Reverse engineering in rubber means analysing a finished vulcanised part, such as an O-ring, to learn what it is made of and how it behaves, so that a compound with equivalent performance can be developed. It is routine in the industry for benchmarking, for qualifying a second source of an existing part, and for investigating a failure. It is a science of inference. The result is an estimate of the composition and a set of properties to match, not a recovered recipe.
What can be found
- Polymer type, often confirmed by two independent methods. Infrared spectroscopy of the polymer and pyrolysis-based methods are the main routes. Thermal analysis (glass transition by DSC) supports the identification.
- Rough composition by weight: polymer, carbon black, mineral filler, ash and volatile or extractable materials, using thermogravimetric analysis (TGA) with solvent extraction.
- Filler identity: carbon black versus mineral, and which minerals, from the ash residue. Acid-soluble fractions point to oxides and carbonates, and others to silica or titanium dioxide.
- Plasticiser and extractables: by solvent extraction followed by infrared and gas chromatography with mass spectrometry.
- Elements: sulfur, fluorine, chlorine and others by elemental analysis.
- Physical profile: hardness, density, tensile properties, compression set, swell in fluids.
- Cure-state hints: swelling-based crosslink density estimates and thermal analysis.
What cannot be recovered
- The exact recipe. Quantities and some ingredients, especially curatives, accelerators and coagents, change during cure and are hard or impossible to identify.
- Specific supplier grades. Two polymers that behave alike can be sold by different makers. The analysis shows type and some features such as fluorine content, not the brand.
- Mixing and moulding details: mixing order and temperatures, cure and post-cure times and mould design.
- Proprietary additives present at very low levels.
For these reasons the work ends with a trial formulation that is mixed, cured and tested against the original, and adjusted until properties match.
A typical workflow
- Define the goal. Match a property set, identify a failure, benchmark, or qualify a second source. The goal sets the depth of analysis.
- Collect and prepare the sample. Record the history of the part. Use enough material, from representative areas, and remove coatings and metal.
- Basic physical tests. Hardness, density and dimensions without destroying the sample.
- Extraction. A solvent such as acetone removes the soluble organics, which are weighed and analysed separately.
- TGA. Heating in an inert atmosphere and then in air separates volatile, polymer, carbon black and ash fractions.
- Polymer identification. Infrared of the extracted rubber and/or pyrolysis with infrared or gas chromatography, with DSC as a cross-check.
- Ash analysis. Dissolving the ash in successive acids and analysing the solutions or the solids to identify inorganic fillers and acid acceptors.
- Elemental analysis. Sulfur, fluorine and others.
- Reconstruction. Convert the weight fractions to phr using the polymer as 100, and propose a recipe using knowledge of what compounders normally use.
- Trial mixes and comparison. Mix, cure, test and iterate.
Special points for fluoro-rubbers and HNBR
- Fluoro-rubbers decompose at high temperature and release hydrogen fluoride. Pyrolysis, TGA and ashing must be done in equipment and with ventilation designed for corrosive fumes, and the lab needs a safety plan.
- Fluorine content and cure system strongly affect performance, so they are measured early.
- In HNBR, acrylonitrile content and the level of unsaturation can be estimated from infrared and are key to matching low-temperature and fluid behaviour.
- Cure systems are hard to analyse after cure, so comparing rheometer curves and compression set of trial compounds is a practical route.
Scope and responsibility
Reverse engineering is a legitimate technique, but how it is used matters. Work only on samples you or your customer own or have the right to analyse. Do not use it to copy another party's confidential or patented formulation, or to claim approvals that belong to another manufacturer. Approvals for drinking water, food contact, oil and gas or aerospace belong to a specific formulation and its manufacturer, and a matching compound has to be qualified on its own.
Pascalor Decode follows these rules. See the Decode page for the service scope and the questions we ask first.
Sources and further reading
- S. Dasgupta, R. Mukhopadhyay, K. C. Baranwal and A. K. Bhowmick, Reverse Engineering of Rubber Products: Concepts, Tools, and Techniques, CRC Press, 2014. The standard text on the subject.
- P. A. Ciullo and N. Hewitt, The Rubber Formulary, 1999.
- ASTM D297, D3677, D3452, D6370; ISO 9924; ISO 4650; ASTM E1356 and ISO 22768.