Browsing by Author "Chandrasekaran, Sridhar"

Now showing 1 - 12 of 12

Controlled resistive switching characteristics of ZrO2-based electrochemical metallization memory devices by modifying the thickness of the metal barrier layer
(2018-04-01)
Chandrasekaran, Sridhar
;
Simanjuntak, Firman Mangasa
;
Tseng, Tseung-Yuen
;
電子工程學系及電子研究所
;
電機工程學系
The effects of varying the thickness of the TiW barrier layer on the switching characteristics of a ZrO2-based electrochemical metallization memory (ECM) device were investigated. The thickness of the TiW barrier layer may alter the resistive switching characteristics of Cu/TiW/ZrO2/TiN ECM devices. Devices made without a TiW barrier layer exhibit unstable cycle-to-cycle behavior. The switching stability of ZrO2 devices improves after inserting a TiW layer. However, the insertion of TiW beyond critical thickness leads to switching degradation. We suggest that an appropriate TiW barrier layer thickness is necessary for achieving good switching performance. (c) 2018 The Japan Society of Applied Physics.
Effect of barrier layer on switching polarity of ZrO2-based conducting-bridge random access memory
(2017-09-11)
Chandrasekaran, Sridhar
;
Simanjuntak, Firman Mangasa
;
Tsai, Tsung-Ling
;
Lin, Chun-An
;
Tseng, Tseung-Yuen
In this study, metal diffusion barrier-dependent switching polarity in ZrO2-based conducting-bridge random access memory was investigated. The device without the barrier layer (BL) exhibited nonpolar switching characteristics. However, inserting TiW BL resulted in positive reset failure. This phenomenon depends on the size and shape of the conducting bridge and also on the defects that contribute to the formation and rupture of the bridge. Consequently, the properties of the conducting bridge govern the device switching performance. Cu- and oxygen vacancy-based conducting bridge during N-Set for a device with and without the BL was proposed. The effect of the insertion of BL on the switching performance was also discussed. The absence of BL resulted in switching instability and poor nonvolatility. By contrast, a device with BL exhibited enhanced uniformity and nonvolatility, and the retention was more than 10(5) s at 200 degrees C. Published by AIP Publishing.
Enhanced Switching Properties in TaOx Memristors Using Diffusion Limiting Layer for Synaptic Learning
(2020-01-01)
Jung, Pei-Yu
;
Panda, Debashis
;
Chandrasekaran, Sridhar
;
Rajasekaran, Sailesh
;
Tseng, Tseung-Yuen
To move towards a new generation powerful computing system, brain-inspired neuromorphic computing is expected to transform the architecture of the conventional computer, where memristors are considered to be potential solutions for synapses part. We propose and demonstrate a novel approach to achieve remarkable improvement of analog switching linearity in TaN/Ta/TaOx/Al2O3/Pt/Si memristors by varying Al2O3 layer thickness. Presence of the Al2O3 layer is confirmed from the Auger Electron Spectroscopy study. Good analog switching ratio of about 100x and superior switching uniformity are observed for the 1 nm Al2O3 based device. Multilevel capability of the memristive devices is also explored for prospective use as a synapse. More than 104 and 4 x 10(4) cycles nondegradable dc and ac endurances, respectively, alongwith 10(4) second retention are achieved for the optimized device. Improved linearities of 2.41 and -2.77 for potentiation and depression, respectively are obtained for such 1 nm Al2O3-based devices. The property of gradual resistance changed by pulse amplitudes confirms that the TaOx memristors can be potentially used as an electronic synapse.
Enhanced Synaptic Linearity in ZnO-Based Invisible Memristive Synapse by Introducing Double Pulsing Scheme
(2019-11-01)
Chandrasekaran, Sridhar
;
Simanjuntak, Firman Mangasa
;
Panda, Debashis
;
Tseng, Tseung-Yuen
;
電子工程學系及電子研究所
The synaptic plasticity of indium tin oxide (ITO)/ZnO/ITO highly transparent (more than 88 %) analog switching resistive memory device is investigated. Highly stable analog switching behavior for more than 2500 cycles with a good memory window of approximately two orders makes it suitable for synapse application. The synaptic response is investigated by applying identical electrical pulses. The potentiation and depression of the device used the conventional identical single-pulse scheme to perform high nonlinearity (0.83) and decaying training epochs. However, the linearity and the training epochs are improved to 0.44 by implementing the identical double-pulse scheme. The proposed double-pulse scheme offers a broad dynamic range (200) having 320 conductance states. This invisible structure and double-pulse scheme can be highly suitable for the neuromorphic computing devices.
