E-beam Lithography Process Research Articles

Topological Insulator Bismuth-Antimony Alloy Contact for Two-Dimensional Semiconductor Based Electronics

Two-dimensional (2D) semiconductors are considered as the most promising candidate for channel materials in next-generation electronic devices owing to their atomically thin nature [1]. In order to minimize the overall chip size, it is imperative to reduce both the channel length and the contact area of the devices. However, the reduction in contact area inevitably leads to an increase in the contact resistance (RC), and recent studies have been devoted to solving this problem. In particular, the significant achievements in reducing the RC to near quantum limits in 2D semiconductor based field-effect transistors (FET) were reported using semi-metal as bismuth (Bi) and antimony (Sb) as contact metals [2,3]. In spite of these advancements, research on the scaling-down of the metal contact with the exploration of alternative contact materials to shrink overall chip footprint remains limited.To overcome these limitations, we propose the topological insulator (TI) as a contact metal for 2D-based electronics. TI is known to enable electron transport without backscattering at the surfaces owing to its unique surface band structures [4]. Result from these properties, the RC increase with a contact area reduction is lower compared to conventional metals, indicating high potential for TIs as an interconnector in extreme scaling regimes [5]. Additionally, the Dirac cone band structure on TIs surface [6] is expected to suppress metal-induced gap states (MIGS) similar to semi-metal contacts [2-3].In this study, a 2D semiconductor FETs was fabricated with Bi-Sb alloy as the electrode material. Bi-Sb alloys are known to exhibit the TI phase with Sb composition ranging from 7% to 22% [7]. A 20 nm thick Ti-phase Bi-Sb alloy was synthesized using molecular beam epitaxy (MBE) on exfoliated MoS2 flakes with pre-patterned contact areas using a standard e-beam lithography process. The device was then completed by capping the TI contacts with gold. A systematic investigation of the electrical properties of the device showed that the on-current with the contact of the Ti-phase Bi-Sb alloy is enhanced compared to the bismuth-only configuration. Through this research, we aim to improve the contact characteristic of 2D based electronics in nano-scale regimes, and provide insights that can contribute to the development and commercialization of next-generation electronic devices.

Room-temperature infrared photoluminescence and broadband photodetection characteristics of Ge/GeSi islands on silicon-on-insulator

In this study, molecular beam epitaxial growth of strain-driven three-dimensional self-assembled Ge/GeSi islands on silicon-on-insulator (SOI) substrates, along with their optical and photodetection characteristics, have been demonstrated. The as-grown islands exhibit a bimodal size distribution, consisting of both Ge and GeSi alloy islands, and show significant photoluminescence (PL) emission at room temperature, specifically near optical communication wavelengths. Additionally, these samples were used to fabricate a Ge/GeSi islands/Si nanowire based phototransistor using a typical e-beam lithography process. The fabricated device exhibited broadband photoresponse characteristics, spanning a wide wavelength range (300-1600 nm) coupled with superior photodetection characteristics and relatively low dark current (∼ tens of pA). The remarkable photoresponsivity of the fabricated device, with a peak value of ∼11.4 A W-1(λ∼ 900 nm) in the near-infrared region and ∼1.36 A W-1(λ∼ 1500 nm) in the short-wave infrared (SWIR) region, is a direct result of the photoconductive gain exceeding unity. The room-temperature optical emission and outstanding photodetection performance, covering a wide spectral range from the visible to the SWIR region, showcased by the single layer of Ge/GeSi islands on SOI substrate, highlight their potential towards advanced applications in broadband infrared Si-photonics and imaging. These capabilities make them highly promising for cutting-edge applications compatible with complementary metal-oxide-semiconductor technology.

Optimizing Josephson Junction Reproducibility in 30 kV E-Beam Lithography: An Analysis of Backscattered Electron Distribution.

This paper explores methods to enhance the reproducibility of Josephson junctions, which are crucial elements in superconducting quantum technologies, when employing the Dolan technique in 30 kV e-beam processes. The study explores the influence of dose distribution along the bridge area on reproducibility, addressing challenges related to fabrication sensitivity. Experimental methods include e-beam lithography, with electron trajectory simulations shedding light on the behavior of backscattered electrons. Wedescribe the fabrication of various Josephson junction geometries and analyze the correlation between the success rates of different lithography patterns and the simulated distribution of backscattered electrons. Our findings demonstrate a success rate of up to 96.3% for the double-resist 1-step low-energy e-beam lithography process. As a means of implementation strategy, we provide a geometric example that takes advantage of simulated stability regions to administer a controlled, uniform dose across the junction area, introducing novel features to overcome the difficulties associated with fabricating bridge-like structures.

