The epoxy resin was Type 618 made by Shanghai Resin Factory of China, tensile modulus: 1.97 GPa, tensile strength: 68.10 MPa. The SiO₂ partiles were spherical particles supplied by Zhejiang Hongsheng Rial Technilogy Co., Ltd. The particle size of the SiO₂ nanoparticles was 20 - 50 nm, the specific surface area was 160 ± 20 m²/g, the bulk density was smaller than 0.11 g/cm³ and the purity being greater than 98.5 wt%.

TheSiO₂ nanoparticles were put in acetone solution. The concentration of SiO₂ nanoparticles was set at 0 wt%, 5 wt%, 10 wt% or 15 wt%, and the mixture was ultra sonicated by an ultrasonic cell disruptor BILON92-IID (Shanghai Bilang)for about 6 hours in order to get a homogenous solution. Then the mix of Bisphenol A epoxy (Type 618 made by Shanghai Resin Factory of China) was added to the solution and stirred at 40˚C for 12 hours.

The composites were fabricated with the SiO₂ nanoparticles filled-epoxy resin through a special box. The mix of Bisphenol A epoxy with SiO₂ nanoparticles and agent Tri-methyl-hexamethylene-diamine (Type 593 made by Shanghai Resin Factory of China) in the volume proportion of 3:1.

The composite coupons for testing were prepared according to the length × width is 9 mm × 9 mm, and the thickness is 5 mm.

The compression tests were conducted under both quasi-static and high strain rates. Strain rate is the rate of change in strain (deformation) of composites with respect to time. For the strain rate at high rates of straining,

where is strain rate, is the strain, is the original length, is the length at each time, is the speed at which the ends are moving away from each other.

For the strain rate in quasi static, the approximate strain can be calculated from follow equation,

where is strain rate, is the speed of compression, is the length of sample. The compression test under the quasi-static condition was performed on an MTS810.23 system at a speed of 1 mm/min.

The high strain rate compression tests were carried out on a split Hopkinson pressure bar (SHPB) apparatus. The schematic illustration and principle of the SHPB employed in this study are shown in Figures 1-3 respectively. A small specimen is between two elastic bars of the same cross sectional area and modulus, which are named as the incident bar and transmission bar separately. The material of the striker and pressure bars is steel subjected to maraging with extremely high yield strength in order to withstand a very high impact velocity. An elastic stress pulse is imparted to the incident bar by impacting it with a striker bar of the same cross sectional area and modulus. The impact of the striker bar generates as elastic stress wave with twice the length of the striker bar and propagates through the incident bar with the velocity of sound in the bar media and passes through the specimen while deforming it. The particle velocity imparted on the incident bar is half the impact velocity of the striker bar. The stress in the bar is given by

where is the density of the bar material, is the bar sonic velocity and is the particle velocity.

When the elastic wave reaches the specimen-incident bar interface, part of it will be reflected back, and part of it will be transmitted through the specimen and pass through the transmission bar. To acquire the direct incident pulse, the reflected pulse and the transmitted pulse, strain gauges are mounted on both the incident and transmission bars. One-dimensional wave propagation is assumed to analyze the strain signals from the strain gauges (Type BF(BH)350-8KA made by Shanghai Shendi Informational Technology Co., Ltd.).

If, and represent the modulus, cross section area and density of the bar and, and

those of specimen then the equations for the strain rates, strain and stress of the specimen are as follows:

where, represents the longitudinal wave velocity in the bar;

represents the specimen length;

and are the strain gauge signals of the reflected and the transmitted pulses, respectively.

The average stress and strain in the specimen can be known as a function of time and the mechanical properties of the composite, such as strain rates and stress-strain behavior, along the loading direction can be determined from Equations (4)-(6).

The SHPB employed in this study was equipped with a momentum trapping device, a gas gun of 14.5 mm inner diameter. The diameter of the incident and transmission bars was the same (diameter 14.5 mm and length 600 mm). Compressed nitrogen gas was employed to operate the gas gun and propel the striker bar to move a distance of 200 mm. The impact velocity of the gas gun (strike bar) could be changed by changing gas pressure to get required strain rates. The schematic and principle of split Hopkinson pressure bar are shown in Figure 2 and Figure 3 respectively.

