The fabrication of Ti₃C₂ and Ti₃CN MXenes by electrochemical etching

Abstract

2D MXenes are well-known for their outstanding performance in electrochemical energy storage and many other applications owing to their high conductivity and specific surface area. An obstacle to the wider synthesis of MXenes for research and industrial applications is the use of hazardous hydrofluoric acid (HF) during their synthesis. Herein, we developed the electrochemical etching process for the synthesis of Ti₃C₂ and Ti₃CN MXenes by using aqueous tetrafluoroboric acid as the electrolyte, thus only involving a very low concentration of HF. The effect of electrical potential and temperature on the etching rate is studied and compared to chemical etching with HBF₄. A mechanism based on the selective anodic dissolution of aluminium from Ti₃AlC₂ and Ti₃AlCN with the tetrafluoroborate ion is proposed. The MXene formation was confirmed by Raman spectroscopy, X-ray photoelectron spectroscopy, X-ray diffraction and electron microscopy. The MXene flakes from the electrochemical etching process have larger lateral dimensions compared to chemically etched MXene flakes as a result of the suppression of the HF decomposition and rapid etching rate. The electrodes of lithium-ion supercapacitors made from electrochemically etched Ti₃C₂ and Ti₃CN exhibited cycle performance and rate capabilities comparable to HF-etched MXenes.

---

1. Introduction

MXenes, two-dimensional transition metal carbides, nitrides, and carbonitrides, have received a tremendous amount of research attention since its discovery in 2011.¹ The high specific surface area, hydrophilicity, conductivity, and mechanical strength of MXenes make them promising for applications including energy storage,^2–7 electromagnetic interference shielding,^8–10 and catalysis.¹¹ MXenes can be described as a hexagonal close packed structure with the composition M_n+1X_nT_x (n = 1 to 4; where M is an early d block transition metal; X is carbon or nitrogen; T is surface terminatal groups such as =O, –OH, –Cl, and –F, depending on the chemical etchant).¹² Terminal group free MXenes (M_n+1X_nT₀) are determined to have D_3d symmetry.¹³ However, the presence of the terminal groups distorts the symmetry of the unit cell. Hence, the literature generally refers to the unit cell of M_n+1X_nT₀ as pseudo-P6₃/mmc.¹⁴

MXenes are fabricated from corresponding MAX phases with the chemical composition of M_n+1AX_n, where A is either group 3 or 4 elements (e.g., Al, Ga, Si) lying between the MXene layers.^15–18 MXenes have very diverse chemical compositions due to the different combinations of M, A, and X elements. To date, more than 50 MXenes have been reported to be successfully synthesised.¹² The family of MXenes is further expanded when two transition metals are mixed in an MXene structure, in which the transition metal atoms may either be arranged randomly across all the M layers, or they may form ordered structures.^19–21 Given that the ratio of M and X in these double transition metal MAX precursors can be controlled as needed, the number of possible MXenes can be limitless.

Conventionally, the synthesis of MXene requires the use of hydrofluoric acid, either on its own or mixed with other acids (e.g., HCl, H₂SO₄),^{1,17,18,22,23} or the use of reagents that release HF in situ (e.g., LiF + HCl, known as the MILD approach, and NH₄HF₂) to remove the A layer from the MAX precursor.^24–26 However, HF is extremely hazardous to humans and the environment. Furthermore, the aggressiveness of HF causes unavoidable damage to the basal plane of the MXene layer, resulting in a small flake size of the delaminated MXenes. A few hydrothermal approaches have been reported to etch the MAX phase to MXene. Gentile et al. reported the hydrothermal etching of Ti₃C₂ using HBF₄ as an etchant.²⁷ Yet, the etching mechanism was still based on the decomposition of HBF₄ to HF upon heating to 60 °C. Another example, inspired by the Bayer process used in bauxite refining for the isolation of aluminium oxide,²⁸ is a fluorine-free synthesis of Ti₃C₂T_xvia alkali treatment reported by Li et al., which involves the use of 27.5 M of NaOH at 270 °C for 12 hours.²⁹ Peng et al. reported a hydrothermal etching of MAX phases (Ti₃AlC₂ and Nb₂AlC) by 12 M HCl and 0.5 M NaBF₄ for up to 32 hours at 180 °C.³⁰ Alternatively, non-aqueous approaches such as molten salt and halogen etching have also been reported.^31–36 However, the abovementioned approaches require harsh reaction conditions, making them costly and difficult to scale up. They may also damage, oxidise or hydrolyse MXene nanosheets.

Amongst all other approaches, electrochemical etching of MAX phases is particularly promising. Electrochemical fabrication of 2D materials is well studied for graphene, 2D-MoS₂, and mono-elemental Xenes.^37,38 In van der Waals solids, upon applying electrical potential, the ion intercalation or gas evolution between the 2D material layers causes the interlayer spacing to expand, overcoming the van der Waals forces and ultimately exfoliating the bulk layered material into its 2D form. The major advantage of electrochemical exfoliation is its environmental friendliness. Using graphene as the example, electrochemical exfoliation involves milder chemicals as the electrolyte (e.g., Na₂SO₄ or other sulfate salts) in contrast to the Hummers' method (concentrated H₂SO₄ and H₃PO₄ in combination with KMnO₄).^39,40 Despite the success of electrochemical exfoliation in the synthesis of graphene and other 2D materials, there are very few reports on the electrochemical fabrication of MXenes. The main obstacle originates from the unique interlayer interaction of the MXene layer in the MAX phase precursor. Unlike other bulk laminated materials, the M–A–M chemical bonding between the MXene layers in the bulk MAX phase precursor is significantly stronger than the physical van der Waals force between the layers in graphite. Thus, the electrochemical exfoliation processes need to chemically selectively etch the A layer in addition to expanding the interlayer distance. Furthermore, unlike the carbon in the strong sp²-hybridised framework in graphene, the transition metal in MXenes is prone to oxidize if the voltage applied is over its oxidation potential (c.f. for the anodic exfoliation of graphite, a voltage of up to +10 V can be applied).³⁹ Nonetheless, since the semi-covalent M–X bond is more thermodynamically stable than the weakly metallic M–A and A–A bonds, the selective oxidation of the A element is feasible under suitable conditions. The electrochemical etching of MAX phases has been reported using various types of HCl, NaCl, LiTFSI, and NH₄OH/NH₄Cl electrolytes.^41–46 However, the MXenes synthesised in those studies were mixed with carbide-derived carbon. This suggests that those electrolytes do not entirely remove the carbide generated during etching. Furthermore, the reaction setups reported in the literature are not scalable; working electrodes comprise MAX phases of ceramic pieces produced by cutting the ceramic or pressing the MAX phase powders into pellets.^43,45,46 However, the rate of the reaction would unavoidably be hindered by the limited surface area of the electrode. There are also reports of depositing MAX phase powders onto conductive material (e.g., carbon fiber clothes).^42,44 However, a binder is needed to ensure good contact. Furthermore, the use of CFC and the MXene produced via this method is always mixed with the fiber residues caused by carbon fiber destruction during the washing process.

