Flexural progressive damage and failure behavior of carbon-aramid/epoxy hybrid woven composites

Abstract

Two kinds of carbon aramid/epoxy hybrid woven composite specimens with different fiber orientations were prepared. The progressive flexural damage behaviors of the composites were studied. The failure process was monitored in real time by acoustic emission during the test, and the characteristics of the acoustic emission signals originating from the damage were deeply studied. In addition, the internal damage initiation/evolution and failure mechanisms were characterized by X-ray micro-computed tomography. The results show that composite specimens exhibit higher strength and obvious quasi-brittle damage behavior when carbon fiber orients along the loading direction, and the macroscopic failure is mainly shear fracture, which propagates in the direction of thickness, and the damage is dominantly distributed above the neutral plane of specimen. When aramid fiber orients along the loading direction, composite specimens show high ductility and the failure modes are mainly ply fracture at the bottom, delamination (inter-ply delamination and intra-ply delamination) and tensile fracture of the tows. The combination of acoustic emission and X-ray micro-computed tomography analysis provides an insight for further elucidation of the progressive damage initiation/evolution and failure mechanism of composites.

Keywords

Carbon aramid/epoxy hybrid woven composite progressive damage analysis AE micro-CT

Introduction

Carbon fiber-reinforced composites are attractive engineering materials due to their superior mechanical properties.¹ However, the limited failure strain hinders their wide application. Extensive researches have shown that hybridization is believed to be one of an effective approach to adjust the properties of composites.^2,3 Hybrid composites are made of two or more different types of reinforcing fibers combined with suitable binding matrix such as resin to meet specific requirements that cannot be achieved with single fibers composites.^4,5 Introducing aramid fiber to carbon fiber-reinforced composites can improve the stiffness of aramid fiber composites and enhance the toughness of carbon fiber composites, thus effectively expanding the application range of composites.⁶

In recent years, the application of carbon aramid/epoxy hybrid woven composites have been continuously increased in aerospace, automobile and sports products, requiring better knowledge of their mechanical performance.^7,8 Sun et al.⁹ studied the low-velocity impact responses of carbon aramid/epoxy hybrid woven twill composites. It is found that placing high stiffness carbon fiber as reinforcement in highly stressed regions can enhance the performance of composites. Song¹⁰ investigated the tensile and bending properties of carbon aramid/epoxy woven hybrid composites. The results show that the lamination position and the continuous cumulative count of reinforcements significantly influence the strength and stiffness of composites. Hashim et al.¹¹ studied the tensile and low cycle fatigue properties of carbon aramid/epoxy hybrid composites at different fiber orientation. It is found that the orientation of fibers has a significant effect on the mechanical properties of composites. According to previous research, it is found that a lot of researchers have focused on the structural characteristics and mechanical properties of carbon aramid/epoxy hybrid woven composites, but there is a lack of study on failure mechanism during damage, especially progressive damage evolution of composites. Zhang et al.¹² used the progressive experiment method to study the structural characteristics and damage initiation/evolution of three-dimensional (3D) braided composites subjected to bending load. The results show that the bending resistance of the longitudinal load-bearing specimen is much higher than that of the transverse load-bearing specimen. Tian et al.¹³ investigated the progressive damage and failure behavior of 3D braided composites under biaxial tension by using an improved finite element method. It is found that the difference of failure modes in the damage evolution process of composites have great influence on the correlation of biaxial failure. Furthermore, owing to the complexity of the application environment, it is subjected to loads from various directions in service, and inevitably produces different degrees of damage. The damage of composites causes a significant reduction of the mechanical properties, resulting in a decrease of load carrying capacity. Therefore, it is necessary to pay attention to the research on the progressive damage and failure behavior of carbon aramid/epoxy hybrid woven composite structures.

