Kinetics-driven high power Li-ion battery with a-Si/NiSi_x core-shell nanowire anodes

Abstract

We report amorphous-Si nanowire shell anodes, supported by NiSi_xnanowire cores, by catalyst-free two-step SiH₄ chemical vapor deposition, where the metallic core acts as a mechanical supporter and a kinetically unlimited charge supplier. We have achieved highly reversible capacitance of over 3000 mAh g⁻¹ even at a 2C rate, with stable cyclic retention which stems from the altered electrochemical reactions with relatively small volume expansion routes by a kinetic effect.

---

Introduction

Anode architectures, particularly when they are three-dimensional at the nanometre scale, are closely related with cell performances in Li-ion batteries.^1–6 Therein, the achievable electrochemical capacity and the power characteristics are inherently determined by a series of phase transitions involved in the anode during lithiation/delithiation. The electrochemical Li–Si system forms various Li-rich Li_xSi intermetallics (0 ≤ x ≤ 4.4), correspondingly symptomatic of the largest known gravimetric charge capacity of up to 4200 mAh g⁻¹ in the bulk limit, thus can represent an attractive anodic system for the high capacity Li-ion battery.^1,3–5

However, the system undergoes notoriously large volume-changes up to 400% during the cyclic lithiation and delithiation, resulting in poor cyclic retention in its cell capacity and output power.^1,3–5 It is suggested that this severe anode pulverization upon the cyclic volume changes can be alleviated by adopting one-dimensional Si nanocrystals, such as Si nanowires (NWs), based on the promises of their affordable accommodation of large volume-changes and the efficient charge collection capability.^4,5,7–11 Yet, monolithic single-crystalline Si NWs exhibit cyclic degradation in their cell capacity particularly at the high charging/discharging rate, thus limiting their integration into the high power cells.^5,12–18

Here, we report amorphous-Si (a-Si) shell anodes, supported by NiSi_x NW cores, by a catalyst-free two-step SiH₄ chemical vapor deposition, where the metallic core acts as a mechanical supporter and a kinetically unlimited charge supplier. Spontaneous silicide NWs, such as NiSi_x, FeSi, CoSi, TaSi₂^19–27 NWs, have been recently investigated, and among others NiSi_x NWs are of particular interest because their growth can be easily integrated with silicon processing technology. Besides, NiSi_x exhibits the lowest resistivity and makes good contacts with Si.^25–27 In this regard, we designed the core-shell NW anode architecture in this work. We have achieved the highly reversible capacity over 3000 mAh g⁻¹ even at 2C rate, with stable cyclic retention which stems from the altered electrochemical reactions with relatively small volume expansion routes by a kinetic effect.

Results and discussion

a-Si NW shells, supported by metallic NiSi_x NW cores (a-Si/NiSi_x NW), were grown by a simple two-step SiH₄ CVD process: (i) the self-catalytic NiSi_x NWs were grown on Ni thin films, (ii) followed by a-Si shells deposition on the NiSi_x NWs, as illustrated in Fig. 1(a). We start off the NiSi_x NW synthesis by thermal evaporation of Ni films on Ag-coated stainless steel (SUS) substrates of a circular shape disk in 15mm diameter. Ag acts as a diffusion barrier of Fe from SUS into the Ni films. The Ni/Ag/SUS substrates were annealed with oxygen at 400 °C, and then reacted with 50 Torr of SiH₄ (10% diluted in H₂) at 420 °C, as described in Ref. 26,27. The a-Si shell deposition was conducted by using thermal decomposition of SiH₄ (10% diluted in H₂) at 550 °C for 30 min. Here we note that a-Si shells are formed, when NiSi_x NWs are oxygen-exposed prior to the SiH₄ flows, whereas the crystalline Si shell (c-Si) was formed otherwise, as shown in Fig. S1.† We also prepared a-Si shell NWs on the insulating c-Si core NWs, and describe a series of comparative cyclic cell performances in greater details,²⁸ and discuss them around the roles of the metallic core for altered electrochemical reactions.^29,30

Fig. 1 (In color) (a) The schematics of the NW heterostructure growth in this study. (b) TEM image of an individual a-Si/NiSi_x NW. The inset shows a plan-view SEM image of a-Si /NiSi_x NWs grown on SUS substrate. (c) HRTEM image of an individual a-Si/NiSi_x NW at the core region and the corresponding FFT-DP along the [1–10] zone axis (inset). (d) HRTEM image of an individual a-Si/NiSi_x NW at the shell region and the corresponding FFT-DP (inset). (e) HAADF image of a-Si/NiSi_x NW. (f and g) EDX elemental mapping images of Si (f) and Ni (g) in (e).

