Investigating lithium intercalation and diffusion in Nb-doped TiO2 by first principles calculations

https://doi.org/10.1016/j.jtice.2021.06.001Get rights and content

Highlights

  • Lithium intercalation and diffusion in Nb-doped TiO2 were investigated using density functional theory.

  • Three common polymorphs were considered: anatase, rutile, and TiO2(B).

  • Nb-doping enhances lithium intercalation in TiO2, but slightly increases the energy barrier for lithium diffusion.

  • The band gap narrows due to Nb-induced dopant states and the fermi level shifts towards the conduction band, improving material conductivity.

  • Nb-doped rutile shows significantly improved electronic properties compared to other studied polymorphs.

Abstract

1) Background

Recent research has shown growing interest in developing titanium dioxide (TiO2) based anode for lithium ion batteries due to its high theoretical specific capacity, safety, chemical stability, and abundance. Niobium-doped (Nb-doped) TiO2 anode was proposed for lithium ion batteries and demonstrated to improve cycling stability and cycling performance.

2) Methods

In this study, first principles calculation based on density functional theory (DFT) was used to reveal and understand the mechanism of Nb-doped TiO2 outperforming pristine TiO2 as the anode material of lithium ion batteries. The lithium intercalation energy, lithium diffusion energy barrier, electron density difference mapping, and electronic structure of Nb-doped TiO2 at dilute lithium concentration were computed and compared to those of pristine TiO2.

3) Significant Findings

For all three investigated polymorphs: anatase, rutile, and TiO2(B), Nb-doping enhances the lithium intercalation process by lowering the intercalation energy, but slightly increasing the energy barrier of lithium diffusion due to stronger interaction between the intercalated lithium and polaron induced by Nb dopant in TiO2. From the analysis of electronic structure, new energy states are formed, induced by doping with Nb. These new states narrow the band gap and shift Fermi levels towards the conduction band, thus facilitating improvement of electronic conductivity for all three phases studied.

Introduction

Rechargeable batteries are indispensable in modern society and lithium ion batteries have been attracted a great deal of attention due to outstanding properties of lithium metal, such as its strong electropositive nature and light weight. In the 1970s, Whittingham proposed a pioneering battery system using lithium and TiS2 as the electrodes and LiClO4 dissolved in a mixture of tetrahydrofuran and dimethoxyethane as electrolyte solution [1]. However, even though using lithium anode has great advantages of high energy density and rate capability, dendritic lithium grows on the interface between lithium anode and liquid electrolyte, which may induce internal short circuits and lead to battery explosion [2]. To overcome such a safety issue of lithium anode, several ideas have been tested and the concept of Li-ion cells was developed [3]. In the 1990s, Sony Corporation successfully invented the commercial lithium ion cell using graphite and LiCoO2 as the electrodes, which featured higher working potential and better energy density with safer electrode materials [4]. Lithium ion batteries have become a promising choice for its higher energy density, stability, better specific power, and longer cycle life, especially when compared to the conventional batteries such as the nickel-metal hydride battery [5].

In past decades, researchers have devoted their efforts to search for suitable materials as the anode of lithium ion battery, including carbon-based, titanium-based, alloy/de-alloy, and conversion materials [6], [7], [8]. Titanium dioxide (TiO2) is considered a potential candidate for anode material due to its chemical stability, low cost, non-toxicity, and abundance. Furthermore, TiO2 possesses an outstanding theoretical specific capacity (335 mAhg−1), which is almost twice to that of Li4Ti5O12 (175 mAhg−1) and comparable to that of graphite (372 mAhg−1), and is safer when compared to the two aforementioned materials [9]. Among various TiO2 polymorphs, anatase, rutile, and TiO2(B) have been extensively investigated through experimental and computational studies for their applicability as the anode material of lithium ion battery [9], [10], [11]. Rutile is the most common phase of TiO2 in nature and the most thermodynamically stable form among other phases. However, it has the largest density and is the most unfavorable for lithium intercalation among these three polymorphs [12]. In the case of anatase, experimental studies confirmed that lithium is able to intercalate into octahedral interspaces and lithium intercalated anatase undergoes phase transformation [10] and phase separation [13]. TiO2(B) has a unique structure which provides channels for lithium diffusion along a-, b-, and c-axis with lower energy diffusion barriers and this makes it a promising material among other TiO2 polymorphs [9].