Bending characteristic is one of the important mechanical properties of composite structures. The analysis of bending damage mechanisms is particularly important because various composite structures are subjected to bending loading in applications.^14–16 In addition, in view of the complexity of progressive damage and structure of woven composites, it is necessary to use some efficient methods for improving the comprehension of progressive damage evolution and failure mechanism. In general, typical non-destructive testing (NDT) technology such as acoustic emission (AE), ultrasonic, digital image correlation (DIC), infrared thermography (IRT) and X-ray micro-computed tomography (Micro-CT) can be used to characterize the deformation or damage of composites. However, a single NDT method fails to provide enough information for the structural damage of composites. To more directly and effectively describe the damage initiation and evolution of woven composites, the combination of multiple NDT methods are needed. As a dynamic Non-destructive technique, AE has been widely used to monitor the damage of composites.^17,18 Sobhani et al.¹⁹ used AE to study the damage mechanism of composite laminates subjected to buckling and post-buckling behavior. The contribution of each damage mechanism to the overall failure was evaluated based on AE hits and energy. Fotouhi et al.²⁰ investigated the damage modes of thin-ply uni-directional carbon/glass hybrid composites under tensile load with AE technology. The relationship between AE signals and the corresponding damage can be established, and the failure modes of composites can be identified based on AE amplitude and energy. Xu et al.²¹ used AE technology to identify various damage modes of hygrothermal aging adhesive composite joints. Meanwhile, the correlation of various damage modes was studied by clustering analysis of AE characteristic parameters.

Micro-CT can be used to study the internal microstructure and damage of composites by acquiring 3D images of internal structures.^22,23 Kaouache et al.²⁴ investigated the microstructure of high-density polyethylene/short hemp fiber composites under tensile test by using Micro-CT observation. It is found that as the strain level increases, the micro-cracks in the thickness direction increase. Ullah et al.²⁵ carried out the Micro-CT analysis of the deformation and damage mechanisms of carbon fiber woven composites subjected to bending loading. They observed various damage modes including matrix cracks, delamination, fiber fracture and tow debonding. Castaneda et al.²⁶ studied the role of z-binders on the damage tolerance of 3D woven composites subjected to the out-plane tension and fatigue loading by AE and Micro-CT. It is found that the z-binders have significant effects on the transverse reinforcement of the mechanical behavior of the 3D woven composites.

This paper addresses a new method that combines AE and Micro-CT to determine the deformation and damage of carbon aramid/epoxy hybrid woven composites under three-point bending loading. The mechanical properties, progressive damage and failure behavior of woven composites with different fiber orientation were discussed. More importantly, the progressive internal damage patterns of composite specimens were characterized and evaluated according to 3D reconstructed images generated by micro-CT.

Experimental procedure

Materials and specimens

The carbon aramid/epoxy hybrid composite laminates were prepared from 20 layers of double-axial carbon/aramid hybrid twill woven fabrics (H3K-CAT5, 200 g/m²) and epoxy resin by vacuum assisted resin infusion (VARI) method with the help of vacuum pressure (−0.097 MPa). The mixing ratio for 3329 A epoxy resin to 3329B hardener in weight was 10:4 and fiber volume fraction in composites laminates was about 60%. In order to ensure the proper wetting of the laminates, a vacuum reflow device was designed in Figure 1 and repeated infusion of resin matrix was obtained. The carbon aramid/epoxy hybrid composite laminates were cured at room temperature for 48 h, and then dried at 130 °C for 8 h in the oven. Considering the view field of Micro-CT scanner, the as-prepared composite laminates (300 mm × 200 mm) were cut into two types of strip specimens with the size of 60 mm × 20 mm by using a small water-cooled cutting machine. The thickness of the specimen is 4.4 ± 0.1 mm. Laying method and sketch map of composite specimens are shown in Figure 1. Carbon fiber orients along the loading direction of Specimen A and aramid fiber orients along the loading direction of Specimen B. A total of four specimens for each type of composite specimen were employed.

Progressive three-point bending test

The progressive three-point bend tests were performed on the machine (LD24) according to the ASTM standard D790 with the crosshead speed of 2 mm/min and the span was 48 mm. The flexural tests were divided in three phases: the first is initial damage, the second is damage evolution and the third is failure. The internal damage of specimen was monitored by using an AE instrument (DS2-8A) during the whole loading progress. AE signals were monitored by two broad spectral band sensors (RS-54A) which installed on the specimen surface with tape. The frequency range that allows high-precision measurement is 100–900 kHz. Silicone grease was covered the surface between the AE transducer and the specimens to improve the acoustic coupling quality. To reduce the influence of noise signals from the machine and the surrounding area, the acquisition threshold was set to 40 dB through repeated tests. Before the testing, the support roller noise could be removed by spectral analysis. Furthermore, there is no obvious damage of the specimens at the support roller and as most AE signals were generated between the support rollers, the support roller noise can be ignored. The sampling frequency was 3 MHz. The experimental setups for bending testing of specimens are illustrated in Figure 2.