Fig. 1(b) shows the representative microstructures of a-Si/NiSi_x NWs by transmission electron microscope (TEM), where the average diameter of a-Si/NiSi_x NW is 190 nm with the 20 nm thick NiSi_x core NWs. Scanning electron microscope (SEM) images in the inset illustrate that the shell NWs were conformally deposited onto the whole NWs in the substrate – see also Fig. S2 in Supporting Information.†High-resolution TEM (HRTEM) images and corresponding fast fourier transform diffraction patterns (FFT-DP) in Fig. 1(c) and 1(d) confirm that Si/NiSi_x NW features amorphous Si NW shells and single-crystalline NiSi₂ NW cores.³¹ High-angle annular dark field (HAADF) images, where the intensity is roughly proportional to square of atomic number, and energy dispersive X-ray spectroscopy (EDX) mapping images in Fig. 1(e), (f) and (g) also demonstrate that core/shell NW structures are clearly discernable, and the elemental distribution of Ni is locally limited in the core region within the instrumental resolution.

The a-Si/NiSi_x NW anodes were incorporated into a coin-type half cell (CR2016 coin-type). The cell uses Li metal foils as the counter and the reference electrode, and 1M LiPF₆ dissolved in a 1 : 1 (by volume, provided by Techono. Semichem. Co.) mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) as the electrolyte. The cell assembly was carried out in an Ar-filled glove box with less than 0.1 ppm each of oxygen and moisture. The charge capacity of the a-Si/NiSi_x NW cell was measured over the potential range of 0.01 to 2.00 V (vs.Li/Li⁺) during the initial 50 cycles in the galvanostatic mode. The measured discharge capacity is carefully normalized into the gravimetric capacity, as described in Fig. S3.†

Fig. 2 shows the cyclic retention of two types of NWs: a-Si/NiSi_x NW cells and a-Si/c-Si NW cells. It is found that a-Si/NiSi_x NWs retain almost the 83% of the first discharge capacity after the 50th cycle with the high charge capacity above 3000 mAh g⁻¹. For a check, we tested the bare NiSi_x NW anodes in the galvanostatic mode; however, we did not observe any apparent potential plateau over the potential range of 0.01 to 3.00 V – also see Fig. S4.† This suggests that the bare NiSi_x NWs alone are electrochemically inactive in our experimental conditions. We have also tested the rate dependent capacity by charging the cell at a constant rate of 0.1C and discharging it at the various rate from 0.1C to 2C (1C≡1 h per half-cycle). Fig. 2(b) shows that the discharge capacity of a-Si/NiSi_x NWs is maintained over 94% of the first discharge capacity, and weakly depends on the C-rate up to 2C. The discharge capacity of a-Si/c-Si NWs, on the other hand, fades significantly and strongly depends on the C-rate. It is obvious that a combination of a-Si shells and metallic NiSi_x NW cores is responsible for the unprecedented reversible lithiation/delithiation in a-Si shells at the high charge/discharge rate.

Fig. 2 (In color) (a) The discharge capacity retention for the a-Si/NiSi_x NW (red circles) and the a-Si/c-Si NW (blue circles), respectively. The inset shows the specific capacity for the a-Si/NiSi_x NW (red close circles-charge and red open circles-discharge) and the a-Si/c-Si NW (blue close circles-charge and blue open circles-discharge), respectively. (b) The rate capability for the a-Si/NiSi_x NW (red circles) and the a-Si/c-Si NW (blue circles), respectively.

Fig. 3(a–c) present HRTEM and HAADF images of a-Si/c-Si NWs after the 50 cycles, where they become porous and pulverized upon the cyclic lithiation/delithiation. By striking contrast, a-Si/NiSi_x NWs maintain the initial core/shell NW structures. Apparently, the structural robustness in a-Si/NiSi_x NWs, guaranteed from a mechanical supporter of NiSi_x cores, is responsible for the enhanced cyclic retention and rate capability, as shown in Fig. 3(d–f).

Fig. 3 (In color) (a–f) TEM images of individual a-Si/c-Si NW and a-Si/NiSi_x NW after 50th cycling. (a) Low magnitude TEM image of a-Si/c-Si NW. (b) HRTEM image of a-Si/c-Si NW at the surface. (c) HAADF image of a-Si/c-Si NW. (d) Low magnitude TEM image of a-Si/NiSi_x NW. (e) HRTEM image of a-Si/NiSi_x NW at the surface. (f) HAADF image of a-Si/NiSi_x NW. (g) The differential capacity of the 10th cycle for the a-Si/NiSi_x NW (red line), the a-Si/c-Si NW (blue line).

