Secondary Logo

Journal Logo


Bench Test Validation of a Dynamic Posterior Leaf Spring Ankle-Foot Orthosis

Bedard, Gary G. BSc, CO, FAAOP; Motylinski, Jennifer MPO; Call, Benjamin MPO; Gao, Fan PhD; Gray, Leslie MEd, CPO

Author Information
Journal of Prosthetics and Orthotics: January 2016 - Volume 28 - Issue 1 - p 30-37
doi: 10.1097/JPO.0000000000000081
  • Free


The posterior leaf spring ankle-foot orthosis (PLS AFO) emerged in the earliest years of the transition from leather- and metal-fabricated lower-limb orthoses into the present practice of thermoplastic devices. Thermoplastics facilitate the production of total contact orthoses that are lightweight and discreet with the capability of fitting into standard shoes.

Wide ranges of orthotic designs capitalize on the flexible nature of thermoplastic materials to provide biomechanical benefits to the patient. The first reference for the use of polypropylene in a lower-limb orthosis was described by Yates.1,2 He described polypropylene as a rigid and molecularly stable thermoplastic material with outstanding chemical resistance. He further stated that polypropylene is tough, lightweight, and odorless and yields high impact strength. In addition, the material can be vacuum formed to create a total contact fit without fear of patient tissue reaction. Yates2 presented an orthosis fabricated with polypropylene for the specific application of patients with drop foot, similar to the PLS AFO.

The biomechanics of the thermoplastic PLS AFOs were earliest described by Sarno and Lehneis3 at New York University. The thermoplastic PLS AFO was stated to provide flexibility and allow eccentric plantarflexion of the ankle at initial contact. The PLS was tailored to the patient so that tibial progression is not inhibited during stance phase. In addition, during the swing phase, the footplate design resisted plantarflexion of the ankle to prevent drop foot.

The thermoplastic PLS AFO is capable of the aforementioned biomechanical properties owing specifically to its unique design. An example of a polypropylene PLS AFO is shown in Figure 1.

Figure 1
Figure 1:
Posterior leaf spring ankle-foot orthosis (PLS AFO) with c-shaped posterior leaf fabricated with polypropylene. Figure 1A, Carbon-infused polypropylene PLS AFO with a flat blade posterior leaf.

The point of emphasis of the PLS design is the trimline posterior to the ankle malleoli. This posterior c-shaped cross-section spring is fabricated to the ankle shape and posterior shank radius of the patient and can vary in width. A study performed by Ramsey4 aimed to develop standards for the fabrication regarding trimline dimensions of the PLS to allow optimization of biomechanics during gait; however, most orthotists rely on clinical judgment for posterior leaf and PLS AFO fabrication.

Deficits in the c-shaped PLS design do exist. Mollan and James5 explained that the c-shaped PLS design will buckle once a certain limit of stress is reached. Once the leaf spring buckles, an opaque area appears on the plastic and is a definitive sign of fatigue failure and ultimate failure of the device. The alternate PLS design that Mollan and James5 offered is a flat blade as the PLS on the PLS AFO. The theory of the flat blade is to provide an area to measure and quantify the response of the orthosis to flexion stress.

Some authors hypothesized that the PLS portion of the PLS AFO would offer energy return during gait; however, when Ounpuu et al.6 evaluated the PLS using joint kinematics and kinetics, they found that no storage and return were observed at the terminal stance with standard c-shaped polypropylene PLS. This fact has led to the use of carbon fiber orthoses in clinical practice. These orthoses either are completely fabricated using thermoset carbon fiber laminates or use a carbon fiber element in the posterior leaf portion of the device. Most of the carbon fiber orthoses prescribed are off-the-shelf designs and lack the capability of being vacuum formed to produce a total contact orthosis that is beneficial in orthotic design. In a research study by Bartonek et al.,7 gait analysis, including kinematics and kinetics, was used to compare the PLS AFO with a carbon fiber orthosis (CFO) (dual carbon fiber spring AFO) in children with hemiplegia. The study concluded that during the ankle third rocker of gait, the CFO produced greater range of motion (ROM), angular velocity, and power generation in comparison with the thermoplastic PLS AFO. The increase was accredited to the fact that the carbon spring in the CFO was loaded during loading response, thus storing energy, and returned that energy during stance phase to cause a forward push of the tibia and increased push off during pre-swing.

