Biodiesel oxidation stability is the fuel’s resistance to degradation by atmospheric oxygen, measured as an induction period in hours at 110 °C. FAME oxidises much faster than petroleum diesel because its fatty acid chains contain carbon–carbon double bonds, and the methylene groups adjacent to those bonds readily lose a hydrogen atom to initiate a radical chain reaction. EN 14214 requires a minimum of 8 hours for B100; phenolic antioxidants extend this by interrupting the chain.
Understanding the mechanism matters because it explains why some interventions work and others do not — and why treating late costs far more than treating early.
The oxidation mechanism
Autoxidation proceeds in three stages.
Initiation. A hydrogen atom is abstracted from a bis-allylic methylene — the CH₂ group sitting between two double bonds — forming an alkyl radical. This position is vulnerable because the resulting radical is resonance-stabilised across the adjacent double bonds. Heat, light, and trace metals all accelerate initiation.
Propagation. The alkyl radical reacts with oxygen to form a peroxy radical, which abstracts hydrogen from another FAME molecule, producing a hydroperoxide and a new alkyl radical. The chain is now self-sustaining. One initiation event can consume many molecules.
Termination. Radicals combine to form stable non-radical products — but these are the oligomers, polymers and insolubles that cause the practical problems.
The relative rate of hydrogen abstraction rises steeply with unsaturation. Compared to a saturated methyl stearate baseline, methyl oleate (one double bond) oxidises roughly an order of magnitude faster, methyl linoleate (two) roughly two orders faster, and methyl linolenate (three) faster still. This is why feedstock composition dominates baseline stability.
Feedstock determines the starting point
| Feedstock | Dominant fatty acid profile | Typical untreated induction period |
|---|---|---|
| Coconut, palm kernel | Highly saturated | Well above 8 h |
| Palm, tallow | Saturated/monounsaturated | Often above 8 h |
| Rapeseed | Monounsaturated (oleic) | 4–8 h |
| Used cooking oil | Variable, often polyunsaturated | 1–5 h, highly variable |
| Soybean | Polyunsaturated (linoleic) | 1–4 h |
| Sunflower | Polyunsaturated (linoleic) | 1–4 h |
Figures are indicative. Actual results depend on processing quality, prior storage and any antioxidant naturally present — tocopherols in vegetable oils survive processing to varying degrees and contribute a natural but unreliable induction period.
Used cooking oil deserves specific attention. It arrives already partially oxidised from cooking, its composition varies batch to batch, and its polymer content is elevated before processing even begins. UCO-derived FAME is typically the hardest to stabilise and the most variable.
What oxidation does to the fuel
Degradation is progressive, and the visible symptoms appear well after the chemistry has started.
Peroxide formation — the first products. Peroxides attack elastomers and seals, and are themselves initiators for further oxidation.
Acid number rise — hydroperoxides decompose to aldehydes and then carboxylic acids. Rising total acid number corrodes fuel system metals and is a reliable early indicator. Tracking acid number across batches often reveals degradation before an induction period test is scheduled.
Viscosity increase — oligomerisation and polymerisation raise viscosity, degrading injector spray pattern and atomisation.
Gum and insoluble formation — the polymers eventually exceed solubility and precipitate. This is what plugs filters and fouls injectors. Related to filter performance: filter blocking tendency improvers.
Sediment and darkening — the visible endpoint, by which point substantial degradation has occurred.
Accelerating factors
Temperature. Reaction rate roughly doubles for every 10 °C increase. A tank in direct summer sun ages fuel dramatically faster than one in shade.
Oxygen contact. Tank breathing draws air in on every temperature cycle. High-turnover fixed-roof tanks breathe more than they appear to.
Trace metals. Copper is the strongest catalyst, iron significant, and nickel, manganese and cobalt all measurably accelerate degradation.1 They operate at very low concentrations by cycling through oxidation states, catalysing the decomposition of hydroperoxides into fresh radicals — which is why a small amount of metal can consume a large amount of antioxidant. Brass and bronze fittings anywhere in the system are a liability. Metal deactivators chelate these species and are frequently essential alongside antioxidants rather than optional.
There is a feedback loop worth understanding: oxidation raises acid number, acidic fuel corrodes copper components, dissolved copper accelerates oxidation further. Studies using copper coupons have shown that antioxidant treatment slows the release rate of copper during corrosion, meaning the antioxidant is partly protecting itself by limiting the catalyst entering the fuel.2 Breaking the loop early is far cheaper than intervening once it is running.
Water. Promotes hydrolysis of esters to free fatty acids, and supports microbial growth whose acidic metabolites accelerate degradation further. FAME is hygroscopic relative to mineral diesel, making water management harder. See bacteria management.
Light. Photo-oxidation initiates radicals directly. A minor factor in bulk storage, a significant one in sampling and laboratory handling.
Processing residuals. Monoglycerides, free glycerol, residual methanol and free fatty acids all reduce stability. Meeting the EN 14214 glyceride limits comfortably rather than marginally pays back in shelf life.
How antioxidants work
Phenolic antioxidants are chain-breaking, hydrogen-donating inhibitors. The phenolic hydroxyl group donates a hydrogen atom to a peroxy radical, converting it to a stable hydroperoxide and leaving a phenoxy radical that is resonance-stabilised and does not propagate the chain. The chain terminates.