Herein, we report a low-HF synthesis of Ti₃C₂ and Ti₃CN MXenes via electrochemical etching, using HBF₄ as the electrolyte. The proposed mechanism allows the Ti₃AlC₂ and Ti₃AlCN precursors to undergo selective anodic dissolution of aluminium with the tetrafluoroborate ion. With the addition of the etching pathway provided by the presence of electrical potential, the etching of MXene is quicker and occurs at a lower temperature, minimising the release of HF. The role of electrical potential in accelerating the etching was studied. Furthermore, the influence of temperature for both electrochemical and chemical etching with HBF₄ was studied. As a result of the suppression of HF decomposition and etching time, the electrochemically etched MXene exhibits a larger lateral size compared to the MXene produced by chemically etching in HF or HBF₄. The energy storage performance of the electrochemically etched Ti₃C₂ and Ti₃CN in lithium-ion batteries is tested and mimics the superior cycle performance and rate capabilities of chemically etched MXene. The electrochemically etched Ti₃C₂ and Ti₃CN anodes exhibit superior cycle performance and rate reversibility for lithium-ion supercapacitors. For clarification, the electrochemically etched Ti₃C₂, Ti₃CN and MXene will be labeled EE-Ti₃C₂, EE-Ti₃CN, and EE-MXene, respectively. MXenes that are chemically etched by HBF₄ will be labeled CE-MXene. MXene etched by HF and HCl will be labeled HF/HCl-MXene.

2. Results and discussion

2.1 Mechanism

Fig. 1a schematically shows the setup of the reaction. A graphite crucible was used as the reaction container and the current collector for the MAX phase powder. This approach allowed the etching of the MAX phase in its powder form without using a binder, as all MAX phase particles settled at the bottom of the graphite crucible during the etching process. Pt wire was used as the cathode. However, it is worth mentioning that the etching can also be carried out with graphite as the counter electrode, which suggests that the hydrogen evolution does not have to occur on the surface of platinum. For the electrochemical etching, the mechanism was proposed as follows:

Anode: Ti₃AlC₂(s) +3BF₄⁻(aq.) → Ti₃C₂(s) + Al³⁺(aq.) + 3F⁻(aq.) + BF₃(g) + e⁻

Cathode: H⁺(aq.) + e⁻ → ½H₂(g)

Overall: Ti₃AlC₂(s) + 3HBF₄(aq.) → Ti₃C₂(s) +3BF₃(g) + AlF₃(s) + 3/2H₂(g)

Fig. 1 (a) Schematic illustration of the setup. (b) XRD of EE-Ti₃C₂ etched with an applied voltage of +1 V at 35 °C with different etching times. (c) XRD pattern of CE-Ti₃C₂ at 35 °C with different etching times. (d) Amperometric monitoring of the electrochemical etching of EE-Ti₃C₂ at various temperatures. (e) Cyclic voltammetry curve of 1 g of Ti₃AlC₂ at various temperatures. Scan rate: 10 mV s⁻¹. (f) Etching time required to completely etch Ti₃AlC₂ to Ti₃C₂ by electrochemical etching and electrochemical etching at various temperatures.

The anodic polarisation of the MAX phase triggers the dissolution of aluminium accompanied by the release of BF₃ gas, whilst a proton is reduced to hydrogen gas on the surface of the cathode. Note that +1 V was applied to all electrochemical etching to minimize the oxidation of the MXene. The initial concentration of HBF₄ in both electrochemical and chemical etching was 24%.

2.2 XRD analysis

Fig. 1b gives the XRD pattern of the electrochemically etched Ti₃C₂ and shows the change of the crystal lattice upon etching the MAX phases to give MXene. Upon electrochemically etching with HBF₄ for 28 hours at 35 °C, the (002) peak of the Ti₃AlC₂ precursor at 9.52° was shifted to 8.63°. According to Bragg's law and the reciprocal-space metric tensor equation of hexagonal crystal lattices, this corresponds to a change of d-spacing from 9.31 Å to 10.26 Å, equivalent to a change of c-LP from 18.62 Å to 20.52 Å. The increased interlayer spacing was also accompanied by an increase of FWHM from 0.10 to approximately 0.6, which is attributed to the decrease of crystallinity upon the removal of aluminium which allows the introduction of surface termination and water intercalant. Peaks observed between 30 and 50° correspond to carbide and intermetallic impurities derived from the MAX phase precursor. After electrochemical etching, these peaks are no longer observable, indicating the effective removal of these impurities. Instead, a broad peak was observed, corresponding to the higher-order peaks of the multilayer MXene.⁴⁷ The XRD showing similar behavior for EE-Ti₃CN can be found in Fig. S1.† Whilst electrochemical etching at 35 °C only required 28 hours for the 100% conversion from MAX phases to MXene, the time required for the complete etching with HBF₄ in the absence of voltage at 35 °C was 60 hours (Fig. 1c). Note that for some MXene samples, two (002) peaks were observed due to the different amounts of water and ions intercalated between the MXene layers.⁴⁸

