Difference between revisions of "- Saturation"

From Advanced Projects Lab
Jump to: navigation, search
(Saturation)
 
(2 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
===Saturation===
 
===Saturation===
Previously, we considered the interaction between the spins and lattice and found the characteristic relaxation time <math>T_1</math>. That is, when the spin temperature <math>T_{_S}</math> is greater than  the lattice temperature <math>T</math> how long does it take for the spin temperature to return to the lattice temperature? Answer: <math>T_1</math>.  
+
Previously, we considered the interaction between the spins and lattice and found the characteristic relaxation time <math>T_1</math>. That is, when the spin temperature <math>T_{_S}</math> is different than  the lattice temperature <math>T</math> how long does it take for the spin temperature to return to the lattice temperature? Answer: <math>T_1</math>.  
  
 
We then discussed spin-spin interactions which led to the notion of inhomogeneous broadening by a spatially-varying local magnetic field. This led to the concept of a phase-memory time constant <math>T_2</math>. That is, when two nearly-resonant spins are excited then allowed to freely decay (RF field off) how long before the two spins are out of phase? Answer: <math>T_2</math>.
 
We then discussed spin-spin interactions which led to the notion of inhomogeneous broadening by a spatially-varying local magnetic field. This led to the concept of a phase-memory time constant <math>T_2</math>. That is, when two nearly-resonant spins are excited then allowed to freely decay (RF field off) how long before the two spins are out of phase? Answer: <math>T_2</math>.
Line 6: Line 6:
 
We now investigate the ''steady-state behavior'' of the spin-lattice system with an applied transverse RF field.
 
We now investigate the ''steady-state behavior'' of the spin-lattice system with an applied transverse RF field.
  
In the absence of the transverse RF field, the differential equation describing the time variation of <math>n</math>, the excess number of nuclei in the lower state is <math>\tfrac{dn}{dt}=\tfrac{n_{_0}-n}{T_{_1}}</math>. This makes sense because if <math>T_S>T</math> then <math>\tfrac{dn}{dt}</math> is greater than zero (<math>n<n_{_0}</math>) and there's a net flow of nuclei into the lower state with a characteristic time constant <math>T_{_1}</math>.
+
In the absence of the transverse RF field, the differential equation describing the time variation of <math>n</math>, the excess number of nuclei in the lower state, is <math>\tfrac{dn}{dt}=\tfrac{n_{_0}-n}{T_{_1}}</math>. This makes sense because if <math>T_S>T</math> then <math>\tfrac{dn}{dt}</math> is greater than zero (<math>n<n_{_0}</math>) and there's a net flow of nuclei into the lower state with a characteristic time constant <math>T_{_1}</math>.
  
When the transverse RF field is present, another term must be added to account for the induced upward transitions corresponding to a net absorption of energy so that we get <math>\tfrac{dn}{dt}=\tfrac{n_{_0}-n}{T_{_1}}-2nP</math>, where <math>P</math> is the probability per unit time that a nuclei makes an upward transition. A steady state is reached when <math>\tfrac{dn}{dt}=0</math> so that a steady-state value of <math>n_{_{ss}}</math> is given by <math>\tfrac{n_{_{ss}}}{n_{_0}}=\tfrac{1}{1+2PT_{_1}}</math>.
+
When the transverse RF field is present, another term must be added to account for the induced upward transitions corresponding to a net absorption of energy so that we get <math>\tfrac{dn}{dt}=\tfrac{n_{_0}-n}{T_{_1}}-2nP</math>, where <math>P</math> is the probability per unit time that a nuclei makes an upward transition. A steady state is reached when <math>\tfrac{dn}{dt}=0</math> so that a steady-state value of <math>n_{_{ss}}</math> is given by <math>\tfrac{n_{_{ss}}}{n_{_0}}=\tfrac{1}{1+2PT_{_1}}</math>. It can be seen that <math>n_{_{ss}}</math> decreases as <math>P</math> or <math>T_{_1}</math> increases.
  
we must now evaluate the transition probability <math>P</math>. If a transverse RF field is applied with the correct sense of rotation at the resonant frequency the probability of a transition in unit time between the states is given (with the help of Fermi's golden rule) by  
+
We must now evaluate the transition probability <math>P</math>. If a transverse RF field is applied with the correct sense of rotation at the resonant frequency the probability of a transition in unit time between the states is given (with the help of Fermi's golden rule) by  
  
 
<math>P_{_{m\rightarrow m'}}=\tfrac{1}{2}\gamma^2H_{_1}^1|<m|I|m'>|^2g(\nu)</math>, where <math><m|I|m'></math> is the appropriate matrix element of the nuclear spin operator.
 
