纳米尺度NiO忆阻器的机制和应用

发布时间：2018-01-07 16:13

本文关键词：纳米尺度NiO忆阻器的机制和应用　出处：《清华大学》2016年博士论文　论文类型：学位论文

【摘要】：忆阻器在新型非挥发性存储器、逻辑电路、神经网络等很多领域有广泛的应用前景。器件的成功商业化,很大程度上得益于对它的内在机理有一个比较深刻的理解。对忆阻器的机制的研究,有利于它在应用中的性能提升。另一方面,当忆阻器应用在交叉杆结构中时,会存在所谓的串扰问题,这一问题的解决对忆阻器的成功应用十分重要。Ni O是最具竞争力的忆阻材料之一,本论文在纳米尺度对它的双极性忆阻行为的机制,以及一种可用于解决串扰问题的忆阻行为进行了研究,主要内容包括以下两个方面:一直以来,氧空位模型被认为是忆阻器的主要机理。但最近的一个理论工作表明,在Ni O的双极性忆阻行为中,阳离子空位和阴离子空位的共存是至关重要的。可是,这一双缺陷模型目前尚没有实验的支持。我们利用脉冲激光沉积法在不同氧压下生长了一系列Ni O薄膜样品,利用导电原子力显微镜测量它们的忆阻行为,利用球差校正的扫描透射电子显微镜和电子能量损失谱表征了一个高氧压和一个低氧压生长的样品的离子空位。结果显示,高氧压生长的样品表现出明显的忆阻行为,它的阳离子空位过剩于阴离子空位决定了Ni O的双极性忆阻行为。这一工作首次为双缺陷模型提供了实验支持,并提供了一种研究双缺陷对忆阻行为影响的方法。固有整流的忆阻行为(Intrinsically Rectifying-Memristive Behavior,IR-MB)被认为是一个能有效解决串扰问题的方法。目前所报道的IR-MB都属于界面主导型,即,低阻态形成的肖特基势垒会抑制反向电压下的潜行电流。并且,之前报道的IR-MB主要集中在宏观样品上,而实际应用中的器件通常在纳米尺度。因此,我们利用超薄多孔氧化铝模板制备了高密度、高均匀性的Ni O纳米点阵。利用导电原子力显微镜测量,Ni O纳米点表现出IR-MB。实验和理论分析表明,这种IR-MB源于Ni O纳米点内形成的内建同型同质结,不同于之前所报道的界面型IR-MB。这是第一个报道由内建同质结主导的IR-MB的工作,并且这一类IR-MB是在纳米器件中观察到的。我们模拟评估了利用Ni O纳米点的IR-MB解决串扰问题的有效性,发现它可允许的最大交叉杆存储量可达3 Mbit,比一般的界面型IR-MB所允许的存储量更大。
[Abstract]:The memristor model in non-volatile memory, logic circuit, many fields such as neural network has a wide application prospect. The device successfully commercialized, largely due to the inherent mechanism of it have a more profound understanding. To investigate the mechanism of the memristor, is conducive to its performance in applications improved. On the other hand, when the memristor used in the cross bar structure, there will be a so-called crosstalk problem, to solve this problem is very important for the successful application of memristor.Ni O is the most competitive one of the memristor mechanism, this thesis in nano scale bipolar memristor on it a kind of behavior, and can be used to solve the problem of crosstalk memristance behavior was studied, the main contents include the following two aspects: all along, the oxygen vacancy model is considered to be the main mechanism of the memristor. But a recent theoretical work table, in the Ni O The bipolar memristor behavior, coexisting cation vacancies and anion vacancies is essential. However, the double defect model there is no experimental support. We use pulsed laser deposition in different oxygen pressure growth under a series of Ni O films, measuring their memristance behavior using conductive atomic force using spherical aberration corrected microscope, scanning transmission electron microscopy and electron energy loss spectroscopy to characterize the ion vacancy a high oxygen pressure and a low oxygen pressure growth samples. The results showed that the high oxygen pressure growth samples showed obvious memristive behavior, its cation vacancies over anion vacancies determines bipolar Ni O memristive behavior. This work provides the first experimental support for the double defect model, and provides a method of double defects on the memristor behavior. The inherent rectifier memristance behavior (Intrinsical Ly Rectifying-Memristive Behavior, IR-MB) is considered to be one of the methods can effectively solve the crosstalk problem. At present, as reported by the IR-MB interface are dominant, that is, the Schottky barrier formed by the low resistance state will inhibit sneak current reverse voltage. And the previously reported IR-MB mainly concentrated in the macroscopic sample, and the actual device the application is usually at the nanometer scale. Therefore, we use ultra-thin porous alumina templates for preparation of high density, high uniformity Ni O nano array. Using conductive atomic force microscope, Ni O nano IR-MB. showed experimental results and theoretical analysis show that the IR-MB derived from Ni O dots formed in the built-in the same type of interface type IR-MB. homojunctions, different reports before this is the first reported by the built-in homojunction led IR-MB work, and this kind of IR-MB is observed in nano devices. We die We evaluated the effectiveness of using Ni O nano dot IR-MB to solve the crosstalk problem, and found that the maximum allowable maximum cross bar storage can reach 3 Mbit, which is larger than that of the general interface IR-MB.

【学位授予单位】：清华大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TN60

【相似文献】