Thursday, August 14, 2008

Pulse Sequences to Minimize Acoustic Ringing

Acoustic ringing can be a real problem at low frequencies. Its effects can be minimized in a simple one-pulse spectrum by either throwing away the initial bad data and applying a large phase correction or by using backward linear prediction to calculate the lost data. There are a number of pulse techniques one can use when collecting the data to cancel out the ringing. Two such examples are shown in the figure below. The bottom spectrum is the result of a simple one pulse experiment. The middle spectrum was collected using Bruker's "aring" pulse program and the top spectrum was acquired using Bruker's "aring2" pulse program. In all three cases a simple Fourier transform was applied with exponential line broadening. The spectra were collected with the same number of scans on a 300 MHz instrument.

3 comments:

Anonymous said...

You may also try increasing the acquisition delay (ad for Varian I believe and de for Bruker) from a standard 6 usec to 20-40 usec. This way you can allow the electronic ringing to subside a bit before opening up the receiver. This of course may also introduce large first order phase offsets.

Anonymous said...

Hi,

Your blog is awesome! Very useful.

What is the frequency range of acoustic ringing. Does it depend on the material?

Steve P.

Glenn Facey said...

Steve,

Thank you for your question. The acoustic frequencies are much lower than the rf frequency and mechanical in nature as a result of the forces generated by putting current through devices in a strong magnetic field. I'm not sure what is the precise range of frequencies. The acoustic ringing disappears when the probe is removed from the magnet. It does not depend on your sample but rather the materials and construction of the NMR probe.

Glenn