Getting Into NMR Without The Superconducting Magnet

Exploring the mysteries of quantum mechanics surely seems like an endeavor that requires room-sized equipment and racks of electronics, along with large buckets of grant money, to accomplish. And while that’s generally true, there’s quite a lot that can be accomplished on a considerably more modest budget, as this as-simple-as-it-gets nuclear magnetic resonance spectroscope amply demonstrates.

First things first: Does the “magnetic resonance” part of “NMR” bear any relationship to magnetic resonance imaging? Indeed it does, as the technique of lining up nuclei in a magnetic field, perturbing them with an electromagnetic field, and receiving the resultant RF signals as the nuclei snap back to their original spin state lies at the heart of both. And while MRI scanners and the large NMR spectrometers used in analytical chemistry labs both use extremely powerful magnetic fields, [Andy Nicol] shows us that even the Earth’s magnetic field can be used for NMR.

[Andy]’s NMR setup couldn’t be simpler. It consists of a coil of enameled copper wire wound on a 40 mm PVC tube and a simple control box with nothing more than a switch and a couple of capacitors. The only fancy bit is a USB audio interface, which is used to amplify and digitize the 2-kHz-ish signal generated by hydrogen atoms when they precess in Earth’s extremely weak magnetic field. A tripod stripped of all ferrous metal parts is also handy, as this setup needs to be outdoors where interfering magnetic fields can be minimized. In use, the coil is charged with a LiPo battery for about 10 seconds before being rapidly switched to the input of the USB amp. The resulting resonance signal is visualized using the waterfall display on SDR#.

[Andy] includes a lot of helpful tips in his excellent write-up, like tuning the coil with capacitors, minimizing noise, and estimating the exact resonance frequency expected based on the strength of the local magnetic field. It’s a great project and a good explanation of how NMR works. And it’s nowhere near as loud as an MRI scanner.

14 thoughts on “Getting Into NMR Without The Superconducting Magnet

  1. The old “The Amateur Scientist” in Scientific American had an article on how to build a magnetic resonance instrument…in the 1950s or early 60s…using vacuum tubes and an oscilloscope. Of course, it couldn’t generate images; the scope just showed the phenomenon of magnetic resonance in hydrogen atoms.

    The entire collection of The Amateur Scientist is available on CD-ROM. Yes, it’s that old.

  2. Stick it on a non-ferrous cart and timestamp the readings in sync with a GPS receiver, and you have a budget SNMR/MRS setup. Eat that metal detectors, that’s some proper geo-fizz right there!

  3. When NMR scans went commercial for medical use, the word “nuclear” was left out of the name for marketing reasons. They figured, probably correctly, that renaming it would make it easier to convince people to get into the machines. Either way, it has no connection to nuclear weapons or power, both of which involve fission or fusion of nuclei; neither of those happen in NMR imaging.

    1. Vaguely remember college organic chemistry using NMR to identify products of synthesis by matching spectra to those in a big book of organic molecules….very cool

  4. That’s just really cool that you can measure subatomic phenomena with nothing but a magnetic field, a coil of wire, a capacitor, a high impedance input low noise amplifier, and a device capable of processing the signal. The $100 audio amp could possibly be cut out and replaced by a cheaper amplifier with a JFET input, a few op amps, and a 24 bit soundcard IC.

    1. It’s fascinating to realize that the peak on the spectrogram is actually an incredible packet of information about the chemical environment. It’s even wilder when you realize that all you need to do after receiving the free induction decay is send a 2.5V 1ms sine pulse back into the coil at the same frequency, and then you have some low-energy selective control over the behavior of the atoms, which you can then sample again.

      About cheaper amp solutions – an AD620 instrument amplifier module, possibly two in series, a mini +15V -15V SMPS step-up module, plus any cheap USB sound card, costs about $30 USD and is quite portable. That was how I was first able to find a signal.

      After I realized I could use something like the UMC202HD, I just found it more convenient and reliable to use a completely prebuilt solution. I was happy to abstract the amplifier and ADC part away because it requires skills and experience that I don’t have! There are other benefits too, like next-day replaceability, which means less downtime if your system breaks.

      I do wonder what it would have cost to replace the amplifier/ADC board on the Magritek Terranova EFNMR system. $3000 USD, and a month-plus delivery time, maybe?

  5. In the environmental field, proton precession magnetometers are used to locate buried ferrous metal (usually drums). My recollection of their operation is that the spin axis of hydrogen nuclei tend to align with the earth’s magnetic field. They are tweaked into alignment with a pulse of a solenoid coil and allowed to precess back into alignment with the earth’s magnetic field. The rate of precession is proportional to the ambient magnetic field and is detected by the instrument. The display unit is nanoTesla and a typical field strength was 50,000 nanoTesla. I’m thinking they resolved to 50 or a 100 nanotesla and could easily detect the concentration of magnetic field over a wire pin flag or your boot’s steel toes. They were used to collect data over an area which was gridded and contoured, which then told you where the points of interest were. They were not set up to use as a conventional metal detector.
    Where do you get hydrogen nuclei? They were filled with a common hydrocarbon like lighter fluid or kerosene. You couldn’t ship or fly with them full, so you’d just pick up some hydrocarbons at your destination.
    A physicist coworker of mine objected to the name ‘proton precession magnetometer’ and insisted they were actually quark spin state detectors. I have no idea.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.