Experimental Analysis of Steel–Concrete Bond Strength Under Varying Material and Geometric Parameters

Abstract

This study presents the outcomes of a comprehensive experimental investigation focused on the bond behavior between reinforcing steel bars and tremie concrete, assessed through standardized pull-out tests. The objective was to evaluate the influence of some key parameters: reinforcement bar diameter, concrete age (and associated compressive strength), steel fiber content, and a bentonite coating on rebar surfaces. Experiments were conducted under laboratory conditions according to relevant standards. Slip between the reinforcement and tremie concrete was measured using a sophisticated high-precision optical laser device, enabling accurate assessment of bond characteristics. A large, i.e., a statistically sufficient, number of specimens was tested, allowing the results to be analyzed using the ANOVA technique to determine the statistical significance of each parameter. The results show that, under most test conditions, the influence of the bentonite suspension coating on the bond strength was not statistically significant. Similarly, variations in the bar diameter and fiber content showed no statistically significant impact within the tested ranges. In contrast, concrete age (compressive strength) exhibited a statistically significant influence, confirming that concrete maturity is a dominant factor in bond development. The results contribute to a better understanding of the bond mechanisms in reinforced concrete and can assist in optimizing design strategies where bond performance is critical.

1. Introduction

Researchers and scientists worldwide are actively engaged in the development of novel approaches and the enhancement of existing methods aimed at ensuring the structural resilience of buildings subjected to dynamic loading. Among the most critical types of dynamic actions are seismic loads, which appear to be increasing in both frequency and intensity in recent years. In response to these trends, numerous recent studies have focused on enhancing the seismic resilience of structural systems. Xu et al. [1] recently published a comprehensive review of various techniques currently employed to improve the seismic performance of building structures. Mata et al. [2], for example, conducted an in-depth analysis of the seismic behavior of composite moment-resisting frames incorporating slender built-up columns.

In the context of reinforced concrete structures subjected to dynamic (seismic) loading, the bond between the embedded reinforcement and the surrounding concrete also plays a crucial role, alongside the intrinsic quality of the constituent materials. The integrity of this bond directly influences the energy dissipation capacity, ductility, and overall seismic resilience of the building structures. Often, failure of reinforced concrete structures occurs through the weakening of the bond property [3]. Accordingly, current standards, especially those concerning seismic design, require the exclusive use of ribbed reinforcement bars in concrete.

As reported by several authors, e.g., [3,4,5,6], the bond between the reinforcement and surrounding concrete is ensured through three primary mechanisms: adhesion, friction between the reinforcement bar and the concrete, and mechanical interlocking of the ribs with the surrounding concrete. Adhesion, which mainly develops during the early stages of cement hydration through the binding of hydrating cement particles to the steel surface, is the dominant mechanism at low load levels. As the load increases, adhesion fails, and the bond is maintained primarily by friction and mechanical interlocking [3,5]. Naaman [7] determined that, at higher load levels, mechanical interlocking of the reinforcement ribs with the surrounding concrete becomes the predominant mechanism for ensuring the bond strength between the embedded reinforcement and concrete.

Due to the above-mentioned importance of a high-quality bond between the surrounding concrete and the embedded reinforcement in terms of the load-bearing capacity, stability, and durability of reinforced concrete structures, many researchers frequently investigate the influence of various concrete and reinforcement parameters on the quality of the bond between the reinforcement and concrete. The magnitude of the pull-out force F_pull and/or the bond stress τ between the embedded reinforcement and the surrounding concrete is generally measured for this purpose. The bond stress τ is usually determined using Equation (1), where Φ and l_e stand for the nominal diameter of the reinforcement bar and the embedment length (i.e., the length over which the reinforcement is in direct contact with the surrounding concrete), respectively. The maximum bond stress τ_max stands for the peak bond stress occurring at the maximum pull-out force F_max, as defined by Equation (2).

$$\tau ={\displaystyle \frac{F_{pull}}{\pi \Phi l_e}}$$

(1)

$${\tau }_{max}={\displaystyle \frac{F_{max}}{\pi \Phi l_e}}$$

(2)

A comprehensive review was recently conducted by Vembu et al. [3] on the influence of individual concrete and reinforcement parameters on the quality of the bond between reinforcement and the surrounding concrete matrix.

