Pub Date : 2025-02-04DOI: 10.1109/TED.2025.3533390
Y. Y. Guo;R. Degraeve;T. Ravsher;D. Garbin;P. Roussel;G. S. Kar;E. Bury;D. Linten;I. Verbauwhede
In this article, we demonstrate physical reservoir computing (RC) in ovonic threshold switching (OTS) devices. We show that SiGeAsSe OTS is suitable as a physical reservoir because of the nonlinear change in the number of delocalized defects. With the combination of phase space reconstruction (PSR), our algorithm can project data into high-dimensional spaces, thereby enhancing the distinguishability of the data. Such ability is suitable for high-accuracy authentication and classification. Our algorithm can be implemented using both crossbar arrays or individual devices, and achieves a significantly low (0.08%) equal error rate (EER) on gait authentication in simulation. Furthermore, we validated our concept by successfully implementing the algorithm on nine hardware OTS devices and achieved an EER of 4.2% on gait authentication. The low leakage current level of OTS, the fast learning of RC, and interval-based readout responses all contribute to a significantly reduced energy consumption of our proposed method.
{"title":"Ovonic Threshold Switching for Ultralow Energy Physical Reservoir Computing","authors":"Y. Y. Guo;R. Degraeve;T. Ravsher;D. Garbin;P. Roussel;G. S. Kar;E. Bury;D. Linten;I. Verbauwhede","doi":"10.1109/TED.2025.3533390","DOIUrl":"https://doi.org/10.1109/TED.2025.3533390","url":null,"abstract":"In this article, we demonstrate physical reservoir computing (RC) in ovonic threshold switching (OTS) devices. We show that SiGeAsSe OTS is suitable as a physical reservoir because of the nonlinear change in the number of delocalized defects. With the combination of phase space reconstruction (PSR), our algorithm can project data into high-dimensional spaces, thereby enhancing the distinguishability of the data. Such ability is suitable for high-accuracy authentication and classification. Our algorithm can be implemented using both crossbar arrays or individual devices, and achieves a significantly low (0.08%) equal error rate (EER) on gait authentication in simulation. Furthermore, we validated our concept by successfully implementing the algorithm on nine hardware OTS devices and achieved an EER of 4.2% on gait authentication. The low leakage current level of OTS, the fast learning of RC, and interval-based readout responses all contribute to a significantly reduced energy consumption of our proposed method.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1112-1117"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-04DOI: 10.1109/TED.2025.3534157
Wenbo Ye;Junmin Zhou;Han Gao;Haowen Guo;Yitian Gu;Xinbo Zou
In this study, the device properties of gallium nitride (GaN) enhancement-mode (E-mode) recessed-gate high-electron-mobility transistor (HEMT) are thoroughly characterized and investigated for low-noise amplifier (LNA) applications. Through low-damage argon-based neutral beam etching (Ar-NBE) technology, the recessed-gate HEMT achieves a positive voltage threshold (${V} _{text {TH}}$ ) of 0.5 V, a maximum transconductance (gm) of 148 mS/mm, and an on-state gate leakage current (${I} _{text {G}}$ ) of 2.39 nA/mm. The device reveals a 1.48-dB minimum noise figure (NFmin), a 14.43-dB associated gain (${G} _{text {a}}$ ), and a 40.2-$Omega $ equivalent noise resistance (${R} _{text {N}}$ ), at a working frequency of 2 GHz. As the frequency increases to 3.5 GHz, the NFmin slightly increases to 1.95 dB. In addition, the device obtained a cutoff frequency (${f} _{text {T}}$ /${f} _{text {MAX}}$ ) of 9.6/27.8 GHz and an input third-order interception point (IIP3) of 10.3 dBm at 2 GHz. This work provides a promising strategy for the implementation of high-performance E-mode LNAs.
