Experimental Study on the Impact Resistance Performance of Civil Air Defense RC Walls Protected by Honeycomb Sandwich Panels

Abstract

Reinforced concrete (RC) walls are extensively used in civil air defense engineering and are susceptible to low-speed impacts with high energy and massive weight during service. Therefore, it is crucial to consider the impact resistance of these walls and explore effective protection methods. The honeycomb structure, known for its energy-absorbing properties, has been widely utilized in aerospace, automotive, and maritime industries. However, there is a need for more research on applying honeycomb structures in energy-absorbing protection devices in civil engineering. This study proposes the use of honeycomb sandwich panels to protect civil air defense RC walls, creating a Honeycomb Sandwich Panel Composite RC wall (HSP-RC wall). Through pendulum impact experiments, we investigated the dynamic response of RC walls, both standard and HSP-RC walls with varying honeycomb parameters, under high-energy impacts. The goal was to evaluate the impact resistance of these RC walls, analyze the deformation differences among different HSP-RC walls, and examine the influence of honeycomb parameters on the impact protection effectiveness. The research results indicate that honeycomb sandwich panels can provide impact protection for RC walls by absorbing energy, and their protection effect is related to the parameters of the honeycomb core layer. This research result can be applied to RC structures that bear impact loads, achieving effective protection for RC structures.

1. Introduction

The walls in civil air defense projects are often made of cast-in-place reinforced concrete, and the effects of various impact loads must be considered. However, more research must be conducted on the out-of-plane, high quality, low-speed impact of RC walls in civil air defense engineering. Given that civil air defense RC walls are susceptible to such impact effects during their service, targeted protection measures are needed for some important RC walls to improve their impact resistance performance.

Regarding the investigation of the impact resistance of reinforced concrete (RC) structures, most research concentrated on RC beams [1,2,3,4], columns [5,6,7,8,9], and slabs [10,11], with comparatively less focus on RC walls. Studies concerning the impact resistance of RC walls encompass both standard RC walls [12] and cantilever RC walls [13,14]. Yong [13] carried out extensive impact tests on cantilever RC walls (used as rockfall barriers), employing impact devices, and introduced a displacement-based DB model to predict the wall deflection and the tensile strain in the steel bars at the base of the wall, aiding in the design of RC rockfall barriers. Arnold [14] performed pendulum tests on cantilever RC walls and developed a displacement-based analysis model to assess and identify crucial design parameters for cantilever RC barrier walls based on deflection and strain data.

In enhancing the impact resistance of RC structures, apart from utilizing high-performance concrete, prestressed reinforced concrete, and fiber-reinforced concrete, most external protective measures for structures involve composite fiber material constraints, steel plates, and steel wire mesh. In addition to RC beams and columns, the research objects of impact protection are mostly RC slabs [15,16], masonry walls [17,18], stone fall protection RC walls [19] and bridge piers [20,21,22].

Zineddin [15] studied the dynamic behavior and failure of RC slabs protected by different types of steel mesh protective layers under impact loads through drop hammer tests. Radnic [16] studied the dynamic response of RC slabs with added CFRP reinforcement strips at the bottom using a small mass drop hammer test (drop hammer mass of 4.208 kg). Compared with ordinary RC slabs, the ultimate bearing capacity of RC slabs reinforced with CFRP was only slightly improved under static and impact loads. In the event of stamping shear failure, CFRP-reinforced RC slabs do not significantly contribute to the ultimate bearing capacity of the plates under static and impact loads.

Mason [17] conducted pendulum impact tests to investigate masonry walls’ out-of-plane impact resistance reinforced by CFRP. Mohd [18] used ABAQUS/Explicit to study the explosion protection of masonry walls reinforced with steel wire mesh and CFRP strips and evaluated their protective effects from the perspectives of displacement and damage. Perera [19] studied the dynamic behavior of a cantilever RC wall protected by a Gabion Cushion after impact. The protective effect can be evaluated from the perspective of the deflection and damage reduction of RC walls, and it is extremely important to seek better out-of-plane impact protection measures for walls and slabs and systematically study their protective effects and design methods for the impact protection of RC components. In civil engineering, research on using porous materials for the energy absorption protection of reinforced concrete structures includes porous energy absorption devices for protecting RC piers [20,21] and foam aluminum for protecting RC columns [22].

