2020Unpublished venueRequires access

Spin‐Torque‐Transfer (STT) MRAM Engineering

Denny D. Tang, Chi‐Feng Pai

Open publisher page 1 citations

Abstract

Spin-torque transfer magnetoresistive random-access memory (STT-MRAM) is far more energy efficient than field MRAM. This chapter describes the memory operation and the performance of STT-MRAM. It discusses energy barriers as a function of a magnetic tunnel junction (MTJ) film stack and device structure. The chapter focuses on the switching properties. The discussion starts from a simple uniform magnetization reversal model, called the Macrospin model, in which the magnetization of the entire MTJ free layer magnetization is assumed to precess in unison under spin current. Then, the chapter discusses two STT-MRAM device reliability issues: tunnel barrier degradation and data retention. It covers an MgO tunnel barrier degradation model and the relation between the thermal energy barrier and the data retention time performance at the chip level. The chapter also discusses the 1 MTJ-1 transistor MRAM cell design and scaling. It covers the SPICE model for memory chip-level circuit simulation.

About this research paper

What this paper is about

Spin-torque transfer magnetoresistive random-access memory (STT-MRAM) is far more energy efficient than field MRAM. This chapter describes the memory operation and the performance of STT-MRAM. It discusses energy barriers as a function of a magnetic tunnel junction (MTJ) film stack and device structure. The chapter focuses on the switching properties. The discussion starts from a simple uniform magnetization reversal model, called the Macrospin model, in which the magnetization of the entire MTJ free layer magnetization is assumed to precess in unison under spin current. Then, the chapter discusses two STT-MRAM device reliability issues: tunnel barrier degradation and data retention. It covers an MgO tunnel barrier degradation model and the relation between the thermal energy barrier and the data retention time performance at the chip level. The chapter also discusses the 1 MTJ-1 transistor MRAM cell design and scaling. It covers the SPICE model for memory chip-level circuit simulation.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Spin-torque transfer magnetoresistive random-access memory (STT-MRAM) is far more energy efficient than field MRAM. This chapter describes the memory operation and the performance of STT-MRAM. It discusses energy barriers as a function of a magnetic tunnel junction (MTJ) film stack and device structure. The chapter focuses on the switching properties. The discussion starts from a simple uniform magnetization reversal model, called the Macrospin model, in which the magnetization of the entire MTJ free layer magnetization is assumed to precess in unison under spin current. Then, the chapter discusses two STT-MRAM device reliability issues: tunnel barrier degradation and data retention. It covers an MgO tunnel barrier degradation model and the relation between the thermal energy barrier and the data retention time performance at the chip level. The chapter also discusses the 1 MTJ-1 transistor MRAM cell design and scaling. It covers the SPICE model for memory chip-level circuit simulation.

Key concepts: Magnetoresistive random-access memory, Tunnel magnetoresistance, Spin-transfer torque, Magnetization, Universal memory, Torque, Electrical engineering, Spin (aerodynamics)

Related papers

Back to paper searchBrowse research topicsOriginal source
Spin‐Torque‐Transfer (STT) MRAM Engineering — Research Paper | ScholarLens