Fiber Reinforced Polymer Composites Applications in USA
DOT - Federal Highway Administration
Benjamin Tang, PE
HIBT-10, Bridge Specialist Group
1200 New Jersey Avenue SE
Washington, DC 20590
January 6, 1997
Published in the First Korea/U.S.A. Road Workshop Proceedings, January 28-29, 1997
History on Fiber Reinforced Polymer Composites
"Although the concept of fiber reinforced materials can be traced back to the use of straw as reinforcement in bricks manufactured by the Isrealites in 800 B.C., and in more recent times to the use of short glass fiber reinforcement in cement in the United States in the early 1930's, fiber reinforced resin matrix materials (or fiber reinforced composites as we know them today) were not developed until the early 1940's". In this paper the term composites is used in the context of advanced fiber reinforced polymers (FRP's).
After World War II, US manufacturers began producing fiberglass and polyester resin composite boat hulls and radomes (radar cover). The automotive industry first introduced composites into vehicle bodies in the early 1950s. Because of the highly desirable light weight, corrosion resistance, and high strength characteristics in composites; research emphasis went into improving the material science and manufacturing process. That effort led to the development of two new manufacturing techniques known as filament winding and pultrusion, which helped advance the composite technology into new markets. There was a great demand by the recreation industry for composite fishing rods, tennis rackets, ski equipment and golf clubs. The aerospace industry began to use composites in pressure vessels, containers, and non-structural aircraft components. The US Navy applied composites in mine sweeping vessels, crew boats and submarine parts. The domestic consumers began installing composite bath tubs, covers, railings, ladders and electrical equipment. The first civil application in composites was a dome structure built in Benghazi in 1968, and other structures followed slowly.
The Composites Institute, that tracks the domestic economy, reported the total consumption figures for 1994 and 1995 at more than three billion pounds each year. Of the nine industries reported, the transportation industry claimed 31% of the total market, thus, being the biggest consumer of composites in both years. The preliminary figures for 1994 was 945.6 million pounds followed by 1995 with 984 million pounds. It is predicted that orders for composites in the transportation sector will exceed one billion pounds for 1996.
With the increasing demand for composites, new and improved manufacturing processes such as pultrusion, resin transfer molding, and filament winding were developed and implemented in the early 1990s. With these enhancements in place, the current focus is to rebuild the US transportation infrastructure using FRP composites for maintenance and rehabilitation of existing bridges as well as new construction.
Civil Structural Applications Using Composites
Many pedestrian bridge projects have been constructed throughout the United States using pultruded composite structural shapes. In states vulnerable to high seismic activity, concrete bridge columns are being retrofitted with a composite filament winding to increase ductility. Plate bonding using thin composite laminates to strengthen concrete and steel bridge members has been demonstrated. Composite prestressed piles are being applied to civil and marine structures in some of the coastal states.
The first US advanced composite vehicular bridge superstructure was dedicated into service on December 4, 1996 in Russell, Kansas. The deck panels were shop-fabricated with composite honeycomb cells which are sandwiched between two face sheets. These panels were then joined together with epoxy in the field. Demonstration bridge projects are being developed in other states such as Delaware, West Virginia and California. Continued research projects using composite reinforcing bars in concrete slabs are being studied in New Hampshire, Washington, D.C. and Michigan. Composite prestressing tendons and stay cables are being developed in Pennsylvania, Michigan, South Dakota and California.
There are three basic manufacturing techniques in producing composite structural products, with many variations and patented processes: 1) The pultrusion process involves a continuous pulling of the fiber rovings and mats through a resin bath and then into a heated die. The elevated temperature inside the die cures the composite matrix into a constant cross-section structural shape. 2) The filament winding process can be automated to wrap resin-wetted fibers around a mandrel to produce circular or polygonal shapes. 3) The layup process engages a hand or machine buildup of mats of fibers that are held together permanently by a resin system. This method enables numerous layers of different fiber orientations to be built up to a desired sheet thickness and product shape.
Characteristics of Composites
The mechanical properties of composites depend on many variables such as fiber types, orientations, and architecture. The fiber architecture refers to the preformed textile configurations by braiding, knitting, or weaving. Composites are anisotropic materials with their strength being different in any direction. Their stress-strain curves are linearly elastic to the point of failure by rupture. The polymeric resin in a composite material, which consists of viscous fluid and elastic solids, responds viscoelastically to applied loads. Although the viscoelastic material will creep and relax under a sustained load, it can be designed to perform satisfactorily. Composites have many excellent structural qualities and some examples are high strength, material toughness, fatigue endurance, and light weight. Other highly desirable qualities are high resistance to elevated temperature, abrasion, corrosion, and chemical attack.
