Browsing by Author "Wei, Jeng-Hua"

Now showing 1 - 7 of 7

A carbon nanotube field effect transistor with tunable conduction-type by electrostatic effects
(2006-07-01)
Chen, Bae-Horng
;
Wei, Jeng-Hua
;
Lo, Po-Yuan
;
Wang, Hung-Hsiang
;
Lai, Ming-Jinn
In this paper, we report a conduction-type-tunable carbon nanotube field effect transistor (CNT-FET) with double-gate structure (DG CNT-FET). In this study, a specially designed narrow top-gate is created to modulate the energy band in the middle region of a single CNT. In the proposed DG device structure, the top-gate and bottom-gate biases exhibit independent modulation behaviors. Depending on whether a positive or negative bias is applied to the top-gate, the CNT-FET can be operated in either n- or p-type conduction. Energy band diagram conducive to the physical mechanisms of the proposed DG CNT-FET device structure is proposed. Based on the proposed hypothesis, ambipolar CNT-FETs can indeed be converted to n- or p-type-like behaviors. (c) 2006 Elsevier Ltd. All rights reserved.
Effect of oxygen absorption on contact resistance between metal and carbon nano tubes (CNTs)
(2006)
Tsui, Bing-Yue
;
Weng, Chien-Li
;
Chang, Chih-Lien
;
Wei, Jeng-Hua
;
Tsai, Ming-Jinn
Contact resistance (R-c) between metal and carbon nanotubes (CNTs) is studied extensively. Metal oxide formation at interface due to oxygen absorption plays very important role. Chemically inert metals such as Pt and Au results in the lowest R-c. Adding Ta into Pt can solve the adhesion issue while keeps low R-c. Those metals can break metal oxide and form metal carbide are also preferred.
Effects of synthetic antiferromagnetic coupling on back-hopping of spin-transfer torque devices
(2020-08-17)
Chen, Kuan-Ming
;
Cheng, Chih-Wei
;
Wei, Jeng-Hua
;
Hsin, Yu-Chen
;
Tseng, Yuan-Chieh
A synthetic antiferromagnetic (SAF) layer is a key component in spin-transfer torque magneto-resistive random-access memory devices. This study reveals that slight fluctuations in SAF coupling at the margin of the reference layer and hard layer (i.e., concurrent reversal) can lead to write errors in the form of back-hopping (BH). It appears that variable BH behavior can be attributed to competition between antiparallel (AP) -> parallel (P) and P -> AP transitions associated with SAF coupling. Our conclusions are supported by careful analysis of switching phase diagrams and measurements of self-heating and voltage-controlled magnetic anisotropy. We also observed that one form of coupling provided higher perpendicular magnetic anisotropic energy and thermal stability, which is likely due to the Dzyaloshinskii-Moriya interaction (DMI) effect. Thus, minimizing variations in DMI by optimizing SAF coupling is crucial for minimizing write error rates.
Extremely Compact Integrate-and-Fire STT-MRAM Neuron: A Pathway toward All-Spin Artificial Deep Neural Network
(2019-01-01)
Wu, Ming-Hung
;
Hong, Ming-Chun
;
Chang, Chih-Cheng
;
Sahu, Paritosh
;
Wei, Jeng-Hua
This work reports the complete framework from device to architecture for deep learning acceleration in an all-spin artificial neural network (ANN) built by highly manufacturable STT-MRAM technology. The most compact analog integrate-and-fire neuron reported to date is developed based on the back-hopping oscillation in magnetic tunnel junctions. This novel device is unique because it performs numerous essential neural functions simultaneously, including current integration, voltage spike generation, state reset, and 4-bit precision. The device itself is also a stochastic binary synapse, and thus eases the implementation of the compact all-spin ANN with high accuracy for online training.