The principle behind using honeycomb sandwich materials to protect reinforced concrete (RC) structures lies in their ability to absorb energy through plastic deformation and fracture when subjected to impact loads. Previous studies [23,24,25] have demonstrated that honeycomb composite materials effectively control the damage extent and impact energy transfer during accidents by rapidly absorbing energy. This local deformation characteristic of honeycomb structures under impact loads can be leveraged to manage damage range and enhance safety protection in civil structures. Additionally, honeycomb sandwich structures possess higher strength. Therefore, this study proposes using honeycomb sandwich structures to protect RC walls and analyzes the impact resistance performance of Honeycomb Sandwich Panel Composite RC walls (HSP-RC walls).

2. Scaled Specimens and Pendulum Impact Experiments

This section introduces the pendulum impact experiments of RC walls under honeycomb sandwich panel protection, including the scaled civil air defense RC walls, honeycomb sandwich panels, pendulum impact experiment system, and experiment parameters.

2.1. Structure and Materials of the RC Wall

Referring to the RC wall in a civil air defense project in Qingdao, a 1:2 scale model was used to design RC wall specimens, as shown in Figure 1. The thickness of the reinforcement protective layer is 20 mm, and the diameters of the longitudinal, distribution, and tie bars are each 8 mm. The longitudinal reinforcement ratio, denoted as ρ, is 0.726%.

The compressive strength of the concrete, f_c0, is 42.24 MPa, while its tensile strength, f_t0, is 2.39 MPa. The rebars used are all HRB400 grade with diameters of 8 mm, 20 mm, 22 mm, and 25 mm. From tensile tests, the yield strength f_yo, ultimate strength f_uo, and elongation are 425 MPa, 603 MPa, and 20.7%, respectively.

2.2. Honeycomb Sandwich Panel

Composite panels on both sides of the honeycomb core can form a honeycomb sandwich panel. In the honeycomb sandwich structure of this experiment, the honeycomb core layer is a classic hexagonal aluminum-based honeycomb structure. This type of honeycomb structure has three orthogonal directions. The two in-plane directions (direction L and direction W), the structural parameters of the honeycomb and the arrangement pattern of the honeycomb holes are shown in Figure 2a. The thickness direction of the honeycomb aluminum is the out-of-plane direction, which is represented as the T direction, as shown in Figure 2b. The honeycomb sandwich panel formed by bonding aluminum alloy panels on both sides of the honeycomb core is shown in Figure 2b, and the connection between honeycomb and panels is through adhesive bonding. H represents the thickness of the honeycomb aluminum core layer of the specimen, and h represents the thickness of the aluminum alloy panel.

Here are the specific dimensions of the honeycomb sandwich panels utilized in the experiment. The thickness H of the honeycomb core layer is 40 mm. The thickness h of both surface panels is 3 mm. The edge lengths l of the honeycomb holes are 1 mm, 1.5 mm, 2 mm, 2.5 mm, and 3 mm, respectively. The honeycomb wall thickness t is 0.04 mm. The parameters of aluminum foil substrate are shown in

Table 1. Parameters of honeycomb aluminum foil material.

Average Thickness	Tensile Strength	Yield Strength	Percentage Elongation	Density	Poisson’s Ratio
0.04 mm	299 MPa	27.6 MPa	3%	2690 kg/m3	0.33

.

Considering the local characteristics of the impact, the overall in-plane dimensions L_L and W_W of the honeycomb aluminum sandwich panels are 1000 mm, as shown in Figure 3a.

2.3. HSP-RC Wall

The HSP-RC wall has an energy-absorbing covering layer made by pasting a honeycomb aluminum sandwich panel on the front of an original RC wall. A resin adhesive forms the adhesion between the honeycomb aluminum sandwich layer and the RC wall. The properties of the adhesion after curing are shown in

Table 2. Performance parameters of resin adhesive.

Colloidal Properties					Bonding Ability
Tensile Strength (MPa)	Elastic Modulus (MPa)	Elongation %	Bending Strength (MPa)	Compressive Strength (MPa)	Tensile Shear Strength of Steel-Steel (MPa)	Tensile Bonding Strength of Steel-C40 Concrete (MPa)
32.53	1675	1.73	41.57	76.86	11.65	6.53

.

The position of the honeycomb aluminum sandwich layer pasted in the center area of the front of the RC wall is shown in Figure 3a. The curing process after adding a sandwich layer to the RC wall, according to the reinforcement and pasting steps in the Chinese National Standard [26], is shown in Figure 3b. After two weeks of static curing, the HSP- RC walls are subjected to pendulum tests.

