Three-Dimensional Magnetic Memory Device

这项提议的目标是探索3-D磁记录，以生产未来的高性能存储设备。对数据的需求不断增加，这种需求将继续呈指数级增长。然而，今年，研究人员第一次目睹了传统纵向磁介质中记录的数据随着面密度增加到大约100Gbit/in2而变得高度不稳定。这些已知的替代技术大多是二维的，有望将超顺磁性的极限推迟到1太比特/英寸以上。然而，为了将超顺磁性极限推迟到基本上超过1太比特/英寸标记，将需要在第三(垂直)维度上堆叠记录层。垂直堆叠是这项提议的主要主题--3-D磁记忆的概念的基础。提出了3-D存储设备的几种实现方式。将从理论和实验两个角度对替代实现背后的物理进行比较研究。通过这些实验，提出了使用聚焦离子束(FIB)来制作尺寸小于100 nm的测试结构。重点将研究三个组成部分：1)介质，2)写入和3)读取过程。所提出的用于访问数据的机制之一利用先前开发的(用于垂直记录)方法来使用3D记录介质下面的“软”磁垫层(SUL)来控制强磁场。在写入过程中，SUL的使用允许在整个3-D介质厚度上显著增加记录场。在读回过程中，SUL的“柔软性”强烈影响敏感场(对于互易性原理)，因此将被用作识别单场平面(2-D层)的机制。为了最小化符号间干扰并提高稳定性，建议在所有三个维度上对记录介质进行图案设计。还将利用基于Landau-Lifshits-Gilbert的微磁模型和基于蒙特卡罗的温度模拟对三维磁记录的物理进行理论研究。磁力显微镜(MFM)和光学克尔显微镜将被用来研究记录介质中的晶间相互作用。最后，为了学习如何有效地分析和读回来自3D介质块的信息，建议利用信号处理技术，例如“约束反卷积”。本研究所需的基本Co/Pt基3-D介质薄膜将在内部使用共溅射沉积方法制备。利用光学光刻和随后的FIB微调，将对纳米级的位单元进行测试。为了进行3-D磁记录的比较研究，希捷科技承诺提供更多的记录换能器(用于在FIU进行进一步的FIB微调)和各种形式的记录介质(通过基于电子束的光刻形成图案)，并帮助提供其他表征方法。此外，为了支持这项工作，希捷已提出捐赠Balzers公司的13室溅射系统。这个项目的最终目标是制定指导方针，以设计一个合适的3-D介质，并了解读写过程的物理过程。它的成功取决于将数据存储方面的工程经验与对磁记录设备的基本物理的理解和对先进数据识别方法的知识充分结合起来。总部设在最大的少数群体服务机构之一和最年轻的研究机构(FIU)，这个具有强大产业联系的联合项目有望促进FIU的研究举措，并有力地促进未被充分代表的群体参与高级研究。在这些研究工作中，私人投资机构已经建立了邀请顶尖研究人员在FIU举行座谈的传统。由于目前的共同努力，金融情报室对纳米级信息系统的研究兴趣显著增加。仅在一年的时间里，参与纳米磁性设备研究的学生(主要是少数人)的数量从1人增加到12人。至于技术影响，由于该项目的性质，它将有助于数十亿美元的数据存储行业的发展。从短期来看，类似于今天流行的闪存，3-D磁存储器可以用作USB兼容设备，具有显著更高的数据容量。从长远来看，3-D磁存储器可能会取代传统的磁硬盘和其他存储技术，在适当的情况下甚至可能转变为单芯片计算。最后，3-D存储沿着垂直一体化的预期未来总体趋势发展得很好。

Intellectual Merit The objective of this proposal is to explore 3-D magnetic recording in order to produce future high-performance memory devices. There is increasing demand for data and this demand will continue to exponentially grow. However, this year for the first time, researchers witnessed that the recorded data in conventional longitudinal magnetic media becomes highly unstable as the areal density increases beyond approximately 100 Gbit/in2. Most of these known alternative technologies are of 2-D nature and promise to defer the superparamagnetic limit beyond one terabit/in2. However, to defer the superparamagnetic limit substantially beyond the one terabit/in2 mark, it will be necessary to stack recording layers in the third (vertical) dimension. The vertical stacking underlies the concept of 3-D magnetic memory - the primary subject of this proposal. Several implementations of a 3-D memory device are proposed. The physics underlying the alternative implementations will be comparatively studied from both theoretical and experimental perspectives. Through these experiments, it is proposed to use focused ion beam (FIB) to fabricate test structures with sub-100-nm dimensions. The focus will be on the study of three components: 1) medium, 2) write and 3) read processes. One of the proposed mechanisms to access data takes advantage of an earlier developed (for perpendicular recording) method to control strong magnetic fields using a "soft" magnetic underlayer (SUL) under the 3-D recording medium. During the write process, the use of the SUL allows to considerably increase the recording field across the entire thickness of the 3-D medium. During the readback process, the "softness" of the SUL strongly influences the sensitivity field (for the Reciprocity Principle) and thus will be used as a mechanism to identify a uni-field plane (2-D layer). To minimize the inter-symbol interference and improve stability, it is proposed to pattern the recording medium in all three dimensions. The physics of 3-D magnetic recording will be also investigated theoretically with Landau-Lifshits-Gilbert-based micromagnetic modeling and Monte-Carlo-based temperature simulations. Magnetic force microscopy (MFM) and optical Kerr Microscopy will be used to study the intergranular interactions in the recording medium. Finally, to learn how to efficiently analyze and read back information from the bulk of the 3-D medium, it is proposed to take advantage of a signal processing technique such as "constrained deconvolution". Basic Co/Pt-based 3-D medium thin-films necessary for this study will be fabricated inhouse using co-sputter deposition. With optical lithography and following FIB trimming, nanoscale bit cells will be tested. To conduct a comparative study of 3-D magnetic recording, Seagate Technology commits to provide additional recording transducers (for further FIB trimming at FIU) and various forms of recording media (patterned via E-beam-based lithography), and help with additional characterization methods. In addition, to support this work, Seagate has offered to donate a 13-chamber sputtering system by Balzers Corporation. The ultimate goal of this project is to develop guidelines to design an adequate 3-D medium and understand the physics of write and read processes.Broader Impacts This project is interdisciplinary in nature. Its success depends on the adequate integration of the engineering experience in data storage with the understanding of the basic physics of magnetic recording devices and knowledge of advanced data recognition methods. Based at one of the largest minority-serving and one of the youngest research institutions (FIU), this joint project with strong industrial ties promises to boost the research initiatives at FIU and strongly promote the involvement of underrepresented groups in advanced research. Throughout these research efforts, PIs have established a tradition of inviting leading researchers to offer colloquia at FIU. The interest in research on Nanoscale information systems at FIU has significantly grown due current joint efforts. In one year alone, the number of students (mostly minority) involved in the study of nanoscale magnetic devices has grown from one to twelve. As for the technological impact, due to the nature of this project, it will contribute to the advancement of the multi-billion-dollar data storage industry. From a short-term perspective, similar to popular flash memory today, 3-D magnetic memory could be used as a USB-compatible device with significantly higher data capacity. In the long-term perspective, 3-D magnetic memory may replace conventional magnetic hard-drives and other memory technologies, and under the right circumstances may even transform into single-chip computing. Finally, 3-D memory goes well along the expected general future trend of vertical integration.