User Guide

The XPS Analyzer pipeline is organized into six modular stages. Each stage operates independently and communicates through validated Pydantic data models, ensuring type safety and data integrity throughout the analysis.

Pipeline Overview

graph TD
    A[Raw Data] --> B[Data Loader]
    B --> C[Preprocessing / Calibration]
    C --> D[Background Subtraction]
    D --> E[Peak Fitting]
    E --> F[Quantification]
    D --> G[Visualization]
    E --> G
    B --> H[Export]
    F --> H

    style A stroke-width:2px
    style H stroke-width:2px

Module Reference

Module	Page	Purpose
data_loader	Data Loading	File parsing, format detection, spectrum structuring
preprocessing	Energy Calibration	Binding energy shift correction
analysis.background	Background Subtraction	Shirley, Tougaard, and Linear background models
analysis.peak_fitting	Peak Fitting	Voigt/Gaussian/Lorentzian deconvolution, doublet fitting
analysis.quantification	Atomic Quantification	RSF-based concentration calculation
export	Data Export	CSV, Excel, and JSON serialization

Code Conventions

• Immutability: All spectral operations return new model instances via model_copy(deep=True). Set inplace=True to override.
• Validation: All models use Pydantic v2 validators. Invalid data raises typed exceptions at construction time.
• Configuration: Global defaults in config/default_settings.toml can be overridden per call.