Diffraction tomography is a promising, quantitative, and nondestructive three-dimensional (3D) imaging method that enables us to obtain the complex refractive index distribution of a sample. The acquisition of the scattered fields under the different illumination angles is a key issue, where the complex scattered fields need to be retrieved. Presently, in order to develop terahertz (THz) diffraction tomography, the advanced acquisition of the scattered fields is desired. In this paper, a THz in-line digital holographic diffraction tomography (THz-IDHDT) is proposed with an extremely compact optical configuration and implemented for the first time, to the best of our knowledge. A learning-based phase retrieval algorithm by combining the physical model and the convolution neural networks, named the physics-enhanced deep neural network (PhysenNet), is applied to reconstruct the THz in-line digital hologram, and obtain the complex amplitude distribution of the sample with high fidelity. The advantages of the PhysenNet are that there is no need for pretraining by using a large set of labeled data, and it can also work for thick samples. Experimentally with a continuous-wave THz laser, the PhysenNet is first demonstrated by using the thin samples and exhibits superiority in terms of imaging quality. More importantly, with regard to the thick samples, PhysenNet still works well, and can offer 2D complex scattered fields for diffraction tomography. Furthermore, the 3D refractive index maps of two types of foam sphere samples are successfully reconstructed by the proposed method. For a single foam sphere, the relative error of the average refractive index value is only 0.17%, compared to the commercial THz time-domain spectroscopy system. This demonstrates the feasibility and high accuracy of the THz-IDHDT, and the idea can be applied to other wavebands as well.

Relativistic few-cycle mid-infrared (mid-IR) pulses are unique tools for strong-field physics and ultrafast science, but are difficult to generate with traditional nonlinear optical methods. Here, we propose a scheme to generate such pulses with high efficiency via plasma-based frequency modulation with a negatively chirped laser pulse (NCLP). The NCLP is rapidly compressed longitudinally due to dispersion and plasma etching, and its central frequency is downshifted via photon deceleration due to the enhanced laser intensity and plasma density modulations. Simulation results show that few-cycle mid-IR pulses with the maximum center wavelength of $7.9\;\unicode{x3bc} \mathrm{m}$ and pulse intensity of ${a}_{\mathrm{MIR}}=2.9$ can be generated under a proper chirp parameter. Further, the maximum energy conversion efficiency can approach 5.0%. Such a relativistic mid-IR source is promising for a wide range of applications.

The converging lens is one of the key components in high-resolution terahertz imaging. In this Letter, a binary diffractive lens is proposed for the scanning imaging system working at 278.6 GHz, in which a convergent beam with a waist diameter of 0.65 mm is generated, and 1 mm lateral imaging resolution is realized. This low-cost terahertz lens, constituted by concentric rings with different radii, is optimized by stimulated annealing algorithm and fabricated by three-dimensional printing. Compared with the conventional transmissive convex lens, higher resolution and enhanced imaging quality are achieved via smaller focal spot of the illumination beam. This type of lens would promote terahertz imaging closer to practical applications such as nondestructive testing and other scenarios.

提出了一种由三维打印技术制作的用于0.3 THz波束调控的二元相位型衍射透镜，可产生轴向焦深为25 mm、腰斑直径为1 mm的太赫兹聚焦波束。将该聚焦波束应用于连续太赫兹波计算机辅助层析成像，不仅将成像分辨率提高到1.5 mm，而且由于扩大了系统景深，提高了成像保真度。这类利用三维打印技术制作的二元衍射元件可有效应用于低频太赫兹波段的光束调控，为太赫兹成像系统的性能优化以及聚焦元件的设计制备提供一种有效途径。

The resolution of the spatial light modulator (SLM) screen and the encoding algorithm of the computer-generated hologram are the primary limiting factors in the generation of large topological charge vortex beams. This paper attempts to solve these problems by improving both the hardware and the algorithm. Theoretically, to overcome the limitations of beam waist radius, the amplitude profile function of large topological charge Laguerre–Gaussian (LG) beam is properly improved. Then, an experimental system employing a 4K phase-only SLM is set up, and the LG beams with topological charge up to 1200 are successfully generated. Furthermore, we discuss the effect of different beam waist radii on the generation of LG beams. Additionally, the function of the LG beam is further improved to generate an LG beam with a topological charge as high as 1400. Our results set a new benchmark for generating large topological charge vortex beams, which can be widely used in precise measurement, sensing, and communication.