Nonlinear constitutive models for nano-scale heat conduction

We present a first-principle based approach that leads from a many-particle description to a nonlinear, stochastic constitutive relation for the modeling of transient heat conduction processes at the nano-mechanical scale. By enforcing statistical consistency, in that the statistics of local energy is consistent with that from an all-atom description, we identify the driving force as well as the model parameters in these generalized constitutive models. The connections to established generalized constitutive relations, including Cattaneo-Vernotte-type models, will be demonstrated.