Unlocking the decoding of unknown magnetic nanobarcode signatures

Abstract

Magnetic nanowires (MNWs) rank among the most promising multifunctional magnetic nanomaterials for nanobarcoding applications owing to their safety, nontoxicity, and remote decoding using a single magnetic excitation source. Until recently, coercivity and saturation magnetization have been proposed as encoding parameters. Herein, backward remanence magnetization (BRM) is used to decode unknown remanence spectra of MNWs-based nanobarcodes. A simple and fast expectation algorithm is proposed to decode the unknown remanence spectra with a success rate of 86% even though the MNWs have similar coercivities, which cannot be accomplished by other decoding schemes. Our experimental approach and analytical analysis open a promising direction towards reliably decoding magnetic nanobarcodes to expand their capabilities for security and labeling applications.

Fig. 1. A schematic depicting magnetic nanobarcodes synthesis procedure, dimensions are not drawn to scale. (a) The ion irradiation of a raw polycarbonate foil (grey), (b) chemical etching (light blue) of the ion irradiated polycarbonate, (c) evaporation of Ti (blue) then Au (yellow) for a back contacts, (d) electrodeposition of MNWs (pink), and (e) a combination wet and dry etch process to remove the back contacts. (f) The chemical structure of polycarbonate. Further details are in ESI.

Fig. 2. The backward remanence magnetization (BRM) spectra of the individual MNWs arrays: (a) FeCo and (b) Ni of various diameters as marked. (c–l) BRM spectra for several combinations (according to the legends). The individual BRM spectra were linearly summed (recreated curves) and compared with the measured BRM of the combinations (exp. data). (b) Shows H_1/2 (the field at the center of the BRM spectrum, where BRM = 0) and σ (the dispersion parameter), which are used to describe the BRM spectra below.

Fig. 3. Magnetic parameters that were tailored to produce unique BRM spectra, (a) coercivity (H_c) and half BRM field (H_1/2), (b) squareness (M_sr/M_s), and (c) the broadening parameter (σ). See ESI for hysteresis loops (H_c, M_sr, M_s) and Fig. 2b for BRM spectra (H_1/2, σ). The maximum error bar (n = 2) for subfigures (a–c) are shown.

Fig. 4. RMS error as a function of the assumed number of magnetic nanobarcodes (N) in unknown combinations, normalized to the RMS error of the best fit to N = 1. As the N increases, the RMS error decreases for combination samples until N equals the real number of the magnetic nanobarcodes in each unknown combination. Here all unknown combinations were prepared with two magnetic nanobarcodes. The combinations shown each had varying diameters with (a) both FeCo, (b) both Ni, and (c) one Ni and one FeCo magnetic nanobarcodes. The unsuccessful decodings were shown by arrows.