Friction products: Carbon fiber stopping power

Low weight, thermal shock resistance, and extreme strength drive growth of carbon-reinforced friction products and materials in brake applications.

Consider the energy generated when the brakes of a Boeing 767 engage during landing: The plane has a maximum mass of 158,000 kg (350,000 lb) and a typical landing speed of 178 mph (287 kph). According to the Center for Advanced Friction Studies (Carbondale, Ill.), 450 million joules of energy or more than 737,000 ft-lb — must be absorbed and dissipated by the rotors and stators of the aircraft’s brake assemblies. For that reason, Gerard Taccini, product manager of HITCO Carbon Composites Inc. (Gardena, Calif.), says, “Carbon plays a big role in today’s friction market.” In the North American market alone, the friction products and materials market was worth $6.9 billion in 2006 and is expected to register a compound annual growth rate of 2.6 percent over the next five years, reaching $7.7 billion by 2012, according to a recent report from market research firm BCC Research (Wellesley, Mass.). Driven by growth in the aerospace sector as well by the inherent performance characteristics of the material, carbon-reinforced composites, such as carbon/carbon (C/C), are expected to offer the best opportunity for growth in the North American friction market, says BCC. C/C is a composite made from a carbon fiber reinforcement impregnated with a carbon matrix by a lengthy and expensive process that involves exposure to extreme heat (pyrolysis). The result is an extremely lightweight composite characterized by a very low coefficient of thermal expansion (CTE) and excellent resistance to the extremely high temperatures generated in braking applications.

“The low weight of C/C composites is critical,” says HITCO’s product manager Scott Ostergren. C/C brake discs, also known as rotors, are four times lighter than conventional metal discs made of either steel or grey-cast iron. “Due to the significant reduction in the forces of rotating mass within all the brake systems on an aircraft — eight assemblies on an Airbus A300, for example — when using C/C, the design of the support structure for the aircraft does not have to be as robust, further reducing the overall aircraft weight.”

Weight reduction also is critical in auto racing applications. “The high-performance braking of C/C allows the driver to decelerate later when approaching a curve without the fear of brake fade,” explains Ostergren. After braking, acceleration is increased due to C/C’s substantial savings in the unsprung weight (weight unsupported by the car’s suspension). “So while the car does not actually go faster,” he notes, “lap times can be reduced.” Metal brakes, on the other hand, can fade and warp at high temperatures. The high peak temperatures generated by C/C disc and pad materials can be an issue, however, which is one of the reasons why C/C automotive brake applications have been limited to open wheel configurations, such as those used in Formula One racing. C/C brake discs also suffer low friction coefficient in cold and damp conditions and therefore are not suitable for production cars.

Carbon fiber-reinforced silicon carbide (C/SiC), however, has been used as a performance braking option on high-end luxury cars for several years and is gaining ground in OEM-installed applications and aftermarket sectors for both production cars and motorcycles as these sectors increasingly seek to improve fuel efficiency.

Stopping power for aircraft

According to HITCO’s Taccini, “Any aircraft developed today will have C/C brakes because of the tremendous weight savings carbon has over sintered steel.” However, it is the superior physical properties of C/C materials that play the most critical role in stopping the aircraft. Unlike automobile brakes, which require only two brake pads, aircraft applications employ a heat stack configuration consisting of alternating rotors and stators with a pressure plate and end plate at either end. “The rotors have key slots on the outer diameter that match up with splines located on the inner side of the wheel, while the stators, pressure plate and end plate have key slots on the inner diameter that match up with the splines on the torque tube,” explains Ostergren. As the wheel spins, the rotors spin with the wheel, and when the brake assembly is actuated, the pistons press against the pressure plate, which presses the components of the heat stack together.”

As the discs rub together, the kinetic energy of the aircraft is absorbed by the heat stack and transformed into heat — temperatures can reach 3000°C (5432°F). Carbon/carbon has a heat capacity approximately 2.5 times greater than that of steel, and at high temperatures, it is nearly twice as strong as steel.

Forming a material with this com-bination of intrinsic strength and thermal capacity involves a complex, multistep manufacturing process. The physical properties of C/C composites depend greatly on fiber selection and orientation, as well as the densification process, which is why these processes are considered highly proprietary.