In order to further investigate the electrochemical roles of the metallic NiSi_x cores, we tried to compare the differential capacity plots for a-Si/NiSi_x NWs with those of a-Si/c-Si NWs, as in Fig. 3(g). After the 1st cycle, which is called the formation or activation process, a-Si/c-Si NWs and a-Si/NiSi_x NWs display two reduction peaks around 260 mV and 80 mV, as marked by A and B, and they are associated with the formations of intermetallic compounds, Li_xSi. Reportedly, the 260 mV and 80 mV peaks are assigned to the formation of amorphous Li_xSi (a-Li_xSi) from a-Si and c-Si, respectively.^32–34 Although other small reduction peaks below 80 mV, marked as C and C′, are not well identified, they are usually assigned to either routes for a-Li_xSi to a-Li_x+ySi (fully lithiated amorphous lithium silicides) transition or a-Li_xSi to c-Li_zSi (fully lithiated crystalline lithium silicides) transition. Identification of the fully lithiated phase whether it is a-Li_xSi to c-Li_zSi is critical, since the a-Li_xSi to c-Li_zSi or vice versa among the series of lithiation/delithiation reactions involves the largest volume changes, which is in turn the major source for the deterioration in the anodic reversibility.^34–37 During the discharge process, a-Si/c-Si NWs display the dominant oxidation peak at 450 mV, marked as F, assigned to the dissociation of c-Li_zSi to a-Li_xSi or a-Si, indicating that its reduction peak below 80 mV corresponds to the nucleation of a crystalline lithium silicide (c-Li_zSi).^34–36 By contrast, a-Si/NiSi_x NWs exhibit two dissimilar oxidation peaks between 250 mV and 500 mV, marked as D and E, which are assigned to the dissociation of a-Li_x+ySi to a-Li_xSi or a-Si.^32,34,37 This finding can be back-traced to assign another phase transition, C′ during the Li charging to the formation of a-Li_x+ySi. In other words, the full lithiation into a-Li_x+ySi in a-Si/NiSi_x, instead of c-Li_zSi in a-Si/c-Si NWs, alleviates severe physical pulverization and leads to the stable reversibility in the cyclic cell performance. We speculate that kinetically unlimited electron supply from NiSi_x NW core coupled with the properly absorbed volume change by a-Si shell provides alternative electrochemical reactions of a-Si/NiSi_x NWs to skip over the nucleation of c-Li_xSi by a kinetic effect. We measured average resistivity of NiSi_x NWs and c-Si NWs to be 10⁻³ Ω·cm, and 10³ Ω·cm. It is known that monolithic Si crystals usually undertake temporal local-concentration of Li ions in Si matrices, followed by the rapid and spatially inhomogeneous evolution of fully lithiated crystalline lithium silicides (c-Li_zSi), which in turn results in severe Si pulverizations. Instead, we argue that uniform distribution of the electrochemical potential in our a-Si/NiSi_x NWs, which is provided by the metallic NiSi_x cores is responsible for the reversible reaction route mediated by fully lithiated amorphous lithium silicides.^3,32,36

Acknowledgements

M.-H.J. acknowledges the support from the Nano Original & Fundamental Technology R&D Program (2010-0019195), the Priority Research Centers Program (2010-0029711), the Basic Research Program (2010-0001873 and 2010-0017853), the Mid-career Researcher Program (2010-0027627) and the WCU Program (R31-2008-000-10059-0) through the National Research Foundation (NRF) of Korea funded by the Ministry of Education, Science, and Technology (MEST), and POSCO. Y.-M.K. acknowledges the Converging Research Center Program (2010K001097) and the Basic Science Research Program (20090072972) through the NRF of Korea funded by the MEST.

