Next Article in Journal
An Automated Method for Extracting and Analyzing Railway Infrastructure Cost Data
Next Article in Special Issue
Degradation Effect of Moisture on Mechanical Properties of Kevlar/PVB Composites with TiO2 Nanoparticles
Previous Article in Journal
AppSimV: A Cyber–Physical Simulation and Verification Platform for Software Applications of Intelligent Buildings
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Buildings
Volume 13
Issue 10
10.3390/buildings13102403
buildings-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Response of Reinforced Concrete Beams under the Combined Effect of Cyclic Loading and Carbonation

by
Linxuan Zhu
1,
Zhijun Zhou
1,*,
Chaoran Chen
2 and
Kangchao Wang
3
1
School of Highway, Chang’an University, Xi’an 710064, China
2
Gansu Province Transportation Planning, Survey & Design Institute Co., Ltd., Lanzhou 730030, China
3
China Civil Engineering Construction Co., Ltd., Beijing 100038, China
*
Author to whom correspondence should be addressed.
Buildings 2023, 13(10), 2403; https://doi.org/10.3390/buildings13102403
Submission received: 9 August 2023 / Revised: 10 September 2023 / Accepted: 16 September 2023 / Published: 22 September 2023
(This article belongs to the Special Issue Research on Durability and Aging on Materials and Structures in Buildings)
Browse Figures
Versions Notes

Abstract

:
To compare the deterioration mechanism of reinforced concrete beams between the combined effect of cyclic loading and carbonation and the sum of both individual factors, an optimized test procedure was introduced in this study. The macroscopic and microscopic results showed that the decrease in carbonation resistance of concrete could be attributed to the changes in pore structures and crack patterns introduced by cyclic loading. However, the carbonation process of flexural tensile concrete corresponding to different test procedures presented different trends. It indicated that the combined action of carbonation and fatigue damage was more serious than the damage caused by the effect of superposition. Finally, a theoretical carbonation model of concrete subjected to the combined damage was proposed and validated by comparing it with previous experimental results. The research findings are significant for improving the accuracy of evaluation of residual service life of reinforced concrete bridges and early warning of durability protection.

1. Introduction

The deterioration of concrete in diverse aggressive environments significantly impacts the performance of reinforced concrete (RC) structures. Meanwhile, concrete properties are mainly affected by carbonation when exposed to the atmospheric environment. Consequently, considerable research has been conducted to investigate the damage mechanisms and relevant improvements [1,2,3,4,5,6]. In addition, considering concrete in structures also bearing load, further investigations have been carried out to assess the impact of static loads on concrete carbonation [7,8,9,10,11]. It is widely recognized that there is a positive correlation between axial tensile stress and carbonation depth. However, a certain threshold exists for the axial compressive loads [12]. Unlike static loads, the traffic loading cycles can contribute to the expansion of microcracks and pores, accelerating the diffusion of CO2 in concrete [13]. Empirical evidence indicates that damaged and cracked concrete is more susceptible to corrosion than intact concrete [14,15,16]. The passive film over the rebar deteriorates, and electrochemical reactions initiate when the aggressive ion concentration reaches a threshold. As the volume of rust production increases, cracks within the concrete propagate, leading to decreased bond performance and a reduction in the bearing capacity of RC structures [17,18,19,20,21]. The phenomenon of accelerated carbonation due to fatigue loading has also been substantiated through carbonation tests conducted on fatigued concrete [22,23,24,25].
However, there are still some limitations in existing studies focusing on the deterioration mechanism of concrete subjected to carbonation and fatigue damage. Most current research used the one-time alternating test method, that is, carried out carbonation tests nafter fatigue loading or carried out fatigue loading after exposure to CO2 [26]. For instance, Song et al. [27] experimentally investigated the carbonation process of concrete with different initial degrees of fatigue damage and various environmental factors. Jiang et al. [28] conducted carbonation tests on fatigue-damaged concrete with uniform compressive, gradient compressive and tensile damage patterns. Those studies mainly concentrated on performance deterioration due to simple mechanical load and carbonation superposition rather than the more severe combined damage [29].
It should be noted that the corrosion in this paper refers to carbonation corrosion. Therefore, given the inconsistency between test procedures and the actual service conditions of structures, it is imperative to develop a new testing method that simulates the combined effects of fatigue damage and carbonation. This would enable a more accurate prediction of initial corrosion of steel bars in RC structures.
In this study, an optimized test method was employed to investigate the response of RC beams under the simultaneous action of cyclic loading and CO2 penetration. Disparities in the concrete carbonation process between the combined effect and the superposition of those corrosion factors were also revealed through macroanalysis and microanalysis. Finally, the improved carbonation model corresponding to the combined damage was established and verified. The research findings can provide a reference for the design thickness of concrete cover and the well-timed implementation of durability improvements for RC structures in the atmospheric environment.

2. Methodology

2.1. Materials and Specimens

The 42.5 ordinary Portland cement (complies with the Chinese standard GB 175-2007 [30]) was employed in this study with a specific surface of 363 m2/kg. The natural river sand (which complies with the Chinese standard JGJ 52-2006 [31]) obtained from Weihe was chosen as the fine aggregate with a fineness modulus of 3.40 and an apparent density of 2700 kg/m3. The coarse aggregate (complies with the Chinese standard JGJ 52-2006 [31]) consisted of continuously graded crushed gravel with a 5–15 mm size range and an apparent 3100 kg/m3 density.
A total of five RC beams were designed and cast. One beam underwent monotonic loading to determine its static capacity. Additionally, two beams were cyclically loaded before carbonation, while the remaining two beams were subjected to the combined action of cyclic loading and carbonation. In addition, several cubic concrete specimens were cast as a blank control group, exposed exclusively to CO2. The concrete mix proportions are shown in Table 1. The dimensions of the RC beam and the details of the reinforcements are given in Figure 1. Both the concrete beams and cubic specimens were cured for 28 days in a standard curing room, maintained at a temperature of 23 ± 2 °C and a relative humidity of 95%.

