The graphene layers anchored onto the silicon surface use a novel approach to accommodating the well-known problematic volume expansion of silicon: a sliding process between adjacent graphene layers. This allows the designers to avoid providing a void space in the electrode to accomodate silicon expansion.

When paired with a commercial cathode (LiCoO₂), the SAIT Si anodes allowed a full Li-ion cell to reach a volumetric energy density of 972 and 700 Wh L⁻¹ at first and 200^th cycle, respectively, 1.8 and 1.5 times higher than those of current commercial lithium-ion batteries.

These volumetric energy densities work out to gravimetric densities of 242.0 and 169.6 Wh kg⁻¹ at the first and 200^th cycles, respectively—these are ~22% and 0% times higher than those of a graphite-based control full cell, respectively.

In addition, the team noted, despite the high volumetric energy density due to the dense graphene-coated Si particle packing, safety is unlikely to be an issue because the volume change of the anode is <10% of the entire cell volume.

(a) A low-magnification TEM image of Gr–Si NP. (b) A higher-magnification TEM image for the same Gr–Si NP from the white box in a. (Insets) The line profiles from the two red boxes indicate that the interlayer spacing between graphene layers is ~3.4 Å, in good agreement with that of typical graphene layers based on van der Waals interaction. (c) A high-magnification TEM image visualizing the origins (red arrows) from which individual graphene layers grow. (d) A schematic illustration showing the sliding process of the graphene coating layers that can buffer the volume expansion of Si. Source: Son et al. Click to enlarge.

The theoretical gravimetric capacity of silicon (Si) reaches almost 4,000 mAh g⁻¹. This unparalleled value has stimulated the battery community to invest considerable research efforts because the high gravimetric capacity enables one to increase the energy densities of lithium-ion batteries (LIBs) significantly, and thus bring future LIB applications, such as electrical vehicles, to a reality. In the past decade, diverse advanced electrode structures and binder designs were developed to resolve chronic capacity fading issues originating from the large volume change of Si, leading to substantially improved cycling performance even over thousands of cycles. In spite of the promising gravimetric value and substantial progress in cycle life, most of Si anodes demonstrated to date have focused primarily on the gravimetric capacity but have not offered a similar promise in their volumetric capacity because existing electrode designs rely on pre-defined void space to accommodate the volume expansion of Si.

In many LIB applications including portable electronics, however, the volumetric energy density is a critical parameter in determining battery performance. Together with a relatively inferior cycle life, weak volumetric energy density is presently a major bottleneck in implementing Si anodes in commercial cells. To meet this critical demand, Si anode technology needs to be revisited with different electrode designs that offer stable cycling performance while the electrode volume is minimized.

… In an attempt to address the limitation of previous conductive coatings as well as to achieve good cycling performance with a significantly higher volumetric energy density, in this study, we adopt multilayer graphene directly grown on the Si surface as a coating material.

—Son et al.

Growing graphene directly on silicon via CVD has been challenging because typical conditions require a reducing atmosphere that tends to strip the native silicon oxide layer off the Si surface, and then drives a reaction between Si and decomposed carbon precursors to form SiC, the researchers noted.

Avoiding the formation of SiC is key; SiC formation is fatal in Si anode operations because SiC is an electrical insulator with poor defect characteristics. Moreover, SiC is inactive in reacting with Li ions and consequently hinders Li ion diffusion into the Si phase.

After experimentation, the SAIT team hit upon including CO₂ (a mild oxidant) in the CVD process along with methane. The inclusion of CO₂ allows avoids the formation of SiC and also lowers the growth temperature. The CO₂also generates more catalytic sites on the surface.

Electrochemical testing showed that graphene coating made a significant difference in the cycling performance and rate capability.

The layered structure of graphene allows interlayer sliding upon the volume expansion of Si as well as a highly conductive percolating network, resulting in an unprecedented volumetric energy density of an LIB full cell with decent cycle life. Overall, the unique 2D character of graphene and the atom-level engineering of its interface with Si to avoid unwanted SiC formation will allow Si anode technology to make a meaningful step towards its wide commercialization.

—Son et al.

Patents covering the new technology have been applied for in Korea, China, Europe and the United States.

Resources

• In Hyuk Son, Jong Hwan Park, Soonchul Kwon, Seongyong Park, Mark H. Rümmeli, Alicja Bachmatiuk, Hyun Jae Song, Junhwan Ku, Jang Wook Choi, Jae-man Choi, Seok-Gwang Doo & Hyuk Chang (2015) “Silicon carbide-free graphene growth on silicon for lithium-ion battery with high volumetric energy density” Nature Communications 6, Article number: 7393 doi: 10.1038/ncomms8393