Research on the impact of waste rubber particles and polyvinyl alcohol (PVA) fibers on the static mechanical properties and impact resistance of concrete

2.3.1. Methods for testing compressive, tensile, and flexural strengths

The test was carried out using a WDW-300KN type electro-hydraulic servo testing machine. According to the requirements of GB/T 50081-2019 (3737. GB/T 50081–2019. 2019. Standard for test methods of physical and mechanical properties of concrete. Chinese Standard.), each group of 3 specimens, adjusts the loading speed on the control computer and tests Cubic compressive strength and axial compressive strength at a loading speed of 6.0 kN/s and flexural strength at 0.5 kN/s, and test split tensile strength by controlling it within the range of 0.05~0.08 MPa/s range. Throughout the test, it is of the utmost importance to maintain vigilance concerning the overall shape of the specimen. In the event of damage occurring, it is essential to make the necessary adjustments to the instrument on time and to duly record the peak load.

2.3.2. Method to test impact resistance

As illustrated in Figure 2, the specimens were impact tested using a self-designed drop hammer impact test set-up by the American Concrete Institute’s (ACI 544 (3838. American Concrete Institute (ACI). 1996. ACI 544: Fiber reinforced concrete. Farmington Hills (MI): American Concrete Institute.)) recommended concrete drop hammer impact test method. The test beam was simply supported at both ends, with an effective span of 300 mm. Before the test, the specimen was placed on the impact test device without eccentricity, and a steel pad was placed at the center of the span on the upper surface of the specimen to moderate the degree of stress concentration on the impact surface. During the test, after the completion of each impact loading, the surface cracking and damage of the specimen were observed. When the first crack appears in the specimen, record the number of impacts at this moment, i.e., the number of impacts of the initial crack ; when the main crack runs through the entire cross-section of the specimen, it is defined as the destruction of the specimen, and record the number of impacts at this moment as the number of destructions . In this paper, we use the number of impacts , the energy absorbed by impact , and the ductility ratio to express the impact resistance of the concrete and the ductility indexes, which are respectively calculated according to Equations [1] and [2].

Where W_i — he maximum impact energy absorbed by the first and last cracks of different sample groups (J); N_i — number of impacts, divided into N₁ (number of initial cracking impacts) and N₂ (number of final cracking impacts); m — mass of the impact ball, 4.5 kg; g — acceleration of gravity, 9.8 m/s²; h — impact height, 457 mm.

2.3.3. Microstructure characterization

A scanning electron microscope model S-3000N manufactured by HITACHI, Japan, was used in this experiment. The scanning electron microscope was used to observe microscopic features such as hydration products, microcracks, and interfacial transition zones in fiber-rubber concrete and to quantify the distribution of microcracks in the interfacial transition zone of the fiber-rubber matrix. The specimens were approximately 10 mm × 10 mm in size and then placed in anhydrous ethanol to prevent continued hydration of the cement stone. Before testing, the specimens were removed from the anhydrous ethanol and dried, placed on an SEM specimen carrier metal sheet coated with conductive adhesive, and then gold-plated using an ion sputtering device.

3.1.1. Compressive strength

Table 5 and Figure 3a illustrate the mean compressive strength of the specimens at 28 days. The cubic compressive strength and axial compressive strength of PFC0 exhibited a gradual decline with the increase in rubber substitution rate. The cubic compressive strength of PFC0, PFC0.1, and PFC0.2 exhibited a decline of 4.51%, 11.81%, and 22.52%, respectively, at 1%, 3%, and 5% substitution of rubber particles in comparison to NC. Similarly, the axial compressive strength exhibits a decrease of 2.35%, 5.41%, and 9.41% at the same level of substitution. As the percentage of rubber substitution increased from 0% to 5%, the concrete cubic compressive strength and axial compressive strength of PFC0.15 samples exhibited a decline of 8.50%, 14.05%, and 28.80%, respectively. Axial compressive strength showed a similar downward trend, with a reduction of 9.64%, 5.16%, and 10.76%, respectively. Prior studies have also mentioned this phenomenon (3939. Uygunoğlu T, Topcu IB. 2010. The role of scrap rubber particles on the drying shrinkage and mechanical properties of self-consolidating mortars. Constr. Build. Mater. 24(7):1141-1150. https://doi.org/10.1016/j.conbuildmat.2009.12.027). Given that the strength of rubber is inferior to that of fine aggregate and that rubber is an organic polymer with elastic properties, whereas the cement matrix is an inorganic material, the properties of the two substances are inherently disparate. The contact surface between rubber and the cementitious matrix is characterized by numerous weak points, which result in relatively poor adhesion and a reduced load-bearing capacity.