Improving linearity by introducing Al in HfO2 as a memristor synapse device
(2019-11-01)
Chandrasekaran, Sridhar
;
Simanjuntak, Firman Mangasa
;
Saminathan, R.
;
Panda, Debashis
;
Tseng, Tseung-Yuen
Artificial synapse having good linearity is crucial to achieve an efficient learning process in neuromorphic computing. It is found that the synaptic linearity can be enhanced by engineering the doping region across the switching layer. The nonlinearity of potentiation and depression of the pure device is 36% and 91%, respectively; meanwhile, the nonlinearity after doping can be suppressed to be 22% (potentiation) and 60% (depression). Henceforth, the learning accuracy of the doped device is 91% with only 13 iterations; meanwhile, the pure device is 78%. A detailed conduction mechanism to understand this phenomenon is proposed.
Neutral oxygen irradiation enhanced forming-less ZnO-based transparent analog memristor devices for neuromorphic computing applications
(2020-04-09)
Simanjuntak, Firman Mangasa
;
Ohno, Takeo
;
Chandrasekaran, Sridhar
;
Tseng, Tseung-Yuen
;
Samukawa, Seiji
Surface oxidation employing neutral oxygen irradiation significantly improves the switching and synaptic performance of ZnO-based transparent memristor devices. The endurance of the as-irradiated device is increased by 100 times, and the operating current can be lowered by 10 times as compared with the as-deposited device. Moreover, the performance-enhanced device has an excellent analog behavior that can exhibit 3 bits per cell nonvolatile multistate characteristics and perform 15 stable epochs of synaptic operations with highly linear weight updates. A simulated artificial neural network comprising 1600 synapses confirms the superiority of the enhanced device in processing a 40 x 40 pixels grayscale image. The irradiation effectively decreases the concentration of oxygen vacancy donor defects and promotes oxygen interstitial acceptor defects on the surface of the ZnO films, which consequently modulate the redox process during rupture and rejuvenation of the filament. This work not only proposes the potential of ZnO-based memristor devices for high-density invisible data storage and in-memory computing application but also offers valuable insight in designing high-performance memristor devices, regardless of the oxide system used, by taking advantage of our neutral oxygen irradiation technique.
Peroxide induced volatile and non-volatile switching behavior in ZnO-based electrochemical metallization memory cell
(2017-09-20)
Simanjuntak, Firman Mangasa
;
Chandrasekaran, Sridhar
;
Pattanayak, Bhaskar
;
Lin, Chun-Chieh
;
Tseng, Tseung-Yuen
We explore the use of cubic-zinc peroxide (ZnO2) as a switching material for electrochemical metallization memory (ECM) cell. The ZnO2 was synthesized with a simple peroxide surface treatment. Devices made without surface treatment exhibits a high leakage current due to the self-doped nature of the hexagonal-ZnO material. Thus, its switching behavior can only be observed when a very high current compliance is employed. The synthetic ZnO2 layer provides a sufficient resistivity to the Cu/ZnO2/ZnO/ITO devices. The high resistivity of ZnO2 encourages the formation of a conducting bridge to activate the switching behavior at a lower operation current. Volatile and non-volatile switching behaviors with sufficient endurance and an adequate memory window are observed in the surface-treated devices. The room temperature retention of more than 10(4) s confirms the non-volatility behavior of the devices. In addition, our proposed device structure is able to work at a lower operation current among other reported ZnO-based ECM cells.
Role of nanorods insertion layer in ZnO-based electrochemical metallization memory cell
(2017-12-01)
Simanjuntak, Firman Mangasa
;
Singh, Pragya
;
Chandrasekaran, Sridhar
;
Lumbantoruan, Franky Juanda
;
Yang, Chih-Chieh
An engineering nanorod array in a ZnO-based electrochemical metallization device for nonvolatile memory applications was investigated. A hydrothermally synthesized nanorod layer was inserted into a Cu/ZnO/ITO device structure. Another device was fabricated without nanorods for comparison, and this device demonstrated a diode-like behavior with no switching behavior at a low current compliance (CC). The switching became clear only when the CC was increased to 75 mA. The insertion of a nanorods layer induced switching characteristics at a low operation current and improve the endurance and retention performances. The morphology of the nanorods may control the switching characteristics. A forming-free electrochemical metallization memory device having long switching cycles (> 10(4) cycles) with a sufficient memory window (10(3) times) for data storage application, good switching stability and sufficient retention was successfully fabricated by adjusting the morphology and defect concentration of the inserted nanorod layer. The nanorod layer not only contributed to inducing resistive switching characteristics but also acted as both a switching layer and a cation diffusion control layer.
