This paper presents a W-band balun based on the mode converter from the TE10 mode in a rectangular waveguide (RWG) to the TE20 mode in a substrate integrated waveguide (SIW). The proposed balun integrates a WR-10 waveguide, a tapered waveguide section, a finline, and a SIW. The finline acts as the key coupling structure, which efficiently extracts the TE10 mode from the RWG and excites it into the SIW. By taking full advantage of the finline’s inherent broadband and low-loss characteristics, the proposed design effectively combines the high performance of conventional waveguides with the integration merits of planar circuits. The realized balun achieves unbalanced-to-balanced signal conversion over the frequency range of 82–107 GHz, corresponding to a relative bandwidth of 26%. It exhibits excellent broadband performance, simple configuration, and high integration potential, demonstrating great application value in W-band millimeter-wave systems.

Array detection chip is one of the key components of terahertz (THz) systems, with important applications in nondestructive testing, perspective imaging, and high-speed communication. This letter proposes an optical readout THz stacked metamaterial array chip operating at a frequency near the atmospheric window (0.22 THz). It adopts the design of upper and lower chips separation and then bonding to achieve the stacked structure. The upper chip combines with metamaterial cantilever pixels to achieve sensing and execution functions, while the lower chip undertakes auxiliary and support functions. The incident THz radiation can be absorbed and converted to mechanical energy of pixels, then read out in parallel at high speed by the optical system. The stacked mechanism reduces the size and thickness of the sensing/execution structure and improves the radiation-thermal-mechanical sensitivity of the chip. Spectral measurements show that its absorptivity is 97%. Meanwhile, the time measurement results indicate that the chip can quickly respond to THz radiation, with a response time of 1.62 ms.

In this letter, an improved matching technique, namely impedance following matching technique (IFMT), is proposed for the design of a broadband and high-efficiency terahertz frequency doubler. The input matching network's impedance dispersion is intentionally increased to follow the SBDs' impedance variations and generate an inflection point in the midband. This technique realizes generally conjugate matching at both band edges and reduced mismatch in the midband, while the output-matching compensates for the remaining mismatch in the midband. Based on IFMT, a GaN-based terahertz frequency doubler was designed, fabricated, and measured. The measurement results demonstrate that the efficiency exceeds 8.6% over the 155-220-GHz band, with a peak efficiency of 21.3%. These results indicate that IFMT enables broadband operation while incurring less efficiency loss, thus exhibiting certain performance merits over previously reported broadband terahertz doublers.

This paper presents a thermal analysis of the 0.34-THz Extended Interaction Klystron (EIK) high-frequency structure and proposes a time-resolved forward-prediction network (TR-FPN-RNN) as a fast surrogate for time-domain particle-in-cell (PIC) simulations. The heat source distribution and trajectory-related beam transport characteristics, quantified by beam transmission efficiency and interception ratio, can be rapidly predicted by the proposed method. The heat sources of the 0.34 THz EIK high-frequency structure are analyzed with the output power of 92.8 W, and a gain of 36.67 dB. The coincident results verify the practicability of the TR-FPN-RNN method, and the averaged errors are below 2

To address the challenges of strong large-signal nonlinearity, difficulty in frequency extrapolation, and lack of physical constraints in traditional models for gallium nitride (GaN) power amplifier chips, this paper proposes a physics-informed neural network (PINN) behavioral model based on data-driven large-signal waveforms and Fourier reconstruction. The model constructs a multi-parallel subnetwork incorporating RF, DC, and fundamental time-domain channels using K/KI parameters. It not only enforces the non-negativity constraint of drain current through the Softplus activation function but also embeds the large-signal Fourier analytical equations as constraints into the loss function. Experimental results demonstrate that the model achieves high prediction accuracy with training at partial frequency points, its internal analytical parameters fully comply with microwave physical laws, and it has been successfully integrated into commercial simulation software to enable schematic-level large-signal simulation.