The goal of this current validation study was to explore the use of a new polypropylene material prepregnated with strands of discontinuous carbon fiber in a PLS AFO design that incorporates a flat blade posterior spring segment (Figure 1A). The new carbon-infused polypropylene composite was designed for orthotic purposes with the aim of producing a material that will offer energy storing and return properties during gait yet have the ability to be vacuum formed during fabrication of the orthosis for a custom total contact fit for the patient. An additional benefit of this material is the ability to change the tibial angle of the shank component of the AFO post fabrication to meet the patient's footwear and gait characteristics. Heat modifications directly to the brace are not an option for the previously discussed off-the-shelf or custom thermoset carbon laminate AFOs; therefore, modifications to the footwear must be made instead.

This is not the first study of this new material, as Gao and Bedard8 systematically evaluated the effects of designs and heat treatment on the material properties of homopolymer polypropylene, copolymer polypropylene, and carbon-infused polypropylene composite. Stiffness and ROM testing revealed that overall ankle ROM after annealing was comparable with that for corrugated design. Carbon-infused polypropylene demonstrated significantly less ROM compared with other materials, proving that adding carbon fibers into the homopolymer significantly improves its mechanical properties.8

The experimental design of the current study will follow similar protocol to Gao and Bedard,8 yet this procedure was initially validated by Gao et al.9

In this study, carbon-infused polypropylene, with varied ply and densities of prepregnated discontinuous carbon fiber, was tested against an identical PLS AFO flat blade design fabricated in homopolymer polypropylene to compare the dynamic nature of the materials. Consistent with previous studies, it was hypothesized that the PLS AFOs fabricated with carbon prepregnated polypropylene composite demonstrate more dynamic mechanical properties as indicated by increased stiffness and energy storing properties (decreased index of hysteresis) than a standard thermoplastic homopolymer polypropylene.



A master positive model was rectified for use in fabricating the 12 test PLS AFOs. The master model was rectified with a flat raised area initiating below the ankle axis and extending up into the posterior mid-calf. The plaster modification served as the molding platform for the integral posterior flat blade element of the AFO. The width at the minimum point in the leaf spring was the same as the M-L measurement at the apices of the malleoli. The plane of the flat section was arranged horizontal to the long axis of the foot.10 The long axis of the foot was defined as a line bisecting the calcaneus and second toe on the plantar aspect. In this manner, the flexural property of the flat blade allows the tibia to progress over the foot in the natural line of progression of the transmalleolar axis. The pattern of the spring blade appeared as an inverted “Y” with the ears extending underneath each malleoli when outlined with an indelible pencil. The shape for the polypropylene spring element and the carbon-based spring element were derived from a foam pattern. Closed-cell thermoplastic foam was heated and formed over the indelible tracing on the AFO master model outlining the inverted Y-shaped plaster build up. The pencil ink transferred to the foam and was trimmed and edge finished for use as a pattern for the blade fabrication coupon to be replicated in polypropylene and carbon for all the sample orthoses. The 3-dimensional (3D)-shaped foam pattern was flatted on the respective polypropylene and composite sheet and an outline was drawn. The pattern was cut with a reciprocating saw and the edges were rounded with standard finishing cones on a rotary power tool.

The rectified master positive model was vacuum formed with 1/8-in polyethylene. The formed plastic was trimmed with a segmented anterior trimline along the anterior double butt seam to allow for removal from the positive model and to serve as a master mold to produce a series of identical plaster positive models with the associated rectification. The replicate positive models were then used to fabricate each test AFO in the same fabrication period. In that manner, all fabrication protocols could be reviewed and implemented with the least variability in the laboratory environment.

