Most stabiliser mistakes in purchasing are not pricing mistakes, they are category mistakes. A buyer searches for "antioxidant", compares price per kilogram, and orders whichever grade meets the assay at the lowest cost. Six months later a part has ozone cracks, a white sole has turned cream, or a copper conductor has cooked its own insulation from the inside out. None of those are antioxidant failures in the ordinary sense. They are failures of the wrong chemistry applied to the wrong failure mode.
Rubber and plastics do not age for one reason, they age for five - thermal oxidation, ozone attack, flex fatigue, metal-ion catalysis, and plain physical loss of the additive itself. Each is governed by a different chemistry, and no single molecule handles more than two of them well. This guide maps those five failure modes onto the sixteen stabiliser grades we stock, with the loadings and the goods-in specifications that decide whether protection is still there in year five.
Five failure modes and the chemistry that answers each
The first question is never "how much antioxidant", it is "what is going to kill this part". The answer fixes the chemical family before any dosage discussion begins.
| Failure mode | What drives it | Chemistry that answers it |
|---|---|---|
| Thermal-oxidative ageing | Heat plus oxygen, hydroperoxide chain branching | Hindered phenol as chain breaker, plus phosphite or thioester as peroxide decomposer |
| Ozone cracking | Ozone attack on unsaturated backbones under strain | p-Phenylenediamine antiozonants, consumed sacrificially at the surface |
| Flex fatigue and dynamic crack growth | Repeated deformation with oxygen and ozone acting together | p-Phenylenediamine, with microcrystalline wax as a static second line |
| Metal-ion catalysis | Copper, manganese or iron residues decomposing hydroperoxides | Hydrazide metal deactivators, and mercaptobenzimidazole chelation |
| Loss of the additive itself | Volatilisation, extraction, blooming, migration into contact media | Higher molecular weight, lower polarity, non-migrating structure |
Thermal-oxidative ageing: one primary and one secondary, never one alone
Thermal oxidation runs as a chain reaction. Heat and shear generate a radical, the radical takes up oxygen to form a peroxyl radical, the peroxyl radical abstracts hydrogen and leaves a hydroperoxide plus a fresh radical, and the cycle repeats. A hindered phenol breaks the cycle by donating hydrogen to the peroxyl radical and forming a stable phenoxyl radical, which is why it is called a primary or chain-breaking antioxidant.
A phosphite or a thioester attacks the same problem from the other end. It reduces the hydroperoxide itself to a harmless alcohol before it can split into two new radicals. That is the secondary or preventive role. A phenol used alone accumulates hydroperoxides it cannot remove; a phosphite used alone cannot stop the radicals already running. The pair is standard practice, not a premium option.
For polyolefins the working total is 0.05 to 0.3 percent stabiliser, usually split so that the phenolic fraction carries long-term protection and the phosphite fraction carries melt stability through extrusion. Our phosphites divide that work differently:
- Antioxidant 618 (CAS 3806-34-6), a distearyl pentaerythritol diphosphite of MW 733, for colour and processing stability in PE, PP, PS, ABS, EVA, PC and PVC, normally blended with a hindered phenol
- Antioxidant 9228 (CAS 154862-43-8), a high-molecular-weight diphosphite of MW 853 for phenol-free PP, multi-pass extrusion PE, PC, PET, PA and PES, chosen when colour retention outweighs cost
- Antioxidant TNPP (CAS 26523-78-4), a liquid triaryl phosphite for PE, PP, polystyrene, SBR and ABS, and for suppressing gel formation during polymerisation, with FDA clearance under 21 CFR 178.2010 and 177.2600
- Antioxidant TTDP (CAS 77745-66-5), the phenol-free liquid trialkyl phosphite discussed in the regulatory section below
- Antioxidant L-20, a high-molecular-weight phenol-free liquid phosphorus blend for anti-yellowing in PU foams, TPU coatings, epoxy resins and shoe materials, supplied without a single CAS number because it is a formulated blend