{"title":"1.48-dB-Noise Figure E-Mode Recessed-Gate GaN MOSHEMT by Argon-Based Neutral Beam Etching for LNA Applications","authors":"Wenbo Ye;Junmin Zhou;Han Gao;Haowen Guo;Yitian Gu;Xinbo Zou","doi":"10.1109/TED.2025.3534157","DOIUrl":"https://doi.org/10.1109/TED.2025.3534157","url":null,"abstract":"In this study, the device properties of gallium nitride (GaN) enhancement-mode (E-mode) recessed-gate high-electron-mobility transistor (HEMT) are thoroughly characterized and investigated for low-noise amplifier (LNA) applications. Through low-damage argon-based neutral beam etching (Ar-NBE) technology, the recessed-gate HEMT achieves a positive voltage threshold (<inline-formula> <tex-math>${V} _{text {TH}}$ </tex-math></inline-formula>) of 0.5 V, a maximum transconductance (gm) of 148 mS/mm, and an <sc>on</small>-state gate leakage current (<inline-formula> <tex-math>${I} _{text {G}}$ </tex-math></inline-formula>) of 2.39 nA/mm. The device reveals a 1.48-dB minimum noise figure (NFmin), a 14.43-dB associated gain (<inline-formula> <tex-math>${G} _{text {a}}$ </tex-math></inline-formula>), and a 40.2-<inline-formula> <tex-math>$Omega $ </tex-math></inline-formula> equivalent noise resistance (<inline-formula> <tex-math>${R} _{text {N}}$ </tex-math></inline-formula>), at a working frequency of 2 GHz. As the frequency increases to 3.5 GHz, the NFmin slightly increases to 1.95 dB. In addition, the device obtained a cutoff frequency (<inline-formula> <tex-math>${f} _{text {T}}$ </tex-math></inline-formula>/<inline-formula> <tex-math>${f} _{text {MAX}}$ </tex-math></inline-formula>) of 9.6/27.8 GHz and an input third-order interception point (IIP3) of 10.3 dBm at 2 GHz. This work provides a promising strategy for the implementation of high-performance E-mode LNAs.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1035-1040"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-04DOI: 10.1109/TED.2025.3531824
Allen L. Garner;N. R. Sree Harsha
The space-charge-limited current (SCLC) in a vacuum diode is given by the Child-Langmuir law (CLL), whose electric potential ${phi }text {(}{x}text {)}propto text {(}{x} /{D}text {)}^{{4} /{3}}$ , where x is the spatial coordinate across the gap and D is the gap separation distance. For a collisional diode, SCLC is given by the Mott-Gurney law (MGL) and ${phi }text {(} {x}text {)}propto text {(}{x} /{D}text {)}^{{3} / {2}}$ . Here, we apply a capacitance argument for SCLC and use the transit time from a recent exact solution for collisional SCLC to show that ${phi }text {(}{x}text {)}propto text {(}{x}/{D}text {)}^{xi }$ for a general collisional gap, where ${4} / {3}le xi le {3}/ {2}$ . Furthermore, $xi $ is strictly a function of $nu {T}$ , where $nu $ is the collision frequency and T is the electron transit time. Using this definition of $xi $ , we estimate the spatial dependence of the electron velocity and use the gap capacitance to derive an analytic equation for collisional SCLC that agrees within ~4.5% of the exact solution that requires solving parametrically through T. This analytic equation for general $xi $ asymptotically recovers the CLL as ${nu }to {0}$ and the MGL as ${nu } to infty $ . Matching these limits shows that ${xi }approx {1.40}$ and ${V}propto {D}^{{2}}{nu }^{{2}}$ at the transition from a vacuum to a collisional diode for any device condition.