Inverse-designed broadband low-loss grating coupler on thick lithium-niobate-on-insulator platform

A grating coupler on 700-nm-thick Z-cut lithium-niobate-on-insulator platform with high coupling efficiency, large bandwidth, and high fabrication tolerance is designed and optimized by inverse design method. The optimized grating coupler is fabricated with a single set of e-beam lithography and etching process, and it is experimentally characterized to possess peak coupling efficiency of −3.8 dB at 1574.93 nm, 1 dB bandwidth of 71.7 nm, and 3 dB bandwidth of over 120 nm, respectively.

Impact of Parasitic Gate Capacitance on RF Performance in GaN-Based HEMTs for X-Band Applications

GaN-based high-electron-mobility transistors (HEMTs) are intensively researched for high-power radio frequency (RF), low noise, and aerospace applications, thanks to their high breakdown electric field, high carrier mobility and density at the hetero-interface, and wide bandgap [1-2]. For the improvement of RF performance, various approaches have been reported, including T-gate structure [3-4], surface passivation [5], n+-regrown source and drain contact [6], thin barrier structure [7], and graded-channel [8]. Among them, the T-gate structure has been extensively employed for the reduction of the gate resistance (Rg ) and thereby improvement of the RF performance. However, the usage of the wide T-gate structure enlarges parasitic gate capacitance components, which deteriorates the RF performance. As a result, one should pay careful attention to the dimension of the wide T-gate structure to reduce Rg as much as possible, not to enlarge the parasitic gate capacitance components. This motivates us to explore the optimum dimension of the wide T-gate structure in GaN-based HEMTs for X-band RF applications.To investigate the detailed impact of the gate resistance and parasitic gate capacitance components, we fabricated LG = 0.15 μm GaN-based HEMTs with various dimensions of the T-gate head from 0.5 to 0.8 μm during an e-beam lithography process step. The unit gate width (WG ) and number of gate finger (NF ) were 100 μm and 2, respectively.For comparison of the device characteristics with respect to the dimension of the gate head, DC and RF measurements were conducted for the fabricated GaN-based HEMTs with various dimensions of T-gate head. The DC characteristics in terms of the threshold voltage (VTH ), the maximum transconductance (gm,max ), and the saturation drain current (Id,sat ) were almost identical, regardless of the dimension of the T-gate head. On the contrary, the current-gain cut-off frequency (fT ) and maximum oscillation frequency (fMAX ) were improved by 26.5% and 21.0%, respectively, when the T-gate head size was reduced from 0.8 to 0.5 μm.In an effort of explore the origin of the dependence of the T-gate head on the RF performance, we have carried out small-signal modeling to extract the parasitic gate capacitance and gate resistance components from the measured RF data. As the T-gate head size was reduced from 0.8 to 0.5 μm, the gate-source parasitic capacitance component (Cgs_par ) was reduced by 19%, while the gate resistance component was increased by 32%. In our device design and fabrication, it turned out that the reduction of Cgs_par was more beneficial than the increase of Rg , leading to the improved RF performance with the gate head size decrease.In this work, our systematic research revealed that, unlike earlier reports [3-4], the reduction of the parasitic gate capacitance was essential for the improvement of the RF performance as opposed to the reduction of the gate resistance in LG = 0.15 μm GaN-based HEMTs for X-band RF applications. Acknowledgement This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MIST) (No. NRF-2021M3C1C3097672).

Analytic study of exposure contrast over feature edge in electron-beam lithography

In electron-beam (e-beam) lithography, the location of a feature edge may vary with experiment due to the stochastic nature of the e-beam exposure and resist-development processes. From the viewpoint of consistent reproducibility of a circuit pattern, it is essential to enhance the stability of a feature edge in the e-beam lithographic process. A fundamental metric affecting the stability is the exposure contrast over the feature edge, and therefore, it is important to understand the dependency of exposure contrast on significant parameters. However, the computer simulation for the dependency analysis is time-consuming and needs to be repeated. In this study, a new method has been developed by deriving closed-form mathematical expressions of exposure contrast for the cases of a single feature and a uniform pattern of multiple features. The mathematical expressions enable the fast analysis of exposure contrast without simulation, and therefore, can serve as a useful tool in e-beam lithography.