Due to the limited experimental conditions, it is necessary to unload when carrying out CT-scanning. The maximum load is taken after the second loading, and the first loading stage is in the elastic stage. It was found that the repeated loading has little impact on the overall mechanical response of the specimens, which can be ignored. The specimens were scanned by using an X-ray tomography system (Skyscan 1172 micro-focus X-ray system) as shown in Figure 3. The X-ray source current of the Micro-CT scanner was set to 128 µA, the power was 76 W. The voxel size was fixed at 6.8 µm. The specimen was fixed on a rotational table, which rotated by a small angle (rotated 180° and each rotation step is 0.2°) relative to the X-ray beam to obtain the tomographic radiographs. A total of 960 two dimensional (2D) images were captured one by one during this process. Finally, the 3D image was reconstructed through the image processing unit (pretreating software NRecon and reconstruction software CTvox).

Results and discussions

Mechanical properties of composite specimens

All results of bending testing are summarized in Table 1. The bending strength (S) can be expressed by

S = \frac{3 F L}{2 b h^{2}}

(1)

where F is the maximum load, L(mm) is the span of specimen during the test, b (mm) and h (mm) is the width and thickness of the specimen, respectively.

Table 1.

Failure load related data of two types of specimens.

Two types of specimens	Load (kN)				Average load (kN)	Standard deviation (kN)
Two types of specimens	Test 1	Test 2	Test 3	Test 4	Average load (kN)	Standard deviation (kN)
Specimen A	3.19	2.9	3.05	2.9	3.01	0.139
Specimen B	1.93	1.88	1.91	2.01	1.93	0.056
	Bending strength (MPa)				Average strength (MPa)	Standard deviation (MPa)
Specimen A	593	539	567	539	559	25.94
Specimen B	359	350	355	374	359	10.34

The ratio of thickness to span ratio t/L of fabricated specimens is roughly equing to 0.09 (<1/10), and the bending strength is calculated by Eq(1), which is derived from the Euler-Bernoulli beam theory. According to the Timoshenko beam theory incorporating the shear deformation applied to short thick beams, the contribution of shear deformation to the considered bending strength is negligible when t/L < 1/10. Therefore, the calculated result of the bending strength S based on the Euler-Bernoulli beam model is acceptable. In Table 1, the average failure load is 3.01 kN and 1.93 kN for Specimen A and B, and the corresponding standard deviation is 0.139 kN and 0.056 kN, respectively. In the same order, the average strength is 559 MPa and 359 MPa, and the corresponding standard deviation is 25.94 MPa and 10.34 MPa, respectively. It indicates that the experimental tests have good repeatability. From Table 1, the bending strength of Specimen A is significantly larger than that of Specimen B, which is related to the main load-carrying part of Specimen A is carbon fiber. Generally, calculating average strength values require the use of specimen where the width is an integer multiple of the repetitive twill unit cell. During the cutting process of specimens, half of such a repetitive structure is inevitably included in width, the 0 deg or 90 deg fibers may be overrepresented in the outer layer. To eliminate this computational error, two types of specimens are accurately designed and the width of small size specimen is increased up to 20 mm. The corresponding standard deviations of flexural strength are less than 5% and the bending strength value can be representative in such a case.

Figure 4(a) and (b) show the flexural load-displacement curves and schematic of progressive experiment stage of Specimen A and B, respectively. Based on the failure characteristics of the laminate composites, the repeated flexural tests are divided into three stress levels (Points A, B and C) as shown in Figure 4(b), revealing damage initiation/evolution and failure mode of composites. 1st, 2nd and 3rd load indicate loading up to point A, B, and C, respectively. The point A is selected after the elastic limit. Point B is located near the maximum load. Point C is selected near the failure of specimens.

