Influence of Silane-Modified Coal Gangue Ceramsite on Properties of Ultra-High-Performance Concrete

Abstract

In this study, a kind of sustainable ultra-high-performance concrete (UHPC) was designed by using coal gangue ceramsite (CGC) and a modified Andreasen–Andersen model. However, when CGC lightweight aggregate with high water absorption is used in UHPC with a low water–cement ratio, CGC has an adverse effect on the working performance of UHPC and may lead to the decrease of mechanical properties. This study found that a 5% silane coupling agent KH560 can make CGC hydrophobic, and cause its contact angle to increase from 0° to 111.32°. Adding 100% hydrophobic modified CGC into UHPC will significantly improve its working performance, with the highest increase of 38.51%. At the same time, the addition of 20% modified CGC can further improve the compressive strength of UHPC (28 days reached 150.1 MPa), reduce the internal porosity by 21.4%, and make the interface bond more compact. In addition, the hydration degree of UHPC has also been improved, a result caused by the cement obtaining more free water for a more complete hydration reaction. This study can provide a new scheme for solving the problem of the solid waste of coal gangue.

1. Introduction

Ultra-high-performance concrete (UHPC) is the most groundbreaking cement-based engineering material to emerge over the past thirty years. Thanks to its remarkably high compressive strength (>120 MPa) and flexural strength (>15 MPa) [1], UHPC is particularly well suited for specialized projects, including long-span bridges, super high-rise buildings, and militarized protective structures [2,3]. Typically, common UHPC utilizes natural sand, river sand, or quartz sand as aggregates, resulting in a density ranging from 2500 to 2700 kg/m³, which is about 10% higher than that of traditional concrete [4]. However, with the decreasing of natural sand and gravel resources and a growing demand for various lightweight functional UHPC materials, the preparation of UHPC using various artificial lightweight aggregates has emerged as a new trend in recent years [5]. Currently, the lightweight aggregates frequently employed in UHPC preparation include expanded clay, shale ceramsite, and pumice. Prior research has demonstrated that incorporating lightweight aggregates helps lower the apparent density of UHPC [6,7]. However, these aggregates typically exhibit high water absorption and porosity. Some studies [8] reveal that even after soaking for 50 days, lightweight aggregates can still take in a certain volume of water. When used in UHPC with a low water–cement ratio, they are prone to disrupting the water distribution and hydration reaction within the slurry, thus potentially weakening the working performance and mechanical properties of UHPC [9]. Therefore, further investigation is warranted to address the challenges posed by the low mechanical properties and high water absorption of lightweight aggregates on UHPC performance.

Coal gangue is a black waste rock generated during coal mining, with the global coal industry producing about 350 million tons of inactive coal gangue annually. Over the past several decades, Europe has amassed 1.75 billion tons of coal gangue waste in landfills [10,11]. Currently, China’s coal gangue reserves have surpassed 7 billion tons and are growing at a rate of 300–350 million tons each year. There are over 2600 large-scale coal gangue piles in China, covering a total area of approximately 15,000 hectares [12]. The chemical makeup of coal gangue is similar to that of clay, being rich in carbon and sulfur and having a high ignition loss. Only within a certain temperature range can it generate a sufficient amount of molten substances with suitable viscosity and expansibility. Based on this feature, China has developed CGC through sintering and applied it to concrete. CGC boasts low apparent density, a round shape, and high mechanical properties [13,14,15,16]. Meanwhile, CGC exhibits corrosion resistance, freeze–thaw resistance, and other favorable properties [17,18], making it suitable for thermal wall insulation in a variety of buildings [18]. At present, while some studies [19] have found that the brittleness of CGC can be significantly reduced when the volume fraction of steel fiber does not exceed 1.5%, the mechanical properties of concrete mixed with CGC are usually worse than those of conventional concrete containing natural aggregate due to the porosity and low strength of CGC aggregate [20]. At the same time, research on the application of CGC in UHPC remains scarce. Only Chen et al. [21] have successfully prepared UHPC with excellent performance using saturated pre-wetted CGC. Moreover, there has been no reported research on how to improve the influence of high water absorption of CGC on concrete performance.

