1. Introduction
During the hardening of cement mortars and concretes, cracking of the material may occur due to its shrinkage [
1,
2]. Shrinkage of cement materials occurs as a result of a decrease in the volume of cement phases and can last for years [
3]; however, the most intense phase occurs in the first 3–4 months. According to the mechanism of the shrinkage of cement materials, it may be one of three types: plastic shrinkage, autogenous shrinkage, and drying shrinkage. The first occurs, plastic shrinkage, occurs during the plastic state of concrete, i.e., before the final setting time [
4]. It is associated with the evaporation of pore water due to the dispersion of a water film from the surface of the concrete [
5]. Autogenous shrinkage is a macroscopic change in volume due to capillary pressure caused by self-drying and the chemical process of cement hydration [
6]. Drying shrinkage occurs during the drying of concrete and is caused by the loss of moisture into the environment due to evaporation [
7].
High autogenous shrinkage that develops at an early age can lead to cracking and thus affect the mechanical properties and durability of hardened cement-based materials [
6]. In addition, hardening accelerators are added to many concrete mixtures, which increases the hydration rate of these mixtures. The increase in the hydration rate causes a high heat of hydration, which can increase autogenous shrinkage and subsequent cracking in concrete [
8]. Shrinkage-induced cracking of cementitious materials creates a pathway for aggressive agents (e.g., chlorides and sulfates) to penetrate the material, thereby deteriorating the serviceability of the structure [
9,
10].
To reduce the autogenous shrinkage of cement-based materials, coarse natural aggregate is replaced with recycled aggregate [
11,
12]; superplasticizers [
13] and viscosity modifiers [
6,
14] are used; various fibers are included in the composition [
15,
16,
17]; curing conditions are changed or internal curing is used [
18]; and additional cementitious materials are used [
19,
20]. However, these measures do not completely eliminate shrinkage. The most effective way to combat shrinkage is to compensate for it using expansive agents [
21,
22,
23].
In practice, several types of expansive agents are mainly used. One such admixture is magnesium oxide (MgO) [
24], the hydration of which results in the formation of magnesium hydroxide (Mg(OH)
2). The volume of magnesium hydroxide is greater than the volume of MgO before hydration, which means that MgO can be useful as a shrinkage-reducing agent in cementitious materials [
25]. When 10% of cement was replaced by reactive MgO, the expansion caused by the formation of Mg(OH)
2, resulting in a 93% reduction in autogenous shrinkage [
26]. Another expansive admixture is CaO. One study [
27] found that replacing 2.5%, 5.0% and 7.5% of cement with CaO resulted in a 42%, 47% and 80% reduction in autogenous shrinkage, respectively, in 28 days. Expansive agents of the sulfoaluminate type are known to include the expanding components C
4A
3S (4CaO·3Al
2O
3·SO
3), C
3A (3CaO·Al
2O
3), and CA (CaO·Al
2O
3), and the main product of expanding hydration is ettringite (3CaO·Al
2O
3·3CaSO
4·32H
2O).
Other studies have shown that high-calcium fly ash from thermal power plants can serve additives [
28]. The authors of [
29] reported that replacing cement with class C fly ash leads to lower autogenous shrinkage (concrete with 40% fly ash had a shrinkage strain of 100 μm/m, and concrete with 60% fly ash had a shrinkage strain of 50 μm/m) compared to mixtures without fly ash (with a shrinkage strain of 200 μm/m). When adding shale ash (containing 10% CaO) to cement, shrinkage decreased on average by 1.3 and 1.8 times with a shale ash content of 30% and 40%, respectively [
30]. Fly ash, containing a large amount of free lime, has binding properties, but during hardening, it greatly increases in volume, which limits its use both as a binder and as a partial replacement for cement in concrete [
31]. Thus, the selection of solutions to eliminate the negative characteristics of high-calcium fly ash for use in cement materials solves one of the most important problems of modern construction science—the development of new building materials that can be manufactured using various types of waste [
32,
33].
The work aims to study the possibility of compensating for autogenous shrinkage of cement materials with an expansive agent based on high-calcium ash from the Nazarovo Power Station (Krasnoyarsk region, Russia).