Carbon fiber orients along the loading direction of Specimen A and composite specimen exhibits brittle failure characteristics. Hence, the load-displacement curves are almost linear before the ultimate fracture as shown in Figure 4(a). The curve enters the third stage after point B, which drops sharply with some fluctuations. This trend may be due to critical structural damage. From Figure 4(a), the load-displacement curve of Specimen B loses its linearity when the load reaches about 0.8 kN. As the bending loading continues, the stiffness of the Specimen B reduces rapidly, and the load-displacement curve drops significantly. Compared with Specimen A, Specimen B exhibits high ductility.

Progressive damage analyses for composites specimen

AE characteristics of composite specimens

The progressive damage of composite specimens can be evaluated through AE monitoring in real time, and the initiation and evolution of damage can be tracked. The analysis of AE parameters such as peak frequency, amplitude, hits, duration and energy are an effective means to identify different damage mechanisms occurring in composite materials and to follow their evolution until failure.²⁷ AE signal that exceeds the threshold and causes a channel to acquire data is called a hit. Amplitude and energy correspond to the maximum voltage and the released strain energy of AE signal waveform. Generally, the internal damage accumulation of the specimen is accompanied by the occurrence of AE signals. The duration is defined as the time interval between the first time the hit signal crosses the threshold and finally falls to the threshold. Due to a large amount of AE data for repeated experiments, two typical specimens were selected for AE characteristic parameter analysis. Figure 5 shows AE relative energy, cumulative hits versus displacement for Specimen A and B at 1st load (a), 2nd load (b) and 3rd load (c), respectively. In the first loading stage, AE signals of Specimen A show a low cumulative hit and relative energy and the AE activity is dominated by low relative energy events around 0–30 mV*ms during this stage. However, the medium energy events (about 6500 mV*ms) are detected at displacement of around 1.4 mm as shown in Figure 5(a). From Figure 5(a), for Specimen B, due to the high toughness of aramid fiber, there is no obvious damage and AE cumulative hits remain at a very low level. In the second loading stage, few AE signals are generated of Specimen A prior to 1.7 mm as shown in Figure 5(b). This phenomenon clearly demonstrates Kaiser effect²⁸ of AE. As the load increases, AE relative energy and hits significantly increase and high-energy AE signals are generated. For Specimen B, at the displacement of about 4.0 mm, there is a sharp increase in AE high-energy signals as shown in Figure 5(b). This phenomenon may be related to the accumulation of micro-damage inside the specimen, and the occurrence of damage such as matrix cracking, fiber debonding, and delamination. In addition, compared with Specimen A, Specimen B has less high-energy signals at this stage. This is related to the damage pattern of the aramid fiber. From Figure 5(c), in the third loading stage, AE hits significantly increase for two types of specimens and the maximum hit of Specimen B reaches 40000. Meanwhile, some AE signals with medium energy of Specimen B are generated at the displacement of about 0.5 mm, this is related to the damage failure mode of Specimen B in the second loading stage. At the beginning of the third loading stage, a large amount of micro-damage accumulations are induced and Felicity effect can be observed. However, intensification and fluctuation of AE energy is found at the displacement of 3–6 mm, indicating that failures are accompanied by multiple damage modes.

Figure 6 depicts the distribution of the AE amplitude and duration with increasing displacement for composite specimens. In this graph, green cylinders correspond to a range of amplitudes from 40–60 dB, blue cylinders correspond to 60–80 dB, and red cylinders correspond to 80–100 dB, respectively. Many researchers have successfully linked the amplitude range to the damage mechanism of composite materials, such as matrix cracking (i.e., 30–60 dB), fiber pull-out (i.e., 60–80 dB), delamination (i.e., 60–85 dB) and fiber breakage (i.e., 80–97 dB), respectively.^27,29,30 From Figure 6, in the first loading stage, AE activity is limited to low amplitude events. From Figure 6(a), in the case of Specimen A, a high duration AE signal with the amplitude of about 60 dB is generated, which may be noise signal or random damage at the edge of the specimen. In the second loading stage, the amplitude levels and duration of AE signals show a significant increasing trend, particularly for Specimen A in Figure 6(b), indicating more severe damage such as matrix cracking, fiber debonding and delamination are generated. For Specimen B, AE signals with low-amplitude and high-duration are generated as shown in Figure 6(b). Moreover, some AE signals with medium amplitude and duration, high-amplitude and short-duration are generated, which correspond to different damage modes. In the third loading stage, AE signals with medium and low amplitude and duration still exist, but AE signals with high-amplitude and high-duration increase remarkably. Furthermore, from Figure 6(c), compared with Specimen A, AE signals of Specimen B have much higher duration and the highest value of duration reaches up to 200000 µs. This indicates the serious micro-damage accumulation of Specimen B.