Silane coupling agent (SCA) is a type of silicone compound featuring two chemical groups within its molecule, and after hydrolysis, these groups can undergo chemical reactions with both inorganic substances and organic polymers, enabling SCA to serve as a “bridge” connecting organic and inorganic materials [22]. For example, Dong et al. modified the surface of rubber aggregate using SCA [23] and compared it with concrete containing unmodified aggregate. The results revealed that, under identical conditions, the compressive strength and splitting tensile properties of concrete treated with SCA improved by 10–20%. At present, numerous researchers employ SCA to modify inorganic materials and additives in concrete, aiming to improve the bonding effect between inorganic materials and cementitious components, thereby optimizing various macro-properties [22]. Examples of such inorganic materials include recycled concrete aggregate [24], fly ash floating beads [25], and basalt fiber [26]. From the above research, it is evident that modifying inorganic materials with SCA is an effective approach. Theoretically, modifying CGC aggregate with SCA can improve the properties of CGC concrete. In addition, although there is some research on the modification of inorganic materials, no studies have explored the content of SCA-modified CGC in UHPC.

Therefore, based on the preceding analysis, this study first uses CGC fine aggregate (0.075 mm~2.36 mm) and combines cement, fly ash, and silica fume as cementing materials. The mix proportion of UHPC was determined through the MAA model, supplemented by insights from previous experiments. To mitigate the adverse impact of CGC’s high water absorption on UHPC performance, CGC was soaked in a KH-560 solution of SCA for surface hydrophobic modification. Firstly, the effectiveness of aggregate modification was assessed via contact angle and water absorption tests. Infrared spectrometry and scanning electron microscopy were used to observe changes in functional groups and the morphology of the modified aggregate. Then, the modified aggregate was mixed into UHPC at various mass substitution rates to evaluate alterations in working performance, apparent density, and mechanical properties. For each group, three specimens were prepared to test the flexural and compressive strength at different ages. The interface, pores, and hydration products of UHPC, both before and after modification, were analyzed using a scanning electron microscope, X-ray diffraction (XRD), thermogravimetry, and mercury intrusion porosimetry. This research aims to facilitate the diversified utilization of CGC in the future.

2. Materials and Methods

2.1. Raw Materials

The cement used in the experiment was P.O 42.5R. The fly ash was purchased from Yangzi Petrochemical in Nanjing, China; it has a pozzolanic and micro-aggregate effect, and can also reduce CO₂ emission by replacing part of the cement [27,28]. Silica fume was purchased from Ekeng silicone in Shanghai, China. The fineness and high activity of silica fume can significantly improve the compressive strength and tensile strength of concrete. In this study, the amount of fly ash and silica fume accounts for 15% and 20% of the cementitious materials, respectively. The coal gangue ceramsite (CGC) came from Dongchen Group in Huainan, China, as shown in Figure 1. In this experiment, CGC with particle sizes ranging from 0.075 mm to 2.36 mm and an apparent density of 1780 kg/m³ was used as aggregate. The chemical compositions of the above-mentioned materials are shown in

Table 1. Chemical composition of cement, fly ash, silica fume and coal gangue ceramsite.

Ingredient	CaO	SiO2	Al2O3	Fe2O3	MgO	SO3	K2O	TiO2	LOI
Cement	57.03	21.35	6.45	3.69	4.01	3.10	0.75	0.28	2.51
Silica fume	0.54	90.45	0.56	0.44	0.87	0.27	1.57	0.01	3.58
Fly ash(I)	13.45	55.39	19.05	2.46	0.1	0.22	2.75	0.07	9.87
CGC	1.45	68.39	18.93	4.62	0.91	0.17	2.63	0.74	1.67

.

As shown in Figure 2, the steel fiber used in the experiment has a length of 13 mm, a diameter of 0.2 mm, a density of 7850 kg/m³ and a tensile strength of more than 2200 MPa. In this study, the amount of steel fiber is 1.5% of the volume ratio of cementitious materials. The water reducer used is polycarboxylic acid superplasticizer (PCE).

2.2. Grading Design

As an aggregate, CGC provides lower strength than quartz sand. Therefore, the modified Andreasen and Andersen models are used to design the UHPC in this experiment to ensure that the mechanical strength of the UHPC can reach the standard and theoretically make the mixture possess the minimum porosity. The equation of this model is as follows (1):

$$P\left(D\right)={\displaystyle \frac{D^q-D_{min}^q}{D_{max}^q-D_{min}^q}}$$

(1)

where P(D) represents the proportion of solids smaller than the particle size d in the total solids (UHPC formula); D represents the particle size (μm) of the particles; D_min represents the smallest particle size (μm) in the total solids; D_max represents the largest particle size (μm) in the total solids; and q represents the distribution modulus, which can be adjusted by different particle size distributions. According to previous research [29,30,31,32], when the mixture is composed of geopolymer, with cement as the binder, and is rich in fine particles, it is suggested that the $q$ value is between 0.20 and 0.25. In this experiment, in order to obtain a suitable packing density and refer to similar literatures, it was decided to fix the $q$ value at 0.23 [33,34].