The processes typically begin with preparation of multiple layers of fiber fabric. Messier-Bugatti (Vélizy-Villacoublay, France), which manufactures carbon discs at its locations in Villeurbanne, France, and Walton, Ky., begins its process with long pre-oxidized polyacrylonitrile (PAN)-based fibers, while HITCO uses pitch-based fibers from Cytec Industries Inc. (West Paterson, N.J.). Pre-oxidized PAN, or oxi-PAN, fibers have been heat stabilized to the point that they will no longer melt or burn, but do not yet have the high carbon content typically associated with carbon fibers used as composite reinforcements. Oxi-Pan fibers have 9 to 14 percent oxygen content, hence the name. Fiber forms can range from randomly oriented chopped mat to woven fabrics. Generally, the fibrous materials are formed in a mold to produce a disc-shaped blank or preform that is then thermally treated to eliminate noncarbon elements. During this process, the blank loses more than half of its initial weight, and its carbon content increases from 65 to 99 percent. The resulting porous structure, however, has poor physical properties and therefore requires further processing — typically isothermal chemical vapor deposition (CVD) is used, either alone or in combination with liquid phenolic impregnation (LPI).

In the LPI process, fibers are infiltrated with a phenolic resin and then carbonized at temperatures up to 4500°F (2500°C), resulting in a C/C composite with a density of 1.5 to 1.6 g/cc. During the isothermal CVD process that follows (also known as chemical vapor infiltration or CVI), porous carbonized structures are placed in a CVD furnace, which is evacuated of air and heated to about 1000°C (1832°F). CVD furnaces are large, double-walled, cylindrical vessels with gas-tight closures. Under high vacuum, a hydrocarbon gas, such as natural gas or a mixture of methane and other gases, is introduced into the furnace. The gas diffuses, or “cracks,” and deposits pyrolytic carbon into the porous carbon fiber structure.

“The pyrolytic carbon matrix keeps the carbon filaments together, preventing fiber pullout that can lead to excessive wear,” explains Taccini. “Pyrolytic carbon has very high temperature resistance and will not degrade under high energy loading like conventional organic matrix composites.”

In the isothermal CVD process, densification occurs at uniform temperature for 30 to 40 days. The process may be repeated until a specific density and porosity is obtained. Discs are then machined, drilled and treated with an antioxidant or special paint. The latter protects them, during brake operation, against oxidation (conversion of carbon to carbon dioxide gas) effected by temperature, humidity and certain chemicals, such as the de-icing products used by most airlines.

Refining the C/C process

Production costs associated with the manufacture of C/C friction products are very high: Beyond the large investment that must be made in manufacturing equipment, there is the recurring cost of energy required to maintain high processing temperatures over lengthy cycle times. Over the past five years, major producers of friction products and materials, including Aircraft Braking Systems Corp. (ABSC, Akron, Ohio), DACC Custom Composites (Changwon, South Korea), Goodrich Corp. (Charlotte, N.C.), HITCO, Honeywell International (Morristown, N.J.) and SGL Carbon (Wiesbaden, Germany) have designed and patented a number of process variations, including advancements in antioxidant systems, process automation, fiber orientation, densification methods and process flow, all in an effort to reduce manufacturing time and costs while maintaining thermal and strength properties.

ABSC has developed and patented a process in which pitch- or PAN-based fiber strands are continuously fed into a mold and then compressed into a preform and needled. Prior to densification, a very high-temperature-tolerant filler, such as aluminum oxide, boron carbide, silicon carbide or pitch, is introduced to increase the surface area to which carbon can bind during the CVD process. Introduced by way of dry powder with a particulate size of less than 800 microns in diameter or liquid slurry with particles of 50 microns or less in diameter, these fillers invade gaps and crevices in the preform and reportedly speed densification.

DACC, which produces C/C aircraft brake discs and automotive brake discs and clutch assemblies, patented a method for manufacturing C/C composites using a combination of liquid impregnation and thermal gradient CVD. Themal gradient CVD differs from conventional isothermal CVD in that heat sources placed on opposite sides of the substrate differ in temperature by several hundred degrees. The resulting temperature gradient causes an increase in the deposition rate.

