Crack Classification in Steel-RC and GFRP-RC Beams with Varying Reinforcement Ratio Using AE Parameters

Abstract

The main aim of this chapter is to monitor the cracking and damage assessment in steel-reinforced concrete (steel-RC) and glass fibre polymer-reinforced concrete (GFRP-RC) beams along with varying percentages of tension reinforcement ratio. Beam specimens measuring (150 × 230 × 2100) mm were tested using a four-point bending flexural test using a universal testing machine together with an AE monitoring system. Acoustic emission (AE) has been applied for the early monitoring of steel-RC and GFRP-RC structures using AE parameters such as cumulative AE hits, average frequency, rise angle, amplitude, duration and AE XY plots to evaluate the micro and macro cracking in the steel-RC and GFRP-RC beams specimens. The most popular applications of AE signal in structural health monitoring are specified on crack monitoring, quantifying the degree of damage, and crack classification. In this research, the results indicated that the average frequency and rise angle parameter of AE signal are applied to classify the types of cracks (flexural or shear cracks) that occur in steel-RC and GFRP-RC beams along with varying percentages of tension reinforcement ratio subjected to flexural loading. As a result of these findings, the AE approach may be used to examine crack monitoring and crack classification in steel and GFRP-RC beams with different percentages of tension reinforcement ratios.

1. Introduction

The most common problem associated with coastal infrastructure in metro cities like Mumbai, Bangalore, and Chennai is corrosion, which leads to cracking (micro-macro) and resulting in gradual ageing of the structure and its components, as a result of climate change and sea-level rises [1]. Moreover, the National Association of Corrosion Engineers (NACE) has detailed the adverse impact on the Indian economy for the corrosion of reinforcement in structural components about U.S. $26.1 billion (2.4% of the nations).

Gross domestic products (GDP) is spent annually for the corrosion of infrastructure by the government of India [2]. Numerous strategies have been proposed to postpone corrosion in reinforced concrete structures to minimise these huge expenditures. Commonly suggested methods include the use of stainless steel [3] and epoxy-coated rebars [4] in place of conventional steel bars, admixing corrosion inhibitors [5], self-healing compounds in concrete [6], and use of polymer concrete. It is important to note that these methods only delay the onset of corrosion and do not prevent it entirely. Recently some researchers have also suggested the use of self-healing micro-capsules for corrosion protection of metal [6, 7]. Apart from that, these specific methods have scalability issues in structural applications and are not cost-effective. Steel corrosion is a major problem in the construction industry, and several approaches have been tried to combat it, but they have proven to be either expensive or ineffective. Civil engineers from all over the world are challenged and in search of new and non-corroded affordable construction materials as well as innovative approaches and systems to problem-solving [8].

In many civil engineering applications, fibre-reinforced polymer (FRP) composites reinforcement has been introduced as an alternative or substitute material that can replace traditional reinforcing steel [9, 10, 11]. Apart from all this, FRP is a non-corrosive material consisting of a polymer matrix reinforced with fibres [12]. The fibres are usually aramid, basalt, carbon, and glass, although other fibres such as asbestos or paper, or wood have been sometimes used. On the other hand, the polymer is usually an epoxy, vinyl ester, or polyester thermosetting plastic and phenol-formaldehyde resins are still in use. FRPs are commonly used in the aerospace, automotive, marine, and construction industries. Glass fibre polymer (GFRP) bars have been predominantly suggested for engineering applications vis-à-vis economy and specific strength properties among other fibres [13].

Being non-corrosive, GFRP composite bars has many advantages such as high strength-to-weight ratio, electromagnetic-neutrality, light-weight, ease of handling, high longitudinal tensile strength, and non-magnetic characteristics, are easily constructed, and can be tailored to satisfy performance requirements [14, 15]. FRP composites have been used as internal reinforcement in concrete, bridge decks, modular structures, formwork, and external reinforcement for strengthening and seismic upgrading in modern construction and restoration of structures due to their favourable qualities. GFRP bars can be utilised in place of steel rebars in harsh exposure situations such as coastal settings, as well as in a variety of other structural applications such as wharves, box culverts, dry docks, and retaining walls [16]. Furthermore, due to their linearly elastic stress-strain relationship up to failure, GFRP reinforcing bars react differently from typical steel reinforcing bars. Furthermore, as compared to steel-reinforced concrete members, the lower modulus of elasticity of GFRP reinforcing bars produces a significant drop in flexural stiffness of GFRP-RC members after cracking and, as a result, greater deflection/deformations under service or loading circumstances [17, 18].

As a result, the serviceability limit state is frequently used to guide the design of GFRP-RC flexure members. In addition, relevant design codes and guidelines for the use of GFRP bars in RC structures have been developed [8]. More research is still needed to provide the required confidence through a better understanding of the flexural behaviour of GFRP-RC [8] using the non-destructive acoustic emission (AE) technique. The AE technique (AET) is considered as one of the most promising techniques from various types of Non Destructive Testing (NDT) methods [19]. The AE method and other NDT methods differ in two main features. First, in AE, the energy signal originates from the sample itself making its own signal, in response to stress. Second, the AE can detect the dynamic process because of its capability to detect movement or strain, whereas most of the other methods can detect existing geometrical discontinuities or fractures [20]. Thus, AE techniques have been applied to detect the crack location [21, 22] to quantify the degree of damage [23] and to determine the crack classification [24] in concrete structures. These efforts, which will greatly improve our understanding of how concrete members reinforced with GFRP bars should be analysed, as well as the combination of these techniques, is expected to overcome the shortcomings of the respective techniques, increasing the efficiency of structural inspection and allowing for more frequent monitoring of structures.