Saje and Lopatič [5] analyzed the influence of conventional ribbed steel reinforcement and various types of basalt reinforcement on the maximum bond strength τ_max between embedded reinforcement and both normal-strength and high-strength concrete. They reported that, for specimens with both steel and basalt reinforcement with a 12 mm diameter, the compressive strength of the concrete significantly affected the bond strength between the concrete and reinforcement. In both cases, the values of τ_max were approximately 70–80% higher when high-strength concrete was used than when normal-strength concrete was used. Specimens with embedded basalt reinforcement achieved approximately 30% lower τ_max values than those with ribbed steel reinforcement bars.

Gangolu et al. [8] investigated the influence of rib geometry on bond strength τ_max in high-strength concrete. They found that the τ_max value for smooth (plain) bars was about 40–50% lower than that for bars with spiral ribs. Furthermore, a spiral rib orientation yielded higher τ_max values than ribs oriented transversely (i.e., perpendicular to the longitudinal axis of the bar). The effect of the reinforcement bar diameter on τ_max was found to be minimal, whereas increasing the embedment length l_e resulted in a decrease in τ_max.

Similar findings were reported by Bashir et al. [9], who examined the effects of Φ and l_e, as well as various rib geometries, on the value of τ_max. They found that, although the pull-out force F_max generally increased with larger bar diameters and longer embedment lengths, the bond strength τ_max tended to decrease with increasing bar diameter and embedment length, regardless of the rib type or orientation. Comprehensive research on the bonding properties between rebar and recycled concrete under different degrees of corrosion was recently performed by Su [10], which showed that increasing the degree of steel corrosion initially improves the bond performance of recycled concrete, but this effect does not continue as the corrosion degree further increases. A negative impact of corrosion on the bond between reinforcement and concrete was also reported by Bidari et al. [11]. As concluded by Tao et al. [12], the bond strength between shaped steel and concrete increases as the concrete strength, cover thickness, steel fiber volume ratio, and stirrup ratio increase. The bond–slip behavior between steel reinforcement and epoxy resin concrete was investigated by Chen et al. [13]. Their studies indicated that, when compared to ordinary concrete, the bond strength was higher in the case of epoxy resin concrete. Stochino et al. [14] analyzed the influence of the presence of recycled aggregate in concrete on the mechanical behavior of composite ribbed slabs. Although the inclusion of recycled aggregate slightly reduced the basic mechanical properties of the concrete itself, it resulted in an improvement in the bond between the embedded steel sheet and the surrounding concrete, thereby enhancing the overall mechanical performance of these structural elements.

Despite numerous studies and findings recently reported in the literature, the influence of individual parameters of embedded reinforcement bars and surrounding concrete on the bond behavior between reinforcement and different types of concrete still remains under-researched. Moreover, the results from different studies are often contradictory, mainly due to the very limited number of test specimens used, which does not allow for a robust or advanced statistical analysis of the obtained results. For example, several studies indicate that an increase in the bar diameter results in a lower bond strength, while many researchers have reported contradictory results [3]. Similar contradictions in findings are also observed, for instance, in the case of the influence of steel fibers in concrete on the bond between reinforcement and concrete [3]. According to Wang et al. [15], the bond strength increases from 46.57 MPa to 56.92 MPa when the steel fiber content increases from 0% to 2%.

Addressing the outlined issue necessitates comprehensive experimental research involving a substantial sample size. The resulting dataset will facilitate an in-depth analysis of the influence of various parameters through the application of advanced statistical methodologies.

This paper presents the results of a comprehensive experimental study performed in the concrete laboratories of Igmat and Irma, Slovenia. Due to the large number of test specimens, an advanced statistical method—analysis of variance (ANOVA)—was used to analyze the experimental results, with the aim of determining the potential statistically significant influence of various reinforcement and concrete parameters on the bond behavior between embedded reinforcement and the surrounding concrete. Specifically, the influence of the reinforcement bar diameter, concrete age (and thus its compressive strength), and steel fiber content in the concrete matrix was analyzed in detail. A specific type of concrete, namely, tremie concrete, was studied, which is used for underwater foundations, piles, diaphragm walls, caissons, etc., ensuring proper placements without segregation or washout. Consequently, the effect of a bentonite coating on the reinforcement bars on the bond between the reinforcement and tremie concrete was also analyzed. The properties of all materials used were determined through experimental methods specifically suited for this type of concrete, in compliance with relevant standards under controlled laboratory conditions. The bond between the concrete and reinforcement was evaluated using the standardized pull-off test, which is widely recognized as a reliable method for assessing the quality of the concrete–reinforcement interface [3].