Fig. 1d shows the amperometry of electrochemical etching carried out at 35 °C, 45 °C, and 50 °C, respectively. In general, a correlation between the aluminium content in the MAX phases with the etching current can be seen. Initially, a high current is always observed; this is due to the anodic dissolution of aluminium oxide impurities mixed with the MAX phase precursor. After that, the current becomes stable and gradually decreases as the etching time increases. This is because of the decrease of aluminium content in the MAX phases as etching time increases. Furthermore, the drop in etching current may be attributed to the fact that the anodic dissolution of the aluminium closer to the surface occurs at a faster rate than the aluminium in the core of the crystal. When the etching is completed, the current will soon reach a plateau, as all the aluminium in the MAX phase is consumed by the anodic dissolution. At this stage, the current flow detected is mainly contributed to the introduction of surface terminations, as a higher amount of O and F content is detected for over-etched MXene (further discussed in the XPS section). The presence of this current plateau is also an indication that both Ti₃C₂ and Ti₃CN are stable and have a slow oxidation rate under the applied conditions. The time required for the electrochemical etching at various temperatures to reach the stable plateaus coincided with the time required to completely diminish the MAX precursor peaks at that temperature as per the XRD analysis, making it particularly useful for monitoring the etching process. Furthermore, it is worth mentioning that when no MAX phase is present in the graphite crucible, no current flow is observed. This further confirmed the proposed mechanism that the redox reaction occurs at the MAX phase surface, resolving the debate that the electrochemical etching is triggered by the electrochemical decomposition of HBF₄ to HF.

Upon integration of the current flow throughout the electrochemical etching process, each Ti₃AlC₂ crystal unit consumes 4.5 electrons (4.7 electrons for Ti₃AlCN) when it is etched for 28 h at an etching temperature of 35 °C (calculation in the ESI†). The value is expected to be slightly overestimated as part of the current flow may originate from the water electrolysis on the surface of the incompletely etched MXene. Yet, it concurs with the proposed mechanism that each aluminium requires 3 electrons for anodic dissolution, with the remaining 1.5 electrons contributing to the oxidation of the Ti₃C₂ surface.

2.3 Effect of voltage

Although the presence of an applied voltage facilitates the etching, controlling the magnitude is critical. Cyclic voltammetry was carried out to study the effect of voltage on the etching efficiency (Fig. 1e). At 35 °C, the rise of current occurs around +0.8 V, indicating that this is the minimum voltage for triggering the electrochemical etching at that temperature. The CV curve mimics a typical capacitor-like shape due to the electrical double layer formed on the surface of the semi-etched MAX phases. It is worth noting that the minimal voltage (V_min) required to trigger the electrochemical etching decreases as temperature increases (V_min at 45 °C = 0.7 V, V_min at 50 °C = 0.65 V). This coincides with the Nernst equation that reduction potential decreases as temperature increases. The fact that no current was detected in the control experiment when no MAX phase was present indicates that the flow of electrons is attributed to the redox reaction of the MAX phases with the electrolyte. Furthermore, it showed that HBF₄ does not undergo other side reactions under the etching voltage. Nevertheless, the increase of voltage beyond +1 V would, in fact, decrease the etching efficiency, despite an increase of current being observed. This is because when the voltage applied exceeds +1 V, the current detected is no longer solely associated with the anodic dissolution of aluminium of the MAX phases but also from the water electrolysis. Also, vigorous gas evolution may disturb the MAX phases from settling at the bottom of the crucible and thus terminating the etching. Therefore, despite water splitting playing an important role in facilitating the electrochemical exfoliation of other 2D materials, such as graphene and MoS₂, the vigorous evolution of gas on the surface of MAX phases only causes an isolation of the MAX phase surface from the electrode and the electrolyte.

2.4 Effect of temperature on etching

The time taken to remove the original (002) peak by electrochemical etching (t_EE) and chemical etching (t_CE) with HBF₄ at temperatures between 35 and 50 °C was measured using the diminishing intensity of the (002) of the MAX phase (Fig. 1f). It was found that both t_EE and t_CE decreased with increasing temperature. Yet, the effect of temperature was approximately twice as high for chemical etching time than for electrochemical etching. For instance, at 35 °C, t_EE and t_CE were observed to be 28 h and 60 h, respectively. By increasing the etching temperature to 50 °C, t_EE and t_CE reduce to 14 h and 30 h, respectively. Note that the etching time for chemical etching with HBF₄ at 60 °C reported by Gentile et al. is only present as a reference and was not counted in the fitting.²⁷ This difference in temperature dependence may be explained by the different etching mechanisms; the chemical etching with HBF₄ relies solely on the decomposition of HBF₄ to generate HF and the thermal energy to overcome the activation barrier in order to achieve the etching, whereas for the electrochemical etching with HBF₄, the presence of voltage provides an additional driving force by increasing the Fermi potential of the aluminium in the MAX phase precursor, allowing the anodic dissolution of aluminium to occur without the generation of HF. Compared to solely chemical etching with HBF₄, the extra reaction pathway provided by the electrochemical etching is concluded to be the main reason for the increased efficiency of etching at low temperatures. In addition, the electrolyte residue after electrochemical etching was dried by heating in a PTFE beaker. A white precipitate was formed after all the HBF₄ residue was evaporated. The FTIR spectrum of the white precipitate corresponded to aluminium oxide, further confirming the anodic dissolution of aluminium during the etching (Fig. S2†).

Fig. 2a shows the Raman spectra of the EE-Ti₃C₂ and EE-Ti₃CN and their MAX phase precursors. For EE-Ti₃C₂ that has undergone completed etching, peaks at 202 and 661 cm⁻¹ originating from the A_1g (Ti, Al, C, O, F) and C (A_1g) modes of Ti₃AlC₂ MAX phases are no longer observable. Conversely, new peaks were detected at 210 and 727 cm⁻¹ corresponding to the A_1g (Ti, C, O, F) and C (A_1g) modes of EE-Ti₃C₂. A broad peak was detected from 230 to 470 and 580 to 730 cm⁻¹, the literature attributes the region of this peak to the in-plane vibration of the terminal group on the MXene surface.¹³ The peak at around 123 cm⁻¹ is the resonant peak which is only detected when an excitation wavelength of 785 nm is used. For EE-Ti₃CN, the A_1g (C, N) peak appeared at 714 cm⁻¹. However, the A_1g (Ti, C, N, O, F) and A_1g (Ti, C, Al, N) peaks were not observed. This might be attributed to the peak broadening induced by the random arrangement of carbon and nitrogen atoms in the X plane. No peak was observed in the carbon region, suggesting that the EE-MXene is free from carbide-derived carbon.