Figure 7 shows the amplitude and peak frequency versus displacement for Specimen A and B at different loading stages, respectively. On the basis of the previous research,^29,31 AE signals with frequency range of 70–120 kHz may be associated with the matrix cracking, the mid-frequency range (130–180 kHz) may be related to delamination. With the aggravation of the damage degree, AE signals with 190–260 and 270–300 kHz may be associated with fiber/matrix debonding and fiber failure, respectively. From Figure 7(a), in the first loading stage, AE signals of Specimen A are mainly distributed within 120 kHz, indicating that there may be damage of micro-matrix crack, accompanied by slight crack growth. For Specimen B, only a few AE signals with frequency below 120 kHz are generated. In the second loading stage, AE signals of the two types of specimens significantly increase and the frequency is mainly focused on 50–200 kHz, as shown in Figure 7(b), showing that the damage is mainly matrix cracking and delamination caused by crack growth. Meanwhile, for Specimen A, some AE signals with frequency between 200 and 300 kHz are generated, and the corresponding amplitude level is relatively low. This phenomenon indicates that there may be fiber failure damage caused by matrix cracking in Specimen A. For Specimen B, few AE signals with frequency from 200 to 270 kHz are generated. In addition, it is found that AE signals with high-amplitude correspond to relatively low frequency, as shown in Figure 7(b), which may be related to the failure mode of the specimen. In the third loading stage, the amplitude and frequency of AE signals for Specimens A and B significantly increase, especially for Specimen B, and more AE signals with frequency above 200 kHz are generated. At the same time, a large number of AE signals with medium amplitude and frequency ranges are generated, indicating that more serious damage has occurred.

To sum up, the progressive damage initiation and evolution in carbon aramid/epoxy hybrid woven composites could be described favorably by studying the AE characteristics in term of relative energy, amplitude distribution, cumulative hits, duration and peak frequency. In order to verify the correctness of the analysis, we further studied the internal meso-damage morphology of composite specimens by using Micro-CT observation.

Micro-CT analyses of the internal structure

To further understand the mechanism of progressive damage of woven composites, Micro-CT technique was used to realize 3D visualization of the internal damage. A series of 2D cross-sectional slices are cut out from the original and the progressive damage patterns of two types of specimens are presented in Figures 8 and 9, respectively. By comparing these images, the initiation/evolution of the damage can be found. In these images, the difference in the X-ray absorption between the fiber/matrix material and voids are used to identify cracks and damage. From Figure 8(a), at the elastic state (point A), the crack is visible, and the micro-damage for matrix cracks is found within Specimen A along y direction. Meanwhile, the occurrence of damage results in a significant rise of AE energy, as shown in Figure 5(a). Besides, there is no significant damage in the x direction within Specimen A. For Specimen B in the Figure 9(a), there no significant damage within specimen except for some original void defects and only irregular micro-damage occurs at the outer edge. After the second loading (from point B to point C), for Specimen A in Figure 8(b), the original damage is propagated and become increasingly more severe. These damages include the kink band under the crosshead, matrix cracking and transverse cracks, etc. The kink band formation may come from fiber bending because of the roller imprint in the three points bending loading.^32,33 For Specimen B, significant matrix cracks and delamination are found at the bottom as shown in Figure 9(b). Meanwhile, AE signals with high-amplitude and high-energy are generated, as shown in Figures 5(b) and 6(b). Afterward, a third loading is performed (reaching point C). From Figure 8(c), it can be observed that the internal damage such as the evolution of delamination and matrix cracking, as well as tow debonding and fiber breakage of Specimen A is more serious. At the same time, the deterioration of damage results in the increase of AE signals, and the severe damage behavior leads to the increase of AE signals with high energy and amplitude as shown in Figures 5(c) and 6(c). As the deformation increases, the specimen is subjected to high tensile stress below the neutral plane, and the crack propagates along the in-plane warp, resulting in more delamination. For Specimen B in Figure 9(c), damages such as matrix cracking, delamination, and fiber breakage occur at this stage. Meanwhile, the damage evolves into ply fracture along the x direction.