2. Materials and Methods

2.1. Materials

2.1.1. Concrete

For specimen preparation, tremie concrete was used, with the binder component consisting of two types of cement, i.e., CEM I 42.5 N SRO (20% of the total cement content) and CEM III/B 32.5 N—LH/SR (80%), in a total amount of 470 kg/m³. Microsilica in suspension was added at a dosage of 3.2% of the total cement mass in the fresh concrete mixture. The effective water-to-binder ratio (w/b)_eff was determined according to the following equation:

$${(w/b)}_{\mathrm{eff}}= {\displaystyle \frac{added water+water in suspension+water in chemical admixtures}{total cement content+dry portion of microsilica in suspension}}$$

(3)

Crushed limestone aggregate from the Calcit Kamnik separation plant was used, with nominal fractions of 0/2 mm (30% of the total aggregate content), 2/4 mm (29%), and 4/8 mm (16.5%) and combined 8/16 mm and 16/32 mm fractions (24.5%), along with limestone filler in the amount of 15% of the cement mass. Three chemical admixtures were used: a high-range water-reducer (superplasticizer), a set retarder, and a stabilizer. The dosages of these admixtures were determined experimentally to achieve (1) a target workability of 550 ± 100 mm and (2) an air content in the fresh concrete mixture of less than 3.0%. The workability and air content were determined using standard slump flow and air content experimental techniques, respectively (refer to

Table 1. Results of fresh concrete experiments.

Property	Label	Result	Standard
Concrete temperature	Tb	25.5 °C	EN 12350-1:2019 [16]
Consistency/workability		600/620 mm	EN 12350-8:2019 [17]
Air content	Ac	1.4% v/v	EN 12350-7:2019 [18]
Density	r	2368 kg/m3	EN 12350-6:2019 [19]
Water-to-binder ratio	(v/v)eff	0.417	SIST 1026:2016, NC [20]

).

In the case of fiber-reinforced concrete, short hooked-end steel fibers were added, which do not affect the workability or placeability of fresh concrete in the case of their amount of up to 0.5% vol. The fibers were produced from wire, with an average maximum tensile strength of R_m = 841 N/mm². All the fibers were 16 mm in length and 0.5 mm in diameter (Figure 1).

All selected fresh concrete properties were determined under standard laboratory conditions at a relative humidity of RH 60 ± 5% and an ambient temperature of 20 ± 2 °C in accordance with the applicable standards (Table 1). The results of these measurements are also presented in the table and represent the average values of three specimens.

After casting, all specimens were cured under standard laboratory conditions. At concrete ages of 28 and 90 days, selected hardened concrete properties were determined using a Zwick Z400 universal testing device (Zwick/Roell, Ulm, Germany) in accordance with the relevant standards (refer to

Table 2. Results of hardened concrete experiments after 28 and 90 days.

Property	Label	28 Days	90 Days	Standard
Compressive strength	fcm	68.4 MPa	79.3 MPa	EN 12390-3:2019 [21]
Elastic modulus	E	42.6 Gpa	42.5 Gpa	DIN 1048 [22]
Water permeability	e	9.7 mm	8.0 mm	EN 12390-8:2019 [23]

). These results are presented in Table 2 and represent the average values obtained from three specimens. The compressive strength and water permeability of the concrete was determined using 150 mm cube specimens, while 100/100/400 mm prismatic specimens were used to obtain the concrete elastic modulus.

2.1.2. Reinforcement

Ribbed steel reinforcement bars of two nominal diameters, Φ12 mm and Φ22 mm, were used in this study. The fundamental properties of the reinforcement bars, determined through a standard tensile test in accordance with EN 6892-1:2020 [24], as well as the rib ratio f_R, representing the proportion of ribs relative to the cross-sectional area of the reinforcement bar and determined according to EN ISO 15630-1:2019 [25], are summarized in

Table 3. Basic characteristics of reinforcement bars.

