Table of Contents
Fetching ...

Enhancing Hole Mobility in Monolayer $WSe_{2}$ p-FETs via Process-Induced Compression

He Lin Zhao, Sheikh Mohd Ta-Seen Afrid, Dongyoung Yoon, Zachary Martin, Zakaria Islam, Sihan Chen, Yue Zhang, Pinshane Y. Huang, Shaloo Rakheja, Arend M. van der Zande

Abstract

Understanding the interactions between strain, interfacial mechanics, and electrical performance is critical for designing beyond silicon electronics based on hetero-integrated 2D materials. Through combined experiment and simulation, we demonstrated and analyzed the enhancement of hole mobility in p-type monolayer $WSe_{2}$ field effect transistors (FETs) under biaxial compression. We tracked FET performance versus strain by incrementing compressive strain to $WSe_{2}$ channels via sequential AlOx deposition and performing intermediate photoluminescence and transport measurements. The hole mobility factor increased at a rate of 340 $\pm$ 95 %/%$ε$, and the on-current factor increased at a rate of 460 $\pm$ 340 %/%$ε$. Simulation revealed that the enhancement under compression arises primarily from a reduction in inter-valley scattering between the $Γ$--K valence bands, and the rate is robust against variations in carrier density, impurity density, or dielectric environment. These findings show that compressive strain is a powerful technique for enhancing performance in 2D p-FETs and that it is multiplicative with defect and doping engineering.

Enhancing Hole Mobility in Monolayer $WSe_{2}$ p-FETs via Process-Induced Compression

Abstract

Understanding the interactions between strain, interfacial mechanics, and electrical performance is critical for designing beyond silicon electronics based on hetero-integrated 2D materials. Through combined experiment and simulation, we demonstrated and analyzed the enhancement of hole mobility in p-type monolayer field effect transistors (FETs) under biaxial compression. We tracked FET performance versus strain by incrementing compressive strain to channels via sequential AlOx deposition and performing intermediate photoluminescence and transport measurements. The hole mobility factor increased at a rate of 340 95 %/%, and the on-current factor increased at a rate of 460 340 %/%. Simulation revealed that the enhancement under compression arises primarily from a reduction in inter-valley scattering between the --K valence bands, and the rate is robust against variations in carrier density, impurity density, or dielectric environment. These findings show that compressive strain is a powerful technique for enhancing performance in 2D p-FETs and that it is multiplicative with defect and doping engineering.

Paper Structure

This paper contains 21 sections, 12 figures.

Figures (12)

  • Figure 1: Process induced strain engineering in monolayer WSe2 p-FETs.a, Concept illustration of enhanced hole mobility in monolayer WSe2 induced by biaxial compressive strain. b, Exploded render of the important layers and experimental design in the process-strained WSe2 FET. c, Optical image of the p-FET array. d, a cross-sectional STEM at the contact at the position marked in (c). Each layer is annotated. e, Representative Photoluminescence (PL) spectra versus energy from a single WSe2 channel. Grey and black spectra come from the channel after TOS doping, and 6 nm SiOx passivation. Shades of blue represent the spectra after incremental stressor depositions. f, PL peak energy (left) and estimated biaxial compression e(right), as a function of AlOx thickness. Error bars represent spatially aggregated quartile statistics across 5 FET channels.
  • Figure 2: Spatially resolved strain analysis.a, hyperspectral maps photoluminescence peak position of one FET channel throughout doping, passivation, and 3 incremental stressor depositions. Brightness corresponds to peak intensity, and color bar hue corresponds to peak energy. b, top and side illustration showing the setup and dimensions of the finite-element models and maps of the simulated $\varepsilon_{xx}$$\varepsilon_{yy}$ strain tensor components. c, Example PL maps of two damaged FET channels at 90 nm stressor thickness. The cracked channel shows sharp non-discontinuities, while the edge delamination shows PL amplitude dimming and peak energy inversion at the edge. d, Cross-section STEM image showing the edge of the channel in a damaged device, showing prominent delamination.
  • Figure 3: Strain-enhancement of WSe2 p-FET performance.a, Current density ($I_\text{ds}$) versus overdrive voltage ($V_\text{OD}$) transfer characteristics of a representative WSe2 p-FET for increasing stressor thicknesses (blue shades). The left and right axes show current density in linear and log scales. b, Corresponding Current density versus drain voltage $V_{ds}$ output characteristics at constant VOD for increasing stressor thicknesses. The inset shows output characteristics for different VOD at a single 60 nm AlOx thickness. c, Plot of measured (green points) and computed (green band) field-effect mobility factor $(\mu/\mu_0)_\text{FE}$ versus compressive (negative) biaxial strain. The mobility factor tracks the change in mobility relative to the initial mobility in each p-FET. Green points show averaged measured mobility factor across 12 WSe2 p-FETs, with errors representing quartile statistics in both mobility and strain uncertainty. The green band represents the variability in computed mobility rising from uncertainty in the assumed experimental parameters. The mean initial mobility $\mu_\text{0,FE} =$ 3.5±1cm2/V· s. The mobility factor strain-tuning rate is 340%/%e. d, Plots of corresponding quartile statistics for the VT shift ($V_\text{T}-V_\text{T,0}$) and on-current factor ($I_\text{on}/I_\text{on,0}$) versus compressive biaxial strain. VT,0 = --30 V, Ion,0 = 0.4±0.16 μA/μm. The VT shift rate is 50 V/%e and the on-current factor strain-tuning rate is 460 %/%e.
  • Figure 4: Strain-induced hole mobility enhancement from modifying inter-valley scattering.a, Computed valence band structure along the G--K direction of monolayer WSe2 under varying biaxial strain. Shown in the color bar on the right, in all panels, compressive, unstrained, and tensile strain regimes are represented by blue, black, and red, respectively. b, Energy of local valence band maxima at the G (square) and K (circle) valleys versus biaxial strain. c, Left, intrinsic scattering rates versus energy under varying strain, including both intra-valley and inter-valley scattering; right, intrinsic scattering rates versus biaxial strain at an energy of 200 meV comparing the relative intrinsic scattering rate without (cross) and with (circle) inter-valley scattering allowed. d, Computed mobility factor versus biaxial strain, comparing the intrinsic mobility (circle) and intrinsic + extrinsic mobility (square). The inset shows the relative computed initial mobility. Fig. \ref{['fig:illusts']}c--d assume $T$ = 300 K, $p = 5 \times 10^{12}\,\text{cm}^{-2}$, $n_\text{imp} = 2.5 \times 10^{12}\,\text{cm}^{-2}$, with SiO$_2$ as the dielectric environment.
  • Figure 5: Parametric analysis of strain-modulated hole mobility enhancement in WSe2.a, Mobility factor versus biaxial compressive strain for different impurity concentrations. b-c, Parametric analysis versus carrier concentration $(p)$, impurity concentration $(n_\mathrm{imp})$, and different dielectric environments of b, unstrained initial hole mobility $\mu_0$ (green) and c, mobility factor strain-tuning rate $\frac{\partial(\mu/\mu_0)}{\partial \varepsilon}$ (1/%e) and evaluated at --0.25% compressive strain. Color represents different strain types: biaxial (black), uniaxial armchair (purple), and uniaxial zigzag (blue). In all plots, unless specifically being varied, analyses assume $T$ = 300 K, $p = 10^{13}$ cm$^{-2}$, $n_{\text{imp}} = 2.5 \times 10^{12}$ cm$^{-2}$, and a SiO$_2$ dielectric.
  • ...and 7 more figures