) of 57 mV indicates an ideal reversible reaction. A derived j_pa/j_pc of 1 and a ΔE_p of 57 mV are considered quasi-reversible. The CV of the electrode samples is shown in Figure 3. The untreated electrode does not show any peaks corresponding to the V(II)/V(III) redox species, as shown in Figure 3b, indicating that the untreated electrode does not catalyze the redox reaction of V(II)/V(III) species, only a hydrogen evolution peak is observed. On the other hand, the untreated carbon paper does show the V(IV)/V(V) redox species as seen in the figure.

 


The cyclic voltammograms of different electrodes at varying scan rates are shown. Panels (a) and (b) represent the positive and negative sides of the untreated carbon paper electrode, respectively. Similarly, panels (c) and (d) show the negative and positive sides of the heat-treated carbon paper electrode. Panels (e) and (f) correspond to the MX-0.1 electrode, while panels (g) and (h) depict the MX-0.5 electrode.

As for the heat treatment process, Figures 3c and d show that the heat-treated carbon paper electrodes exhibit redox peaks on the negative side of the CV, whereas the MXene-coated carbon paper electrodes, with the V(II)/V(III) redox couple, show a significant enhancement from no activity to a pronounced response. The cathodic peak of the heat-treated carbon paper shifts to a lower potential on the negative side, and no peaks were observed above scan rates of 4 mV·s⁻¹. In Figure 3c, compared to the untreated carbon paper, the heat-treated carbon paper shows large non-Faradaic currents on the positive side, which is also observed for MX-0.1 and MX-0.5 electrodes. This can be attributed to the higher electrolyte wettability of the heat-treated carbon paper surface, leading to an increased electrochemical active surface area and higher capacitance compared to the untreated MX-0.1 and MX-0.5 electrodes.

By analyzing the relationship between anodic peak current, scan rate (v), and peak separation, the kinetics of the electrode towards the vanadium species were investigated, as shown in Figure 4. To determine the diffusion coefficient, j_pa was plotted against the square root of scan rate (v¹/²), which resulted in a linear curve. The slope of this curve (Figures 4a and c) was used in Eqn (2). The observed linearity, along with a peak separation greater than 200 mV, indicates an irreversible electrode process. Furthermore, to determine the reaction rate constant k₀, the natural logarithm of j_pa was plotted against (E_p - E₀'), which gives the intercept.

 


The relationship between current density and the square root of scan rate (panels a and c) shows a linear correlation. Additionally, there is a linear relationship between the natural logarithm of peak current density and peak separation. Subfigures (a) and (b) correspond to the negative side, while subfigures (c) and (d) correspond to the positive side.

3.Research Summary

In this work, we successfully demonstrated a simple drop-casting technique to coat MXene suspension onto untreated carbon electrodes. This method requires no heat treatment, binders, or any other pre-treatment steps. Our results show that this technique provides a simple and viable alternative for electrode modification, yielding electrochemical performance comparable to that achieved with standard heat treatment methods. SEM image analysis reveals that MXene particles exhibit minimal aggregation and are evenly distributed over the carbon paper surface. However, increasing the MXene loading may reduce the performance of the carbon paper. MX-0.1, with a MXene coating density of 0.1 mg cm⁻², effectively preserved the exposed surface area of the carbon paper, while higher loadings (MX-0.5 and MX-1.0) reduced the exposed surface area as MXene flakes filled the pores. Therefore, further studies are needed to determine the optimal coating density to enhance battery performance. The introduction of Ti₃C₂Tₓ MXene was found to introduce various terminal groups, including oxygen, chlorine, and fluorine, on the electrode surface. Using MX-0.1 as the anode and cathode, the battery achieved a capacity of 118 mAh, which is comparable to that of heat-treated carbon paper electrodes, with an energy efficiency (EE) of 68% and coulombic efficiency (CE) of 67%. In conclusion, the drop-casting technique for coating Ti₃C₂Tₓ MXene onto carbon paper is an attractive alternative to heat treatment. This simple modification enhances electrochemical performance and eliminates the need for complex pre-treatment steps. The results of this study open new possibilities for simple electrode modification techniques in various applications. For future research, it will be crucial to thoroughly examine long-term battery cycling, explore different current densities, and further optimize the coating density of MXene-coated carbon electrodes to assess stability and catalytic activity. Additionally, a deeper understanding of how MXene interacts with carbon paper and vanadium electrolytes is vital for advancing our knowledge in this field.

4.Canrd Brief Introduce

Canrd use high battery R&D technology(core members are from CATL) and strong Chinese supply chain to help many foreign companies with fast R&D. We provide lab materials, electrodes, custom dry cells, material evaluation, perfomance and test, coin/pouch/cylindrical cell equipment line, and other R&D services.
Email: contact@canrd.com    Phone/Wechat/WhatsApp: +86 19867737979

Canrd Official Web     Canrd Company Vedio     Canrd Company profile

Website : www.canrud.com

 




没有评论:

发表评论