Mango Butter
INCI Mangifera Indica Seed Butter. Saponified with NaOH it is listed as Sodium Mangoseedate, with KOH as Potassium Mangoseedate. Also: Mangifera indica seed butter, Mango kernel fat, Mango kernel butter.
Published values
| Saponification value | 167.53–188.44 mg KOH/g |
| Used by the calculator | 178.0 mg KOH/g (midpoint) |
| SAP for NaOH | 0.1269 g/g |
| Iodine value | 44.84–59.41 g I₂/100 g |
| INS (SAP − IV) | 126 |
| Melting point | 17.64–34.32 °C |
| Unsaponifiable matter | 0.94–1.9 % |
A bar made of 100 % Mango Butter
Conventional qualities summed from the fatty-acid groups; a single-oil bar is a reference point, not a recipe.
Fatty-acid profile
| Fatty acid | Published range, % | Used | |
|---|---|---|---|
| Lauric acid | C12:0 · saturated | 4.03–9.83 | 6.9 |
| Myristic acid | C14:0 · saturated | 1.85–7.11 | 4.5 |
| Palmitic acid | C16:0 · saturated | 0.54–6.81 | 3.7 |
| Stearic acid | C18:0 · saturated | 20.11–40.43 | 30.3 |
| Oleic acid | C18:1 | 31.14–37.19 | 34.2 |
| Linoleic acid | C18:2 | 3.7–16.91 | 10.3 |
| α-Linolenic acid | C18:3 n-3 | 5.34–14.27 | 9.8 |
| Arachidic acid | C20:0 · saturated | 2.26–11.69 | 7.0 |
Where these numbers come from
Table 1 (physical properties incl. melting point), Table 2 (chemical properties: SAP, iodine, unsaponifiable matter), Table 3 (complete fatty-acid composition), all broken down by the four mango varieties studied.
Regulatory identity standard for the refined commercial product: saponification value 185-198, iodine value (Wijs) 32-57, unsaponifiable matter not more than 1.5%, butyro-refractometer reading at 40 C 43.7-51.6 (or refractive index 1.4550-1.4604). Used as corroboration only; the structured fields use Diomande et al.'s measured values.
All structured fields are from a single complete primary source, Diomande et al. 2021 (World Journal of Advanced Research and Reviews, open access), which measured four Ivorian mango varieties (Amelie, Brooks, Kent, Keitt) individually - the min/max spans the four varieties' own measured values, not a literature compilation. Table 1: melting point 17.64-34.32 C (a wide spread - two of the four varieties melt well below what is usually called a 'butter', reflecting real cultivar variation in this kernel fat). Table 2: saponification number 167.53-188.44 mg KOH/g; iodine value (Wijs) 44.84-59.41 gI2/100g; unsaponifiable matter 0.94-1.90% of total lipid. Table 3: complete fatty-acid profile (8 acids; each variety's own values sum to approximately 100%): lauric 4.03-9.83%, myristic 1.85-7.11%, palmitic 0.54-6.81%, stearic 20.11-40.43%, arachidic 2.26-11.69%, linolenic 5.34-14.27%, linoleic 3.70-16.91%, oleic 31.14-37.19%. Cross-checked against the FSSAI (India) 'Mango Kernel Fat' regulatory standard (Chapter 2.2, item 8): saponification value 185-198, iodine value (Wijs) 32-57, unsaponifiable matter not more than 1.5% - overlaps this paper's measured ranges reasonably well (the FSSAI standard governs the refined commercial product; Diomande et al. analyzed raw kernel oil from named cultivars, hence the slightly wider/lower spread). No melting point or fatty-acid table is given by the FSSAI standard (it uses a butyro-refractometer/refractive-index identity test instead), so those two fields rely on Diomande et al. alone. DISPUTED, PENDING CORROBORATION (flagged 2026-09-21): this is a single-source profile from four Ivorian cultivars, and it disagrees with the mainstream description of mango kernel fat on three fields. It reports lauric 4.03-9.83 % and myristic 1.85-7.11 %, where mango kernel fat is usually described as carrying little or none of either, and alpha-linolenic 5.34-14.27 % against a more usual figure below 2 %. Those values give the fat a cleansing number of about 11 in the soap-quality convention, which is not how mango butter behaves. Treat the profile as one laboratory's measurement of four specific cultivars rather than as a general figure for the ingredient, until a second primary source corroborates it.
This record is part of the open ingredient database (ODbL). Found an error or a better primary source? The dataset is on GitHub; corrections with a document link are welcome.