As the quantity of PVA fibers increases, the cubic compressive strength and axial compressive strength demonstrate a pattern of initial growth followed by a decline. As the fiber content is increased from 0% to 0.2%, the cubic compressive strength of PFC0, PFC0.1, and PFC0.2 demonstrates an increase of 2.28%, 3.48%, and 2.76%, respectively, in comparison to NC. Similarly, the axial compressive strength exhibited an increase of 2.59%, 4.94%, and 1.41%, respectively. When the specimen is compressed, the bridging effect of PVA fibers allows them to withstand the compressive stress associated with cracking development, reducing lateral deformation and increasing compressive strength. PVA fibers are spread randomly and systematically across the cement paste and aggregates, providing a spatial network that prevents cracking. PVA fibers fill the matrix’s major pores, when combined with aggregates, improving compressive strength. Furthermore, excessive fiber diversity could cause flaws in the matrix by causing fiber grouping. This occurrence has resulted in a decrease in the performance of the PFC0.20 sample compared to the PFC0.15 sample.

3.1.2. Flexural strength

The flexural strength results are demonstrated in Table 5 and Figure 3b. With increasing PVA fiber doping, the flexural strength of the samples increases and then decreases. The maximum strength of 6.75 MPa is obtained for PFC0.15; the strengths of PFC0.10, PFC0.15, and PFC0.20 are increased by 1.8%, 2.9%, and 1%, respectively, as compared to that of NC. The PVA fibers have a strong holding power with cement matrix, good affinity to concrete material, and good dispersion. Therefore, it can be connected to fill the pore cracks, uniformly distributed in the concrete and the common force, while the three-dimensional distribution of fibers changes the crack expansion path so that the specimen’s internal tensile stress redistribution, thereby improving the specimen’s flexural capacity.

As the rate of rubber aggregate substitution increased, a corresponding decrease in flexural strength was observed. The flexural strength of PFC0 concrete exhibited a decline of 1.22%, 4.12%, and 8.84%, respectively, in comparison to NC when the rubber admixture was increased from 0 to 5%. The flexural strength of the PFC0.15 specimens exhibited a notable decline when the rubber mixing amount was increased from 0 to 5%. This decline was observed in all three specimens, with the PFC0.15-R1, PFC0.15-R3, and PFC0.15-R5 specimens displaying a reduction of 13.78%, 3.41%, and 6.96%, respectively. The air-entraining effect of rubber results in an increase in air content within the specimen with an increase in rubber content (4040. Mohtasham Moein M, Saradar A, Rahmati K, Arman Shirkouh H, Sadrinejad I, Aramali V, Karakouzian M. 2022. Investigation of impact resistance of high-strength portland cement concrete containing steel fibers. Mater. 15(20):7157. https://doi.org/10.3390/ma15207157). This reduction in compactness, coupled with the low strength of the rubber material itself, leads to a reduction in bearing capacity and bonding ability with cement mortar. The presence of weak surfaces between the rubber and cement mortar also contributes to a reduction in flexural strength. The findings indicate that incorporating an optimal dosage of PVA fibers into rubber concrete can effectively mitigate the detrimental impact of rubber particles on the flexural strength of concrete.