The crack is initiated at the top and bottom of Specimen A and B, respectively, as shown in Figures 8(a) and 9(b). Combined with the characteristics of the AE signals, it can be concluded that as the load increases, the crack further develops in the thickness direction. Moreover, the transverse cracks are induced within Specimen A along the x direction. Finally, other damage mode such as tow debonding, delamination and fiber breakage are found as shown in Figure 8(c). Simultaneously, in the x direction, damage propagates from the ply crack to the surrounding and delamination is induced. During crack propagation, the warp and weft will experience delamination and tow debonding. The process of damage propagation in fiber tows can be found in Ullah et al.²⁵ For Specimen B from Figure 9(b), the crack propagates from the initial tow debonding along the yarn path and matrix cracks and delamination are initiated. As the load increases, the crimped tows below the neutral plane of specimen attempt to straighten along the loading direction due to the tension, which is prevented by the transverse carbon fiber tows. Thus, the damage begins with tows debonding at the bottom, and then propagates to the bulk along the thickness direction as shown in Figure 9(b). Subsequently, more matrix cracks and transverse cracks are induced and propagated. As the bottom yarn breaks, it will cause serious damage such as matrix cracking, delamination, ply fracture (x direction) and fiber breakage.

Characterization of failure damage mode for composites specimen

To further understand the damage mechanism, the surface of the final fracture of the two samples were scanned with help of Micro-CT and the 2D cross section of different positions and directions are obtained as shown in Figures 10 and 11. The failure modes of Specimen A and B exhibit significant differences. In general, the kink band under the loading head, fiber bundle crack/fracture and delamination are main failure modes of bending specimen surface. Besides, the fiber bundle crack/fracture is directly related to the tensile stress under bending loading. For Specimen A, under the action of three-point bending load, due to the 3D stress, the damage begins with the kinks on the top, which propagates in the direction of thickness, and the damage is dominantly distributed above the neutral plane of specimen. Moreover, from Figure 10(a) and (b), the major failure modes are kink bond, matrix cracking, tow debonding, delamination (inter-ply delamination and intra-ply delamination), and fiber bundle breakage.

Unlike the failure modes of Specimen A, for Specimen B in Figure 11, there is no catastrophic and continuous crack damage. From Figure 11, the macroscopic failure damage of Specimen B is dominated by the tow stripping and fiber breakage at the bottom. At the same time, as the load increases, there is also kink band at the top. Besides, the damage is dominantly distributed under the neutral plane of specimen as shown in Figure 11(a), which is related to the aramid fiber as the main fiber direction of Specimen B. Specimen B shows a large deflection increase and exhibits a low bending strength as shown in Figure 4(a). The prominent failure modes are ply fracture at the bottom, delamination (inter-ply delamination and intra-ply delamination) and tensile fracture of the tow. According to the analysis of CT images for Specimens A and B, it is found that the cracks propagation and damage patterns of composites with different fiber orientations are different and Micro-CT can provide effective information about the internal damage evolution/initiation and failure mechanism of composite specimens. Furthermore, in order to more fully understand the damage mechanism of two different types of specimens, it is a better way to study the deformation, damage initiation/evolution and failure mechanism of carbon aramid/epoxy hybrid woven composites by combining AE and Micro-CT technology, which can accurately and effectively describe the internal damage evolution of composites.

Conclusions

In this study, the progressive damage and failure behavior of carbon aramid/epoxy hybrid woven composites with different fiber orientations were investigated by using AE and Micro-CT technology. The results are summarized as follows:

Different fiber orientations have significantly effect on the mechanical properties and failure behavior of carbon/aramid hybrid woven composites. When carbon fiber orients along the loading direction, composite specimens show higher bending strength and brittle failure characteristics, while composite specimens exhibit higher toughness when aramid fiber orients along the loading direction.

