XPS Analysis Platform

ChemFitX provides an online XPS analysis workflow for researchers working with X-ray photoelectron spectroscopy data from surfaces, thin films, catalysts, oxides, sulfides, polymers, carbon materials, semiconductors and related materials. The platform combines survey-spectrum inspection, C 1s binding-energy calibration, high-resolution spectrum preprocessing, background correction, peak deconvolution, chemical-state guidance, elemental-composition calculations and export tools in one browser-based workspace.

Upload one or more XPS spectra, move between survey and high-resolution datasets, inspect the binding-energy axis, apply justified preprocessing, correct the background, add and fit peak components, review fit quality and compare fitted binding energies with an internal reference database. Survey spectra can also be used for sensitivity-factor-based elemental composition when the required peaks, areas and correction factors are appropriate.

The XPS files are read and analyzed locally in your web browser by this application. The uploaded spectra are not sent to ChemFitX servers during the analysis.

What the XPS Analysis Platform Can Do

Import multiple XPS datasets in CSV, TSV, TXT, DAT, ASC, XY, XLSX and XLS formats.
Display survey and high-resolution spectra with the conventional high-to-low binding-energy axis direction.
Apply a C 1s binding-energy offset to all loaded spectra from a user-selected experimental C 1s reference position.
Apply optional Savitzky-Golay, Gaussian or moving-average smoothing to high-resolution spectra.
Calculate and inspect Shirley, linear, polynomial, approximate Tougaard, multi-point anchor, rubber-band, SNIP, rolling-minimum or zero backgrounds.
Modify a calculated background interactively before subtraction when manual adjustment is needed.
Add peak components manually and fit high-resolution spectra with a genetic-algorithm search.
Allow peak position, height and FWHM to vary independently, or lock selected binding-energy positions.
Repeat complete genetic searches with Auto Fit and retain the lowest-error solution from the configured trials.
Create reusable session fit models and add linked spin-orbit partner peaks with fixed splitting, equal FWHM and a chosen height ratio.
Report fitted binding energy, height, FWHM, peak area, area percentage and fit-quality statistics including R2, RMSE, AIC and BIC.
Provide nearest chemical-state reference suggestions for fitted high-resolution peaks.
Identify candidate survey elements from approximate binding-energy matches and allow manual survey assignments.
Calculate survey elemental composition using background-corrected areas, relative sensitivity factors and optional transmission factors.
Include or exclude carbon independently from the atomic-percent normalization denominator.
Calculate mass percentage and normalized atomic ratios from the selected survey composition entries.
Search the built-in binding-energy database and display selected survey reference lines.
Calculate a modified Auger parameter from photoelectron binding energy and Auger kinetic energy.
Estimate an information depth from 3 lambda cos(theta) when an appropriate IMFP and take-off angle are supplied.
Estimate simple overlayer thickness from substrate attenuation when suitable reference intensity and attenuation length are available.
Compare loaded datasets by integrated area, number of fitted components, dominant fitted binding energy, dominant area percentage, R2 and calibration offset.
Export a publication-ready plot image and export analysis data and results to CSV or Excel.

Supported XPS Data and File Formats

Supported files

The platform accepts CSV, TSV, TXT, DAT, ASC, XY, XLSX and XLS files. Each spectrum should provide binding-energy and intensity values that can be interpreted as a numerical XPS trace. Multiple spectra can be loaded into one session so that a survey scan, C 1s reference spectrum and other high-resolution core-level scans can be analyzed together.

Survey and high-resolution spectra

ChemFitX uses the dataset context to show the controls that are relevant to the active spectrum. Survey datasets expose elemental-identification and composition tools, the C 1s dataset exposes the energy-calibration controls, and high-resolution datasets expose smoothing, background correction and peak fitting.

Binding-energy axis convention

The spectrum plot displays XPS binding energy from high to low energy. This is a common presentation convention for XPS, but users should still verify that the imported energy column has the intended units and direction before interpreting peaks.