For the specified particle size, if the residual sum of squares function minimizes the difference between the theoretical design curve and the experimental curve, each raw material is considered as the optimal content [35]. Therefore, the model is optimized by the following Equation (2):

$$\mathrm{R}\mathrm{S}\mathrm{S}={\sum }_{i=1}^n{\left[P_{exp}\left(D_i\right)-P_{the}\left(D_i\right)\right]}^2\to min$$

(2)

Among them, $P_{exp}$ is an experimental mixture and $P_{the}$ is a theoretical mixture designed according to the MMA model.

According to the above MAA model, the mixture ratio model of UHPC with CGC as aggregate is finally obtained, as shown in Figure 3.

2.3. Hydrophobic Modification of Aggregate

The silane coupling agent (SCA) KH-560 (γ-propyl trimethoxysilane) used in this experiment was purchased from Shanghai Yuanye Biotechnology Co., Ltd. in Shanghai, China. KH-560 is a colorless transparent silane, which contains epoxy functional groups and is soluble in organic solvents (such as acetone, benzene, and ethylene) and water. In order to promote the dispersion of the SCA on the surface of the CGC, it is necessary to carry out hydrolysis in the solution of water and alcohol (ethanol, isopropanol, etc.) in advance. If the amount of SCA is too small, the linking effect with the aggregate surface is poor, which affects the hydrophobic effect of the surface; however, when the amount of SCA is too large, the thick coating may affect the bond strength between the aggregate and cementitious materials. Therefore, when preparing the SCA solution, the mass fraction of KH-560 (2.5 wt% and 5 wt%) is taken as the control variable, and the appropriate dosage of KH-560 is determined according to the test numbers KH0, KH2.5, and KH5.

Firstly, water and alcohol were mixed and stirred according to the mass ratio of 4:6 [22], and a certain amount of acetic acid was added to adjust the PH of the solution to 4, which is conducive to a better hydrolysis of SCA in an acidic environment, and then KH-560 with a certain mass ratio was added. The prepared solution was placed on a magnetic stirrer and continuously stirred for 20 min to achieve hydrolysis. After hydrolysis, CGC was added and the mixture was left to soak for 24 h. The soaked CGC was then dried so that the hydrophobic modified aggregate needed for the experiment was obtained. Then the hydrophobic effect of different concentrations of SCA was tested using a contact angle test, and the changes of surface morphology and functional groups of modified aggregate were observed through infrared analysis and by scanning electron microscope, so as to confirm the required mass ratio of KH-560.

2.4. Mixing Ratio

The dried modified CGC replaces the unmodified aggregate in the UHPC with the mass ratio of 20%, 40%, 60%, 80%, and 100%. For the foundation mixture ratio without modified aggregate, it is recorded as M0, and the mixture ratio with 20% to 100% modified aggregate is recorded as M20, M40, M60, M80, and M100, respectively. The mixing ratio is summarized in

Table 2. UHPC mixing ratio (kg/m³).

Mixture	M0	M20	M40	M60	M80	M100
Cement	310	310	310	310	310	310
FA	71	71	71	71	71	71
SF	95	95	95	95	95	95
CGC	395	316	237	158	79	0
MCGC	0	79	158	237	316	395
Steel Fiber	22	22	22	22	22	22
PCE	7	7	7	7	7	7
Water	100	100	100	100	100	100

.

2.5. Contact Angle Experiment

The hydrophobic modification effect of aggregate surface was measured by JC2000C1 contact angle goniometer (Netzsch, Selb, Germany), and the maximum error was 0.1°. Droplets were deposited on the surface of the sample by a vertically downward syringe under the action of gravity. This angle can be captured by a high-resolution camera and analyzed by analysis software. When the contact angle is less than 90°, it indicates a lyophilic state, and when it is greater than 90°, it indicates a lyophobic state. The test was repeated three times for each group of samples.

2.6. Water Absorption

In order to test the change of water absorption of the aggregate after hydrophobic modification, the dry water absorption of the saturated surface of the coal gangue ceramsite aggregate was tested. An appropriate amount of aggregate was put into a container, 20 mm of clear water was poured into the sample, and it was left to stand for 24 h. Then the aggregate was dried to a constant weight. The calculation formula is as follows (3) (accurate to 0.1%):

$$w_a={\displaystyle \frac{m_{l1}-m_{l0}}{m_{l0}}}\times 100\%$$

(3)

where $w_a$ represents the saturated surface dry water absorption (%); $m_{l1}$ represents the saturated surface dry sample mass (g); $\mathrm{a}\mathrm{n}\mathrm{d} m_{l0}$ represents the mass (g) of the dried sample.

