Paste phase composition

The chemical composition of pastes determined with electron microprobe and electron scaiming microscopy revealed that the CVS ratio in C-S-H was between 1.7 and 2.0. In the set of experimental results the local maximum values were 1.75 and 1.95 respectively [20]. The ratios of other atoms to calcium are approximately as follows Na 0.01 Mg 0.01-0.08 A1 0.04-0.08 S 0.01-0.03 K 0.01-0.02 Fe 0.01-0.03 [21-23], The composition does not change either with time of curing— except of S/C—or with w/c it seems to be related to the chemical composition of alite and belite. 

Taylor and Newbury [24] presented the results of analysis in the system of coordinates (A1+Fe)/Ca-Si/Ca and obtained a triangle which is shown in Fig. 4.7. Near the apexes, where there is most of points, the chemical composition corresponds to the following products CH, C-S-H and hydrogamets. Taylor and Newbury are of the opinion that the points in middle position, not near the apexes, correspond to the mixtures of phases. In the pastes matured for a short time the AFm phase was found [25]. The differences between the inner and outer C-S-H was not observed, perhaps except of some minor components their percentage in onter product is a tittle lower [25, 26]. The Ca/Si ratio higher than 2, results presumably from the presence of C-S-H—AFm mixtures in nanometric scale. This is possible because of the layer structure of both phases this perfect mixture of gels can be thns explained [9]. On the other side the monophase mixtures of CH with C-S-H are observed under the microscope only in the pastes with very low w/c [9]. 

A typical CH content in the pastes prodnced from Portland cement is in the range 18-25 % after 3 month of hardening. AFt is present only at early age. Using some assumptions and methods Taylor [27] proposed a mass balance for the paste prepared at w/c=0.5 and stored one year at 11%RH. The following volume ratios of particular phases were proposed clinker 0.05 CH 0.11 CaCOj 0.01 C-S-H 

This is the consequence of low mobility of Fe ions which do not migrate but are located in hydrates formed in situ from the ferrite phase [27]. Apart from the iron ions in octahedral sites there are the AF+, Tf + and ions the latter ones are concentrated inbrownmillerite in clinker. The magnesium ions show also a low mobility in pore solution they do not produce bracite but hydrotalcite [Mg Fe, 2(011)2] (C03)o, 25(OH)o5 [27]. The formation of hydrotalcite in situ after brownmillerite is promoted by the highest concentration of magnesium in this phase. 

---

Changes in (a) paste phase composition and (b) temperature during 30 min of paste mixing [10]. 

Influence of H 04/Pb0 ratio on phase composition and crystal morphology of the paste Phase composition... 

There is relatively an unanimity opinion about the role of paste phase composition on the strength of concrete. The high strength is attributed to the C-S-H phases, as in the case of ambient temperature, as well to 1.1 nm tobermorite [136, 155]. On the other hand the a-C2SH is causing the significant strength decrease [1, 156, 178, 199], For this reason the 10-20% quartz addition to the paste causes the decrease of the mechanical properties of material, however, the quartz addition of 30-40% is favourable [1]. 

Hydration gives rise to effects on pore filling and the consequent enhancement of mechanical performance (low-porosity pastes are stronger than high-porosity ones). The first fast hydration step is followed by a relatively dormant period that may last 6 months or more, depending on temperature, particle size, and aqueous phase composition. In order to control the hydration step, alkyl sulfonate salts surface-active substances (SAS) are used. 

Several studies on the quaternary systems of CaO-SiOj-HiO with NajO or K,0 have been reported (K19,S52,M52). Alkali greatly lowers the concentrations of CaO in the solution and raises those of Si02. The solid phase compositions are difficult to study. Determinations based on changes in concentration on adding CH to alkali silicate solutions are subject to considerable experimental errors, while direct analyses of the solid are difficult to interpret because the alkali cations are easily removed by washing. Suzuki ei al. (S52) considered that they were adsorbed. Macpheeef /. (M52) reported TEM analyses of the C-S-H in washed preparations obtained by reaction ofCjS (lOg) in water or NaOH solutions (250 ml). The C-S-H obtained with water had a mean Ca/Si ratio of 1.77 that obtained with 0.8 M NaOH had a mean Ca/Si ratio of 1.5 and a mean NujO/SiOj ratio of 0.5. These results do not appear to be directly relevant to cement pastes. The pore solutions of the latter may be 0.8 M or even higher in alkali... 

The experimental considerations applying to calcium silicate pastes (Sections 5.1 and 5.2) are equally relevant to cement pastes. Of the methods so far used in attempts to determine the degrees of reaction of the individual clinker phases as a function of time, QXDA (C39,D12,T34,P28) has proved much the most satisfactory. Procedures are essentially as for the analysis of a clinker or unreacted cement (Section 4.3.2), but it is necessary to take account of overlaps with peaks from the hydration products, and especially, with the C-S-H band at 0.27-0.31 nm. The water content of the sample must be known, so that the results can be referred to the weight of anhydrous material. If a sample of the unhydrated cement is available, and its quantitative phase composition has been determined, it may be used as the reference standard for the individual clinker phases in the paste. 

