Acoustic Emission Assessment of Corroded RC Columns Jointly Reinforced with Concrete Canvas and CFRP

Abstract

To explore the crack behaviors of corroded concrete columns jointly reinforced by concrete canvas (CC) and carbon fiber reinforced polymer (CFRP), a total of six specimen sets with different reinforcement forms and corrosion rates are designed and tested by acoustic emission (AE) technology. The assessed methods are AE characteristic parameters, RA-AF values, b values, and AE event spatial location map. The results show that the presence of CC changes the failure morphology of the specimen and improves the uneven deformation of concrete in the core area, which proves that the jointly reinforced method exhibits high plasticity deformation ability. The crack mode of the jointly reinforced specimen undergoes multiple transformations between tensile cracks and shear cracks, before ending in the final crush failure. The fluctuations in both amplitude and frequency of the b values of the jointly reinforced specimens increase significantly with time. The crack activity of concrete is more intense in this period, which proves that the jointly reinforced method ensures a more complete failure. Moreover, the increases in corrosion rate reduced the probability of shear cracks in the core concrete and further increased the failure degree of the specimen. The present study demonstrated that AE monitoring can effectively capture the characteristics of the cracking process of jointly reinforced concrete.

1. Introduction

Fiber reinforced polymer (FRP) is progressively used in reinforced concrete (RC) structures as the common reinforcement method, owing mainly to its attractive properties, including excellent tensile strength, lightweight, resistance to electrochemical corrosion, and ease of tailoring [1,2,3,4,5]. To date, rich research has been carried out on the application of FRP to chiefly strengthen damaged RC structures. One of the primary outcomes is that wrapping FRP around RC columns can effectively improve the bond strength between steel and concrete by providing lateral stress, thereby significantly improving compressive strength [6,7,8].

However, FRP materials also have their own limitations, including high initial costs, lack of ductility [9], and the aging and erosion of adhesive in FRP permit the intrusion of the corrosive medium into the concrete and thus decrease its anti-corrosion ability [10]. Therefore, a new model needs to be proposed to solve the above problems to ensure structural safety.

Innovation in RC structures has introduced concrete canvas (CC) to civil and military engineering due to its high density, corrosion resistance, and convenient construction. Some beneficial applications of CC include prefabricated shelter coverings, slope protections, tracks for vehicles or pedestrians, lining protections, and utilizations in the defense sector [11]. Chen et al. [12] found that 3D spacer fabric can arrest crack propagation, absorb impact energy, and improve its integrity. Niu et al. [13,14] found that the mechanical properties of the RC columns reinforced with CC and CFRP are better than those reinforced with only CFRP through axial compression tests.

One obvious feature of the reinforced structure is invisible bonding damage and covered concrete substrate cracking. Therefore, the damage needs to be detected in advance or in real time to ensure structural safety and to predict the residual life of FRP composite-RC structures in service. It can be instead detected utilizing acoustic emission (AE), a concomitant phenomenon in the process of material deformation and crack propagation [15]. AE technology is often used for damage monitoring of concrete and large structures [16], identification of failure mechanism [17,18], location of crack source [19], quantification of damage degree [20,21], etc. Przemysław et al. [22] the brittle microfracture was the most important mechanism responsible for damage growth leading to the final failure. The AE signals and AE energies were very high. Charlotte et al. [23] found that the moment of concrete macro-cracking can be derived from a sudden increase of the cumulative AE events and cumulative AE energy. This study observed the applications of CC and CFRP in concrete application and carried out axial compression tests on corroded reinforced concrete columns reinforced with these two materials. Moreover, it studied their cracking process using acoustic emission technology, providing a basis for future engineering applications and follow-up research on concrete durability in structures.

2. Experimental Program

2.1. Specimen Details

To begin, a total of six specimens were constructed in three groups with different reinforcement schemes, and their performance was monitored using acoustic emission. These RC columns were reinforced with CC and CFRP in the presence of corrosion induced by the electrochemical approach. Detailed geometrical information on the columns is shown in Figure 1, where the size of the specimen was 150 × 300 mm (diameter × height). The diameter of the longitudinal reinforcement was 10 mm. The diameter of the stirrup was 6 mm with an outer diameter of 120 mm. The thickness of the protective layer was 15 mm. The mechanical properties of the reinforcement and CFRP are listed in

Table 1. Mechanical properties of the reinforcement and CFRP.

