SpectralBench
Automatic peak detection and material assignment for Raman spectra
Identifying materials from a Raman spectrum typically involves comparing peak positions against reference databases, spectral atlases, or published literature — a slow process that demands experience with Raman shift tables and knowledge of which bands belong to which materials.
SpectralBench automates Raman peak identification: upload your Raman data, and the peak identifier detects scattering bands, matches them against a curated database of over 65 Raman band entries, and returns ranked assignments with confidence scores. The tool covers organic compounds, polymers, minerals, carbon materials, and semiconductors — all without leaving your browser. Once you have identified your peaks, use the unit converter to express Raman shifts in wavelength or energy units, or the preprocessing tools to remove fluorescence background before analysis.
Upload a Raman spectrum in any supported format — JCAMP-DX, SPC, Bruker OPUS, CSV, or TXT. SpectralBench parses the file client-side, automatically detects that the spectrum is Raman based on the spectral range and characteristics, and renders it on an interactive chart. The peak detector applies a prominence-based algorithm to locate scattering bands, filtering out baseline noise and fluorescence artifacts based on your sensitivity settings.
Each detected peak is matched against a curated database of over 65 Raman band entries spanning alkanes, aromatics, carbonyls, amines, sulfur compounds, phosphates, carbon materials (D, G, 2D bands), minerals, oxides, and semiconductors. Assignments are ranked by confidence — high, medium, or low — based on how well the observed Raman shift matches the expected ranges for each material or functional group.
Click any peak in the results table to highlight it on the chart. Expand a row to see alternative assignments and overlapping bands. Adjust the prominence threshold to detect weaker Raman bands, or reduce the max peaks count to focus on the strongest scattering features. All processing runs in your browser; your spectral data is never uploaded.
Raman spectroscopy measures inelastic scattering of monochromatic light (typically from a laser) by molecular vibrations. When a photon interacts with a molecule, it can lose or gain energy corresponding to a vibrational mode — producing Stokes or anti-Stokes Raman scattering. The frequency shift between the incident and scattered light, measured in wavenumbers (cm⁻¹), is characteristic of specific molecular bonds and crystal structures.
Unlike FTIR, which measures absorption of infrared light by molecular dipole changes, Raman spectroscopy is sensitive to polarizability changes. This makes Raman particularly strong for detecting symmetric vibrations (C=C, S-S, C-C backbone), non-polar bonds, and lattice modes of crystalline materials — bonds that are often weak or invisible in infrared spectra. The two techniques are complementary: bands strong in Raman tend to be weak in IR, and vice versa.
SpectralBench's Raman database covers the most commonly encountered bands in organic chemistry, polymer science, mineralogy, carbon nanomaterials, and semiconductor characterization. Each entry includes the expected Raman shift range, relative intensity, band shape, and a description of the vibration to help with interpretation.
The Raman peak database covers a wide range of materials and compound classes encountered in research and industrial laboratories:
The reference table below summarizes key Raman shift ranges used in peak identification. SpectralBench's full database contains over 65 entries with detailed sub-ranges, intensity descriptors, and band shape information.
| Band | Raman Shift (cm⁻¹) | Material / Group |
|---|---|---|
| TiO₂ Anatase | 144 | Very strong, sharp — anatase polymorph fingerprint |
| Quartz (SiO₂) | 464 | Si-O-Si symmetric stretch — mineral identification |
| Phenylalanine | 1003 | Ring breathing mode — biological tissue marker |
| Carbonate (CO₃²⁻) | 1085 | Symmetric stretch — calcite, aragonite, dolomite |
| Carbon D Band | 1350 | Disorder/defect band in sp² carbon materials |
| Carbon G Band | 1580 | E₂g mode — graphene, graphite, CNTs, carbon fibers |
| C=C Stretch | 1620–1680 | Alkenes — strong Raman due to high polarizability |
| C-H Stretch | 2800–3100 | CH₃ and CH₂ symmetric/asymmetric stretches |
| Feature | SpectralBench | RRUFF | OpenSpecy |
|---|---|---|---|
| Auto peak detection | Yes | No | Limited |
| Material assignment | Automatic | Manual lookup | Spectral matching |
| Runs offline | Yes (client-side) | No | No |
| Data privacy | Never uploaded | N/A (no upload) | Uploaded to server |
| Material coverage | Organic, mineral, carbon, semiconductor | Minerals only | Polymers, minerals |
| Confidence scores | Yes | No | Correlation scores |
| Cost | Free | Free | Free |
Upload your Raman data file to SpectralBench in any supported format (JCAMP-DX, SPC, OPUS, or CSV). The tool automatically detects that your spectrum is Raman based on the spectral range, then matches detected peaks against a curated database of over 65 Raman band entries covering organic compounds, polymers, minerals, carbon materials, and semiconductors. Results are ranked by confidence so you can focus on the most likely assignments first.
A strong Raman band near 1580 cm⁻¹ is the G band (graphitic band) of carbon materials. It arises from the E₂g vibrational mode of sp² carbon in graphene, graphite, carbon nanotubes, and carbon fibers. Together with the D band near 1350 cm⁻¹, the G/D ratio is widely used to assess the crystallinity and defect density of carbon materials.
Raman and FTIR are complementary techniques based on different selection rules. Raman scattering is sensitive to polarizability changes (symmetric vibrations, non-polar bonds like C=C, S-S), while FTIR absorption requires dipole moment changes (polar bonds like O-H, C=O). SpectralBench uses separate databases for each modality and automatically detects which one to apply based on your spectrum.
Yes. The Raman database includes entries for common minerals and inorganic materials — carbonates (calcite, aragonite near 1085 cm⁻¹), silicates (quartz at 464 cm⁻¹), oxides (TiO₂ anatase at 144 cm⁻¹, rutile at 447 cm⁻¹), and lattice modes of crystalline solids. Upload your mineral Raman spectrum and get automatic peak assignments.
Automated identification is highly accurate for well-resolved peaks from pure compounds and known materials. Complex mixtures, fluorescence background, and low signal-to-noise spectra may produce multiple candidate assignments — SpectralBench shows all possibilities ranked by confidence so you can apply your domain knowledge to select the best match. Preprocessing tools like baseline correction can improve results for fluorescent samples.
RRUFF is an excellent mineral Raman database but requires manual peak-by-peak lookup and focuses on mineralogy. OpenSpecy offers spectral matching but requires an internet connection and uploads your data to a server. SpectralBench combines automated peak detection with material assignment in a single workflow, runs entirely in your browser (no data upload), and covers organic compounds, polymers, carbon materials, and semiconductors in addition to minerals.
The database covers Raman shifts from approximately 100 cm⁻¹ to 3100 cm⁻¹, spanning lattice modes and metal oxides (100–600 cm⁻¹), fingerprint vibrations (600–1500 cm⁻¹), double bonds and aromatic rings (1500–1700 cm⁻¹), and C-H / O-H stretching (2800–3100 cm⁻¹). Over 65 entries cover alkanes, aromatics, carbonyls, amines, sulfur compounds, phosphates, carbon materials, semiconductors, and more.
Identify functional groups from FTIR spectra with 80+ database entries.
Match your Raman spectrum against a reference library with HQI scoring.
Baseline correction and fluorescence removal to improve Raman peak detection.
Overlay your Raman spectrum against references with difference and mirror views.