He Guan

This paper presents a K-band four-channel phased array beamformer with temperature compensation in 65 nm CMOS for 5G and satellite communications. The beamformer includes a four-way power divider/combiner, four RF channels, and digital control circuits. Each RF channel comprises a receive chain, a transmit chain, and a pair of receive/transmit (TX/RX) single-pole double-throw (SPDT) switches. The receive chain consists of a low-noise amplifier (LNA), a six-bit reflective-type phase shifter (RTPS), a drive amplifier (DA), two temperature-compensation attenuators (TCAs), and a six-bit attenuator (ATT); the transmit chain integrates a power amplifier (PA), two TCAs, a six-bit RTPS, a DA, and a six-bit ATT. Measurements show the chip exhibits 0–4.5 dB gain, noise figure (NF) < 7.8 dB, root mean square (RMS) phase error < 3.5°, and RMS gain error < 0.4 dB in receive mode operating in 19–23 GHz. In transmit mode operating in 21–23 GHz, it provides 6–10 dB gain range, RMS phase error < 3.4°, RMS gain error < 0.25 dB, and output power at 1 dB compression point (OP1dB) > 6.5 dBm. In addition, the receive and transmit gain variations are within 0.8 dB and 0.4 dB, respectively, when temperature ranges from −55 °C to 85 °C. With a compact footprint of 3.5 × 4.8 mm², the beamformer consumes 110 mW (receive) and 190 mW (transmit) DC power per channel.

This study employs a systematic numerical investigation framework centered on TCAD simulation to address thermal management challenges in InP-based In0.53Ga0.47As planar Gunn diodes. Through comprehensive two-dimensional thermal-electrical co-simulation, we evaluate substrate materials (InP, Si, diamond) and thermal films (diamond, graphene) on device performance. The simulation results indicate that the diamond substrate can significantly reduce the operating temperature of the device, increase the oscillation frequency, and enhance the output power. Both diamond and graphene thermal films effectively mitigate thermal resistance, improving output characteristics through temperature reduction exceeding 150 K. Furthermore, we propose a novel diamond-graphene composite thermal management structure that achieves a significant temperature reduction from 607.3 K to 373.3 K (ΔT = 234 K), simultaneously enhancing oscillation frequency to 203.6 GHz and power conversion efficiency to 0.80%. These findings establish material-driven thermal optimization principles for high-frequency terahertz sources, suggesting significant potential for device performance enhancement.

This article presents the design and implementation of a 6–18 GHz GaN monolithic microwave integrated circuit (MMIC) power amplifier (PA). A two-stage cascaded reactive matching network structure based on transistor stacking technology is employed to achieve circuit gain, and a multi-cell combination is used in the final stage to simultaneously achieve high power and high efficiency. For demonstration, a prototype of the proposed PA with an area of 4.5 × 3.4 mm² is fabricated in a 0.1 µm GaN-on-Si high-electron-mobility transistor (HEMT) process. The measured results of the GaN PA show a small signal gain of 25–29 dB, an output power of 40.8–42.5 dBm, and a power-added efficiency (PAE) of 27–38% in the operating frequency range of 6–18 GHz.

In this article, we propose a synthesis method for wideband high-selectivity bandpass filters (BPFs) based on the tri-mode lumped-element resonator in heterogeneous low-temperature co-fired ceramic (LTCC) substrates. The tri-mode lumped-element resonator is analyzed to generate three resonances and two controllable transmission zeros without using cross couplings, showing a simple topology. A Remez-like numerical iterative algorithm is adopted to construct the polynomials of the tri-mode resonator in the bandpass domain, which can eliminate the need for complex transmission-function polynomial derivations and high-order iterations, allowing direct calculation of lumped-element values by the polynomials without optimization. Furthermore, the extracted-pole technique is used to expand the filter design. A comprehensive synthesis workflow is proposed for high-order filtering responses by integrating the extracted-pole technique and the tri-mode lumped-element resonator. The resultant filter is capable of achieving the desired responses with a wide bandwidth and multiple transmission zeros in a simple and practically achievable topology. For demonstration, a BPF covering 5G-n77 band (3.3–4.2 GHz) is implemented using a heterogeneous LTCC substrate. The measured result shows a low loss of 0.76dB and the circuit size is only $1.3\times 1.09\times 0.53$ mm³ (or $0.016\lambda _{{0}} \times 0.014\lambda _{{0}} \times 0.007\lambda _{{0}}$ ), which is suitable for 5G applications.

