Channel Layer Research Articles

Enhancement Mode Ga2O3 Field Effect Transistor with Local Thinning Channel Layer

β−Ga2O3 field−effect transistors (FETs) were fabricated with and without local thinning to change the threshold voltage. A 220 nm Ga2O3 layer was mechanically exfoliated from a Cr−doped gallium oxide single crystal. Approximately 45 nm Ga2O3 was etched by inductively coupled plasma to form the local thinning. The threshold voltage of the device with etched local thinning increased from −3 V to +7 V compared to the unetched device. The effect of the local thinning was analyzed by device simulation, confirming that the local thinning structure is an effective method to enable enhancement−mode Ga2O3 FETs.

Device feasibility and performance improvement methodologies for thin film transistors using In-Ga-Sn-O channels prepared by atomic-layer deposition

Thin film transistors (TFTs) using In-Ga-Sn-O (IGTO) active channel layers, which were prepared by atomic-layer deposition (ALD) techniques, were fabricated and characterized with exploring the optimum IGTO channel compositions and the process conditions for the formation of gate-stack structures. The introduction of an O3 oxidant was confirmed to secure a wider process window, which was explained by hydrogen incorporation from the protection layer/gate insulator stacked layers into the IGTO active channel. The cationic compositions of the IGTO thin films were controlled by subcyclic ratio design of each precursor. When the number of In-Ga precursor subcycles increased from 4 to 6, the Ga/Sn ratio increased from 3.5 to 4.0, leading to marked improvements in the VON characteristics of the fabricated IGTO TFTs. It was found from a comparative analysis on the device characteristics that the Ga/Sn ratio in the IGTO channel had a more dominant effect on the VON characteristics rather than the In/Ga ratio. The device using the channel with a cationic composition of 5.8:4.0:1.0 (In:Ga:Sn) exhibited the field-effect mobility of 19.1 cm2/V s, the subthreshold swing of 0.22 V/dec, and the threshold voltage (VTH) of 0.22 V, which corresponded to the best device characteristics among the fabricated devices. Furthermore, the VTH shifts were examined to be +0.4 and −0.4 V in positive and negative bias stress conditions, respectively, demonstrating excellent gate-bias stress stabilities. The obtained results suggest the device feasibility and fabrication process validity for the TFT applications using IGTO active channel layers prepared by ALD techniques.

Effects of the stepped sidewall morphology on the ON-state performance for vertical GaN trench-gate MOSFETs

Vertical GaN trench-gate MOSFETs with ∼130 nm stepped sidewalls in the p-GaN channel layer are studied and two significant influences have been observed compared to the devices with smooth sidewalls. The first effect is the degraded channel mobility of 14.6 cm2 (V·s)−1 to 4.5 cm2 (V·s)−1 which can be attributed to the increased probability of surface acoustic phonons scattering under inversion conditions. Another impact is that only when a drain voltage (V DS) is applied over 30 V can the devices switch on. It can be speculated that the stepped sidewall has a horizontal channel and the transverse electric field is inadequate to drive the electrons at low V DS. According to the investigation, when devices need to be switched at a high V DS, the stepped sidewall morphology should be taken into consideration.

Numerical analysis on polymer electrolyte membrane fuel cell performance enhancement with novel selective-patterned gas diffusion layers

To improve the performance of polymer membrane electrolyte fuel cells (PEMFCs), many studies dealing with flow fields, including those related to the channel and gas diffusion layer (GDL), have been conducted. Especially, the GDL, which is an important component that determines the PEMFC performance, has been used as a diffusion media which covers all reaction sites between the channel and catalyst layer. The common rectangular shape of the GDL enables the diffusion of the reactant under the rib. However, it results in the deterioration of mass transport under the channel. In this study, GDLs that have a shape like a rib are devised, which enable the diffusion under the rib and better mass transport under the channel. Then, the selective-patterned GDL are applied to PEMFCs having different cathode channel types of serpentine, parallel, and interdigitated channels. The PEMFCs having serpentine and parallel channels with selective-patterned GDLs show better performance under all voltage conditions. The performance of serpentine and parallel channel PEMFC was improved 6.0% and 6.7% respectively at 0.5V voltage condition. The interdigitated channel PEMFC having a selective-patterned GDL does not show the remarkable improvement obtained for other channel types using a selective-patterned GDL and the improvement was only 1.6% at 0.5V voltage condition. However, for the volume power density, all analysis results of cases having a selective-patterned GDL are better than the cases with a normal GDL and the improvement degreed of volume power density of serpentine, parallel and interdigitated channel PEMFC was 14.0, 14.7 and 9.2% respectively at 0.5V voltage condition.