2.4. Pendulum Experiment of HSP-RC Walls

2.4.1. Pendulum Experiment System

The pendulum experiment system is consistent with reference [12], but with different specimens, depicted in Figure 4a, featuring the central point at the front of the RC wall as the impact position, with the impact direction perpendicular to the wall, coordinated by the experimental setup. The clamp plate, connected to the reaction frame, constrains the top beam, while the bolt, attached to the fixed ground beam, secures the bottom beam. The model’s boundary conditions are illustrated in Figure 4a. The pendulum has a cylindrical hammer head with a 10 cm diameter, equipped with a sensor to measure impact force. To prevent severe stress concentration and local damage due to the small contact area between the impact head and the RC wall, a 50 mm thick square iron plate (340 mm × 340 mm) is placed between them.

As shown in Figure 4b, two high-speed cameras are used to capture the impact response of the RC wall. Camera-1, positioned two meters from the side, records the mid-span displacement and velocity at 3644 frames per second. Camera-2, located five meters from the back, captures the crack propagation on the wall’s rear at 550 frames per second.

2.4.2. Impact Load and Constraints in the Experiment

The HSP-RC wall pendulum impact test consists of six civil air defense RC wall specimens, one of which is an original RC wall, and the other five of which are HSP-RC walls, with different honeycomb hole side lengths. The specimen number and the value of the honeycomb hole edge length of the specimen are summarized in

Table 3. Summary of wall specimen parameters.

Specimen Number	RC Wall	HSP-RC Wall-1	HSP-RC Wall-1.5	HSP-RC Wall-2	HSP-RC Wall-2.5	HSP-RC Wall-3
l	/	1 mm	1.5 mm	2 mm	2.5 mm	3 mm

Note: “/” indicates no addition of honeycomb sandwich panel.

.

The six wall specimens were tested under identical impact loads, with an impact mass (m) of 2 tons and impact velocities (v) of 2.3 m/s. The impact energy (E_k) for each test was 5290.0 J.

2.5. Impact Damage and Protection Effect of HSP-RC Wall

The dynamic behavior and failure of the HSP-RC walls after impact can be divided into three parts. The first one is the deformation and cracking of the protected structure (the RC wall); the second is the deformation and damage of the honeycomb aluminum sandwich panel; and the third is the damage of the adhesive layer connecting the honeycomb sandwich panel and the RC wall. The dynamic response and deformation failure division of the entire HSP-RC wall after the pendulum falls and impacts the specimen are shown in Figure 5.

Therefore, the next two sections will analyze the results of the impact experiment, study the dynamic behavior and the impact resistance performance of the HSP-RC wall, and discuss the protective performance of the honeycomb sandwich panels for the RC wall.

3. Impact Resistance Performance of HSP-RC Walls

This section focuses on the force and RC walls’ displacement under impact, analyzes the damage to RC walls from the perspective of concrete cracks, and then evaluates the protective effect of honeycomb sandwich layers and the impact resistance performance of HSP-RC walls.

3.1. Impact Force

During the complete impact process of a pendulum test, the pendulum will experience rebound after the first impact on the specimen. After the rebound, it will impact again until the energy is exhausted. The impact force–time curves between the hammer head and the specimens of the first impact on the HSP-RC wall’s honeycomb sandwich structure layer and the first impact on the RC wall during the experiment are shown in Figure 6, Figure 7, Figure 8, Figure 9, Figure 10 and Figure 11, respectively.

In all pendulum impact tests, the duration of the entire impact process during the first impact is about 40–45 ms. When the edge length of the honeycomb hole is 2 mm, the duration of the impact process is the shortest: about 40 ms. According to Figure 6, Figure 7 and Figure 8, when l increases from 1 mm to 2 mm, the duration of the impact process decreases from 45 ms to 40 ms. When l is greater than 2 mm, the entire impact process is 45 ms.

Under the same load and boundary conditions, the HSP-RC walls’ impact force-time curve can be divided into several stages, including the initial peak stage (stage number 1), zero value stage (stage number 2), fluctuation descent stage (stage number 3), plateau stage (stage number 4), and descent stage (stage number 5). Taking HSP-RC Wall-3, which has the most typical impact force variation, as an example, its impact force stages are shown in Figure 12.