2.7. Fourier Infrared Exchange Spectrometer

The modified aggregate was analyzed by FT-IR with a Fourier transform infrared spectrometer (Nicolet STD-11202624D, Boca Raton, FL, USA). Through the infrared spectrum data, we were able to study the changes of organic groups on the surface of the coal gangue ceramsite before and after modification; the wave number range was 500~4000 cm⁻¹.

2.8. Working Performance Experiment

The influence of the substitution rate of modified CGC on its working performance was determined by detecting the fluidity of the UHPC with different proportions. The newly mixed slurry was put into a truncated cone circular die (the diameter of the upper opening was 70 mm, the diameter of the lower opening was 100 mm, and the height was 60 mm) twice, and vibrated with a tamping rod. Finally, it was placed on a jumping table and bounced 25 times. A ruler was used to measure the diameters in two directions perpendicular to each other at the bottom, and the results were averaged, which was the desired fluidity result.

2.9. Apparent Density

The apparent density of the UHPC was measured according to the Chinese national standard GB/T 50081-2019 [36], and the apparent volume of the specimen was calculated as follows (4):

$$V_{\mathrm{a}}={\displaystyle \frac{m_d-m_w}{{\rho }_w}}$$

(4)

where $V_{\mathrm{a}}$ is the apparent volume (m³) of the specimen, including the solid volume of the specimen and the closed pore volume inside the specimen; $m_d$ is the mass of the dried specimen (kg); $m_w$ is the mass of the specimen in water (kg); ${\rho }_w$ is the density of water, and when the water temperature is 20 °C, the value is 998 kg/m³.

The apparent density of the UHPC can be calculated according to the following Equation (5):

$${\mathsf{\rho }}_{\mathrm{a}}={\displaystyle \frac{m_d}{V_a}}$$

(5)

where ${\mathsf{\rho }}_{\mathrm{a}}$ is the apparent density of the specimen, and the result is accurate to 10 kg/m³.

2.10. Mechanical Property Test

The influence of the substitution rate of modified CGC on its mechanical properties was determined by testing the flexural strength and compressive strength of the UHPC with different proportions after curing for 28 days. The specimen size was 40 × 40 × 160 mm, and it was cured in a standard environment until the age test, with three samples in each group. The average value of the measurement was taken as the result of its compressive and flexural strength. Due to experimental limitations, tensile and split tensile strength tests were not performed in this study. Future work will consider these properties to provide a more comprehensive evaluation of the UHPC performance.

2.11. X-Ray Diffraction (XRD)

The cured 28 d sample was soaked in anhydrous ethanol for 24 h, then dried in a vacuum oven at 40 °C for 12 h, and the sample was ground into powder. The crystal composition of the sample before and after adding the modified aggregate was detected by Rigaku-2500 X-ray diffractometer (Rigaku Corporation, Tokyo, Japan).

2.12. Thermogravimetric (TG) Analysis

The samples cured for 28 days were analyzed by thermogravimetry to characterize the main components of the samples. Before the test, the dried samples were ground, and then tested by STA 449C synchronous thermal analyzer (Netzsch, Selb, Germany). The weight of each sample used in the test was 25 mg, and the protective gas used in the test was nitrogen. The sample was heated from room temperature to 1000 °C at a heating rate of 20 °C/min.

2.13. Mercury Intrusion Porosimetry (MIP)

The samples, cured for 28 days, were cut into particles and put into alcohol to stop the hydration reaction. Before testing, they were dried in a drying oven, and then put into AutoPore IV 9510 mercury intrusion porosimeter (Micromeritics Instrument Corp., Norcross, GA, USA) to measure their porosity and pore structure.

2.14. Scanning Electron Microscope (SEM)

SEM was used to observe the micro-morphology changes of the CGC surface before and after modification and the UHPC before and after adding the modified aggregate, mainly aiming at the changes of concrete interface after replacing the modified aggregate.

3. Results and Discussion

3.1. Hydrophobic Modification

3.1.1. Contact Angle

The contact angle is an important parameter for assessing the hydrophobicity of a sample’s surface. The impact of varying doses of KH-560 solution on hydrophobic modification is depicted in Figure 4.

It is clear that the aggregate exhibits increasing hydrophobicity as the concentration of the SCA solution rises. When the concentration of SCA is 2.5% and 5%, the contact angles measure 59.78° and 111.32°, respectively, indicating that the aggregate develops noticeable hydrophobicity when the SCA concentration reaches 5%. Therefore, for this experiment, the SCA solution with a concentration of 5% was selected as the hydrophobic solution ratio.