This paper proposes a 220-GHz endfire Vivaldi antenna excited mode converter that enables efficient conversion from the rectangular waveguide (RWG) TE10 mode to the substrate integrated waveguide (SIW) TE20 mode. The proposed converter consists of a RWG, an endfire Vivaldi antenna, a grounded slotline, and an SIW section. The Vivaldi radiating arms feature a tapered slot profile that gradually flares along the propagation direction, forming an endfire configuration that effectively couples the TE10-mode guided wave in the RWG into the slotline. Simulation results show that the converter achieves an insertion loss better than 2 dB and a return loss better than 10 dB from 202.8 to 228.8 GHz, while providing suppression better than 40 dB and 65 dB for the TE10 and TE30 modes, respectively.

A novel K-band transceiver component is introduced in this letter, featured with high-density integration, wide band, and miniaturization. The advanced system-in-package (SiP) technology and double-sided slotted (DS) packaging technology are employed to integrate the core chip and GaAs transceiver chip circuits with four channels together, and therefore to achieve the lightening and thinning of the component. After structure design and thermal design, an 8 x 12 tile style module in K-band has been manufactured and tested. Test results show that the active transmission gains are more than 5 dB, with the input and output reflection loss more than 10 dB from 20.5 to 25 GHz within the relative bandwidth of 21.7%. For the total 96 channels, phase consistency is less than 8.5 degrees, and the amplitude consistency is better than 0.59 dB, demonstrating excellent channel consistency. Additionally, the total dimensions of the fabricated module are only 104 x 78 mm, and the total weight is only 290 g. The architecture design and processing technologies indicate significant potential for miniaturizing and high-density integrated tile-style phased array antennas.

The trade-off between efficient mixing and low energy consumption remains a potential limitation of static mixers. This paper proposes a biomimetic symmetrical helical static mixer (HSSM) based on the internal spiral configuration of snails, aiming to synergistically enhance flow restructuring and efficient scalar transport. This structure induces periodic secondary flow and coherent vortex coupling through gradient curvature and spatially symmetric layout, achieving synergistic optimization of hydrodynamic intensification and pressure drop regulation. Three mixing unit arrangements (HSSM-A, B, C) were designed, and the reliability of the mathematical model was validated through a combination of experimental and CFD numerical simulations. Compared to Kenics mixers, HSSM reduces pressure drop and friction factor by 14.2%–37.8%, decreases G-value by 7.4%–21.1%, and increases mixing index by 15.3%–58.1%. Compared to LPD mixers, it lowers the CoV value by 53.2%–71.2% and achieves a maximum mixing degree of 99.48%. Through numerical simulation, the hydraulic resistance, shear generation characteristics, turbulence evolution, and concentration field development were systematically investigated. Mechanistic analysis indicates that the symmetric spiral structure induces stable counter-rotating vortex pairs and periodic secondary flows, thereby converting axial transport into coupled axial-radial convection. This ordered vortex coupling enhances scalar gradient stretching and interface renewal while suppressing large-scale flow separation, achieving efficient scalar homogenization and controllable energy dissipation. Among all configurations, HSSM-C exhibits the best overall transport performance under complete mixing constraints. The proposed design provides a structural strategy for low-energy, high-efficiency mixers.

A radial power combining frequency doubler architecture based on circularly polarized TE11 and TE21 modes is presented in this article. In the proposed architecture, the input divider operates with the circularly polarized TE11 mode and generates $N$ fundamental signals with a progressive phase shift of $2\pi $ / $N$ . After frequency doubling, the phase progression of the generated second-harmonic signals is doubled to $4\pi $ / $N$ , which matches the phase condition required by a circularly polarized TE21-mode radial combiner. Therefore, the desired second-harmonic signals can be coherently combined by the TE21-mode radial combiner. Furthermore, the operating principle of the circularly polarized TE21-mode septum polarizer is analyzed, and the minimum number of combining paths required for the proposed architecture is derived from the modal excitation condition. It is shown that at least five paths are required to uniquely excite the desired single-handed circularly polarized TE21 mode. To verify the proposed topology, a 170-GHz five-way radial power combining frequency doubler was designed, fabricated, and measured. The measured results show an output power of 100–203 mW over 163–178 GHz. An average power combining efficiency (PCE) of 69% is achieved at an input power of 750 mW.

