Microstructural Features of Freezing and Thawing-Creep Damages for Concrete Mixed with Fly Ash

The macroscopic morphology characteristics, pore structure characteristics, and microfracture morphology of concrete with fly ash subjected to the freeze-thaw-creep effect were analyzed via scanning electron microscopy (SEM). The results revealed that the macrosection of a specimen subjected to freeze-thaw cycling evolves from a regular to an irregular morphology in which the degree of fragmentation increases. Four specimen pore structure types characterized by single holes, nonconnected hole clusters, connected hole clusters, and fly ash holes, respectively, were identified. The microfracture morphology of the concrete was found to include five types of brittle fractures—river, step, cascade, hemispherical, and irregular patterns—and two types of ductile fractures—dimple and peak forest patterns. Two sets of experiments in which (1) the fly ash content (m = 35%) was kept constant and the number of freeze-thaw cycles increased, and (2) the number of freeze-thaw cycles (n = 120) was kept constant, and the fly ash content was increased were carried out. In both cases, the number of connected hole clusters increased and a surrounding skeleton structure with a needle filamentous or flaky appearance was produced. In addition, the degree of deterioration of the pore structure increased and the fracture features changed from brittle to ductile.


Introduction
The deeper underground space is a complicated environment associated with high ground stress, high ground temperature, and high karst water pressure, which poses significant challenges to engineering personnel and scientific researchers [1][2][3][4][5]. Concrete is one of the most widely used building materials, where applications have extended beyond conventional buildings to underground, aquatic, and marine structures. Concrete materials can incur damage as a result of environmental factors such as high and low temperatures, high pressure, corrosion, wind load, freezing and thawing, and rain and snow. Such factors can produce deformations in engineering structures and can eventually lead to structural instability. Accordingly, the physical-mechanical response of concrete materials and structures under one or more factor conditions has become a subject of intense research in the fields of building materials and structural engineering [6,7]. Numerous studies have looked at the property changes and damage mechanisms of concretes in freezing and thawing environments [8,9], leading to a number of techniques for improving the antifreezing and thawing properties of these materials. However, many problems must still be resolved, including the physical-mechanical behaviors [10,11], damage mechanisms [12,13], and micromechanical behaviors [14,15] of different concrete materials (highstrength, high-performance, very-high performance, fiber concrete, etc.) under the coupling of freezing and thawing and other factors (load, corrosion, etc.).
With the continuous increase in energy demand, mineral resources extraction has gradually shifted to deeper strata, and the development depth of vertical shafts has increased consequently, along with the increase in the cross-sectional size of the shaft [16][17][18]. For the development of deep and thick unstable aquifer mines, the artificial ground freezing method is widely used [19][20][21]. When constructing using the freezing method, the freezing shaft lining is crucial, and its main function is to block groundwater and withstand the temporary load to ensure the safety and smooth operation of shaft sinking construction [22][23][24][25]. The structural forms of freezing shaft lining primarily include single-layer reinforced concrete, double-layer reinforced concrete, and reinforced concrete composite shaft lining. Currently, concrete materials are essential in the construction of all freezing shaft lining.
Rheology, which must always be taken into consideration in the design and construction of concrete engineering structures, is one of the more important factors determining the degree of failure and damage experienced by concrete engineering structures after long-term service. Accordingly, there has been a significant amount of research on the rheological properties of concrete materials and structures [26][27][28]. Understanding the creep deformation features and the microstructural change laws of concrete materials under freezing and thawing conditions plays a critical role in improving the freezing and thawing resistance and durability of concretes.
Analysis of the microcomponents and morphological structures of materials is the key to understanding their mechanical properties and damage mechanisms [29]. Based on preliminary research [30][31][32] and the scanning electron microscope, the overall morphology and structural features of the micropores and microtopographic features of concrete mixed with fly ash by the freezing and thawing-creep action were studied. The results constitute a foundation for understanding the micromechanical mechanisms of the damage fracturing for concrete mixed with fly ash by the freezing and thawing-creep action.  Table 1. River sand with a fineness modulus of 2.8 and an apparent density of 2,769 kg/m 3 was used as a fine aggregate; crushed limestone with a size of 5-20 mm and an apparent density of 2,719 kg/m 3 was used as a coarse aggregate. Carboxylic acid provided by Zhejiang Wulong Chemical Co. Ltd was used as a slushing agent for the concrete. The mix proportions used in the concrete are summarized in Table 2.