TiO2 has several advantages as the anode material, but there still exists a few shortcomings that need to be overcome, such as a high band gap (3–3.4 eV) resulting in low electronic conductivity and small diffusion coefficient causing the poor performance of lithium diffusion [14]. Several methods have been proposed to improve these issues, such as morphology control [9, 15, 16], size reduction [11, 17, 18], hydrogenation process [19, 20], and doping/coating [21], [22], [23], [24], [25], [26], [27]. Morphology control or size reduction increases the reaction surface area and decreases the lithium diffusion distance [15, 17]; while hydrogenation process generates oxygen vacancies to improve the electronic conductivity [12]. Modifying TiO2 via doping cations or anions is also aimed to enhance the electrochemical properties of TiO2. It is found that coalloying with anionic elements (e.g. nitrogen) narrowed the band gap via new valence states produced by the dopant in the band gap and overlapped with those of TiO2 [28]. On the other hand, doping with cationic elements (e.g. niobium) enhanced the charge transfer by introducing additional charge carriers and therefore improving low conductivity of TiO2 [29, 30]. Doping with cationic elements also narrowed the band gap by addition of donor levels and tends to form localized midgap states [31]. Usui et al. [32] reported that the performance of lithium ion batteries with niobium-doped (Nb-doped) rutile as anode material showed ~30% improvement in cycling performance and rate capability compared to Nb-doped anatase. Furthermore, both Nb-doped polymorphs provided better cycling performance and rate capability at various current densities than their corresponding pristine TiO2, but the detailed mechanism is still unclear. Such advantages and mysteries of Nb-doped TiO2 are of interest and worthy of further elucidation of the mechanism for improved performance as anode material.

First principles calculation is commonly used to understand the capability and reveal the mechanism of a material as electrodes of lithium ion batteries from the perspective of atoms and electrons. The lithium intercalation energy and lithium diffusion barrier from nudged elastic band (NEB) for anatase were calculated using GGA+U functional and the results were in good agreement with those from experiments [33]. The lithium diffusion pathways in rutile were investigated using GGA functional and computational results showed that the diffusion pathway along c-direction had the lowest lithium diffusion energy barrier than other pathways [34]. For TiO2(B), the computational results using GGA functional showed that there are three possible lithium intercalation sites (denoted as A1, A2 and C) and the diffusion pathway from a C site to its nearest neighboring C site had the lowest energy barrier for lithium diffusion [35]. Dawson and Robertson compared lithium intercalation properties in the three aforementioned TiO2 polymorphs using sX hybrid functional and found that lithium intercalation is most energetically preferred in anatase [36]. However, only a few computational studies investigated Nb-doped TiO2 and most of them focused on the electronic properties of a certain Nb-doped TiO2 polymorph [37] or applications of Nb-doped TiO2 as photocatalyst [38] or in dye-sensitized solar cell (DSSC) [39].

In this study, we aimed to systematically investigate the electronic properties of Nb-doped anatase, rutile, and TiO2(B) using GGA+U functional, as well as the lithium intercalation energy and lithium diffusion energy barrier in these three Nb-doped TiO2 polymorphs. The calculation results of pristine and Nb-doped TiO2 polymorphs were compared and discussed in order to understand the reasons or mechanism for the improved performance of Nb-doped TiO2 as the anode material of the lithium ion battery.

Section snippets

Methods

In this study, CASTEP [40] was employed for all calculations using spin-polarized method within the generalized gradient approximation (GGA) with Perdew-Burke-Ernzerhof (PBE) density functional [41]. Ultrasoft pseudopotentials from standard database implemented in the Materials Studio package were used. The value of plane-wave cutoff energy (620 eV) for TiO2 from our previous work [12] was confirmed to be suitable for all three TiO2 polymorphs with niobium (Nb) doping through convergence

Lattice parameters of Nb-doped TiO2

The unit cell lattice parameters of three TiO2 polymorphs with Nb doping from calculation (full geometry optimization) and experiment are summarized in Table 1. When compared to the calculation results of pristine TiO2, the increase in lattice parameter a for anatase after Nb doping is significantly larger than that of lattice parameter c, which is consistent with the experimental observation [48]. The lattice parameter a of rutile after Nb doping slightly increases and this is identical to

Conclusions

Density functional theory calculations were performed to investigate the effect of Nb doping on the performance of TiO2 as the anode material of lithium ion batteries using the same computational methodology and scale-similar models for three TiO2 polymorphs: anatase, rutile, and TiO2(B). Detailed lithium intercalation and diffusion, electron density difference mapping, and electronic structure were utilized to reveal and elucidate the reason and mechanism of the improved performance of

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors are grateful to funds from the Ministry of Science and Technology of Taiwan (MOST 106–2221-E-008–088-MY3) and computational resources from the National Center for High-Performance Computing (NCHC).

References (72)

  • J. Huo et al.

    TiO2/carbon nanofibers doped with phosphorus as anodes for hybrid Li-ion capacitors

    J Power Sources

    (2020)
  • D.P. Opra et al.

    Doping of titania with manganese for improving cycling and rate performances in lithium-ion batteries

    Chem Phys

    (2020)
  • M.V. Koudriachova et al.

