Friday, August 1, 2008

What Does an RF Pulse Actually Look Like?

NMR books and papers always show pulse sequences where the pulses are represented by perfect rectangles indicating perfect rectangular manifolds of monochromatic radiation. This is not the case in the real world. Despite the very impressive timing specifications given by instrument companies which greatly simplify the designing of pulse programs, the pulses at the output of the amplifiers are subject to the imperfect response of electronic components in the console. This manifests itself as pulses with imperfect edges (i.e. measurable rise and decay times). The figure below illustrates this point. The top two panels show 5 microsecond and 1 microsecond pulses measured with a 1 GHz digital oscilloscope at the output of the amplifier of a very modern NMR spectrometer. One can see that to a good approximation, the 5 microsecond pulse is rectangular whereas the 1 microsecond pulse shows an obvious rise and decay.The bottom panel shows an expansion of the beginning and end of the 5 microsecond pulse. The pulses seen by the sample will be even worse due to the response of the capacitors in the tank circuit of the probe.

6 comments:

Magdalena said...

Hi,

How does coil affect the RF pulse shape?

When measuring the reflected signal of the sample with an oscilloscope, why are there high spikes on edges?

Your blog is very helpful.

Magdalena

Glenn Facey said...

Magdalena,

I have never observed that , so cannot say.

Glenn

Anonymous said...

Hi Glenn,
Thank you very much for this wonderful blog.
I have two doubts. Please help me if you find some time.
1. How is the phase of a RF pulse defined?
2. How is it applied in RF to the sample, from this coil-shape?
Thanks and Regards
Vinod

Glenn Facey said...

Dear Anonymous,

1. The phase of the pulse is defined in the pulse program, usually in a phase list. In the case of Bruker pulse programs; 0=x, 1=y, 2=-x and 3=-y.

2. I am not sure what coil shape you refer to. The pulses are delivered to the sample via the coil in which the sample sits. The simplest configuration is a horizontal solenoid coil in a vertical superconducting magnet. During the application of the pulse, the sample experiences an oscillating transverse magnetic field. This is also true for a vertically oriented Helmholtz coil. See the BLOG entry for March 17, 2008. http://u-of-o-nmr-facility.blogspot.com/2008/03/probe-coil-geometry.html.

Glenn

Anonymous said...

Thank you very much Glenn.
But now I've more doubts. Please help me if time permits.
1. How many radio-frequencies are there in a pulse? (is a pulse monochromatic?)
2. If the pulse is monochromatic, how is it able to (quantitatively) excite nuclei of different chemical equivalence?
3. If the pulse contains many radio-frequencies, what is the range over which it does quantitative excitation?

Thanks and regards
Vinod

Glenn Facey said...

Anonymous,

Thank you for your comment. To answer your questions....

1. Radio frequency pulses are monochromatic but since they have finite width, a Fourier analysis of the pulse contains many frequencies. In very simplistic terms, the Fourier transform of the pulse determines its excitation profile.

2. and 3. See this post:
http://u-of-o-nmr-facility.blogspot.com/2009/12/defining-excitation-profile.html

Glenn