Study on the Bonding Properties of Reinforced Reef Limestone Concrete and Its Influencing Factors

Abstract

Reinforced concrete structures play a pivotal role in island and reef engineering projects. Given the resource constraints typical of island regions, substituting traditional manufactured sand aggregate with reef limestone not only reduces reliance on river sand but also addresses the issue of disposing of waste reef limestone slag generated during excavation. However, the performance characteristics of reef limestone concrete, particularly its bond strength with reinforcing steel, warrant further investigation. This is particularly true for the bond–slip behavior of the reinforcement. This study aims to elucidate the effects of various parameters on the bond performance between steel and reef limestone concrete through central pullout tests. These parameters include the type and diameter of the reinforcement, bond length, and loading rate. The investigation encompasses the analysis of load–slip curves, bond failure modes, and variations in bond stress. Additionally, using the Abaqus software, a numerical simulation was conducted to analyze the mesoscopic stress characteristics, thereby revealing the mechanisms of bond formation and failure modes between steel reinforcement and reef limestone concrete. The results indicate that the bond–slip curve for reef limestone concrete reinforced with ribbed rebars and Glass Fiber-Reinforced Polymer (GFRP) rebars can be broadly categorized into four phases: minor slip, slip, decline, and residual, with the residual phase exhibiting a wave-like pattern. The predominant failure modes in reef limestone concrete are either pulling out or splitting. The bond stress in reef limestone concrete decreases with an increase in rebar diameter and bond length; conversely, it increases with the loading rate, although the ultimate slip decreases. The mesoscopic failure characteristics of reinforced reef limestone concrete, as simulated in Abaqus, are consistent with the experimental outcomes.

1. Introduction

Reef limestone concrete is a type of concrete that utilizes fragments of reef limestone as aggregate, a material that has found practical applications in several coastal nations [1,2]. Reef limestone primarily forms through biogenic sedimentary processes, including the demise of reefs and the accumulation of calcium carbonate structures. These processes yield a porous and structurally heterogeneous rock. The high porosity not only reduces the density of rock but also impacts its load-bearing capacity and durability. Owing to its unique biogenic genesis, reef limestone exhibits characteristics such as high porosity—often as much as 50%—a low density of approximately 1.8 g/cm³, and comparatively reduced mechanical strength. These properties significantly differentiate reef limestone concrete from traditional terrestrial rock aggregate concrete in various aspects [3,4,5]. Although the reduced density and increased porosity may result in diminished structural strength, insufficient for the high load demands of island construction, reef limestone concrete offers significant advantages in terms of environmental conservation, resource efficiency, and cost control [6,7]. Employing reef limestone concrete reduces reliance on traditional aggregates, especially in resource-limited island and coastal regions. Additionally, it effectively addresses the disposal issues of discarded reef limestone in these areas, offering a sustainable choice of building material.

In recent years, spurred by advancements in national maritime strategies, numerous scholars both domestically and internationally have focused on the properties of reef limestone concrete, aiming to enhance its utilization in engineering applications [8,9,10]. Howdyshell [11] conducted a survey on using reef limestone in concrete, finding its incorporation as an aggregate feasible. Beshr [12] noted that reef aggregates affect the cementation process, consequently reducing the compressive strength of reef limestone concrete compared to standard concrete. Huang [13] examined the impacts of reef coarse aggregates and crushed stone on concrete, using uniaxial tests and digital imaging to observe crack patterns. The results highlighted that reef concrete is more brittle, predominantly failing through cracks that penetrate the aggregates. Ma [14] investigated the uniaxial dynamic mechanics of reef concrete, measuring parameters such as compressive strength, fractal dimension, and energy dissipation. Early-stage reef concrete was found to exhibit higher strength, reaching 0.9 times the standard compressive strength after seven days, with the fractal dimension ranging from 2.027 to 2.302, showing a positive correlation with the logarithm of the strain rate. Utilizing true triaxial Hopkinson bar experimental techniques, Fan [15] explored the dynamic mechanical properties and failure characteristics of coral limestone concrete under bidirectional stress constraints in a water-saturated state. The findings indicate an increase in dynamic compressive strength with the strain rate, significantly amplified by lateral stress constraints, especially in water-saturated conditions.

Research findings indicate that reef concrete is already widely adopted. To expand its applications, reef limestone concrete is often utilized alongside steel reinforcement [16,17,18,19]. Wang [20] conducted a central pullout test to examine the adhesion between carbon fiber-reinforced materials and reef concrete. The study identified that the bond load–slip curve can be segmented into four stages, revealing that bond strength decreases with fiber reinforcement diameter and anchor length but increases with the concrete’s inherent strength. Further, Wang [21] investigated the effects of coral sand powder (CSP) and preservatives (SBT(R)-ZX(V) and SBT(R)-RMA(II)) on the corrosion of steel bars in coral aggregate seawater concrete (CASC) within marine environments. The study proposed and validated a time model for steel corrosion in CASC, aiming to predict crack initiation and estimate service life. Nie [22] performed pullout tests using various concrete types, steel bars, bond length ratios, concrete strengths, and protective layer thicknesses. Results indicated that the bond strength between SCAC and ECR surpasses that of traditional concrete, although it decreases as the bond length ratio increases. Yuan [23] assessed the flexural behavior of concrete beams (CAC) constructed with seawater, sea sand, and coral aggregate, reinforced with non-corrosive BFRP steel bars. The study found that these beams exhibited inferior flexural properties compared to those made with natural aggregate. Lastly, Liang [24] explored the potential performance enhancement of BFRP (basalt fiber-reinforced polymer) reinforced ECC–coral aggregate concrete beams through four-point bending tests. The tests focused on failure modes, bending capacity, mid-span deflection, and crack width under various reinforcement ratios for both coral aggregate concrete beams and ECC–coral aggregate concrete beams.