2. AE monitoring technique

AE has been recently recognised as one of the most reliable passive tools for in-situ health monitoring of civil engineering RC structures [19]. It employs surface-mounted AE-sensors to capture the energy bursts in the form of the transient elastic-stress waves. These elastic waves are generated due to the rapid release of energy during deformation or crack propagation in RC structures during any type of loading [20, 25]. These AE sensors convert transient elastic waves into electrical signals. In AET, various AE parameters are extracted and used to correlate to damage initiation and progression in various kinds of infrastructures as well as localise and quantify it. Some of the key AE parameters reported for damage analysis are cumulative AE hits and AE counts, AE energy (MARSE), and AE signal strength are shown in Figure 1. Acoustic emission hits in AE bursts are described as the number of times an AE transient signal crosses the threshold value of the anticipated signals in a structure. As the cumulative AE hits and counts increase, it points towards damage progression in the structure and gives information about the intensity of the AE event. AE energy is the transient elastic energy released during an AE event and is measured as the area under the AE signal. A significant jump in the AE signal energy in the form of a ‘Knee” indicates severe damage in the form of macro-cracking in the structure [21, 22].

Figure 1.

Waveform of AE signal and its various parameter [26].

Recent studies have also demonstrated the potential of AE techniques to detect the onset and propagation of damage/cracking in RC structures under flexural loading. Characterisation of cracks in plain and reinforced concrete beams subjected to flexural loading up to failure has also been reported [25]. It has been reported that as the level of damage in the RC beam increases, an increase in AE parameters of AE hits, counts, AE energy, rise time, and duration has been observed. AE parameters of average frequency (AF) and rise angle (RA) have been correlated with the cracking pattern and its type-tensile or shear cracks [27]. It has been observed that AE-based Ib-value along with RA and AF has been successfully used for the evaluation of flexural deformation of RC beams under cyclic loading [27]. AET has also been recently used for monitoring the fracture behaviour of different types of composite concrete beams [28]. AE not only determines accurately the onset of cracking and monitors the development of fracture but also indicates various kinds of damage and fracture modes in the form of de-bonding and concrete cracking in these beams. Hence, it can be concluded that AET has been established as a potential NDT tool for monitoring the performance of RC structures under loading when subjected to various types of damages. In this work, the efficacy of AET to understand and compare the failure pattern of steel and GFRP reinforced concrete is explored to establish its effectiveness as a potential NDT tool for concrete structures.

In this study, steel-RC and GFRP-RC beams with varying tension reinforcement were prepared and tested under the four-point bending test associated with AE equipment. The main objectives of this chapter are to examine the behaviour of cracks at each stage of the mechanical behaviour of the RC beams from loading to failure using the AE parameter analysis-based method. Moreover, this chapter also attempts to examine the effect of the changes in the varying percentage tension reinforcement ratio of steel-RC and GFRP-RC beam and level of damage on the parameters (cumulative AE hits, amplitude, rise angle, and A-FRQ) of the AE parameter analysis-based method. Furthermore, this chapter also aims to classify the crack types and classification of damage level occurs in two differently reinforced concrete beams along with varying percentage tension reinforcement ratio.

3. Experimental program and methodology

3.1 Specimen details and test matrix

RC beams having an effective length of 2000 mm with an overhang of 50 mm on each side with 150 mm × 230 mm cross-sectional dimensions were cast (Figure 2) using design mix proportions of 1:1.47:2.54 of cement, sand, and coarse aggregates with water-cement ratio of 0.46 using IS code method [30, 31]. The average compressive strength of the concrete used in both steel and GFRP reinforced concrete beams was experimentally obtained as 35.9 MPa. Moreover, mechanical properties of steel [32] and GFRP bars [33] were determined experimentally using Universal Testing Machine (Hung Ta Make, Taiwan with 1000 kN capacity) as shown in Table 1.

Figure 2.

Longitudinal and X-section details [29].

Bar diameter	Ultimate tensile strength (MPa)		Elastic modulus (GPa)		Ultimate strain
(mm)	Steel	GFRP	Steel	GFRP	Steel	GFRP
8	530	866	201	41.3	0.0049	0.021
10	544	1092	205	43.7	0.0048	0.025
12	566	1219	210	41.20	0.0048	0.029

Table 1.

Mechanical properties of the reinforcing bars.

In the present study, two sets of beams were cast-one reinforced with traditional steel bars of Fe-500 grade bars denoted as (S-series) and the other reinforced with GFRP bars denoted as (G-series). It is important to note that S-series beams had both longitudinal as well as transverse reinforcement made of Fe 500 steel whereas G-series beams had both longitudinal as-well-as transverse reinforcement entirely made of GFRP bars.

