We have studied the magnetic and transport properties of Co2MnGa (CMG) thin films grown on MgO(100) substrates in terms of their chemical evolution from amorphous to ordered L21 phases at the substrate temperature Ts during the thin film deposition. Interestingly, the chemical order and magnetic properties sharply change depending on Ts. The CMG film deposited at Ts = 550 °C exhibits the L21-ordered structure and the magnetization of 3.5 μB/f.u., while the CMG film deposited at Ts = 300 °C shows a B2-ordered structure and a relatively lower magnetization of 3 μB/f.u., possibly due to the Mn–Mn antiferromagnetic interactions. A metallic behavior of the electrical resistivity appeared in the CMG film deposited at Ts = 550 °C, whereas the semiconducting behavior appeared in the CMG films deposited at 300 °C and room temperature. Moreover, we found that the absolute value of α = d(Δρ)/d(T1/2) in the low-temperature range below about 20 K is a measure to evaluate the degree of the chemical disorder. In a Hall effect measurement, the L21-ordered CMG film obtained at Ts = 550 °C shows a sizable anomalous Hall resistivity of 15 µΩ cm. This study unveils the relation between Ts and atomic ordering, providing a new pathway for optimizing the chemical order.
We report a magnetic imaging method using atomic force microscopy to measure a locally induced anomalous Nernst effect. A tip contact creates a local temperature gradient on a sample surface controlled by a neighboring Joule heating wire. We demonstrate imaging of magnetic domains in a nanowire of the ferromagnetic Weyl semimetal Co2MnGa with a spatial resolution of a sub-hundred nanometer at room temperature.
This letter reports high-performance VCMA-MRAM devices fabricated on 12-inch wafer. By leveraging a bottom-pinned MTJ stack and synthetic antiferromagnet engineering, the optimized structure reduces stray field to 1.5 mT, enabling highly symmetric switching with a VCMA coefficient of 71.2 fJ/V$\cdot $ m. Meanwhile, a thicker MgO barrier enhances device robustness and endurance exceeding $\mathbf {{10}}^{\mathbf {{14}}}$ cycles. Moreover, the devices sustain fast voltage-driven switching down to 0.6 ns while maintaining a low write energy of 30 fJ/bit, surpassing prior state-of-the-art MRAM counterparts. Our results establish bottom-pinned VCMA-MRAM as a promising candidate for ultra-fast, energy-efficient, and high-endurance embedded nonvolatile memory.
The spin–orbit field and interfacial exchange field (IEF) are two major interface phenomena, and the detection and manipulation of these fields can enable a variety of nanoscale spintronics devices. Optimizing the IEF, which governs the spin-dependent scattering asymmetry at (ferromagnetic insulator)|(normal metal) interfaces, will pave the way for next-generation nanoscale, low-power insulator spintronics devices. Here, we demonstrate an experimental pathway to detect an IEF between insulating ferromagnet EuS and non-magnetic Cu using magnetoresistance (MR) measurements, and show that the spin-dependent scattering at the common interface can lead to a significant current-in-plane MR in Py Cu EuS trilayer Hall-bar device. Our experiment suggests that simple magnetoresistance measurements can be used to experimentally detect the IEF and thereby the magnetic state of a ferromagnetic insulator.
The increasing computational and energy‐efficiency demands of artificial intelligence are driving the development of new computing paradigms based on emerging memory devices. Probabilistic bits (p‐bits) based on magnetic tunnel junctions (MTJs), benefiting from energy efficiency, CMOS compatibility, and fast sampling, show broad potential in stochastic neural networks. However, unavoidable device‐to‐device variations in MTJ switching probabilities hinder the realization of scalable p‐bit chips. Here, we integrate a high‐density array of voltage‐controlled magnetic anisotropy MTJs (VCMA‐MTJs) and employ a tail truncation strategy to compensate for their variations in probabilistic switching ranges, enabling chip‐level p‐bit functionality. Our VCMA‐MTJ chip (VM‐chip) successfully implements restricted Boltzmann machines (RBM) and deep belief networks (DBN) for image recognition, anomaly detection, image completion, and image generation, demonstrating the practical utility of this chip in stochastic neural networks. Furthermore, we propose a dual‐control strategy that tunes both the amplitude and duration of the voltage pulse, significantly enhancing network performance and performing comparably to conventional software‐generated pseudo‐random numbers. Our work enhances the tolerance of probabilistic computing to variations in MTJ switching probabilities, offering flexibility in device selection and array integration. This approach may provide a promising pathway toward scalable p‐bit hardware and cost‐effective manufacturing in future probabilistic computing systems.