<math>P_{_{m\rightarrow m'}}=\tfrac{1}{2}\gamma^2H_{_1}^1|<m|I|m'>|^2g(\nu)</math>, where <math><m|I|m'></math> is the appropriate matrix element of the nuclear spin operator.
Line 27: Line 27:
  
 
<math>T_{_S}=T\tfrac{n_{_0}}{n_{_{ss}}}=\frac{T}{Z}</math>.
 
<math>T_{_S}=T\tfrac{n_{_0}}{n_{_{ss}}}=\frac{T}{Z}</math>.
 +
 +
The spins can easily be heated up to extremely high temperatures. For example, at room temperature for <math>^1H</math>, a transverse field of 0.1 Gauss, <math>T_{_1}=1 \text{sec}</math> (cold water) and <math>T_{_2}=10^{-4}</math> seconds, <math>Z\approx 0.05</math>  giving a spin temperature <math>T_{_S}=T/Z\approx5400K</math>.
 +
 +
These high spin temperatures don't require the absorption of much energy from the transverse field. The amount of energy required to raise a spin system, similar to that discussed above, to infinity (the excess number of spins in the lower state equal to zero) is approximately 1 [https://en.wikipedia.org/wiki/Erg erg] or 100 nJ and (from the Wikipedia link above on the Erg) is about the energy a housefly expends doing one pushup.
 +
 +
The rate of approach of a spin system to the steady state while experiencing a steady RF transverse field is described by a solution having the factor <math>e^{\tfrac{-t}{T_{_1}Z}}</math>. That is, the return to equilibrium of a driven spin system occurs exponentially (as seen with the undriven case) with a characteristic time of <math>T_{_1}Z</math> rather than just <math>T_{_1}</math>.

Latest revision as of 12:27, 18 February 2019

Saturation

Previously, we considered the interaction between the spins and lattice and found the characteristic relaxation time . That is, when the spin temperature is different than the lattice temperature how long does it take for the spin temperature to return to the lattice temperature? Answer: .

We then discussed spin-spin interactions which led to the notion of inhomogeneous broadening by a spatially-varying local magnetic field. This led to the concept of a phase-memory time constant . That is, when two nearly-resonant spins are excited then allowed to freely decay (RF field off) how long before the two spins are out of phase? Answer: .

We now investigate the steady-state behavior of the spin-lattice system with an applied transverse RF field.

In the absence of the transverse RF field, the differential equation describing the time variation of , the excess number of nuclei in the lower state, is . This makes sense because if then is greater than zero () and there's a net flow of nuclei into the lower state with a characteristic time constant .

When the transverse RF field is present, another term must be added to account for the induced upward transitions corresponding to a net absorption of energy so that we get , where is the probability per unit time that a nuclei makes an upward transition. A steady state is reached when so that a steady-state value of is given by . It can be seen that decreases as or increases.

We must now evaluate the transition probability . If a transverse RF field is applied with the correct sense of rotation at the resonant frequency the probability of a transition in unit time between the states is given (with the help of Fermi's golden rule) by

, where is the appropriate matrix element of the nuclear spin operator.

For it can be shown that . For the case of this reduces to

so that we have

.

If an RF field is applied, whose amplitude is large, becomes quite small; the spin temperature becomes very high and the spin system is said to be saturated. Remembering that we can write

, where is the value of the saturation factor at the maximum of the lineshape function .

Recall that, in thermal equilibrium, the lattice temperature is defined in terms of as . Similarly, when the system is not in thermal equilibrium, the spin temperature is related to as . These two results yield

.

The spins can easily be heated up to extremely high temperatures. For example, at room temperature for , a transverse field of 0.1 Gauss, (cold water) and seconds, giving a spin temperature .

These high spin temperatures don't require the absorption of much energy from the transverse field. The amount of energy required to raise a spin system, similar to that discussed above, to infinity (the excess number of spins in the lower state equal to zero) is approximately 1 erg or 100 nJ and (from the Wikipedia link above on the Erg) is about the energy a housefly expends doing one pushup.

The rate of approach of a spin system to the steady state while experiencing a steady RF transverse field is described by a solution having the factor . That is, the return to equilibrium of a driven spin system occurs exponentially (as seen with the undriven case) with a characteristic time of rather than just .