The length of the zero value stage is related to the honeycomb parameter (the length of honeycomb hole l in this experiment) of the honeycomb sandwich structure. The zero value stage only appears when l ≥ 1.5 mm; as l increases, the zero value stage increases significantly. When l is 3 mm, the duration of the zero value stage is the longest; it lasts about 2.5 ms. During the entire impact contact process of specimens with different honeycomb sandwich panels, the maximum value of the impact force occurs in the fluctuation descent stage. During the fluctuation descent stage, the peak value of the impact force gradually decreases in a fluctuating manner until reaching the plateau stage. A bigger l will cause a longer duration of the fluctuation descent stage, about 13 ms when l is 3 mm, and make the range of the impact force’s fluctuation larger. The duration of the plateau stage of the impact force will vary slightly, fluctuating around 20 ms. The duration time of the descent stage will be consistent.

The duration time of the first impact on the original RC wall and different HSP-RC walls was not significantly different. The maximum impact force during the impact of a pendulum on HSP-RC walls was much greater than that of an original RC wall, with the former being about 3–5 times that of the latter. The maximum impact force is related to the material of the impact contact surface and the material properties of the two objects being impacted. The impact resistance of different HSP-RC walls is strongly correlated with the honeycomb parameter of the honeycomb core layer, denoted as l in this experiment. The plateau value reflects the specimen’s impact resistance, so compared with the original RC, the impact resistance of HSP-RC walls is greatly improved. The impact resistance of different HSP-RC walls is highly correlated with the honeycomb parameter of the honeycomb core layer, which is the l in this experiment. The first three stages of impact force reflect the absorption of impact energy by the sandwich panel, while the platform stage reflects the overall impact resistance performance of the HSP-RC wall.

Taking specimen HSP-RC wall-3 as an example, after the pendulum falls, a total of three consecutive impacts occur during the entire pendulum impact process. The impact force–time histories during these three impacts are shown in Figure 13.

When the HSP-RC wall is subjected to another impact, there is no significant change in the total impact time compared to the first impact. However, the duration of the zero value and fluctuation decrease stages increases, and the plateau stage disappears. At the same time, the zero value stage during the re-impact process is divided into two parts, and the peak value of the impact force appeared between the two zero value stage parts. According to the Hopkinson compression bar experiment, the honeycomb structure will have dynamic progressive buckling under impact compression. Therefore, it is speculated that the zero value stage of the impact force is caused by the dynamic progressive buckling of the honeycomb core under the impact. And the initial progressive buckling defect of the honeycomb core layer will have already appeared after the first high-energy impact. Therefore, the zero value segment lasts longer than the first impact during the re-impact process.

3.2. Impact Displacement

3.2.1. Mid-Span Displacement of the RC Wall

In the pendulum impact tests, the mid-span displacement of the different RC walls during the first impact process was measured, as shown in Figure 14.

After adding a honeycomb sandwich panel to the surface of the RC wall, the honeycomb sandwich panel layer had varying degrees of protection against the overall deformation of the RC wall under impact. The larger l of the honeycomb hole will cause a lower relative density and stiffness of the honeycomb core. It will provide a stronger protective ability against the mid-span deformation of the RC wall.

3.2.2. Mid-Span Residual Deflection of the RC Wall

All specimens were left to stand for 2 weeks to measure the residual mid-span deflection of the RC wall after the pendulum impact experiment. The mid-span residual deflection of the RC wall is shown in Figure 15.

Honeycomb sandwich panels can effectively reduce the mid-span residual deflection of RC walls. The larger the l of the honeycomb core, the higher the porosity, and the better the protective effect on RC walls’ permanent deformation. It should be noted that the slope between the residual deflection and honeycomb aperture is not constant. With a honeycomb hole edge length of 2 mm as the boundary, the slope of deflection reduction for l ≥ 2 mm is smaller than that for honeycomb hole edge length smaller than 2 mm. As the l of the honeycomb core increases, its improvement effect on residual deflection on RC walls will decrease.

3.3. Surface Cracks on RC Walls

This section mainly introduces actual surface cracks in RC walls under the protection of different honeycomb sandwich layers. The cracks on the RC wall back surface are divided, including the the crack propagation process, and the most severe wall cracks occurring during the impact process (widest crack).

3.3.1. Cracks on the RC Wall Side

There were no cracks in the area where the wall connects with the upper beam and the lower beam in any of the specimens. There were no complete through cracks on the side of the HSP-RC wall specimens, and the side cracks of all specimens are shown in Figure 16.

There were six major cracks on the side of the original RC wall. When the edge length of the honeycomb hole was 1 mm or 2 mm, there were four major cracks on the HSP-RC wall. When the edge length of the honeycomb hole was 1.5 mm, there were five major cracks on the HSP-RC wall. When the edge length of the hole was 2.5 mm or 3 mm, there were three major cracks on the HSP-RC wall. Therefore, the honeycomb sandwich panel layer can reduce the number of cracks on the side of the RC wall. Within the honeycomb hole side length range used in this experiment, the larger the honeycomb hole edge length, the better the crack reduction effect.

