▎ 摘　　要

One of the critical issues for the fabrication of desirable sensing materials has focused on the construction of an effective continuous network with a low percolation threshold. Herein, graphene-based elastomer composites with a segregated nanostructured graphene network were prepared by a novel and effective ice-templating strategy. The segregated graphene network bestowed on the natural rubber (NR) composites an ultralow electrical percolation threshold (0.4 vol %), 8-fold lower than that of the NR/graphene composites with homogeneous dispersion morphology (3.6 vol %). The resulting composites containing 0.63 vol % graphene exhibited high liquid sensing responsivity (6700), low response time (114 s), and good reproducibility: The unique segregated structure also provides this graphene-based elastomer (containing 0.42 vol % graphene) With exceptionally high stretchability, sensitivity (gauge factor approximate to 139), and good reproducibility (similar to 400 cycles) of up to 60% strain under Cyclic tests. The fascinating performances highlight the potential applications of graphene elastomer composites with an effective segregated network as multifunctional sensing materials.