On the origin of the significant difference in lithiation behavior between silicon and germanium

doi:10.1016/j.jpowsour.2014.04.011

Journal of Power Sources

Volume 263, 1 October 2014, Pages 252-258

https://doi.org/10.1016/j.jpowsour.2014.04.011 Get rights and content

Highlights

•
We examine and compare the lithiation behavior of Si and Ge using DFT calculations.
•
Li diffusivity is greater and less concentration-dependent in Ge as compared to Si.
•
Li diffusion is subject to Li–host interaction and host lattice rigidity/dynamics.
•
We reveal the origin of the superior rate performance of Ge-based anodes.
•
High performance anodes can be designed via fine-tuning of Si–Ge alloys.

Abstract

Silicon and germanium are both recognized as a promising anode material for high-energy lithium-ion batteries. Si is best known for its superior energy storage capacity, while Ge exhibits better rate capability and cycleability. To better understand the underlying reasons behind their lithiation behavior differences, particularly the enhanced Li transport in Ge, we examine and compare Li-host lattice interactions and dynamics using density functional theory calculations. At the onset of lithiation, an isolated Li interstitial is found to form polar covalent bonds with four nearest host atoms, while the degree of covalency is noticeably greater for Li–Si than Li–Ge bonds. The relatively stronger Li–Si interaction, along with the stiffer Si lattice tend to be responsible for the suppressed Li mobility (D_Li = 10⁻¹³ cm² s⁻¹) in c-Si, as compared to the c-Ge case (D_Li = 10⁻¹¹ cm² s⁻¹). With continued lithiation, D_Li in a-Li_xSi increases significantly from 10⁻¹² to 10⁻⁷ cm² s⁻¹ (x = 0.14–3.57); contrarily, D_Li in a-Li_xGe is around 10⁻⁷ cm² s⁻¹ and less concentration dependent. Our analysis shows that the rapid Li diffusion in a-Li_xGe is directly related to the facile atomic rearrangements of host Ge atoms even at the early stages of lithiation.

Graphical abstract

Introduction

Li-ion batteries (LIBs) have received tremendous attention as they power a wide range of applications from portable devices, electric vehicles to various renewable energy systems [1], [2], [3]. Currently, the most adopted anode material is graphite, which has good cycleability but the dendrite formation raises safety concerns and the capacity is rather limited (372 mAh g⁻¹) especially at high charge/discharge rates. Therefore, in order to satisfy the ever-increasing energy density/power capability requirements and stringent safety standards, there is an imminent need to find new electrode materials with superior lithiation properties. Among the alternatives considered, Si stands out the most because of its impressive capacity (4200 mAh g⁻¹ for Li₂₂Si₅ [4], [5]), safe thermodynamic potential and abundance. Second only to Si, Ge has a relatively high theoretical capacity of 1624 mAh g⁻¹ (Li₂₂Ge₅ [6]), a higher electrical conductivity compared to Si [7], and a superior rate capability, up to 1000 C (full lithiation in 1/1000 h) [6]. However, the understanding and development of Ge-based anodes have gained much less attention likely because of its higher price relative to Si.

Being in the same column in the periodic table, Si and Ge share many similarities, including the disadvantages of undergoing large structural changes and volume expansion upon lithiation, which can consequently lead to early capacity fading. To overcome this drawback, many ongoing studies have focused on utilizing Si and Ge of different forms, such as thin films [8], [9], [10], [11], [12], nanoparticles [13], [14], nanowires [15], [16], [17], [18], and alloys/composites with active/inactive elements [19], [20], [21], [22]. In both Si and Ge cases, nanostructuring seems to have positive impacts on enhancing the rate capability and reducing/preventing electrode pulverization, thereby improving the cycleability. In comparison to Si, Ge of comparable nano-architecture is able to withstand much faster charging rates with noticeably less crack formation [23], [24]. Furthermore, there appears to be subtle differences in their responses to electrochemical lithiation/delithiation, as demonstrated by recent in-situ characterization [24]. On the theoretical side, there have been many studies employing density functional theory (DFT) to examine Li incorporation in Si (crystalline/amorphous bulks [25], [26], [27] and nanowires [28], [29]) and a few on Ge [30], [31], [32]. Nonetheless, the fundamental understanding regarding the nature and origin of their dissimilar responses to lithiation is still limited; to the best of our knowledge, no atomistic study has been reported to investigate the likely overlooked differences between Si and Ge as anode material, especially regarding their lithiation dynamics.