The progressive damage evolution and failure mechanisms in carbon aramid/epoxy hybrid woven composites with different fiber orientations could be described favorably by analyzing the AE characteristic parameters (amplitude, relative energy, counts, duration and peak frequency). For Specimen A, carbon fiber orients along the loading direction, the amplitude and energy of AE signals increase significantly in the damage evolution stage. Finally, AE signals significantly increase with the aggravation of damage and more signals with medium and low amplitude and long duration are generated. For Specimen B, aramid fiber withstands the main load and AE signals are mainly concentrated in the 3rd loading stage.

Micro-CT provides an effective means for characterizing the internal damage of composite specimens. For Specimen A, the kink bands are the initiation of damage, causing matrix cracks, which then continue into the bulk. In the failure stage, the main damages are delamination, tow debonding and fiber breakage. In the case of Specimen B, the interface failure mainly occurs in 2nd loading stage, and the origin of the damage is thought to be the tow stripping at the bottom. In the third loading stage, the serious damages are induced and main damages such as delamination, matrix cracking, and fiber fracture can appear at the bottom, and matrix cracks appear at the top of the specimen.

The combination of AE and Micro-CT analysis provides theoretical reference and technical support for understanding the progressive damage initiation/evolution and failure mechanism of composites.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors gratefully thank the support of the National Natural Science Foundation of China (grant no. 11502064 and 11572109) and Key projects of science and technology research in Colleges and Universities of Hebei Province (grant no. ZD2017006).

ORCID iD

Wei Zhou

References

Ullah

Harland

Silberschmidt

VV.

Damage and fracture in carbon fabric reinforced composites under impact bending. Compos Struct 2013; 101: 144–156.

Kim

HS.

Design optimization and manufacture of hybrid glass/carbon fiber reinforced composite bumper beam for automobile vehicle. Compos Struct 2015; 131: 742–752.

Bulut

Erkliğ

Yeter

Hybridization effects on quasi-static penetration resistance in fiber reinforced hybrid composite laminates. Compos Part B Eng 2016; 98: 9–22.

Sayer

Bektaş

Sayman

An experimental investigation on the impact behavior of hybrid composite plates. Compos Struct 2010; 92: 1256–1262.

Yeter

Erkliğ

Bulut

Hybridization effects on the buckling behavior of laminated composite plates. Compos Struct 2014; 118: 19–27.

Pincheira

Canales

Medina

, et al. Influence of aramid fibers on the mechanical behavior of a hybrid carbon-aramid-reinforced epoxy composite. Proceedings of the IMechE 2018; 232: 58–66.

Sugahara

Yang

, et al. A study on the energy absorption properties of carbon/aramid fiber filament winding composite tube. Compos Struct 2015; 123: 301–311.

Batista

ACMC

Tinô

SRL

Fontes

, et al. Anisotropy and holes in epoxy composite reinforced by carbon/glass and carbon/aramid hybrid fabrics: experimental and analytical results. Compos Part B Eng 2017; 125: 9–18.

Sun

Tang

Rong

ZJ.

An experimental investigation on the low-velocity impact response of carbon-aramid/epoxy hybrid composite laminates. J Reinf Plast Comp 2017; 36: 422–434.

10.

Song

JH.

Tensile and bending behavior for woven carbon-aramid/epoxy hybrid composites with various lamination structures by the VARTM method. Fibers Polym 2016; 17: 1269–1276.

11.

Hashim

Majid

DLA

Mahdi

E-S

, et al. Effect of fiber loading directions on the low cycle fatigue of intraply carbon-kevlar reinforced epoxy hybrid composites. Compos Struct 2019; 212: 476–483.

12.

Zhang

P-f

Zhou

Yin

H-f

, et al. Progressive damage analysis of three-dimensional braided composites under flexural load by micro-CT and acoustic emission. Compos Struct 2019; 226: 111196.

13.

Tian

Yan

, et al. Progressive damage and failure analysis of three-dimensional braided composites subjected to biaxial tension and compression. Compos Struct 2018; 185: 496–507.