3.1.2. Water Absorption

To evaluate the hydrophobic modification effect of CGC, we measured the water absorption change of dry aggregates after reaching a saturated surface following hydrophobic treatment. As illustrated in Figure 5, KH0 represents unmodified aggregates, while KH2.5 and KH5 represent aggregates modified with different SCA concentrations, respectively.

Figure 5 clearly shows that SCA effectively reduces aggregate water absorption. As the SCA concentration increases, aggregate water absorption decreases accordingly. Compared to unmodified aggregates, those soaked in 2.5% and 5% SCA solutions exhibited a 13.1% and 27.9% reduction in water absorption, respectively. Therefore, it was proved that the hydrophobic modification effect of the aggregate was remarkable.

3.1.3. FTIR Analysis

To further determine the changes of groups contained in the CGC before and after modification, infrared spectroscopy was conducted in this experiment, as depicted in Figure 6.

The infrared spectrum reveals that the absorption peak at 1059 cm⁻¹ stems from the antisymmetric stretching vibration of Si-O-Si [37]. The peak around 847 cm⁻¹ corresponds to the bending vibration absorption of Si-OH, while the peak at 786 cm⁻¹ is the symmetric stretching vibration of the Si-O bond. Additionally, the absorption band near 1617 cm⁻¹ shows the tensile vibration peak of the -OH group. As the concentration of the modifying solution increases, the peak shape of the -OH group moves to lower wavenumbers, likely due to the accelerated consumption of -OH bonds with higher KH-560 content [38]. Furthermore, a new absorption peak emerges at 2951 cm⁻¹, attributed to the stretching vibration peak of the C-H bond in the propyl group [39], and all these indicate that CGC was successfully modified and hydrophobic groups were introduced.

SCA KH-560 is a low-molecular-weight organosilicon compound with a unique structure. During the coupling process, the alkoxy group at one end of KH-560 undergoes hydrolysis to form a silicon hydroxyl group, which can undergo a condensation reaction with the hydroxyl group on the surface of CGC, establishing strong hydrogen bonds and Si-O-Si covalent bonds, thus increasing the interaction between the SCA and the aggregate and allowing the SCA to be firmly adsorbed onto the ceramsite aggregate surface [40]. Meanwhile, the other end of the KH-560 molecule contains hydrophobic epoxy groups (-OR), which can make SCA, when adsorbed onto the surface of inorganic materials, to form a hydrophobic layer, thus reducing the contact area between the surface of inorganic materials as well as water and improving the hydrophobicity of the surface. The hydrolytic condensation reaction equation is presented as follows (6):

(6)

-OR represents a hydrophobic group.

3.1.4. SEM

Figure 7 displays the surface conditions of CGC before and after modification, respectively. In comparison to the unmodified aggregate shown in Figure 7a, it can be seen that the surface of the modified aggregate in Figure 7b is distinctly coated with a layer of SCA solution, which makes the aggregate hydrophobic.

3.2. Working Performance

The modified aggregate is substituted for the unmodified aggregate in UHPC on a mass-ratio basis, and according to the replacement ratio, the original ratio and substitution ratio are labeled as M0, M20, M40, M60, M80, and M100. In accordance with the relevant standard GBT2419-2005 [41], the fluidity of each ratio was measured, and the average flow diameter was recorded, as illustrated in Figure 8, and the influence of modified aggregate on the working performance of UHPC was analyzed.

From Figure 8, it is evident that as the content of modified aggregate rises, the fluidity of the UHPC gradually improves, demonstrating that the SCA effectively enhances the UHPC’s workability. The fluidity of M0 is only 148 mm, while when the mass substitution rate of modified aggregate hits 100%, the fluidity reaches 205 mm, marking a 38.51% increase. In contrast to the UHPC system incorporating unmodified CGC and steel fibers studied by Shan et al. [42] (flowability of approximately 180 mm), the use of SCA-modified CGC in this study significantly boosts the flowability of the UHPC, thereby ensuring favorable workability.

The enhanced workability primarily stems from the hydrophobicity imparted to the ceramsite aggregate surface through SCA modification. Given a constant water–cement ratio, as the proportion of modified aggregate in the UHPC increases, the amount of free water that the aggregate can absorb gradually decreases, so more free water exists between cementitious materials and aggregate [22], ultimately resulting in improved workability of the UHPC.