The design of RC beams is based on the [8]. Steel-RC beams were designed as under reinforced [34] whereas GFRP-RC beams were designed as over reinforced [8]. The reinforcement ratio (ρ = 100A_st/bd%) for each set of beams was varied as 0.33, 0.52, and 1.1% based on volumetric calculations. The steel-RC and GFRP-RC beams were identified according to the series. The arrangement is in the form of A-B-C, where A is the steel or GFRP-RC beam type, B is the steel or GFRP reinforcement ratio, and C is the name of the specimen which is denoted as numeric numbers 1, 2, 3. The reinforcement details of both S- and G-series are shown in Table 2. Three specimens of each beam series were cast to ensure repeatability of results but only one beam per type of each reinforcement ratio is explained in this research effort.

Series Code		ρ (%)	Asc	Ast
S-0.33-1 S-0.33-2	G-0.33-1 G-0.33-2	0.33%	2–8 mm Ø	2–8 mm Ø
S-0.33-3	G-0.33-3
S-0.52-1 S-0.52-2 S-0.52-3	G-0.52-1 G-0.52-2 G-0.52-3	0.52%	2–8 mm Ø	2–10 mm Ø
S-1.11-1 S-1.11-2 S-1.11–3	G-1.11-1 G-1.11-2 G-1.11–3	1.11%	2–8 mm Ø	2–12 mm Ø

Table 2.

Reinforcement details in the steel-RC and GFRP-RC beams.

Asc = area of compression reinforcement; Ast = area of tension reinforcement.

The experimental investigation involves testing of steel and GFRP reinforced concrete beams in four-point flexural loading which was displacement controlled at a rate of 0.01 mm/s (Figure 3). The loads were applied at L/3 from both supports using a steel spherical roller with a hydraulically controlled load cell (Figure 3). Mid-span deflections were measured using a Linear variable differential transformer (LVDT) attached underside of the RC beam and the load-deflection data was recorded by a high, speed data acquisition system. Before the actual AE monitoring, the AE sensors were checked for sensitivity using the pencil lead break test (PLB). After a successful PLB test, the wave velocity of concrete was set to 3.5 × 10⁶ mm/s. To acquire AE signals, a threshold of 45 dB was set initially with a preamplifier gain of 40 dB as input. AE-win software was used to acquire the signals originating due to bending and subsequent cracking. The mechanical performance of the steel and GFRP reinforced beams was compared by studying load-deflection characteristics, failure modes, and the progression of visible cracking patterns and moment carrying capacity. For AE monitoring of the steel and GFRP-RC beams, six AE sensors (R6α, PAC Make) with a resonant frequency of 60 kHz were attached to the front (3 Nos) and the back-face surface of the beam (3 Nos) as shown in Figure 3. The AE sensors were attached to the beams using a Vaseline gel and held in position using cello tape till the end of the experiment of steel and GFRP-RC beams. AE signals were recorded continuously during the entire duration of the loading of the beam. From the recorded AE signals, various AE waveform parameters of amplitude and number of AE hits, their expanse, and spread obtained using AE X-Y event plots have been used to study the variation in fracture and failure pattern of steel-RC and GFRP-RC beams. The speckle pattern shown in Figure 3, is used for digital image correlation (DIC) analysis which is the future scope of the work.

Figure 3.

Acoustic emission monitoring setup [29]. (a) Schematic. (b) Actual beam with AE sensor.

4. Results and discussions

4.1 Flexural performance of steel reinforced beams

The load-deflection plot of steel-RC beams is broadly classified into three regions un-cracked elastic, cracked-elastic, and plastic zones (Figures 4 and 5). Initially, the applied loads as well as deflection are small and follow a linear relationship.

Figure 4.

Load v/s deflection plots for S-series beams.

Figure 5.

Load v/s deflection plot for G-series beams.

This zone I is named uncracked elastic zone as shown in Figure 6. With further increase in loading, a significant change and reduction in stiffness of the beam are observed with the development of hairline cracks at cracking load (P_cr) of (5.58, 7.59, and 9.54) kN with a deflection (δ_cr) of (0.61, 0.60, and 0.43) mm for S-0.33-1, S-0.52-1, and S-1.11-1 beam, respectively (Table 3). These cracks progress along the sides of the beam at constant stiffness. The cracks initiate and start becoming visible at a load of (16, 20, and 30) kN for S-0.33-1, S-0.52-1, and S-1.11-1 beam, respectively, in the tensile zone of the beam. With further increase in loading, the cracks start propagating and appear in the form of distributed flexural and shear cracks leading to steel yielding at a load (P_y) of (28.93, 44.18, and 75.09) kN, with a deflection (δ_y) of (6.27, 6.85, and 4.94) mm in S-0.33-1, S-0.52-1, and S-1.11-1 RC beams, respectively as shown in Table 3. This part of the load-deflection plot from P_cr to P_y is termed as cracked-elastic zone II.

Figure 6.

Load v/s deflection plots for S-series beams (one beam per type).

