Microstructural Aspects of the Fracture of Hardened Cement Paste

For microstructural studies, samples of hardened cement paste were fractured, impregnated with epoxy using a procedure that does not produce drying cracks, and examined using a scanning electron microscope. Cracks generally appear to pass around unhydrated cement grains and along calcium hydroxide cleavage planes. Cracks are often branched or forked at their tips, and a few microcracks are sometimes observed just ahead of the tips. Perhaps more importantly, the crack path is often offset, with the interjacent material between these offset crack segments forming a bridge across the crack. Even though the crack plane is most likely continuous in three dimensions, these bridging materials may provide local regions of ligamentary tractions across the crack, thereby providing a plausible explanation for the rising fracture resistance ( R -curve behavior) of hardened cement paste.     

Early-age acoustic emission measurements in hydrating cement paste: Evidence for cavitation during solidification due to self-desiccation

In this study, the acoustic emission activity of cement pastes was investigated during the first day of hydration. Deaired, fresh cement pastes were cast in sealed sample holders designed to minimize friction and restraint. The majority of acoustic emission events occurred in lower water to cement ratio pastes, while cement pastes with higher water to cement ratios showed significantly less acoustic activity. These acoustic events occurred around the time of setting. A layer of water on the surface of the cement pastes substantially reduced acoustic emission activity at the time of setting. According to these experimental results, the acoustic emission measured around setting time was attributed to cavitation events occurring in the pores of the cement paste due to self-desiccation. This paper shows how acoustic emission might be used to indicate the time when the fluid–solid transition occurs in a cement paste, often referred to as time-zero. Knowledge of time-zero is fundamental for determining when mechanical properties develop and in calculations of residual stresses.     

Influence of chloride ions and pH level on the durability of high performance cement pastes

The resistance to chemical attack of low water to binder ratio pastes containing silica fume was studied by soaking small paste disks in three different pH controlled solutions, with or without sodium chloride, for periods of up to three months. The pastes were made using water to binder ratios of 0,25 and 0,38. The three solutions in which the paste disks were soaked were the following: 3% NaCl (by weight) at a pH level of 8,5,0% NaCl at 8,5, and 0% NaCl at 4,5. After three months of exposure, the results show that the pH level of the aggressive solution is the most important factor controlling the durability of cement pastes subjected to chemical attack. The total porosity and the depth of decalcification was found to increase with the decrease of the pH level. It was also found that the3water to binder ratio does not significantly affect the deterioration processes, but only influences the kinetics of these processes. The decrease of the water to binder ratio reduces significantly the rate of deterioration. Chloroaluminate crystals were observed only in the cement pastes having a water to binder ratio of 0,38.     

In situ double torsion fracture studies of cement mortar and cement paste inside a scanning electron microscope

The characteristics of microcracking in cement and high strength mortar have been investigated by fracture experiments of small (33×11×1.1mm) double torsion (DT) type specimens, inside the specimen chamber of a conventional scanning electron microscope. The control of incremental crack growth, at less than a micron, is finer than has been reported before. The cracking path is tortuous and quite branched both around and through fine aggregate particles, thus absorbing significantly more energy than a corresponding straight crack. In the vicinity of the main crack tip a zone of diffuse microcracking was observed which may be regarded as a “process zone”. These two observations suggest that fracture mechanics models utilising discrete straight single cracks should only be used with circumspection. In cement paste there also appeared to be a transition from predominantly intergranular to transgranular fracture with increased age of hydration.     

Decalcification shrinkage of cement paste

Decalcification of cement paste in concrete is associated with several modes of chemical degradation including leaching, carbonation and sulfate attack. The primary aim of the current study was to investigate the effects of decalcification under saturated conditions on the dimensional stability of cement paste. Thin (0.8 mm) specimens of tricalcium silicate (C₃S) paste, white portland cement (WPC) paste, and WPC paste blended with 30% silica fume (WPC/30% SF) were decalcified by leaching in concentrated solutions of ammonium nitrate, a method that efficiently removes calcium from the solid while largely preserving silicate and other ions. All pastes were found to shrink significantly and irreversibly as a result of decalcification, particularly when the Ca/Si ratio of the C-S-H gel was reduced below ∼ 1.2. Since this composition coincides with the onset of structural changes in C-S-H such as an increase in silicate polymerization and a local densification into sheet-like morphologies, it is proposed that the observed shrinkage, here called decalcification shrinkage, is due initially to these structural changes in C-S-H at Ca/Si ∼ 1.2 and eventually to the decomposition of C-S-H into silica gel. In agreement with this reasoning, the blended cement paste exhibited greater decalcification shrinkage than the pure cement pastes due to its lower initial Ca/Si ratio for C-S-H gel. The similarities in the mechanisms of decalcification shrinkage and carbonation shrinkage are also discussed.     

