Engineering oxygen vacancies in Na2Ti3O7 for boosting its catalytic performance in MgH2 hydrogen storage

Rational design of high-eciency catalysts plays a critical role in improving the hydrogen storage performances of the MgH 2 . Herein, ower-like Na 2 Ti 3 O 7 catalyst with rich oxygen vacancies (Na 2 Ti 3 O 7 O v ) was synthesized from Ti 3 C 2 -MXene and demonstrated to remarkably enhance the hydrogen storage of MgH 2 . Specically, with an addition of 5 wt.% Na 2 Ti 3 O 7 -O v , the initial dehydrogenation temperature of the MgH 2 + 5Na 2 Ti 3 O 7 -O v composite reduced substantially from 287 °C (for MgH 2 ) to 183 °C. Moreover, the MgH 2 + 5Na 2 Ti 3 O 7 -O v composite exhibited fast hydrogen ab/desorption kinetics and superb reversible hydrogen storage performance with a retention rate of 90.1 % after 10 cycles attributed to the higher structural stability of Na 2 Ti 3 O 7 -O v . Both experimental and theoretical results conrm that the oxygen vacancies in Na 2 Ti 3 O 7 -O v reduce the reaction activation energy during MgH 2 dehydrogenation, hence accounting for the excellent hydrogen sorption kinetics. This work would lead to new design and development of advanced defect-based nano-catalysts for the MgH 2 hydrogen storage system.


Introduction
Hydrogen has been commonly considered as a sustainable and clean energy carrier [1][2][3] . Unfortunately, the e cient and safe storage of hydrogen with high gravimetric and volumetric capacity poses a major economical bottleneck for the emerging hydrogen economy 4,5 . Solid-state hydrogen storage technology, such as metal hydride 6,7 , metal-organic frameworks [8][9][10] , or complex hydrides 11 , has deemed to be a promising method for hydrogen storage due to its high-volume hydrogen storage capacity, safety, free of high pressure and heat insulation vessels 12 . Among them, magnesium hydride (MgH 2 ) has been widely regarded as a most promising hydrogen storage material owing to its high gravimetric hydrogen storage capacity (7.6 wt.%), excellent de/rehydrogenation reversibility, low cost, and environmentally friendly [13][14][15] . However, the sluggish kinetics of hydrogen ab/desorption of MgH 2 severely limits the practical application of MgH 2 for hydrogen storage [16][17][18] .
Catalyst doping has proven to be an e cient pathway to enhance the reaction kinetics for hydrogen storage in MgH 2 and hence reduce the operating temperatures of the MgH 2 hydrogen storage system [19][20][21] . Among all catalysts investigated, Ti-based materials are commonly recognized as the best-performing ones and have been widely used to improve the hydrogen storage performances of the MgH 2 system, via signi cantly reducing the activation energy of de/ab-hydrogenation of MgH 2 while unchanging the enthalpy and entropy in the hydride formation process 22 . In particular, titanate materials, such as NiTiO 3 23 , K 2 Ti 6 O 13 24 , TiVO 3.5 25 , etc., have attracted more attentions in MgH 2 hydrogen storage system.
These materials were commonly loaded to MgH 2 via a ball-milling process using sacri cial agents that would in situ decompose into catalytic-inert substance. Thus, the rational design of highly e cient catalysts and development of more stable Na 2 Ti 3 O 7 in the MgH 2 hydrogen storage system would play a vital role in the hydrogen storage system.
As evidently demonstrated in the literature, defect engineering is an effective strategy to expose more active sites and tune the structural regularity and band structure of nanostructured catalysts 26,27 (003) and (101) planes of Na 2 Ti 3 O 7 (Fig. 1g), respectively. X-ray dispersive analysis (EDS) con rms that Na, Ti, and O elements are distributed uniformly throughout the ower-like structure (Fig. 1h). For comparison, the Na 2 Ti 3 O 7 with different amounts of oxygen vacancies were synthesized and presented in Fig. S3 By contrast, the pure MgH 2 exhibited a sluggish de/re-hydrogenation rate even at a high temperature of 300 °C (Fig. S8).
To explore the kinetics of hydrogen absorption, the isothermal rehydrogenation of the dehydrogenated MgH 2 + 5Na 2 Ti 3 O 7 -O v were examined at different temperatures (Fig. 2d). As clearly shown in the gure,  (Fig. 2f).
Furthermore, the energy pro les of dissociation and decomposition of MgH 2 molecule on the MgH 2 + 5Na 2 Ti 3 O 7 -O v and MgH 2 + 5Na 2 Ti 3 O 7 composites were also calculated and are presented in Fig. 4c (Table S5). Additionally, the △G of MgH 2 dissociation on the Na 2 Ti 3 O 7 -O v (101) surface (0.501 eV) is also lower than that on Ti (110) surface (0.529 eV) and more detailed results are shown in Fig. S17-S20

Methods
Synthesis of Ti 3 C 2 . First, Ti 3 AlC 2 powders were synthesized according to our previous report. [20] speci cally, covered by an airtight hood accompanying in a glove box was employed to prevent contact with air moisture and oxygen. The morphologies of materials were studied on a scanning electron microscope (SEM, Zeiss Supra 50 VP) and a transmission electron microscope (TEM, JEM-2100). X-ray photoelectron spectra (XPS) were collected on a PHI quantera SXM spectrometer with a monochromatic Al Kα X-ray source (1486.60 eV). The reference C1s peak (284.8 eV) was used to calibrate the binding energies. Raman spectra were recorded on a Renishaw-in via using a 532 nm laser. Differential scanning calorimetry (DSC, STDQ600) measurements were implemented to analyze the decomposition activation energy, where the sample was heated from room temperature to 500°C in 50 mL min − 1 Ar ow at a heating rate of 5, 10, 20, and 40°C min − 1 , respectively.
De/re-hydrogenation performances. The de/re-hydrogenation behavior and temperature-programmed desorption (TPD) were measured by a Sieverts' apparatus (purchased from Zhejiang University). In a typical test, 0.1 g of the MgH 2 + 5Na 2 Ti 3 O 7 composite was loaded into a stainless-steel tube reactor and all the operations were performed in a vacuum glove box. During the TPD experiment, the sample was heated from ambient temperature to 350°C at a heating rate of 3°C min − 1 . In isothermal experiments, under constant hydrogen pressure of 2.2 MPa the samples were heated to the desired temperature and maintained at the temperature during the following tests.

Declarations
Data availability All data are available from the authors, please refer to author contributions for speci c data sets. Source data are provided as a Source Data le. Source data are provided with this paper.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.