3.3. Apparent Density

To assess the impact of modified aggregate on the apparent density of the UHPC, the apparent density of samples cured for 28 days was measured, as depicted in Figure 9.

When no modified aggregate was incorporated, the apparent density of the UHPC after 28 days of curing was 2189 kg/m³. Upon adding modified aggregate, within a mass substitution rate of 40%, the apparent density increased, and it peaked at a 20% dosage, reaching 2210 kg/m³, which is 1% higher than M0. With the mass substitution rate of modified aggregate surpassing 40%, the apparent density started to decline. At a 100% substitution rate, the apparent density dropped to only 2090 kg/m³, 4.5% lower than that of M0.

The measurement of the UHPC’s apparent density is related to its internal pore volume. When the mass substitution rate of modified aggregate is below 40%, it is speculated that the internal pore structure is refined, with a reduction in pore quantity, so the apparent volume is reduced, which leads to an increase in apparent density. Conversely, when the mass substitution rate exceeds 40%, there is an uptick in the number of internal pores. As a result, the apparent volume of the UHPC expands, causing its apparent density to decline.

3.4. Mechanical Properties

To evaluate the impact of modified aggregate replacement rate on the mechanical properties of the UHPC, the flexural and compressive strengths of samples cured for 28 days were measured. The changes in flexural strength are illustrated in Figure 10a, while the compressive strength variations are shown in Figure 10b. The flexural and compressive strength of the M0 group without modified aggregate reached 20.6 MPa and 139.64 MPa.

The strength test data diagram reveals that as the mass substitution rate of modified aggregate increased, the 28-day mechanical properties of the UHPC initially improved and then decreased. The optimal mechanical properties were achieved at a 20% substitution rate, with flexural and compressive strengths reaching 23.9 MPa and 150.1 MPa, respectively, which are 16% and 7% higher than M0. Compared to the 28-day compressive strength of 143 MPa reported by Zhao et al. [43] using calcined coal gangue aggregate, this study further substantiates the significant effectiveness and feasibility of silane-modified aggregates in enhancing UHPC’s mechanical performance. At a 40% substitution rate, flexural and compressive strengths still improved by 6% and 4%, respectively. However, when the substitution rate reached 60%, the mechanical properties started to fall below those of M0 and continued to decrease with increasing content. When the substitution rate reached 100%, flexural and compressive strengths dropped to only 14.6 MPa and 108 MPa, representing decreases of 29% and 23%, respectively.

Drawing on the relevant literature and in light of the experimental observations regarding mechanical properties, it can be deduced that as the reaction progresses, when the substitution rate of modified aggregate remains below 40%, two key factors contribute to the enhancement of UHPC’s mechanical properties. On the one hand, the bond strength between the aggregate and the cement matrix gradually improves. The SCA not only boosts the aggregate’s hydrophobicity but also reacts with hydroxyl groups in the cement paste to produce new structured compounds [22], thus reducing the interfacial porosity and improving the mechanical properties of UHPC, which is consistent with Xu’s conclusion [44]. On the other hand, the hydration reaction of cement-based materials is also enhanced [45], and after silane modification, the hydrophobic film on the aggregate surface may mitigate the early-stage “rapid consumption” of water, preserving more for sustained hydration later. This optimizes the hydration environment in concrete, ensuring a uniform release of water, better internal humidity control, and a more stable secondary hydration process that extends the reaction time. The combined effect of these factors strengthens UHPC’s mechanical properties. However, when the modified aggregate content exceeds 40%, excessive aggregates may lead to poor interfacial bonding, reducing interface contact performance [46]. At the same time, too much free water does not react completely, which leads to a decrease in strength.

3.5. Micro-Mechanism Analysis

3.5.1. XRD Analysis

X-ray diffraction (XRD) was conducted to investigate the alterations in the crystalline phases of UHPC mixtures (M0, M20, and M100) after 28 days of standard curing, as depicted in Figure 11.

The XRD pattern reveals that the UHPC incorporating hydrophobic-modified aggregate largely retains the same crystal structure as the control group. The most prominent characteristic peaks in all three samples are observed at 29.5° and 32.6°, corresponding to tricalcium silicate (C₃S) and dicalcium silicate (C₂S), respectively, which are due to the existence of incompletely hydrated cement particles and the low water–cement ratio of UHPC. Calcium hydroxide is detected at 18.1° and 47.5°, while ettringite is identified at 8.9°. Although the peak values of these hydration products exhibit some changes with the addition of modified aggregate, the variations remain relatively subtle. The most notable peak change occurs in the Calcium-Alumino-Silicate-Hydrate (C-A-S-H) gel at 27.8°. C-A-S-H gel is the primary binding phase formed during the hydration of cementitious materials, contributing significantly to the strength and durability of concrete. The peak intensity significantly increases with the incorporation of modified aggregate, suggesting an augmentation in the amount of UHPC hydration products. Therefore, the data indicate that the addition of hydrophobic modified aggregate will not change the crystal structure in UHPC, but only affect the amount of hydration products.