Of the 12 test orthoses, three AFOs were fabricated from natural (no coloring agent) homopolymer polypropylene with a doubler of the same polypropylene in the spring area. The host material for the overall orthosis and for the flat blade spring element was composed of a 3/16-in sheet polypropylene. The remaining nine test AFOs were fabricated from a commercial carbon-infused polypropylene in a 3/16-in gauge. The vacuum forming compatible commercial composite is composed of five plies with dual carbon plies with a total of 0.6 oz/yd2 of carbon. The stiffness of commercial composite host sheet was validated in a previous study.8

The posterior blade coupon was bonded into the host orthosis during the vacuum-forming process to create an integral spring. All of the sheet materials used in the fabrication of the test orthoses were a polypropylene resin base, either as the neat sheet material or the carbon-infused composite. Polypropylene has the feature whereby the material will chemically bond to itself when at melt temperature. Therefore, the dynamic flat blade spring becomes an integral element in the orthosis without the need of added adhesive or mechanical fastening.

The carbon spring coupon was composed of variants with different carbon densities, layers, and resin elements. The specification of the experimental carbon leaf spring prepregnated components offers a range of carbon content from 1.0 to 2.0 oz/yd2 and a gauge of 1/8 and 3/16 in. The leaf spring material is consolidated in the same manner as the commercial variety with the identical homopolymer food grade polypropylene resin. The plies of resin and carbon are varied to offer a prospective range of stiffness. Three composite test orthoses were fabricated from the three leaf spring prepregnated composite variants for a total of nine prepregnated test orthoses.

Polypropylene composite-leaf spring (PC-LS) 1.5: 1/8 in carbon leaf spring laminate with 1.5 oz/yd2 carbon content in seven plies

  1. 0.033 in polypropylene (PP)
  2. 0.5 oz/yd2 veil
  3. 0.033 in PP
  4. 0.5 oz/yd2 veil
  5. 0.033 in PP
  6. 0.5 oz/yd2 veil
  7. 0.033 in PP

PC-LS 1.0: 3/16 in carbon leaf spring with 1.0 oz/yd2 carbon content in five plies

  1. 0.62 in PP
  2. 0.5 oz/yd2 veil
  3. 0.62 in PP
  4. 0.5 oz/yd2 veil
  5. 0.62 in PP

PC-LS 2.0: 3/16 in leaf spring laminate with 2.0 oz/yd2 carbon content in nine plies

  1. 0.033 in PP
  2. 0.5 oz/yd2 veil
  3. 0.033 in PP
  4. 0.5 oz/yd2 veil
  5. 0.033 in PP
  6. 0.5 oz/yd2 veil
  7. 0.033 in PPO
  8. .5 oz/yd2 veil
  9. 0.033 in PPO

The sheet materials were heated in a PDQ BT-1 Infrared Oven (OTS Corp. Weaverville, NC, USA). Attainment of proper melt temperature level for the carbon-infused polypropylene entailed a heat soak technique to minimize a temperature gradient through the thickness of the material and overheating of the top surface. The quartz tube heaters used in the oven provide a medium infrared wave length that only penetrates the top surface of polypropylene by a few microns. Consequently, the core and the bottom of the sheet material are heated by conduction from the concentration of infrared energy on the sheet surface. A heat soak method involves monitoring the surface temperature of the material with a handheld infrared noncontact thermometer (model 22-325; Radio Shack Corp, Fort Worth, TX, USA). Once the selected melt heat range is attained on the surface, the sheet tray is pulled out of the oven to allow the core and bottom to continue heating but allowing the surface of the composite to cool. In this manner, the surface is not overheated from the concentration of the IR energy at the sheet surface and a more uniform full thickness temperature profile is attained. The sheet tray is pushed back into the oven, and once the surface temperature is again at the selected melt temperature, the sheet material is removed and molded. The host composite sheet and the spring blade fabrication coupon were all heated at the same time.

The polypropylene host sheetstock and polypropylene blade coupon were heated to a surface temperature of 360°F. The carbon-impregnated materials were heated to a surface temperature of 380° to 410 °F. Sheetstock molding temperature was measured with a handheld Radio Shack IR thermometer (model 22-325; Radio Shack Corp).