A broadband high-brightness quantum-dot double solid immersion lens single photon source

High-brightness single photon sources (SPSs) are key components for practical quantum information processing systems. Although the performances of recently reported high-brightness SPSs are excellent, it remains challenging to match the emission wavelength of a quantum dot (QD) to the cavity since the high-Q cavity structures have narrow spectral bandwidths. Here, we propose a highly bright and broadband QD SPS that can be deterministically fabricated with a simple yet precise method. The optimized GaAs-polymer double solid immersion lens structure is capable of a brightness of 88% at 0.5 NA and has an operation band of 65 nm with a brightness of over 80% from numerical simulations. Experimentally, we achieved a brightness of 51.6% ± 2% and pure single photon emission [g(2)(0) = 0.029 ± 0.005] at saturation. We believe that our result can pave the way to a practical high-brightness QD SPS, considering its simple QD geometry together with its low cost and precise deterministic fabrication without using expensive and complicated e-beam lithography and dry etching processes.

Comprehensive modeling of the extrinsic transconductance for InxGa1-xAs/In0.52Al0.48As quantum-well high-electron-mobility transistors

This paper presents the comprehensive modeling of extrinsic transconductance (gm_ext) in the saturation regime for InxGa1-xAs/In0.52Al0.48As quantum-well (QW) high-electron-mobility transistors (HEMTs) with Lg from 10 μm to sub-100 nm, covering the ballistic and mobility relevant regimes. First, we fabricated and characterized two types of lattice matched (LM) channel In0.53Ga0.47As QW HEMTs on a 3-inch InP substrate. One of these was fully fabricated with an i-line stepper, yielding Lg from 10 μm to 0.5 μm, and the other was fabricated in a mix & match manner with a combination of an i-line stepper and an e-beam lithography equipment, where T-shaped gates were implemented using a tri-layer resist stack of ZEP520A-PMGI-ZEP520A in the e-beam lithography process. Using devices with a wide range of Lg values, we examined the transconductance (gm) characteristics in saturation and attempted to correlate them to analytical expressions capable of describing the carrier transport properties of the devices using only physical parameters, such as the gate capacitance (Cgi) apparent mobility (μn_app), and saturation velocity (vsat), and the series resistances (Rs and Rd). In this way, we assessed the dependences of the extrinsic transconductances of the fabricated devices on their gate lengths, with the model parameters of Cgi = 0.65 μF/cm2, μn_app = 9400 cm2/V⋅s, vsat = 4.3 × 107 cm/s and Rs = 147 Ω⋅μm. These values were observed to be well-matched with those of the physical parameters. Most importantly, we extended the proposed approach to realistically project the increase in gm_ext that could be accomplished by improving certain device parameters.

Nanoampere‐Level Piezoelectric Energy Harvesting Performance of Lithography‐Free Centimeter‐Scale MoS2 Monolayer Film Generators (Small 24/2022)

Energy Harvesting In article number 2200184, Yong Soo Cho and co-workers successfully apply an interdigitated electrode for a centimeter-scale monolayer MoS2 film having preferred domain structure, without a complicated etching or e-beam lithography process. The fabricated device exhibits superior piezoelectric energy harvesting performance with 400.4 mV and 40.7 nA.

Creating hot spots within air for better sensitivity through design of oblique-wire-bundle metamaterial perfect absorbers

Better sensitivity of a biosensor could boost up the detection limit of analytes, thus a must in the fields of bio-sensing and bio-detection. To further enhance the sensitivity of a biosensor, in this work, we design an oblique-flat-sheet metamaterial perfect absorber (MPA) to concentrate the hot spots within air between the oblique flat sheet and the continuous ground metal, thus enabling fully interaction between analytes and hot spots. The corresponding field distributions in simulation corroborated our assumption and its sensitivity could be up to 1049 nm/RIU. Then, we fabricated the sample by e-beam lithography process for a seed layer and simply tilting the sample during deposition to obtain oblique flat sheets. When considering the stochastic nature of the deposited multiple oblique flat sheets, we modified the metallic upper resonator of the MPA from the single oblique-flat-sheet into randomly distributed oblique-wire-bundle (OWB) and in simulation, its sensitivity is boosted up to 3319 nm/RIU. In experiments, the measured sensitivity is 1329 nm/RIU under different concentrations of glucose solutions that is four times larger than the 330 nm/RIU of the planar MPA. The higher sensitivity was attributed to that the OWB MPA could provide hot spots within air not only between OWB and grounded metal but also among wires. Moreover, the OWB could also trap and concentrate the analytes locally.

