Doi: 10

Boron nitride nanotubes (BNNTs) [ 1,2 ] and boron nitride nanosheets (BNNSs) [ 3 ] have gained increasing attention for their structural similarity to carbon nanotubes (CNTs) and graphene, respectively. These BN nanostructures are wide-bandgap materials ( ∼ 6 eV), and their electronic properties are different from those of CNTs and graphene. In addition, BNNTs and BNNSs offer intriguing properties that are not available in carbon nanostructures. For example, the bandgap of BNNTs is not sensitive to the change of their diameters, chiralities, and number of tubular shells. Furthermore, their bandgap is tunable by doping, the giant Stark effect, and structural deformation. [ 1,2 ] However, due to their electrically insulating nature, BNNTs and BNNSs cannot be used as conduction channels in electronic devices and sensors. Doping of BNNTs with carbon [ 4 ] and fl uorine [ 5 ] has led to phase segregation and defective BNNTs, respectively. Obviously doping has failed to convert BNNTs into semiconductors. More recently, bent BNNTs have been shown to conduct a minute current, [ 6 , 7 ] which is attributed to their radial deformation that was predicted to result in a reduced bandgap. The insulating nature of BNNTs and BNNSs prevents them from being used as conduction channels in electronic devices. On the other hand, semiconductors, such as Si, are indispensable for electronic devices including fi eld-effect transistors (FETs). However, future FETs will encounter various fundamental limitations, [ 8,9 ] including i) high power consumption due to leakage in the semiconducting channels; ii) short channel effects as the conduction length approaches the scale of the depletion layer width, and iii) high contact resistance at the semiconducting channels. Novel device concepts have been proposed to overcome these issues including the exploration