Magnetoelastic Sensor Optimization for Improving Mass Monitoring

Abstract

Magnetoelastic sensors, typically made of magnetostrictive and magnetically-soft materials, can be fabricated from commercially available materials into a variety of shapes and sizes for their intended applications. Since these sensors are wirelessly interrogated via magnetic fields, they are good candidates for use in both research and industry, where detection of environmental parameters in closed and controlled systems is necessary. Common applications for these sensors include the investigation of physical, chemical, and biological parameters based on changes in mass loading at the sensor surface which affect the sensor's behavior at resonance. To improve the performance of these sensors, optimization of sensor geometry, size, and detection conditions are critical to increasing their mass sensitivity and detectible range. This work focuses on investigating how the geometry of the sensor influences its resonance spectrum, including the sensor's shape, size, and aspect ratio. In addition to these factors, heterogeneity in resonance magnitude was mapped for the sensor surface and the effect of the magnetic bias field strength on the resonance spectrum was investigated. Analysis of the results indicates that the shape of the sensor has a strong influence on the emergent resonant modes. Reducing the size of the sensor decreased the sensor's magnitude of resonance. The aspect ratio of the sensor, along with the bias field strength, was also observed to affect the magnitude of the signal; over or under biasing and aspect ratio extremes were observed to decrease the magnitude of resonance, indicating that these parameters can be optimized for a given shape and size of magnetoelastic sensor.

Keywords: geometry; magnetoelastic; magnetostrictive; mass; monitoring; resonance; sensor; wireless.

Figure 1

Illustration of the detection of a magnetoelastic sensor by measuring the change in its resonance spectrum. The magnetoelastic sensor and detection coil are inductively coupled, which records the sensor’s resonance spectrum by measuring the coil’s impedance spectrum.

Figure 2

Microscopic images of the edge/corner profiles (A) and edge (B) of a mechanically sheared rectangular sensor that has been annealed. The images feature the 5 mm edge of a 12.7 × 5 mm sensor.

Figure 3

Block-diagram of the key instrumentation and interfaces in this experimental setup.

Figure 4

The DC bias coil (A) and detection coil (B) used in this experiment. During measurement, the detection coil is placed in the DC bias coil, such that the two coils are concentric. The exposed portion of the chamber slide is not used, as the sensor is only placed in the chamber that is completely inserted into the detection coil.

Figure 5

A diagram showing the 8 locations (A–H) where drops were deposited on the surface of each 12.7 mm × 5 mm sensor. Each letter represents a different location. With ‘H’ being at the origin center of the sensor, the rest of the letters (A–G) are mirrored across the four quadrants of the sensor surface to show symmetry.

Figure 6

Schematic representation of magnetic field orientation and angles of maximum displacement in this study. The bias and activation fields are applied parallel and coincident with the 0° alignment of each shape, which is denoted by the darker color configuration. The lighter color configuration shows the same edge/corner that was initially aligned with 0° but is now aligned with the maximum displacement angle for that shape. The direction of applied fields remained the same throughout the experiments.

Figure 7

Charts illustrating the range of sensitivities that exist across the sensor surface in 3D (A) and as 2D lines along the length of the sensor (B). The rectangular sensors were assumed to behave symmetrically. The horizontal axes in both figures represent the entire length of the sensor, with the center of the sensor defined as the origin.

Figure 8

Changes in the sensor’s resonance frequency at varying directions of magnetic field. Three shapes of sensors were evaluated: rectangular (A, 12.7 mm × 5 mm), square (B, length = 12.7 mm), and equilateral triangle (C, base = 14.6 mm, height = 12.7 mm). Angular increments were selected based on the rotational symmetry of the shape. (n = 5; error = ± standard deviation).

Figure 9

Plots of a single square sensor’s resonances at 0° (A), 45° (B), 90° (C), and 135° (D) rotations from the normal orientation.

Figure 10

A plot of the results of the DC bias field optimization experiment (A) and an example of the under- and over-biasing effects on the resonance spectrum (optimal bias at 1.71 kA/m) (B). The sensors were all fabricated at an aspect ratio of 2.5 length over width. (n = 5; error = ± standard deviation).

Figure 11

The responses of rectangular sensors fabricated at lengths 9 mm (A) and 12.7 mm (B) with varying widths were plotted against the surface area of those sensors. (n = 4; error = ± standard deviation).

Figure 12

The response of rectangular sensors fabricated at lengths 9 mm (A) and 12.7 mm (B), normalized for their surface area, plotted against the aspect ratio of those sensors (n = 4, error = ± standard deviation).