Novel method of converting metallic-type carbon nanotubes to semiconducting-type carbon nanotube field-effect transistors
(2006-04-01)
Chen, Bae-Horng
;
Wei, Jeng-Hua
;
Lo, Po-Yuan
;
Pei, Zing-Way
;
Chao, Tien-Sheng
Depending on the chirality, single-walled carbon nanotubes (SWNTs) can be either metallic or semiconducting. Thus far, the production of SWNTs, irrespective of synthesis methods, still yields a mixture of both types, with the metallic type being prevalent. However, semiconducting-type SWNTs are needed for carbon nanotube field-effect transistors (CNT-FETs) as well as many sensors. This is because only the semiconducting-type SWNTs can be effectively modulated by the gate voltage. In contrast, the lack of field effect in metallic-type SWNTs adversely impacts their applications in high-performance electronic devices. In this study, we demonstrate for the first time a novel plasma treatment method that allows us to convert metallic-type carbon nanotubes to semi conducting-type CNT-FETs. On the basis of our experimental results, we believe that the ion bombardment during Ar plasma treatment attacks both metallic- and semiconducting-type nanotubes; however, the metallic-type carbon nanotubes are more vulnerable to the attack than those of the semiconducting type, and are subsequently transformed into the latter type.
NV-BNN: An Accurate Deep Convolutional Neural Network Based on Binary STT-MRAM for Adaptive AI Edge
(2019-01-01)
Chang, Chih-Chcng
;
Wu, Ming-Hung
;
Lin, Jia-Wei
;
Li, Chun-Hsien
;
Parmar, Vivek
Binary STT-MRAM is a highly anticipated embedded non-volatile memory technology in advanced logic nodes <28 nm. How to enable its in-memory computing (IMC) capability is critical for enhancing AI Edge. Based on the soon-available STT-MRAM, we report the first binary deep convolutional neural network (NV-BNN) capable of both local and remote learning. Exploiting intrinsic cumulative switching probability, accurate online training of CIFAR-10 color images (similar to 90%) is realized using a relaxed endurance spec (switching <= 20 times) and hybrid digital/IMC design. For offline training, the accuracy loss due to imprecise weight placement can be mitigated using a rapid non-iterative training-with-noise and fine-tuning scheme.
Prospect of cobalt-mix-tetraethoxysilane method on localized lateral growth of carbon nanotubes for both p- and n-type field effect transistors
(2006-09-01)
Chen, Bae-Horng
;
Lin, Horng-Chih
;
Huang, Tiao-Yuan
;
Wei, Jeng-Hua
;
Hwang, Chien-Liang
To apply carbon nanotubes (CNTs) to nanoelectronics, researchers must effectively control the placement and manipulation of massive numbers of CNTs. In this work, we demonstrate a precise growth of single-walled carbon nanotubes (SWNT) on preassigned locations with only cobalt as catalyst. This is in contrast to the laborious and time-consuming physical manipulation of numerous nanotubes one at a time used in the conventional approach. Laterally grown CNTs were accomplished in preassigned areas using an integrated circuit compatible process in this study. In order to synthesize SWNT as the channel of a field effect transistor (FET), the cobalt-mix-tetraethoxysilane (CMT) solution and catalytic chemical vapor deposition were used. The CMT solution has the unique property of well dispersing the cobalt catalyst and uniformly embedding it in predetermined locations after the CMT solution has solidified. Our results show that laterally grown bundled-CNTs could be formed in atmospheric chemical vapor deposit ion with ethanol, by properly controlling the temperature of process, the process time, and the hydrogen reduction time. Since our bundled-CNTs were exposed to the air, all the as-grown CNT-FETs we manufactured exhibit p-type characteristics. Furthermore, we also demonstrate air-stable n-type CNT-FETs successfully without resorting to any additional and complicated annealing process by simply depositing a silicon nitride film on the as-grown p-type devices by plasma enhanced (PECVD) at 390 degrees C. The use of prepatterned catalyst islands, CVD method, and flexibility of simultaneously manufacturing both n- and p-type CNT-FETs may open a new era for applications of CNT-based nanoelectronics. (c) 2006 American Vacuum Society.