Skip to contentUnited States Department of Transportation - Federal Highway AdministrationSearch FHWAFeedback


<< PreviousContentsNext >>

High Performance Concrete Pavements
Project Summary



As part of a comprehensive research program investigating the use of glass fiber-reinforced plastic (GFRP) reinforcing bars in concrete pavements, the West Virginia University initiated a research study consisting of three research elements: (1) construction and evaluation of GFRP-doweled pavement sections; (2) investigation of moisture diffusion in GFRP dowels; and (3) construction and evaluation of a CRCP pavement section containing GFRP rebars. The use of GFRP bars in lieu of steel bars is expected to alleviate the problems associated with steel corrosion and steel distress. At the conclusion of the project, a workshop and technology transfer initiative are scheduled to be held.

Study Objectives

As mentioned above, three separate investigations are being conducted under this research project. Specific objectives of each investigation are described below.

GFRP Dowel Study

The objectives of this study are to (Li 2004; Vijay, GangaRao, and Li 2004):

  1. Evaluate the use of GFRP dowel bars as load transfer devices in JPCP under HS-25 static and fatigue loads and compare their response with JPCP consisting of steel dowels. This includes the conduct of laboratory and field evaluations of JPCP with both GFRP and steel dowels, as well as the conduct of analytical modeling of dowel response in terms of maximum bending deflection, relative deflection, and bearing stress.
  2. Evaluate the long-term performance of field rehabilitation of JPCP using GFRP dowels.
Moisture Diffusion Study

This is a laboratory study with the following objectives (Gupta, Rana, and GangaRao 2004):

  1. Measure the diffusion coefficients and equilibrium moisture contents, and establish the mechanism of water diffusion through neat and glass fiber-reinforced DERAKANE 411-350 Momentum vinyl ester resin bars.
  2. Reduce the diffusion coefficient through GFRPs by dispersing nanoclay in the matrix.
  3. Study the effect of clay loading on the diffusion properties of GFRPs by varying the amount of clay (1, 2, and 5 percent by weight).
  4. Study the effect of alkaline and saline solutions on the diffusion of water through GFRPs.
  5. Take scanning electron microscope (SEM) images of damage at the surface of GFRP after prolonged immersion in distilled water.
  6. Determine the changes in tensile strength and impact strength of vinyl ester-clay nanocomposites with varying amounts of nanoclay and correlate the results with available theories.
  7. Use dynamic mechanical analysis (DMA) to study the structural changes due to immersion in distilled water for extended periods.
GFRP Rebars in CRCP Study

The objectives of this study are to (GangaRao, Chen, and Vijay 2003; Choi and Chen 2005):

  1. Design and construct a CRCP that used GFRP bars for longitudinal reinforcement.
  2. Evaluate the performance of GFRP-reinforced CRCP in relation to conventional steel-reinforced CRCP during the winter and summer seasons in terms of:
    • Crack spacing and crack width.
    • Crack-cluster ratio and percentage of "Y" cracking.

    Both the GFRP-reinforced CRCP and the conventional steel-reinforced CRCP will be constructed on the same project and will share the same pavement thickness, base design, traffic loadings, and environmental conditions.

  3. Check the existing design and construction methodologies given by the 1993 AASHTO Pavement Design Guide for the performance of conventional steel-reinforced CRCP and suggest necessary modifications for CRCP with GFRP reinforcement.

Project Design and Layout

GFRP Dowel Study

The study of GFRP dowels consists of both laboratory and field investigations. Laboratory tests conducted in this study include static and repeated load applications on 279.4- and 304.8-mm (11- and 12-in.) thick slabs containing both 25.4- and 38.1-mm (1.0- and 1.5-in.) diameter steel and GFRP dowels placed at 152.4- and 304.8-mm (6- and 12-in.) spacings.

For the field investigation, GFRP dowel bars were included the Corridor H project on Rt. 219, near Elkins WV. Both 25.4- and 38.1-mm (1.0- and 1.5-in.) diameter GFRP dowels were installed in the field at 152.4-,203.2-,228.6-, and 304.8-mm (6-, 8-, 9-, and 12-in.) spacings. Load calibrated field tests were conducted on these pavements in 2002 and 2003 using a WVDOT truck. The 38.1-mm (1.5-in.) diameter GFRP dowel bars were also utilized for pavement rehabilitation near the junction of Rt. 857 and Rt. 119 on University Avenue in Morgantown, WV. Field data collected through automatic data acquisition system included strain and joint deflections (which were used for assessing joint load transfer efficiency), relative deflections across the joints, and overall pavement performance.

Moisture Diffusion Study

Diffusion of water through GFRPs is being studied gravimetrically by performing transient water uptake experiments on glass fiber-reinforced and nonreinforced nanocomposites in distilled water, alkaline solutions, and saline solutions. A diffusion mechanism will be proposed for both glass fiber-reinforced and nonreinforced nanocomposites based on the experimental results. Equations will also be proposed to predict the changes in diffusion behavior with temperature. SEM images will be taken to estimate the damage of the fiber surface after extended exposure to distilled water, tensile tests and fracture tests will be carried out on nonreinforced nanocomposites with various clay loadings, and DMA tests will be carried out to study changes in glass transition temperature (Tg), storage modulus and loss modulus with clay loading. Results of these laboratory experiments will be correlated with available theories.