The development of island and reef engineering projects in our country continues to rely extensively on the use of steel reinforcement. However, the bond performance between steel rebars and reef limestone concrete necessitates further exploration. Current research has predominantly focused on the bonding characteristics of BFRP (basalt fiber-reinforced polymer) and CFRP (carbon fiber-reinforced polymer) rebars with reef concrete, while studies on the bond–slip behavior of economically viable and corrosion-resistant GFRP (Glass Fiber-Reinforced Polymer) rebars with reef limestone concrete are comparatively sparse [25,26,27]. Moreover, the impact of various anchoring parameters on the bond performance between steel rebars and reef limestone concrete has not been systematically and comprehensively studied, limiting practical guidance for their application in actual island and reef construction projects. This study employs the central pullout test method aimed at investigating the effects of different types of steel rebars, rebar diameters, bond lengths, and loading rates on the bond performance with reef limestone concrete. The research will analyze the relationship curves between load and slip, bond failure modes, and variations in bond stress. Concurrently, utilizing Abaqus numerical simulations, the study will examine the characteristics of mesoscopic stress distribution to elucidate the bonding mechanisms and failure modes between the steel rebars and reef limestone concrete.

2. Materials and Test Methods

2.1. Reef Limestone Aggregate Grading

The reef limestone aggregate used in the experiments is derived from reef limestone in the South China Sea, as shown in Figure 1. Reef limestone is primarily composed of calcareous cementation, predominantly consisting of CaCO₃ [28]. Currently, reef limestone is available in large rock bodies and smaller fragments that have undergone weathering and erosion. There is a scarcity of intermediate-sized reef limestone aggregates in nature, which cannot be directly sourced. The absence of intermediate-sized reef limestone as aggregate in concrete mixing can lead to a loss of workability in the concrete. Therefore, it is necessary to crush the reef limestone rock bodies. By breaking down large blocks into the required aggregate sizes, it is possible to meet the demands for mixing reef limestone concrete.

The moisture content of the undisturbed reef limestone is relatively high. This occurs because the reef limestone contains many independent and disconnected pores, preventing the natural loss of water. If the wet reef limestone is directly crushed and mixed with concrete, the resulting strength reduction is insufficient to meet the high load requirements of island and reef construction, thus hindering practical engineering applications. Therefore, it is necessary to dry the reef limestone and remove impurities before crushing. The PE111X150-1 crusher was used to crush the reef limestone. The particle size at the crusher’s outlet ranged from 5 to 25 mm. Due to the low strength of the reef limestone, more powdery impurities were produced during the crushing process. If not removed, these impurities affect the stability of the reef limestone concrete. Therefore, after crushing, the reef limestone aggregate was screened using a ZBSX-92A impact-type standard pendulum instrument. The pendulum instrument operated at vibrations of 221 and 147 times per minute, with a pendulum stroke of 25 mm and a motor speed of 1400 rpm. After electric sieving, manual sieving continued until no debris fell from the sieve holes. The material remaining on each sieve was weighed. The allowable quality error before and after screening did not exceed 0.5%. Before the next sieving test, a brush was used to remove fine particles adhering to the sieve and base, ensuring minimal test error. The mesh diameters were 0.15, 0.3, 0.6, 1.18, 2.36, 4.75, 9.5, 13.2, and 16 mm. In the final screening results, particles ranging in size from 4.75 to 16 mm were classified as coarse aggregates, while those from 0.15 to 4.75 mm were classified as fine aggregates. The continuously graded reef limestone coarse and fine aggregate gradation is shown in Figure 2.

Following the sieving process, the bulk densities of the coarse and fine reef limestone aggregates were measured using a measuring cylinder and an electronic scale. The aggregates fell naturally into the cylinder until full, forming a conical shape at the top, which was then leveled off before weighing. The bulk density was calculated using Equation (1), with the measured bulk density of the coarse aggregate at 873 kg/m³ and that of the fine aggregate at 1392 kg/m³. The bulk density of coarse aggregate measured in the five tests was 873 kg/m³, 881 kg/m³, 865 kg/m³, 870 kg/m³, and 876 kg/m³, respectively, and the standard deviation was 6.0415. The bulk density of fine aggregate measured in the five tests is 1392 kg/m³, 1396 kg/m³, 1390 kg/m³, 1385 kg/m³, and 1397 kg/m³, and the standard deviation was 4.8476.

$${\rho }_{bu}=\frac{(m_t-m_v)\times 1000}{V}$$

(1)

where ${\rho }_{bu}$ represents the bulk density (kg/m³), $m_t$ is the total mass of the sample and the measuring cylinder (kg), $m_v$ is the mass of the measuring cylinder alone (kg), and V is the volume of the measuring cylinder (L).

The method for testing the apparent density of reef limestone aggregates followed the standards outlined in “Lightweight Aggregates and Test Methods” (GB/T17431.2-2010) [29]. Using the calculation described in Equation (2), the apparent densities for the coarse and fine aggregates were determined to be 1939.6 kg/m³ and 2698.5 kg/m³, respectively. This indicates the mass of the aggregates per unit volume, including the pores that do not permeate the aggregate.

$${\rho }_{ap}=\frac{m\times 1000}{V_t-V_r-500}$$

(2)

where ${\rho }_{ap}$ represents the apparent density (kg/m³), $m$ is the mass of the drying sample quality (kg), $V_r$ is the volume of circular metal plate (mL), and $V_t$ is the total volume of the experiment, the circular metal plate, and the water (mL).

