Understanding the Lithiation of the Sn Anode for High-Performance Li

Subscriber access provided by READING UNIV

Article

Understanding the Lithiation of Sn Anode for High Performance Li-ion Batteries with Exploration of Novel LiSn Compounds at Ambient and Moderately High Pressure Raja Sen, and Priya Johari ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b11173 • Publication Date (Web): 26 Oct 2017 Downloaded from http://pubs.acs.org on October 29, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Understanding the Lithiation of Sn Anode for High Performance Li-ion Batteries with Exploration of Novel Li-Sn Compounds at Ambient and Moderately High Pressure Raja Sen and Priya Johariú Department of Physics, School of Natural Sciences, Shiv Nadar University, Greater Noida, Gautam Budhha Nagar, UP 201 314, India. E-mail: [email protected],[email protected] Abstract Volume expansion and elastic softening of Sn anode on lithiation result in mechanical degradation and pulverization of Sn, affecting the overall performance of Li-Sn batteries. It can however be overcome with the help of void-space engineering by using a Lix Sn phase as pre-lithiated anode, where an optimal value for x is desired. Currently, Li4.25 Sn is known as the most lithiated Li-Sn compound. But, recent studies have shown that at high pressure several exotic and unusual stoichiometries can be obtained, that may even survive decompression from high-to-ambient pressure with improved mechanical properties. With a belief that hydrostatic pressure may help in realizing Li-richer (x > 4.25) Li-Sn compounds as well, we performed extensive calculations using evolutionary algorithm and density functional theory to explore all stable and low energy metastable Li-Sn compositions at pressure ranging from 1 atm to 20 GPa. This not only helped us in enriching the chemistry of Li-Sn system, in

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

general, but also in improving our understanding of the reaction mechanism in Li-Sn batteries, in particular, and guiding a route to improve the performance of Li-ion batteries through synthesis of Li-rich phases. Besides the experimentally known Li-Sn compounds, our study reveals the existence of five unreported stoichiometries (Li8 Sn3 , Li3 Sn1 , Li4 Sn1 , Li5 Sn1 , and Li7 Sn1 ) and their associated crystal structures at ambient and high pressure. While Li8 Sn3 has been identified as one of the most stable Li-Sn compound in the entire pressure range (1 atm–20 GPa) with R3m symmetry, the Lirich compounds like Li3 Sn1 -P2/m, Li4 Sn1 -R¯3m, Li5 Sn1 -C2/m, and Li7 Sn1 -C2/m are predicted to be metastable at ambient pressure and found to get thermodynamically stable at high pressure. Here, the discovery of Li5 Sn1 and Li7 Sn1 opens up the possibility to integrate them as a prelithiated anode for efficiently preventing electrochemical pulverization as compared to the experimentally known highest lithiated compound –Li17 Sn4 .

Keywords Lithium-Tin Compounds, Li-ion Battery, Evolutionary Algorithm, Crystal Structure Prediction, Density Functional Theory.

1

INTRODUCTION

In search of new anode materials for high performance lithium-ion batteries (LIBs), Sn and Sn based composites have received significant attention because of their higher capacity (990 mAh g≠1 for Li4.4 Sn) and abundant availability in nature. 1,2 Although the neighbors of Sn in group IV, i.e., Si and Ge, are also well known for their extraordinary theoretical capacities (4200 mAh g≠1 for Li4.4 Si and 1600 mAh g≠1 for Li4.4 Ge), 1,2 however, studies reveal Sn also to be a good choice because of its better electrical conductivity and large interstitial space which makes Li diffusion more favorable in it as compared to Si and Ge. 2,3 Moreover, Sn

2


Page 2 of 31

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


is ductile in nature while Si and Ge are brittle, which makes the mechanical integrity of Sn better than the preceding group IV elements. 4,5 Besides having several advantages, the application of Sn as an anode is still far from commercialization because of huge irreversible capacity loss, particle fracture, and electrochemical pulverization due to drastic variation in volume during lithiation-delithiation process (≥ 257%). 2,6–8 However, it is shown that such anomalous volume expansion during lithiation develops a compressive stress (in the range of GPa) in Sn anode, 9–12 and thus, there remains a large possibility of formation/appearance of high pressure phases during cycling of intermetallic re-chargeable batteries. The formation of high pressure phases (which may metastably exist at ambient pressure) in the chargingdischarging cycle is a well known phenomenon for intermetallic materials and has already been demonstrated by Darwiche et al., 13 Villevieille et al., 14 and Obrovac et al. 15 Thus, in order to gain a deep insight into charging-discharging mechanism of a Li-Sn battery, it is important to understand the whole process at the atomistic level, by studying all possible stable and low-energy metastable Li-Sn phases at ambient pressure. Simultaneously, it is also important to find the pressure range in which these low-energy metastable phases get thermodynamically stable, to guide their synthesis. Under pressure many unusual stoichiometric which are basically impossible at ambient pressure, can easily be synthesized, that may even survive decompression from high to ambient pressure and exhibit superior mechanical properties. For example, Zeng et al. 8 have shown recently that a high pressure phase of Li15 Si4 , i.e., —-Li15 Si4 , which is stable between 7-18 GPa with space group Fdd2, can be retained under ambient conditions, with substantially improved mechanical properties as compared to its –-phase. Furthermore, pressure may also provide novel Li-rich compounds, 16 which we believe that when quenched at ambient pressure can be used as a pre-lithiated anode. This can help in improving the performance of Li-ion batteries through void space engineering. Recently, Li et al. 17 have shown through a systematic study of charge-discharge process for three thermodynamically stable phases of Lix Si (x = 4.4, 3.75, and 2.33), along with nitride3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