2. Materials and Methods
2.1. Materials
The materials used for manufacturing the specimens were Portland cement CEM I 42.5; sea sand of fraction 0–5 mm with a fineness modulus of 2.8; distilled water; calcium nitrate (chemical reagent of analytical grade) in the form of an aqueous solution of 37.5% concentration; and high-calcium fly ash from the Nazarovo Power Station obtained by burning brown coals from the Kansk-Achinsk basin. Chemical analysis of fly ash was carried out (ZAO Regional Analytical Center “Mekhanobr Engineering Analit”). Mg, Fe, Si were determined by atomic emission spectrometry with inductively coupled plasma, Na was determined by the flame spectrophotometric method, S was determined by the titrimetric method, loss on ignition was determined by the gravimetric method, Ca was determined by the complexometric method, and chlorine was determined by the turbodimetric method. According to the chemical composition data (
Table 1), the fly ash had a high content of calcium oxide and could therefore be used as an expansive agent.
Calcium nitrate is an explosion-proof and non-flammable product. Calcium nitrate is not a carcinogen. It is easily regenerated and disposed of. However, when working with this additive, it is necessary to use personal protective equipment—a suit, a respirator, rubber gloves, and shoes.
2.2. Methods of Investigating High-Calcium Fly Ash
The composition of the hydration products of high-calcium fly ash from Nazarovo Power Station was studied using X-ray diffraction analysis (XRD) and differential thermal analysis (DTA). The test specimens were prepared as follows. The original fly ash was mixed with distilled water and stored in a desiccator above water. After 28 days, the specimens were dried, crushed, passed through a 025 sieve, and subjected to vacuum drying for 3 h at a residual pressure of 3.3 Pa. The dried specimens were ground in an agate mortar until they passed through a 005 sieve. The analysis was carried out on a Dron 7 X-ray diffractometer with the following parameters: CuKα radiation, λ = 0.15406 Å and a 2θ shooting range from 6 to 86° with a step of 0.02° and an exposure of 5 s. The decoding of X-ray patterns (phases and their percentage content) was carried out in the QuanCem program (OAO Burevestnik, Russia). The analysis is based on the dependence of the intensity of the diffraction maxima of the phase being determined on its content in the sample. A semi-quantitative analysis was carried out, in which the exact amount of the substance was not measured. The results of the analyses were expressed as an approximate estimate of the measured amount of the substance based on the height or area of the X-ray pattern peaks. This estimate was characterized as a numerical value. Differential thermal analysis was performed using the Thermoscan-2 device.
The degree of fly ash expansion during hydration was determined using Le Chatelier rings. The fly ash paste was made manually. The fly ash was first mixed with solutions of some mineral electrolyte salts with a concentration required to obtain a given percentage of the additive based on dry matter, and then water was added until an approximately equal consistency was obtained for all compositions, which was assessed visually and by the mixing force. Electrolyte salt additives were used to assess the possibility of regulating the degree of fly ash stone expansion. Of the tested additives, such as MgCl2, CaCl2, Al2SO4, NaCO3, Ca(NO3)2, only calcium nitrate had a significant effect on fly ash stone, increasing its degree of expansion. Fly ash volume change tests were conducted based on the Russian State Standard GOST 30744-2001 “Cements. Methods of testing with using polyfraction standard sand” methodology. Two Le Chatelier rings were mounted on glass plates and filled with fly ash paste simultaneously without compaction. The excess paste was cut off with a knife; the rings were covered from above with glass plates with a weight of 100 g and cured in air at a temperature of (20 ± 2) °C and a relative air humidity of 45–55%. The distance f between the ends of the indicator needles was periodically measured with a caliper with an accuracy of 0.5 mm, and the difference Δf = f − d was calculated, where d was the value of f before the experiment.
The compositions of the fly ash paste are given in
Table 2. A chemical reagent of analytical grade was used as an additive of calcium nitrate.
2.3. Composition and Production of Fine-Grained Concrete Specimens
The shrinkage and strength of fine-grained concrete containing an expansive agent (D) as a partial replacement for cement (10, 20 and 30%) were studied. The composition of additive D varied in terms of Ca(NO
3)
2 content, with three values, 5, 10, and 15% of the fly ash weight. The composition symbols are given in
Table 3.
The composition of fine-grained concrete was characterized by the following ratios: a cement–sand ratio of 1:3 and W/(cement + D) = 0.47 (taking into account water in calcium nitrate solution). The content of D in the binder and Ca(NO
3)
2 in D are indicated in
Table 4.
The specimens were manufactured in accordance with the methodology of the Russian State Standard GOST 30744-2001. Ten experimental compositions of concrete mix were manufactured, differing in the content and composition of D (
Table 3).