The commercial composite has higher sag strength than a homogenous polypropylene sheet because of the 3D discontinuous matrix of carbon fibers that are imbedded into the polypropylene resin. Consequently, a higher molding temperature can be used with the commercial composite to increase the flow of the resin along with the discontinuous fibers to optimize the molding of the melt temperature sheet onto the positive model. Sag strength is the resistance of a melt temperature plastic to resist gravity deformation. It can also be called melt strength and is related to the rheology science of polymer flow, viscosity, and deformation. In respect to polypropylene, sag occurs after the material has reached a sufficient temperature whereby the crystalline elements of the polymer chains have released and thus allowing flow to occur.

Polypropylene typically cannot sustain a high molding temperature because of its lower sag strength and the physical inability to hand carry the mold temperature sheetstock from the flat oven tray to the positive model for hand draping and vacuum pressure application. The increased sag strength of the commercial composite demonstrated the process benefit of stabilizing the sheet not only in the finished product by the increase in physical stiffness but also during the fabrication process with an increase in sag strength when the material is in a melt temperature state of flow. The addition of discrete layers of carbon veil in the composite mechanically inhibits gravity sag at melt temperature. Consequently, the wall thickness of the composite has the ability to be greater than neat polypropylene sheet in comparative process handling.

The 12 test PLS AFOs were all vacuum formed on a Becker Orthopedic VacStation BT. The samples were produced at sea level with one atmosphere of molding pressure (29.92 in Hg or 14.7 pounds per square inch [PSI]) and 410 cu ft/min flow from the surge tank, which is incorporated directly in the VacStation BT. The vacuum pump was a Welch Duoseal oil immersion vacuum pump (Duoseal model 1402; Welch-Llmvac, Niles, IL, USA). Each of the positive models was covered with two layers of thermoforming nylon to act as a bleeder layer to assist with the evacuation of air under the encapsulated plastic or composite. A closed cell foam collar was used on the muzzle of the molding manifold to enhance the vacuum seal and prevent the loss of surge pressure when the vacuum valve was opened. The melt temperature blade coupon was placed on the nylon encased positive model first, then the host sheet was molded over the whole positive. Two cotton strings were laid across the blade coupon with a short extension past the coupon. The string was bonded between the host layer and the blade coupon and acted as an evacuation route for any possible air that may have been trapped between the layers. In the 12 test AFOs, no air entrapment was visible in the finished orthoses when a focused light was placed behind the double layer of material for inspection; thus, a full chemical bond was executed.

All 12 of the AFOs had the same finished trimline to maintain consistency across the test series of orthoses. A 1/8-in polyethylene mask was vacuum formed over the first finished AFO on the master positive model to match the trimline to the subsequent test orthoses. The mask was placed on the outer surface of each subsequent unfinished molded AFO for trimline transfer.

Quantifying the physical characteristics of the four blade materials in the 12 finished PLS orthoses with a bench test apparatus provided a realistic material evaluation with transfer to actual clinical performance. Certified thermoplastic sheet materials are not offered in the orthotics and prosthetics (O&P) field as compared with the certified format as found with metal raw materials. The characteristics of thermoplastic materials are impacted by processing parameters, which in turn impacts the performance of a finished orthosis. Consequently, the bench testing of finished orthosis of a particular design offers some transfer insight into general clinical services. Thermoplastic materials pass through two heat histories from gas polymerization into raw solid bead form, then through the extrusion process, whereby the raw bead form is converted into flat sheetstock. The conversion into sheet form is the first-generation heat history. The second heat history occurs during vacuum thermoforming into the final orthosis. The heat history of sheet extrusion and final vacuum forming can greatly change the orientation of the long thermoplastic macromolecules of the thermoplastic resin as well as the ratio of amorphous and crystalline molecular structure, which all can impact the physical characteristics of the clinical product.

