Low-velocity impact response and infrared radiation characteristics of thermoplastic/thermoset composites

Zhibin ZHAO, Zhengwei YANG,b,*, Wei ZHANG, Dejun LIU, Yin LI,Jinshu CHEN

a School of Missile Engineering, Rocket Force University of Engineering, Xi’an 710038, China

b School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an 710049, China

c China Aerodynamics Research and Development Center, Mianyang 621000, China

KEYWORDS Infrared thermography;Low-velocity impact;Thermoplastic composites;Thermoset composites;Damage characterization;Failure mechanism

Abstract The low-velocity impact response and infrared radiation characteristics of composites have rarely been focused on simultaneously. This study aims to investigate the low-velocity impact response and infrared radiation characteristics of the glass fiber reinforced thermoplastic polypropylene and carbon fiber reinforced thermosetting epoxy resin laminates wildly used in the aircraft industry.The impact tests were conducted at five energy levels.Characterization parameters such as impact load,displacement,and absorbed energy were measured.The damage evolution and damage modes of the laminates were analyzed through active and passive thermography,ultrasonic C-scan, and optical microscope. The results indicate that Thermosets (TS) laminates exhibit better impact resistance,while Thermoplastics(TP)laminates show higher delamination ductility,and the maximum contact force of TP laminates is much smaller than that of the TS laminates under lowvelocity impacts, but the low bending stiffness and low ductility of the TP matrix cause the difference in energy absorption level between the two not significant. The temperature characteristic changes of passive infrared thermography heat maps could characterize the damage mode of the laminates.The correlation between the heat maps and the impact characteristic curves is explained;the fluctuation of the impact characteristic curves is directly related to the hot spot characteristics changes of the heat maps.More frequent curve fluctuations correspond to a larger and brighter hot spot on the heat map,which peaks at the maximum impact load after the impact force versus time curve fluctuation cutoff point, the maximum center displacement of the impact force versus displacement curve, and the maximum absorbed energy of the absorbed energy versus time curve.

1. Introduction

With high specific strength, high specific modulus, and outstanding fatigue resistance, Fiber-Reinforced Polymer Composites (FRPC) laminates are widely used in the aircraft,automotive, and defense industries.1-3FRPC laminates can be classified into Thermoplastics (TP) and Thermosets (TS)according to the resin reversibility in the curing process. TP resins can be melted and reshaped, while TS resins form cross-linked structures in the curing process that prevent remelting and reshaping.4In the past few decades,TS composites have attracted more attention than TP ones in engineering applications because of their outstanding mechanical properties.5For example, thermosetting epoxy resin is the most widely used CFRP in the aircraft industry.However,TP composites have shown immense potential in engineering applications due to their recyclability, excellent fracture toughness,and high impact tolerance.4,6As a standard TP composite,glass fiber reinforced polypropylene composite is widely used in the aircraft, automotive, and naval industries due to its low cost and excellent performance.7However,FRPC is sensitive to even Low-Velocity Impacts(LVI)and sustains substantial damage without surface manifestation, reducing the structural strength and threatening aircraft’s safe operation.8-10Therefore, it is necessary to study the LVI response of composite materials and the damage detection methods.

Many works on the LVI response of TP/TS composite materials revealed that the damage modes of composites under LVI include matrix cracking, delamination, debonding, fiber pull-out,and fiber breakage,which occur in succession or coincide as the impact energy increases.11-14Srinivasan et al.15investigated and characterized the damage modes and damage propagation process of TP/TS composites following LVI and Compression-After-Impact (CAI) using Scanning Electron Microscope (SEM) and ultrasound. However, the mechanical characteristics (force, deflection, and energy) were not analyzed due to instrument limitations. Carrillo et al.16found through experiments that the energy absorption mechanism of aramid fabric/polypropylene laminates was different from that of the plain layered aramid fabric laminates and considered the application of TP resin for improving impact protection performance and reducing cost. Vieille et al.5compared the LVI response characteristics and damage mechanisms of three composites with different resins(epoxy resin,polyphenylene sulfide, and polyetheretherketone (PEEK)) using ultrasonic C-scan and microscopy and found that PEEK composites exhibited better impact resistance performance among all the three composites. Arikan and Sayman17conducted single and repeated impact experiments on the Eglass fiber reinforced polypropylene and epoxy composites and found that,with similar perforation energy,the maximum contact force of TP composites was much smaller than that of the TS composites, and TP composites showed better impact resistance under repeated impacts. Recently, Kazemi et al.6compared the impact characteristics of the thermoplastic Elium®resin, liquid at room temperature, FRPC laminates and two epoxy resin laminates, and found that Elium®resin laminates showed better ductility and absorbed more energy through more elastoplastic deformation, achieving a 40%higher energy absorption and a 240% higher structural integrity. Comparative studies on the LVI response of carbon fiber reinforced thermosetting epoxy resin and glass fiber reinforced thermoplastic polypropylene composites are still rare or lack details. Further study is needed.