For the phenolic half, the choice is decided more by where the part will spend its life than by any difference in radical-trapping chemistry:
- Antioxidant 330 (CAS 1709-70-2), MW 775, a three-armed hindered phenol for PP, HDPE, spandex fibres, PA and specialty rubbers, chosen where extraction and migration resistance matter
- Antioxidant 3114 (CAS 27676-62-6), MW 784, a hindered phenol on a triazine isocyanurate core for long-term heat ageing of PP, PE, PA, PS, ABS, PVC and polyesters, US food-contact approved
- Antioxidant 1098 (CAS 23128-74-7), MW 637, the first-choice stabiliser for polyamide 6 and 66 mouldings, fibres and films, and a good fit for POM, polyesters and polyurethanes
- Antioxidant 245 (CAS 36443-68-2), a lower-molecular-weight asymmetric hindered phenol at MW 587 for styrenics including HIPS, ABS, MBS and SBR, plus POM, PA, TPU and spandex
- Antioxidant 535 (CAS 125643-61-0), a liquid hindered phenol of MW 391 for storage stabilisation of polyether polyols and anti-scorch in flexible PU foam, where a solid would be awkward to meter and slow to dissolve
- Antioxidant 445 (CAS 10081-67-1), a diphenylamine antioxidant of MW 406 that carries heat-ageing protection across a very wide rubber set - NR, BR, SBR, NBR, CR, EPDM, ACM and HNBR - and is equally at home in PE, PP, PS, PU foam, adhesives and lubricants
Ozone and flex fatigue: only p-phenylenediamines will do
Ozone attacks the double bonds of unsaturated rubbers such as NR, SBR, BR and NBR, and it does so far more aggressively when the rubber is under strain. Static strain opens straight cracks perpendicular to the stress; repeated flexing grows them into the familiar chevron pattern. No hindered phenol prevents this, because the reaction is not a radical chain reaction - it is a direct addition across the double bond that phenols cannot interrupt.
p-Phenylenediamines work on a completely different principle: sacrifice. The molecule migrates continuously to the rubber surface, reacts with ozone far faster than the polymer does, is consumed in the process, and is replenished from the bulk. Protection is a diffusion-controlled supply line, and that single mechanism explains nearly every practical rule about them.
- Antidegradant IPPD (CAS 101-72-4) is the strongest antiozonant in our range. Its molecular weight of 226 keeps it mobile enough to keep feeding the surface, and at equal loading it outperforms 6PPD on ozone crack growth. Work at 0.5 to 2 phr, and up to 3 phr in heavy-duty dynamic goods.
- The price is bloom and stain. IPPD is a p-phenylenediamine, and its reaction products form a brown surface film, so it belongs on tyre sidewalls, conveyor covers and industrial hose covers - anywhere performance is visible and appearance is not. It is also a skin sensitiser, so keep it out of compounds handled bare-handed.
- Wax is not a substitute but a multiplier. Microcrystalline wax blooms to the surface and forms a physical barrier that cuts the ozone flux, slowing consumption of the IPPD beneath it. Together the two give roughly double the protection of either alone, but only under static conditions - under flexing the film fractures and the wax contribution collapses.
- Where appearance cannot be sacrificed, the answer is a non-staining system rather than a weaker antiozonant. That is the subject of the next section.
Non-staining and non-discolouring: MB and the white-goods problem
White and light-coloured articles rule out the whole amine family. A white sole, a translucent hose, a natural-rubber latex thread or a pale automotive seal cannot contain a stain-forming p-phenylenediamine, no matter how effective it is. What is left is a smaller, quieter set of chemistries - and one of them does work no phenol can do.
Antidegradant MB (CAS 583-39-1), chemically 2-mercaptobenzimidazole, is a non-staining, non-discolouring antidegradant at MW 150. It has three properties that make it hard to replace:
- Its mercapto group complexes copper and manganese ions directly, so it protects against catalytic ageing rather than just radical attack. This is why wire and cable insulation, EPDM weatherstrip and copper-contact elastomers are its natural applications.
