Effect of aggregate characteristics on properties of cemented sand and gravel

1 Introduction

Cemented sand and gravel (CSG) is a new type of dam building material; it is formed by mixing, spreading, and vibrating a small amount of cemented materials (cement, fly ash) with non-screened and non-washed sand and gravel materials on the construction site [1]. Compared with concrete, the amount of adhesive used in CSG is only one-third of that of concrete; thus, CSG has different material properties from ordinary concrete [2]. Therefore, different studies should be conducted on them. In addition, CSG dams formed by paving and rolling CSG materials, which are intermediate between gravity dams and earth-rock dams, have now been applied worldwide. It mainly uses riverbed sand and gravel or artificial crushed stones as coarse aggregates; among them, natural sand and stone materials have a more rounded particle shape, while artificial crushed stone materials have more edges and corners due to the differences in performance between natural sand and artificial crushed stone materials, it is necessary to reasonably mix aggregates when designing the mix proportion of CSG. Therefore, it is necessary to quantify the impact of aggregate characteristics on the performance of CSG materials.

As an important component of concrete, aggregates play a skeleton role, and their characteristics have an impact on the strength and durability of concrete materials [3]. The basic characteristics of aggregates include mechanical properties, physical properties, and morphological characteristics, among which the morphological characteristics of aggregates directly affect the arrangement, stacking, and contact forms of aggregate particles [4]. Huang [5] proposed from an experimental perspective that only when the proportion of coarse aggregates in the sample reaches a certain amount, it will have an impact on the mechanical properties. Ye et al. [6] found that the complexity of the shape of coarse aggregate has a significant impact on the final compressive destructive morphology of the sample but has no significant impact on the tensile strength of the material. Zhao et al. [7] prepared concrete using three types of coarse aggregates with elliptical, layered, and irregular shapes and studied the mechanical properties of concrete containing different forms of coarse aggregates. StrzałKowski and Garbalińska [8] believe that crushed basalt aggregates can cause significant aeration in the concrete, and the compressive strength of concrete made from crushed aggregates is significantly lower than that made from natural circular gravel aggregates. Li and Liu [9] studied the transport law of chloride ions inside concrete under freeze–thaw cycles using a circular aggregate model. Jia’s [10] research shows that aggregate morphological characteristics have a significant impact on the generation and development of concrete cracks, and a polygonal mesoscopic model of recycled concrete has been established to simulate the actual frost damage of recycled concrete. De Larrard et al. [11] used two types of aggregates with different shape characteristics to prepare concrete and studied the impact of aggregate characteristics on concrete carbonation. The study showed that aggregate shape has a significant impact on the variation coefficient of carbonation depth. Lee et al. [12] used roundness to quantify the angularity of coarse aggregate and studied the impact of the angularity of coarse aggregate on the mechanical properties of concrete based on crushed stone and pebble materials. The study showed that the compressive strength and tensile strength of crushed stone coarse aggregate are higher than those of pebble coarse aggregate, while the elastic modulus of pebble coarse aggregate is higher than that of crushed stone coarse aggregate.

Most studies mainly focus on the influence of aggregate shape on the mechanical properties of ordinary concrete materials, but there is little research on other characteristics in the field of CSG materials. Therefore, this article studies the influence of aggregate shape characteristics on the frost resistance of CSG and takes aggregate gradation, fractal dimension, and aggregate shape characteristic value as quantitative analysis factors by means of numerical simulation. The influence of CSG on the strength of materials properties and the material degradation mechanism is explored, providing a reference for the engineering application of CSG.

2 Finite-element random aggregate model

In this article, the generation process of random aggregates is performed using the Monte Carlo method [13,14], that is, generating a group of random variables that exhibit uniform distribution on the range [ 0 , 1 ] . Let the probability density function of X be

Random variables in other arbitrary ranges can be obtained based on random variable transformations in the [ 0 , 1 ] range. For example, a uniformly distributed random variable Y within an arbitrary [ a , b ] range can be obtained by Y = a + ( b − a ) X . Thus, random variables that satisfy uniform distribution across various ranges can be generated.

From a mesoscopic perspective, the CSG material is considered a three-phase composite material composed of coarse aggregate, mortar matrix, and the interface transition zone between coarse aggregate and mortar matrix. At the same time, on a two-dimensional plane, it is assumed that artificial crushed stone aggregates are treated as irregular polygons and natural sand gravel aggregates are treated as circles. In this study, by generating aggregate random circles and adhesive random circles with boundaries, the inner and outer circles were divided into quadrants. The number of corner points for each quadrant was determined, and the corner point coordinates were formed. Finally, the corner points were connected to generate polygons. [15].