The void ratio of the broken reef limestone was calculated as 58% using Equation (3).

$$v=(1-\frac{{\rho }_{bu}}{{\rho }_{ap}})\times 100$$

(3)

where v represents the void ratio (%), ${\rho }_{bu}$ is the bulk density of the coarse aggregate (kg/m³), and ${\rho }_{ap}$ is the apparent density of the coarse aggregate (kg/m³).

According to the aggregate design specifications, it is necessary to conduct tests for moisture content and water absorption on the crushed reef limestone aggregates. The various properties of the crushed reef limestone aggregates are presented in

Table 1. Physical properties of reef limestone aggregate.

Coarse aggregate	Bulk density (kg/m3)	Apparent density (kg/m3)	Void ratio (%)	Moisture content (%)	Water absorption (%)
	873	1939.6	59	13.2	22.9
Fine aggregate	Bulk density (kg/m3)	Apparent density (kg/m3)	Fineness modulus	Moisture content (%)	Water absorption (%)
	1392	2698.5	2.5	2.9	3.7

.

2.2. Preparation of Reinforced Reef Limestone Concrete

Before commencing the fabrication of reef limestone concrete specimens with steel reinforcement, the plastic molds with a diameter of 100 mm had to undergo pretreatment, as shown in Figure 3. This involves drilling holes on both sides of the mold, with diameters matching those of the steel bars, to ensure smooth insertion of the reinforcement without leakage of cement slurry during the casting process. Moreover, in the non-bonded regions of the mold, PVC sleeves were affixed using epoxy resin. The steel bars were cut according to the design requirements and placed into the molds. Following the proportions in

Table 2. Reef limestone concrete proportions.

Name	Portland Cement	Coarse Aggregate	Fine Aggregate	Groundwater	Fly Ash	Slag	Anti-Crack Waterproofing Agent
Contents (kg/m3)	780	700	300	250	70	150	15

, a mixture of coarse and fine reef limestone aggregates, ordinary Portland cement, water, and supplementary cementitious materials, such as fly ash and slag, was prepared to produce C30-grade reef limestone concrete. Prior to preparation, the reef limestone aggregates were immersed in water-filled containers for 12 h to remove harmful ions and impurities. To reduce the probability of shrinkage and cracking, and to enhance the impermeability and durability of the concrete, polycarboxylate superplasticizers were added during the mixing process. After thorough mixing of the constituents, the mixture was poured into the molds and subjected to vibration for compaction. The concrete was left to set in the molds for 24 h before demolding, then the specimens were transferred to a standard curing room for an additional 28 days of curing, as shown in Figure 4.

In this study, the reinforcing bars used included HRB400-grade ribbed steel bars, HPB300-grade plain round steel bars, and GFRP bars. Both the ribbed and plain round steel bars were sourced from a metal products factory in Henan, while the GFRP bars were obtained from Zhejiang Anjie Composite Materials Company. The ribbed GFRP bars, composed of epoxy resin, had a diameter of 10 mm. The rib width and length of these reinforcing bars complied with the requirements of relevant national standards.

To conduct the pullout tests, the reinforcing bars were precisely cut into short segments, each measuring 450 mm in length. During the cutting process, special attention was paid to preventing the bending of the steel bars, ensuring they remained on the same horizontal plane from the free end to the loading end, thereby guaranteeing the accuracy and validity of the experiments. The morphology of the reinforcing bars is illustrated in Figure 5. The mechanical properties of the reinforcing bars are presented in

Table 3. Strength indicators of steel bars.

Steel Bar Model	Yield Strength ${\mathit{f}}_{\mathit{y}}$ (MPa)	Tensile Strength ${\mathit{f}}_{\mathit{s}\mathit{t}}$ (MPa)	Elongation ${\mathit{\delta }}_{\mathit{s}\mathit{t}}$ (%)	Elastic Modulus $\mathit{E}$ (GPa)
HPB300	370	552	27.5	220
HRB400	472	630	22.5	208
GFRP	~	973	1.87	54.8

.

The pullout specimens used in this study were designed according to the Canadian Standards Association (CSA) standard. The designed dimensions are 100 mm × 100 mm × 100 mm. PVC sleeves were used to separate the bonded and unbonded segments of the steel reinforcement in the reef limestone concrete. The sleeves were primarily placed at the free end and the loading end of the specimens, with their lengths determined by the bonding requirements. The specific placement of the bonding sleeves is illustrated in Figure 6.

Current research findings indicate that the bonding performance of steel-reinforced reef limestone concrete is influenced by factors such as the diameter of the reinforcing bars, bond length, type of reinforcement, concrete strength, and loading rate. However, the strength of the concrete has a relatively minor impact on the bond strength between the reinforcing bars and the concrete. Taking these factors into consideration, the experimental parameters for this study were designed as follows:

• Type of reinforcement: Plain round bars, ribbed bars, and GFRP bars were selected for investigation.
• Reinforcing bar diameter: Given the widespread use of ribbed bars in construction, diameters of 8 mm, 10 mm, and 12 mm were chosen for the bond tests of steel-reinforced reef limestone concrete.
• Bond length: Bond lengths of 30 mm, 40 mm, 50 mm, and 60 mm were selected for the bond tests between the reinforcing bars and the reef limestone concrete.
• Loading rate: Five loading rates, namely 0.01 mm/min, 0.1 mm/min, 1 mm/min, 2 mm/min, and 5 mm/min, were selected for investigation.