3.1.3. Splitting tensile strength

Table 5 and Figure 3c illustrates that the splitting tensile strengths of the specimens exhibit a gradual decline with an increase in the rubber substitution rate. In comparison to the NC specimens, the split tensile strengths of the PFC0-R1, PFC0-R3, and PFC0-R5 specimens exhibited a reduction of 3%, 7.71%, and 14.2%, respectively, at rubber substitution rates of 1%, 3%, and 5%. The reduction in specimen strength was more pronounced when the rubber substitution rate was 5%. The observation of the PFC0.15 specimens revealed a decrease in split tensile strength, with the PFC0.15-R1, PFC0.15-R3, and PFC0.15-R5 specimens exhibiting a reduction of 17.79%, 16.73%, and 18.86%, respectively, as the rubber component grew from 0 to 5%. This is due to the hydrophobic nature of rubber, which impairs its ability to combine with other materials during the mixing process. The resulting contact surface bonding performance is inadequate, leading to a notable reduction in strength with an increase in rubber substitution rate.

As the content of PVA fibers incorporated into the specimen increases, the splitting tensile strength exhibits a trend of initial growth followed by a subsequent decline. When the PVA fiber doping level is 0.15%, the splitting tensile strength of the PFC0.15 specimen is 5.62 MPa, which is 12.28% higher than that of the NC specimen. Conversely, the split tensile strength of the PFC0.20 specimen was observed to be 4.45% lower than that of the PFC0.15 specimen. This is because the fiber can serve as a bridging role, thereby inhibiting the development of cracks when they occur. Nevertheless, when the PVA fibers are excessive, they can readily interlock with one another within the concrete, which ultimately results in a decline in strength.

3.2.1. Failure modes

Figure 4 illustrates the damage patterns of the samples after impact. As illustrated in Figure 4a, there is only one radial serrated main crack on the surface of the NC specimen. This crack quickly penetrated through the specimen after its appearance, splitting the specimen into two halves, which is a clear case of brittle impact damage. Figure 4b PFC0-R3 sample, although radial cracks also appeared after impact, the initial crack impact count was significantly higher than that of NC. This indicates that the rubber’s deformation resistance played a role before the sample cracked. However, since rubber is a hydrophobic material, its bonding with cementitious stone is poor, and when cracks occur in the sample, they will quickly penetrate through. Figure 4c PFC0.15-R3 specimen: due to the combined action of PVA fibers and rubber particles, the initial crack is narrow and weak. As the number of impacts increases, the main crack begins to extend radially until it penetrates the entire specimen. PVA fibers can bridge the specimens between cracks and prevent the formation of cracks.

3.2.2. Impact resistance test findings

The impact test results and impact ductility ratios of fiber concrete are presented in Table 6. Table 6 display that the fiber-rubber concrete holds greater initial cracking impact resistance, ultimately cracking impact resistance and ductility ratios than NC.

Figures 5-7 show the relevant impact performance metrics, including the number of impacts (N₁, N₂), energy absorbed by initial and final cracking impact (W₁, W₂), toughness ratio (μ), and number of post-cracking impacts (N₂-N₁). The addition of PVA fibers improves the impact performance of PFC. The rubber and PVA fibers mixtures produced a synergistic effect, the incremental impact energy consumption of the mixtures was higher than that of the single mixtures, and the final cracking impact energy consumption of the specimens of group PFC0.20-R5 increased by 40.5% compared to the specimens of group PFC0.20, and 68.75% compared to that of the specimens of group PFC0-R5, which showed a positive mixing effect. The reason for this is that the PVA fibers themselves have a high modulus of elasticity, which enhances the impact resistance of the concrete and increases the number of impact cracks. Rubber powder can fill the tiny pores in the concrete, increase the compactness of concrete, rubber, and cement matrix synergistic resistance to impact loading, but also give full play to its elasticity, improve the deformation capacity of concrete after impact cracking, enhance the ductility and toughness of concrete, making it more resistant to impact. The N₁ and N₂ of PFC0.20-R5 group specimens were the highest, 69 and 72, respectively, which were 154.3% and 157.1% higher than those of the NC group, respectively. μ of PFC0.15-R3 group specimens was the largest.

