Field and Laboratory Performance of an FRP-Reinforced Concrete

Stone, D., D. Koenigsfeld, J. Myers, and A. Nanni. "Durability of GFRP Rods, Laminates and Sandwich Panels Subjected to Various Environmental Conditioning." Proceedings - Second International Conference on Durability of Fiber Reinforced Polymer (FRP) Composites for Construction, Montreal, Quebec, Canada, May 29-31, 2002.

DURABILITY OF GFRP RODS, LAMINATES, AND SANDWICH PANELS SUBJECTED TO VARIOUS ENVIRONMENTAL CONDITIONING

Danielle K. Stone,1 Daniel Koenigsfeld,2 John Myers,3 and Antonio Nanni4 1

Ph.D. Candidate/GRA, Dept. of Civil Engineering, The University of Missouri – Rolla Undergraduate Research Assistant, Dept. of Civil Engineering, The University of Missouri – Rolla 3 Assistant Professor, Dept. of Civil Engineering, University of Missouri – Rolla 4 V&M Jones Professor, Dept. of Civil Engineering, University of Missouri – Rolla

2

ABSTRACT This paper outlines the performance of glass fiber-reinforced polymeric (GFRP) materials subjected to various environmental conditioning. The study is performed in conjunction with the construction of one bridge using FRP-reinforced concrete panels and three bridges using GFRP sandwich panels. GFRP rods utilizing a urethane-modified vinyl ester resin and GFRP laminates and sandwich panels utilizing an isophthalic polyester resin were subjected to environmental conditions designed to simulate their in-situ environments. Two different conditioning regimens were utilized for each type of GFRP material considered. First, to replicate the exposure of the GFRP rods to an alkaline environment, a solution containing calcium, sodium and potassium hydroxides was formulated with a pH of 12.6; an elevated temperature of 60°C (140°F) was utilized to accelerate absorption. The rods were also subjected to a series of freeze-thaw, high temperature, and high relative humidity cycles in an environmental chamber to duplicate seasonal effects in the mid-west United States. The tensile and interlaminar shear properties of GFRP rods subjected to these conditions are compared to the properties of control specimens. GFRP laminates and GFRP sandwich panels made from the same materials were also subjected to two separate conditioning regimens. One series consisted of the aforementioned conditioning in an environmental chamber while the other consisted of immersion in a 60°C (140°F) saline solution. The tensile properties of the single-ply GFRP laminates are compared to that of the control specimens. For the FRP sandwich panels, small-scale beams were tested in four-point bending to compare the ultimate load and failure mode. Additional studies are on-going and will added to body of knowledge when completed.


INTRODUCTION Although glass fiber-reinforced polymer (GFRP) materials have been steadily gaining acceptance in the construction industry, there remains a need to verify their longterm durability. In conjunction with a project to construct four bridges utilizing GFRP materials, a durability study was undertaken to examine the performance of these materials following various conditioning regimes. The bridges utilize GFRP materials in the form of rods for reinforcement in concrete and honeycomb sandwich panels for use as bridge or bridge deck panels. For each of these GFRP material applications, two conditioning regimens were chosen to simulate in-situ exposures.

GFRP REINFORCING RODS Conditioning Regimens First, to replicate the exposure of the GFRP rods to an alkaline environment such as would be encountered when used as reinforcement for concrete, a solution containing calcium, sodium and potassium hydroxides was formulated. The following percentages by weight were dissolved in distilled water to produce a solution with a pH of 12.6,

0.16%Ca ( OH ) 2 + 1%Na ( OH ) + 1.4%K ( OH )

[1]

To accelerate absorption, the rods were immersed in the alkaline solution at an elevated temperature of 60°C (140°F). The following correlation between the exposure to an alkaline solution at elevated temperature developed by Litherland et al. was reported by Vijay et al. (1999) N = 0.098 e0.0558T C

[2]

where N is the predicted age in natural days, C is the number of days of exposure to the alkaline solution at an elevated temperature, T, in Fahrenheit. For this research program, specimens were conditioned for either 21 or 42 days, which corresponds to natural aging of 14 or 28 years, respectively, The rods were also subjected to a second conditioning scheme, consisting of a series of freeze-thaw, high temperature, and high relative humidity cycles. This environmental conditioning was performed in an environmental chamber to duplicate seasonal effects in the mid-west United States. Table 1 details the temperature and humidity information for each of the cycle types along with the total number of cycles to which the specimens were exposed.


