The History of the O-Ring
An O-ring is a loop of elastomer with a round (O-shaped) cross section used as a mechanical seal. They are designed to be seated in a groove and compressed during assembly between two or more parts, creating a seal at the interface. The joint may be static, or (in a few circumstances) have relative motion between parts and O-ring (rotating pump shafts and hydraulic cylinders, for example). Joints with motion usually require lubrication of the O-ring to reduce wear. This is often accomplished with the fluid being sealed. O-rings are one of the most popular seals used in machine design because they are inexpensive and easy to make, reliable, and have simple mounting requirements. They can seal tens of megapascals (thousands of psi) pressure. In some cases, O-rings are used with back-up rings.
The O-ring was invented in 1936 by then 72-year-old Danish-born Niels Christensen. During World War II, the US government “bought” critical war-related patents after finding out big businesses were in violation of Christensen’s patent right. Christensen got a lump sum payment of $75,000 for it.
Theory and Design: Successful O-ring joint design requires a rigid mechanical mounting that applies a predictable deformation to the O-ring. This introduces a calculated mechanical stress at the O-ring contacting surfaces. As long as the pressure of the fluid being contained does not exceed the contact stress of the O-ring, leaking cannot occur. The seal is designed to have a point contact between the O-ring and sealing faces. This allows a high local stress, able to contain high pressure, without exceeding the yield stress of the O-ring body. The flexible nature of O-ring materials accommodates imperfections in the mounting parts. O-rings are available in a large number of standard sizes and materials. Manufacturers or reference books supply application and machining data for the mounting. O-rings are one of the most common and important elements of machine design.
Material: O-ring selection is based on chemical compatibility, application temperature, sealing pressure, lubrication requirements, quality, quantity and cost. Typical O-ring materials: Nitrile (NBR or Buna-N), Silicone (VMQ), ®Fluorocarbon (Viton, FKM), Perfluoroelastomer (Kalrez® (FFKM), Fluorosilicone (FVMQ), Ethylene Propylene (EPM, EPDM, EP, EPR), Neoprene (CR, Chloroprene), Polyurethane (AU, EU).
Other Seals:
There are variations in cross-section design other than circular. These include the O-ring with an x-shaped profile, commonly called the X-ring, Q-ring, or by the trademarked name Quad Ring. When squeezed upon installation, they seal with 4 contact surfaces small contact surfaces on the top and bottom. This contrasts with the standard O-ring’s comparatively larger single contact surfaces top and bottom. X-rings are most commonly used in reciprocating applications, where they provide reduced running and breakout friction and reduced risk of spiraling when compared to O-rings.
Square cuts:
There are also rings with a square profile, commonly called square-cuts, lathe cuts, or Square rings. When O-rings were selling at a premium because of the novelty and complexity in manufacturing, lack of efficient manufacturing processes and high labor content, Square rings were used as an economical substitution for O-rings. The square ring is typically manufactured by molding a tube which is then lathe-cut. The physical sealing performance of square rings in static applications should be superior to that of O-rings because of larger contact surfaces, but it is offset by LESS compression. It is inferior to o-rings in dynamic applications. Square rings can also be more difficult to install than O-rings.
Challenger Disaster: The failure of an O-ring seal was determined to be the cause of the space shuttle Challenger disaster on January 28, 1986. A contributing factor was cold weather prior to the launch. This was famously demonstrated on television by a Cal Tech professor (physics) Richard Feynman, when he placed a small O-ring into his ice water, and subsequently showed its loss of pliability before an investigative committee.