Material Type	Strain at Ultimate Strength (×10−6)	Yield Strength (MPa)	Ultimate Strength (MPa)	Elastic Modulus (MPa)
6 mm hoop	2019	422	528	209,014
10 mm rebar	2622	527	646	200,991
CFRP	162	-	1848	233,077

. CFRP adopts CFS-I-300 g produced by Kaben Engineering Technology Co., Ltd. (Beijing, China). The type of reinforcement in the CFRP layer is transverse fabric and the weight is 300 g/m². The specimen parameters are further shown in

Table 2. Specimen parameters.

Specimen Type	Specimen Code	Theoretical Corrosion Rate (%)	CC Layers	CFRP Layers
URC group	URC-0	0	0	0
	URC-5	5	0	0
CRC group	CRC-5	5	0	2
JRC group	JRC-0	0	1	2
	JRC-5	5	1	2
	JRC-10	10	1	2

. The concrete canvas examined here consisted mainly of 3D mesh cloth containing cement powder within it. The 3D mesh cloth was chiefly woven with outer fabric and inner fiber, with a thickness of 20 mm. The 42.5 fast-hardening sulfoaluminate cement was employed. The CC was produced by Haishan New Materials Co., Ltd. (Qingdao, China). The physical characteristics of CC are shown in Figure 2. The commercial concrete containing the coarse aggregates with a continuous gradation of particle size of 5~31.5 mm crushed stone was used to construct the columns. The average compression strength of the 150 × 150 × 150 mm standard concrete prisms was 39.1 MPa after 28 days of curing at room temperature.

2.2. Reinforcement and Corrosion Induction Methodology

A schematic diagram of the reinforcement process of the specimen is shown in Figure 3. In each specimen, the exposed reinforcement at the top was cut off. The specimens in the RC group were reinforced with CFRP only by the traditional on-site wet sticking method. The lap length was 150 mm. In order to avoid damage at the column end, a layer of CFRP with a width of 50 mm was pasted on the top and bottom of the specimen. The specimen of the JC group was a column reinforced with CC and CFRP. First, the cement powder was poured into the 3D mesh cloth by the vibration method. Second, it was tied tightly and watered before being left until hardened. Finally, the CFRP was wrapped around the outer surface of CC with a lap length of 150 mm. All the corroded RC columns were placed on wooden pedestals in a cell filled with sodium chloride solution of 5% concentration by weight. The accelerated corrosion device is shown in Figure 4. When the power-on time reached the time calculated according to Faraday’s law [24] (The power-on times of the specimen with a corrosion rate of 5% and 10% were 372.7h and 746.6 h, respectively), the specimens were taken out so that the rust products on their surface could be cleaned and then the specimens were left to dry for 48 h.

2.3. Measurement Systems

A schematic diagram of the AE acquisition system is shown in Figure 5. All the column specimens were axially tested using a hydraulic servo presser machine. The machine had a capacity of 5000 kN, with assembled computer data acquisition system, which could directly record the load-displacement and load-strain curves. A loading rate of 0.26 mm/min was set during the test.

For measurement, AE signals emitted from the crack process of concrete during loading were collected by a PAC-DiSP system (manufactured by Physical Acoustics Corporation, Beijing Representative Office, Beijing, China). PAC products have been widely used in the AE research fields due to their high reliability and accuracy [25,26]. PAC R15 sensors, with an effective operation frequency ranging from 50 to 400 kHz (reported by PAC), were used here to gather AE signals. Six centrally symmetrical AE sensors were used for each column and the locations of the sensors were 30 mm vertically spaced from both ends. Silicon grease was used as the coupling agent between the sensors and the concrete surface. The gain of the preamplifier was 40 dB. The acquisition threshold was set to 40 dB to ensure the background noise was not gathered. Before the axial compression test, a lead break test was carried out to verify the responses of the sensors and the PAC-DiSP system.

