Spectroscopy BENCHMARK
UV-Vis absorption: 6.2% mean error across 50 molecules with topology-aware FLUX formulas. IR: <1% error on 32 NIST molecules. NMR: 0.3–0.5 ppm MAE. Full methodology published.
UV-Vis absorption: 6.2% mean error across 50 molecules with topology-aware FLUX formulas. IR: <1% error on 32 NIST molecules. NMR: 0.3–0.5 ppm MAE. Full methodology published.
How FluxMateria predicts spectral properties from first principles
Seven FLUX formulas dispatch based on chromophore topology. All derived from first-principles geometry.
Vibrational frequencies from harmonic force constants derived from FLUX bond energies. 32 NIST reference molecules covering all major functional groups. No empirical scaling factors.
Chemical shifts from electronic shielding environments. 10 SDBS reference molecules with ¹H, ¹³C, ¹&sup9;F, ³¹P, and ¹¹B nuclei. 0.3–0.5 ppm MAE.
All molecules, predictions, and experimental references
| # | Molecule | SMILES | Category | Predicted (nm) | Experimental (nm) | Error | Status |
|---|---|---|---|---|---|---|---|
| 1 | Benzene | c1ccccc1 | fused_aromatic | 253.1 | 254 | 0.4% | PASS |
| 2 | Naphthalene | c1ccc2ccccc2c1 | fused_aromatic | 306.7 | 275 | 11.5% | FAIL |
| 3 | Anthracene | c1ccc2cc3ccccc3cc2c1 | fused_aromatic | 382.9 | 375 | 2.1% | PASS |
| 4 | Tetracene | c1ccc2cc3cc4ccccc4cc3cc2c1 | fused_aromatic | 481.4 | 473 | 1.8% | PASS |
| 5 | Pentacene | c1ccc2cc3cc4cc5ccccc5cc4cc3cc2c1 | fused_aromatic | 607.1 | 578 | 5.0% | PASS |
| 6 | Phenanthrene | c1ccc2c(c1)ccc1ccccc12 | fused_aromatic | 294.0 | 293 | 0.3% | PASS |
| 7 | Pyrene | c1cc2ccc3cccc4ccc(c1)c2c34 | fused_aromatic | 320.3 | 335 | 4.4% | PASS |
| 8 | Fluorene | c1ccc2c(c1)Cc1ccccc1-2 | fused_aromatic | 286.8 | 265 | 8.2% | PASS |
| 9 | Biphenyl | c1ccc(-c2ccccc2)cc1 | fused_aromatic | 286.8 | 250 | 14.7% | FAIL |
| 10 | Styrene | C=Cc1ccccc1 | fused_aromatic | 253.1 | 248 | 2.1% | PASS |
| 11 | p-Terphenyl | c1ccc(-c2ccc(-c3ccccc3)cc2)cc1 | fused_aromatic | 320.6 | 280 | 14.5% | FAIL |
| 12 | Chrysene | c1ccc2ccc3ccc4ccccc4c3c2c1 | fused_aromatic | 320.3 | 320 | 0.1% | PASS |
| 13 | Toluene | Cc1ccccc1 | subst_aromatic | 253.1 | 261 | 3.0% | PASS |
| 14 | Phenol | Oc1ccccc1 | subst_aromatic | 253.1 | 270 | 6.3% | PASS |
| 15 | Aniline | Nc1ccccc1 | subst_aromatic | 253.1 | 280 | 9.6% | PASS |
| 16 | Anisole | COc1ccccc1 | subst_aromatic | 253.1 | 270 | 6.3% | PASS |
| 17 | Nitrobenzene | [O-][N+](=O)c1ccccc1 | subst_aromatic | 253.1 | 269 | 5.9% | PASS |
| 18 | Chlorobenzene | Clc1ccccc1 | subst_aromatic | 253.1 | 264 | 4.1% | PASS |
| 19 | p-Nitroaniline | Nc1ccc([N+](=O)[O-])cc1 | subst_aromatic | 379.9 | 381 | 0.3% | PASS |
| 20 | 1-Naphthol | Oc1cccc2ccccc12 | subst_aromatic | 328.6 | 294 | 11.8% | FAIL |
| 21 | 2-Naphthol | Oc1ccc2ccccc2c1 | subst_aromatic | 328.6 | 328 | 0.2% | PASS |
| 22 | 1-Aminonaphthalene | Nc1cccc2ccccc12 | subst_aromatic | 340.8 | 318 | 7.2% | PASS |
| 23 | Pyridine | c1ccncc1 | heterocyclic | 253.1 | 257 | 1.5% | PASS |
| 24 | Quinoline | c1ccc2ncccc2c1 | heterocyclic | 304.7 | 313 | 2.7% | PASS |
| 25 | Isoquinoline | c1ccc2cnccc2c1 | heterocyclic | 304.7 | 317 | 3.9% | PASS |
| 26 | Indole | c1ccc2[nH]ccc2c1 | heterocyclic | 276.1 | 280 | 1.4% | PASS |
| 27 | Carbazole | c1ccc2c(c1)[nH]c1ccccc12 | heterocyclic | 292.5 | 292 | 0.2% | PASS |
| 28 | Pyrrole | c1cc[nH]c1 | heterocyclic | 207.1 | 210 | 1.4% | PASS |
| 29 | Thiophene | c1ccsc1 | heterocyclic | 238.7 | 235 | 1.6% | PASS |
| 30 | Acridine | c1ccc2cc3ccccc3nc2c1 | heterocyclic | 416.2 | 400 | 4.0% | PASS |