Considering the above factors, 15 groups of specimens were designed, each containing 3 specimens, resulting in a total of 45 specimens. After casting and demolding, the specimens were cured in a standard curing room for 28 days before conducting the pullout tests. The specific specimen numbering and parameters are shown in

Table 4. Center drawing specimen number and parameter design.

Test Piece Number	Reinforcement Type	Steel Bar Diameter (mm)	Bond Length (mm)	Loading Rate (mm/min)	Number of Test Pieces
P-R10-L50-V0.4	Plain round bars	10	50	1.0	3
H-R10-L50-V0.4	Ribbed bars	10	50	1.0	3
G-R10-L50-V0.4	GFRP bars	10	50	1.0	3
H-R8-L50-V0.4	Ribbed bars	8	50	1.0	3
H-R10-L50-V0.4	Ribbed bars	10	50	1.0	3
H-R12-L50-V0.4	Ribbed bars	12	50	1.0	3
H-R10-L30-V0.4	Ribbed bars	10	30	1.0	3
H-R10-L40-V0.4	Ribbed bars	10	40	1.0	3
H-R10-L50-V0.4	Ribbed bars	10	50	1.0	3
H-R10-L60-V0.4	Ribbed bars	10	60	1.0	3
H-R10-L50-V0.01	Ribbed bars	10	50	0.01	3
H-R10-L50-V0.1	Ribbed bars	10	50	0.1	3
H-R10-L50-V1	Ribbed bars	10	50	1	3
H-R10-L50-V2	Ribbed bars	10	50	2	3
H-R10-L50-V5	Ribbed bars	10	50	5	3

(P: plain round bar, H: ribbed bar, G: GFRP bar, R: Steel bar diameter, L: Bond length, V: loading rate).

2.3. Pullout Test and Calculation Principles

In this study, the pullout tests were conducted using a hydraulic universal testing machine manufactured by Hualong Company, as shown in Figure 7. The testing machine was equipped with displacement and load sensors, enabling real-time monitoring and recording of data changes during the experimental process. The hydraulic universal testing machine had a maximum tensile force of 1000 kN, making it capable of performing high-strength tensile tests. The loading system of the testing machine was highly flexible, supporting both displacement control and stress control, as well as a combination of these two control methods, making it suitable for experiments that require precise control of deformation.

As the universal testing machine was primarily designed for tensile and compressive tests on specimens, an additional loading steel frame was required for the pullout tests, as shown in Figure 8. The loading steel frame was fabricated from high-strength steel plates and threaded steel rods. The top and bottom of the frame were composed of high-strength steel plates, while the four sides were secured with four threaded steel rods. A hole with a diameter of 40 mm was reserved at the bottom of the steel plate to allow the reinforcing bar to pass through smoothly. Using a custom-made clamp, the upper steel plate was attached to the fixture on the crossbeam of the universal testing machine, while the lower reinforcing bar was secured using the lower weighing fixture. During the experiment, the bottom crossbeam moved while the upper part remained stationary, thereby achieving the pullout of the steel-reinforced concrete.

The universal testing machine primarily collects data on the displacement of the concrete itself. However, as this study required establishing the slip relationship at the interface between the reinforcing bar and the reef limestone concrete, YHD-50 displacement sensors were installed at the free end of the reinforcing bar, as shown in Figure 9. A pre-fabricated displacement sensor fixing bracket secured the displacement sensors on both sides of the reinforcing bar. By adjusting the bottom end of the displacement sensor, it was possible to check the proper positioning of the sensors. The sensors were connected in a full-bridge configuration, and data were collected via a static strain gauge. After the specimen and displacement sensors were properly fixed, loading and data recording proceeded simultaneously, allowing for the measurement of the relative slip between the reinforcing bar and the reef limestone concrete.

The operational principle of the displacement gauge is analogous to that of both static and dynamic resistance strain bridges. When a displacement change, denoted as ΔL, occurs in the displacement gauge, the mechanical transmission mechanism induces a corresponding movement in the dual contacts on the variable resistor. This movement, through the displacement sensor and the strain bridge, leads to a change in the external bridge resistance, represented as ΔR. Consequently, this change in displacement is converted into an electrical signal output, which can be described using the general relationship formula for resistance strain gauges, as illustrated in Equation (4).

$$\epsilon =\frac{\Delta R}{\xi R}$$

(4)

where ε represents the strain value expressed in microstrain (10⁻⁶), ΔR denotes the change in resistance induced by displacement (Ω), R is the resistance of the bridge arms in the Wheatstone bridge (Ω), and ξ is the sensitivity coefficient of the displacement gauge (mv·mm⁻¹).

The loading rate for the pullout test was set at 1 mm/min, using a displacement-controlled method. During the pullout test, it was imperative to synchronize the data acquisition from the displacement sensors with the loading process. The loading was terminated when the pullout force reached 35% of the maximum load.

Bond stress between the reinforcing bar and the reef limestone concrete was calculated using the average bond stress method, considering the pulling force from the loading test system, the diameter of the reinforcing bar, and its surface area. The calculation formula is provided in Equation (5).

$$\tau =\frac{P}{\pi d{l}_n}$$

(5)

where $\tau$ is the average bond stress of reinforced concrete (MPa), P is the test pressure of the universal testing machine (N), d is the diameter of the drawn steel bar (mm), and ln is the bonding length of the steel bar (mm).

By calculating the average bond stress for the three specimens in each loading scenario, the mean bond stress of the reinforcing bar in reef limestone concrete specimens was obtained. The average bond stress and failure modes of the reinforced reef limestone concrete specimens are presented in

Table 5. Average bond stress and failure mode of reinforced reef limestone concrete specimens.