2.2. Load Application

2.2.1. Equipment for Loading Test

The loading equipment in the environmental chamber can simulate the mechanical properties of the specimen in real-world conditions, such as high temperatures in summer, low temperatures in winter, and rainy and snowy weather. As depicted in Figure 2, the loading equipment consisted of three main components: the loading system, the control system, and the data collection system. The loading system comprised a reaction frame, an actuator, a load-distributing girder, a fixture, and a displacement sensor, among other elements. The electromagnetic actuator was bolted to the middle of the reaction frame, and the distribution girder was connected to the bottom of the actuator. This arrangement facilitated the precise transmission of the load to the specimen following the instructions from the control system. Subsequently, the force or displacement signals were promptly fed back through the sensors to ensure the accuracy of the test. The strain of concrete and steel during the test was also collected by the stress-strain analysis system.
In contrast to the loading system, the control system and data collection system were positioned outside the environmental chamber connected to the equipment via oil-pressure pipes and sensor lines. This setup provided convenience during operation and prevented interference from the high temperature and humidity inside the chamber.
In our tests, the maximum load the equipment could apply to the specimen, both static and dynamic, was 300 kN. The length of the load-distributing girder was 500 mm, and the RC beam was subjected to a four-point flexural loading.

2.2.2. Calculation of Applied Loading

A constant amplitude sinusoidal cyclic loading was employed in this study to simulate the effect of vehicle loads. The maximum load, denoted as Fmax, was determined based on a specified load ratio, Smax. The ultimate load Fu was obtained from the static failure test of the reference beam B1, which yielded a value of 46 kN. The values of Smax selected for the specimens were 0.3 and 0.5. Equation (1) was utilized to calculate the corresponding maximum load, while the minimum load Fmin was set at 0.1 of the ultimate load. The cubic concrete specimens were subjected to an unloaded test, where Smax was equal to 0.
During the testing procedure, the beams initially experienced a mean load between Fmax and Fmin. Subsequently, fatigue damage was accumulated through constant amplitude sinusoidal cyclic loading at 4 Hz. Further details of the experimental groups and the corresponding loads can be found in Table 2.
F m a x = F u ⋅ S m a x ,

2.3. Accelerated Carbonation Test

The accelerated carbonation test process is dissipated in Figure 3. After the completion of the required curing periods, a specified cyclic load was applied to specimens R2 and R3 for 28 days, respectively. The undamaged specimens R4 and R5 were then placed sequentially on the loading device in the environmental chamber for combined damage. Simultaneously, the concrete cubes and the damaged beams B2 and B3 were also introduced into the chamber for carbonation. According to the China National Standard GB/T 50082-2009 [32], the concentration of CO2 was maintained at (20 ± 3)%, and the values of the relative humidity and temperature were controlled at (70 ± 5)% and (20 ± 2) °C, respectively.
Carbonation resistance is one of the important durability indicators of concrete [12]. In this study, the carbonation resistance of concrete was evaluated by quantifying the carbonation depth. Figure 3 shows a split surface sprayed with phenolphthalein solution. The carbonated regions exhibited a colorless appearance, while the non-carbonated area remained red. The carbonation depth was measured by a concrete carbonation depth ruler. Before measurement, the irregular concrete surface was smoothed by gentle abrasion with fine abrasive paper. Subsequently, place the base plane of the ruler against the concrete surface on one side of the sample hole. The instrument was then moved along the hole wall until the stylus stops at the carbonation front (the junction of the colorless and red areas). Finally, the read number of the pointer on the dial was the measured carbonation depth. The sampling locations at the bottom of RC beams are also depicted in Figure 4. Two sample holes were dry drilled at each location using an 8 mm diameter rotary impact drill. Finally, the mean carbonation depth derived from each pair of sample holes was considered the carbonation depth corresponding to different carbonation durations.

2.4. Microscopic Tests

The structure and composition of the specimen subjected to carbonation and cyclic loading were analyzed qualitatively and quantitatively by scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP), as shown in Figure 5. Samples were extracted from the tensile zone of the concrete beam, corroded by the combination of carbonation and cyclic loading, to observe the changes in microstructure structure and corrosion products over different exposure durations. Additionally, after 28-day carbonation of fatigue-damaged specimens, samples were obtained from the tensile zone to compare further the pore structure corresponding to different exposure conditions.
The sample of SEM was a concrete block with a particle size of 3 mm. It was initially fixed in the middle of two pieces of conductive adhesive on the disc. Gold spraying was then conducted on the prepared sample for 150 s to enhance its electroconductibility. This step enabled the better observation of microstructural changes in the concrete before and after corrosion using SEM.
In contrast, the MIP was conducted on a concrete block with a volume of 1 cm3 that was removed from a degraded concrete specimen. The selected samples were prepared, cleaned, and dried to a constant weight. Secondly, the dried sample was placed into the proper penetrometer and sealed. Thirdly, the initial test was carried out under low-pressure and subsequent tests under high-pressure conditions.

3. Results and Analysis

3.1. Macro-Damage Behavior

3.1.1. Crack Width of RC Beams

Figure 6 illustrates the maximum crack width observed in the flexural tensile zone for all corroded RC beams. Notably, the crack widths of B2 and B3 were measured under the dynamic action of cyclic loading without carbonation, whereas B4 and B5 were tested under the combined effect of both factors.
The first graph demonstrates a rapid increase in crack width during the first 14 days. This can be attributed to the flexural tensile stress exceeding the tensile strength of the concrete, leading to the formation of new microcracks and the propagation of existing cracks. Thereafter, the crack width only slightly increased with the number of loading cycles. There are two main reasons for the downward trend in crack width development. Firstly, as the loading cycles increase, the tensile stress is primarily borne by the steel bars rather than the concrete. Secondly, the accumulation of slight changes in the microstructure can also cause macroscopic changes. For example, the carbonization products resulted in the partial blocking of the internal pores and cracks in concrete after 14 days, which can be verified below in Section 3.2.1.
Figure 6 also indicates that the bending cracks were mainly concentrated in the pure bending zone of the RC beam, expanding from the bottom to the center of the beam after thousands of loading cycles. Besides, the number and spacing of cracks distributed in the bottom of B4 were less than those of B5, and the maximum crack width of B5 (0.12 mm) was about three times that of B4 (0.05 mm) after 28 days. The same observation applied to B2 and B3, indicating that the number and width of flexural cracks increased with increasing load levels.
In addition, the crack width curve of B2 lay above that of B4 for equal carbonation ages, a trend mirrored in B3 and B5. Further, the difference in crack width between the specimens subjected to different corrosion procedures increased with longer carbonation durations. It provided further validation of the point that carbonation slowed down the crack expansion.