| 31 | Formaldehyde | C=O | carbonyl | 292.5 | 294 | 0.5% | PASS |
| 32 | Acetaldehyde | CC=O | carbonyl | 292.5 | 290 | 0.9% | PASS |
| 33 | Acetone | CC(C)=O | carbonyl | 292.5 | 280 | 4.5% | PASS |
| 34 | 2-Butanone | CCC(C)=O | carbonyl | 292.5 | 280 | 4.5% | PASS |
| 35 | Butanal | CCCC=O | carbonyl | 292.5 | 290 | 0.9% | PASS |
| 36 | Propanal | CCC=O | carbonyl | 292.5 | 290 | 0.9% | PASS |
| 37 | 3-Pentanone | CCC(=O)CC | carbonyl | 292.5 | 280 | 4.5% | PASS |
| 38 | Cyclohexanone | O=C1CCCCC1 | carbonyl | 292.5 | 285 | 2.6% | PASS |
| 39 | Benzaldehyde | O=Cc1ccccc1 | aromatic_carbonyl | 253.1 | 250 | 1.2% | PASS |
| 40 | Acetophenone | CC(=O)c1ccccc1 | aromatic_carbonyl | 253.1 | 245 | 3.3% | PASS |
| 41 | Benzophenone | O=C(c1ccccc1)c1ccccc1 | aromatic_carbonyl | 286.8 | 253 | 13.4% | FAIL |
| 42 | Anthraquinone | O=C1c2ccccc2C(=O)c2ccccc21 | aromatic_carbonyl | 286.8 | 325 | 11.8% | FAIL |
| 43 | Fluorescein | OC(=O)c1ccccc1-c1c2ccc(=O)cc2oc2cc(O)ccc12 | aromatic_carbonyl | 821.6 | 490 | 67.7% | FAIL |
| 44 | 1,3-Butadiene | C=CC=C | conjugated | 207.5 | 217 | 4.4% | PASS |
| 45 | 1,3,5-Hexatriene | C=CC=CC=C | conjugated | 249.0 | 258 | 3.5% | PASS |
| 46 | trans-Stilbene | /C(=C\c1ccccc1)c1ccccc1 | conjugated | 286.8 | 295 | 2.8% | PASS |
| 47 | trans-Azobenzene | c1ccc(/N=N/c2ccccc2)cc1 | conjugated | 286.8 | 320 | 10.4% | FAIL |
| 48 | Cinnamaldehyde | O=C/C=C/c1ccccc1 | conjugated | 253.1 | 287 | 11.8% | FAIL |
| 49 | 1,3,5,7-Octatetraene | C=CC=CC=CC=C | conjugated | 290.5 | 290 | 0.2% | PASS |
| 50 | Vanillin | O=Cc1ccc(O)c(OC)c1 | conjugated | 375.7 | 308 | 22.0% | FAIL |
PASS = within 10% of experimental λmax. All 50 experimental test cases shown. Experimental values from NIST UV-Vis database and standard literature.
Accuracy breakdown by chromophore type
| Category | Count | Mean Error | Median Error | ≤10% Pass Rate |
|---|---|---|---|---|
| heterocyclic | 8 | 2.1% | 1.6% | 8/8 (100%) |
| carbonyl | 8 | 2.4% | 2.6% | 8/8 (100%) |
| fused_aromatic | 12 | 5.4% | 4.4% | 9/12 (75%) |
| subst_aromatic | 10 | 5.5% | 6.3% | 9/10 (90%) |
| conjugated | 7 | 7.9% | 4.4% | 4/7 (57%) |
| aromatic_carbonyl | 5 | 19.5% | 11.8% | 2/5 (40%) |
| Overall | 50 | 6.2% | 3.9% | 40/50 (80%) |
32 NIST WebBook reference molecules
10 SDBS reference molecules, 5 nuclei
How FluxMateria spectroscopy compares to established approaches
| Metric | FluxMateria | TD-DFT (B3LYP) | ZINDO | ML (GNN) |
|---|---|---|---|---|
| UV-Vis λmax Error | 6.2% | 20–40 nm | 30–50 nm | 10–20 nm |
| Time per Molecule | ~25 ms | Minutes–hours | Seconds | ~100 ms |
| Fitted Parameters | 0 | Many (functional) | ~20 | Millions |
| Training Data | None | None | None | 50K+ spectra |
| Interpretable? | Yes | Yes | Partial | No |
TD-DFT and ZINDO errors reported as absolute nm differences; FluxMateria as relative %. Both represent typical literature ranges. FluxMateria achieves competitive accuracy with no training data.
All 7 spectroscopy formulas derive from a single geometric axiom — every prediction deterministic and reproducible.
Working in the application, validation in progress
All four additional spectroscopy types are implemented and available in the FluxMateria application. Quantitative benchmarks against experimental data are in progress and will be published as validation is completed.
Honest documentation of where predictions are strongest and where gaps remain
Primary data sources for experimental validation
Generate predicted spectra for your molecules with peak assignments and full interpretability.