The authors would like to point out an error in one of the equations. The overpotential (η) used in eqn (2) of the main manuscript has been defined as η = ϕs − ϕe − ULi (see left column of the fourth page). Instead the overpotential term should be, η = ϕs − ϕe − ULi + (Δμe−/F).1,2 Here, ϕs is the solid phase potential, ϕe is the electrolyte potential, ULi is the open circuit potential of lithium metal, F indicates the Faraday constant, and Δμe− is the change in electrochemical potential due to the mechanical stress induced effects. The form of eqn (2) shown in the published version of the manuscript takes into account the impact of mechanical stress induced change in electrochemical potential (Δμe−) in the pre-exponential factor, but the effect of Δμe− was erroneously neglected in the overpotential (η) term, which appears inside the exponential.

The correction of the overpotential term does not change the qualitative results shown in the paper; however, it allows better quantitative predictions. To demonstrate how this correction in the overpotential term of the Butler–Volmer equation affects some of the results, we recreate the two key figures in the manuscript, namely, Fig. 4 and 6, with the updated form of overpotential (η). The modified figures are shown below. It is evident from the corrected figures that the relative magnitude of current density at the peak and the valley change due to incorporation of the extra term into the overpotential expression. However, the overall trend did not change significantly. Such as, enhanced dendrite growth at higher current densities is still evident from the corrected version of Fig. 4. Similarly, it can still be concluded from the modified Fig. 6 that high modulus electrolytes demonstrate improved suppression of the dendritic protrusion.

1 aBarai, Pallab1 aHiga, Kenneth1 aSrinivasan, Venkat uhttps://appliedenergyscience.lbl.gov/publications/correction-lithium-dendrite-growth