In terms of the impact damage detection methods for composite materials, the non-contact Infrared Thermography(IRT) with fast detection speed and intuitive detection results has received extensive attention in recent years. In particular,IRT supports online monitoring,which is of great significance for the real-time observation of the composite materials damage initiation and evolution process during LVI and the study on the mechanism of damage propagation. Meola,18-20Boccardi,21,22and Jakubczak23et al have done an amount of work on IRT online monitoring. Infrared cameras were used for monitoring the thermal phenomena of composites with different fibers or matrix under LVI. They found that the impact response of composites includes thermo-elastic effect and thermo-plastic effect. At the same time, some breakthroughs have been made in heat map processing. Afterward, Maierhofer24also applied IRT to the online monitoring of different Fiber-Reinforced Polymer (FRP) materials under LVI and found that the thermal images were suitable for monitoring the LVI damage expansion process. The monitoring results might be further applied for testing new composites. In previous work, we adopted the active/passive IRT methods to preliminarily investigate the damage detection and damage evolution of CFRP/GFRP LVI and fatigue tests, in which the advantages of IRT in damage detection and mechanism research were verified.25-28In addition, the damage mode of thermosetting composites with different fibers were characterized by infrared thermal imaging under LVI.29However, the relationship between the infrared heat maps and impact characteristics curves of composite materials under LVI and the damage mode characterization of IRT in thermoplastic composites have not been analyzed.

In this paper, a comparative experimental study is carried out on the LVI characteristics and infrared radiation characteristics of TP polypropylene and TS epoxy laminates. The impact tests are performed at five energy levels,and the results include characterization parameters such as impact force, displacement, and energy absorption. The impact characteristics of the laminates are analyzed and evaluated through active/passive thermography, ultrasonic C-scan, and optical microscope. A unified characterization of the TP/TS composites damage modes with IRT and studying the applicability of active thermography can promote the application of IRT in composite materials. Meanwhile, the relationship between infrared heat maps and impact characteristics curves is investigated, and the correlation between the two is elaborated. This research has certain reference significance for designers and engineers to improve the impact resistance and damage tolerance of TP/TS composites.

2. Material and methods

2.1. Material (Sample preparation)

The laminate specimens are E-glass fiber reinforced polypropylene and T800 carbon fiber reinforced epoxy resin (YH69)composite materials. The layup of the specimens is all[45/0/-45/90]4s; the dimension is 150 mm × 100 mm × 4.8 m m with a ply thickness of 0.15 mm. The Rapid Scan 2 roller type ultrasonic C-scan equipment with the scanning accuracy of 0.8 mm was applied for factory inspection, and the results show that the specimens have good surface quality and no internal defects such as delamination and inclusions.

2.2. Experimental methods

According to the ASTM D7136/D7136M—12 standard,30Instron-Dynatup 9250HV impact testing machine (Fig. 1(a))with the impact velocity accuracy of±0.002 was used to carry out the LVI test.The testing machine comprises an impactor,a pneumatic clamping device, a pneumatic rebound braking device, and a digital acquisition unit (Fig. 1(b)). The impactor’s weight is 12.527 kg,and the hammerhead is a hemispherical hammerhead with a diameter of 16 mm. The specimen is placed in the frame holder with a 150 mm×100 mm open window in the center.Four cylindrical rubber heads are used to fix the specimen and prevent vibration during impact. The Infra Tec Vhr 680 infrared camera with the measurement accuracy of±1.5 ℃(0-100 ℃)(or±2 ℃(＜0 ℃or ＞100 ℃))is placed vertically on the backside of the laminates’ impact face for online monitoring during the LVI test, and the acquisition frame rate is set to 50 Hz. Mainly focuses on the assessment of low-velocity impact damage of the laminates; hence, four lower energy levels(5,10,15,20 J)and one higher energy level(35 J)are used in the test,and each energy level test contains at least 2 specimens.In addition,the pulsed and ultrasonic infrared thermography detection equipment performed after impact test are shown in Fig. 1(c) and (d).

