Figure 2. (a) Variation of Li capacity with the number of cycles during half cycles of Li embedding and delithiation; (b) Nominal stresses measured during half cycles of Li embedding and delithiation; (c) Coulombic efficiencies and irreversible stresses; (d) Stresses present in the sample at the beginning of each cycle;

The irreversible capacity, i.e., the difference in Li capacity for the Li embedding/delithiation process, is described by the Coulombic efficiency (CE). The cumulative irreversible stress in the initial cycle, mentioned in the article, is two times higher than the stress induced by the reversible stress, and the surface effect of the irreversible stress (film thickness effect) suggests that the irreversible component is mainly determined by one or more phenomena occurring near the surface of the film. The high capacity and consequent high stress recorded during the embedded lithium half-cycle in the first cycle and the relatively low capacity and stress reversal during the delithiation lead to a large difference between the capacity as well as the stress during the first two half-cycles, with very low Coulombic efficiency (CE) values in the first cycle. The correlation between electrochemical behavior and stress development during the first 20 cycles and the fact that the irreversible stresses are only from surface phenomena suggest that a large amount of irreversible stresses are directly related to the formation of the SEI layer.

The stress evolution in the initial cycle is also shown in more detail in Figures 2b-2d, the author designed a graphitic carbon film with a lattice coefficient parallel to the substrate so that the stress development is only related to the change of the film. During discharge (half cycle of lithium insertion), a net compressive stress will be generated, and then it will be reversed during charging (half cycle of lithium removal). In the first lithium insertion cycle, a large part of the compressive stress is not reversed (equivalent to the initial residual stress). The magnitude of the irreversible stress is significantly reduced in the second cycle, and this trend continues until the difference between the stresses of the two half cycles in the later period is negligible, the stress development during the delithiation/intercalation process is almost completely reversed. In this experiment, The maximum in-plane stress change caused by the insertion of lithium is about -0.25 GPa (reversible stress, Figure 2b), the irreversible net compressive stress formed in the initial cycle is much higher (about -0.5 GPa, see Figure 2d). The author believes that this irreversible stress is an important part of the characteristics of feedback materials.

4.2 Sources of irreversible stress

4.2.1 Surface effects of irreversible stresses

The authors verified by assumptions that the irreversible stress originates from the surface layer of the graphitic carbon layer, independently of the thickness of the CVD C-film. By varying the thickness study, if the irreversible stress appears inside the film, the corresponding stress thickness is proportional to the film thickness; conversely, if this contribution originates from the surface layer of the graphitic carbon layer, the stress thickness is independent of the film thickness.

Figure 3. (a) Actual stress thickness for irreversible stresses with different thicknesses of film

4.2.2 Stress changes under different voltages

Figure 4. (a) Potential and nominal stress (corresponding to the residual stress of the deposited film) evolution with time for the first cycle;
(b)-(e) Stress evolution in different potential ranges (embedded lithium/delithium half-cycle)

It is known that below ~0.25 V the actual Li embedding is predominant and the rapid change in compressive stress can be seen in the potential, and once the potential reaches ~1.0 V, a protective SEI layer is formed, which almost completely suppresses the solvated ionic co-embedding. Figures 4b-e summarize the stress contributions measured in different voltage ranges during lithium embedding and delithiation in cycles 1 and 2. The authors show that most of the compressive stresses in the lithium embedding half-cycle occur below 0.25 V, the stresses occur for actual lithium ion embedding, and that this voltage range is also accompanied by the ongoing formation of the SEI layer during the first few cycles, i.e., the cause of the small amount of irreversible compressive stresses observed after the second cycle is also the formation of the SEI layer; whereas, any stresses at higher potentials (> 0.25 V ) in the first cycle are directly related to SEI formation or possible co-embedding of solvated ions.

Figure 5. (a) Stress variation in the first embedded lithium half-cycle; (b) Stress variation in the second embedded lithium half-cycle; (c) Stress variation in the first embedded lithium half-cycle (Al2O3-coated CVD C-sheet)

Figure 5a stress variations are interpreted as Li embedding + SEI formation + embedding of solvated ions; Figure 5b, as Li embedding + continued slow formation of passivated SEI; and Figure 5c, as actual Li embedding in graphite (with the Al2O3 coating blocking SEI formation), which further confirms that the main source of irreversible stress is SEI formation.

Figure 6. (a) Schematic representation of the formation of the SEI film and accompanying stresses on the thin-film graphite electrodes; (b) expected behavior in the graphite particles and the surrounding SEI layer

6. Summary

In this research, an optical stress sensing device is used to measure the stress in the film in real time (in-situ). The first few electrochemical cycles of graphite electrodes against Li metal are reported for the first time. Discuss the evolution of the huge irreversible compressive stress generated in the thin-film graphite electrodes. The relationship between the irreversible stress and the formation of SEI is determined from multiple angles, which provides important guidance for the study of the electrochemical behavior and actual mechanical properties of lithium-ion batteries.

7. References

Mukhopadhyay A, Tokranov A, Xiao X, et al. Stress development due to surface processes in graphite electrodes for Li-ion batteries: A first report[J]. Electrochimica Acta, 2012, 66(none):28-37.