**1. Introduction**

Critical infrastructure in sectors, such as energy, communications and government, are highly vulnerable to the threat of improvised explosives, as demonstrated by the ongoing terrorist attacks around the world. To ensure the security and stability of societies, governments are faced with a serious challenge and there is an urgent need for a costeffective testing tool to support the study of blast effects in the near zone of structures, so that new blast-resistant structures can be developed, validated and deployed more quickly.

Blast testing has long been the primary means of studying close-in blast loading [1–3]. The explosive is capable of exerting a pressure of several tens to hundreds of megapascals on the specimen within a short period of time after detonation, and such test conditions are difficult to replicate by other means. Schenker [4] conducted full-scale blast tests on concrete slabs to obtain dynamic response test data for concrete elements and verified these via numerical calculations. Dharmasena et al. [5] used stainless steel honeycomb sandwich

**Citation:** Xiong, Z.; Wang, W.; Yu, G.; Ma, J.; Zhang, W.; Wu, L. Experimental and Numerical Study of Non-Explosive Simulated Blast Loading on Reinforced Concrete Slabs. *Materials* **2023**, *16*, 4410. https://doi.org/10.3390/ ma16124410

Academic Editor: Miguel Ángel Sanjuán

Received: 31 May 2023 Revised: 13 June 2023 Accepted: 14 June 2023 Published: 15 June 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

panels and investigated the extent of damage to specimens at 100 mm blast distance and trinitrotoluene (TNT) doses of 1.0, 2.0 and 3.0 kg respectively. However, this research tool is characterized by the difficulty of measuring dynamic mechanical parameters, large scatter, low repeatability and poor visibility of component damage. Therefore, non-explosive methods that are safe, controllable and can be studied in the laboratory, such as the gas gun, the shock tube and blast simulator, has emerged. Since the blast shock wave causes damage to the human brain at pressures lasting a few milliseconds in the approximate range of 68.95 to 690.48 Pa [6], the gas gun is well suited for shock wave testing in this low-pressure range. Bartyczak [7] improved the existing gas gun method and investigated the impact resistance of helmet materials. Schleyer [8] summarized a series of tests with shock tubes on large structures, which proved that the tubes were suitable for simulating the loads generated by far ranges explosions. In order to investigate the effect of concrete strength on the dynamic response of concrete slabs under blast loading, Thiagarajan [9] conducted impact tests on four types of concrete slabs using a shock tube to simulate air blast loading.

To solve the problem of not being able to load full-scale components with gas guns, and to compensate the limitations of shock tubes for near-zero blast studies, the first blast simulation device [10] has been developed (2006) at the University of California, San Diego (UCSD) for the investigation of the blast resistance of different walls, columns and composite structures. Oesterle [11] investigated the impact resistance of concrete masonry walls containing different reinforcements at impulse of 1000–20,000 Pa· s. Gram [12] carried out impact and blast tests on reinforced concrete columns and compared the results of the two tests, which showed good similarity in deformation and failure modes of the specimens. Rodrigueznikl [10] first captured the formation of shear cracks and the spalling of the concrete protective layer when studying reinforced concrete columns under 6800–15,700 Pa· s impulse. This impulse corresponds to the blast load of 560 kg of high explosive on a vehicle at 0.9 m from the ground and 3.5–6.1 m blast distance. Freidenberg [13] carried out impact tests on high-strength prototype walls, all of which simulated blast distances of several meters and charge sizes of several hundred kilograms equivalents, and produced planar shock waves similar to those produced by car bombs. Huson [14] demonstrated the use of water bags to apply loads to composite sandwich members and member joints. Wolfson [15] briefly discuss the dynamic response and damage patterns of honeycomb structures under the action of a proximity blast by firing different types of impact modules. The European Commission Joint Research Centre (JRC) has proposed the development of explosion simulation capabilities similar to those of the United States. The European Laboratory for Structural Assessment (ELSA) has constructed a new testing facility called the electronic blast simulator (e-BLAST) [16–18], which uses servo-controlled drive technology to simulate the loading of structures by air shock waves without the use of explosives. The facility uses three synchronized electric linear motors to accelerate the impact module and load the structure under study. Elastic materials are placed between the impact mass and the structure to maintain consistency with relevant air shock wave pressures. This method will not only describe the effects of explosions on structural systems but also evaluate reinforcement and retrofitting techniques for buildings and bridges to resist terrorist bomb attacks. It will also help investigate the problem of progressive collapse, where local damage spreads disproportionately and leads to overall destruction, as seen in the Oklahoma City bombing.

Currently, there is a lack of research on RC slabs using blast simulators, and the range of impact intensities for simulated blast loading is not sufficient to cover a wide range of blast environments. To address this issue, this paper conducted nine impact tests on reinforced concrete slabs using the newly developed VMLH Blast Simulator. The study also included numerical calculations to investigate the effects of rubber shape, impact velocity, and the bottom and top thicknesses of the pyramidal rubber on the impact loading. Additionally, two equivalence criteria are proposed and used to compare the test results with ideal blast loading. The VMLH Blast Simulator has the ability to safely and precisely

apply blast-like loading to components within a controlled laboratory setting, addressing the limitations of traditional explosion testing in terms of data collection, testing duration, and expenses.