C 1s Binding-Energy Calibration

The platform provides a C 1s calibration workflow for applying a constant binding-energy offset to all loaded spectra. Enter or pick the observed C 1s reference position, enter the justified target binding energy, and apply the resulting offset to the complete XPS dataset.

The default target value is 284.8 eV, which is commonly used for adventitious carbon in many workflows. The application explicitly treats this as a convention rather than a universally valid reference. If your experiment uses another reference, change the target value to the value justified by your acquisition and sample conditions.

When a calibration is applied, the same energy shift is propagated to the loaded spectra, fitting ranges, backgrounds, fitted peak positions and survey assignments so that the session remains internally consistent.

Optional Smoothing of High-Resolution XPS Spectra

Smoothing is available as optional preprocessing for high-resolution spectra. ChemFitX provides Savitzky-Golay, Gaussian and moving-average methods with configurable window settings. Savitzky-Golay processing also includes a selectable polynomial order.

Use smoothing only when it improves signal readability without materially changing peak position, width, asymmetry or area. Because peak fitting operates on the processed signal, excessive smoothing can change the quantities later used for chemical-state interpretation and component-area analysis.

XPS Background and Baseline Correction

A physically reasonable background is important before interpreting high-resolution component areas. ChemFitX provides several background models so that users can choose a method appropriate to the spectrum and the analytical convention being followed.

Shirley background – iterative background commonly used for many core-level regions.
Linear background – a two-point linear baseline across the selected binding-energy region.
Polynomial edge fit – polynomial background fitted from edge behavior.
Approximate Tougaard 2-, 3- and 4-PIESCS-style backgrounds – practical loss-kernel implementations that require careful parameter validation.
Multi-point anchor baseline – uses several anchor regions across the selected interval.
Rubber-band lower-hull baseline – follows a lower-envelope style baseline.
SNIP iterative baseline – iterative peak-clipping style background estimation.
Rolling-minimum baseline – local rolling-window minimum estimate.
Zero background – no background subtraction.

The background calculation can be limited to a user-defined binding-energy region or applied to the full active region. For a calculated background, the Modify Background mode exposes draggable handles that can be moved vertically before the correction is applied. After subtraction, the fitting engine uses the corrected spectrum.

Background choice can substantially influence integrated peak areas and fitted component fractions. The approximate Tougaard variants in this platform should be validated against the method, parameters and software convention used in your laboratory before quantitative publication.

Peak Deconvolution and Genetic-Algorithm Fitting

High-resolution spectra can be deconvoluted into multiple peak components. Add components at approximate binding-energy positions, configure the starting height and FWHM, choose which parameters may vary, and then run the fitting routine.

Parameters that can be fitted

Peak position (binding energy).
Peak height.
Peak FWHM.
Selected binding-energy positions can be locked when a justified constraint is required.

Genetic-algorithm optimization

The fitting engine uses a genetic-algorithm search. Users can control the population size, number of iterations, mutation probability and crossover probability. Auto Fit repeats complete genetic searches from the current component model for the selected number of trials and keeps the solution with the lowest residual sum of squares.

Repeated global searches can reduce dependence on one initial optimization path, but they do not remove model-selection ambiguity. A low numerical residual is not, by itself, proof that the chosen chemical components are physically correct.

Reusable models and linked doublets

Fit models can be saved within the current browser session and applied again. A spin-orbit doublet helper can add a linked partner to a selected main component using a user-defined energy splitting and partner-to-main height ratio. The linked partner follows the main component during fitting with equal FWHM and the specified splitting and intensity relationship.

Doublet constraints should be based on the physics and chemistry of the core level being analyzed. The helper does not determine the correct splitting or branching ratio automatically for every element or chemical state.

Understanding XPS Peak-Fit Results

Binding energy

The fitted binding energy is the optimized peak-center position after any session energy calibration. Chemical-state interpretation should consider the calibration method, sample charging, local chemistry, multiplet structure, satellites and comparison with appropriate literature or evaluated databases.