3. Results and discussion

The damage characteristics of TP/TS composite laminates,the heat map sequence during the impact process, and the impact response curves under LVI were analyzed. As the analysis of the heat map sequence and the impact response curves need to be conducted in conjunction with the macroscale and mesoscale damage characteristics of the laminates, the laminates’ damage characteristics after LVI are discussed first.

3.1. Damage feature

According to the experimental plan,after the LVI test with different energy levels, the specimen surface was visually inspected before the ultrasonic C-scan. The results are shown in Figs. 2 and 3.

According to Fig. 2, the two composites show different damage characteristics with the increase of impact energy.When the impact energy is lower than 20 J, the TP laminate surface shows dents and bulges of different depths. There are dents and matrix cracking on the TS laminate’s impact face,while no visual damage exists on the back face. When the impact energy is 35 J, there are many fiber breakages and matrix cracking on the back face of the TP laminate, while only matrix cracking appears on the back face of the TS laminate.Overall,TP laminates’damage is much more severe than that of the TS laminates, which shows that the impact resistance of TP laminates is weaker than that of the TS laminates.Moreover, apparent matrix cracking can be observed on the TS laminates’ impact face compared with the TP laminates,which is closely related to its brittleness and the axial stress of the laminates caused by impact.31According to the ultrasonic C-scan results in Fig. 3, as the impact energy increases,the laminates’ delamination area gradually increases, and the layer farther from the impact face generates more severe delamination damage because of the stress waves generated during the impact. The stress waves propagating along the thickness direction to the back face of the laminate reflect and generate tensile stress, resulting in matrix cracking. With the extension of the interlayer matrix cracking, delamination follows.With the attenuation of the stress wave,the delamination area gradually decreases from the back face in the thickness direction.32Combined with the results of visual inspection, it was concluded that the TS laminate has minor damage when the impact energy is 5 J, indicating that the impact energy at this time is near the impact energy threshold of the TS laminate.The TP laminates’delamination area is larger than that of the TS laminates with 10, 15, 20 J impact energy while smaller with 35 J impact energy, indicating that the initial delamination resistance of the TP laminates is weaker than that of the TS laminates. However, the TP laminates show excellent delamination resistance with the increase of impact energy. All of this agrees with Ref. 5, attributing to the excellent interlaminar fracture toughness of the thermoplastic resin.

The Aosvi M330 optical microscope with the inspection accuracy of±0.1 μm was used to observe the laminate surface and fracture and get further insight into the laminates’damage modes. The microscopic details of laminate surfaces with 10 J and 35 J impact energy are shown in Fig. 4. With 10 J impact energy, there was almost no damage to the TP laminate’s impact face,but fiber-matrix debonding appeared on the back face. The TS laminate showed minor fiber breakage on the impact face due to its rigidity, but no damage appeared on the back face.With 35 J impact energy,a large number of fiber breakages and fiber-matrix debonding were found on the back face of the TP laminate while still no damage on the impact face, indicating that the TP laminate may not have matrix cracking, which agreed with the findings of Reyes and Sharma.33Meanwhile, apparent matrix cracking and fiber breakages appeared on the impact face of the TS laminate. In contrast, matrix cracking was found along the sublayer fiber direction on the back face.

To further investigate the damage mechanism after impact,the specimens’vertical pit center is cut to observe the cross section via metallographic microscope. The result is shown in Fig. 5. The TS laminates delaminate at the interface between adjacent layers with the matrix cracking extension in Fig. 5(a).The delamination starts from the layer close to the impact face and spreads to the back face, and the damage extension direction is consistent with the direction of the fiber layer.34The micrographs of the TP/TS laminate cross section under 35 J impact energy are shown in Fig.5(b)and(c),respectively.In Fig. 5(b), delamination and matrix cracking can be observed, while fiber breakage and delamination damage can be observed in Fig. 5(c), which is consistent with the results shown in Figs. 2 and 3.