3.5.2. TG-DTG Analysis

To examine how the addition of hydrophobic-modified aggregate affects the hydration products of UHPC, a TG-DTG analysis was performed on UHPC samples from the M0, M20, and M100 groups after 28 days of curing. The following TG curve in Figure 12a and the DTG curve in Figure 12b were obtained.

The DTG curve reveals three obvious endothermic peaks. The first, occurring between 50 °C and 200 °C, primarily results from the dehydration of hydration products in UHPC, such as the C-S-H gel and ettringite (AFt) [47]. The second endothermic peak, observed between 320 °C and 530 °C, is mainly attributed to the dehydration and decomposition of Ca(OH)₂; The third peak, appearing between 650 °C and 750 °C, is mainly caused by the decarbonization and decomposition of crystalline calcium carbonate (CaCO₃). When comparing M0, M20, and M100, the most notable differences emerge in the 50 °C and 200 °C range. As the content of modified aggregate increases, the amounts of C-S-H gel and AFt in UHPC rise, leading to greater mass loss. Additionally, regarding the CH dehydroxylation peak, its intensity strengthens with an increasing proportion of modified aggregate. At the same time, compared to the CH loss in the M0 group between 320 °C and 530 °C, the weight loss in M20 and M100 increased by 4.7% and 8.6% respectively, indicating that as the modified aggregate content rises, more CH is consumed, resulting in the formation of additional C-S-H gels. Therefore, it can be seen that the hydration degree of UHPC is further improved with the increase of modified aggregate.

3.5.3. Pore Structure Analysis

MIP was employed to analyze the pore evolution in UHPC with different ratios of modified aggregate, namely M0, M20, and M100, as illustrated in the following figure. Figure 13a depicts the variation in porosity; the pore content and pore size distribution of each classification are demonstrated in Figure 13b–d which show the changes in cumulative pore volume.

As evident from Figure 13a, with an increasing mass substitution rate of modified aggregate, the porosity of UHPC initially decreases and then rises. Specifically, the porosity of M0, M20, and M100 are 15.49%, 12.18%, and 18.92%, respectively. UHPC with 20% modified aggregate exhibits the lowest porosity. From a porosity perspective, 20% modified aggregate is the best mass substitution ratio in UHPC.

Pores in concrete can be broadly categorized into four categories: gel pores (<10 nm), less harmful pores (10–50 nm), capillary pores (50–200 nm), and macropores (>200 nm) [48]. As depicted in Figure 13b,c, the gel pores in UHPC decrease as the content of modified aggregate increases, and the pore distribution in M0 and M20 is primarily concentrated in gel pores and less-harmful pores. The addition of an appropriate amount of modified aggregate also optimizes the content of less-harmful pores in M20 to some extent. However, the pore distribution in M100 is mainly centered around less-harmful pores and macropores, and macropores are the key pores that negatively impact the mechanical properties of cement-based materials. When the substitution rate of modified aggregate reaches 100%, the macroporous content in UHPC rises, resulting in the lower mechanical properties of M100. From the change in cumulative pore volume depicted in Figure 13d, it is clear that adding a suitable quantity of modified aggregate helps refine the pores and enhance the mechanical properties of UHPC, as evidenced by the left-shift of the cumulative pore volume curve for M20. However, an excessive amount of modified aggregate leads to more harmful pores, causing the cumulative pore volume curve for M100 to shift to the right, so the mechanical properties of UHPC will decrease.

The change in pore structure reveals that the incorporation of hydrophobic-modified aggregate exerts a significant effect on the internal pore characteristics of UHPC. With the overall water-binder ratio held constant, when the modified aggregate content reaches 20%, the UHPC system exhibits a more favorable pore distribution. On the one hand, the surface of the modified aggregate demonstrates an excellent combination of affinity and hydrophobicity, which enhances the interfacial bonding strength between aggregate and cement matrix [49]. On the other hand, the structure of the interface transition zone is optimized, facilitating the further hydration of cement particles [45], thus reducing the number of harmful pores in the UHPC, further refining the gel pores, and lowering the overall porosity. However, when the replacement rate of modified aggregate is increased to 100%, several issues arise. First of all, the interface bonding between the modified aggregates is relatively weak, making it prone to the formation of micro-cracks and gaps, which subsequently increases interface defects. Secondly, due to the high proportion of aggregate replacement, some areas become enriched with cement paste, resulting in a local retention of free water, which gradually evaporates during the later curing process, thus forming new pores in the material. Under the combined effect of these factors, the porosity of the UHPC increases, and the rise in macropores is also a contributing factor [47].