GFRP Rebars in CRCP Study

Analytical modeling and finite element (FE) analysis of CRCP reinforced with GFRP bars have been carried out at WVU (Vijay and GangaRao 1999; Chen and Choi 2002, 2003, 2004). Based on the laboratory and FE analysis results, designs for a GFRP-reinforced CRCP section and a conventional steel-reinforced CRCP control section have been developed, but a suitable project site has not yet been identified.

Current Project Status, Results, and Findings

Some preliminary findings and recommendations are available from the study of GFRP dowels (Li 2004; Vijay, GangaRao, and Li 2004). These are presented below.

Findings from GFRP Dowel Study
  1. In this research, GFRP dowels were found to be acceptable alternatives to traditional steel dowels for transferring joint loads in JPCP pavements. Joints with GFRP dowels provided adequate deflection load transfer efficiency exceeding the minimum values recommended by AASHTO (75 percent) and ACPA (60 percent) in both laboratory tests and field tests.
  2. GFRP dowel-concrete interfaces in slab #1 and slab #4 after 5 million load cycles were found to be in excellent condition with no visible damage, microcracking, or separation between the GFRP dowel and surrounding concrete .
  3. The greater flexibility of the GFRP dowels results in a shorter distance to the point of inflection (no bending in the GFRP dowel) compared to steel dowels. The required length of 38.1-mm (1.5-in.) GFRP dowels is 65 percent of that for a steel dowel with same diameter. The required length of a 25.4-mm (1-in.) GFRP dowel is 69 percent of that for a steel dowel with same diameter. The bearing stress at the joint face is greater with the GFRP dowel compared to a steel dowel.
  4. Under static load testing, slabs with smaller diameter GFRP dowel and shorter spacings provided lower relative deflections than GFRP dowels with larger diameter and longer spacings. During fatigue loading up to 5 million cycles, the relative deflection of the slab with smaller GFRP diameter bars and shorter spacings and a 10.16-mm (0.4-in.) joint width increased from 0.3251 mm (0.0128 in.) to 2.0396 mm (0.0803 in.) over a range from 2 million to 5 million cycles, whereas the relative deflection of the slab with larger GFRP diameter bars and longer spacings and a 6.35-mm (0.25-in.) joint width decreased from 0.635 mm (0.025 in.) to 0.4318 mm (0.017 in.) over the range from 2 million to 5 million cycles.
  5. The LTE of slabs with 25.4-mm (1-in.) diameter GFRP bars spaced at 152.4-mm (6-in.) intervals was 72 percent after 5 million load cycles. The LTE of slabs on a strong base and with 38.1-mm (1.5-in.) diameter GFRP bars spaced at 304.8-mm (12-in.) intervals was greater than 80 percent, whereas the same set-up on slabs with poor base support conditions exhibited an LTE of 55 percent. Hence, it is very important to ensure adequate support conditions.
  6. The LTE affects the slab integrity and slab stresses, whereas relative deflection affects ride comfort and impact on slab at joints. It is important to consider both LTE and relative deflections when the performance of JPCP is evaluated. For example, at 5 million cycles, the slab with 38.1-mm (1.5-in.) diameter GFRP dowels spaced at 304.8-mm (12-in.) intervals exhibited 55 percent LTE, but its relative deflection (0.4318 mm [0.017 in.]) was less than that of the slab with 25.4-mm (1-in.) diameter GFRP dowels spaced at 152.4-mm (6-in.) intervals (LTE of 72 percent and a relative deflection of 2.0396 mm [0.0803 in.]).
Findings from GFRP CRCP Study

The concrete stresses induced in a GFRP-reinforced concrete slab due to concrete shrinkage or temperature variations have been analytically calculated in comparison to those induced in a steel reinforced concrete slab, and the validity of the analytical model has been verified numerically and experimentally. The analytical solution indicates that a low Young's modulus of GFRP rebars results in a stress reduction in concrete slabs. The thermal stress in concrete can be either tensile or compressive, depending on temperature variations, the coefficient of thermal expansions (CTEs) of the concrete, and the type of GFRP reinforcement.

From comparison with the FE calculation, the analytical solution of average axial concrete stresses is shown to be valid throughout the longitudinal (x-) direction, especially in the vicinity of the slab's middle section, where the maximum average axial stresses in concrete appeared. Meanwhile, the comparison also reveals the applicability of the FE method to the CRCP analysis. A 5-ft CRCP segment FE model was built to simulate the behavior of the CRCP and to study the effects of the CRCP-design considerations (such as the CTE of concrete, the subbase friction, and the bond-slip between concrete and reinforcement) on the stress development and crack width in the CRCP. It is shown that using concrete with a lower CTE reduces the concrete's tensile stress level when exposed to a temperature drop, and a weaker bond between the concrete and reinforcement also decreases the concrete tensile stress level in the CRCP and the tensile reinforcement stress level at its cracks. With a higher subbase friction, the CRCP is more likely to crack from its bottom area, since the concrete stress level at the bottom will be higher than that at the top. This suggests that one can control crack spacing and crack widths of GFRP-reinforced CRCP by increasing the amount of reinforcement, reducing the bond between the concrete and reinforcement, or reducing the bond between the slab and the base.