Parameters	Pcr (kN)	δcr (mm)	Py (kN)	δy (mm)	Ppeak (kN)	Pult (kN)	δult (mm)	δA ult (mm)
S-0.33-1	5.58	0.61	28.93	6.27	35.18	33.43	55.51	53.13
S-0.33-2	4.41	0.54	30.84	10.23	33.70	31.58	51.23
S-0.33-3	5.38	0.99	24.97	6.82	30.71	29.82	52.65
G-0.33-1	7.89	1.37	33.6	30.33	51.32	48.94	68.68	65.14
G-0.33-2	6.41	0.99	39.94	45.63	49.34	44.13	65.27
G-0.33-3	6.22	0.97	33.4	32.49	47.68	34.13	61.48
S-0.52-1	6.99	0.55	45.85	7.34	50.43	46.10	43.10	42.59
S-0.52-2	5.81	0.52	43.75	6.54	48.26	38.40	41.77
S-0.52-3	6.14	0.53	46.26	9.53	48.80	46.08	42.90
G-0.52-1	8.01	0.91	51.8	40.47	57.73	56.74	60.10	55.52
G-0.52-2	7.56	0.81	54.33	38.77	59.32	51.96	54.26
G-0.52-3	8.77	0.75	48.19	29.37	57.60	53.69	52.20
S-1.11-1	9.54	0.43	81.44	5.50	88.94	84.85	30.9	31.42
S-1.11-2	10.17	0.71	75.42	6.03	83.41	72.52	31.02
S-1.11–3	9.75	0.34	75.87	11.04	82.81	78.01	32.36
G-1.11-1	10.31	0.34	57.46	14.06	83.71	76.98	34.47	34.52
G-1.11-2	9.55	0.76	66.84	20.78	79.51	70.33	33.39
G-1.11–3	12.9	1.03	63.70	15.95	81.81	64.57	35.71

Table 3.

Comparison of flexural parameters of steel and GFRP reinforced beams.

In zone III named plastic zone, the concrete section is cracked and ineffective in resisting the loads and the entire load is taken by steel and yields. It is marked by an increase in the mid-span deflection 23.22, 14.47, and 12.48 mm with a minor increase in load up to a peak load (P_Peak) of (35.18, 50.80, and 88.96) kN, pointing towards larger strain at the level of steel and increase in curvature of the cracked section with an increase in the percentage of steel. Further due to the strain hardening of steel, the beams fail at an ultimate load (P_u) of (33.43, 44.96, and 82.75) kN with a deflection (δ_u) of (55.51, 43.1, and 30.9) mm in for S-0.33-1, S-0.52-1, and S-1.11-1 RC beams, respectively. In general, it is observed that with the increase in the reinforcement ratio, the ultimate load-carrying capacity increases by approximately 34% in S-0.52-1 and 42% in S-1.11-1 beams as compared to S-0.33-1 indicating higher load carrying capacity with an increase in tensile reinforcement. Another important observation is a significant increase in the area under the load-deflection plot with an increase in the reinforcement ratio. It is also observed that plastic zone III reduces drastically with an increase in steel. The failure takes place at the much lower strain in the S-1.11-1 RC beam. All S-series RC beams specimens failed by steel yielding and followed by concrete crushing.

4.2 Flexural performance of GFRP reinforced beams

The mechanical behaviour of GFRP-RC beams in flexure is visibly different from steel-RC beams (Figure 7). Broadly, the behaviour of GFRP-RC beams exhibits a bi-linear load-deflection response up to the failure without any yielding or ductility as experienced by steel reinforced beams. Initially, the load-deflection curve is perfectly linear and this zone is un-cracked elastic zone I. A noticeable decrease in the stiffness of the beam is observed with the formation of hairline cracks at a load (P_cr) of (7.89, 8.0 and 11.42) kN with a deflection (δ_cr) of (1.31, 0.9, and 0.39) mm in G-0.33-1, G-0.52-1, and G-1.11-1 RC beams, respectively. These minor cracks progress along the sides of the beam at constant stiffness. With an increase in the reinforcement ratio, initial bending stiffness also increases. As the load further increases, the cracks initiate and become visible at a load of (15, 20, and 33) kN for G-0.33-1, G-0.52-1, and G-1.11-1 RC beam in the pure bending region. Further with the increase in loading, cracks progress towards the compression zone, and this zone II is named cracked-elastic zone. The GFRP beams exhibit elastic response in this zone with the progression of flexural cracks. This trend continues till a first drop in the load-carrying capacity is observed at (33.6, 45.72, and 57.18) kN with a deflection of (30.33, 32.01, and 16.37) mm with increasing reinforcement ratio, pointing towards initiation of concrete crushing in G-0.33-1, G-0.52-1, and G-1.11-1 RC beams.

Figure 7.

Load v/s deflection plot for GFRP reinforced beams (one beam per type).

The failure is a typical flexural failure in the form of vertical flexural cracks in the pure bending zone along with their simultaneous spreading towards the entire length of the beam. Further the crushing of concrete progress with a sharp increase in the load-carrying capacity in zone III (concrete crushing zone). The beam continuous to carry load linearly with an increase in deflection until the second drop in load is observed at (41.7, 47.63, and 73.99) kN with a deflection of (52.68, 40.74, and 28.41) mm in G-0.33-1, G-0.52-1, and G-1.11-1 RC beams, respectively. The effective concrete section is highly reduced due to cracking and ineffective in resisting the tensile load and the beam fail at peak load of (P_Peak) (51.32, 60.47, and 83.71) kN for in G-0.33-1, G-0.52-1, and G-1.11-1 RC beam with the ultimate deflection (δ_u) (68.68, 60.09, and 34.47) mm. All the GFRP reinforced beams fail typically by concrete crushing since they are designed as over-reinforced beams to prevent the failure by GFRP rupture as expected in under-reinforced GFRP beams.

