High Performance Carbon Fibers

The First Carbon Fibers

The synthetic carbon industry had its official beginning in 1886 with the creation of the National Carbon Company. Based in Cleveland, Ohio, the company would eventually merge with Union Carbide in 1917 to form Union Carbide & Carbon Corp., which changed its name to Union Carbide Corp. in 1957. The carbon products division of Union Carbide Corp. became the independent UCAR Carbon Company in 1995, and was renamed GrafTech International Holdings in 2002.

Electricity was mostly a lab curiosity until the late 1800s, when carbon arc lamps began lighting the streets of major U.S. cities. The lamps were composed of two carbon rods connected to a current source and separated by a short distance. A blazing hot path of charged particles—the “arc”—formed between the two rods, giving off an intense light. National Carbon got its start by producing carbon electrodes for streetlamps in downtown Cleveland.

In 1879, Thomas Edison invented the first incandescent light bulb, which uses electricity to heat a thin strip of material, called a filament, until it glows. He may also have created the first commercial carbon fiber. To make his early filaments, Edison formed cotton threads or bamboo slivers into the proper size and shape and then baked them at high temperatures. Cotton and bamboo consist mostly of cellulose, a natural linear polymer made of repeating units of glucose. When heated, the filament was “carbonized,” becoming a true carbon copy of the starting material—an all-carbon fiber with the same exact shape. Tungsten wire soon displaced these carbon filaments, but they were still used on U.S. Navy ships as late as 1960 because they withstood ship vibrations better than tungsten.

Near the end of World War II, Union Carbide began investigating a replacement for tungsten wire in vacuum tubes by carbonizing rayon, another cellulose-based polymer (like cotton) that became popular in clothing. The end of the war brought an end to the government’s funding for this project, but carbon fibers were still raising interest in the commercial sector. Barnebey-Cheney Company, in 1957, briefly manufactured carbon fiber mats and tows (rope-like threads without the twists) from rayon and cotton. These were used as high temperature insulation and filters for corrosive compounds. A year later, Union Carbide developed a carbonized rayon cloth and submitted it to the U.S. Air Force as a replacement for fiberglass in rocket nozzle exit cones and re-entry heat shields.

While finding a certain degree of success in their respective niches, all of these early carbon fiber materials had poor mechanical properties, making them unsuitable for structural use. It took a chance discovery to set the age of high performance carbon fibers in motion.

Back to top

Discovery of High Performance Carbon Fibers:
Roger Bacon’s “Perfect Graphite”

The modern era of carbon fibers began in 1956, when Union Carbide opened its Parma Technical Center just outside Cleveland. The complex was one of the major laboratories of Union Carbide’s basic research program, modeled after the university-style corporate labs that became popular in the late 1940s and 1950s. They gathered young, bright scientists from a variety of backgrounds and let them loose on their favorite projects, giving them an extraordinary degree of autonomy.

With a freshly minted Ph.D. in physics, Roger Bacon joined the Parma staff in 1956. “I got into carbon arc work, studying the melting of graphite under high temperature and pressures,” Bacon recalls. “I took on the job of trying to determine the triple point of graphite. That’s where the liquid, solid, and gas are all in thermal equilibrium.” The equipment was akin to the early carbon arc streetlamps, only operating at much higher pressures. Small amounts of vaporized carbon would travel across the arc and then deposit as liquid. As Bacon decreased the pressure in the arc, he noticed that the carbon would go straight from the vapor phase to the solid phase, forming a stalagmite-like deposit on the lower electrode. “I would examine these deposits, and when I broke one open to look at the structure, I found all these whiskers,” he says. “They were imbedded like straws in brick. They were up to an inch long, and they had amazing properties. They were only a tenth of the diameter of a human hair, but you could bend them and kink them and they weren’t brittle. They were long filaments of perfect graphite.”

The year was 1958, and Bacon had demonstrated the first high performance carbon fibers. In fibrous forms, carbon and graphite are the strongest and stiffest materials for their weight that have ever been produced. Bacon demonstrated fibers with a tensile strength of 20 Gigapascals (GPa) and Young’s modulus of 700 GPa. Tensile strength measures the amount of force with which a fiber can be pulled before it breaks; Young’s modulus is a measure of a material’s stiffness, or its ability to resist elongation under load. For comparison, steel commonly has a tensile strength of 1-2 GPa and Young’s modulus of 200 GPa.

Carbon fibers are polymers of graphite, a pure form of carbon where the atoms are arranged in big sheets of hexagonal rings that look like chicken wire. Bacon’s graphite whiskers were sheets of graphite rolled into scrolls, with the graphite sheets continuous over the entire length of the filament.

“After studying the heck out of these things, I finally published a paper in the Journal of Applied Physics in 1960,” Bacon says. The paper has since become a milestone, partially because some have claimed that Bacon may have been the first person to produce carbon nanotubes—hollow cylinders of graphite with diameters on the order of single molecules. Their incredible properties have made nanotubes one of the hottest areas of research in recent years, promising to revolutionize just about every area of science. Sumio Iijima published a paper in 1991 that is often regarded as the first discovery of carbon nanotubes; it reported on a method that produced both tubes and scrolls. The process is similar to Bacon’s, suggesting that he too may have prepared nanotubes along with his whiskers, although he didn’t know it at the time. “I may have made nanotubes, but I didn’t discover them,” he says.

