3.1. Determination the Impact of Activation of Portland Cement in VLD on Chemical Composition by IR-Spectroscopy Method
The IR spectra of portland cement powders processed within different times in VLD and other optimal parameters of machine operation were obtained. The initial portland cement powders were processed for 1, 3, 5 and 7 min. The IR spectrograms selected immediately after activation of the investigated powders are shown in
Figure 3. Furthermore, the effect of the VLD treatment on the chemical composition of portland cement included in a mixture of portland cement and MF plasticizer was further investigated.
As shown in
Figure 3, visible differences in the IR spectra of the initial and processed portland cement powders are identified at any duration in the VLD. For example, visible spectrum increasing and widening is observed at 3645 cm
−1; 1125 cm
−1 и 885 cm
−1. A narrow peak at 3645 cm
−1 relates to valence vibrations of the –OH group included in the crystalline structure of silicates (water fixed in the crystallographic structure) [
36]. Activation leads to a reduction of this absorption band, suggesting that water is removed from the portland cement powders. This assumption is supported by the temperature of the portland cement powder processed in VLD (
Figure 4). From the data presented, it is evident that the temperature of portland cement powder up to 5 min of processing in VLD is growing almost linearly at a rate of 18.4 °C/min. Then, the growth rate of the temperature is significantly reduced (3.7 °C/min). The relationship between the temperature of the portland cement powder and the processing duration in the VLD has the following form:
It is seen from relationship (3) that the maximal heating-up temperature of portland cement powder processed in VLD above 8 min can reach 138 °C. The experimentally established heating-up temperature (the processing time in VLD at least 2 min) provided thermal degradation of such portland cement components as gypsum stone.
The absorption band of control sample is recorded in the range 3410–3420 cm
−1 associated the valence oscillations of the -OH group by hydrogen bonds [
37]. Compared to the control composition, the activation of 1 min reduces significantly the intensity of the maximum characteristic of the related -OH -groups. The longer activation time of more than 1 min leads to the disappearance of the band associated with the dehydration of the samples due to the increased temperature during processing in VLD.
The absorption band observed in the investigated infrared spectra at 1474 cm−1 is caused by the oscillation of CO3 groups in calcium carbonate crystals. The activation of portland cement practically does not affect the change in intensity of the specified band. This leads to the conclusion that there is no effect of the treatment in VLD on the calcium carbonate content in portland cement. In the range at 1125 cm−1 there are valence oscillations of the Si-O bond, and there is an increase in the spectrum with the duration of activation.
The absorption bands on the portlandcement in the area 500–885 cm
−1 by IR spectrogram correspond to the mixed complex combinations of Al-O and Ca-O, which is the case for the mixture of mono-, calcium dialuminate and α-Al
2O
3. The lack of absorption bands of -OH valence bands and the increase in the number of Al-O and Ca-O bonds in activated compositions indicate an increase in the adsorption capacity of activated portland cement. The characteristic anomaly area according to
Figure 3 is shown in
Table 1.
The IR spectra of powders were investigated within 60 days of storage from the time of their activation in order to determine the changes occurring in portland cement processed in VLD during natural atmospheric conditions. The duration of storage of portland cement powders’ storage period for 60 days is determined by the requirements of Russian standard 30515-2013, conforming to European standards EN 197-1:2011 and EN 197-2:2000. According to these standards, the portland cement storage period is 60 days in the manufacturer’s packing. The obtained results are presented in
Table 2.
After the 60-day portland cement storage period, characteristic bands are observed on IR spectrograms which tally with the samples taken immediately after activation. However, deviations in anomaly areas are observed at 1474 cm−1. More significant carbonization of portland cement (over 90% in comparison to the control composition) is identified, which increases with the duration of processing in VLD.