Test Method.
Concrete specimens were subjected to freezing and thawing-creep deformation testing. Each specimen had the dimensions 100 × 100 × 300 mm 3 and had been subjected to curing for more than 90 d. There were three specimens in each group. The uniaxial compressive strength of cube concrete specimens (100 × 100 × 100 mm 3 ) with different fly ash replacement ratios were 58.1 MPa (m = 0%), After each set of 30 rapid freezing-and-thawing cycles (GBT50082-2009 China), the specimens were subjected to creep deformation testing. A Changchun Kexin YAS-5000 microprocessor-controlled electro-hydraulic servo compression-testing machine was used to carry out creep deformation tests under progressive loading. Each specimen was first loaded to 30% of its strength value at a loading rate of 2 kN/s, after which the loading was maintained at the strain stability state. Following this, the load value was progressively increased by 10% of the uniaxial compressive strength, with the loading maintained at the strain stability state, until the specimen was damaged. The testing units used for the creep deformation testing are shown in Figure 1.
SEM testing was then applied to the concrete specimen fragments. The testing units for the SEM analysis are shown in Figure 2. The serial no. of the SEM test system is TESCAN VGEA3 provided by the Civil Structure Laboratory of Xuzhou Institute of Engineering. Figure 3 shows the macrofracture features of concrete specimens with fly ash replacements of m = 0, 20, 35, and 50% and following n = 0, 30, 60, 90, and 120 freezing-and-thawing cycles (magnification 15x).

Macro Features of Creep Fractures of Fly Ash Concretes after Freezing and Thawing.
The macrofractures of the specimens evolved from a smooth to a concave-convex irregular morphology as the   cycles of freezing and thawing; after n = 60, 90, and 120 cycles, however, the specimen developed an unsmooth fracture surface with unfavorable integrity. The postfreezing-and-thawing fly ash concrete specimens experienced more deterioration in fracture surface smoothness and integrity than did the specimens with no

Geofluids
fly ash, with the degree of deterioration increasing with the percentage of fly ash replacement. For example, the m = 0 and 20% specimens had relatively smooth fracture surfaces with favorable integrity after n = 30 cycles (Figures 3(b) and 3(g), respectively), whereas the m = 35% specimen had a relatively rougher fracture surface and unfavorable integrity (Figure 3(l)) and the m = 50% specimen had an irregular fracture surface, large amounts of fracturing, and even worse integrity (Figure 3(q)).

Pore Structure Features of Fly Ash Concrete after Freezing
and Thawing-Creep Deformation. The size, quantity, and distribution of pores in a concrete have a direct effect on its physical and mechanical properties. Previous research has shown that the pore structure features of a concrete are closely related to its antifrost properties [33]. Based on SEM observation and analysis of the pore structure features of the fly ash concrete specimens following freezing and thawing and creep deformation, the following four types of pore structure were identified: Single-pore: in these structures, individual air bubbles remained in the cement paste and failed to escape during the concrete hardening process, resulting in the formation of single pores that were distributed within the specimen in a dispersive manner, as shown in Figure 4(a).
Disconnected pore clusters: in these structures, multiple concentrations of air bubbles remained in the cement paste during the concrete hardening process (alternatively, large air bubbles were deformed or divided as a result of extrusion of cement paste), resulting in the formation of pore clusters, as shown in Figure 4(b).
Connected pore clusters: during the concrete hydration process, products can fill and obstruct large pores, resulting   4 Geofluids in the formation of connected pore clusters. Alternatively, disconnected pore clusters can incur damage to their pore walls as a result of external environmental effects such as freezing and thawing or corrosion, resulting in the formation of connected pore clusters, as shown in Figure 4(c). The presence of honeycomb-type connected pore clusters has a detrimental effect on the mechanical properties of concrete materials. Fly ash (FA) pores: although most fly ash ball particles are solid particles, some are hollow or semi-hollow. Pores can be generated when the surfaces of such ball particles are fractured, as shown in Figure 4(d).
To further analyze the influence of freezing and thawing on concrete pore structure features, SEM images of concretes (m = 35%) following different numbers of freezing-andthawing cycles were analyzed ( Figure 5, magnification 5000x). The number of harmful pores with diameters in the range of 200-5,000 nm increased with the number of freezing-and-thawing cycles, leading to loosened frameworks and degraded pore structures. At n = 0, there were fewer pores surrounded by a compact framework structure, smoother pore walls, and a pattern of independent pores. After n = 60 and 90 cycles, the number of pores increased significantly, the pore walls became extremely irregular, and the pores became mutually connected; in the process, a needlesilk framework developed around the pores, and the pore clusters adopted a sponge-type framework. After n = 120 cycles, the number of small pores increased significantly and the pore clusters and framework assumed a rough and loose composition. Figure 6 shows changes in the pore structure features with increasing fly ash replacement (n = 120). It is seen that increasing the replacement roughened the fracture surface structure and increased the number of pores and filaments. Relative to concrete specimens with no fly ash, the specimens with fly ash developed a more unfavorable pore structure and experienced a higher degree of pore structure deterioration. Step pattern: crystals with complicated components can be ruptured, torn, or cut off under external forces, resulting in the formation of step patterns, as shown in Figure 7(b).
Irregular pattern: the weak surfaces of a concrete can generate fractures and damage under external force, resulting in the formation of irregular shapes and patterns, as shown in Figure 7(c). With their intergranular fracture features, these types of fracture are of a typical brittle type.
Semispherical surface pattern: this fracture pattern is rarely observed in rocks. In cases in which the weak interface between a spherical fine aggregate and cement jelling material is tensioned or sheared under an external force, fractures and separation of fine aggregates can occur, leaving a semispherical pattern, as shown in Figure 7(d). In this case, the spherical surface will be smooth and a typical brittle fracture will occur.

