Even though the battery being tested in this study is a different lithium ion chemistry than the lithium iron phosphate most of us are using, there are still some good concepts to take away from the paper.
First, the internal structure of the battery is put under mechanical stress as the battery state of charge changes and ions migrate back and forth between the anode and cathode. In their conclusion, they attributed much of the physical swelling of the prismatic cell to cracking, electrode de-lamination, and lithium plating. They noted that the plating problem in the braced cell was greatly reduced because the layers inside the battery were less prone to separate, and this led to a reduction in swelling and a greater retention of charge capacity.
Unfortunately, their cell brace did not allow for measurement of the pressures applied to the cell. It does appear that it is an inflexible brace, so the pressure probably changed with state of charge and age. Interestingly, they say that both the braced and unbraced cell cathodes suffered from cracking. However, it was the delamination and lithium plating that caused most of the capacity loss, and these factors were improved by bracing.
So essentially, by building a battery cell fixture that applies pressure to the cells within the limits of the manufacturers specifications, you are keeping the materials inside the battery tightly connected as the individual layers expand and contract. By doing that, you are preserving contact for ion exchange across the layers and preventing lithium from plating the surface of the materials, and that keeps the lithium in circulation in the battery and helps preserve its capacity.
Its nice to have some experimental data to support the practice of putting cells under some pressure as they are operating.
Good find.