Grating couplers beyond silicon TPA wavelengths based on MPW

Silicon photonic integrated circuits (PICs) at 2.2–2.5 μm wavelengths have great potential in applications of nonlinear optics, spectroscopy, and light detection and ranging, due to the merits of negligible two-photon absorption and moderate Rayleigh scattering from the waveguide sidewall roughness. As key components in silicon PICs, grating couplers at such wavelengths have been previously studied. However, previous devices were mainly developed based on the e-beam lithography (EBL) process, being challenging for high-volume and low-cost fabrication. Here, we demonstrated focusing subwavelength grating couplers at the spectral range of 2.2–2.5 μm based on a multi-project wafer (MPW) service. A mean coupling efficiency of −7.77 dB with a standard deviation of 0.5 dB at a center wavelength of 2.36 μm and a mean 3 dB bandwidth of 85 nm was experimentally achieved. Moreover, we studied the reproducibility of the fabricated grating couplers from die to die, as well as the grating design and fiber coupling tolerance. Our study is expected to provide a useful guideline for developing short-wavelength mid-IR devices based on MPW.

Estimation of critical dimension and line edge roughness using a neural network

While electron-beam (e-beam) lithography is widely employed in the pattern transfer, the proximity effect makes features blurred, and the stochastic nature of the exposure and development processes causes the roughness in the feature boundaries. In an effort to reduce the proximity effect and line edge roughness (LER), it is often necessary to estimate the critical dimension (CD) and LER. In our previous study, the e-beam lithographic process was modeled using the information extracted from SEM images for the estimation of CD and LER. This modeling involves several parameters to be determined and tends to require a long computation time. In this study, the possibility of improving the accuracy of the CD and LER estimation using a neural network (NN) is investigated. In the NN-based estimation, the explicit modeling of the e-beam lithographic process can be avoided. This paper describes the method of estimating the CD and LER using a NN, including the issues of training, tuning, and sample reduction and presents results obtained through an extensive simulation.

How to solve problems in micro- and nanofabrication caused by the emission of electrons and charged metal atoms during e-beam evaporation

We discuss how the emission of electrons and ions during electron-beam-induced physical vapor deposition can cause problems in micro- and nanofabrication processes. After giving a short overview of different types of radiation emitted from an electron-beam (e-beam) evaporator and how the amount of radiation depends on different deposition parameters and conditions, we highlight two phenomena in more detail: First, we discuss an unintentional shadow evaporation beneath the undercut of a resist layer caused by the one part of the metal vapor which got ionized by electron-impact ionization. These ions first lead to an unintentional build-up of charges on the sample, which in turn results in an electrostatic deflection of subsequently incoming ionized metal atoms toward the undercut of the resist. Second, we show how low-energy secondary electrons during the metallization process can cause cross-linking, blisters, and bubbles in the respective resist layer used for defining micro- and nanostructures in an e-beam lithography process. After the metal deposition, the cross-linked resist may lead to significant problems in the lift-off process and causes leftover residues on the device. We provide a troubleshooting guide on how to minimize these effects, which e.g. includes the correct alignment of the e-beam, the avoidance of contaminations in the crucible and, most importantly, the installation of deflector electrodes within the evaporation chamber.

3D modeling of electron-beam lithographic process from scanning electron microscope images

Computational lithography is typically based on a model representing the lithographic process where a typical model consists of three components, i.e., line spread function, conversion formula (exposure-to-developing rate conversion), and noise process (exposure fluctuation). In our previous study, a practical approach to modeling the e-beam lithographic process by deriving the three components directly from SEM images was proposed. However, a 2D model of a substrate system was employed; i.e., the exposure variation along the resist-depth dimension was not considered. In this study, the possibility of improving the accuracy of modeling using a 3D model is investigated. The 3D model is iteratively determined by modeling the critical dimension estimated based on the model to those measured in SEM images. This paper describes the 3D modeling approach and new optimization procedures and discusses in detail the results from an extensive simulation for an accuracy analysis of the 3D modeling approach.