Hybrid visible-blind ultraviolet photodetector based on NiO thin-film phototransistor and p-NiO/n-Si heterojunction diode

A hybrid visible-blind ultraviolet (UV) photodetector, which combines a p-NiO/n-Si heterojunction diode and NiO-based thin-film phototransistor, was proposed, and its optoelectronic properties were studied. The hybrid photodetector exhibited a high quantum efficiency and photocurrent gain. The p-NiO/n-Si heterojunction separates the photogenerated electron–hole pairs, thereby reducing the recombination probability of the electron–hole pairs and enhancing photocurrent gain. In the hybrid photodetector, the ratio of the photo current to dark current (IP/ID) increased with an increase in the drain source voltage (VDS) because more photogenerated holes were extracted. Responsivity decreased with increasing positive gate voltage (VG) owing to the depleted channel layer. Conversely, the responsivity increases with increasing negative VG because more holes are accumulated at the surface of the NiO channel layer. The UV (280 nm)/visible (500 nm) rejection ratio is as high as 700–850, indicating that the prepared hybrid device is a good visible-blind photodetector. Detectivity increased from 6.72 × 1012 to 2.23 × 1013 Jones by decreasing VG from 10 to −10 V. Band diagrams of p-NiO/n-Si heterojunction diodes and NiO-based thin-film phototransistors were proposed to elucidate the mechanism of the photocurrent gain in the prepared hybrid photodetector.

Using Pruning-Based YOLOv3 Deep Learning Algorithm for Accurate Detection of Sheep Face.

Simple SummaryThe identification of individual animals is an important step in the history of precision breeding. It has a great role in both breeding and genetic management. The continuous development of computer vision and deep learning technologies provides new possibilities for the establishment of accurate breeding models. This helps to achieve high productivity and precise management in precision agriculture. Here, we demonstrate that sheep faces can be recognized based on the YOLOv3 target detection network. A model compression method based on K-means clustering algorithm and combined with channel pruning and layer pruning is applied to individual sheep identification. In addition, the results show that the proposed non-contact sheep face recognition method can identify sheep quickly and accurately.Accurate identification of sheep is important for achieving precise animal management and welfare farming in large farms. In this study, a sheep face detection method based on YOLOv3 model pruning is proposed, abbreviated as YOLOv3-P in the text. The method is used to identify sheep in pastures, reduce stress and achieve welfare farming. Specifically, in this study, we chose to collect Sunit sheep face images from a certain pasture in Xilin Gol League Sunit Right Banner, Inner Mongolia, and used YOLOv3, YOLOv4, Faster R-CNN, SSD and other classical target recognition algorithms to train and compare the recognition results, respectively. Ultimately, the choice was made to optimize YOLOv3. The mAP was increased from 95.3% to 96.4% by clustering the anchor frames in YOLOv3 using the sheep face dataset. The mAP of the compressed model was also increased from 96.4% to 97.2%. The model size was also reduced to 1/4 times the size of the original model. In addition, we restructured the original dataset and performed a 10-fold cross-validation experiment with a value of 96.84% for mAP. The results show that clustering the anchor boxes and compressing the model using this dataset is an effective method for identifying sheep. The method is characterized by low memory requirement, high-recognition accuracy and fast recognition speed, which can accurately identify sheep and has important applications in precision animal management and welfare farming.