Tough
Fractures. Tough fractures, which are generated as a result of plastic deformation, have irregular fracture surfaces. Two tough fracture types-dimple and fungling patterns-were identified in the SEM images and then analyzed.
(1) Dimple Pattern. Figure 8(a) shows an SEM image of the topographical features of a dimple fracture surface. On this  (2) Fungling Pattern. This topographic feature is widely observed in fly ash concrete specimens after freezing and thawing. Fungling fractures are formed through a mechanism similar to that causing dimples via the tearing of pores as a result of external force. However, fungling patterns are often formed in regions with higher pore densities in which the pore or pore cluster walls are relatively thin, enabling plastic deformation of the pore walls under the effect of ten-sile stress. The walls are eventually pulled out, forming a fungling pattern (Figure 8(b)).
To further analyze the influence of freezing and thawing on the topographic features of concrete fractures, SEM images of concretes (m = 35%) subjected to increasing numbers of freezing-and-thawing cycles were analyzed (Figure 9). At n = 0, (Figure 9(a)), the fractures featured hidden river patterns that were distributed with irregular short and small tearing edges, indicating a change in the concrete damage features from brittle to ductile fractures. After n = 60, 90, and 120 cycles, as shown in Figures 9(c)-9(e), respectively, fungling patterns developed on the fractures (ductile fracture    Geofluids feature). The fungling patterns became denser as the number of freezing-and-thawing cycles increased. Figure 10 shows the changes in the fracture topographic features as the amount of fly ash replacement was increased (n = 120). At m = 0% (Figure 10(a)), the intergranular fracturing of the semispherical patterns was uniform. At m = 20% (Figure 10  Step pattern River pattern

Geofluids
in the fractures. These results indicate that increasing the fly ash replacement in a concrete will change its postfreezing-and-thawing microfracture features from brittle to ductile.

Conclusions
The results of this study suggest that, as the number of freezing-and-thawing cycles increases, the macrofractures of a concrete specimen will evolve from a smooth to an irregular morphology in which there are more cracks on the fractures. In this process, the specimens will be gradually divided into many small regions surrounded by cracks. As the fly ash replacement is increased, the specimen will develop an improved fracture degree.
In our specimens, we identified four types of concrete pore structure-single-pore, disconnected pore clusters, connected pore clusters, and fly ash pores. At a constant fly ash replacement (m = 35%), increasing the number of freezing-and-thawing cycles reduced the number of disconnected pore clusters, while increasing the number of connected clusters produced a needle-silk or flower piece framework around the pores with an enhanced degree of pore structure deterioration. The same results were obtained by increasing the fly ash replacement and subjecting the samples to a constant number of freezingand-thawing cycles (n = 120).
The microtopographic features of concrete fractures are primarily brittle (river, step, stacking, semispherical surface, or irregular patterns) or tough (dimple or fungling patterns). With the fly ash replacement held constant (m = 35%), the concrete fracture features changed from brittle to ductile as the number of freezing-and-thawing cycles increased; the same results were obtained when the fly ash replacement was increased, while the number of freezing-and-thawing cycles remained constant (n = 120).

Data Availability
All data, models, and code generated or used during the study appear in the submitted article.

Conflicts of Interest
The authors declare that there is no conflict of interest regarding the publication of this paper.