The physical test sample adopts a size of 100 mm × 100 mm × 100 mm CSG cube test block, and the established numerical simulation model has a size of 100 mm × 100 mm. The particle size range of aggregates is 5–32.5 mm, and the number of aggregates within each particle size range is determined [16,17]. First, three-dimensional aggregates are generated based on the obtained three-dimensional grading curve to determine the distribution of aggregates. Then, the plane is intercepted from the three-dimensional model, and the total aggregate area of each particle size range is counted on the plane so as to obtain the two-dimensional aggregate quality cumulative distribution function. On this basis, a two-dimensional aggregate particle number that meets a specific grading is generated. The specific process is shown in Figure 1.

Figure 1

Method for determining the two-dimensional aggregate quality accumulation distribution function under arbitrary gradation.

The number of aggregates is calculated in each particle size range using the following formula:

where A gg is the aggregate area with a particle size of [ D s , D s + 1 ] ( mm 2 ) , P 2 A is the aggregate accumulation distribution function of a certain particle size, D max is the maximum value of aggregate particle size ( mm ) , D m in is the minimum value of aggregate particle size ( mm ) , R agg is the area ratio of coarse aggregate, typically taken as 0.35–0.45, and A con is the area of concrete ( mm 2 ) .

The influence of aggregates on the mesoscopic parameters of CSG is temporarily not considered during the numerical simulation in this article. The mesoscopic material model and parameters are selected in reference to the literature [18].

3 Effect of aggregates on mechanical properties of CSG

Using aggregate shape as the control variable, the influence of different aggregate characteristics on the mechanical properties of CSG is explored. Taking an aggregate content of 40% as an example, the aggregate is compounded according to the standard fuller grading, and a random aggregate model is established with different aggregate shapes, as shown in Figure 2.

Figure 2

Random aggregate model considering the aggregate shape: (a) LC1, (b) LC2, (c) LC3, (d) LC4, and (e) LC5.

In this article, the fractal dimension [19] is selected to characterize the irregularity of aggregate shape. For CSG, the shape of the pebble is smooth and regular, while the cross-section of the gravel is usually complex and tortuous. The shape characteristics of pebbles and crushed stones have significant differences, so box dimension is used to characterize them. Let F be any non-empty bounded subset on R , N δ be the size δ the minimum number of boxes that can cover the set of F , then the box dimension D x can be calculated by the following formula [20]:

where δ represents the measurement scale of units in continuous distribution, and N δ indicates the number of measurement scales, that is, the number of small squares covering the aggregate.

In order to quantitatively explore the differences in shape between different aggregates, feature parameter models for composite regular circular and irregular polygonal aggregates are established using the entropy weight method, and the complexity of the aggregate shape is evaluated. The principle of the entropy weight method is to determine the weight value of indicators based on the amount of information provided by the observed values of each indicator, and the entropy weight method can objectively calculate the weights of various indicators based on information entropy theory [21].

The steps to determine the index weight coefficient using the entropy weight method are as follows:

1. Normalize x ij

   (4) P ij = x ij / ∑ i = 1 n x ij .

2. Calculate the entropy of the j th index

   (5) e j = − k ∑ i = 1 n P ij ln ( P ij ) .

   In the equation, k > 0 , e j > 0 , where k = 1 / ln n

3. Calculate the coefficient of difference for indicator x ij

   (6) g j = 1 − e j .

4. Determine weight

According to the aforementioned calculation steps, the weight coefficient of regular circular aggregates is obtained as ω 1 , and the weight coefficient of irregular polygonal aggregates is ω 2 , and on this basis, comprehensively the complex aggregate shape characteristic values are obtained considering the shape φ . The calculation formula is as follows:

where D x 1 is the fractal dimension of circular aggregate, D x 2 is the fractal dimension of irregular polygon aggregate, ω 1 is the weight coefficient of circular aggregate, and ω 2 is the weight coefficient of irregular polygonal aggregates.

The fractal dimension of circular aggregate and polygon aggregate in LC1–LC5 is calculated according to formula (3), and the organized calculation results are shown in Table 1.

