Why a score changes
Most revisions come from one of four things: a better source for the molecular formula or mass; a clearer picture of which feedstock a commercial grade is actually made from; a synthesis route that turns out to be run two different ways in industry; or an error in how the method was applied. The method itself is fixed and published on how we calculate NOI — what changes is what we know about the molecule.
Entries that have been checked by hand, fragment by fragment, are marked as hand-audited on the ingredient’s own page, so you can tell a reviewed value from a computed one at a glance. Everything else has been produced by the documented method and reviewed, but not individually re-derived.
Review conventions
These are the judgement calls that come up again and again. They are applied consistently across the whole database, so an entry that looks surprising is usually one of these rather than an oversight.
- Mineral counter-ions count as natural origin. The metal or mineral part of a salt — the sodium, potassium, magnesium, calcium or zinc — comes from a mined or brine source, not from petrochemistry, so its mass counts as natural origin.
- Ammonium counter-ions are petrochemical. Ammonia is made industrially from hydrogen out of natural gas, so an ammonium ion carries a petrochemical mass, unlike the mineral counter-ions above.
- Silicones score 0. Silicon for silicone chemistry is refined from quartz, but the route to it is an industrial synthesis that has no natural-origin pathway under this method, so silicones are scored 0 by convention across the database. The one exception is protein carried by a silicone: hydrolysed silk or collagen bound into a silicone polymer keeps its natural-origin mass, and only the silicone part scores 0.
- Peptides are natural origin. A peptide is built from amino acids made by fermentation or protein hydrolysis, so its amino-acid units count as natural origin, beta-alanine included, even though bulk beta-alanine is mostly made synthetically. Amino acids that are only ever made synthetically, such as norleucine, are the exception, and caps added from petrochemical reagents, such as acetyl groups, still count as petrochemical.
- Two commercial routes means route-dependent. Where an ingredient is genuinely made both from a natural feedstock and from a petrochemical one, and the label cannot tell you which you have, the entry is marked route-dependent and shown as a range rather than as a single number you could not rely on. Ask your supplier which route their grade uses.
- Odd-carbon oxo-alcohol blends are petrochemical. Alcohols with odd carbon numbers, and the isomer blends sold alongside them, are the signature of the oxo process running on petrochemical olefins; natural fatty chains are even-numbered. Those fragments are counted as petrochemical.
- Acetyl, methyl and formyl groups are petrochemical. These small capping groups are added from acetic, formic and methylating reagents made industrially from petrochemical feedstocks, so their mass is counted as petrochemical even when the rest of the molecule is plant-derived.
- Terpenes from pine, citrus and essential oils are natural origin. Turpentine terpenes — pinene, camphene and the myrcene made from them — citrus limonene and the terpenes of essential oils are plant feedstocks, so the camphor, linalool, geraniol, citronellal, menthol and other aroma chemicals made from them keep that natural-origin mass. Terpenoids assembled from petrochemical building blocks, such as citral made from isobutylene or linalool made from acetone and acetylene, are petrochemical; where both kinds are sold, the entry is route-dependent.
Changelog
— Peptide and protein-silicone rulings, and 256 new ingredients
Two rulings, applied across the database: peptides count as natural origin even where they contain beta-alanine, and protein bound into a silicone keeps its natural-origin mass. Both are set out in the conventions above.
| Ingredient | Was | Now | Why |
|---|---|---|---|
| Carnosine | 0.61 | 1.00 | Its beta-alanine is an amino acid like its histidine. Anserine and Acetyl Carnosine already treated it that way. |
| Polysilicone-14 | 0 | 0.70 | Its hydrolysed silk keeps its natural-origin mass; only the silicone part scores 0. The supplier does not publish the proportions, so the value follows the patent for this kind of polymer and is marked low confidence. |
| Polysilicone-16 | 0 | 0.70 | As above, for hydrolysed collagen. |
- 256 new ingredients, among them 34 peptides and some 200 plant extracts, oils and ferments; the database now holds 7,145.
- Helianthus Annuus Petal Extract showed the official description of the whole-plant extract; it has been removed.