3.3.2. Crack Propagation on the Back of the RC Wall

The crack propagation process on the back of HSP-RC Wall-1.5 after the pendulum impact is illustrated in Figure 17. Initially, the impact load caused several horizontal long cracks and some vertical short cracks to appear on the back of the RC wall, with minimal crack width. As the impact progressed, these cracks gradually widened, reaching their maximum width during the peak of the impact. Subsequently, as the pendulum rebounded, the crack widths on the back of the wall gradually decreased and stabilized at a fixed value.

Figure 17 shows that the main transverse cracks on the back of the HSP-RC Wall-1.5 specimen occurred within 7.28 ms, corresponding to the initial peak stage, zero value stage and fluctuation descent stage of the impact force. Differing from the loop failure crack of the RC Wall specimen around the contact area first (shown in Figure 18), the cracks on the back of HSP-RC Wall-1.5 were mainly transverse cracks, and there were no loop-connected cracks in the contact center. There were far fewer micro-expansion cracks than in the RC Wall specimen.

3.3.3. Maximum Crack during Impact

The most serious cracks of all specimens during the first impact of the pendulum are summarized in Figure 19. In the process of the first impact on the wall after releasing the pendulum, all the cracks of the specimens first appeared in the mid-span of the RC wall, and the number of through cracks on the RC wall back protected by the honeycomb sandwich panel was smaller than that of the original RC wall. The cracks in the original RC wall showed a diffusion trend from the impact center to the whole wall, while there were more horizontal cracks in the RC wall protected by a honeycomb sandwich panel.

Combined with the crack propagation process, the addition of the honeycomb sandwich panel changed the transmission mode of the impact load. Due to the honeycomb sandwich panel, the failure behavior of the RC wall changed from the original impact local failure (circumferential cracks first appeared in the impact area) to the overall failure (similar to beams, horizontal through cracks appeared in the back of the mid-span).

The maximum crack width of each specimen during impact is quantified and summarized in

Table 4. Maximum crack width during impact.

Specimen Number	RC Wall	HSP-RC Wall-1.5	HSP-RC Wall-2	HSP-RC Wall-2.5	HSP-RC Wall-3
Pixel resolution	1.6644 mm/pixel	1.1594 mm/pixel	1.2739 mm/pixel	1.4733 mm/pixel	1.298 mm/pixel
Pixels	4.5 pixel	2.5 pixel	2.5 pixel	2 pixel	2 pixel
Crack width	7.49 mm	2.89 mm	3.18 mm	2.95 mm	2.60 mm

through image processing technology and identification calculation. It can be seen that the addition of a honeycomb aluminum sandwich panel effectively prevented the growth of RC wall cracks’ width, and the maximum width of the wall cracks was reduced by more than half during the whole impact process.

3.4. Damage to the Honeycomb Sandwich Panel and the Interface

3.4.1. Deformation of the Honeycomb Sandwich Panel

Under the low-speed impact of high energy and mass, the deformation of the honeycomb sandwich panel on the surface of the RC wall mainly occurred on the upper surface. In addition, the deformation of several honeycomb sandwich panels was the same, except that the pit’s depth and the concave area’s external dimensions were different. The bottom surface of the honeycomb sandwich panel was in contact with the constrained RC wall, which can be regarded as being without deformation. Taking the HSP-RC Wall-2.5 as an example, the impact plastic deformation of the upper surface of the honeycomb sandwich panel with an edge length of 2.5 mm is shown in Figure 20. The yellow box in the picture represents the contact area between the iron plate and the honeycomb sandwich panel, which is compressed as a whole and presents a state of overall plane sinking. There is no significant deformation outside the red box. The area between the yellow and red boxes is the transition zone between the iron plate pressing area and the surrounding undeformed area.

Referring to Dahai’s [27] description of the impact process of the honeycomb sandwich panel in the drop weight test, it can be inferred that the impact deformation process acting on the sandwich panel under the flat indenter impact is as shown in Figure 21. After the iron plate (equivalent to the flat indenter) makes contact with the honeycomb sandwich panel, the direct contact area of the sandwich panel subsides, the rest of the sandwich panel surface does not deform, and the deformation transits in the area around the contact area. After the impact contact, the hammer and iron plate rebound, the upper panel and core of the sandwich layer experience a slight recovery, and the bottom surface remains unchanged, with the RC wall having no local deformation.