#### **2. Experiment Program**

#### *2.1. Non-Explosive Blast Simulator*

A new impact-based facility has been developed that vertically fires multi-mass impact modules to load large components at precise high velocities (VMLH). The VMLH Blast Simulator is powered by a high-speed gas-hydraulic actuator, as shown in Figure 1. Compared to traditional chemical explosion tests, this equipment has the advantage of easy data acquisition, high reproducibility and reliability, especially without the fireball generated by the explosion, and the ability to use high-speed cameras to capture the deformation–destruction evolution of the component.

**Figure 1.** Diagram of the VMLH Blast Simulator.

Before the test, the nitrogen cylinder is inflated and stored. During the test, the high pressure nitrogen is released instantaneously, and the piston rod and impact module connected by the brittle bolt are accelerated together. Then, during the acceleration process, a number of high-speed switching valves in the hydraulic system are opened simultaneously, and the oil in the impact cylinder flows back rapidly; when the impact module reaches the set speed value, the oil pressure rises, producing a buffering effect on the piston rod to slow down. At this point, the brittle bolt breaks, and the impact module, which is separated from the piston rod, starts a short free fall movement until it hits the specimen vertically. The test process is shown in Table 1.

The VMLH Blast Simulator is designed to simulate the effects of an explosion on a component or structure. Similar to other blast simulators [12,17], it achieves this by creating non-elastic collisions between the impact module and the specimen. This is carried out by adding a shock-absorbing layer between them, which transmits the load to a reinforced concrete slab in the form of waves. In this article, silicone rubber is arranged between the impact module and the specimen to adjust the shape of the force during the collision. When the impact module falls on the rubber, the rubber undergoes elastic deformation, producing lateral waves or shear waves. These waves propagate along the rubber until they reach the edge of the reinforced concrete slab and are reflected back. The rubber also produces longitudinal waves or compression waves, which propagate through the reinforced concrete slab, reach the lower surface of the slab, and are reflected back. These waves cause vibration and deformation of the slab. When the explosion load is applied to the component, a pressure wave called the explosion wave will be generated. This pressure wave will propagate inside the component and the surrounding medium, causing damage to the component and the surrounding environment. Under the action of explosion load, the stress and deformation inside the component will undergo drastic changes, which may lead to the fracture, deformation or collapse of the component.

**Table 1.** Test procedure of the VMLH Blast Simulator.

## *2.2. Test Specimen*

The dimension of these slabs are 1200 mm length, 900 mm width and 180 mm depth. They are reinforced in both directions with a 10 mm steel bar reinforcement mesh with 100 mm spacing and 25 mm of concrete cover. All the RC slabs were cast from the same batch of commercial concrete with a strength class of C40. Five 150 mm cubes were cast and tested. The average strength of the concrete cubes after 28 days of curing is 45.6 MPa. The type of reinforcement is HRB400E, which has a yield strength of 435 Mpa and a Young's modulus of 209 Gpa. The dimensions of the slab, in addition to the reinforcement detailing and supporting conditions, are shown in Figure 2.

**Figure 2.** Preparation of RC slabs. (**a**) Dimensions and reinforcement; and (**b**) supporting structure.

#### *2.3. Impact Cushion*

Silicone rubber (referred to as rubber for short) is well-established, inexpensive to prepare, and can be reused in trials. The silicone rubber is placed on the surface of the specimen. On one hand, it prevents direct contact between the impact module and the specimen, which may cause damage to the metal material. On the other hand, the existence of the rubber achieves flexible contact with the specimen, avoiding small angle deviations that may occur during the falling process of the impact module, which may cause uneven loading of the specimen. More importantly, the viscoelastic material properties of the rubber determine the waveform of the load transmitted by the impact. Therefore, rubber was chosen as the impact cushion, and two shapes were designed, as shown in Figure 3. The planar rubber size is 500 mm × 500 mm with a thickness range of 20~100 mm. The pyramidal rubber size is 100 mm × 100 mm, and the height is 50 mm, where the values of bottom thickness *h*<sup>1</sup> and upper thickness *h*<sup>2</sup> are 20 mm and 30 mm, respectively.

**Figure 3.** Impact cushion and installation. (**a**) Planar rubber; and (**b**) pyramidal rubber.

#### *2.4. Experimental Setup*

A set of displacement sensors was used to measure the impact module speed, as shown in Figure 1. Two accelerometers (SD1407) were attached to the impact module for measuring acceleration, with a range of 5000 g and a sensitivity of 2.2~2.64 pc/g, as shown in Figure 4a. Four load cells (KD3050) were used to measure the impact forces, with a range of 5000 g and a sensitivity of 19.6~19.88 mv/kN, as shown in Figure 4b. These were placed in the holes of the mounting platform, followed by the cover plate and pre-tightened with a screw. To prevent the mounting platform from oscillating significantly during the test, it was secured using two strapping ropes, as shown in Figure 3.

The test data were acquired using the super dynamic signal test and analysis system (DH5960), and the PCO high-speed camera was used to record the impact module from the acceleration to impact with the specimen mass, as shown in Figure 4c. The data acquisition system was set up with six channels, two of which were accelerometers and four were load cells, and the sampling rate was set to 500 kHz. The high-speed camera was connected to a computer with the Camware4 commercial software installed to manage the recording process, as shown in Figure 4d. The high-speed camera used in this test has a frame rate of up to 4500 fps and a resolution of 680 × 1200 pixels.

**Figure 4.** Composition of the measurement system. (**a**) Accelerometer sensors; (**b**) Impact force sensors; (**c**) Data acquisition system; and (**d**) High-speed cameras.