In this paper, on the basis of DFT calculations, we examine how Li diffusion is affected by its interaction with the pure Si and Ge matrices, analyze the dynamic behaviors of Li as well as the host lattice atoms, and look into the impacts of Ge-alloying on anode performance. The fundamental findings explain the origin of the lithiation behavior differences between Si and Ge, particularly the significantly enhanced Li transport in Ge, and thus assist the rational design of the next-generation high performance Si- and Ge-based anodes.

Section snippets

Computational methods

The calculations reported herein were performed on the basis of density functional theory (DFT) within the generalized gradient approximation (GGA-PW91) [33], as implemented in the Vienna Ab-initio Simulation Package (VASP) [34], [35], [36]. Spin polarization of the Li–Si (Ge) system was also examined, but appears to be unimportant. The projector augmented wave (PAW) method with a planewave basis set was employed to describe the interaction between ion cores and valence electrons. The PAW

Results and discussion

We first examined how the room-temperature diffusivity of Li varies with Li content (x) in a-Li_xSi alloys using AIMD simulations [Fig. 1]. Here, a-Li_xSi alloys were considered, instead of their crystalline counterparts, because Si lithiated beyond the first charge cycle is most likely to remain in the amorphous state due to the sizable kinetic barrier for recrystallization at room temperature [8], [39]. For each alloy, three samples are averaged to calculate the mean-square displacements (MSD)

Conclusion

We examined and compared the dynamic lithiation processes in Si and Ge using DFT calculations. The variations of room-temperature Li diffusivity (D_Li) with Li content (x) in a-Li_xSi and a-Li_xGe were evaluated using DFT-based MD simulations. For Li diffusion in the crystalline matrix, D_Li is predicted around ×10⁻¹³ cm² s⁻¹ in c-Si and ×10⁻¹¹ cm² s⁻¹ in c-Ge. With increasing x, D_Li in a-Li_xSi tends to rise by orders of magnitude from ×10⁻¹² cm² s⁻¹ (x = 0.14) to ×10⁻⁷ cm² s⁻¹ (x = 3.57), whereas D

Acknowledgments

This work was partially supported by the Robert A. Welch Foundation (F-1535) and SK Innovation Co., Ltd. We would like to thank the Texas Advanced Computing Center for use of their computing resources.

References (47)

P. Limthongkul et al.
Acta Mater.
(2003)
H.J. Jung et al.
Solid State Commun.
(2003)
S. Huang et al.
J. Power Sources
(2011)
V.L. Chevrier et al.
J. Alloys Compd.
(2010)
G. Kresse et al.
Comput. Mater. Sci.
(1996)
C.J. Wen et al.
J. Solid State Chem.
(1981)
V.B. Shenoy et al.
J. Power Sources
(2010)
X.R. Wang et al.
Surf. Sci.
(2002)
Z. Cui et al.
J. Power Sources
(2012)
M. Armand et al.
Nature
(2008)

A.S. Arico et al.

Nat. Mater.

(2005)

J.-M. Tarascon

Phil. Trans. R. Soc. A

(2010)

R.A. Sharma et al.

J. Electrochem. Soc.

(1976)

B.A. Boukamp et al.

J. Electrochem. Soc.

(1981)

J. Graetz et al.

J. Electrochem. Soc.

(2004)

E.M. Conwell

Proc. Inst. Radio Eng.

(1952)

C. Yu et al.

Adv. Energy Mater.

(2012)

B. Laforge et al.

J. Electrochem. Soc.

(2008)

B. Gao et al.

Adv. Mater.

(2001)

J. Graetz et al.

Electrochem. Solid State Lett.

(2003)

C.K. Chan et al.

Nat. Nanotechnol.

(2008)

L.F. Cui et al.

Nano Lett.

(2009)

H.T. Nguyen et al.

Adv. Energy Mater.

