Shear Strain Profiles

The shear strain profiles for each of the specimens using epoxy types A, B, and C are shown in Figs. 15-17, Figs. 18-20 and Figs. 21-23, respectively. For Figs. 15-17 the shear profiles provided are for the side of the double shear test speci­mens that experienced the highest shear strain. In all of the specimens shown in Figs. 18-23, the uppermost curves, representing the shear strain profiles at fail­ure, have a downward concave shape in contrast with the convex shape exhib­ited by all of the other shear strain profile curves. This deviation in the shape of the curve is expected at the onset of debonding failure, due to the fact that the initiation of separation between the GFRP plate and the UHPC prism starting from the free end would cause an instantaneous drop in shear strain in SG-1 and SG-6 relative to SG-2 and SG-7, respectively, as a result of the sudden

FIG. 14—Shear interface slippage behaviour comparison for all epoxy adhesive types used. Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014 Downloaded/printed by Rochester Institute Of Technology pursuant to License Agreement. No further reproductions authorized.

release of fixed constraint. Therefore the strain data values collected by SG-1 and SG-6 at failure are not representative of the peak shear strain reached at those locations at the instant directly prior to debonding. In general, the shear strain profiles exhibited load-strain trends similar to a polynomial relationship

(y = Ax2 + Bx + C) closer to the free end while the trend is better described by a power relationship (y = Axn + C) closer to the fixed end. This is due to the changes in shear strain over the length of the bonded area, where the rate of strain increase is very stagnant at a distance from the free end but changes dra­matically as the distance decreases.

FIG. 18—Shear strain profiles for specimen B-1. Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014 Downloaded/printed by

Summary of Relative Performance

The relative performance of the three different types of epoxy adhesives used for bonding of the coarse silica sand aggregates from both double shear testing as well as tension pull-out testing is summarized in Table 5, where the higher

ranking is associated with the smallest number. While experimental error is to be expected for all research to be performed in a laboratory environment, the analysis conducted did show good consistency within each subgroup examined. Overall, it was evident that epoxy types A and B performed better than epoxy

FIG. 22—Shear strain profiles for specimen C-2. Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014 Downloaded/printed by

type C. Though epoxy type A performed better under shear loading, with epoxy type B showing higher performance under tension loading, it must be taken into account that the specimens with epoxy type A did not reach their full shear capacity at the interface due to premature shear failure at the bolted connec­tions. Comparison between the distribution of data values collected from the shear and tension tests also showed greater consistency and smaller variance during the shear testing. From these results, epoxy type A was chosen as the epoxy adhesive for bonding of the coarse silica sand aggregates at the GFRP – UHPC interface, where the specimens demonstrated the best and most reliable performance under both shear and tension loading.

Conclusions

From the results obtained in this experimental program, the following conclu­sions were made:

TABLE 5—Relative performance of epoxy adhesives in double shear and tension pull-out tests. Epoxy Type Double Shear Test Tension Pull-out Test A 1 2 B 2 1 C 3 3

– Specimens using epoxy type A for bonding the coarse silica sand aggre­gates at the bond interface performed best during the double shear bond testing, where testing was interrupted by premature connection failure without any interface debonding.

– Specimens using epoxy type B for bonding the coarse silica sand aggre­gates at the bond interface performed best during tension pull-out testing.

– Failure in the specimens with epoxy type A and B occurred due to a com­bination of aggregate fracture as well as separation in the epoxy layer rather than solely in the epoxy layer, as was the case for the specimens using epoxy type C.

– Shear interface slippage behaviour is better represented using strain val­ues obtained along the length of the specimen rather than differential dis­placement between the UHPC prism and the GFRP plate due to the higher number of data points used in the assessment of the load-slippage behav­iour as well as the elimination of potential shifts in the mounting appara­tus that could influence the data collected.

Acknowledgments

The writers would like to thank the following companies for their generous don­ations of materials used: Fyfe Co. LLC, Lafarge Canada, Hardwire LLC, and

Sika Canada Inc. In addition, we would like to acknowledge the University of

Calgary as well as the Natural Sciences and Engineering Research Council of

Canada (NSERC) for their financial support towards this research.