FWHM

FWHM describes the fitted peak width in electronvolts. Peak width can reflect intrinsic lifetime broadening, instrumental resolution, chemical heterogeneity, unresolved components and model choice. It should not be interpreted as a unique chemical-state identifier.

Peak area and area percentage

The platform calculates an area for each fitted component and the percentage of the summed fitted-component area represented by that peak. These component percentages are useful for comparing fitted species within the same carefully modeled region, but they are not automatically equivalent to bulk composition or absolute concentration.

R2, RMSE, AIC and BIC

The fit summary reports R2 and RMSE as goodness-of-fit measures and AIC/BIC as information criteria related to residual error and model complexity. These values can help compare candidate component models fitted to the same data, but they should be interpreted together with residual structure, physically realistic widths, chemically plausible peak positions and justified constraints.

Chemical-State Suggestions and Binding-Energy Database

After fitting a high-resolution spectrum, ChemFitX compares fitted peak positions with its internal binding-energy reference entries and lists the nearest candidate assignments. The platform also provides a searchable binding-energy database for common core levels and chemical environments.

These matches are guidance only. XPS chemical-state assignment can depend on sample charging, spin-orbit structure, satellites, multiplets, final-state effects, peak asymmetry and the full chemistry of the sample. Final assignments should therefore be compared with evaluated references and appropriate literature rather than accepted solely from the nearest numerical binding energy.

Survey Peak Identification and Elemental Composition

Survey spectra can be used to locate candidate elemental features and estimate surface elemental composition. The platform compares survey features with an internal list of approximate core-level positions, supports a configurable matching tolerance and allows manual survey assignments when a suitable reference line has been verified.

Expected-element suggestions

High-resolution spectra already loaded in the same session can be used to suggest elements that may be expected in the survey scan. These are suggestions only. An element enters the composition calculation only when a corresponding survey feature is matched or manually assigned and is enabled for quantification.

Relative sensitivity factors and transmission

For each included element, ChemFitX uses a background-corrected survey area together with a relative sensitivity factor (RSF) and an optional transmission factor. The composition calculation is based on the sensitivity-corrected quantity A/(S*T), followed by normalization across the selected elements.

A generic Wagner/PHI-style Al K-alpha RSF set is available for convenience, and custom user/instrument factors can be entered. For quantitative work, use factors and transmission corrections appropriate to your instrument, X-ray source, analyzer settings and adopted quantification procedure.

Including or excluding carbon

Carbon can be included or excluded from the atomic-percent normalization denominator independently of using C 1s as the energy-calibration reference. This is useful when carbon is only adventitious/reference contamination rather than an intended sample constituent. If carbon is an intentional component of the material, it can remain included.

Atomic percent, mass percent and ratios

The composition table reports the observed binding energy, integrated area, RSF, transmission factor, sensitivity-corrected area, atomic percentage, mass percentage and optional normalized atomic ratio. These values describe the selected survey quantification model and should not be treated as absolute bulk composition without considering XPS sampling depth and matrix/instrument effects.

Advanced XPS Calculators

Modified Auger parameter

The modified Auger parameter is calculated as alpha’ = BE + KE from a user-supplied photoelectron binding energy and Auger-electron kinetic energy. It can provide useful complementary chemical-state information when both measurements are appropriate and consistently calibrated.

Information depth estimate

The information-depth calculator reports the simple estimate 3 lambda cos(theta), where lambda is the user-supplied inelastic mean free path and theta is the take-off angle used by the platform. This is an estimate, not an automatically determined IMFP or effective attenuation length.

Simple overlayer-thickness estimate

The overlayer calculator uses d = lambda cos(theta) ln(I0/I), where I0 is a reference substrate intensity, I is the attenuated measured substrate intensity and lambda is the supplied effective attenuation length. The result is only as meaningful as the reference intensity, attenuation length and simple uniform-overlayer assumptions used.