Published data on the characteristics of thermoplastic resin that is converted into sheet material is generated under the guidelines of ASTM D4101, whereby the thermoplastic test coupons are injection molded from the original bead form and then annealed to optimize the crystalline structure of the resin. Physical test coupons formulated under ASTM D4101 represent the idealized physical characteristics of a specific resin formula but may not be representative of the actual molecular structure found in the extruded thermoplastic sheet nor the finished clinical product derived from the same resin. Thus, bench testing of finished orthoses produced in a prescribed manner will offer the clearest translation of a design into clinical practice and the intended impact on gait characteristics.


An AFO plaster positive model was segmented at the ankle joint for use as a surrogate leg in the test apparatus. The mandrel extending from the surrogate leg was attached to a mounting jig with adjustable components that allowed precise positioning in both horizontal and vertical directions (Figure 2).

Figure 2
Figure 2:
Experimental setup for testing mechanical properties of the posterior leaf spring ankle-foot orthosis (PLS-AFO).

A motorized device with an inline torque sensor (Transducer Tech Inc, Newark, CA, USA) and optical encoder developed in an earlier study were used to move the AFO within the prescribed range of resistance torque. The motor was controlled by a motor drive under speed mode in which both the speed and direction of the motor rotation were modulated.

A NI PCI [National Instruments Peripheral Component Interconnect] card was used for data acquisition and motor control and a custom graphical interface written in LabVIEW (National Instrument Inc, Austin, TX, USA) was used to set testing parameters including speed, range of resistance torque, and display registered signals.

The PLS AFO was donned on the surrogate leg, inserted in a modified operative shoe that was firmly attached to the apparatus footplate, and secured via Velcro bands (Figure 2). The center of the motor shaft represented the anatomical surrogate leg ankle joint axis. The PLS AFO was passively moved by the motor at a prescribed speed of 5°/s with peak resistance torques of 20 Nm in both ankle dorsiflexion and plantarflexion. The motor stopped when the prescribed resistance torque was reached, then it switched movement direction (e.g., when it moved in dorsiflexion and reached the resistance torque of 20 Nm, it stopped and then moved in plantarflexion).

The resistance torque and corresponding angular position were registered simultaneously via the uniaxial torque sensor and motor encoder at a rate of 1000 Hz. Each test lasted 120 seconds and repeated three times for each sample.

Although dynamic testing can analyze kinetic and kinematic functions of AFOs on actual users during gait, differing pathologies and gait disabilities might render comparisons between subjects difficult to interpret. Although bench testing may not account for different patient groups, literature suggests that bench testing analysis of AFOs enables more accurate control of experimental conditions and improves repeatability.11 Therefore, this study will analyze material properties using bench testing.


The torque and position signals were digitally filtered using a fourth-order zero-lag low-pass Butterworth filter with cutoff frequency set at 5 Hz. The torque component attributed to the footplate due to gravity was subtracted from the original torque signal. The torque-angle relationship was plotted to establish the hysteresis loops (Figure 3). Index of hysteresis was calculated as the ratio of the area within the loop to the area below the ascending limb and presented in percentage. Each hysteresis loop was further divided into loading (plantarflexion) and unloading (dorsiflexion) (Figure 3).

Figure 3
Figure 3:
Sample plot of the ankle resistance torque (Nm) and position (degrees).8

The multiple loading and unloading curves were pooled and averaged respectively. The following mechanical properties of the joint were calculated based on the averaged torque-angle curves: positional torque (ankle angle measured at 10 Nm resistance on both loading and unloading curves), quasi-static stiffness at 10 Nm resistance torques (both loading and unloading), and energy loss based on the index of hysteresis. Hysteresis is associated with viscoelasticity, and the index serves as a quantification of relative energy loss during unloading.12


Analysis of variance was used to analyze the response variables including the ankle ROM, angular position at 10 Nm of resistance torque, and stiffness for all 12 tested devices. Tukey honestly significant difference (HSD) test was used for post hoc comparisons. The significant level was set at 0.05. All statistics were conducted in MATLAB statistics toolbox (MathWorks, Inc, Natick, MA, USA).