One example of DACC’s process begins with a unidirectional mat produced by winding 320K oxi-PAN fiber on a mandrel. To prevent damage to the surface of the fiber, surface impregnation occurs using a polyvinyl, epoxy or other appropriate resin. A minimum of three layers of carbon fiber mat, each about 0.9 mm/0.4 inch thick, are combined to produce a preform. The preform stack is needle-punched to reinforce the mats in the z-axis. In the finished product, about 10 percent of the fiber is oriented in the z-direction. The needled preform is heat-treated in a vacuum atmosphere at 1700°C (3092°F) to remove noncarbon impurities. According to DACC, fiber ultimately accounts for approximately 45 percent of the composite’s volume.

For densification, DACC uses the thermal gradient CVD process. A wire-like heat source is used to expose the inside of the ring-shaped preform to higher temperatures than are used to heat the ring from the outer edge. During this process, the preforms are typically stacked, with insulating materials placed between them to enable simultaneous densification of multiple preforms. Carbon infiltrates the preform from the inside diameter to the outside diameter. After separating the preform from the hot wire, thermal treatment is performed in an argon atmosphere at 2000°C/3632°F. Preforms are then machined to final part shape and treated with an oxidation inhibitor. Part density is reportedly in the range of 1.6 to 1.9 g/cc. The fiber tow size and fiber volume ratio can be varied to tailor mechanical and thermal properties, such as compression, shear strength and friction coefficient, to meet the specific requirements.

A patented process from Honeywell Aircraft Landing Systems (South Bend, Ind.) focuses on controlling variations in the preform microstructure to optimize wear, strength, toughness and thermal conductivity. According to Honeywell, the overall strength of a brake disc can be enhanced by locating longer 40-mm to 60-mm (1.6-inch to 2.4-inch) fibers in the interior planes of the preform while placing shorter 10-mm to 20-mm (0.39-inch to 0.79-inch) fibers in the outer layers, where they enhance the friction and wear properties of the brake disc. Honeywell’s process employs a robotic chopping and spraying system and resin transfer molding (RTM) to aid in preform densification and to reduce the number of densification cycles.

The U.S. Air Force Research Lab Propulsion Directorate also has developed a process now licensed to SMJ Carbon Technology (both located at Edwards Air Force Base, Calif.). Dubbed the In-situ Densification Process, it impregnates the porous carbon fiber preform with a hydrocarbon feedstock, such as naphthalene. Under pressure, the naphthalene is converted to mesophase pitch, which at higher temperatures is converted to carbon. Because of naphthalene’s low viscosity, the process requires relatively low pressures to induce the material to fill voids in the preform. Moreover, the less viscous material invades the preform faster than other friction-product matrix materials, reportedly reducing cycle time to less than 24 hours. Several cycles are required to reach carbon density and toughness required by brake applications.

Hitting the road with ceramics

C/SiC friction materials, commonly known as carbon/ceramic, provide yet another option for braking applications. SGL Group and automaker Porsche first introduced carbon/ceramic brake discs on the latter’s 911 GT2 sports car in 2001, and now Porsche ships 50 percent of its 911 GT3 models with this option. The European automotive industry has driven the use of carbon/ceramic brake discs, offering them as standard equipment and options in a wide range of luxury cars, including models from Aston Martin, Bentley, Bugatti, Lamborghini and Mercedes-Benz.

Compared to conventional metal brakes, carbon/ceramic discs offer many advantages. “Carbon/ceramic rotors do not fade like conventional iron disc or drum brakes,” explains Stan Hemsted, product manager for friction products at Starfire Systems (Malta, N.Y.), which produces carbon/ceramic brake technology for both the automotive and motorcycle aftermarkets. “The material is also inherently quiet, damping instead of amplifying noise.”

One of C/SiC’s few disadvantages of is its low volumetric heat capacity, which requires discs to be internally vented to aid in cooling efficiency. High surface temperatures can accelerate brake pad wear, and a heat shield or other form of insulation is required to protect hydraulic pistons and other components from the high heat.

Brake pads used with C/SiC discs generally contain ceramic powder combined with metal. While the ceramic provides the necessary hardness, the metal forms a transfer coating on the disc and pad surfaces during a break-in period, creating a film that provides the primary friction surface.