Effect of curing conditions on the total water content of hardened portland cement paste

The effect of various curing conditions on the amount of water held by pastes with low water/cement ratios was investigated. It was concluded from the results that factors significantly influencing the total water content of a cement paste made with ordinary Portland cement and cured at room temperature are: water/cement ratio, curing medium, curing period and curing history.     

Properties and structure of oil shale ash pastes. II: Mechanical properties and structure

Compressive strength, drying-shrinkage and expansion in water of oil shale ash pastes were studied and compared to the corresponding properties of portland cement paste. X-ray diffraction and some scanning electron microscopy runs were also included in the study. It was concluded that the structure and properties of the ash pastes can be described and explained by the same models which have been suggested for portland cement paste. The only exception was the total porosity of the ash paste which remained unchanged with time. A suitable modification in the structural model of the portland cement paste was suggested to allow for this specific behaviour.     

Pore structure of air-entrained hardened cement paste

The effect of air entrainment on the pore structure of hardened cement paste was investigated. Air-entrained and air-free samples of various water-cement ratios and ages were prepared by a well-defined procedure. The first and second-intrusion pore-size distribution curves of the samples were determined by mercury intrusion porosimetry. It was observed that sample preparation technique affects the pore-size distributions of hardened cement pastes. The second-intrusion curves indicated a decrease in the total volume and a reduction in the size of pores that are uniform in cross section with decreasing age and water: cement ratio. The second-intrusion curves of air-entrained and air-free pastes of equal water: cement ratio and age matched with each other. It was concluded that air entrainment introduces only large air voids observable by a naked eye and does not alter the characteristic fine pore structure of hardened cement paste appreciably.     

The pozzolanic reactivity of monodispersed nanosilica hydrosols and their influence on the hydration characteristics of Portland cement

This study reveals that the nanosilica hydrosols with higher specific surface areas had faster pozzolanic reactivity, especially at early ages; moreover, the results are indicative of the accelerating influence of nanosilicas and silica fume on the hydration of cement. Faster initial and final setting times observed for cement pastes containing nanosilicas are consequence of these mechanisms. However, less hydration degree of cement compared to the plain paste was observed at age of 7 days and after. This can be attributed to the entrapment of some of mix water in the aggregates of nanosilicas formed in cement paste environment, making less water available for the progress of cement hydration. The same mechanism is believed to be responsible for the reduction of flowability of cement pastes.     

Effects of densified silica fume on microstructure and compressive strength of blended cement pastes

Some experimental investigations on the microstructure and compressive strength development of silica fume blended cement pastes are presented in this paper. The silica fume replacement varies from 0% to 20% by weight and the water/binder ratio (w/b) is 0.4. The pore structure by mercury intrusion porosimetry (MIP), the micromorphology by scanning electron microscopy (SEM) and the compressive strength at 3, 7, 14, 28, 56 and 90 days have been studied. The test results indicate that the improvements on both microstructure and mechanical properties of hardened cement pastes by silica fume replacement are not effective due to the agglomeration of silica fume particles. The unreacted silica fume remained in cement pastes, the threshold diameter was not reduced and the increase in compressive strength was insignificant up to 28 days. It is suggested that the proper measures should be taken to disperse silica fume agglomeration to make it more effective on improving the properties of materials.     

Retardation of Cs+ and Cl− Diffusion Using Blended Cement Admixtures

Diffusion of cesium chloride through thin plates of hardened cement pastes was studied. Blast-furnace slag and condensed silica fume were used as blending admixtures in an attempt to retard the diffusion of cesium and chloride ions. The curing and diffusion temperatures were varied from 27° to 60°C, and the water/solid ratio was varied from 0.30 to 0.40. Results indicate that the cesium ion diffuses more slowly than the chloride ion in hardened cement paste systems. Blending admixtures caused a further reduction in diffusivity for both ions, which is important for preventing corrosion or restricting radionuclide transport.     

