With the further scaling down of device dimensions, charge trapping memory with high k materials HfO2 serving as capture layer shows good endurance and high storage capacity. Its relatively simple process and complete compatibility with the conventional semiconductor process furthermore make it widely studied. The oxygen vacancies in HfO2 are studied using the first-principles calculation in order to learn their influence on the storage properties of charge trapping memory. Write and erase operations of memory devices are simulated via changing the number of electrons in the super cell with defects. The results show that basically the distance between oxygen vacancies has no effect on charge trapping, but the number of oxygen vacancies does affect it. The more the number of oxygen vacancies, the stronger the electron capture ability is. Moreover, four-fold coordinated oxygen vacancy (Vo4) has lager capability for trapping charge than three-fold coordinated oxygen vacancy (Vo3). The analysis of density of states shows that Vo4 induces a large number of quantum states with deep energy levels which is little affected by distance and has large possibility of trapping charges. The results show that oxygen vacancy defects in HfO2 tetravalent coordination are conducive to improving the storage characteristics of charge trapping memory.
HfO2, as a gate dielectric material for the charge trapping memory, has been studied extensively due to its merits such as high k value, good thermal stability, and conduction band offset relative to Si, etc.. In order to understand the reason why the charge trapping efficiency is improved by high k capture layer with respect to charge trapping type memory, the variation of HfO2 crystal texture induced by oxygen vacancy and the influences of it are investigated using the first principle calculation based on density functional theory. Results show that the distance of the nearest neighbor oxygen atom from oxygen vacancy is markedly reduced after optimization, whereas the decrease of distances between the next nearest neighbor oxygen atom from oxygen vacancy and hafnium is less. The change of local crystal lattice is caused by optimized oxygen vacancy for it significantly changes the local lattice, but rarely influences the far lattice. Deep energy level and density of electron states in conduction band are contributed by Hf atoms, while the density of electron states in valence band is contributed by O atoms. The local density of electron states in each element and the total density of electron states in the optimization system are all larger than those in the system without optimization, and the sum of the local densities of electron states is less than the total density of electron states. The trapped charges are moving mainly around the oxygen vacancy and the adjacent atoms of oxygen in the optimization system, but the charges are without optimization throughout the system. The local energy of charge is increased in optimized defect system, while the local energy of charge is conspicuously reduced in the system without optimization, i.e. lattice variation without saturation characteristic has a large effect on the local energy of charge. Results further prove that the change of crystal lattice induced by oxygen vacancy has strong ability to capture charge, which helps improve the features of memory.
For the three kinds of composite materials, i.e., Cu(111)/HfO2(001), Cu(111)/HfO2(010) and Cu(111)/HfO2(100), the first-principles method based on the density functional theory is adopted to calculate their rates of mismatching of interface model, interface adhesion energies, the electric charge densities, the electron localization functions, and the charge density differences separately. The results indicate that the rate of mismatching of the Cu(111)/HfO2(010) interface model is lowest and its interface adhesion energy is higher than the others’, which means that the Cu(111)/HfO2(010) is most stable. From the analyses of charge densities and electron localization functions of the three interfaces, it can be found that only the Cu(111)/HfO2(010) interface is able to form the connective electronic channel along the vertical direction of the Cu electrode. This indicates that electrons possess the localizabilty and connectivity along the HfO2(010) direction, which corresponds to the switching-on direction of the resistive random access memory (RRAM) device. The charge density difference analysis reveals that the charge density distributions overlap, the electrons transfer mutually and bond at the interface of the Cu(111)/HfO2(010). In addition, based on the model of Cu (111)/HfO2 (010) interface, the formation energies of the interstitial Cu at different positions are also calculated. The results show that the closer to the interface the Cu atom, the more easily it migrates into HfO2. This indicates that the electrochemical reaction takes place more easily under the applied voltage, which results in the formation and rupture of Cu conductive filaments. All the above findings will provide a theoretical guidance for improving the performances of RRAM devices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.