4.3 Effect of longitudinal tension reinforcement ratio on average mid-span deflection

In structural engineering, the term deflection is defined as the movement of a body from its original position under a force, load, or weight of the body itself. The effect of midspan deflection with the same concrete strength is intrinsically related to the longitudinal reinforcement ratio. As the longitudinal reinforcement ratio changes from 0.33, 0.52, and 1.1%, the mid-span deflection decrease in both steel and well as GFRP reinforced concrete beams. For example, in the case of steel-reinforced concrete beams, increasing the longitudinal reinforcement ratio changes from 0.33, 0.52, and 1.1% decreases the mid-span deflection by 53.15, 42.59, and 31.42 mm. On the other hand in the case of GFRP reinforced concrete beams decreases by 65.14, 55.52, and 34.43 mm. Hence, the overall comparison of steel and GFRP reinforced beams under static flexural loading is widely different. GFRP-RC beams exhibit higher deflections and lower crack widths in comparison to steel reinforced beams at the same reinforcement ratios (Figure 8 and Table 4). The deflections are 22.55, 30.35, and 9.57% higher in G-0.33-1, G-0.33-1, and G-1.11-1 RC beams in comparison to S-0.33-1, S-0.52-1, and S-1.11-1 RC beams. This is because of the low elastic modulus of GFRP bars as compared to steel bars.

Figure 8.

Variation in average maximum mid-span deflection Vs ‘ρ’.

RC specimen	Mexp (kN-m)	Mth (kN-m)	Mth/Mexp	RC specimen	Mexp (kN-m)	Mth (kN-m)	Mth/Mexp
S-0.33-1	11.73	7.83	0.66	G-0.33-1	17.11	15.33	0.93
S-0.33-2	11.23	7.83	0.70	G-0.33-2	16.45	15.33	0.93
S-0.33-3	10.24	7.83	0.76	G-0.33-3	15.90	15.33	0.96
Average	11.06	7.83	0.70	Average	16.48	15.33	0.94
S-0.52-1	16.09	12.63	0.78	G-0.52-1	19.25	18.83	0.97
S-0.52-2	16.27	12.63	0.77	G-0.52-2	19.78	18.83	0.95
S-0.52-3	16.81	12.63	0.75	G-0.52-3	19.20	18.83	0.98
Average	16.39	12.63	0.76	Average	19.41	18.83	0.96
S-1.11-1	27.81	23.95	0.86	G-1.11-1	27.91	25.63	0.95
S-1.11-2	29.66	23.95	0.80	G-1.11-2	26.51	25.63	0.96
S-1.11–3	27.61	23.95	0.86	G-1.11–3	27.28	25.63	0.93
Average	28.36	23.95	0.84	Average	27.23	25.63	0.94

Table 4.

Moment capacities of steel and GFRP reinforced beams.

4.4 Effect of longitudinal reinforcement ratio on experimental moment carrying capacities

The experimental moment carrying capacity is a maximum bending moment that can be resisted by a beam or any other structural member before it fails in bending. The effect of experimental moment carrying capacities with the same concrete strength is essentially related to the longitudinal reinforcement ratio.

As the longitudinal reinforcement ratio changes from 0.33, 0.52, and 1.1% the experimental moment carrying capacity increases in both steel and well as GFRP reinforced concrete beams. For example in the case of steel-reinforced concrete beams, increasing the longitudinal reinforcement ratio changes from 0.33, 0.52, and 1.1% increases the experimental moment carrying capacity by 11.06, 16.39, and 28.36 kN-m. On the other hand in the case of GFRP reinforced concrete beams increases by 16.48, 19.41, and 27.23 kN-m as shown in Figure 9 and Table 4. Overall, it is observed that with the increase in the longitudinal reinforcement ratio, the ultimate load-carrying capacity increases by approximately 34% in G-0.33-1 and 42% in G-0.52-1 beams in comparison to S-0.33-1 and S-0.52-1. It is pointing towards higher tensile strength of GFRP bars in comparison to steel reinforced beams. But as the reinforcement ratio increases from 0.52 to 1.1% there is a slightly 5.09% decrease in load-carrying capacity in G-1.11-1 beam as compared to S-1.11-1 beam due to increase brittleness with higher ρ. It points towards a lower modulus of elasticity of GFRP bars as compared to steel bars.

Figure 9.

Variation in an average experimental moment carrying capacities Vs ‘ρ’.

