Graphene plasmons have attracted a lot of attention due to large confinement and small mode volume. However, the graphene-based plasmonic devices are still limited in the practical applications due to relatively small light absorption of graphene and limited light-matter coupling efficiency in general excitation strategy. Here, this work reported a strong plasmonic coupling effect observed in a novel graphene-Bi 2 Te 3 heterostructure on the top of silicon gratings. It is interesting to find that the extinction spectra of the graphene-Bi 2 Te 3 heterostructure has shown three times greater magnitude than that of graphene. This observation is mainly attributed to two factors: first, the coupling efficiency between the graphene and Bi 2 Te 3 second, the higher light absorption in the graphene-Bi 2 Te 3 heterostructure. Moreover, the plasmonic resonance peak of the graphene-Bi 2 Te 3 heterostructure can be easily tuned by changing the grating period just like what happens in the graphene film. In all, this work utilizes the simple silicon grating to couple the light into the graphene-Bi 2 Te 3 heterostructure, and further explores the hybridized Dirac plasmons in the graphene-Bi 2 Te 3 heterostructure. We believe it will stimulate the interest to study the variant plasmonic heterostructure and trigger new terahertz device applications. Graphene plasmons have attracted a lot of attention due to large confinement and small mode volume. However, the graphene-based plasmonic devices are still limited in the practical applications due to relatively small light absorption of graphene and limited light-matter coupling efficiency in general excitation strategy. Here, this work reported a strong plasmonic coupling effect observed in a novel graphene-Bi 2 Te 3 heterostructure on the top of silicon gratings. It is interesting to find that the extinction spectra of the graphene-Bi 2 Te 3 heterostructure has shown three times greater magnitude than that of graphene. This observation is mainly attributed to two factors: first, the coupling efficiency between the graphene and Bi 2 Te 3 ; second, the higher light absorption in the graphene-Bi 2 Te 3 heterostructure. Moreover, the plasmonic resonance peak of the graphene-Bi 2 Te 3 heterostructure can be easily tuned by changing the grating period just like what happens in the graphene film. In all, this work utilizes the simple silicon grating to couple the light into the graphene-Bi 2 Te 3 heterostructure, and further explores the hybridized Dirac plasmons in the graphene-Bi 2 Te 3 heterostructure. We believe it will stimulate the interest to study the variant plasmonic heterostructure and trigger new terahertz device applications.