The sand was poured into the dosing device of the paddle mixer. Water and the required amount of Ca(NO3)2 solution were poured into the mixer bowl, previously wiped with a damp cloth, and then cement and fly ash were added, after which the mixer was started in automatic mode. In this case, cement, fly ash and liquid components were mixed at low speed. After 30 s, sand was fed for 0.5 min, after which mixing continued for another 30 s at high speed. Then, mixing was stopped for 90 s before a final mix at high speed for another 60 s. The total mixing time of the concrete mix was 150 s, not counting the stopping time.
The prepared mix was placed in a three-section metal mold for making prism specimens, measuring 40x40x160 in two layers of mm. First, approximately 300 g of the mix was placed in each section. The placed mix was leveled for the first layer and compacted with 60 jolts (drops from a height of 15 mm) on a jolting apparatus. Then, the compartments were filled with the remaining mix, which was leveled for the second layer and compacted with a similar cycle of 60 jolts. An illustration of the production of specimens is shown in
Figure 1.
The mold with the specimens was covered with a plexiglass plate and placed in a wet storage chamber. After 24 h from the moment of production, the forms were dismantled and the specimens were placed on a grid in a water bath. The water temperature during storage of the specimens was regulated automatically and was (20 ± 1) °C. Every 14 days, half the volume of water in the bath was replaced with fresh water.
2.4. Testing Methodology for Fine-Grained Concrete Specimens
In order to study the influence of D on shrinkage caused by chemical processes in the system, 40 × 40 × 160 mm prism specimens, manufactured as indicated above, were insulated with two layers of water- and steam-impermeable material immediately after stripping. The first layer consisted of aluminum foil on a self-adhesive base; the second layer was made of hydraulic tape. The tightness of such insulation was previously checked by periodically weighing the specimens for more than 200 days. The difference in weight values over the entire period did not exceed 0.2 g.
The linear deformation of the specimens was measured using the device shown in
Figure 2, equipped with clock-type indicators with a division value of 1 µm. The specimens were installed in the measuring device immediately after stripping and waterproofing, i.e., 24 h after manufacture. During the test, the specimens were kept in the laboratory at a temperature of 20 ± 2 °C and a relative air humidity of 60 ± 5%. The expansion–shrinkage deformation was measured on two twin specimens simultaneously.
After the shrinkage test, specimens of 10 experimental compositions were tested at the age of 90 days, first for tension in bending, and the resulting beam halves were tested for compression. Three identical specimens of each composition, made from the same batch and cured under the same conditions, were tested for strength. The results presented were obtained as arithmetic means of three values.
3. Results and Discussion
3.1. Phase Composition of Fly Ash before and after Hydration
The results of the study of the composition of the hydration products of high-calcium fly ash from Nazarovo Power Station are presented in
Table 4, and the X-ray diffraction patterns are shown in
Figure 3.
Fly ash from Nazarovo Power Station was characterized by a significant content of free lime and quartz. The fly ash also contained periclase, aluminum sulfate, and almost all the main minerals of Portland cement clinker. In addition, the fly ash in its original state contained about 5.5% calcium hydroxide, which was obviously formed due to partial slaking of free lime by atmospheric moisture during storage. At the same time, the fly ash contained 34.8% CaOfree, which should provide expansion. As a result of hydration, the content of this lime decreased by more than two times; however, the amount of Ca(OH)2 increased insignificantly, to 6.4%. Here, we can assume firstly that there was a slowdown in the lime hydration reaction in the presence of silicates, aluminates, and calcium ferrites through analogy of the effect of clay impurities on the slaking of air lime; secondly, we can assume the participation of CaO in the formation of calcium hydroaluminosilicates, including in their X-ray amorphous form.
The results of the DTA of fly ash from Nazarovo Power Station are presented in
Figure 4.
Curve 1 has two endothermic peaks: 550 °C (decomposition of Ca(OH)2) and 885 °C (decomposition of CaCO3). As a result of hydration, an endothermic peak at 150 °C appeared on the thermogram (curve 2), which can be attributed to the dehydration of ettringite and calcium aluminosilicates. A wide endothermic depression at a temperature of 300–400 °C may correspond to the dehydration of calcium hydrosilicates. An increase in the endothermic effect at 550 °C indicates an increase in the amount of Ca(OH)2 as a result of the hydration of CaO.
According to the results of X-ray phase and differential thermal analysis, it can be assumed that the expansion of fly ash during hydration occurred as a result of the formation of two phases—Ca(OH)2 and ettringite.
