Difference between revisions of "Static Zero Field Relaxation"
(Created page with "Relaxonomy --> here ---- By convention we define the muon spin polarization <math>\vec{P}</math> to be initially in the <math>z</math> direction when there is no applied...") |
|||
(10 intermediate revisions by the same user not shown) | |||
Line 2: | Line 2: | ||
---- |
---- |
||
By convention we define the muon spin polarization <math>\vec{P}</math> to be initially in the <math>z</math> direction when there is no applied magnetic field ('''ZF'''). The relaxation of <math>\vec{P}</math> is then described by |
By convention we define the muon spin polarization <math>\vec{P}</math> to be initially in the <math>\hat{z}</math> direction when there is no applied magnetic field ('''ZF'''). The relaxation of <math>\vec{P}</math> is then described by |
||
<center><math> g_{zz}(t) \; \equiv \; \langle P_z(0) \, P_z(t) \rangle \; , </math></center> |
<center><math> g_{zz}(t) \; \equiv \; \langle P_z(0) \, P_z(t) \rangle \; , </math></center> |
||
Line 8: | Line 8: | ||
Like "relaxation" in TF, this formulation suffers from several assumptions that are frequently invalid: first, that the average ''local'' magnetic field <math>\langle \vec{B_\mu} \rangle</math> at the muons is also zero; second, that any random local magnetic fields ('''RLMF''') <math>\vec{B_\mu}</math> are distributed uniformly about zero in all three directions. The first assumption is not true for a magnetized ferromagnetic crystal, for example, and the second is probably not true in ''any'' single crystal environment with RLMF. In such cases this description is strictly ''invalid'', but people use it as an approximation anyway. |
Like "relaxation" in TF, this formulation suffers from several assumptions that are frequently invalid: first, that the average ''local'' magnetic field <math>\langle \vec{B_\mu} \rangle</math> at the muons is also zero; second, that any random local magnetic fields ('''RLMF''') <math>\vec{B_\mu}</math> are distributed uniformly about zero in all three directions. The first assumption is not true for a magnetized ferromagnetic crystal, for example, and the second is probably not true in ''any'' single crystal environment with RLMF. In such cases this description is strictly ''invalid'', but people use it as an approximation anyway. |
||
For the case of an isotropically random local field with a Gaussian width <math>\Delta</math>, |
|||
<center> |
|||
<math>{\cal D}(|B|) = {1 \over \sqrt{2\pi} \; \Delta} \exp\left[- {1\over2} \left(B \over \Delta \right)^2\right] </math>, |
|||
</center> |
|||
the resultant "Kubo-Toyabe" static ZF relaxation function (named after Kubo and Toyabe, who first described it) looks like this: |
|||
<center> |
|||
[[Image:KT-GZF.png|400px|inline image (click to see full size)]] |
|||
{| width="80%" |
|||
! align="center" | |
|||
<math>g_{zz}^G(\Delta,t) = {1\over3} + {2\over3} \left(1 - \Delta^2 t^2\right) \; \exp\left[\Delta^2 t^2 \over 2 \right]</math>. |
|||
|} |
|||
</center> |
|||
The static ZF Kubo-Toyabe function was later generalized by Toshimitsu Yamazaki for field distributions representable by a generalized power law, giving |
|||
<center> |
|||
<math>g_{zz}^{\rm TY}(n,\Delta,t) = {1\over3} + {2\over3} \left(1 - \Delta^n t^n\right) \; \exp\left[\Delta^n t^n \over n\right]</math>. |
|||
</center> |
|||
This function should be regarded as strictly phenomenological, like the "''stretched exponential''" (which see). |
|||
<center> |
|||
[[Image:Toshi-ZF.png|400px|inline image (click to see full size)]] |
|||
{| width="80%" |
|||
! align="center" | Yamazaki's generalized relaxation function <math>g_{zz}^{\rm TY}(n,\Delta,t)</math> |
|||
for power laws <math>n=1</math> (red), <math>n=2</math> (black) and <math>n=3</math> (blue). |
|||
|} |
|||
</center> |
Latest revision as of 10:25, 17 September 2022
Relaxonomy --> here
By convention we define the muon spin polarization <math>\vec{P}</math> to be initially in the <math>\hat{z}</math> direction when there is no applied magnetic field (ZF). The relaxation of <math>\vec{P}</math> is then described by
where the lower case <math>g_{zz}(t)</math> denotes the static case of the more general <math>G_{zz}(t)</math> relaxation function.
Like "relaxation" in TF, this formulation suffers from several assumptions that are frequently invalid: first, that the average local magnetic field <math>\langle \vec{B_\mu} \rangle</math> at the muons is also zero; second, that any random local magnetic fields (RLMF) <math>\vec{B_\mu}</math> are distributed uniformly about zero in all three directions. The first assumption is not true for a magnetized ferromagnetic crystal, for example, and the second is probably not true in any single crystal environment with RLMF. In such cases this description is strictly invalid, but people use it as an approximation anyway.
For the case of an isotropically random local field with a Gaussian width <math>\Delta</math>,
<math>{\cal D}(|B|) = {1 \over \sqrt{2\pi} \; \Delta} \exp\left[- {1\over2} \left(B \over \Delta \right)^2\right] </math>,
the resultant "Kubo-Toyabe" static ZF relaxation function (named after Kubo and Toyabe, who first described it) looks like this:
<math>g_{zz}^G(\Delta,t) = {1\over3} + {2\over3} \left(1 - \Delta^2 t^2\right) \; \exp\left[\Delta^2 t^2 \over 2 \right]</math>. |
---|
The static ZF Kubo-Toyabe function was later generalized by Toshimitsu Yamazaki for field distributions representable by a generalized power law, giving
<math>g_{zz}^{\rm TY}(n,\Delta,t) = {1\over3} + {2\over3} \left(1 - \Delta^n t^n\right) \; \exp\left[\Delta^n t^n \over n\right]</math>.
This function should be regarded as strictly phenomenological, like the "stretched exponential" (which see).
Yamazaki's generalized relaxation function <math>g_{zz}^{\rm TY}(n,\Delta,t)</math>
for power laws <math>n=1</math> (red), <math>n=2</math> (black) and <math>n=3</math> (blue). |
---|