3.1.2. Carbonation Depth of Concrete

For RC beams subjected to flexural loading, their lateral section could be divided into a flexural tension zone and a flexural compression zone by a neutral plane. In our tests, flexural compressed concrete experienced relatively low compressive loads, inhibiting carbonation. Considering that the tensile strength of concrete was significantly lower than its compressive strength, there exited a difference in carbonation rate, which further affected steel corrosion in concrete. Consequently, it is necessary to investigate the carbonation depth in the flexural tensile concrete specifically [33].
Figure 7 shows the carbonation depth and carbonation rate of flexural tensile concrete for different test procedures. Generally, all specimens showed an increasing trend in carbonation depth during carbonation. However, fatigue-damaged specimens exhibited greater carbonation depths than non-damaged concrete. After 28-day combined action of carbonation and cyclic loading, the carbonation depth of B2, B3, B4, and B5 were 7.45, 12.75, 8.68, 14.36 mm, respectively, which were 22.5%, 109.7%, 42.8%, and 136.2% higher than that of concrete cube A5 (6.08 mm), respectively. It can also be found from Figure 7b that the carbonation rates of specimens subjected to cyclic loading were consistently higher than those of unstressed specimens. For example, the carbonation rate of concrete cubes after 28-day carbonation was 1.16, while the rates of B4 and B5 were 1.64 and 2.174, respectively. This represented an increase of 41.26% and 133.76%, respectively. It was consistent with previous reports that fatigue damage decreased the carbonation resistance of concrete [13,26].
Different test procedures lead to differences in the carbonation results. As shown in Figure 7b, the carbonation rates of B4 and B5 were positively correlated with exposure time, while those of B2 and B3 increased and then decreased. As for the carbonation depth, the difference between B2 and B4 increased with longer exposure duration. For pre-fatigue damaged specimens with lower load levels, some cracks closed after unloading, hindering the transportation of CO2. Hence, the carbonation depths of B2 and B4 were initially similar.
Conversely, the crack width and length of the specimen under the combined effect of those corrosion factors gradually expanded with an increasing loading period, resulting in a growing disparity in the carbonation depth. However, the influence of the test procedure on specimens with higher load levels showed a different pattern: the carbonation depth of B5 started lower than that of B3 and eventually surpassed it. The carbonization rates of B3 and B5 followed the same trend. The reasons for this phenomenon are as follows. Firstly, the pre-fatigue damaged specimen under a higher load level produced large unrecoverable deformation earlier, resulting in a higher initial carbonation depth. Secondly, due to the lack of newly exposed surfaces, the carbonation rate of specimens with initial fatigue damage decreased with decreasing hydration products. Thirdly, carbonation-introduced autogenous healing also had the same effect.
It should be noticed that the effect of cyclic loading on carbonation was more significant compared to the influence of the test procedure. The carbonation depths of B3 and B5 were greater than those of B2 and B4 in total average. In addition, the carbonization rates of B3 and B5 with larger loading amplitudes were greater than 2 mm/d0.5, while those of B2 and B4 with smaller loading amplitudes did not exceed 1.7 mm/d0.5.

3.2. Micro-Damage Behavior

3.2.1. Microstructure Subjected to Combined Effect of Carbonation and Cyclic Loading

This study exclusively analyzed the microstructural variation of concrete under the combined effect of carbonation and cyclic loading. The reason for this is as follows. First, the concrete microstructures show similarities between the different corrosion procedures employed in our experiments. Second, SEM offers qualitative rather than quantitative microstructural evaluation.
Figure 8a reveals that the initial concrete structures were not dense with voids of microcracks on the surface before corrosion. This is because the concrete is mixed with a variety of materials. Their differential shrinkage in response to temperature fluctuations results in the generation of numerous tiny pores and cracks. After 14 days of combined damage, the number of microcracks increased, with some shots penetrating the aggregate and cement paste, as shown in Figure 8b. This could be attributed to the local tensile stress surpassing the tensile strength of the concrete or the occurrence of fatigue damage. The microscopic morphology of the specimen at 28 days is shown in Figure 8c. The wider cracks and lamellar concrete with the tendency of spalling were observed, indicating a further progression of fatigue damage in the concrete.
Figure 9 presents the changes in the corrosion products. Large quantities of hydrated products, such as calcium hydroxide (Ca(OH)2) and calcium silicate hydrates (C-S-H), were observed when the concrete was not corroded. After 14 days, Ca(OH)2 decreased greatly, whereas numerous carbonation products were generated, i.e., calcium carbonate (CaCO3). These products enveloped some of C-S-H, as shown in Figure 9b. Eventually, Ca(OH)2 was nearly depleted, and C-S-H was completely covered by CaCO3. Moreover, a substantial accumulation of CaCO3 was observed on the inner surface of the pores, resulting in a reduction in pore diameter and connectivity, as shown in Figure 9c.

