This Small Business Innovation Research (SBIR) Phase I project aims to develop a novel type of Magnetoresistive Random Access Memory based on synthetic multiferroics. Magnetic memory commercially used today exploits an electric current for magnetization switching. The bulk of the consumed energy is wasted due to the relatively low switching efficiency. This project investigates the feasibility of using synthetic multiferroics comprising piezoelectric and magnetostrictive materials for magnetization switching. The key advantage of using multiferroics is the ability to control magnetization state by an electric field. The utilization of electric field instead of electric current makes it possible to significantly reduce the energy per switch. There is an urgent need in low-power consuming memory devices, which would be of great benefit for a variety of practical applications including wireless devices such as cellular phones as well as other devices biased by the portative energy sources. It is the goal of this Phase I project to provide the feasibility study of the electric-field driven magnetic memory.

The broader impact/commercial potential of this project includes fundamental advances in portative electronic devices and systems enabling lower power consumption and providing longer operation before battery recharging. The development of fast and low power consuming multiferroic memory has the potential to eventually replace the existing memory elements in a market. If successful, this project will lead to a revolutionary change in the data storage technology by providing fast, scalable and non-volatile memory devices. It is fair to say that the development of low-power consuming memory is of great benefit for society offering better performance for a variety of electronic devices and the overall reduction of electricity consumption.

Project Report

The Andromeda Scientific team is building a new type of magnetic memory utilizing synthetic multiferroics as a switching element. Multiferroics is special type of material combining electrical and magnetic properties (i.e. magnetization can be switched by applying an electric field and, vice versa, electric polarization is a function of the external magnetic field). The utilization of multiferroics allows us to switch between the two magnetic states of the memory element by applying an electric field rather than by passing an electric current. This novel type of multiferroic-based memory is expected to be more energetically efficient than the best existing magnetic memory elements. During the Phase I of the project, Andromeda Scientific has developed a physical model for the synthetic multiferroic element, which incorporates the material parameters (dielectric properties of the piezoelectric material, magnetic properties of the magnetostrictive material) and structure geometry (thickness of the piezoelectric layer, thickness of the ferromagnetic layer, memory element area). Based on the established model, we carried out numerical simulations on the multiferroic element operation: (i) magnetization change as a function of the applied voltage, (ii) dynamic response of the element in different operational regimes. An Optimum material structure (e.g. an optimum ratio between the ferromagnetic and piezoelectric layer thicknesses) has been identified from the numerical simulations. Following the results obtained through the numerical modeling, the multiferroic elements have been fabricated. First, high quality (surface roughness less than 4nm) piezoelectric lead zirconate titanate (PZT) films have been fabricated and tested. Then, the films were covered by Ni layer deposited by the sputtering technique. The material quality check and preliminary DC characterization were obtained through the independent electric and magnetic measurements. Next, magnetization switching in the fabricated multiferroic elements has been studied in a variety of regimes (i.e. different amplitude and frequency of the applied electric field). The testing has been accomplished by using the magneto-optical Kerr effect (MOKE), where the multiferroic element was biased via a high-frequency pulse generator (0.25GHz-5.0GHz). The Andromeda Scientific team collected a large volume of experimental data showing magnetization reversal at the different regimes of switching. The results of the experimental testing confirmed the main conclusions of the theoretical model. Finally, a circuit model has been developed based on the obtained experimental data. The circuit model comprises the physical layout of the multiferroic element and allows us to simulate the characteristics of the complete memory element combining the multiferroic element with the magnetic tunneling junction. We studied the operation of the E-MRAM in different regimes and found the range of parameters most important for specific practical application (fast switching (<100ps) and high thermal stability (Δ>80) for data storage servers, low energy operational mode (<10aJ/switch) for battery biased devices. Critical analysis and a pathway to commercialization have been provided by the obtained theoretical and experimental results. With the help of Dawnbreaker team, we have developed a commercialization strategy, which links the advantages of the E-MRAM technology with the particular market needs and takes into account the capabilities of our company. The project has resulted in a novel technology to be used in magnetoresistive random access memory. The developed technology allows to reduce (~1000 times) electric power consumption in electronic devices (i.e. computers, data storage servers, cell phones) and improve the operational characteristics (i.e. more reliable data storage, longer time before recharging). Overall, the developed technology offers a significant reduction of consumed electricity which, in turn, would further benefit US citizens by improvement in ecology.

Agency
National Science Foundation (NSF)
Institute
Division of Industrial Innovation and Partnerships (IIP)
Type
Standard Grant (Standard)
Application #
1214227
Program Officer
Juan E. Figueroa
Project Start
Project End
Budget Start
2012-07-01
Budget End
2012-12-31
Support Year
Fiscal Year
2012
Total Cost
$140,962
Indirect Cost
Name
Andromeda Scientific LLC
Department
Type
DUNS #
City
Long Beach
State
CA
Country
United States
Zip Code
90815