{"title":"The Implications of Collisions on the Spatial Profile of Electric Potential and the Space-Charge-Limited Current","authors":"Allen L. Garner;N. R. Sree Harsha","doi":"10.1109/TED.2025.3531824","DOIUrl":"https://doi.org/10.1109/TED.2025.3531824","url":null,"abstract":"The space-charge-limited current (SCLC) in a vacuum diode is given by the Child-Langmuir law (CLL), whose electric potential <inline-formula> <tex-math>${phi }text {(}{x}text {)}propto text {(}{x} /{D}text {)}^{{4} /{3}}$ </tex-math></inline-formula>, where x is the spatial coordinate across the gap and D is the gap separation distance. For a collisional diode, SCLC is given by the Mott-Gurney law (MGL) and <inline-formula> <tex-math>${phi }text {(} {x}text {)}propto text {(}{x} /{D}text {)}^{{3} / {2}}$ </tex-math></inline-formula>. Here, we apply a capacitance argument for SCLC and use the transit time from a recent exact solution for collisional SCLC to show that <inline-formula> <tex-math>${phi }text {(}{x}text {)}propto text {(}{x}/{D}text {)}^{xi }$ </tex-math></inline-formula> for a general collisional gap, where <inline-formula> <tex-math>${4} / {3}le xi le {3}/ {2}$ </tex-math></inline-formula>. Furthermore, <inline-formula> <tex-math>$xi $ </tex-math></inline-formula> is strictly a function of <inline-formula> <tex-math>$nu {T}$ </tex-math></inline-formula>, where <inline-formula> <tex-math>$nu $ </tex-math></inline-formula> is the collision frequency and T is the electron transit time. Using this definition of <inline-formula> <tex-math>$xi $ </tex-math></inline-formula>, we estimate the spatial dependence of the electron velocity and use the gap capacitance to derive an analytic equation for collisional SCLC that agrees within ~4.5% of the exact solution that requires solving parametrically through T. This analytic equation for general <inline-formula> <tex-math>$xi $ </tex-math></inline-formula> asymptotically recovers the CLL as <inline-formula> <tex-math>${nu }to {0}$ </tex-math></inline-formula> and the MGL as <inline-formula> <tex-math>${nu } to infty $ </tex-math></inline-formula>. Matching these limits shows that <inline-formula> <tex-math>${xi }approx {1.40}$ </tex-math></inline-formula> and <inline-formula> <tex-math>${V}propto {D}^{{2}}{nu }^{{2}}$ </tex-math></inline-formula> at the transition from a vacuum to a collisional diode for any device condition.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1419-1426"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-04DOI: 10.1109/TED.2025.3534174
Arno Kirchbrücher;Gerrit Lükens;Carsten Beckmann;Jasmin Ehrler;Qi Shu;Jens Wieben;Thorsten Zweipfennig;Holger Kalisch;Andrei Vescan
Conventional AlGaN/GaN metal-insulator–semiconductor heterostructure field-effect transistors (MISHFETs) are affected by trapped charges at the dielectric/AlGaN interface causing instabilities and quasi-permanent shifts of the threshold voltage. In this work, we investigate the charging and especially discharging processes of these interface states in AlGaN/GaN/AlGaN double-heterostructure (DH) MISHFET with an Al2O3 gate dielectric. Appropriately designed, these dopant-free devices contain a polarization-induced 2-D electron gas (2DEG) as well as a 2-D hole gas (2DHG). After applying large gate biases, the Al2O3/AlGaN interface is known to persistently capture electrons from the 2DEG in deep states resulting in a positive threshold voltage shift. Here, we demonstrate that the interface can be actively discharged by carriers from the 2DHG when a sufficiently large negative gate bias is applied, resulting in a negative threshold voltage shift. These mechanisms of charge trapping and detrapping are correlated to the design parameters and biasing conditions of the device. In addition, we investigate low- and high-temperature characteristics and show that these devices can be operated like a memory-type component.