3. Results and Discussion

3.1. Failure Pattern

Different failure patterns were observed in the three specimen groups as shown in Figure 6. (I) For the URC group, the increase in load expanded the corrosion micro-cracks in the direction of the longitudinal reinforcement. The specimen was divided into several prisms until the concrete was crushed (see Figure 6a). (II) For the CFRP-only reinforced group, the outer layer of CFRP partially fractured when experiencing uneven radial deformation causing large-scale failure of CFRP. This is similar to the existing research result of the columns confined with FRP [27] where the FRP failure pattern was “flaky” (see Figure 6b). (III) For the CC and CFRP jointly reinforced JC group, with an increase in load, the expansion rate of concrete was faster than those of uncorroded specimens due to corrosion of reinforcement. The FRP failure pattern was detected as the “velvet” type (see Figure 6c). After the joint reinforcement failed, the height-diameter ratio of the specimen changed, indicating that this reinforcing method can improve the plastic deformation capacity of the specimen. The current failure descriptions state only the damage phenomenon on the surface of the specimens. So, it is necessary to observe the damage process inside the specimen with a non-destructive testing method, which is the subject of discussion next.

3.2. AE Characteristic Parameters

The AE signal comes from the material’s plastic deformation and cracking process. The elastic stress wave generated by cracking in the material can be sensed and recorded by the AE sensor and converted into the waveform as depicted in Figure 7 [28]. Studying the characteristic parameters of AE signals is helpful in inferring the development of internal damage to the specimen. The widely used AE signal characteristic parameters include ringing count, amplitude, energy, frequency, duration time, and rise time.

The internal breaking behavior of the jointly reinforced specimen can be studied by analyzing the variation in the characteristic parameters of the AE signal. Figure 8 shows the variation of stress and AE characteristic parameters with loading time for the studied specimens under axial compression.

For the unreinforced specimen, since there was no reinforcement, the AE energy could be effectively released during the entire loading period. It is worth noting that after the specimen, URC-0, released the maximum crack energy, the energy rate still had a short-term load-holding capability until failure. However, for the unreinforced specimen, URC-5, the corrosion crack existed in the specimen before loading, and the energy rate did not increase significantly at the initial stage of loading.

For the specimen reinforced with CFRP only, CRC-5, the confinement effect of CFRP controlled the expansion of cracks inside the specimen within a specific range and accumulated the energy of the concrete failure. The AE signal at the initial loading stage was not strong. Then, the energy rate rapidly increased and gradually stabilized at around 0.2 × 10³ mV·10⁻⁶. Both the AE signal intensity and the energy rate increased until the yield strength of the specimen was reached. At this time, the microcracks in the specimen began to develop and expand stably, which was accompanied by the breakage of fine CFRP fibers. During the subsequent loading process, the carbon fiber layout was broken due to the uneven deformation of the specimen. The strong multi-segment AE signal indicates a multi-segment failure inside the specimen. When the CFRP fracture strain was reached, the specimen failed and was accompanied by extremely high sound and AE energy. This process quickly released the energy accumulated in the front section, resulting in a rapid increase in the energy rate and a few extremely high peaks in the AE signal. Ma et al. [29] also found that the time range of AE stable developing stage of BFRP-jacketed concrete was longer than that of the unjacketed concrete and the time range increased with an increase in FRP layers.

For the jointly reinforced specimens, JRC-0, JRC-5, and JRC-10, different from that reinforced by CFRP-only, the energy rate of the specimens did not change visibly at the initial loading stage. The accumulated energy and energy rate of AE changed when the specimen was loaded in the elastic-plastic stage. This process differed from that of the specimen with CFRP reinforcement only because CC resisted the uneven deformation of the specimen so that the specimen would not undergo obvious crack propagation before failure. When the specimen approached the peak stress, this process gradually released the energy accumulated in an increasing trend within a short period so that the energy rate increased rapidly, and the AE signal showed a continuously rising peak point. Compared with the uncorroded specimen, JRC-0, the corroded specimens, JRC-5 and JRC-10, had a longer energy release process at this stage, but the maximum energy value was not much different. The influence of corrosion rate on the jointly reinforced specimens is that in the middle and late stages of loading, the failure process of the specimens was longer, and the cumulative energy and energy rate curves changed more obviously.

The accumulated energy represents the total energy released when the specimen is damaged and reflects the damage degree of the specimen. From the result gathered in Figure 8, it can be determined that the accumulated energy of specimen URC-5 is 5.54 times that of specimen CRC-5, and the accumulated energy of specimen JRC-5 is 5.32 times that of specimen CRC-5. It shows that the damage degree of the specimen reinforced by CFRP was small, while the damage degrees of the unreinforced and jointly reinforced specimens were much higher. Adding CC can improve the energy fluctuation in the initial loading stage and ensure efficient energy accumulation. In the failure stage, CC improved the uneven deformation of the specimen and the energy release was transformed from rapid to slow release in the specimen with CFRP reinforced method.