3.2.2. Pore Structure in Corrosion Concrete

The pore structure of concrete can directly reflect the material transport characteristics. Figure 10a shows the pore size distribution of concrete with different test procedures. The most probable pore size of B2, B3, B4 and B5 was 40.29 nm, 62.56 nm, 40.30 nm and 62.59 nm, respectively. These results suggest that not only the peak strength but also the pore size notably increased as the load level increased. The pore size of B3 and B5 were increased by 22.27 nm and 22.29 nm, respectively. The larger pore sizes can be attributed to the interconnection of some of the pores. In addition, the concrete porosity was measured, where it was 7.24%, 11.07%, 10.57%, 11.54% for B2, B3, B4 and B5, respectively. Larger porosity in B3 and B5 was found, which increased by 3.83% and 0.97% compared to B2 and B4, respectively. The generation and growth of tiny voids surrounding the cracks as the external stressing increased was responsible for the increasing porosity. The results validated the view that the porosity of concrete increased with increasing fatigue damage [27].
The influence of the test procedure should also not be overlooked either. The porosity of pre-fatigue damaged concrete after carbonation was smaller than that of the specimen under the combined effect of carbonation and cyclic loading. However, the peak pore sizes were approximately the same. This observation may be indicative of the fact that the presence of carbonation in the pre-damaged concrete contributes to the blocking of the pores. Specimens that were initially damaged were carbonated in the case of unloading, avoiding the generation of new pores. Carbonation products then filled the pores among different-sized particles, leading to the smaller porosity. Conversely, the higher total porosity of concrete under the combined action of carbonation and cyclic loading was attributed to concrete brittleness, which was positively associated with carbonation.
Considering the wide distribution range of pore size in concrete, Wu [34] divided the pore diameter into four regions for simplified analysis: R1 region (<20 nm), R2 region (20–100 nm), R3 region (100–200 nm), and R4 region (>200 nm). The pores belonging to R1, R2, R3, and R4 were defined as harmless pores, low-damage pores, harmful pores, and multi-damage pores, respectively. Figure 10b illustrates the percentage of pore volume in different regions relative to the total pore volume. The majority of the pores in all concrete specimens were composed of low-damage pores and multi-damage pores. Instead, the proportion of harmless pores was relatively small, which suggested that load levels and test procedures had no significant influence on harmless pores. Furthermore, comparing B2 with B3 and B4 with B5, the proportion of low-damage pores decreased by 25.35% and 20.45%, respectively. However, the proportion of multi-damage pores increased by 20.28% and 18.74%, respectively. This also certificated the point that higher load levels resulted in more significant damage to the concrete. Besides, the percentage of harmful and multiple damage pores in B4 increased by 4.17% and 9.37%, respectively, compared to B2. Similarly, the harmful and multiple damage pores in B5 increased by 5.14% and 7.83%, respectively, compared to B3. These findings affirmed that the combined effect of carbonation and cyclic loading exerted a more pronounced influence on concrete deterioration than the sum of those individual corrosion factors.

4. Carbonation Model

As mentioned above, different test procedures lead to varying trends in the carbonation depth and the carbonation rate of concrete. The carbonation resistance of concrete is lower under the combined action of carbonation and cyclic loading. Therefore, it is necessary to establish a reliable carbonation model for predicting the initial corrosion of reinforcement in RC beams.
According to previous research, fatigue damage did not change the proportional relationships between carbonation depths of concrete and square roots of carbonation ages [28]. Thus, an influence coefficient of cyclic loading KD was introduced to qualitatively analyze the effect of cyclic loading on the carbonation depth of concrete, as shown in Equation (2).
x D = k t = K D ⋅ K 0 ⋅ t = K D ⋅ x 0 ,
where xD is the carbonation depth of fatigue-damaged concrete (mm), x0 is the carbonation depth of non-damaged concrete (mm), k is the carbonation rate coefficient, t is the carbonation age (day), K0 is the carbonation coefficient of non-damaged concrete, KD is the influence coefficient of cyclic loading.

4.1. The Carbonation Depth of Non-Damaged Concrete x0

Several prediction models have been proposed by researchers for the carbonation depths of non-damaged concrete, including Zhang and Jiang [35], Gong et al. [36], and Niu [37]. The predicted carbonation depths are compared with the experimental results in Figure 11. Upon comparison, the model proposed by Zhang and Jiang exhibited superior calculated precision, with an average relative error of 0.75. Hence, this model was selected as the fundamental carbonation model for undamaged concrete, as Equation (3) described.
x 0 = 629.25 ( 1 − R H ) 1.1 C 0 ( w / c ⋅ γ c − 0.34 ) c ⋅ γ H D ⋅ γ c t ,
where RH is the relative humidity (%), C0 is the volume concentration of CO2 (%), w is the water mass per volume (kg/m3); c is the cement mass per volume (kg/m3), γc is the coefficient of cement type, γHD is the coefficient of cement hydration.

4.2. The Influence Coefficient of Cyclic Loading KD

To accurately describe the accumulation and development process of fatigue damage in concrete under cyclic flexural loading, it is necessary to define an appropriate fatigue damage variable and ascertain its relationship with the load history. Tanaka et al. [22] defined a parameter to quantitative describe the damage degree in specimens. It was directly related to the stress amplitude and loading cycles, as shown in Equation (4).
D F = σ / f + 0.0431 ( l g N ) 1.24 ,
where DF is the fatigue damage degree, σ is the tensile stress (MPa), f is the flexural strength (MPa), N is the number of cyclic loading.
The fitting process was conducted on our test data of the carbonation depth of non-damaged concrete, indicating that K0 was 1.1056. Hence, Equation (2) can be described as follows:
K D = x D 1.1056 t ,
According to Equations (4) and (5), the results of KD and DF for the corroded specimens B4 and B5 are shown in Table 3. The corresponding least-square regression line of the coefficient KD versus the variable DF is shown as follows, with a coefficient of determination (R2) equal to 0.995.
K D = 0.6243 + 0.035 e 4.67 D F ,
By introducing Equations (3) and (6) into Equation (2), the carbonation depth of concrete under cyclic flexural loading can be obtained:
x D = ( 390.14 + 22.02 e 4.67 [ σ / f + 0.0431 ( lg N ) 1.24 ] ) ( 1 − R H ) 1.1 C 0 ( w / c ⋅ γ c − 0.34 ) c ⋅ γ c ⋅ γ H D t

4.3. Model Validation

Experimental data collected from the literature [25] were also utilized to validate the proposed carbonation model in this paper. Ordinary Portland cement was used, and essential details about the selected concrete materials are summarized in Table 1. After curing for 90 days, all specimens were subjected to repeated loads to achieve the specified fatigue damage degree by controlling the number of load cycles. Then, these damaged concrete specimens were moved into an environmental chamber for 56 days. The arrangement of the fatigue test is shown in Table 4. The reported exposure conditions for those specimens were the same as in our study.
Figure 12 shows the carbonation depth of all fatigue-damaged concrete beams calculated by the proposed model. The results obtained by different methods were almost identical. For fatigue damage degrees of 0.2, 0.4, and 0.6, the average relative errors between the predicted carbonation depth and the reported results were 7.52%, 9.68%, and 7.80%, respectively. Nevertheless, when DF = 0.8, the test results were greater than the prediction values, with the disparity growing over longer exposure periods. The difference in the definition of the degree of fatigue damage was responsible for the increasing relative error.

