
We demonstrate the existence of doubly charged exciton states in strongly screened bilayers of transition metal dichalcogenide (TMD) layers. These complexes are important because they are preformed electron pairs that can, in principle, undergo Bose-Einstein condensation (BEC), in which case they would also form a new type of superconductor, consisting of stable bosons with net charge
(The design of the structure, showing the circuit connections, with the middle layer defined as the ground)
This is the equivalent circuit. The bottom, metallic (Au) layer has two roles; first, it acts as a back gate for the doping of MoSe2 layers, and second, it provides screening that leads to image charge, as shown in figure above. This dual gate structure is aimed to dope the two layers of MoSe2 separately. This structure is commonly used in many interlayer studies. We applied the PDMS dry transfer technique to create the stack; the stacks are built on the gold gate with an hBN layer of approximately 2 nm between the TMD layers, a spacer hBN layer of approximately 8 nm between the metal and the first TMD monolayer, and a thick capping hBN layer. The drain and top gate (source) gates are 130 nm gold deposited by EBPG after the transfer.

(Microscope image of the quaternion sample.)

A metastable state of four fermions in a 2D structure is no longer controversial, in general. In the past decade, numerous reports have confirmed the existence of higherorder excitonic complexes in transition metal dichalcogenide (TMD) monolayer and bilayer structures. These include both intralayer and interlayer excitons, trions (excitons bound to one extra charge), biexcitons (two neutral excitons bound together), hexcitons (three excitons bound together) and charged biexcitons (biexcitons bound to one extra charge). The existence of a state with net charge of 2e, however, has been less well established.
(Illustration of the quaternion four-fermion state, with image charge in the metal layer.)
Our measurements include 1) continuous control of the doping density with both positive and negative carriers, showing the expected population dependencies on the free carrier density, and 2) measurement of the dependence on magnetic field, showing that this new bound state is a spin triplet. These results imply that it is promising to look for superconductivity in this system.

(PL intensity with back gate sweep)