Raman analysis, from instrument file to physical result
Atlas covers the full Raman workflow: parsing spectra and maps from the instrument, material and substrate identification, peak fitting with resonance and instrument broadening handled, physics models for defect density, strain and doping, and layer count, region statistics, and Raman–PL correlation. Every calibration limit is written into the result.
What Atlas does with Raman data
Every analysis runs as a documented Block with recorded parameters, so the same result can be reproduced, reviewed, and compared across samples.
Material-aware from the first parse
Atlas identifies the material from its Raman modes, detects substrate lines such as Si 520 cm⁻¹, and builds a per-pixel quality mask before any physics runs.
Physics, not just peak parameters
Linewidths become defect-density estimates, mode positions become a strain-versus-doping decomposition, and mode separation becomes a layer count, with a warning when that calibration saturates.
Resonance and instrument handled
Resonant bands are deconvolved before inversion and instrument broadening is read from the file header, so a 13.5 cm⁻¹ composite width is not mistaken for a disordered film.
Statistics by region
Film interior, damage halo, and bare substrate are segmented first. Edge profiles and data-driven clusters show where quality changes, not just the average.
Raman and PL on the same pixels
Atlas registers Raman and PL exports from the same field, corrects the offset between instruments, and tests whether both techniques see the same defective population.
Beyond TMDs
Dedicated workflows cover MoS₂, WS₂, and other TMDs, graphene, carbon nanotubes, and GaN stress and doping.
Atlas compared with legacy Raman software
Most labs already own capable Raman software. This is where Atlas adds to it, and where your existing tools are still the better choice.
| Capability | Matter42 AtlasCloud or private deployment | Instrument vendor softwaree.g. LabSpec, WiRE, WITec Suite | Desktop analysis toolse.g. Origin, Fityk, Python scripts |
|---|---|---|---|
| Peak fitting across a full map | Supported. Every pixel, with shared fit windows across samples. | Supported. Map fitting is standard. | Partly supported. Per spectrum, or scripted. |
| Defect density, strain, and doping from the spectra | Supported. Built in, with the calibration source and range stated. | Partly supported. Peak parameters; physical interpretation left to the analyst. | Partly supported. Manual, from published coefficients. |
| Resonance and instrument-broadening corrections | Supported. Applied automatically and reported in the result. | Partly supported. Typically handled by the analyst. | Partly supported. Typically handled by the analyst. |
| Warnings when a model is outside its calibration | Supported. Flags out-of-range estimates and disagreeing estimators. | Partly supported. Fit statistics only. | Partly supported. Fit statistics only. |
| Region segmentation and clustering | Supported. Film, damage halo, substrate, and data-driven clusters. | Supported. Cluster and multivariate analysis in most suites. | Partly supported. Scripted. |
| Correlating Raman with PL or AFM of the same field | Supported. Registers exports across instruments and correlates pixels. | Partly supported. Usually within one instrument’s data. | Not supported. Manual alignment. |
| Reading files from a mixed instrument fleet | Partly supported. LabSpec natively; others through text/CSV export or guided parsing. | Partly supported. Own format first; others through import filters. | Partly supported. Through text/CSV export. |
| Results tied to samples and process history | Supported. Fits, maps, and estimates are stored as structured outputs linked to the file, parameters, sample, and process run. | Partly supported. Project files on the workstation; sample context kept separately. | Partly supported. Saved project files; sample context kept by hand. |
| Instrument control and acquisition | Not supported. Not an acquisition tool. Keep your instrument software for this. | Supported. Core purpose. | Not supported. Not offered. |
- Supported
- Partly supported
- Not supported
Reflects typical, publicly documented workflows in each category as of September 2026. Capabilities vary by product, version, and add-on module. Product names are trademarks of their respective owners.
Where Atlas stops
A result is only useful if you know its limits. These are the current boundaries, stated in the product the same way.
- Acquisition, instrument control, and spectrometer calibration stay in your vendor software.
- Atlas does not count layers once the mode-separation calibration saturates in the multilayer regime; it says so instead.
- Defect-species rankings are marked qualitative when the estimated density falls outside the classifier’s calibration range.
- Absolute doping is not measured, and borrowed strain or doping coefficients are flagged as borrowed.
Files Atlas reads
- Horiba LabSpec maps (.txt)
- CSV and TSV spectral exports
- Two-column single spectra
- Multi-page TIFF hyperspectral stacks
Worked examples
Material spotlight · 12 min read
What Atlas can do with WS₂
A 130 × 130 µm co-registered Raman and PL map of a PFIB-patterned WS₂ flake: segmentation, two independent defect-density estimators, strain–doping decomposition, exciton/trion fitting, edge profiles, and cross-modal correlation — with every calibration caveat Atlas raised.
Material spotlight · 11 min read
What Atlas can do with MoS₂
ALD versus sputtered MoS₂ films across four process conditions: what Raman linewidth, peak separation, strain–doping spread, and defect-density estimators say about disordered large-area films, and how they track FET on/off ratio.
Technical note · 7 min read
How a Raman linewidth becomes a defect density
Inside the simulation-to-calibration pipeline: MLIP phonons, defect supercells, and an inversion that states its own limits.
Technical note · 6 min read
How Raman spectroscopy works
An interactive walkthrough: photon scattering, why the shift is a chemical fingerprint, and what peak positions, widths, and the anti-Stokes ratio tell you.
Questions
- Which Raman file formats does Atlas read?
- Horiba LabSpec maps are parsed natively, along with CSV/TSV spectral exports, two-column single spectra, and multi-page TIFF hyperspectral stacks. Files from other vendors can be exported to text, or decoded through the agent-guided parsing path, which can turn a one-off decode into a reusable parser.
- Does Atlas replace LabSpec or WiRE?
- No. Acquisition, instrument control, and spectrometer calibration stay in your vendor software. Atlas sits downstream: it turns the exported map into defect, strain, and doping results, and keeps them linked to the sample and process that produced them.
- Is it only for 2D materials?
- TMDs are where Atlas started, but dedicated workflows also cover graphene, carbon nanotubes, and GaN stress and doping, and the generic peak-fitting tools work on any Raman spectrum.
Bring your Raman data
Send us a file from your own instrument. We’ll run it through Atlas and walk you through every number and caveat it reports.