Theoretical Design Of Two-Dimensional Silicon Semiconductors And Nanoscale Field Effect Transistors

The continuous scaling of transistor size has brought great improvement in chip speed and integration.Currently,conventional silicon-based integrated circuit technology has entered the sub-5 nm node,and the size and performance of bulk Si-based field-effect transistors（FETs）are approaching their intrinsic physical limits.To further improve device performance and chip integration,it is important to explore and design novel semiconductor materials for the future development of integrated circuits.Benefiting from the characteristics of atomic ultra-thin thickness,smooth and uniform surface,and facilitate stress-regulation,two-dimensional（2D）materials have shown great potential for application in high-performance FETs.While ushering in significant development opportunities,the study of 2D FETs is also facing some significant challenges.On the one hand,while Si-based 2D materials represent ideal candidates replacing conventional bulk Si because of their inherent compatibility with the current semiconductor technology,the currently available Si-based 2D materials,such as intrinsic silicene,exhibits graphene-like semimetallic properties unsuitable for direct application in semiconductor devices.As such,it is important to design novel semiconducting 2D Si and to explore their potential application in 2D FETs.On the other hand,for the lack of proper doping technology for 2D materials,metal electrodes are usuall employed to directily contact with 2D materials in actual 2D FETs,leading to complex contact issues at the interfaces between metal electrodes and 2D semiconductors,such as Fermi level pinning（FLP）and Schottky barriers,which represent a bottleneck restricting the performance of 2D FETs.Therefore,it is also of significant importance to explore the interfacial contacts of 2D Si with different metal electrodes when designing Si-based 2D FETs.Targeting the issues raised above,this dissertation performs thorough investigations on the theoretical design of 2D Si-based semiconductors and nanoscale FETs,which include the following three topics:（1）Theoretical design of novel 2D Si semiconductors.By focusing on the lattice topology,we designed three 2D Si semiconductor materials,i.e.,monolayer hybrid honeycomb-kagome silicon（hhk-Si）,monolayer hybrid kagome-dumbbell silicon（hkd-Si）,and thickness-dependent kagome lattice silicon（KL-Si）.Based on energetic calculations,phonon analysis,and ab initio molecular dynamics（AIMD）simulations,the three proposed novel 2D Si all exhibit great geometric stability.More importantly,the first-principles calculations prove the intrinsic semiconducting properties of the three 2D-Si materials,with band gaps of 1.18 eV,0.80 eV,and 1.36～2.82 eV（depending on thickness）,respectively.In particular,in framework of full phonon scattering,the electron mobility of the hkd-Si is predicted to be as high as 6.9 × 10³ cm²·V^-1·s^-1 at room temperature（300 K）,which is higher than that of all other Si and most nonSi 2D semiconductors reported in literature.By suppressing the out-of-plane lattice vibrations,the electron mobility of the hhk-Si can be as high as 10³ cm²·V^-1·s^-1.In addition,the hole and electron mobilities of the KL-Si gradually increase and decrease with increasing the thickness,respectively,and the results based on the deformation potential method are 813～3455 cm²·V^-1·s^-1 and 63～1260 cm²·V^-1·s^-1,respectively.（2）Theoretical design of 2D Si-based nanoscale FETs.Based on the three proposed 2D-Si semiconductors,we theoretically designed a set of nanoscale 2D FETs,of which the intrinsic transport properties was modeled by means of density functional theory in combination with the non-equilibrium Green’s function method.At 5-nm channel length,all the hhk-Si FETs,hkd-Si FETs,and KL-Si pFETs can meet the requirements of the International Technology Roadmap for Semiconductors（ITRS）for high-performance（HP）devices（in terms of on current,delay time,and power consumption）.In particular,in the case of the hhk-Si FETs,benchmarking with the ITRS standard,the gate length can be scaled down to 2 nm（pFET）and 3 nm（nFET）,respectively,and the scaling performance of the devices is better than most nonSi 2D FETs.In addition,the output characteristics of both the hhk-Si FETs and KL-Si FETs show a phenomenon of giant negative differential resistance,and the current peak-valley ratio is as high as over 106,as a result of serious mismatch of the density of states（DOS）between the source and drain electrodes under bias.Moreover,benefiting from the relatively isolated conduction band and valence band of the hhk-Si,the hhk-Si FETs can break the thermal limit suffered by conventional FETs and achieve an ultra-steep subthreshold swing（SS）of sub-60 mV/dec at room temperature.By comparing the nanoscale FETs based on more than 20 typical Si and non-Si 2D semiconductors,the 2D-Si FETs are very competitive in terms of intrinsic transport properties,which results are useful for the design of next-generation 2D nanoscale FETs.（3）Regulation of the electrical contact interfaces in the 2D Si-based nanoscale FETs.Based on the Schottky barrier FET model,we studied the interfacial properties between the intrinsic hhk-Si and LHD-Si（large honeycomb dumbbell silicene）and different metal electrodes,and simulated the impacts of metal electrodes on the device performance.In the case of the hhk-Si FETs,the Fermi levels of the studied metals（Ti/Ag/Au/Cu/Pt）are always pinned in the band gap,resulting in a Schottky barrier that is difficult to eliminate.Due to the unique band structures of the hhk-Si,the hhk-Si FETs with the Ti/Ag/Au electrodes can achieve the ultra-steep SS of sub-60 mV/dec,leading to the On/Off current ratio over 105.In the case of the LHD-Si FETs,the Fermi levels of the Al/Ag/Cu/Bi/NbTe₂/TaTe₂ electrodes are always pinned near the conduction band,thus leading to n-type Ohmic contact and good n-type FET characteristics,while obvious p-type barriers and poor p-type FET performance were also obtained.Besides,by utilizing the "cold metal" band features of the NbTe₂/TaTe₂,the LHD-Si nFETs can also achieve ultra-steep SS values of sub-60 mV/dec.Furthermore,we also systematically studied the effective modulation of surface passivation on the interfacial issues in the LHD-Si FETs.It is revealed that the F-/H-/OH-surface passivation can significantly suppress the FLP in the LHD-Si FETs,and the pinning factor can be improved from 0.08 to 0.28～0.46.Besides,surface passivation can effectively regulate the contact type（n-type or ptype）between LHD-Si and metal electrodes.The F-,OH-and H-passivated LHD-Si FETs exhibit n-type,p-type,and ambipolar transfer characteristics,respectively,which is equivalent to a "surface doping technology".In summary,by using the first-principles modeling method of density functional theory combined with non-equilibrium Green’s function,this dissertation performs in-depth and systematic investigations on the theoretical design of 2D Si semiconductors and FETs from both materials and devices perspectives.The studies of this dissertation provide theoretical guidelines for the design of 2D Si semiconductors and nanoscale FETs,promoting the research and development of novel nanoelectronic devices.