SBIR Phase I: Exchange Coupling Enabled Scalable High Density Nonvolatile STT-RAM at Fast Speed and Low Power

SBIR 第一阶段：高速、低功耗、支持交换耦合的可扩展高密度非易失性 STT-RAM

• 批准号: 0711564
• 负责人: Eugene Chen
• 金额: $ 10万
• 依托单位: Grandis, Inc
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-07-01 至 2007-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0711564&HistoricalAwards=false
• 关键词: SBIR; Phase; Exchange; Coupling; Enabled

This SBIR project addresses a novel (not heat assisted switching) use of ferromagnetic/antiferromagnetic interfacial exchange coupling to enhance the magnetic anisotropy and thermal stability of the free layer in a MTJ. Spin Torque Transfer Random Access Memory (STT-RAM) solves the write current scaling problem of Magnetic Random Access Memory (MRAM) by using the spin momentum of the tunneling electrons to switch Magnetic Tunnel Junction (MTJ) memory elements. Common MTJ used in STT-RAM suffers from a reduced thermal stability when optimized for lower current STT switching required for a small memory cell size and low power operation. The proposed approach promises lower STT switching current, higher TMR and adequate thermal stability for a high density non-volatile STT-RAM at fast speed and low cost.The success of this effort will enable STT-RAM to achieve performance and density that surpass mainstream semiconductor memories such as SRAM, DRAM and NOR Flash in both embedded and standalone memory markets, and create new sectors in the semiconductor industry. Consumers will buy standalone memories in mobile devices such as cell phones, iPod and digital cameras. Corporations like Qualcomm and Freescale can use this technology in embedded memories.

这个SBIR项目解决了一个新的（非热辅助开关）使用铁磁/反铁磁界面交换耦合，以提高磁各向异性和热稳定性的MTJ中的自由层。 自旋力矩转移随机存取存储器（STT-RAM）通过使用隧穿电子的自旋动量来切换磁性隧道结（MTJ）存储器元件来解决磁性随机存取存储器（MRAM）的写入电流缩放问题。当针对小存储器单元尺寸和低功率操作所需的较低电流STT切换进行优化时，STT-RAM中使用的公共MTJ遭受降低的热稳定性。 该方法的成功将使STT-RAM在性能和密度上超越嵌入式和独立存储器市场上的主流半导体存储器（如SRAM、DRAM和NOR Flash），并在半导体产业中开辟新的领域。消费者将购买手机、iPod和数码相机等移动的设备中的独立存储器。像高通和飞思卡尔这样的公司可以在嵌入式存储器中使用这项技术。