— Terpene and vanillin rulings, text corrections and reaction films
Terpenes made from pine turpentine — myrcene, pinene, camphene and the aroma chemicals built from them — now count as natural origin across the whole database, as set out in the convention above. Many entries already treated them that way; the ones below had counted them as petrochemical.
| Ingredient | Was | Now | Why |
|---|---|---|---|
| Tetramethyl Acetyloctahydronaphthalenes (Iso E Super) | 0 | 0.58 | Its myrcene half comes from pine turpentine; only the small ketone part is petrochemical. |
| Hydroxyisohexyl 3-Cyclohexene Carboxaldehyde (Lyral) | 0 | 0.73 | Its myrcenol part is pine-derived myrcene plus water. |
| 3-Benzylidene Camphor | 0 | 0.63 | Camphor is made from pine α-pinene; only the benzylidene ring is petrochemical. |
| 4-Methylbenzylidene Camphor | 0 | 0.59 | As above. |
| Alpha-Isomethyl Ionone | 0 | varies | Citral from litsea, lemongrass or pine terpenes gives 0.66; citral made from isobutylene gives 0. |
| Hydroxycitronellal | 0 | varies | Citronellal from citronella oil or pine turpentine is natural origin. |
| Tetrahydrolinalool | 0 | varies | Linalool made from pine α-pinene is natural origin. |
- The route tables for Geraniol, Menthol, Linalool, Citral, Citronellol and fifteen of their esters and relatives listed pine- or myrcene-derived material as petrochemical. They now carry that route at its natural value; the headline scores are unchanged.
- The descriptive text for Isopropyl Myristate, Ceramide NP, Sodium Lauroyl Methyl Isethionate and PEG-100 Stearate described their fatty acids or phytosphingosine as petrochemical, contradicting their own scores. The scores were right; the text now matches them.
- Vanillin is now route-dependent. Vanillin made from wood lignin, by fermentation of ferulic acid or eugenol, or from vanilla beans is natural origin; vanillin built from petrochemical guaiacol is not. Vanillyl Butyl Ether (0.65 with natural vanillin) and Veratraldehyde (0.91) follow it, and Ferulic Acid gained a lignin-vanillin route (0.70).
- Two new entries: Vanillyl Butyl Ether and Methyl Dihydrojasmonate (Hedione).
- Reaction-mechanism films (“How it’s made”) were added to 93 ingredient pages; each one ends on that ingredient’s score.
— Database expansion
- The database was expanded to 2,792 entries, widening coverage well beyond the original set of commonly queried ingredients.
- Hub pages were added so ingredients can be browsed by function — emulsifiers, preservatives, humectants and the rest — alongside the full database and the A–Z index.
- Internal scoring jargon was removed from public text. Rationales that referred to numbered rules or to the calibration set now read as plain language; the underlying values are unchanged.
— Methodology rewrite and audit
The scoring method was rewritten and the original 116 entries were re-derived by hand against it. 47 entries were corrected. Most of the corrections ran in the same direction: mineral and peptide ingredients that had been scored 0 were in fact natural origin, and several partly natural molecules had been rounded to a whole number rather than split by mass.
| Ingredient | Was | Now | Why |
|---|---|---|---|
| Copper Tripeptide-1 | 0 | 1.0 | A peptide with a mineral counter-ion; both parts are natural origin. |
| Isopropyl Myristate | 0 | 0.78 | The myristoyl chain is natural; only the isopropyl fragment is petrochemical. |
| Polysorbate 20 | 0.5 | 0.28 | Scored by mass: the sorbitan and lauroyl core against twenty ethylene oxide units. |
| Titanium Dioxide | 0 | 1.0 | A mined mineral, not a petrochemical product. |
| Zinc Oxide | 0 | 1.0 | As above — mineral origin. |
Five of the forty-seven, as examples. Each corrected entry carries its own arithmetic on its ingredient page.
Found something wrong?
Send the ingredient, what looks wrong and a source, and it will be checked. Contact us — reported errors are corrected and the correction is logged on this page.