    Diffusion of Li-ions in rutile. An ab initio study

    Solid State Ion

    (2003)
  • A.J. Gardecka et al.

    Nb-doped rutile titanium dioxide nanorods for lithium-ion batteries

    Solid State Sci

    (2018)
  • T.P. Feist et al.

    The soft chemical synthesis of TiO2(B) from layered titanates

    J Solid State Chem

    (1992)
  • J.W. Pan et al.

    Electronic properties of TiO2 doped with Sc, Y, La, Zr, Hf, V, Nb and Ta

    Chem Phys Lett

    (2015)
  • L. Brohan et al.

    La transformation TiO2(B) → anatase

    Mater Res Bull

    (1982)
  • B. Zachau-Christiansen et al.

    Lithium insertion in different TiO2 modifications

    Solid State Ion

    (1988)
  • A. Ahmadi et al.

    Theoretical study of aluminum nitride nanotubes for chemical sensing of formaldehyde

    Sens Actuators B Chem

    (2012)
  • M.S. Whittingham

    Electrical energy storage and intercalation chemistry

    Science

    (1976)
  • E. Peled

    The electrochemical behavior of alkali and alkaline earth metals in nonaqueous battery systems—The solid electrolyte interphase model

    J Electrochem Soc

    (1979)
  • T. Nagaura et al.

    Lithium ion rechargeable battery

    Progress in Batteries & Solar Cells

    (1990)
  • G.-.A. Nazri et al.

    Lithium batteries: science and technology

    (2003)
  • Y. Ren et al.

    Nanoparticulate TiO2(B): an anode for lithium-ion batteries

    Angew Chem Int Ed

    (2012)
  • R. van de Krol

    In situ X-ray diffraction of lithium intercalation in nanostructured and thin film anatase TiO2

    J Electrochem Soc

    (1999)
  • Y.S. Hu et al.

    High lithium electroactivity of nanometer-sized rutile TiO2

    Adv Mater

    (2006)
  • H.-.L. Yeh et al.

    First-principles study of lithium intercalation and diffusion in oxygen-defective titanium dioxide

    J Phys Chem C

    (2018)
  • M. Wagemaker et al.

    Equilibrium lithium transport between nanocrystalline phases in intercalated TiO2 anatase

    Nature

    (2002)
  • G. Armstrong et al.

    TiO2(B) nanotubes as negative electrodes for rechargeable lithium batteries

    Electrochem Solid-State Lett

    (2006)
  • M. Wagemaker et al.

    The influence of size on phase morphology and Li-ion mobility in nanosized lithiated anatase TiO2

    Chem Eur J

    (2007)
  • X. Lu et al.

    Hydrogenated TiO2 nanotube arrays for supercapacitors

    Nano Lett

    (2012)
  • J. Qiu et al.

    Hydrogenation synthesis of blue TiO2 for high-performance lithium-ion batteries

    J Phys Chem C

    (2014)
  • Y. Wang et al.

    Niobium doped TiO2 with mesoporosity and its application for lithium insertion

    Chem Mater

    (2010)
  • W. Xu et al.

    Niobium-doped titanium dioxide with high dopant contents for enhanced lithium-ion storage

    ChemElectroChem

    (2020)
  • R. Asahi et al.

    Visible-light photocatalysis in nitrogen-doped titanium oxides

    Science

    (2001)
  • M. Fehse et al.

    Nb-doped TiO2 nanofibers for lithium ion batteries

    J Phys Chem C

    (2013)
  • Cited by (12)

    • Nb-doped TiO<inf>2</inf> mesoporous films with improved electrochemical and electrochromic energy storage performance

      2022, Optical Materials
      Citation Excerpt :

      The lattice spacing of 6-NTO sample is 0.367 nm, also corresponding to the (101) plane of tetragonal structured anatase TiO2. It can be seen that the lattice spacing of 6-NTO increases, which mainly due to the large ionic radius of Nb5+ intercalating TiO2 lattice and replacing Ti4+ [46]. Moreover, the distribution of the Nb and Ti ions for 6-NTO sample is studied by the EDS mapping analysis as given in Fig. 5(e).

    • Highly effective Al-doped titanium niobate porous anode material for rechargeable high-rate Li-ion storage performance

      2022, Journal of the Taiwan Institute of Chemical Engineers
      Citation Excerpt :

      Therefore, transition metal oxides have been intensively investigated for their promising applications as negative electrode materials for advanced energy storage devices [2,5]. Typically, Titanium (Ti)-based oxides are considered to be an prominent alternative to replace graphite owing to their appropriate operating potential (1.0–2.0 V vs. Li), high theoretical capacity, high structural and chemical stability [6–8]. Firstly, Goodenough et al. [9] proposed a TiNb2O7 (TNO) monoclinic structure of anode material for practical application in LIBs.

    View all citing articles on Scopus
    View full text