Average loading and unloading stiffness values of homopolymer polypropylene PLS AFOs were 0.905 and 0.832 Nm/°, respectively (P = 0.007, P = 0.001). Average loading and unloading stiffness values of 3/16-in five-ply carbon-infused polypropylene AFOs were 0.746 and 0.638 Nm/°, respectively (P = 0.007, P = 0.001). Average loading and unloading stiffness values of 1/8-in seven-ply carbon-infused polypropylene AFOs were 0.985 and 1.129 Nm/°, respectively (P = 0.007, P = 0.001). Average loading and unloading stiffness values of 3/16-in nine-ply carbon-infused polypropylene AFOs were 1.223 and 1.339 Nm/°, respectively (P = 0.007, P = 0.001) (Figure 4).

Figure 4
Figure 4:
Table of results for range of motion (ROM), hysteresis, stiffness (loading and unloading), and position at 10 Nm Torque (loading and unloading).

Average ROM for polypropylene AFOs was 26.986° (P = 0.002) (Figure 5A). Average ROM for 3/16-in five-ply carbon-infused polypropylene AFOs was 23.64° (P = 0.002). Average ROM for 1/8-in seven-ply carbon-infused polypropylene AFOs was 29.66° (P = 0.002). Average ROM for 3/16-in nine-ply carbon-infused polypropylene AFOs was 25.012° (P = 0.002).

Figure 5
Figure 5:
Range of motion (ROM) of polypropylene (3/16 in) and five-ply (3/16 in), seven-ply (1/8 in), and nine-ply (3/16 in) carbon fiber–infused composites.

Average position at 10 Nm of torque during loading and unloading of the polypropylene AFOs was 12.311° and 9.634°, respectively (P = 0.005, P = 0.139). Average position at 10 Nm of torque during loading and unloading of the 3/16-in five-ply carbon-infused polypropylene AFOs was 14.174° and 11.857°, respectively (P = 0.005, P = 0.139). Average position at 10 Nm of torque during loading and unloading of the 1/8-in seven-ply carbon-infused polypropylene AFOs was 9.328° and 11.982°, respectively (P = 0.005, P = 0.139). Average position at 10 Nm of torque during loading and unloading of the 3/16-in nine-ply carbon-infused polypropylene AFOs was 7.710° and 9.885°, respectively (P = 0.005, P = 0.139) (Figure 6).

Figure 6
Figure 6:
Stiffness of polypropylene (3/16 in) and five-ply (3/16 in), seven-ply (1/8 in), and nine-ply (3/16 in) carbon fiber–infused composites.

The average hysteresis for polypropylene AFOs was 15.98%, as shown in Figure 7A. Average hysteresis for 3/16-in five-ply carbon-infused polypropylene AFOs was 13.55%. Average hysteresis for 1/8-in seven-ply carbon-infused polypropylene AFOs was 14.08%. Average hysteresis for 3/16-in nine-ply carbon-infused polypropylene AFOs was 12.80% (Figure 7B).

Figure 7
Figure 7:
A, Index of hysteresis graph for homopolymer polypropylene. B, Index of hysteresis for nine-ply (3/16 in) carbon-infused polypropylene. Index of hysteresis is a quantification of energy loss during loading-unloading cycle.


Similar to Gao and Bedard,8 the stiffness value in this study is a representation of resistance to rotation moment per angular displacement. Our results showed that both 1/8-in seven-ply and 3/16-in nine-ply carbon-infused polypropylene AFOs were stiffer than 3/16-in homopolymer polypropylene AFOs. The 3/16-in five-ply carbon-infused polypropylene PLS AFO proved to be less stiff yet restricted ROM better than the 3/16-in homopolymer polypropylene PLS AFO. This was an unexpected finding and seems to contradict Gao and Bedard.8 One possible explanation is the Gao and Bedard8 study tested a solid-ankle AFO design, where this study tested a PLS blade design. The 1/8-in seven-ply carbon-infused polypropylene AFOs demonstrated greater stiffness while allowing greater ROM than 3/16-in homopolymer polypropylene AFOs. The increased stiffness of the carbon-infused polypropylene composite materials will produce a smaller index of hysteresis by allowing less deformation under increased torque, thus providing greater dynamic energy than their homopolymer polypropylene counterparts do. The increased stiffness will offer a rigid toe lever during preswing for dynamic push off during gait. Furthermore, thinner carbon-infused polypropylene AFOs can provide greater stiffness than their thicker homopolymer polypropylene counterparts without sacrificing ROM for the user when such an outcome is indicated.

