Sensoron: Fusing Memory and Computing into Spintronics-based Sensors

我们研究了在以下物理系统中进行库群计算的可能性：1铁磁磁性隧道结(MTJ)传感器和2个磁电(ME)存储器。在主题1中，我们使用一种新的混合VSM/AC磁化率计方法来测量MTJ传感器的响应。我们还开发了大规模的模拟来表征集体MTJ传感器的动力学特性。在主题2中，我们开发了世界领先的质量的铂/铬薄膜器件。此外，我们还展示了一个大的反常霍尔效应信号，这为读取Cr2O_3反铁磁畴的内部存储状态开辟了道路。

We investigated possibilities of Reservoir Computing in the physical systems of: ① ferro-magnetic magnetic tunnel junction (MTJ) sensors, and ② magneto-electric (ME) memories.In topic ①, we developed the measurements for the response of MTJ sensors using a novel method for hybrid VSM/ac susceptometer. Also we developed large-scale simulations for characterizing the dynamics of collective MTJ sensors.In topic ②, we developed world-leading quality of Pt/Cr2O3 thin-film devices. Also, we demonstrated a large anomalous Hall effect signal, which opens the way for reading the internal memory state of Cr2O3 antiferromagnetic domains.