{"title":"Threshold Voltage Control in AlGaN/GaN/AlGaN Double-Heterostructure MISHFET Utilizing 2-D Electron and Hole Gases","authors":"Arno Kirchbrücher;Gerrit Lükens;Carsten Beckmann;Jasmin Ehrler;Qi Shu;Jens Wieben;Thorsten Zweipfennig;Holger Kalisch;Andrei Vescan","doi":"10.1109/TED.2025.3534174","DOIUrl":"https://doi.org/10.1109/TED.2025.3534174","url":null,"abstract":"Conventional AlGaN/GaN metal-insulator–semiconductor heterostructure field-effect transistors (MISHFETs) are affected by trapped charges at the dielectric/AlGaN interface causing instabilities and quasi-permanent shifts of the threshold voltage. In this work, we investigate the charging and especially discharging processes of these interface states in AlGaN/GaN/AlGaN double-heterostructure (DH) MISHFET with an Al2O3 gate dielectric. Appropriately designed, these dopant-free devices contain a polarization-induced 2-D electron gas (2DEG) as well as a 2-D hole gas (2DHG). After applying large gate biases, the Al2O3/AlGaN interface is known to persistently capture electrons from the 2DEG in deep states resulting in a positive threshold voltage shift. Here, we demonstrate that the interface can be actively discharged by carriers from the 2DHG when a sufficiently large negative gate bias is applied, resulting in a negative threshold voltage shift. These mechanisms of charge trapping and detrapping are correlated to the design parameters and biasing conditions of the device. In addition, we investigate low- and high-temperature characteristics and show that these devices can be operated like a memory-type component.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1131-1140"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-04DOI: 10.1109/TED.2025.3531322
Omar Kittaneh;Sara Helal;M. A. Majid
This work generalizes the inverse power law-normal (IPL-normal) model for complete data to right-censored data, assuming that the coefficient of variation remains constant and free of stress. The maximum likelihood (ML) estimating equations of the model’s accelerating parameters and the general coefficient of variation are derived using new trivial but fundamental identities. The ML estimating equation of the general coefficient of variation is explicit and generalizes its counterpart for complete data, which was previously introduced. The ML method is compared with the classical least squares (LS) technique. Although the ML method is laborious and numerically sensitive, this article favors ML over LS for a drastic reason that only ML can estimate the general coefficient of variation, but it still recommends using both the methods for some other reasons. The generalized IPL-normal model is used to precisely specify the life model of organic light-emitting diodes based on a standard real data of complete samples of lives which was discussed in several previous works but censored in this work.
{"title":"The Inverse Power Law-Normal Model for Right-Censored Data With Application to Life Prediction of Organic Light-Emitting Diodes","authors":"Omar Kittaneh;Sara Helal;M. A. Majid","doi":"10.1109/TED.2025.3531322","DOIUrl":"https://doi.org/10.1109/TED.2025.3531322","url":null,"abstract":"This work generalizes the inverse power law-normal (IPL-normal) model for complete data to right-censored data, assuming that the coefficient of variation remains constant and free of stress. The maximum likelihood (ML) estimating equations of the model’s accelerating parameters and the general coefficient of variation are derived using new trivial but fundamental identities. The ML estimating equation of the general coefficient of variation is explicit and generalizes its counterpart for complete data, which was previously introduced. The ML method is compared with the classical least squares (LS) technique. Although the ML method is laborious and numerically sensitive, this article favors ML over LS for a drastic reason that only ML can estimate the general coefficient of variation, but it still recommends using both the methods for some other reasons. The generalized IPL-normal model is used to precisely specify the life model of organic light-emitting diodes based on a standard real data of complete samples of lives which was discussed in several previous works but censored in this work.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1229-1234"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521384","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-04DOI: 10.1109/TED.2025.3530867
Zilin Wang;Wenjie Zhao;Zheyao Wang
CuSn-Cu bonding is widely used in 2.5-D integration due to the ease of operation. However, it suffers from high bond resistance due to the high resistivity of Cu-Sn intermetallic compounds (IMCs) and the high contact resistance due to the Kirkendall voids at the Cu/IMC interfaces. We report a low-resistance CuSn-Cu interconnect fabricated by bonding a porous Cu bump and a Cu-Sn pad with a 200-nm Sn layer. The porous Cu bump, fabricated by treating normal Cu-Sn bumps with oxygen and formic acid, is bonded with the Cu-Sn pad through solid-state bonding at 100 °C under a 5-MPa pressure, followed by 250 °C annealing. The bonding pressure presses the porous Cu to distribute uniformly into the Sn layer, avoiding the Kirkendall voids by providing the Cu atoms locally instead of all from the Cu/Sn interfaces for Cu-Sn reactions. The IMC resistance is also reduced as the 200-nm Sn layer forms an ultrathin IMC layer. A large-scale bump array has been bonded successfully, and a single bond resistance of 6.5 m$Omega $ and a specific contact resistivity (SCR) of $1.0times 10^{-}9 ~Omega cdot $ cm2 have been obtained.