Both C/SiC discs and ceramic-imbued pads are more expensive than traditional materials. At one-fourth the cost of a C/C racing brake, the price tag for carbon/ceramic discs is still high, especially to replace a metal disc that costs less and has worked well for years in everyday driving.

The catalysts for growth of C/SiC disc use likely will be a combination of low weight and increased service life. C/SiC brake discs are designed to last the life span of the car, clocking in at an estimated 300,000 km (186,411 miles) vs. 60,000 km (37,282 miles) for conventional metal brake systems, reports Kwangsoo Kim, president of DACC, in a benchmark of brake disc materials. C/SiC discs are highly abrasion-resistant and reportedly offer more than six times the resistance to thermal shock than cast iron discs; twice that of C/C racing brakes — 46,000 Watts per meter (W/m) vs. 6,800 W/m for iron and 16,000 W/m for C/C.

“C/SiC significantly reduces wheel weight,” adds Hemsted. C/SiC discs offer nearly a 65 percent weight reduction over comparable iron discs, which translates into improved handling, acceleration and fuel efficiency.

Also, unlike C/C discs, C/SiC discs have a stable friction coefficient, operating at both low temperatures and temperatures as high as 1000°C (1832°F). Iron discs crack at 700°C (1292°F). Such a wide operational temperature range, combined with the mechanical properties of C/SiC, have opened up possibilities for their use in both aircraft and high-speed train applications.

As with C/C brake disc production, C/SiC producers like Starfire Systems and SGL are working to reduce the price of their friction products through improved production processes. Starfire, which manufactures proprietary polymers and PAN-based carbon weaves, uses the PIP (polymer infiltration and pyrolysis) process to produce its C/SiC products. The process involves soaking a carbon fiber preform with a proprietary ceramic-forming polymer precursor that converts to very hard silicon carbide ceramic during pyrolysis. Starfire sources oxi-PAN fiber from several manufacturers, including Cytec Engineered Materials Inc. (Tempe, Ariz.) and Zoltek Inc. (St. Louis, Mo.).

“PIP is a simpler and less expensive process than melt infiltration,” says Hemsted, adding that the process “also offers greater control over silicon crystal size and provides better overall uniformity,” he adds.

In Starfire’s process, fabric pre-impregnated with the ceramic-forming polymer is cut into plies and stacked to make a preform of the appropriate size and thickness. A 5-mm/0.2-inch thick motorcycle rotor in the company’s trademarked Starblade series, for example, requires 10 plies. The preform then is heated under pressure to produce a near-net shape part. Pyrolysis at 850°C (1562°F) converts the hydrogen-rich polymer to silicon carbide, shrinking the matrix and increasing its density. Then the rotor is re-impregnated with polymer and pyrolized several more times until only 5 percent to 8 percent porosity remains. The PIP process takes about one week to complete a blank, which is then machined to a final disc. The entire process requires two to three weeks to complete.

SGL, which uses a liquid silicon melt infiltration process, has reduced manufacturing costs by moving from batch production to a continuous process. During SGL’s process, near-net shape C/C preforms are heated in carbonization furnaces to temperatures reaching 900°C (1652°F), after which they require siliconization in a high-vacuum furnace. During this stage, the discs are subjected to even higher temperatures and infiltrated with molten silicon. At about 1700°C (3092°F) the silicon reacts with the carbon matrix, generating C/SiC. Reinforcing carbon (short and long fibers, woven fabrics or felts, manufactured by SGL) are treated to protect them against the reaction while firmly linking them to the resulting silicon carbide matrix. The company produces 30,000 carbon/ceramic brake discs per year at its plant in Meitingen, Germany, but reportedly has the potential to expand annual production to 250,000 discs.

Stopping power that won’t stop

Today, carbon fiber-reinforced brake components are commonplace in commercial aircraft, where C/C use is easily justified because its demonstrated superior performance outweighs the high cost of both materials and lengthy production processes. But brake manufacturers now have the potential to market similar products into the automotive market, an arena in which their use is clearly optional. Material and process innovations are moderating cost and enabling composite brake performance in a broad range of driv-ing conditions. As pressure grows to conserve oil and improve fuel economy, automakers keen to lightweight vehicles should find these carbon composites an increasingly acceptable choice.