Some of the advantages in the use of composite structural members include the ease of manufacturing, fabrication, handling, and erection. Project delivery time can be short. It took the Russell county engineer one day to install the deck panels in the first vehicular composite bridge. Composites can be formulated and designed for high performance, durability and extended service life. They have excellent strength-to-weight ratios. If durability can be proven to last 75 years, composites can be economically justified using the life-cycle cost method.
Some of the disadvantages in the use of composites in bridges are high first cost, creep, and shrinkage. The design and construction require highly trained specialists from many engineering and material science disciplines. The composites have a potential for environmental degradation, for examples, alkalis' attack and ultraviolet radiation exposure. There are very little or nonexistant design guidance and/or standards. There is a lack of joining and/or fastening technology. Because of the use of thin sections, there are concerns in global and local buckling. Although the light weight feature may be an advantage in the response to earthquake loading, it could render the structure aerodynamically unstable. In manufacturing with the hand layup process, there is a concern about the consistency of the material properties.
What Is Really in Composites?
As the name implies, advance fiber reinforced polymer composites is made of fiber reinforcements, resin, fillers, and additives. The fibers provide increased stiffness and tensile capacity. The resin offers high compressive strength and binds the fibers into a firm matrix. The fillers serve to reduce cost and shrinkage. The additives help to improve not only the mechanical and physical properties of the composites but also workability. The discussions that follow immediately will explain the basic functions and behaviors of the constituents.
The fiber is an important constituent in composites. A great deal of research and development has been done with the fibers on the effects in the types, volume fraction, architecture, and orientations. The fiber generally occupies 30% - 70% of the matrix volume in the composites. The fibers can be chopped, woven, stitched, and/or braided. They are usually treated with sizings such as starch, gelatin, oil or wax to improve the bond as well as binders to improve the handling. The most common types of fibers used in advanced composites for structural applications are the fiberglass, aramid, and carbon. The fiberglass is the least expensive and carbon being the most expensive. The cost of aramid fibers is about the same as the lower grades of the carbon fiber. "Other high-strength high-modulus fibers such as boron are at the present time considered to be economically prohibitive".
- Glass Fibers
The glass fibers are divided into three classes -- E-glass, S-glass and C-glass. The E-glass is designated for electrical use and the S-glass for high strength. The C-glass is for high corrosion resistance, and it is uncommon for civil engineering application. Of the three fibers, the E-glass is the most common reinforcement material used in civil structures. It is produced from lime-alumina-borosilicate which can be easily obtained from abundance of raw materials like sand. The fibers are drawn into very fine filaments with diameters ranging from 2 to 13 X 10-6 m. The glass fiber strength and modulus can degrade with increasing temperature. Although the glass material creeps under a sustained load, it can be designed to perform satisfactorily. The fiber itself is regarded as an isotropic material and has a lower thermal expansion coefficient than that of steel.
Table 1  Typical Properties E-Glass S-Glass Density (g/cm3) 2.60 2.50 Young's Modulus (GPa) 72 87 Tensile Strength (GPa) 1.72 2.53 Tensile Elongation (%) 2.4 2.9
- Aramid Fibers
These are synthetic organic fibers consisting of aromatic polyamides. The aramid fibers have excellent fatigue and creep resistance. Although there are several commercial grades of aramid fibers available, the two most common ones used in structural applications are Kevlar® * 29 and Kevlar® 49. The Young's Modulus curve for Kevlar® 29 is linear to a value of 83 GPa but then becomes slightly concave upward to a value of 100 GPa at rupture; whereas, for Kevlar® 49 the curve is linear to a value of 124 GPa at rupture (see Table 2). As an anisotropic material, it's transverse and shear modulus are an order of magnitude less than those in the longitudinal direction. The fibers can have difficulty achieving a chemical or mechanical bond with the resin.
Table 2  Typical Properties Kevlar 29 Kevlar 49 Density (g/cm3) 1.44 1.44 Young's Modulus (GPa) 83/100 124 Tensile Strength (GPa) 2.27 2.27 Tensile Elongation (%) 2.8 1.8
- Carbon Fibers
The graphite or carbon fiber is made from three types of polymer precursors -- polyacrylonitrile (PAN) fiber, rayon fiber, and pitch. The tensile stress-strain curve is linear to the point of rupture. Although there are many carbon fibers available on the open market, they can be arbitrarily divided into three grades as shown in Table 3. They have lower thermal expansion coefficients than both the glass and aramid fibers. The carbon fiber is an anisotropic material, and its transverse modulus are an order of magnitude less than its longitudinal modulus. The material has a very high fatigue and creep resistance.
