The goal of this work is to study the Cassie–Baxter state on the microstructure hydrophobic surfaces. The dependence of the energy barrier on the drop size, contact angle, the gap between the pillars, and pillar width is investigated. We consider the drop in three dimensions using a numerical approach to minimize the free energy of the drop in any situations. Numerical results are obtained for the wetting of a hydrophobic surface textured with a square lattice of pillars. According to the results, we found that the curvature increases allowing the liquid to touch the surface below the posts as the drop loses volume or decreased the contact angle with fixed drop volume. In the situations we considered, the pillar diagonal length is very critical and sensitive to the details of the surface patterning when the drop volume is larger.

All-silicon terahertz absorbers have attracted considerable interest. We present a design and numerical study of an all-silicon polarization-insensitive terahertz metamaterial absorber. The meta-atoms of the metamaterial absorber are square silicon rings which can be viewed as gratings. By properly optimizing the structure of the meta-atom, we achieve a broadband absorptivity that is above 90% ranging from 0.77 THz to 2.53 THz, with a relative bandwidth of 106.7%. Impedance matching reduces the reflection of the terahertz waves and the (0, ±1)-order diffraction induce the strong absorption. The absorption of this absorber is insensitive to the polarization of the terahertz wave and has a large incident angle tolerance of up to 60 degrees. The all-silicon metamaterial absorber proposed here provides an effective way to obtain broadband absorption in the terahertz regime. Metamaterial absorbers have outstanding applications in terahertz communication and imaging.

Accurately classifying and identifying non-cooperative targets is paramount for modern space missions. This paper proposes an efficient method for classifying and recognizing non-cooperative targets using deep learning, based on the principles of the micro-Doppler effect and laser coherence detection. The theoretical simulations and experimental verification demonstrate that the accuracy of target classification for different targets can reach 100% after just one round of training. Furthermore, after 10 rounds of training, the accuracy of target recognition for different attitude angles can stabilize at 100%.