XRF Analysis Toolkit

An initial attempt to make a Python package for X-ray fluorescence (XRF) and differential phase contrast (DPC) analysis. It is thin but will grow.

Features

• Data Correction: Deadtime correction and i0 normalization
• Energy Calibration: Flexible energy axis calibration
• Peak Detection: Automated scatter and XRF peak detection
• Physics-Based Identification: XRF line identification using excitation physics
• Spatial Mapping: Element distribution visualization
• DPC Analysis: Beam deflection calculations

Installation

From source:

git clone https://github.com/yourusername/xrf-analysis.git
cd xrf-analysis
pip install -e .

Requirements:

• Python >= 3.8
• numpy >= 1.20.0
• scipy >= 1.7.0
• matplotlib >= 3.3.0
• h5py >= 3.0.0
• xraylib >= 4.0.0

Quick Start

import xrf_analysis as xrf

# Initialize XRF database
xrf.initialize_combined_xrf_database()

# Load and correct data
results = xrf.correct_xrf_data('your_file.h5')
results = xrf.apply_energy_calibration(results, eV_per_channel=10)

# Detect peaks
detection_results = xrf.detect_and_identify_xrf_peaks(results)

# Visualize
xrf.plot_peak_detection_results(detection_results)
xrf.plot_element_distribution(results, 'Fe', 'Ka')

Project Structure

xrf_analysis/
├── core/
│   ├── physics.py          # XRF physics calculations
│   ├── database.py         # XRF line database
│   ├── corrections.py      # Data corrections
│   └── peak_detection.py   # Peak detection algorithms
├── visualization/
│   ├── spectra.py         # Spectrum plotting
│   └── spatial.py         # Spatial distribution plots
└── dpc/
    └── analysis.py        # DPC calculations

Workflow

1. Load and Correct Data

# Load HDF5 file with deadtime and i0 correction
results = xrf.correct_xrf_data('data.h5')

# Apply energy calibration
results = xrf.apply_energy_calibration(
    results,
    eV_per_channel=10,
    energy_offset=0
)

2. Visualize Spectra

# Plot integrated and maximum envelope spectra
xrf.plot_spectrum_with_energy(results)

3. Detect and Identify Peaks

# Detect XRF peaks with scatter peak exclusion
detection_results = xrf.detect_and_identify_xrf_peaks(
    results,
    thresholds=[10, 50, 10],      # [height, distance_eV, prominence]
    tolerance_keV=0.060,           # 60 eV matching tolerance
    detect_scatter=True            # Exclude elastic/Compton peaks
)

# Visualize detection results
xrf.plot_peak_detection_results(detection_results)

4. Map Element Distributions

# Create spatial distribution map
intensities = xrf.plot_element_distribution(
    results,
    element_name='Fe',
    line_family='Ka',
    integration_width_keV=0.15,
    sensor_sum=True,
    scale='log'
)

Advanced Features

Custom Peak Detection Parameters

detection_results = xrf.detect_and_identify_xrf_peaks(
    results,
    thresholds=[20, 100, 15],      # Higher thresholds for cleaner data
    smooth_window=7,                # More smoothing
    bkg_method='polynomial',        # Polynomial background
    bkg_radius=100                  # Larger background filter
)

45° Geometry Correction

# For samples measured at 45° to the beam
xrf.plot_element_distribution(
    results,
    'Cu', 'Ka',
    correct_45deg=True,
    expand_x=True  # Expand to sample surface coordinates
)

DPC Analysis

from xrf_analysis.dpc import calculate_dpc_deflections, plot_dpc_results

# Calculate beam deflections
dpc_results = calculate_dpc_deflections(
    thor_x, thor_y, thor_Tot, i0,
    Lx=10, Ly=10  # Detector dimensions in mm
)

# Visualize DPC results
plot_dpc_results(abs_x, abs_y, dpc_results['beam_x'], 
                 dpc_results['beam_y'], thor_Tot)

API Reference

Core Functions

correct_xrf_data(file_path)

Load and correct XRF data from HDF5 file.

apply_energy_calibration(results, eV_per_channel, energy_offset)

Apply energy calibration to create energy axis.

initialize_combined_xrf_database()

Pre-calculate XRF line database (call once at startup).

detect_and_identify_xrf_peaks(results, **kwargs)

Detect and identify XRF peaks with physics-based filtering.

Visualization Functions

plot_spectrum_with_energy(results)

Plot integrated and envelope spectra.

plot_peak_detection_results(detection_results, y_min)

Visualize peak detection with annotations.

plot_element_distribution(results, element_name, line_family, **kwargs)

Create spatial distribution maps.

License

MIT License - see LICENSE file for details.

Contributing

Contributions welcome! Please open an issue or submit a pull request.

Citation

If you use this package in your research, please cite:

https://github.com/AustralianSynchrotron