Considering the test results in Figs.2-5,the damage modes of TP laminates are dents,delamination,fiber-matrix debonding, and fiber breakages under the five impact energy levels,while the damage modes of the TS laminates include dents,matrix cracking, and delamination. Overall, TS laminates under the same thickness and boundary conditions exhibited better impact resistance, but the TP laminates showed higher delamination ductility under high energy impacts.

3.2. Heat map sequence

According to the analysis in Section 3.1, LVI results in dents on the specimens’ impact face, bulges on the back face, and a certain degree of interior damage. Also, the impact causes slight vibration of the specimen on a macroscopic level.Therefore, Eq. (1) can be obtained through the law of energy conservation:35,36

where Eais the energy absorbed by the specimen,which is consumed in four ways:damage,plastic deformation,friction and slight oscillation;Evis the specimen’s vibration kinetic energy,which characterizes the slight vibration of the specimen after LVI; Efis the energy consumed by friction, which is used to characterize the friction between the punch and the specimen in contact; Epis the energy consumed by plastic deformation,which characterizes the deflection, pits, bulges, and thickness reduction; Edis the energy consumed by damage formation,which characterizes the impact damage such as matrix cracking,delamination,and fiber breakage.According to the mechanism of damage energy release,the formation of damage(Ed)is accompanied by energy release. Due to the heat conduction effect, the temperature field on the material’s surface changes.The vibrational kinetic energy (Ev) is also dissipated in the form of heat due to the damping effect, thereby changing the temperature field of the material surface. Kang and Kim35pointed out that plastic deformation energy (Ep) and damage energy (Ed) are the two most crucial energy absorption mechanisms during the impact process. The object oscillates very slightly during impact, and the corresponding vibrational kinetic energy (Ev) accounts for a tiny proportion of the total absorbed energy (Ea); therefore, Evcan be ignored. Although the friction energy Efoccupies about 10% of the impact energy, it is still one of the lowest energy dissipation modes compared with the plastic deformation property and the damage energy(80%of impact energy),so it can also be ignored in the qualitative characterization of the damage mode.36In conclusion, the change of temperature in the surface of the material during impact mainly comes from the damage.

Time sequence analysis is conducted on the passive thermography monitoring results of TP/TS laminates. The heat map frame before the impact is defined as time t=0.The heat map sequence before and after the impact is extracted. The specimens are labeled according to the impact energy.The heat maps are shown in Figs. 6 and 7. The thermal anomaly zone near the impact site of the specimen is the impact damage area.As the formation of impact damage is accompanied by energy dissipation,the temperature is higher than the background.At 0.04 s after impact, the hot zone began to appear, gradually became apparent,and began to fade again.The extent of damage and energy dissipation increased steadily with the increase of impact energy.The impacted zone generated higher temperatures and a larger thermal anomaly zone. These results are consistent with the ultrasonic C-scan results in Fig. 3.

In the heat map sequence of TP and TS laminates with 10 J impact energy, a dark zone with a temperature lower than the surroundings appeared at the impact point 0.16 s after impact.About 1.00 s later, the dark zone at the impact point disappeared, and a hot zone appeared. The main reason for this phenomenon is that the pits generated by the impact cause the structure’s thickness to decrease,leading to increased local density ρ. According to α = k/(ρc), the material thermal conductivity k and the specific heat capacity c are both constants.The decrease of thermal diffusion coefficient α would decrease the impact point’s ability to approach thermal equilibrium locally.However,in the heat map sequence of TP and TS laminate with 35 J impact energy,hot zones with temperatures significantly higher than the surroundings appeared at the impact point due to the fiber breakage or the extensive internal delamination on the back face of specimens corresponding to the impact point, resulting in high heat dissipation and the significant increase in local temperature.