5. Discussion

Currently, the atmospheric concentration of CO2 is approximately 0.03%. It has been observed that 28 days of accelerated carbonation is equivalent to 51 years of natural outdoor carbonation. According to our proposed carbonation model, it would take approximately 183 years for the carbonation depth to reach the reinforcement surface if the thickness of the concrete cover is 40 mm. This duration significantly exceeds the intended design service life of the RC bridges. However, it is essential to note that the carbonation lifetimes estimated by the model are relatively conservative. This can be attributed to the following factors: (1) Complex Service Environment: The actual operating conditions experienced by the bridge are subject to various complexities, such as drying-wetting cycles, freeze-thaw cycles, and erosion caused by chloride. These environmental conditions can compromise the concrete microstructure, influencing carbonation [38,39,40,41]. (2) Artificial Factors: The impact of construction and maintenance practices must also be considered. The compactness of concrete and the uniformity of the concrete cover thickness on the girders may be compromised due to these human-induced factors. As a result, sections with thinner protective layers may experience failure sooner than expected. (3) Cracks and Carbonation Depth: The carbonation model proposed in this paper does not account for the carbonation depth at cracks. In addition to the internal voids, chloride and carbon dioxide can continuously permeate the concrete through cracks, and pitting may occur on the reinforcement after cracks become deeper and wider, which poses significant challenges for prestressed concrete structures [42,43]. There are also some potential sources of error in the experimental setup. For instance, the accelerated carbonation process maintained a constant temperature and humidity, deviating from the actual atmospheric conditions. Moreover, different mineral admixtures are commonly employed to enhance the durability of concrete, so material parameters should be considered to improve the accuracy of the carbonation model.
Durability issues caused by corrosion can lead to a reduction in the strength of concrete and the corrosion of reinforcement materials, all of which can adversely affect the load-bearing capacity of structures [26]. However, accurately evaluating the corrosion damage by traditional regression methods and numerical simulation is challenging. The possible reason is that there are many time-dependent factors affecting the corrosion. Hence, it is desirable to find an innovative method to solve the corrosion problems of RC structures and overcome the shortcomings of traditional methods. With the advancement of digital technology and the fourth industrial revolution, coupled with the availability of computing power and data, advanced analytical tools like deep learning and artificial neural networks bring powerful algorithms for performing advanced corrosion risk assessment [44,45,46,47]. For instance, the machine learning algorithm has been commonly used to evaluate the mechanical properties and corrosion properties of RC structures [48,49,50,51].
Moreover, the problem that the test conditions are inconsistent with real-world conditions can also be addressed with those tools. RC structures in offshore and marine environments suffer from the combined effect of loading, carbonation, and chloride attack during their lifetimes. Nevertheless, it is not easy to simultaneously test fatigue loading, carbonation, and chloride corrosion in concrete beams because of the complexity of operating and controlling fatigue loading simulation test equipment. Therefore, future studies can focus on the degradation mechanism and mechanical response of different elements of RC structures subjected to the coupling effect of those factors with new artificial intelligence models and algorithms.

6. Conclusions

Corrosion tests were performed on RC beams subjected to the combined effect of carbonation and cyclic loading and the superposition of both corrosion factors in this study. The degradation mechanism of the concrete beams with different test procedures was investigated both macroscopically and microscopically. A carbonation model corresponding to the combined damage was derived from theoretical analysis and indoor test data. Within the scope of our experiments, the following conclusions are drawn:
(1) Macroscopically, the width and the number of bending cracks concentrated at the bottom of the RC beam were positively correlated with loading levels and cycle numbers. Additionally, the carbonization rate was not constant for the same concrete specimen, which was greatly affected by the load level. However, the test procedure also had an important effect on the carbonization process of concrete. When the higher load level was applied, the carbonation resistance of concrete under the combination of carbonation and cyclic loading was worse than that of initially damaged specimens.
(2) Microscopically, cyclic loading did fatigue damage to the concrete, reflecting in a looser structure with numerous cross and parallel cracks, even partial spalling. The combined action of carbonation and cyclic loading resulted in more multiple-damage pores and increased porosity in concrete compared to specimens subjected to the superposition of those individual corrosion factors. The possible reason was that the pore structure of pre-damaged concrete was only affected by carbonation after unloading. The formation of CaCO3 crystals accumulated and filled up the smaller pores or made larger pores even become interconnected, leading to a reduction in size.
(3) Except for the exposure durations, the decrease in carbonation resistance was proportional to the fatigue damage caused by cyclic flexural loading. However, CaCO3 inhibited the initial development of microcracks to a certain extent and encased some uncarbonized cement hydration products, thus decelerating the carbonation rate. Considering the influence of the test procedure on the carbonation of fatigue-damaged reinforced concrete structures, a carbonation model corresponding to combined action was proposed based on theoretical analysis and empirical data. It can provide a reference for the design value of concrete cover for RC structures to be built. In addition, it furnishes a theoretical foundation for determining the optimal timing of intervention in the initial corrosion of reinforcements.

Author Contributions

Conceptualization, L.Z.; methodology, C.C.; validation, L.Z.; formal analysis, L.Z.; investigation, C.C. and K.W.; data curation, K.W.; writing—original draft preparation, L.Z.; writing—review and editing, L.Z. and Z.Z.; supervision, Z.Z.; project administration, Z.Z.; funding acquisition, Z.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by SCIENCE AND TECHNOLOGY OF SHAANXI PROVINCIAL DEPARTMENT OF TRANSPORTATION, grant number 20-13K.

Data Availability Statement

All data, models, and code generated or used during the study are presented in the submitted article.

Acknowledgments

Linxuan Zhu acknowledges the financial support from the China Scholarship Council (CSC, No. 202206567001).

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the study’s design, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.