In this article, a 12-way radial power divider at sub-THz frequencies has been presented. The radial divider is based on circularly polarized TE11 (TE11CP) mode and features low insertion loss. By simplifying the TE11CP-mode transducer and minimizing waveguide flanges, the diameter of the power divider has been reduced to only 28 mm with a height of 17.35 mm. The measured insertion loss averages only 0.3 dB in the frequency range of 209-231 GHz. The isolation between the output ports almost always exceeds 10 dB, and the input return loss is better than 17.8 dB.

In this letter, a W-band power combiner is proposed by exploiting the 3-dB branching waveguide directional couplers and GaN power amplifiers. In the framework, a novel waveguide coupler is realized through an integrated load based on the spoof surface plasmon polaritons (SSPPs), which can effectively realize the absorption of the reflected electromagnetic wave and ensure the high isolation of the output ports. For application and verification, the electromagnetic energy divided by the directional coupler is linked to the grounded coplanar waveguide (GCPW) utilizing waveguide E-plane probes, and the quasi-TEM modes of the wave are attached to the power amplification monoliths using gold wire for energy amplification. Lastly, the two enhanced electromagnetic energy signals are combined by the directional coupler to generate a high-power output electromagnetic wave signal. Experimental results demonstrate that, in the frequency range of 85-100 GHz, the power combiner achieves a maximum output power of approximately 3 W and an average combining efficiency of 84.5%. The proposed prototype possesses the merits of high-temperature tolerance and low-implementation difficulty, which may facilitate further advancements in solid-state devices.

In this letter, a novel waveguide matched load is proposed by exploiting the natural bandstop of spoof surface plasmon polaritons (SSPPs) and the ohmic loss of metal nickel (Ni). The designed waveguide matched load composes of a periodic metallic Ni-based SSPPs and an AlN substrate, which is directly embedded in the center of the E-plane of the standard WR10 rectangular waveguide. Experimental results demonstrate that the electromagnetic waves can be efficiently captured and absorbed within specific frequency ranges, which exhibit an excellent VSWR performance with a return loss better than 18 dB from 85 to 100 GHz. The proposed prototype possesses the merits of high-absorption rate and low-implementation difficulty, which may provide a solution for further breakthroughs in the microwave and millimeter wave solid-state devices.

In this paper we propose a miniaturized large-angle beam scanning phased array antenna using liquid crystal. We innovatively combine the liquid crystal electrically tunable structure with the wide-beam antenna element structure and design an integrated multi-layer antenna structure which realizes large-angle beam scanning within the working bandwidth. The problems of low beam control accuracy and narrow scanning angle of traditional array antenna are effectively addressed. The overall dimensions of the prototype are 74 & times;60 & times;4 mm. Based on the test results of the prototype the gain has reached 20.2 dBi at 27 GHz and the scanning angle was greater than +/- 60 degrees.

Accurate modeling of nonlinear capacitance is significant for physics-based compact modeling of GaN high-electric-mobility transistor (HEMT). For conventional methods, physical modeling of the electron velocity saturation effect (VSE) is often hard to realize in nonlinear capacitance models due to the lack of an effective boundary potential model. In this article, an improved capacitance modeling method that incorporates the VSE within the drift-diffusion regime is proposed, based on the quasi-physical zone division (QPZD) model theory. First, a bias-dependent equivalent length model for the saturation region in the channel induced by VSE, is derived. This model is then integrated into the boundary potential calculation to fully account for VSE. Next, the nonlinear capacitance model is developed by comprehensively considering the VSE in both the effective gate length model and the integral boundary, calculated using the proposed boundary potential model. This approach contributes to the accuracy improvement for the intrinsic capacitance characteristics in the saturation region. Finally, the nonlinear capacitance model, incorporating VSE, is validated using nonlinear capacitance data extracted from multi-bias S-parameters. The verification results demonstrate that the proposed method provides an average accuracy improvement of over 11.5% in nonlinear capacitance characterization.