Φ (mm)	Rp0.2 (MPa)	Rm (MPa)	Agt (%)	fR
12	596.5	716.1	9.04	0.067
22	566.0	699.5	10.12	0.088

. In the table, R_p0.2, R_m, and A_gt represent the 0.2% proof stress, ultimate tensile strength, and total extension at maximum force, expressed as a percentage of the extensometer gauge length, respectively.

The experimental procedures and testing devices used to determine the characteristics shown in Table 3 are described in detail later in this article (Section 2.3.1).

2.1.3. Bentonite Coating

A bentonite coating is commonly used in diaphragm wall construction. In this process, the excavated trench is first filled with a bentonite slurry, after which tremie concrete is placed using the contractor method. The tremie concrete displaces the bentonite slurry, and it is essential to ensure an adequate bond between the concrete and the embedded reinforcement. In this study, the preparation of the bentonite coating followed the guidelines provided by the manufacturer, i.e., 60 kg of dry bentonite per 1 m³ of water. A bentonite coating with a thickness of approx. 1 mm was applied with a brush; this is a commonly used thickness in engineering practice.

2.2. Preparation and Number of Test Samples

2.2.1. Preparation of Test Samples

Prior to testing, test specimens were prepared in accordance with Annex D of the EN 10080:2005 standard [26]. The specimens consisted of concrete cubes with 20 cm sides and a centrally embedded reinforcement bar. Special molds with integrated openings for bar placement were used (Figure 2). The standard prescribes a minimum cube side length of 20 cm or at least 10 times the bar diameter; thus, in the case of 22 mm diameter bars, a minimum side length of 22 cm would be required. However, since no damage to the concrete specimens was observed during testing, it was concluded that the slightly smaller cube size did not affect the test results.

The total length of the reinforcement bar was approximately 1.0 m, with 50 mm protruding below the specimen and 750 mm above. In accordance with the standard, the bond between the concrete and the reinforcement was ensured only along a length of 5Φ, which corresponds to an embedment length of l_e = 60 mm for bars of Φ = 12 mm and l_e = 110 mm for Φ = 22 mm bars. On the remaining bar length—i.e., 140 mm for Φ = 12 mm and 90 mm for Φ = 22 mm—bonding was completely prevented by fitting a plastic sleeve over the bar (Figure 2).

After the proper positioning of the reinforcement bars in the molds, the specimens were filled with concrete. The placement of tremie concrete, such as that used in this study, must be performed without vibration, a requirement that was adhered to during specimen casting. Adequate consolidation, without segregation or bleeding, was achieved through the previously described optimization of the fresh concrete mix and the resulting satisfactory fresh concrete properties.

After casting, the specimens were cured for 24 h under standard laboratory conditions at a relative humidity of RH 60 ± 5% and an ambient temperature of 20 ± 2 °C before demolding. Following demolding, the specimens were cured until testing (i.e., at 28 and 90 days) in accordance with the EN 12390-2:2019 standard [27].

2.2.2. Number of Test Specimens

In total, 45 test specimens (i.e., 20 without a bentonite coating and 25 with a bentonite coating) were prepared to achieve the objective of this study. In

Table 4. Basic information of test specimens.

Parameter	Without Bentonite Coating
Reinforcement diameter (mm)	12
Concrete age (days)	28			90
Amount of fibers (kg/m3)	0	40	80	0	40	80
Number of specimens	4			4
Reinforcement diameter (mm)	22
Concrete age (days)	28			90
Amount of fibers (kg/m3)	0	40	80	0	40	80
Number of specimens	4	4	4
Parameter	With Bentonite Coating
Reinforcement diameter (mm)	12
Concrete age (days)	28			90
Amount of fibers (kg/m3)	0	40	80	0	40	80
Number of specimens	5			5
Reinforcement diameter (mm)	22
Concrete age (days)	28			90
Amount of fibers (kg/m3)	0	40	80	0	40	80
Number of specimens	5	5	5

, basic information of the test specimens used in this study is summarized.