{"title":"Low-Resistance Interconnects by Bonding Porous Cu Bumps and Ultrathin Cu/Sn Pads","authors":"Zilin Wang;Wenjie Zhao;Zheyao Wang","doi":"10.1109/TED.2025.3530867","DOIUrl":"https://doi.org/10.1109/TED.2025.3530867","url":null,"abstract":"CuSn-Cu bonding is widely used in 2.5-D integration due to the ease of operation. However, it suffers from high bond resistance due to the high resistivity of Cu-Sn intermetallic compounds (IMCs) and the high contact resistance due to the Kirkendall voids at the Cu/IMC interfaces. We report a low-resistance CuSn-Cu interconnect fabricated by bonding a porous Cu bump and a Cu-Sn pad with a 200-nm Sn layer. The porous Cu bump, fabricated by treating normal Cu-Sn bumps with oxygen and formic acid, is bonded with the Cu-Sn pad through solid-state bonding at 100 °C under a 5-MPa pressure, followed by 250 °C annealing. The bonding pressure presses the porous Cu to distribute uniformly into the Sn layer, avoiding the Kirkendall voids by providing the Cu atoms locally instead of all from the Cu/Sn interfaces for Cu-Sn reactions. The IMC resistance is also reduced as the 200-nm Sn layer forms an ultrathin IMC layer. A large-scale bump array has been bonded successfully, and a single bond resistance of 6.5 m<inline-formula> <tex-math>$Omega $ </tex-math></inline-formula> and a specific contact resistivity (SCR) of <inline-formula> <tex-math>$1.0times 10^{-}9 ~Omega cdot $ </tex-math></inline-formula> cm2 have been obtained.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1276-1281"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-04DOI: 10.1109/TED.2025.3535461
Justin P. Heimerl;Harrison P. Lee;Christoph Jungemann;John D. Cressler
We numerically investigate a new asymmetric superjunction (SJ) collector profile implemented in the context of high-speed 90-nm silicon-germanium heterojunction bipolar transistors (SiGe HBTs). After reviewing past results on vertical SJ SiGe HBT collector design, we present a new SJ profile, which was generated using Bayesian optimization techniques. This new design provides improvements over both a standard collector profile and a standard SJ profile. The new profile provides a substantial increase in maximum cutoff frequency (${f}_{T}$ ), with a slight increase in open-base-collector–emitter breakdown voltage ($text {BV}_{text {CEO}}$ ). Then, we demonstrate, via hydrodynamic (HD) simulations and numerically solving the Boltzmann transport equation (BTE), that the simulated results are physically grounded. Finally, we address control and manufacturing concerns.
{"title":"Revisiting the Vertical Superjunction in SiGe HBT Performance Optimization","authors":"Justin P. Heimerl;Harrison P. Lee;Christoph Jungemann;John D. Cressler","doi":"10.1109/TED.2025.3535461","DOIUrl":"https://doi.org/10.1109/TED.2025.3535461","url":null,"abstract":"We numerically investigate a new asymmetric superjunction (SJ) collector profile implemented in the context of high-speed 90-nm silicon-germanium heterojunction bipolar transistors (SiGe HBTs). After reviewing past results on vertical SJ SiGe HBT collector design, we present a new SJ profile, which was generated using Bayesian optimization techniques. This new design provides improvements over both a standard collector profile and a standard SJ profile. The new profile provides a substantial increase in maximum cutoff frequency (<inline-formula> <tex-math>${f}_{T}$ </tex-math></inline-formula>), with a slight increase in open-base-collector–emitter breakdown voltage (<inline-formula> <tex-math>$text {BV}_{text {CEO}}$ </tex-math></inline-formula>). Then, we demonstrate, via hydrodynamic (HD) simulations and numerically solving the Boltzmann transport equation (BTE), that the simulated results are physically grounded. Finally, we address control and manufacturing concerns.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1523-1527"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143580929","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this article, a diode magnetron injection gun (MIG) with an external anode for Ka-band gyrotron traveling wave tubes (gyro-TWTs) is designed, whose anode is demountable. The magnet bore radius constrains the maximum radius of an MIG. To adapt to a magnet with smaller bore radius, a compact MIG with smaller radial size is necessary. Compared to the previous conventional Ka-band MIG, the gun effective radius constrained by the magnet bore is reduced from 35 to 19.5 mm, which is a 44% reduction. For a better high-voltage insulation performance of the MIG, studies are conducted on its three-layer dielectric separately: corrugated ceramic is used between the cathode and the anode, and the external anode is immersed in insulating oil. Additionally, a capture structure is designed to suppress backflow electrons. The velocity ratio and transverse velocity spread of the final MIG are 1.36% and 2.19%, respectively. Due to the replaceability of its anode, the MIG has potential for application in multiband or pulsed-continuous wave gyro-TWT operations. To verify the high-voltage stability of the MIG with an external anode, a high-voltage insulation experiment is carried out. The experiment reveals that no breakdown occurs at the operating voltage of 48 kV in the MIG.