Notes and references

1. P. G. Bruce, B. Scrosati and J.-M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930.
2. I. Grigoriants, L. Sominski, H. Li, I. Ifrgan, D. Aurbach and A. Gedanken, Chem. Commun., 2005, 7, 921.
3. S.-H. Ng, J. Wang, D. Wexler, K. Konstantinov, Z.-P. Guo and H.-K. Liu, Angew. Chem., Int. Ed., 2006, 45, 6896.
4. H. Kim and J. Cho, Nano Lett., 2008, 8, 3688.
5. J. R. Szczech and S. Jin, Energy Environ. Sci., 2011, 4, 56.
6. J. R. Dahn, T. Zheng, Y. Liu and J. S. Xue, Science, 1995, 270, 590.
7. C. K. Chan, H. Peng, G. Liu, K. Mcilwrath, X. F. Zhang, R. A. Huggins and Y. Cui, Nat. Nanotechnol., 2008, 3, 31.
8. K. Peng, J. Jie, W. Zhang and S.-T. Lee, Appl. Phys. Lett., 2008, 93, 033105.
9. H. Kim, B. Han, J. Choo and J. Cho, Angew. Chem., Int. Ed., 2008, 47, 10151.
10. J. Graetz, C. C. Ahn, R. Yazami and B. Fultz, Electrochem. Solid-State Lett., 2003, 6, A194.
11. C. K. Chan, X. F. Zhang and Y. Cui, Nano Lett., 2008, 8, 307.
12. A. Magasinski, P. Dixon, B. Hertzberg, A. Kvit, J. Ayala and G. Yushin, Nat. Mater., 2010, 9, 353.
13. B. Gao, S. Sinha, L. Fleming and O. Zhou, Adv. Mater., 2001, 13, 816.
14. L.-F. Cui, Y. Yang, C.-M. Hsu and Y. Cui, Nano Lett., 2009, 9, 3370.
15. S. Zhou, X. Liu and D. Wang, Nano Lett., 2010, 10, 860.
16. L.-F. Cui, R. Ruffo, C. K. Chan, H. Peng and Y. Cui, Nano Lett., 2009, 9, 491.
17. S. Zhang, Z. Du, R. Lin, T. Jiang, G. Liu, X. Wu and D. Weng, Adv. Mater., 2010, 22, 5378.
18. T. Song, J. Xia, J.-H. Lee, D. H. Lee, M.-S. Kwon, J.-M. Choi, J. Wu, S. K. Doo, H. Chang, W. I. Park, D. S. Zang, H. Kim, Y. Huang, K.-C. Hwang, J. A. Rogers and U. Paik, Nano Lett., 2010, 10, 1710.
19. A. L. Schmitt, M. J. Higgins, J. R. Szczech and S. Jin, J. Mater. Chem., 2010, 20, 223.
20. J. R. Szczech and S. Jin, J. Mater. Chem., 2010, 20, 1375.
21. A. L. Schmitt, M. J. Higgins, D. Schmeisser, F. J. Himpsel and S. Jin, Nano Lett., 2006, 6, 1617.
22. Y. Song, A. L. Schmitt and S. Jin, Nano Lett., 2006, 6, 965.
23. K. Seo, K. S. K. Varadwaj, P. Monhanty, S. Lee, Y. Jo, M.-H. Jung, J. Kim and B. Kim, Nano. Lett., 1240, 2007, 7.
24. Y.-L. Chueh, M.-T. Ko, L.-J. Chou, L.-J. Chen, C.-S. Wu and C.-D. Chen 2006, 6, p. 1637.
25. K. Kang, C.-J. Kim and M.-H. Jo, J. Appl. Phys., 2009, 105, 122407.
26. K. Kang, S.-K. Kim, C.-J. Kim and M.-H. Jo, Nano Lett., 2008, 8, 431.
27. C.-J. Kim, K. Kang, Y. S. Woo, K.-G. Ryu, H. Moon, J.-M. Kim, D.-S. Zang and M.-H. Jo, Adv. Mater., 2007, 19, 3637.
28. K. Kang, H.-S. Lee, D.-W. Han, G.-S. Kim, D.-H. Lee, G. Lee, Y.-M. Kang and M.-H. Jo, Appl. Phys. Lett., 2010, 96, 053110.
29. J. Yin, M. Wada, K. Yamamoto, Y. Kitano, S. Tanase and T. J. Sakai, J. Electrochem. Soc., 2006, 153, A472.
30. Graetz, C. C. Ahn, R. Yazami and B. Fultz, Electrochem. Solid-State Lett., 2003, 6, A194.
31. Crystal structure of Si: diamond cubic, lattice constant: 5.431 Å, space group: Fm3m, JCPDS 43–0989; Crystal structure of NiSi₂: diamond cubic, lattice constant: 5.416 Å space group: Fd-3m, JCPDS 27–1402.
32. M. N. Obrovac and L. J. Krause, J. Electrochem. Soc., 2007, 154, A103.
33. J. P. Maranchi, A. F. Hepp and P. N. Kumta, Electrochem. Solid-State Lett., 2003, 6, A198.
34. C. K. Chan, R. Ruffo, S. S. Hong, R. A. Huggins and Y. Cui, J. Power Sources, 2009, 189, 34.
35. T. D. Hatchard and J. R. Dahn, J. Electrochem. Soc., 2004, 151, A838.
36. Y. M. Kang, S. M. Lee, S. J. Kim, G. J. Jeong, M. S. Sung, W. U. Choi and S. S. Kim, Electrochem. Commun., 2007, 9, 959.
37. N. Obrovac and L. Christensen, Electrochem. Solid-State Lett., 2004, 7, A93.

---

Footnote

† Electronic supplementary information (ESI) available: Structure of c-Si/NiSi_x NW, a photograph of the a-Si/NiSi_x NW grown on substrate, normalization method for calculating the gravimetric capacity, and current charge/discharge curve for NiSi_x NW cell in the galvanostatic mode. See DOI: 10.1039/c0sc00628a

---