Looking at the impact of PVA fibers alone, fiber rubber concrete N₁ and N₂, μ with the increase in fiber mixing, 0.15% is the optimal mixing amount. The appropriate amount of PVA fibers have a good bond with concrete and their high tensile strength, consumes impact energy through bridging deformation and improves impact resistance.

3.3.1. Theory

From the experimental data in the previous section, it is known that the number of shocks is discrete and random, which is likewise confirmed by many researchers (4141. Rahmani T, Kiani B, Shekarchi M, Safari A. 2012. Statistical and experimental analysis on the behavior of fiber reinforced concretes subjected to drop weight test. Constr. Build. Mater. 37:360-369. https://doi.org/10.1016/j.conbuildmat.2012.07.068, 4242. Ostle B, Tumer KV, Hicks CR. 1996. Engineering statistics: the industrial experience. New York: Duxbury Press.). To study the damage law of the impact resistance test more objectively, finding a probability distribution model applicable to the number of impacts has become a research focus. After previous studies, it was found that the damage mechanism of the concrete impact test is highly similar to fatigue damage, so to explore the distribution law of the impact resistance of fiber concrete, the researcher applies the Weibull distribution function to the impact test. After previous studies, it was found that the Weibull distribution function is consistent with the damage properties of concrete materials under impact dynamic loading. In recent years, the Weibull distribution has been widely used in concrete; it can effectively describe the change of failure probability with time and use the test data to derive the correct impact law. Therefore, in this chapter, the Weibull distribution function in the theory related to mathematical statistics is used to fit the analysis and life reliability prediction of the number of initial and final cracking impacts of fiber-rubber concrete, respectively. The Weibull-distributed probability density function of the number of impacts N_x for fiber concrete is expressed as follows:

Where: α is the shape parameter of the Weibull distribution; N_a is the scale parameter (characteristic lifetime parameter), >0; N₀ is the position parameter (minimum lifetime parameter), ≥0; N_x is the observed value of the random variable, referring to the number of initial cracking impacts N₁ and the number of final cracking impacts N₂.

The image of the Weibull distribution probability density function is shown in Figure 8. When α = 1, the image becomes an exponential distribution function, when α = 2, the image becomes a Rayleigh distribution function, and when 3 < α < 5, the image is close to a normal distribution curve.

Weibull cumulative distribution (probability of failure) function:

The survival probability function can be expressed as shown below:

From the point of view of safety and reliability, it can be assumed that N₀ = 0. Equations [3], [4] and [5] are simplified as follows:

Upon taking the natural logarithm of both sides of Equation [8] twice yields the following equation.

Let $Y=\text{ln}\left[\text{ln}\left(1/P\left(N_x\right)\right)\right]$, $X=\text{ln}{N}_x$, $\beta =\alpha \text{ln}{N}_a$ Can be obtained:

If X and Y satisfy the linear equation in the above equation, then the impact life N is demonstrated to conform to a two-parameter Weibull distribution. Mathematical statistics correlation theory (4343. Binici B. 2008. Design of FRPs in circular bridge column retrofits for ductility enhancement. Eng. Struct. 30(3):766-776. https://doi.org/10.1016/j.engstruct.2007.05.012) says that the coefficient of determination R² above 0.7 indicates a good correlation; from the size of the correlation coefficient R to determine the correlation relationship, it is considered that 0.5 < | R | < 0.8 for the dominant linear correlation, and | R | > 0.8 is generally referred to as highly linear correlation.

The initial cracking impact times N₁ and destructive impact times N₂ of each group of concrete specimens were arranged in the order of smallest to largest, respectively, and then the expected valuation of the survival probability function Equation [11] was used to calculate the survival probability:

Where i is the number of orders after the number of impacts in order, i = 1, 2, 3, ..., n. n is the total number of samples for each group of specimens, this test n = 3.

3.3.2. Fitting analysis of impact test findings

According to Equations [9-11] and impact test results, the initial and final cracking impact numbers of the fiber-rubber concrete were fitted to obtain the corresponding parameter values α, β and the correlation coefficient R², as shown in Table 7. Figure 9 show the linearly fitted images of the Weibull distribution for the initial and final cracking impact numbers of the specimens.