Fig. 2 (a) Raman spectrum of the EE-Ti₃C₂, EE-Ti₃CN, Ti₃AlC₂, and Ti₃AlCN. Excitation wavelength: 785 nm. (b) XPS survey spectrum of EE-Ti₃C₂ and EE-Ti₃CN. (c) XPS Ti 2p spectrum of EE-Ti₃C₂. (d) XPS C 1s spectrum of EE-Ti₃C₂.

X-ray photoelectron spectroscopy was performed to probe the composition and the oxidation state of the element. The wide scan of the EE-MXene can be found in Fig. 2b. High-resolution core-level scans were performed to get an insight into the oxidation state of each element. Upon successful etching, no Al_2p signal was detected for the EE-MXene. In terms of the termination group introduced during the etching, both fluorine and oxygen were detected. No boron was detected in all the etched samples, indicating that no Ti–O–BF₃ terminal groups were introduced. EE-MXene has a higher binding energy for the Ti 2p signal compared to its MAX phase precursor, as the aluminium in the MAX phases is substituted by O and F atoms, which are more electronegative. For the EE-Ti₃C₂, four pairs of doublets were assigned, corresponding to the titanium metal in the oxidation state of +1 to +3, and X–Ti–F of EE-Ti₃C₂, located at 454.6 (460.8), 455.7 (461.2), 457 (462.7), and 460.4 (466.4) (Fig. 2c). The remaining doublet located at 459.4 (465.4) was assigned to titanium oxide. For the C_1s spectrum (Fig. 2d), 4 singlets were assigned at 281.8, 284.7, 286.2, and 288.4 eV, corresponding to X–Ti, C–C, C–O, and C=O for EE-Ti₃C₂. For the N 1s signal of the EE-Ti₃CN, 4 peaks located at 396.4, 397.1, 399.5, and 402.4 eV are assigned for N–Ti, N–C, CH₃CN, and NH₄NO₃, respectively (Fig. S3†). Other core level scan results of EE-Ti₃CN can be found in Fig. S4 and S5.† A more detailed discussion of the X-ray spectroscopy of MXene can be found in the work by Halim et al.⁴⁹

Table 1 lists the chemical composition of both the electrochemically and chemically etched MXenes. For comparison, MXenes chemically etched with the conventional mixed acid (HF and HCl) method were also evaluated. In general, the three different etching conditions led to a similar degree of fluoride termination. However, HF/HCl-MXene also had chloride termination, giving a higher content of halide termination overall. Both EE-Ti₃C₂ and CE-Ti₃C₂ exhibited a higher oxygen content than HF/HCl-MXene. No significant difference in chemical composition is observed for the CE-Ti₃C₂ and EE-Ti₃C₂ materials. This indicates that the acceleration of etching upon the application of voltage did not increase the oxidation of the EE-MXenes. This matches the current plateau observed from the amperometric monitoring during the electrochemical etching. Nevertheless, a significant increase in both oxygen and fluorine content was observed for the over-etched EE-Ti₃C₂. This shows that despite the electrochemical etching being more selective to the aluminium than the MXene, over-etching would still oxidise the MXene layer.

Table 1 Stoichiometry of Ti₃C₂ and Ti₃CN etched by electrochemical etching of HBF₄, chemical etching of HBF₄, and chemical etching with HF and HCl

Sample	Etching conditions	Ti	C	N	O	OH	F	Cl
EE-Ti3C2	+1 V, 35 °C, 28 h	3	2		0.3	0.8	0.5
EE-Ti3C2	+1 V, 35 °C, 36 h	3	2		0.3	1.1	1
CE-Ti3C2	35 °C, 60 h	3	2		0.4	0.6	0.6
HF/HCl-Ti3C2	35 °C, 24 h	3	2		0.2	0.3	0.6	0.2
EE-Ti3CN	+1 V, 35 °C, 28 h	3	1.2	0.8	0.4	0.8	0.6
HF/HCl-Ti3CN	35 °C, 24 h	3	1.2	0.8	0.3	0.5	0.7	0.3

---

2.5 Morphology

Upon etching, the color of the EE-Ti₃C₂ multilayer powder changed from grey to dark grey for both EE-Ti₃C₂ and EE-Ti₃CN (Fig. S6†). When treated with TMAOH, the EE-Ti₃C₂ water dispersion appeared as a dark green solution. Fig. 3a shows the SEM images of EE-Ti₃C₂. Whilst resembling the laminated structure of the MAX phase precursor, EE-MXenes displayed an accordion-like structure due to the removal of the aluminium and the introduction of surface termination as well as water intercalation. Fig. 3b and c show the AFM images of delaminated EE-Ti₃C₂ and the cross-sections of the flakes. The delaminated EE-Ti₃C₂ flake exhibited a smooth surface, and most of the flakes were observed to have a thickness of 3, 6, and 8 nm. Fig. 3d and e display the optical images of EE-Ti₃C₂ and its lateral size distribution of delaminated EE-Ti₃C₂ and HF/HCl-Ti₃C₂. For the delaminated HF/HCl-Ti₃C₂, most of the flakes had a lateral size of 2–3 μm. For EE-Ti₃C₂, flakes with lateral sizes as large as 9 μm are observable, indicating that the MXene basal plane was preserved during the etching. Nevertheless, it is worth noting that a significant reduction in the lateral size of the delaminated EE-Ti₃C₂ flake is observed for EE-MXenes, which are over-etched. The decrease of plane-to-edge ratio upon over-etching matched the increase of oxygen and fluoride content obtained from the XPS chemical composition. The ultrathin structure of the delaminated EE-Ti₃C₂ flake can also be observed in TEM (Fig. 3f), with a SAED pattern showing diffraction spots corresponding to the 110 (1.49 Å) and 100 (2.56 Å) planes of the hexagonal crystal lattice.⁵⁰ SEM and TEM images of EE-Ti₃CN can be found in Fig. S8 and S9.†

