Complementary field effect transistors based on atomically thin semiconductors for ultra-scaled electronics (CFETS)

Atomically thin body (ATB) semiconductors such as transition metal dichalcogenides (TMDs) hold tremendous promise for scaling of modern electronics because their ultra-thin channel nature provides excellent gate electrostatics. Major semiconductor companies and international roadmaps for devices and systems have cited ATB semiconductors for sub-1-nm technologies beyond 2030. To realize high-performance complementary field effect transistors (CFETs) with ATB semiconductors, both n- and p-type field effect transistors (FETs) with low contact resistance, high ON current, low subthreshold swing, and stable threshold voltage are required. In bulk semiconductors, these characteristics are achieved by substitutional doping doping and forming ultra-clean interfaces through covalent bonding. However, these approaches cannot be applied to ATB semiconductors. Therefore, interfaces that are free of defects between the metal/semiconductor and semiconductor/dielectric must be realised to achieve high-performance n- and p-type FETs based on ATB semiconductors. However, making clean and defect free metal/semiconductor interface on ATB semiconductors for electron/hole transfer has proven to be challenging because atomically thin materials are fragile and easily damaged by the deposition of metals. My breakthroughs (Nature 2019, 2022) in gentle deposition of metals on monolayered TMDs provide ultra-clean van der Waals (vdW) contacts where the Fermi level of semiconductors can be tuned with gate bias so that n- or p-type polarity can be achieved by selecting high or low, respectively, work function metals. Development of vdW contacts for 2D semiconductors have shown high ON-state currents (over milliamperes per micrometre) and contact resistance close to quantum limit (tens of ohm micrometres) for n-type FETs. P-type FETs are more challenging due to issues with depositing high work function metals on ATB semiconductors. The growth of metal films on TMDs tends to give preferential orientations that have low work functions compared to their bulk counterparts. Another overlooked issue for p-type FETs is that many widely used oxide dielectrics cause electron doping on ATB semiconductors, which reduces the overall hole current in the p-type FETs. The interface interaction between ATB semiconductor and dielectric is also problematic for achieving reliable threshold voltage and low subthreshold swings for n-type FETs. My recent work shows that an inert semiconductor/dielectric interface gives low subthreshold swing values (63 mV/dec) approaching the thermal limit and very low threshold voltage for n-type FETs. These challenges remain for p-type FETs based on ATB semiconductors. The main goal of this proposal is to demonstrate high-performance CFETs based on ATB semiconductors that are currently limited by p-type FETs. This will be accomplished by studying the interface between high work function metals and ATB semiconductors, developing remote doping strategies, and optimising the interface between oxide dielectrics and ATB TMDs to achieve high-performance p-type FETs. The proposal builds on my previous breakthroughs in realizing ultra-clean vdW contacts (Nature, 2019, 2022) on ATB semiconductors for high-performance FETs. It also builds on my recent work on ultra-clean interface between ATB semiconductors and CMOS compatible dielectrics (Nature Electronics, under revision). Efficient hole injection into atomically thin semiconductors will also facilitate the development of other novel energy-efficient electronic and optoelectronic devices such as tunnel field effect transistors, valleytronics, and light emitting diodes. Outcomes of the proposed work will accelerate advanced semiconductor development in application areas that are strategically important to the UK.