Test Piece Number	Average Bond Stress Peak (MPa)			Mean (MPa)	Failure Modes
	${\mathit{\tau }}_1$	${\mathit{\tau }}_2$	${\mathit{\tau }}_3$
P-R10-L50-V0.4	7.93	5.42	2.98	5.44	Rebar pullout failure
H-R10-L50-V0.4	21.56	21.78	17.68	20.34	Concrete splitting failure
G-R10-L50-V0.4	17.32	15.89	16.57	16.59	Concrete splitting failure
H-R8-L50-V0.4	26.58	27.06	26.77	26.80	Rebar fracture failure
H-R10-L50-V0.4	21.56	21.78	17.68	20.34	Concrete splitting failure
H-R12-L50-V0.4	18.88	17.63	18.52	18.34	Concrete splitting failure
H-R10-L30-V0.4	25.87	24.98	25.13	25.08	Rebar pullout failure
H-R10-L40-V0.4	22.64	21.95	22.32	22.31	Concrete splitting failure
H-R10-L50-V0.4	21.56	21.78	17.68	20.34	Concrete splitting failure
H-R10-L60-V0.4	16.54	16.93	16.74	16.73	Concrete splitting failure
H-R10-L50-V0.01	18.32	18.61	18.35	18.42	Concrete splitting failure
H-R10-L50-V0.1	21.56	21.78	17.68	20.34	Concrete splitting failure
H-R10-L50-V1	23.96	24.57	24.19	24.24	Concrete splitting failure
H-R10-L50-V2	26.15	26.38	25.76	26.09	Concrete splitting failure
H-R10-L50-V5	29.65	30.81	31.16	30.54	Concrete splitting failure

.

3. Analysis of Test Results of Bond–Slip Failure Characteristics

3.1. Effect of Steel Type

The bond–slip curves obtained from the pullout tests of reef limestone concrete specimens with different types of reinforcing bars are presented in Figure 10. The experimental results reveal significant differences among the bond–slip curves of specimens reinforced with deformed bars, plain round bars, and GFRP bars. For the specimens incorporating ribbed bars and GFRP bars, the bond–slip curves can be divided into four distinct stages. In contrast, for the specimens reinforced with plain round bars, the bond–slip curves can be categorized into three stages.

The bond–slip curve for plain round bars is depicted in Figure 10a. Initially, as loading commences, the bond stress increases rapidly, reaching a peak. Once this peak is attained, the bond stress quickly declines, followed by a gradual reduction until the reinforcing bar is completely pulled out. The bond stress in plain round bars arises primarily from chemical adhesion and friction.

In the initial phase of the bond–slip response, chemical adhesion is compromised due to the onset of bar slippage, leading to a loss of the adhesive bond. In the subsequent stages, the bond stress is predominantly governed by frictional resistance. After reaching the peak stress, the bond stress undergoes a strengthening phenomenon, attributed to the accumulation of frictional forces. Consequently, as slip progresses, the bond stress increases.

The bond–slip curves for specimens with deformed bars and GFRP bars can be divided into four distinct stages:

• Micro-slip stage: In this stage, the pullout force on the reef limestone concrete is minimal, and the slip at the free end is negligible. The bond between the reinforcing bar and the reef limestone concrete is primarily due to chemical adhesion. During this stage, the frictional force and mechanical interlocking between the reinforcing bar and the reef limestone concrete do not yet play a significant role. Chemical adhesion primarily contributes to the bond strength. As the load increases, the concrete between the ribs of the deformed bar begins to sustain compressive damage, evidenced by splitting cracks on the concrete surface.
• Slip stage: After the micro-slip stage, the slip stage commences. In this stage, the reinforcing bar at the free end begins to exhibit noticeable displacement, and cracks start to appear in the concrete near the loaded end. Once cracks form, the chemical adhesion between the reinforcing bar and the reef limestone concrete loses its effectiveness. As the load continues to increase, the concrete between the ribs of the deformed bar begins to experience compressive damage, manifesting as splitting cracks on the concrete surface. This damage progresses from the bottom of the reef limestone concrete towards the top. During this stage, the bond strength is mainly provided by the frictional resistance between the ribs, with the concrete near the ribs being under compression while the regions between the ribs are under tension. The slip rate of the reinforcing bar accelerates, and the load–slip curve shows a rapid increase. When the tensile force between the ribs exceeds the tensile strength of the concrete, large splitting cracks develop, and the concrete essentially loses its load-bearing capacity.
• Descending stage: Following the development of splitting cracks in the reef limestone concrete, its load-bearing capacity is largely compromised. At this point, further application of load does not increase the bond strength; instead, it begins to decrease. During the pullout process, there is still some frictional resistance between the ribs and the concrete, but the primary contributor to the bond strength in this stage is the mechanical interlocking. As the load continues to increase, the concrete at the interlocking locations begins to fail.
• Residual stage: After the failure of the concrete at the interlocking locations between the reinforcing bar and the reef limestone concrete, the pullout force in this stage is mainly composed of the frictional resistance between the reinforcing bar and the reef limestone concrete. The bond stress remains stable within a certain range until the reinforcing bar is completely pulled out, culminating in a pullout failure.

When comparing the bond–slip curves of ribbed bars and GFRP bars, it becomes apparent that the slope of the curve for ribbed bars is steeper than that of GFRP bars. This discrepancy can be primarily attributed to the difference in elastic moduli between the two materials. Due to the higher elastic modulus of ribbed bars, their deformation (displacement) is noticeably smaller under the same stress level. In contrast, GFRP bars, composed primarily of polymers and glass fibers, exhibit a significantly lower elastic modulus, resulting in more pronounced deformation under equivalent loads. Furthermore, under identical bonding conditions, the average peak stress and average bond stress of ribbed bars are 20.44% and 18.43% higher, respectively, than those of GFRP bars. This phenomenon is likely due to the superior tensile strength of ribbed bars compared to GFRP. This enhanced tensile performance enables the steel bars to transfer more force to the surrounding concrete during the bonding process, thereby improving the bonding effect.