Under high energy or a high strain rate, the sandwich panel with air and water as the back support may break locally and appear to crack between the panel and the core layer. The sandwich panel with the RC wall as the back support had good integrity under local impact and only experienced local press deformation. At the end, the sandwich panel had a local depression, and there was subsidence within 10 cm around the flat pit, without cracking between the panel and the core, as shown in Figure 21.

The iron plate acts as a square flat indenter under the impact of the pendulum. Under the impact of the high-energy flat indenter, the area around the indenter on the impact surface of the honeycomb sandwich panel is subject to tensile yield, the upper surface layer and core layer on the impact contact surface are sunken, and the sunken honeycomb core layer is subject to progressive folding and buckling, forming a square flat pit with the same size as the iron plate. The pit is surrounded by a concave transitional area (shown in Figure 22). In the deformed area, the upper surface and core layer consume almost all impact energy.

The average subsidence depth of the flat pit on the upper layer of the sandwich panel after impact and the total length and width of the depression area were measured, as shown in

Table 5. Geometric parameters of the depression area (mm).

Edge Length of Honeycomb Hole	1	1.5	2	2.5	3
Pit subsidence depth	3	4	5	8	9
Size of depression area WW × LL	463 × 410	468 × 416	472 × 423	465 × 437	521 × 520

.

It can be seen that the larger the side length of the honeycomb hole, the greater the pit depth and the larger the transition area in the depression area, the larger the transition area involved in energy consumption. Due to the in-plane orthogonal anisotropy of the honeycomb, when l is small (l ≤ 2.5 mm), the in-plane material properties of the honeycomb differ greatly in L and W directions. Therefore, the width of the transition area along the two main in-plane directions (L and W) is different, and the length and width of the entire depression area are different. However, when l of the honeycomb hole becomes larger, and the honeycomb porosity is larger, the difference in the material properties in the L and W directions gradually reduces, so the length and width of the depression area tend to be the same.

During the impact process of the original RC wall, the dissipation of impact energy depends on the RC wall itself, while the consumption of impact energy of the HSP-RC wall mostly depends on the plastic deformation of the honeycomb sandwich panel. Therefore, under the same high-energy and low-speed impact, the mid-span deformation, displacement, crack width and total crack length of the original RC wall are higher than those of the HSP-RC wall.

3.4.2. Energy Consumption and Recovery Coefficient of the First Collision

The pendulum rebound velocity can reflect the impact kinetic energy consumed by the HSP-RC Wall when the pendulum first impacts the specimen, and the total impact times of the pendulum can reflect the single impact energy consumption of the HSP-RC Wall. The pendulum rebound speed, kinetic energy consumption after the first impact and total impact times on specimens in the experiment are shown in

Table 6. First impact energy consumption of the pendulum.

Specimen Number	HSP-RC Wall-1	HSP-RC Wall-1.5	HSP-RC Wall-2	HSP-RC Wall-2.5	HSP-RC Wall-3	RC Wall
Impact velocity V1	2.30 m/s	2.30 m/s	2.30 m/s	2.30 m/s	2.30 m/s	2.30 m/s
Rebound velocity V2	1.82 m/s	0.849 m/s	0	0.539 m/s	0.562 m/s	0.846 m/s
Energy consumption ΔE	1977 J	4569 J	5290 J	4999 J	4974 J	4574 J
Total number of impacts	5	4	1	3	3	4

.

The honeycomb sandwich panel undergoes irreversible plastic deformation after being impacted, and the collision between the pendulum and the HSP-RC wall is an elastic–plastic collision. The normal recovery coefficients $e=V_2/V_1$ of different HSP-RC walls reflect the differences in collision materials. There are significant differences in the recovery coefficients of different HSP-RC walls during a collision, and it can be inferred from experiments that selecting honeycomb sandwich panels with different parameters can ensure that the value of e is between 0 (when l is 2 mm) and 0.79 (when l is 1.5 mm). A high recovery coefficient may make the rebound speed too fast, leading to secondary damage. A lower recovery coefficient can ensure that the composite structure has a better single dissipation effect on impact energy.