4.5 Modes of failure

The design of both steel-RC and GFRP-RC is based on ACI Code [8, 34]. The GFRP-RC beams were designed as over-reinforced beams with a reinforcement ratio of 0.0052 [8] which was greater than the balanced reinforcement ratio of 0.00308 [34]. The steel-RC beams were designed as under-reinforced beams having a reinforcement ratio (0.0052) less than the balanced reinforcement ratio of 0.02. A compression failure for the GFRP-RC beams and a tension failure for the steel beams were expected during flexural testing. The observed modes of failure of steel-RC and GFRP-RC beams are presented in Figure 10. Steel-RC beam failed by the crushing of concrete after the tension reinforcement yielded (Figure 10a, c, and e) whereas the GFRP-RC beam failed typically in shear followed by concrete crushing (Figure 10b, d, and f) since they are designed as over-reinforced beams to prevent their failure by GFRP rupture. This indicates that even though both steel-RC and GFRP-RC beams have the same area of tension reinforcement (A_st) ratio, GFRP-RC beams experience a different mode of failure as compared to the steel-reinforced concrete beam. Therefore, some modification in the design has to be considered when the GFRP bar is to be used as reinforcement.

Figure 10.

Failure modes in steel-RC and GFRP-RC beams. (a) S-0.33-1. (b) G-0.33-1. (c) S-0.52-1. (d) G-0.52-1. (e) S-1.11-1. (f) G-1.11-1.

A comparison of theoretical and experimental moment carrying capacities of steel-RC and GFRP-RC beams is presented in Table 4. The ratio of M_th/M_exp is less than 1 for both steel-RC and GFRP-RC beams. Thus, for design purposes, the strength reduction factor (Ø) for all over reinforced GFRP reinforced beams is calculated by Eq. (1). The theoretical moment of resistance (M_th) in the case of the GFRP-RC beam is calculated by Eq. (1) [8].

$$M_{\mathit{th}}={\rho }_f{f}_f\left[1-0.59{\rho }_f\frac{f_f}{f^{\prime }c}\right]{\mathit{bd}}^2$$E1

On the other hand, the theoretical moment of resistance (M_th) in the case of the steel-RC beam is calculated by the formula given in Eq. (2) [34]

$$M_{\mathit{th}}={\rho }_f{f}_f\left[1-0.59{\rho }_f\frac{f_y}{f^{\prime }c}\right]{\mathit{bd}}^2$$E2

where M_th = theoretical bending moment resistance, ρ_f = reinforcement ratio, b = width of the beam, d = effective depth of the beam, f’_c = Design characteristic concrete compressive strength. Where, ρ_fb = balanced reinforcement ratio, ρ_f = actual reinforcement ratio and Ø = 0.65 for ρ_f ≥ 1.4 ρ_fb [8]. For under-reinforced steel beams, the strength reduction factor is taken as 0.9 (ACI 319, (2019)). The fracture/failure and cracking pattern in the two differently reinforced beams along with a varying percentage of tension reinforcement ratio are further studied using acoustic emission in Section 5 respectively.

4.6 Damage classification

From the load-deflection plots of steel-RC and GFRP-RC beams, as shown in Figures 12, 14, and 16 the development of cracking patterns as shown in Figures 11, 13, and 15) in the two beams can be classified into three damage levels. Damage level I refers to the phase when the invisible cracking occurs. This damage zone is differentiated from the un-damaged state by the formation of visible hairline cracks and a distinct decrease in the stiffness of the beam is observed. Damage level II refers to the phase between the formation of hairline cracks and the stage of steel yielding in the case of the steel-RC beams and 1st drop-in load in case of GFRP-RC beam and further leading to the formation of distributed flexural and shear cracks. Damage level III refers to the phase between the steel yielding and final failure caused due to concrete crushing in the SB beam. In the case of GB beams, this refers to the phase between 1st drop in load and the final failure due to the crushing of compressive concrete. The formation of crack patterns at different levels of loading and the development of the different damage levels in steel-RC and GFRP-RC beams are shown in Figures 11–16 respectively.

Figure 11.

Typical crack patterns. (a) Damage level I. (b) Damage level II. (c) Damage level III.

Figure 12.

Damage level classification.

Figure 13.

Typical crack patterns. (a) Damage level I. (b) Damage level II. (c) Damage level III.

Figure 14.

Damage level classification.

Figure 15.

Typical crack patterns. (a) Damage level I. (b) Damage level II. (c) Damage level III.

Figure 16.

Damage level classification.

5. Acoustic emission results

5.1 Average frequency (AF) and rise angle (RA)

A parametric analysis was performed between AF and RA values by moving averages based on over 100 AE hits [20, 25]. In the present case, the proportion of AF and RA values are likewise set to 1:200. The plot of Average frequency (AF) and rise angle (RA) values for all three damage levels, for S-series and G-series RC beams, is shown in Figures 17–19 respectively. AF-RA value plot gives a fair indication of the cracking modes in steel-RC and GFRP-RC beams and is used for crack classification. The diagonal line represents the transition line between tensile and shear cracks and is used as a reference line for crack classification.

Figure 17.

Variation in AF Vs RA values at different levels of damages. (a) Damage level I. (b) Damage level II. (c) Damage level III.

Figure 18.

Variation in AF vs. RA values at different levels of damages. (a) Damage level I. (b) Damage level II. (c) Damage level III.

Figure 19.

