GIWAXS index tool in Python

Zihan Zhang (zihan.zhang-1@colorado.edu) & Keith White (Keith.White@colorado.edu)

This is a python code to show the Bragg peak position and the Miller index on a 2D image (q\(_{xy}\), q\(_{z}\)). The program is based on GIWAXS simulation tool. You need to download diffraction.py to use the functions.

The Bragg peak position is calculated using cif file. They are labeled as (h k l) on the image and the experiment data is used as background for comparison.

Fig 1. This is the 2D image generated from the GIWAXS index tool. The background is simulation instead of experiment for illustration. The left row is not labeled because a cutoff of intensity is used to eliminate the peaks with low intensity.

To plot the Miller index and their positions, you will need Bragg_peaks functions to calculate the Bragg peaks first and then call Mindexing function to make the 2D figure. The code and the workflow are attached below.

LibraryDefine the 'Mindexing' functionParametersMain

import numpy as np
import diffraction as diff
import matplotlib.pyplot as plt
import matplotlib.cm as cm
import time as time
from matplotlib.widgets import Button
import scipy.io
from matplotlib.widgets import RangeSlider
import scipy.ndimage as ndimage
import matplotlib.patheffects as path_effects

def Mindexing(data,colorbar,fsize,qrange,Mindexrange,I_miller,Cutoff_I,Mqxy,Mqz):
    simuposi=np.zeros([100,2])
    isimuposi=0
    fig,ax=plt.subplots(figsize=fsize)
    plt.imshow(data, interpolation='nearest', cmap=cm.jet,
                   origin='lower', extent=qrange,
                   vmax=colorbar*data.max(), vmin=data.min())
    plt.xlabel('q$_{xy}$(1/A)',fontsize=16)
    plt.ylabel('q$_{z}$(1/A)',fontsize=16)

    MaxI=0
    for h in Mindexrange:
        for k in Mindexrange:
            for l in Mindexrange:
                if Mqxy[h,k,l]<qxymax and Mqz[h,k,l]>qzmin and Mqz[h,k,l]<qzmax:
                    MaxI=np.maximum(I_miller[h,k,l],MaxI)

    for h in Mindexrange:
        for k in Mindexrange:
            for l in Mindexrange:
                if Mqxy[h,k,l]<qxymax and Mqz[h,k,l]>qzmin and Mqz[h,k,l]<qzmax:
                    if I_miller[h,k,l]>Cutoff_I*MaxI:
                        plt.plot(Mqxy[h,k,l],Mqz[h,k,l], 'ko')
                        simuposi[isimuposi,0]=Mqxy[h,k,l]
                        simuposi[isimuposi,1]=Mqz[h,k,l]
                        isimuposi=isimuposi+1
                        textstr='('+str(h-hkl_dimension)+','+str(k-hkl_dimension)+','+str(-l+hkl_dimension)+')'
                        millerplt=plt.text(Mqxy[h,k,l]/(2*qxymax)+0.5, (Mqz[h,k,l]-qzmin)/(qzmax-qzmin), textstr, 
                                 transform=ax.transAxes, fontsize=10,verticalalignment='top',color='w')
                        millerplt.set_path_effects([path_effects.Stroke(linewidth=3, foreground='black'),
                                    path_effects.Normal()])
    return simuposi

data = II1   # This is the background/experiment data.
data = np.log(II1+1)
colorbar=0.9

fsize=(30,30) # You need a large figure to see the index.
qrange=[-qxymax, qxymax, qzmin, qzmax] # This is the q range of your experiment data.
Mindexrange=np.linspace(0,hkl_dimension,hkl_dimension+1) # Only the left side of the image is labeled
Mindexrange=Mindexrange.astype('int')

Cutoff_I=0.001 # Set the cutoff to eliminate the low intensity peaks

dirr = ''
filename = ''
address = dirr + filename +'.vasp'
thetax=np.pi/2*0
thetay=np.pi/2*0
hkl_dimension=5

a1,a2,a3,positions=diff.read_poscar(address)
Bpeaks,Mqxy,Mqz,I_miller = diff.Bragg_peaks(a1,a2,a3,positions,thetax,thetay,hkl_dimension)
simuposi=Mindexing(data,colorbar,fsize,qrange,Mindexrange,I_miller,Cutoff_I,Mqxy,Mqz)
print(simuposi)

Workflow of this program:

stateDiagram-v2

    CIF --> POSCAR
    POSCAR: POSCAR 
    POSCAR: Lattice Constants
    POSCAR: Atom position
    POSCAR --> Bragg

    Input1 --> Bragg
    Input1: Input
    Input1: (hkl) dimensions
    Input1: Rotation
    Bragg: Bragg peak
    Bragg: Bragg peak postitions in reciprocal space (3D)

    Bragg --> Mindexing
    Input2 --> Mindexing
    Input2: Input
    Input2: Experiment data
    Input2: Cutoff Intensity
    Input2: figure size
    Mindexing --> 2d
    2d: 2d image display