{"title":"Investigation of a Magnetron Injection Gun With an External Anode for Ka-Band Gyro-TWT","authors":"Boxin Dai;Wei Jiang;Binyang Han;Chaoxuan Lu;Yelei Yao;Zewei Wu;Jianwei Zhou;Guo Liu;Jianxun Wang;Yong Luo","doi":"10.1109/TED.2025.3534181","DOIUrl":"https://doi.org/10.1109/TED.2025.3534181","url":null,"abstract":"In this article, a diode magnetron injection gun (MIG) with an external anode for Ka-band gyrotron traveling wave tubes (gyro-TWTs) is designed, whose anode is demountable. The magnet bore radius constrains the maximum radius of an MIG. To adapt to a magnet with smaller bore radius, a compact MIG with smaller radial size is necessary. Compared to the previous conventional Ka-band MIG, the gun effective radius constrained by the magnet bore is reduced from 35 to 19.5 mm, which is a 44% reduction. For a better high-voltage insulation performance of the MIG, studies are conducted on its three-layer dielectric separately: corrugated ceramic is used between the cathode and the anode, and the external anode is immersed in insulating oil. Additionally, a capture structure is designed to suppress backflow electrons. The velocity ratio and transverse velocity spread of the final MIG are 1.36% and 2.19%, respectively. Due to the replaceability of its anode, the MIG has potential for application in multiband or pulsed-continuous wave gyro-TWT operations. To verify the high-voltage stability of the MIG with an external anode, a high-voltage insulation experiment is carried out. The experiment reveals that no breakdown occurs at the operating voltage of 48 kV in the MIG.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1448-1454"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Dynamic resistance degradation, which is severely affected by the trapping effect, is a critical challenge for lateral AlGaN/GaN power devices, especially when operating in high-voltage and high-frequency applications. In this brief, an enhancement-mode p-GaN gate HEMT with a drain-side thin p-GaN (DST) structural design is proposed. The DST design can suppress the dynamic resistance degradation by injecting holes from the drain-side p-GaN. Meanwhile, by thinning the p-GaN layer, the on-state current conduction characteristics of the DST-HEMT can be greatly improved. The thinning process of the drain-side p-GaN is carried out simultaneously with the source/drain ohmic contact region etching process, which is well compatible with the existing process platform. By performing circuit-level dynamic resistance testing, GaN-on-sapphire DST-HEMT achieves minimal dynamic resistance degradation under 1200-V off-state bias conditions, which is comparable to the test results in vertical GaN-on-GaN devices. In addition, the dynamic switching capability of the device is also demonstrated. These results reveal the notable potential of GaN-on-sapphire DST-HEMTs for high-voltage and high-power applications.