There are probably no effective direct methods at present for determining either C-S-H or AFm phases in cement pastes in both cases, this is probably attributable to the low degree of crystallinity. Odler and Abdul-Maula (015) found that determination of AFm phase by QXDA was only semiquantitative. Postulated quantitative phase compositions of cement pastes may, however, be tested by comparing observed and calculated TG curves (Section 7.3.3). 

The essential input data are (a) the bulk chemical composition of the cement, (b) the quantitative phase composition of the cement and the chemical compositions of its individual phases, (c) the fraction of each phase that has reacted, (d) the w/c ratio, (e) the COj content of the paste and an estimate of how it is distributed among phases, and (0 the composition of each hydrated phase for the specified drying condition. If (b) is unknown, it may be estimated as described in Section 4.4, and if (c) is unknown, it may be estimated from the age as described by Parrott and Killoh (P30), or, more simply though less precisely, by using empirical equations (D12,T37). If the phase composition by volume and porosities are to be calculated, densities of phases are also required. 

In principle, it should be possible to calculate the heat of hydration from the quantitative phase compositions of the unreacted mix and of the paste, using standard enthalpies of formation. The sensitivity of such calculations to small errors in the latter data probably renders this approach unsatisfactory with existing data. 

Table 7.5 Coefficients in equation 7.3 for predicting the cumulative heat evolution in a cement paste of given age from the potential phase composition of the cement (for w/c = 0,4 and 21°C) (C38)... 

The phase composition of the paste by weight and by volume, and the non-evaporable water content, were calculated using procedures... 

In Section 7.3.3, a method was described for calculating the quantitative phase composition of a cement paste by weight and by volume for various drying conditions. Fig. 8.5 includes porosities thus calculated for 18-month-... 

Calculations based on reaction stoichiometry and densities of phases support the conclusions from experimental observations that mature pastes of composite cements are more porous than comparable pastes of Portland cements. This is indicated by the results in Table 7.3, 9.4 and 9.6. Similar calculations for 180-day-old pastes of w/s 0.45 indicate free water porosities of about 24% for a typical Portland cement, 35% for a cement with 40% slag, 35% for one with 40% pfa and 32% for one with 30% microsilica. The calculated values are in all cases somewhat higher than observed mercury porosities (F34,F41). 

G65 Grudemo, A., Strength-structure relationships of cement paste materials. Part I. Methods and bctsic data for studying phase composition and microstructure (CBI Research, 6 77), 101 pp., Swedish Cemcnl and Concrete Research Institute, Stockholm (1977) also private communication quoted by L.-O. Nilsson in Report TVBM-1003, Division of Building Materials, University of Lund, Sweden (1980). 

In H2SO4 solution of low rel. dens., the zonal processes should be carefully controlled as they create an inhomogeneity in the phase composition of the paste throughout the plate cross-section, which leads to inhomogeneity of the PAM structure. 

Fig. 3.13. Dependence on formation time of (a) phase composition and (b) chemical composition of paste and active mass, and (c) plate potential under formation current (

Table 3.1 [15]. The volume of the particles increases during the oxidation of PbO to p-Pb02, and decreases during the oxidation of PbS04 and basic lead sulfates to Pb02. Thus, the overall volume change of the crystals will depend on the phase composition of the paste.

Fig. 3.30. Pore volume and pore surface-area distribution for PAM formed from pastes with various phase compositions. Arrows mark the assumed boundary between micro- and...

Fig. 3.31. Dependence of cell capacity on total pore volume (Kp) and on volume of micropores (F p) for PAM prepared from pastes with various phase compositions [36].

The capacity of plates with lead-antimony grids and different phase composition and density of the pastes as a function of the BET surface-area of the PAM (at the fifth cycle) is shown in Fig. 3.32 [36]. In general, the plate capacity increases with increase in BET surface-area. Several plates have comparatively low BET surface-area but high capacity. Hence, the BET surface-area of the PAM is not the sole parameter that determines plate capacity. 

The dependencies of the phase composition of the paste and the cycle-life of the positive plates on the acid-to-oxide ratio used for paste preparation are presented in Fig. 3.33 [34]. The relative intensities of the characteristic diffraction lines for the different phases in pastes prepared using different acid-to-oxide ratios are shown in Fig. 3.33 (a) and (b). The sulfation of the pastes increases with increase in acid content. 

Fig. 3.33. (a), (b) Phase composition of pastes prepared with different amounts of acid at 30 and 80°C (c), (d) Changes in C20 capacity of 12-V/42-Ah automotive batteries with positive plates produced from 3BS, 4BS, and IBS pastes (depending on acid strength and preparation temperature) on cycling at 50% DoD. The battery capacity is limited by the positive plates [34]. 

---