By producing his high strength and high modulus whiskers, Bacon had demonstrated experimentally something that theoreticians had proposed long ago. But the fibers were still just a laboratory phenomenon, not a practical development. “I estimated the cost of what it took to make them, and it was $10 million per pound,” he says. To tap their full potential, manufacturers needed a cheap and efficient way to produce the fibers. Much of the research in the ensuing decades was dedicated to exactly that.

Back to top

Early Applications of Carbon Fibers

As early as 1959—just one year after Bacon’s discovery—scientists at Parma had taken a step toward producing high performance carbon fibers. Curry Ford and Charles Mitchell patented a process for making fibers and cloths by heat-treating rayon to high temperatures, up to 3,000 °C. They had produced the strongest commercial carbon fibers to date, which led to the entry of carbon fibers into the “advanced composites” industry in 1963.

Composites are reinforced materials consisting of more than one component. The industry had been dominated by fiberglass and boron fibers, which were extremely popular in the late 1950s and early 1960s. Boron fibers, which contained a tungsten core, were especially strong and stiff, but they were also expensive and heavy. Carbon fibers were much lighter, so the appearance of relatively affordable carbon composites was a welcome development, and they found widespread use in gaskets and packaging materials.

While the tensile strength of these materials was increasing, all commercial carbon fibers to this point were still of relatively low modulus, despite Bacon’s demonstration of their mechanical potential. The first truly high modulus commercial carbon fibers were invented in 1964, when Bacon and Wesley Schalamon made fibers from rayon using a new “hot-stretching” process. They stretched the carbon yarn at high temperatures (more than 2800° C), orienting the graphite layers to lie nearly parallel with the fiber axis. The key was to stretch the fiber during heat up, rather than after it had already reached high temperature. The process resulted in a ten-fold increase in Young’s modulus—a major step on the way to duplicating the properties of Bacon’s graphite whiskers.

Union Carbide developed a series of high modulus yarns based on the hot-stretching process, beginning in late 1965 with “Thornel 25.” The trade name was derived from Thor, the Norse god for strength, and the Young’s modulus of the fibers—25 million pounds per square inch (psi), which is equivalent to about 172 GPa. The Thornel line continued with increasingly higher levels of modulus for more than ten years.

The U.S. Air Force Materials Laboratory supported much of Union Carbide’s research into rayon-based fibers during this period in an attempt to develop a new generation of stiff, high strength composites for rocket nozzles, missile nose tips and aircraft structures. The fibers were also used in spacecraft heat shields to reinforce phenolic resin—plastics that solidify upon heating and cannot be re-melted. As a missile or rocket returns to the atmosphere, the phenolic resin decomposes slowly while absorbing the heat energy, allowing it to survive the trip through the atmosphere without destroying itself. Carbon fibers kept the phenolic resins intact and they have been an important ingredient in aerospace materials ever since.

Back to top

Polyacrylonitrile (PAN)-based Carbon Fibers

While researchers in the United States were reveling in rayon, scientists overseas were busy creating their own carbon fiber industries based on polyacrylonitrile, or PAN, which had been passed over by U.S. producers after unsuccessful attempts at making high modulus fibers.

A quiet study by Japanese researchers in 1961—largely unknown to Western scientists—demonstrated high strength and high modulus fibers from PAN precursors. Akio Shindo of the Government Industrial Research Institute in Osaka, Japan, made fibers in the lab with a modulus of more than 140 GPa, about three times that of rayon-based fibers at the time. Shindo’s process was quickly taken up by other Japanese researchers, leading to pilot-scale production in 1964. In that same year, just a few months before Bacon and Schalamon debuted their hot-stretching method, William Watt of the Royal Aircraft Establishment in England invented a still higher-modulus fiber from PAN. The British fibers were rapidly put into commercial production.

The secret behind these developments was better precursors. In both Japan and England, researchers had access to pure PAN, with a polymeric backbone that provided an excellent yield after processing. The continuous string of carbon and nitrogen atoms led to highly oriented graphitic-like layers, eliminating the need for hot stretching. Chemical manufacturers in the United States, however, generally inserted other compounds in the polymer backbone that could account for up to 20 percent of the product, making them totally unsuitable for carbonizing.

The Japanese eventually took the lead in manufacturing PAN-based carbon fibers, effectively beating the British at their own game. Japan’s Toray Industries developed a precursor that was far superior to anything seen before, and in 1970 they signed a joint technology agreement with Union Carbide, bringing the United States back to the forefront in carbon fiber manufacturing.

PAN-based fibers eventually supplanted most rayon-based fibers, and they still dominate the world market. In addition to high modulus fibers, British researchers in the mid-1960s also developed a low modulus fiber from PAN that had extremely high tensile strength. This product became widely popular in sporting goods such as golf clubs, tennis rackets, fishing rods and skis; it is also extensively used for military and commercial aircraft.