The change process of TP/TS laminates’ surface temperature is analyzed to visually characterize the heat dissipation effect in the impact process, which includes two aspects: the time series characteristics of the surface temperature under different impact energy and the surface temperature distribution under the same impact energy.The surface temperature difference is used to analyze the surface temperature’s temporal change characteristics under different impact energy to eliminate environmental interference and the surface characteristics of specimens on the surface temperature, as shown in29

According to Fig.8,the temperature difference on the specimens’ surface is 0 before 2 s because the specimens have not been impacted,and there is no heat dissipation.The specimens are damaged, and the heat is released due to impact after 2 s,which cause the local temperature difference on the surface of specimens to rise suddenly. Subsequently, the specimen’s surface temperature field tends to balance due to the continuous dissipation of heat,and the corresponding surface temperature difference gradually decreases.

Fig. 9 shows the maximum surface temperature difference of TP/TS laminates under different impact energy. The maximum surface temperature difference between the laminates increases with the increase of impact energy. However, the TP laminates show a significant nonlinear relationship compared to the TS laminates, specifically as follows: the maximum surface temperature difference of specimens with 5-15 J impact energy change slightly by about 1.0-2.0 °C at the moment of impact. However, the specimens’ surface temperature difference with 20 J and 35 J impact energy increase significantly by about 9 °C and 24 °C, respectively. Combined with the analysis of the ultrasonic C-scan results in Fig. 3 and microscopic inspection results in Fig. 4, the reason for this nonlinear relationship is that the primary damage mode in the specimen changes with the increase of impact energy.However, the TS laminates exhibit an approximately linear relationship, and the maximum surface temperature difference between specimens changes slightly by about 0.5-2.0 ℃, indicating that the modes of impact damage in all TP laminates and TS laminates at 5-15 J impact energy are mainly matrix cracking and delamination. In contrast, the impact damage of TP laminates with 20 J and 35 J impact energy have a large number of fiber breakages and fiber-matrix debonding in addition to matrix cracking and delamination.It is further inferred that the formation of matrix cracking and delamination damage produces less energy dissipation, while fiber breakage and fiber-matrix debonding are accompanied by more significant energy dissipation. Therefore, the temperature rising effect can be used as an effective way to characterize the damage mode in the passive thermography detection of impact damage. The weak temperature rising effect could indicate that the specimen has matrix cracking and delamination;the strong temperature rising effect indicates that the specimen has fiber breakage and fiber-matrix debonding mode.

Given the different damage modes of the specimens under different impact energy, the passive thermography detection results of specimens of TP/TS laminates with 10 J and 35 J impact energy are taken as examples to analyze the distribution characteristics of the surface temperature under the different impact energy levels. Six pixels in the impact damage area and background area of specimens with 10 J and 35 J impact energy are marked,where a1is the pixel at the impact position,a2-a5are the pixels in the impact damage area, and a6is the pixel in the background area, as shown in Fig. 10(a), (d), (g),and (j). The temperature time series curve of each pixel is extracted,as shown in Fig.10(b),(e),(h),and(k).The temperature of the impact damage area rises instantaneously to the maximum after being impacted, but each pixel’s rising rate is different. Fig. 10(c), (f), (i), and (l) shows the temperature change curves of each pixel within 0.2 s of specimens with 10 J and 35 J impact energy after being impacted.For the specimens of TP/TS laminates with 10 J impact energy,the temperature rise rate of pixel a1is significantly lower than that of pixel a2, and the temperature of the corresponding pixel a1is lower than that of pixel a2. For the specimen of TP/TS laminates with 35 J impact energy, the heat dissipation at the impact point is higher than other areas due to fiber breakage and fiber-matrix debonding, so the temperature rise rate of pixel a1is higher than other areas, which further confirms the analysis results of the dark zone at the impact point of the specimen with 10 J impact energy and the bright zone at the impact point of the specimen with 35 J impact energy in heat map sequence.

To further analyze the thermal difference between the damaged area and the background area, the heat maps of TP/TS laminates with 10 J and 35 J impact energy at 0.04 s are taken as examples. As shown in Fig. 11(a), (c), (e) and (g), two temperature measurement lines are defined in the heat map,among which l1passes through the temperature measurement line of the impact point,and l2passes through the background area. The temperature data on the temperature measurement lines l1and l2are extracted in turn,and the temperature difference between the two temperature measurement lines is obtained. The result is shown in Fig. 11(b), (d), (f) and (h).