The channels East of Olympus Mons, Mars

ABSTRACT Multiple layers of lava flows and channels characterize the region adjacent to the eastern slope of Olympus Mons, the largest volcano on Mars. We have mapped this volcanic region to survey and classify individual channel systems and determine their formative processes. As a final output of this mapping effort, we have produced a 1:1.9 million scale channel map that is first published in this paper in both GIS and static formats.

14‐2: Student Paper: Enhanced Electrical Characteristics of Low‐Temperature Processed In‐Ga‐Zn‐O Thin‐Film Transistors with Oxygen Scavenging Layer

In this study, the oxygen scavenging layer (OSL) was introduced between channel and gate insulator to improve the electrical characteristics even at 200"C fabrication of amorphous indium‐gallium‐ zinc oxide thin‐film transistors (a‐IGZO TFTs). The OSL is a layer introduced between the channel layer and the gate insulator layer, which is thinly deposited to absorb oxygen in the effective channel of a‐IGZO with annealing process. MgOx was used to make the OSL. The a‐IGZO TFTs with OSL, even if annealed at a low‐temperature of 200°C, exhibit improved electrical characteristics and stability under positive bias temperature stress (PBTS) compared to those without OSL: field‐effect mobility from 8.15 to 16.08 cm2/Vs, subthreshold swing from 0.45 to 0.42 V/decade, on/off current ratio from 3.19 × 108 to 9.53 × 109. A threshold voltage shift under PBTS at 50 °C for 10,000 sec was decreased from + 8.94 V to + 1.98 V. These improvements are attributed to Mg of OSL absorbing oxygen ions in the effective channel layer of a‐IGZO.

미세액적 제조를 위한 다층 액적 분할 미세유체 시스템 개발

In this study, we present the multilayered symmetrical droplet splitting microfluidic system for preparation of microspheres. The microfluidic device was fabricated by conventional photolithography and PDMS casting. Multiple layers of microfluidic channels for symmetrical droplet splitting were stacked and integrated into a device. Each layer was designed to obtain 16 microdroplets from one droplet by droplet splitting. The droplet size was controlled with flow rate of dispersed phase (DI-water) and continuous phase (Mineral Oil with 3 wt.% SPAN80) by using a syringe pump. The droplet splitting behavior and production rate were analyzed by high-speed camera and inverted microscope in one layer of the microfluidic device. Additionally, the droplet size and size distribution were observed in each layer of the microfluidic device. The droplet size could be controlled by flow control of two phase flows with high uniformity of droplet size less than 5% coefficient of variation.

Modeling Multigate Negative Capacitance Transistors With Self-Heating Effects

A unified model for multigate negative capacitance transistors with self-heating effects (SHEs) is developed. With a metal–ferroelectric–insulator–semiconductor (MFIS) structure, thermal effects result in different temperature rises in the channel and ferroelectric layer. A gate control equation for the double-gate and gate-all-around MFIS structures is developed, on top of which both the current–voltage and charge–voltage characteristics are covered. The formulations describe dependences of ferroelectric layer thickness, with backward compatibility for MOSFETs. A unified thermal network is then proposed for SHE, capturing the essential physics of couplings between ferroelectric temperature and conduction current. The model has been verified with TCAD simulations and implemented for circuit simulations.

P‐15: Student Poster: Charge Trap‐Based Synaptic Transistor Employing In‐Ga‐Zn‐O as Channel and Trap Layers for Bio‐Inspired Neuromorphic Computing

In this paper, indium‐gallium‐zinc‐oxide (IGZO) based charge trap synaptic transistors with Al2O3/IGZO/Al2O3 gate dielectric stacks were proposed. It was found that the annealing effect of the IGZO charge trap layer is one of the key parameters for an excellent memory window and program/erase operations without light stimuli. Synaptic behaviors such as excitatory post‐synaptic current (EPSC), inhibitory post‐synaptic current (IPSC), long‐term potentiation (LTP), and long‐term depression (LTD) are success‐fully implemented.

