Electric-field responsive contrast agent based on liquid crystals and magnetic nanoparticles

Voltage-sensitive optical dyes have revolutionized neuroscience research by depicting neuronal activity in real time and with high spatial resolution.^1–3 However, translation of these results into human use is difficult because of light scattering due to brain tissue and the skull.⁴ A voltage-sensitive MRI contrast agent could similarly transform neuroscience and neurology for humans by providing MRI-readable, functional brain information. We therefore sought such a contrast agent.

Composites of smectic liquid crystals (LCs) and magnetic nanoparticles (MNPs) are known as ferrosmectic materials, and theory describing such materials was first described by Brochard and de Gennes.⁵ In such ferrosmectic composites, the phase and structure of LC molecules affect organization and structure of MNPs (Figure 2A). Previous work has demonstrated that such organizational changes in LC-MNP films can be detected using X-ray diffraction,⁶ and that the ordering response of MNPs can be tuned by altering MNP surface chemistry.^7,8 Additionally, recent research has demonstrated magnetic manipulation of liquid crystals using magnetic nanorods incorporated into LC-MNP composites.⁹ As LCs can be made electric field responsive, and MNP organization can be determined using magnetic resonance imaging, encapsulation of LC-MNP composites has the potential to wirelessly convey local information on the electrical state of an environment via MRI readout. In this work, we describe work characterizing LC-MNP composites under electric field excitation using both X-ray diffraction and MRI.

LC-MNP composites were made by mixing 4-cyano-4’octylbiphenyl (8CB, MW 291.44, Frinton Laboratories, Inc., Vineland NJ) with 2 nm diameter MNPs of Fe₅₂Co₄₈ (30% wt) synthesized via polyol processing.⁷ All MNPs were synthesized in our labs. The liquid crystal 8CB molecule has a length of 2.25 nm. Prior to all experiments, LC and MNPs were mixed and sonicated at 35°C for 30 minutes, followed by sonication at 40°C for 30 minutes. The 35°C sonication step induces the smectic-A-nematic transition (which is known to occur at 35°C), and the 40°C sonication step activates the nematic isotropic transition for 8CB.^7,8 For X-ray diffraction measurements, films of LC-MNP composites were made by depositing 200 $μ$l of the composite onto a glass cover slip (22 mm x 22 mm). Films were ∼200 $μ$m thick, and measurements were taken at 0 V/cm, 50 V/cm, and 100 V/cm applied across the film (film was approximately 1 cm²). A schematic of the X-ray diffraction experimental arrangement is shown in Figure 1. X-ray diffraction measurements are shown in Figure 2B and 2D.

MRI data was collected on composites prepared in a glass vial (5 mm inner diameter) having electrodes on either side of the glass vial. MR Imaging was performed with electric fields of 20 V/m and currents of 12 nA and compared with controls having no applied field (Figure 2C, left and right panels, respectively).

Both X-ray diffraction and MRI were performed on LC-MNP composites, and both techniques were capable of detecting changes in sample ordering based on electric field excitation of the LC-MNP composite (as compared with null experiments involving no electric field excitation). The liquid crystals provide a structural scaffolding upon which MNPs order. X-ray diffraction experiments were performed for 2θ = 2° to 7°. When averaged, changes in sample ordering can be detected using X-ray diffraction (Figure 2B), with increased signal intensity arising from increases in applied voltage. Using XRD, we observe a significant increase in signal from a sample excited at 50 V/cm or 100 V/cm as compared with a null control (Figure 2B). While XRD measurements of composites excited at 50 V/cm and 100 V/cm were within the error of each other, these findings compared with 0 V/cm excitations suggest that increasing electric field excitation yields increasingly ordered LC-MNP composite films.

The measure of the layer spacing (Figure 2D) is more illuminating than the increase in intensity. It shows that the value for the layer spacing moves from being a relatively small spacing (indicating disorder) to being close to the liquid crystal layer spacing at 50 V/cm and 100 V/cm (indicating increased order). This is shown in Figure 2D, which shows the layer spacings obtained when the voltage increases from 0 V/cm to 100 V/cm.

Using MRI, we observe a 5% increase in signal from a sample excited at 20 V/m as compared with a null control (Figure 2C). The contrast is most likely due to local changes in the magnetic field that affect spin relaxation. Ordered MNPs in LC-MNP composite films have a greater overall effect on ensembles of protons in the vicinity of the film, as compared with randomly ordered MNPs in LC-MNP composite films. Such a contrast agent might be used for imaging neuron activation with high temporal and spatial resolution, as electrical fields in and around neurons are often in excess of 20 V/m threshold.

Future experiments will better elucidate the nature of LC-MNP ordering, and will better decipher the effects of LC-MNP ordering as opposed to the changes expected in MNP-only samples under exposure to electric field. Additionally, future experiments will explore the response of LC-MNP composites to electric fields of 50 V/cm and 100 V/cm, as was tested via XRD measurements. Replacing the 8CB smectic liquid crystal with a ferroelectric liquid crystal may further increase the ordering of MNPs in the LC-MNP composite by decreasing the necessary electric field necessary to induce ordering.

Ferrosmectic liquid crystal-magnetic nanoparticle composites show potential as MRI-readable sensors of applied voltage. Here we show that the voltage response of these composites is detectable with X-ray diffraction and MRI. Incorporation of such composites into microscale particles may enable such voltage sensing capabilities to be performed wirelessly using magnetic resonance imaging techniques.