References

  1. Coppola, L.; Beretta, S.; Bignozzi, M.C.; Bolzoni, F.; Brenna, A.; Cabrini, M.; Candamano, S.; Caputo, D.; Carsana, M.; Cioffi, R.; et al. The improvement of durability of reinforced concretes for sustainable structures: A review on different approaches. Materials 2022, 15, 2728. [Google Scholar] [CrossRef]
  2. Das, B.B.; Pandey, S.P. Influence of fineness of fly ash on the carbonation and electrical conductivity of concrete. J. Mater. Civ. Eng. 2011, 23, 1365–1368. [Google Scholar] [CrossRef]
  3. Leemann, A.; Nygaard, P.; Kaufmann, J.; Loser, R. Relation between carbonation resistance, mix design and exposure of mortar and concrete. Cem. Concr. Compos. 2015, 62, 33–43. [Google Scholar] [CrossRef]
  4. Zhang, K.; Xiao, J.; Hou, Y.; Zhang, Q. Experimental study on carbonation behavior of seawater sea sand recycled aggregate concrete. Adv. Struct. Eng. 2021, 25, 927–938. [Google Scholar] [CrossRef]
  5. Galan, I.; Andrade, C.; Castellote, M. Natural and accelerated CO2 binding kinetics in cement paste at different relative humidities. Cem. Concr. Res. 2013, 49, 21–28. [Google Scholar] [CrossRef]
  6. Visser, J.H.M. Influence of the carbon dioxide concentration on the resistance to carbonation of concrete. Constr. Build. Mater. 2014, 67, 8–13. [Google Scholar] [CrossRef]
  7. Castel, A.; François, R.; Arliguie, G. Effect of loading on carbonation penetration in reinforced concrete elements. Cem. Concr. Res. 1999, 29, 561–565. [Google Scholar] [CrossRef]
  8. De Schutter, G. Quantification of the influence of cracks in concrete structures on carbonation and chloride penetration. Mag. Concr. Res. 1999, 51, 427–435. [Google Scholar] [CrossRef]
  9. Song, H.W.; Kwon, S.J.; Byun, K.J.; Park, C.K. Predicting carbonation in early-aged cracked concrete. Cem. Concr. Res. 2006, 36, 979–989. [Google Scholar] [CrossRef]
  10. Talukdar, S.; Banthia, N.; Grace, J.R. Modelling the effects of structural cracking on carbonation front advance into concrete. Struct. Eng. Int. 2015, 6, 73–87. [Google Scholar] [CrossRef]
  11. Tang, G.B.; Yao, Y.; Wang, L.; Cao, Y.; Cui, S.P. Prediction model of concrete carbonation depth under the combined actions of stress. J. Build. Mater. 2020, 23, 304–308. (In Chinese) [Google Scholar]
  12. Tang, J.Z.; Wu, J.; Zou, Z.H.; Yue, A.Y.; Mueller, A. Influence of axial loading and carbonation age on the carbonation resistance of recycled aggregate concrete. Constr. Build. Mater. 2018, 137, 707–717. [Google Scholar] [CrossRef]
  13. Jiang, C.; Huang, Q.H.; Gu, X.L.; Zhang, W.P. Experimental investigation on carbonation in fatigue-damaged concrete. Cem. Concr. Res. 2017, 99, 38–52. [Google Scholar] [CrossRef]
  14. Chen, D.; Mahadevan, S. Chloride-induced reinforcement corrosion and concrete cracking simulation. Cem. Concr. Compos. 2008, 30, 227–238. [Google Scholar] [CrossRef]
  15. Vidal, T.; Castel, A.; François, R. Analyzing crack width to predict corrosion in reinforced concrete. Cem. Concr. Res. 2004, 34, 165–174. [Google Scholar] [CrossRef]
  16. Alonso, C.; Andrade, C.; Rodriguez, J.; Diez, J.M. Factors controlling cracking of concrete affected by reinforcement corrosion. Mater. Struct. 1998, 31, 435–441. [Google Scholar] [CrossRef]
  17. Zandi Hanjari, K.; Kettil, P.; Lundgren, K. Analysis of mechanical behavior of corroded reinforced concrete structures. ACI Struct. J. 2011, 108, 532–541. Available online: https://trid.trb.org/view/1118791 (accessed on 8 August 2023).
  18. Kiani, K.; Shodja, M.H. Prediction of the penetrated rust into the microcracks of concrete caused by reinforcement corrosion. Appl. Math. Model. 2011, 35, 2529–2543. [Google Scholar] [CrossRef]
  19. Malerba, P.; Sgambi, L.; Ielmini, D.; Gotti, G. Influence of corrosive phenomena on bearing capacity of RC and PC beams. Adv. Con. Constr. 2017, 5, 117–143. [Google Scholar] [CrossRef]
  20. Dang, V.H.; François, R. Influence of long-term corrosion in chloride environment on mechanical behaviour of RC beam. Eng. Struct. 2013, 48, 558–568. [Google Scholar] [CrossRef]
  21. Huang, L.; Ye, H.L.; Jin, X.Y.; Jin, N.G.; Xu, Z.N. Corrosion-induced shear performance degradation of reinforced concrete beams. Constr. Build. Mater. 2020, 248, 118668. [Google Scholar] [CrossRef]
  22. Tanaka, K.; Jeon, J.H.; Nawa, T.; Hashida, H. Effect of Repeated Load on Micro Structure and Carbonation of Concrete and Mortar, 8th ed.; DBMC: Strathroy, ON, Canada, 1999. [Google Scholar]
  23. Jiang, C.; Gu, X.; Zhang, W.; Zou, W. Modeling of carbonation in tensile zone of plain concrete beams damaged by cyclic loading. Constr. Build. Mater. 2015, 77, 479–488. [Google Scholar] [CrossRef]
  24. Han, J.; Liu, W.; Wang, S.; Geert, D.S.; Sun, W.; Liang, Y. Carbonation reaction and microstructural changes of metro-tunnel segment concrete coupled with static and fatigue load. J. Mater. Civ. Eng. 2016, 29, 04016216. [Google Scholar] [CrossRef]
  25. Zhou, Y.X. An Experimental Study on Carbonation of Concrete under Flexural Fatigue. Master’s Thesis, Xi’an University of Architecture and Technology, Xi’an, China, 2016. [Google Scholar]
