Lap splice assessment of GFRP rebars in reinforced concrete beams under flexure

Abstract

To increase the knowledge and experience gained from the existing literature regarding the bond behavior of glass fiber-reinforced polymer (GFRP) rebars, a total of sixteen full-scale GFRP-reinforced concrete beams were subjected to a four-point bending test. The GFRP reinforcement comprised a single M16 (No.5) GFRP sand-coated bar without confinement reinforcement in the constant moment region. The research parameters encompassed three different lap-spliced lengths (i.e., 40-, 60- and 80-db) and two distinct concrete clear covers (i.e., 19 mm and 38 mm) to assess their impact on bond strength and failure modes. Strain gauges along the GFRP rebar and small potentiometers positioned at the end of the splice within the constant moment zone allowed detailed measurements. Rebar slippage in lap-spliced specimens was observed to have a limited impact on bond stress transfer capacity and member flexural stiffness until reaching maximum slippage. Concrete cover significantly influenced the behavior of lap-spliced GFRP-RC beams, with 38 mm-cc specimens exhibiting an average 21% higher capacity compared to 19 mm-cc specimens. The ACI expression exhibited an average overestimation of 24% for 19 mm-cc specimens, whereas for 38 mm-cc specimens, it offered a more accurate estimation, with only a 10% over-prediction. However, the computed lap splice length to achieve the required tensile stress was reduced from the one required by ACI due to extra factors implemented in the development length equation. This aligns with prior research findings that identified unconservative values of bond stress when employing the ACI provisions for specific combinations of parameters, highlighting the necessity for refinements in predictive models.