Thus, it is established that due to the heating during processing of the portland cement in VLD, the intensity of the maximum is reduced at 3645 cm−1, characterizing the –OH group. It is shown that the activation of portland cement has almost no effect on the intensity of the absorption band at 1474 cm−1, which characterizes the oscillation of the CO3- group in the calcium carbonate crystals. Furthermore, the infrared spectra of portland cement powders after 60 days of storage from the moment of their activation show an increase in the intensity of the peak corresponding to the CO3- group with an increase in the processing time in the VLD. In addition, the increase in the number of Al-O and Ca-O bonds in the VLD activation compositions indicates an increase in the adsorption capacity of activated portland cement.
Naturally, the physical activation of dispersed materials (powders) should lead to significant impact associated with surface effects (dissolution, adsorption, wetting, etc.). As is known, wetting heat is released by the wetting of a solid body [
38] which is more affected for lyophilic materials than for lyophobic materials. For chemically active components, surface events are expected to occur more intensively. In this context, a study was carried out on the influence of the activation of portland cement in VLD on the heat of its hydration using the differential scanning calorimetry method. Measurements were made for 480 min at different thermostat temperatures: 22 °C, 40 °C, 60 °C and 80 °C. The resulting thermograms are shown in
Figure 5, and the analysis of the received thermograms is summarized in
Table 3.
According to the
Table 3 it can be seen that the total amount of emitted heat in the first 100 min of hydration activated in the VLD is higher in 5,6 times at the temperature of the thermostat 22 °C; 1.43 times at 40 °C; and 5.126 times at 60 °C 2.24 times at 80 °C, as compared to the conventional formulations, accordingly.
Furthermore, the rate of hydration of the compositions produced by activation in VLD is greater than the compositions produced without activation 1.615 times at 22 °C; 1.85 times 40 °C; 2.71 times 60 °C; 2.3 times at 80 °C. A characteristic showing the physical activation of portland cement in VLD are the properties listed in
Table 3 assigned to the surface area of the powdered portland cement. These specific characteristics are presented in
Table 4.
Table 4 shows that the treatment of portland cement in VLD leads to the physical activation of portland cement, which is accompanied by an increase in the amount of heat generated by the hydration of portland cement and the rate of hydration.
3.2. Determination the Impact of Storage Duration of Portland Cement on Kinetics of Mass-Change and Compressive Strength of Fine-Aggregate Concrete
The influence of processing portland cement in VLD and the presence of a plasticizer on the kinetics of mass-change of portland cement during storage in laboratory conditions at a relative humidity of 70 ± 5% and at a temperature of 20 ± 3 °C within 1.5–3 months is determined. The results of the experiment are given in
Figure 6a. The analysis of the obtained experimental data was carried out according to the given characteristic ∆M/S
i, (here, S
i is the specific surface area of the i composition). The results are presented in
Figure 6b.
Figure 6a shows that the activation of portland cement increases the absorption of moisture from the natural atmosphere. Moreover, as the processing time in VLD increases, the quantity of water absorbed by the portland cement powder from the natural atmosphere increases. By 90 days of exposure in the natural atmosphere, the amount of water absorbed by composition No 3 is greater than composition No 1 by 37%, and by 53% for composition No 4. Obviously, this difference related to a change in the specific surface area of the portland cement powder. The evaluation of the quantity of absorbed water shows that this value is almost identical for compositions No 1, No 3 and No 4; the deviations are small and may be related to the physical inaccuracy of measurement (
Figure 6b). Thus, the increase in the amount of water absorbed by portland cement powder when exposed in the natural atmosphere is caused by the increase in the surface area of the powder.