Based on the principle of the entropy weight method and combined with formula (3), the weight coefficients of regular circular aggregates and irregular polygonal aggregates are calculated. The weight coefficient of regular circular aggregates ω 1 is 0.5123, and the weight coefficient of irregular polygonal aggregates ω 2 is 0.4877. According to formula (8), the shape characteristic values of CSG aggregates can be calculated φ based on a random aggregate model. Uniaxial compression numerical simulation tests were conducted to obtain mechanical performance data of CSG. The test results and aggregate characteristic values are organized as shown in Figure 3.

Figure 3

Influence relationship between aggregate shape characteristic values and mechanical properties of CSG. (a) Influence relationship between aggregate shape characteristic value and peak stress of CSG, and (b) influence relationship between aggregate shape characteristic value and elastic modulus of CSG.

As can be seen from Figure 3, with the increase of regular circular natural sand gravel aggregates, the aggregate shape characteristic value increases, and its peak stress and elastic modulus also increase; this result is consistent with the experimental research results in the literature [22,23,24]. Among them, LC1 is made of 100% irregular polygonal artificial crushed stone aggregate, and LC5 is made of 100% circular natural sand gravel aggregate. The above experimental group showed significant differences in mechanical properties, and the main reasons for this phenomenon are analyzed as follows: The first reason is the corner effect of the aggregate. For irregular polygonal aggregates, the ITZ unit area around the aggregate is greater than that of regular circular aggregates with the same particle size. At the same stress level, there will be more irregular polygonal aggregate ITZ units around the sample that exhibit destruction and fracture. The second reason is due to the influence of stress concentration. For irregular polygonal aggregates, their peripheral stress distribution is relatively more concentrated. For regular circular aggregates, their rounded appearance can effectively reduce the stress concentration phenomenon of CSG.

4 Effect of aggregate characteristics on frost resistance of CSG

To further analyze the impact of aggregate shape on frost resistance, 0, 5, and 10 times freeze–thaw cycle tests were designed for comparative analysis. During the freeze–thaw cycle test, the area of the failed unit reflects the quality loss of the sample; meanwhile, the larger the area of the failed unit, the greater the strength loss.

4.1 Failure characteristics during unfreezing–thawing

The mesoscopic component damage of unfreezing–thawing samples at different displacement levels is shown in Figures 4 and 5.

Figure 4

Loading failure characteristics of polygonal aggregate: (a) loading 0.93 mm starts to failure, (b) loading 1.53 mm failure characteristics, (c) loading 1.83 mm failure characteristics, and (d) loading 2.43 mm failure characteristics.

Figure 5

Loading failure characteristics of circular aggregate: (a) loading 0.93 mm starts to failure, (b) loading 1.53 mm failure characteristics, (c) loading 1.83 mm failure characteristics, and (d) loading 2.43 mm failure characteristics.

From the above figure, it can be seen that: As the strain develops, each group of sample transitions from the linear elastic stage to the softening stage, the tearing effect between different materials is more pronounced, and the deformation disharmony increases. The interface with lower tensile strength prioritizes damage after the interface material is damaged, the stress effect on the interface further weakens, and the mortar unit begins to fail. As the mortar is extensively damaged, stress concentration is further strengthened, and some aggregates are damaged. In addition, compared to circular aggregates, polygonal aggregates have more edges and corners, which exacerbates the deformation disharmony between mortar and aggregates, resulting in more pronounced stress concentration. From this, it can be seen that under the same loading step size, polygonal aggregates have more failure units than circular aggregates.

4.2 Failure characteristics after freezing and thawing

4.2.1 Failure characteristics of CSG after five freeze–thaws

Compared to the unfreezing–thawing samples of CSG, the freeze–thaw damage characteristics of the CSG samples after five freeze–thaw cycles have changed: (1) The overall damage characteristics are the same as those of the unfreezing–thawing samples, both of which first generate interface micro-cracks, then extend to the mortar, and then partially damage the aggregate. (2) Due to the effect of freeze–thaw cycles, the parameters of the micro-component materials deteriorate, and the decrease in elastic modulus leads to a decrease in stress at the same displacement. Therefore, the loading displacement at the beginning of damage increases compared to unfreezing–thawing. (3) Due to the deterioration of material parameters, the interface failure units of CSG after freezing and thawing increase, and the corresponding aggregate damage caused by stress concentration decreases (Figures 6 and 7).

Figure 6

Loading failure characteristics of polygonal aggregate samples after freezing and thawing: (a) loading 1.23 mm starts to failure, (b) loading 1.53 mm failure characteristics, (c) loading 1.83 mm failure characteristics, and (d) loading 2.43 mm failure characteristics.