Furthermore, one could hypothesize that the stiffness of 3/16-in five-ply carbon-infused polypropylene AFOs was significantly less than that of 3/16-in nine-ply carbon-infused polypropylene because the latter contains twice the carbon content. Research with variations of material thickness and consistent carbon fiber ply and density should be carried out to validate theorized ROM based on thickness properties. Similar research would be indicated with variations of material density and consistent ply and thickness to isolate the effects of density on material stiffness.

Clinically, using the results of data researched by Ramsey4 stating that the average stiffness factor for slow to normal walking speeds based on Palmer's11,13 and Winter's11,13 studies was 0.0115 Nm/°/kg, it is possible to calculate the amount of material stiffness necessary for optimal ambulation per material and orthotic device. This chart is represented in Figure 8 and depicts weights calculated for both stiffness factors of 0.0115 and 0.01 Nm/°/kg. The 0.01 Nm/°/kg stiffness factor was suggested by Ramsey4 to more accurately depict the stresses placed on a PLS AFO because these devices bend on an axis posterior to the anatomical ankle axis, thus reducing the stiffness requirements of the AFO.

Figure 8
Figure 8:
Chart of stiffness of the devices per material as it relates to a optimal patient weight based on a stiffness factor described by Ramsey.4

Range of motion is critical to orthotic design, and it is essential for orthotists to understand the material properties and allowable ROM of prescribed devices. Limiting ROM could cause adverse effects for the patient, such as muscle weakness or contracture formation. The results of this study showed that the thickness of a material plays a large role in its ROM properties. The 1/8-in seven-ply, thick carbon-infused polypropylene exhibited the largest ROM, followed closely by the 3/16-in homopolymer polypropylene. The other two 3/16-in carbon-infused polypropylene materials, five and nine ply, exhibited a lesser degree of ROM. This finding suggests that the thickness of plastic is essential to design because the 1/8-in seven-ply carbon-infused plastic demonstrated ROM properties similar to homopolymer polypropylene in comparison with the other 3/16-in carbon-infused variants. All ROM values were within 5° of one another for allowable device ROM at 20 Nm resistance torque.

A further ROM value measured in this study was angular position at 10 Nm of resistance torque. This value is a representative of how stiffness affects ROM at moderate torque levels. The trend exhibited through this research indicates that, at moderate torque levels, the stiffer the material, the less angular ROM is allowed regardless of material thickness. This trend was significant for the loading curve (plantarflexion) only and further studies would be needed to verify trends in the unloading curve (dorsiflexion) of the hysteresis.

The index of hysteresis is a quantification of energy loss during the loading-unloading cycle of the orthotic devices.8 The findings of this study were consistent with the hypothesis that the index of hysteresis, or energy loss, would decrease for the carbon-infused polypropylene PLS AFO variants. The index of hysteresis was the greatest for the homopolymer polypropylene PLS AFO and lowest for the 3/16-in nine-ply carbon-infused polypropylene PLS AFO. The 1/8-in seven-ply carbon-infused polypropylene had the largest value for the carbon-infused variants; however, this result is because of the increased ROM allowed by this device owing to the decreased thickness of the material. These results might indicate that because less energy is lost during bench testing loading-unloading cycle for the carbon-infused PLS AFOs, these devices and materials might be able to offer energy to the patient during the gait cycle. This trend is promising for further orthotic studies and research on carbon-infused polypropylene with respect to energy loss.

Limitations existed in this study and are acknowledged. Materials tested were in varied ply, density, and thickness of carbon-infused polypropylene. The varied material designs make it difficult to attribute the dynamic material properties to a specific parameter such as material thickness or ply content. Although the results of this study show definite trends in material properties, further research is required to confirm observations. Limitations occurred during this study because it was difficult to align each orthosis in the exact plane as the previous orthosis. For this reason, the axis of rotation may have differed slightly between test trials. To counter this fact, multiple trials were run per material and device; however, differences and error will undoubtedly occur. To decrease error, 27 representative cycles were averaged per material grouping; however, it must be noted that for the 1/8-in seven-ply carbon-infused polypropylene, three cycles were eliminated and the data presented was an average of 24 representative cycles.