Comparing Multiple XPS Datasets

The Loaded Dataset Comparison table summarizes all spectra in the current session. Depending on the available analysis, it can compare dataset type, number of points, integrated area, number of fitted components, dominant fitted binding energy, dominant component area percentage, fit R2 and applied calibration offset.

This table is useful for organizing a multi-spectrum XPS workflow, but direct quantitative comparison between different core levels requires appropriate sensitivity, transmission and acquisition-condition corrections.

How to Use the ChemFitX XPS Analysis Tool

Upload one or more XPS files and load the spectra into the workspace.
Confirm which dataset is the survey scan, which is C 1s, and which are high-resolution core-level spectra.
If binding-energy calibration is required, open the C 1s spectrum, pick or enter the observed reference position, confirm the justified target value and apply the offset to all spectra.
For a high-resolution spectrum, apply smoothing only if it is genuinely needed.
Choose an appropriate background method, define the fitting region and inspect the calculated background before subtraction.
Add one or more peak components at chemically reasonable starting positions.
Configure which peak parameters may vary and apply justified position or doublet constraints when required.
Run Fit Peaks or Auto Fit and inspect R2, RMSE, AIC, BIC, fitted positions, widths, component areas and residual behavior.
Review chemical-state suggestions as reference guidance rather than automatic identification.
For a survey spectrum, identify candidate elements, verify each assignment, set appropriate RSFs/transmission factors and choose whether carbon should be included in normalization.
Calculate elemental composition and inspect atomic %, mass %, corrected area and atomic ratios.
Use the advanced calculators only when the required physical parameters are known and appropriate.
Export the active spectrum as CSV, export the full analysis workbook to Excel, or export the current spectrum plot as PNG.

Interpreting XPS Results Responsibly

Binding-energy values should be interpreted only after a justified energy-calibration procedure.
The nearest database value is not sufficient to prove an oxidation state or chemical species.
Peak number, line shape, FWHM and constraints should be chemically and physically defensible, not chosen only to maximize numerical fit quality.
Background choice directly affects fitted areas and therefore can affect chemical-state fractions and survey quantification.
Survey atomic percentages are sensitivity-corrected surface-composition estimates and depend on RSFs, transmission, acquisition conditions and the selected quantification windows.
XPS is surface sensitive. Survey composition should not automatically be interpreted as bulk composition.
Generic sensitivity factors are useful for exploratory work but should be replaced by the instrument or validated library factors used by your laboratory for quantitative reporting.
The advanced depth and overlayer calculators require externally supplied physical parameters and simplified assumptions; the platform does not infer those parameters from the spectrum.

Exporting XPS Analysis Results

CSV export

CSV export provides a plot-ready table for the active spectrum. Depending on the current analysis, the exported spectrum can include the experimental intensity together with fitted components, fit sum and residual information.

Excel export

The Excel workbook is designed as a complete session export. It includes a results summary and separate sheets for loaded datasets. Where available, it records elemental composition, peak-fit parameters, fit statistics and export notes, chemical-state suggestions, dataset-comparison information and advanced-calculator results.

PNG export

The current XPS spectrum plot can be exported as a PNG image. Interactive handles used for editing are hidden from the exported plot so that the saved image contains the analytical curves rather than the editing controls.

Data Privacy and Local Browser Processing

The XPS application reads and analyzes uploaded spectra locally in the browser. The analysis code does not upload the XPS files to ChemFitX servers.

Researchers should still follow institutional requirements for unpublished spectra, confidential data, shared computers, browser storage and exported result files.

