In NMR, Why does longitudinal magnetization decrease when....

[OmerKocak]

*** Answered off site. No need for further explanation. ***


I am trying to understand MR Imaging physics.

In NMR, when you put some energy in a system with a static longitudinal net magnetization, you create a transverse component and the longitudinal component decreases (in other words, net magnetization vector tips sideways). I can not understand why this happens because;

The transverse component is a result of phase coherence. When two individual spins get in phase with each other, why does the longitudinal components suffer. They are the same vectors that now have transverse components in the same direction, which should have to effect on their longitudinal components. And furthermore, to tip that magnetization you put some energy in the system, which should make more spins prefer a higher energy state and thus, should result in a higher longitudinal component.

I do not see anyone addressing this issue, so I guess there is something obvious I am missing but I could not figure out how.

 

[Orodruin]

In NMR, when you put some energy in a system with a static longitudinal net magnetization, you create a transverse component

First of all, it needs to be pointed out that you are not just randomly putting energy into the system. If that was all you did there would just be more spins antiparallel to the longitudinal field without any phase coherence and this would also lead to a lower net longitudinal magnetisation without giving any transverse magnetisation. Instead, the RF pulse is specifically designed to flip spins precessing with the appropriate Larmor frequency 90 degrees. This is also why the longitudinal component decreases, the RF pulse turns it into a transversal component.

And furthermore, to tip that magnetization you put some energy in the system, which should make more spins prefer a higher energy state and thus, should result in a higher longitudinal component.

That more spins prefer higher energy a priori means that the longitudinal magnetisation will decrease, not the other way around, since the previous state has a larger number of parallel than antiparallel spins. Increase the number of antiparallel spins and the longitudinal magnetisation decreases. (You could add even more energy in an ordered way using an RF pulse to turn the magnetisation 180 degrees, bit this will not give you more magnetisation, just the same magnetisation in the opposite direction.)