- It decomposes hydroperoxides, which puts it in the secondary role, and it shows a genuine synergy with both amine and phenolic antioxidants.
- It retards cure. MB delays accelerators such as MBT, MBTS, TMTD and the sulfenamides, so a formulator who adds it for ageing protection and does not re-check the rheometer will find the optimum cure time has moved. Watch T90 on every compound change; this is the single most common surprise with MB.
The goods-in specification for MB is correspondingly tight, because at these loadings a fraction of a percent of impurity is a real effect:
| Test | High grade | First grade | Why it matters |
|---|---|---|---|
| Initial melting point, °C, min | 295.0 | 290.0 | Free MBT and other low-melting impurities depress the melting point; a low figure means a short or impure reaction, not a harmless variation |
| Loss on drying, percent, max | 0.30 | 0.40 | Moisture changes weigh-up accuracy and can interfere with cure |
| Ash, percent, max | 0.30 | 0.40 | Inorganic residue, mostly catalyst; shows up as specks in translucent goods |
| Residue on 150 micrometre sieve, percent, max | 0.00 | 0.00 | MB is used at low loadings; oversize particles never disperse and become visible defects |
Molecular weight decides whether the additive is still in the part
Three mechanisms remove a stabiliser from a polymer without any chemistry happening at all: it evaporates during high-temperature processing, it is extracted by water, oil or fuel in service, or it migrates to the surface and blooms. All three are diffusion-limited, and diffusion scales with molecular size. The same functional chemistry therefore behaves completely differently depending on how large the molecule carrying it is.
| Grade | CAS | Molecular weight | Practical consequence |
|---|---|---|---|
| BHT, food grade | 128-37-0 | 220 | Deliberately volatile and mobile. Excellent for oils, fats and fuel, and for headspace protection in packaging; a poor choice for a solid part that must last five years |
| Antioxidant 535 (liquid) | 125643-61-0 | 391 | Liquid phenolic for polyols and PU foam; metered as a liquid and dispersed without melting |
| Antioxidant 445 | 10081-67-1 | 406 | Amine antioxidant for heat ageing; broad rubber compatibility, moderate permanence |
| Antioxidant 1024 | 32687-78-8 | 553 | Metal deactivator plus phenol; the lower melting point of the two deactivators, so easier to disperse, slightly more prone to surface loss |
| Antioxidant 245 | 36443-68-2 | 587 | Balanced permanence for styrenics and polyurethanes |
| Antioxidant 1098 | 23128-74-7 | 637 | Amide-linked phenol; the amide groups also chelate metal traces, which is why it is the default for polyamide |
| Antioxidant MD-697 | 70331-94-1 | 697 | Higher molecular weight than 1024, so better processing retention; different melting point, so check dispersion separately |
| Antioxidant 618 | 3806-34-6 | 733 | High-molecular-weight solid phosphite; resists loss during compounding |
| Antioxidant 330 | 1709-70-2 | 775 | High permanence in extraction-heavy service |
| Antioxidant 3114 | 27676-62-6 | 784 | High permanence, food-contact approved |
| Antioxidant 9228 | 154862-43-8 | 853 | The highest molecular weight phosphite in the range; very low volatility, chosen for colour-critical high-temperature processing |
Two practical consequences follow. First, molecular weight is a sourcing parameter, not a footnote: if a supplier quotes the same chemistry at a materially different molecular weight, that is a different product with a different service life. Second, the food-grade version of a molecule is not a purer version of the same thing - BHT food grade (CAS 128-37-0) is chemically the same molecule as technical BHT, and the difference is entirely the compliance package behind it: INS 321 and E321, FDA 21 CFR 172.115, a JECFA acceptable daily intake of 0 to 0.3 mg per kg body weight, and an EFSA figure of 0.25 mg per kg body weight per day. Same CAS, two different sets of paperwork, and the paperwork is what an auditor reads.