From the FE study, a feasible longitudinal reinforcement design for a 254-mm (10-in.) GFRP-reinforced CRCP has been developed by evaluating concrete stress levels in 1-m (3.5-ft), 1.5-m (5- ft), and 2.4-m (8-ft) long CRCP-slab segments. Using #6 longitudinal GFRP rebars at 152.4-mm (6-in.) spacings at the mid-depth of the slab is shown to be an economically feasible design for GFRP-CRCP on the flexible subbase (or lime-treated clay subbase). In addition, #5 GFRP-rebars spaced at 48 in will be adequate as transverse reinforcement.

Higher CTE, higher Young's modulus, lower tensile strength, and greater drying shrinkage of the concrete appear to reduce the crack spacing in the CRCP. However, even though the crack width is generally narrower as the crack spacing is smaller, it can open wider after the crack formation stabilizes. This occurs when the ambient temperature and air humidity levels drop enough to cause significant concrete volume changes such that the crack width of those smaller crack spacing sections will become larger. Increased subbase friction levels always cause shorter crack spacings followed by a narrower crack width for all cases. From the mechanistic analysis results, either of the GFRP-CRCPs with the limestone, sandstone, and siliceous river gravel concrete on asphalt-stabilized subbase seems to perform satisfactorily without raising the GFRP-reinforcement ratio. In addition, the subbase that can provide a bond-slip of about 178.9 kPa/mm (5,000 lbf/in2/in.) is able to guarantee a satisfactory performance of the GFRP-reinforced CRCP with granite concrete.

The designs of a 279.4-mm (11-in.) GFRP-reinforced CRCP and a conventional steel-reinforced CRCP sections were prepared in 2003. The final designs of the sections may vary depending upon the location of the experimental sections.

Points of Contact

Hota V. S. GangaRao, Ph.D., P.E., Professor
(304) 293-7608, ext. 2634

Hung-Liang (Roger) Chen, Ph.D., Professor
(304) 293-3031, ext. 2631

P. V. Vijay, Ph.D., Assistant Professor
(304) 293-7608, ext. 2660

Constructed Facilities Center
College of Engineering and Mineral Resources
West Virginia University
Morgantown, WV 26506


Chen, H. L., and J. H. Choi. 2002. "Effects of GFRP Reinforcing Rebars on Shrinkage and Thermal Stresses in Concrete," Proceedings of the 15th ASCE Engineering Mechanics Conference, Columbia University, New York, NY.

---. 2004. "Effects of GFRP Reinforcing Rebars on Shrinkage and Thermal Stresses in Continuously Reinforced Concrete Pavement," submitted to ACI Structural Journal, American Concrete Institute, Farmington Hills, MI.

Choi, J. H., and H. L. Chen. 2003. "Design Considerations of GFRP-Reinforced CRCP." Field Applications of FRP Reinforcement: Case Studies , SP-215. American Concrete Institute, Farmington Hills, MI.

---. 2005. Development and Deployment of Second Generation FRP Composite Applied to Concrete Pavements (Task 3: Continuously Reinforced Concrete Pavement). Draft Final Report for USDOT/FHWA Project # DTFH61-99-X-0078. Federal Highway Administration, Washington, DC.

GangaRao, H. V. S., H. L. Chen, and P. V. Vijay. 2003. Field Implementation and Performance Evaluation of CRCP with FRP Composite Bars. West Virginia University Work Plan. Federal Highway Administration, Washington, DC.

Gupta, R., H. Rana, and H. V. S. GangaRao. 2004. Moisture Diffusion through Neat and Glass-Fiber Reinforced Vinyl Ester Resin Containing Nano-Clay. Draft Final Report, CFC 04-102. Federal Highway Administration, Washington, DC.

Li, H. 2004. Evaluation of Jointed Plain Concrete Pavement (JPCP) with FRP Dowels. Master's Thesis, West Virginia University, Morgantown.

Vijay, P. V., and H. V. S. GangaRao. 1999. Development of Fiber Reinforced Plastics for Highway Application: Aging Behavior of Concrete Beams Reinforced with GFRP Bars. CFC-WVU Report No. 99-265 (WVDOH RP #T-699-FRP1). West Virginia Department of Transportation, Charleston.

Vijay, P. V., H. V. S. GangaRao, and H. Li. 2004. Evaluation of Jointed Plain Concrete Pavements (JPCP) with FRP Dowels. Draft Final Report, CFC 04-101. Federal Highway Administration, Washington, DC.

<< PreviousContentsNext >>
Updated: 04/07/2011

United States Department of Transportation - Federal Highway Administration