Table 1. Environmental Chamber Cycles. Cycles Freeze-Thaw High Temperature High Relative Humidity (60% – 100%) High Relative Humidity (60% – 100%)

Temperature Range (ºC) -17.8 to 4.4 26.7 to 48.9

Total Number 200 480

15.6

160

26.7

80

Several rod types and sizes were utilized for this portion of the research program. Two FRP bar manufacturers supplied rods; one provided only 6.4-mm (0.25-in) diameter rods, identified herein as HB2, and the other provided 6-mm (0.25-in), 9.5-mm (0.375-in), and 12.7-mm (0.5-in) diameter rods, identified herein as M2, M3, and M4, respectively. Interlaminar Shear Properties The interlaminar shear properties of GFRP rods subjected to these conditions were determined and are compared to the properties of control specimens. ASTM standard test method D4475 was utilized to determine the apparent horizontal shear strength of the GFRP rods. Specimens were evaluated prior to any conditioning, after 4 cycles of exposure in the environmental chamber, and after 21 and 42 days of exposure to the alkaline solution. Prior to conditioning the specimens were cut to the recommended length, four times the diameter of the given bar, and the ends of the specimens were sealed with silicone to prevent unwanted absorption of moisture. Six specimens of each type were evaluated under three-point bending over a span length of three times the diameter of the given bar. ASTM D4475 dictates that the loading rate of the specimens should be 1.3 mm (0.05 in) per minute. The failure load, P, and the diameter, d, of the bar are used to calculate the apparent horizontal shear strength, S, of the bar using the following equation: S = 0.849

P d2

[3]

It should be noted that the shear strength values obtained via this test are to be used only for comparative purposes, whereby the change in properties of rods subjected to different exposure regimens can be identified. A typical load versus mid-span deflection curve is illustrated in Figure 1; the curve is for one of the M4 control specimens.


14 12

Load (kN)

10 8 6 4 2 0 0.00

2.00

4.00

6.00

8.00

10.00

Displacement (mm) Figure 1. Typical Load versus Mid-span Deflection Curve for Shear Strength Tests.

Failure for nearly all of the specimens was indicated by a combination of vertical (parallel to the loading head of the machine propagating along the longitudinal axis of the bar) and horizontal (perpendicular to the loading head of the machine propagating along the longitudinal axis of the bar) cracking. Figure 1 contains a picture of the failed specimen as well, illustrating the failure mode described. Table 2 details the average failure load and shear strength values for each type of bar tested along with the standard deviation of the shear strength and the percent residual shear strength of each of the conditioned rods compared to the appropriate control specimen. Figure 2 illustrates the residual shear strength percentages. Specimen identification consists of the bar type, either HB2, M2, M3, or M4, followed by the conditioning regimen; C for control, 4EC for four environmental cycles, or 21 or 42 for 21 and 42 days of alkaline conditioning, respectively. It should be noted that the shear strength testing of the M321, M342, M421, and M442 specimens is still in progress and will be reported in the final version of this paper.