protected Li4.4 Si, that the use of any of the pre-lithiated Lix Si phases over the unmodified Si, enhances the overall performance of the batteries. However, amongst studied phases, superior discharging capacity can be obtained only with Li-rich Li4.4 Si. This significant achievement has been accomplished due to the fact that void space can be engineered with Lix Si, because a pre-lithiated Lix Si anode can create void space after delithiation, which can accommodate the volume expansion during the next lithiation process. Of course, an optimal value of x is always expected for this purpose. The improvement in the performance of Li-ion batteries through void space manipulation has also been demonstrated by several other experimental studies. 18–20 This again extends the importance of studying high pressure phases in order to look for some Li-rich phase that can provide possibility to introduce more void space through delithiation as compared to Li17 Sn4 . We believe that this will certainly help in preventing the electrochemical pulverization of anode in the subsequent lithiationdelithiation steps, and eventually, improving the performance of Li-Sn batteries in a better way. We, therefore, in current work aim to investigate the Li-Sn binary phase diagram at ambient and moderately high pressure (up to 20 GPa) by performing an extensive search on the Li-Sn systems using the evolutionary algorithm based approach, 21–23 in conjunction with first-principles density functional theory (DFT) and density functional perturbation theory (DFPT) based calculations. At ambient pressure, besides the experimentally reported well known Li-Sn compounds, this work reveals several new stable and metastable Li-Sn compounds with quite diverse and unusual crystal structures, and thereby, provides a comprehensive study to develop a better understanding of reaction mechanism in Li-Sn batteries, in particular, and a richer Li-Sn chemistry, in general. Very interestingly, our calculations predict one of the most stable compound at ambient pressure–Li8 Sn3 , which is currently not present in the available Li-Sn phase diagram, along with several Li-rich metastable and stable phases like Li3 Sn1 , Li4 Sn1 , Li5 Sn1 , Li7 Sn1 at ambient and high pressure, respectively. It is noteworthy to mention that many of these novel high pressure Li-rich compounds can 4


Page 4 of 31

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


be synthesized using hydrostatic pressure, which may also survive decompression at ambient pressure, and thus, can help in optimizing the anode for high-performance Li-Sn batteries. We believe that our study will provide a basis for future experimental work.

2

METHODS

We used evolutionary algorithm based technique as implemented in the USPEX code, 21–23 together with DFT and DFPT to find stable and metastable Li-Sn compounds as well as their ground state structures, at ambient and high pressure. Computational details regarding USPEX and DFT calculations are provided in Section 1 of the supporting information (SI), together with the details related to calculation of various properties, as well.

3 3.1

RESULTS AND DISCUSSIONS Thermodynamic Stability and Phase Diagram of Li-Sn Compounds

To explore the stable and metastable Li-Sn compounds in the pressure ranging from 1 atm to 20 GPa, we computed convex hull and pressure-composition phase diagram, depicted in Figure 1. It can be noted from the phase diagram (Figure 1(b)) that within the given pressure range, Li exhibits two different stable phases: Im3m (1 atm–7.6 GPa) and Fm3m (7.6 GPa–20 GPa), while Sn possesses three stable phases viz. Fd3m (1 atm–0.8 GPa), I41 /amd (0.8 GPa–7.2 GPa), and I4/mmm (7.2 GPa–20 GPa). This phase sequence for pure Li and Sn are also in accordance with the previous experimental and theoretical data. 24–27 In order to draw the thermodynamic convex hull at different pressure, enthalpy of appropriate phase of pure Li and Sn are considered. It should also be noted here that since we considered maximum of 40 atoms per unit cell in our calculations, this restricts us to obtain the stoichiometry Li17 Sn4 (F43m, Z = 20 ). However, in order to account the effect of Li17 Sn4 on the thermodynamical 5