Figure 7

Loading failure characteristics of circular aggregate samples after freezing and thawing: (a) loading 1.23 mm starts to failure, (b) loading 1.53 mm failure characteristics, (c) loading 1.83 mm failure characteristics, and (d) loading 2.43 mm failure characteristics.

4.2.2 Failure characteristics of CSG after 10 freeze–thaws

After ten freeze–thaw cycles, the freeze–thaw damage characteristics of the CSG sample further changed: (1) The overall damage characteristics are consistent with the above. First, micro-cracks are generated at the interface, which then extends to the mortar, and finally, some aggregates are damaged. (2) Due to the effect of freeze–thaw cycles, the parameters of the micro-component materials deteriorate, and the decrease in elastic modulus leads to a decrease in stress at the same displacement. Therefore, the loading displacement at the beginning of damage increases compared to unfreezing–thawing. However, under the same loading displacement, after ten freeze–thaw cycles, the CSG increases in stress concentration in the remaining units due to the increase in failure units; therefore, the damage is more severe. (3) After the freeze–thaw cycle, the degree of damage of irregular polygonal aggregate samples is more severe than that of circular aggregate CSG samples, and this phenomenon is similar to unfreezing–thawing. Therefore, when designing the mix ratio of CSG, the aggregate grading and shape should be further optimized based on the actual situation of the engineering site (Figures 8 and 9).

Figure 8

Loading failure characteristics of circular aggregate samples after freezing and thawing: (a) loading 1.23 mm starts to failure, (b) loading 1.53 mm failure characteristics, (c) loading 1.83 mm failure characteristics, and (d) loading 2.43 mm failure characteristics.

Figure 9

Loading failure characteristics of polygonal aggregate samples after freezing and thawing: (a) loading 1.23 mm starts to failure, (b) loading 1.53 mm failure characteristics, (c) loading 1.83 mm failure characteristics, and (d) loading 2.43 mm failure characteristics.

The strength of concrete decreases with the increase of freeze–thaw cycles, with the most prominent being tensile strength and flexural strength. That is, with the increase of freeze–thaw cycles, the tensile strength and flexural strength of concrete decrease rapidly, while the compressive strength decreases more slowly [25]. Similarly, as the number of freeze–thaw cycles of CSG materials increases, their compressive strength gradually decreases. The process of freezing and thawing of pore water inside CSG can be seen as a continuous process of loading and unloading the pore wall. The small pores inside gradually increase, and some of them transform into large pores. There is a lot of gravity water inside the large pores, which first breaks down during the freezing process, leading to a gradual increase in frost heave. In addition, due to the relatively small amount of cementitious material used in CSG compared to concrete, the failure of CSG in mortar under freeze–thaw conditions occurs faster and more completely, resulting in faster material failure.

5 Conclusion

In areas where natural aggregates are scarce, artificial crushed materials need to be supplemented to meet the requirements of CSG dam construction. This article considers the differences between natural and artificial aggregates, quantifies the characteristics of aggregates, and designs uniaxial compression and rapid freeze–thaw test plans for different aggregate characteristics, obtains the stress characteristics (strength, stress–strain curve) and durability indicators (strength loss, quality loss) of CSG poured with different groups of circular–polygonal blended aggregates. The damage characteristics of CSG materials were revealed from the perspective of aggregate shape, providing reference significance for the application of CSG materials in cold regions. The main conclusions obtained are as follows:

1. From the perspective of the shape characteristics of coarse aggregate, due to the large number of edges and corners of polygonal aggregates, the deformation of mortar and aggregates becomes increasingly uncoordinated, and the phenomenon of stress concentration is obvious; therefore, the mechanical properties are poor. As the proportion of circular aggregates increases, the peak stress and elastic modulus of CSG increase, indicating that adding circular aggregates will improve the overall mechanical properties of CSG materials.

2. Through freeze–thaw cycling tests, it was found that cracks often initiate from the interface transition zone and then extend to the mortar, and finally, the aggregate undergoes damage. Moreover, the quality and strength losses of crushed stone aggregates are greater than those of circular aggregates, indicating that the addition of crushed stone aggregates is detrimental to the frost resistance of CSG.

3. When designing the mix ratio of CSG, the proportion of circular and polygonal aggregates should be optimized based on the actual situation of the engineering site to obtain better application performance.