Self-biased magnetoelectric Ni/LiNbO₃/Ni composites: from material characterisation to a battery-free implantable node
Year: 2026
Type: article
Abstract: Self-biased magnetoelectric (ME) composites, which convert a magnetic field into electricity with no static bias, are attractive for powering battery-free, deeply implanted microsystems through tissue without heating it, while returning their data. This paper develops a self-biased Ni/LiNbO₃/Ni composite in which sputter-deposited nickel acts both as the magnetostrictive phase and as the electrodes; an intrinsic residual stress (≈ 50 MPa, confirmed by temperature-dependent VSM) supplies the bias and removes the static field, permanent magnets and epoxy bonding of conventional ME laminates. A unified equivalent circuit, a Mason model reduced near resonance to a Butterworth–Van Dyke network closed by a magneto-mechanical gyrator and a piezoelectric transformer, all elements extracted in closed form from impedance and VSM data, predicts the deliverable power within a factor of two across four composites over two decades of output, and is ready for end-to-end co-simulation in SPICE (Simulation Program with Integrated Circuit Emphasis). Guided by a fill-factor and skin-depth design study, an optimised sample raises the output from ≈ 1 V / 16 µW to 8.5 V / 0.33 mW self-biased (1.5 mW under a small bias). This power is shown, by circuit simulation on the measured source and by a measured-power budget, to run a complete battery-free node: the ME link supplies a pressure sensor, its analogue-to-digital converter and a microcontroller, a 13.56 MHz backscatter channel returns the data, and the harvested power lets a semi-active uplink clear the range limit of passive backscatter at 10 cm. A dosimetric and thermal analysis shows the low operating frequency to be a primary safety enabler (induced field and SAR scale as f and f²): the ME link deposits 50 to 130 times less specific absorption rate than an iso-footprint 13.56 MHz inductive link operated at the field its own standard permits.
Language: English
Subfield: Mechanical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,325
OpenAlex ID: W7225378045