3.2. Determination of the Degree of Expansion of Fly Ash during Hydration
The test results are shown in
Figure 5 and
Figure 6. It was found that the fly ash from Nazarovo Power Station itself underwent negligible expansion, which seems strange, judging by the significant amount of free lime in the fly ash. For example, the fly ash from Berezovskaya GRES with a comparable content of CaO
free greatly increased in volume during hydration [
31]. An explanation for this is given in the work [
34], where the morphology of free lime particles in fly ash from Nazarovo Power Station and Berezovskaya GRES is compared. In the first fly ash, CaO was in a special state, wherein an individual particle looked like a cluster of much smaller subparticles. The lime in the Berezovskaya fly ash was completely devoid of this property.
As can be seen from
Figure 5 and
Figure 6, the minor expansion of fly ash during hydration can be significantly increased by adding Ca(NO
3)
2. The higher the calcium nitrate content, the greater the expansion of the fly ash paste. Presumably, the effect of Ca(NO
3)
2 on expansion can be explained by the reaction of this salt with calcium aluminates to form expanding compounds. It is known that when tricalcium aluminate interacts with a calcium nitrate solution with a concentration of at least 1.22 mol/L, calcium hydronitroaluminate of the trinitrate form 3CaO·Al
2O
3·3Ca(NO
3)
2·16–18H
2O is formed, which can produce expansion similar to calcium hydrosulfoaluminate. In our case, the content of the Ca(NO
3)
2 additive in the solution with mixing water was 0.64, 1.28 and 1.46 mol/L, which corresponds to 5, 10, and 15% content of dry additive within the total fly ash mass. With a 5% additive content, the solution concentration was significantly less than 1.22 mol/L. At the same time, the increase in expansion deformation compared to pure fly ash was insignificant. In the other two cases, the solution concentration exceeded the specified value of 1.22 mol/L and, accordingly, we saw a sharp increase in deformation.
The obtained results show the possibility of using a “Nazarovo Power Station fly ash + calcium nitrate” complex as an expansive agent (D) in concrete to compensate for its shrinkage.
3.3. Effect of Expansive Additive on Strength of Fine-Grained Concrete
The results of the strength tests are given in
Table 5.
As can be seen from
Table 5, the replacement of up to 20% of cement with fly ash from Nazarovo Power Plant (compositions 0D
0, 10D
0, and 20D
0) with a sufficiently long curing period had virtually no effect on the strength of concrete, either in bending or in compression. The difference in the results was within the margin of error of the experiment. There were also few differences in the strength of other specimens containing 5–15% Ca(NO
3)
2 in the expansion additive. At the same time, the strength of these compositions was on average 9% lower than the strength of specimens that did not contain Ca(NO
3)
2. At the same time, calcium nitrate was an accelerator of cement hardening. Its decreasing effect on the strength of concrete in this case can possibly be explained by the expansion effect. The results do not reveal any noticeable effect of the calcium nitrate content in the expansion additive on the strength. The proportion of complex additive D in the binder had a somewhat greater effect on the strength.
3.4. Effect of Expansive Agent on Autogenous Shrinkage of Fine-Grained Concrete
The effect of fly ash on the shrinkage of fine-grained concrete (compositions 0D
0, 10D
0 and 20D
0) is positive (
Figure 7). Shrinkage, although insignificantly, is reduced. This effect increases with an increase in the fly ash content in the binder. This result should probably be attributed to the expanding effect of fly ash from the Nazarovo Power Plant.
The presence of calcium nitrate in the fly ash reduces shrinkage to a greater extent (compositions 10D0, 10D5, 10D10 and 10D15) and even causes a slight expansion of the concrete, which turns into shrinkage after reaching maximum expansion at the age of about 3 days. The transition through 0 to the negative region of linear strain increment values occurs after 8–10 days. During this period, the concrete gains sufficient strength, and subsequent shrinkage is not dangerous in terms of crack formation. With the same amount of expansive agent, the shrinkage is smaller, the more Ca(NO3)2 in the additive. Noticeable expansion occurs starting with a 10% content of calcium nitrate.
The results of determining the autogenous shrinkage of fine-grained concrete depending on the content of Ca(NO
3)
2 in the expansive agent (with the replacement of cement with 20% expansive agent) are shown in
Figure 8.
The results shown in
Figure 8 confirm the previously made conclusions about the effect of calcium nitrate on the expansion of concrete in the initial stages of hardening and subsequent shrinkage. Moreover, in the case of shrinkage after expansion, it turns out to be greater, the higher the content of Ca(NO
3)
2 and, accordingly, the higher the expansion.
The effect of the Ca(NO
3)
2 content in the expansive agent on the autogenous shrinkage of concrete when replacing cement with expansive agent by 30% is shown in
Figure 9.
