Hao et al. (2009) showed that ribbed-surface GFRP bars with rib spacing equal to the bar diameter exceeded the bond strength of the sand-coated bar.  However, Morales Arias et al. (2012) showed that sand-coated GFRP bars provided higher bond strength. Is there any confirmation?  

Logically, the ribbed bars are supposed to provide better bond because they rely on interlocking and bearing in addition to friction and adhesion.

I would not say it is a general trend to have the bond strength of ribbed GFRP bars higher than that of the sand-coated GFRP bars. The judgment should be based on testing GFRP bars of the same diameter and mechanical properties (tensile modulus and strength) with different surface configurations while maintaining the other parameters such as embedment length and concrete strength unchanged. Considering the wide range of the mechanical properties of the current available GFRP bars and the differences in the manufacturing and coating techniques, the conclusion will be limited to the tested products.

I have tested many bars of different surface configurations and I can not confirm a trend. In some tests I had the pullout failure due to peeling of the sand-coating or the ribs. The mechanical bond may have a significant contribution in ribbed bars, but on the other hand the uniform distribution in case of sand-coated bars may enhance the performance.

The following paper provides an extensive testing on bond of Carbon, Glass, and Aramid FRP bars. Please check it. (Okelo, R. and Yuan, R.L. 2005 “Bond Strength of Fiber Reinforced Polymer Rebars in Normal Strength Concrete.” ASCE Journal of Composites for Construction, Vol. 9, No. 3, June 1, pp. 203-213). They concluded that: (1) surface deformations similar to the ones on steel are very effective means of bond enhancement; (2) other surface deformations or indentations obtained by stressing an external helicoidal strand and deep dents (groves) are acceptable means of bond improvement; (3) sand coating and surface texture are other possibilities of obtaining better bond performance compared to a smooth surface; (4) the mode of failure of bond is mainly dependent on the concrete compressive strength, the shape (and composition) of the bar’s surface, the cover thickness, and embedment length.

You may also look to the free-end slip of the different FRP bars which may help you in comparing the performance.

Wish that helps. Good luck!

Thanks Ehab for your sincere reply. Currently I am editing a paper comparing the bond characteristics of both sand-coated and ribbed-surface GFRP. Based on about 200 pullout tests covering different parameters as you mentioned while each parameter was tested on 5 identical specimens. The test results showed that the bond strength of sand-coated bars showed enhanced results compared to the ribbed-surface bas of the same parameters for some cases and showed lower bond strength for others. I cannot obtain a remarkable trend. I was bias to ribbed-surface bars while the test results were bias to the sand-coated bars. In addition, I did not find any distinct trend in literature. So, I raised this query. 

By the way, the tests were conducted at Ryerson University, Toronto, ON, Canada.

The CSA-S806-12 provides a factor k5 (bar surface profile factor) for development length calculation equals 1.05 for ribbed-surface bars and 1 for sand-coated bars. This confirms implicitly that the sand-coated bars provide higher bond strength which entail lower development length.

Since we are followers on the research gate, we can obtain and exchange other findings.  

Of course we can share the findings. Good luck Hamdy!