In the HSP-RC wall low-speed impact experiment, when the thickness h of the honeycomb sandwich panel’s surface layer remains 3 mm, and the hole edge length l of the honeycomb core layer is 2 mm, the sandwich panel will have the best energy consumption under a single collision. In this case, the composite structure of an “aluminum alloy–honeycomb aluminum–aluminum alloy–RC wall” is similar to a laminated plate and absorbs all the energy at once, which can ensure that the pendulum only experiences one impact. Most of the impact energy is consumed by honeycomb sandwich panels, which can effectively prevent rebound and secondary impact damage. In this situation, the force–time curve has the shortest duration. Therefore, in some people’s engineering projects, important structures must avoid secondary damage from impact. The reasonable selection and use of honeycomb sandwich panels as protective layers can prevent the rebound phenomenon of impact loads.

3.4.3. Destruction of the Interface between Sandwich Panels and RC Walls

During the pendulum impact process, no severe cracking occurred between the layers of the specimens. The cracking of the adhesive layer between the RC wall and the honeycomb sandwich panel reflects whether the connecting adhesive layer is an energy-consuming area. After the impact, if the honeycomb sandwich panel can be completely removed from the RC wall without damaging the integrity of the RC wall and honeycomb sandwich panel, it is judged that the interface adhesive layer is cracked.

Table 7. Damage record of interface adhesive layer.

Specimen Number	HSP-RC Wall-1	HSP-RC Wall-1.5	HSP-RC Wall-2	HSP-RC Wall-2.5	HSP-RC Wall-3
Adhesive cracking	Yes	Yes	No	No	No

records the cracking of the adhesive layer. After removing the honeycomb sandwich panel, the disconnection of the interface bonding layer is mostly due to cracks between the bonding layer and the honeycomb sandwich panel, and a small number of locations are due to cracks in the bonding layer and wall concrete.

The honeycomb sandwich panel composite RC wall appears as a layered material; it tends to dissipate energy at the weak layer when it is impacted. The small edge length of the honeycomb hole makes the honeycomb sandwich panel strong and then causes the interface adhesive layers to also serve as energy dissipation areas. The cracking of the bonding surface is also related to the strong shear resistance of the honeycomb sandwich panel without significant bending deformation. The RC wall has bending deformation, and the mismatch between the deformation of the sandwich panel and the RC wall is part of the reason for interlayer cracking.

When l ≥ 2, the entire HSP-RC wall mainly consumes energy depending on the honeycomb core and the upper layer under the impact load. The bottom layer, connected to the RC wall and the interface adhesive layer of the RC wall, hardly consumes any energy.

Compared with ordinary RC walls, due to the airtightness of the alloy sheet and the colloid connection between the honeycomb sandwich panel and the RC wall, the HSP-RC wall, which does not display interlayer cracking, can ensure that the whole composite structure does not cause problems such as gas permeation.

4. Impact Theory Model

According to reference [28], the pressing process of a sandwich panel under a large mass impact can be considered quasi-static pressing. The ideal impact depression area is illustrated in Figure 23a. Considering the area in the first quadrant (within the circle) as an example, R represents half the width of the iron plate (square indenter head), and δ is the depth of penetration of the square indenter head. The structural models of regions A and C are established along the x-axis, as shown in Figure 23b. Additionally, the structural models of corner regions B and C are established along the diagonal of region C, as shown in Figure 23c.

The σ_p in Figure 23 is the platform stress of the honeycomb core under low-speed and out-of-plane impact, which can be obtained according to Equation (1) [27].

$${\sigma }_p=5.6{\left({\displaystyle \frac{t_c}{l_c}}\right)}^{5/3}{\sigma }_c$$

(1)

In Equation (1), t_c and l_c are the thickness and edge length of the honeycomb hole wall, respectively, and σ_c is the yield stress of the honeycomb matrix material.

Based on the half model of the impact area, we established the expression of the displacement field, as shown in Equation (2).

$$W(x,y)=\left\{\begin{array}{ll}\delta & 0<x<R, 0<y<R\\ \delta {\left(1-{\displaystyle \frac{x-R}{\xi -R}}\right)}^2& R<x<\xi , 0<y<R\\ \delta {\left(1-{\displaystyle \frac{y-R}{\xi -R}}\right)}^2& 0<x<R, R<y<\xi \\ \delta {\left(1-{\displaystyle \frac{\sqrt{x^2+y^2}-\sqrt{2}R}{\xi -R}}\right)}^2& R<x<\xi , R<y<\xi \end{array}\right.$$

(2)