Table 3  Typical Properties High Strength High Modulus Ultra-High Modulus Density (g/cm3) 1.8 1.9 2.0 - 2.1 Young's Modulus (GPa) 230 370 520 - 620 Tensile Strength (GPa) 2.48 1.79 1.03 - 1.31 Tensile Elongation (%) 1.1 0.5 0.2
Since its tensile strength decreases with increasing modulus, its strain at rupture will also be much lower. Because of the material brittleness at higher modulus, it becomes critical in joint and connection details, which can have high stress concentrations. As a result of this phenomenon, carbon composite laminates are more effective with adhesive bonding that eliminates mechanical fasteners.
The resin is another important constituents in composites. The two classes of resins are the thermoplastics and thermosets. A thermoplastic resin remains a solid at room temperature. It melts when heated and solidifies when cooled. The long-chain polymers do not chemically cross link. Because they do not cure permanently, they are undesirable for structural application. Conversely, a thermosetting resin will cure permanently by irreversible cross linking at elevated temperatures. This characteristic makes the thermoset resin composites very desirable for structural applications. The most common resins used in composites are the unsaturated polyesters, epoxies, and vinyl esters; the least common ones are the polyurethanes and phenolics.
- Unsaturated Polyesters
The unsaturated polyester amounts to about 75% of all polyester resins used in USA. It is produced by the condensation polymerization of dicarboxylic acids and dihydric alcohols. The formulation contains an unsaturated material such as maleic anhydride or fumaric acid which is a part of the dicarboxylic acid component. The formulation affects the viscosity, reactivity, resiliency and heat deflection temperature (HDT). The viscosity controls the speed and degree of wet-out (saturation) of the fibers. The reactivity affects cure time and peak exotherm (heat generation) temperatures. High exotherm is needed for a thin section curing at room temperature and low exotherm for a thick section. Resiliency or flexible grade composites have a higher elongation, lower modulus, and HDT. The HDT is a short term thermal property which measures the thermal sensitivity and stability of the resins.
The advantages cited in the unsaturated polyester are its dimensional stability and affordable cost. Other advantages include ease in handling, processing, and fabricating. Some of the special formulations are high corrosion resistant and fire retardants. This resin is probably the best value for a balance between performance and structural capabilities.
The epoxies used in composites are mainly the glycidyl ethers and amines. The material properties and cure rates can be formulated to meet the required performance. Epoxies are generally found in marine, automotive, electrical and appliance applications. The high viscosity in epoxy resins limits it use to certain processes such as molding, filament winding, and hand lay-up. The right curing agent should be carefully selected because it will affect the type of chemical reaction, pot life and final material properties. Although epoxies can be expensive, it may be worth the cost when high performance is required.
- Vinyl Esters
The vinyl ester resins were developed to take advantage of both the workability of the epoxy resins and the fast curing of the polyesters. The vinyl ester has higher physical properties than polyesters but costs less than epoxies. The acrylic esters are dissolved in a styrene monomer to produce vinyl ester resins which are cured with organic peroxides. A composite product containing a vinyl ester resin can withstand high toughness demand and offer excellent corrosion resistance.
Polyurethanes are produced by combining polyisocyanate and polyol in a reaction injection molding process or in a reinforced reaction injection molding process. They are cured into very tough and high corrosion resistance materials which are found in many high performance paint coatings.
The phenolic resins are made from phenols and formaldehyde, and they are divided into resole and novolac resins. The resoles are prepared under alkaline conditions with formaldehyde/phenol (F/P) ratios greater than one. On the contrary, novolacs are prepared under acidic conditions with F/P ratios less than one. Resoles are cured by applying heat and/or by adding acids. Novolacs are cured when reacting chemically with methylene groups in the hardener. The phenolics are rated for good resistance to high temperature, good thermal stability, and low smoke generation.
Since resins are very expensive, it will not be cost effective to fill up the voids in a composite matrix purely with resins. Fillers are added to the resin matrix for controlling material cost and improving its mechanical and chemical properties. Some composites that are rich in resins can be subject to high shrinkage and creep and low tensile strength. Although these properties may be undesirable for structural applications, there may be a place for their use.
