Casimir-Lifshitz force with graphene: theory versus experiment, role of spatial non-locality and of losses

Pablo Rodriguez-Lopez,Mauro Antezza
2024-07-26
Abstract:We calculate the Casimir-Lifshitz force (CLF) between a metallic sphere and a graphene-coated SiO$_2$ plane and compare our finding with the experiment and theory by M. Liu \emph{et al.}, PRL {\bf 126}, 206802 (2021), where a non-local and lossless model for the graphene conductivity (GC) has been used and shown to be compatible with the experimental results. Recently, that conductivity model has been shown to be not correctly regularized [<a class="link-https" data-arxiv-id="2403.02279" href="https://arxiv.org/abs/2403.02279">arXiv:2403.02279</a>], to predict nonphysical results in the non-local regime, and being correct only in its local limit, where its expression is identical to the local Kubo conductivity model (once also losses are introduced). To compare the experimental results with the correctly regularized Kubo theory and to clarify the effective role played by the graphene non-locality and losses in that experiment, we calculated the CLF using three different models for the GC: the correct general non-local Kubo model, the local Kubo model, and the non-regularized and lossless model used by M. Liu \emph{et al.}. For the parameters of the experiment, the predictions for the Casimir-Lifshitz force using the three models are practically identical, implying that, for the experiment, both non-local and lossy effects in the GC are negligible. This explains why the GC model used in M. Liu \emph{et al.} provides results in agreement with the experiment. We find that the experiment cannot distinguish between the Drude and Plasma prescriptions. Our findings are relevant for present and future comparisons with experimental measurement of the Casimir-Lifshitz force involving graphene structures. Indeed, we show that an extremely simple local Kubo model, explicitly depending on Dirac mass, chemical potential, losses and temperature, is largely enough for a totally comprehensive comparison with typical experimental configurations.
Mesoscale and Nanoscale Physics,Quantum Physics
What problem does this paper attempt to address?
### What problems does this paper attempt to solve? This paper aims to explore and explain the non - locality and loss effects of graphene in the Casimir - Lifshitz force (CLF) experiments. Specifically, the authors focus on the following points: 1. **Comparison between theory and experiment**: - The researchers recalculated the CLF between a metal sphere and a SiO₂ plane coated with graphene and compared it with the experimental results published by M. Liu et al. in 2021. - M. Liu et al. used a non - local and loss - free graphene conductivity model for theoretical prediction and found that it was in agreement with the experimental results. 2. **Correctness of the conductivity model**: - Recent studies have pointed out that the graphene conductivity model used by M. Liu et al. was not properly regularized, resulting in non - physical results predicted in non - local cases. - To verify this, the authors of this paper used three different graphene conductivity models to recalculate the CLF: the general non - local Kubo model, the local Kubo model, and the unregularized and loss - free model used by M. Liu et al. 3. **Effects of non - locality and loss**: - By comparing the prediction results of these three models, the researchers hope to clarify the specific effects of the non - locality and loss effects of graphene on the CLF. - Under the experimental parameters, the prediction results of these three models for the CLF are almost the same, indicating that in this experiment, the non - locality and loss effects can be ignored. 4. **Ability to distinguish different models**: - The study also explored whether the experiment can distinguish between the Drude model and the Plasma model, and the results show that this distinction cannot be achieved under the current experimental conditions. ### Summary The main objective of this paper is to recalculate the CLF using different graphene conductivity models, verify and explain why the theoretical model of M. Liu et al. can be in agreement with the experimental results, and at the same time reveal that under specific experimental conditions, the effects of non - locality and loss on the CLF can be ignored. This research provides important theoretical guidance for future CLF experiments involving graphene structures.