Suboxide interface induced digital-to-analog switching transformation in all Ti-based memristor devices
(2020-08-17)
Chang, Lung-Yu
;
Simanjuntak, Firman Mangasa
;
Hsu, Chun-Ling
;
Chandrasekaran, Sridhar
;
Tseng, Tseung-Yuen
Oxidation of TiN is a diffusion-limited process due to the high stability of the TiN metallic state at the TiN/TiO2 junction. Hence, the TiN/TiO2/TiN device being the inability to form a suitable interfacial layer results in the exhibition of abrupt current (conductance) rise and fall during the set (potentiation) and reset (depression) processes, respectively. Interfacial engineering by depositing Ti film served as the oxygen gettering material on top of the TiO2 layer induces a spontaneous reaction to form a TiOx interfacial layer (due to the low Gibbs free energy of suboxide formation). Such an interface layer acts as an oxygen reservoir that promotes gradual oxidation and reduction during the set and reset processes. Consequently, an excellent analog behavior having a 2-bit per cell and robust epoch training can be achieved. However, a thick interfacial layer may degrade the switching behavior of the device due to the high internal resistance. This work suggests that interfacial engineering could be considered in designing high-performance analog memristor devices.
Switching Failure Mechanism in Zinc Peroxide-Based Programmable Metallization Cell
(2018-10-19)
Simanjuntak, Firman Mangasa
;
Chandrasekaran, Sridhar
;
Lin, Chun-Chieh
;
Tseng, Tseung-Yuen
;
電子工程學系及電子研究所
The impact of peroxide surface treatment on the resistive switching characteristics of zinc peroxide (ZnO2)-based programmable metallization cell (PMC) devices is investigated. The peroxide treatment results in a ZnO hexagonal to ZnO2 cubic phase transformation; however, an excessive treatment results in crystalline decomposition. The chemically synthesized ZnO2 promotes the occurrence of switching behavior in Cu/ZnO2/ZnO/ITO with much lower operation current as compared to the Cu/ZnO/ITO (control device). However, the switching stability degrades as performing the peroxide treatment for a longer time. We suggest that the microstructure of the ZnO2 is responsible for this degradation behavior and fine tuning on ZnO2 properties, which is necessary to achieve proper switching characteristics in ZnO2-based PMC devices.
The impact of TiW barrier layer thickness dependent transition from electro-chemical metallization memory to valence change memory in ZrO2-based resistive switching random access memory devices
(2018-08-30)
Chandrasekaran, Sridhar
;
Simanjuntak, Firman Mangasa
;
Aluguri, Rakesh
;
Tseng, Tseung-Yuen
;
電子工程學系及電子研究所
The effect of TiW metal barrier layer thickness on voltage-current characteristics of the Cu/TiW/ZrO2/TiN conductive bridge random access memory device was systematically investigated. The change of reset behavior from abrupt decrease to gradual decrease with increasing TiW thickness was observed. Electronic conduction during the forming process was also analyzed to obtain detailed information about the effect of TiW layer thickness on the nature of the conduction phenomenon. The temperature coefficient of resistance of the conductive filament confirms that an electro-chemical metallization (ECM) based conduction was observed in the devices made with a thinner TiW layer. On the other hand, valence change memory (VCM) based conduction was observed with a thick TiW layer. A conduction mechanism is proposed to explain the ECM to VCM conduction transformation phenomenon.
ZnO2/ZnO bilayer switching film for making fully transparent analog memristor devices
(2019-05-01)
Simanjuntak, Firman Mangasa
;
Chandrasekaran, Sridhar
;
Lin, Chun-Chieh
;
Tseng, Tseung-Yuen
;
電子工程學系及電子研究所
Hydrogen peroxide treatment induces the phase transformation of hexagonal ZnO to cubic ZnO2 on the surface of the ZnO switching memory film; this oxidation process effectively reduces the concentration of n-type donor defects (oxygen vacancies and zinc interstitials) in the switching film. The chemically oxidized ZnO2 layer not only lowers the operation current of the device but also can serve as an oxygen bank to improve the endurance of the memristor. The oxidation reaction of peroxide treatment can be easily controlled to achieve an analog behavior with good switching uniformity. The analog memristor device is able to perform two-bit per cell and synaptic operations. Based on the experimental synaptic data, an image processing of 7 x 9 pixels using a simulated artificial neural network comprises 63 synapses is evaluated to mimic the visual cortex function of the brain.