Fig. 3 (a) SEM image of EE-Ti₃C₂ multilayer powder. (b and c) AFM images and the cross-section of the delaminated EE-Ti₃C₂ flake. (d and e) Optical images of delaminated EE-Ti₃C₂. (d and e) Lateral size distribution of delaminated EE-Ti₃C₂ and HF/HCl-Ti₃C₂ flake. Overlapped flakes were excluded to avoid incorrect estimations. (f) TEM image of delaminated EE-Ti₃C₂ flake.

2.6 Li-ion supercapacitor

To study the intrinsic electrochemical properties of the EE-MXene, the energy storage performance of the EE-MXene was demonstrated and compared with that of the HF/HCl-MXene. The cyclic voltammetry curves of the half cell with EE-Ti₃C₂ and EE-Ti₃CN with lithium foil as a counter electrode are displayed in Fig. 4a. A reversible anodic peak at around +1 V was observed for all types of MXenes which is attributed to the Li ion insertion/extraction into/from the MXene layers. The peak current, I_max, of CV curves for both EE-Ti₃C₂ and EE-Ti₃CN increases as scan rates increase, with a relationship obeying the power law

I = aν^b

with b, in the plot of log(i_max) vs. log(v), being close to 1. This indicates that the charge storage mechanism is dominated by the capacitive effect.⁵¹Fig. 4c displays the galvanostatic charge/discharge profile of EE-Ti₃C₂ and EE-Ti₃CN at a current density of 0.1 A g⁻¹. The charge and discharge capacity for the first cycle was observed to have a coulombic efficiency of approximately 50%; this is attributed to the formation of a solid electrolyte interface upon the first charge and discharge cycle. Yet, the coulombic efficiency increases rapidly to 99.9% and stabilised after a few cycles, with a specific capacity of 175 mA h g⁻¹ for EE-Ti₃C₂, 195 mA h g⁻¹ for EE-Ti₃CN, 190 mA h g⁻¹ for HF/HCl-Ti₃C₂ and 210 mA h g⁻¹ for HF/HCl-Ti₃CN respectively. A tilted slope was observed during the charge and discharge process, indicating pseudocapacitance behavior. Fig. 4d and e show the cycle performance of all MXenes at a rate of 0.1 and 1 A g⁻¹, respectively. Before the charge/discharge at 1 A g⁻¹, pre-cycling was done at 0.1 A g⁻¹ for 20 cycles as an activation process. For all MXenes, no capacity losses were observed at 0.1 A g⁻¹ after 250 cycles at 0.1 A g⁻¹. In fact, an increase in capacity was observed and became more significant at a high current density. In general, all MXenes only reach their maximum capacity around a few decades of cycles. At 1 A g⁻¹, the peak capacity was observed as follows: EE-Ti₃C₂ = 140 mA h g⁻¹, EE-Ti₃CN = 165 mA h g⁻¹, HF/HCl-Ti₃C₂ = 142 mA h g⁻¹, HF/HCl-Ti₃CN = 166 mA h g⁻¹. All MXenes exhibit superior cyclic performance, with a capacity retention of approximately 80% after 2000 cycles. The trend of increase in the capacity of MXene during cycling may be attributed to that of the irreversible expansion of the interlayer spacing caused by lithiation. Thus, more lithium can be accommodated between the MXene layers. This trend was also observed for other MXene energy storage systems.^31,52–54

Fig. 4 (a) CV curves of EE-Ti₃C₂ and EE-Ti₃CN from 2.5 V to 0.05 V vs. Li/Li⁺, obtained at a scan rate of 0.5 mV s⁻¹. (b) CV curves of EE-Ti₃C₂ at various scan rates. (c) Galvanostatic charge–discharge profiles of EE-Ti₃C₂ and EE-Ti₃CN at a current density of 0.1 A g⁻¹. (d) Cycle performance of EE-Ti₃C₂, EE-Ti₃CN, HF/HCl-Ti₃C₂, and HF/HCl-Ti₃CN at a current density of 0.1 and (e) 1 A g⁻¹. (f) Rate capability test of EE-Ti₃C₂ and EE-Ti₃CN and their HF/HCl etched analogs at 0.1, 0.2, 0.5, and 1 A g⁻¹.

Both, EE-Ti₃C₂ and EE-Ti₃CN, showed slightly lower capacity than their HF/HCl etched analogues. Such difference is attributed to the small amount of chloride termination in the HF/HCl-MXene, which is bulkier than the hydroxide and fluoride functional groups, creating wider spaces for the accumulation of lithium ions.⁵⁵ Furthermore, the differences in flake size are important; smaller flakes should increase the ionic conductivity (more pathways) while reducing the electronic conductivity (higher ohmic resistance). The delaminated EE-MXene flakes showed a slightly larger lateral size in contrast to HF/HCl-MXene, making the latter more accessible to lithium deposition.⁵⁶ All MXenes exhibited good rate capability; the capacity decreased by approximately 60% when the rate increased from 0.1 to 1 A g⁻¹. The capacity was then recovered when the rate decreased back to 0.1 A g⁻¹ without any loss of capacity, indicating good rate reversibility (Fig. 4d). The difference in capacity of all MXenes became smaller at high current capacity. This is because the current density is so high that the redox-active components were not able to undergo complete lithiation and delithiation. Therefore, at high current density, the redox contribution becomes less dominant and the electrode exhibits EDL-like behaviour, which is more dependent on the specific surface area.

