The potential of silicon anode based lithium ion batteries

A system-level perspective

Lithium ion batteries (LIBs) have been successfully deployed in a myriad numbers of consumer electronics and are increasingly adopted in electric vehicles. The development of high energy density LIBs is critical for meeting the existing and anticipated energy requirements of consumer electronics and electric vehicles. In that regard, silicon (Si) is considered as a potential next-generation anode material for LIBs and is projected to provide large increase in energy density. Despite over 5000 journal articles on Si anode in the past decade, there is a lack of clarity on the extent of practical improvement in energy density that can be accomplished by switching the anode from graphite to Si in LIBs. Issues related to initial loss of capacity and cyclability of Si anode have been reported extensively in these articles. Experimental data have shown that up to 40% increase in gravimetric energy density can be achieved using Si anode. However, such increase in energy density is achieved when you allow the LIBs to swell beyond permissible limits. Unlike graphite which expands only ∼10% when charged, Si expands 300–400% when charged. Such large volume change of Si will lead to swelling of LIBs if the amount of Si in Si-carbon composite (SCC) exceeds a threshold level that is required to avoid external dimensional change of the anode. The porosity of anode should be adjusted according to amount of Si in the SCC anode. Swelling of LIBs is an important practical issue and has major safety and performance implications. While swelling of LIBs arising from any source is undesirable, a volume expansion tolerance of up to ∼5% is provided by module manufacturers to permit the swelling from generation of gases from decomposition of electrolyte. As such, LIBs battery packs used in laptops or cell phones or electric vehicles have minimal space within the device to allow large volume expansion. Thus, given the problem associated with swelling of LIBs, it is intuitive that the improvement in energy density should be derived with the assumption that external dimensional change of anode is constrained. Based on this real-world assumption, we performed a simple, theoretical, system-level analysis on improvements that Si anode based LIBs can yield over conventional graphite based LIBs for three types of commercial cathodes – lithium cobalt oxide (LCO), lithium nickel manganese cobalt oxide (NMC), and lithium nickel cobalt aluminum oxide (NCA), and at a constant cathode thickness of 70 μm.

Limits of energy density: The theoretical analysis performed using well-established experimental data reveals that the amount of Si in SCC anode that would maximize the volumetric energy density of anode is limited to 11.68 wt.%. The value of Si in SCC that would maximize the volumetric energy density is independent of thickness of anode, cathode chemistry but is dependent on the composition of anode, and porosity of the lithiated anode. The theoretical boundary for gravimetric and volumetric energy density of SCC anode based LIBs was obtained for NCA based cathode and the values were ∼14% and ∼21%, respectively, higher than graphite based LIBs. This theoretical bound is for anode optimized for energy density and not for power density. For Si anode based LIBs to have the same power capabilities as graphite based LIBs, it should have similar porosity of anode at the charging state. The level of improvement in energy density will drop down to as low as ∼8% once practical acceptable lithiated porosity is accounted. These improvement levels are significantly lower than the current projected benefits based on either weight or the initial volume.

So, why is there a discrepancy between experimental and theoretical work? Most scientific papers (for valid reasons) do not report the amount of swelling of the anode or LIB. While the anodes of coin cell and low capacity pouch cells (<100 mAh) that are typically used by LIB researchers undergoes reversible volume changes during cycling; the volume change is minimal and thus do not get reported. The volumetric energy density/capacity reported in scientific articles are based on volume of cells in the discharged state and not in the fully charged state. The swelling issues become obvious and problematic when LIBs are made using high-capacity (>5 Ah) commercial pouch cells.

Based on our work, we believe that ∼8% improvement in energy density of Si anode based LIBs is good enough to justify the incorporation of Si into anodes and thus, the presence of Si anode as next generational anode material in prominent technology roadmaps is rightly justified. However, expecting Si to contribute to a large increase in energy density needs to be reexamined. Exploration should continue on other technology options (e.g., beyond lithium ion technologies, flow batteries) that can provide significant improvements in energy density. Other anode materials that can provide higher specific capacity than graphite such as hard carbons, composite alloys, metal oxides, etc. will continue to be attractive. Higher capacity cathodes and approaches that can enable thicker cathode will increase the utilization of Si anode and thus will provide higher level of improvement on a cell level. A full article on this work is published in Scientific Reports (http://dx.doi.org/10.1038/srep27449).

Disclaimer: The views and opinions expressed in this article are those of the authors and do not necessarily reflect the position of SABIC.