Further analysis of the bond–slip curves indicates that the residual stress of GFRP bars is higher than that of ribbed bars. This observation can be attributed to the frictional wear experienced by the surface of GFRP bars during the slip process in reef limestone concrete. As the pullout force increases gradually, this wear accumulates on the surface of the GFRP bars, effectively filling the gaps between the bar and the concrete interface. As the pullout force continues, the undamaged portions of the GFRP bars begin to engage with the bonding zone of the concrete. With the continued increase in load, the accumulated wear debris imposes circumferential pressure on the GFRP bars. The interaction between this circumferential pressure and the concrete interface results in additional frictional forces, attributed to the so-called ‘wedging effect’.

The failure modes of reef limestone concrete with different types of bars exhibit significant differences. Specifically, plain bars primarily exhibit pullout failure, whereas ribbed bars and GFRP bars exhibit splitting failure, as illustrated in Figure 11. This phenomenon is closely related to the tensile forces exerted by the bars and the mechanical interaction with the ribs of the concrete. Owing to the absence of mechanical interlocking between the plain bars and the reef limestone concrete, the failure mode of plain bars is characterized by pullout failure. In such cases, no splitting cracks appear on the reef limestone concrete, and the bars remain free of concrete residue. Specimens exhibiting splitting failure undergo instantaneous failure, characterized by a loud ‘popping’ sound, followed by the direct failure of the concrete. In the initial stage of splitting failure, cracks rapidly propagate with the increase in tensile stress. When the tensile stress reaches the ultimate peak, the specimens emit a loud noise, and the reef limestone concrete undergoes splitting failure, leading to the disintegration of the specimens and the formation of penetrating splitting cracks. The specimens are split into several pieces, accompanied by the sound of brittle fracture. Some specimens, despite being cracked, retain their structural integrity.

3.2. Effect of Steel Bar Diameter

The diameter of the bars significantly impacts the concrete [30,31]. Consequently, this experiment examines the influence of bar diameter. In the three groups of pullout tests, aside from the varying diameters, all other parameters and indicators remain consistent, employing bars of Φ8 mm, Φ10 mm, and Φ12 mm. The bond–slip curves for reef limestone concrete specimens with varying bar diameters in the pullout tests are depicted in Figure 12. Figure 13 displays the average bond stress for reef limestone concrete specimens across different bar diameters.

When the diameter of the bars increases from 8 mm to 12 mm, the peak stress and average bond stress of the reef limestone concrete are reduced by 30.81% and 31.56%, respectively. This suggests that the bond strength of the bars is inversely proportional to their diameter. The primary reason for this phenomenon is that larger diameter bars necessitate longer bond lengths to achieve optimal bond stress, as discussed in Section 3.3. Increased bond lengths result in reduced bond stress. Consequently, to maximize bond stress, it is essential to appropriately match the bar diameter with the bond length. Additionally, the Poisson effect results in the conversion of a portion of the radial force of the bars into longitudinal force, thus diminishing the radial force. As the diameter of the bars increases, the radial force also increases, which negatively impacts the mechanical interlocking force of the bars, and consequently, affects the bond performance between the bars and the concrete. In the pullout tests, the ‘shear lag effect’ is observed, which means that when the bars are subjected to tension, the internal and surface movement patterns of the bars differ, resulting in a non-uniform stress distribution on the surface of the bars. This effect results in uneven stress distribution across the cross-section of the bars, leading to measured bond stress that is lower than the average bond stress. As the diameter of the bars increases, the shear lag effect becomes more pronounced, leading to a discrepancy between the calculated bond stress and the actual bond stress, which is detrimental to the improvement of bond stress. Therefore, precise control of the bar diameter and understanding its relationship with bond performance becomes particularly important.

Furthermore, research [32] has demonstrated that excessively large bar diameters can induce bleeding during the casting process, leading to a partial loss of bond strength between the bars and the concrete. This reduction in contact area between the bars and the surrounding concrete further diminishes the bond performance with the reef limestone concrete.

The failure mode of reef limestone concrete with varying bar diameters consistently manifests as splitting failure, as illustrated in Figure 14. When the bar diameters were 8 mm and 10 mm, the bars and reef limestone concrete split into two pieces. However, with a 12 mm bar diameter, the number of fractured pieces in the reef limestone concrete specimens increased to three, accompanied by a relatively loud fracturing sound during the loading process.

3.3. Effect of Bond Length

Concerning the influence of bond length, this experiment involved preparing four groups of specimens with varying bond lengths to analyze their impact on the bond performance of reef limestone concrete. Figure 15 displays the average bond stress of reef limestone concrete specimens with varying bar bond lengths. Based on the analysis of the experimental results, it is evident that the bond performance between the bars and the reef limestone concrete diminishes progressively as the bond length increases. The shortest bond length, in comparison to the longest, exhibits a 25.6% reduction in bond stress. This reduction is attributable to the non-uniform distribution of bond stress. As the bond length increases, the non-uniformity of the bond stress becomes more pronounced, leading to reduced average bond stress.

