(562h) Fabrication and Characterization of Ionomer-Gated MoTe2 Field Effect Transistors

The semiconductor-to-metal transition in monolayer MoTe₂ has been predicted to occur under uniaxial strains as low as 0.3% and has been experimentally demonstrated in few-layer MoTe₂ under 0.2% in-plane tensile strain (S. Song et al., NanoLetters, 16, 2016, 188-). Most of the current approaches to straining two-dimensional (2D) crystals require global strain of a flexible substrate or direct deformation of suspended flakes using, e.g., an AFM tip. We propose a new approach to access the strain-induced phase transition in MoTe₂ whereby individual MoTe₂ devices can be addressed electrically. The proposed method employs an MoTe₂ field-effect transistor (FET) with an ion gate, where the polymer electrolyte is a custom single-ion conductor. The anions in this ionomer are covalently bound to the polymer backbone, while the cations are free to respond to the applied field. In our device design, when a positive gate bias is applied, cations drift away from the gate to the surface of the MoTe₂ crystal and establish an electrostatic double layer (EDL) at the electrolyte/semiconductor interface, and a depletion layer at the gate/electrolyte interface. The electrostatic imbalance that is created induces strain in the ionomer, which can be transferred to a free-standing MoTe₂ channel and thereby induce a semiconducting-to-metallic phase transition. This strain response has been previously studied in the field of ionic polymer metal composites (IPMCs) for applications in biomimetic actuators and artificial muscle, but has not yet been applied to 2D electronics. MoTe₂ crystal FETs are fabricated by electron beam lithography (EBL) on MoTe₂ flakes exfoliated onto p-doped SiO₂/Si; 10 nm Ti/100 nm Au contacts are deposited by e-beam evaporation. MoTe₂ FET transfer and output characteristics will be reported with the channel current modulated by an ionomer gate. EDL formation and dissolution is measured by changes in the channel current (I_DS) as a function of the ionic gate bias (V_G).

This work is supported by the National Science Foundation under Grant #DMR-1607935.