Frequency Sweep NMR

The target used in the upcoming G_Eⁿ experiment, E02-013, in Hall A is the polarized ³He target. The ³He nuclei are polarized by the spin-exchange mechanism via optically polarized rubidium. (The detailed process of Rb polarization is discussed elsewhere.) There are two methods currently employed to determine the polarization of the helium viz. NMR (Nuclear Magnetic Resonance) and EPR (Electron Paramagnetic Resonance). Here, we shall talk about Frequency Sweep (FS) NMR polarimetry.

The basic principle of NMR relies on sweeping either the magnetic field, or the RF through resonance. In the case of frequency sweep NMR we sweep the RF while holding the magnetic field (H₀)¹ constant. The expression is symmetric with respect to these two parameters.

The way NMR was performed so far is as follows (field sweep NMR):

• Set the RF at 91kHz. (This gives the resonance at H₀ ~ 28G for ³He (~ 21.5 G for water). These quantities are related through the formula: ω = γ H₁ where γ is the gyromagnetic ratio (3.24 kHz/G for ³He; 4.26 kHz/G for water), and H₁ is the RF field.)

• Sweep the holding field H₀ from 25G to 32G (18G to 25G for water) and back; so we sweep through resonance twice (up and down sweeps).

• Plot the data and fit the resonance peak with the square-root of a Lorentzian; extract the peak heights which correspond to the polarization of the substance (after proper calibration of course).

See here [PDF] for a treatise on FS NMR. It follows Abragam in the initial stages. The AFP condition is derived.

For the RF sweep, we keep the holding field H₀ constant at 28G (21.36G) and sweep the RF from 87kHz to 95kHz and back. The sweep rate is pretty much determined by the AFP condition, and we found that 4KHz/s works really well (in terms of polarization loss).

This method was tested at University of Kentucky by Dr. Wolfgang Korsch (my advisor) and at The College of William and Mary by Dr. Todd Averett.