2.3. Experimental Methods

2.3.1. Determination of Basic Characteristics of Reinforcement Bars and Rebar Ratio

The basic properties of the reinforcement bars used, presented in Table 3, were determined using a standard tensile test according to the EN 6892-1:2020 standard [24]. The test was conducted using a Zwick Z400 universal testing machine equipped with special grips designed to ensure adequate clamping of the reinforcement bars, thereby completely preventing any slippage at the gripping points during the test. The reinforcement bar specimens were loaded until failure in accordance with Method A224 of the EN 6892-1:2020 standard [24], which prescribes displacement-controlled (or strain-controlled) loading of the specimen. The tensile test setup in accordance with the EN 6892-1:2020 standard [24] is shown in Figure 3.

Deformations during the test were measured very accurately using advanced laserXtens Array HP Zwick optical (Zwick/Roell, Ulm, Germany) sensors (shown on the left side of Figure 3) with a resolution of 0.11 µm. The results of the tensile test indicate that the reinforcement bars used conform to S500B-grade reinforcing steel.

The rebar ratio f_R was determined using a specialized measurement system, PSARON HTI RIB3D, along with its dedicated software (PSARON RIB3D 2025, version 8.1.0). The optical measurement system enables 3D imaging of reinforcement bars with diameters up to 50 mm, providing an automatic measurement of their key properties (nominal diameter, rib orientation, spacing and height, and ribbed surface area). Figure 4 shows the procedure for determining f_R on a ribbed reinforcement bar.

2.3.2. Determination of Pull-Out Force Between Concrete and Embedded Reinforcement

In order to achieve the objective of this study, the pull-out test was carried out in accordance with Annex D of the EN 10080:2005 standard [26], which is known as the most commonly used method to determine the bond strength between reinforcement and concrete [3]. For testing, the specimen was mounted into a suitable adapter such that the load was applied at the upper, longer end of the reinforcement bar in an upward direction (Figure 5). Accordingly, a supporting system was used to fully restrain any upward movement of the concrete specimen during testing.

Displacement, or slip, between the reinforcement bar and the surrounding concrete was measured using a specialized optical laser system, representing a significant improvement over the standard method prescribed in the EN 10080:2005 standard [26], which recommends the use of mechanical dial gauges (refer to Figure 3).

The relative slip d between the reinforcement and the concrete was determined by measuring the change in distance between two reference points (Figure 5) using universal Zwick optical laser system software (version Test Expert 1.8). The moving reference point was fixed to the reinforcement bar (upper point in Figure 5), while the fixed reference point was attached to a custom steel plate measuring 50 × 50 mm, rigidly mounted on the bottom face of the concrete cube (lower point in Figure 5). This setup provided a higher resolution and, consequently, a very accurate measurement of bar displacement during the test.

2.3.3. Analysis of Variance and Hypothesis Testing

In addition to basic statistical analyses, an advanced statistical technique—analysis of variance (ANOVA)—was used to assess the influence of individual parameters on the maximum pull-out force F_max and the corresponding bond stress τ_max, as well as to evaluate the test results. A more detailed theoretical background of the ANOVA method can be found elsewhere in the literature, e.g., in [28].

We calculated the F-statistic and the critical F-value (denoted as F_crit) and formulated the following null (H₀) and alternative (H₁) hypotheses:

• H₀—The influence of the analyzed parameter (i.e., concrete age, reinforcement bar diameter, fiber content in concrete, or prior bentonite coating) on the value of F_max (or τ_max) is statistically significant.
• H₁—The influence of the analyzed parameter (i.e., concrete age, reinforcement bar diameter, fiber content in concrete, or prior bentonite coating) on the value of F_max (or τ_max) is not statistically significant.

If F ≥ F_crit, the null hypothesis cannot be rejected, indicating that the parameter has a statistically significant influence on F_max (or τ_max). If F < F_crit, the null hypothesis is rejected, implying that the parameter does not have a statistically significant effect. A confidence level of α = 0.05 was chosen, as is common in similar studies.

3. Results

3.1. General Presentation of d–F_pull (d–τ) Diagram

A typical diagram of the development of the pull-out force F_pull (or bond stress τ) as a function of slip d between the embedded reinforcement bar and the surrounding concrete is presented in Figure 6. The maximum pull-out force F_max (and the maximum bond stress τ_max) corresponds to the peak point on the curve shown in the figure.

