Abstract
With the drive towards a greener construction industry, focus has been placed on minimizing the emission of CO2 from cement clinker production. This led to an increase in the utilization of blended cements in the construction industry. Currently, it is common practice to replace up to ~20 % of the cement clinker with supplementary cementitious materials to reduce the amount of CO2 emission. However, the aim is to further increase this substitution to further improve the environmental situation. As a result, more research and a better understanding of their effective performance are required. One such aspect where additional studies are required for such materials is in their interactions with different admixtures, especially at higher addition of the supplementary cementitious materials. Therefore, the purpose of this investigation is to study the rheological impact of different superplasticizers, commonly employed in conventional ordinary Portland cement systems on a selection of blended cements.
For investigation purposes, five different superplasticizers across the whole range were selected; a lignosulfonate, a naphthalene based polycondensate and three different polycarboxylates with varying side chain lengths and charge densities. They were initially dosed according to commercial usage (0.2 % by weight of cement) to reflect the practicality of these polymers in the blended cements as compared to that conventionally employed.
As supplementary cementitious materials, calcined marl and fly ash were chosen and utilized at replacement percentages of 20, 40 and 60 wt.% respectively. A low w/c ratio of 0.36 was selected to prevent bleeding and segregation of the cement pastes, particularly in the case of fly ash loaded systems. In situations where little flow was observed, much higher dosages of the superplasticizers were employed (up to 0.8 %by weight of cement).
Method wise, the rheological properties of the cement pastes were measured quantitatively by a Physica MCR 300 rheometer to ensure. Such a measurement method is generally favored over manual measurements such as the mini slump test, etc as it can give accurately a concise and pinpoint measurement independent of the user. The heat released up to 24 h was measured by the TAM air isothermal calorimeter, while the packing density and water demand of the cement pastes were analyzed employing centrifugal consolidation subjected to a compaction energy of 4,000 rpm for a period of 5 min.
The main findings in this work are as follow:
Generally, the blended cements showed similar trends in their interactions with superplasticizers as the ordinary Portland cement. The polycarboxylate based superplasticizers (NRG, SX and SPN) were more efficient in dispersing all cement systems than the naphthalene based (NAPh) and lignosulfonate (LS). Additionally, the lignosulfonate superplasticizer increased the flow resistance of the cement pastes (decrease fluidity of the paste) with increasing dosages.
In the fly ash systems, an increase in fly ash content in the cement resulted in an increase in fluidity of the neat cement paste. No bleeding was detected even when up to 60 wt.% replacement of cement clinker by fly ash was performed. The fly ash particles were relatively inert and acted as fillers. Minimal dissolution of fly ash particles occur in the paste and this is relatively negligible as compared to the bulk reaction of the cement clinkers during early age. Upon addition of superplasticizers, the trend in performance of the superplasticizers mimicked that of an ordinary Portland cement and the plasticizing effectiveness was as follow: NRG > SX > SRN > NAPH > LS. On the other hand, the slump flow retention (ability of maintaining flow resistance over time) was SRN > SX > NRG > LS ≈ NAPh. Fly ash particles interacted with the superplasticizers. However, the impact on the interaction with clinker phases was low and any adsorption was reversible. NRG proved to be a very good plasticizer, where no rheological data could be measured in a cement
containing 60 wt.% fly ash when 0.2 %bwob of this polymer was added due to segregation of the paste.
Comparing a fly ash cement (pre-blend in the cement mill while grinding) and a manually blended cement each containing ~20 wt.% fly ash, the flow resistance of the neat cement pastes were totally different. The manually blended cements displayed a better much better flow than the former, which possessed rheological properties similar to that of an ordinary Portland cement. This difference can be attributed to the availability of effective surface area from the fly ash which was
exposed to undergo interaction with superplasticizers or water molecules in presence of clinker phases.
In the calcined marl systems, the calcined marl was shown to be a strong adsorber of water. The affinity or the water retaining capacity of the calcined marl per unit mass decreased as the solid replacement percentage increase in presence of a constant amount of water (w/c = 0.36). In general, an increased replacement of cement clinker with calcined marl resulted in a decrease in flow of the neat cement pastes and the paste stopped flowing when 60 wt.% calcined marl was utilized. Upon addition of superplasticizers, calcined marl competed with the cement clinkers for superplasticizers. The extent of competition was masked by the uptake of water and also varied according to the characteristic of the superplasticizers. SX, possessing an intermediate side chain length and charge density proved to be a much better dispersing agent here. The plasticizing effectiveness of the
superplasticizers was SX > NRG > SRN > NAPh > LS, similar to an ordinary Portland cement or in the case of cement containing fly ash with the exception of SX polymer. The effectiveness of the dispersing agents on slump retention was SRN ≈ SX > NRG > LS > NAPh.
In the investigation of heat released during hydration of cements, pure fly ash showed negligible amount of heat released up to 24h. In cements containing fly ash, the hydration of the cements was retarded with increasing fly ash content, confirming the dilution effect of fly ash on the system. The effect of superplasticizers (degree of retardation and change in total heat released) was similar to that observed in an ordinary Portland cement. Pre-blended fly ash cement displayed similar hydration profile as the ordinary Portland cement since it has been finer ground (Blaine 454 vs. 382 m2/kg) to compensate for the sluggish reaction of FA
Pure calcined marl, on the other hand, displayed thrice the amount of heat released as that for a pure fly ash sample. This, in comparison to the heat evolved in cement hydration was however, negligible. When blended with cement, the calcined marl contributed to the hydration of the cement paste. In the case of calcined marl addition, no retardation was observed, up to a loading of 40 wt.% calcined marl. However, a replacement by 60 wt.% calcined marl displayed a change in hydration profile, indicating that hydration of calcined marl dominated at higher replacement percentages. Superplasticizers only affect and retard the hydration of the cement pastes at threshold superplasticizer dosages of 0.4 %by weight of cement.