Abstract
The primary objective of this report was to find out why calcined marl replacing cement in mortars leads to such a high strength and how it possibly can lead to longer term strength increase after the calcium hydroxide generally thought to be required for pozzolanic reactions has been depleted.
The strength of all mortars with calcined marl replacing cement from 0, 20, 35 , 50 and 65 vol% increases from one year to years curing. The increase is the highest for the mortar with the highest cement replacement in spite of not containing any calcium hydroxide to sustain a pozzolanic reaction since it was 90 days old. After two years the compressive strength is about equal for mortars where 35, 50 and 65 vol% cement has been replaced by calcined marl, considering the standard deviations.
The strength gain of the mortar with the 65 vol% cement replacement can partly be explained by the higher relative pore refinement of this mortar (-3.5 % since 90 days) compared to mortars with 35 and 50% replacement that only had a relative capillary pore refinement of about 1 % from 90 days to 2 years. However, the reason for the pore refinement is not found.
Some hypotheses are put forward that could explain the strength increase or pore refinement in spite of lack of calcium hydroxide to sustain classical pozzolanic reaction;
1) There are two types of CSH in the system. One with high C/S from hydration of cement and one with low C/S from the direct pozzolanic reaction between calcined marl and calcium hydroxide. The one with higher C/S has higher solubility of Ca2+and is considered weaker mechanically than the one with lower C/S. Slowly the two different CSHs will equilibrate, and one ends up with an overall stronger CSH with intermediate C/S.
2) The solubility of Ca2+ from CSH with high C/S and pH is so high that it might lead to a direct further pozzolanic reaction with unreacted calcined marl.
3) When calcined marl reacts, both silicate and aluminate is released, forming CSH and CAH, as well as possibly intermediate products like C2ASH8. Crystalline CAH generally demands a higher atomic Ca/Al than amorphous CSH demand Ca/Si, and the CSH is more flexible in its Ca/Si. The craving for Ca by the aluminate could lead to the de-calcification of inner CSH as observed, as well as lowering the Ca/Si of the overall CSH that will adapt to that by increased polymerization of silicates. Inclusion of aluminate in the CSH structure (i.e. bridging of dimers) will also contribute to a higher degree of polymerization of CSH. In total this may lead to a higher strength binder.
It is recommended to make paste sample for further studies by 27Al and 29Si MAS NMR (nuclear magnetic resonance), XRD (X-ray diffraction), or other analytic techniques that otherwise might be disturbed by the aggregate of mortars. The objective would be to find out whether or not extensive polymerization of the CSH gel with inclusion of aluminum in the structure can explain the strength increase for high replacements of cement by calcined marl. It would also reveal any other compounds containing aluminum or silicon. Other analytic techniques might be disturbed by the aggregate of mortars.