The bond–slip curves for the pullout tests are depicted in Figure 16. As the bond length increases, the bond strength of the reef limestone concrete decreases correspondingly. As the bond length increases from 30 mm to 40 mm, the peak stress and average bond stress decrease by 12.48% and 11.05%, respectively. When the bond length increases from 40 mm to 50 mm, the peak stress and average bond stress decrease by 4.77% and 8.83%, respectively. As the bond length further increases from 50 mm to 60 mm, the peak stress and average bond stress decrease by 22.26% and 17.74%, respectively. This phenomenon can be primarily attributed to the non-linear distribution of bond stress. In the reef limestone concrete specimens, a longer bond length results in a shorter high-stress zone. The bond stress distribution is non-uniform, leading to reduced average bond stress. In this experiment, at a 30 mm bond length, the failure mode is pullout, accompanied by the extraction of a small amount of reef limestone concrete.

In this experiment, the failure modes of reef limestone concrete with different bond lengths primarily include pullout failure and splitting failure, as illustrated in Figure 17. Experimental observations revealed that at a bond length of 30 mm, the bond failure mode of the reinforced concrete is pullout failure. However, at bond lengths of 40 mm, 50 mm, or 60 mm, the failure mode shifts to splitting failure.

3.4. Effect of Loading Rate

The use of reinforced concrete, dynamic loads, wind loads, and impact loads, in addition to static loads, significantly impacts the service life and bond performance of reinforced concrete [33,34]. Being a strain-rate sensitive material, concrete is particularly responsive to the loading rate, which significantly influences the failure mode and the critical failure point of the material. Consequently, the loading rate is a crucial factor in determining the bond characteristics of concrete. Investigating the effect of loading rate on the bond characteristics of reef limestone concrete not only deepens our theoretical understanding of material behavior but also offers vital reference information for optimizing the design and construction of reinforced concrete structures. This research can assist engineers in designing more durable and safer structures, particularly in environments frequently subjected to dynamic loads. In this study, five loading rates were established: 0.01 mm/min, 0.1 mm/min, 1 mm/min, 2 mm/min, and 5 mm/min. In this experimental phase, all variables aside from the loading rate, including bond length, bar diameter, and concrete strength, were held constant.

From Figure 18, it is evident that when the bond length, diameter, and concrete strength of the reinforced reef limestone concrete remain constant, the effect of loading rate on the bond stress is generally positive. As the loading rate increases, the corresponding bond stress for the same slip value also rises. With an increase in loading rate, the slope of the bond–slip curve during the slip stage also becomes steeper. At a slip value of 1 mm, the bond stresses corresponding to loading rates of 0.01 mm/min, 0.1 mm/min, 1 mm/min, 2 mm/min, and 5 mm/min measure 3.74 MPa, 9.85 MPa, 11.72 MPa, 15.61 MPa, and 19.38 MPa, respectively.

From Figure 19, it is apparent that the bond–slip curves of reef limestone concrete under different loading rates are generally similar, featuring micro-slip, slip, and descending stages. In the initial loading stage, the bar slip is minimal, and the bond stiffness between the bars and the reef limestone concrete is significant. As the pullout load increases, the bond–slip relationship begins to exhibit a linear trend. During this phase, the chemical adhesion force between the bars and the reef limestone diminishes, concurrently increasing the mechanical bond force. Upon entering the slip stage, the pullout force and bond stress transition into a nonlinear phase characterized by a decreasing growth rate of the pullout force and the onset of cracks in the central region of the concrete.

For the 10 mm diameter deformed bars, the average bond strengths of reef limestone concrete at loading rates of 0.01 mm/min, 0.1 mm/min, 1 mm/min, 2 mm/min, and 5 mm/min are 18.42 MPa, 20.34 MPa, 24.24 MPa, 26.09 MPa, and 30.54 MPa, respectively. The loading rate not only affects the slip stage of the bond–slip curve but also influences the peak stress. When the loading rate increases from 0.01 mm/min to 0.1 mm/min, the peak stress and average bond stress of reef limestone concrete increase by 17.03% and 10.42%, respectively. When the loading rate increases from 0.1 mm/min to 1 mm/min, the peak stress and average bond stress of reef limestone concrete increase by 12.8% and 19.17%, respectively. When the loading rate increases from 1 mm/min to 2 mm/min, the peak stress and average bond stress of reef limestone concrete increase by 6.43% and 7.62%, respectively. When the loading rate increases from 2 mm/min to 5 mm/min, the peak stress and average bond stress of reef limestone concrete increase by 18.1% and 17.05%, respectively.

Figure 20 illustrates the bond failure modes observed during the central pullout tests of reinforced reef limestone concrete under various loading rates. The experimental findings show that the failure mode of the reef limestone concrete consistently manifested as splitting across all tested loading rates. At lower loading rates, the specimens primarily split into two segments. As the loading rate increased, the number of fragments into which the specimens split also increased.

4. Numerical Simulation of Failure Characteristics of Reinforced Reef Limestone Concrete Interface

4.1. Model Establishment and Material Constitutive

In the central pullout tests on reinforced reef limestone concrete, failure predominantly occurs at the interface, with only macroscopic failure modes visible. Changes in stress and internal damage at the bonding interface and within the concrete remain unobservable. Numerical simulations of these tests were conducted using ABAQUS/CAE software, enabling comparison with experimental results and facilitating micromechanical insights into the failure mechanisms of reinforced reef limestone concrete. In the simulations, standard steel rebars were used; the reef limestone concrete blocks were modeled as 100 mm cubes, and the rebar diameter was set at 10 mm. The reef limestone concrete was characterized using the Concrete Damaged Plasticity model available in the ABAQUS software suite. The rebar was modeled using a bilinear idealized elastoplastic constitutive relationship, while the interface between the steel and reef limestone concrete was simulated with a cohesive zone model. Material parameters, derived from mechanical tests, are detailed in the accompanying

Table 6. Mechanical parameters of test materials.