Sn/(Li+Sn) 0

(a)

0.2

0.4

0.6

0.8

1

0

-0.2 -0.4 0

(b)

I41/amd

Fd3m

P2/m

Cmmm P-3m1

P = 1 atm

I41/amd

P4/mbm

I4/mmm

Sn

Pbam P2/m

∆H (eV/atom)

Li2Sn5 Li1Sn1

Pm3m

-0.2

P21/m

-0.4

R3m P3m1

P = 5 GPa 0

Li7Sn3 Li5Sn2 Li13Sn5

R3m P2/m

-0.25

Li8Sn3 Li3Sn1

Fm3m P3m1

-0.5

P = 10 GPa

2:5

1:1

7:2 3:1 8:3

7:3

5:1

4:1 13:5 5:2

P = 20 GPa

1 atm

2

4

Li7Sn1

Fm3m

Im3m

17:4

Li5Sn1

P1

C2/m

-0.5

Li

Li4Sn1

C2/m

-0.25

-0.75

Li7Sn2

I4/m

R3m

0

7:1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

6

Sn

8

10 12 Pressure (GPa)

14

Li 16

18

20

Figure 1: (a) Convex hull for the Li-Sn system at 1atm, 5, 10, and 20 GPa pressure. Thermodynamically stable phases of Li-Sn (except Li17 Sn4 ) are represented with blue circles, while red squares are used to identify the low-energy metastable phases. Meanwhile, the same is represented by green circle and yellow square, respectively, for the Li17 Sn4 . (b) Pressurecomposition phase diagram of the Li-Sn system ranging from 1atm to 20 GPa. The stable and metastable phases are shown by solid bold and thin dash lines, respectively. stability of other compounds, we explicitly considered the formation enthalpy of Li17 Sn4 (calculated separately) to draw convex hull at the pressure values of 1 atm, 5, 10, and 20 GPa. In figure showing convex hull (Figure 1(a)), thermodynamically stable phases of Li-Sn (except Li17 Sn4 ) are represented with blue circles, while red squares are used to identify the low-lying local minima on all the convex hull. Meanwhile, the same is represented by green circle and yellow square, respectively, for the Li17 Sn4 . 3.1.1

Li-Sn Compounds at Ambient Pressure

On analysing the convex hull for 1 atm pressure (Figure 1(a)), we found that our ab initio evolutionary search correctly predicts most of the experimentally known Li-Sn compositions and their respective phases, except for Li7 Sn2 . In case of Li7 Sn2 experimentally known phase is Cmmm, while we found P3m1 phase to be more stable than Cmmm. However, the difference in the enthalpy for both phases is just 6 meV/atoms, which is in agreement with results of Geneser et al. 28 From our calculations, the experimentally known stoichiometries (phases) such as Li1 Sn1 (P2/m), Li13 Sn5 (P¯3m1), Li7 Sn2 (P¯3m1), and Li17 Sn4 (F¯43m), are 6


Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


found to lie on convex-hull tie-line, while the other experimentally known stoichiometries like Li2 Sn5 (P4/mbm), Li7 Sn3 (P21 /m), and Li5 Sn2 (R¯3m) are found marginally above the convex-hull tie-line (≥ 1 – 3 meV/atom), representing these phases to be metastable at ambient pressure and temperature, as none of the structures possess dynamical instability in their respective phonon dispersion curves (Figure S1 in SI). The decomposition energy of these experimentally identified compounds in terms of measured vertical length from the convex hull are tabulated in Table T1 of SI. Finally, in order to give a firm insight in our theoretical findings and to cross check the thermodynamic stability of experimentally reported Li-Sn compounds, we compared our results with the convex hull build upon the experimental structures available in the open quantum materials database (OQMD). 29,30 Except for Li2 Sn5 (after inclusion of formation enthalpy of Li17 Sn4 in the convexhull), our results are found to be in good agreement with the OQMD based convex hull. In case of Li2 Sn5 , the decomposition energy is less than 3 meV/atom. This energy difference lies close to the boundary line of numerical accuracy in our calculations, therefore, we believe that such discrepancy can be ignored. Besides the above mentioned known compositions in the Li-Sn system, very interestingly, our calculations revealed one of the most stable stoichiometry (next to Li13 Sn5 ) of Li-Sn, i.e., Li8 Sn3 (R3m) which currently does not exist in the phase-diagram for Li-Sn, neither at ambient conditions nor at high temperature or pressure. While, our calculations predict this composition to be stable through out the investigated pressure range. This composition, however, was previously predicted by Gasior et al. in 1996 using electromotive force methods, 31 but it’s structure and phase were not known until now and thus, this composition is still missing in the Li-Sn phase diagram. But, our results strongly convince us for its existence which is also supported by recent work of Morris et al. 32 , and thus, we believe that careful measurements at ambient conditions can perhaps discover this composition experimentally as well. Furthermore, our calculations also acknowledge four yet unknown Li-rich compositions: 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Li3 Sn1 (P2/m), Li4 Sn1 (R¯3m), Li5 Sn1 (C2/m), and Li7 Sn1 (C2/m). In case of 1 atm pressure, all these stoichiometries do not lie on the convex hull tie-line but the absence of imaginary frequencies in phonon dispersion curves (Figure S1 & S2 in SI) and calculation of mechanical properties (which we will discuss in later section), confirm these compounds to be metastable at 1 atm pressure. In order to not to ignore the crucial role played by ion-dynamics in deciding the ground state energy of materials, especially of light materials, we re-calculated the formation enthalpies at 1 atm pressure by considering the vibrational contributions (zero-point energy). The results presented in Table T2 of SI, clearly indicate the negligible effect of zero-point energy (ZPE) on the formation enthalpies for all Li-Sn compounds (≥ smaller by three orders of magnitude), which is not even changing the order of stability of the compounds. We therefore have neglected the contribution of ZPE while addressing the relative stability of Li-Sn compounds at higher pressure. 3.1.2

