PSI - Issue 10

I. Karatasios e l. / Structural Int grity Procedia 00 (2018) 00 – 000

I. Karatasios et al. / Procedia Structural Integrity 10 (2018) 211–218

216 6

0.0 0.5 1.0 1.5 2.0 2.5 3.0

y = 0.45x - 0.3373 R² = 0.9667

0 4-point bending strength of stone mortar system - STRM (MPa) 2

4

6

8

3-point bending strength of join mortars - RM (MPa)

Fig. 2. Correlation of flexural and adhesive strength of mortar mixtures as they were determined by 3-point and 4-point flexural bending of mortar specimens and joined stone fragments respectively.

15

Stone 3PB Mortars 3PB Stone-Mortar 4PB

12

12

9

6.1

5.5

5.4

6

3.04

2.74

2.5

Strength (MPa)

3

1

0.8

0

High strength Medium Strength Low Strength

Fig. 3. Comparison of internal cohesion of stone and mortars specimens in terms of 3-point flexural bending as well as, the level of adhesive strength of mortar mixtures 4-point flexural bending on joined stone fragments. The adhesion strength of join mortars at 28 days, as this is determined by four-point flexural bending strength, is close to one third of the three-point bending strength of mortar mixtures. Overall, the adhesion strength of different binders is related to their microstructure, since the later affects the total surface area available for adhesion to the stone surface. It is well known that the more porous the binder the lower the total surface area available. In terms of adhesion performance, porosity affects also the mechanical properties of the join mixtures and especially the bending strength that is indicative of the internal cohesion of the binder. Therefore, ternary mixtures of lime-cement-pozzolan exhibited the lower adhesion strength, very similar to that of NHL mixtures. In contrast, the cement mixtures exhibited the higher adhesion strength. However, in terms of microstructural compatibility, the use of different binders allowed the modification of the open porosity values at 28 days, in order to match better those of natural stones (Fig.5). By modifying the binder composition it is possible to control and adjust the adhesion of different mixtures, aiming to fulfill the strength requirements of natural stone fragments of varying strength and thus, produce compatible repair mixtures. In absolute values, the bond strength between mortar and stone was primary depended on the binder composition. It is worth noting NHL binder failed to provide good adhesion to the stone substrate during application, leading to increased number of failed specimens. In contrast, the use of lime-pozzolan mixtures exhibited increased workability and reduced water demand for standard consistency at 16 cm. However, lime-based mixtures had quite high porosity