A significant decrease in the amount of water absorbed by portland cement powder is observed for portland cement processed in VLD jointly with the MF plasticizer (
Figure 6). Moreover, the dependence on the processing time in VLD maintains the trends, that is, as the processing time in VLD increases, the quantity of water absorbed decreases. An obvious assumption for determining the cause of these deviations in the absorption of portland cement powder is the placement on the surface of the particles of portland cement of a plasticizer MF, which in this case should have less ability to absorb water from the natural atmosphere. The experimental test confirmed the assumption that MF plasticizer powder has less ability to absorb water from the atmosphere (
Figure 6b). Comparison of data of
Figure 6a,b shows that the water absorption of portland cement powder with the MF plasticizer is higher than for MF plasticizer powder only. This indicates the presence of a surface of portland cement powder not covered by the MF «free» plasticizer surface, the fraction of which can be calculated by the formula:
where
—the amount of water absorbed by the powder produced by the treatment in VLD of portland cement and MF plasticizer;
—the quantity of water absorbed by the powder of portland cement;
—the quantity of water absorbed by the powder of MF plasticizer. The results of the Formula (4) calculation are presented in
Table 5.
Table 5 shows that the average proportion of free surface of portland cement particles treated with MF plasticizer in VLD is, on average, 34%. The variation of
in time is linked only to random changes in experimental data.
The absorption of water from the natural atmosphere by portland cement should certainly have a negative impact on its strength. In this connection, studies have been carried out on the hardening kinetics of fine-grained concrete from portland cement stored in open air for 0–90 days. The results of the tests are presented in
Table 6.
Table 6 shows that storage of portland cement in air results in a significant reduction in the compression strength of the resulting fine-grained concrete. Thus, the reduction in compression strength is up to 36% in compositions No 1 and No 4, and up to 43% in compositions No 2–3, and No 5–6.
The introduction of MF during the grinding of portland cement in VLD and the further storage of obtained cement in open air can slow down the rate of strength reduction of fine-grained concrete. Thus, the decrease in strength is 26–34%, depending on the activation time (compositions No 5–6).
Table 6 shows that the introduction of plasticizer reduces the water–cement ratio, which naturally increases the strength of the material (according to the Bolomei–Skramtaev law [
39,
40]). In this context, it is interesting to estimate the activity of portland cement in fine-grained concrete (comparing of the compositions No 1–3 and No 4–6). This can be estimated by a formula that is a consequence of the Bolomei–Skramtaev law:
where
is the activity (strength) of portland cement;
—concrete strength;
is thecement–water ratio; index «1» и «2» complies with concrete without (compositions No 1–3) and with plasticizer (compositions No 4–6), respectively; sign «+» is chosenas the rule of application in accordance with the rule of application of the Bolomei–Skramtaev law for fine-grained concrete with plasticizer
.
According to
Table 6, the ratio of
for compositions No 1 and No 4 is 1.33, for compositions No 2 and No 5—
and for compositions No 3 and No 6—
. These values display that the contribution of portland cement to the strength of concrete without plasticizer is higher than with a plasticizer. This result is appropriate, as portland cement hydration products in concrete without plasticizer have no spatial difficulty in placing. However, this has negative consequences, both in the formation of a less dense cement stone and in the formation of a cement stone containing fewer clinker stocks, which is a prerequisite for a low durability [
41,
42].
3.3. Determination the Impact of Activation of Portland Cement in VLD on Phase Composition of Nanomodified Cement Stone
The spectrograms of cement stones obtained without activation and with activation of portland cement in VLD were explored for determination the impact of activation exposure against chemical composition change. Spectrograms of cement stone samples received by traditional method are shown in
Figure 7. Quantitative values on basic wavenumberare presented in
Table 7.
IR spectrograms of samples of cement stone processed by the activation of portland cement in VLD is illustrated in
Figure 8. Estimated absorption areas according to data in
Figure 8 are shown in
Table 8.