An antifatigue and self-healable ionic polyurethane/ionic liquid composite as the channel layer for a low energy cost synaptic transistor

The developing trends of the traditional transistor include being synaptic and flexible, which could down low the energy cost and realize the daily application of flexible electronics, respectively. However, only bendable synaptic transistor has been developed, which is mainly restricted by the rigid components. Among which, the channel layer has not been replaced by any stretchable material yet. In this work, we synthesized an ionic polyurethane (iPU), and blend it with ionic liquid (IL) to construct a flexible channel layer with high transparency, stability, antifatigue and stretchability. The self-healing process can be completed at ambient and operating temperatures, and the self-healing speed was accelerated by the elevated temperature. The recovery ratio of the iPU/IL film was 95% after cyclic stretch to 50% strain for 100 cycles, which proved good antifatigue property. The stable combination between iPU and IL was verified by the two-dimensional Raman spectra, and the interconnected paths of IL around hard domains were discovered, which was probably due to the strong interaction between ions and hard segments. Spin coating technology was applied to construct the iPU/IL channel layer on a Si matrix, and an ionic conductive transistor was manufactured subsequently, in which the conductive mechanism is similar with a true synapse. The resulted transistor showed an on-off ratio of 2.1 × 102, and its energy dissipation was only 64 nJ when simulating the signal transmission of a nerve synapse, which was close to the power cost in other synaptic transistors.

Graphene Oxide-Chitosan Network on a Dialysis Cellulose Membrane for Efficient Removal of Organic Dyes.

Currently, water pollution is a significant health problem for both humans and animals due to large amounts of dye-containing wastewater. Thus, polymer composite membranes (PCMs) are considered as efficient adsorption/filtration membranes that can be utilized for removing organic dyes from contaminated water/wastewater. In this study, the goal is to explore the modification of the interfacial dialysis cellulose (DC) surface through molecular interactions of an active graphene oxide-chitosan (GO-CTS) composite hydrogel (GCCH) network without the use of an external cross-linker toward an effective dye removal ability using a simple casting process and a low-cost adsorption technique, resulting in the formation of a PCM, i.e., GO/CTS/DC membrane (GCD-mems). Concomitantly, the incorporation of the GCCH network (as an active hybrid network) and DC (as a supporting material) is considered as a promising approach toward a dye-removing PCM. As a result, the GCD-mems showed that cellulose robustly interacted via the chemical bonds of the GCCH network by maintaining the three-dimensional (3D) porous layer structures, and the functional surface of the membrane was enhanced toward specific groups for an effective dye removal approach. In addition, there is a significant improvement in dye removal performance after modification of the interfacial DC surface through molecular interactions of GCCH, i.e., high adsorption capacities of cationic and anionic dye molecules on the GCD-mems, compared to the relevant GO-based adsorbents. Also, the dye flux and rejection of the GCD-mems can simultaneously remove both methylene blue and Congo red. In the adsorption, it is appropriate with the pseudo-second-order and Langmuir models corresponding to chemical adsorption and monolayer approaches, as well as physical sieving through the 3D layers of porous channels of GCD-mems during the filtration process. Moreover, the structural stability and sustainability of the PCMs are enhanced during the recycling process, and the use of ethanol in the recycling process further simplifies the process and reduces the cost of the PCMs. Thus, the GCD-mems are encouraged as potential candidates that can be applied directly in the removal of dyes from the wastewater of textile industries or selective dialysis applications.

Realization of Enhanced Long‐Term Visual Memory for Indium–Gallium–Zinc Oxide‐Based Optical Synaptic Transistor

AbstractAmorphous indium–gallium–zinc oxide (IGZO)‐based optical synaptic transistor using visible light as the signal that shows a clear difference between long‐term memory (LTM) and short‐term memory (STM) is introduced in this study. However, since oxide semiconductors, including IGZO, do not sense visible light due to their high band gap energy of around 3.0 eV, defects are generated intentionally within IGZO channel layer to enable sensitivity to the visible light by inserting two layers of mesh‐structured titanium dioxide (m‐TiO2) inside the channel layer. Additionally, a layer of thin film‐structured hafnium dioxide (HfO2) is inserted inside the channel to increase the persistent photoconductivity (PPC) effect. This increased PPC effect causes the device to maintain the current long after the light signal is removed. Therefore, the average difference between LTM and STM is achieved ≈110 times, showing a clear difference between LTM and STM. Additionally, 721% of maximum paired‐pulse facilitation (PPF) is achieved. Thus, the device introduced in this study is believed to lead the coming technology such as interactive appliances including displays and robots.