Controlled synthesis of periodic arrays of ZnO nanostructures combining e-beam lithography and solution-based processes leveraged by micro X-ray fluorescence spectroscopy

In this work, a comprehensive study on how to combine purely chemical, bottom-up solution-based synthesis with advanced e-beam lithography for the controllable production of ZnO-nanostructure periodic arrays on Si and SiO 2 substrates is presented. The study was complemented with micro-X-ray fluorescence spectroscopy demonstrating the latter's great potential as a rapid, non-destructive and enabling nanometrology technique that can leverage the development of novel nanofabrication processes. The study resulted in a fabrication process framework that not only can help in relating the process parameters to the desired geometrical feature of the periodic nanoarchitectures, but that is also compatible with standard microfabrication techniques and large-scale production. • Controllable fabrication of periodic arrays of ZnO nanostructures • Successful merging of low-cost hydrothermal technique with e-beam lithography for the fabrication of periodic ZnO nanostructures • Use of micro-XRF as a non-destructive nanometrology tool leveraging the definition of the process window • Clear definition of process window for cost-efficient production of custom-designed ZnO nanostructure arrays

Efficient Large-Scale Removal of Poly(methyl methacrylate) Layers by Using a Methyl Isobutyl Ketone Solution

In this study, we proposed a fast and effective technique to remove residual poly(methyl methacrylate) (PMMA) after its use as a supporting layer for transferring graphene films on a large scale with high productivity. We utilized a methyl isobutyl ketone (MIBK) solution, which is commonly used to develop PMMA layers, for an e-beam lithography process. This was done because MIBK molecules tend to inflate the PMMA layer, weakening the link between PMMA and the graphene surface. Field-effect transistors were fabricated on the transferred graphene, with which we addressed the Dirac points and their carrier motilities. We observed that due to the reduced p-doping effects, the Dirac points were located closer to the zero-gate bias compared to those obtained using the acetone treatment. Finally, the average device mobility reached 5400 and 3900 cm2/Vs for holes and electrons, respectively. These values are more than five times the values obtained using the conventional acetone treatment.

Broadband lumped-element Josephson parametric amplifier with single-step lithography

We present a lumped-element Josephson parametric amplifier (JPA) fabricated using a straightforward e-beam lithography process. Our strongly coupled flux-pumped JPA achieves a gain of 20 dB with a bandwidth of 95 MHz around 5 GHz, while the center frequency is tunable by more than 1 GHz, with the additional possibility for rapid tuning by varying the pump frequency alone. Analytical calculations based on the input-output theory reproduce our measurement results closely.

High density GaAs nanowire arrays through substrate processing engineering

GaAs nanowires (NWs) vertically aligned were successfully fabricated through substrate processing engineering. High-density vertical GaAs NWs are grown on n-type Si (111) substrate by molecular beam epitaxy. Systematic experiments indicate that substrate pretreatment is crucial to vertical epitaxial growth of one-dimensional (1D) nanomaterials. The substrates etched using diluted buffered oxide etch (BOE) were explored to improve the NW density and vertical. We also find that the substrate processing engineering strongly affect the morphology of GaAs NWs. Finally, we demonstrate fabrication of GaAs NW arrays on Si surface by field-emission scanning electron microscopy (FE-SEM). This single-step process indeed offers a simple and cost-effective way to obtain a large area of GaAs NW arrays without using e-beam lithography (EBL) and/or nanoimprint lithography (NIL) processes. This work provided a new approach for hight density NW arrays.

Simulation and experimental study on developed profiles in the positive polymer resist PMMA

Theoretical and experimental study on developed profiles in the positive tone PMMA (polymethyl-methacrylate) resist and e-beam lithography process parameters was performed. E-beam lithography control system Elphy Quantum (Raith) installed on Scanning Electron Microscope Quanta FEG (FEI) with a field emission cathode and Gaussian intensity distribution has been used for the conducted experiments. Simulation results obtained by different models concerning the geometry of the developed resist profiles in PMMA were compared and verified with experimental data for improvement characterization of PMMA resist. The estimated models presented are useful for prediction and optimization of the developed resist profiles at electron beam lithography.

Fabrication of Broadband Antireflective Sub-Micro Structures on 4H-SiC by Mesh Patterning Etching

An approach to fabricating pseudoperiodic antireflective subwavelength structures on silicon carbide by using modified resist as etching mask is demonstrated. The etched submicron structure is a mesh-trenched pattern thus as the so called “mesh patterning etching”. The mesh patterning process is more time-efficient than the e-beam lithography or nanoimprint lithography process. The influences of the reactive-ion etching conditions and post bake temperature of the resist film to the sub-micron structure profile and its corresponding surface reflectance have been systematically investigated. Under optimal experimental conditions, the average reflectance of the silicon carbide in the range of 390-800 nm is dramatically suppressed from 40% to around 8% after introducing the mesh patterning sub-micro structures.