In this test, R² is 0.689 at minimum and 0.997 at maximum, and the coefficient of determination R² of most of the specimens reaches above 0.800, which shows that there is a strong linear link between X=lnN_x and Y=ln[ln(1/P(N_x))]. Therefore, it can be inferred that the number of impacts of fiber-rubber concrete specimens can obey the two-parameter Weibull distribution function well.

3.3.3. Reliability analysis of impact life

By Equations [8-10] the value of N under different failure probabilities can be calculated using the following equation:

Combined with the regression fitting coefficients α and β in Table 7, the specific values of P for the failure probability function F(N_x) are taken as 5%, 15%, and 30%, respectively, and the predicted values of the initial and final cracking impact life of each group of specimens under different failure probabilities can be obtained, as shown in Table 8. Then the natural logarithm is taken to obtain lnN₁ and lnN₂, and the line graphs of the impact life of each group at varying failure probabilities are constructed, as illustrated in Figure 10. As the content of PVA fiber and rubber particles increases, the expected impact life for both initial and final cracks is enhanced for each group, which is consistent with the experimental results at the same failure probability.

3.4.1. Micro-morphological analysis of specimens

Figure 11 depicts the micro-morphological photographs of the NC, PFC0.15, PFC0.2-R3, and PFC0.20-R5 specimens. As illustrated in the figure, there are many defects and harmful pores in the cement-hardened paste in the NC specimen (Figure 11a). PVA fibers are evenly distributed on the cement soil surface of the PFC0.15, PFC0.20-R3, and PFC0.20-R5 specimens. Due to the bridging effect of the fibers, the PVA fibers are effectively bonded to the mortar interface to form a micro-reinforcement system, which is beneficial to the bending toughness of the specimens (Figure 11b, 11c and 11d). The rubber particles are randomly dispersed in the concrete matrix and evenly distributed in the mortar matrix. The amount of rubber particles can be appropriately increased. The rubber powder can fill some small pores in the concrete, thereby improving the toughness of the concrete (Figure 11c and 11d). However, rubber is an organic material with poor compatibility. With the increase in rubber content, the amount of bubbles introduced into the concrete increases, and the pores inside the concrete increase, which leads to a decrease in the compressive strength of the concrete.

3.4.2. Microscopic analysis of the rubber-mortar interface ITZ

Figure 12 illustrates the morphology of hydration products in the interfacial transition zone (ITZ) of the rubber-mortar matrix interface numbered PFC0.2-5. It can be observed that a small number of short columnar and clustered C-S-H and a considerable number of CH crystals (square or rectangular) are generated inside the mortar. CH has low strength and poor stability, while C-S-H gel can connect various hydration products to form a three-dimensional spatial structure (4444. Chen J, Yuan Y, Zhu Q, Duan J. 2023. High-temperature resistance of high-strength concrete with iron tailing sand. J. Build. Eng. 63:105544. https://doi.org/10.1016/j.jobe.2022.105544). The formation of short columnar and clustered C-S-H leads to an increase in the volume of hydration products. Due to the small number of C-S-H, the pore volume between the rubber and mortar matrix interface is increased, making the hydration products show a relatively void structure, thereby reducing the ITZ strength of PFC0.2-R5.

3.4.3. Analysis of the mechanism of crack-resistant reinforcement of PVA fibers

PVA fibers exhibit a three-dimensional random distribution in concrete, forming a spatial network that enhances the overall load transfer performance. As illustrated in Figure 13, PVA fibers primarily bond and work synergistically with concrete in two ways: one is that PVA fibers effectively prevent cracking; the other is that the chemical bonding force between the surface of the PVA fibers and the concrete transmits stress to the matrix on both sides of the crack while consuming some of the stress for the tensile deformation failure of the fibers themselves, thereby preventing further crack propagation, enhancing their tensile performance, and thereby improving flexural strength and impact resistance.