Air void morphology in fresh cement pastes

Two different procedures are used in conjunction with low-temperature scanning electron microscopy to image the air voids in cement pastes at very early ages. The first procedure isolates the air voids from cement paste after less than 30 min of hydration, and allows them to be imaged apart from the paste. The second procedure involves quenching the fresh cement paste specimens in liquid nitrogen after 5 min of hydration. In both cases, a distinct air void shell is apparent even at these short hydration times. The shell appears to be made up of small (1-5 μm) mineral particles. The second method confirms the presence of a water-rich transition zone around the air voids in the quenched pastes, consistent with earlier studies. Foam stability studies show that sodium oleate gives more stable foams than sodium dodecyl benzene sulfonate, but is more sensitive to the presence of calcium ions.     

Effect of magnesium and sulfate ions on durability of silica fume blended mixes exposed to the seawater tidal zone

The effect of silica fume on deterioration resistance to sulfate attack in seawater within tidal zone and simulated wetting-drying condition has been studied in Portland cement concretes and pastes containing silica fume (SF) with/without ground granulated blast furnace slag (GGBS). Changes in the compressive strength and capillary water absorption of specimens as a function of SF content have been investigated combined with phases determination by means of scanning electron microscopy and X-ray energy dispersion analysis. The strength change factors (SCFs) of specimens with SF (the more SF content, the higher strength loss) were greater than that of the mixes without SF or cured under tap water. Mg²⁺ ion originated attack found to be the dominating deterioration mechanism as confirmed by X-ray and chemical analyses.Further, the incorporation of GGBS with SF mixes in different exposure conditions led to the worst performance in all of the test environments. Lower cement content and hydration rate accompanied with particular chemical composition of GGBS made concrete and paste specimens to be more susceptible to deleterious seawater environment.     

Cement pastes of low water to solid ratio: An investigation of the polymerization of silicate anions in the presence of a superplasticizer and silica fume

Polymerization of silicate anions was investigated by the molybdate method in cement pastes with water to solid ratio of 0.28. Two series of samples - with and without superplasticizer admixture - were prepared using ordinary portland cement (OPC) in which 0, 5, 10, and 15 percent by weight was replaced by condensed silica fume (CSF). During the hydration ranging from 4 hours up to 50 days, the free lime content and insoluble residue were also determined. The 28 day compressive strengths of the hardened composites were between 79 and 108 MPa. The results from the molybdate complex confirm that the presence of CSF in hydrating blends and its pozzolanic activity influences the size dispersity of silica anions by increasing the proportion of polymers. It is also suggested that in the cement pastes of w/s ratio of 0.28, the conversion level of CSF by pozzolanic reaction decreased. Finally, a comparison is made between the polymerization characteristics of the 0.28 w/s-ratio pastes and pastes of w/s = 0.68 which have been hydrated for 6.3 years.     

Physicochemical characteristics of acrylic-acid polymer-impregnated cement pastes

Various Portland cement pastes were made using water cement ratios of 0·20, 0·25, 0·35 or 0·40 and then cured for 1, 3, 7, 28, 90 or 180 days. These pastes were impregnated with acrylic acid monomer under vacuum and the monomer-impregnated samples were then treated at two different temperatures, 40 or 60°C, for the polymerization process, using benzoyl peroxide as initiator. Several physicochemical studies were carried out on each cement paste; these studies include compressive strength tests, bulk density, compressive strength versus gel/space ratio relationships, polymer load, X-ray diffraction analysis and differential thermal analysis. Results have indicated that compressive strength improvement in acrylic acid-polymer impregnated cement pastes is mainly dependent on initial water/cement ratio, curing time and gel/space ratio. The results of X-ray diffraction analysis and differential thermal analysis indicated that the intrusion of polymer into the cement paste matrix does not affect the phase composition of the Portland cement hydration products.     

