Merge pull request #11 from UppASD/switching2

Switching2

Author: johanhellsvik

content/switching/TASK2/damping/damping.sh

@@ -1,4 +1,4 @@
-#!bin/bash
+#!/bin/bash
 for alpha in 0.2 0.4 0.6 0.8 1.0
 do
  mkdir Damping$alpha

content/switching/switching.rst

@@ -8,71 +8,68 @@ speed and with lower energy consumption has obvious benefits for our society. He
 
 The presentation is avalable as :download:`pdf <Presentation/Lecture_switching_UppASD_school.pdf>`
 
-TASK1
+TASK1: External field
 -----------------
 * The magnetization direction of sample can be manipulated by applying an external magnetic field with the desired orientation. In this exercise you will apply a magnetic field to switch the magnetization direction of macrospin and bcc Fe from **+z** to **-z** direction. 
 
-All the input files are in the Folder ``TASK1``, use the flag ``hfield`` in ``inpsd.dat`` to control the strength as well as the direction of magnetic field.
-Here you will switch the magnetization direction of 
-a macrospin and bcc Fe  from **+z** to **-z** direction, compare the switching 
+1. Use the flag ``hfield`` in ``inpsd.dat`` to control the strength as well as the direction of magnetic field.
+Here you will use the inputs in the folder ``macrospin`` to switch the magnetization direction of 
+a macrospin from **+z** to **-z** direction, compare the switching 
 time with external field ``-1000 ,-1500 ,-2000, -2500, -3000 Tesla``.
-You can directly use the script ``switching.sh`` to complete this simulation. Finally, use the plotting script ``plot.gnu`` or ``plot.py`` to plot the 
+You can directly use the script ``switching.sh`` to complete this exercise. Finally, use the plotting script ``plot.gnu`` or ``plot.py`` to plot the 
 average magnetization of **z** component as a function of time
 for each magnetic field and compare them. Keep in mind that the external field in this simulation
 is incrediblely huge, here we use huge magnetic field to accelerate our simulation. 
 
-After that, you can try different Jij in ``jfile`` of bccFe and compare 
+2. After that, use the inputs in the folder ``bccFe``.  Set the temperature to 0K and the external field to ``-2000T``, then you can try different Jij in ``jfile`` of bccFe and compare 
 whether different set of Jij affects the switching process? Read the maunal and find out where you need to modify in the ``jfile``.
 
 Do you think the switching process of bccFe behaves like a macrospin? Try to explain it 
 from the output ``moment.bccFe100.out``.
 
-TASK2
+TASK2: Damping and magnetic anisotropy
 ------------------------
 * Damping enters the LLG equation as a phenomenological term, which denotes the energy and angular momentum dissipation from the magnetic system to the environment. The magnetic switching is heavily dependent on the damping. 
 In this exercise you will compare the switching time of a macrospin with different damping. 
 
-Use the flag ``damping`` in ``inpsd.dat`` to control the damping parameter in LLG equation. Here you still switch the magnetization direction of 
+The inputs for this simulation are in the folder ``damping``. Use the flag ``damping`` in ``inpsd.dat`` to control the damping parameter in LLG equation. Here you still switch the magnetization direction of 
 the macrospin from **+z** to **-z** direction, compare the switching 
-time with damping ``0.2 ,0.4, 0.6, 0.8, 1.0``. Finally, plot the 
+time with damping ``0.2 ,0.4, 0.6, 0.8, 1.0``. You can directly use the script ``damping.sh`` to complete this exercise. Finally, plot the 
 average magnetization of **z** component as a function of time
-for system with each damping and see how different damping affects the switching process(Here you can try to modify the python plot script in **TASK1**). 
+for system with each damping and see how different damping affects the switching process and explain it from LLG equation(Here you can create a python plot script by yourself). 
 
 Keep in mind that the damping value in real materials is 
 usually in the order of **1e^⁻3** or **1e^⁻4**, here we use huge 
 damping to accelerate our simulation.
 