Sensitive Detection of Sweat Cortisol Using an Organic Electrochemical Transistor Featuring Nanostructured Poly(3,4-Ethylenedioxythiophene) Derivatives in the Channel Layer.

In this study, we examined the influence of functionalized poly(3,4-ethylenedioxythiophene) (PEDOT) nanostructures decorated on the channel layer of an organic electrochemical transistor (OECT) for the detection of sweat cortisol, an adrenocorticosteroid stress hormone. The OECT device featured a bilayer channel confined by a PEDOT:polystyrenesulfonate (PSS) underlayer and a nanostructure-decorated upper layer engineered from the monomers EDOT-COOH and EDOT-EG3 through template-free electrochemical polymerization. This molecular design allowed antibody conjugation using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide/N-hydroxysulfosuccinimide coupling through the carboxylic acid side chain, with EDOT-EG3 known to minimize nonspecific binding of biomolecules. We also engineered an OECT device having a channel area without any nanostructures to gain insight into the effect of the nanostructures on cortisol sensing. Our new nanostructure-embedded OECT device facilitated real-time detection of cortisol at concentrations ranging from 1 fg/mL to 1 μg/mL with a detection limit of 0.0088 fg/mL with good linearity (R2 = 0.9566), in addition to excellent selectivity toward cortisol among other structurally similar interfering compounds and high stability and reproducibility. With its rapid response for the detection of 100 ng/mL cortisol-spiked artificial sweat, this nanostructure-decorated OECT device has potential clinical practicality and utility in wearable sensors for future healthcare applications.

Vertically stacked multilayer atomic-layer-deposited sub-1-nm In2O3 field-effect transistors with back-end-of-line compatibility

In this work, we demonstrate vertically stacked multilayer sub-1-nm In2O3 field-effect transistors (FETs) with surrounding gate in a back-end-of-line (BEOL) compatible low-temperature fabrication process. A typical bottom-gated single layer In2O3 FET with maximum on-state current (ION) of 890 μA/μm at VDS = 0.8 V and an on/off ratio over 106 is achieved with a channel length (Lch) of 100 nm. The effects of HfO2 capping and O2 annealing are systematically studied, which is critical to realizing the multilayer FETs. Each atomically thin In2O3 channel layer with a thickness (TIO) of 0.9 nm is realized by atomic layer deposition (ALD) at 225 °C. Multilayer FETs with a number of In2O3 layers up to 4 and 1.2 nm-thick HfO2 between each individual layer are fabricated. An enhancement of on-state current (ION) from 183 μA in a single layer In2O3 FET to 339 μA in a 4 layer device with an on/off ratio of 3.4 × 104 is achieved, demonstrating the key advantage of the multilayer FETs to improve the current. Several critical features, such as large-area growth, high uniformity, high reproducibility, ultrathin body, flexibility, and BEOL compatibility, have turned ALD In2O3 into a noteworthy candidate for next-generation oxide semiconductor channel materials.

Fiber Orientation Analysis of Overflow Water-Assisted Injection Molding with Short Glass Fiber Reinforced Polypropylene

The fiber orientation is playing an important performance indexed for glass fiber reinforced polypropylene for water-assisted injection molding. Based on the viscoelastic constitutive equation (White-Metzner) and the fiber orientation model (iARD-RPR), the effects of fiber mass content, water injection delay time, water injection pressure, and melt temperature, which are on the fiber orientation along the flow direction and shear rate distribution of the melt, were investigated. Studies found that the orientation degree of the fiber along the flow direction was reduced with the increase of the fiber mass content, the extension of the water injection delay time, and the improvement of the melt temperature and that the orientation degree of the fiber along the flow direction was raised with the increase of the water injection pressure flow in the laminar flow state, but it was reduced with the increase in the turbulent state. It can be further learned from the shear rate distribution that decreasing fiber mass content, reducing the water injection delay time, lower melt temperature, and increasing water injection pressure in laminar flow conditions will increase the shear rate in the channel layer and the shear rate gradient along the thickness direction of the melt, while the water injection pressure in the turbulent state is on the contrary.