3. Conclusions

This work has demonstrated that applying voltage can significantly increase the efficiency of the etching of MAX phases to produce MXene sheets with a large lateral size. The applied voltage allows the aluminium in both the Ti₃AlC₂ and Ti₃AlCN MAX phases precursor to undergo anodic dissolution. This extra reaction pathway significantly reduces the etching time and the etching temperature required for etching in HBF₄, and reduces the release of HF during the etching. Temperature accelerated both chemical and electrochemical etching but also may lead to over-etching and nanosheet damage. Compared to the conventional chemical approaches which use HF-based etchants, the electrochemical etching with HBF₄ is milder, faster, and safer to handle, with a high yield of 85%. Furthermore, unlike conventional HF approaches, in which temperature and concentration of the etchant are the only variables that control the reaction kinetics, voltage can provide further driving force to the reaction. This process may eventually be used to fabricate MXenes that cannot be etched by using HF. The simple experimental setup and mild reaction conditions of the electrochemical etching allow scale up for large-volume manufacturing. As the result of reduced etching time and HF exposure, the EE-MXene exhibits a bigger aspect ratio and less halide termination compared to HF-etched MXene. For energy storage applications, EE-MXene demonstrated the pseudocapacitive nature of energy storage in lithium-ion supercapacitors and exhibited superior cyclability and rate performance.

4. Experimental section

4.1 Electrochemical etching of MAX phases

MAX phase powder (1 g, 500 mesh, Nanoplexus) and HBF₄ (250 mL, 24% in H₂O) were added to a graphite crucible. A platinum wire was used as the counter electrode. The suspension was rested for 30 minutes until the electrolytes reached the desired temperature and all the MAX phase powders settled at the bottom of the crucible. +1 V was applied for 28 hours at 35 °C. After electrochemical etching, the MXene was washed with deionized water via vacuum filtration. The MXene was dried at 80 °C under vacuum for 15 hours. The dried EE-MXene was stored at 3 °C until further characterization application. The average yield of EE-Ti₃C₂ and EE-Ti₃CN obtained was between 85% and 75%.

4.2 Chemical etching of MAX phases with HBF₄

The chemical etching with HBF₄ was carried out similar to the electrochemical etching, but without applying voltage.

4.3 Chemical etching of Ti₃AlC₂ with HF and HCl

The chemical etching of Ti₃AlC₂ with HF and HCl was carried out as reported elsewhere.⁵⁷ In brief, 1 g of Ti₃AlC₂ was slowly added to the chemical etchant and stirred at 300 rpm for 24 hours at 35 °C. The chemical etchant was a mixture of HF (48%, 2 mL) and HCl (37%, 12 mL), diluted with deionized water (volumetric ratio of HCl : H₂O : HF = 6 : 3 : 1). The resulting suspension was washed with deionized water several times via centrifugation until the pH of the suspension became neutral. The solid was collected via vacuum filtration and stored at 3 °C until further analysis.

4.4 Delamination

100 mg of EE-MXene was added to 10 mL of TMAOH (0.15 M) and stirred for 15 hours at 35 °C. The dispersion was washed with DI water until the pH of the dispersion became neutral. The suspension was sonicated for 1 hour to facilitate further delamination. The suspension was submerged in an ice bath with N₂ flow during the sonication to suppress the rate of oxidation. The dispersion was allowed to settle overnight at 3 °C, the supernatant was collected for further analysis.

4.5 Materials characterisation

Raman spectroscopy was carried out to study the vibrational mode of the MAX phase and MXene using an inVia confocal Raman microscope (Renishaw, Wotton-under-Edge, Gloucestershire, UK). X-ray diffraction patterns were obtained using a D8 autochanger (Bruker, Brighton, Sussex, UK) using Cu K_α radiation with a step size of 0.02° in the range of 3–90°. Atomic force microscopy was carried out to measure the thickness of the delaminated MXene flakes using a Nanowizard 4 XP Nanoscience atomic force microscope (Bruker, Brighton, Sussex, UK). SEM and TEM images were obtained to study the morphology using a Tescan Mira scanning electron microscope (Tescan, Kohoutovice, Czech Republic) and Talos transmission electron microscope (Thermofisher, Loughborough, UK), respectively. X-ray photoelectron microscopy was carried out to obtain the elemental composition and oxidation state of the MAX phase and MXene using an ESCA2SR spectrometer (ScientaOmicron GmbH) using monochromated Al K_α radiation (1486.6 eV, 20 mA emission at 300 W, 1 mm spot size) with a base vacuum pressure of ∼1 × 10⁻⁹ mbar. Charge neutralisation was achieved using a low-energy electron flood source (FS40A, PreVac). Binding energy scale calibration was performed by defaulting the C–C in the C 1s photoelectron peak to be 284.7 eV. Analysis and curve fitting were performed using Voigt-approximation peaks using CasaXPS.

4.6 Electrochemical testing

The EE-MXene electrodes were prepared by mixing the MXene multilayer powders with super P carbon black, and polyvinylidene fluoride (PVDF) binder in N-methyl-2-pyrrolidinone (NMP) with a gravimetric ratio of 7 : 2 : 1 and the slurry was pasted on copper foil with a thickness of 100 μm and dried at 80 °C under vacuum overnight. A microporous membrane (Celgard 2400) was used as the separator. Cell assembling was carried out in an argon-filled glove box with the concentration of moisture and oxygen below 1 ppm. The electrolyte was 1 M LiPF₆ in ethylene carbonate and dimethyl carbonate in a 1 : 1 volumetric ratio. Galvanostatic charge–discharge tests were carried out using a CT3001A battery tester (Landt, Vestal, New York, USA) from 0.05 V to 2.5 V. Current densities of 100, 200, 500, and 1000 mA g⁻¹ were applied. Cyclic voltammetry experiments were carried out using an IviumStat (Ivium Technologies, Eindhoven, Netherlands).

5 Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its ESI.†

6 Author contributions

K. C. C. performed the experimental work under the supervision of M. B. and I. K. at the University of Manchester and Y. G. at Drexel University, T. Z. synthesized titanium aluminium carbonitride. The analysis was performed by K. C. C. with contributions from all authors. The manuscript was drafted by K. C. C. with input from all authors.