The material of the failed O-ring was Viton (registered trade name with DuPont) and the manufacturer of that particular O-ring was Morton-Thiokol in Utah, USA. Viton is not a good material in cold temperature applications. When an O-ring is frozen there is a TG (transition-to-glass) point where it will not bounce back. Even when O-ring does not reach TG point, the O-ring once compressed, like most any other material in cold temperature, will take longer time to bounce back to its original shape, Viton doubly so. The O-rings (and all other seals) do their work by creating positive push against the surface in turn blocking the leak. The night before the launch showed freezing temperature. Concerned with this, the NASA technician has measured the temperature prior to launch. The ambient temperature was within the launch range, and the shuttle got the green light to launch. However, NASA did not account for the fact the temperature “at the O-ring” was still below the launch range. The air was warm enough, but the O-ring hadn’t thawed yet. What Dr. Feynman observed during his video forensic investigation was he saw a puff of black smoke come out of the side of the solid fuel rocket. He deduced that that must have been failed zinc-oxide putty and O-ring bits at the joint expelled out by hot gas. The tiny leak when the flame reached it acted as a torch against the external tank and booster. Freed booster struck the main tank and pierced the tank’s side. Liquid hydrogen and liquid oxygen fuels ignited. The Challenger was completely destroyed 73 seconds after the launch.
The rubber industry has gone through its share of transformation after the accident. All O-rings come with batch number and serial date, just like the medicine industry, to control and track precise distribution. O-rings can be recalled off the shelf if needed.
As far back as 1839, Charles Goodyear first improved the elastic properties of natural rubber by heating with sulfur (vulcanization). It was not until the 1930s that the macromolecule model of rubber was understood. After World War II and through the 1950s rapid developments in synthetic polymers were made. Most commercial high-performance elastomers trace their origins to the 1960s and 1970s.
Polymers are long chains of repeating chemical units, or monomers. The chemical skeletal structures may be linear, cyclic or branched. When one monomer is polymerized, the resultant polymer is called a homopolymer. Examples include polyethylene, polystyrene and polytetrafluoroethylene (PTFE). Copolymers (or dipolymers) are derived from the polymerization of more than one type of monomer. The distribution of monomers in these copolymers can be statistical, random or alternating. Examples include ethylene-propylene and fluorocarbon elastomers (vinylidene fluoride and hexafluoropropylene). Terpolymers are three-monomer-unit polymers, such as ethylene-propylene-diene (EPDM) and specialty fluorocarbon grades.
Rubber is composed of long chains of randomly oriented molecules. These long chains are subject to entanglement and cross-linking. The entanglement has a significant impact on the viscoelastic properties such as stress relaxation. When a rubber is exposed to stress or strain energy, internal rearrangements such as rotation and extension of the polymer chains occur. These changes occur as a function of the energy applied and the duration and rate of application, as well as the temperature at which the energy is applied.
A rubber response to an applied energy can be energy storage (elastic) or energy dissipation (viscous). For sealing elastomers, the elastic component of response is most important. An applied stress induces a corresponding strain which creates contact stress (or sealing force). As the polymer chains rearrange to reduce this internal energy, or stored force, a loss of sealing force occurs.
Rubber products are typically cured at high temperature and pressure. The addition of curatives and accelerators forms cross-links between the polymer chains or backbone. It is this network of cross-links that largely determines the physical properties of tensile, elongation and compression set.
Fillers play a large role in rubber technology. Carbon black and silica fillers can serve to improve the hardness, abrasion resistance, tensile properties and tear strength. Non-black fillers, such as titanium dioxide and barium sulfate can offer pigmenting properties for part identification, as well as improved stability in strong oxidizing environments. However, the viscoelastic response and hysteresis losses are greatly enhanced by the use of fillers.
The physical properties of an elastomer vary with the test conditions especially temperature. The rate of application of a load also has an effect, as does previous stress history.
Plastics are rigid long-chain polymers which are not usually connected or cross-linked. Plastics can 
either be thermoplastic-meaning they can be heated and cooled without changing propertiesor – thermoset, where an increase in temperature changes the chemical structure and properties. As a class, plastics have low elongation and high elongation set.
Elastomers are flexible long-chain polymers which are capable of cross-linking. Cross-linking chemically bonds polymer chains which can prevent reversion to a non-cross-linked polymer at elevated temperatures. The cross-link is the key to the elastic, or rubbery, properties of these materials. The elasticity provides resiliency in sealing applications.
Thermoplastic elastomers (TPEs) often combine the properties of elastomers with the ease of processability of thermoplastics. They are the result of a physical combination of soft, elastic polymer segments and hard, crystalline segments which are capable of cross-linking. Thermoplastic elastomers are generally classified by their structure rather than their chemical makeup.