The regulatory layer: two liquid phosphites changing status
For twenty years the default liquid phosphite has been TNPP, tris(nonylphenyl) phosphite (CAS 26523-78-4). That position is now under pressure for a reason that has nothing to do with its performance. In January 2025 the substance was added to the ECHA Candidate List of substances of very high concern, on the grounds that it meets the criteria for an endocrine disruptor to the environment when it contains 0.1 percent or more of branched and linear 4-nonylphenol. Nonylphenol itself is already restricted under REACH Annex XVII.
The procurement detail that matters is that the restriction is tied to a residual, not to the molecule as a name. Commercial TNPP specifications commonly allow up to 3.0 percent free nonylphenol, so two drums sold under the same CAS number can carry materially different regulatory exposure. If you are in a market that is watching this listing, ask for the free nonylphenol figure on the certificate of analysis rather than the product name on the label.
Antioxidant TTDP (CAS 77745-66-5), triisotridecyl phosphite, is the phenol-free answer. It is a trialkyl phosphite rather than a triaryl one, with no nonylphenol in the structure, and it is positioned as a direct swap:
- Phosphorus content 4.9 percent against 4.30 percent for TNPP, so less material does the same job
- Colour of 30 APHA or better, acid value of 0.50 mg KOH/g or below, refractive index 1.4550 to 1.4650 at 25 °C and specific gravity 0.882 to 0.900
- Low volatility and good hydrolysis resistance, the two weaknesses that most often decide whether a phosphite survives a compounding line
- Typical use at 0.1 to 2.0 percent across rigid and flexible PVC, ABS, SBS, SIS, SEBS, SBR and PU foam, where it also suppresses the nitrogen-oxide yellowing that shows up in foam and coatings
Antioxidant L-20 covers the same phenol-free ground from the other direction - a high-molecular-weight liquid phosphorus blend, used for anti-yellowing in PU foam and shoe materials, TPU coatings and binders, and epoxy resins and curing agents.
What belongs on the purchase order
Stabilisers fail at goods-in more often than they fail in service, and the reason is that the specification is written as a name and a percentage instead of a set of numbers. Three groups of tests cover nearly everything:
- Identity and purity. Initial melting point and assay by HPLC or GC. For MB, the melting point is the fastest read on purity, because low-melting residues pull it down immediately.
- Volatile and inorganic residues. Loss on drying and ash. Both are cheap tests and both predict dispersion and speck problems.
- Physical form. Residue on a 150 micrometre sieve for powders, and for liquid phosphites the colour in APHA, the acid value, the phosphorus content, the refractive index and the specific gravity. For TNPP, free nonylphenol as well.
Two tests that are specific to this family are worth adding. Metal deactivators are bought for their chelation behaviour, so a copper-contact service life test belongs in qualification even though it is not on the certificate. And MB is bought for non-staining protection, so every compound change should re-check T90 on the rheometer, because MB retards cure and a small assay shift moves the optimum.
Selection in one page
| If the part fails by | Lead chemistries | Typical loading | Watch out for |
|---|---|---|---|
| Heat, long-term in a polyolefin | 330, 3114, 1098, plus 618 or 9228 as secondary | 0.05 to 0.3 percent total | Total loading, not phenolic loading, drives cost |
| Heat, in a polyamide | 1098, with a phosphite for processing | 0.1 to 0.5 percent | Amide chemistry is preferred over simple esters in PA |
| Ozone cracking on a dynamic rubber | IPPD, with microcrystalline wax | 0.5 to 2 phr, up to 3 phr | Stain, bloom, skin sensitisation |
| Colour and translucency must survive | MB, with phenolic and phosphite partners | 0.5 to 2 phr in rubber | Cure retardation - check T90 |
| Copper or metal contact | 1024 or MD-697, or MB | 0.1 to 0.2 percent | Dispersion differs between the two deactivators |
| Fuel, oil, water or steam contact | 330, 3114, 9228 | 0.1 to 0.5 percent | Extraction, not volatility, is the failure route |
| Liquid system: polyol, foam, coating, adhesive | 535, TNPP replaced by TTDP, L-20 | 0.1 to 2.0 percent | Nonylphenol content if TNPP is retained |
Grades we supply
We supply all sixteen stabilisers discussed above. Quotations are written against the specification you send us, so the melting point, the assay, the sieve residue and - for the liquid phosphites - the colour, acid value and free nonylphenol figures are all confirmed in writing before shipment.
