# Design and construction of a small embeddable nuclear magnetic resonance sensor utilizing 3D-printed components

## TL;DR

This paper describes a low-cost, 3D-printed NMR sensor that can be embedded in various materials to detect composition changes.

## Contribution

A novel, cost-effective NMR sensor design using 3D printing and neodymium magnets for improved construction and performance.

## Key findings

- The sensor achieved a homogeneous magnetic field of 180 mT using optimized magnet separation.
- It successfully detected T2 lifetimes in samples like milk, yogurt, and cement paste.
- The sensor's performance was validated through pulse calibration and signal processing techniques.

## Abstract

This paper presents the design and construction of a cost-effective embeddable nuclear magnetic resonance sensor using 3D printing to improve the construction process. The sensor comprises two 25.4 mm diameter x 3 mm thick neodymium-iron-boron disk magnets and an elliptical radio frequency coil. Magnetic field simulations were employed to determine the optimal separation between magnets, achieving a relatively homogeneous B0 field of 180 mT at the center of the array. Custom 3D-printed parts ensured precise magnet alignment and facilitated coil fabrication. The sensor was encased within a Faraday cage constructed from a printed circuit board to mitigate external electromagnetic interference. A remote tuning circuit was developed to tune the coil to 7.66 MHz. Initial testing involved using an eraser sample to determine the required 90° and 180° pulse amplitudes and duration. The sensor’s performance was further validated under immersion conditions in milk, yogurt, and fresh cement paste, using the Carr-Purcell-Meiboom-Gill technique. The signals obtained were processed by fitting the data to an exponential decay function to obtain the T2 lifetimes and their corresponding signal intensities, and by Inverse Laplace Transformation to obtain the T2 lifetime distribution. Results indicate the sensoŕs capability to detect variations in samples having different compositions.