
What a compound actually contains
Silicone rubber is not sold as a resin. It is sold as a compound: a high-consistency gum mixed with filler, a structure control additive, a cure system and a short list of minor ingredients, delivered as a soft sheet or a strip that a moulding shop can cut and press.
The gum is a polydimethylsiloxane with a small fraction of vinyl side groups, typically 0.05 to 0.5 mol percent, and a molecular weight around 500,000 to 700,000. That gives a Mooney viscosity of 30 to 80, which is what makes the compound handleable on a mill and mouldable in compression or injection.
Everything else exists to turn that gum into a useful material. Fumed silica raises the strength and the hardness. A structure control additive stops the silica from continuing to thicken the compound on the shelf. The cure system decides whether it is a peroxide or a platinum addition cure, and the post-cure decides how much of the network is finished before the part ships.
Filler: the ingredient that sets hardness
Fumed silica is the default reinforcing filler, at 20 to 45 parts per hundred of gum. It is made by flame hydrolysis, it has a surface area of 150 to 300 square metres per gram, and its silanol groups hydrogen-bond to the silicone chain. That bonding is the reinforcement.
Hardness tracks the loading almost linearly at about 1.7 Shore A per part of silica, so a 50 Shore A compound carries roughly 32 parts and an 80 Shore A compound roughly 45. Beyond about 50 parts the compound becomes difficult to mix and to mould, and the tear strength has already collapsed.
The reinforcement comes with a side effect. Fumed silica keeps thickening the compound for days after mixing, a process called crepe hardening, until it becomes too stiff to mould. Structure control additives, usually a hydroxyl-terminated siloxane and a small amount of water, block the remaining silanol groups and stabilise the viscosity.
Precipitated silica is cheaper and gives a higher hardness for the same loading, but it is more hydrophilic, harder to disperse and gives a lower tear strength. It appears in cheap general purpose compounds and in oil-resistant grades, not in a seal that has to last.
A typical 50 Shore A formulation
The numbers below are a conventional peroxide-cure general purpose compound. They are not a recipe to copy, because the gum and the filler grade determine the result, but they show the proportions.
| Ingredient | Parts by weight | What it does |
|---|---|---|
| VMQ gum, 0.15 mol % vinyl | 100 | The elastomer |
| Fumed silica, 200 m2/g | 32 | Reinforcement and hardness |
| Structure control additive | 4 | Stops crepe hardening |
| Hydroxyl silicone oil | 3 | Softener, improves tear |
| Dicumyl peroxide, 40 percent | 1.5 | Cure |
| Iron oxide, heat stabiliser | 1.0 | Service life above 200 C |
| Pigment | 0.5 | Colour |
Table 1 - A conventional 50 Shore A peroxide-cure VMQ compound. The functional ingredients are the gum, the silica and the cure; the rest is what makes it last.
Two entries are worth a note. The hydroxyl silicone oil is a softener that reduces hardness without reducing the silica, and it improves tear because it lowers the compound modulus at the crack tip. The iron oxide is a heat stabiliser that scavenges the radicals which would otherwise continue to crosslink the rubber in service at 200 degrees.
Tear strength peaks in the middle of the range
Tear strength rises from about 22 kN/m at 30 Shore A to a peak of 42 at 60 Shore A, then falls back to 26 at 80. The shape surprises people who assume harder is tougher, and it has two causes.
At the soft end, the compound has too little filler to deflect a crack, and the network is weak. At the hard end, the filler loading is so high that the material is brittle: the crack runs through the compound rather than being blunted by it. The peak is where there is enough reinforcement to blunt a crack and enough rubber to deform at the tip.
If an application needs tear strength, it is a formulation problem before it is a hardness problem. A 50 Shore A compound with 3 parts of hydroxyl silicone oil and a higher vinyl content will out-tear a 50 Shore A compound with more silica and less oil, at the same hardness on the durometer.
Cure systems: peroxide or platinum
Peroxide cure is the default for moulded goods. A peroxide decomposes at 150 to 170 degrees, abstracts a hydrogen from the vinyl or the methyl group, and forms a carbon-carbon crosslink. It is cheap, tolerant of fillers and pigments, and it works in thick sections.
Platinum addition cure is the default for liquid silicone and for anything that has to be clear, odourless or food contact. A platinum catalyst adds a silicon-hydrogen bond across a vinyl group, with no by-products. It is faster, cleaner, and far more sensitive to contamination: sulphur, tin, phosphorus and amine compounds poison the catalyst at parts per million.
| Peroxide cure | Platinum addition | |
|---|---|---|
| Cure temperature | 150 to 175 C | 120 to 160 C |
| By-products | Volatile fragments, smell | None |
| Post-cure needed | Yes, to remove fragments | Less, but still usual |
| Compression set | Good | Better |
| Contamination sensitivity | Low | Very high, sulphur and tin poison it |
| Typical use | General moulded goods, thick sections | LSR, medical, food, clear parts |
Table 2 - Peroxide against platinum cure. The platinum system is better on every performance measure and worse on every handling measure.
