NbSe2 Nanosheets/Nanorolls Obtained via Fast and Direct Aqueous Electrochemical Exfoliation for High-Capacity Lithium Storage

Layered transition-metal dichalcogenides (LTMDs) in two-dimensional (2D) form are attractive for electrochemical energy storage, but research efforts in this realm have so far largely focused on the best-known members of such a family of materials, mainly MoS2, MoSe2, and WS2. To exploit the potential of further, currently less-studied 2D LTMDs, targeted methods for their production, preferably by cost-effective and sustainable means, as well as control over their nanomorphology, are highly desirable. Here, we report a quick and straightforward route for the preparation of 2D NbSe2 and other metallic 2D LTMDs that relies on delaminating their bulk parent solid under aqueous cathodic conditions. Unlike typical electrochemical exfoliation methods for 2D materials, which generally require an additional processing step (e.g., sonication) to complete delamination, the present electrolytic strategy yielded directly exfoliated nano-objects in a very short time (1–2 min) and with significant yields (∼16 wt %). Moreover, the dominant morphology of the exfoliated 2D NbSe2 products could be tuned between rolled-up nanosheets (nanorolls) and unfolded nanosheets, depending on the solvent where the nano-objects were dispersed (water or isopropanol). This rather unusual delamination behavior of NbSe2 was explored and concluded to occur via a redox mechanism that involves some degree of hydrolytic oxidation of the material triggered by the cathodic treatment. The delamination strategy could be extended to other metallic LTMDs, such as NbS2 and VSe2. When tested toward electrochemical lithium storage, electrodes based on the exfoliated NbSe2 products delivered very high capacity values, up to 750–800 mA h g–1 at 0.5 A g–1, where the positive effect of the nanoroll morphology, associated to increased accessibility of the lithium storage sites, was made apparent. Overall, these results are expected to expand the availability of fit-for-purpose 2D LTMDs by resorting to simple and expeditious production strategies of low environmental impact.


S1.2. Cathodic exfoliation experiments
The electrolytic delamination of NbSe2 was carried out in a two-electrode set-up under aqueous cathodic conditions, using a platinum foil piece as the counter electrode (anode).
In a typical experiment, 100 mg of bulk NbSe2 powder were compacted onto a circular piece of graphite foil (10 mm in diameter) by means of a hydraulic press (5 tons applied for 1 min). The resulting NbSe2/graphite foil electrode and the platinum foil piece were immersed in an aqueous 0.3 M KNO3 solution (25 mL) at a distance of ~2 cm from each other and connected to a DC power supply (E3633A apparatus, from Keysight Technologies) via crocodile clips. Only about one half of the NbSe2/graphite foil piece was actually immersed in the aqueous electrolyte, the emerged half being held with the crocodile clip. Almost immediately upon application of a negative voltage (-10 V) to the NbSe2/graphite foil electrode, a reddish-brown substance was seen to release from it and to get dispersed in the electrolyte. At the same time, gray particles also detached from the cathode and sedimented at the bottom of the electrolytic cell (see Movie S1 in The Supporting Information). After one minute of cathodic treatment, the bias voltage was turned off and the previously emerged half of the NbSe2/graphite foil piece was immersed S-3 in the electrolyte to treat the corresponding fraction of NbSe2 material (again, a bias voltage of -10 V applied for one minute). Finally, the electrolytic solution containing the reddish-brown dispersion was collected (the sedimented gray particles were discarded) and processed to recover this material for subsequent use. To this end, the dispersion was first sedimented either by allowing it to rest undisturbed overnight or by centrifuging it at 100 g for 10 min. Centrifugation was conducted with the electrolytic dispersion in glass vials that in turn were inserted into 50 mL centrifuge tubes. This was done to avoid direct contact of the dispersion with the polypropylene material of the centrifuge tubes, as sedimentation of the reddish-brown product caused it to strongly adhere to the latter, which prevented its recovery. Then, the sedimented material was re-suspended in pure water by a brief treatment with a vortex mixer. Following three consecutive sedimentation/re-suspension cycles, the reddish-brown product was subjected to a final sedimentation step, and the sediment was finally collected and dried under a vacuum. For the subsequent studies, this dried product could be readily dispersed in water and isopropanol via a brief treatment (1-2 min) with a vortex mixer or a bath sonicator.