- Antidegradant IPPD (CAS 101-72-4), p-phenylenediamine antiozonant for dynamic NR, IR, SBR, BR and NBR goods, tyres, cables and mountings
- Antidegradant MB (CAS 583-39-1), non-staining mercaptobenzimidazole for heat ageing and copper protection in wire, cable, latex and light-coloured goods, also a heat stabiliser for PP and PE
- Antioxidant 445 (CAS 10081-67-1), diphenylamine antioxidant for heat-ageing protection across NR, BR, SBR, NBR, CR, EPDM, ACM and HNBR, and for PE, PP, PS, PU foam, adhesives and lubricants
- Antioxidant 1024 (CAS 32687-78-8), hindered phenol plus hydrazide metal deactivator for XLPE and PE cable insulation, copper-contact polyolefins, NBR fuel hoses, PA, POM and EPDM; FDA 21 CFR 178.2010
- Antioxidant MD-697 (CAS 70331-94-1), higher-molecular-weight metal deactivator for wire and cable insulation, metal-contact parts and long-term thermal stabilisation of polyolefins, polyesters, PA and PU
- Antioxidant 330 (CAS 1709-70-2), high-molecular-weight hindered phenol for PP, HDPE, spandex fibres, PA and specialty rubbers
- Antioxidant 3114 (CAS 27676-62-6), triazine-core hindered phenol for long-term heat ageing of PP, PE, PA, PS, ABS, PVC and polyesters; US food-contact approved
- Antioxidant 245 (CAS 36443-68-2), asymmetric hindered phenol for HIPS, ABS, MBS, SBR, POM, PA, TPU, spandex, RIM and PVC
- Antioxidant 1098 (CAS 23128-74-7), first-choice polyamide stabiliser for PA6 and PA66 mouldings, fibres and films, also POM, polyesters, polyurethanes and adhesives
- Antioxidant 535 (liquid) (CAS 125643-61-0), liquid hindered phenol for polyether polyol storage stabilisation and anti-scorch in flexible PU foam, elastomers and lubricants
- Antioxidant 618 (CAS 3806-34-6), diphosphite secondary antioxidant for colour and processing stability in PE, PP, PS, ABS, EVA, PC and PVC
- Antioxidant 9228 (CAS 154862-43-8), high-molecular-weight diphosphite for phenol-free PP, multi-pass extrusion PE, PC, PET, PA and PES
- Antioxidant TNPP (liquid) (CAS 26523-78-4), liquid aryl phosphite for PE and PP processing, PA6 molecular weight maintenance, adhesives and elastomer polymerisation; FDA 21 CFR 177.2600
- Antioxidant TTDP (liquid) (CAS 77745-66-5), phenol-free liquid trialkyl phosphite as a direct TNPP replacement for rigid and flexible PVC, ABS, SBS, SEBS, SBR, PU foam, lubricants and wire and cable compounds
- Antioxidant L-20, high-molecular-weight phenol-free liquid phosphorus blend for anti-yellowing in PU foam and shoe materials, TPU coatings and binders, epoxy resins and curing agents
- Antioxidant BHT, food grade (CAS 128-37-0), food-grade hindered phenol for fats, oils, baked and fried foods under FDA 21 CFR 172.115 and INS 321, and for food-contact packaging plastics
For further reading, our earlier guides cover the functional split between stabiliser classes in polymer antioxidants, the light-stabiliser side in HALS grades, the absorber side in replacing UV-328, and the acid scavengers that sit alongside these systems in metallic stearates.
Send us the failure mode, the polymer, the service temperature and whether the part touches metal, food or skin, and we will propose the family, the loading and the goods-in specification as one package.