In Situ Strain Maps during Lithiation of a Graphite Anode

The swelling and shrinking associated with insertion and de-insertion of lithium into battery electrodes on each cycle can cause battery fracture, degradation, and failure.

Lithium ion transport within the electrodes can be significantly affected, which in turn affects the battery’s life and durability. Previous strain measurements in Li battery electrodes have used model electrodes or have been at a macroscopic level. Such measurements cannot be used to validate proposed microscopic models that aim to understand and control battery life: performance, degradation, and durability. In this paper, we report the first in-situ measurements of micro-scale strains in a commercial lithium battery electrode during lithium insertion into the graphite. The experimentally measured deformation and strain field maps display both dilation and contraction. (In the video images below, strain values have been scaled, so their absolute values are not meaningful.)

Through a combination of experiment and analytical modeling, we explain the observed but unexpected contraction during lithiation as due to the substantial stiffening of graphite as lithium inserts. This phenomenon was recently predicted by us using DFT calculations, but it had been ignored in previous electrode stress models. In situ quantification of local strains shows that increased graphite crystallite volume during lithiation is accommodated primarily by a significant decrease in the electrode porosity. Such a change in porosity can substantially impact battery performance and lead to lithium battery aging, degradation and failure; however, this effect on battery durability has also generally been ignored in models of lithium-ion batteries.

The surprising cycle-to-cycle material and microstructural changes, as elucidated here, will have a profound influence on our understanding of the relationship between chemistry, electrochemistry, and mechanics in Li battery electrodes The consistency of the experimental strain maps with the diffusion-induced stress model relies on our previous prediction that the modulus of lithiated graphite increases with Li concentration.

In fact, the modulus of a material can increase with Li insertion (as with graphite) or decrease with Li insertion (as with Si). In this sense, the phenomenon is quite different from the analogous thermal stress problem, where the underlying material does not change significantly with temperature, and a constant modulus is a valid assumption. We suggest that caution should be exercised using a diffusion-induced stress model to predict the stress generated due to Li diffusion in Li-ion batteries in cases where material mechanical properties vary as a function of Li concentration.

In these operando frames are shown (a) a sequence of images of lithiated graphite (Lishen 18650 battery) labeled 2, 3, 4, 6, 7, 8, 9, 12, 14, 17, 20, 25, 30, 35, 40, 45, 50.

At the top of each image after the first is a pair of numbers, such as 2-4. This means that image 4, which you are viewing on the left, is being compared to image 2 (the first frame in the set).

The comparison is being done with Digital Image Correlation (DIC) software, http://en.wikipedia.org/wiki/Digital_image_correlation. This software “is an optical method that employs tracking & image registration techniques for accurate 2D and 3D measurements of deformation, displacement and strain from the digital images.”

The results are shown on the right image of each frame. Each gridpoint on the right image is associated with the corresponding point on the left image, and it has an arrow, which shows the displacement that occurred at that gridpoint between, for example, images 2 and 4. The strain at that point in the vertical direction is the derivative of the displacement with respect to the vertical direction. It is indicated by the color, where purple represents zero strain and red represents the maximum strain. (Strain values have been scaled so their absolute values are not meaningful in these images.)

We see in each frame that regions converting from red (stage 2, LiC₁₂) to gold (stage 1, LiC₆) dilate, while regions converting from blue/black (dilute stage 2) to red contract. This is a result of the fact predicted by Qi et al that graphite stiffens (higher modulus) as it is lithiated.