S1.3. Characterization techniques
The materials were characterized by X-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM), scanning transmission electron microscopy (STEM), atomic force microscopy (AFM), energy-dispersive X-ray (EDX) spectroscopy, Raman spectroscopy and X-ray photoelectron spectroscopy (XPS). XRD patterns were recorded with a D5000 diffractometer (Siemens), using Cu K radiation, a step size of 0.015º and of 250-300 kHz were employed. To image exfoliated NbSe2 nano-objects by AFM, the delaminated material was dispersed in isopropanol, drop-cast (10-20 L) onto a freshly cleaved highly oriented pyrolytic graphite (HOPG) substrate and dried under vacuum at room temperature overnight (drying under ambient conditions led to molecularly thin islands of the alcohol on the HOPG surface, which could be largely removed under vacuum conditions). The recorded AFM images were analyzed with SPIP software (Image Metrology). Raman spectra were acquired with a Renishaw inVia Qontor instrument, working at a laser excitation wavelength of 532 nm (green line). To minimize damage to the sample, the incident laser power was set below 0.5 mW. XPS was carried out on a SPECS system equipped with a Phoibos 100 hemispherical electron energy analyzer. The spectra were recorded at a take-off angle of 90º, working at a pressure below 10 -7 Pa and using a monochromatic aluminum X-ray source operated at a voltage of 14.00 kV and a power of 175 W. The photoexcited electrons were analyzed in the constant pass energy mode, using a pass energy of 50 eV for survey spectra and 10 eV for high resolution core-level spectra. The surface charging effect was compensated by the use of an electron flood gun operated at 0.4 eV and 0.10 mA. CasaXPS software was used for data processing. Specimens for both XPS and Raman spectroscopy were prepared in the form of continuous, thin films by drop-casting NbSe2 dispersions onto stainless steel discs, which were allowed to dry at room temperature. In the case of the commercial NbSe2 powder, a pellet was prepared by means of a hydraulic press.

S1.4. Electrochemical measurements
The cathodically delaminated NbSe2 materials were tested as electrodes for lithium storage in a coin cell configuration. The working electrodes were prepared by mixing 54 wt% of delaminated NbSe2 as the active material, 16 wt% CNTs and 20 wt% Super C65 as the conductive additives, and 10 wt% PVDF as the binder. A small volume of NMP was added to the latter components and the mixture was transformed into a homogeneous slurry with the aid of a high-shear mixer. The slurry was then cast onto a 24 cm 2 sheet of copper foil and dried at 120 ºC for 3 h. Circular discs 10 mm in diameter were finally cut from the coated copper foil. The total mass loading of the working electrode (i.e., the       S-10 selenium in another less electrically conducting allotropic form, which would become positively charged upon photoemission and consequently shifted to higher binding energy. Indeed, selenium shows different allotropic forms; most of them are nonconducting while the most thermodynamically stable one is electrically conductive [3]. According to the Raman results (see main text) both conducting (crystalline, gray t-Se) and non-conducting (amorphous, red Se) are detected in the surface of the NbSe2 materials.
The fact that some Se in NbSe2 form is detected in the material processed in isopropanol from the Se 3d spectrum (Fig. S7c) but not from the Nb 3d spectrum (Fig. 2e in the main text) can be explained by the difference in the probing depths of the corresponding XPS signals (see Fig. S4). As the kinetic energy of the Se 3d XPS electrons is ~150 eV higher than that of those ejected from Nb 3d core level, the latter band is more surface-specific, and thus more suitable for the detection of surface oxides, while the NbSe2 material underneath is better detected in the Se 3d band.
S-11   10 (*) It is not clear whether the reported values are calculated with respect to the active material or to the complete electrode. In case they were given per mass of electrode, the values would have to be multiplied by a factor of 1.43 to express them as relative to the active material.

S4. Cathodic delamination of other LTMDs
S-13 S6. Additional information of the cyclability of the cathodically delaminated NbSe2 materials for lithium storage Figure S9. Cyclability of NbSe2 nanorolls (orange trace) and nanosheets (red trace) in terms of capacity at a current density of 0.5 A g -1 , including a greater number of cycles (more than 3000) than in the main text. The gravimetric capacity figures are given relative to the total mass of the NbSe2-based electrode; they would be a factor of ~1.85 larger if given relative to the mass of active material (i.e., mass of NbSe2 only).