In this paper, a broadband mode converter operating in the terahertz frequency range is presented, which converts microstrip quasi-TEM mode to SIW TE 20 mode. The converter consists of a microstrip line, a grounded slotline, and an SIW. Initially, the quasi-TEM mode of the microstrip line couples into the grounded slotline. Subsequently, since both the quasislotline mode and the SIW TE 20 mode are odd modes with similar electric field distributions. The longitudinal slotline effectively excites the $\text{TE}_{20}$ mode in the SIW. The propagation characteristics of the converter were investigated through simulations, achieving an insertion loss of $\mathbf{1. 6 8} \mathbf{- 2. 4 5} \mathbf{ d B}$ and a return loss better than $\mathbf{1 5}$ dB over the frequency range of $195-235 \text{GHz}(18.3 \%)$. The converter introduced in this paper holds significant potential for widespread applications in terahertz planar integration.

Diamond surface quality is critical to improve the power and reliability for the hydrogen-terminated diamond (H-diamond) metal-oxide-semiconductor field-effect transistors (MOSFETs). In this Letter, significant surface microdefects were identified in the H-diamond after heat treatment at less than 500 degrees C. These microdefects may be caused by thermal bubbling of free hydrogen introduced during the diamond hydrogenation process and are, therefore, referred to as hydrogen-induced defects in this paper. Based on the peak intensity of the grazing incidence x-ray diffraction spectra, the long-range ordering of the H-diamond surface lattice degrades with the heat treatment temperature. To explain the effect of these hydrogen-induced defects on the drain breakdown of H-diamond MOSFETs, a critical electric field model considering hydrogen bubbling effect is proposed. It indicates that mitigation of hydrogen-induced defects would improve the long-range ordering of H-diamond surface lattice and, thus, increase the H-diamond breakdown strength. Next, two kinds of Al2O3/H-diamond MOSFETs with the same device structure are fabricated by depositing Al2O3 at 150 and 450 degrees C, respectively. It is verified that the H-diamond MOSFET with Al2O3 deposited at 150 degrees C exhibits a drain soft-breakdown strength 4.75 times greater than that of the device with Al2O3 deposited at 450 degrees C. By mitigating the hydrogen-induced defects in the H-diamond surface, a record drain soft-breakdown strength of 1.1 MV/cm is reached for the Al2O3/H-diamond MOSFET. These results would be helpful in improving the power and reliability of H-diamond devices.

An improved 2X-though de-embedding technique is proposed in this article. By adding a matching calibration component, it addresses feed line length limitation in the traditional 2X-through method, as well as the tolerance introduced by truncation of the time-domain signal.

A stacked metamaterial MEMS (meta-MEMS) chip is proposed, which can perfectly absorb electromagnetic waves, convert them into mechanical energy, drive movement of the optical micro-reflectors array, and detect millimeter waves. It is equivalent to using visible light to image a millimeter wave. The meta-MEMS adopts the design of upper and lower chip separation and then stacking to achieve the “dielectric-resonant-air-ground” structure, reduce the thickness of the metamaterial and MEMS structures, and improve the performance of millimeter wave imaging. For verification, we designed and prepared a 94 GHz meta-MEMS focal plane array chip, in which the sum of the thickness of the metamaterial and MEMS structures is only 1/2500 wavelength, the pixel size is less than 1/3 wavelength, but the absorption rate is as high as 99.8%. Moreover, a light readout module was constructed to test the millimeter wave imaging performance. The results show that the response speed can reach 144 Hz and the lens-less imaging resolution is 1.5 mm.