3.5.4. SEM Analysis

The microstructural transformations of UHPC before and after modification are illustrated in the figure. Figure 14a,b presents SEM images of the UHPC interface in the absence of modified aggregate. Figure 15a,b depicts the SEM images showcasing the interface changes in UHPC when 20% modified aggregate is incorporated; Figure 15c,d reveals the morphology of hydration products within UHPC containing the modified aggregate.

As evident from Figure 14a,b, the surface of the UHPC without modified aggregate exhibits a notable number of pores, and some fine, round, and smooth-surfaced fly ash particles remain incompletely reacted. The interface is relatively looser, with a poor connection between the cementitious materials and aggregates, and cracks and gaps are present in the interface transition zone (ITZ). Aggregate is not fully enveloped and bonded by the gel, which compromises the compactness of the UHPC. By comparing Figure 15a,b, it becomes clear that the addition of modified aggregate renders the UHPC interface denser, with a reduction in unreacted fly ash particles. At the same time, the increased production of gel enhances the bonding effect with the aggregate. The aggregate is thoroughly wrapped, and the bonding is more robust, resulting in fewer micro-cracks and an improved interface degree between the substrates. Figure 15c,d reveals the presence of numerous hydration products in the UHPC.

SEM images reveal that incorporating 20% modified aggregate significantly enhances the hydration reaction degree of cementitious materials in the UHPC, leading to the formation of a greater quantity of C-S-H gels. At the same time, the ITZ between the aggregate and matrix becomes more tightly bonded, resulting in the improved density of the UHPC, which aligns with the pore test results. Related studies indicate that when concrete aggregates exhibit high hydrophilicity, a water film tends to adsorb on their surface, and during cement hydration, this can promote the growth of loose, large crystals like calcium hydroxide at the interface, thereby compromising interfacial bonding strength [50]. However, the hydrophobic groups introduced after modifying aggregates with the SCA prevent the formation of a continuous water film on the aggregate surface, thus avoiding the growth of loose large crystals at the interface and improving the bonding strength of the ITZ [49]. This also elucidates why adding an appropriate amount of modified aggregate can improve the mechanical properties of UHPC.

4. Conclusions

In this study, the application of coal gangue ceramsite in UHPC is successfully designed through an MAA model. At the same time, the influence of silane coupling agent KH560 solution on the working properties, mechanical properties, and microstructure of CGC after it was hydrophobically modified by UHPC was studied. According to the experimental results in this paper, the following conclusions can be drawn.

• A 5% concentration of SCA solution can make CGC obviously hydrophobic. It is found that the water absorption of aggregate decreases by 27.9% and the contact angle reaches 111.32°. The changes of functional groups and morphology of the aggregate surface can be obviously observed by infrared analysis and scanning electron microscope.
• The working performance of UHPC was significantly improved with the increase of modified CGC content. When the substitution rate reached 100%, the fluidity increased by 38.51% compared with the reference group, which could effectively solve the influence of CGC super water absorption on the working performance of UHPC. However, the improvement of working performance is limited at low dosage, which shows that the effect of modified CGC on improving the fluidity of UHPC is more significant at a high substitution rate.
• When the substitution rate of modified CGC is less than 40%, the mechanical properties of UHPC 28d are improved. When the dosage is 20%, the flexural strength and compressive strength of UHPC are improved by 16% and 7%, respectively, compared with the M0 group without modified aggregate.
• Through XRD and TG-DTG experiments, it is found that the addition of modified aggregate is beneficial to improve the hydration degree of UHPC. The addition of hydrophobic aggregate can prevent too much free water from being absorbed by the aggregate and thus participate in the reaction of cement slurry to improve the degree of hydration. At the same time, MIP and SEM experiments show that adding 20% modified aggregate is beneficial in reducing the porosity of UHPC, optimizing the interfacial bonding performance, and improving its mechanical properties. However, when the content is further increased by more than 40%, the internal porosity of UHPC increases and the mechanical properties decrease. Although MIP and TG-DTG did not analyze the intermediate group, the trend showed that the porosity gradually increased from M20 to M100, and the interfacial bonding strength gradually decreased, which was consistent with the mechanical strength results.