The data in the graphs in
Figure 9 are similar to what took place with a lower content of the expansive agent. Calcium nitrate increases the degree of expansion of concrete, but subsequent shrinkage increases faster in the case of a higher content of Ca(NO
3)
2.
The latter circumstance has no explanation yet. The effect of the proportion of cement replacement by the expansive agent on expansion and autogenous shrinkage is shown in
Figure 10 and
Figure 11.
Increasing the proportion of the expansive agent in the binder at a constant content of calcium nitrate leads to an increase in the initial expansion of the concrete. After reaching the maximum value of expansion of the specimens, shrinkage begins. The time to reach this maximum is practically independent of the amount of expansive agent and is about 3 days.
Table 6 shows the dosages and recipe for complex additive D for obtaining shrinkage-free or expanding concrete.
4. Conclusions
The composition of an expansive agent for concrete, consisting of fly ash from Nazarovo Power Plant and calcium nitrate at 5–15% of fly ash weight, has been developed and studied. Based on the results of the study, the following conclusions can be drawn.
The content of Ca(NO3)2 can be used to regulate the degree of expansion of the agent during hydration and, accordingly, to control shrinkage, to obtain shrinkage-free or expanding concrete. The shrinkage–expansion deformations of concrete can also be regulated by the amount of cement replaced by the expansive agent. In this case, fly ash provides a pozzolanizing effect, and calcium nitrate is an activator of fly ash expansion. Without an activator, the degree of expansion of this ash is insufficient.
Replacement of 10% of cement with a complex additive containing 5% calcium nitrate reduces shrinkage by two times. If the calcium nitrate content in this additive is from 10 to 15%, a slight expansion occurs, compensating for shrinkage. With 20–30% replacement of cement with an additive, depending on its calcium nitrate content, it is possible to obtain shrinkage-free concrete (with a Ca(NO3)2 content of less than 5%) or an expanding composition (with a Ca(NO3)2 content of 5 to 15%). The expansion deformation increases with an increase in the content of the activating agent in the additive.
The effect of fly ash on concrete shrinkage is positive. Replacing part of the cement (up to 20%) with fly ash from Nazarovo Power Plant somewhat reduces concrete shrinkage, probably due to a slight expansion of the fly ash. This effect increases with an increase in the fly ash content in the binder.
The presence of calcium nitrate in the fly ash causes significant expansion of the fly ash and, accordingly, reduces the shrinkage of concrete, and with a Ca(NO3)2 content in fly ash of 10% or more, concrete experiences expansion in the initial stages of hardening for up to 3 days. After this, shrinkage deformation occurs. The transition of this deformation through 0 to the negative range of values occurs in 8–15 days. During this period, the concrete gains sufficient strength, and subsequent shrinkage is not dangerous in terms of cracking. With the same amount of expansive agent, more Ca(NO3)2 in the additive induces less shrinkage. An increase in the proportion of expansive agent in the binder with a constant content of calcium nitrate leads to an increase in the initial expansion of concrete. After reaching the maximum value of expansion of the specimens, shrinkage begins. The time taken to reach this maximum is practically independent of the amount of expanding additive and is about 3 days. The transition of spontaneous deformation through 0, from expansion to shrinkage, after the first 8–15 days of hardening can be explained as follows. Cement stone swells in water and shrinks in air due to the predominant content of the gel component. At first, the gel, which has not yet gained sufficient strength, is subject to the expanding action of the additive, but this action is limited in time by the hydration process of free lime as well as the formation of ettringite. Upon completion of this chemical process, normal shrinkage occurs, which is characteristic of cement compositions.
Compositions without the use of calcium nitrate, when replacing 10–30% of cement with fly ash only, do not show a decrease in strength compared to the reference mix. The use of an expansive agent containing 5–15% Ca(NO3)2 reduces the strength of concrete by an average of 9%, despite the fact that calcium nitrate is a hardening accelerator. This is explained by the expanding effect of this salt on the fly ash from Nazarovo Power Plant. At the same time, there was no noticeable effect on strength, either from the content of calcium nitrate in the expanding additive or from the proportion of the additive itself in the binder. After 3 months of hardening, the samples did not show signs of structural damage, cracks, flaking, efflorescence, etc.
Further research on this topic could consider the sustainability aspect. It would be useful to address the environmental impact of using high-calcium fly ash and calcium nitrate, including potential leaching of nitrates. In addition, researchers can focus on the need for optimization studies on the ideal proportion of Ca(NO3)2 and fly ash to achieve the desired expansion without compromising the structural integrity of the concrete.