The three major types of fillers used in the composite industry are the calcium carbonate, kaolin, and alumina trihydrate. Other common fillers include mica, feldspar, wollastonite, silica, talc, and glasses. When one or more fillers are added to a properly formulated composite system, the improved performance includes fire and chemical resistance, high mechanical strength, and low shrinkage. Other improvements include toughness as well as high fatigue and creep resistance. Some fillers cause composites to have lower thermal expansion and exotherm coefficients. Wollastonite filler improves the composites' toughness for resistance to impact loading. Aluminum trihydrate improves on the fire resistance or flammability ratings. Some high strength formulations may not contain any filler because it increases the viscosity of the resin paste. High viscosity resins may have a problem wetting out completely for composite with heavy fiber reinforcement. A filler should not be used with fiber volume greater than 50% for the sheet molding composite production method.
A variety of additives are used in the composites to improve the material properties, aesthetics, manufacturing process, and performance. The additives can be divided into three groups -- catalysts, promoters, and inhibitors; coloring dyes; and, releasing agents. Their roles are as simple as their names imply, and they need no further discussion here.
Professor Steenkamer and his coauthors at the University of Delaware stated it well when they wrote: "The development of a composite is a complex process that requires the simultaneous consideration of various parameters such as component geometry, production volume, reinforcement and matrix types and relative volumes, tooling requirements, process and market economics, etc. Every decision made during the product development process is intricately related to a set of three interacting decision's areas (i.e., materials, processing, and configuration)".
The development of the advanced composite technology is an engineer's dream for innovative design and application. The characteristics of a composite can be tailored and designed to meet any desired specifications. Most of the information and design data available on composites are in the aerospace applications, but they are protected under the guise of proprietary systems and/or military classified documents. Unlike conventional isotropic materials of steel and concrete, there are no readily available design charts and guidelines to help the structural engineer. When it comes to working with composites as opposed to conventional materials, as the author has discovered, the difference can be as dramatic as night and day.
The challenge in applying composites is for one to understand the behavior of not only the constituents in the composites but also the completed end product in the way they respond to an applied load. Since a separate design specification for composites bridges is not yet available, existing bridge design guidelines may have to be used with some caution. Under the current American Association of State Highway and Transportation Officials (AASHTO) LRFD bridge design specifications, the philosophy is one of a probability-based limit state approach. The four basic limit states that is applicable in bridge design with advanced composite materials are the service, fatigue and fracture, strength, and extreme event.
The service limit states dictate the level of deformation and crack width under normal service conditions for a bridge to perform satisfactorily during its service life. The fatigue and fracture limit states restrict the stress range under normal service conditions within an expected number of load cycles. They are to limit crack growth under repetitive loading and to prevent fracture during the design life of a bridge. The strength limit states are to ensure that both the global and local strength and stability are provided to resist the statistically significant load combinations as experienced by a bridge during its design life. Some overstress and structural damage may be inevitable, but the overall integrity of the structure will not be compromised. The extreme event limit states ensure the structural survival of a bridge during a major earthquake or unusual collision force.
Based on some of the design data obtained from completed composite bridge structures to date, the deflection and/or local buckling govern composite design. With the inherent low section modulus of a composite structural member and critical high stress demand in structural applications, a designer should consider the following features carefully in his design:
- Avoid abrupt thickness change in components
- Take advantage of geometrical shapes
- Take advantage of hybrid systems
- Use bonded assemblies and joints
- Provide good details for connected joints
- Avoid abrupt thickness change in components
In steel or concrete design, an increase in the plate or flange thickness will usually keep the stresses under control. Although this concept also works for composites, it is inefficient for a composite member to follow suit by increasing its overall part thickness. Because composites are viscoelastic materials, it is undesirable to create high stress risers. An understanding in the stress flow of a structural member will help a designer tailor the parts' thicknesses locally and avoid abrupt changes in its geometry.
- Take advantage of geometrical shapes
In most design using composites, the stress level is very low. An optimal design in composites balances the stress, deflection, and stability with the use of flanges, ribs, stiffeners, honeycomb or box-cells, or tubes to maximize the stiffness of the section. By placing flanges farther apart at the top and bottom of a hollow core, the section modulus can be designed to span longer structures. By proportioning and orienting the cells adequately, local buckling can be eliminated and material stiffness can be increased.
- Take advantage of hybrid systems
By taking advantage of the high stiffness in concrete and the high strength in composites, concrete filled carbon composite tubes for piles and main superstructure members in bridges are found to be very cost effective. Structural timber beams reinforced with composites in strategic locations have demonstrated an increase in the beam capacity. Prestressing tendons in concrete beams and decks are being studied in South Dakota and Michigan. Pultruded carbon FRP composite laminates bonded to steel beams and concrete slabs are being considered for strengthening of bridges. Composite fiberglass rods replacing reinforcing bars in concrete bridge decks are being studied in West Virginia. With any of these hybrid systems, the designer should account for the difference in the strains of each material affecting the compatibility of the total unit.