The chemical composition of the hydration products of the cement stone samples, obtained by various methods, has a response rate of 3642; 3400; 1415; 1102; 950; 874 cm
−1. The absorption band at 3642 cm
−1 was caused by oscillations of –OHgroups and identifies calcium hydrosilicates of different structures (portlandite, xenolite and other hydrosilicates of similar structure). The number of such hydrosilicates in compositions No 2, No 3, produced without activation, decreases by 8–25%, compared to composition 1. The combined activation of CNT TUBALL and portland cement in VLD also reduces the number of hydrosilicates in the cement stone by 8%, compared to the unmodified composition. The modification of the CNT Graphistrength does not affect the content of the mentioned hydrosilicates. The number of hydrosilicates in this group is reduced by 61–69% in compositions No 4–5 (
Table 8).
Haчaлoφopмы
Koнeцφopмы
The presence of anomalies at reflections of 1400–1600 cm
−1, as well as a wide-banded spectrum in the region of 3400–3500 cm
−1, indicates the presence of submicrystallinehydrosilicates of a tobermorite-like structure [
43]. The number of hydrosilicate data in the compositions No 2–3 is decreasing compared to the reference composition, while the introduction of the MF additive makes it possible to increase the number of sub-microcrystalls of hydrosilicates of a tobermonite-like structure by 6–32% (
Table 7). The activation of portland cement together with CNT and MF in VLD increases the number of mentioned hydrosilicates by 22–55% (
Table 8). The clearer resolution of the spectrum in this area indicates a higher degree of roundness of the mentioned hydrosilicates in MF-added compositions.
Haчaлo φopмы
Koнeц φopмы
Absorption bands at 1415–1473cm
−1 indicate oscillations of the –OH groups at the vertices of silica–acid tetrahedrons, either belong to calcium carbonate, or may indicate the presence of both components [
44]. The insertion of CNT in various methods for producing cement stone slightly reduces the amount of the silicate phase in comparison with preparations without additives. The activation of portland cement in the VLD together with the CNT and the MF additive (compositions No 4, No 5
Table 8) contributes to the formation of the silicate phase by 48–66% more than in the composition No 1. In compositions No 4, and No 5,
Table 7, the number of the silicate phase increases by 17–49%, compared to composition No 1. The calcium hydrosilicates of the two-dimensional structure and the one-dimensional ones in the form of chains ν(SiO), observed in absorption bands of 1000–1100 cm
−1 increase in compositions No 2 and No 3,
Table 7 (39% for CNT Graphistrength, and 47% for CNT Tuball as compared to No 1). The additional introduction of MF significantly increases the number of mentioned calcium hydrosilicates (84–261%).
With the activation of portland cement in VLD, the introduction of CNT TUBALL causes a decrease in the calcium hydrosilicates of the two-dimensional structure, and the introduction of the CNT Graphistrength causes an increase of 28%. The activation of portland cement results in a 39% increase in the hydrosilicate data (compositions No 1
Table 7 and
Table 8). In compositions No 4–5 of
Table 8, the quantity of calcium hydrosilicates of the two-dimensional structure does not increase significantly.
The valence oscillations of Si(OH), as well as the groups of calcium hydrosulfoaluminate (HSAC) occur at 950 cm
−1 [
45]. The modification of the cement stone CNT shows a decrease in the intensity of the reflection band, and hence the modification of the CNT results in a reduction of the HSAC group by 21–23% (compositions No 2–3 of the
Table 7). The activation of portland cement in the VLD together with the CNT does not reduce the amount of GAC. Regardless of the method for producing cement stone, the introduction of the MF additive makes it possible to significantly reduce the amount of HSAC.
Weak reflections at 874 cm
−1 are characteristic of –(Si
4O
10) groups, that is, the silicate phase [
46]. The introduction of CNT increases the number of silicate phases having –(Si
4O
10) groups. For example, the modification of cement stone CNT increases the quantity of this silicate phase by 6–21% in the compositions of
Table 2,
Table 3 and
Table 7, and 5–5.1% in
Table 2,
Table 3 and
Table 8. The introduction of the MF additive makes it possible to substantially increase the number of silicate phases having –(Si
4O
10)-groups, especially in the compositions produced by VLD activations by 85–28%.