(2011)

Cited by (46)

High-rate and low-temperature performance of germanium nanowires anode for lithium-ion batteries
2021, Journal of Electroanalytical Chemistry
Citation Excerpt :
Although germanium is inferior to silicon in terms of theoretical gravimetric specific capacity (1624 mAh g−1 vs. 3579 mAh g−1)1, their volumetric specific capacities are rather close (7366 and 8334 mAh cm−3). At the same time, germanium exhibits electrical conductivity two orders of magnitude higher than that of silicon (partially due to a smaller bandgap [3]) and shows room-temperature Li-ion diffusivity 400 times greater than that of silicon [4,5]. However, like for other high-capacity materials, a major drawback of germanium is a huge (more than 350%) volume expansion at lithiation.
The original version of a negative electrode of lithium-ion battery is prepared. To this end the arrays of Ge nanowires were electrodeposited from an aqueous solution onto a titanium substrate with drop-like indium crystallization centers. The nanowires were characterized by SEM, TEM, EDX, XPS, galvanostatic cycling, CV, and EIS. Electrochemical tests were carried out in a wide temperature range with using the propylene-carbonate-based electrolyte. The electrochemical performances of the nanowires were shown to be dependent on the current density upon electrodeposition. The nanowires with diameter 20 to 40 nm and length up to 0.5 μm are covered with a thin oxide layer, consisting of both stoichiometric and non-stoichiometric germanium oxides. The electrodes demonstrate excellent features, namely room-temperature specific capacity ca. 1300 mAh g⁻¹ at 1C and 4C rate, and about 850 mAh g⁻¹ at 24C, as well as good low-temperature performance, 255 mAh g⁻¹ at −50 °C.
Recent progress on germanium-based anodes for lithium ion batteries: Efficient lithiation strategies and mechanisms
2020, Energy Storage Materials
Citation Excerpt :
In addition to the phase transition study, efforts have been devoted towards the Li diffusion and mechanical properties of Ge anode materials. Li-host lattice interactions and dynamics were calculated by density functional theory (DFT) method (Fig. 6) [44]. DFT calculation indicates that with an increasing Li content x, the diffusion coefficient of Li ion, DLi+ in a-LixGe remains relatively constant with a value of approximately 107 cm2s−1, exhibiting no concentration dependence.
Germanium-based materials with extremely high theoretical energy capacities have gained a lot of attention recently as potential anodes for lithium ion batteries. These materials can also offer improved Li insertion/extraction kinetics and cycling performance, providing a promising candidate of anode to meet the increasing demand for batteries with higher energy and power densities. In this review, we summarized the recent research progress on germanium-based anodes, from the fundamental understanding of the lithiation process to different kinds of germanium-based anodes and their facile and lower-cost preparation methods. The impact of the morphology of these germanium-based anodes on the electrochemical performance is also outlined. This review could serve as a handy reference, as well as a starting point, for future research related to high capacity germanium-based anodes for lithium-ion batteries.
The Interaction of Ge-Co-In Nanostructures with Lithium in Wide Temperature Range: Eis Study
2024, SSRN
First-Principles Dynamics Investigation of Germanium as an Anode Material in Multivalent-Ion Batteries
2023, Nanomaterials
Research Progress on Anode Materials for Low Temperature Metal-ion Batteries
2023, Cailiao Daobao/Materials Reports
Lithiophilic Nanowire Guided Li Deposition in Li Metal Batteries
2023, Small

View all citing articles on Scopus

View full text

On the origin of the significant difference in lithiation behavior between silicon and germanium

Highlights

Abstract

Graphical abstract

Introduction

Section snippets

Computational methods

Results and discussion

Conclusion

Acknowledgments

Acta Mater.

Solid State Commun.

J. Power Sources

J. Alloys Compd.

Comput. Mater. Sci.

J. Solid State Chem.

J. Power Sources

Surf. Sci.

J. Power Sources

Nature

Nat. Mater.

Phil. Trans. R. Soc. A

J. Electrochem. Soc.

J. Electrochem. Soc.

J. Electrochem. Soc.

Proc. Inst. Radio Eng.

Adv. Energy Mater.

J. Electrochem. Soc.

Adv. Mater.

Electrochem. Solid State Lett.

Nat. Nanotechnol.

Nano Lett.

Adv. Energy Mater.