3.3. Rise Time/Amplitude (RA) and Average Frequency (AF)

The cracking of concrete is an essential factor that causes the deterioration of concrete structures. Different cracking types display AE signals with different characteristics. According to AE characteristic parameters, material failure types can be identified. Tensile cracks occur with a short rise time and larger amplitude during fracture propagation, exhibiting a low rise time/amplitude (RA) value. In contrast, AE signals with a long rise time and low AF (average frequency) value are usually released from shear cracks [30]. RA and AF can be computed by the following equations:

RA = rise time/amplitude

(1)

AF = counts/duration

(2)

For the unreinforced concrete structures, the relationship between the crack mode and RA and AF values is shown in Figure 9a. It indicates that a shear crack usually has a higher RA value and lower AF value than a tensile crack. For FRP reinforced concrete structures, the failure mode of FRP confined concrete is crushing cracks. Therefore, the crack pattern shift is from tensile crack to crush crack for FRP confined concrete, as seen in Figure 9b.

The relationship between RA and AF values of each specimen along different loading times is shown in Figure 10, where each point represents a mean value of 50 AE events for consistent interpretation. It can be found that at the initial stage of loading, the crack mode of each specimen was mainly tensile crack. For the unreinforced specimens, URC-0 and URC-5, the data points develop rightward with time. Corrosion made higher RA value and more scattered data points of the specimen, URC-0, revealing its evolution of crack patterns from tensile crack to tensile-shear mixed crack type. For the specimen reinforced with CFRP only, CRC-5, the distribution range of AF value was larger than that of the specimen, URC-0, which means more counts and less duration. During the loading process, most of the crack patterns of the specimen were shear cracks, while only a few were crush cracks. This shows that CFRP can reduce the number of shear cracks in the concrete inside the specimen and delay the time of failure, which is consistent with the conclusion of Ma [31].

The confining pressure effects provided by CFRP in the jointly reinforced specimens, JRC-0, can be ignored at the initial stage of loading. At this stage, the crack mode was dominated by tensile cracks and supplemented by shear cracks. The distribution range of AF value was smaller than the unreinforced specimen and the specimen reinforced with CFRP during the whole loading process. The distribution range of the RA value gradually increased, which represents that the crack mode of the specimen developed from tensile crack to tensile crushing mixed crack. For the specimens, JRC-5 and JRC-10, it was found that the crack mode of the specimen was dominated by mixed cracks and supplemented by shear cracks. This demonstrates that the corrosion in the specimen would reduce the probability of shear cracks in the core concrete. It is worth noting that the sudden increase in AE events corresponds to the increase in RA value and the decrease in AF value in general. However, for this test, when the AE events of the specimen had multiple cracks at the elastic-plastic stage, RA values did not gradually increase but underwent multiple transformations between tensile cracks and shear cracks, resulting in the final crush failure. This shows that the existence of CC caused more crush cracks in the specimen, which made the failure more complete, hence improving its ductility.

3.4. B Value

The magnitude-frequency relationship (MFR) was used in earthquake seismology to describe the relationship between the occurrence frequency and the magnitude of earthquakes. Some researchers correlated that acoustic emission and earthquakes have the exact crack mechanism, and the magnitude frequency relationship is also used in the AE field [32]. Sagar et al. [33] proved that the maximum likelihood estimate could be used as a robust approach to obtaining an accurate variation of b values [19]:

$$b=\frac{20\ast {\log }_{10}e}{\overline{A}-A_{\min }}$$

(3)

where $\overline{A}$ is the arithmetic mean amplitude and A_min is the threshold magnitude.

In this test, each b value corresponds to 500 times AE events. Therefore, the density of the b value is positively related to the number of AE events per unit of time. Early research has indicated that b increases when microcracks develop. Inversely, when macrocracks begin to form, b decreases to a relatively low value.