Based on previous research [29], finally, the expression for the impact force is obtained as Equation (3):

$$F={\displaystyle \frac{4}{3}}{N}_0\left[{\left(2N_0\delta /{\sigma }_{\mathrm{p}}\right)}^{1/2}-R\right]+{\displaystyle \frac{2\pi }{3}}{N}_0\delta +{\displaystyle \frac{1}{6}}{\sigma }_p\left[\pi \left(2N_0\delta /{\sigma }_{\mathrm{p}}\right)+4R{\left(2N_0\delta /{\sigma }_{\mathrm{p}}\right)}^{1/2}+8R^2\right]$$

(3)

The impact force is only related to R, δ and the deformation area of the upper layer of the honeycomb sandwich panel (i.e., the flat pit and the transition area), and the transition area is related to δ and R, so the expression for the impact force can be transformed into Equation (4).

$$F=f\left(\delta \right)=K_C{\delta }^n$$

(4)

The pendulum may rebound after the first impact on the honeycomb sandwich panel, using an energy balance model:

$${\displaystyle \frac{1}{2}}M{V}_I^2-{\displaystyle \frac{1}{2}}M{V}_2^2=E_b+E_s+E_c$$

(5)

In Equation (5), E_b represents buckling deformation energy dissipation, E_s represents shear energy dissipation, E_c represents contact energy dissipation, and V₁ and V₂ represent impact velocity and rebound velocity, respectively.

Ignoring the global deformation and the deformation of the honeycomb sandwich panel’s bottom layer supported by the RC wall, it can be inferred from Equations (4) and (5) that

$${\displaystyle \frac{1}{2}}M\left(V_I^2-V_2^2\right)={\displaystyle {\int }_0^{\delta }f(\delta )d\delta }$$

(6)

It can be seen that under the protection of honeycomb sandwich panels, the rebound speed of the impact body under a single collision is determined by the depth of the honeycomb sandwich panel indentation caused by the impact.

5. Conclusions

This article proposes a method for using a honeycomb sandwich panel for the impact protection of RC walls. Based on the original RC wall impact test, honeycomb sandwich panel composite RC wall impact tests with different honeycomb parameters were conducted using a large-energy pendulum experiment system, and the results of this series of tests were analyzed to study the protective effect of different honeycomb sandwich panels on RC walls under low-speed impact. The main conclusions of this article are as follows.

(1) The impact resistance of the HSP-RC wall formed by adding honeycomb sandwich panel protection to the RC wall was greatly improved. Honeycomb sandwich panels significantly reduce the crack width and number of RC walls, as well as the displacement and residual deflection of the RC walls. Honeycomb sandwich panels can change the impact load transmission path to the RC wall, thereby affecting the expansion and distribution of cracks on the RC wall, changing the failure mode of the RC wall and effectively reducing the local and global deformation of the RC wall. The ultimate failure mode of the HSP-RC wall tends to be a global bending failure.

(2) Compared with the impact force–time curve of the original RC walls under low-speed impact, the impact force–time curve of the HSP-RC walls shows a fluctuation descent stage and a zero value stage. The longer the edge length of the honeycomb hole, the longer the fluctuation descent stage will be, and the larger the range of the fluctuation force value will be. The duration of the zero value stage increases from scratch with the increase in the edge length of the honeycomb hole and then shows a positive correlation with it. The impact force platform value of the HSP-RC wall will be much greater than that of the original RC wall.

(3) The recovery coefficient of the HSP-RC walls is related to the structural parameters of the honeycomb core layer. A reasonable selection of structural parameters of honeycomb sandwich panels can achieve a one-time consumption of all impact energy to avoid secondary impact. Excessive rigidity of the honeycomb sandwich panels will lead to cracking of the adhesive layer between the sandwich panel and the RC wall. Therefore, it is necessary to avoid selecting honeycomb sandwich panels with a relatively high density and stiffness.

(4) The protective effect of honeycomb sandwich panels on RC walls is related to their structural parameters. Within the range of honeycomb parameters selected in the experiment, when the edge length of the honeycomb hole is longer, the honeycomb sandwich panel consumes energy better, which means that the honeycomb sandwich panel can have a better protective effect for the RC wall under impact and reduce the destruction of the RC wall.

Due to the fact that this article mainly studies the impact protection performance of honeycomb sandwich panels through large-scale pendulum impact tests, the research cost was high, and the number of samples used in the test was limited (seven different RC wall specimens). Therefore, the relationship between the impact protection performance of honeycomb sandwich panels and honeycomb parameters obtained has certain limitations. In order to obtain a more comprehensive understanding of the influence of various honeycomb parameters on the impact protection performance of sandwich panels, numerical research will be conducted based on experiments in the future to obtain the influence of various parameters on the protection performance of sandwich panels and guide subsequent engineering applications.