Variation in AF vs. RA values at different levels of damages. (a) Damage level I. (b) Damage level II. (c) Damage level III.

In damage level I, the plot of AF-RA for the S-0.33-1, S-0.52-1, and S-1.11-1 RC beam (Figures 17(a),18(a), and 19(a)) shows the development of cracks initially due to tensile cracking whereas for the G-0.33-1, G-0.52-1, and G-1.11-1 RC beam, it is in shear cracking mode (Figures 17(a), 18(a), and 19(a)). A high average AF value of 32.2, 53.73, and 50.64 kHz with a lower RA value of 579.9, 398.78, and 205.0 μs values were noticed in the S-0.33-1, S-0.52-1, and S-1.11-1 RC beam as against lower average AF value of 15.0, 16.20, and 26.73 kHz and a higher RA value of 6432.5, 7324.80, and 10752.52 μs for the G-0.33-1, G-0.52-1, and G-1.11-1 RC beam (Tables 5–7), respectively. These values indicate that the steel-RC beams can resist and bridge the cracks better owing to the perfect bond between concrete and steel and the high modulus of elasticity of steel bars as compared to GFRP reinforced concrete beams. In this damage level, the hairline cracks were visible in both beams.

Damage level	Rise angle		Average frequency
	S1	G1	S1	G1
I	579.9	6432.5	32.2	15.0
II	427.9	5157.2	41.1	25.5
III	1327.0	3317.6	37.3	42.2

Table 5.

Variation in RA and AF values in S-0.33-1 and G-0.33-1 RC beams.

Damage level	Rise angle		Average frequency
	S2	G2	S2	G2
I	398.78	7324.8	53.73	16.20
II	315.21	5956.17	56.72	27.65
III	1225.257	5130.68	40.99	43.66

Table 6.

Variation in RA and AF values in S-0.52-1 and G-0.52-1 RC beams.

Damage level	Rise angle		Average frequency
	S3	G3	S3	G3
I	205	10752.52	50.64	26.73
II	238.79	7829.34	50.71	26.81
III	454.922	5255	47.41	39.4

Table 7.

Variation in RA and AF values in S-0.1.11-1 and G-1.11-1 RC beams.

With the increasing load in damage level II, invisible cracking is observed in the steel-RC beam at higher AF and smaller RA value indicating tensile cracking mode whereas, in the case of the GFRP-RC beam, a reversed trend is observed with a slight increase in average AF value and drop in RA value pointing towards a shift from shear to tensile cracking in the GFRP-RC beam (Figures 17(b),18(b), and 19(b)). Further, in damage level III, a slight decrease in the average AF value of 37.3, 40.99, and 47.41 kHz with a minute increase in the RA value of 1327.0, 1225.25, and 454.92 μs value was noticed in the S-0.33-1, S-0.52-1, and S-1.11-1 RC beam. A plausible explanation for this can be attributed to the reduction in the cross-sectional area due to the yielding of steel bars in the steel-RC beam.

On the other hand, a continuous increase in AF value of 42.2, 43.66, and 39.40 kHz with a significantly lower RA value of 3317.6, 5130.68, and 5255 μs was noticed in the G-0.33-1, G-0.52-1, and G-1.11-1 RC beam (Figures 17(c),18(c) and 19(c)). AFRA plots at the failure point suggest that the steel-RC beam experiences flexural cracks localisation as shown in Figure 10a,c, and e, whereas the GFRP-RC beam experiences shear cracking as also observed visually as shown in Figure 10b,d, and f.

Hence, AF-RA plots can be exploited to predict the initiation and progression of invisible and visible crack formation in concrete as indicated by the density of dots in the plot of the AFRA value. Hence, the AE plot of AF and RA value can effectively demonstrate the variation in initiation and progression of damage, classification of cracking, and failure modes in steel as well as GFRP-reinforced beams.

5.2 AE XY plots monitoring

To further understand the sequences of AE events, their locations within the steel-RC and GFRP-RC beam have been plotted at different times in the form of event maps. The locations of the AE events during the entire process of damage are presented in X-Y plots and these AE XY-event plot profiles were obtained by using AE-win software (Figures 20–28). It may be recalled that the cracks can be located only in the zones covered by the AE sensors. Every crack is labelled as an event recorded by three or more sensors. The red spawn (dots) in AE plots represents the location of each AE event recorded to indicate the frontal surface condition of the steel-RC and GFRP-RC beams at different stages of damage under flexural loading. AET is supposed to increase the efficiency of structural inspection by indicating the initiation and progression of cracking and the surface strains in the two types of beams respectively.

Figure 20.

(a) Beam sample (b) AE XY plots at damage level I.

Figure 21.

(a) Beam sample (b) AE XY plots at damage level II.

Figure 22.

(a) Beam sample (b) AE location- XY plots at damage level III.

Figure 23.

(a) Beam sample (b) AE XY plots at damage level I. (c) Beam span (m).

Figure 24.

(a) Beam sample (b) AE XY plots at damage level II.

Figure 25.

(a) Beam sample (b) AE location- XY plots at damage level III.

Figure 26.

(a) Beam sample (c) AE XY plots at damage level I.

Figure 27.

(a) Beam sample (c) AE XY plots at damage level II.