  26. Liu, D.; Wang, C.; Gonzalez-Libreros, J.; Guo, T.; Cao, J.; Tu, Y.M.; Elfgre, L.; Sas, G. A review of concrete properties under the combined effect of fatigue and corrosion from a material perspective. Constr. Build. Mater. 2023, 369, 130489. [Google Scholar] [CrossRef]
  27. Song, L.; Liu, J.L.; Cui, C.X.; Yu, Z.W.; Fan, Z.W.; Hou, J. Carbonation Process of Reinforced Concrete Beams under the Combined Effects of Fatigue Damage and Environmental Factors. Appl. Sci. 2020, 10, 3981. [Google Scholar] [CrossRef]
  28. Jiang, C.; Huang, Q.H.; Gu, X.L.; Zhang, W.P. Modeling the effects of fatigue damage on concrete carbonation. Constr. Build. Mater. 2018, 191, 942–962. [Google Scholar] [CrossRef]
  29. Yao, Y.; Wang, L.; Wittllmann, F.H.; De Belie, N.; Schlangen, E.; Gehlen, C.; Li, J. Recommendation of RILEM TC 246-TDC: Test methods to determine durability of concrete under combined environmental actions and mechanical load. Mater. Struct. 2017, 50, 155. [Google Scholar] [CrossRef]
  30. GB 175-2007; Common Portland Cement. Standards Press of China: Beijing, China, 2007. (In Chinese)
  31. JGJ 52-2006; Standard for Technical Requirements and Test Method of Sand and Crushed Stone (Or Gravel) for Ordinary Concrete. China Building Materials Press: Beijing, China, 2006. (In Chinese)
  32. GB/T 50082-2009; Standard for Test Methods of Long-Term Performance and Durability of Ordinary Concrete. China Architecture Publishing & Media Co., Ltd.: Beijing, China, 2009. (In Chinese)
  33. Liu, Z.; Van den Heede, P.; De Belie, N. Effect of the mechanical load on the carbonation of concrete: A review of the underlying mechanisms, test methods, and results. Materials 2021, 14, 4407. [Google Scholar] [CrossRef]
  34. Wu, Z.W.; Lian, H.Z. High Performance Concrete; China Railway Publishing House: Beijing, China, 1999. [Google Scholar]
  35. Zhang, Y.; Jiang, L.X. A practical mathematical model of concrete carbonation depth cased on the mechanism. Ind. Build 1998, 1, 16–19+47. (In Chinese) [Google Scholar]
  36. Gong, L.S.; Su, M.Q.; Wang, H.L. Multi coefficient carbonation equation of concrete and its application. Concr. Reinf. Concr. 1985, 6, 10–16. (In Chinese) [Google Scholar]
  37. Niu, D.T. Durability and Life Prediction of Concrete Structures; Science Press Pub. Co.: Beijing, China, 2003. [Google Scholar]
  38. Xu, T.; Zhou, Z.; Wang, M.; Zhu, L.X.; Tian, Y.Q.; Han, D.D. Damage mechanism of pier concrete subjected to combined compressive stress, freeze-thaw, and salt attacks in saline soil. Constr. Build. Mater. 2022, 324, 126567. [Google Scholar] [CrossRef]
  39. Sun, W.; Mu, R.; Luo, X.; Miao, C. Effect of chloride salt, freeze-thaw cycling and externally applied load on the performance of the concrete. Cem. Concr. Res. 2002, 32, 1859–1864. [Google Scholar] [CrossRef]
  40. Liu, F.; Zhang, T.; Luo, T.; Zhou, M.; Zhang, K.; Ma, W. Study on the Deterioration of Concrete under Dry–Wet Cycle and Sulfate. Materials 2020, 13, 4095. [Google Scholar] [CrossRef]
  41. Liu, X.; Ma, E.; Liu, J.; Zhang, B.; Niu, D.; Wang, Y. Deterioration of an industrial reinforced concrete structure exposed to high temperatures and dry-wet cycles. Eng. Fail. Anal. 2022, 135, 106150. [Google Scholar] [CrossRef]
  42. Wang, H.; Dai, J.; Sun, X.; Zhang, X. Characteristics of concrete cracks and their influence on chloride penetration. Constr. Build. Mater. 2016, 107, 216–225. [Google Scholar] [CrossRef]
  43. Wang, X.; Val, D.; Zheng, L.; Jones, M.R. Influence of loading and cracks on carbonation of RC elements made of different concrete types. Constr. Build. Mater. 2018, 164, 12–28. [Google Scholar] [CrossRef]
  44. Wei, Z.; Fernandes, H.; Herrmann, H.G.; Tarpani, J.R.; Osman, A. A deep learning method for the impact damage segmentation of curve-shaped cfrp specimens inspected by infrared thermography. Sensors 2021, 21, 395. [Google Scholar] [CrossRef]
  45. Alamri, A.H. Application of machine learning to stress corrosion cracking risk assessment. Egypt. J. Pertol. 2022, 31, 11–21. [Google Scholar] [CrossRef]
  46. Valença, J.; Mukhandi, H.; Araújo, A.G.; Couceiro, M.S.; Júlio, E. Benchmarking for Strain Evaluation in CFRP Laminates Using Computer Vision: Machine Learning versus Deep Learning. Materials 2022, 15, 6310. [Google Scholar] [CrossRef]
  47. Jia, H.; Qiao, G.; Han, P. Machine learning algorithms in the environmental corrosion evaluation of reinforced concrete structures—A review. Cem. Concr. Compos. 2022, 133, 104725. [Google Scholar] [CrossRef]
  48. Cai, R.; Han, T.; Liao, W.; Huang, J.; Li, D.; Kumar, A.; Ma, H. Prediction of surface chloride concentration of marine concrete using ensemble machine learning. Cem. Concr. Res. 2020, 136, 106164. [Google Scholar] [CrossRef]
  49. Okazaki, Y.; Okazaki, S.; Asamoto, S.; Chun, P.J. Applicability of machine learning to a crack model in concrete bridges. Comput. Aided Civ. Infrastruct. Eng. 2020, 35, 775–792. [Google Scholar] [CrossRef]
  50. Boğa, A.R.; Öztürk, M.; Topcu, I.B. Using ANN and ANFIS to predict the mechanical and chloride permeability properties of concrete containing GGBFS and CNI. Compos. B Eng. 2013, 45, 688–696. [Google Scholar] [CrossRef]