Li-Sn Compounds at High Pressure

It should be noted that since the metastable compounds, especially the low-energy ones, are synthesizable under certain thermodynamical conditions, 33,34 we believe that most of the above discussed Li-rich compositions can be synthesized. Therefore, to guide synthesis of these novel stoichiometries (Li3 Sn1 , Li4 Sn1 , Li5 Sn1 , Li7 Sn1 ) by determining the pressure range in which these compounds stay stable, we also performed calculations to study Li-Sn compositions at high pressure (up to 20 GPa). We trust that if synthesized, there remains a high possibility for these compounds to survive decompression from high to ambient pressure with enriched mechanical properties. 8 We propose to use these high pressure quenched Li-rich Li-Sn phases as pre-lithiated anode for void space engineering to improve the performance of the battery. In case of pre-lithiated anode, void space can be created after extraction of Li during the initial delithiation process, which can effectively help in accommodating the volume expansion during the subsequent lithiation cycles. 18–20 This has been recently demonstrated by Li et al. 17 that, in general, the use of pre-lithiated Lix Si phases over the un8


Page 8 of 31

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


modified Si enhances the overall performance of the batteries while, in particular, a superior discharging capacity can be obtained with the use of Li-rich compound like Li4.4 Si. Thus, we believe that if our newly discovered high pressure Li-rich compounds like Li5 Sn1 and Li7 Sn1 can be synthesized, decompressed at ambient pressure (as they are dynamically stable at ambient pressure), and then used as pre-lithiated reagent, a high performance Li-Sn battery can be realized. In fact, even if these Li-rich quenched phases do not return to the same phase (stoichiometry) after one delithiation/lithiation cycle, the void space or the enlarged volume of the anode will however be there to limit the extreme change in the volume, and thus, shall not effect the performance of the battery. Furthermore, the study of high pressure phase will also enrich our understanding of the reaction mechanism during lithiation of Sn, since high pressure phases are observed to form during cycling of intermetallic rechargeable batteries. For example, a high pressure phases, i.e., cubic-Na3 Sb, which is well known to be synthesized only between 1-9 GPa, has been shown to form (using XRD) at the end of the discharge of Na-Sb cell by Darwiche et al. 13 Villevieille et al. 14 also observed the formation of high pressure NiSb2 phase during delithiation of Ni-Sb anode. Similar results have also been reported by Obrovac et al. 15 for the Na-Sn system, where the group observed the formation of metastable NaSn3 phase during sodiation of Sn anode. All these studies further indicate the importance and need of studying high pressure phases as well as compositions of Li-Sn in order to develop a proper understanding of reaction mechanism during charging-discharging process in a Li-Sn battery.

To investigate the high pressure (1atm9.8GPa

shown in Figure S3 of SI.

The most fascinating results are observed beyond 10 GPa pressure (see convex hull for 20 GPa and phase diagram). Our calculations disclose a diverse chemistry of Li-Sn compounds by predicting the stability of some novel Li-rich Li-Sn compounds. Except for Li2 Sn5 and Li7 Sn3 , all metastable stoichiometries such as Li5 Sn2 , Li3 Sn1 , Li4 Sn1 , and Li7 Sn1 become stable at this range of pressure (10

Understanding the Lithiation of the Sn Anode for High-Performance Li

Recommend Documents