Figure 28.

(a) Beam sample (b) AE location- XY plots at damage level III.

It is apparent from Figures 20(a),23(a), and 26(b) that, during the damage level I i.e., at cracking loads of ∼5.58, 6.99, and 9.54 kN for S-0.33-1, S-0.52-1, and S-1.11-1 RC beam and ∼7.89, 8.01, and 10.31 kN for G-0.33-1, G-0.52-1, and G-1.11-1 RC beam respectively. It is indicated by the appearance of AE events in the XY-plot at the same instant which suggests the formation of invisible cracks in steel-RC and GFRP-RC at the same location as shown in Figures 20(b),23(b), and 26(b). Hence, it can be concluded that invisible cracking which is not visible to the naked eye can be reliably displayed by AE XY- event plots.

Further, with an increase in loading, it is visually observed that at a yield load (P_y) of ∼28.93, 45.85, and 81.44 kN in S-0.33-1, S-0.52-1, and S-1.11-1 RC beam the earlier invisible crack starts to become visible and eventually coalesce together to form visible cracks propagating vertically upwards. This indicates the progression of damage to level II as shown in Figures 21(a),24(a), and 27(a). These cracks do not result in sudden failure of the beam as their propagation is arrested by the presence of shear stirrups. On the contrary, in the case of the GFRP-RC beams, owing to the elastic behaviour of GFRP bars, the beam continues to carry load linearly and invisible cracking is observed at a load of ∼33.6, 51.80, and 57.46 kN. On the other hand, the AE event plot shows the congregation of red dots pointing towards the coalescence of invisible cracks into visible cracks at approximately (1.2, 1.2, and 1.1) m and (1.3, 1.2, and 1.1) m distance from the left support for S-0.33-1, S-0.52-1, and S-1.11-1 RC and G-0.33-1, G-0.52-1, and G-1.11-1 RC beams. This is in a close match with the actually cracked beam and the same can also be observed in the actual beam sample and compared with the AE X-Y plots (Figures 21(b),24(b), and 27(b)).

With further loading i.e. damage level III, the steel bars yield leading to flexural failure followed by concrete crushing (Figures 22(a), 25(a), and 28(a)). The same can also be confirmed with the actual beam sample. Moreover, the G-0.33-1, G-0.52-1, and G-1.11-1beams fail typically in shear followed by concrete crushing at damage level III. The corresponding ultimate load of ∼48.94, 56.74, and 76.98 kN in G-0.33-1, G-0.52-1, and G-1.11-1beams and is also depicted by extremely dense AE event plots at the same locations at 1.33, 1.33, and 1.33 m from the left support in Figures 22(b), 25(b), and 28(b). A clear trajectory of transverse vertical cracks is also observed from the AE event plot. This is also validated by the image of the actually cracked beams. Thus, AE events maps provide a reliable and real-time indication of the initiation and progression of cracking inside concrete in steel as well as GFRP-RC beams undergoing flexural loading.

Hence, it can be concluded that AE has the potential to serve as an online non-destructive monitor that can map the progression of cracking in RC structures. Various AE parameters like cumulative AE hits and their amplitudes, the plot of AF, and RA can effectively capture the initiation and progression of cracking in the steel and GFRP-RC beams, much before these are visible to the naked eye. Moreover, AE XY-plot has ample potential to serve as an effective tool to monitor cracking in the terms of AE XY-plot, and much before the actual cracking is visible to the naked eye. Hence, the advanced AE tools can be effectively used in conjunction with NDE of various types of RC structures incorporating steel as well as GFRP bars.

6. Conclusion

The focus of this chapter is to critically investigate and differentiate between the flexural and cracking behaviour of steel-RC and GFRP-RC beams utilising a non-destructive technique of AE. The following conclusions are drawn from the study:

• The load-deflection plot of steel and GFRP reinforced beams shows contrasting profiles. With increasing reinforcement ratio, steel-reinforced beams show typically an increase in ultimate load-carrying capacity, shrinking plastic zone with reduced ductility, and failure taking place at the much lower strains. All the steel-RC beams failed in flexure followed by concrete crushing. On the other hand, GFRP reinforced beams exhibit bi-linear load-deflection response up to the failure without any yielding or with higher ultimate load-carrying capacities and higher deflections as compared to steel reinforced beams indicating enhanced ductility and ultimate strength. All the GFRP-RC beams failed in shear and were followed by concrete crushing.

• Surface-mounted passive AE sensors give an early warning of initiation of micro-cracking inside concrete well depicted by low amplitude AE hits with the progression of failure in the form of macro cracking displayed by AE hits of larger amplitudes in steel as well as GFRP reinforced concrete beams. The amplitude and the cumulative number of AE hits are well indicative of the initiation and progression of invisible and visible cracking in the RC beam in the form of well-defined AE activity phases. A striking contrast in the AE activity is observed in steel and GFRP reinforced beams.

• AE XY-event plots give a pictorial representation of various stages of onset, initiation, and progression of cracking in steel and GFRP reinforced concrete beams in terms of their origin, location their rates of growth. A close matching and coherence are observed between the AE XY-event map with the micro and macro-cracks visually observed in the actual beams at various stages of loading.