MXene-Based Lithium-Ion Anodes: An Outlook on Computational Predictions and Experimental Validation.

First-principles studies have produced extensive predictions of MXene-anode capacities, voltages, and diffusion barriers over the past decade. Yet, a gap remains between these predicted metrics and the performance observed in operating cells. We suggest that this gap reflects not a limitation of computation itself but the absence of a sustained feedback loop between computational and experimental work. Computational predictions are often communicated and cited as performance forecasts even though they describe idealized systems (bare or uniformly terminated surfaces, equilibrium conditions, and the absence of electrolyte, interphase, and cycling history) that differ from the environment of an operating anode. Focusing on MXene lithium-ion anodes, we examine where predicted and measured capacities diverge, distinguishing predictions that are inaccurate from predictions that are simply applied beyond the conditions they were constructed to describe. We then consider the physics that standard protocols are not designed to capture and outline three coordinated directions for reconnecting computation to operating reality: coordinated theory-experiment studies as a structural foundation, machine-learning interatomic potentials to reach the relevant time scales, and explicit electrolyte modeling to recover interfacial energetics often omitted from vacuum-reference calculations. We propose that the progress of MXene-anode research be measured by how closely computation and experiment inform each other.