- Use bonded assemblies and joints
Much work needs to be done in developing good joints to assemble the composite members. The successful use of the epoxy adhesive technology from the aerospace industry has been transferred to many recent civil structural applications. The concept of using epoxied shear transfer toggle strips has been demonstrated in two composite bridges in the United Kingdom. Plate bonding using epoxy adhesive on thin laminates to strengthen civil structures is seen as a promising application. The column wrapped with carbon tows (sheets) will be as strong as the epoxy bonded overlapping splice. The ability to advance the composite technology in civil structures will depend on the integrity and durability of these joints.
- Provide good details for connected joints
Discontinuities within a structural system can be a designer's nightmare. Special attention must be given to the local stress flow, overall load path, and joint lines that create weak links or porosity introduced during the manufacturing process. Other irregularities introduced during the cutting/drilling and fit up process must be evaluated. It is important to select proper fasteners. Certain composites with high flexural modulus are very brittle and have a tendency to granulate; they would not be suitable with screws. The ability to connect the components into a structural system will enable composites to go far in civil applications. It needs a technological breakthrough from the current thinking of using nuts and bolts to connect its members.
There has been a great deal of development in the composite technology. It is an exciting time for civil engineers to be involved with composites. Both the US Government and private institutions are funding many demonstration bridge projects to show that advanced composite materials can be applied to rebuilding our highway infrastructure.
There is much to be learned about composites. Manufacturers, bridge owners, government officials, academia, researchers and contractors need to work together. Trade secrets should be honored and respected to the extent that information is provided to bridge owners to understand and evaluate the behavior of their structures. The bridge owner has been and will continue to be held responsible for the safety of the traveling public during the service life of a structure. Current laws require bridge owners to inventory and rate their bridges. The owners need to know the pertinent information and data that are used in bridge design, manufacturing, and fabricating. The AASHTO, ASTM, and American Concrete Institute have established numerous technical committees to develop design specifications, guidelines, standards, testing methods and methodologies. The Composites Institute which represents the composite industry has been coordinating the industry's effort to develop product design manuals, improve manufacturing processes, and collect test data. Numerous universities are offering research and design courses in composites. There are ample opportunities for civil engineers to participate and contribute to this growing technology.
- SIEBLE, F. & KARBHARI, V; A paper titled Advanced Composites for Civil Engineering Applications in the United States, University of California, San Diego, CA.
- BALLINGER, Craig; "Composites poised to make inroads as Highway Structural Materials," The Journal: Roads and Bridges; April 1994 Issue, pp. 40-44.
- SMITH, W; Resin Systems; Delaware Composites Design Encyclopedia (DCDE), Volume 3; 1990.
- LINDSAY, Karen; What is a Composite?; The Journal on Composites Design and Application, Fall 1995 Issue; pp 29-30.
- ZWEBEN, C.; Introduction to Mechanical Behavior and Properties of Composites Materials; DCDE, Volume 1; 1989.
- ABEREGG, Dale; SMC Design Guidelines for Structural Applications, The Journal: CDA Summer 1996 Issue; pp 22.
- "Plastics and Composites in Construction," The ENR Journal, November 7, 1994.
- STEENKAMER, D.A.; WILKINS, D.J.; KARBHARI, V.M.; Resin Transfer Molding: Part 1 - Materials and Preforming; Nov. 1992; CCM, UDEL, Newark, DE.
- ABEREGG, Dale, SMC Design Guidelines for Structural Applications, The Journal: CDA Summer 1996 Issue, pp. 22.
- WOLFE, Joanne, Tips for Using Fasteners, The Journal: CDA Summer 1996 Issue, pp. 34.
- LINDSAY, Karen; State-of-the-Industry: 1995-1996; The Journal: CDA February 1996 Issue, pp. 12-16.
- FHWA Training Manual in LRFD Design of Highway Bridges, Modjeski and Masters, Inc. October 1994.
- REISCH, Marc S.; Advanced Polymer Composite Makers Look Toward Infrastructure Market; March 25, 1996, C & EN.
- PODOLNY, Walter, Jr.; Winds of Change and Paradigms of Obsolescence; Proc., The National Steel Bridge Symposium, October 15-17, 1996, Chicago, Illinois; The National Steel Bridge Alliance, Chicago, Illinois.
* Kevlar® is a registered trade mark of E.I. du Pont de Nemours & Co.