2019 Fiscal Year Research-status Report
Development of Ultimate Low-Noise Magnetic Field Sensors Using Novel Sensing Layer in Magnetic Tunnel Junctions for Sensing-Driven Society
| Project/Area Number |
19K15429
|
|---|
| Research Institution | Tohoku University |
|---|
Principal Investigator |
アルマダウィ ミフタ 東北大学, 先端スピントロニクス研究開発センター, 助教 (90729576)
|
|---|
| Project Period (FY) |
2019-04-01 – 2021-03-31
|
| Keywords | spintronics / sensor / noise / YIG / PtMn / deep learning |
|---|
| Outline of Annual Research Achievements |
This work aims to (1) analyze the microscopic origins of magnetic noise using advanced analyses of magnetic domain images, and (2) develop new magnetic tunnel junction (MTJ) sensing devices that decrease noise by 1000 times. For goal (2), new MTJ sensing layer structures will be developed.
(1) We modified an existing magneto‐optical Kerr effect (MOKE) microscope with a low‐noise scientific CMOS camera. We also developed new characterization computer programs. We explored the domain evolution at very low magnetic fields, of new sensing layer structures developed in path (2). We also developed new algorithms for noise processing based on a novel deep-learning approach.
(2) Using sputtering growth, we developed two new hybrid structures for sensing layers:
(2-a) We optimized stoichiometric high‐quality Y3Fe5O12 (YIG) films with low surface roughness, optimizing growth conditions (gas pressure, deposition power, oxygen ratio), post-procession (crack-relief trenches, post-annealing), and substrate/buffer combinations (Si and Gd3Ga5O12; Pt, Ta, SiO2). We could make polycrystalline crack-free YIG films with roughness Ra<0.4 nm. We also found a very low linewidth of ferromagnetic resonance (~30 Oe).
(2-b) We also optimized exchange-biased L10-PtMn/fcc-NiFe sensing layers, with optimized control of flat surfaces, and magnetoresistance ratios in MgO MTJs.
We published one papper, have two accepted papers, presented one conference presentation, and two accepted presentations have been canceled due the pandemic.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
A lot time and resources were spent on preparing the requirements of research.
The challenges were: time to acquire sputtering targets, acquiring low-cost low-noise camera, optimizing growth conditions, and computer programing.
However, after fixing the initial challenges, the work progressed more rapidly.
|
| Strategy for Future Research Activity |
Path (1): We will develop more analysis techniques for MOKE to be used in characterizing domain dynamics in ultra-low field region, such as real-space resolved fourier‐transform, lock‐in detection, and first‐order reversal curve (FO
RC). Furthermore, we will develop a new setup for combined MOKE and magnetotranport measurement.
Path (2): We will fabricate MTJs with hybrid sensing layers of PtMn/NiFe, and YIG/NiFe. Furthermore We will measure sensing and magnetic noise properties.
Publications: Two papers in refereed journals: (1) on the YIG/FM sensing layer in J. Phys. D: Appl. Phys., and (2) on the complete sensor devices in Appl. Phys. Lett. Presentation at 2020 MMM international conf.
|