-* To switch a magnetic system, one needs to overcome the energy barrier, which is determined by system’s magnetic anisotropy and the volume of the system. In actual device, the energy barrier can stablize system's magnetic order. However, huge energy barrier will induce higher power consumption to manipulate the magnetism in device. In this assignment, you will explore the magnetic switching process of macrospin with different magnetic anisotropy. 
+* To switch a magnetic system, one needs to overcome the energy barrier, which is determined by system’s magnetic anisotropy and the volume of the system. In actual device, the energy barrier can stablize system's magnetic order. However, huge energy barrier induces higher power consumption to manipulate the magnetism in device. In this assignment, you will explore the magnetic switching process of macrospin with different magnetic anisotropy. 
 
-Use the flag ``anisotropy`` in ``inpsd.dat`` combined with an external
+The inputs for this simulation are in the folder ``anisotropy``.  Use the flag ``anisotropy`` in ``inpsd.dat`` combined with an external
 file named ``kfile`` storing the magnetic anisotropy
 parameter of system to manipulate the magnetic anisotropy parameter.
 Here still switch the magnetization direction of 
 macrospin from **+z** to **-z** direction, compare the switching 
-time with anisotropy ``K1=-0.02 ,-0.2, -0.4``. Read the maunal and find out where you need to modify in the ``kfile``.
-Plot the  average magnetization of **z** component as a function of time
+time with anisotropy ``K1=-0.02 ,-0.2, -0.4``. You can directly use the script ``switching_K.sh`` to complete this exercise. Read the manual and find out where you need to modify in the ``kfile``.
+Plot the average magnetization of **z** component as a function of time
 for system with each magnetic anisotropy and see the effect of anisotropy on switching.
 
 
-
-TASK3
+TASK3: Precessional switching
 ------------------------
 In this exercise, you will simulate the precessional switching which has been introduced in the talk. 
 
 Here you will use the flag ``do_bpulse`` in ``inpsd.dat`` to control the type of field pulse(The pulse type like square pulse, square pulse with exponential head and tail,
-Gaussian pulse,exponentially decaying pulse have been implemented in UppASD). Here you will use the Gaussian pulse in this exercise, so set ``do_bpulse 2``. 
-Then modify the parameter of pulse center and width of pulse for Gaussian in the file named ``bpulsefile``.
-Plot the average magnetization of **z** component as a function of time and compare the switching process of different pulse.
-
+Gaussian pulse,exponentially decaying pulse have been implemented in UppASD). Here you will use the Gaussian pulse in this exercise, so set ``do_bpulse 2``. Run the simulation and plot the average magnetization of **z** component as a function of time.
+If you have time, try to compare the switching process with different damping and parameters for Gaussian pulse in the file named ``bpulsefile``.
 
 
-TASK4 
+TASK4: Spin transfer torque(STT) 
 ------------------------
 * Magnetic memory devices have been studied extensively in the past years. The IBM racetrack memory being one of the prime examples of this. This device relies on the movement of a magnetic texture connecting two magnetic domains with different orientation. This texture is known as a domain wall. To move this wall a spin polarized current can be applied to the system exerting a torque over the texture forcing it to move. In this exercise, you will simulate the precessional switching. 
 
 In this task, use the flag ``Initmag 4`` and ``restartfile ./restart.DOMAIN.DW`` to initialize spin configuration of 
 the system. Then turn on the STT flag by ``stt A`` and define the polarized direction of current by ``jvec jx jy jz``. Visualize the domain wall motion dynamics of the
-system by UppASD GUI, for that you need to set ``do tottraj Y`` to get the trajectory of each spin at each sampling step. You can play with the flag ``jvec`` to manipulate the direction of domain motion.
+system by UppASD GUI, for that you need to set ``do tottraj Y`` to get the trajectory of each spin at each sampling step. You can play with the flag ``jvec`` to manipulate the direction of domain motion(Play with different strength and vectors of current).