7 Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

This work was supported by the Graphene Nownano Centre for Doctoral Training program of the University of Manchester and the Turing Scheme offered by the UK Government. We acknowledge the technical support provided by the XRD Facility and Henry Royce Institute at the University of Manchester for the XPS analysis (funded by EPSRC, grant agreement EP/P025021/1). We also acknowledge the Royal Academy of Engineering, Morgan Advanced Materials and the European Union's Horizon 2020 Research and Innovation program under grant agreement no. 881603. The work at Drexel University was supported by the U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, grant no. DE-SC0018618.

References

1. M. Naguib, et al., Two-dimensional nanocrystals produced by exfoliation of Ti₃AlC₂, Adv. Mater., 2011, 23, 4248–4253.
2. B. Anasori, M. R. Lukatskaya and Y. Gogotsi, 2D metal carbides and nitrides (MXenes) for energy storage, Nat. Rev. Mater., 2017, 2, 16098.
3. R. Ma, et al., Self-Supporting, Binder-Free, and Flexible Ti₃C₂T_x MXene-Based Supercapacitor Electrode with Improved Electrochemical Performance, ACS Nano, 2022, 16, 9713–9727.
4. S. J. Kim, et al., High mass loading, binder-free MXene anodes for high areal capacity Li-ion batteries, Electrochim. Acta, 2015, 163, 246–251.
5. Y. Xie, et al., Prediction and Characterization of MXene Nanosheet Anodes for Non-Lithium-Ion Batteries, ACS Nano, 2014, 8, 9606–9615.
6. M. Naguib, et al., MXene: a promising transition metal carbide anode for lithium-ion batteries, Electrochem. Commun., 2012, 16, 61–64.
7. D. Er, J. Li, M. Naguib, Y. Gogotsi and V. B. Shenoy, Ti₃C₂ MXene as a high capacity electrode material for metal (Li, Na, K, Ca) ion batteries, ACS Appl. Mater. Interfaces, 2014, 6, 11173–11179.
8. F. Shahzad, et al., Electromagnetic interference shielding with 2D transition metal carbides (MXenes), Science, 2016, 353, 1137–1140.
9. A. Iqbal, P. Sambyal and C. M. Koo, 2D MXenes for Electromagnetic Shielding: A Review, Adv. Funct. Mater., 2020, 30, 2000883.
10. J. Lipton, et al., Scalable, Highly Conductive, and Micropatternable MXene Films for Enhanced Electromagnetic Interference Shielding, Matter, 2020, 3, 546–557.
11. Z. W. Seh, et al., Two-Dimensional Molybdenum Carbide (MXene) as an Efficient Electrocatalyst for Hydrogen Evolution, ACS Energy Lett., 2016, 1, 589–594.
12. Y. Gogotsi, The Future of MXenes, Chem. Mater., 2023, 35, 8767–8770.
13. A. Sarycheva and Y. Gogotsi, Raman Spectroscopy Analysis of the Structure and Surface Chemistry of Ti₃C₂T_x MXene, Chem. Mater., 2020, 32, 3480–3488.
14. H. W. Wang, M. Naguib, K. Page, D. J. Wesolowski and Y. Gogotsi, Resolving the Structure of Ti₃C₂T_x MXenes through Multilevel Structural Modeling of the Atomic Pair Distribution Function, Chem. Mater., 2016, 28, 349–359.
15. M. Khazaei, et al., Insights into exfoliation possibility of MAX phases to MXenes, Phys. Chem. Chem. Phys., 2018, 20, 8579–8592.
16. J. Halim, et al., Synthesis and Characterization of 2D Molybdenum Carbide (MXene), Adv. Funct. Mater., 2016, 26, 3118–3127.
17. M. Alhabeb, et al., Selective Etching of Silicon from Ti₃SiC₂ (MAX) to Obtain 2D Titanium Carbide (MXene), Angew. Chem., Int. Ed., 2018, 57, 5444–5448.
18. M. Anayee, et al., Role of acid mixtures etching on the surface chemistry and sodium ion storage in Ti₃C₂T_x MXene, Chem. Commun., 2020, 56, 6090–6093.
19. M. Han, et al., Tailoring electronic and optical properties of MXenes through forming solid solutions, J. Am. Chem. Soc., 2020, 142, 19110–19118.
20. I. Persson, et al., Tailoring Structure, Composition, and Energy Storage Properties of MXenes from Selective Etching of In-Plane, Chemically Ordered MAX Phases, Small, 2018, 14, 1703676.
21. B. Anasori, et al., Control of electronic properties of 2D carbides (MXenes) by manipulating their transition metal layers, Nanoscale Horiz., 2016, 1, 227–234.
22. M. Naguib, et al., New two-dimensional niobium and vanadium carbides as promising materials for Li-ion batteries, J. Am. Chem. Soc., 2013, 135, 15966–15969.
23. J.-C. Lei, X. Zhang and Z. Zhou, Recent advances in MXene: preparation, properties, and applications, Front. Phys., 2015, 10, 107303.
24. M. Alhabeb, et al., Guidelines for Synthesis and Processing of Two-Dimensional Titanium Carbide (Ti₃C₂T_x MXene), Chem. Mater., 2017, 29, 7633–7644.
25. Y. J. Kim, et al., Etching Mechanism of Monoatomic Aluminum Layers during MXene Synthesis, Chem. Mater., 2021, 33, 6346–6355.
26. M. Anayee, et al., Kinetics of Ti₃AlC₂ Etching for Ti₃C₂T_x MXene Synthesis, Chem. Mater., 2022, 34, 9589–9600.
27. A. Gentile, et al., Critical Analysis of MXene Production with In-Situ HF Forming Agents for Sustainable Manufacturing, ChemElectroChem, 2022, 9, 202200891.