This mechanism explains three practical realities:
Antioxidants are consumed. They are sacrificial. The induction period is essentially a measure of how long the antioxidant supply lasts under test conditions. Once depleted, oxidation proceeds at the uninhibited rate.
They prevent, they do not reverse. Adding antioxidant to fuel that has already oxidised does not remove existing peroxides, acids or polymers. It slows further degradation from wherever the fuel currently is. Early dosing is far more valuable than remedial dosing.
Response is non-linear. The first increment of antioxidant delivers more hours than the second. Doubling the dose does not double the induction period. This makes dose–response testing on your specific feedstock the only economic way to set treat rate.
Common actives
| Antioxidant | Notes |
|---|---|
| TBHQ | Generally among the most effective per unit mass in FAME; good solubility |
| Pyrogallol | Very effective; solubility and colour considerations |
| Propyl gallate | Effective; solubility limitations in some matrices |
| BHT | Widely available and inexpensive; less effective per unit mass than TBHQ in FAME |
| BHA | Moderate effectiveness |
Typical treat rates run 100–1,000 ppm depending on feedstock and target margin. Blended packages combining a primary phenolic with a metal deactivator often outperform either component alone where trace metals are present.
For a fuller treatment of one common active, see TBHQ as a fuel antioxidant.
Building a shelf-life strategy
- Establish your baseline. Test untreated FAME from your current feedstock. Without this, you are dosing blind.
- Run a dose–response curve. Test at 100, 250, 500 and 1,000 ppm. Identify the point where additional dose stops buying meaningful hours.
- Set a target with margin. For extended storage, aim for 10–12 hours at despatch rather than the 8-hour minimum, because induction period declines in storage.
- Dose at production. As early in the fuel’s life as possible.
- Address trace metals. Add a metal deactivator where copper or iron contact exists.
- Manage water. Regular tank draining, and biocide treatment where microbial activity is established.
- Monitor the trend. Track induction period and acid number across batches and through storage. The trend tells you more than any single result.
Frequently asked questions
How long does biodiesel last in storage?
Commonly quoted at around 6 months for B100 under reasonable conditions, but this varies enormously with feedstock, antioxidant treatment, temperature and water management. Well-stabilised palm-derived FAME stored cool and dry lasts considerably longer; untreated soy-derived FAME can degrade within weeks.
Why does biodiesel oxidise faster than diesel?
FAME contains carbon–carbon double bonds with vulnerable adjacent methylene groups. Petroleum diesel is predominantly saturated hydrocarbon without these reactive sites.
What is a good oxidation stability value for biodiesel?
EN 14214 requires minimum 8 hours. For material facing extended storage, 10–12 hours at despatch gives working margin.
Can you fix biodiesel that has already oxidised?
Only partially. Antioxidants inhibit further degradation but do not reverse existing acid, peroxide or polymer formation. Severely degraded fuel may require reprocessing or blending down.
Does biodiesel oxidation stability decline in storage?
Yes. The antioxidant pool depletes over time, so induction period falls throughout storage. This is why dosing to the specification minimum at despatch leads to failures downstream.
Do metal deactivators replace antioxidants?
No. They address a different mechanism — chelating catalytic metals rather than terminating radical chains. Where trace metals are present, the two are complementary and both are needed.
Where to go next
Shelf life is set by feedstock, treat rate and storage conditions together. The only way to know your effective treat rate is to test your own FAME across a dose range.
Rodanco develops biodiesel antioxidants and runs Rancimat dose–response screening in our laboratory, with metal deactivators where trace metal catalysis contributes.
See also our renewables capability and full antioxidants range.
Rodanco B.V., Alkmaar, the Netherlands. Precision chemistry for fuel producers, blenders and refiners — screened in our own laboratory, proven in field application.
Related articles
References
- Effect of Metal Contaminants and Antioxidants on the Oxidation Stability of Argemone mexicana Biodiesel — investigates Fe, Ni, Mn, Cu and Co as accelerants of free-radical oxidation. Waste and Biomass Valorization, Springer. https://link.springer.com/article/10.1007/s12649-019-00886-5
- Effect of antioxidants on oxidation stability of biodiesel derived from vegetable and animal based feedstocks — includes copper coupon corrosion testing and the effect of antioxidant on copper release rate. https://ir.xtbg.ac.cn/bitstream/353005/7911/2/Effect%20of%20antioxidants%20on%20oxidation%20stability%20of%20biodiesel%20derived%20from%20vegetable%20and%20animal%20based%20feedstocks.pdf
- ÖNORM EN 14214:2019 preview — 8 h minimum oxidation stability requirement. ANSI Webstore. https://webstore.ansi.org/preview-pages/ON/preview_ONORM+EN+14214_2019.pdf
- Solubility and Antioxidant Potential of a Pyrogallol Derivative for Biodiesel Additive. Molecules, MDPI. https://www.mdpi.com/1420-3049/24/13/2439
About this article
Written by the Rodanco technical team. Rodanco B.V. operates an in-house laboratory in Alkmaar, the Netherlands, running screening, compatibility testing and tailored formulation work, and a field service team conducting on-site trials and dosing optimisation across European and global energy markets.
Standards notice. Values quoted from CEN standards are given for guidance. Standards are periodically revised and national annexes differ. Always verify against the current published edition before contractual or specification use.
Last reviewed: August 2026