Scientific Scope and Limitations

ChemFitX is an analysis and interpretation aid. Final XPS conclusions remain the responsibility of the researcher.
The default 284.8 eV C 1s reference is a common convention, not a universally valid absolute reference.
Smoothing can alter peak shape, width and area if applied too strongly.
Approximate Tougaard backgrounds in this platform should be validated against the specific method and parameters used for quantitative publication.
Peak deconvolution is model dependent. Different plausible component models may fit the same spectrum with similar residual errors.
Genetic-algorithm optimization searches parameter space but cannot determine the chemically correct number or identity of components by itself.
Chemical-state suggestions are approximate database matches and do not prove species identity.
Spin-orbit splitting and component intensity ratios should be constrained using appropriate literature or validated reference values for the specific core level.
Survey quantification depends on RSFs, transmission, background selection, peak integration and instrument conditions.
Absolute depth profile, IMFP/EAL, work function, film thickness and absolute elemental concentration are not inferred automatically from the uploaded spectra.
The simple information-depth and overlayer-thickness calculators require user-supplied physical parameters and simplified models.
For publication-quality interpretation, compare peak positions, line shapes and quantitative factors with evaluated references such as the NIST XPS Database and the sensitivity-factor resources used by your instrument or established analysis software.

Frequently Asked Questions

What is XPS analysis?

X-ray photoelectron spectroscopy analyzes the kinetic energies of photoelectrons emitted from a material under X-ray irradiation and is widely used to study surface elemental composition and chemical states. Interpretation depends on energy calibration, spectral background, line shape, sensitivity factors and sample chemistry.

Can ChemFitX calibrate XPS binding energy?

Yes. The platform provides a C 1s reference workflow that calculates a constant energy offset from the observed and target reference positions and applies that offset to all loaded spectra.

Is 284.8 eV always the correct C 1s reference?

No. It is a common adventitious-carbon convention, but it is not universally valid. Use the reference value justified for your sample and measurement protocol.

Which background methods are available?

The platform includes Shirley, linear, polynomial, approximate Tougaard variants, anchor, rubber-band, SNIP, rolling-minimum and zero-background options.

How are XPS peaks fitted?

Peak components are optimized with a genetic-algorithm search. Users can choose which peak positions, heights and FWHM values may vary, and Auto Fit can repeat complete searches and retain the lowest-error result.

Does the platform automatically identify oxidation states?

No. It provides nearest binding-energy reference suggestions. Final chemical-state assignment requires the full spectral context, calibration, satellites, multiplets, line shape, chemistry and appropriate literature.

What fit-quality values are reported?

The platform reports R2, RMSE, AIC and BIC together with fitted peak parameters and component areas.

Can I fit spin-orbit doublets?

The platform includes a linked-partner helper that can apply a user-defined splitting, equal FWHM and fixed partner-to-main height ratio. The correct constraint values must come from appropriate physical or literature information.

Can ChemFitX calculate XPS atomic percentages?

Yes. Survey composition is calculated from selected background-corrected areas, RSFs and optional transmission factors, followed by normalization across the included elements.

Can carbon be excluded from survey composition?

Yes. Carbon can be excluded from the atomic-percent normalization while C 1s can still be used for energy calibration.

What is the modified Auger parameter calculator?

It adds a selected photoelectron binding energy and Auger kinetic energy to obtain the modified Auger parameter, which can complement binding-energy information for some chemical-state analyses.

Does ChemFitX determine film thickness automatically?

No. The overlayer calculator uses a simple attenuation equation and requires a suitable reference substrate intensity and attenuation length supplied by the user.

Are my XPS files uploaded to ChemFitX?

For this application, the files are read and analyzed locally in the browser and are not uploaded to ChemFitX servers by the analysis code.

What can I export?

You can export the active plot-ready spectrum as CSV, a complete multi-dataset analysis workbook as Excel, and the current XPS plot as PNG.

Scientific References

NIST X-ray Photoelectron Spectroscopy Database – evaluated binding energies, chemical shifts and Auger data for XPS interpretation.
Briggs, D.; Grant, J. T. Surface Analysis by Auger and X-ray Photoelectron Spectroscopy – reference text for XPS methodology and interpretation.

Instrument-manufacturer or validated analysis-software sensitivity-factor libraries – use the RSF set appropriate to your instrument and acquisition conditions for quantitative composition.