*5.2. Shrinkage and Drying*

One of the first preliminary observations on the influence of the drying process on the characteristics of the cracking pattern were made by Bazant et al. [110,111]. Concrete and cement pastes were the subject of research. They noticed that as a result of the drying process, a network of parallel cracks is formed, whose spacing and width of opening is proportional to the depth of penetration of drying. With typical test samples, most cracks are too thin to be observed with the naked eye. It was also found that the characteristics of the cracking pattern predominate in the direction of the development of discontinuous microcracking rather than in the direction of continuous macrocracks.

Wang et al. [106] studied plastic shrinkage behavior in cement composites with the addition of fly ashes and various types of dispersed fibers. The structure of air pores using a mercury-intrusion porosimetry and the cracking pattern using the image analysis were studied. The maximum crack opening width was measured as well as the crack area. It was observed that the addition of dispersed fibers generally increases the number of large pores in cement paste, which affects both bleeding behavior and crack distribution in the material. It was found that, depending on the type of fibers and their properties, the crack area can be reduced by 30–40%, with 0.1% (by volume) of fiber addition. It has a very big influence especially on the mechanical properties of the cement composite in the early stages of maturation.

Plastic shrinkage behavior of cement pastes with the addition of alkali-resistant glass fibers was studied by Bakhshi and Mobasher [107]. The addition of glass fibers in the amount of 0.06%, 0.11%, 0.17% and 0.23% by volume was studied. The tests were carried out in low-pressure conditions, and additionally the sample size, *w*/*c* ratio and conditions of initial care were also a variable factor. The morphology of the cracking pattern was analyzed using the image analysis. It was found that the drying process can be divided into two main stages, i.e., stage I—constant drying rate period, and stage II —falling drying rate period. The research confirmed that by using glass fibers it is possible to control the process of formation and propagation of cracks during the maturation process of the cement matrix. While it was not observed that the presence of fibers had an impact on the cracking pattern in the stage I, in stage II the presence of fibers reduced the diffusivity values. Scientists concluded that the reason for this was the bridging effect of the glass fibers. The results showed that in the case of 0.11% and 0.23% of fibers addition, the areal fraction of cracks was reduced by 22% and 61%, respectively. The maximum crack opening width was also reduced by 47% and 71%, respectively, compared to classical cement paste.

Bisschop and Wittel [8] conducted a study to evaluate the morphology of the cracking patterns in classical cement paste with the *w*/*c* = 0.5. The study determined the effect of sample thickness and methods of drying on the crack density, crack length and the crack penetration depth. The latter parameter was quantified using a fluorescent resin impregnation technique. It was found that for samples < 14 mm thick, the single-sided drying resulted in lower values of the crack penetration depth (15–20% of the sample thickness), compared to double-sided drying (20–35% of the sample thickness). In the case of samples that were dried in an Environmental Scanning Electron Microscope chamber at 25% humidity, 20 ± 0.2 ◦C, at 590 Pa, for 5 h, it was noticed that the cracking pattern was characterized by multiple crack cuts at an angle of about 120◦. Many dead-end cracks were also found. In case of drying in a climatic chamber at normal atmospheric pressure it was found that the cracks intersections resemble more a 90◦–180◦ system.

Lura et al. [113] studied the geometric features of the cracking pattern in cement paste. The observations were carried out on samples with a diameter of 10 mm, in which there was a centrally located steel rod with a diameter of 1.5 mm, 3 mm or 6 mm. The factor causing the cracking was the autogenous shrinkage of the cement paste. The crack detection was performed by impregnation with galium and analysis was performed by means of an optical and scanning electron microscopy. It was observed that as the thickness of the cement paste envelope around the steel rod increased (a rod diameter decreased), the samples showed the negative fracture geometry, which resulted in a stable crack growth. The length of the crack was greater the smaller the diameter of the rod was used.

Xuan et al. [114] examined the shrinkage cracking characteristics of concrete produced from the construction waste, mainly crushed concrete and masonry aggregates. The influence of four factors, i.e., masonry content, moisture content, degree of compaction and cement content, on the characteristics of the cracking pattern was examined. Two cracks characteristics, i.e., the crack width and the crack spacing, were studied. It was found that the crack width decreases with decreasing amount of cement, decreasing the compaction degree, increasing moisture and increasing the masonry content. In the case of the crack spacing it was found that this parameter decreases with decreasing the amount of cement, increasing the compaction degree, increasing moisture and decreasing the masonry content.