The determined b values of each specimen are shown in Figure 11. It can be found that b does not decrease monotonously but fluctuates constantly. This means that the proportion of AE energy of concrete was unstable with great fluctuation during the whole loading process. For the unreinforced specimens, URC-0 and URC-5, it is observed that the b value tends to decrease first and then increase. This indicates that many large cracks occurred in the specimen during loading. After this part of the concrete was damaged, other parts of the concrete would continue to bear pressure until the b value dropped significantly, declaring, therefore, that the specimen had failed. Corrosion would cause the specimen to enter the plastic stage ahead of time, thus affecting the frequency of the b value such that its decrement was higher. For the specimen reinforced with CFRP only, CRC-5, it is observed that the b value has an upward trend at the initial stage of loading, which indicates that energy was accumulated at this time, and microcracks were dominant in the specimens. In the middle and late stages of loading, the b value suddenly drops, implying that the crack degree of the concrete at this stage was much larger than before, and the specimen had local macro cracks. After that, the value of b varied up and down with the development of time. This reveals that macro cracks and micro-cracks occurred at the same time in the specimen at the later stage of loading.

For the jointly reinforced specimens, JRC-0, JRC-5, and JRC-10, the fluctuation frequencies of the b value of each specimen were slow at the initial stage of loading. But the fluctuation amplitude and frequency increased significantly with time, indicating that the crack development was accelerated in the later period, and more AE signal responses were generated. Hence, the crack activity of concrete became more frequent. Compared with CRC-5, the fluctuation frequency of b of the jointly reinforced specimens is higher, but the amplitude is smaller in the middle and late stages of loading. This indicates that the damage of the jointly reinforced specimens was more complete when it failed. With the increase of corrosion rate, the frequency and amplitude of the b value fluctuation of the jointly reinforced specimens increase, and hence, the damage degree is further increased.

3.5. AE Event Spatial Location Map

The spatial positioning and development trend of the AE signal can reflect the expansion direction of the internal cracks in the concrete [34]. The number and position of AE events under different stress levels can be intuitively seen from the spatial distribution of AE signals, indicating the evolution process of damage and cracking inside the specimen. Figure 12 shows the spatial distribution of AE signals of each specimen under different stress levels.

Three general observations can be attained: (I) For URC group specimens, AE events appeared at both ends of the column at the initial loading stage, indicating that the cracks first formed at both ends of the column and then gradually developed to the middle. (II) For RC group specimens, the AE signal of the specimen was lesser at the initial loading stage. Similar to the URC group, the crack gradually developed toward the middle with increasing stress. When the ultimate strength was reached, the AE signals appeared densely along the crack development direction. The “flaky” damage characteristic of CFRP was manifested in some sparse AE signals generated on the surface of the specimen. The damage process was fast and the AE signal strength was very high. (III) For JRC group specimens, it is different from the CRC group specimen since the AE signal was scattered in the entire specimen during the loading period, indicating that small micro-cracks were uniformly generated inside the specimen. When the ultimate strength was reached, the AE signals mostly appeared in the middle of the specimen, but the number was large and widely distributed. The “velvet” damage characteristic of CFRP was noticed at the dense AE signal points on the surface of the specimen. The failure process was slow while the peak AE signal strength was not as good as that of the CRC specimen.

4. Conclusions

The axial compression test with AE monitoring of corroded RC columns jointly reinforced with concrete canvas (CC) and carbon fiber reinforced polymer (CFRP) was carried out. The failure process of the specimens was studied by AE parameter analysis, RA-AF value analysis, b value analysis, and AE event spatial location map. Some conclusions can be made as follows:

• The added CC remedies better the stress concentration of the specimen due to local failure during the loading process, makes the failure more complete, changes the failure mode of CFRP from “flaky” to “velvet”, and makes the sound of fiber failure duller.
• When the ultimate strength is reached, the AE signals of the JRC group specimen are widely distributed, and the number is large, while the CRC group specimen is narrow.
• From the AE characteristic parameters, it is observed that the CC addition can ensure an efficient accumulation and release of energy throughout the loading stage. The change range of the energy rate of the specimen increases due to corrosion in the late loading stage.
• The crack mode of the jointly reinforced specimens undergoes multiple transformations between tensile cracks and shear cracks, resulting in the final crush failure. The corrosion will reduce the probability of shear cracks in the core concrete.
• The fluctuations in both amplitude and frequency of the b values of the jointly reinforced specimens increase significantly with time. The crack activity of concrete is more intense in this period, which proves that the jointly reinforced method ensures a more complete failure.