As can be seen in Figure 6, the d–F_pull (or d–τ) diagram can be logically divided into four characteristic phases, clearly indicated in Figure 6:

• Phase 1: An initial increase in the pull-out force F_pull (or bond stress τ), during which the full bond is maintained between the reinforcement and the concrete (i.e., the relative slip d = 0).
• Phase 2: A linear increase in slip d with increasing pull-out force F_pull (or bond stress τ) up to the maximum values F_max (or τ_max).
• Phase 3: A nearly linear decrease in F_pull (or τ) with a further increase in d.
• Phase 4: A pronounced increase in d, with a less steep decline in F_pull (or τ).

It is estimated that, during Phase 1, the bond between the reinforcement and concrete is primarily governed by adhesion; in Phases 2 and 3, the dominant mechanisms are friction between the bar and the concrete, and mechanical interlock of the ribs into the concrete. During Phase 4, bonding is mainly ensured by mechanical interlock alone.

3.2. Influence of Different Parameters on F_max and τ_max Values

Figure 7 and Figure 8 show the influence of the reinforcement bar diameter Φ, concrete age, and amount of fibers on the F_max and τ_max values, respectively. The graphs present the average values of four results (for specimens without a coating) and five results (for specimens with a bentonite coating), along with the standard deviations from these averages.

As expected, increasing the diameter of the reinforcement bar resulted in a significantly higher maximum pull-out force F_max, both for bars without the bentonite coating and for those with the coating. For specimens without the coating, increasing the bar diameter from 12 mm to 22 mm led to an average increase in F_max of 273%. For specimens with the bentonite coating, this increase was very similar and reached 268%. In contrast, the increase in bond stress τ_max with increasing bar diameter was significantly lower, i.e., 11% without the coating and 9% with the bentonite coating. It is noticeable that, in both diameter groups, the presence of a bentonite coating resulted in somewhat lower values of F_max and τ_max. For the 12 mm bars, F_max and τ_max were on average 21% lower with the bentonite coating, while for the 22 mm bars, they were 22% lower.

The next group of results, presented in Figure 7 and Figure 8, demonstrates the effect of concrete age—and thus its compressive strength—on the maximum pull-out force F_max and maximum bond stress τ_max. In this case, the diameter of the reinforcing bars Φ was 12 mm. The presented values represent the mean results of four tests in the uncoated condition and five tests in the coated condition, including the corresponding standard deviations. As shown in Table 2, the average compressive strength of the concrete was 68.4 MPa at 28 days and 79.3 MPa at 90 days. With increasing concrete age from 28 to 90 days, both F_max and τ_max increased in specimens with and without the bentonite coating. On average, the increase in F_max and τ_max was 28% for the coated bars and 57% for the uncoated bars. At both testing ages, the application of the bentonite coating to the reinforcement resulted in somewhat lower F_max and τ_max values. Specifically, at 28 days, the coated bars exhibited a 27% reduction in both the maximum pull-out force and bond stress compared to the uncoated bars. At 90 days, this reduction increased to 56%.

The last group of results demonstrates the influence of the steel fiber content in the concrete on the F_max and τ_max values. In this case, the diameter of the reinforcing bars was 22 mm. The results represent the mean values of four tests (without prior coating) and five tests (with prior coating), along with the corresponding standard deviations from the mean. In the case of the bentonite coating, a slight increase (approx. 4%) in both F_max and τ_max was observed when 40 kg/m³ of steel fibers was added compared to the fiber-free specimens. However, in the uncoated condition, a reduction of around 6% in both F_max and τ_max was recorded with the same fiber content. Interestingly, with a further increase in the fiber content from 40 kg/m³ to 80 kg/m³, a decrease in both F_max and τ_max was observed. Specifically, in the case with the coating, the reduction amounted to approximately 11%, while in the uncoated condition, the reduction was about 13%.

In all cases, the presence of a bentonite coating on the reinforcing bars led to a reduction in both F_max and τ_max. In the fiber-free concrete, this reduction averaged 29%, whereas in the fiber-reinforced concrete, the average reduction was approximately 15%.