The pixel temperature of TP/TS laminates in the damaged area is higher than the pixel temperature in the background area. For the TP/TS laminate specimens with 10 J impact energy, the temperature difference curves of l1and l2are both bimodal, with two maximum values and one minimum value,indicating the existence of dark zones. For the specimen with 35 J impact energy,although the temperature difference curves of TP/TS laminates are single-peaked with only one maximum value, the temperature rise effects of TP/TS laminates are different at the impact point. The TP laminate shows a strong temperature rise due to many fiber breakages and fiber-matrix debonding, while the TS laminate shows a weak temperature rise. According to the microscopic results of Fig. 5, although TS laminate with 35 J impact energy does not have a large number of fiber breakages at this time, there are a large number of delamination and minor fiber breakages.The characteristics of TS laminate with 35 J impact energy is similar to TP laminate with 15 J impact energy.According to Fig.12,a large amount of fiber-matrix debonding and minor fiber breakages occur.

According to the heat map sequence and temperature evolution analysis,the dark zones at the impact,the weak temperature rise effect, and the bimodal temperature measurement line could characterize the appearance of matrix cracking and small delamination; the bright zones, the weak temperature rise effects, and the single-peak temperature line could characterize the occurrence of a large number of matrix cracking, delamination, fiber-matrix debonding and minor fiber breakages; the hot zones, the strong temperature rise effects,and the single-peak temperature line could characterize the appearance of a large number of fiber breakages and fiber-matrix debonding.

A thermal excitation source composed of pulsed flash lamps arranged in parallel was used in the experiments to verify the applicability of pulsed thermal imaging in detecting TP/TS composite materials. The results are shown in Fig. 13. Since pulsed thermography is sensitive to thermal insulation damage, the hot spots corresponding to fiber breakage are often covered by matrix cracking and delamination damage. Therefore, the hot spots in the heat map represent the presence of matrix cracking and delamination damage.37There is a dark area in the central zone of the hot spot with a lower temperature than the surrounding area. The reasons are: the impact behavior causes the thickness of the laminate to be thinner,and the reduction of the interlayer gap causes the density of the central area to be higher than the density of the surrounding area.The difference in density affects the heat transfer rate,which leads to a lower temperature in the impact center than in the surrounding area, forming a dark zone. The range of matrix cracking is small on the surface, resulting in discontinuous surface hot spots.

Comparing the pulse thermography results of the two composites in combination with Fig.2 shows that the visible damage of TP laminates is relatively more severe, but the damage detected by the pulse infrared method is not apparent.The TS laminates have almost no damage on the inspection surface,but outstanding damage detection results under pulse excitation. This phenomenon is probably due to the fact that the TP laminate exhibits a ductility similar to metal under LVI.The pits caused by impact are not much different from the bulge’s height on the back face of the specimens, so the thickness of the impact zone changes slightly,and the thermal resistance capacity shows almost no change. The heat accumulation in the impact zone is not much different from the rest of the specimens under pulse excitation, so the pulse detection effect is poor. However, due to the TS laminates’poor overall ductility, energy is dissipated mainly through structural damage and destruction when facing impact behavior. After the impact, even with barely visible impact damage on the surface, a large amount of internal damage already occurs. The appearance of damage causes the unequal thickness of the laminate and the increase in thermal resistance.As a result, the TS composite has a better detection effect.

3.3. Mechanical characteristic curves

Typical impact response characteristic curves of composite materials mainly include impact force versus time (f-t) curve,impact force versus displacement (f-d) curve, and absorbed energy versus time (e-t) curve.