  51. Felix, E.F.; Possan, E.; Carrazedo, R. Analysis of training parameters in the ANN learning process to mapping the concrete carbonation depth. J. Build. Pathol. Rehabilit. 2019, 4, 16. [Google Scholar] [CrossRef]
Buildings 13 02403 g001
Figure 1. Dimension and reinforcement detailing of RC beams.
Figure 1. Dimension and reinforcement detailing of RC beams.
Buildings 13 02403 g001
Buildings 13 02403 g002
Figure 2. Equipment for loading test.
Figure 2. Equipment for loading test.
Buildings 13 02403 g002
Buildings 13 02403 g003
Figure 3. Process of Accelerated Carbonation Test.
Figure 3. Process of Accelerated Carbonation Test.
Buildings 13 02403 g003
Buildings 13 02403 g004
Figure 4. The sampling locations.
Figure 4. The sampling locations.
Buildings 13 02403 g004
Buildings 13 02403 g005
Figure 5. Microscopic tests: (a) SEM; (b) MIP.
Figure 5. Microscopic tests: (a) SEM; (b) MIP.
Buildings 13 02403 g005
Buildings 13 02403 g006
Figure 6. The width and distribution of cracks. (a) SEM; (b) MIP.
Figure 6. The width and distribution of cracks. (a) SEM; (b) MIP.
Buildings 13 02403 g006
Buildings 13 02403 g007
Figure 7. The results of the carbonation test: (a) The carbonation depths of specimens; (b) The carbonation rates of specimens.
Figure 7. The results of the carbonation test: (a) The carbonation depths of specimens; (b) The carbonation rates of specimens.
Buildings 13 02403 g007
Buildings 13 02403 g008
Figure 8. Microstructure of concrete corresponding to different corrosion durations: (a) 0 days; (b) 14 days; (c) 28 days.
Figure 8. Microstructure of concrete corresponding to different corrosion durations: (a) 0 days; (b) 14 days; (c) 28 days.
Buildings 13 02403 g008
Buildings 13 02403 g009
Figure 9. Corrosion products of concrete corresponding to different corrosion durations: (a) 0 days; (b) 14 days; (c) 28 days.
Figure 9. Corrosion products of concrete corresponding to different corrosion durations: (a) 0 days; (b) 14 days; (c) 28 days.
Buildings 13 02403 g009
Buildings 13 02403 g010
Figure 10. The parameters of pore structure: (a)The pore size distribution; (b) Regional cumulative pore volume distribution.
Figure 10. The parameters of pore structure: (a)The pore size distribution; (b) Regional cumulative pore volume distribution.
Buildings 13 02403 g010
Buildings 13 02403 g011
Figure 11. Comparison of carbonation models [35,36,37].
Figure 11. Comparison of carbonation models [35,36,37].
Buildings 13 02403 g011
Buildings 13 02403 g012
Figure 12. Comparison of concrete carbonation depth corresponding to different fatigue degrees: (a) DF = 0.2; (b) DF = 0.4; (c) DF = 0.6; (d) DF = 0.8.
Figure 12. Comparison of concrete carbonation depth corresponding to different fatigue degrees: (a) DF = 0.2; (b) DF = 0.4; (c) DF = 0.6; (d) DF = 0.8.
Buildings 13 02403 g012
Table 1. Mix proportion of concrete.
Table 1. Mix proportion of concrete.
ResourceWater
(kg/m3)		Cement
(kg/m3)		Fine Aggregate
(kg/m3)		Coarse Aggregate
(kg/m3)		Water-Cement Ratio
Experiment16146068411160.35
Reference 117035067511030.49
1 The data are from Zhou et al. [25].
Table 2. Summary of the corrosion conditions corresponding to specimens.
Table 2. Summary of the corrosion conditions corresponding to specimens.
SpecimenDimension
(mm)		ProcedureSmaxFmax/Fmin
(kN)		Average Load
(kN)		Load
Amplitude
AA1–A5100 × 100 × 100Carbonation0---
BB1100 × 150 × 1700Static failure----
B2SC *0.34.6/13.89.24.6
B3SC0.54.6/2313.89.2
B4CC **0.34.6/13.89.24.6
B5CC0.54.6/2313.89.2
* SC stands for the superposition of corrosion factors, i.e., B2 and B3 were firstly subjected to cyclic loading for 28 days before carbonation; ** CC stands for the combined effect of corrosion factors, i.e., B4 and B5 were corroded by the simultaneous action of cyclic loading and carbonation.
Table 4. Arrangement of Fatigue Tests.
Table 4. Arrangement of Fatigue Tests.
Load LevelFatigue Damage DegreesResidual Strains (μm)Loading Cycles
0.25–0.75000
0.25338,047
0.410669,266
0.615993,216
0.8212109,212
Table 3. Results of KD and DF.
Table 3. Results of KD and DF.
Specimenσ/fItemExposure Durations (days)
37142128
B40.3xD2.223.725.627.318.68
KD1.15931.27171.35851.44281.4837
DF0.60170.63000.65360.66760.6776
B50.5xD3.526.089.2511.9314.36
KD1.83822.07852.23602.35470.8376
DF0.77170.80000.82362.45460.8476
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhu, L.; Zhou, Z.; Chen, C.; Wang, K. Response of Reinforced Concrete Beams under the Combined Effect of Cyclic Loading and Carbonation. Buildings 2023, 13, 2403. https://doi.org/10.3390/buildings13102403

AMA Style

Zhu L, Zhou Z, Chen C, Wang K. Response of Reinforced Concrete Beams under the Combined Effect of Cyclic Loading and Carbonation. Buildings. 2023; 13(10):2403. https://doi.org/10.3390/buildings13102403

Chicago/Turabian Style

Zhu, Linxuan, Zhijun Zhou, Chaoran Chen, and Kangchao Wang. 2023. "Response of Reinforced Concrete Beams under the Combined Effect of Cyclic Loading and Carbonation" Buildings 13, no. 10: 2403. https://doi.org/10.3390/buildings13102403

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop