O&P Library > Orthotics and Prosthetics > 1986, Vol 40, Num 4 > pp. 44 - 58

Orthotics and ProstheticsThis journal was digitally reproduced with permission from the American Orthotic & Prosthetic Association (AOPA).

Funding for this project was provided by the American Academy of Orthotists and Prosthetists through a grant from the US Department of Education (grant number H235K080004). However, this does not necessarily represent the policy of the Department of Education, and you should not assume endorsement by the Federal Government. For more information about the Academy please visit our website at

You can help expand the
O&P Virtual Library with a
tax-deductible contribution.

View as PDF

with original layout

Evaluation of High Strength Materials for Prostheses

Virgil W. Faulkner, C.P.O. *
Martha Field, M.S. *
John W. Egan, M.S. *
Norman G. Gall, M.D. *


The weight of a prosthesis has always been a problem for prosthetic researchers. According to Mooney, most below-knee prostheses, laminated in the normal prosthetic laboratory, weigh about five pounds. Below-knee prostheses are usually attached to the limb by a strap around the thigh or with wedges pressing inwards above the condyles of the femur. With normal gravitational forces, this weight creates a friction between the residual limb and the prosthetic socket interface that may cause skin breakdown.

The weight of a prosthesis may cause excessive muscle work that will result in high energy consumption for amputees. Mooney states that, "a standard prosthesis requires approximately 12 percent more energy consumption" and "energy consumption is the key to successful ambulatory activities."

Ganguli, et al. stated that, "with respect to energy expenditure, the degree of departure from normal performance standards in the below-knee amputee fitted with a patellar tendon bearing (PTB) prosthesis is quite high." Cummings, et al. states that, "a distally applied weight of 2 1/2 pounds would be expected to add five to ten percent to the energy requirement of ambulation." Fisher and Gullickson5 state that below-knee amputees "walk 36 percent slower, expending two percent more Kcal/min and 41 percent more Kcal/mtr than the normal person." Waters, et al. found that vascular below-knee amputees walk 41 percent slower and expended 55 percent more Kcal/mtr/Kcal/kg than non-amputees.

The need for lighter weight prostheses is often cited in the literature and occasionally an innovative procedure will surface; however, when the technology differs from that in current practice, the prosthetic clinic team has difficulty adapting to it. The procedure described by Wilson9 was not familiar to the prosthetist; as a consequence, this very lightweight prosthesis is not commonly fabricated.

Prostheses are normally excessively heavy, which tends to increase residual limb trauma and energy expenditure with the likelihood of less successful prosthetic function. It is the intent of the clinic team to provide an appliance that will stand up under the strain of constant use. With these considerations in mind, the Rehabilitation Engineering Lab (REL) at the University of Texas Health Science Center at San Antonio (UTHSCSA) proposed to determine if a material could be designed which would utilize normal prosthetic laboratory techniques, yet allow the prosthetist to produce a below-knee prosthesis weighing less than two pounds and having the strength to adequately support normal ambulation loads.


Aramid® fibers and carbon fibers were selected as new materials to be used as a reinforcement for the lamination of prostheses because:

  •  Aramid® fibers have a very high tensile strength (Figure 1) and the elongation to break ratio is very low (Figure 2).
  • Fig. 1 and Fig. 2

  •  Carbon fibers exhibit an excellent modulus and their density is lower than many other materials currently used for strength in prostheses (Tables 1 and 2).
  • Table 1. and Table 2.

The tensile strength of Aramid® and carbon fibers is far superior to nylon, the material normally used by many prosthetists. The nearly linear stress/strain curve to failure of Kevlar® 29 (Aramid® fiber) is similar to that of glass, but unlike those of other organic fibers (Fig. 3 ). Because it is relatively insensitive to fiber surface defects, the tensile strength of Kevlar® 29 is uniform along the length of the fibers.

Research work in the area of orthotics and prosthetics using carbon fibers has been directed primarily toward orthotics. In 1976, N.A.S.A. published a technical brief in which they described a new, lightweight brace constructed of fiber reinforced polymer materials. Also in 1976, the Southwest Research Institute published a final technical report prepared by S.R. McFarland and G.C. Grimer11 in which they reported producing a pair of bilateral long leg braces from carbon fiber filaments. These braces weighed approximately IV2 pounds each, including the footplate which was formed of steel.

The orthoses produced by N.A.S.A. and the braces produced by McFarland at Southwest Research Institute both employed a very lengthy process which requires placing layers of composite materials on an intercore and laminating these materials together to be used as struts for the orthosis. Neelham, in his paper, "Carbon Fiber Reinforced Plastic Applied to Prosthetics and Orthotics," described a process similar to the one employed by N.A.S.A. and Southwest Research to fabricate a harness for externally powered upper extremity prostheses that were fitted to thalidomide damaged children. He also fabricated a thoracolumbosacral orthosis and a bilateral hip-knee-ankle-foot orthosis.

The fabrication process and the technology needed to fabricate these orthoses and prostheses require extensive retraining in laboratory techniques for prosthetists and orthotists in this country. New machines and tools would have to be installed. Richard Striebinger, in a letter to S.R. McFarland dated February, 1983, stated that his group at the Rensselaer Polytechnic Institute in New York had fabricated an orthosis in a sandwich construction using graphite, Kevlar® 29, and an epoxy matrix along with a foam core. This process, like the others, requires a long, complicated curing process under vacuum at room temperature.

Hittenberger and Putzi, at the V.A.M.C. lab in Seattle, Washington, reported they had developed a laminating procedure for lightweight prostheses which requires one of the laminations to be split and a foam core removed. This produced a prosthesis that weighed approximately 1 1/2 pounds. However, the lab procedures, as described, require the prosthetist to cut the prosthesis posterially along the sagittal line. This would tend to weaken the prosthesis in an area that receives very high stress and might cause it to break. The "Ultralight Below Knee Prosthesis" requires a "hand draped" vacuum formed fabrication procedure and polypropylene polymers. These are split posteriorly and later welded together. While the prostheses are ultralight when compared to conventional systems, the process requires new technology, additional tools and machines, and the end product is prone to failure, due to the high stress placed on the ankle-foot components during the forming process, and because of inappropriate heating and cooling of the plastic. This procedure has not gained acceptance by the prosthetic profession.


This project was designed to study the following objectives:

  •  To establish manufacturing techniques and criteria for knitting Aramid® and carbon fibers into stockinette materials suitable for lamination in prosthetic laboratories.
  •  To determine which fiber or combination of fibers would make the strongest and lightest weight prosthesis.
  •  To determine the best polymer (acrylic—epoxies—polyesters) for laminating these fibers in prostheses.


Carbon fibers and yarns are made by several companies in the United States, however, most of these products cannot be knitted into materials that are suitable for normal prosthetic applications because the fibers are so soft. In their natural state the fibers must be braided into heavy bulky strands to eliminate breakage during the knitting process. These bulky braids result in an undesirable uneven surface on the completed prosthesis.

Aramid® fibers and yarns in a variety of sizes are manufactured in the United States. Most of these are suitable for knitting purposes. In addition, the Otto Bock Company of Minnesota has developed a lamination technique using carbon fibers in a mat form,16 but reinforcement materials in a mat form are not normally used in the prosthetic lab. The superior properties of Aramid® and carbon fibers have prompted several companies to develop an assortment of fabrics to be used for prosthetic laminating.

Aramid® fibers are used in Aralon.®* This product is described as a high strength "stockinette" made of high technology fibers next in strength to that of carbon. The manufacturer claims that Aralon® "produced a prosthesis over 40 percent stronger and almost half the weight of conventional prostheses and that Aralon® is 2 1/2 times superior in ratio of fiber strength to weight than nylon." It also is claimed to have superior impact and fatigue resistance and excellent thermal stability with little change in dimension over normal temperature ranges. Aralon® is said to be compatible with both polyester and epoxy resins, and stretches like regular "stockinette." Carbon fibers in combination with glass and Aramid® were knitted into a stockinette material for this project by IPOS.**

A stockinette material made from a combination of carbon and glass fibers** has been available for several years, but most prosthetic facilities have not used it because it is very expensive, the glass fibers are health hazards to work with, and the knitted material when laminated does not have a smooth appearance. It is claimed this carbon fiber material is compatible with an acrylic resin, trade-named Carbon Acryl.®** According to the manufacturer, Carbon Acryl® has an additive that makes it very compatible with the carbon fibers and causes a "chemical bond" during lamination.

  • The following yarn specifications were obtained for knitting and testing by the Knit-Rite Company of Kansas City, Missouri:
    —Aramid®: Kevlar® 29—14/1 and 20/1
    —Carbon: Pyron—4/10 w.c. and 2/32 w.c. Panex (retired)—30Y800, 30Y300 and 30R —Glass: Fiberglass—150—1/0-1 —Nylon: Stretch nylon—1/100 Type 66 D-4 Perma-Set (The above yarns were knit in stockinette and rib stitch by Knit-Rite, Inc., of Kansas City, Missouri, as outlined in Figure 4.)
  • Fig. 4

Using the knitted stockinette materials from Knit-Rite and IPOS (Fig. 5 ), we laminated a series of test models using the new stockinette material with:

  •  IPOS Carbon Acryl® acrylic resin
  •  Epocast 502 epoxy resin
  •  Laminac 4110 polyester.

To restrict variability in strength measurements due to physical and geometrical factors, a cylindrical aluminum mold with two flat sides of equal proportions (Fig. 6 ) was machined and fabricated to be used as the model for laminating all of the test laminations. Coupons measuring two and one half centimeters by five centimeters were cut from each of the laminations (Fig. 7 ). These coupons were tested for strength using the Instron.® The Instron® conventionally measures strength and flexoral properties of plastics. It conforms to the American National Standard K6575-1971. This testing method has been approved for use by agencies of the Department of Defense to replace Method 1031 of Federal Test Methods Standard 406 and for listing in the DoD Index of Specifications and Standards. The instrument provides a graphic readout of the force (measured in Newtons) required to fracture the coupons.***


Using each of the different materials with each of the different resins, a series of test models were laminated under vacuum pressure over the custom designed aluminum mold.

  • Acrylic Resins—Using the custom made mold, we laminated the stockinette made from the Aramid,® nylon, and glass fibers separately and in combination using the acrylic resins, as follows:
    —Over the custom mold, we pulled a poly vinyl alcohol sheet (PVA) and applied vacuum under the PVA to insure good mold clarification.
    —We applied the stockinette, and over this stockinette we pulled a PVA bag to hold the acrylic resin.
    —Vacuum was applied under this bag to insure good mold conformity.
    —The laminating resin was prepared by combining 250 grams of carbon Acryl® with enough hardening powder to effect a cure time of 30 minutes.
    —This mixture was then poured into the PVA bag, allowed to impregnate the stockinette, and then cured.
    —From the laminated model, we cut two coupons, 2 1/2 cm by 5 cm.
    —To test the strength of the coupons, they were placed in the Instron,® using a three-point bending apparatus on supports spaced 30mm apart. A downward force was applied exactly at the center of the coupon at a rate of 10mm descent per minute. The strength of the material was measured as peak force at fracture.
  •  Epoxy Resin—Using the above described procedures, we laminated a new series over the custom made model using epoxy resin.
  •  Polyester Resin—Using the above described procedure, we laminated a new series over the custom made model using polyester resin.

At this time, our project has produced more than 300 laminated coupons using the various combinations of fibers. The strongest coupons obtained from the various combinations of Aramid,® carbon, nylon, and glass fibers are listed in Table 3. .


Coupons of standardized width and length, but variable thickness, were tested in a three-point transverse loading apparatus using the Instron® for administering a measured load. Thickness, maximum transverse breaking force, and the standardized width and length parameters were then compiled, and the transverse strength computed according to the formula,Equation 1.

S— is the maximum stress incurred by an "extreme fiber" most distant from the central bending axis;
F— is the transverse load in New-tons;
L— is the span between the two supports (30mm in this experiment);
z— is the "section modulus" characteristic of the cross-section geometry. For these coupons it is equal to: 1/6 * width * Thickness.

Therefore,Equation 2.

Coupons were grouped to the type resin and fiber combination; (See Fig. 8 and 9 and Fig. 10 ).

The appropriate individual transverse strength measurements were then pooled, and means and standard deviations computed (Table 3. ). The relatively large standard deviations in some of the groups are due in part to the nature of the laminating process currently in widespread use. When woven tubular stockinettes are pulled over a particular prosthesis shape, the orientation and overlap of fiber layers becomes arbitrary within certain bounds set by the stockinette manufacturer's knitting pattern. Accordingly, when test coupons are cut from the laminated prostheses, there is no way to control for direction or degree of offset of fiber layers. Since this element of randomness would creep into all tests, it was concluded that a mean strength estimate would reflect a fairly respresentative number for an "average" prosthesis made in this clinically typical manner.

To illustrate a comparison of "typical" prostheses weights using any of the several possible combinations, we choose a model below-knee prosthesis laminated in nylon/polyester by a local prosthetic facility. The facility was unaware that the below-knee prosthesis was to be used for this research project.

The finished prosthesis, including the socket, was first coated with a castable ure-thane elastomer produced by Smooth-on, Inc., of Gillette, New Jersey. After curing, the elastomer was carefully removed without stretching, then cut into eleven pieces

in such a way that they would lay approximately flat. These pieces were then measured using a 2-D digitizing planimeter. This area figure was taken to represent the total surface area of the prosthesis, excluding the plantar surface of the prosthetic foot. For this prosthesis, the area totaled some 2,300 square centimeters.

Next, circular corings were taken in various areas or "zones" on the prosthesis: four in the socket wall, and three in various places down the leg. For the socket cores, which penetrated both the outer prosthetic wall and the socket inner wall, two distinct layers of hardened composite were visible. Measurements of layer thickness were made. Three zones emerged: Zone 1i, the inner lay-up thickness of the socket wall itself; Zone 1o, the outside wall thickness of the socket; and Zone 2o, the outside wall thickness everywhere else in the leg.

A set of nylon/polyester coupons were tested to obtain a figure for the material's strength (as maximum stress). By calculating the equivalent breaking force required to break a nylon/polyester coupon with thickness equal to that of each zone in the prosthesis, a "Design Break Force" figure was obtained for each zone (Fig. 11 ). Then, using the stress numbers determined for each test material, an estimated thickness could be calculated for any new material used to build a prosthesis having a similar "Design Break Force" for each zone. Furthermore, knowing the density of each composite, the surface area (2,300 cm) and thickness of material requred, a weight figure was generated giving the minimum weight of composite materials required in an equivalent "typical" prosthesis (Fig. 4 ).


Knitted combinations of high-strength yarns were laminated with different resins and laboratory tested in order to obtain a material which could be used for making lightweight, high-strength prostheses and orthoses by facilities using techniques and equipment readily available to them.

This project has established knitting specifications for stockinette manufacture using Aramid® and cotton yarns. These yarns and the combinations tested may not be the most suitable for prosthetic laminations because of the many variables, i.e., price, availability, combinations not tested, and the fact that newer and stronger fibers are waiting to be discovered.

Although prototypes of prostheses have been made by the Rehabilitation Engineering Laboratory, the actual clinical work still needs to be done. However, the results of this research indicate that materials have been identified which have potential and should be tested further using controlled experimental designs.


The researchers would like to thank the following individuals and organizations for their help which made this project possible:

  •  National Institute of Health Research
  •  The Veterans Administration Research and Rehabilitation Department.
  •  Dr. Barry Norling, Department of Restorative Dentistry, UTHSCSA
  •  Dr. William Stavinoha, Department of Pharmacology, UTHSCSA
  •  Cono Farias, Photographer, Department of Radiology, UTHSCSA
  •  David Gipson, Orthotic Technician, Department of Physical Medicine and Rehabilitation

*Manufactured by Comfort Manufacturing Company of Burlington, New Jersey
**IPOS Komman Ditgesellschaft, Luner Renn Bahn 14.D2120 Luneberg
***1 Newton = 102 grams.


  1. Staros, A., "Materials and External Prostheses," Bull. Pros. Res., 10 (8): 77-91, 1986.
  2. Mooney, V., "Innovations in Care of the Amputee," Text Med., 75: 98-102, 1975.
  3. Ganguli, S., Datta, S.R., Chatter, J.E. Bimroy, J., "Metabolic Cost of Walking at Different Speeds with Patellar Tendon Bearing Prostheses," J App Physiol, 36 (4): 440-443, 1974.
  4. Cummings, V., March, H., Steve, L., Robinson, K.G., "Energy Cost of Below Knee Prostheses Using Two Types of Suspension," Arch Phy Med Rehabil, 60: 293-297, 1979.
  5. Fisher, S.V., Gullickson, G., "Energy Cost of Ambulation in Health and Disability: A Literature Review," Arch Phys Med Rehabil, 59: 124-133, 1978.
  6. Waters, R., Perry, J., Antonelli, D., Hislop, H., "Energy Cost of Walking Amputees: The Influence of Level of Amputation," J Bone & Jt Surg, 15: 42-46, 1976.
  7. Committee on Prosthetic/Orthotic Education, The Geriatric Amputee, Principles of Management, N.A.S. 1974.
  8. Murphy, E.F., "The Fitting of Below Knee Prostheses," in Human Limbs and Their Substitutes, Hafner Publishing Co., New York, 1968.
  9. Wilson, A., Pritham, C, Stills, M.," Manual for an Ultralight Below Knee Prosthesis, Moss Rehabilitation Center, Temple University, 1976.
  10. N.A.S.A. Tech. Brief B75-10303, Lightweight Orthotic Braces, 1975.
  11. McFarland, S.R., Grimer, G.C., Application of Lightweight Composite Materials for Use in Artificial Limbs and Limb Braces.
  12. Neelham, R.L.P., "Carbon Fiber Reinforced Plastic Applied to Prosthetics and Orthotics," Biomed Engin, 3: 305-314, 1981.
  13. Personal Letter from R. Striebinger to S. McFarland, The Southwest Research Institute, San Antonio, Texas.
  14. Hittenberger, D., Putzi, R., "A Laminated Ultralight Prosthesis," Orthotics and Prosthetics, Vol., 39, No. 1, pp. 41, 46, Spring, 1985.
  15. Quigley, M., Irons, G.P., Technical Report, Rehabilitation Engineering Center, Rancho Los Amigos, Downey, California, 1977.
  16. Roman, E.J., and Mott, L., "An Acrylic Lamination Technique for an Ultralight Below Knee Prosthesis," Orthotics and Prosthetics, Vol. 34, No. 4, pp. 32-39, December, 1980.
  17. Sanders, G.T., Lower Limb Amputations: A Guide to Rehabilitation, F.A. Davis Company, Philadelphia, 1986.

O&P Library > Orthotics and Prosthetics > 1986, Vol 40, Num 4 > pp. 44 - 58

The O&P Virtual Library is a project of the Digital Resource Foundation for the Orthotics & Prosthetics Community. Contact Us | Contribute