Table 2. Interlaminar Shear Strength Test Results. Sample

Failure Load (kN)

Shear Strength (MPa)

HB2C HB24EC HB221 HB242 M2C M24EC M221 M242 M3C M34EC M4C M44EC

2.28 2.46 2.23 2.21 3.44 3.22 3.25 3.16 5.93 6.31 11.52 11.28

48.04 51.72 47.02 46.53 72.49 67.87 68.33 66.63 55.49 59.05 60.66 59.40

Standard Deviation Percent in Shear Strength Residual Shear (MPa) Strength 2.58 100.0 2.14 107.7 3.27 97.9 2.97 96.9 3.66 100.0 3.79 93.6 1.82 94.3 3.79 91.9 9.68 100.0 7.25 106.4 3.29 100.0 5.26 97.9

110 Residual Shear Strength

100 90 80 70 60 50 40 30 20 10 0 Control

Environmental Cycles HB2

M2

21 Day Alkaline M3

M4

Figure 2. Residual Shear Strength.

42 Day Alkaline


Results to this point indicate that degradation of the rods subjected to environmental cycles and the alkaline solution is minimal. The maximum degradation experienced, approximately 8%, was by the M2 rods conditioned for 42 days in the alkaline solution. Little, if any, degradation was experienced by the rods exposed to the environmental cycles. Tensile Properties Tensile specimens were only subjected to one of the conditioning regimens. The tensile capacity of all four types of rods was evaluated after 42 days of exposure to the alkaline solution and compared to the properties of the control specimens. To avoid failure of the specimen at the grips due to the relatively low transverse strength of the FRP materials, the ends of the specimens were encased in steel pipe using an expansive grout. A gripping length of 38.1 cm (15 in) was used based on work conducted by Micelli et al. (2001). Furthermore, an overall specimen length of 40db plus two times the gripping length, where db is the diameter of the bar, was used based on provisional specifications for FRP rods testing that are under review by ACI committee 440K. A typical stress versus strain plot from the tensile testing is illustrated in Figure 3. As expected, in all cases linearelastic behavior was exhibited until failure. Three rods of each type were tested, with the average results detailed in Table 3 and illustrated in Figure 4. 400 350

Stress (MPa)

300 250 200 150 100 50 0 0.000

0.002

0.004

0.006

0.008

0.010

Strain (mm/mm) Figure 3. Stress versus Strain plot for the tensile tests.

0.012


Table 3. Tensile Test Results. Load (kN) Control 26.52 HB2 33.41 M2 62.39 M3 99.57 M4 42 Day Alkaline 15.83 HB2 29.97 M2 49.20 M3 90.97 M4

Stress (MPa)

Modulus (GPa)

Percent Residual Stress

Percent Residual Modulus

837.71 1054.90 875.63 786.00

45.86 36.40 37.23 36.12

100.0 100.0 100.0 100.0

100.0 100.0 100.0 100.0

499.87 869.43 689.48 718.43

34.07 33.64 32.49 33.43

59.7 82.4 78.8 91.4

74.3 92.4 87.3 92.6

Figure 4. Residual Strength and Modulus from Tensile Tests. 110 100 Percent Residual

90 80 70 60 50 40 30 20 10 0 Stress (Control)

Stress (42 Day Alkaline) HB2

M2

Modulus (Control) M3

Modulus (42 Day Alkaline)

M4

A comparison was made between the values guaranteed as minimum design values by the manufacturers of the rods and the values obtained in the laboratory for the control specimens. The guaranteed tensile strength and tensile modulus of the HB2 rods are 825


MPa (120 ksi) and 40.8 GPa (5.92x106 psi), respectively. The control HB2 specimens are satisfactory in terms of both strength and modulus. For the M2, M3, M4 rods, the guaranteed tensile properties are 725 MPa (105 ksi) for tensile strength and 42 GPa (5.92x106 psi) for tensile modulus. The M2, M3 and M4 bars all exhibited satisfactory strength values, however the modulus values were slightly below the guaranteed values. Residual values of tensile strength and modulus can be compared to the recommendations put forth by ACI (2001), which outline an environmental reduction factor, CE, to be used to account for the long-term strength of the bars. For GFRP bars exposed to earth and weather, this factor is equal to 0.7. All of the rods tested in this study meet this recommendation, except for the HB2 rods, which exhibited a strength of approximately 60 percent of the control strength following the alkaline conditioning.