Fig.14 shows the f-t curves of TP/TS laminates with different impact energy levels. The first sudden load drop point on the curve is called Hertzian failure, representing the rapid propagation of laminate delamination damage.38-40The impact force at this time is usually called the Hertzian force FH. After the sudden drop, the force continues to fluctuate and peaks at Fm. These fluctuations represent the formation process of laminate damage,41,42so the damage degree of laminate continues to increase until the impact force is offset completely.The Hertzian force FHunder different impact energy is approximately the same for the same composite material. FHof the TP laminate (TP/E-glass) is about 4.00 kN, and that of the TS laminate(TS/T800 C)is about 7.00 kN.In addition,FHof glass fiber reinforced epoxy resin laminate (TS/E-glass)is about 6.75 kN by using a variety of energies (20, 40, 50,60 J) to impact, which is similar to FHof carbon fiber reinforced TS laminate. Therefore, it can be inferred that FHis mainly related to the matrix, possibly because the laminate counteracts impact loads mainly by generating delamination damage before substantial fiber fracture damage occurs and the delamination is mainly connected with the transverse matrix cracking.43The reason why FHof carbon fiber reinforced TS laminate is larger than that of glass fiber reinforced TS laminate may be the difference in fiber strength. Both matrix and fibers may significantly influence the magnitude of FH.

Under the five impact energy levels,the TS laminates’maximum peak load is 76.6%, 57.4%, 50.0%, 47.1%, and 41.8%,respectively,which are higher than those of TP laminates.This superiority is because of the good rigidity and load-bearing capacity offered by the TS resin and high modulus carbon fiber. Moreover, the better impact toughness of the TP laminates causes the Fmdifference proportion to decrease as the impact energy increases.For example,with 35 J impact energy,the connection between heat maps and the f-t curves is established and shown in Fig. 15. As the frequency of curve fluctuations increases, the damage degree of the specimen and the energy dissipation caused by the damage also continues to increase, and the size and brightness of the hot spot area in the heat map also increase(the hot spot is the temperature field of laminate impact area). The curve fluctuation lasts for certain time after Fmoccurs, and the laminate damage is no longer produced after the fluctuation. Therefore, the hot spot area and the brightness of the heat map peak at the fluctuation cutoff.

According to TP/TS laminates’ damage states during the impact process, the f-d curve can be divided into three stages:the elastic deformation stage,the damage extension stage,and the rebound stage. According to Fig. 16, the impact force depends linearly on the displacement in the elastic deformation stage. The force at this time is less than FH. With the gradual increase of impact energy,the curve slope in the elastic stage of laminate increases, increasing the laminate stiffness.17,44After the Hertzian failure occurs in the damage extension stage, the curve fluctuates sharply, which is a gradual extension of the damage.According to the test results in Fig.16 and the related analysis, the greater the impact energy, the longer the damage expansion stage, and the greater the damage produced after the impact.Finally,when the specimen deflection piece reaches the maximum value Lm,the punch begins to enter the rebound stage. In the rebound stage, the contact force between the punch and the specimen gradually decreases while the rebound velocity gradually increases. At the end of the rebound stage,the specimen’s final central displacement Lfdoes not return to the origin, which indicates irreversible plastic deformation of laminate. The higher the impact energy, the greater the damage and irreversible plastic deformation produced. The area enclosed by the f-d curve is the absorbed energy of laminate.45

Comparing the f-d curves of TP/TS laminates at five impact energy levels reveals that the maximum impact displacement and final center displacement of laminate increase with the increase of impact energy. The difference between the maximum center displacement of the two types of laminates is 1.9,2.3,2.3,2.6,and 3.5 mm,respectively,which indicates that TP laminate produces greater deformation and damage at the same impact energy level.Severe damage leads to smaller elastic deformation, while the energy required for the rebound is stored in the elastic deformation,46resulting in the larger final central displacement of the TP laminate. The connection between heat map sequence and f-d curve at 35 J impact energy level is shown in Fig.17.The curve’s fluctuation is the same as the f-t curve in Fig. 14, which also implies the generation and development of the damage. Therefore, the relationship between heat map characteristics and fluctuation of the f-d curve is also the same as the relationship between heat maps and the f-t curves. According to Fig. 17, the fluctuation is cut off at the maximum center displacement, after which the punch starts to rebound, at which time the heat spot area and brightness of the heat map near the maximum central displacement reach their peak due to the end of the damage event.Subsequently,the energy is gradually dissipated in the rebound stage,the hot spot area gradually decreases,and the brightness gradually decreases.