{"title":"Suppression of Dynamic Resistance Degradation in 1200-V GaN-on-Sapphire E-Mode GaN HEMTs by Drain-Side Thin p-GaN Design","authors":"Wenfeng Wang;Feng Zhou;Junfan Qian;Can Zou;Weizong Xu;Fangfang Ren;Dong Zhou;Dunjun Chen;Yuanyang Xia;Leke Wu;Yiheng Li;Tinggang Zhu;Youdou Zheng;Rong Zhang;Hai Lu","doi":"10.1109/TED.2025.3534746","DOIUrl":"https://doi.org/10.1109/TED.2025.3534746","url":null,"abstract":"Dynamic resistance degradation, which is severely affected by the trapping effect, is a critical challenge for lateral AlGaN/GaN power devices, especially when operating in high-voltage and high-frequency applications. In this brief, an enhancement-mode p-GaN gate HEMT with a drain-side thin p-GaN (DST) structural design is proposed. The DST design can suppress the dynamic resistance degradation by injecting holes from the drain-side p-GaN. Meanwhile, by thinning the p-GaN layer, the on-state current conduction characteristics of the DST-HEMT can be greatly improved. The thinning process of the drain-side p-GaN is carried out simultaneously with the source/drain ohmic contact region etching process, which is well compatible with the existing process platform. By performing circuit-level dynamic resistance testing, GaN-on-sapphire DST-HEMT achieves minimal dynamic resistance degradation under 1200-V <sc>off</small>-state bias conditions, which is comparable to the test results in vertical GaN-on-GaN devices. In addition, the dynamic switching capability of the device is also demonstrated. These results reveal the notable potential of GaN-on-sapphire DST-HEMTs for high-voltage and high-power applications.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1537-1540"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143580901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-04DOI: 10.1109/TED.2024.3520544
Yue Hu;Yuxin Wang;Chuanxu Li;Hao Ren;Liang Lou
This article proposes a temperature-humidity assembled sensor based on AlScN surface acoustic wave (SAW) technology for simultaneous monitoring of temperature and humidity. The assembled sensor includes two SAW resonators with frequencies of 475 and 493.675 MHz. The AlScN thin film serves as the piezoelectric film for the temperature and humidity sensor. Graphene oxide (GO) is used as the humidity-sensitive film in the sensor. The humidity sensor can detect a frequency shift of 587.5 kHz within the range of 20%–90% RH, with sensitivities of 4.75 kHz/%RH at low humidity (20%–50% RH), 9.33 kHz/%RH at medium humidity (50%–80% RH), and 16.3 kHz/%RH at high humidity levels (80%–90% RH). The assembled sensor demonstrates excellent stability and repeatability, with a response time and recovery time of 8.5 and 6 s, respectively. The temperature and humidity sensors have a similar temperature coefficient of frequency (TCF). The sensor measures ambient temperature using the temperature sensor and employs a decoupling algorithm to mitigate the impact of temperature on the humidity sensor, improving its accuracy.
{"title":"SAW Temperature and Humidity Assembled Sensor Based on AlScN Piezoelectric Thin Film","authors":"Yue Hu;Yuxin Wang;Chuanxu Li;Hao Ren;Liang Lou","doi":"10.1109/TED.2024.3520544","DOIUrl":"https://doi.org/10.1109/TED.2024.3520544","url":null,"abstract":"This article proposes a temperature-humidity assembled sensor based on AlScN surface acoustic wave (SAW) technology for simultaneous monitoring of temperature and humidity. The assembled sensor includes two SAW resonators with frequencies of 475 and 493.675 MHz. The AlScN thin film serves as the piezoelectric film for the temperature and humidity sensor. Graphene oxide (GO) is used as the humidity-sensitive film in the sensor. The humidity sensor can detect a frequency shift of 587.5 kHz within the range of 20%–90% RH, with sensitivities of 4.75 kHz/%RH at low humidity (20%–50% RH), 9.33 kHz/%RH at medium humidity (50%–80% RH), and 16.3 kHz/%RH at high humidity levels (80%–90% RH). The assembled sensor demonstrates excellent stability and repeatability, with a response time and recovery time of 8.5 and 6 s, respectively. The temperature and humidity sensors have a similar temperature coefficient of frequency (TCF). The sensor measures ambient temperature using the temperature sensor and employs a decoupling algorithm to mitigate the impact of temperature on the humidity sensor, improving its accuracy.","PeriodicalId":13092,"journal":{"name":"IEEE Transactions on Electron Devices","volume":"72 3","pages":"1315-1322"},"PeriodicalIF":2.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143521539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：