14.

Hao

, et al. Experimental investigation on deformation and strength of carbon/epoxy laminated curved beams. Polym Test 2012; 31: 520–526.

15.

Audibert

Andreani

A-S

Lainé

, et al. Mechanical characterization and damage mechanism of a new flax-Kevlar hybrid/epoxy composite. Compos Struct 2018; 195: 126–135.

16.

Zhou

Yin

H-F

Shang

Y-J

, et al. Failure behavior and damage visualization of thick carbon/aramid hybrid woven composites under flexural loading conditions. Nondestruct Test Eva 2020; 35: 139–157.

17.

Zhou

Liu

Z-h

, et al. Acoustic emission behaviors and damage mechanisms of adhesively bonded single-lap composite joints with adhesive defects. J Reinf Plast Comp 2015; 34: 84–92.

18.

Liu

Yang

Peng

XQ.

Delamination analysis of carbon fiber composites under hygrothermal environment using acoustic emission. J Compos Mater 2017; 51: 1557–1571.

19.

Sobhani

Saeedifar

Najafabadi

, et al. The study of buckling and post-buckling behavior of laminated composites consisting multiple delaminations using acoustic emission. Thin Wall Struct 2018; 127: 145–156.

20.

Fotouhi

Suwarta

Jalalvand

, et al. Detection of fibre fracture and ply fragmentation in thin-ply UD carbon/glass hybrid laminates using acoustic emission. Compos Part A Appl S 2016; 86: 66–76.

21.

Liu

, et al. Damage mode identification of adhesive composite joints under hygrothermal environment using acoustic emission and machine learning. Compos Struct 2019; 211: 351–363.

22.

Skarżyński

Suchorzewski

Mechanical and fracture properties of concrete reinforced with recycled and industrial steel fibers using digital image correlation technique and X-ray micro computed tomography. Constr Build Mater 2018; 183: 283–299.

23.

Baran

Straumit

Shishkina

, et al. X-ray computed tomography characterization of manufacturing induced defects in a glass/polyester pultruded profile. Compos Struct 2018; 195: 74–82.

24.

Kaouache

Addiego

Hiver

J-M

, et al. In situ mechanical characterization of short vegetal fibre-reinforced high-density polyethylene using X-Ray tomography. Macromol Mater Eng 2013; 298: 1269–1274.

25.

Ullah

Harland

Silberschmidt

VV.

Characterization of mechanical behavior and damage analysis of 2D woven composites under bending. Compos Part B Eng 2015; 75: 156–166.

26.

Castaneda

Wisner

Cuadra

, et al. Investigation of the Z-binder role in progressive damage of 3D woven composites. Compos Part A Appl S 2017; 98: 76–89.

27.

Woo

Kim

TW.

High strain-rate failure in carbon/kevlar hybrid woven composites via a novel SHPB-AE coupled test. Compos Part B Eng 2016; 97: 317–328.

28.

Vladimir

Petr

Zhang

YN.

Spectral characteristics of acoustic emission in carbon Fiber-Reinforced composite materials subjected to cyclic loading. Adv Mater Sci Eng 2018; 53: 818–823.

29.

Zhou

Zhao

W-Z

Zhang

Y-N

, et al. Cluster analysis of acoustic emission signals and deformation measurement for delaminated glass fiber epoxy composites. Compos Struct 2018; 195: 349–358.

30.

Saidane

Scida

Assarar

, et al. Damage mechanisms assessment of hybrid flax-glass fibre composites using acoustic emission. Compos Struct 2017; 174: 1–11.

31.

Saravanakumar

Lakshminarayanan

Arumugam

Effect of thickness and denting behavior of glass/epoxy laminates subjected to quasi-static indentation (QSI) loading under acoustic emission monitoring. J Nondestruct Eval 2018; 37: 63.

32.

Zhang

Sun

Liu

, et al. Off-axis bending behaviors and failure characterization of 3D woven composites. Compos Struct 2019; 208: 45–55.

33.

Subagia

IDGA

Kim

Tijing

LD.

Effect of stacking sequence on the flexural properties of hybrid composites reinforced with carbon and basalt fibers. Compos Part B Eng 2014; 58: 252–258.