3.3. Statistical Analysis of Influence of Different Parameters on F_max and τ_max Values

The following section presents the results of a statistical analysis evaluating the influence of the prior coating of the reinforcing bars on the values of F_max and τ_max, conducted using the analysis of variance (ANOVA) technique. A statistically significant effect of the coating is considered to be present when the critical F-value (F_crit) exceeds the calculated F-statistic.

As shown in Figure 9, the prior coating of the reinforcing bars had a statistically significant effect on the reduction in F_max and τ_max only in two specific cases: (i) specimens aged 90 days with a bar diameter of 12 mm and (ii) specimens aged 28 days with a bar diameter of 22 mm. In all other cases, the effect of the coating on F_max and τ_max was not statistically significant.

It can be clearly realized from Figure 10 that, both in the case without the prior coating of the reinforcing bars (Figure 10a) and with the bentonite-coated bars (Figure 10b), increasing the concrete age from 28 to 90 days and increasing the bar diameter from 12 mm to 22 mm had a statistically significant effect on the increase in the maximum pull-out force F_max.

Similarly, the increase in concrete age from 28 to 90 days had a statistically significant effect on the increase in the maximum bond stress τ_max in both the coated and uncoated conditions. In contrast, unlike its effect on F_max, the influence of the bar diameter on τ_max was not statistically significant, regardless of the presence of the coating (Figure 11).

The addition of steel fibers to the concrete did not have a statistically significant effect on either the maximum pull-out force F_max (Figure 10) or the maximum bond stress τ_max (Figure 11) in any of the tested cases. This is consistent with the findings of some other authors [3], while other studies reported an increase in the bond strength with a higher content of steel fibers in the concrete [15].

4. Discussion

An adequate bond between reinforcement and the surrounding concrete is of critical importance in most structural applications. Accordingly, numerous authors have investigated the influence of various parameters on the bond quality between reinforcement and different types of concrete [2,3,4,5,6,7,8,9,10,11,12,13,14,15,29,30]. However, mainly due to the typically limited number of test specimens, the findings of different studies are often contradictory. Additionally, the influence of individual factors in the case of tremie concrete—particularly the effect of a preliminary bentonite coating on reinforcing bars—on the bond between concrete and reinforcement remains relatively poorly investigated.

To enable a more comprehensive analysis of the influence of concrete and reinforcement parameters on bond behavior, an extensive experimental campaign was conducted using a large number of specimens, and the results of this study are presented in this article. This approach allowed for the application of an advanced statistical method to assess the impact (i.e., statistically significant influence) of individual parameters on the bond strength between tremie concrete and reinforcement. The use of a novel, advanced measurement system based on laser-based, non-contact slip monitoring between concrete and reinforcement further enhances the relevance and accuracy of the reported results.

The results presented in the previous sections show that the bond stress values between the embedded ribbed reinforcement and the surrounding tremie concrete generally fall within the range of 10–17 MPa, which corresponds well with the results reported in the literature [3].

The application of the bentonite coating to the reinforcing bars resulted in a reduction in both the maximum pull-out force F_max and the bond stress τ_max in all analyzed cases. This reduction ranged, on average, between 15% and 56%. However, further detailed statistical analysis of the results revealed that statistically significant reductions in F_max and τ_max due to the coating were observed only in two cases: 90-day-old specimens with 12 mm diameter bars and 28-day-old specimens with 22 mm diameter bars. In all other cases, the presence of the bentonite coating did not result in statistically significant differences, likely due to the relatively high variability (variance) within each set of test results, which was appropriately taken into account using the AnoVa technique. This is a very important and useful finding that suggests that, in general, the presence of a bentonite coating on reinforcement does not have an important impact on the bond between embedded reinforcement and the surrounding tremie concrete.

While the increase in the bar diameter from 12 mm to 22 mm resulted in a statistically significant increase in F_max in both the coated and uncoated conditions, the effect of the bar diameter on τ_max was not statistically significant. These findings are consistent with those reported by Gangolu et al. [6], while Bashir et al. [9], Eligehausen et al. [29], and Tayeh et al. [30] observed a slight decrease in the bond stress τ_max with increasing bar diameter.