This article presents a 24–30-GHz wideband transmit–receive (T/R) front-end module (FEM) for 5G millimeter-wave (mm-Wave) applications in the 150-nm GaN-on-SiC HEMT process. The power amplifier in the transmitting chain is constructed of two fully symmetrical push–pull amplifiers that are designed using transformer-based baluns. To improve efficiency, the harmonic control technique is used in the output impedance matching network, where the second-harmonic impedance can be optimized without affecting the fundamental response. A high-order stacked-transistor single-pole double-throw (SPDT) switch is designed with a low insertion loss of 0.75dB and a high input 1-dB compression point (IP ${}_{\mathrm {1dB}}$ ) of over 44dBm, ensuring almost no gain compression in transmitting mode. The low-noise amplifier (LNA) in the receiving chain is designed into a three-stage topology, achieving a flat gain and low-noise figure (NF) within the operating bandwidth. In measurements, the proposed FEM operating across 24–30 GHz demonstrates a maximum saturated power ( $P_{\mathrm {sat}}$ ) of 37.5dBm and peak power-added efficiency (PAE) of 32% in the transmitting mode. An error vector magnitude (EVM) of 5.75% and an average output power of 25.7 dBm with a 64-quadrature amplitude modulation (QAM) 5G FR2 modulated signal of 100-MHz bandwidth at 27 GHz are realized. In the receiving mode, the FEM demonstrates a NF of less than 3.1 dB, a flat small-signal gain of 20–21.3 dB, and an output 1-dB compression point (OP ${}_{\mathrm {1dB}}$ ) of 20.4 dBm.

The etching process is pivotal in the fabrication of GaN-based devices, directly influencing their performance. This study presents an enhanced low-damage hybrid cyclic etching technology for AlGaN/GaN heteroepitaxy structures, combining oxygen plasma oxidation with HCl etching to achieve atomic-scale precision. By carefully controlling the oxidation layer thickness, this method preserves a low surface roughness of 0.228 nm and achieves an etching rate of 0.45 nm/cycle with excellent consistency. Compared to conventional ICP dry etching and KOH wet etching evaluated under identical experimental conditions, the proposed technique significantly reduces etching-induced-damage. When applied to the ohmic contact formation of AlGaN/GaN HEMTs using the same metallization scheme and annealing process, devices fabricated with the new hybrid etching method demonstrated a significantly lower contact resistance of 0.76 Ω7mm compared to those using conventional ICP etching (1.21 Ω7mm) or no additional etching after SiN removal (1.37 Ω7mm). Furthermore, these devices exhibited a threshold voltage of -6.5 V, a saturation drain current density of 880.55 mA/mm, and a knee voltage of approximately 4 V. This method also demonstrates potential for recessed gate etching and post-dry-etching damage repair, offering a promising solution for low-roughness surfaces processing with broad application prospects.

Gallium nitride high-electron mobility transistors (GaN-on-Si HEMTs) on silicon substrates exhibit significant potential in the field of radio frequency devices due to its high frequency, high-speed, and low-cost advantages. Nevertheless, the thermal stress caused by the mismatch in lattice parameters and the coefficient of thermal expansion between the Si substrate and GaN epitaxial layer constrains the device performance. Due to different temperature distributions in different scenarios, there is an urgent need for a model that can balance the contradiction between epitaxial growth and device operation for the joint design of GaN-on-Si HEMTs’ epitaxial structures. This work presents thermal stress formulas specifically designed for GaN epitaxial growth and device operation, based on Townsend’s theory of multilayer film curvature and stress. Furthermore, our work innovatively proposes a multiscenario coupled GaN-on-Si HEMTs’ epitaxial structure design model based on multiobjective optimization. The model represents a combined objective function incorporating thermal stress experienced during device operation and epitaxial growth, as well as the total thermal resistance and bulk thermal conductivity of the epitaxial layer. Through simulation validation, the model has been proved to effectively calculate the optimized epitaxial structure parameters across multiple scenarios to meet diverse requirements. And it can effectively reduce the operating temperature. This model can accelerate the process of GaN epitaxial design and provide an effective method for enhancing the performance of GaN-on-Si HEMTs.

This article presents, for the first time, the theoretical analysis and design methodology for realizing broadband high-efficiency continuous class-F/F⁻¹ (CCF/F⁻¹) power amplifiers (PAs) with active second harmonic injection technique. By introducing an auxiliary PA operated at the second harmonic frequency of the desired band, the optimal phase shift parameter between the drain voltage and current of the CCF/F⁻¹ PA can be properly selected. In this way, the total drain efficiency (DE), rather than just the DE of the main PA, can be improved. That is to say, the power consumption of the auxiliary PA has been considered carefully. The calculated results show that the maximum total DEs in the CCF and CCF⁻¹ modes are increased to 98.16% and 93.16%, respectively. At the same time, the maximum output powers are also enhanced, being 1.323 and 1.319 times larger than those in the traditional CCF and CCF⁻¹ modes, respectively. On the other hand, the specific phase relationship between the main PA and auxiliary PA at the input node is derived. Based on the detailed design guidelines, using two GaN HEMT transistors, a proof-of-concept CCF PA with active second harmonic injection technique working across 1.1–1.65GHz (a fractional bandwidth of 40%) has been designed and fabricated. Experimental results show that the implemented PA achieves the measured output power of 41.2–42.1dBm, the total DE of 73.7%–84.9% with an average of 78.4%, and the total power-added efficiency (PAE) of 68.6%–79.1%.