Compression tests of epoxy composites with various SiO₂ nanoparticles contents were conducted at different strain rates. As above-mentioned, at least five samples were tested at each strain rate (including quasi-static state). Figure 4 shows the typical signals obtained from the input and output bar for 5 wt% SiO₂ nanoparticles contents at strain rate of 1600/s conditions. According to the one dimensional stress wave propagation theory of the SHPB apparatus, the stress strain curves under high strain rates can be calculated. The results are shown in Figures 5-8. It can be observed that the stress-strain curves are sensitive to strain rate for different SiO₂ nanoparticles contents.

Figure 5 shows the stress-strain curves of epoxy without SiO₂ nanoparticles at various strain rates. Figure 5 indicates that the compressive failure stress increases, the compressive stiffness decreases and the failure strain increases with increasing strain rate at high strain rate compression.

Figure 6 shows the stress-strain curves of epoxy composites with 5 wt% SiO₂ nanoparticles at various strain rates. Similar to Figure 5 which is for the composites without SiO₂ nanoparticles, the compressive failure stress and compressive failure strain are increasing with increasing strain rate. And the stiffness of composites is increased with the SiO₂ nanoparticles content increasing, because the SiO₂ nanoparticles can increase the interface strain in the resin.

Figure 7 and Figure 8 are the stress-strain curves of epoxy composites with 10 wt% SiO₂ nanoparticles and 15 wt% SiO₂ nanoparticles at various strain rates, respectively. Similar with Figure 6, the compressive failure stress increases as the strain rate grows, the compressive failure strain also increases when strain rate increases at high strain rates compression. Meanwhile, the compressive stiffness of composites with SiO₂ nanoparticles is larger than that of the composites without SiO₂ nanoparticles. However, it can be found that the compressive failure stress are decline after the SiO₂ nanoparticles content is more than 10 wt%.

Figure 9 shows the compressive failure stress of epoxy composites with various SiO₂ nanoparticles contents under different strain rates. It can be seen that the compressive failure stress increases with the strain rate, and it is increases along SiO₂ nanoparticles contents at first and then decrease after the SiO₂ content is 15 wt%. For high strain rate loading, the failure stress of epoxy composites with various SiO₂ nanoparticles contents is much larger than that in quasi-static loading.

Figure 10 shows the compressive failure strain of epoxy composites with various SiO₂ nanoparticles contents under different strain rates. Similar to Figure 12, it can be seen that the compressive failure strain increases with the strain rate and SiO₂ nanoparticles contents.

The photographs in Figures 11(a)-(d) display the fracture features of epoxy without SiO₂ nanoparticles at various strain rates under compressive loading. It is found that breakage of epoxy specimen occur at various strain rates. As the strain rate increasing, the specimen is compressed into several pieces of fragment which is shown in Figure 12(d). The obvious beginning crack happens at the strain rate of 800/s, and drastic at the strain rate of 1400/s. The failure state of the sample without SiO₂ nanoparticles at various strain rates under in-plane loading is shown in the figures. It is can be observed that the specimen is cracked because of the radicalized stress transmission though the little epoxy block.

Figures 12(a)-(d) shows compression fractures of epoxy/SiO₂ nanoparticles (5 wt%) composites at various strain rates. The crazes appear at the strain rate of 800/s, and the apparent crack arise at the strain rate of 1400/s. However, the specimen keeps as an entirety, instead of being the fragments, which is quite different from the state of the sample without SiO₂ nanoparticles at the same strain rate. And then, at the strain rate of 2000/s, the composites fractured. The crack happens under high strain rate compression.

The photographs in Figure 13 illustrate compression fractures of epoxy/SiO₂ nanoparticles (10 wt%) composites at various strain rates. It is found that the failure modes under quasi-static state and high strain rates are rather different. The performance of crack of is evidence under the strain rate of 0.001/s. On the contrary, the morphological feature under the high strain rates is rather stable, which the shape of the specimen hardly change under 800/s and 1400/s expect a single craze, furthermore, the integrity of the sample is well kept ever under 2000/s.

Compression fractures of epoxy/SiO₂ nanoparticles (15 wt%) composites at various strain rates is shown in Figures 14(a)-(d). Similar with the situation of epoxy/SiO₂ nanoparticles (5 wt%) composites, the serious crack happens under low strain rate, which shows several crevices and is loose texture in appearance. On the opposite, under the high strain rates, the regular crazes occur, and there are hardly tiny cracks that can be noticed. The samples after test still keep hard quality. There is no obvious deformation, because the suitable content SiO₂ nanoparticles can increase the interface strength of composites, which indicates that the composites can absorb more energy with smaller deformation.