In-Zn-Sn-O thin film based transistor with high-k HfO2 dielectric

Indium zinc tin oxide (IZTO) semiconducting thin films have been investigated as active channel layers for thin-film transistor (TFT) applications. The IZTO layer was deposited by radio frequency magnetron sputtering (RF-MS) using a sintered IZTO ceramic target with an In:Zn:Sn metal atomic ratio of 40:50:10 on the HfO2 layer either deposited using an atomic layer deposition (ALD) or RF-MS. The annealing treatments after IZTO and HfO2 film depositions were performed in air using a tube furnace and in an oxygen atmosphere using rapid thermal annealing at 400 °C, respectively. Optical transparency measured by UV-Vis spectroscopy showed that the HfO2 films deposited by either RF-MS or ALD were transparent in the visible region and had a similar band gap energy of about 4.9 eV. Electrical characteristics of IZTO based TFT were comparatively evaluated for the samples with IZTO/ALD or RF-MS HfO2/n++-Si structure. The IZTO TFT sample with 60 nm-thick HfO2 deposited by ALD had an on/off current ratio value of 7.0 × 106 and a field-effect mobility value of 6.2 cm2/Vs, which was superior to that with RF-MS deposited HfO2.

β‐(Al0.17Ga0.83)2O3/Ga2O3 Delta‐Doped Heterostructure MODFETs with an Ultrathin Spacer Layer and a Back‐Barrier Layer: A Comprehensive Technology Computer‐Aided Design Analysis

The output characteristics of β‐(Al0.17Ga0.83)2O3/β‐Ga2O3‐based heterostructure modulation‐doped field‐effect transistors (MODFETs) with an ultrathin spacer layer and a back‐barrier layer are fitted with experimental measurements using a Silvaco ATLAS technology computer‐aided design (TCAD) simulation environment, and the calibration of the physical model and material parameters is realized. The effects of spacer layer thickness, barrier layer thickness, Si‐δ doping density, and insertion of a β‐Ga2O3 cap layer on the transfer and transconductance characteristics are examined. It is found that a β‐Ga2O3 cap layer on the top of the heterostructure can increase the sheet carrier density in the heterostructure. A breakdown analysis is also carried out to reveal the effects of several layers on the off‐state characteristics. A range of channel layer thicknesses from 15 to 25 nm is found to be the optimum range to avoid a high off‐state leakage current and earlier breakdown voltage.

Near infrared light-sensing semi-transparent organic phototransistors with soluble benzothiadiazole-based conjugated polymer films

Here we report semi-transparent organic phototransistors (OPTRs), which consist of poly[{3-(5-(7-methyl-5,6-bis(octyloxy)benzo[c][1,2,5]thiadiazol-4-yl)thiophen-2-yl)-6-(5-methylthiophen-2-yl)}-co-{2,5-bis(2-octyldodecyl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione}] (PDPP-8OBT) as a sensing channel layer, for the detection of particular near infrared (NIR) light (905 nm) that is used in the light and distance ranging (LiDAR) systems. Two different thicknesses (45 and 75 nm) were chosen for the examination of NIR light-sensing performances, while both ca. 20 nm-thick silver source/drain electrodes and transparent indium-tin oxide (ITO) gate electrodes were applied to achieve the semi-transparency of devices. Results showed that the present devices could be operated in a mode of typical p-channel transistor and were sensitive to the NIR light. The semi-transparent OPTRs with the 75 nm-thick PDPP-8OBT layers showed better NIR light-sensing performances with the high photosensitivity of ca. 200 %. The performance of devices was slightly varied by running about 50 cycles of transfer curves between forward and backward sweeps, whereas the NIR light-sensing characteristics were well kept during the cycle tests. It is expected that the present NIR-OPTRs with the PDPP-8OBT sensing channel layers can be applied as a practical detector for LiDAR systems through further optimization of device stability.