Post-curing at 200 degrees for four hours is not optional on a peroxide compound. It drives off the volatile fragments, completes the crosslink and improves compression set substantially. A part that skips the post-cure meets the drawing as moulded and changes shape in service.
Additives, mixing and maturation
The five parts that decide whether it lasts
Minor ingredients are a small fraction of the compound and most of the difference between a two year part and a twenty year part.
- Heat stabilisers: iron oxide, cerium oxide or a titanium compound, at 1 to 3 parts. They scavenge the radicals that would continue to crosslink the rubber at 200 degrees and turn it brittle.
- Oil resistance: a fluorosilicone blend or a phenyl group on the chain. Neither is cheap, and both change the cure.
- Mould release: a fatty acid ester or a zinc stearate at 0.2 to 0.5 parts. Too much and the part cannot be printed or bonded.
- Flame retardants: platinum compounds, zinc borate or a phosphorus additive, at the cost of compression set.
- Colour: a pigment masterbatch, which can shift hardness by 1 to 3 Shore A and can affect cure if it carries a sulphur or a tin contaminant.
The release agent is the ingredient that causes the most trouble downstream. A compound that demoulds beautifully because it carries 1 part of zinc stearate will not accept a silicone adhesive or a pad print without a primer, and the moulding shop discovers that after the parts are made.
Mixing: order, temperature and maturation
High consistency silicone is mixed on a two-roll mill or in an internal mixer, and the order of addition matters more than the machine.
- Gum and structure control additive first, until the gum is smooth.
- Filler added in thirds, with the band cut and folded between additions. All at once and the compound crumbles.
- Softener and process aids after the filler is fully incorporated.
- Cure system last, at the lowest possible temperature, because the peroxide will scorch if the mill gets above 60 to 70 degrees.
The compound then matures for 24 hours before use. Maturation lets the structure control additive finish its work and the silica network reach equilibrium, which is why a freshly mixed compound moulds differently from the same compound a day later.
Contamination control is the other half of mixing. A few milligrams of a sulphur-cured rubber in a platinum compound stops it curing altogether, and the failure looks like a bad catalyst batch. Mills, knives and trays are dedicated or cleaned to a standard, not wiped.
What to test on the compound, not on the part
Testing the moulded part tells you what happened. Testing the compound tells you what will happen, and it catches a bad batch before it becomes a bad lot.
- Hardness on a 6 mm slab, ISO 7619-1, 1 second reading. The part test is a confirmation, not a measurement.
- Specific gravity, which catches a filler loading error faster than hardness does.
- Mooney viscosity at 100 degrees, ML 1+4, which catches crepe hardening and a mixing error.
- Cure rheometry on a moving die rheometer: scorch time, t90 and torque rise. A flat curve means a dead catalyst or an expired peroxide.
- Compression set on a standard button, 22 hours at 175 degrees, which is the single best predictor of seal life.
The rheometer curve is the most informative single test and the one most often skipped. A compound that reaches the right torque but takes twice the expected time to get there will mould, and it will have a worse set and a shorter life than the batch that behaved.
Can I make a 20 Shore A compound?
Only by reducing the filler and adding softener, and the result has low tear, high set and a tacky surface. Below about 25 Shore A a silicone gel or a foam is usually a better answer than a solid compound.
Why does my compound get harder on the shelf?
Crepe hardening: the fumed silica continues to form a network. A structure control additive slows it, and a compound older than its shelf life will need re-milling before use.
What is the difference between fumed and precipitated silica?
Fumed silica gives higher tear and better dispersion at a higher cost. Precipitated silica gives more hardness per part and is cheaper, but it is more hydrophilic and gives a lower tear strength.
Does pigment affect cure?
It can. Some pigments carry sulphur, tin or phosphorus traces that poison a platinum cure, and most shift hardness by 1 to 3 points. A colour change on a critical part needs a cure check, not an assumption.
Why is post-curing necessary?
It removes the volatile fragments from a peroxide cure, completes the crosslink and improves compression set. A part that skips it changes dimension and hardness in service.
Can I blend two compounds to hit a hardness?
Yes, on a mill, and it is common for a short run. The blend needs its own hardness and cure check, because two compounds with different cure systems do not blend linearly.
Need a compound tuned to a temperature or a chemical?
Tell us the service temperature, the medium and the hardness. We will tell you whether a standard compound will do it or whether the formulation has to change, and what that costs.