Additional suggested research would be to perform studies on material properties of PLS AFOs during gait, instead of bench testing. The plaster model that the PLS AFOs were mounted on during bench testing is not a good replica of patient soft tissue. Properties of the device would change if clinically studied. The materials were tested at a constant test speed, unlike typical stresses placed on an orthotic device during gait. Further research should include data collection in a clinical gait laboratory.


The research results followed the proposed hypothesis: PLS AFOs fabricated with carbon prepregnated polypropylene composite demonstrated more dynamic mechanical properties, as indicated by increased stiffness and decreased index of hysteresis, than homopolymer polypropylene orthoses, thus decreasing energy loss and providing a rigid toe lever at preswing. The nine-ply carbon-infused polypropylene PLS AFO demonstrated a 35.14% increase in stiffness over the corresponding homopolymer polypropylene orthosis and a 21.81% reduction in hysteresis for the broad expanse of the test results for all the test orthoses. Limitations existed in the study and further research with human subjects within a clinical gait laboratory is recommended to match bench testing to potential improvements in locomotion.


1. Yates G. A method for the provision of lightweight aesthetic orthopaedic appliances. Orthopaedics 1968; 1: 153–162.
2. Yates G. Molded plastics in bracing. Clin Orthop Relat Res 1974; 102: 46–57.
3. Sarno JE, Lehneis HR. Prescription considerations for plastic below-knee orthoses. Arch Phys Med Rehabil 1971; 52 (11): 503–510.
4. Ramsey JA. Development of a method for fabricating polypropylene non-articulated dorsiflexion assist ankle foot orthoses with predetermined stiffness. Prosthet Orthot Int 2011; 35 (1): 54–69.
5. Mollan RB, James WV. A new flexible design of drop-foot orthosis. Injury 1977; 8 (4): 310–314.
6. Ounpuu S, Bell KJ, Davis RB 3rd, DeLuca PA. An evaluation of the posterior leaf spring orthosis using joint kinematics and kinetics. J Pediatr Orthop 1996; 16 (3): 378–384.
7. Bartonek A, Eriksson M, Gutierrez-Farewik EM. Effects of carbon fibre spring orthoses on gait in ambulatory children with motor disorders and plantarflexor weakness. Dev Med Child Neurol 2007; 49 (8): 615–620.
8. Gao F, Bedard GB. Effects of materials, design and heat treatment on SAFO mechanical properties: a preliminary report. J Prosthet Orthot 2013; 25 (3): 143–150.
9. Gao F, Carlton W, Kapp S. Effects of joint alignment and type on mechanical properties of thermoplastic articulated ankle-foot orthosis. Prosthet Orthot Int 2011; 35 (2): 181–189.
10. Milliron M, Loesel S, Micheli L, Edington C. Comparison of methods for calculation of foot progression angle. Med Sci Sports 1992; 24 (5): S173.
11. Palmer ML. Sagittal Plane Characterization of Normal Human Ankle Function Across a Range of Walking Gait Speeds. Boston, MA: Department of Mechanical Engineering, Massachusetts Institute of Technology; 2002.
12. Kubo K, Kanehisa H, Kawakami Y, Fukunaga T. Influence of static stretching on viscoelastic properties of human tendon structures in vivo. J Appl Physiol (1985) 2001; 90 (2): 520–527.
13. Winter DA. Biomechanics and Motor Control of Human Gait: Normal, Elderly and Pathological. 2nd ed. Waterloo, Ontario, Canada: University of Waterloo Press; 1991.

ankle-foot orthosis; posterior leaf spring; AFO; PLS; carbon fiber; prepreg; polypropylene; hysteresis

© 2016 by the American Academy of Orthotists and Prosthetists.