Back to top

Improvements in Carbon Fiber Production and Materials

Any material containing carbon can be “carbonized” by heating it to around 1,000° C, producing a substance that is roughly 99 percent carbon. Upon further heating, typically to about 2,500 °C, such a material can be converted to 100 percent carbon, while transforming the internal structure from a poorly ordered to a more ordered form. But not all carbon materials heat-treated to these high temperatures are truly graphitic. Only certain carbons start with an adequately ordered structure to form nearly perfect graphite crystals, and only these graphitic substances can approach the excellent properties of pure graphite—high thermal and electric conductivities combined with high stiffness (Young’s modulus).

PAN and rayon are both non-graphitizing materials, so carbon fibers from these precursors will never be truly graphitic, even after heat treatment to high temperatures. To make the next generation of carbon fibers, scientists needed a new starting material. Once again, research at the Parma Technical Center led the way.

Leonard Singer came to Parma in the mid-1950s with little experience in carbon or graphite. He was attracted to the “utopian flavor” of the place, and he planned to continue his work with electron paramagnetic resonance. He was using this research technique to study the underlying mechanism of carbonization, which involved heating various petroleum- and coal-based materials. Heating organic substances like these inevitably leads to the formation of a pitch—a tar-like mixture of hundreds of branched compounds with different molecular weights. Pitch is an important high carbon organic precursor used in the manufacture of a number of carbons and graphites.

Two Australian scientists had recently made an important discovery involving pitches. Most pitches are isotropic, having identical properties in all directions, but these researchers showed how a pitch can be polymerized slightly further to orient the molecules in a layered form. “This happens because of the existence of a liquid crystal state, which is also called a mesophase,” Singer says. “That really solved the orientation mystery which had been bothering me for a long time.” Fiber research was going on all around him at Parma, so Singer couldn’t help but be pulled in. “It occurred to me that one could probably make a fiber out of this,” he says. “That’s when I decided to try orienting a fiber by elongation of the carbonaceous mesophase.”

Singer and his assistant, Allen Cherry, designed a “taffy-pulling” machine that applied stress to the viscous mesophase to align the molecules, and then heated the material to convert it to a highly oriented carbon fiber. The process worked, and subsequent analyses verified that they had made highly-oriented graphitizable carbon fibers.

The physical properties of these graphitized mesophase pitch fibers were astounding. Not only did they have an ultrahigh elastic modulus, approaching 1,000 GPa, but these were also the first carbon fibers with ultrahigh thermal conductivity. This made them especially useful for any application where stiffness and heat removal were important — such as aircraft brakes and electronic circuits. Most mesophase pitch-based fibers did not achieve the high tensile strengths of some PAN and rayon fibers, except in the laboratory.

Singer’s initial discovery was made in 1970, but a patent for both the fiber and the process was not issued until 1977. The patent was an incredible amount of work, a 42-page document with 47 illustrations.

Pitch is a fairly inexpensive raw material. However, depending on the form and properties of the desired product, the cost of the final product—mat, strands or cloth—can vary widely. On the one hand, the mesophase pitch-based carbon fibers used in aircraft brakes and reinforced concrete are relatively inexpensive. On the other hand, due to the extremely high graphitizing temperatures required, the ultrahigh modulus, high thermal conductivity fibers required in satellites and other spacecraft can be expensive.

Back to top

Carbon Fibers Today

All commercial carbon fibers produced today are based on rayon, PAN or pitch. Rayon-based fibers were the first in commercial production in 1959, and they led the way to the earliest applications, which were primarily military. PAN-based fibers have replaced rayon-based fibers in most applications, because they are superior in several respects, notably in tensile strength. Fibers from PAN fueled the explosive growth of the carbon fiber industry since 1970, and they are now used in a wide array of applications such as aircraft brakes, space structures, military and commercial planes, lithium batteries, sporting goods and structural reinforcement in construction materials. In the late 1970s, Union Carbide formed a separate division as its primary carbon fiber producer; the business has since been sold to Amoco and then to Cytec, which is among a group of major carbon fiber manufacturers that spans the globe.

Pitch-based fibers are unique in their ability to achieve ultrahigh Young’s modulus and thermal conductivity and, therefore, have found an assured place in critical military and space applications. But their high cost has kept production to a minimum; only a few Japanese companies in addition to Cytec are currently making commercial mesophase fibers. A lower modulus, non-graphitized mesophase-pitch-based fiber, which is much lower in cost, is used extensively for aircraft brakes.

The cost of making carbon fibers has been reduced drastically in the last 20 years, and researchers are bringing that cost down every day. As they do, many of the applications once considered impossible will become reality. Carbon fibers are used sparingly in automotive applications, but someday entire body panels may be made from them. All high speed aircraft have carbon fiber composites in their brakes and other critical parts, and in many aircraft they are used as the primary structures and skins for entire planes. Carbon fibers could even be used to develop earthquake-proof buildings and bridges.

Back to top

Further Reading

• A history of innovation and growth (GrafTech International, Inc.)
  

Back to top