GFRP SANDWICH PANELS Conditioning Regimens GFRP laminates and GFRP sandwich panels utilizing an isophthalic polyester resin were also subjected to two separate conditioning regimens. One series consisted of the aforementioned conditioning in an environmental chamber. The second exposure regime consisted of immersion in a 60°C (140°F) saline solution for 42 days; the saline solution consisted of sodium chloride, NaCl, dissolved in tap water to create a 15% solution by weight. Tensile Properties A series of GFRP laminates were prepared using the same fibers and resin used to comprise the GFRP sandwich panels. Once cured the flat laminate sheet was cut into 38.1 mm (1.5 in) by 381 mm (15 in) specimens; the laminate sheet was approximately 1.5 mm (0.06 in) thick. GFRP laminate specimens were tested prior to and after the aforementioned conditioning. Prior to exposure to the conditioning regimens and in order to prevent moisture absorption, the edges of the laminate specimens were sealed with silicone. Six specimens were tested for each of the three conditions, with the average results summarized in Table 4. The results are also illustrated in Figure 5. Table 4. Tensile Properties of GFRP Laminates. Control Percent Residual Environmental Cycles Percent Residual 42 days Saline Conditioning Percent Residual

Load (kN) 16.63

17.33 15.85

Stress (MPa) 296.34 100.0 298.51 100.7 272.99 92.1

E (GPa) 15.56 100.0 16.28 104.6 15.57 100.1


110 100

Percent Residual

90 80 70 60 50 40 30 20 10 0 Tensile Stress Control

Tensile Modulus

Environmental Cycles

42 Day Saline

Figure 5. Residual Strength and Modulus of GFRP Laminates.

As illustrated by the results in Table 4 and Figure 5, approximately 8% degradation occurred in the strength of laminate, whereas no degradation in the modulus of the laminates was exhibited. Flexural Behavior For the GFRP sandwich panels, small-scale beams, measuring approximately 0.13 m (5 in) deep by 0.20 m (8 in) wide by 0.91 m (36 in) long, were tested in four-point bending. One specimen was exposed to the aforementioned conditioning in an environmental chamber, while the exposure regime for the second specimen consisted of immersion in a 60°C (140°F) saline solution for 42 days. A comparison between the ultimate load and failure mode of the two conditioned specimens and the control specimen is presented. For the specimen subjected to the environmental cycles, nothing was done to the specimen prior to conditioning. However, for the specimen immersed in the saline solution, small 7.94-mm (0.31-in) holes were drilled approximately at the neutral axis through the entire width of the specimen; this was done in order to create the worst-case scenario of flooding of the entire core.

After conditioning of all of the specimens was completed, testing of all three specimens was conducted at one time. The specimens were tested under four-point bending with two equal loads applied 317.5 mm (12.5 in) from the supports, leaving a 203.2-mm (8in) constant moment region in the center of the beam. The span-to-depth ratio for the


specimens was approximately 2.6. A picture of the test setup is illustrated in Figure 6. Loading of the specimens was conducted using a constant displacement rate until failure of the specimen occurred.

Figure 6. Test setup for the small-scale beam GFRP sandwich panel specimens.