The e-t curves of laminates are divided into three stages,as shown in Fig. 18. In Stage I, the laminate absorbs energy through the crater caused by elastic deformation, absorbing a relatively small amount of energy.47In Stage II,the laminate absorbs impact energy by generating elastic-plastic deformation and various types of damage, at which time the curve is parabolic and increases rapidly to the maximum absorbed energy Em.In Stage III,when the punch center’s displacement reaches the maximum and the velocity becomes 0,the laminate gradually separates from the surface by feeding the previously accumulated strain energy back to the punch. Thus, the absorbed energy curve decreases rapidly to a constant value Efand then remains constant. The above process shows that rebound energy Ermakes the absorbed energy of laminate smaller than the impact energy under LVI, while the diversity of damage modes makes the e-t curve of laminate nonlinear.6,48The parameters mentioned in Figs. 14, 16, and 18 are shown in Table 1.

At the first four impact energy levels(5,10,15,20 J),the TP laminate can absorb more energy. The absorbed energy of the TS laminate is 24.6%, 18.6%, 15.5%, and 19.4%, which are lower than the TP laminate, while the absorbed energy of the TS laminate is 21.4%,which is higher than the TP laminate when the impact energy rises to 35 J. This difference may be due to the fact that the TP laminate sustains more damage than the TS laminate and absorbs more energy through the damage at lower impact energy. When the impact energy is higher, the TS laminate absorbs more energy because of its higher strength and higher load-bearing capacity. The above discussion shows that the TP laminates have better energy absorption capacity than the TS laminates under the same conditions when subjected to lower impact energy, but the opposite is true for higher impact energy. However, the maximum contact force of the TP laminates is only 2/3 of that of the TS laminates. The low bending stiffness and ductility of TP laminate cause that the difference in energy absorption level between the two is not significant.

Table 1 Key parameters of TP/TS laminates under different impact energy levels.

The relationship between the heat map sequence and the e-t curves at 35 J impact energy level is shown in Fig. 19. As the impact energy increases, the absorbed energy of the laminate is also increased gradually. The specimen absorbs a large amount of energy through the damage. The higher the absorbed energy,the greater the degree of damage.Therefore,the hot spot’s area and brightness on the heat map also increase and peak near the maximum absorbed energy.

4. Conclusions

This paper experimentally studies the LVI response characteristics of TP and TS laminates. Based on the results obtained,the following conclusions are drawn:

(1) From the perspective of impact resistance, the TS laminates, under the identical circumstances, exhibit better impact resistance, while the TP laminates show higher delamination ductility, and the maximum contact force of TP laminates is much smaller than that of the TS laminates under LVI. However, the low bending stiffness and low ductility of the TP matrix cause that the difference in energy absorption level between the two is not significant.

(2) The damage detection effects on TP/TS laminates of pulse thermography are different,the TS laminates show better detection ability than the TP laminates. And in the passive thermal monitoring,for composites with different matrix,the dark zone at the impact area,the weak temperature rise effect, and the bimodal temperature measurement line could characterize the appearance of matrix cracking and small delamination. The hot zone,the weak temperature rise effect,and the unimodal temperature measurement line could characterize the appearance of severe matrix cracking, severe delamination, fiber-matrix debonding, and minor fiber breakages. The hot zone, the strong temperature rise effect,and the unimodal temperature measurement line could characterize the appearance of a large number of fiber breakages and fiber-matrix debonding.

(3) The Hertzian forces are different for each composite.The Hertzian force of TS laminate is greater than TP laminate. The Hertzian force not only depends on the type of matrix but also relates to the type of fiber. The effect of secondary factors (such as layup method, specimen thickness, manufacturing process, and impactor shape) on Hertzian force is not clear, which needs further exploration.

(4) The fluctuation of the impact characteristic curve(f-t,fd) is directly related to the hot spot characteristic changes in heat maps.The more frequent the curve fluctuation,the larger the hot spot and the higher the brightness in the heat map. The hot spot area and brightness peak at the maximum impact load after the f-t curve fluctuation cutoff point, the maximum center displacement of the f-d curve, and the maximum absorbed energy of the e-t curve.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The ultrasonic C-scanning equipment was provided by China Aircraft Strength Research Institute.This study was supported by the Major Research Plan of the National Natural Science Foundation of China (No. 92060106), the National Natural Science Foundation of China(No.52075541),the China Postdoctoral Science Foundation (No. 2019M650262), and the Natural Science Foundation of Shaanxi Province, China(No. 2020JM-354).