Sorption kinetics of superabsorbent polymers (SAPs) in fresh Portland cement-based pastes visualized and quantified by neutron radiography and correlated to the progress of cement hydration

Water sorption of two superabsorbent polymers in cement-based pastes has been characterized by neutron radiography. Cement pastes with W/C of 0.25 and 0.50 and one additionally containing silica fume (W/C = 0.42) were investigated. The SAPs differed in their inherent sorption kinetics in extracted cement pore solution (SAP 1: self-releasing; SAP 2: retentive).Desorption from SAP 1 started very early after paste preparation. Hence, its individual non-retentiveness governs its behavior only.SAP 2 released water into all matrices, but its kinetics were different. In the paste with the highest W/C, some moderate water release was recorded from the beginning. In the other two pastes, SAP 2 retained its stored liquid during the dormant period, i.e., up to the percolation threshold. Intense desorption then set in and continued throughout the acceleration period.These findings explain the pronouncedly higher efficiency of SAP 2 as internal curing admixture as compared to SAP 1.     

Autogenous relative humidity change and autogenous shrinkage of high-performance cement pastes

In this paper, the effects of water to cementitious material ratio (w/cm), silica fume (SF) and ground blast-furnace slag (GBFS) on autogenous relative humidity (RH) change and autogenous shrinkage (AS) of high-performance cement pastes were studied. The mechanism of self-desiccation caused by mineral admixture and reduction of w/cm were studied by the parameters of mineral admixture self-desiccation-effect coefficient k and efficient w/cm r_e proposed. Furthermore, the relationship between autogenous RH and AS of high-performance paste was established. The results indicate that w/cm is a chief factor that affects autogenous RH change and AS of cement pastes. The lower the w/cm of paste is, the more reduction the autogenous RH and the increment of AS are. SF increases autogenous RH reduction and AS increment of cement paste at early ages, and GBFS increases autogenous RH reduction and AS increment at later ages. The effect of mineral admixtures on autogenous RH change of paste resulting from self-desiccation can be reflected effectively by the nonlinear equation with the parameters of k and r_e. There exists a good linear correlation between autogenous RH change and AS of cement pastes.     

Evaluation of the pozzolanic activity of fluid catalytic cracking catalyst residue (FC3R). Thermogravimetric analysis studies on FC3R-Portland cement pastes

Spent fluid catalytic cracking catalyst (FC3R) from a petrol refinery has shown a great pozzolanic activity in lime pastes as have been demonstrated in previous studies. Based on these results, the pozzolanic activity of the FC3R in Portland cement pastes has been investigated. This evaluation has been carried out by means of thermogravimetry (TG) of cured FC3R-Portland cement pastes. The influence of water/binder ratio and the replacement percentage of FC3R on the pozzolanic reaction were investigated. Due to the chemical composition of FC3R that is similar to metakaolin (MK), and knowing that MK has a high pozzolanic activity, the latter was used as a material of comparison in the study of the water/binder ratio influence. The scope of this study is the determination of pozzolanic activity of FC3R when incorporated to Portland cement, and the evaluation on amount and nature of pozzolanic products. FC3R has shown a similar reactivity to MK, yielding similar pozzolanic products: CSH, CAH and CASH. The optimum replacing percentage in Portland cement pastes was in the 15-20% range.     

Cracking processes in steel fiber reinforced cement paste

Fracture patterns produced when a crack advanced from a notch in cement paste specimens reinforced with steel fibers were studied by SEM methods. The specimens were small compact tension specimens with precast notches that could be wedge loaded within the SEM chamber in a moist environment. Steel fibers were positioned either in an array of five parallel fibers spaced 2 mm apart and across the expected crack propagation path, or else were in random orientation on the plane being observed. The cracks induced by wedge loading were found to be geometrically complex and certainly could not be described as simple straight cracks as assumed in various models. On intersection with fibers oriented perpendicularly to them, the cracks tended to displace laterally and branch into a number of microcracks; on intersection with fibers at less than perpendicular angles, the tendency was for the crack to change course and run parallel to the inclined fiber. Often at perpendicular intersections the crack appeared to be arrested in the matrix 10 to 40 μm ahead of the actual fiber interface, and then produced a “pseudo-debonding crack” parallel to the fiber but some distance away from the actual interface. These cracking patterns are considered to be influenced by the microstructure of the cement paste near the interface, which is clearly different from that of the bulk cement paste.     

Some comments on the microstructure of hardened cement pastes

The effect of water/cement ratio of hardened cement pastes in the range 0.23 to 0.71 on the micropore structure as revealed by mercury intrusion and nitrogen adsorption porosimetry is discussed. The relationship between these micropores and the solid minerals as revealed by electron microscopy is also discussed.