The amount of steel fibers in the concrete had no statistically significant effect on either the pull-out force F_max or the bond stress τ_max. This indicates that the presence of fibers in tremie concrete does not influence the bond between the concrete and reinforcement. This result is expected, as the fiber content does not substantially affect the properties or quality of the interfacial transition zone between the reinforcement and the cementitious matrix, which plays a key role in bond performance.

In contrast, as concrete ages, its mechanical properties improve, which enhances the interfacial transition zone and, consequently, the bond strength between the reinforcement and the cement matrix. This is reflected in the statistically significant increase in both F_max and τ_max with concrete age, from 28 to 90 days. Similar finding were reported by Wang et al. [15], who observed a positive correlation between compressive strength and bond strength in the case of ultra-high-performance manufactured sand concrete.

5. Conclusions

This paper presents the results of a comprehensive experimental study aiming to determine the influence of various parameters (including a prior bentonite coating on embedded reinforcement bars) related to tremie concrete and embedded ribbed reinforcement on the maximum pull-out force (F_max) and bond stress (τ_max) between the reinforcement and surrounding tremie concrete. Pull-out tests were conducted in accordance with Annex D of the standard EN 10080:2005 [26], with the most significant modification being the use of an advanced optical measurement system for monitoring deformations and displacements during testing, allowing for the determination of deformation/displacement values accurately and precisely. Additionally, the use of an advanced statistical technique enabled a more detailed analysis of the influence of various parameters on the quality of the bond between the embedded reinforcement and the surrounding tremie concrete.

Based on the presented results, the following key conclusions and findings can be drawn:

• The prior bentonite coating of reinforcement bars had a statistically significant effect on the reduction in both F_max and τ_max only in two cases: 90-day-old specimens with 12 mm diameter reinforcement and 28-day-old specimens with 22 mm diameter reinforcement. In all other cases, the coating did not result in a statistically significant reduction in F_max or τ_max.
• In the uncoated condition, increasing the reinforcement bar diameter from 12 mm to 22 mm led to an average increase in F_max of 273% and in τ_max of 11%. In the coated condition, the same increase in the bar diameter resulted in an average increase in F_max of 268% and in τ_max of 9%. While the increase in the bar diameter had a statistically significant effect on F_max in both the coated and uncoated conditions, its influence on τ_max was not statistically significant.
• In the uncoated condition, increasing the concrete age from 28 to 90 days resulted in a 57% increase in both F_max and τ_max. In the coated condition, this increase averaged 28%. In both cases, the increase in concrete age had a statistically significant positive effect on F_max and τ_max.
• The addition of fibers to the concrete generally led to a slight decrease in F_max and τ_max. However, this effect was not statistically significant in either the coated or uncoated condition.

The results presented in this study are derived exclusively from the standardized pull-out test methodology. While this approach enables a controlled evaluation of bond strength, it provides a limited representation of the complex mechanisms governing the interaction at the steel–concrete interface, especially in the case with a bentonite coating on the surface of the embedded reinforcement. To achieve a more in-depth understanding of bond behavior and associated failure modes, the implementation of advanced experimental techniques such as scanning electron microscopy (SEM), ultrasonic testing methods, or digital image correlation (DIC) is recommended.

Furthermore, the present investigation considers only a single type of steel fiber and a specific type of reinforcing bar that lacks any visible signs of corrosion. The inclusion of a broader range of fiber types, particularly with varying aspect ratios (length to diameter), as well as different rib geometries of the reinforcing bars and varying degrees of corrosion damage, would significantly enhance the scientific rigor and practical relevance of the study. Such parameters are known to substantially influence bond performance and should be systematically evaluated in future research. Moreover, other concrete types should be considered, in addition to the tremie concrete used in this study.

However, the results of this study reveal that it would be pertinent to further investigate the influence of additional parameters on the bond quality between concrete and reinforcement using advanced statistical analysis techniques. Moreover, other concrete types can also be considered (in addition to tremie concrete). This approach, which is based on a relatively large number of test specimens and advanced statistical analysis, would significantly contribute to enhancing the quality, durability, and, most importantly, load-bearing capacity of (dynamically loaded) reinforced concrete structures and improve insights into the effects of different parameters on the bond behavior between reinforcement and various concrete types.