Material Properties	Density ρ (g/cm3)	Elastic Modulus E (GPa)	Poisson’s Ratio μ
Rebar	7.85	210	0.35
Reef limestone concrete	2.34	30	0.2

.

Since the reinforced reef limestone concrete represents a solid component, it was modeled using hexahedral elements. The interface between the rebar and the reef limestone concrete was simulated with an adhesion model. To further enhance the fidelity of the experimental replication, a fixed load was applied to the base boundary of the reef limestone concrete for stabilization. Subsequently, a displacement load was applied to the free end of the rebar, effectively simulating the pullout process by replicating the movement of rebar, as shown in Figure 21.

Refer to the test factors outlined in Section 3, and select diameters (8 mm, 10 mm, 12 mm) and bond lengths (30 mm, 40 mm, 50 mm, 60 mm) for analysis and comparison. Detailed information is provided in

Table 7. Simulation working condition table.

Test Piece Number	Steel Bar Diameter d (mm)	Reinforcement Type	Bonding Length L (mm)
LD1	8	Smooth round bar	50
LD2	10	Smooth round bar	50
LD3	12	Smooth round bar	50
LA1	10	Smooth round bar	30
LA2	10	Smooth round bar	40
LA3	10	Smooth round bar	50
LA4	10	Smooth round bar	60

.

4.2. Analysis and Discussion of Calculation Results

4.2.1. Comparison of Cohesion of Reinforced Reef Limestone Concrete

The results of the simulated cohesive strength of reinforced reef limestone concrete, compared with experimental data, are presented in

Table 8. Cohesion of reinforced reef limestone concrete.

Test Piece Number	Ptest (kN)	Pcal (kN)	Error (%)	δtest (mm)	δcal (mm)	Error (%)
LD1	32.98	32.42	1.69	4.57	4.34	5.03
LD2	31.93	30.84	3.41	5.03	5.19	3.18
LD3	35.09	36.07	2.79	5.26	5.37	2.09
LA1	23.53	24.71	5.01	3.41	3.58	4.98
LA2	27.21	27.69	1.76	4.52	4.63	2.43
LA3	31.93	31.48	1.4	5.26	5.31	0.95
LA4	31.77	31.52	0.79	5.37	5.49	2.23

.

As indicated in the table, the simulation results closely align with the experimental outcomes. The material constitutive model and mesh configuration used in the simulation are well-suited to the characteristics of reinforced reef limestone concrete. The study reveals that the error rates for the experiments consistently remain below 10%. Specifically, the error rates for pressure and strain in reinforced reef limestone concrete are approximately 6%.

4.2.2. Reef Limestone Concrete Stress Cloud Diagram

(1) LD group

The specimens in the LA group displayed failure modes similar to those in the reinforced reef limestone concrete of the LD group, as illustrated by the stress contour map of specimen LA3 (refer to Figure 22). The bond–slip characteristics were comparable to those of the LD group, with the primary distinction occurring in the sliding phase. During this phase, a longer bond length in the reef limestone concrete specimens resulted in a shorter high-stress region. This uneven distribution of bond stress contributed to a lower average bond stress, a consequence of the nonlinear distribution of bond stress. A comparison of the average bond stresses across groups LA1, LA2, LA3, and LA4 revealed a successive decrease, aligning with the experimental findings.

(2) LA group

The specimens in the LA group displayed failure modes similar to those in the reinforced reef limestone concrete of the LD group, as illustrated by the stress contour map of specimen LA3 (refer to Figure 23). The bond–slip characteristics closely mirrored those of the LD group, with the notable exception occurring during the sliding phase. During this phase, a longer bond length in the reef limestone concrete specimens corresponded to a shorter region of high stress. This uneven stress distribution resulted in a lower average bond stress, due to the nonlinear distribution of bond stress. A comparison of the average bond stresses across groups LA1, LA2, LA3, and LA4 revealed a progressive decrease, aligning with the experimental findings.

5. Conclusions

This paper details the creation and testing of 45 reinforced reef limestone concrete specimens, organized into 15 groups. The study primarily explores the effects of rebar type, bond length, rebar diameter, and loading rate on the load–slip curve, bond failure mode, and bond stress. The key findings are summarized as follows:

• Significant differences are observed in the bond–slip curves among ribbed rebars, plain rebars, and GFRP rebars. The bond–slip curve for ribbed and GFRP rebars can be divided into four stages: micro-slip, slip, decline, and residual stages, with the residual stage in reef limestone concrete exhibiting a wavy pattern. The predominant failure modes are pullout and splitting, with none reaching tensile strength. The bonding forces primarily involve chemical adhesion, friction, and mechanical interlock, with friction and mechanical interlock being the most significant contributors.
• The bond stress of reinforced reef limestone concrete decreases with increasing rebar diameter. Specifically, increases from 8 mm to 10 mm and from 10 mm to 12 mm result in significant reductions in peak stress and average bond stress. The average bond stress for 12 mm diameter rebar is notably lower than that for 8 mm diameter rebar.
• The average bond stress gradually decreases with increasing bond length. The stress reductions are particularly noticeable as the bond length increases from 30 mm to 60 mm, with the nonlinear distribution of bond stress considered the primary cause of this phenomenon.
• The average bond stress progressively increases with an increase in loading rate. Studies across various loading rates (0.01 mm/min to 5 mm/min) show that as the loading rate increases, the bond stress between the rebar and reef limestone concrete also increases, while the ultimate slip decreases.