28. B. Dash, B. R. Das, B. C. Tripathy, I. N. Bhattacharya and S. C. Das, Acid dissolution of alumina from waste aluminium dross, Hydrometallurgy, 2008, 92, 48–53.
29. T. Li, et al., Fluorine-Free Synthesis of High-Purity Ti₃C₂T_x (T=OH, O) via Alkali Treatment, Angew. Chem., Int. Ed., 2018, 57, 6115–6119.
30. C. Peng, et al., A hydrothermal etching route to synthesis of 2D MXene (Ti₃C₂, Nb₂C): enhanced exfoliation and improved adsorption performance, Ceram. Int., 2018, 44, 18886–18893.
31. H. Dong, et al., Molten Salt Derived Nb₂CT_x MXene Anode for Li-ion Batteries, ChemElectroChem, 2021, 8, 957–962.
32. L. Liu, et al., Exfoliation and Delamination of Ti₃C₂T_x MXene Prepared via Molten Salt Etching Route, ACS Nano, 2022, 16, 111–118.
33. M. Li, et al., Element Replacement Approach by Reaction with Lewis Acidic Molten Salts to Synthesize Nanolaminated MAX Phases and MXenes, J. Am. Chem. Soc., 2019, 141, 4730–4737.
34. R. A. Vaia, et al., Halogen etch of Ti₃AlC₂ MAX phase for MXene fabrication, ACS Nano, 2021, 15, 2771–2777.
35. H. Shi, et al., Ambient-Stable Two-Dimensional Titanium Carbide (MXene) Enabled by Iodine Etching, Angew. Chem., Int. Ed., 2021, 60, 8689–8693.
36. P. Urbankowski, et al., Synthesis of two-dimensional titanium nitride Ti₄N₃ (MXene), Nanoscale, 2016, 8, 11385–11391.
37. Y. Yang, et al., Electrochemical exfoliation of graphene-like two-dimensional nanomaterials, Nanoscale, 2019, 11, 16–33.
38. A. M. Abdelkader, A. J. Cooper, R. A. W. Dryfe and I. A. Kinloch, How to get between the sheets: a review of recent works on the electrochemical exfoliation of graphene materials from bulk graphite, Nanoscale, 2015, 7, 6944–6956.
39. K. Parvez, et al., Exfoliation of Graphite into Graphene in Aqueous Solutions of Inorganic Salts, J. Am. Chem. Soc., 2014, 136, 6083–6091.
40. D. C. Marcano, et al., Improved synthesis of graphene oxide, ACS Nano, 2010, 4, 4806–4814.
41. S. Yang, et al., Fluoride-Free Synthesis of Two-Dimensional Titanium Carbide (MXene) Using A Binary Aqueous System, Angew. Chem., 2018, 130, 15717–15721.
42. S.-Y. Pang, et al., Universal Strategy for HF-Free Facile and Rapid Synthesis of Two-dimensional MXenes as Multifunctional Energy Materials, J. Am. Chem. Soc., 2019, 141, 9610–9616.
43. M. R. Lukatskaya, et al., Room-Temperature Carbide-Derived Carbon Synthesis by Electrochemical Etching of MAX Phases, Angew. Chem., 2014, 126, 4977–4980.
44. M. Song, S. Y. Pang, F. Guo, M. C. Wong and J. Hao, Fluoride-Free 2D Niobium Carbide MXenes as Stable and Biocompatible Nanoplatforms for Electrochemical Biosensors with Ultrahigh Sensitivity, Adv. Sci., 2020, 7, 2001546.
45. W. Sun, et al., Electrochemical etching of Ti₂AlC to Ti₂CT_x (MXene) in low-concentration hydrochloric acid solution, J. Mater. Chem. A, 2017, 5, 21663–21668.
46. J. Chen, et al., Simplified Synthesis of Fluoride-Free Ti₃C₂T_xvia Electrochemical Etching toward High-Performance Electrochemical Capacitors, ACS Nano, 2022, 16, 2461–2470.
47. M. Shekhirev, C. E. Shuck, A. Sarycheva and Y. Gogotsi, Characterization of MXenes at every step, from their precursors to single flakes and assembled films, Prog. Mater. Sci., 2021, 120, 100757.
48. S. Célérier, et al., Hydration of Ti₃C₂T_x MXene: An Interstratification Process with Major Implications on Physical Properties, Chem. Mater., 2019, 31, 454–461.
49. J. Halim, et al., X-ray photoelectron spectroscopy of select multi-layered transition metal carbides (MXenes), Appl. Surf. Sci., 2016, 362, 406–417.
50. I. Persson, et al., How Much Oxygen Can a MXene Surface Take Before It Breaks?, Adv. Funct. Mater., 2020, 30, 1909005.
51. V. Augustyn, et al., High-rate electrochemical energy storage through Li⁺ intercalation pseudocapacitance, Nat. Mater., 2013, 12, 518–522.
52. M. R. Lukatskaya, et al., Cation Intercalation and High Volumetric Capacitance of Two-Dimensional Titanium Carbide, Science, 2013, 342, 1502–1506.
53. O. Mashtalir, M. R. Lukatskaya, M. Q. Zhao, M. W. Barsoum and Y. Gogotsi, Amine-assisted delamination of Nb₂C MXene for Li-ion energy storage devices, Adv. Mater., 2015, 27, 3501–3506.
54. B. Zhang, J. Zhu, P. Shi, W. Wu and F. Wang, Fluoride-free synthesis and microstructure evolution of novel two-dimensional Ti₃C₂(OH)₂ nanoribbons as high-performance anode materials for lithium-ion batteries, Ceram. Int., 2019, 45, 8395–8405.
55. S. Kajiyama, et al., Enhanced Li-Ion Accessibility in MXene Titanium Carbide by Steric Chloride Termination, Adv. Energy Mater., 2017, 7, 1601873.
56. K. Maleski, C. E. Ren, M. Q. Zhao, B. Anasori and Y. Gogotsi, Size-Dependent Physical and Electrochemical Properties of Two-Dimensional MXene Flakes, ACS Appl. Mater. Interfaces, 2018, 10, 24491–24498.
57. M. Liu, et al., Deformation of and Interfacial Stress Transfer in Ti₃C₂ MXene-Polymer Composites, ACS Appl. Mater. Interfaces, 2022, 14, 10681–10690.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta03457k

---