Manual lay-up is utilized for the manufacturing of the specimens; each of the specimens had slightly different dimensions. The moment of inertia values for the smallscale beams were approximated as 1776.48 cm4 (42.68 in4), 1366.9 cm4 (32.84 in4), and 1620.81 cm4 (38.91 in4) for the control, environmental cycle, and saline exposed beams, respectively. Figure 7 illustrates the load divided by the moment of inertia of the specimen plotted versus mid-span deflection in order to normalize the results for all three specimens. Failure of the control specimen occurred at a load of approximately 76.21 kN (17.13 kips). Failure loads for the environment cycle and saline exposed specimens were 66.25 kN (14.89 kips) and 62.33 kN (14.01 kips), respectively. Failure of all three panels occurred by delamination of the core from the top and bottom faces of the panel on only one end of the panel. In the case of the control and environmental cycle exposed specimens, “buckling” of the core material occurred whereby one side of the core material bulged out approximately 12.7 mm (0.5 in) at approximately one-quarter of the span. Figure 8 illustrates the buckling-type behavior exhibited by the control and environmental cycle exposed specimens. It should be noted that the variability in the manufacturing process could have influenced both the failure load and the failure mode of the specimens tested. The same trends exhibited in the testing of the GFRP laminates are reiterated by the results of the small-scale beam testing. The ultimate normalized load exhibited by the environmental cycle exposed specimen is approximately 13% higher than that exhibited by


the control specimen and for the saline exposed specimen the normalized load is approximately 10% lower than the control. Furthermore, the stiffness exhibited by both conditioned specimens is higher than that exhibited by the control specimen.

4

Load/Moment of Inertia (N/cm )

60.00 50.00 40.00 30.00 20.00

Control Environmental Cycles 42 Day Saline

10.00 0.00 0.00

2.00

4.00

6.00

8.00

10.00

Deflection (mm)

Figure 7. Normalized Load versus Mid-span Deflection for the Small-Scale GFRP Sandwich Panel Beams.

Figure 8. Buckling-type Behavior Exhibited by the Core Material.

Future testing will be conducted in order to eliminate the variability in the manufacturing process from affecting the test results. Testing of the conditioned specimens will be conducted prior to conditioning up to approximately 40 to 50% of ultimate load.


Then following conditioning the specimens will be tested to failure. In this way the number of specimens necessary is reduced and the variability of the manufacturing process is eliminated from the variables. The comparison drawn between the two tests on the same specimen lends confidence to the results and assures that the geometric and material properties are identical in each case and that the only changes that are exhibited are due to the conditioning to which the specimen was exposed.

CONCLUSIONS

Based on testing conducted on GFRP reinforcing rods for concrete and GFRP sandwich panels for use as bridge and bridge deck panels the following conclusions can be drawn: • Interlaminar shear strength of GFRP rods is effected only minimally, if at all, by exposure to accelerated aging in an environmental chamber or an alkaline solution at an elevated temperature. • Both tensile strength and tensile modulus of GFRP rods are affected by exposure to an alkaline solution at an elevated temperature. Degradation was generally within the recommended reduction factors offered by ACI (2001). • When exposed to accelerated aging or a saline solution at an elevated temperature the tensile strength of the GFRP laminates was not adversely affected. In fact in most cases, the modulus and strength of the laminates increased due to the conditioning. • The flexural behavior of GFRP sandwich panel beams exposed to accelerated aging in an environmental chamber or exposed to a saline solution at an elevated temperature is different from the behavior of the control specimen. The modulus of both conditioned specimens was higher than that of the control specimen and the strength of the environmentally conditioned specimen was higher than that of the control, while the strength of the saline conditioned specimen was lower. • It is possible that the variability in the manufacturing process could have affected the results of the flexural testing. Subsequent testing will be conducted by testing the specimens to be conditioned prior to any conditioning and then re-testing them after conditioning.

REFERENCES

ACI Committee 440. Guide for the Design and Construction of Concrete Reinforced with FRP Bars. 440.1R-01, American Concrete Institute, Farmington Hills, MI, 2001, pp. 41. Micelli, F. and A. Nanni. Mechanical Properties and Durability of FRP Rods. CIES report 00-22, March 2001.


Vijay, P.V., and Ganga Rao H.V.S., (1999), “Accelerated and Natural Weathering of Glass Fiber Reinforced Plastic Bars”, Proc. FRPRCS-4, November 1-4th, 1999, Baltimore, USA, pp. 605-614.